Economic Analysis for the
Final Stage 2 Disinfectants
and Disinfection Byproducts
Rule
-------
Office of Water (4606-M) EPA 815-R-05-010 December 2005 www.epa.gov/safewater
-------
Contents
Executive Summary
Executive Summary ES-1
ES.l Need for the Rule ES-1
ES.2 Consideration of Regulatory Alternatives ES-2
ES.3 Summary of the Stage 2 DBPR ES-3
ES.4 Systems Subjectto the Stage 2 DBPR ES-8
ES.5 National Benefits and Costs of the Stage 2 DBPR Preferred Regulatory
Alternative ES-10
ES.5.1 Derivation of the Stage 2 DBPR Compliance Forecast and Consequent
Reductions in DBFs ES-13
ES.5.2 Derivation of Benefits ES-18
ES.5.3 Derivation of Costs ES-22
ES.6 Estimated Impacts on Household Costs ES-23
ES.7 Comparison of Costs and Benefits for Four Regulatory Alternatives ES-24
ES.8 Conclusions ES-29
Chapter 1. Introduction
1. Introduction 1-1
1.1 Summary of the Stage 2 DBPR 1-1
1.2 Document Organization 1-7
1.3 Calculations and Citations 1-8
Chapter 2. Need for the Rule
2. Need forthe Rule 2-1
2.1 Introduction 2-1
2.1.1 Description of the Issue 2-1
2.2 Public Health Concerns to Be Addressed 2-2
2.3 Regulatory History 2-2
2.3.1 Statutory Authority for Promulgating the Rule 2-2
2.3.2 1979 Total Trihalomethane Rule 2-3
2.3.3 1989 Total Coliform Rule 2-4
2.3.4 1989 Surface Water Treatment Rule 2-4
2.3.5 1996 Information Collection Rule 2-4
2.3.6 1998 Interim Enhanced Surface Water Treatment Rule 2-5
2.3.7 1998 Stage 1 Disinfectants and Disinfection Byproducts Rule 2-6
2.3.8 2000 Proposed Ground Water Rule 2-6
2.3.9 2001 Arsenic Rule 2-7
2.3.10 2001 Filter Backwash Recycling Rule 2-7
2.3.11 2002 Long Term 1 Enhanced Surface Water Treatment Rule 2-7
2.3.12 2005 Long Term 2 Enhanced Surface Water Treatment Rule 2-7
2.4 Economic Rationale 2-8
Chapter 3. Baseline Conditions
3. Baseline Conditions 3-1
3.1 Introduction 3-1
Final Economic Analysis for the Stage 2 DBPR i December 2005
-------
3.2 Data Sources 3-2
3.3 Surface Water Analytical Tool 3-3
3.4 Industry Profile 3-5
3.4.1 Public Water System Categorization 3-7
3.4.2 Systems, Plants, and Population Subject to the Stage 2 DBPR 3-9
3.4.2.1 Plant Baseline 3-9
3.4.2.2 Population Baseline 3-17
3.4.3 Water Treatment Plant Design and Average Daily Flows 3-21
3.4.4 Number of Households Served 3-23
3.5 Influent Water Quality Characterization 3-24
3.5.1 Summary of Available Influent Water Quality Data 3-24
3.5.2 Regional Differences in Water Quality 3-26
3.6 Treatment Characterization for the Pre-Stage 1 Baseline 3-31
3.7 DBP Occurrence for the Pre-Stage 1 Baseline 3-36
3.7.1 Description of ICRand SWAT DBP Data 3-36
3.7.1.1 ICRDBPData 3-36
3.7.1.2 SWAT DBP Data 3-37
3.7.2 Pre-Stage 1 DBP Occurrence for Large Surface Water Plants 3-37
3.7.3 Pre-Stage 1 DBP Occurrence in Large Ground Water Plants 3-38
3.7.4 Pre-Stage 1 DBP Occurrence for Medium Surface and Ground Water Plants
3-42
3.7.5 Pre-Stage 1 DBP Occurrences for Small Surface and Ground Water Plants
3-42
3.8 Uncertainties in Development of the Pre-Stage 1 Baseline 3-43
Chapter 4. Consideration of Regulatory Alternatives
4. Consideration of Regulatory Alternatives 4-1
4.1 Introduction 4-1
4.2 Process for Development of Regulatory Alternatives 4-2
4.3 Regulatory Alternatives Considered 4-3
Chapter 5. Compliance Forecast and Predicted Changes in DBP Levels
5. Compliance Forecast and Consequent Reduction in Chlorination DBFs 5-1
5.1 Introduction 5-1
5.2 Overview of Methodologies used in the Primary Analysis 5-1
5.3 Compliance Forecast Methodology 5-5
5.3.1 Tools for Surface and Ground Water Systems 5-5
5.3.2 Accounting for the Stage 1 DBPR 5-6
5.3.3 Operational Safety Margins 5-10
5.3.4 Accounting forthe IDSE 5-11
5.3.4.1 Analysis of Spatial Variability in Large and Medium Surface Water
Systems 5-12
5.3.4.2 Modifying the Operational Safety Margin 5-16
5.3.4.3 Incorporating Potential Impacts of the IDSE into the Compliance
Forecast 5-17
5.3.5 Methodology for Incorporating SWAT and ICR Matrix Method Results into the
Compliance Forecast 5-18
5.3.6 Compliance Forecast Simulation Model 5-20
5.4 Compliance Forecast Results 5-21
5.5 Reduction in National Average TTHM and HAA5 Levels 5-35
Final Economic Analysis for the Stage 2 DBPR ii December 2005
-------
5.5.1 Overview of Methodology 5-35
5.5.2 Reductions for Large and Medium Surface Water Systems 5-35
5.5.2.1 SWAT Methodology 5-35
5.5.2.2 The ICR Matrix Method 5-36
5.5.2.3 Combining SWAT and ICR Matrix Method Results 5-43
5.5.3 Reductions for Small Surface Water Systems 5-45
5.5.4 Reductions for Large and Medium Ground Water Systems 5-46
5.5.5 Reductions for Small Ground Water Systems 5-46
5.5.6 Results for All Systems 5-51
5.6 Reduction in Frequency of Peak TTHM and HAA5 Concentrations 5-57
5.6.1 Methodology and Assumptions 5-57
5.6.2 Results 5-61
5.7 Uncertainties in the Compliance Forecast and Subsequent DBF Reduction 5-61
5.7.1 Uncertainty in DBF data 5-64
5.7.1.1 Representativeness of the ICR data 5-64
5.7.1.2 Uncertainty in the subset of ICR data used for the ICR Matrix
Method 5-65
5.7.2 Uncertainty in the Delta Approach 5-65
Chapter 6. Benefits Analysis
6. Benefits Analysis 6-1
6.1 Introduction 6-1
6.1.1 Overview of Methodology for Quantifying Stage 2 DBPR Benefits 6-2
6.1.2 Summary of National Benefits of the Stage 2 DBPR 6-6
6.2 Problem Identification and Assessment of Potential Hazard 6-7
6.2.1 Cancer 6-7
6.2.1.1 Epidemiological Evidence of DBP Carcinogenicity 6-7
6.2.1.2 Toxicological Evidence of DBP Carcinogenicity 6-26
6.2.1.3 Issues with Human and Animal Cancer Data Concordance 6-32
6.2.1.4 Conclusions 6-33
6.2.2 Reproductive and Developmental Health Effects 6-34
6.2.2.1 Epidemiological Evidence of Adverse Reproductive and Developmental
Health Effects 6-35
6.2.2.2 Toxicological Evidence of Adverse Reproductive and Developmental
Health Effects 6-53
6.2.2.3 Conclusions 6-59
6.3 Exposure Assessment 6-60
6.3.1 Population Exposed 6-60
6.3.2 Routes of Exposure 6-61
6.3.2.1 Special Exposure Issues for Pregnant Women 6-62
6.3.3 Exposure Reduction 6-62
6.3.3.1 Reducing Exposure to All Levels of DBFs 6-63
6.3.3.2 Reducing Exposure to Peak DBP Occurrences 6-64
6.4 Benefits of the Stage 2 DBPR: Reduced Incidence of Adverse Effects 6-65
6.4.1 Reduced Incidence of Bladder Cancer Cases 6-65
6.4.1.1 Annual Cancer Cases Ultimately Avoidable 6-66
6.4.1.2 Annual Cancer Cases Avoided Accounting for Cessation Lag 6-72
6.4.1.3 Adjustments in Annual Cancer Cases Avoided to Account for the Rule
Implementation Schedule 6-74
6.4.2 Reduced Incidence of Reproductive and Developmental Effects 6-76
Final Economic Analysis for the Stage 2 DBPR Hi December 2005
-------
6.4.3 Other Health-Related Benefits 6-77
6.4.4 Non-Health-Related Benefits 6-77
6.4.5 Potential Increases in Health Risks 6-77
6.5 Valuation of Health Benefits for the Stage 2 DBPR 6-81
6.5.1 Value of Reductions in Potential Adverse Reproductive and Developmental
Health Effects 6-81
6.5.2 Value of Reductions in Bladder Cancer Cases 6-82
6.5.3 Value of Benefits Resulting from the Stage 2 DBPR for the Preferred
Alternative 6-87
6.5.4 Comparison of the Value of Benefits for Regulatory Alternatives 6-91
6.6 Uncertainties 6-93
6.7 Sensitivity Analysis for Other Factors 6-94
6.8 Potential Fetal Losses Avoided 6-97
6.8.1 Reproductive Effects Illustrative Calculation 6-98
6.8.2 Value of Potential Reductions in Fetal Losses Avoided 6-100
Chapter 7. Cost Analysis
7. Cost Analysis 7-1
7.1 Introduction 7-1
7.1.1 Overview of Methodology for Quantifying Stage 2 DBPR Costs 7-1
7.1.2 Cost Summary 7-3
7.2 Labor Rates and Laboratory Fees 7-11
7.3 Non-Treatment Costs for Systems and States/Primacy Agencies 7-13
7.3.1 Rule Implementation 7-14
7.3.2 Initial Distribution System Evaluations 7-14
7.3.3 Monitoring Plans 7-15
7.3.4 Additional Routine Monitoring 7-16
7.3.5 Operational Evaluations 7-17
7.3.6 Results (One-Time and Yearly Costs) 7-17
7.4 Technology Unit Costs 7-18
7.4.1 Treatment Technologies Used to Estimate Costs 7-19
7.4.2 Alternatives to Treatment 7-30
7.4.3 Uncertainty in Unit Costs 7-31
7.5 The Stage 2 DBPR Cost Model 7-31
7.5.1 Probability Analysis to Estimate Nominal Treatment Costs 7-31
7.5.2 Projections and Discounting to Produce Annualized Costs 7-34
7.5.3 Methodology for Estimating Household Costs 7-35
7.6 Results 7-36
7.6.1 Number of Plants Making Treatment Technology Changes 7-36
7.6.2 One-Time Costs 7-36
7.6.3 Total Annual Costs 7-36
7.6.4 Household Cost Results 7-41
7.7 Non-Quantified Costs 7-47
7.8 Uncertainty Analysis 7-47
7.9 Comparison of Regulatory Alternatives 7-50
Chapter 8. Economic Impact Analysis
8. Economic Impact Analysis 8-1
8.1 Introduction 8-1
8.2 Regulatory Flexibility Act and Small Business Regulatory Enforcement Fairness Act 8-1
Final Economic Analysis for the Stage 2 DBPR iv December 2005
-------
8.2.1 Determining Significant Impacts on Small Entities 8-2
8.2.2 Summary of the SBREFA Process 8-6
8.3 Small-System Affordability 8-7
8.3.1 Affordability Threshold 8-8
8.3.2 Affordable Compliance Treatment Technologies 8-9
8.3.3 Funding Options for Disadvantaged Systems 8-13
8.4 Feasible Treatment Technologies for All Systems 8-14
8.4.1 ICR Treatment Studies 8-15
8.4.2 BAT Evaluation Using SWAT 8-16
8.4.3 BATs for Consecutive Systems 8-17
8.5 Effect of Compliance with the Stage 2 DBPR on the Technical, Managerial, and
Financial Capacity of Public Water Systems 8-18
8.5.1 Requirements of the Stage 2 DPBR 8-19
8.5.2 Systems Subjectto the Stage 2 DBPR 8-20
8.5.3 Impact of the Stage 2 DBPR on System Capacity 8-20
8.5.4 Rationale for Scores 8-20
8.5.5 Derivation of Stage 2 DBPR Scores 8-23
8.5.5.1 Familiarization with the Stage 2 DBPR 8-23
8.5.5.2 Conducting an Initial Distribution System Evaluation 8-24
8.5.5.3 Compliance with MCLs for TTHM and HAA5 8-24
8.5.5.4 Stage 2 Monitoring Plan 8-26
8.5.5.5 Additional Routine Monitoring 8-26
8.5.5.6 Operational Evaluations 8-26
8.5.6 Summary 8-27
8.6 Paperwork Reduction Act 8-27
8.7 Unfunded Mandates Reform Act Analysis 8-28
8.7.1 UMRA Requirements and their Impact on the Stage 2 DBPR 8-28
8.7.2 Social Benefits and Costs 8-30
8.7.3 Disproportionate Budgetary Effects 8-31
8.7.4 Macroeconomic Effects 8-37
8.7.5 Consultation with Small Governments 8-37
8.7.6 Consultation with State, Local, and Tribal Governments 8-37
8.7.7 Regulatory Alternatives Considered 8-38
8.7.8 Impacts on Small Governments 8-38
8.8 Indian Tribal Governments 8-38
8.9 Impacts on Sensitive Subpopulations 8-43
8.9.1 Protecting Children from Environmental Health Risks and Safety Risks . . . 8-43
8.10 Environmental Justice 8-44
8.11 Federalism 8-45
8.12 Actions Concerning Regulations That Significantly Affect Energy Supply, Distribution,
or Use 8-45
Chapter 9. Comparison of Benefits and Costs of the Stage 2 DBPR
9. Comparison of Benefits and Costs of the Stage 2 DBPR 9-1
9.1 Introduction 9-1
9.2 Summary of National Benefits, Costs, and Net Benefits of the Stage 2 Preferred
Regulatory Alternative 9-1
9.2.1 National Benefits Summary 9-4
9.2.2 National Cost Summary 9-8
9.2.3 National Net Benefits 9-11
9.3 Comparison of Regulatory Alternatives 9-14
Final Economic Analysis for the Stage 2 DBPR v December 2005
-------
9.3.1 Comparison of Reductions in DBF Occurrence 9-14
9.3.2 Comparison of Benefits and Costs 9-15
9.3.3 Cost-Effectiveness 9-20
9.4 Effect of Uncertainties on the Estimation of Net National Benefits 9-23
9.5 Summary of Conclusions 9-27
Chapter 10. References
Final Economic Analysis for the Stage 2 DBPR vi December 2005
-------
Appendices
Appendix A: Surface Water Compliance Forecasts Using SWAT
Appendix B: Ground Water Plant Compliance Forecasts
Appendix C: Supplemental Compliance Forecasts
Appendix D: Rule Activity Schedule
Appendix E: Annual Cancer Cases Avoided as a Result of the Stage 2 DBPR
Appendix E2:Calculation of PAR, Attributable Cases and Cases Avoided for the Colon
and Rectal Cancer Sensitivity Analyses
Appendix F: Valuation of Stage 2 DBPR Benefits
Appendix G: Illustrative Calculation for Quantifying Potential Reproductive and
Developmental Benefits of the Stage 2 DBPR
Appendix H: National Costs for Non-Treatment Related Rule Activities
Appendix I: Unit Costs for Technologies Considered in the Stage 2 DBPR
Appendix J: Stage 2 DBPR Cost Projections
Appendix K: Benefit and Cost Models
Appendix L: Quality Assurance Supplemental Information
Appendix M: Ground Water Systems Adding Disinfection Under the Ground Water Rule
Appendix N: Cost Effectiveness Analysis Using a Quality-Adjusted Life Years Approach
Final Economic Analysis for the Stage 2 DBPR vii December 2005
-------
Exhibits
Executive Summary
Exhibit ES. 1 Summary of Stage 2 DBPR Requirements ES-4
Exhibit ES.2 Implementation Timeline for the Stage 2 DBPR ES-5
Exhibit ES.3 Stage 2 DBPR Population-Based Compliance Monitoring Requirements ES-7
Exhibit ES.4a Number of Disinfecting Systems Subject to Non-Treatment-Related Rule Activities ES-9
Exhibit ES.4b Non-Treatment Rule Activities for Systems Installing Disinfection to Comply with the
Ground Water Rule ES-10
Exhibit ES.5 Summary of Estimated National Benefits and Costs of the Stage 2 DBPR
Preferred Regulatory Alternative ($ Million / Year) ES-12
Exhibit ES.6 Tools Used to Develop the Stage 2 DBPR Compliance Forecasts ES-14
Exhibit ES.7a Plants Making Treatment Technology Changes, Preferred Regulatory Alternative ES-17
Exhibit ES.Vb Estimated Reduction in Average TTHM and HAAS from Pre-Stage 2 to Post-Stage 2,
Preferred Regulatory Alternative ES-18
Exhibit ES.8 Summary of Annual Household Cost Increases ES-23
Exhibit ES.9 Comparison of Benefits for All Regulatory Alternatives (SMillions) ES-25
Exhibit ES.10 Comparison of Costs for All Regulatory Alternatives (SMillions) ES-25
Exhibit ES. 11 Comparison of Annualized Mean Net Benefits for All Regulatory Alternatives
(SMillions) ES-26
Exhibit ES. 12 Cost Per Discounted Case Avoided, by Discount Rate and Regulatory Alternative
(SMillions) ES-28
Exhibit ES.13 Cost Effectiveness Analysis Using MILYs Saved from Cases of Bladder Cancer
Avoided,by Rule Alternative, 3 and 7 Percent Discount Rates ES-29
Chapter 1. Introduction
Exhibit 1.1 Comparison of Stage 1 and Stage 2 DBPR Compliance Calculations 1-3
Exhibit 1.2 IDSE Standard Monitoring Requirements 1-4
Exhibit 1.3 Stage 2 Population-Based Monitoring Requirements 1-5
Exhibit 1.4 Stage 2 DBPR Implementation Schedule 1-6
Chapter 3. Baseline Conditions
Exhibit 3.la SWAT Components 3-4
Exhibit 3.1b SWAT Inputs and Outputs 3-6
Exhibit 3.2 Derivation of the Stage 2 DBPR Plant Baseline 3-13
Exhibit 3.3 Derivation of the Stage 2 DBPR Population Baseline 3-18
Exhibit 3.4 Design Flows and Average Daily Flows per Plant (MGD) 3-22
Exhibit 3.5 Number of Households Subject to the Stage 2 DBPR 3-23
Exhibit 3.6 ICR Large System Influent Water Quality Parameters—Summary of Pre-Stage 1
Plant-Mean Data 3-25
Exhibit 3.7 Cumulative Distribution of TOC in Influent Water of Large System ICR Plant-Mean
Data 3-27
Exhibit 3.8 Cumulative Distribution of Bromide in Influent Water of Large System ICR Plant-Mean
Data 3-28
Exhibit 3.9 Medium and Small System Influent Water Quality Parameters- Summary of Pre-Stage 1
Plant-Mean Data 3-29
Exhibit 3.10 Influent Water TOC Distribution for ICR Surface Water Systems 3-30
Exhibit 3.11 Influent Water TOC Distribution for ICR Ground Water Systems 3-30
Exhibit 3.12 Influent Water TOC Distribution for Ground Water Systems Derived from the Ground
Final Economic Analysis for the Stage 2 DBPR
vm
December 2005
-------
Water Supply Survey 3-31
Exhibit 3.13a Pre-Stage 1 DBPR Treatment Technologies-in-Place for CWS Surface Water Plants 3-33
Exhibit 3.13b Pre-Stage 1 DBPR Treatment Technologies-in-Place for NTNCWS Surface Water
Plants 3-34
Exhibit 3.14a Pre-Stage 1 DBPR Treatment Technologies-in-Place for CWS Ground Water Plants 3-35
Exhibit 3.14b Pre-Stage 1 DBPR Treatment Technologies-in-Place for NTNCWS Ground Water
Plants 3-35
Exhibit 3.15 Summary of Pre-Stage 1 DBP Occurrence for Large Surface Water Plants, DS Average
Data 3-39
Exhibit 3.16 Cumulative Distributions of TTHM Data Predicted by SWAT, Pre-Stage 1
(DS Average) 3-40
Exhibit 3.17 Cumulative Distributions of HAA5 Data Predicted by SWAT, Pre-Stage 1
(DS Average) 3-40
Exhibit 3.18 Cumulative Distributions of Bromate Data Predicted by SWAT, Pre-Stage 1 (Finished
Water) 3-41
Exhibit 3.19 Cumulative Distributions of Chlorite Data Predicted by SWAT (Finished Water) ... 3-41
Exhibit 3.20 Summary of Pre-Stage 1 DBP Occurrence for Large Ground Water Plants, ICR Data 3-42
Exhibit 3.21 Summary of Pre-Stage 1 DBP Occurrence Data for Small Systems, DS Average
Data 3-43
Exhibit 3.22 Summary of Uncertainties Affecting Stage 2 DBPR Baseline Estimates 3-44
Chapter 4. Consideration of Regulatory Alternatives
Exhibit 4.1 Comparison of Hypothetical Compliance Calculations for Stage 1 and Stage 2
Regulatory Alternatives 4-7
Chapter 5. Compliance Forecast and Predicted Changes in DBP Levels
Exhibit 5.1 Tools Used to Develop the Stage 2 DBPR Compliance Forecasts 5-2
Exhibit 5.2 Compliance Evaluation of Screened ICR Surface and Ground Water Plants 5-9
Exhibit 5.3 Predicted Increase in Percent Making Treatment Technology Changes based on Spatial
Variability Analysis 5-13
Exhibit 5.4a Analysis of Variability for Stage 2 Non-Compliant Plants 5-13
Exhibit 5.4b Cumulative Distribution of ICR LRAAj^^ - ICR LRAA2ndHI for Stage 2 Non-Compliant
Plants (TTHM data) 5-14
Exhibit 5.4c Cumulative Distribution of ICR LRAAj^^ - ICR LRAA2ndHI for Stage 2 Non-Compliant
Plants (HAA5 Data) 5-15
Exhibit 5.5 Compliance Analysis of ICR Screened Plants at Different Operational Safety
Margins 5-17
Exhibit 5.6 Predicted Percent of Plants Making Treatment Technology Changes to Meet Stage 1 and
Stage 2 Regulatory Alternatives for the ICR Matrix Method and SWAT 5-19
Exhibit 5.7 Uniform Distributions for Incorporating Results from SWAT and the ICR Matrix Method
into the Compliance Forecast for Surface Water Systems 5-20
Exhibit 5.8 Compliance Forecast Exhibits for the
Stage 2 DBPR Preferred Alternative 5-22
Exhibit 5.9 Plants in CWSs and NTNCWSs Making Treatment Technology Changes From Stage 1
For Stage 2 DBPR Regulatory Alternatives 5-24
Exhibit 5. lOa Pre-Stage 2 DBPR Treatment Technologies-in-Place for CWS Surface Water Plants 5-25
Exhibit 5.1 Ob Pre-Stage 2 DBPR Treatment Technologies-in-Place for NTNCWS Surface Water
Plants 5-25
Exhibit 5.1 la Treatment Technology Selection Deltas for CWS Surface Water Plants, Percentage of
Plants, Preferred Alternative 5-26
Final Economic Analysis for the Stage 2 DBPR
IX
December 2005
-------
Exhibit 5.1 Ib Treatment Technology Selection Deltas for CWS Surface Water Plants, Number of
Plants, Preferred Alternative 5-26
Exhibit 5.1 Ic Treatment Technology Selection Deltas for NTNCWS Surface Water Plants, Percentage
of Plants, Preferred Alternative 5-27
Exhibit 5.1 Id Treatment Technology Selection Deltas for NTNCWS Surface Water Plants, Number of
Plants, Preferred Alternative 5-27
Exhibit 5.12a Post-Stage 2 DBPR Treatment Technologies-in-Place for CWS Surface Water Plants,
Percentage of Plants, Preferred Alternative 5-28
Exhibit 5.12b Post-Stage 2 DBPR Treatment Technologies-in-Place for CWS Surface Water Plants,
Number of Plants, Preferred Alternative 5-28
Exhibit 5.12c Post-Stage 2 DBPR Treatment Technologies-in-Place for NTNCWS Surface Water
Plants, Percentage of Plants, Preferred Alternative 5-29
Exhibit 5.12d Post-Stage 2 DBPR Treatment Technologies-in-Place for NTNCWS Surface Water
Plants, Number of Plants, Preferred Alternative 5-29
Exhibit 5.13a Pre-Stage 2 DBPR Treatment Technologies-in-Place for CWS Ground Water Plants 5-30
Exhibit 5.13b Pre-Stage 2 DBPR Treatment Technologies-in-Place for NTNCWS Ground Water
Plants 5-30
Exhibit 5.14a Treatment Technology Selection Deltas for CWS Ground Water Plants, Percentage of
Plants, Preferred Alternative 5-31
Exhibit 5.14b Treatment Technology Selection Deltas for CWS Ground Water Plants, Number of
Plants, Preferred Alternative 5-31
Exhibit 5.14c Treatment Technology Selection Deltas for NTNCWS Ground Water Plants, Percentage
of Plants, Preferred Alternative 5-32
Exhibit 5.14d Treatment Technology Selection Deltas for NTNCWS Ground Water Plants, Number of
Plants, Preferred Alternative 5-32
Exhibit 5.15a Post-Stage 2 DBPR Treatment Technologies-in-Place for CWS Ground Water Plants,
Percentage of Plants, Preferred Alternative 5-33
Exhibit 5.15b Post-Stage 2 DBPR Treatment Technologies-in-Place for CWS Ground Water Plants,
Number of Plants, Preferred Alternative 5-33
Exhibit 5.15c Post-Stage 2 DBPR Treatment Technologies-in-Place for NTNCWS Ground Water
Plants, Percentage of Plants, Preferred Alternative 5-34
Exhibit 5.15d Post-Stage 2 DBPR Treatment Technologies-in-Place for NTNCWS Ground Water
Plants, Number of Plants, Preferred Alternative 5-34
Exhibit 5.16a ICR Matrix Method for the Stage 2 DBPR Preferred Alternative (80/60 LRAA, IDSE)-
20 Percent Safety Margin 5-38
Exhibit 5.16b ICR Matrix Method for a Stage 2 DBPR Preferred Alternative (80/60 LRAA, IDSE)- 25
Percent Safety Margin 5-39
Exhibit 5.16c ICR Matrix Method for Regulatory Alternative 2 (80/60 SH) 5-40
Exhibit 5.16d ICR Matrix Method for Regulatory Alternative 3 (40/30 RAA) 5-41
Exhibit 5.17 TTHM and HAA5 Levels for Stage 2-Compliant Plants Using Chloramines and/or an
Advanced Treatment Technology 5-42
Exhibit 5.18 Inputs to Monte Carlo Simulation Model: Estimated DBP Reduction from SWAT and
ICR Matrix Method 5-44
Exhibit 5.19 Inputs to the Monte Carlo Simulation Model: Uniform Distributions Based on ICR
Matrix Method-to-SWAT Multiplier 5-45
Exhibit 5.20 TTHM and HAA5 Levels for Stage 2-Compliant Ground Water Plants Using
Chloramines and/or an Advanced Treatment Technology 5-47
Exhibit 5.21 a ICR Matrix Method for Ground Water Plants for the Stage 2 DBPR Preferred
Alternative 5-48
Exhibit 5.21b ICR Matrix Method for Ground Water Plants for Regulatory Alternative 2 5-49
Exhibit 5.2Ic ICR Matrix Method for Ground Water Plants for Regulatory Alternative 3 5-50
Exhibit 5.22 Reduction in Average TTHM and HAA5 Concentrations from Pre-Stage 1 to
Final Economic Analysis for the Stage 2 DBPR
December 2005
-------
Pre-Stage 2 5-52
Exhibit 5.23 Reduction in Average TTHM and HAA5 Concentrations from
Pre-Stage 2 to Post-Stage 2, Preferred Alternative 5-53
Exhibit 5.24a Reduction in Average TTHM and HAA5 Concentrations from Pre-Stage 2 to Post-Stage
2, Regulatory Alternative 1 5-54
Exhibit 5.24b Reduction in Average TTHM and HAA5 Concentrations from Pre-Stage 2 to Post-Stage
2, Regulatory Alternative 2 5-55
Exhibit 5.24c Reduction in Average TTHM and HAA5 Concentrations from Pre-Stage 2 to Post-Stage
2, Regulatory Alternative 3 5-56
Exhibit 5.25a ICR Matrix Method for Peak Locations for the Stage 2 DBPR, 20 Percent Safety Margin,
Large Surface and Ground Water Plants 5-58
Exhibit 5.25b ICR Matrix Method for Peak Locations for the Stage 2 DBPR, 25 Percent Safety Margin,
Large Surface and Ground Water Plants 5-59
Exhibit 5.26 Frequency of Occurrence of Peak Locations for ICR Surface and Ground Water Plants
Using Chloramines and/or Advanced Treatment Technologies 5-60
Exhibit 5.27 Predicted Percent of Distribution System Sampling Locations with Peaks for Pre-Stage 1,
Pre-Stage 2, and Post-Stage 2 Conditions,
20 Percent safety Margin 5-61
Exhibit 5.28 Summary of Uncertainties in the Compliance Forecast 5-63
Exhibit 5.29 Potential Impact of Uncertainties on the Compliance Forecast and DBP Reduction
Analysis 5-64
Chapter 6. Benefits Analysis
Exhibit 6.1 Summary of Quantified Benefits for the Stage 2 DBPR 6-6
Exhibit 6.2 Venn Diagram of Bladder Cancer in the U.S. Population 6-10
Exhibit 6.3 Summary of Epidemiology Studies for Bladder Cancer Associated with Chlorinated
Drinking Water and EPA Calculated PAR Values 6-12
Exhibit 6.4 Summary of Epidemiology Studies from Villanueva et al. (2003) for Bladder Cancer
Associated with Chlorinated Drinking Water used in Developments of the PAR
Analysis 6-14
Exhibit 6.5 Estimated OR for Ever-Exposed, Both Sexes Category from
Villanueva et al. (2003) Meta-Analysis 6-15
Exhibit 6.6 Summary of Epidemiology Studies from Villanueva et al. (2004) for Bladder
Cancer Associated with Chlorinated Drinking Water used in Developments
of the PAR Analysis 6-15
Exhibit 6.7 Summary of Estimated OR Values Associated with Average TTHM Exposures
for Both Sexes from Villanueva et al. (2004) 6-17
Exhibit 6.8 Detailed Data on OR as a Function of Average TTHM Exposure Level Provided by
Kogevinas and Villanueva (2005) 6-18
Exhibit 6.9 Estimates of Pre-Stage 1 Annual Bladder Cancer Cases Attributable to DBFs 6-18
Exhibit 6.10 Summary of Bladder Cancer Epidemiology Studies and Review/Meta-analysis
Studies Reviewed for Stage 2 DBPR 6-20
Exhibit 6.11 Summary of EPA's Cancer Risk Assessments as currently presented on IRIS for Specific
DBFs 6-27
Exhibit 6.12 Quantification of Cancer Risk for BDCM, Bromoform, DBCM, and DCAA,
Pre-Stage 2 Baseline 6-29
Exhibit 6.13 Summary of Reproductive/Developmental Epidemiology Studies 6-37
Exhibit 6.14 Odds Ratios (and 95 Percent Confidence Intervals :) Calculated by
Reif et al. (2000) for Reproductive and Developmental Health Endpoints at
TTHM Levels of > 80 (ig/L versus < 80 (ig/L and > 60 (ig/L versus < 60 (ig/L 6-50
Exhibit 6.15 PAR Values (and 95 Percent Confidence Intervals :) Calculated by
Final Economic Analysis for the Stage 2 DBPR
XI
December 2005
-------
Reif et al. (2000) for Reproductive and Developmental Health Endpoints at
TTHM Levels of > 80 (ig/L versus < 80 (ig/L and > 60 (ig/L versus < 60 (ig/L
(Values are Percentages) 6-51
Exhibit 6.16 Availability of Reproductive and Developmental Toxicology Studies for
Specific DBFs 6-55
Exhibit 6.17 Reproductive and Developmental Health Effects Associated with DBFs in
Toxicological Studies 6-56
Exhibit 6.18 Estimated Population Exposed to DBFs in Drinking Water 6-60
Exhibit 6.19 National Average TTHM1 Reduction Estimates 6-63
Exhibit 6.20 Comparison of Range of Estimates of Stage 2 Cases Ultimately Avoidable
for Three PAR Approaches and DBP Reductions 6-71
Exhibit 6.21a Comparison of Alternative Cessation Lag Models: Estimates of Annual Cases
Avoided by Year Following Exposure Reduction (TTHM as an indicator) 6-75
Exhibit 6.21b Estimates of Annual Cases Avoided by Year For Three Cessation Lag Models,
Considering Rule Implementation Schedule (TTHM as an Indicator) 6-76
Exhibit 6.22a Predicted Chlorite Plant-Mean Concentration for Pre-Stage 2 and Post-Stage 2 .... 6-79
Exhibit 6.22b Predicted Chlorite Monthly Average Concentrations for Pre-Stage 2 and
Post-Stage 2 6-79
Exhibit 6.23a Predicted Bromate Plant-Mean Concentrations for Pre-Stage 2 and Post-Stage 2 ... 6-80
Exhibit 6.23b Predicted Bromate Monthly Average Concentrations for Pre-Stage 2 and
Post-Stage 2 6-80
Exhibit 6.24 VSL, WTP, and Morbidity Increment Price Level Updates 6-84
Exhibit 6.25 Value of Morbidity Increment, VSL, and WTP by Year, Adjusted for Income
Elasticity 6-86
Exhibit 6.26a Non-Discounted Stream of Benefits from the Stage 2 DBPR Preferred Regulatory
Alternative, All Systems, WTP Curable Lymphoma, TTHM as Indicator 6-88
Exhibit 6.26b Non-Discounted Stream of Benefits from the Stage 2 DBPR Preferred Regulatory
Alternative, All Systems, WTP Chronic Bronchitis, TTHM as Indicator 6-89
Exhibit 6.27 Benefits Summary for the Stage 2 DBPR, Preferred Regulatory Alternative
(Millions, 2003$) 6-90
Exhibit 6.28 Benefits Summary for the Stage 2 DBPR, Preferred Regulatory Alternative
(Millions, 2003$) 6-91
Exhibit 6.29 Number and Annualized Value of Estimated Bladder Cancer Cases Avoided for
All Stage 2 DBPR Regulatory Alternatives, Villanueva et al. (2003) for
Baseline Risk (Millions, 2003$) 6-92
Exhibit 6.30 Uncertainties and Possible Effect on Estimate of Benefits 6-94
Exhibit 6.31 Annualized Value1 of Estimated Bladder Cancer Cases Avoided for the Primary Analysis,
and Estimated Colon and Rectal Cancer Cases Avoided for the
Sensitivity Analysis (Millions, 2003$) 6-97
Exhibit 6.32 Summary of the Fetal Loss Human Epidemiology Studies 6-99
Chapter 7. Cost Analysis
Exhibit 7.1 Stage 2 DBPR Cost Model Inputs and Outputs 7-2
Exhibit 7.2a Baseline Systems Subject to Non-Treatment-Related Rule Activities 7-4
Exhibit 7.2b Non-Treatment-Related Rule Activities for Systems Installing Disinfection to
Comply with the Ground Water Rule 7-5
Exhibit 7.3 Number and Percent of Plants Making Treatment Technology Changes for the
Stage 2 DBPR 7-7
Exhibit 7.4 Initial Capital and One-Time Costs for the Stage 2 DBPR (SMillions) 7-8
Exhibit 7.5a Total Annualized Costs for Stage 2 DBPR Rule Activities ($Millions/Year,
3 Percent Discount Rate) 7-9
Final Economic Analysis for the Stage 2 DBPR
xn
December 2005
-------
Exhibit 7.5b Total Annualized Costs for Stage 2 DBPR Rule Activities ($Millions/Year,
7 Percent Discount Rate) 7-10
Exhibit 7.6a System Wage Rates by Standard Size Categories 7-11
Exhibit 7.6b System Wage Rates by Monitoring Size Categories 7-12
Exhibit 7.7 Summary of System Costs for Non-Treatment Related Stage 2 DBPR Rule
Activities (One-Time and Yearly) 7-18
Exhibit 7.8a Treatment Technologies for Surface Water Plants 7-21
Exhibit 7.8b Treatment Technologies for Disinfecting Ground Water Plants 7-23
Exhibit 7.9 Household Cost Inputs 7-25
Exhibit 7.10a Capital Unit Costs ($/Plant) for CWS Surface Water Plants 7-27
Exhibit 7. lOb Annual O&M Unit Costs ($/Plant/Year) for CWS Surface Water Plants 7-27
Exhibit 7. lOc Household Unit Treatment Costs ($/Household/Year) for CWS Surface
Water Plants 7-28
Exhibit 7.1 la Capital Cost ($/Plant) for CWS Disinfecting Ground Water Plants 7-29
Exhibit 7.1 Ib Annual O&M Costs ($/Plant/Year) for CWS Disinfecting Ground Water Plants 7-29
Exhibit 7.1 Ic Household Unit Treatment Costs ($/Household/Year) for CWS Disinfecting
Ground Water Plants 7-30
Exhibit 7.12 Uniform Distributions for Incorporating the ICR Matrix Method-to-SWAT
Multiplier into the Compliance Forecasts for Surface Water Systems 7-33
Exhibit 7.13 Total Initial Capital Costs (SMillions) and Yearly O&M Costs ($Millions/Year) ... 7-38
Exhibit 7.14a Total Annualized Costs at 3 Percent Social Discount Rate (SMillions) 7-39
Exhibit 7.14b Total Annualized Costs at 7 Percent Social Discount Rate (SMillions) 7-40
Exhibit 7.15 Annual Household Cost Increases 7-42
Exhibit 7.16a Household Cost Distributions, All Surface Water Systems Subject to the Rule 7-43
Exhibit 7.16b Household Cost Distributions, All Ground Water Systems Subject to the Rule 7-44
Exhibit 7.17a Household Cost Distributions, Surface Water Systems Making Treatment
Technology Changes 7-45
Exhibit 7.17b Household Cost Distributions, Ground Water Systems Making Treatment
Technology Changes 7-46
Exhibit 7.17c Household Cost Distributions, Small Systems Making Treatment Technology
Changes (Surface and Ground) 7-47
Exhibit 7.18 Cost Uncertainty Summary 7-49
Exhibit 7.19 Total Annualized Cost for the Stage 2 DBPR Regulatory Alternatives (SMillions) . . 7-50
Chapter 8. Economic Impact Analysis
Exhibit 8.1 Annualized Compliance Cost as a Percentage of Revenues for All Small Entities .... 8-5
Exhibit 8.2 Derivation of Available Expenditure Margin 8-9
Exhibit 8.3 Affordability Analysis Inputs 8-10
Exhibit 8.4a Affordable Compliance Treatment Technologies and Household Unit Treatment
Costs ($/HH/Year) for Surface Water Systems 8-11
Exhibit 8.4b Affordable Compliance Treatment Technologies and Household Unit Treatment
Costs ($/HH/Year) for Ground Water Systems 8-11
Exhibit 8.4c Distribution of Household Unit Treatment Costs for Plants Adding Treatment 8-12
Exhibit 8.5 SWAT Model Predictions of Percent of Large Plants in Compliance with
TTHM and HAA5 Stage 2 MCLs after Application of Specified Treatment
Technologies 8-17
Exhibit 8.6 Estimated Impact of the Stage 2 DBPR on Small System Capacity
(0 = no impact, 1 = minimal impact, and 5 = very significant impact) 8-21
Exhibit 8.7 Estimated Impact of the Stage 2 DBPR on Large System Capacity
(0 = no impact, 1 = minimal impact, and 5 = very significant impact) 8-22
Exhibit 8.8 Summary of Average Annual Burden Hours and Labor Costs 8-28
Final Economic Analysis for the Stage 2 DBPR
xm
December 2005
-------
Exhibit 8.9 Public and Private Costs for the Stage 2 DBPR
(Annualized at 3 and 7 Percent, SMillions) 8-30
Exhibit 8.10 Total Annualized Benefits and Costs of Regulatory Alternatives
(SMillions, 2003$) 8-31
Exhibit 8.1 la Number of Small Disinfecting Systems by State 8-33
Exhibit 8.1 Ib Percent of Small Disinfecting Systems by State 8-34
Exhibit 8.12a Total Annualized Cost of Compliance for CWSs (3 and 7 Percent Discount Rates)
(SMillions) 8-35
Exhibit 8.12b Annualized Cost of Compliance for NTNCWSs (3 and 7 Percent Discount Rates)
(SMillions) 8-35
Exhibit 8.13 Percentages and Costs by Public and Private Sector
(Costs Annualized at 3 and 7 Percent) 8-36
Exhibit 8.14 Annual Cost of Compliance for Tribal Systems by System Type and Size
(Annualized at 3 Percent) 8-41
Exhibit 8.15 Increase in Energy Usage as a Result of the Stage 2 DBPR 8-48
Exhibit 8.16 Sample Calculation for Determining Increase in Energy Usage:Chloramines 8-49
Chapter 9. Comparison of Benefits and Costs of the Stage 2 DBPR
Exhibit 9. la Summary of Benefit and Cost Estimates by Year for the Stage 2 Preferred Regulatory
Alternative Using Lymphoma WTP (SMillions) 9-2
Exhibit 9. Ib Summary of Benefit and Cost Estimates by Year for the Stage 2 Preferred Regulatory
Alternative Using Bronchitis WTP (SMillions) 9-3
Exhibit 9.2 Summary of Nonqualified National Benefits of the Stage 2 DBPR 9-5
Exhibit 9.3 Summary of Annual Bladder Cancer Cases Ultimately Avoidable for the Stage 2 DBPR
Preferred Regulatory Alternative 9-6
Exhibit 9.4 Estimated Annualized National Benefits for the Stage 2 Preferred Regulatory Alternative
(SMillions) 9-7
Exhibit 9.5a Annualized Costs for Stage 2 DBPR Preferred Regulatory Alternative Rule Activities
($Millions/Year, 3% Discount Rate) 9-9
Exhibit 9.5b Annualized Costs for Stage 2 DBPR Preferred Regulatory Alternative Rule Activities
($Millions/Year, 7 Percent Discount Rate) 9-10
Exhibit 9.6 Annualized Mean Net Benefits for the Stage 2 Preferred Regulatory Alternative
(SMillions) 9-12
Exhibit 9.7 Estimated Annualized National Costs and Benefits for the Stage 2 Preferred Regulatory
Alternative with Uncertainty Measured as a Percent of the Mean (SMillions) 9-13
Exhibit 9.8 Estimated Breakeven Points (Number of Bladder Cancer Cases Avoided) for the Stage 2
Preferred Regulatory Alternative 9-13
Exhibit 9.9 Comparison of DBP Reduction (of Annual Plant Mean TTHM Data) 9-15
Exhibit 9.10 Comparison of Number and Annualized Value of Estimated Bladder Cancer Cases
Avoided for All Regulatory Alternatives (SMillions) 9-16
Exhibit 9.11 Comparison of Costs for All Regulatory Alternatives (SMillions) 9-17
Exhibit 9.12 Comparison of Mean Net Benefits for All Regulatory Alternatives (SMillions) .... 9-18
Exhibit 9.13 Incremental Net Benefits for All Regulatory Alternatives ($ Millions) 9-19
Exhibit 9.14 Incremental Cost Per Case Avoided1 for All Regulatory Alternatives by Discount Rate
(SMillions) 9-22
Exhibit 9.15 Benefit Cost Ratios for All Regulatory Alternatives 9-23
Exhibit 9.16 Effects of Uncertainties on National Estimates 9-25
Final Economic Analysis for the Stage 2 DBPR
xiv
December 2005
-------
Acronyms and Notations
AAM
ACS
AIPC
AMWA
ARBRP
ASDWA
AD
AO
ATSDR
AUX1
AUX8
AVG1
AVG2
AWWA
AWWARF
BAT
BCAA
BCAN
BDCAA
BDCM
BLS
CATT
CCR
CDBG
CDC
CDHS
CI
CKA
CL2
CLM
CLO2
COI
CPI
CWS
CWSS
DBAA
DBAN
DBCAA
DBCM
DBPR
DBF
DCAA
DCAN
DNA
DOC
DS Average
DSE
DS Maximum
DWSRF
Annual Average of the Maximum
American Cancer Society
All Indian Pueblo Council
Association of Metropolitan Water Agencies
Arsenic Rule Benefits Review Panel
Association of State Drinking Water Administrators
Advanced Disinfectants
Advanced Oxidants
Agency for Toxic Substances and Disease Registry
Auxiliary Database 1
Auxiliary Database 8
Average sample point number 1
Average sample point number 2
American Water Works Association
American Water Works Association Research Foundation
Best Available Technology
Bromochloroacetic Acid
Bromochloroacetonitrile
Bromodichloroacetic Acid
Bromodichloromethane
Bureau of Labor Statistics
Cases Attributable to DBFs
Consumer Confidence Report Rule (1998)
Community Development Block Grant
Centers for Disease Control and Prevention
California Department of Health Services
Confidence Interval
Chernoff-Kavlock Assay
Chlorine
Chloramines
Chlorine Dioxide
Cost of Illness
Consumer Price Index
Community Water System
Community Water Systems Survey
Dibromoacetic Acid
Dibromoacetonitrile
Dibromochloroacetic Acid
Dibromochloromethane
Disinfectants and Disinfection Byproducts Rule
Disinfection Byproduct
Dichloroacetic Acid
Dichloroacetonitrile
Deoxyribonucleic Acid
Dissolved Organic Carbon
Distribution System Average Sample Point
Distribution System Equivalent Sample Point
Distribution System Maximum Sample Point
Drinking Water State Revolving Fund
Economic Analysis for the Stage 2 DBPR
xv
December 2005
-------
EA
EBCT
EC
ECI
ED10
EPA
ES
FACA
FBRR
FLR
FR
FRFA
FTE
GAC
GAC10
GAC20
GDP
GIS
GW
GWR
GWSS
GWUDI
HAAS
HAA6
HAA9
HAN
ICMA
ICR
ICRSS
ILSI
IDSE
IDSE SMP
IESWTR
IPCS
IRFA
IRIS
kg
KWh/y
LED10
LF
LH
LOAEL
LRAA
LT IESWTR
LT2ESWTR
MBAA
MCAA
MCAN
MCL
MCLG
M-DBP
Economic Analysis
Empty Bed Contact Time
Enhanced Coagulation
Employment Cost Index Information
Effective Dose for 10% response
Environmental Protection Agency
Enhanced Softening
Federal Advisory Committees Act
Filter Backwash Recycling Rule (2001)
Full Liter Resorption
Federal Register
Final Regulatory Flexibility Analysis
Full-Time Equivalent
Granular Activated Carbon
Granular Activated Carbon—10-Minute Contact Time
Granular Activated Carbon—20-Minute Contact Time
Gross Domestic Product
Geographical Information System
Ground Water
Ground Water Rule
Ground Water Supply Survey
Ground Water Under the Direct Influence of Surface Water
Haloacetic Acids [five] [sum of monochloroacetic acid, dichloroacetic acid,
trichloroacetic acid, monobromoacetic acid, and dibromoacetic acid]
Haloacetic Acids [total of six]
Haloacetic Acids [total of nine]
Haloacetonitrile
International City/County Management Association
Information Collection Rule (1996)
Information Collection Rule Supplemental Survey
International Life Sciences Institute
Initial Distribution System Evaluation
Initial Distribution System Evaluation Standard Monitoring Program
Interim Enhanced Surface Water Treatment Rule (1998)
International Programme on Chemical Safety
Initial Regulatory Flexibility Analysis
Integrated Risk Information System
Kilogram
Kilowatt Hours per Year
Lower Bound on the Effective Dose for 10% response
Lag Function
Luteinizing Hormone
Lowest-Observed-Adverse-Effect-Level
Locational Running Annual Average
Long Term 1 Enhanced Surface Water Treatment Rule (2002)
Final Long Term 2 Enhanced Surface Water Treatment Rule
Monobromoacetic Acid
Monochloroacetic Acid
Monochloroacetonitrile
Maximum Contaminant Level
Maximum Contaminant Level Goal
Microbial-Disinfectants/Disinfection Byproducts [Advisory Committee]
Economic Analysis for the Stage 2 DBPR
xvi
December 2005
-------
MF
MG
MOD
mg/kg-day
mg/L
MHI
mJ/cm2
MLE
MRDL
MRDLG
MRRR
MW
mWh
MX
NAICS
NCI
NCSL
NCWS
NDMA
NDWAC
NF
ng
NGA
NHEERL
NLC
NOAEL
NODA
NOM
NPDWR
NRWA
NTNCWS
NTU
03
OGWDW
OR
OMB
O&M
PAR
POE
POU
ppb
ppm
PUC
PSC
PV
PWS
PWSID
RAA
RFA
RfD
RIA
Microfiltration
Million Gallon
Million Gallons per Day
Micrograms per Liter
Milligrams per Kilogram per Day
Milligrams per Liter
Median Household Income
Millijoules per centimeter square
Maximum Likelihood Estimation
Maximum Residual Disinfectant Level
Maximum Residual Disinfectant Level Goal
Maximum Relative Risk Reduction
Megawatt
Megawatt Hours
3-Chloro-4-(dichloromethyl)-5-hydroxy-2(5H)-furanone
North American Industry Classification System
National Cancer Institute
National Conference of State Legislatures
Noncommunity Water System
N-nitrosodimethylamine
National Drinking Water Advisory Council
Nanofiltration
Nanograms
National Governors' Association
National Health and Environmental Effects Research Laboratory (EPA)
National League of Cities
No-Observed-Adverse-Effect-Level
Notice of Data Availability
Natural Organic Matter
National Primary Drinking Water Regulations
National Rural Water Association
Nontransient Noncommunity Water System
Nephelometric Turbidity Unit
Ozone
Office of Ground Water and Drinking Water
Odds Ratio
Office of Management and Budget
Operations and Maintenance
Population Attributable Risk
Point-of-Entry
Point-of-Use
Parts per Billion
Parts per Million
Public Utilities Commission
Public Services Commission
Present Value
Public Water System
Public Water System Identification
Running Annual Average
Regulatory Flexibility Act
Reference Dose
Regulatory Impact Analysis
Economic Analysis for the Stage 2 DBPR
xvi i
December 2005
-------
RR
RUS
RSI
SAB
SBA
SBAR
SBREFA
SCADA
SD
SDS
SDWA
SDWIS
SEER
SER
SH
SIC
SMP
sss
sw
SWAT
Stage 1 DBPR
Stage 2 DBPR
SWTR
TBAA
TCAA
TCAN
TCR
THM
TMF
TNCWS
TOC
TOX
TTHM
TWO
T&C
UF
UMRA
use
USDA
USEPA
uv
UVA
VSL
WEC
WHO
WITAF
w(t)
WTP
Relative Risk
Rural Utility Service
Risk Sciences Institute
Science Advisory Board
Small Business Administration
Small Business Advocacy Review
Small Business Regulatory Enforcement Fairness Act
Supervisory Control and Data Acquisition
Sprague-Dawley
Simulated Distribution System
Safe Drinking Water Act (1974)
Safe Drinking Water Information System
Surveillance, Epidemiology, and End Results
Small Entity Representatives
Single Highest
Standard Industrial Codes
Standard Monitoring Program
System-Specific Study
Surface Water
Surface Water Analytical Tool
Stage 1 Disinfectants and Disinfection Byproducts Rule (1998)
Final Stage 2 Disinfectants and Disinfection Byproducts Rule
Surface Water Treatment Rule (1989)
Tribromoacetic Acid
Trichloroacetic Acid
Trichloroacetonitrile
Total Coliform Rule (1989)
Trihalomethane
Technical, Managerial, and Financial
Transient Noncommunity Water System
Total Organic Carbon
Total Organic Halides
Total Trihalomethanes [four] [sum of chloroform, bromodichloromethane,
dibromochloromethane, and bromoform]
Technical Workgroup
Technology and Cost
Ultrafiltration
Unfunded Mandates Reform Act
United States Code
United States Department of Agriculture
United States Environmental Protection Agency
Ultraviolet [Light Disinfection]
Ultraviolet-254 Absorbance
Micrograms per Liter
Value of a Statistical Life
Whole embryo culture
World Health Organization
Water Industry Technical Action Fund
cessation lag weighting factor
Willingness to Pay
Economic Analysis for the Stage 2 DBPR
xvm
December 2005
-------
Health Risk Reduction and Cost Analysis (HRRCA)
Under the Safe Drinking Water Act (SDWA) Amendments of 1996, when
proposing a national primary drinking water regulation that includes an maximum contaminant level
(MCL), the Environmental Protection Agency (EPA) must conduct a health risk reduction and cost
analysis (HRRCA). A HRRCA contains seven requirements, all of which are addressed in this Economic
Analysis (EA) for the Stage 2 Disinfectants and Disinfection Byproducts Rule (DBPR). The table below
shows where the HRRCA requirements are discussed in this document.
HRRCA Crosswalk to the Economic Analysis for the Stage 2 DBPR
HRRCA Requirement
Quantifiable and nonquantifiable health risk
reduction benefits
Quantifiable and nonquantifiable health risk
reduction benefits from co-occurring contaminants
Quantifiable and nonquantifiable costs
Incremental costs and benefits associated with
MCL alternatives
Effects of the contaminants on the general
population and sensitive subpopulations
Increased health risk that may occur as a result of
compliance
Other relevant factors (quality and uncertainty of
information)
Addressed in Economic Analysis
Chapters (All sections and exhibits)
Chapter 9 (Sections 9.2.2; Exhibits 9.1-9.4, 9.6,
9.7-9.10, and 9. 12)
Chapter 6 (Section 6.4.3 and 6.4.4)
Chapter 7 (All sections and exhibits)
Chapters (Sections 8.2, 8.3, and 8.6-8.8; Exhibits
8.1, 8.7-8.9 and 8.1 1-8.13)
Chapter 9 (Sections 9.1 .2 and 9.1 .3, 9.4.2;
Exhibits 9.2-9.3, 9.5, 9.7, and 9.1 1)
Chapter 6 (Section 6.5.4; Exhibit 6.28)
Chapter 7 (Section 7.9; Exhibit 7.19)
Chapters (Section 8.7.2; Exhibit 8.9)
Chapter 9 (Section 9.3; Exhibits 9.9-9.15)
Chapter 6 (Sections 6.2 and 6.3.2.1 ; Exhibits 6.2-
6.17)
Chapters (Sections 8.9 and 8.10)
Chapter 6 (Section 6.4.5)
Chapter 3 (Section 3.8; Exhibit 3.22)
Chapter 5 (Section 5.7; Exhibit 5.28)
Chapter 6 (Section 6.6; Exhibit 6.29)
Chapter 7 (Section 7.8; Exhibit 7.18)
Chapter 9 (Section 9.4; Exhibits 9.16)
Economic Analysis for the Stage 2 DBPR
xix
December 2005
-------
Executive Summary
This document presents the Economic Analysis (EA), prepared by the U.S. Environmental
Protection Agency (EPA), of the benefits and costs of the final Stage 2 Disinfectants and Disinfection
Byproducts Rule (DBPR). Executive Order 12866 requires federal agencies to conduct an analysis of the
benefits and costs of proposed and final rules that cost over $100 million annually. Although EPA's
analysis of the Stage 2 DBPR has determined that its annual costs are most likely below this threshold,
EPA has chosen to publish a complete EA for this rule.
ES.l Need for the Rule
Over 48,000 public water systems (PWSs), serving more than 260 million people in the United
States, chemically disinfect their water to kill or inactivate microbial contaminants (USEPA 200 Ic). This
is an essential public health measure. Chemical disinfection, however, may pose health risks of its own.
Disinfection byproducts (DBPs) result from reactions between chemical disinfectants and naturally
occurring compounds in source waters. Research has shown that chlorinated waters and DBPs may be
associated with increased risk of bladder and other cancers. While there are uncertainties in the
quantitative relationship between the incidence of these cancers and the occurrence of DBPs in drinking
water, EPA believes that additional reductions in these DBP levels in drinking water will reduce the
incidence of bladder cancer and, possibly, other cancers.
In addition, results from toxicology and, particularly, epidemiology studies published in the last
several years suggest a potential increased risk for pregnant women and their fetuses who are exposed to
DBPs in drinking water. The studies have shown that early-term miscarriage, stillbirth, low birth weight,
and some birth defects may be associated with drinking water containing DBPs. (These studies are
discussed in detail in Chapter 6.) Uncertainties pertain to the extent to which reproductive and
developmental effects may be associated with DBP exposure, which DBPs may be of greatest concern,
what levels of DBPs may pose a risk, and at what period of development fetuses may be at the greatest
risk. Although the levels of DBPs potentially associated with specific adverse reproductive and
developmental effects are not known and no causal link has been established, EPA believes the evidence
supports concern for these potential hazards and warrants regulatory action.
In a separate but concurrent action, EPA is promulgating the Long Term 2 Enhanced Surface
Water Treatment Rule (LT2ESWTR) to improve control of microbial contaminants, particularly
Cryptosporidium, in surface water and to ensure that microbial protection is not compromised by efforts
to reduce exposure to DBPs. Together, the Stage 2 DBPR and LT2ESWTR represent the final stage of a
two-stage strategy that was developed in a regulatory negotiation effort in 1992 and 1993.1 They reflect
recommendations presented by the Stage 2 Microbial and Disinfection Byproducts (M-DBP) Federal
Advisory Committee Agreement in Principle, signed in September 2000 (USEPA 2000n).
lrrhe key outcomes of that regulatory negotiation effort were recommendations to proceed with rules
addressing DBPs and microbial pathogens in two stages and to collect relevant information from public water
supplies for use in the development of these rules and the analysis of their impacts. This two-stage approach was
subsequently incorporated into the 1996 Safe Drinking Water Act (SDWA) Amendments. The first stage of the
M-DBP rulemaking process culminated with the joint promulgation of the Stage 1 DBPR and the Interim Enhanced
Surface Water Treatment Rule (IESWTR) by EPA in December 1998.
Final Economic Analysis for the Stage 2 DBPR ES-1 December 2005
-------
ES.2 Consideration of Regulatory Alternatives
The M-DBP Advisory Committee met from March 1999 to December 2000 to evaluate whether
and to what degree EPA should promulgate revised or additional DBF standards to protect public health.
The Advisory Committee carefully considered extensive new data on the occurrence and health effects of
DBFs, as well as costs and potential impacts on PWSs, and concluded that a targeted protective public
health approach should be taken to address exposure to DBFs beyond the requirements of the Stage 1
DBPR. While there had been substantial research to date, the Advisory Committee also concluded that
significant uncertainty remained regarding the risk associated with DBFs in drinking water.
After extensive deliberations, the Advisory Committee recommended maintaining the MCLs for
total trihalomethanes (TTHM) and haloacetic acids [total of five] (HAAS) at 0.080 mg/L (80 (ig/L) and
0.060 mg/L (60 (ig/L) respectively, but changing the compliance calculation from the system-wide
running annual average (RAA) calculation to a locational running annual average (LRAA) calculation.
Systems would also be required to conduct an Initial Distribution System Evaluation (IDSE) to select
compliance monitoring sites that reflect high TTHM and HAAS concentrations. The revised compliance
determination would require MCLs of 0.080 mg/L for TTHM and 0.060 mg/L for HAAS calculated as
LRAAs to be met at individual monitoring sites identified through the IDSE. The committee also
provided recommendations for simultaneous compliance with the LT2ESWTR so that the reduction of
DBFs would not compromise microbial protection. The M-DBP Agreement in Principle (available on the
web at http://www.epa.gov/safewater/disinfection/st2agreement.html) summarizes the recommendations
from the Advisory Committee (USEPA 2000m).
This EA considers four regulatory alternatives derived from a larger number discussed by the M-
DBP Advisory Committee, including the Preferred Alternative that EPA is promulgating in the Stage 2
DBPR:
• Preferred Alternative: MCLs of 80 micrograms per liter (i-ig/L) for TTHM and 60 i-ig/L for
HAAS, measured as an LRAA. MCL of 10 i-ig/L for bromate. Compliance monitoring is
preceded by the IDSE.
• Alternative 1: MCLs of 80 |ig/L for TTHM and 60 |ig/L for HAAS, measured as an LRAA.
MCL of 5 i-ig/L for bromate.
Alternative 2: MCLs of 80 |ig/L for TTHM and 60 |^g/L for HAAS, measured as a single
highest (SH) value. MCL of 10 |ig/L for bromate.
• Alternative 3: MCLs of 40 |ig/L for TTHM and 30 |ig/L for HAAS, measured as an RAA.
MCL of 10 |ig/L for bromate.
For comparison with the Preferred Alternative, EPA also developed estimates of the benefits and
costs for Alternatives 1, 2, and 3 in this document.
Final Economic Analysis for the Stage 2 DBPR ES-2 December 2005
-------
ES.3 Summary of the Stage 2 DBPR
The requirements of the Stage 2 DBPR apply to all community water systems (CWSs) and
nontransient noncommunity water systems (NTNCWSs)—both ground and surface water systems2—that
add a disinfectant other than ultraviolet light (UV), or that deliver water that has been treated with a
disinfectant other than UV. New since the Stage 1 DBPR, the Stage 2 DBPR formally defines
consecutive systems and includes provisions specific to consecutive systems to ensure the same level of
heath protection for people served in consecutive systems as people served by non-consecutive systems.
Each Stage 2 DBPR rule activity for the Preferred Regulatory Alternative is described below and
illustrated in the flow chart in Exhibit ES. 1. Exhibit ES.2 displays the compliance schedule for each rule
activity. Note that consecutive systems of any size must comply with the requirements of the Stage 2
DBPR on the same schedule as required for the largest system in the combined distribution system.
Initial Distribution System Evaluations
For many systems, compliance monitoring will be preceded by an Initial Distribution System
Evaluation (IDSE) to identify Stage 2 DBPR compliance monitoring locations that represent distribution
system sites with high TTHM and HAAS levels. The IDSE consists of either standard monitoring or a
system specific study (SSS), unless a system meets the criteria for a 40/30 certification or a very small
system waiver. To meet the criteria for a 40/30 certification, systems must have low Stage 1 monitoring
results (every individual compliance sample is less than or equal to 40 |^g/L and 30 |^g/L for TTHM and
HAAS, respectively) and no TTHM or HAAS monitoring violations during a 2-year eligibility period.
Systems can qualify for a very small systems waiver if they serve fewer than 500 people and have TTHM
and HAAS data. In addition, NTNCWSs serving fewer than 10,000 people are not required to conduct an
IDSE.
Compliance with Stage 2 DBPRMCLs
The Stage 2 DBPR changes the way sampling results are averaged to determine compliance. The
determination for the Stage 2 DBPR is based on an LRAA (i.e., compliance must be met at each
monitoring location) instead of the system-wide RAA used under the Stage 1 DBPR.
Monitoring Plans
Systems must develop a Stage 2 DBPR monitoring plan that includes monitoring locations,
monitoring dates, and compliance calculation procedures. States have the option to implement a
procedure for addressing modifications to wholesale system and consecutive system monitoring plans on
a case-by-case basis.
2 For the purposes of this EA, "surface water" is equivalent to the definition of subpart H systems used in
the Stage 2 DBPR rule language and includes systems that provide ground water under the direct influence of
surface water (GWUDI).
Final Economic Analysis for the Stage 2 DBPR ES-3 December 2005
-------
Exhibit ES.1 Summary of Stage 2 DBPR Requirements
Systems Subject to the Stage 2 DBPR
(All surface water and ground water CWSs and NTNCWSs that apply a chemical disinfectant to their water,
or deliver such water, including consecutive systems.)
Rule Implementation
(All systems subject to the rule must perform rule implementation activities such as reading the rule, training, etc.)
Initial Distribution System Evaluation (IDSE)
(All systems subject to the rule must perform an IDSE, or meet the criteria not to perform an IDSE.)
Systems not performing an IDSE
Systems performing an IDSE
and submitting an IDSE plan
Systems
receiving a
40/30 certification
Systems receiving
a very small
system waiver
NTNCWSs
serving
< 10,000 people
Systems conducting
Standard Monitoring
Systems conducting a
System Specific Study
Systems may or may not
have to select new Stage 2 DBPR
monitoring sites
Systems submit an
IDSE report recommending
Stage 2 monitoring sites
Stage 2 DBPR Monitoring Plans
(All systems subject to the rule must develop a Stage 2 DBPR monitoring plan that includes monitoring locations, monitoring
dates, and compliance calculation procedures)
Ensure Compliance with Stage 2 DBPR MCLs
(All systems subject to the rule must meet Stage 2 DBPR MCLs. Systems may or may not have to make treatment or
operational changes.)
Routine Monitoring Requirements
(Monitoring requirements for the Stage 2 DBPR are based on system type and population served [not number of plants per
system, as for the Stage 1 DBPR]. Systems subject to the Stage 2 DBPR may have fewer or more routine monitoring
requirements compared to those already required by the Stage 1 DBPR.)
Operational Evaluations
(All systems subject to the rule that exceed the operational evaluation level must perform an operational evaluation
and submit a report to the State within 90 days.)
Final Economic Analysis for the Stage 2 DBPR ES-4
December 2005
-------
Exhibit ES.2 Implementation Timeline for the Stage 2 DBPR
Schedule 1
Systems serving
> 100,000 1
Schedule 2
Systems serving
50,000 to 99,9991
Schedule 3
Systems serving
10,000 to 49,9991
Schedule 4
Systems serving
< 10,000 1
2006
2007
2008
| LT2 Crypto monitoring
!
IDSE Plar
October 1
IDJ
Ap
2006
| IDSE mon.
i Due
,2006
IDSE
Janua
2009
Rep
ry1
LT2 Crypto monitoring
t
3E Plan Due
ril 1,2007
IDSE Plar
October 1
IDS
Ap
2007
| IDSE
mon.
| LT2
i Due
,2007
1
5E Plan D
-111,2008
IDJ
Jul
Crypto
IDSE
ort Du
,2009
SERep
y1,20
2010
Treatment
^
2011
2012
i2013 | 2014
_
i
Installation | Possible Extension '
i
Be
Ap
n T
gin Compliance
111,2012 I
Treatment Installation
>" ^
X
\
ort
09
Due
monitoring
mon.
E. Coli mon.
je
2008
IDSE
IDJ
Jar
Begin Cor
October 1
u
I
Possible Extension 2 i
n
npl
, 2(
r
ance
)12
2015
Treatment Installation 1 Possible Extension 2 '
1 1
I n
3E Report Du
many 1, 2010
Crypto
mon. t
2009
I
a
mon.i
1 1
Begin Compliance
October 1,201 3
Treatment Installation 1 Possible Extension 2 1
,',"''' I 1
IDSE Report
July 1,2010
2010
Due
2011
2012
I
Begin Co
October 1
2013
f
npliance
,2013
2014
2015
11ncludes all systems that are part of a combined distribution system that have a largest system with this population.
2 A State may grant up to an additional 2 years for systems to comply if the State determines that additional time is necessary for capital improvements.
3 Subpart H systems that must conduct Cryptosporidium monitoring have an additional 12 months to comply with the Stage 2 DBPR MCLs.
The IDSE plan is either a Standard Monitoring plan or an SSS plan. The IDSE report is either for Standard Monitoring or the SSS.
Final Economic Analysis for the Stage 2 DBPR
ES-5
December 2005
-------
Routine Monitoring Requirements
EPA has adopted a population-based monitoring approach for the Stage 2 DBPR, where
compliance and IDSE monitoring requirements are based only on source water type and retail population
served. This is a change from the plant-based approach used in the 1979 TTHM rule and the Stage 1
DBPR. EPA's decision to use a population-based approach for all systems is based on improved public
health protection, flexibility, and simplified implementation. Exhibit ES.3 presents the new, population-
based Stage 2 DBPR compliance monitoring requirements.
Operational Evaluations
Because Stage 2 DBPR MCL compliance is based on individual DBP measurements at a location
averaged over a four-quarter period, a system could find higher TTHM or HAAS levels than the MCL
values, while at the same time maintaining compliance with the Stage 2 DBPR. This is because the high
concentration could be averaged with lower concentrations at a given location. For this reason, the Stage
2 DBPR includes a provision for "operational evaluations" as follows:
A system has exceeded an operational evaluation level at any monitoring location when the
sum of the two previous quarters' compliance monitoring results plus twice the current
quarters result, divided by 4, exceeds 80 i-ig/L for TTHM or 60 i-ig/L for HAAS.
If an operational evaluation level is exceeded, the system must conduct an "operational
evaluation" and submit a written report of the evaluation to the State/Primacy Agency no later than 90
days after being notified of the analytical results that caused the excursion.
Requirements for Consecutive Systems
The Stage 2 DBPR includes provisions for consecutive systems, which are PWSs that receive
finished water from another public water system (a wholesale system). Consecutive systems face
particular challenges in providing water that meets regulatory standards for DBFs and other contaminants
whose concentration can increase in the distribution system. Moreover, previous regulation of DBP
levels in consecutive systems varies widely among States. In consideration of these factors, the Stage 2
DBPR provides monitoring, compliance schedule, and other requirements specifically for consecutive
systems. These requirements are intended to facilitate compliance by consecutive systems with MCLs for
TTHM and HAAS under the Stage 2 DBPR and help to ensure that consumers in consecutive systems
receive equivalent public health protection.
Final Economic Analysis for the Stage 2 DBPR ES-6 December 2005
-------
Exhibit ES.3 Stage 2 DBPR Population-Based Compliance Monitoring
Requirements
System Size
(Population Served )
Distribution System Sample
Locations1
Highest
TTHM
Locations
Highest
HAA5
Locations
Existing
Stage 1
Compliance
Locations2
Total
Sample
Locations per
System5
Monitoring
Frequency3
Systems Using Surface Water in Whole or in Part4
<500
500-3,300
3,301-9,999
10,000-49,999
50,000-249,999
250,000-999,999
1 Mil-4, 999,999
> 5,000,000
1
1
1
2
3
5
6
8
1
1
1
1
3
4
6
7
NA
NA
NA
1
2
3
4
5
2
2
2
4
8
12
16
20
per year
per quarter
per quarter
per quarter
per quarter
per quarter
per quarter
per quarter
Systems Using Only Ground Water
<500
500 - 9,999
10,000-99,999
100,000-499,999
> 500,000
1
1
2
3
3
1
1
1
2
3
NA
NA
1
1
2
2
2
4
6
8
per year
per year
per quarter
per quarter
per quarter
1 Locations must be based on the system's recommendations for Stage 2 DBPR compliance monitoring locations in
its report to the State/Primacy Agency, unless the State/Primacy Agency requires different or additional locations.
Locations should be distributed throughout the distribution system to the extent possible.
2 Alternate between highest HAAS LRAA and highest TTHM LRAA locations among the existing Stage 1 average
resident time compliance locations. If the number of existing Stage 1 compliance locations is fewer than the
specified number for Stage 2, alternate between highest HAAS LRAA locations and highest TTHM LRAA locations
from the IDSE.
3 All systems must monitor during the month of highest DBP concentrations. Systems on quarterly monitoring must
take dual sample sets approximately every 90 days.
4 For the purposes of this EA, "surface water" systems are equivalent to "subpart H" systems and include systems
that use GWUDI.
5 Systems on quarterly monitoring must take dual sample sets every 90 days at each monitoring location, except for
subpart H systems serving 500-3,300 people. Systems on annual monitoring and subpart H systems serving 500-
3,300 people are required to take individual TTHM and HAAS samples (instead of a dual sample set) at the
locations with the highest TTHM and HAAS concentrations, respectively. Only one location with a dual sample set
per monitoring period is needed if the highest TTHM and HAAS concentrations occur at a same location and month
(if monitored annually).
NA = Not Applicable
Final Economic Analysis for the Stage 2 DBPR
ES-7
December 2005
-------
ES.4 Systems Subject to the Stage 2 DBPR
Exhibit ES.4a shows the baseline number of systems subject to the rule and the estimated number
that will perform various rule activities (implementation, IDSE monitoring, monitoring plans, and
operational evaluations).3 This baseline is derived from EPA's Safe Drinking Water Information System
(SDWIS) inventory, 4th quarter 2003 data. The systems are subdivided by type (CWS or NTNCWS),
source water type (either disinfecting ground water only, or surface water and mixed-source) and size
(small or large, based on population served). The number of ground water systems in column A
represents the subset of all ground water systems that currently disinfect.
As shown in column B, EPA estimates that all disinfecting CWSs and NTNCWSs will have to
perform at least minimal implementation activities (reading and understanding the rule, training, etc.).
All systems will also have to develop Stage 2 monitoring plans as shown in column F. The number of
systems performing IDSE monitoring (shown in column D), however, is only a fraction of all systems
because some will choose to perform studies or will receive waivers from IDSE requirements.
EPA has established a population-based monitoring approach for the Stage 2 DBPR, where
monitoring requirements are no longer based on the number of plants per system as under the Stage 1
DBPR but rather on the population served. As a result, the number of Stage 2 compliance samples
required per year for any particular system may stay the same, decrease, or increase from Stage 1
requirements. Appendix H includes information on changes in monitoring burden.
EPA expects that some number of Stage 2-compliant systems will find TTHM and HAAS levels
high enough to trigger the requirement for an operational evaluation. Column H shows the estimated
number of systems that may require operational evaluations.
In addition to those ground water systems that currently disinfect, EPA predicts that some
systems will install disinfection to comply with the anticipated Ground Water Rule (GWR). Exhibit
ES.4b shows the number of systems predicted to install disinfection based on information in the GWR EA
(USEPA 2000g) and system inventory data. Because the GWR is expected to be promulgated within 8
months after the Stage 2 DBPR is promulgated, EPA expects new systems adding disinfection to meet
GWR requirements to simultaneously achieve compliance with Stage 2 MCLs. Therefore, these systems
are not included in the treatment baseline. The IDSE will likely not apply to these systems because they
are expected to install disinfection after the IDSE requirement period is complete. Systems installing
disinfection for the GWR will, however, be required to prepare monitoring plans and in some cases,
monitor DBFs for the first time under Stage 2. Exhibit ES.4b shows that all newly disinfecting ground
water systems will prepare monitoring plans. All newly disinfecting systems will also need to conduct
compliance monitoring for the first time.
3 The baseline number of plants-as opposed to systems-that may have to make treatment technology
changes to meet rule requirements is different and discussed in section ES.5.1.
Final Economic Analysis for the Stage 2 DBPR ES-8 December 2005
-------
Exhibit ES.4a Number of Disinfecting Systems Subject to Non-Treatment-Related
Rule Activities
System Size
(Population Served)
Stage 2
DBPR
System
Baseline
A
Number and Percent of Systems Performing Various Rule Activities
Implementation
B C=B/A*100
IDSE Monitoring
D E=D/A*100
Stage 2 Monitoring
Plans
F G=F/A*100
Operational
Evaluations
H I=H/A*100
Surface Water and Mixed CWSs
<1 0,000
> 10,000
National Totals
9,397
2,406
11,803
9,397 100%
2,406 100%
11,803 100%
7,771 83%
2,038 85%
9,809 83%
8,160 87%
2,406 100%
10,566 90%
97 1%
327 14%
424 4%
Disinfecting Ground Water Only CWSs
<1 0,000
> 10,000
National Totals
28,806
1,423
30,229
28,806 100%
1,423 100%
30,229 100%
2,707 9%
258 18%
2,966 10%
11,801 41%
1,423 100%
13,225 44%
0 0%
0 0%
0 0%
Surface Water and Mixed NTNCWSs
<1 0,000
> 10,000
National Totals
771
6
777
771 100%
6 100%
777 100%
0 0%
5 83%
5 1%
0 0%
6 100%
6 1%
0 0%
0 0%
0 0%
Disinfecting Ground Water Only NTNCWSs
<1 0,000
> 10,000
National Totals
GRAND TOTAL
5,480
4
5,483
48,293
5,480 100%
4 100%
5,483 100%
48,293 100%
0 0%
1 24%
1 0%
12,780 26%
0 0%
4 100%
4 0%
23,800 49%
0 0%
0 0%
0 0%
424 1%
Note: Detail may not add to totals due to independent rounding.
Non-treatment-related rule activities, in addition to those shown in the table, also include routine compliance monitoring. Some systems
are expected to take more samples and some less from Stage 1 to Stage 2 depending on the number of plants in their systems. Overall
the Stage 2 DBPR results in an increase in the total number of compliance samples taken from the Stage 1 DBPR. See Exhibit H.8a,
column H, for the change in total samples for the different size categories.
Sources: (A), (B), (D), (F), and (H): Appendix H, Exhibit H.15a.
Final Economic Analysis for the Stage 2 DBPR
ES-9
December 2005
-------
Exhibit ES.4b Non-Treatment Rule Activities for Systems Installing Disinfection
to Comply with the Ground Water Rule
System Size
(Population Served)
Baseline No. of
Systems Adding
Disinfection for the
GWR
A
Stage 2 Monitoring
Plans
B
C=B/A*100
Surface Water and Mixed CWSs
<1 0,000
> 10,000
National Totals
0
0
0
0
0
0
-
-
-
Ground Water Only CWSs
<1 0,000
> 10,000
National Totals
1,030
13
1,042
1,030
13
1,042
100%
100%
100%
Surface Water and Mixed NTNCWSs
<1 0,000
> 10,000
National Totals
0
0
0
0
0
0
-
-
-
Ground Water Only NTNCWSs
<1 0,000
> 10,000
National Totals
GRAND TOTAL
1,509
1
1,510
2,552
1,509
1
1,510
2,552
100%
100%
100%
100%
Notes:
Detail may not add to totals due to independent rounding.
Non-Treatment-Related Rule Activities, in addition to those shown in
the table, include routine compliance monitoring for all systems.
Sources:
(A), (B), (D): Appendix H, Exhibit H.12b.
ES.5 National Benefits and Costs of the Stage 2 DBPR Preferred Regulatory Alternative
EPA has determined from its analysis of the available animal toxicological studies and human
epidemiological studies that the Stage 2 DBPR could provide benefits resulting from reduced incidence of
cancer, particularly bladder cancer, and reduced incidence of potential adverse reproductive and
developmental effects. The main category of benefits that EPA has quantified is the expected range of
avoided new cases of bladder cancer each year, including both fatal and non-fatal cases. In addition, EPA
has estimated the monetized value of avoiding these fatal and non-fatal bladder cancer cases.
The major steps in deriving and characterizing cancer cases avoided are the following: (1)
Estimate the current and future annual bladder cancer cases from all causes; (2) Estimate how many cases
can be attributed to DBP occurrence and exposure; and (3) Estimate the reduction in future cases
corresponding to anticipated reductions in DBP occurrence and exposure due to the Stage 2 DBPR. For
step 2, EPA has developed three approaches to estimating the number of bladder cancer cases attributable
to DBFs. Taken together, the three approaches provide a reasonable estimate of the range of potential
risk. For simplicity's sake, one estimate, based on a 2003 meta-analysis by Villanueva et al., is carried
through the full benefits analysis.
To assign a monetary value to avoided bladder cancer cases, EPA used the value of a statistical
life (VSL) for fatal cases and used two alternate estimates of willingness-to-pay to avoid non-fatal cases
(one based on curable lymphoma and the other based on chronic bronchitis). Exhibit ES.5 summarizes
Final Economic Analysis for the Stage 2 DBPR
ES-10
December 2005
-------
the estimated total number of bladder cases avoided and the resulting monetized benefits for the Stage 2
DBPR Preferred Regulatory Alternative. The cases avoided and monetized benefits are based on
reductions in average TTHM and HAAS concentrations that result from making treatment technology
changes.
There are several categories of unquantified potential health and non-health benefits resulting
from rule implementation that could contribute to the overall value of the benefits of the Stage 2 DBPR.
Two important categories are colon and rectal cancers, and reproductive and developmental effects. In
addition to bladder cancer, human epidemiology studies on chlorinated surface water have also reported
associations with colon and rectal cancers. In the Stage 1 DBPR, EPA concluded that early studies
suggested a small possible increase in rectal and colon cancers from exposure to chlorinated surface
waters. Since the Stage 1 DBPR, the database of studies on colon and rectal cancers continues to support
a possible association, but evidence remains mixed. For these reasons, EPA performed a sensitivity
analysis to determine potential reductions in rectal and colon cancer cases as a result of the Stage 2
DBPR.
Scientific knowledge about the association of reproductive and developmental health effects with
DBP exposure is not known well enough to quantify these risks or the benefits of reduced DBP exposure
in the primary benefits analysis. Nevertheless, although the results from different studies are mixed and
no causal link has been established between adverse reproductive or developmental effects and
chlorinated drinking water, a weight of evidence evaluation of the health effects data suggests a potential
association. EPA believes additional benefits from the Stage 2 DBPR could come from reducing
potential reproductive and developmental effects risks and that it is therefore important to provide some
quantitative indication of the possible magnitude of these benefits. To do this, EPA completed an
illustrative calculation of potential benefits for one specific reproductive effects endpoint (fetal loss).
Unquantified benefits are discussed in detail in Chapter 6 and are summarized in Exhibit 9.2.
EPA's national cost estimate includes costs incurred by CWSs and NTNCWSs for rule
implementation, the IDSE, preparing Stage 2 monitoring plans, conducting additional routine monitoring,
operational evaluations, and treatment technology changes (which account for the majority of the national
costs) as well as estimated costs to States/Primacy Agencies. Exhibit ES.5 summarizes the total national
annualized cost estimate for the Stage 2 DBPR Preferred Regulatory Alternative. Note that the exhibit
presents two estimates for national costs and monetized benefits, depending upon the discount rate used
for present value calculations and annualizing one-time costs.4
Sections ES.5.1 through ES.5.3 summarize the methods used to estimate the number of plants
making treatment technology changes to comply with the rule, and the benefits and costs resulting from
these treatment technology changes. Chapters 3, 5, 6, and 7 and the appendices provide a more complete
discussion of all data and calculations used to derive the results in Exhibits ES.5.
4 For the Stage 2 DBPR cost and benefit analyses, calculations are made using two discount rates-3 and 7
percent-to represent current policy evaluation methodologies. Chapters 6 and 7 provide additional information on
the derivation of these rates.
Final Economic Analysis for the Stage 2 DBPR ES-11 December 2005
-------
Exhibit ES.5 Summary of Estimated National Benefits and Costs of the Stage 2 DBPR
Preferred Regulatory Alternative ($ Million / Year)
Type of Cost or Benefit
Estimated Number of Bladder Cancer Cases Avoided per Year
Includes both fatal and non-fatal cases. Expected value (90% confidence bounds) Based
on Villanvueva et al. 2003 for baseline risk estimates, smoking/lung cancer cessation lag
model, and TTHM as an indicator of all DBFs.
Benefits and
Costs Based on
Annualization
Discount Rate of
3%
Benefits and
Costs Based on
Annualization
Discount Rate of
7%
Annualized Monetized Benefits of Bladder Cancer Cases Avoided
WTP for Lymphoma as the basis for non-fatal cases,
smoking/lung cancer cessation lag model (90% confidence bounds)
WTP for Chronic Bronchitis as the basis for non-fatal cases,
smoking/lung cancer cessation lag model (90% confidence bounds)
Annualized Total Costs (90% confidence bounds)
Annualized Monetized Benefits of Bladder Cancer Cases Avoided
WTP for Lymphoma as the basis for non-fatal cases,
smoking/lung cancer cessation lag model (90% confidence bounds)
WTP for Chronic Bronchitis as the basis for non-fatal cases,
smoking/lung cancer cessation lag model (90% confidence bounds)
Annualized Total Costs (90% confidence bounds)
Surface Water
< 10,000
9.0
$49.3
($7.5 -$11 3.8)
$24.5
($5.3 - $54.5)
$12.2
($7.2 -$17.3)
$39.6
($6.0 - $91 .4)
$19.7
($4.3 - $43.7)
$11.5
($7.0 -$16.2)
> 10,000
260.5
$1 ,430.3
($21 7.6 -$3304.0)
$712.7
($154.3 -$1581 .2)
$35.7
($21.0 -$51.1)
$1,165.5
($177.4 -$2690.6)
$580.3
($85.5 - $873.2)
$36.1
($21 .3 -$51.1)
Disinfecting Ground Water
< 10,000
3.5
$19.3
($2.9 - $44.6)
$9.6
($2.1 -$21.4)
$18.1
($16.1 -$20.2)
$15.5
($2.4 - $35.8)
$7.7
($1.7-$17.1)
$17.0
($15.0 -$19.0)
> 10,000
5.8
$31.9
($4.9 - $73.7)
$15.9
($3.4 - $35.3)
$11.1
($10.2 -$11. 9)
$25.9
($3.9 - $59.9)
$12.9
($2.8 - $28.6)
$10.5
($9.6 - $1 1 .3)
State/
Primacy
Agencies
$1.7
$1.7
Total
278.9
(102.9-540.9)
$1 ,530.8
($232.9 -$3536.1)
$762.8
($165.2 -$1692.3)
$78.8
($56.2 -$102.2)
$1 ,246.5
($189.8 -$2877.7)
$620.7
($134.6 -$1375.8)
$76.8
($54.6 - $99.4)
Notes:
Detail may not add due to independent rounding. Monetized benefits and costs are discounted and annualized in 2003 dollars. Costs are for CWSs and NTNCWSs and include treatment and non-
treatment costs.
Sources:
Benefits for the Stage 2 DBPR are estimated using three approaches for estimating baseline risk, three different cessation lag models, and either TTHMs or HAASs as an indicator. Nominal
benefits presented here are estimated using Villanueva et al. (2003) for baseline risk, TTHMs as an indicator, and the Smoking/Lung Cancer cessation lag model. Using TTHM or HAAS as an
indicator for all DBPs produces similar results. Because Villaneuva et al. (2003) and the Smoking/Lung Cancer cessation lag model result in benefits estimates that are in between the other
alternatives, results are presented as a representative comparison to costs. Chapter 6 presents results for the full range of alternative approaches to estimating baseline risk and cessation lag.
The 90 percent confidence bounds around cost address uncertainty in the compliance forecast methodology, potential impacts of the IDSE, and unit costs. 90 percent confidence bounds around
cases avoided accounts for uncertainty in the compliance forecast, PAR, cessation lag model form, and predicted DBP reduction from the compliance forecast. 90 percent confidence bounds
around monetized benefits also reflect uncertainty in VSL and WTP inputs. Causality has not been established, so lower bound of potential risks may be as low as zero.
Totals for bladder cancer cases avoided and monetized benefits from Exhibit 6.28. Detail for source and size provided in Appendices E and F.
Annualized total costs and state/primacy agency costs derived from Exhibits 7.5a and 7.5b.
Final Economic Analysis for the Stage 2 DBPR
ES-12
December 2005
-------
ES.5.1 Derivation of the Stage 2 DBPR Compliance Forecast and Consequent Reductions in DBFs
Changes in concentrations of DBFs are the direct result of changes in treatment technologies.
Therefore, it is important that EPA uses consistent methodologies for forecasting treatment technology
changes and predicting reductions in DBFs (specifically, the reductions in TTHM and HAAS
concentrations that are used in the benefits analysis). This section summarizes the tools used and key
assumptions for the Stage 2 DBPR compliance forecasts and the consequent reductions in TTHM and
HAAS concentrations.
Since the rule was proposed, EPA has modified the compliance forecast methodology in an
attempt to further quantify uncertainties in the analysis. Specifically, EPA has developed a second
method to predict the number of surface water plants making treatment technology changes and
consequent reductions in TTHM and HAAS concentration, which supplements the Surface Water
Analytical Tool (SWAT) predictions. EPA has also quantified uncertainty in the potential impacts of the
Initial Distribution System Evaluation (IDSE). Uncertainties are characterized using Monte Carlo
simulation in the cost and benefits models.
Predictive Tools Used to Develop the Compliance Forecast
EPA uses different methods for different system sizes and source water types to develop the
compliance forecasts, as shown in Exhibit ES.6. Because extensive data were available from the
Information Collection Rule (ICR), analysis tools and methods drawing specifically on those data were
used to develop compliance forecasts for large surface and ground water systems. For large surface water
systems, EPA used two different methodologies, both drawing from ICR data: the Surface Water
Analytical Tool (SWAT) and the ICR Matrix Method. The ICR Matrix Method uses TTHM and HAAS
distribution system data from the ICR to predict how many plants will need to make treatment technology
changes for a specific regulatory alternative. SWAT uses a series of decision rules and algorithms to
predict the number of plants making treatment technology changes and the type of treatment they will
install for a specific regulatory alternative, based on source water quality and existing treatment as
reported in the ICR database. The ICR Matrix Method and SWAT produce different results; thus, results
from both are incorporated into the cost model using a Monte Carlo simulation model to account for
uncertainties in both methods.
The forecast for large ground water systems was generated using the ICR Ground Water Delphi
process, which convened a group of experts to evaluate plant configurations and predict technology
selection for ground water plants that did not meet rule requirements. Compliance forecasts for large
surface and ground water systems were used to generate forecasts for medium and small systems, making
adjustments to account for different operational and water quality characteristics that exist in the latter.
Final Economic Analysis for the Stage 2 DBPR ES-13 December 2005
-------
Exhibit ES.6 Tools Used to Develop the Stage 2 DBPR Compliance Forecasts
System Size
(Population Served)
Large (> 100,000 people)
Medium (10,000 to 99,999 people)
Small (<1 0,000 people)
Source Water Category
Surface Water
The Surface
Water Analytical
Tool (SWAT)
Extrapolation
from SWAT
Extrapolation
from SWAT,
adjusted to deal
with small
system-specific
issues
ICR Matrix
Method
Extrapolation
from ICR
Matrix Method
Extrapolation
from ICR
Matrix Method
Disinfecting Ground Water
ICR Ground Water Delphi
Group
Extrapolation from large
ground water system results
Extrapolation from large
ground water system results,
adjusted to deal with small
system-specific issues
Tools Used to Predict Changes in DBF Levels
For the benefits analysis, EPA needs information on the changes in both average and peak
TTHM and HAAS levels that result from implementation of the Stage 2 DBPR. Estimates of bladder
cancer cases avoided and the colon and rectal cancer sensitivity analysis are based on reductions in
average levels, while the illustrative analysis of potential developmental and reproductive health benefits
is based on reductions in occurrences of peak concentrations.
To predict changes in average TTHM and HAAS levels for surface water systems, EPA uses two
methods: the ICR Matrix Method and SWAT. As noted in the previous section, the ICR Matrix Method
evaluates distribution system data to identify plants that would need to make treatment technology
changes to meet a specific regulatory alternative. To predict average DBP concentrations occurring after
treatment technology changes, EPA used TTHM and HAAS occurrence data for those surface water
plants already using chloramines and/or advanced technologies and in compliance with Stage 2 at the
time of the ICR. The predicted average TTHM and HAAS levels for all surface water plants is a
weighted average for plants that do and do not change treatment technology.
SWAT is a model that uses a series of decision rules and algorithms to predict (1) which surface
water plants need to change treatment technology to meet a specific regulatory alternative, (2) which
treatment technology those plants will select based on a least cost decision tree, and (3) resulting changes
in the national average TTHM and HAAS levels in distribution systems. As with the compliance forecast,
SWAT and the ICR Matrix Method produce different results for DBP levels; thus, both are incorporated
into a Monte Carlo simulation model to further quantify uncertainty in the national benefits estimate.
ICR ground water plant data were not robust enough to develop a ground water model similar to
SWAT; therefore, the ICR Matrix Method is the only approach used to predict reductions in average
TTHM and HAAS levels for these systems.
Final Economic Analysis for the Stage 2 DBPR
ES-14
December 2005
-------
To predict changes in the occurrence of peak TTHM and HAAS concentrations, EPA used only
the ICR Matrix Method5. Similar to the way in which it is used to evaluate changes in average TTHM
and HAAS concentrations, the ICR Matrix Method evaluates distribution system data to identify plants
that would need to make treatment technology changes to meet a specific regulatory alternative. To
predict occurrence of peaks after treatment technology changes, EPA analyzed TTHM and HAAS
occurrence data for those surface water plants already using chloramines and/or advanced technologies at
the time of the ICR. The predicted occurrence of peaks for all plants is a weighted average for plants that
do and do not make treatment technology changes.
Accounting for the Stage 1 DBPR
For cost and benefit analyses, the compliance forecast and consequent reduction in DBFs needs
to represent treatment technology changes from the pre-Stage 2 baseline (i.e., after implementation of the
Stage 1 DBPR). The best data available to characterize large plants are from the ICR, which were
collected before the Stage 1 DBPR compliance deadlines and likely represent pre-Stage 1 conditions6.
The compliance forecast, therefore, needs to account for treatment technology changes as a result of the
Stage 1 DBPR before predicting changes that are needed for the Stage 2 DBPR. Similarly, the post-Stage
2 TTHM and HAAS predictions need to take into account changes as a result of the Stage 1 DBPR before
predicting reductions that will occur as a result of the Stage 2 DBPR.
EPA uses a "delta" compliance forecast method that was developed by the Microbial /
Disinfection Byproducts (M-DBP) Technical Working Group (TWG) to characterize the incremental
changes due to Stage 2 relative to Stage 1. The method has four steps. First, EPA characterizes treatment
technologies and TTHM and HAAS occurrence for the Pre-Stage 1 baseline. Second, EPA predicts
treatment technology changes and subsequent reduction in TTHM and HAAS levels from Pre-Stage 1
baseline to post-Stage 1 DBPR conditions. Third, treatment technology changes and subsequent
reductions in TTHM and HAAS levels are predicted from pre-Stage 1 baseline to post-Stage 2 DBPR
conditions. Lastly, results from Step 2 are subtracted from Step 3 to calculate the incremental treatment
technology change and TTHM/HAA5 reduction from post-Stage 1 DBPR to post-Stage 2 DBPR
conditions.
This delta method was selected by the M-DBP TWG over a more direct, two step approach (i.e.,
predict pre-Stage 2 conditions and then use the pre-Stage 2 conditions to predict impacts for Stage 2)
because modeling tools are not able to predict the treatment technology selection or TTHM, HAAS,
bromate, and chlorite levels at the plant level. The delta approach compensates, at the national level, for
potential errors in treatment technology selection and resulting TTHM and HAAS concentrations
predicted for Stage 2. The TWG believed that using the delta approach reduces the impact of uncertainty
in SWAT predictive equations for TTHM and HAAS. The delta approach is used with both the ICR
Matrix Method and SWAT analyses.
5 Although the SWAT model was calibrated to national average TTHM and HAAS concentrations in
distribution systems and validated against industry treatment technology predictions, it was not calibrated to plant-
level DBF predictions and, thus, could not be used to assess changes in occurrence of peak levels. See Appendix A
for more information on SWAT.
6 There is uncertainty in using the ICR data to represent pre-Stage 1 conditions because some plants may
have begun making changes prior to the ICR in anticipation of the Stage 1 DBPR (McGuire et al., 2002). See
Section 3.8 for a full discussion of uncertainties in ICR data.
Final Economic Analysis for the Stage 2 DBPR ES-15 December 2005
-------
Accounting for the IDSE
Because the purpose of the IDSE is to identify Stage 2 compliance monitoring locations with high
DBF levels, it is possible that systems may measure higher DBF levels at Stage 2 compliance monitoring
sites than were measured under the ICR. This suggests that the number of plants predicted to make
treatment technology changes, the level of treatment they select, and the resulting reductions in TTHM
and HAAS levels, all based on ICR data, could be underestimated.
The M-DBP TWO recommended that the Stage 1 and Stage 2 compliance forecast methodology
incorporate an operational safety margin of 20 percent to represent the operational level (i.e., 80 percent
of the MCL) at which systems typically take some action to ensure consistent compliance with a new
drinking water standard and the level at which systems target new treatment technologies to meet the
standard. EPA believes that this safety margin already accounts for the impacts of the IDSE for some
systems, including small systems, ground water systems, and those using chloramines7. EPA believes,
however, that the 20 percent safety margin is not sufficient to account for the potential impacts of the
IDSE on large and medium surface water systems because spatial variability of DBP levels and
distribution system complexity are greatest in these systems. Since the proposal, EPA developed a
methodology that analyzed ICR data from surface water plants to assess the extent of spatial variability of
TTHM and HAAS levels and used this as a basis for quantifying the impacts of the IDSE for large and
medium surface water systems.
Results
Exhibit ES.Va shows the mean number and percent of plants expected to make advanced
treatment technologies changes to meet the requirements of the Stage 2 DBPR. Advanced technologies
include alternative disinfectants such as ozone, UV, and chlorine dioxide, and DBP precursor removal
technologies such as granular activated carbon adsorption and nanofiltration. The 90 percent confidence
intervals around the mean estimate for surface water systems account for alternative compliance forecast
methodologies (SWAT and the ICR Matrix Method) and uncertainty in the potential impacts of the IDSE.
Because one method (instead of two) was used to predict ground water plants making treatment
technology changes, Exhibit ES.7a only presents a mean value for these plants.
Exhibit ES.7b shows the reduction in the national average TTHM and HAAS concentrations
occurring in drinking water distribution systems as a result of treatment technology changes to meet Stage
2 DBPR requirements. The 90 percent confidence intervals around the mean estimate for surface water
systems account for alternative methodologies (SWAT and the ICR Matrix Method) and the potential
impacts of the IDSE.
7EPA believes that the 20 percent safety margin accounts for potential impacts of the IDSE for small
systems because their distribution systems are not as complex when compared to large systems. EPA also believes
that the safety margin accounts for the IDSE for ground water systems because the year-to-year variability in source
water quality (and thus, TTHM and HAAS formation) is low. Chloramine systems generally observe lower spatial
and temporal variability in TTHM and HAAS distribution system levels (USEPA 2005k); thus, EPA believes the 20
percent safety margin accounts for potential impacts of the IDSE for these systems.
Final Economic Analysis for the Stage 2 DBPR ES-16 December 2005
-------
Exhibit ES.7a Plants Making Treatment Technology Changes, Preferred
Regulatory Alternative
System Size (Population
Served)
Stage 2 DBPR
Plant Baseline
A
Number of Plants Making
Treatment Technology Changes
Mean
B
5th %ile
C
95th %ile
D
Percentage of Plants Making
Treatment Technology Changes
Mean
E=B/A
5th %ile
F=C/A
95th %ile
G = D/A
Primarily Surface Water CWSs
<1 0,000
> 10,000
National Totals
3,996
2,555
6,552
352
373
724
196
174
371
506
582
1,088
8.8%
14.6%
11.1%
4.9%
6.8%
5.7%
12.7%
22.8%
16.6%
Primarily Ground Water CWSs
<1 0,000
> 10,000
National Totals
40,376
7,044
47,419
1,170
145
1,314
/"
/"
2.9%
2.1%
2.8%
//
Primarily Suface Water NTNCWSs
<1 0,000
> 10,000
National Totals
760
6
766
68
1
69
38
0
39
98
1
100
9.0%
14.6%
9.0%
5.0%
6.8%
5.0%
12.9%
22.8%
13.0%
Primarily Ground Water NTNCWSs
<1 0,000
> 10,000
National Totals
Grand Total All Plants
5,480
4
5,483
60,220
153
0
154
2,261
1,877
2,655
2.8%
2.1%
2.8%
3.8%
//
/
3.1%
4.4%
Note: Detail may not add to totals due to independent rounding. Treatment changes include adding advanced
technologies and/or chloramines.
Sources:
(A) Exhibit 3.2, column AB. Represents 4th quarter 2003 SDWIS data. System baseline converted to plant baseline
through four steps: 1) link purchased systems to their respective sellers, 2) estimate percent of ground water systems
that disinfect, 3) categorize systems by primary source, and 4) multiply the system inventory by estimate of mean plants
per system to produce plant inventory.
(B) - (D) Exhibit 7.3. The 90 percent confidence intervals for surface water systems represent alternative compliance
forecast methodologies (SWAT and the ICR Matrix Method) and uncertainty in the potential impacts of the IDSE.
Final Economic Analysis for the Stage 2 DBPR
ES-17
December 2005
-------
Exhibit ES.7b Estimated Reduction in Average TTHM and HAAS from Pre-Stage 2
to Post-Stage 2, Preferred Regulatory Alternative
Source Water
Type
SW
GW
System Size
(Population Served)
Large (> 10,000)
Small (< 10,000)
Large (> 10,000)
Small (< 10,000)
All Systems*
Population
A
160,935,736
8,422,403
65,152,168
28,514,211
263,024,518
TTHM
Pre-Stage 2
Level (ug/L)
B
35.5
35.5
13.2
14.7
27.69
Post-Stage 2 Level (ug/L)
Mean
5th
95th
C
32.2
32.9
13.0
14.4
25.53
33.7
33.8
26.44
30.7
32.0
24.60
Percent Reduction
Mean
5th
95th
D = (B - C) / B
9.2%
7.2%
1.4%
2.0%
7.8%
5.1%
4.7%
4.5%
13.5%
9.7%
11.2%
Source Water
Type
SW
GW
System Size
(Population Served)
Large (> 10,000)
Small (< 10,000)
Large (> 10,000)
Small (< 10,000)
All Systems*
Population
A
160,935,736
8,422,403
65,152,168
28,514,211
263,024,518
HAAS
Pre-Stage 2
Level (ug/L)
E
25.0
25.0
7.0
7.8
18.67
Post-Stage 2 Level (ug/L)
Mean
5th
95th
F
22.5
23.1
6.6
7.4
16.96
23.7
23.8
17.71
21.1
22.4
16.12
Percent Reduction
Mean
5th
95th
G = (E - F) / E
9.9%
7.6%
4.5%
6.3%
9.2%
5.2%
4.7%
5.2%
15.5%
10.5%
13.7%
Notes: Detail may not add due to independent rounding.
All TTHM and HAAS concentrations represent the mean of all plant-mean concentrations. Total for "All Systems" is calculated using a Monte Carlo analysis.
Results for large SW systems for the preferred alternative represent the combined results for the 20% and 25% safety margins.
* TTHM and HAAS concentrations for all systems are the population-weighted values
Sources:
(A) SDWIS 2003 3rd quarter freeze, community water system population (Exhibit 3.3, CWSs only).
(B) and (E) Exhibit 5.22
(C) and (F) Outputs from the benefits Monte Carlo simulation model. Confidence bounds for large and medium SW systems account for uncertainties in compliance
forecast methodologies and potential impacts of the IDSE. Confidence bounds for small SW systems account for uncertainties in compliance forecast methodologies.
ES.5.2 Derivation of Benefits
Three categories of potential benefits are addressed in this EA: those associated with reductions
in the incidence of bladder cancer, those associated with decreasing cases of colon and rectal cancers, and
those associated with reductions in the incidence of adverse reproductive and developmental health
effects. The primary benefits analysis in this EA is based on reductions in bladder cancer cases. Potential
benefits associated with reduced incidence of colon and rectal cancers are quantified in a sensitivity
analysis, while potential benefits associated with decreasing adverse developmental and reproductive
health effects (specifically, fetal losses) are presented as an illustrative calculation.
EPA used similar approaches to estimate the number of annual bladder cancer cases avoided (the
primary benefits analysis), the number of annual colon and rectal cancer avoided (the sensitivity
analysis), and the number of annual fetal losses avoided (the illustrative calculation). The major steps in
deriving and characterizing cases avoided are:
• Estimate the current and future annual cases of illness from all causes
• Estimate how many cases can be attributed to DBP occurrence and exposure
• Estimate the reduction in future attributable cases corresponding to anticipated reductions in
DBP occurrence and exposure due to the Stage 2 DBPR
All benefit calculations were performed using the Stage 2 DBPR Benefits Model (USEPA 2005h).
Final Economic Analysis for the Stage 2 DBPR
ES-18
December 2005
-------
For bladder cancer, EPA computed the monetized benefits of the Stage 2 DBPR by multiplying
the estimated number of bladder cases avoided by the estimated monetary value associated with avoiding
both fatal and non-fatal cases of bladder cancer. The value of a statistical life (VSL) was used for fatal
bladder cancers, while two alternate estimates of willingness to pay (WTP) to avoid non-fatal bladder
cancer are used (one based on avoiding a case of curable lymphoma and the other based on avoiding a
case of chronic bronchitis). EPA also computed the benefits for the reduction in colon and rectal cancer
by using the same VSL and WTP estimates. EPA recognizes that there could be additional value
associated with the number of potential avoided fetal losses estimated in the illustrative calculation.
However, the Agency is unable at this time either to develop a specific estimate of this value or to use a
benefit transfer method to estimate the value from studies that address other endpoints because of
associated uncertainty (see Section 6.8 for a full discussion of this issue).
Bladder Cancer
To calculate potential benefits from reduced incidence of bladder cancer cases, EPA began by
estimating the number of new bladder cancer cases occurring per year from all causes. The National
Cancer Institute's Surveillance, Epidemiology, and End Results (SEER, 2004) program provides data on
cancer rates (new cases per 100,000 population per year) as a function of age. EPA used this information
in conjunction with population-by-age data from the 2000 U.S. Census to estimate the number of new
cases of bladder cancer. Results show that the number of new bladder cancer cases per year starts to
increase at about age 35 and peaks at 1,500 to 2,000 cases from about age 66 to 85. Although the annual
rate of bladder cancer does not decline much after age 85, the incidence of new bladder cancers does,
which represents the overall decline in the number of individuals alive after that age. The resulting total
number of new bladder cancer cases per year, 56,506, is slightly lower than that currently estimated by
the American Cancer Society (ACS).8 This likely represents EPA's use of the census population data
from 2000.
To estimate the baseline number of cases attributable to DBP exposure, EPA used three different
approaches:
• Using the range of Population Attributable Risk (PAR) values derived from consideration of
5 individual epidemiology studies used for the Stage 1 EA and the Stage 2 proposal EA
(yields apre-Stage 1 range of best estimates for PAR of 2% to 17%).
• Using the Odds Ratio (OR) of 1.2 from the Villanueva et al. (2003) meta-analysis that reflects
both sexes, ever exposed population from the studies considered (yields a pre-Stage 1 best
estimate for PAR of-16%)
• Using the Villanueva et al. (2004) pooled data analysis to develop a dose-response
relationship for OR as a function of Average TTHM. The dose-response relationship was
modeled as linear with an intercept of OR = 1.0 at TTHM exposure level = 0 (yields a pre-
Stage 1 best estimate for PAR of-17%)
Taken together, the three approaches provide a reasonable estimate of the range of potential risk.
For the sake of simplicity, EPA carried only one these approaches, that based on Villanueva et al. (2003),
through the entire benefits model. EPA notes that the existing epidemiological evidence has not
conclusively established causality between DBP exposure and any health risk endpoints, so the lower
bound of potential risks may be as low as zero.
The American Cancer Society estimated in 2004 that 60,240 new cases of bladder cancer would occur in
the U.S. population that year (ACS website, 2004).
Final Economic Analysis for the Stage 2 DBPR ES-19 December 2005
-------
To quantify the reduction in cases, EPA assumes that there is a linear relationship between
average DBF concentration and bladder cancer risk. Thus, percent reductions in national average DBFs
are used to determine the percent reductions in bladder cancer cases attributable to DBFs. Predicted
reductions in national average TTHM and HAAS levels resulting from predicted treatment technology
changes to comply with Stage 2 were used as indicators of overall chlorination DBF reductions. The
baseline cases attributable to DBFs multiplied by the percent reductions in TTHM or HAAS
concentrations result in the estimated annual bladder cancer cases "ultimately avoidable" for the Stage 1
and Stage 2 rules.
Over the long run, the annual cases ultimately avoidable (derived as described above) will be
attained. They will not be achieved instantaneously, however. Research shows that a lag period (referred
to as "cessation lag") exists between the point in time when reduction in exposure to a carcinogen occurs
and the point in time when the full risk reduction benefit of that exposure reduction is realized by affected
individuals. Because there is no epidemiological or other empirical data available that specifically
address the rate of achieving bladder cancer benefits resulting from DBF reductions, EPA uses data from
three epidemiological studies that address the rate of risk reduction following exposure reduction to other
carcinogens (cigarette smoke and lung cancer, cigarette smoke and bladder cancer, and arsenic and
bladder cancer) to generate three possible cessation lag functions for bladder cancer and DBFs.
The cessation lag functions are used to project the number of bladder cancer cases avoided each
year after implementation as a result of the Stage 2 DBPR over a 100-year period. A 100-year period was
selected as the timeframe after which effectively all of the exposed population is composed of individuals
exposed only to post-Stage 2 levels for their entire lifetime. At that time (and from that point forward)
the annual bladder cancer cases ultimately avoidable are achieved for the exposed population. The
projected number of cases avoided each year is then further adjusted forward in time to reflect when
systems are expected to install new treatment to reduce DBFs based on the rule implementation schedule.
Although a 100-year cessation lag period is modeled, annual avoided cases of bladder cancer are
calculated primarily for the first 25 years after rule promulgation. A 25-year time period was used to
coincide with the estimated life span of capital equipment and a time lag of five to ten years for
technology installation after rule promulgation.
The final step in the benefit calculation is to monetize the average annual cases avoided. This is
done by applying economic values for avoided illnesses and deaths. EPA has estimated that 74 percent of
bladder cancer cases are non-fatal (USEPA, 1999a). The value of avoiding non-fatal bladder cancer cases
is based on people's WTP for incremental reductions in the risk they face of contracting cancer. The
metric of WTP to avoid an increased risk includes the desire to avoid treatment costs, pain and
discomfort, productivity losses, and any other adverse consequences related to a non-fatal case of bladder
cancer. Because specific estimates of WTP for avoiding non-fatal bladder cancer are not available, EPA
estimated values from two other non-fatal illnesses: curable lymphoma and chronic bronchitis. Both are
considered valid estimates of WTP for non-fatal cancer.
For fatal bladder cancer cases, VSL is used to capture the value of benefits. The VSL represents
an estimate of the monetary value of reducing risks of premature death from cancer. Therefore, it is not
an estimate of the value of saving a particular individual's life. Rather, it represents the sum of the values
placed on small individual risk reductions across an exposed population. Other economic factors are
taken into consideration when calculating benefits over time, such as income growth and social discount
rates.
Final Economic Analysis for the Stage 2 DBPR ES-20 December 2005
-------
There are several areas of uncertainty with respect to quantified benefits for bladder cancer.
Many are described qualitatively in the analysis, while others are incorporated explicitly as follows:
• There is uncertainty in the percent reduction in TTHM and HAAS concentrations resulting
from predicted treatment technology changes (i.e., the compliance forecasts). Uncertainty in
SWAT and potential impacts of the IDSE are quantified in the primary analysis.
• Three approaches were used to estimate the baseline number of bladder cancer cases
attributable to DBF exposure. For the sake of simplicity, one approach using data from
Villanueva et al. (2003) was carried through the full benefits model.
• The estimated PAR values from the Villanueva et al. (2003) meta-analysis include confidence
bounds resulting from statistical uncertainty in the odds ratio underlying the PAR calculation.
The confidence bounds from Villanueva et al. (2003) capture a significant portion of the
confidence intervals of the other two approaches.
Three independent cessation lag models derived from three different epidemiological studies
are used in the model. Also, two functional forms are used for each of these data sets and
uncertainty in the parameters of those functions is included in the analysis.
EPA uses two alternatives for valuing non-fatal bladder cancer.
Colon and Rectal Cancers
Human epidemiology studies on chlorinated surface water have reported associations with colon
and rectal cancers. Colon and rectal cancers combined are the third most common type of new cancer
cases and deaths in both men and women in the U.S., excluding skin cancers. Therefore, any benefit from
reducing the incidence of colon and rectal cancers could be significant; hence EPA includes a quantitative
sensitivity analysis for benefits from avoiding colon and rectal cancers due to Stage 2 DBP reductions.
EPA estimated the reduction in colon and rectal cancer in a similar manner to bladder cancer
cases. Background incidence data were available from the SEER cancer registry and two quality studies
were chosen to estimate a PAR value for DBFs. Using the percent reductions in DBFs and the smoking
and lung cancer cessation lag model, the number of colon and rectal cancer cases avoided annually were
estimated and monetized with the same VSL and WTP estimates as for bladder cancer.
Developmental And Reproductive Health Effects
As noted previously, EPA believes additional benefits from this rule could come from reductions
in developmental and reproductive health effects, although the relationship of these effects to DBP
exposure is not known well enough to quantify risks or benefits in the primary analysis. EPA was able to
do an illustrative calculation of the potential benefits of reducing the risk of fetal loss, the non-cancer
effect for which the most epidemiological data exist in relation to DBP exposure. Because approximately
one million of the six million pregnancies each year in the United States end in a miscarriage or stillbirth
(Ventura et al. 2000), avoiding even a small risk attributable to DBP exposure by reducing DBP levels
may result in a significant number of avoided fetal losses.
EPA estimated the reduction in potential fetal losses in a similar manner to bladder cancer cases.
A range of possible PAR values for relating annual fetal losses to DBP exposure was obtained from
available epidemiological studies. Reductions in the number of peak DBP events due to the Stage 1
DBPR and the Stage 2 DBPR were estimated. Reductions in exposure to peak DBFs were assumed to be
proportional to reductions in peak DBP events. Like the analysis of bladder cancer, there is uncertainty in
Final Economic Analysis for the Stage 2 DBPR ES-21 December 2005
-------
fetal loss PAR values, reflected in the range of values used in the analysis. There are other important
uncertainties in this illustrative calculation, including the assumed proportional relationship between
reduction in fetal losses and reduction in exposure to peak levels due to the Stage 2 DBPR.
ES.5.3 Derivation of Costs
To estimate the total national costs of the Stage 2 DBPR, EPA calculated the incremental costs to
be incurred by PWSs and States/Primacy Agencies from the Stage 1 DBPR to the Stage 2 DBPR. Cost
analyses for PWSs include identifying treatment process improvements that systems may make, as well as
estimating the costs to implement the rule, conduct IDSEs, prepare monitoring plans, perform additional
routine monitoring, and conduct operational evaluations (referred to as "non-treatment" activities in this
document). The cost analysis for States/Primacy Agencies includes estimates of the labor burdens that
they would face, such as training employees in the requirements of the Stage 2 DBPR, responding to
PWS reports, and record keeping. Cost calculations are performed using the Stage 2 DBPR Cost Model
(USEPA 2005i). The methodology for estimating treatment and non-treatment related costs for systems
is discussed in the next several paragraphs, followed by a discussion of uncertainties. (A complete
discussion of the cost analysis is provided in Chapter 7.)
All treatment costs are based on mean unit cost estimates for advanced technologies and
chloramines. Technology unit cost estimates are in the form of "dollars per plant" for initial capital and
ongoing operation and maintenance (O&M) activities. Derivations of unit costs for a wide range of
system sizes are described in detail in the document, Technologies and Costs Document for the Final
Long Term 2 Enhanced Surface Water Treatment Rule and Final Stage 2 Disinfectants and Disinfection
Byproducts Rule (USEPA 2005n). EPA combines the compliance forecast results (i.e., predicted number
of plants changing technology and which technologies those plants select, derived as described in Section
ES.5.1) with unit costs to produce total national treatment cost estimates.
Non-treatment costs for implementation, the IDSE, monitoring plans, additional routine
monitoring, and operational evaluations are based on estimates of labor hours for performing these
activities and on additional laboratory costs. For all non-treatment cost calculations, EPA used the Stage
2 DBPR system baseline shown in Exhibit ES.4a plus the estimate of additional ground water systems
expected to install disinfection for the GWR as shown in ES.4b (for routine monitoring and monitoring
plan costs only).
EPA recognizes that systems vary with respect to many of the attributes that are used as input
parameters to the Stage 2 DBPR cost model (e.g., plants per system, population served, flow per
population, labor rates). In most cases, there is insufficient information to characterize fully the
variability on a national scale. EPA believes that mean values for the various input parameters are
adequate to generate EPA's best estimate of national costs for the rule.
EPA has quantified several large areas of uncertainty in costs by quantifying uncertainty in
compliance forecasts, as noted in section ES.5.1. There is uncertainty in the national average unit capital
and O&M costs for the various technologies expected to be implemented in response to the Stage 2
DBPR. This uncertainty has been incorporated into the cost model (using Monte Carlo simulation
procedures). The national costs of the Stage 2 DBPR summarized in Exhibit ES.5 show both the
expected values and the 90 percent confidence bounds on the national cost estimates obtained from the
cost model.
Final Economic Analysis for the Stage 2 DBPR ES-22 December 2005
-------
ES.6 Estimated Impacts on Household Costs
EPA assumes that systems will, to the extent possible, pass cost increases on to their customers
through increases in water rates. Exhibit ES.8 presents estimated annual household cost increases for the
Stage 2 DBPR Preferred Regulatory Alternative. The top half of the exhibit shows summary statistics for
all households served by systems subject to the rule, including those that will not make treatment
technology changes but will incur other minimal costs, such as for rule implementation or additional
routine monitoring. The bottom half shows statistics just for those households served by systems actually
making treatment technology changes to comply with the rule (see Exhibit ES.7a for estimates of the
percent of plants making treatment technology changes). Because treatment technology changes
represent the majority of rule costs, this provides insight into how the rule will affect that segment of the
population most impacted by the rule.
Exhibit ES.8 Summary of Annual Household Cost Increases
Households Served by All Plants
All Systems
All Small Systems
SW< 10,000
SW> 10,000
GW< 10,000
GW> 10,000
Total Number of
Households Served
101,553,868
14,261,241
3,251 ,893
62,137,350
11,009,348
25,155,277
Mean Annual
Household
Cost Increase
$ 0.62
$ 2.20
$ 4.58
$ 0.46
$ 1.49
$ 0.13
Median Annual
Household
Cost Increase
$ 0.03
$ 0.10
$ 0.79
$ 0.02
$ 0.02
$ 0.00
90th Percentile
Annual
Household Cost
Increase
$ 0.36
$ 0.79
$ 2.69
$ 0.35
$ 0.39
$ 0.03
95th Percentile
Annual
Household Cost
Increase
$ 0.98
$ 2.57
$ 7.24
$ 1.81
$ 0.99
$ 0.08
Percentage of
Annual
Household Cost
Increase < $12
99%
97%
95%
99%
98%
1 00%
Percentage of
Annual
Household Cost
Increase < $120
100%
100%
99%
100%
100%
100%
Households Served by Plants Adding Treatment
All Systems
All Small Systems
SW< 10,000
SW> 10,000
GW< 10,000
GW> 10,000
Total Number of
Households Served
10,161,304
591 ,623
285,911
9,060,119
305,712
509,562
Mean Annual
Household
Cost Increase
$ 5.53
$ 46.48
$ 43.05
$ 2.83
$ 49.69
$ 5.97
Median Annual
Household
Cost Increase
$ 0.80
$ 18.47
$ 13.79
$ 0.80
$ 16.65
$ 1.37
90th Percentile
Annual
Household Cost
Increase
$ 10.04
$ 168.85
$ 173.53
$ 6.98
$ 109.86
$ 26.82
95th Percentile
Annual
Household Cost
Increase
$ 22.40
$ 197.62
$ 177.93
$ 11.31
$ 197.62
$ 33.84
Percentage of
Household Cost
Increase < $12
92%
38%
47%
96%
31%
79%
Percentage of
Household Cost
Increase < $120
99%
89%
85%
100%
92%
100%
Notes: Detail may not add to total due to independent rounding. Number of households served by systems adding treatment will be higher than households
served by plants adding treatment because an entire system will incur costs even if only some of the plants for that system add treatment (this would result in
Source: Exhibit 7.15
As shown in Exhibit ES.8, the mean, median, and 90th percentile annual household cost increases
for all systems are $0.62, $0.03, and $0.36 per year, respectively. The mean, median, and 90th percentile
household cost increases for those served by plants making treatment technology changes are $5.53,
$0.80, and $10.04, respectively. Households in small systems served by plants making treatment
technology changes will experience the highest household cost increases because they must spread
technology costs over a smaller customer base.
Final Economic Analysis for the Stage 2 DBPR
ES-23
December 2005
-------
EPA analyzed the affordability of the Stage 2 DBPRto determine if variance technologies are
needed for small systems.9 The analysis was performed by comparing total expected household cost (i.e.,
the current annual household cost of water plus the household cost increase) to an "affordability
threshold" equal to 2.5 percent of median household income. The results of EPA's analysis show that, for
each of the three categories of small systems, there are affordable compliance technologies available to
small systems and that variance technologies are not required.10
ES.7 Comparison of Costs and Benefits for Four Regulatory Alternatives
Section ES.2 described the four regulatory alternatives considered in this economic analysis.
Benefits and costs for these four alternatives are summarized in Exhibit ES.9 and ES.10, respectively.
EPA recognizes that the quantified benefits based on reduced cases of bladder cancer could be zero for all
alternatives since causality has not yet been established between exposure to chlorinated water and
bladder cancer.
The regulatory alternative EPA chose, which specifically targets the highest DBP levels and
potential risks in the distribution systems, is also the least-cost alternative by a substantial margin (Exhibit
ES. 10). Estimated costs for Alternative 1 are approximately three times those for the Preferred
Alternative because of the more stringent bromate standard. Quantified benefits based on bladder cancer
cases avoided are nearly the same for Alternative 1 and the Preferred Alternative (Exhibit ES.9) because
the benefits of avoiding potential cancer cases by lowering the bromate standard in Alternative 1 are
unquantified. EPA did not favor this alternative because of a concern that lowering the bromate level to 5
(ig/L could have adverse effects on microbial protection (see Chapter 4 for a full discussion). The range
of quantified benefits increases significantly with Alternatives 2 and 3. However, these alternatives do
not include the risk targeting strategy of the Preferred Alternative. EPA has estimated that a large portion
of the surface water systems covered by the rule would have to switch from their current treatment
practice to more expensive advanced technologies to comply with these alternatives. The associated costs
presented in ES.10 show mean estimated values between $421 and $634 million per year at a 3 percent
discount rate. The M-DBP Advisory Committee did not favor Alternatives 2 and 3 because it believed
that the health effects data are not certain enough to warrant such a drastic shift in the nation's drinking
water treatment practices.
9 Section 1415(e)(l) of SDWA allows States to grant variances to small water systems in lieu of complying
with an MCL if EPA determines that no nationally affordable compliance technologies exist for that system
size/water quality combination. These variances also may be granted only where EPA has identified a variance
technology under Section 1412(b)(15) for the contaminant, system size, and source water quality in question. EPA
is conducting a rigorous review of the methodology for the small system affordability analysis.
10 Only 15 small systems, all of which serve fewer than 500 people, are expected to install treatment
technologies (i.e., integrated membranes with chloramines and granulated activated carbon (GAC) with advanced
disinfectants) that are above the affordability threshold. EPA believes, however, that the number of plants in small
systems predicted to add advanced technologies (including GAC) is overstated for two reasons: 1) Stage 2 DBPR
requirements for small systems are similar to Stage 1 DBPR requirements and may not trigger compliance violations,
as explained in the minimal impacts sensitivity analysis in Chapter 7, and 2) distribution system modifications are
not considered in the compliance forecast. A more detailed discussion is provided in Chapter 8.
Final Economic Analysis for the Stage 2 DBPR ES-24 December 2005
-------
Exhibit ES.9 Comparison of Benefits for All Regulatory Alternatives ($Millions)
/
Average Annual Number
of Cases Avoided
Annualized Mean Benefits
of Cases Avoided
(90% Confidence Bounds)
Discount Rate,
WTP for Non-
Fatal Cases
3%, Lymphoma
7%, Lymphoma
3%, Bronchitis
7%, Bronchitis
Regulatory Alternative
Preferred
279
(103-541)
$1,531
($233-3,536)
$1,246
($190-2,878)
$763
($165-1,692)
$621
($135-1,376)
A1
250
(127-397)
$1,377
($209-3,180)
$1,126
($172-2,600)
$686
($149-1,522)
$561
($122-1,243)
A2
939
(483-1,466)
$5,167
($786-11,936)
$4,227
($644 - 9,758)
$2,575
($558-5,712)
$2,105
($457 - 4,665)
A3
1296
(675-1,988)
$7,130
($1,085-
16,468)
$5,832
($888-13,464)
$3,552
($769 - 7,880)
$2,904
($630 - 6,436)
Notes: Average annual avoided cases is based upon the 25-year period of analysis, so it is lower than the maximum cases to be
avoided following the cessation lag period. Values are discounted and annualized in 2003$. Based on TTHM as an indicator,
Villanueva et al. (2003) for baseline risk, and smoking/lung cancer cessation lag model. Assumes 26 percent of cases are fatal, 74
percent are non-fatal (USEPA 1999a). The cessation lag is explained in detail in Appendix E.
The 90 percent confidence bounds for cases avoided reflect uncertainty in PAR, reduction in average TTHM and HAAS
concentrations, and cessation lag estimates. The 90 percent confidence bounds for benefits reflect uncertainty in monetization
inputs relative to mean cases. EPA recognizes that benefits may be as low as zero since causality has not yet been established
between exposure to chlorinated water and bladder cancer.
Sources: Exhibit 6.28
Exhibit ES.10 Comparison of Costs for All Regulatory Alternatives ($Millions)
Regulatory
Alternative
Preferred
A1
A2
A3
Annualized Total Regulation Costs
Discounted at 3%, 25 Years
Mean
Value
$78.8
$254.1
$421.7
$634.2
90 Percent
Confidence Bound
Lower
(5th %tile)
$56.2
$166.4
$367.8
$536.4
Upper
(95th
%tile)
$102.2
$346.2
$477.8
$736.2
Discounted at 7%, 25 Years
Mean
Value
$76.8
$241.8
$406.4
$613.1
90 Percent
Confidence Bound
Lower
(5th %tile)
$54.6
$158.1
$353.9
$517.7
Upper
(95th
%tile)
$99.4
$330.0
$461.3
$712.8
Note: The 90 percent confidence bounds reflect uncertainty in compliance
forecasts and unit treatment costs.
Sources: Appendix J.
For the Preferred Alternative, see Exhibit J.2as for 3% and
J.2awfor7%.
For Alternative 1, see Exhibit J.3i for 3% and J.3m for 7%.
For Alternative 2, see Exhibit J.4i for 3% and J.4m for 7%.
For Alternative 3, see Exhibit J.5i for 3% and J.5m for 7%.
Final Economic Analysis for the Stage 2 DBPR
ES-25
December 2005
-------
A comparison of alternatives can also be made based upon net benefits: the difference between
the annualized costs and the annualized monetized benefits. Exhibit ES. 11 shows that the Preferred
Alternative has higher net benefits than Alternative I11, but lower net benefits than Alternatives 2 and 3
using either estimate of WTP for non-fatal bladder cancer. These net benefits do not include the
unquantified benefits.
Exhibit ES.11 Comparison of Annualized Mean Net Benefits for All Regulatory
Alternatives ($Millions)
WTP for Non-Fatal
Bladder Cancer
Cases
Lymphoma
Bronchitis
Rule
Alternative
Preferred
A11
A2
A3
Preferred
A11
A2
A3
Annualized Value
3%, 25
Years
$ 1,452
$ 1,122
$ 4,746
$ 6,495
$ 684
$ 432
$ 2,153
$ 2,918
7%, 25
Years
$ 1,170
$ 885
$ 3,821
$ 5,219
$ 544
$ 319
$ 1,698
$ 2,291
Notes: All values are discounted and annualized in 2003$.
Based on TTHM as an indicator, Villanueva et al. (2003) for
baseline risk, and smoking/lung cancer cessation lag model.
Assumes 26 percent of cases are fatal, 74 percent are non-
fatal (USEPA 1999a). EPA recognizes that benefits may be
as low as zero since causality has not yet been established
between exposure to chlorinated water and bladder cancer.
Footnote 1: Alternative 1 appears to have fewer benefits than
the Preferred Alternative because it does not incorporate the
IDSE, as explained in Chapter 4. Furthermore, this EA does
not quantify the benefits of reducing the MCL for bromate
(and potentially associated cancer cases), a requirement that
is included only in Alternative 1.
Source: Exhibit 9.12
Alternative 1 appears to have fewer benefits than the Preferred Alternative because it does not incorporate
the IDSE, as explained in Chapter 4. Furthermore, this EA does not quantify the benefits of reducing the MCL for
bromate (and potentially associated cancer cases), a requirement that is included only in Alternative 1.
Final Economic Analysis for the Stage 2 DBPR
ES-26
December 2005
-------
A further consideration is the relative cost effectiveness of each regulatory alternative of the
Stage 2 DBPR. Cost effectiveness is generally used for determining which alternatives meet a certain
criterion-a threshold for either maximum costs or minimum cases avoided. The rule is cost effective if it
achieves the target level of cases avoided below an acceptable cost; or if it yields an acceptable level of
benefits below the maximum allowable cost. The most cost effective regulatory alternative would
achieve the greatest benefits for a given expenditure, or would impose the least cost for achieving a given
level of benefit. Cost effectiveness analysis (CEA) usually produces a ratio consisting of a cost measure
and an effectiveness measure, i.e., dollars per bladder cancer case avoided. The problem with CEA is
that, as with analysis of benefit cost ratios, it does not take into account the scale of benefits or costs.
Therefore, when considering alternatives such as those in the Stage 2 DBPR, CEA does not provide
enough information for choosing a regulatory alternative, and is only part of a complete analysis as
presented in this EA. In general, EPA recommends that decisions as to whether a specific control strategy
is justified be based on a complete comparison of benefits and costs.
However, the CEA provided in Exhibit ES. 12 does provide useful information. This exhibit
shows the cost per bladder cancer case avoided for each alternative in order of increasing regulatory cost,
using discount rates of 3 and 7 percent. As a threshold for cost effectiveness, consider the WTP measures
for avoidance of non-fatal lymphoma (mean of $4.49 million per case avoided in 2003$) or chronic
bronchitis (mean of $0.80 million per case avoided in 2003$), which serve in this EA as surrogates for a
measure of the amount society is willing to pay to avoid a non-fatal case of bladder cancer.12 If an
alternative is cost effective, it will have an average cost per case avoided that is equal to or less than the
value of the WTP estimate. The Preferred Alternative meets this criterion: at discount rates of 3 and 7
percent, its unit costs of $0.33 and $0.41million, respectively, are less than the WTP values of $4.49 and
$0.80 million, respectively. Furthermore, the Preferred Alternative is more cost effective than the other
alternatives: it has the lowest cost per case avoided. Alternatives 2 and 3 are also below the minimum
threshold of $0.80 million per case avoided. As described previously in this section, Alternative 1 has a
relatively high cost per case avoided because it includes the cost for more stringent bromate reductions
but does not include the benefits of the reduction.
12The WTP values, shown here in 2003$, could be used to develop an annualized WTP value for the most
accurate comparison to the annualized cost per case avoided values presented in Exhibit ES. 12. First, they would be
increased over the period of analysis to reflect the elasticity of WTP in response to increases in real income. Second,
the WTP value would be weighted differentially over the period of analysis to reflect in the annualized value the
difference in the number of cases avoided, which varies on an annual basis. Third, because the cases avoided in the
CEA ratio include both fatal and non-fatal cases, the VSL would be incorporated into a weighted average (to reflect
that 26% of cases are fatal) with the WTP value after it, too, was increased over time to reflect the above 2
considerations. However, each of these factors would increase the threshold used in this analysis; therefore
annualized WTP values are not calculated because the Preferred Alternative, and most of the other alternatives, have
costs per case avoided that are already below the lowest of the thresholds ($0.80 million in 2003$).
Final Economic Analysis for the Stage 2 DBPR ES-27 December 2005
-------
Exhibit ES.12 Cost Per Discounted Case Avoided, by Discount Rate and
Regulatory Alternative ($Millions)
Rule Alternative
Preferred
Alternative 1
Alternative 2
Alternative 3
Cost Per Case Avoided
3%
$
0.33
$ 1.18
$
$
0.52
0.57
7%
$ 0.41
$ 1.42
$ 0.63
$ 0.69
Notes: Values are discounted and annualized in 2003$. Based on
TTHM as an indicator, Villanueva et al. (2003) for baseline risk, and
smoking/lung cancer cessation lag model. Assumes 26 percent of
cases are fatal, 74 percent are non-fatal (USEPA 1999a). EPA
recognizes that benefits may be as low as zero since causality has not
yet been established between exposure to chlorinated water and
bladder cancer.
1) The cost effectiveness ratios are a conservative estimate in that the
regulatory costs in the numerator are not adjusted by subtracting the
medical costs associated with cases avoided to produce a net cost
numerator. Adjustment of the numerator in this CEA would not alter
the relative cost effectiveness of the alternatives or change their
rankings, because it involves the subtraction of a constant. In the case
where thresholds of maximum public expenditure or minimum cases to
be avoided are prescribed, defining the numerator more precisely by
making such adjustments would be appropriate.
Source: Exhibit 9.14
An alternative type of CEA considers the cost per life year saved from avoided premature
mortality. In Appendix N to the Stage 2 DB PR Economic Analysis, a Quality Adjusted Life Years
(QALYs) analysis is developed. Specifically, Exhibit ES.13 presents an approach to QALYs termed
Morbidity Inclusive Life Years (MILYs), which incorporates life years saved from avoided morbidity (a
time equivalent for lost quality of life during illness) with life years saved from avoided mortality.
Exhibit ES. 13 shows that the Preferred Alternative, which is the least stringent but has a targeted strategy
(IDSE) of reducing DBP exposure, has the lowest cost per MILY, i.e., it is more cost effective than the
other alternatives.
The cost per MILY ratios can also be compared to prima facie cost per MILY thresholds, with the
understanding that the thresholds are arbitrary values, often derived by reference to the cost per QALY
(or MILY) for interventions that public health specialists agree are justified. The Harvard Cost Utility
Analysis database presents a median cost-utility ratio of $31,000 per QALY (or MILY) (2002$) for
respiratory and cardiovascular interventions, while Tengs et al. (1995) report a median cost per life-year
saved for life-saving interventions of $48,000 (1993$). The health economics literature often uses either
$50,000 or $100,000 per QALY (or MILY) as a threshold with ratios less than these values considered
prima facie cost effective.
The Preferred Alternative is cost effective compared to the highest of these thresholds ($100,000)
at a discount rate of 3 percent; it is not cost effective when considered at a 7 percent discount rate
($118,000 vs. $100,000 per MILY), although it is more cost effective than the other alternatives.
Final Economic Analysis for the Stage 2 DBPR
ES-28
December 2005
-------
Exhibit ES.13 Cost Effectiveness Analysis Using MILYs Saved from Cases of
Bladder Cancer Avoided, by Rule Alternative, 3 and 7 Percent Discount Rates
Rule
Alternative
Net Cost
(ECOI)
Net Cost
(TCOI)
(Million $)
A
B
MILYs
(Years)
C
Cost per
MILY
(ECOI)
Cost per
MILY
(TCOI)
($)
D = A*106/C
E = B*106/C
3 Percent
Preferred
Alternative 1 1
Alternative 2
Alternative 3
$ 43
$ 222
$ 302
$ 469
$ 45
$ 224
$ 307
$ 476
$ 725
$ 652
$ 2,449
$ 3,379
$ 59,946
$ 340,584
$ 123,259
$ 138,754
$ 62,001
$ 342,641
$ 125,315
$ 140,810
7 Percent
Preferred
Alternative 1 1
Alternative 2
Alternative 3
$ 60
$ 227
$ 350
$ 536
$ 61
$ 227
$ 352
$ 538
$ 510
$ 462
$ 1,733
$ 2,390
$ 118,394
$ 491,295
$ 202,150
$ 224,024
$ 119,540
$ 492,449
$ 203,303
$ 225,178
Abbreviations: MILYs = Morbidity Inclusive Life Years; ECOI = Enhanced Cost of Illness;
TCOI = Traditional Cost of Illness
Notes: All values are discounted and annualized in 2003$. Based on TTHM as an indicator,
Villanueva et al. (2003) for baseline risk, and smoking/lung cancer cessation lag model. Some
numbers may not add correctly due to rounding. EPA recognizes that benefits may be as low as
zero since causality has not yet been established between exposure to chlorinated water and
bladder cancer.
Footnote 1: Alternative 1 appears to have fewer benefits (MILYs) than the Preferred Alternative
because it does not incorporate the IDSE, as explained in Chapter 4. Furthermore, this EA
does not quantify the benefits of reducing the MCL for bromate (and potentially associated
cancer cases), a requirement that is included only in Alternative 1.
Source: Exhibit N.17
ES.8 Conclusions
EPA is finalizing the Stage 2 DBPR to reduce the potential risks that byproducts of chlorination
pose to consumers of drinking water. Disinfection itself is important for protecting the public against
waterborne microbes, and is practiced by over 48,000 PWSs in the United States. The chemicals
commonly used, however, can react with substances in the source water to create potentially harmful
DBFs. These DBFs include TTHM and HAAS, which are potentially associated with increased incidence
of bladder and possibly other cancers. DBFs may also be associated with potential adverse reproductive
and developmental effects such as early-term miscarriage, stillbirth, low birth weight, and some birth
defects. There is uncertainty in the scientific literature regarding the extent to which DBFs contribute to
the incidence of these adverse effects in the exposed population and EPA notes that existing
epidemiological evidence has not conclusively established causality between DBF exposure and any
health risk endpoints. Nevertheless, EPA believes that the weight of evidence warrants concern for these
potential hazards and justifies additional regulatory action beyond the Stage 1 DBPR.
Final Economic Analysis for the Stage 2 DBPR
ES-29
December 2005
-------
1. Introduction
This document presents an analysis of the costs and benefits of the Stage 2 Disinfectants and
Disinfection Byproducts Rule (DBPR). The analysis is performed in compliance with Executive Order
12866, Regulatory Planning and Review (USEPA 1993), which requires that the Environmental
Protection Agency (EPA) estimate the economic impact of rules costing over $100 million annually in an
Economic Analysis (EA) and to submit the analysis in conjunction with publishing the rule.
This chapter provides a summary of the Stage 2 DBPR in Section 1.1. Section 1.2 outlines the
organization of this EA and Section 1.3 provides information regarding supporting calculations and
citations in this EA.
1.1 Summary of the Stage 2 DBPR
The requirements of the Stage 2 DBPR apply to all community water systems (CWSs) and
nontransient noncommunity water systems (NTNCWSs) that add a disinfectant other than ultraviolet light
(UV) or that deliver water that has been treated with a disinfectant other than UV. New since the Stage 1
DBPR, EPA has included requirements specifically for consecutive systems to ensure uniform regulation
of consecutive systems in all States. The Stage 2 DBPR defines consecutive systems as public water
systems that receive finished water from another public water system (a wholesale system).
The Stage 2 DBPR builds on the 1979 Total Trihalomethanes Rule and the 1998 Stage 1 DBPR
by requiring reduced levels of DBFs in distribution systems. Each rule activity for the Preferred
Regulatory Alternative and the associated rule schedule are described below. Note that consecutive
systems of any size must comply with the requirements of the Stage 2 DBPR at the same time as the
largest system in the combined distribution system.
The numerical maximum contaminant levels (MCLs) for the Stage 2 DBPR are the same as for
the Stage 1 DBPR MCLs: 80 micrograms per liter (• g/L) for total trihalomethanes (TTHM), and 60 • g/L
for haloacetic acids (five) (HAAS). The Stage 2 DBPR is designed to reduce high TTHM and HAAS
levels in the distribution system by changing compliance monitoring and calculation requirements. The
compliance determination for the Stage 2 DBPR is based on a locational running annual average (LRAA)
instead of the system-wide running annual average (RAA) used under the Stage 1 DBPR. LRAAs are
essentially RAAs calculated separately for each sample location in the distribution system. With the
Stage 2 LRAA requirement, the TTHM and HAAS MCLs must be met at each monitoring location, while
the Stage 1 RAA requires a system to average results over all monitoring locations. Exhibit 1.1 provides
a comparison of Stage 1 and Stage 2 DBPR compliance calculations.
For many systems, compliance monitoring will be preceded by an initial distribution system
evaluation (IDSE) to identify sample locations for Stage 2 compliance monitoring that represent
distribution system sites with high TTHM and HAAS levels. Systems may perform an IDSE either by
completing a system specific study (SSS) or conducting standard monitoring , unless a system meets the
criteria for a 40/30 certification or a very small systems waiver. To meet the criteria for a 40/30
certification, systems must have low Stage 1 monitoring results (every individual compliance sample is
less than or equal to 40 • g/L and 30 • g/L for TTHM and HAAS, respectively) and no TTHM or HAAS
monitoring violations during a 2-year eligibility period. Systems can qualify for a very small systems
waiver if they serve fewer than 500 people and have TTHM and HAAS data. In addition, NTNCWSs
serving fewer than 10,000 people are not required to conduct an IDSE.
Final Economic Analysis for the Stage 2 DBPR 1-1 December 2005
-------
The Stage 2 DBPR changes the way in which compliance monitoring requirements are
determined. Stage 1 compliance monitoring for TTHM and HAAS is based on a system's population
served, source water type, system type, and number of plants treating water in that system. This "plant-
based" approach is grounded in the assumption that larger systems have more treatment plants and thus
greater system complexity. While this is generally true, the plant-based approach created
disproportionately burdensome monitoring requirements for some systems where the number of plants did
not represent system size, such as larger systems with very large plants or smaller systems with many
disinfecting wells. Moreover, a plant-based approach can complicate monitoring of purchased water
systems, particularly complex ones with multiple connections. For these reasons, EPA has developed a
"population-based" monitoring approach for the Stage 2 DBPR, whereby the monitoring requirements are
based on population served and source water type (not plants per system). Exhibit 1.2 shows the new,
population-based standard monitoring requirements for the IDSE. Exhibit 1.3 presents the new,
population-based Stage 2 DBPR compliance monitoring requirements. EPA believes that the new Stage 2
population-based approach makes monitoring requirements more protective of public health, simpler, and
more equitable for systems of the same size and type.
Systems must develop a Stage 2 DBPR monitoring plan that includes monitoring locations,
monitoring dates, and compliance calculation procedures. States have the option to implement a
procedure for addressing modifications to wholesale system and consecutive system monitoring on a
case-by-case basis.
The Stage 2 DBPR is being promulgated simultaneously with the Long Term 2 Enhanced Surface
Water Treatment Rule (LT2ESWTR) to address complex risk trade-offs between DBFs and microbial
pathogens. The schedule for the Stage 2 DBPR is summarized in Exhibit 1.4. Note that the compliance
deadlines are based on population served. For consecutive and wholesale systems, the compliance
schedule is based on the population served by the largest system in a combined distribution system.
Because Stage 2 DBPR MCL compliance is based on individual DBP measurements at a location
averaged over a four-quarter period, a system could find higher TTHM or HAAS levels than the MCL
values, while at the same time maintaining compliance with the Stage 2 DBPR. This is because the high
concentration could be averaged with lower concentrations at a given location. For this reason, the Stage
2 DBPR includes a provision for "operational evaluations" as follows:
• A system has exceeded an operational evaluation level at any monitoring location when the
sum of the two previous quarters' compliance monitoring results plus twice the current
quarters result, divided by 4, exceeds 80 • g/L for TTHM or 60 • g/L for HAAS.
If an operational evaluation level is exceeded, the system must conduct an "operational evaluation" and
submit a written report of the evaluation to the State/Primacy Agency no later than 90 days after being
notified of the analytical results that caused the excursion.
Final Economic Analysis for the Stage 2 DBPR 1-2 December 2005
-------
Exhibit 1.1 Comparison of Stage 1 and Stage 2 DBPR Compliance Calculations
Stage 1 DBPR
First Quarter
Distribution System
Sampling Location
Second Quarter
Third Quarter
Fourth Quarter
Average of All Samples
Average of All Samples
Average of All Samples
Average of All Samples
V
Running Annual Average (RAA) of Quarterly Averages
MUST BE AT OR BELOW COMPLIANCE LEVELS
Stage 2B DBPR (Note that some sampling locations may change as a result of the IDSE.)
First Quarter
Second Quarter
Third Quarter
Fourth Quarter
First Quarter
Second Quarter
Third Quarter
Fourth Quarter
Locational Running Annual
Average
MUST BE AT OR BELOW
COMPLIANCE LEVELS
First Quarter iS
Second Quarter €
Third Quarter in
Fourth Quarter Jl
Locational Running Annual
Average
" MUST BE AT OR BELOW
COMPLIANCE LEVELS
First Quarter A
Second Quarter A
Third Quarter A
Fourth Quarter A
Locational Running Annual
Average
MUST BE AT OR BELOW
COMPLIANCE LEVELS
First Quarter
Second Quarter
Third Quarter
Fourth Quarter
Locational Running Annual
Average
"MUST BE AT OR BELOW
COMPLIANCE LEVELS
Final Economic Analysis for the Stage 2 DBPR
1-3
December 2005
-------
Exhibit 1.2 IDSE Standard Monitoring Requirements
System Size
(Population Served)
Number of Distribution System Sites1 (by
location type) per System
Near
Entry
Point 2
Average
Residence
Time
High
TTHM
High
HAAS
Total
Number of
Sites per
System
Number of
Monitoring
Periods and
Frequency for the
1 -year IDSE
period
Systems Using Surface Water in Whole or in Part3
< 500 consecutive
systems
< 500 non-consecutive
systems
500 - 3,300
consecutive systems
500 - 3,300 non-
consecutive systems
3,301 -9,999
10,000-49,999
50,000 - 249,999
250,000 - 999,999
1 million - 4,999,999
> 5,000,000
1
-
1
-
-
1
3
4
6
8
-
-
-
-
1
2
4
6
8
10
1
1
1
1
2
3
5
8
10
12
-
1
-
1
1
2
4
6
8
10
2
2
2
2
4
8
16
24
32
40
1 (during peak
historical month)4
1 (during peak
historical month)4
4 (every 90 days)
4 (every 90 days)
4 (every 90 days)
6 (every 60 days)
6 (every 60 days)
6 (every 60 days)
6 (every 60 days)
6 (every 60 days)
Systems Using Only Ground Water
< 500 consecutive
systems
< 500 non-consecutive
systems
500 - 9,999
10,000-99,999
100,000-499,999
> 500,000
1
-
-
1
1
2
-
-
-
1
1
2
1
1
1
2
3
4
-
1
1
2
3
4
2
2
2
6
8
12
1 (during peak
historical month)4
1 (during peak
historical month)4
4 (every 90 days)
4 (every 90 days)
4 (every 90 days)
4 (every 90 days)
1 Samples must be taken at locations other than existing Stage 1 monitoring locations. Dual sample sets (Le., a
TTHM and an HAAS sample) must be taken at each monitoring location during each monitoring period. Sampling
location must be distributed throughout the distribution system.
2 If the number of entry points to the distribution system is fewer than the specified number of sampling locations,
additional samples must be taken equally at high TTHM and HAAS locations. If there is an odd extra location
number, a sample at a high TTHM location must be taken. If the number of entry points to the distribution system
is more than the specified number of sample locations, samples must be taken at entry points to the distribution
system having the highest water flows.
3 For the purposes of this EA, "surface water" systems are equivalent to "subpart H" systems and include systems
that use ground water under the direct influence of surface water (GWUDI).
4 The peak historical month is the month with the highest TTHM or HAAS levels or the warmest water temperature.
Final Economic Analysis for the Stage 2 DBPR
1-4
December 2005
-------
Exhibit 1.3 Stage 2 Population-Based Monitoring Requirements
System Size
(Population Served )
Distribution System Sample
Locations1
Highest
TTHM
Locations
Highest
HAA5
Locations
Existing
Stage 1
Compliance
Locations2
Total
Sample
Locations per
System5
Monitoring
Frequency3
Systems Using Surface Water in Whole or in Part4
<500
500-3,300
3,301-9,999
10,000-49,999
50,000-249,999
250,000-999,999
1 Mil-4,999,999
> 5,000, 000
1
1
1
2
3
5
6
8
1
1
1
1
3
4
6
7
1
2
3
4
5
2
2
2
4
8
12
16
20
per year
per quarter
per quarter
per quarter
per quarter
per quarter
per quarter
per quarter
Systems Using Only Ground Water
<500
500 - 9,999
10,000-99,999
100,000-499,999
> 500,000
1
1
2
3
3
1
1
1
2
3
1
1
2
2
2
4
6
8
per year
per year
per quarter
per quarter
per quarter
36 based on the system's recomrm
state/Pr
ndatipns for Stage 2 DBPR cor
tate/Pr " ~
iphance monr
line rent or a
ring locations in
:ionaf locations.
Locations must .._,_,_.._ ._. _,__, .
its report to the State/Primacy Agency, unless the State/Primacy Agency requires i
Locations should be distributed throughout the distribution system to the extent possible.
Alternate between highest HAAS LRAA and highest TTHM LRAA locations among the existing Stage 1 average
resident time compliance locations. If the number of existing Stage 1 compliance locations is fewer than the
specified number for Stage 2, alternate between highest HAAS LRAA locations and highest TTHM LRAA locations
from the IDSE.
All systems must monitor during the month of highest DBP concentrations. Systems on quarterly monitoring must
take dual sample sets approximately every 90 days.
For the purposes of this EA, "surface water" systems are equivalent to "subpart H" systems and include systems
that use GWUDI.
Systems on quarterly monitoring must take dual sample sets every 90 days at each monitoring location, except for
subpart H systems serving 500-3,300 people. Systems on annual monitoring and subpart H systems serving 500-
3,300 people are required to take individual TTHM and HAAS samples (instead of a dual sample set) at the
locations with the highest TTHM and HAAS concentrations, respectively. Only one location with a dual sample set
per monitoring period is needed if the highest TTHM and HAAS concentrations occur at a same location and month
(if monitored annually).
Final Economic Analysis for the Stage 2 DBPR
1-5
December 2005
-------
Exhibit 1.4 Stage 2 DBPR Implementation Schedule
Schedule 1
Systems serving
> 100,000 1
Schedule 2
Systems serving
50,000 to 99,9991
Schedule 3
Systems serving
10,000 to 49,9991
Schedule 4
Systems serving
< 10,000 1
2006
2007
| LT2 Crypto
t
IDSE Plar
October 1
IDJ
Ap
2006
i Due
,2006
| LT2
2008
monitoring
IDSE mon.
IDSE
Janua
2009
Rep
ry 1
Crypto monitoring
!
3E Plan Due
ril 1,2007
I
IDSE Plar
October 1
IDJ
Ap
2007
IDSE
mon.
LT2
i Due
,2007
!
3E Plan D
ril 1,2008
IDJ
Jul
Crypto
IDSE
2010
2011
2012 | 2013
1
| 2014
Treatment Installation | Possible Extension 2
* * * i
ort Du
,2009
I
SERep
y 1, 20
Ł
Be
Ap
n T
gin Compliance
111,2012 I
u
1
_l
Treatment Installation , Possible Extension 2 •
ort
09
Due
monitoring
mon.
Ł. Co// mon.
je
2008
IDSE
IDJ
Jar
I
Begin Compl
October 1, 2C
ance
)12
Treatment Installation
•/ •*• • > ' ',
I n
3E Report Du
luary 1, 2010
Crypto
mon. 4
2009
1
mon.]
2015
Possible Extension 2 '
L _l
I I
Begin Compliance
October 1,201 3
Treatment Installation
/ / f
IDSE Report
July 1,2010
2010
Due
2011
2012
Begin
Octob
Possible Extension 2 I
L^ _l
n
Co
er1
2013
f
npliance
,2013
2014
2015
11ncludes all systems that are part of a combined distribution system that have a largest system with this population.
2 A State may grant up to an additional 2 years for systems to comply if the State determines that additional time is necessary for capital improvements.
3 Subpart H systems that must conduct Cryptosporidium monitoring have an additional 12 months to comply with the Stage 2 DBPR MCLs.
Final Economic Analysis for the Stage 2 DBPR
1-6
December 2005
-------
1.2 Document Organization
The rest of this EA is organized into the following chapters:
• Chapter 2 identifies public health concerns addressed by the rule and provides a 20-year
regulatory history that includes a description of relevant National Primary Drinking Water
Regulations (NPDWRs). It also explains the statutory authority for promulgating the Stage 2
DBPR and the economic rationale for choosing a regulatory approach.
Chapter 3 characterizes conditions that exist (including system inventory, treatment, and
water quality data) before systems make changes to meet the Stage 2 DBPR requirements.
• Chapter 4 reviews alternative regulatory approaches that EPA considered during the
development of the rule and presents the rationale for selecting the Preferred Regulatory
Alternative.
• Chapter 5 summarizes the methodology used to develop the compliance forecasts and
predictions of reductions in DBFs. It also contains compliance forecast results for the Stage 1
DBPR and the Stage 2 DBPR Preferred Regulatory Alternative.
• Chapter 6 reviews available epidemiological and toxicological data related to DBFs. The
public health and economic benefits of this rule, as well as several sensitivity analyses, are
provided in this chapter.
• Chapter 7 presents an estimate of the costs of implementing the rule to industry, households,
and States/Primacy Agencies. It also compares the costs of the four regulatory alternatives.
Chapter 8 discusses analyses performed to evaluate the effects of the rule on different
segments of the population and considers various executive orders and requirements,
including the Regulatory Flexibility Act (RFA) and Unfunded Mandates Reform Act
(UMRA).
• Chapter 9 compares the rule's benefits and costs to evaluate whether projected benefits
exceed costs. The results for the Preferred Regulatory Alternative are discussed and
compared to the regulatory alternatives considered.
Final Economic Analysis for the Stage 2 DBPR 1-7 December 2005
-------
1.3 Calculations and Citations
This EA presents results from detailed and complex analyses. To help the reader track the
various calculations and analyses, the following are provided:
A detailed reference section (see Chapter 10).
A row on most tabular exhibits throughout the document that gives the formulas used to
compute the contents of each column.
• Sources for elements of exhibits throughout the document that are not calculated in the
exhibits themselves.
Supporting electronic files (Stage 2 DBPR cost model; Stage 2 DBPR benefits model; Stage
2 DBPR Surface Water Analytical Tool (SWAT) supporting files, and the ICR Matrix
Method supporting files).
Final Economic Analysis for the Stage 2 DBPR 1-8 December 2005
-------
2. Need for the Rule
2.1 Introduction
This chapter first identifies the issue to be addressed by the Stage 2 Disinfectants and
Disinfection Byproducts Rule (DBPR) (section 2.1.1) and then summarizes in section 2.2 the public
health concerns addressed by the rule. Section 2.3 provides the regulatory history leading up to the Stage
2 DBPR, and section 2.4 addresses the economic rationale for choosing this regulatory approach.
2.1.1 Description of the Issue
Over 48,000 public water systems (PWSs), serving more than 260 million people in the United
States, chemically disinfect their water to kill or inactivate microbial contaminants (USEPA 200 Ic). This
is an essential public health measure. Chemical disinfection, however, may pose health risks of its own.
Disinfection byproducts (DBFs) result from reactions between chemical disinfectants and naturally
occurring compounds in source waters. Research has shown that chlorinated waters and DBFs may be
associated with increased risk of bladder and other cancers. While there are uncertainties in the
quantitative relationship between the incidence of these cancers and the occurrence of DBFs in drinking
water, EPA believes that additional reductions in these DBF levels in drinking water will reduce the
incidence of bladder cancer and, possibly, other cancers.
In addition, results from toxicology and, particularly, epidemiology studies published in the last
several years suggest a potential increased risk for pregnant women and their fetuses who are exposed to
DBFs in drinking water. The studies have shown that early-term miscarriage, stillbirth, low birth weight,
and some birth defects may be associated with drinking water containing DBFs. (These studies are
discussed in detail in Chapter 6.) Uncertainties pertain to the extent to which reproductive and
developmental effects may be associated with DBF exposure, which DBFs may be of greatest concern,
what levels of DBFs may pose a risk, and at what period of development fetuses may be at the greatest
risk. Although the levels of DBFs potentially associated with specific adverse reproductive and
developmental effects are not known and no causal link has been established, EPA believes the evidence
supports concern for these potential hazards and warrants regulatory action.
In a separate but concurrent action, EPA is promulgating the Long Term 2 Enhanced Surface
Water Treatment Rule (LT2ESWTR) to improve control of microbial contaminants, particularly
Cryptosporidium, in surface water and to ensure that microbial protection is not compromised by efforts
to reduce exposure to DBFs. Together, the Stage 2 DBPR and LT2ESWTR represent the final stage of a
two-stage strategy that was developed in a regulatory negotiation effort in 1992 and 1993.1 They reflect
recommendations presented by the Stage 2 Microbial and Disinfection Byproducts (M-DBP) Federal
Advisory Committee Agreement in Principle, signed in September 2000 (USEPA 2000n).
lrThe key outcomes of that regulatory negotiation effort were recommendations to proceed with rules
addressing DBFs and microbial pathogens in two stages and to collect relevant information from public water
supplies for use in the development of these rules and the analysis of their impacts. This two-stage approach was
subsequently incorporated into the 1996 Safe Drinking Water Act (SDWA) Amendments. The first stage of the
M-DBP rulemaking process culminated with the joint promulgation of the Stage 1 DBPR and the Interim Enhanced
Surface Water Treatment Rule (IESWTR) by EPA in December 1998.
Final Economic Analysis for the Stage 2 DBPR 2-1 December 2005
-------
2.2 Public Health Concerns to Be Addressed
EPA's primary mission is to protect human health and the environment. In carrying out this
mission, EPA must often make regulatory decisions based on incomplete or uncertain information. The
Agency believes it is appropriate and prudent to take action to protect public health when evidence
indicates that exposure to a contaminant could present significant risks to the public, rather than take no
action until risks are unequivocally proven. Additionally, the 1996 Amendments to the Safe Drinking
Water Act (SDWA) require EPA to address DBP and microbial risks by certain statutory deadlines.
An important consideration in assessing public health risks is the number of people who may be
exposed to a particular contaminant. More than 260 million people in the United States potentially are
exposed to DBFs via drinking water because they are served by public water systems (PWSs) that add
chemical disinfectants (see Exhibit 3.3 for the Stage 2 population baseline). While effective in
controlling many harmful microorganisms, chemical disinfectants also form DBFs, some of which may
pose health risks. Because of the large number of people potentially exposed to DBFs, EPA is concerned
about any health risks that may be associated with DBFs. Information on these risks can come from two
types of studies-epidemiological and toxicological.
Epidemiological studies have investigated the relationship between exposure to chlorinated
drinking water and cancer. These studies suggest an association between bladder, rectal, and colon
cancers and exposure to chlorinated drinking water. Numerous toxicology studies have shown several
DBFs (such as bromodichloromethane, bromoform, dichloroacetic acid, and bromate) to be carcinogenic
in laboratory animals (see Chapter 6 for details).
Other epidemiological studies indicate a potential link between DBP exposure and adverse
reproductive and developmental health effects, particularly early-term miscarriage (see Chapter 6 for
discussion). In addition, toxicological studies have shown that several DBFs cause adverse reproductive
and developmental health effects in laboratory animals. EPA believes that these studies together provide
evidence that DBFs may present potential public health risks to pregnant women and their fetuses.
Research, therefore, supports EPA's conclusion that chlorinated drinking water could potentially
be a source of health risk to the general public. There is uncertainty in the scientific literature regarding
the extent to which DBFs contribute to the incidence of these adverse effects in the exposed population.
Nevertheless, EPA believes that the weight of evidence warrants concern for these potential hazards and
justifies additional regulatory action beyond the Stage 1 DBPR.
2.3 Regulatory History
2.3.1 Statutory Authority for Promulgating the Rule
The primary responsibility for regulating the quality of drinking water lies with EPA. The
SDWA establishes this responsibility and defines the mechanisms at the Agency's disposal to protect
public health. EPA sets standards by identifying which contaminants should be regulated and by
establishing the maximum levels of the contaminants allowed in drinking water.
Section 1412(b)(l) of the 1996 SDWA reauthorization mandated new drinking water
requirements. EPA's general authority to set Maximum Contaminant Level Goals (MCLGs) and develop
Final Economic Analysis for the Stage 2 DBPR 2-2 December 2005
-------
the National Primary Drinking Water Regulations (NPDWRs) was modified to apply to contaminants that
"may have an adverse effect on the health of persons," are "known to occur or there is a substantial
likelihood that the contaminant will occur in public water systems with a frequency and at levels of public
health concern," and for which, "in the sole judgment of the Administrator, regulation of such
contaminant presents a meaningful opportunity for health risk reductions for persons served by public
water systems" (SDWA 1412(b)(l)(A)).
To regulate a contaminant, EPA sets an MCLG at a level at which no known or anticipated
adverse health effects occur. MCLGs are established solely on the basis of protecting public health and
are not enforceable. EPA simultaneously sets an enforceable Maximum Contaminant Level (MCL) as
close as technologically feasible to the MCLG, while taking costs into consideration. If it is not feasible
to measure the contaminant at levels presumed to have impacts on health, a treatment technique can be
specified in place of an MCL. For water systems, compliance with a drinking water regulation means
either not exceeding the MCL or meeting treatment technology requirements.
Additionally, EPA identifies maximum concentrations of residual disinfectants that can occur in
water without harming human health and sets maximum residual disinfectant level goals (MRDLGs) and
maximum residual disinfectant levels (MRDLs). PWSs maintain residual levels of disinfectants in the
distribution system, following treatment, to ensure consumer protection from microbial contaminants.
Like MCLGs, MRDLGs are not enforceable, while MRDLs are.
In addition to the general authorities cited above, SDWA 1412(b)(2)(C) requires specifically that
EPA promulgate the Stage 2 DBPR.
The Administrator shall promulgate an Interim Enhanced Surface Water Treatment Rule,
a Final Enhanced Surface Water Treatment Rule, a Stage 1 Disinfectants and Disinfection
Byproducts Rule, and a Stage 2 Disinfectants and Disinfection Byproducts Rule in
accordance with the schedule published in Volume 29, Federal Register, Page 6361
(February 10, 1994), in Table III. 13 of the proposed Information Collection Rule.
(SDWA 1412(b)(2)(C))
The following sections summarize the development of relevant NPDWRs over the past 20 years.
2.3.2 1979 Total Trihalomethane Rule
Under the Total Trihalomethane Rule (44 Federal Register (FR) 68624, November 29, 1979),
EPA set an MCL for TTHM (the sum of the concentrations of chloroform, bromoform, bromodichloro-
methane, and dibromochloromethane) of 0.10 milligrams per liter (mg/L) as a running annual average
(RAA) of quarterly measurements. This standard applied to CWSs using surface or ground water that
served at least 10,000 people and that added a disinfectant to the drinking water during any part of the
treatment process. This 1979 rule was superseded by the 1998 Stage 1 DBPR (section 2.3.7) with which
all CWSs and NTNCWSs must have complied by January 2004.
Final Economic Analysis for the Stage 2 DBPR 2-3 December 2005
-------
2.3.3 1989 Total Coliform Rule
The Total Coliform Rule (TCR) (54 FR 27544, June 29, 1989) applies to all PWSs. Because
monitoring PWSs for every possible pathogenic organism is not feasible, coliform organisms are used as
indicators of possible contamination. Coliforms are easily detected in water and are used to indicate a
system's vulnerability to pathogens. In the TCR, EPA set an MCLG of zero for total coliforms. EPA
also set a monthly MCL for total coliforms and required testing of total-coliform-positive cultures for the
presence of E. coli or fecal coliforms. E.coli and fecal coliforms indicate more immediate health risks
from sewage or fecal contamination and are used as the indicator of an acute MCL violation. Coliform
monitoring frequency is determined by population served, the type of system (community or
noncommunity) and the type of source water (surface water or ground water). In addition, the TCR
required sanitary surveys every 5 years (or 10 years for noncommunity systems using disinfected ground
water) for systems that collect fewer than 5 routine total coliform samples per month (typically systems
serving fewer than 4,100 people).
2.3.4 1989 Surface Water Treatment Rule
Under the Surface Water Treatment Rule (SWTR) (54 FR 27486, June 29, 1989), EPA set
MCLGs of zero for Giardia lamblia, viruses, andLegionella and established requirements for all PWSs
using surface water or GWUDI as a source. The SWTR includes treatment technique requirements for
filtered and unfiltered systems that are intended to protect against the adverse health effects associated
with Giardia lamblia, viruses, and Legionella, as well as many other pathogenic organisms. These
requirements include:
Maintenance of a disinfectant residual in water entering and within the distribution system.
Removal or inactivation of at least 99.9 percent (3 logs) of Giardia and 99.99 percent (4 logs)
of viruses.
• For filtered systems, meeting a turbidity performance standard for the combined filter effluent
of 5 nephelometric turbidity units (NTUs) as a maximum and 0.5 NTU in 95 percent of
monthly measurements, based on 4-hour monitoring for treatment plants using conventional
treatment or direct filtration (with separate standards for other filtration technologies). These
requirements were enhanced by the 1998 Interim Enhanced Surface Water Treatment Rule
(IESWTR) and the 2002 Long Term 1 Enhanced Surface Water Treatment Rule
(LT1ESWTR).
• Watershed control programs and other requirements for unfiltered systems.
2.3.5 1996 Information Collection Rule
The Information Collection Rule (ICR) (61 FR 24354, May 14, 1996) applied to PWSs serving
more than 100,000 people. A more limited set of ICR requirements covered ground water systems
serving 50,000 to 100,000 people.
The ICR authorized EPA to collect occurrence and treatment information from water treatment
plants to help evaluate the possible need for changes to microbial requirements and microbial treatment
practices and to help evaluate the need for future regulation of disinfectants and DBFs. The ICR provided
Final Economic Analysis for the Stage 2 DBPR 2-4 December 2005
-------
EPA with information on the national occurrence of (1) chemical byproducts that form when disinfectants
used for microbial control react with naturally occurring compounds and ions present in source water; and
(2) disease-causing microorganisms including Cryptosporidium, Giardia, viruses, and coliform bacteria.
The ICR also mandated the collection of data on how water systems currently treat for contaminants. The
ICR monthly sampling data provided 18 months of information on the quality of the influent and treated
water, including pH, alkalinity, turbidity, temperature, calcium, total hardness, total organic carbon,
ultraviolet254 (UV) absorbency, bromide, ammonia, and disinfectant residual. These data provide some
indication of the "treatability" of the water, the occurrence of contaminants, and the potential for DBF
formation. The data collected under the ICR have been analyzed to help develop the LT2ESWTR and
Stage 2 DBPR.
2.3.6 1998 Interim Enhanced Surface Water Treatment Rule
The IESWTR (63 FR 69478, December 16, 1998) enhances the 1989 SWTR. It applies to PWSs
serving at least 10,000 people and using surface water or GWUDI as a source. These systems began
compliance with the IESWTR in January 2002. The purpose of the IESWTR is to improve control of the
protozoan Cryptosporidium and to address tradeoffs between the risks of microbial pathogens and those
of DBFs. The requirements and guidelines include:
• An MCLG of zero for Cryptosporidium.
• Removal of 99 percent (2 logs) of Cryptosporidium for systems that use filters.
• For filtered systems, a turbidity performance standard for the combined filter effluent of 1
NTU as a maximum and 0.3 NTU as a minimum in 95 percent of monthly measurements,
based on 4-hour monitoring for treatment plants using conventional treatment or direct
filtration.
Continuous monitoring of individual filter effluent in conventional and direct filtration plants
and recording turbidity readings every 15 minutes when these filters are on-line.
• A disinfection benchmark to assess the level of microbial protection provided before facilities
change their disinfection practices to meet the requirements of the Stage 1 DBPR.
• Inclusion of Cryptosporidium in the definition of GWUDI and in the watershed control
requirements for unfiltered PWSs.
• Covers for all new finished water storage facilities.
• A primacy provision that requires States to conduct sanitary surveys for all surface water
systems, including those serving fewer than 10,000 people.
The IESWTR was promulgated concurrently with the Stage 1 DBPR, described in the next setion,
so that systems could coordinate their response to the risks posed by DBFs and microbial pathogens.
Final Economic Analysis for the Stage 2 DBPR 2-5 December 2005
-------
2.3.7 1998 Stage 1 Disinfectants and Disinfection Byproducts Rule
The Stage 1 DBPR (63 FR 69390, December 16, 1998) applies to all CWSs and NTNCWSs that
add a chemical disinfectant to their water. Certain requirements designed to provide protection against
acute health effects from chlorine dioxide also apply to transient noncommunity water systems
(TNCWSs). Compliance for surface water and GWUDI systems serving at least 10,000 people began in
January 2002. Surface water and GWUDI systems serving fewer than 10,000 people and all ground
water systems were required to comply by January 2004.
The Stage 1 DBPR sets MRDLGs for chlorine (4 mg/L as chlorine (C12)), chloramines (4.0 mg/L
as C12), and chlorine dioxide (0.8 mg/L as C1O2); and MCLGs for bromodichloromethane (0 mg/L),
bromoform (0 mg/L), dibromochloromethane (0.06 mg/L), dichloroacetic acid (0 mg/L), trichloroacetic
acid (0.3 mg/L), bromate (0 mg/L), and chlorite (0.8 mg/L). The rule sets MRDLs for chlorine (4.0 mg/L
as C12), chloramines (4.0 mg/L as C12), and chlorine dioxide (0.8 mg/L as C1O2); and MCLs for TTHM
(0.080 mg/L), HAAS (0.060 mg/L), bromate (0.010 mg/L), and chlorite (1.0 mg/L). The MRDLs and
MCLs, except those for chlorite and chlorine dioxide, are calculated as RAAs. For conventional surface
water and GWUDI systems, a treatment technique—enhanced coagulation/softening—is specified for the
removal of DBF precursors.
As noted in section 2.3.6, the Stage 1 DBPR was promulgated concurrently with the lESWTRto
coordinate the control of DBFs and microbial contaminants.
2.3.8 2000 Proposed Ground Water Rule
The proposed Ground Water Rule (65 FR 30194, May 10, 2000) addresses fecal contamination in
ground water systems. It also builds on the TCR through provisions based on further evaluation of E. coll
monitoring results measured under the TCR. Key components of the approach for protection of ground
water included in the proposed rule are:
Sanitary surveys for all ground water systems.
Hydrogeologic sensitivity assessments to identify ground water wells that are susceptible to
fecal contamination.
• Triggered source water monitoring (based on TC monitoring) for an indicator of fecal
contamination for all systems that do not achieve 4-log treatment, and in addition, routine
source water monitoring for an indicator of fecal contamination for systems that have been
determined to draw from sensitive ground water sources.
Correction of significant deficiencies and fecal contamination by eliminating the source of
contamination, correcting the deficiency, providing an alternative source of water, or
providing inactivation and/or removal of 99.99 percent (4 logs) of viruses.
Compliance monitoring to ensure that disinfection treatment is reliably operated when it is
used.
Final Economic Analysis for the Stage 2 DBPR 2-6 December 2005
-------
2.3.9 2001 Arsenic Rule
The Arsenic Rule (66 FR 6976, January 22, 2001) increases the level of public health protection
against exposure to arsenic in drinking water. The rule revises the MCL for arsenic in drinking water
from 0.05 mg/L to 0.010 mg/L and sets an MCLG of 0 mg/L for all CWSs and NTNCWSs. Clarification
on how compliance is demonstrated for many inorganic and organic contaminants in drinking water is
also given. All existing CWSs and NTNCWSs must comply with the Arsenic Rule by January 23, 2006.
2.3.10 2001 Filter Backwash Recycling Rule
The Filter Backwash Recycling Rule (FBRR) (66 FR 31086, June 8, 2001) regulates systems that
return filter backwash to the treatment process. The rule applies to surface water and GWUDI systems
that use direct or conventional filtration and recycle spent filter backwash water, sludge thickener
supernatant, or liquids from dewatering processes. The rule requires that these recycled liquids be
returned to a location such that all steps of a system's conventional or direct filtration are employed. The
rule also requires systems to notify the State that they practice recycling. Finally, systems must collect
and maintain information for review by the State.
2.3.11 2002 Long Term 1 Enhanced Surface Water Treatment Rule
The LT1ESWTR (67 FR 1812, January 14, 2002) enhances the 1989 SWTR requirements for
small systems. LT1ESWTR enhances control of Cryptosporidium and other disease-causing microbes for
surface water and GWUDI systems that serve fewer than 10,000 people. Key provisions in the
LT1ESWTR are very similar to those for the IESWTR, but provide additional flexibility for small
systems.
2.3.12 2005 Long Term 2 Enhanced Surface Water Treatment Rule
Made final in concert with the Stage 2 DBPR, the LT2ESWTR strengthens control of
Cryptosporidium, and applies to all PWSs that use surface water or GWUDI as a source. It incorporates
system-specific treatment requirements based on a "Microbial Framework" approach that targets high-risk
systems. This approach involves assigning systems to different categories (or "bins") based on the levels
of Cryptosporidium found in the source water. Additional treatment requirements, if any, are linked to
the level of Cryptosporidium. A system will choose technologies and management practices from a
"toolbox" of options appropriate to its bin.
Medium and large systems (those serving at least 10,000 people) that filter are required to
conduct Cryptosporidium source water monitoring for 24 months to determine their bin classification.
Small systems (those serving fewer than 10,000 people) that filter will monitor E. coli bacteria in their
source water biweekly for 12 months. Based on their E. coli results, they may be required to monitor
Cryptosporidium as well.
In addition to requirements for filtered systems, the LT2ESWTR requires unfiltered systems to
continue to meet the filtration avoidance criteria under the 1989 SWTR and provide inactivation at 4 logs
(99.99 percent) for virus, 3 logs (99.9 percent) for Giardia, and 2 to 3 logs (99 to 99.9 percent) for
Cryptosporidium (depending on results of Cryptosporidium monitoring of source water). Building on the
Final Economic Analysis for the Stage 2 DBPR 2-7 December 2005
-------
SWTR requirements, inactivation requirements for unfiltered systems subject to the LT2ESWTR must be
met using a minimum of two disinfectants.
Also, the LT2ESWTR requires systems with uncovered finished water reservoirs to cover the
reservoirs or treat reservoir discharge to the distribution system to achieve 4-log virus inactivation, 3-log
Giardia inactivation, and 2-log Cryptosporidium inactivation.
2.4 Economic Rationale
This section addresses the economic rationale for choosing a regulatory approach. Such a
rationale is required by Executive Order Number 12866, Regulatory Planning and Review (USEPA
1993), which states:
[EJach agency shall identify the problem that it intends to address (including, where
applicable, the failures of the private markets or public institutions that warrant new
agency action) as well as assess the significance of that problem. (Section 1, b(l))
In addition, Office of Management and Budget (OMB) guidance dated January 11, 1996, states
that "in order to establish the need for the proposed action, the analysis should discuss whether the
problem constitutes a significant market failure" (USEPA 1996b).
In a perfectly competitive market, prices and quantities are determined solely by the aggregated
decisions of buyers and sellers. Such a market occurs when many producers of a product are selling to
many buyers, and both producers and consumers have perfect information on the characteristics and
prices of each firm's products. Barriers to entry in the industry cannot exist, and individual buyers and
sellers must be "price takers"; i.e., their individual decisions cannot affect the price. Several properties of
the public water supply do not satisfy the conditions for a perfectly competitive market and, thus, lead to
market failures that require regulation.
First, many water systems are natural monopolies. A natural monopoly exists when it is
impossible for more than one firm in each area to recover the costs of production and survive. There are
high fixed costs associated with reservoirs and wells, transmission and distribution systems, treatment
plants, and other facilities. For other potential suppliers to enter the market, they would have to provide
the same extensive infrastructure to realize similar economies of scale and be competitive. A splitting of
the market with increased fixed costs (for example, two supplier networks in a single market) usually
makes this situation unprofitable for one or both suppliers. The result is a market suitable for a single
supplier and one that is hostile to alternative suppliers. In such natural monopolies, suppliers have fewer
incentives for providing high-quality services or maintaining competitive prices. In these situations,
governments often intervene to help protect the public interest.
Because PWSs are legal, as well as natural, monopolies, they often are subject to price controls, if
not outright public ownership. While customers may demand improvements in water quality, the
regulatory regime may not transmit that demand to the water supplier or allow the supplier to raise its
price to recover the cost of the improvements. If consumers do not believe that their drinking water is
safe enough, they cannot simply switch to another water utility. Other options for obtaining safe drinking
water (e.g., buying bottled water or installing point-of-use filtration) most often cost consumers more than
the purchase from public water supplies. Therefore, the water supplier may have little incentive to
improve water quality.
Final Economic Analysis for the Stage 2 DBPR 2-8 December 2005
-------
Second, the public may not understand the health and safety issues associated with drinking water
quality. Understanding the health risks potentially posed by trace quantities of drinking water
contaminants involves analysis and synthesis of complex toxicological and health sciences data.
Therefore, the public may not be aware of the risks it faces. EPA has implemented a Consumer
Confidence Report Rule (CCR) (63 FR 44512 August 1998) that makes water quality information more
easily available to consumers. This rule requires CWSs to publish an annual report on local drinking
water quality. Consumers, however, still have to analyze this information for its health risk implications.
Even if informed consumers are able to engage water systems in a dialogue about health issues, the
transaction costs of such interaction (measured in personal time and monetary outlays) present another
significant impediment to consumer expression of risk reduction preferences.
SDWA regulations are intended to provide a level of protection from exposure to drinking water
contaminants. The regulations set minimum performance requirements to protect consumers from
exposure to contaminants. They are not intended to restructure market mechanisms or to establish
competition in supply; rather, they establish the level of service to be provided that best represents public
preference for safety. The federal regulations reduce the high information and transaction costs by acting
on behalf of consumers in balancing risk reduction and the social costs of achieving this risk reduction.
Final Economic Analysis for the Stage 2 DBPR 2-9 December 2005
-------
3. Baseline Conditions
3.1 Introduction
To quantify the effects of the Stage 2 Disinfectants and Disinfection Byproducts Rule (DBPR), it
is necessary to have a baseline against which to compare the set of regulatory alternatives. The baseline
is a characterization of the industry and its operations under the conditions expected to exist before
systems make changes to meet requirements of the Stage 2 DBPR. The baseline allows a consistent
comparison of public health impacts (developed in Chapter 6) and the economic and financial impacts
(developed in Chapters 7 and 8) of each regulatory alternative.
Development of the baseline consists of the following processes:
• Compiling an industry profile
Characterizing the relevant properties of the raw water treated by the industry
Characterizing the types and frequency of advanced treatment technologies being used at
water treatment plants
• Characterizing disinfection byproduct (DBF) occurrence in finished water and in the
distribution system
The appropriate baseline for assessing the impacts of Stage 2 would be conditions following
implementation of Stage 1. However, the compliance deadline for the Stage 1 DBPR occurred only
recently (January 2004) for small surface water systems and all ground water systems and in January
2002 for large and medium surface water systems. Thus, the observed water quality data and occurrence
of treatment technologies presented in this chapter represent pre-Stage 1 conditions. Predictions of pre-
Stage 2 baseline (as well as post-Stage 2) conditions are presented in Chapters 5, 6, and 7.
Some characteristics of the pre-Stage 1 baseline are modeled to allow for consistent comparison
to post-Stage 1 and post-Stage 2 conditions. This chapter presents DBP occurrence and treatment
technologies in place for large surface water plants as predicted using a tool developed by the
Environmental Protection Agency (EPA), the Surface Water Analytical Tool (SWAT).
Sections 3.2 and 3.3 describe the data sources and tools used to characterize the pre-Stage 1
baseline. Section 3.4 characterizes the water industry, including the baseline estimates of treatment plants
and population subject to the Stage 2 DBPR. Influent water quality is summarized in section 3.5, and
section 3.6 describes the types of treatment technologies used by systems prior to the Stage 1 DBPR.
Treated water quality, as it relates to the pre-Stage 1 baseline, is presented in section 3.7. Lastly, section
3.8 itemizes and estimates the effects of uncertainties in the baseline analysis.
This chapter presents an analysis at a level of detail and precision appropriate to support
subsequent analyses and regulatory decisions for the Stage 2 DBPR. Therefore, it does not give an
exhaustive review of the water supply industry, source waters, or industry practices.
Final Economic Analysis for the Stage 2 DBPR 3-1 December 2005
-------
3.2 Data Sources
Several data sources were used to characterize the baseline and to predict treatment technology
changes and water quality for different regulatory alternatives. The Safe Drinking Water Information
System-Federal Version (SDWIS/FED1) data (4th Quarter Freeze Year 2003 data) is used to create system
and population baselines (USEPA 2003t). SDWIS is EPA's national regulatory compliance database for
the drinking water program. It includes information on the nation's 170,000 public water systems
(PWSs) and on violations of drinking water regulations. A second key source of data used to develop the
industry profile is the Third Edition of the Water Industry Baseline Handbook (Baseline Handbook)
(USEPA 2001c) published in May 2001, which compiles data derived from the 1995 Community Water
System Survey (CWSS) and SDWIS. The 1995 CWSS was a mail survey that covered ground and
surface water systems of all sizes (based on population served). The survey was based on a two-phased,
stratified sample design. Phase 1 was a telephone screening survey that provided a sampling frame for
the main data collection in Phase 2. The survey sample in Phase 2 was stratified according to water
system size (residential population served), ownership (public, private, or ancillary), and primary water
source (ground or surface). A total of 3,681 systems covering a range of source water types and system
sizes were randomly selected to receive the main survey questionnaire. Of these, 1,980 systems
responded. See the EPA Report, "Community Water System Survey, Volume 2" (USEPA 1997c), for
more information on the 1995 CWSS sample design and data evaluation.
EPA also used the December 2000 document, "Geometries and Characteristics of Water Systems
Report" (Model Systems Report) (USEPA 2000c). In this document, EPA analyzed 1995 CWSS data to
develop equations relating flow and population, among other things.
The data source providing the most comprehensive information on influent water quality,
treatment processes, and finished water quality came from the 1996 Information Collection Rule (ICR),
which applied to all PWSs serving at least 100,000 people, with a more limited set of ICR requirements
pertaining to ground water systems serving 50,000 to 100,000. The purpose of the ICR was to collect
DBP and microbial occurrence and treatment information to help evaluate the need for microbial and
DBP rules. The ICR gathered plant-level data from approximately 300 water systems over 18 months
(July 1997-December 1998). These data characterize the source waters and the water quality at each step
in the treatment process and at points in the distribution system. The water quality data include
information about the DBFs formed when chemical disinfectants react with naturally occurring
compounds present in source water. In addition, the ICR collected treatment and process train data that
were used in the predictive analyses described in this chapter.
The American Water Works Association (AWWA) submitted several comments in response to
the proposed Stage 1 DBPR that underscore the necessity of the ICR in developing the Stage 2 DBPR.
AWWA stated, "Promulgation of the Stage 2 D/DBPR and LT2ESWTR [Long Term 2 Enhanced Surface
Water Treatment Rule] is contingent upon completion of necessary health effects research and analysis of
the ICR data" (USEPA 1998b) and "AWWA believes that the data from both the ICR and complimentary
research will ensure that a scientific database will be created to make important and cost effective
decisions on the direction of both the final ESWTR and Stage 2 of the D/DBPR" (USEPA 1997d). In
addition, AWWA concurred with the appropriateness of the phased approach to allow for analysis of ICR
data in its comments on the 1994 rule versions. Comments on the 1994 rule from AWWA explain that
data collected through the ICR would be used to determine the occurrence of DBFs and DBP precursors
as well as treatment capabilities associated with DBP control in developing the Stage 2 DBPR (USEPA
1994b).
1 Throughout this document, the acronym "SDWIS/FED" is shortened to "SDWIS."Refer to EPA's website for more
information on SDWIS (http://www.epa.gov/safewater/sdwisfed/sdwis.htm)
Final Economic Analysis for the Stage 2 DBPR 3-2 December 2005
-------
For medium systems (serving 10,000 to 99,999 people) and small systems (serving fewer than
10,000 people), several additional data sources were used to characterize the source water and finished
water quality:2
ICR Supplemental Surveys
The National Rural Water Association (NRWA) Survey
The Ground Water Supply Survey
Small surface and ground water plant data collected by various States (several States
provided DBF data to EPA)
• The Water Utility Database (WATER:\STATS, AWWA 2000)
Data from these sources were also used to help predict treatment technologies changes to comply
with regulatory alternatives. These data are presented in detail in the Occurrence Assessment for the
Final Stage 2 Disinfectants and Disinfection Byproducts Rule (USEPA 2005k) (Occurrence Document).
The Occurrence Document and Appendix L also discuss the data quality of each of the sources used in
this Economic Analysis (EA).
3.3 Surface Water Analytical Tool
Although observed DBF data are available for pre-Stage 1 conditions, finished water quality is
modeled to give a consistent basis to compare with the pre-Stage 2 and post-Stage 2 predictions. The
SWAT is the main tool developed by EPA to model DBF occurrence for different regulatory alternatives.
SWAT uses a series of algorithms and decision rules to predict the type of treatment a plant will use and
the resulting DBF occurrence, given a specific regulatory alternative and source water quality based on
ICR data. Additional description of SWAT is provided in Appendix A and the SWAT Operations
Manual (USEPA 2000a).
SWAT was designed to provide answers to two broad questions:
What treatment technologies will large surface water treatment plants implement (given a
pre-determined, least-cost decision tree) to comply with a defined set of disinfection and DBF
compliance criteria?
What is the predicted finished and delivered water quality (particularly DBF levels) produced
by large surface water treatment plants after implementation of a range of treatment
technologies for a given set of disinfection criteria?
SWAT has four major components (Exhibit 3.la), including:
ICR Auxiliary Database 8 (AUX8) - It is a Microsoft Access™ database that contains both
inputs and outputs of the SWAT program. Inputs include ICR influent water quality and
2 Although the language in EPA rules generally does not include systems serving exactly 10,000 people in
the "small" category, this document places them in the small category to be consistent with the system and
population data categories from the Baseline Handbook.
Final Economic Analysis for the Stage 2 DBPR 3-3 December 2005
-------
plant process train data. Outputs consist of treatment technologies predicted for compliance,
treated water quality, and modified process train data.
Decision Tree Program—This part of SWAT determines how a treatment plant is modified to
comply with defined regulatory alternatives. First, the program determines if an individual
plant can be modified using the least expensive (and typically least effective) treatment
technology to comply with the regulatory alternative. If not, the program moves to the next
lowest-cost treatment technology. This process continues until the plant achieves
compliance. The program receives inputs from the database (AUX8), and uses the Water
Treatment Plant Model (described in the next bullet) to estimate treated water quality before
and after predicted treatment technology changes, and sends results back to the database.
Water Treatment Plant Model—This model is the main predictive component of SWAT. It
generates predictions of treated water quality (e.g., DBP levels) for the water treatment
process trains defined by the Decision Tree Program. Predictive modules of the model were
calibrated using the central tendency of the ICR data.
User Interface—A Windows™ interface enables the user to specify the disinfection and DBP
criteria, as well as numerous other assumptions for a SWAT run.
Exhibit 3.1 a SWAT Components
User Interface
Inputs
(eg, MX criteria)
ICR Auxiliary Database 8
Treatment Plants and Water
Quality (ICRData)
Modified Plants and Selected
Outputs (eg, DBP Levels)
Water Treatment
Rant Model
Decision Tree
Program
The Water Treatment Plant Model and the Decision Tree Program work together to predict DBP
occurrence levels and treatment plant modifications. The Water Treatment Plant Model computes DBP
concentrations that represent the treatment process train for a plant, influent water quality characteristics,
and specific treatment constraints. TTHM and HAAS concentrations are predicted for two distribution
system conditions-maximum water residence time and average water residence time.3 If the DBP
3 Empirical equations are used to estimate DBP formation based on residence time (see Appendix A for
discussion of these equations).
Final Economic Analysis for the Stage 2 DBPR
December 2005
-------
concentrations do not meet regulatory constraints at the average residence time (for Stage 1) or at
maximum residence time (for Stage 2), the Decision Tree Program modifies the process train to meet the
specified water quality objectives using a least-cost decision sequence. For the pre-Stage 2 DBPR
baseline runs (conditions following implementation of the Stage 1 DBPR) and Stage 2 DBPR regulatory
alternative runs, these criteria are based on specified maximum contaminant levels (MCLs) for DBFs. If
a plant's predicted DBP occurrence exceeds an MCL, the Decision Tree Program chooses a more
effective treatment technology (i.e., the next higher-cost option) in the decision tree, and the Water
Treatment Plant Model generates a new DBP prediction. This selection process continues until a selected
treatment technology results in the plant meeting the regulatory alternative. SWAT was also run for a
pre-Stage 1 baseline, where the model only predicted the DBP levels based on the inputted process train
and water quality data. This run was to serve as a comparison to the ICR data to evaluate and calibrate
predictions (see Appendix A).
SWAT was run using actual data from the ICR on influent water quality, treatment trains, and
related characteristics of 273 ICR surface water plants. All SWAT results are based on a 12-month period
using input data from months 7-18 (January 1998-December 1998). The number of months with valid
data input varied among plants; therefore, the output of SWAT does not contain 12 months of data for
every plant, as is discussed in greater detail in Appendix A.
Exhibit 3.1b summarizes the inputs and outputs used in the SWAT modeling process. Appendix
A describes the treatment technology selection process of the decision tree, the assumptions contained
within SWAT, and the analysis of uncertainty. For further programming details, refer to The Surface
Water Analytical Tool (SWAT) Version 1.1-Program Design and Assumptions (USEPA 2000a).
3.4 Industry Profile
This section provides the water industry characterization used to derive costs and benefits for the
Stage 2 DBPR. It is organized as follows:
• Section 3.4.1 is a background section with terminology and definitions used to characterize
the water industry baseline. It also identifies distinctions that are important for regulatory
analysis.
• Section 3.4.2 presents the baseline numbers of systems, plants, and population subject to the
Stage 2 DBPR.
Section 3.4.3 presents mean plant design and average daily flows.
Section 3.4.4 estimates the total number of households subject to the Stage 2 DBPR.
Final Economic Analysis for the Stage 2 DBPR 3-5 December 2005
-------
Exhibit 3.1 b SWAT Inputs and Outputs
Input data
Source water quality
Treatment plant
characteristics
PH
Temperature (average and annual minimum)
Total organic carbon (TOC)
Ultraviolet254 (UV) absorbance
Bromide
Alkalinity
Hardness (total and calcium)
Ammonia
Turbidity
Flow (average and peak hourly)
Presence of, sequence of, and parameters (e.g., volumes and
retention times) for unit processes, including rapid mix, flocculation,
settling basin, filtration, contact tank, reservoir, granulated activated
carbon, membranes, and ozone chambers
Dosages and chemical feeds (e.g., alum, ammonium sulfate,
ammonia, CO2, NaOH, CI2(gas), CIO2, ferric chloride, lime, ozone,
potassium permanganate, soda ash, SO2, and H2SO4)
Average and maximum distribution system residence times
Compliance measures
(DBFs)
Finished water
concentration
Distribution system
average concentration
Distribution system
maximum concentration
For each plant, and for each month for which data are available for
that plant, the DBP concentration at the entry point to the distribution
system is calculated, representing a residence time of 0. These
monthly values can then be used in compliance calculations.
For each plant, and for each month for which data are available for
that plant, the DBP concentration in the distribution system is
calculated based on the average distribution system residence time
reported by the system. These monthly values can then be used in
different compliance calculations.
For each plant, and for each month for which data are available for
that plant, the DBP concentration in the distribution system is
calculated based on the maximum distribution system residence time
reported by the system. These monthly values can then be used in
different compliance calculations.
Source: Appendix A.
Final Economic Analysis for the Stage 2 DBPR
3-6
December 2005
-------
3.4.1 Public Water System Categorization
Categorization of water systems is important because system size, ownership, and
retail/wholesale relationships dictate the way in which costs and benefits are estimated. This section
explains the water system categories as defined by EPA's National Primary Drinking Water Regulations
(NPDWRs) and describes further subdivisions according to water source, size (population served), and
ownership for regulatory analysis purposes.
PWS Type
NPDWRs apply to all PWSs. A PWS is a system that provides water for human consumption
through pipes or other constructed conveyances and that has at least 15 service connections or regularly
serves an average of at least 25 individuals per day for at least 60 days per year. PWSs are categorized as
follows:
• A Community Water System (CWS) is a PWS that has at least 15 service connections
used by year-round residents or regularly serves at least 25 year-round residents.
A Noncommunity Water System (NCWS) is a PWS that is not a CWS. NCWSs are
subdivided into two categories:
A Nontransient Noncommunity Water System (NTNCWS) is a NCWS that
regularly serves at least 25 of the same people more than 6 months per year.
A Transient Noncommunity Water System (TNCWS) is a NCWS that does not
regularly serve at least 25 of the same people more than 6 months per year.
Source Water Type
For the purposes of regulatory analysis, systems are typically categorized according to the source
of their water. Types of sources include surface water (reservoirs, lakes, or flowing streams), ground
water under the direct influence of surface water (GWUDI), ground water (aquifers not under the
influence of surface water), and treated water that is purchased from other systems. For the purposes of
this document, "surface water" includes GWUDI sources.4
In SDWIS and the Baseline Handbook (USEPA 200Ic), systems are assigned a source type using
the following hierarchy, in descending order: Surface Water, Purchased Surface Water, Ground Water,
and Purchased Ground Water. The presence of the first source in this list determines the source
assignment for that system. As a result, all "mixed systems" (systems with both a ground and surface
water source) are placed in the surface water system category. Based on an analysis in the Model
Systems Report (USEPA 2000c), it is estimated that 21 percent of surface water systems obtain some of
their water from ground water sources. Approximately one-third of these, or 8 percent of all surface
water systems in SDWIS and the Baseline Handbook, receive the majority of their flow from ground
water.
Other data sets classify systems differently. For example, the ICR Auxiliary Database 1 (AUX1)
classifies systems that receive more than 80 percent of their source water from surface water as surface
4 EPA also refers to the grouping of surface water and GWUDI systems as "subpart H" systems in the Stage
2 DBPR rule language. Surface water and GWUDI systems are grouped together because they fall under the same
requirements in the Safe Drinking Water Act (SDWA) regulations.
Final Economic Analysis for the Stage 2 DBPR 3-7 December 2005
-------
water systems. Systems that rely on ground water for more than 80 percent of their supply are considered
ground water systems. Systems that receive more than 80 percent of their supply from another system are
considered purchased water systems. All other systems are considered mixed. The 1995 CWSS data are
classified by primary source (the source that provides more than 50 percent of average flow to the
distribution system). In cases where there are three different sources (e.g., surface, ground, and
purchased), systems in the 1995 CWSS are classified by the largest source.
This EA begins with numbers from the SDWIS database. These numbers are then reclassified
according to primary source water type based on the CWSS data.
Ownership
Systems are categorized in SDWIS and in the Baseline Handbook according to three ownership
types: "private," "public," and "other." Private systems are owned by private corporations or individuals.
Public systems are owned by public entities such as municipalities, counties, or special districts and have
access to capital and means of financing that are not available to private systems. The "other" category
contains systems where ownership is not reported in SDWIS. These distinctions become important in
calculating household costs (see Chapter 7) and in assessing Unfunded Mandates Reform Act (UMRA)
requirements (see Chapter 8).
Purchased Water and Wholesale System Types
Systems are typically categorized according to whether they treat water themselves or purchase
treated water from other systems. The Stage 2 DBPR defines a consecutive system as a PWS that buys or
otherwise receives some or all of its finished water from one or more wholesale systems. A wholesale
system is defined as a PWS that treats and then sells or otherwise delivers finished water to another PWS.
Treatment modifications are generally not made by consecutive water systems, but are instead made by
the associated wholesale systems. Costs of these treatment modifications are typically passed on to the
consecutive systems in the form of water rate increases.
Population Served
The number of people served by systems (indicating system size) is a key parameter used to
calculate benefits and costs of drinking water regulations. This EA defines two types of system
populations: the retail customers of a system who buy water directly from the system, and the wholesale
customers of a system who are served by a second system that purchases treated water from the first.
Systems are categorized in SDWIS and the Baseline Handbook by retail population served. Systems in
the 1995 CWSS database are classified by total population (wholesale and retail).
Systems are categorized by population in order to group them for analyses. Mean estimates are
used for each of the nine population size categories in this EA. Although some variability is lost by
characterizing systems by nine population size categories, EPA believes the level of analysis provides
adequate information to characterize national costs and benefits. This approach is consistent with other
regulations developed by EPA.
Final Economic Analysis for the Stage 2 DBPR 3-8 December 2005
-------
3.4.2 Systems, Plants, and Population Subject to the Stage 2 DBPR
To estimate costs and benefits attributable to the Stage 2 DBPR, EPA has developed the
following industry baselines:
• The System Baseline (Appendix H) is used to estimate non-treatment costs incurred by
systems for rule implementation, IDSEs, Stage 2 monitoring plans, additional routine
monitoring, and operational evaluations. Systems are categorized based on the
population breakouts used for Stage 2 DBPR monitoring requirements which are
different than the standard nine size categories used in the rest of this EA. The derivation
of the system baseline is in Appendix H, section H. 1.
• The Plant Baseline (Exhibit 3.2) is used to estimate treatment costs (based on predictions
of plants changing to various advanced treatment technologies).
• The Population Baseline (Exhibit 3.3) is used to estimate cancer cases avoided as a result
of the Stage 2 DBPR and subsequent monetized benefits.
The purpose of this section is to define these baselines and describe how they were derived.
3.4.2.1 Plant Baseline
Exhibit 3.2, presented at the end of this subsection, shows the derivation of the baseline number
of treatment plants subject to the Stage 2 DBPR (i.e., the plant baseline). The derivation is described
below in four steps. Step 1 involves modifying the surface water system inventory to better represent the
size and number of plants that exist by "linking" purchasing surface water systems to their respective
sellers. Only surface water systems were modified in this step (the number of purchasing ground water
systems is such a small proportion of all ground water systems that linking them to sellers was not
expected to change the characterization of the ground water plant baseline). Step 2 removes systems
which do not disinfect from the baseline. In step 3, the system inventory was reclassified from the
SDWIS source water categorization to the primary source water type (i.e., the source type that provides
more than 50 percent of the water to a system). The final step, step 4, involves converting the system
inventory to a treatment plant inventory based on estimates of average treatment plants per system.
Step 1: Modify the Surface Water System Inventory by Linking Buyers and Sellers
Because population served is used directly to estimate the volume of water treated, the type of
system population reported is key to defining an accurate treatment plant baseline. As noted in section
3.4.1, system populations in SDWIS represent retail populations only. In other words, system
populations reported in SDWIS do not include the populations of those consecutive systems to whom
they sell water (purchased water systems are considered separate, stand-alone systems). More than half of
the surface water systems are consecutive, stand-alone systems. Purchased-water systems comprise a
much lower proportion of ground water systems (approximately five percent).
The advantage of classifying systems by retail population as done in SDWIS is that it
appropriately accounts for both the total number of individual PWSs in the United States and the total
population served by all of those systems. However, a disadvantage (especially for surface water CWSs)
when estimating national costs of regulations is that it does not directly account for the fact that the water
delivered by the consecutive systems to their retail customers is actually treated by other systems. It is
important to recognize that the total flow of surface water is actually treated by fewer than half of the
Final Economic Analysis for the Stage 2 DBPR 3-9 December 2005
-------
surface water systems accounted for in SDWIS. Because of economies of scale, the cost of treatment (in
cents per gallon) is less for systems treating larger flows than it is for systems treating smaller flows. For
example, it is typically more expensive to build and operate two treatment plants serving 5,000 people
than one treatment plant serving 10,000 people. Failing to account for the fact that surface water is
actually treated in larger quantities at a smaller number of systems than SDWIS suggests could result in
an upward bias in national cost estimates of rules that affect a substantial portion of surface water
systems.
To rectify this bias, an analysis was performed to "link" consecutive surface water systems to
their respective wholesale system using data from SDWIS (each purchased system lists the PWS
identification number(s) for systems that sell water to it) . If a consecutive system could be linked to a
wholesaler, that system was removed from the system count and its population was added to the
population of the wholesale system.
The methodology used to link the SDWIS 2003 system inventory is described in detail below:
If a system has multiple sources, (e.g., it has a primary source of surface water in addition
to a purchased surface water source), it was assumed to be adequately represented as a
non-purchased surface water system, and was not linked to its seller (i.e., only 100-
percent-purchased-surface-water systems were linked).
• For systems that purchase water, all sellers were identified using SDWIS data (SDWIS
has a table that lists the PWS identification number (PWSID) for each seller).
If a purchased surface water system (System P) purchases all of its water from one non-
purchased surface water system (System S), its population was added to that of System S,
and it was removed from the inventory of purchased systems.
If the purchased surface water system buys water from multiple non-purchased systems, it
was assigned to the most directly related non-purchased seller with the largest population.
For example, a purchased system (System C) purchases from a non-purchased system
(System Bl) and a purchased system (System B2), which in turn purchases from a non-
purchased system (System A). In this case, System C was linked to System Bl; in other
words, the population of System C was added to that of System B1. It was not linked to
either System B2 or System A, even if those systems were larger.
Some purchased systems have what is referred to as "cascading provider relationships."
For instance, a purchased system, System C, may purchase water from another system,
System B. System B does not treat its own water but, instead purchases water from a
non-purchased system, System A. For this analysis, the populations of both Systems B
and C were added to the population of System A, and Systems B and C were removed
from the inventory of unlinked systems.
• When the purchased system and its seller are not of the same type (e.g., a CWS
purchasing from a NTNCWS), they were not linked and are counted as separate, unlinked
purchased systems. Systems purchasing from systems of different ownership type (e.g., a
public water system purchasing from a private water system), however, were linked.
If the ID number of the seller did not correspond to an active water system, the purchased
system was counted as a separate, unlinked, purchased system.
Final Economic Analysis for the Stage 2 DBPR 3-10 December 2005
-------
In a few cases, the seller could not be found, i.e., a purchased system (e.g., System C)
cannot be linked to a non-purchased system. These purchased systems were counted as
separate, unlinked, purchased systems.
Results of the linking exercise for surface water CWSs and NTNCWSs are shown in Exhibit 3.2,
columns F through J.
As shown in Exhibit 3.2, the sum of the totals of columns F and G, there are approximately 2,700
purchasing systems remaining unlinked in the inventory. These include surface water systems that could
not be linked (e.g., many surface water NTNCWSs purchase water from CWSs and were not be linked),
purchased surface water systems with a non-purchased ground water source, and the unlinked purchased
ground water systems. For the purposes of estimating treatment costs in this EA, EPA includes the
remaining unlinked purchased water systems in the system inventory and evaluates them as if they are
treating water themselves. As described previously, evaluating purchasing systems as if they are stand-
alone, treating systems could result in an upward bias in national cost estimates. This bias is greatly
reduced, however, by the linking effort described in this step.
While EPA believes that linking consecutive systems with their wholesalers will improve the
accuracy of cost estimates, it is possible that purchased systems may be out of compliance even when the
wholesaler is in compliance, thereby obligates them to incur treatment costs that are not being captured by
this approach. EPA believes that the number of these systems, however, is small and will not have a
measurable effect on the costs or benefits of the Stage 2 DBPR.
Step 2: Remove Systems which do not Disinfect
The Stage 2 DBPR applies only to systems which disinfect their water. Therefore systems which
do not disinfect were removed from the baseline. The inventory is reduced by the percent disinfecting
(shown in Exhibit 3.2, column K) to produce the results shown in columns L through P.
The estimate of percent disinfecting in column P comes from several sources. The percent of
ground water CWSs providing disinfection is derived from 1995 CWSS results, as summarized in Table
B1.3.3 of the Baseline Handbook. The percent of ground water NTNCWSs that disinfect was derived
from Ground Water Disinfection and Protective Practices in the United States (USEPA 1996a). These
data sources do not include systems that may add disinfection to correct a significant deficiency under the
Ground Water Rule (GWR). Because the GWR is expected to be promulgated within 8 months after the
Stage 2 DBPR is promulgated, EPA expects new systems adding disinfection to meet GWR requirements
to simultaneously achieve compliance with Stage 2 MCLs. Therefore, these systems are not included in
the treatment baseline.
Step 3: Re-classify Systems by Primary Source Water Type
The characterization of distribution system DBP levels and predictions of treatment technology
changes are very different for ground and surface water plants. This is mainly because ground water
sources generally have lower DBP precursor (e.g., TOC) concentrations than surface water; thus, DBP
levels and predicted changes to meet Stage 2 DBPR requirements are generally less for plants treating
ground water than for plants treating surface water.
As noted in section 3.4.1, all mixed systems (even those that are primarily ground water) are
grouped with the 100 percent surface water systems in SDWIS. If EPA applied the compliance forecasts
for surface water plants to systems that are primarily served by ground water sources, costs could be
overstated. Therefore, systems were reclassified by primary source. This is consistent with
Final Economic Analysis for the Stage 2 DBPR 3-11 December 2005
-------
recommendations in the Arsenic NDWAC Final Report (National Drinking Water Advisory Council
2001).
SDWIS does not contain information on whether or not a system is mixed or the relative
proportions of surface and ground water flow used, indicating only whether it is served by all ground
water or by at least some proportion of surface water. Therefore, to reclassify by primary source, EPA
used flow data from the 1995 CWSS to estimate the proportion of surface water and mixed CWSs that
received more than 50 percent of their flow from a ground water source (percentages are shown in Exhibit
3.2, column Q). These systems, originally classified as surface water CWSs in SDWIS, were re-assigned
to the ground water CWS category. Note that this adjustment was not made for NTNCWSs because these
systems are most often a single building or in a small area, and are less likely to be served by more than
one source type.
Step 4: Convert System Inventory to Plant Inventory
The 1995 CWSS data (question 18 from the CWSS questionnaire) were used to estimate the
number of treatment plants per system for both surface and ground water CWSs for all system sizes. The
analysis produced a distribution of plants per system within each system size category. For analyses in
this EA, EPA uses the mean plant per system estimate (presented in column W of Exhibit 3.2). For
NTNCWSs, EPA assumed a 1:1 plant per system ratio for all sizes and source water types because these
systems are most often a single building or located in a small area.
Final Economic Analysis for the Stage 2 DBPR 3-12 December 2005
-------
Exhibit 3.2 Derivation of the Stage 2 DBPR Plant Baseline
Step 1 : Use modified inventory for surface water systems
System Size
(population served)
No. of Purchased
Systems from SDWIS
Public
A
Private
B
No. of Non-Purchased
Systems from SDWIS
Public
C
Private
D
Total No. of
Systems from
SDWIS
E=A+B+C+D
Remaining
Unlinked Systems
Public
F
Private
G
No. of Linked
Systems
Public
H
Private
I
Total
Number of
Linked
Systems
J = F+G+H+I
Surface Water and All Mixed CWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
413
855
655
1,189
862
704
111
59
0
4,848
293
630
385
302
139
91
18
9
0
1,867
183
433
333
940
955
878
174
188
15
4,099
196
294
97
157
86
100
31
25
3
989
1,085
2,212
1,470
2,588
2,042
1,773
334
281
18
11,803
23
41
21
53
48
32
7
7
0
232
8
24
11
23
15
2
0
0
0
83
149
416
309
862
947
967
222
238
20
4,130
193
290
98
153
90
103
36
28
3
994
373
771
439
1,091
1,100
1,104
265
273
23
5,439
Ground Water-Only CWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
156
613
326
332
113
34
1
1
0
1,576
106
252
103
83
19
6
0
0
0
569
1,243
4,294
2,794
4,197
2,117
1,040
113
52
3
15,853
10,395
9,569
1,613
1,257
412
200
28
12
0
23,486
11,900
14,728
4,836
5,869
2,661
1,280
142
65
3
41,484
156
613
326
332
113
34
1
1
0
1,576
106
252
103
83
19
6
0
0
0
569
1,243
4,294
2,794
4,197
2,117
1,040
113
52
3
15,853
10,395
9,569
1,613
1,257
412
200
28
12
0
23,486
11,900
14,728
4,836
5,869
2,661
1,280
142
65
3
41,484
Surface Water and All Mixed NTNCWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
15
31
10
17
8
2
0
1
0
84
36
44
16
12
3
2
0
0
0
113
90
124
38
26
5
0
0
0
0
283
90
118
42
37
9
1
0
0
0
297
231
317
106
92
25
5
0
1
0
777
15
31
10
16
8
2
0
1
0
83
31
42
14
12
3
2
0
0
0
104
90
123
39
25
6
0
0
0
0
283
90
116
43
38
8
1
0
0
0
296
226
312
106
91
25
5
0
1
0
766
Ground Water-Only NTNCWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
Grand Total, All Systems
11
17
7
8
5
3
0
0
0
51
6,559
10
17
3
1
1
0
0
0
0
32
2,581
1,703
3,166
1,224
432
28
5
1
0
0
6,559
26,794
6,872
4,141
798
411
40
3
0
1
0
12,266
37,038
8,596
7,341
2,032
852
74
11
1
1
0
18,908
72,972
11
17
7
8
5
3
0
0
0
51
1,942
10
17
3
1
1
0
0
0
0
32
788
1,703
3,166
1,224
432
28
5
1
0
0
6,559
26,825
6,872
4,141
798
411
40
3
0
1
0
12,266
37,042
8,596
7,341
2,032
852
74
11
1
1
0
18,908
66,597
Sources:
(A) - (D) SDWIS 4th Quarter 2003 Frozen Database, systems with an other ownership designation were considered public
(F) - (I) Analysis of data in the SDWIS 4th Quarter 2003 Frozen Database
Final Economic Analysis for the Stage 2 DBPR
3-13
December 2005
-------
Exhibit 3.2 Derivation of the Stage 2 DBPR Plant Baseline (Continued)
Step 2: Calculate Disinfecting Systems
System Size
(population served)
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
Grand Total, All Systems
Percent
Disinfecting
K
Remaining Disinfecting
Unlinked Systems
Public
L=K*F
Private
M=K*G
No. of Linked Disinfecting
Systems
Public
N=K*H
Private
O=K*I
Linked
Disinfecting
Systems
P=L+M+N+O
LINKED Disinfecting Surface Water and All Mixed CWSs
100%
100%
100%
100%
100%
100%
100%
100%
100%
23
41
21
53
48
32
7
7
0
232
8
24
11
23
15
2
0
0
0
83
149
416
309
862
947
967
222
238
20
4,130
193
290
98
153
90
103
36
28
3
994
373
771
439
1,091
1,100
1,104
265
273
23
5,439
Disinfecting Ground Water-Only CWSs
53%
78%
84%
80%
87%
97%
86%
96%
100%
82
478
274
265
98
33
1
1
0
1,231
56
196
87
66
16
6
0
0
0
427
656
3,345
2,347
3,345
1,838
1,004
98
50
3
12,685
5,489
7,454
1,355
1,002
358
193
24
12
0
15,886
6,283
11,473
4,062
4,678
2,310
1,235
123
63
3
30,229
LINKED Disinfecting Surface Water and All Mixed NTNCWSs
100%
100%
100%
100%
100%
100%
100%
100%
100%
15
31
10
16
8
2
0
1
0
83
31
42
14
12
3
2
0
0
0
104
90
123
39
25
6
0
0
0
0
283
90
116
43
38
8
1
0
0
0
296
226
312
106
91
25
5
0
1
0
766
Disinfecting Ground Water-Only NTNCWSs
29%
29%
29%
29%
29%
29%
29%
29%
29%
-
3
5
2
2
1
1
0
0
0
15
1,561
3
5
1
0
0
0
0
0
0
9
624
494
918
355
125
8
1
0
0
0
1,902
19,000
1,993
1,201
231
119
12
1
0
0
0
3,557
20,733
2,493
2,129
589
247
21
3
0
0
0
5,483
41,918
Sources:
Sources:
(K) Percentage of ground water CWSs that disinfect is estimated using
percentage of treatment in place from the Third Edition of the Baseline
Handbook (Table B1.3.3), originally derived from the 1995 CWSS.
Final Economic Analysis for the Stage 2 DBPR
3-14
December 2005
-------
Exhibit 3.2 Derivation of the Stage 2 DBPR Plant Baseline (Continued)
Step 3: Re-allocate such that systems are categorized by primary source water type
System Size
(population served)
% SW that
are
Primarily
GW
Q
No. of Disinfecting,
Purchased Systems by
Primary Source
Public
R
Private
S
No. of Disinfecting,
Non-Purchased
Systems by Source
Public
T
Private
U
Total No. of
Disinfecting,
Linked Systems
V=R+S+T+U
LINKED Primarily Surface Water CWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
3.7%
9.6%
0.0%
5.9%
12.0%
10.0%
8.9%
14.0%
0.0%
8.4%
22
37
21
50
42
29
6
6
0
214
Primarily Ground Water CWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
-
-
-
-
83
481
274
268
104
36
1
2
0
1,250
8
22
11
22
13
2
0
0
0
77
143
376
309
811
833
870
202
205
20
3,770
186
262
98
144
79
93
33
24
3
922
359
697
439
1,027
968
994
241
235
23
4,983
56
199
87
68
18
6
0
0
0
433
662
3,385
2,347
3,396
1,951
1,100
117
83
3
13,045
5,496
7,482
1,355
1,011
368
203
27
15
0
15,958
6,297
11,547
4,062
4,742
2,442
1,346
146
101
3
30,686
LINKED Primarily Surface Water NTNCWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
15
31
10
16
8
2
0
1
0
83
31
42
14
12
3
2
0
0
0
104
90
123
39
25
6
0
0
0
0
283
90
116
43
38
8
1
0
0
0
296
226
312
106
91
25
5
0
1
0
766
Primarily Ground Water NTNCWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
Grand Total, All Sys.
-
-
-
-
-
-
3
5
2
2
1
1
0
0
0
15
1,561
3
5
1
0
0
0
0
0
0
9
624
494
918
355
125
8
1
0
0
0
1,902
19,000
1,993
1,201
231
119
12
1
0
0
0
3,557
20,733
2,493
2,129
589
247
21
3
0
0
0
5,483
41,918
Sources:
(Q) Percentage of SW systems that are primarily GW from "Geometries and Characteristics of
Public Water Supplies" (USEPA 2000c), Exhibit 2.9.
(R) For surface water, R=L*(1-Q); for ground water, R=L+((Q for SW)*(L for SW)).
Final Economic Analysis for the Stage 2 DBPR
3-15
December 2005
-------
Exhibit 3.2 Derivation of the Stage 2 DBPR Plant Baseline (Continued)
| Baseline Number of Plants Subject to the Stage 2 DBPR
Step 4: Convert system inventory to plant inventory
System Size
(population served)
Plants per
System
W
Disinfecting, Purchased Plants
Public
X=W*R
Private
Y=W*S
Disinfecting, Non-
Purchased Plants
Public
Z=W*T
Private
AA=W*U
Total No.of
Disinfecting Plants
AB = X+Y+Z+AA
LINKED Primarily Surface Water CWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
1.0
1.1
1.1
1.1
1.3
1.3
2.4
2.6
3.2
-
22
41
23
55
55
37
15
16
0
264
8
24
12
24
17
2
0
0
0
87
143
414
340
892
1,083
1,131
485
532
64
5,086
186
288
108
158
103
121
79
63
10
1,115
359
767
483
1,129
1,258
1,292
579
610
74
6,552
Disinfecting Primarily Ground Water CWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
1.0
1.3
1.5
1.6
2.1
4.0
4.9
9.1
9.1
-
85
636
411
428
214
144
7
18
0
1,942
57
262
130
108
38
24
0
0
0
619
675
4,468
3,520
5,433
4,019
4,401
575
759
27
23,879
5,606
9,876
2,032
1,617
759
813
134
141
0
20,979
6,423
15,242
6,093
7,587
5,030
5,382
716
918
27
47,419
LINKED Primarily Surface Water NTNCWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
-
15
31
10
16
8
2
0
1
0
83
31
42
14
12
3
2
0
0
0
104
90
123
39
25
6
0
0
0
0
283
90
116
43
38
8
1
0
0
0
296
226
312
106
91
25
5
0
1
0
766
LINKED Disinfecting Primarily Ground Water NTNCWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
Grand Total, All Sys.
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
-
-
3
5
2
2
1
1
0
0
0
15
2,304
3
5
1
0
0
0
0
0
0
9
819
494
918
355
125
8
1
0
0
0
1,902
31,150
1,993
1,201
231
119
12
1
0
0
0
3,557
25,947
2,493
2,129
589
247
21
3
0
0
0
5,483
60,220
Sources:
(W) Derived from Question 18 of the 1995 CWSS, calculations based on classification of systems by primary
source. Methodology to be included in subsequent drafts of the Geometries Document (USEPA 2000c).
Final Economic Analysis for the Stage 2 DBPR
3-16
December 2005
-------
3.4.2.2 Population Baseline
The population baseline is used in the Stage 2 DBPR benefits analysis to help derive the cases of
bladder cancer avoided as a result of treatment technology changes resulting from the Stage 2 DBPR (see
chapter 6). Because the benefits of the rule are a function of treatment technology changes and
subsequent DBP reductions, the population baseline must be consistent with the plant baseline. Thus, the
derivation of the Stage 2 DBPR population baseline is similar to that of the Stage 2 DBPR plant baseline
(with the exception of Step 4- convert system inventory to plant inventory; this is not needed given that
the population served by all plants in a size category and all systems in a size category are the same).
Note that because NTNCWs are most often businesses such as restaurants, schools, campgrounds,
etc., their population generally duplicates the population served by CWSs. Total population served by
disinfecting systems as derived in this section is just the total of the CWS population served, or
264,513,763 (169,358,139 + 95,155,624) from Exhibit 3.3, column Q.
Step 1: Modify the Surface Water System Inventory by Linking Buyers and Sellers
As with the plant baseline, EPA has modified the system-level population data in SDWIS to add,
for surface water systems, populations served by purchasing systems to the population served by their
wholesale seller. The overall effect of this step shifts population into higher system size categories;
however, it does not alter the total population served by all surface water systems as reported in SDWIS.
Section 3.4.2.1 provides the rationale and detailed methodology used to modify the surface water
inventory.
Step 2: Remove Population of Systems that do not Disinfect
The Stage 2 DBPR applies only to systems that disinfect their water. Therefore, systems that do
not disinfect were removed from the baseline. The inventory is reduced by the percent disinfecting
(shown in Exhibit 3.3, column F) to produce the results shown in columns G through K.
The estimate of percent disinfecting in column P is derived in section 3.4.2.1
Step 3: Re-classify Population by Primary Source Water Type
As with the plant baseline, EPA modified population data from SDWIS to represent populations
served either by primarily ground or primarily surface water systems. Given that SDWIS does not
contain information on whether or not a system is mixed or the relative proportions of surface and ground
water flow used, EPA used flow data from the 1995 CWSS to estimate the proportion of surface water
and mixed CWSs that received more than 50 percent of their flow from a ground water source
(percentages are shown in Exhibit 3.3, column L). This population, originally classified as served by
surface water CWSs in SDWIS, was re-assigned to the ground water CWS category. This adjustment was
not made for NTNCWSs because these systems are most often a single building or in a small area, and are
less likely to be served by more than one source type.
Final Economic Analysis for the Stage 2 DBPR 3-17 December 2005
-------
Exhibit 3.3 Derivation of the Stage 2 DBPR Population Baseline
Step 1 : Use linked inventory for SW systems
System Size
(population served)
Population Served by
Unlinked Systems
Public
A
Private
B
Population Served by
Linked Systems
Public
C
Private
D
Total
Population
Served
E=A+B+C+D
LINKED Surface Water and All Mixed CWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
1,124
10,817
15,425
107,441
287,030
693,894
476,236
2,432,694
0
4,024,661
399
5,925
7,381
47,998
76,258
38,135
0
0
0
176,096
8,260
114,084
229,498
1,735,746
5,746,521
22,634,702
15,181,764
63,779,714
48,565,698
157,995,987
10,743
68,769
65,606
284,439
545,907
2,537,058
2,591,527
9,798,126
7,062,250
22,964,425
20,526
199,595
317,910
2,175,624
6,655,716
25,903,789
18,249,527
76,010,534
55,627,948
185,161,169
Ground Water-Only CWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
9,640
165,625
239,129
594,474
586,870
644,110
87,933
140,000
0
2,467,781
6,456
64,176
70,939
146,219
97,393
82,397
0
0
0
467,580
73,773
1,180,056
2,022,042
7,850,024
11,954,771
21,421,779
7,191,107
11,222,309
3,933,533
66,849,394
604,212
2,054,329
1,111,269
2,201,328
2,347,681
4,180,506
1,955,231
2,108,763
0
16,563,319
694,081
3,464,186
3,443,379
10,792,045
14,986,715
26,328,792
9,234,271
13,471,072
3,933,533
86,348,074
LINKED Surface Water and All Mixed NTNCWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
752
7,481
6,659
29,057
49,231
72,000
0
169,846
0
335,026
1,578
9,531
8,568
18,649
13,003
43,055
0
0
0
94,384
4,285
27,491
25,534
45,178
24,537
0
0
0
0
127,025
4,486
27,624
29,560
60,403
38,642
13,000
0
0
0
173,715
11,101
72,127
70,321
153,287
125,413
128,055
0
169,846
0
730,150
Ground Water-Only NTNCWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
Grand Total, All Systems
545
3,984
4,510
13,823
23,800
55,200
0
0
0
101,862
6,929,330
487
3,080
1,950
1,350
4,800
0
0
0
0
11,667
749,727
87,724
806,717
824,809
657,954
145,634
137,008
66,000
0
0
2,725,846
227,698,252
344,860
845,693
535,712
649,238
207,114
36,200
0
110,000
0
2,728,817
42,430,276
433,616
1,659,474
1,366,981
1,322,365
381,348
228,408
66,000
110,000
0
5,568,192
277,807,585
Sources:
(A-D) for Surface Water CWSs & NTNCWSs: "Linked" system inventory derived from
SDWIS 4th Quarter Year 2003 Freeze data. See section 3.4.2.2 for a description of linking
methodology.
Final Economic Analysis for the Stage 2 DBPR
3-18
December 2005
-------
Exhibit 3.3 Derivation of the Stage 2 DBPR Population Baseline (Continued)
Step 2: Remove population which doesn't disinfect
System Size
(population served)
Percent
Disinfecting
F
Population Served by
Remaining Unlinked,
Disinfecting Systems
Public
G=A*F
Private
H=B*F
Population Served by Linked,
Disinfecting Systems
Public
I=C*F
Private
J=D*F
Total Population
Served by
Disinfecting Systems
K=G+H+I+J
LINKED Surface Water and All Mixed CWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
100%
100%
100%
100%
100%
100%
100%
100%
100%
-
1,124
10,817
15,425
107,441
287,030
693,894
476,236
2,432,694
0
4,024,661
399
5,925
7,381
47,998
76,258
38,135
0
0
0
176,096
8,260
114,084
229,498
1,735,746
5,746,521
22,634,702
15,181,764
63,779,714
48,565,698
157,995,987
10,743
68,769
65,606
284,439
545,907
2,537,058
2,591,527
9,798,126
7,062,250
22,964,425
20,526
199,595
317,910
2,175,624
6,655,716
25,903,789
18,249,527
76,010,534
55,627,948
185,161,169
Disinfecting Ground Water-Only CWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
53%
78%
84%
80%
87%
97%
86%
96%
100%
-
5,090
129,022
200,868
473,796
509,403
621,566
75,886
134,960
0
2,150,591
3,409
49,993
59,589
116,537
84,537
79,513
0
0
0
393,577
38,952
919,264
1,698,515
6,256,469
10,376,741
20,672,017
6,205,925
10,818,306
3,933,533
60,919,722
319,024
1,600,322
933,466
1,754,458
2,037,787
4,034,188
1,687,364
2,032,848
0
14,399,458
366,475
2,698,601
2,892,438
8,601,260
13,008,469
25,407,284
7,969,176
12,986,113
3,933,533
77,863,349
LINKED Surface Water and All Mixed NTNCWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
100%
100%
100%
100%
100%
100%
100%
100%
100%
-
752
7,481
6,659
29,057
49,231
72,000
0
169,846
0
335,026
1,578
9,531
8,568
18,649
13,003
43,055
0
0
0
94,384
4,285
27,491
25,534
45,178
24,537
0
0
0
0
127,025
4,486
27,624
29,560
60,403
38,642
13,000
0
0
0
173,715
11,101
72,127
70,321
153,287
125,413
128,055
0
169,846
0
730,150
Disinfecting Ground Water-Only NTNCWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
Grand Total, All Systems
29%
29%
29%
29%
29%
29%
29%
29%
29%
-
-
158
1,155
1,308
4,009
6,902
16,008
0
0
0
29,540
6,539,818
141
893
566
392
1,392
0
0
0
0
3,383
667,441
25,440
233,948
239,195
190,807
42,234
39,732
19,140
0
0
790,495
219,833,230
100,009
245,251
155,356
188,279
60,063
10,498
0
31,900
0
791,357
38,328,955
125,749
481,247
396,424
383,486
110,591
66,238
19,140
31,900
0
1,614,776
265,369,444
Sources:
(F) Percentage of ground water CWSs that disinfect is estimated using percentage of treatment in place from the
Third Edition of the Baseline Handbook (Table B1.3.3), originally derived from the 1995 CWSS.
Final Economic Analysis for the Stage 2 DBPR
3-19
December 2005
-------
Exhibit 3.3 Derivation of the Stage 2 DBPR Population Baseline (Continued)
| Baseline Number of People Subject to the Stage 2 DBPR
Step 3: Re-allocate such that systems are categorized by primary source water type
System Size
(population served)
%SW
that are
Primarily
GW
L
Population Served by
Remaining Unlinked,
Disinfecting Systems
Public
M
Private
N
Population Served by Linked,
Disinfecting Systems
Public
O
Private
p
Total
Population
Served by
Disinfecting
Systems
Q=M+N+O+P
LINKED Primarily Surface Water CWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
3.7%
9.6%
0.0%
5.9%
12.0%
10.0%
8.9%
14.0%
0.0%
-
1,082
9,779
15,425
101,102
252,586
624,505
433,851
2,092,117
0
3,530,447
384
5,356
7,381
45,166
67,107
34,322
0
0
0
159,716
7,954
103,132
229,498
1,633,337
5,056,938
20,371,232
13,830,587
54,850,554
48,565,698
144,648,931
10,346
62,167
65,606
267,657
480,398
2,283,352
2,360,881
8,426,388
7,062,250
21,019,046
19,767
180,434
317,910
2,047,262
5,857,030
23,313,410
16,625,319
65,369,059
55,627,948
169,358,139
Disinfecting Primarily Ground Water CWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
5,132
130,060
200,868
480,135
543,847
690,956
118,271
475,537
0
2,644,806
3,424
50,562
59,589
119,368
93,688
83,327
0
0
0
409,957
39,258
930,216
1,698,515
6,358,878
11,066,324
22,935,487
7,557,102
19,747,466
3,933,533
74,266,779
319,421
1,606,924
933,466
1,771,240
2,103,296
4,287,894
1,918,010
3,404,585
0
16,344,837
367,234
2,717,762
2,892,438
8,729,622
13,807,155
27,997,663
9,593,384
23,627,588
3,933,533
93,666,379
LINKED Primarily Surface Water NTNCWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
-
752
7,481
6,659
29,057
49,231
72,000
0
169,846
0
335,026
1,578
9,531
8,568
18,649
13,003
43,055
0
0
0
94,384
4,285
27,491
25,534
45,178
24,537
0
0
0
0
127,025
4,486
27,624
29,560
60,403
38,642
13,000
0
0
0
173,715
11,101
72,127
70,321
153,287
125,413
128,055
0
169,846
0
730,150
Disinfecting Primarily Ground Water NTNCWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
Grand Total, All Systems
-
-
158
1,155
1,308
4,009
6,902
16,008
0
0
0
29,540
6,539,818
141
893
566
392
1,392
0
0
0
0
3,383
667,441
25,440
233,948
239,195
190,807
42,234
39,732
19,140
0
0
790,495
219,833,230
100,009
245,251
155,356
188,279
60,063
10,498
0
31,900
0
791,357
38,328,955
125,749
481,247
396,424
383,486
110,591
66,238
19,140
31,900
0
1,614,776
265,369,444
Note: Detail may not add due to independent rounding.
Sources:
(L) Percentage of SW systems that are primarily GW from "Geometries and Characteristics of
Public Water Supplies" (USEPA 2000c), Exhibit 2.9.
(M) For surface water, M=G*(1-L); For ground water, M=G+((G for SW)*(L for SW)).
(O) For surface water, O=I*(1-L); For ground water, O=l+((l for SW)*(L for SW)).
(P) For surface water, P=J*(1-L); For ground water, P=J+((J for SW)*(L for SW)).
Final Economic Analysis for the Stage 2 DBPR
3-20
December 2005
-------
3.4.3 Water Treatment Plant Design and Average Daily Flows
Treatment technology costs depend on the volume of water treated per day. The cost analysis
described in Chapter 7 uses two types of treatment plant flow: (1) design flow, which is the maximum
capacity at which the plant was intended to operate, expressed in millions of gallons per day (MGD), and
(2) average daily flow, which is the flow produced by a treatment plant in one day, an average derived
from 365 days of flow measurements, expressed in MGD. Design flows are used to estimate the capital
costs of the treatment technology that will be installed to meet the requirements of the Stage 2 DBPR.
Average daily flows are used to estimate the annual cost of ongoing operations and maintenance (O&M).
To estimate flows for different sized systems, EPA developed the following regression equations:
Surface Water: Design Flow (MGD) = 0.36971 ^-97757/ 1,000
Average Daily Flow (MGD) = 0.10540 X102058/ 1,000
Ground Water: Design Flow (MGD) = 0.39639 .Y097708/ 1,000
Average Daily Flow (MGD) = 0.06428 X1 -07652/ 1,000
Where X= mean population served per system.5
These equations are based on 1995 CWSS data. Their derivation is presented in detail in the Model
Systems Report (USEPA 2000c) and summarized in the Baseline Handbook (USEPA 200Ic). The
equations are used in this EA to estimate mean flows per plant for each size category, using the mean
population served per plant. (The mean population served per plant can be calculated by dividing the
total population for a given size category presented in Exhibit 3.3, column Q, by the baseline number of
plants in that size category as presented in Exhibit 3.2, column V.). Exhibit 3.4 shows the population per
system, the number of plants per system, and the design and average flows per plant.
This EA uses a single regression equation to estimate flows for either public or privately owned
systems. There is, however, a slight difference in the flow characteristics for these two ownership types,
as discussed in the Model Systems Report (USEPA 2000c). The use of different flow equations for
public and private systems would not affect total national costs, although per-household costs may be
slightly affected. EPA has evaluated the equations and believes that the differences are small and would
have a negligible effect on household costs.
5 Equations are from the 3nd Edition of the Baseline Handbook as derived in December 2000 Model Systems
Report (USEPA 2000c).
Final Economic Analysis for the Stage 2 DBPR 3-21 December 2005
-------
Exhibit 3.4 Design Flows and Average Daily Flows per Plant (MGD)
System Size
(Population
Served)
Average No. of
Population Served per
System
Average No. of
Plants/System
Design Flows (MGD)
Per Plant
Average Daily Flow (MGD) Per
Plant
LINKED, Primarily Surface Water CWSs
<100
1 00-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1 ,000,000+
X
55.0
258.9
724.2
1 ,994.2
6,050.7
23,463.6
68,866.1
278,426.9
2,418,606.4
Y
1.0
1.1
1.1
1.1
1.3
1.3
2.4
2.6
3.2
Z = 0.36971 X° 97757/1000Y
0.019
0.077
0.210
0.565
1.415
5.325
8.263
29.886
200.952
AA= 0.10540 X102058/1000Y
0.006
0.028
0.079
0.223
0.587
2.340
3.804
1 4.609
107.803
Disinfecting Primarily Ground Water CWSs
<100
1 00-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1 ,000,000+
X
58.3
235.4
712.0
1 ,840.9
5,654.6
20,806.8
65,649.2
234,214.8
1,311,177.7
Y
1.0
1.3
1.5
1.6
2.1
4.0
4.9
9.1
9.1
Z = 0.39639 X° 97708/1000Y
0.021
0.062
0.162
0.384
0.893
1.642
4.119
7.685
41 .355
AA= 0.06428 X1 07652/1000Y
0.005
0.017
0.050
0.131
0.342
0.716
2.012
4.261
27.216
LINKED, Primarily Surface Water NTNCWSs
<100
1 00-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1 ,000,000+
X
49.1
231.2
663.4
1 ,684.5
5,016.5
25,611.0
-
169,846.0
-
Y
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
Z = 0.36971 X° 97757/1000Y
0.017
0.076
0.212
0.527
1.532
7.541
-
47.930
-
AA= 0.10540 X1 02058/1000Y
0.006
0.027
0.080
0.207
0.630
3.327
-
22.937
-
Disinfecting Primarily Ground Water NTNCWSs
<100
1 00-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1 ,000,000+
X
50.4
226.1
672.7
1,552.1
5,153.4
20,764.4
66,000.0
110,000.0
-
Y
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
Z = 0.39639 X° 97708/1000Y
0.018
0.079
0.230
0.520
1.679
6.554
20.286
33.417
-
AA= 0.06428 X107652/1000Y
0.004
0.022
0.071
0.175
0.637
2.856
9.918
17.188
-
Note: Formulas may not produce exact results due to independent rounding (average people per plant includes fractions).
Source: Equations relating mean population to flow are from the Baseline Handbook (USEPA 2001 c).
X is the total population for the size category (Exhibit 3.3, column Q) divided by the total number of systems for the size category
(Exhibit 3.2, column V).
Final Economic Analysis for the Stage 2 DBPR
3-22
December 2005
-------
Comparable analyses relating average daily and design flow to population was not performed for
the NTNCWSs. Other drinking water rules have evaluated flows for NTNCWSs according to service
categories (e.g., schools, restaurants, hotels, industry) instead of size. EPA considered using this method
for evaluating NTNCWSs for the Stage 2 DBPR, but decided against it for the following reasons:
• Service category flows are based on mean population served for all systems in that category,
regardless of source water type. EPA expects that surface water and GWUDI sources would
be more prevalent in larger NTNCWSs, but has no basis for developing revised population
estimates for each service category by source.
The prediction of treatment technology selection in Chapter 7 is a function of population
served and does not directly apply to service categories that may include a wide range of
water system sizes and flows (e.g., schools can be very small local buildings or metropolitan
high schools).
EPA, therefore, applied the CWS regression equations to NTNCWSs, recognizing that this may
over-estimate flows and, therefore, costs. This over-estimation is addressed as part of the uncertainties
summarized in section 3.8. Note that because the ratio of plants per system was assumed to be 1:1 for all
NTNCWSs, plant flows equal system flows. Mean plant flows for CWSs and NTNCWSs may differ
from each other because of the difference in mean population per plant within each size category.
3.4.4 Number of Households Served
The number of households served by CWSs expected to be subject to the Stage 2 DBPR is
estimated by dividing the population for each system size category by the average number of people per
household (2.59) (U.S. Census Bureau 2001). As shown in Exhibit 3.5, PWSs serve about 102 million
households.
Exhibit 3.5 Number of Households Subject to the Stage 2 DBPR
System Size (Population
Served)
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Total
National Total
Number of Households Served
Linked, Primarily Surface
Water
7,632
69,666
122,745
790,449
2,261,402
9,001,317
6,419,042
25,239,019
21,477,972
65,389,243
Primarily Disinfecting
Ground Water
141,789
1 ,049,329
1,116,772
3,370,510
5,330,948
10,809,909
3,704,009
9,122,621
1,518,739
36,164,625
101,553,868
Note: Detail may not add due to independent rounding.
Source: Calculated by dividing the total population served (Exhibit 3.3, column Q)
by 2.59, the average number of people per household (U.S. Census Bureau 2001).
Note: Detail may not add due to independent rounding.
Source: Calculated by dividing the total population served (Exhibit 3.3, column Q) by 2.59, the average
number of people per household (U.S. Census Bureau 2001).
Final Economic Analysis for the Stage 2 DBPR
3-23
December 2005
-------
3.5 Influent Water Quality Characterization
3.5.1 Summary of Available Influent Water Quality Data
Predictions of compliance forecasts assume that a system will choose a treatment technology for
the Stage 2 DBPRthat best addresses its water quality improvement needs. The quality of the source
water plays a key role in evaluating treatment alternatives to meet regulatory requirements. This section
provides an overview of influent water quality on a national level.
Exhibit 3.6 summarizes the influent water quality data that were collected under the ICR for large
surface and ground water systems. These data represent the parameters that most affect DBF formation in
disinfected waters. The median, 90th percentile, and range provide some insight into the variability in
plant means among all large surface and ground water plants.
Monthly plant data collected during the last 12 months of the ICR collection period (January
1998-December 1998) were averaged to estimate a "plant-mean" value for each plant. Only the last 12
months were evaluated because they appear to be of higher quality than data collected during the first 6
months of the survey. In addition, using all 18 months of data could skew results (data from the last 2
quarters of the year would be counted twice).
ICR summary data in Exhibit 3.6 are grouped according to "Ground" or "Surface" water plants
types. This designation is based on the source water type reported by the plant for each month from July
1997 to December 1998. The types of sources recorded were surface water, ground water, mixed, or
purchased. Most plants reported on one source type for all months, but some plants reported surface
water for some months and mixed for others. These plants were considered surface water plants. One
ground water plant reported ground water for some months and mixed for others—this plant was
considered a ground water. Analyses of all plants includes "blended," "mixed," and "purchased" plant-
types from the ICR database. These plant types make up a small portion (less than 10 percent) of the
total—most ICR plants are categorized as either surface or ground water plants.
Data for these influent water quality parameters were part of the input data for SWAT. They
were also used by the Delphi Group, a group of industry experts gathered to advise EPA on technical
issues surrounding the rule, and the small-system experts (both for surface water and ground water) to
assess treatment alternatives. For a complete characterization and discussion of these parameters, see
Chapter 3 of the Occurrence Document (USEPA 2005k).
Final Economic Analysis for the Stage 2 DBPR 3-24 December 2005
-------
Exhibit 3.6 ICR Large System Influent Water Quality Parameters—
Summary of Pre-Stage 1 Plant-Mean Data
Parameter
Alkalinity
(mg/L as CaCO3)
Bromide (mg/L)
PH
Temperature (°C)
Total Hardness
(mg/L as CaCO3)
Total Organic
Carbon
(mg/L as C)
Turbidity
(Nephelometric
Turbidity Units)
UV254 Absorbance
(cm'1)
Source
Type
Surface
Ground
Surface
Ground
Surface
Ground
Surface
Ground
Surface
Ground
Surface
Ground
Surface
Ground
Surface
Ground
Number
of
Plants
336
121
320
118
336
119
334
121
326
116
307
103
328
116
306
104
Mean
of
Plant
Means
81
159
0.055
0.103
7.6
7.3
16.0
19.9
117
194
3.14
1.46
18.2
1.3
0.098
0.062
Median of
Plant Means
79
156
0.027
0.066
7.7
7.4
16.1
20.1
109
181
2.71
0.19
6.7
0.2
0.079
0.009
90th Percent! le of
Plant Means
165
264
0.115
0.190
8.2
8.0
20.7
26.3
251
352
5.29
3.36
34.0
2.6
0.176
0.266
Range of Plant
Means
2.75-273
1.00-415
0-1.325
0-1.325
6.0-8.5
4.1 -8.8
3.7-27.7
9.5-30.5
3.1 -501
3.6-778
0-21.4
0-16.1
0.06-529
0.03-38.7
0-0.880
0-0.606
Note: The maximum surface water bromide mean value, 3.13 milligrams per liter (mg/L), is not shown. This value was
calculated based on a one-month reported bromide concentration of 28 mg/L, which EPA assumes to be a reporting
error. (Laboratories often report bromide values in ug/L, rather than mg/L; this value may not have been converted to
mg/L.) All the other values for that plant in the last 12 months of the ICR were below 0.1 mg/L.
Source: ICR AUX1 database (USEPA 2000h). Represents distribution of plant-mean data as calculated using ICR
monthly data from the last 12 months of the ICR (January 1998 - December 1998). Only plants with reported data for
at least 9 of the 12 months are included in this summary table. Does not include blended, mixed, or purchased
plants.
Final Economic Analysis for the Stage 2 DBPR
3-25
December 2005
-------
A key influent water quality parameter related to Stage 2 DBPR compliance is TOC. TOC is a
measure of organic content in the water and is generally a good indicator of the concentrations of total
trihalomethanes (TTHM) and haloacetic acid (HAAS) precursors. The distribution of plant-mean TOC
concentrations for plants with surface water sources covers a large range (from 0 to 21.4 mg/L (Exhibit
3.6)); however, 90 percent of the plants had mean TOC concentrations below 5.3 mg/L. Exhibit 3.7
shows the distribution of plant-mean TOC concentrations for surface water and ground water plants for
the subset of plants shown in Exhibit 3.6. For ground water plants, 70 percent had mean TOC
concentrations below 1 mg/L; the highest values were close to those for surface water plants. A large
percentage of ICR ground water plants (approximately 25 percent) are located in Florida, where high
levels of TOC occur in ground water.
Bromide in source water can affect the amount and type of DBFs formed, shifting the distribution
of DBFs more to the brominated species. Also, bromide can react with ozone and chlorine dioxide to
form bromate, another byproduct of concern. As shown in Exhibit 3.6, most of the plant-mean bromide
levels are relatively low (the 90th percentiles were 0.122 and 0.190 mg/L for surface water and ground
water sources, respectively). Exhibit 3.8 shows the distribution of mean bromide concentrations for large
surface water and ground water plants.
There is no extensive data set similar to the ICR that provides comparable influent water quality
data for medium and small systems. Therefore, as noted in section 3.2, several alternative data sources
were used to characterize these systems and compare their water quality to the large systems. These data
sources include the ICR supplemental survey (ICRSS), NRWA data, AWWA WATER:\Stats data, and
data from individual states. The ICRSS is a survey meant to compliment the ICR data set. It is a survey
of raw source water quality and DBF concentrations from 40 random plants each from the small, medium,
and large size categories. The NRWA surveyed 117 random small plants nationwide and determined
treatment process, source water quality, and DBF concentrations. The WATER:\STATS database was
compiled by AWWA and contains source water quality, treatment processes, and DBF concentrations for
872 member plants of mostly medium and large size categories. The State data include DBF monitoring
data from 10 States representing 562 small surface water systems and 2,336 small ground water systems.
The Occurrence Document (USEPA 2005k) provides an overview of each alternative data set and
compares source water quality for medium and small systems. Exhibit 3.9 provides a summary of
influent water quality data from these sources for medium and small surface and ground water systems.
Appendices A and B provide additional detail regarding influent water quality data that are relevant to
compliance forecast analyses for surface and ground water systems, respectively.
3.5.2 Regional Differences in Water Quality
EPA evaluated ICR data for surface and ground water systems to determine if there were
differences in influent water quality among regions. Exhibits 3.10 and 3.11 show the range of average
TOC concentrations by State for surface and ground water systems, respectively, using ICR data. Exhibit
3.12 shows average TOC concentrations by State for ground water systems using Ground Water Supply
Survey (GWSS) data. Surface water systems did not exhibit any notable regional trends; however, ICR
data and GWSS data show that Florida has very high TOC concentrations compared to other States.
Florida also has the largest proportion of large ground water systems of all the States. The ICR Ground
Water Delphi Group estimated that, of the large and medium ground water plants that will need to make
changes to comply with the Stage 2 DBPR (which includes a requirement for the IDSE), more than 80
percent are in Florida (see Appendix B, Exhibit B.4 for compliance forecast data on ground water
systems).
Final Economic Analysis for the Stage 2 DBPR 3-26 December 2005
-------
Exhibit 3.7 Cumulative Distribution of TOC in Influent Water of Large System
ICR Plant-Mean Data
50%
40%
30%
20%
10%
» SW ICR TOC Plant-Means (N=307)
n GW ICR TOC Plant-Means (N=103)
0% J
8 10 12 14 16
Plant-Mean TOC (mg/L as C)
18
20
22
24
Source: ICR AUX1 database (USEPA2000h).
Final Economic Analysis for the Stage 2 DBPR
3-27
December 2005
-------
Exhibit 3.8 Cumulative Distribution of Bromide in Influent Water of Large System
ICR Plant-Mean Data
100% -r
90%
c
01
Ol
Q.
• SW ICR Bromide Plant-Means (N=320)
n GW ICR Bromide Plant-Means (N=118)
0.2
0.4
0.6 0.8
Plant-Mean Bromide (mg/L)
1.2
1.4
Source: Each data point in the distribution represents the mean value of monthly data collected at a single plant over
a 12-month period (January 1998-December 1998). Only plants with reported data for at least 9 of the 12 months are
included in this summary table (USEPA 2000h).
Final Economic Analysis for the Stage 2 DBPR
3-28
December 2005
-------
Exhibit 3.9 Medium and Small System Influent Water Quality Parameters-
Summary of Pre-Stage 1 Plant-Mean Data
Data Source/Size Category
N
Mean of
Plant-
Means
Median of
Plant-
Means
90th
Percentile of
Plant-Means
Range of
Plant-Means
Source Water Alkalinity (mg/L as CaCO3)
NRWA Small Surface Water
(SW) Plants
ICR Supplemental Survey (ICR
SS) Medium SW Plants
ICR SS Small SW Plants
95
40
38
81
82
66
74
74
55
146
159
123
0-281
4.8 - 240
4.4 - 249
Source Water Bromide (mg/L)
NRWA Small SW Plants
ICR SS Medium SW Plants
ICR SS Small SW Plants
95
40
38
0.063
0.050
0.02
0.021
0.016
0
0.107
0.092
0.044
0-1.72
0-0.53
0 - 0.27
Source Water pH
NRWA Small SW Plants
ICR SS Medium SW Plants
ICR SS Small SW Plants
78
40
36
7.3
7.6
7.3
7.5
7.6
7.4
8.1
8.2
8.0
3.8-8.8
5.9-8.4
5.8-8.3
Source Water TOC (mg/L as C)
NRWA Small SW Plants
ICR SS Medium SW Plants
ICR SS Small SW Plants
WATERASTATS Medium SW
Plants
WATERASTATS Medium GW
Plants
96
40
38
102
51
3.0
3.6
2.4
5.6
2.3
2.6
3.7
2.1
3.2
0.79
5.4
5.5
4.5
6.4
7.0
0.3-9.0
0.2-7.9
0.1 -7.1
0-200
0-25
Source Water Turbidity (NTU)
NRWA Small SW Plants
ICR SS Medium SW Plants
ICR SS Small SW Plants
76
40
36
7.8
13
6.2
4.1
5.9
3.5
18
33
13
0.1 -65
1 - 103
0.3-43
Source Water UV-254 (cm 1)
NRWA Small SW Plants
ICR SS Medium SW Plants
ICR SS Small SW Plants
96
40
38
0.082
0.093
0.074
0.074
0.083
0.051
0.127
0.171
0.113
0.01 -0.23
0.03-0.21
0.02 - 0.44
Note: ICR SS data are the plant-means for plants that took at least three-fourths of the total possible samples for
each parameter. Only plants that had both a Winter and Summer sample are included in the NRWA data for this
analysis.
Final Economic Analysis for the Stage 2 DBPR
3-29
December 2005
-------
Exhibit 3.10 Influent Water TOC Distribution for ICR Surface Water Systems
No Data
AJ TOC < 1 to mg/L
jj] TOC >= 1 to 2 mg/L
1 TOC >= 2 to 3 mg/L
ID TOC >= 3 to 4 mg/L
TOC >= 4 mg/L
Source: ICR AUX1 Database (USEPA 2000h); mean of all plant-means for each State.
Exhibit 3.11 Influent Water TOC Distribution for ICR Ground Water Systems
No Data
A] TOC < 1 to mg/L
B] TOC >= 1 to 2 mg/L
TOC >= 2 to 3 mg/L
TOC >= 3 to 4 mg/L
TOC >= 4 mg/L
Source: ICR AUX1 Database (USEPA 2000h); mean of all plant-means for each State.
Final Economic Analysis for the Stage 2 DBPR
3-30
December 2005
-------
Exhibit 3.12 Influent Water TOC Distribution for Ground Water Systems Derived
from the Ground Water Supply Survey
No Data
A] TOC < 1 to mg/L
i] TOC >= 1 to 2 mg/L
TOC >= 2 to 3 mg/L
nj TOC >= 3 to 4 mg/L
TOC >= 4 mg/L
Source: Ground Water Supply Survey (USEPA 1983)
3.6 Treatment Characterization for the Pre-Stage 1 Baseline
This section summarizes treatment conditions for the pre-Stage 1 baseline estimate of treatment
technologies-in-place for the baseline. Chapter 5 provides further detail on treatment technologies and
compliance forecast methodology used in this EA.
Although cost analyses in Chapter 7 are performed for each of the nine system size categories
separately, treatment characterizations are predicted according to the following aggregated categories by
population served:
• Small systems
• Serving fewer than 100 people
• Serving 100 to 999 people
• Serving 1,000 to 9,999 people
• Medium systems—serving 10,000 to 99,999 people
• Large systems—serving 100,000 or more people
Small systems were stratified by the three population categories shown above to represent
differences in the number of systems needing to change treatment technologies and the technology
Final Economic Analysis for the Stage 2 DBPR
3-31
December 2005
-------
options available to each category. The treatment characterizations presented here and in Chapter 7 show
the nine population size categories used for costing, but present one compliance forecast (as a percentage
of the total baseline number of plants) for each of the population size categories listed above.
Exhibits 3.13 and 3.14 summarize the pre-Stage 1 DBPR baseline treatment technologies in place
for surface and ground water treatment plants, respectively6. For plants in large and medium ground
water systems, ICR treatment data were used to derive the estimated percent of plants using each
treatment technology as no other model or data source exists to characterize treatment technologies. For
plants in large and medium surface water systems, SWAT-predicted results from the "initial plant run"
(USEPA 200 Ib) are used to characterize the percent of plants using each treatment technology in Exhibit
3.13. SWAT-predicted results were used instead of available ICR-observed data to allow for consistent
comparison of pre-Stage 1 data to modeled pre-Stage 2 and post-Stage 2 data. (If observed data were
used for pre-Stage 1 treatment technology-in-place estimates, differences between pre-Stage 1 and pre-
Stage 2 results would represent potential inconsistencies in observed vs. predicted data, not just the
expected treatment technology change from pre-Stage 1 to pre-Stage 2. The SWAT Model uses a subset
of the ICR plants, so while percentages are similar, they are not exact.) In addition, the SWAT initial
plant run is used to calculate DBF reductions from pre-Stage 1 to post-Stage 1 for use in the benefits
models.
For all small systems, the only significant use of advanced treatment technologies was reported in
the NRWA database for small surface water systems (approximately 3.6 percent are estimated to be using
microfiltration/ultrafiltration (MF/UF), as shown in Exhibit 3.14). The percent using each treatment
technology is based on evaluation of CWS data; EPA assumed that NTNCWSs use similar treatment
technologies for the size categories shown.
6As described in Appendix A, the treatment technologies used to characterize the pre-Stage land pre-Stage
2 baselines are different than those presented in the Stage 1 RIA. New tools to characterize treatment technologies-
in-place have been made available since the Stage 1 DBPR was promulgated, notably the SWAT Model.
Final Economic Analysis for the Stage 2 DBPR 3-32 December 2005
-------
Exhibit 3.13a Pre-Stage 1 DBPR Treatment Technologies-in-Place for CWS Surface Water Plants
System Size
(Population
Served)
<100
100-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
=1,000,000
Total Plants, %
(Population
Served)
<100
100-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
=1,000,000
Total Plants, %
No Advanced
Treatment
Technologies1 CL2
A
96.4% 346
96.4% 739
466
96.4% 1 ,089
1,213
53.4% 689
309
53.4% 326
39
79.6% 5,216
No Advanced
Treatment
Technologies1 CLM
B
0.0% 0
0.0% 0
0
0.0% 0
0
31 .6% 408
183
31.6% 193
23
12.3% 808
GAC10 + AD
CL2
M
0.0% 0
0
0.0% 0
0
0.0% 0
CLM
N
0.0% 0
0
0.0% 0
0
0.0% 0
Chlorine Dioxide
CL2
C
/
0.0% 0
0
0.0% 0
0
5.1% 65
29
5.1% 31
4
2.0% 129
CLM
D
0.0% 0
0.0% 0
0.0% 0
0.0% 0
3.0% 39
17
3.0% 18
2
1 .2% 77
GAC20
CL2
O
0.0% 0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
CLM
P
0.0% 0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
UV
CL2
E
CLM
F
GAC20 + AD
CL2
Q
0.0% 0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
CLM
R
0.0% 0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
Ozone
CL2
G
0.0% 0
0
0.0% 0
0
3.2% 42
19
3.2% 20
2
1 .3% 82
CLM
H
0.0% 0
0
0.0% 0
0
1 .9% 25
11
1.9% 12
1
0.7% 49
Membranes
CL2
S
0.0% 0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
CLM
T
0.0% 0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
MF/UF
CL2
I
3.6% 13
3.6% 28
17
3.6% 41
45
0.2% 3
1
0.2% 1
0
2.3% 150
CLM
J
0.0% 0
0.0% 0
0
0.0% 0
0
0.1% 2
1
0.1% 1
0
0.1% 3
GAC10
CL2
K
/
0.9% 12
5
0.9% 6
1
0.4% 24
CLM
L
0.5% 7
3
0.5% 3
0
0.2% 14
TOTAL
CL2
A+C+E+G+I+K+M+O+Q+S
100.0% 359
100.0% 767
483
100.0% 1,129
1,258
62.8% 811
364
62.8% 383
46
85.5% 5,601
CLM
B+D+F+H+J+L+N+P+R+T
0.0% 0
0.0% 0
0
0.0% 0
0
37.2% 481
216
37.2% 227
27
14.5% 951
Note: Detail may not add to totals due to independent rounding.
1"No Adv" includes conventional, non-conventional, and softening plants.
Source: Surface water systems serving <10,000 people: National Rural Water Survey (USEPA2001a). Surface water systems serving 10,000 or more people: SWAT initial plant run (USEPA2001b).
Percentage using chloramine is taken from the Occurrence Document (USEPA 2005k) and ICR AUX1 data (USEPA 2000h).
Final Economic Analysis for the Stage 2 DBPR Proposal
3-33
December 2005
-------
Exhibit 3.13b Pre-Stage 1 DBPR Treatment Technologies-in-Place for NTNCWS Surface Water Plants
System Size
(Population
Served)
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
=1,000,000
Total Plants, %
(Population
Served)
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
=1,000,000
Total Plants, %
No Advanced
Treatment
Technologies1 CL2
A
96.4% 218
96.4% 301
102
96.4% 89
24
53.4% 3
0
53.4% 1
0
96.1% 737
No Advanced
Treatment
Technologies1 CLM
B
0.0% 0
0.0% 0
0
0.0% 0
0
31 .6% 2
0
31.6% 0
0
0.2% 2
GAC10 + AD
CL2
M
0.0% 0
0
0.0% 0
0
0.0% 0
CLM
N
0.0% 0
0
0.0% 0
0
0.0% 0
Chlorine Dioxide
CL2
C
0.0% 0
0
0.0% 0
0
5.1% 0
0
5.1% 0
0
0.0% 0
CLM
D
0.0% 0
0.0% 0
0.0% 0
0.0% 0
3.0% 0
0
3.0% 0
0
0.0% 0
GAC20
CL2
O
0.0% 0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
CLM
P
0.0% 0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
uv
CL2
E
/
CLM
F
GAC20 + AD
CL2
Q
0.0% 0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
CLM
R
0.0% 0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
Ozone
CL2
G
0.0% 0
0
0.0% 0
0
3.2% 0
0
3.2% 0
0
0.0% 0
CLM
H
0.0% 0
0
0.0% 0
0
1 .9% 0
0
1.9% 0
0
0.0% 0
Membranes
CL2
S
0.0% 0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
CLM
T
0.0% 0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
MF/UF
CL2
I
3.6% 8
3.6% 1 1
4
3.6% 3
1
0.2% 0
0
0.2% 0
0
3.6% 27
CLM
J
0.0% 0
0.0% 0
0
0.0% 0
0
0.1% 0
0
0.1% 0
0
0.0% 0
GAC10
CL2
K
/
0.9% 0
0
0.9% 0
0
0.0% 0
CLM
L
0.5% 0
0
0.5% 0
0
0.0% 0
TOTAL
CL2
A+C+E+G+I+K+M+O+Q+S
100.0% 226
100.0% 312
106
100.0% 92
25
62.8% 3
0
62.8% 1
0
99.7% 765
CLM
B+D+F+H+J+L+N+P+R+T
0.0% 0
0.0% 0
0
0.0% 0
0
37.2% 2
0
37.2% 0
0
0.3% 2
Note: Detail may not add to totals due to independent rounding.
1"No Adv" includes conventional, non-conventional, and softening plants.
The NTNCWS technology distribution is assumed to be the same as the CWS technology distribution presented in Exhibit 3.13a.
Source: Surface water systems serving <10,000 people: National Rural Water Survey (USEPA 2001a). Surface water systems serving 10,000 or more people: SWAT initial plant run (USEPA 2001 b).
Percentage using chloramine is taken from the Occurrence Document (USEPA 2005k) and ICR AUX1 data (USEPA 2000h).
Final Economic Analysis for the Stage 2 DBPR Proposal
3-34
December 2005
-------
Exhibit 3.14a Pre-Stage 1 DBPR Treatment Technologies-in-Place for CWS Ground Water Plants
System Size
(Population Served)
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
71,000,000
Total Plants
No Adv1 CL2
A
100.0% 6,423
100.0% 15,242
6,093
100.0% 7,587
5,030
92.3% 4,968
661
92.3% 847
25
98.9% 46,878
No Adv1 CLM
B
0.0% 0
0.0% 0
0
0.0% 0
0
5.4% 290
39
5.4% 49
1
0.8% 379
UVCL2
C
0.0% 0
0.0% 0
0
0.0% 0
0
0.0% 0
UVCLM
D
0.0% 0
0.0% 0
0
0.0% 0
0
/
/
0.0% 0
Ozone CL2
E
0.0% 0
0.0% 0
0
0.0% 0
0
0.8% 41
6
0.8% 7
0
0.1% 54
Ozone CLM
F
0.0% 0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
GAC20 CL2
G
0.0% 0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
GAC20 CLM
H
0.0% 0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
Membranes CL2
I
0.0% 0
0.0% 0
0
0.0% 0
0
1.5% 83
11
1.5% 14
0
0.2% 108
Membranes
CLM
J
0.0% 0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
TOTAL USING CL2
K=B+D+F+H+J
100.0% 6,423
100.0% 15,242
6,093
100.0% 7,587
5,030
94.6% 5,093
677
94.6% 869
26
99.2% 47,040
TOTAL USING CLM
L = B+D+F+H+J
0.0% 0
0.0% 0
0
0.0% 0
0
5.4% 290
39
5.4% 49
1
0.8% 379
Note: Detail may not add to totals due to independent rounding
1"No Adv" includes conventional, non-conventional, and softening plants.
Source: Ground water systems serving <10,000 people - limited data available. Assumed only cholrine usage and no advanced technologies; Ground water systems serving 10,000 or more people - based on ICR data for 130 large GW
plants.
Exhibit 3.14b Pre-Stage 1 DBPR Treatment Technologies-in-Place for NTNCWS Ground Water Plants
System Size
(Population Served)
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
=1,000,000
Total Plants
No Adv1 CL2
A
100.0% 2,493
100.0% 2,129
589
100.0% 247
21
92.3% 3
0
92.3% 0
0
100.0% 5,483
No Adv1 CLM
B
0.0% 0
0.0% 0
0
0.0% 0
0
5.4% 0
0
5.4% 0
0
0.0% 0
UVCL2
C
0.0% 0
0.0% 0
0
0.0% 0
0
0.0% 0
UVCLM
D
0.0% 0
0.0% 0
0
0.0% 0
0
/
0.0% 0
Ozone CL2
E
0.0% 0
0.0% 0
0
0.0% 0
0
0.8% 0
0
0.8% 0
0
0.0% 0
Ozone CLM
F
0.0% 0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
GAC20 CL2
G
0.0% 0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
GAC20 CLM
H
0.0% 0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
Membranes CL2
1
0.0% 0
0.0% 0
0
0.0% 0
0
1.5% 0
0
1.5% 0
0
0.0% 0
Membranes
CLM
J
0.0% 0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
TOTAL USING CL2
K=B+D+F+H+J
100.0% 2,493
100.0% 2,129
589
100.0% 247
21
94.6% 3
0
94.6% 0
0
100.0% 5,483
TOTAL USING CLM
L = B+D+F+H+J
0.0% 0
0.0% 0
0
0.0% 0
0
5.4% 0
0
5.4% 0
0
0.0% 0
Note: Detail may not add to totals due to independent rounding
The NTNCWS technology distribution is assumed to be the same as the CWS technology distribution presented in Exhibit 3.14a.
1"No Adv" includes conventional, non-conventional, and softening plants.
Source: Ground water systems serving <10,000 people - limited data available. Assumed only cholrine usage and no advanced technologies; Ground water systems serving 10,000 or more people - based on ICR data for 130 large GW
plants.
Final Economic Analysis for the Stage 2 DBPR Proposal
3-35
December 2005
-------
3.7 DBF Occurrence for the Pre-Stage 1 Baseline
For pre-Stage 1 DBPR conditions, observed DBF data are available from the ICR for large
systems; and from the NRWA survey, WATER:\STATS, and other data sources for medium and small
systems (see section 3.2 for a summary of data sources). The pre-Stage 1 DBPR baseline DBF
occurrence is also predicted using SWAT. In Chapter 5, SWAT-modeled DBF occurrence is used instead
of observed (ICR) values so that any changes shown in the analysis would be a result of treatment
technology changes and operating conditions and not differences between observed and modeled data.
Section 3.7.1 provides background information describing the ICR and SWAT DBF data used in
this EA. Section 3.7.2 summarizes pre-Stage 1 DBF occurrence data for large surface water systems.
Pre-Stage 1 DBPR occurrence levels for large ground water systems are provided in section 3.7.3. DBP
occurrence for medium systems is discussed in section 3.7.4, followed by presentation of small system
data in section 3.7.5. Section 3.8 describes the uncertainties in observed and predicted DBP data.
3.7.1 Description of ICR and SWAT DBP Data
3.7.1.1 ICR DBP Data
The analysis of ICR DBP data in this EA is consistent with the methodology used in the
Occurrence Document (USEPA 2005k), Chapter 5, and Appendix A. A brief description of the data and
assumptions used in the analyses are provided below.
Sampling Period
Consistent with the influent water quality date summarized in Section 3.5, DBP data represent
that last 12 months (four quarters) of the ICR collection period (January - December, 1998). Data
collected appear to be of higher quality than data collected during the first 6 months of the survey. In
addition, 6 quarters were not used because 2 quarters would be counted twice and could skew results
Plant Source Water Type
DBP data for ground and surface water plants are analyzed in this section. Consistent with
influent water quality data described in Section 3.5, plant-source water type designation is based on the
source water type reported by the plant for each month from July 1997 to December 1998. The types of
sources recorded were surface water, ground water, mixed, or purchased. Most plants reported on one
source type for all months, but some plants reported surface water for some months and mixed for others.
These plants were considered surface water plants. One plant reported ground water for some months and
mixed for others—this plant was considered a ground water plant.
Distribution System Sampling Locations
Quarterly TTHM and HAA5 data were collected at the following distribution system sampling
locations (note that these locations are each associated with one ICR plant):
• Average 1 (AVG 1) and Average 2 (AVG 2)—two sample locations in the distribution system,
each representing an approximate average residence time, as designated by the water system.
• Distribution System Maximum (DS Maximum)—the sample location in the distribution
system that has the longest residence time, as designated by the water system.
Final Economic Analysis for the Stage 2 DBPR 3-36 December 2005
-------
• Distribution System Equivalent Location (DSE)—a sample location in the distribution system
that has a known residence time, where no additional disinfectant has been added between the
plant and sample location, and where there has been no blending with water from other
plants.
Initial Plant Screening
All ICR plants (there are approximately 500 plants in the ICR database) were screened to ensure
that at least 3 of 4 quarters have TTHM and HAAS data for at least 3 of 4 distribution system locations.
Note that the total number of plants that meet the minimum screening criteria (311 plants, of which 213
are surface water plants, 83 are ground water plants, and 15 are either blended, mixed, or purchased water
plants) represents more than 60 percent of all large plants that participated in the ICR data collection
effort.
The initial plant screening was done to minimize biases in RAA and LRAA calculations (e.g.,
LRAAs could be skewed if data from multiple quarters is missing). While the screening process is
intended to reduce biases in data analysis, EPA recognizes that biases in the RAA and LRAA
calculations may still exist. First, missing data points from locations may skew the quarterly average.
For example, a plant with less than 3 of 4 quarters of data for the maximum residence time location (but
having at least 3 of 4 quarters of data for all other locations, allowing it to be included in the analysis)
would probably have an RAA that is skewed low. Second, missing quarterly data could skew the yearly
average. For example, because higher DBF levels are typically seen during the warmest months, missing
data in the warmest quarter may lower the annual average at that location. The screening criteria
described above were selected to strike a balance between minimizing biases in RAA and LRAA
calculations and maximizing the number of plants evaluated.
3.7.1.2 SWAT DBF Data
SWAT produces monthly estimates of DBF occurrence for surface water plants at two
distribution system locations:
• Distribution System Average (DS Average)—theoretical location with average residence time
(calculated by averaging the residence times reported by the water system for the four
locations listed above).
• Distribution System Maximum (DS Maximum)—theoretical location with the maximum
residence time (highest residence time reported for the four locations above).
Data from these locations are compared and summarized in subsequent sections.
3.7.2 Pre-Stage 1 DBF Occurrence for Large Surface Water Plants
Exhibit 3.15 summarizes the TTHM, HAA5, bromate, and chlorite occurrence for pre-Stage 1
baseline conditions. Pre-Stage 1 occurrence is shown for both observed ICR data and SWAT-predicted
data. Exhibits 3.16 through 3.19 show the cumulative distributions of the same plant-mean and individual
observations (monthly DBP concentrations) for SWAT data. SWAT plant-mean data represent plant-
mean concentrations at the DS Average (average residence time) location. ICR plant-mean data represent
the average of four distribution system locations (AVG1, AVG2, DSE, and DS Maximum). Bromate and
Final Economic Analysis for the Stage 2 DBPR 3-3 7 December 2005
-------
chlorite data represent finished-water concentrations from both ICR and SWAT data sets, although ICR
chlorite data show the maximum finished-water concentration at each plant, rather than the plant-mean.
Statistical calculations of individual observations are for SWAT monthly data and ICR quarterly data.
Exhibit 3.15 reveals differences between SWAT-predicted and ICR observed data. Although the
predicted SWAT data is calibrated to the national averages reported in the ICR, differences still exist
because of differences in calculated versus actual residence times, uncertainty in the SWAT predictive
equations, uncertainty in variability in the sampling data, and the fact that SWAT essentially follows a
single slug of water through the distribution system, while the ICR measures instantaneous values of
different parcels of water. See section 3.8 for a discussion of uncertainty in each data set.
3.7.3 Pre-Stage 1 DBF Occurrence in Large Ground Water Plants
ICR data were the only source of pre-Stage 1 DBF data for large ground water systems. There
are limited or no data on bromate and chlorite, since these DBFs were monitored only by plants using
ozone or chlorine dioxide, and only one ground water plant in the ICR used these disinfectants (USEPA
2005k). TTHM and HAA5 data for these plants are summarized in Exhibit 3.20. DBF levels in ground
water are significantly less than in large surface water plants (see Exhibit 3.15); mean TTHM levels in
ground water plants are less than half those in large surface water plants, compared to observed or
modeled surface water data. Ground water data are more skewed than surface water data; there is a much
bigger difference between the median and the mean values for ground water.
Final Economic Analysis for the Stage 2 DBPR 3-38 December 2005
-------
Exhibit 3.15 Summary of Pre-Stage 1 DBP Occurrence for
Large Surface Water Plants, DS Average Data
Parameter
Plant-Mean Data
N
Mean
Median
90th
%ile
Range
Individual Observations
N
Mean
Median
90th
%ile
Range
TTHM (ug/L)
Pre-Stage 1
(ICR)
Pre-Stage 1
(SWAT)
213
273
42
49
40
42
70
9
0-117
3-207
3,083
2,784
42
49
37
39
78
100
0-177
0-356
HAAS (ug/L)
Pre-Stage 1
(ICR)
Pre-Stage 1
(SWAT)
213
273
29
36
24
30
52
71
0-116
1-146
3,083
2,784
29
36
23
27
55
77
0-188
0-294
Bromate (ug/L) (Ozone plants only)
Pre-Stage 1
(ICR)
Pre-Stage 1
(SWAT)
14
15
2.6
6.3
2.2
1.8
5.4
24.3
0.02-7.2
0.2-28
157
156
2.6
6.1
1.9
1.9
6.7
18
0-14.6
0.1-62
Chlorite (ug/L) (Chlorine dioxide plants only)
Pre-Stage 1
(ICR)
Pre-Stage 1
(SWAT)
18
22
429
819
465
812
701
1394
2.2-1,105
140-1,680
192
177
435
636
435
700
830
1330
0-1,719
42-1,680
Note: For TTHM and HAAS data, SWAT data are from the DS Average location and ICR data are the average of four
distribution system locations for the last 12 months of the ICR collection period (January 1998-December 1998). For
bromate and chlorite data, finished water data from both SWAT and ICR are used. All SWAT data are based on
monthly predicted observations, ICR TTHM and HAAS data are based on quarterly observations, and ICR chlorite
and bromate data are based on monthly observations. For ICR data, only individual observations used to calculate
plant means are shown. For ICR data, only plants that have data for 3 of the last 4 quarters were included, and, for
ICR TTHM and HAAS data, only plants with at least 3 of the 4 required distribution system samples each quarter were
included.
Sources: SWAT Initial Plant Run and Run 300 (USEPA2001b); ICR AUX1 Database (USEPA2000h), screened data.
Final Economic Analysis for the Stage 2 DBPR
3-39
December 2005
-------
Exhibit 3.16 Cumulative Distributions of TTHM Data Predicted by SWAT, Pre-
Stage 1 (DS Average)
100%
90%
80%
3> 70%
0.
>
3
i
o
60%
50%
40%
30%
20%
10%
100
150 200 250
Plant Mean TTHM (ug/L)
300
350
400
Note: DS Average data from SWAT
Source: SWAT Initial Plant Run (USEPA2001b).
Exhibit 3.17 Cumulative Distributions of HAAS Data Predicted by SWAT, Pre-
Stage 1 (DS Average)
100%
S> 70%
1
8 60%
s.
g 50%
• 'Plant Mean
"Monthly
40%
30%
20%
10%
0%
50
100
Plant Mean HAAS (ug/L)
150
200
Note: DS Average data from SWAT
Source: SWAT Initial Plant Run (USEPA2001b)
Final Economic Analysis for the Stage 2 DBPR
3-40
December 2005
-------
Exhibit 3.18 Cumulative Distributions of Bromate Data Predicted by SWAT, Pre-
Stage 1 (Finished Water)
20
Bromate (ug/L)
Note: Finished water data from SWAT for ozone plants only.
Source: SWAT Initial Plant Run (USEPA2001b).
Exhibit 3.19 Cumulative Distributions of Chlorite Data Predicted by SWAT
(Finished Water)
200
400
600
800 1000
Chlorite (ug/L)
1200
1400
1600
1800
Note: Finished water data from SWAT. Includes chlorine dioxide plants only.
Source: SWAT Initial Plant Run (USEPA2001b).
Final Economic Analysis for the Stage 2 DBPR
3-41
December 2005
-------
Exhibit 3.20 Summary of Pre-Stage 1 DBP Occurrence for Large Ground Water
Plants, ICR Data
Parameter
Plant-Mean Data
N
Mean
Median
90th
%ile
Range
Individual Observations
N
Mean
Median
90th
%ile
Range
TTHM (ug/L)
Pre-Stage 1
82
15.4
6.8
37
0-123
1,196
15.6
5.8
45
0-300
HAAS (ug/L)
Pre-Stage 1
82
8.5
2.2
22
0-71
1,196
8.8
1.5
26
0-124
Source: AUX1 database (USEPA 2000h),screened data.
3.7.4 Pre-Stage 1 DBP Occurrence for Medium Surface and Ground Water Plants
DBP occurrence data for medium ground water and surface water plants are limited. Plant-mean
data on TTHM are available from WATER:\STATS, a database compiled by the American Water Works
Association (AWWA 2000). Graphs of WATER:\STATS data in Appendix A show that DBP levels and
water quality parameter levels are similar in medium and large surface water plants. Graphs for ground
water in the Occurrence Document (USEPA 2005k) also show similarities between medium and large
ground water plants. Therefore, EPA assumed that DBP occurrence for medium surface water and
ground water plants is roughly equivalent to DBP occurrence for large surface water and ground water
plants, respectively.
3.7.5 Pre-Stage 1 DBP Occurrences for Small Surface and Ground Water Plants
The small-system experts used NRWA survey data and TTHM data submitted to EPA from eight
States to assess pre-Stage 1 DBP occurrence levels for small surface and ground water plants. Exhibit
3.21 summarizes the TTHM and HAAS data from these two data sets.
Although Exhibit 3.21 shows that TTHM levels from the State data set are higher than the levels
from the NRWA data set, NRWA data are considered more reliable and representative of national pre-
Stage 1 DBP occurrence than the State surface water data. (For further characterization of small surface
and ground water plant data sets, refer to Chapter 3 of the Occurrence Document (USEPA 2005k)).
Therefore, NRWA observed data were used to describe occurrence for small surface water plants.
Final Economic Analysis for the Stage 2 DBPR
3-42
December 2005
-------
Exhibit 3.21 Summary of Pre-Stage 1 DBP Occurrence Data for Small Systems,
DS Average Data
Parameter
(source)
Plant-Mean Data
N
Mean
Median
90th
%ile
Range
Individual Observations
N
Mean
Median
90th
%ile
Range
TTHM (ug/L)
Pre-Stage 1
(NRWA
survey, SW
systems)
Pre-Stage 1
(State data,
SW systems)
Pre-Stage 1
(State data,
GW systems)
96
562
2,336
83
99
17
62
65
3
179
215
46
0-328
0-687
0-655
384
N/A
N/A
85
N/A
N/A
64
N/A
N/A
169
N/A
N/A
0-451
N/A
N/A
HAAS (ug/l_)
Pre-Stage 1
(NRWA
survey, SW
systems)
96
45
34
84
0-262
384
44
34
90
0-475
Source: Pre-Stage 1 data: NRWA data (USEPA 2001 a) are weighted averages of data at locations having average and
maximum residence times in the distribution system. Average residence time data are weighted three times more than
maximum residence time data to make data equivalent to DS Averages calculated for ICR TTHM and HAAS. NRWA
plant-mean data include only those plants that had data for both sampling periods and for both distribution system
locations. Only those individual observations that were used to calculate plant-mean data are shown here. State data
(USEPA 2005k) are a mixed data set from eight States for surface water and seven States for ground water; N/A indicates
no individual observations were available for this data set.
3.8 Uncertainties in Development of the Pre-Stage 1 Baseline
There is uncertainty in this baseline analysis due to measurement error and incomplete
information that could result in either an over-estimate or under-estimate of the benefits and/or costs as
presented in Chapters 6 and 7. These uncertainties were not modeled as the impacts of these uncertainties
is unknown and EPA believes these uncertainties have less of an effect than those which are modeled in
Chapters 6 and 7.
Exhibit 3.22 presents key uncertainties and an estimate of the effects that each may have on
subsequent analyses. Note the effects on benefits and costs is unknown for most of the uncertainties
listed in Exhibit 3.22. A detailed discussion of each uncertainty follows the exhibit.
Final Economic Analysis for the Stage 2 DBPR
3-43
December 2005
-------
Exhibit 3.22 Summary of Uncertainties Affecting Stage 2 DBPR
Baseline Estimates
Uncertainty
Uncertainty in baseline
data inputs used to
generate the industry
baseline (SDWIS and
1 995 CWSS data)
CWS flow equations for
NTNCWSs
Uncertainty in use of
chloramines and
advanced treatment
technologies
Uncertainty in observed
data and predictive tools
used to characterize
DBP occurrence for the
pre-Stage 1 baseline
Section
Where
Estimates
are
Presented
3.4
3.4.3
3.6
3.7
Effect on Benefit Estimate
Under-
estimate
Over-
estimate
Unknown
Impact
X
No impact on benefits
X
X
Effect on Cost Estimates
Under-
estimate
Over-
estimate
X
Unknown
Impact
X
X
X
Uncertainty in the Industry Baseline
EPA recognizes that there is uncertainty related to the various data sources used to define the
system inventory for the Stage 2 DBPR. The uncertainty in the system inventory data inputs is not
quantified in this EA; however, a qualitative discussion of the identified uncertainties is provided below.
As noted above, SDWIS and the 1995 CWSS are the primary sources of system inventory data.
SDWIS is EPA's primary drinking water database, containing data for over 170,000 PWSs. SDWIS
stores State-reported information on each water system, including name, ID number, number of people
served, type of system (year-round or seasonal), and source of water (ground water or surface water),
along with monitoring and violation information. In 1998, EPA began a major effort to assess the quality
of its drinking water data in SDWIS. The results, published in Data Reliability Analysis of the EPA Safe
Drinking Water Information System/Federal Version, found that the quality of the required inventory data
was high (USEPA 2000e). Thus, EPA believes that uncertainty in the system inventory data from
SDWIS with respect to numbers of systems, source information, and size classification is low.
The 1995 CWSS was developed to gather data on water systems in the United States. A total of
3,681 systems covering a range of source water types and system sizes were statistically selected to
receive the main survey questionnaire. Of these, 1,980 systems responded. These responses were
weighted to maintain statistical representation of the total universe of CWSs. The EPA report Community
Water System Survey (USEPA 1997c) provides information on the 1995 CWSS survey design and data
evaluation.
Final Economic Analysis for the Stage 2 DBPR
3-44
December 2005
-------
The 1995 CWSS was the primary data source used to develop the following industry baseline
characteristics:
• Percent of ground water systems that disinfect
• Percent of SDWIS surface water systems that use primarily ground water
• Treatment plants per system
• Average and design flow based on population served (presented in section 3.4.3)
Because the CWSS is a survey of CWSs, estimates based on the data will contain uncertainty
because of sampling errors. To help define these uncertainties, the CWSS report provides the confidence
intervals on certain parameters. The report does not, however, contain data for percent disinfecting,
percent of SDWIS surface water systems providing ground water, and treatment plants per system that are
used in this EA. The confidence intervals for similar parameters can provide some information on
uncertainty. For example, an analysis of the percent of ground water systems with no treatment (which
uses similar data to the analysis of percent disinfecting), yielded 95 percent confidence intervals of less
than + 10 percent.
For average and design flow regression equations, one measure of uncertainty is the R-value for
the regressions. The regressions both for average daily flow and for design flow had very high R-values
(0.97 and 0.90, respectively), indicating a low level of uncertainty.
Uncertainties in Flow Equations for NTNCWs
As noted in Section 3.4.3, the relationship between population served and average daily flow and
design flow was not modeled for NTNCWSs. As a surrogate, the population-flow equations derived for
CWSs based on data collected during the 1995 CWSS were used. EPA recognizes that the CWS flow
equations likely overestimate average daily and design flows for NTNCWSs since NTNCWSs generally
only operate for part of the day and may not have high volume water uses such as showering or washing
clothes. This overestimation could lead to an overestimation in treatment technology costs and therefore,
national costs of the rule. The number of NTNCWSs relative to CWSs, however, is small, so EPA
anticipates that the impact of this overestimate will be minor.
Uncertainties in pre-Stage 1 Use of Advanced Treatment Technology
The estimated use of chloramines and advanced treatment technologies prior to the
implementation of the Stage 1 DBPR is based on different data sources depending on system size. For
large surface and ground water systems, use of treatment technologies is based on ICR data. EPA expects
that the treatment technology characterization for large systems has a relatively high degree of certainty
because the ICR database represents a census of all plants serving more than 100,000 people. For
medium and small systems, use of chloramines or advanced treatment technologies is based on data
gathered during the 1995 CWSS. As described earlier in this section, the CWSS is a statistically designed
survey and is expected to contain sample error. Thus, the estimated pre-Stage 1 treatment technologies in
place for medium and small systems are less certain than the estimated pre-Stage 1 treatment technologies
in place for large systems.
Uncertainties in ICR DBF Data
The are several sources of uncertainty in the DBP data collected under the ICR. The American
Waterworks Association Research Foundation (AWWARF) has compiled a thorough description of the
Final Economic Analysis for the Stage 2 DBPR 3-45 December 2005
-------
ICR data collection challenges and ultimate quality of the data in a publication, Information Collection
Rule Data Analysis (the AWWARF ICR Report) (McGuire et al. 2002). Data quality controls were
developed by a group of industry experts and strictly enforced; thus, EPA believes that the data quality in
the ICR database is very high.
One key area of uncertainty that is addressed in the AWWARF ICR Report relates to the
representativeness of all data collected during the ICR. The authors try to answer the question, how does
the water quality during the year of ICR data collection (1998) represent past years and adequately be
used to predict future DBF occurrence? Weather and rainfall during the ICR sample period were
compared to historical data to make this assessment.
On a nationwide basis, 1998 was hotter and wetter than normal. Approximately 75 percent of the
country experienced warmer than usual temperatures during 1998. Overall it was also a very wet year
with an average rainfall of 32.6 inches compared to the average of 27.2 inches. Twenty-two percent of
the country experienced wetter-than-normal conditions. Increased rainfall could bias the results,
increasing levels of constituents that derive mainly from runoff such as TOC. Other constituent could be
lower than normal such as bromide, which tends to rise during droughts. It should also be noted that
these trends are for the national data and that on a regional basis the trends may be different. For
example, even though 1998 was a very wet year on a national basis several mid-Atlantic states
experienced severe droughts during the summer of 1998. Chapter 3 of the AWWARF ICR Report
(McGuire et al. 2002) provides additional details for this assessment.
It is unknown how year-to-year variability in source water quality will affect estimated DBF
occurrence. The year of data collection (1998) could represent a worst-case, best-case, or typical year
depending on water-quality trends for a given plant. It is likely that some plants may experience higher
DBF occurrence in future years than what is represented in the ICR database.
Because of the nature of distribution system monitoring, the representativeness of a single grab
sample is uncertain. Pereira et al. (2004) showed that DBF levels can fluctuate even on a daily basis.
Over a 1 week sampling period where samples were taken every 6 hours that the coefficient of variability
ranged from 6 to 20 percent depending on the DBF measured and the sampling location. One grab
sample collected at a discreet point in time for the ICR does not represent this potential variability. In
addition to hourly variations, the ICR data were not required to be collected at evenly spaced intervals.
Thus, there is uncertainty in assuming that a single data point represents typical occurrence over the entire
quarter.
Based on comparisons of ICR and historical DBF databases, researchers suspect that plants
changed their treatment technology in anticipation of the Stage 1 DBPR prior to the ICR data collection
period (McGuire et al. 2002). The ICR then is not likely the true pre-Stage 1 baseline. EPA believes,
however, that because costs and benefits for the Stage 2 DBPR are based on treatment technology
changes from a predicted pre-Stage 2 baseline, the impact of this uncertainty is small.
Uncertainties in SWAT Predictions
Part II of Appendix A is dedicated to the discussion of uncertainties in the SWAT model. Major
areas of uncertainty in the SWAT predictions for the pre-Stage 1 baseline are (1) the uncertainty in ICR
observed data, upon which the SWAT model is based, (2) uncertainty in predictive equations for DBP
formation, (3) uncertainty in the SWAT compliance determination, and (4) uncertainty in SWAT
treatment technology selection. See Appendix A for a detailed discussion of uncertainties and
information on how the SWAT model was validated.
Final Economic Analysis for the Stage 2 DBPR 3-46 December 2005
-------
4. Consideration of Regulatory Alternatives
4.1 Introduction
To address the public health concerns presented in Chapter 2 and discussed in more detail in
Chapter 6, the Environmental Protection Agency (EPA) convened the Microbial-Disinfectants /
Disinfection Byproducts (M-DBYP) Advisory Committee under the Federal Advisory Committees Act
(FACA) to explore a number of regulatory alternatives for the Stage 2 Disinfectants and Disinfection
Byproducts Rule (DBPR). The M-DBP Advisory Committee was composed of representatives from the
following groups:
All Indian Pueblo Council, Pueblo Office of Environmental Protection
American Water Works Association
Association of Metropolitan Water Agencies
Association of State Drinking Water Administrators
Chlorine Chemistry Council
Clean Water Action
Conservation Law Foundation
Environmental Council of the States
International Ozone Association
National Association of County and City Health Officials
National Association of People with AIDS
National Association of Regulatory Utility Commissioners
National Association of State Utility Consumer Advocates
National Association of Water Companies
National Environmental Health Association
National League of Cities
National Resources Defense Council
National Rural Water Association
Physicians for Social Responsibility
Unfiltered Systems
U.S. Environmental Protection Agency
Water and Wastewater Equipment Manufacturers Association
M-DBP Advisory Committee deliberations, which began the Spring of 1999 and culminated in
December 2000 with an Agreement in Principle (USEPA 2000m), are documented on EPA's website in
the form of meeting summaries (USEPA 2000n).
This chapter describes the process for developing regulatory alternatives, then summarizes the
four Stage 2 DBPR regulatory alternatives considered in this Economic Analysis (EA). Among the four
alternatives is the Preferred Alternative, which represents the recommendation of the M-DBP Advisory
Committee.
Final Economic Analysis for the Stage 2 DBPR 4-1 December 2005
-------
4.2 Process for Development of Regulatory Alternatives
The process that led to the development of the Stage 2 DBPR began with the initiation of a
negotiated rulemaking by EPA in 1992 to address public health concerns related to disinfectants,
disinfection byproducts (DBFs), and microbial pathogens. The Regulatory Negotiation Committee met
from November 1992 through June 1993. The Committee included representatives of State and local
public health and regulatory agencies, public water systems, elected officials, consumer groups, and
environmental groups. As a result of its deliberations, the Committee recommended the development of
three sets of rules:
• A two-stage approach for regulations to control risks from DBFs.
• A two-stage approach for regulations to control risks from microbial contaminants (the
Interim Enhanced Surface Water Treatment Rule (IESWTR) and the Long Term Enhanced
Surface Water Treatment Rule).
• An information collection rule to support the above.
The Information Collection Rule (ICR) was promulgated in May 1996. The DBF and microbial
regulatory process recommendations of the Committee were subsequently incorporated into the 1996 Safe
Drinking Water Act (SDWA) Amendments as statutory requirements. The Stage 1 DBPR and the
IESWTR were both promulgated by EPA in December 1998.
Results from the 1996 ICR, which were gathered between July 1997 and December 1998,
provided the M-DBP Advisory Committee and EPA with a wealth of information on large water systems,
their treatment processes, and the quality of their source and finished waters. This information was used
to develop and run the Surface Water Analytical Tool (SWAT) (a model that uses a series of algorithms
and decision rules to predict treatment technology changes and DBP occurrence for regulatory
alternatives; see Appendix A). The output from SWAT formed much of the basis for estimates of
national cost and exposure to DBFs for the regulatory alternatives under consideration. Additional data
were obtained from a survey conducted by the National Rural Water Association (NRWA) of 120
smaller systems—serving fewer than 10,000 people—as well as from a variety of State data sources.
The M-DBP Advisory Committee considered several key questions during the negotiation
process, including:
• What health effects will the Stage 2 DBPR address?
• Should disinfectants and DBFs not regulated under the Stage 1 DBPR now be regulated?
• Should standards for disinfectants and DBFs set under the Stage 1 DBPR be amended?
• Should monitoring requirements under the Stage 1 DBPR be amended?
• Should compliance standards be calculated differently than those in the Stage 1 DBPR?
• What are the risk tradeoffs that need to be considered?
Final Economic Analysis for the Stage 2 DBPR 4-2 December 2005
-------
EPA used SWAT to develop rough estimates of costs and exposure reductions for over a
hundred possible rule alternatives. Of these, the M-DBP Advisory Committee focused its attention on
those alternatives that would reduce peaks in DBFs that may occur throughout the distribution system.
4.3 Regulatory Alternatives Considered
Four Stage 2 DBPR regulatory options are considered in this EA. They include what is referred
to as the Preferred Alternative, representing the recommendation of the M-DBP Advisory Committee,
and three other alternatives studied by the Committee, but not selected for reasons noted below. Though
not selected as a preferred option, the Committee considered each of those other three options as
alternatives worth careful consideration; EPA carried them through the benefits and cost calculations for
comparison with the Preferred Alternative. EPA chose the least-cost alternative that targets the highest
risks as the Preferred Alternative.
The goal of the M-DBP Advisory Committee was to increase the stringency of the total
trihalomethanes (TTHM) and haloacetic acids (HAAS) compliance standards by reducing peak
concentrations of DBFs in distribution systems. The Advisory Committee debated three different
compliance determination approaches. The first, a running annual average (RAA), bases compliance on
the average of all samples taken over a 12-month period, and allows certain monitoring locations to have
DBP levels higher than the maximum contaminant level (MCL) as long as the average does not exceed
the MCL. The second approach, a locational running annual average (LRAA), bases compliance on the
average of all samples taken at each specific monitoring location over a 12-month period, and requires the
average of all samples at individual monitoring locations to be no higher than the MCL. The third, a single
highest (SH) value, bases compliance on each individual sample meeting the MCL, and requires all
monitoring locations never to have DBP levels higher than the MCL. Compared to RAA compliance
options, the options involving LRAA and SH compliance measures focus on reducing peak
exposures—and the potential inequalities in exposure resulting from them—to customers served in some
parts of distribution systems. The LRAA has the added benefit of reducing average DBP exposures.
The following discussion provides details of the four main regulatory alternatives considered.
Preferred Alternative
The Stage 1 DBPR set MCLs for total TTHM at 80 ug/L and HAAS at 60 ug/L, each measured
as a RAA based on quarterly averages of all samples taken. The Stage 2 Preferred Alternative retains
these MCL values, but modifies how compliance is determined for TTHM and HAAS under the Stage 1
DBPR. Its components include:
• MCL of 80 micrograms per liter (ug/L) TTHM measured as an LRAA
MCL of 60 ug/L HAAS measured as an LRAA
• MCL of 10 ug/L bromate measured as an RAA, based on monthly samples taken at the
finished water point (no change from the Stage 1 DBPR)
• Compliance monitoring preceded by the Initial Distribution System Evaluation (IDSE)
Final Economic Analysis for the Stage 2 DBPR 4-3 December 2005
-------
Under the Preferred Alternative for the Stage 2 DBPR, systems are required to identify the
compliance monitoring sites that best represent high TTHM and HAAS levels through the IDSE and to
monitor at these locations. The purpose of this alternative is to control the levels of TTHM and HAAS at
locations in the distribution system with the highest levels of these DBFs as determined from the IDSE in
order to meet the MCLs throughout the entire distribution system.
Alternative 1
This alternative has the same TTHM and HAAS requirements as the Preferred Alternative, but
has a lower MCL for bromate. Its components include:
• MCL of 80 ug/L TTHM measured as an LRAA
• MCL of 60 ug/L HAAS measured as an LRAA
• MCL of 5 ug/L bromate measured as an RAA, based on monthly samples taken at the
finished water point
Members of the M-DBP Advisory Committee did not favor this alternative because they were
concerned that lowering the bromate level to 5 ug/L could have adverse effects on microbial protection.
In addition, the more stringent bromate standard led to estimated costs that are approximately three times
those for the Preferred Alternative. Alternative 1 would probably cause some systems to stop using
ozone or not consider ozone for microbial protection—developments that the M-DBP Advisory
Committee and EPA did not want to encourage because ozone is more effective than chlorine against
Cryptosporidium (Clark et al. 1994).
Alternative 2
This alternative measures TTHM and HAAS concentrations as a SH value for the MCL and
maintains the MCL for bromate. Its components include:
• MCL of 80 ug/L TTHM measured as the SH value for any sample taken
• MCL of 60 ug/L HAAS measured as the SH value for any sample taken
• MCL of 10 ug/L bromate measured as an RAA (no change from the Stage 1 DBPR)
This alternative is more stringent than the Preferred Alternative and Alternative 1. Under
Alternative 2, no TTHM or HAAS sample can exceed the MCL. EPA has estimated that a large portion
of the surface water systems covered by the rule would have to switch from their current treatment
practice to more expensive advanced treatment technologies to comply with this alternative. The M-DBP
Advisory Committee did not favor this alternative because it believed that the health effects data are not
certain enough to warrant such a drastic shift in the Nation's drinking water treatment practices. In
addition, Alternative 2 does not include the risk targeting strategy of the Preferred Alternative.
Final Economic Analysis for the Stage 2 DBPR 4-4 December 2005
-------
Alternative 3
This alternative reduces MCLs for TTHM and HAAS and maintains the MCL for bromate,
but does not modify how compliance is measured under the Stage 1 DBPR. Its components include:
• MCL of 40 ug/L TTHM measured as an RAA
• MCL of 30 ug/L HAAS measured as an RAA
• MCL of 10 ug/L bromate measured as an RAA (no change from Stage 1 DBPR MCL)
This alternative reduces the average level of TTHM and HAAS in the distribution system, but
does not necessarily reduce peaks in the distribution system as an LRAA compliance strategy is expected
to do. As with Alternative 2, a large portion of the surface water systems covered by the rule would have
to switch from their current treatment practices to expensive advanced treatment technologies to comply
with this alternative. Similarly, the M-DBP Advisory Committee did not favor this alternative because it
believed that the health effects data are not certain enough to warrant such a drastic shift in the Nation's
drinking water treatment practices and because it does not include the risk targeting strategy of the
Preferred Alternative.
Exhibit 4.1 shows how compliance would be determined for the Stage 1 DBPR and for each of
the Stage 2 regulatory alternatives described above when applied to a hypothetical large surface water
system. This hypothetical system has one treatment plant and measures TTHM in the distribution system
in four locations per quarter (the calculations shown would be the same for HAAS). Note that the
measured concentrations of TTHM and HAAS are the same in all cases. In this example, the system is
in compliance with the Stage 1 DBPR, but would be in violation of all four Stage 2 DBPR regulatory
options.
In addition to the four regulatory options addressed above, the M-DBP Advisory Committee
considered other changes in the approach to regulating DBFs to be worth noting. Both the current Stage
1 DBPR and the Stage 2 DBPR alternatives considered in this analysis use TTHM and HAAS as the
specific chlorination DBFs measured for compliance. The M-DBP Advisory Committee determined that
TTHM and HAAS are reliable indicators for all halogenated DBFs that exist in chlorinated drinking
water, including known DBFs that are unmeasurable and others that have yet to be identified. The
Committee considered having EPA modify the group of indicators to include a total of six haloacetic acids
(HAA6), a total of nine haloacetic acids (HAA9), or other chlorination DBFs. However, the M-DBP
advisory committee did not recommend that EPA expand the DBP indicators to include HAA6 or HAA9.
Fewer plants measured for HAA9 under the ICR because of analytical method problems for detecting
HAA9 (USEPA 1999c).
The Stage 1 DBPR also set a bromate MCL at 10 ug/L measured as an RAA, based on monthly
measurements for ozone systems and a chlorite MCL at 1.0 mg/L based on measurements required for
chlorine dioxide systems. The M-DBP Advisory Committee debated whether the bromate MCL should
be lowered (Regulatory Alternative 1). The Stage 1 DBPR set the MCL for bromate at 10 ug/L partly
because that was the limit of EPA's analytical capability at that time. New methods now exist to
measure lower concentrations of bromate, which would allow a lower limit to be set. However, the
Committee was concerned that a lower bromate MCL might discourage systems from switching to (or
continuing to use) ozone to increase microbial protection. Unlike chlorine, ozone is effective in the
Final Economic Analysis for the Stage 2 DBPR 4-5 December 2005
-------
disinfection of Cryptosporidium—a focus of the Long Term 2 Enhanced Surface Water Treatment Rule
(LT2ESWTR). Therefore, the M-DBP Advisory Committee recommended that EPA not change the
bromate MCL. The M-DBP Advisory Committee did not discuss the chlorite standard, and EPA does
not believe it needs to be revised.
Last, it should be noted that reductions in exposure to DBFs could also be achieved through
treatment techniques in lieu of or in addition to setting MCLs. For example, reducing organic precursor
compounds, measured as Total Organic Carbon (TOC), by such means as enhanced coagulation has been
shown to lower DBP formation. The M-DBP Advisory Committee considered regulatory alternatives for
reducing precursors and determined that the Stage 1 DBPR reduced TOC to a sufficient degree. Further,
removing ever-smaller quantities of these compounds will be more difficult, less efficient, and increasingly
costly. While analysis of ICR data shows that some systems could improve performance in this way (and
SWAT incorporated that into its decision tree), a regulatory requirement for reducing TOC levels was
deemed unnecessary.
Final Economic Analysis for the Stage 2 DBPR 4-6 December 2005
-------
Exhibit 4.1 Comparison of Hypothetical Compliance Calculations for Stage 1 and
Stage 2 Regulatory Alternatives
r TBasis of Compliance
I llviolation of MCL
Stage
TTHM
Wo exc
Q1
Q2
Q3 '
Q4
1 DBPR
MCL = 80ug/
eedance of /
Loc. 1
100
75
,55
60
L measured
\ncL
Loc. 2
40
50
45
55
as an RAA
Loc. 3
50
40
55
40
Loc. 4
50
100
110 •'
75
RAA
Qtrly Avq.
60 :
66
66
58
r 63
Preferred Stage 2 DBPR Alternative and Alternative 1*
TTHM MCL = 80 ug/L measured as an LRAA
LRAA at Location 4 exceeds MCL
Q1
Q2
Q3
Q4
LRAA
LQG..11
/too
75
55
60
Loc. 21
40
50
45
55
LQC, 31
,/§G
40
y55
40
73 j_ 48 |_ 46
Loc. 41
50
100
110
75
84
The Preferred Alternative and Alternative 1 have the same TTHM MCL;
they differ only in regard to the bromate MCL.
Footnote 1: Based on the IDSE, new locations targeted for high DBPs.
Alternative 2
TTHM MCL = 80 ug/L measured as a single highest value
Three samples at Locations 1 and 4 exceed MCL
Q1
Loc. 1
100
Loc._2_
40
J_oc_3_
50
75 J _50 | 40_ _ _
_ ^5_ _ I _ _45_ _l _55_ _ _
"" ~ o" ~ r ~ ~s~ ~n ~4o"""'
Loc_4_J
50
100
110
75
Alternative 3
TTHM MCL = 40 ug/L measured as an RAA
RAA exceeds MCL
Q1
Q2
Q3 .
Q4
Loc. 1
100
75
,55
60
Loc. 2
40
50
45 .
55
Loc. 3
50
40
,55
40
Loc. 4
50
100
110 ,
75
RAA
Qtrlv Ava
60
66
66 .-'
58
63
Final Economic Analysis for the Stage 2 DBPR
4-7
December 2005
-------
5. Compliance Forecast and Consequent Reduction in Chlorination DBFs
5.1 Introduction
The compliance forecast represents the changes systems are predicted to make in treatment
technologies to comply with a new drinking water regulation. Treatment technology changes to meet the
Stage 2 Disinfectants and Disinfection Byproducts Rule (DBPR) result in costs incurred by the water
systems as well as reductions in concentrations of disinfection byproducts (DBFs) that determine the
benefits achieved by the rule.
Section 5.2 provides an overview of the methodologies used to develop the compliance forecasts
and consequent reduction in the levels of DBFs. Section 5.3 provides details on the derivation of the
compliance forecast, and Section 5.4 presents the forecast results for the Stage 1 DBPR and the Stage 2
DBPR Preferred Alternative (forecasts for the other regulatory alternatives are presented in Appendix C).
Predicted reductions in the levels of two key classes of DBFs, total trihalomethane (TTHM) and five
haloacetic acids (HAA5), are discussed in Sections 5.5 and 5.6. Uncertainties in the compliance forecast
and DBP reduction estimates are summarized in Section 5.7.
In support of this chapter:
• Appendices A and B explain the derivation of the compliance forecasts (i.e., the number of
plants making treatment technology changes and which treatment technologies they select)
for surface water and disinfecting ground water systems, respectively.
• Appendix C provides supplemental compliance forecasts for the Stage 1 DBPR and Stage 2
DBPR regulatory alternatives.
Note that "compliance forecast" and "technology selection forecast" are used interchangeably in
this Economic Analysis (EA) to denote the overall percentage of plants predicted to change treatment
technology, along with the specific technologies those plants are predicted to select to achieve compliance
with the Stage 1 or Stage 2 DBPRs.
5.2 Overview of Methodologies used in the Primary Analysis
Changes in concentrations of DBFs are the direct result of changes in treatment technologies.
Therefore, it is important that EPA use consistent methodologies for forecasting treatment technology
changes and predicting reductions in DBFs (specifically, the reductions in TTHM and HAA5
concentrations that are used in the benefits analysis). This section summarizes the tools used and key
assumptions for the Stage 2 DBPR compliance forecasts and the consequent reductions in TTHM and
HAA5 concentrations.
Since the rule was proposed, EPA has modified the compliance forecast methodology in an
attempt to quantify uncertainties in the analysis. Specifically, EPA has developed a second method to
predict the number of surface water plants making treatment technology changes and consequent
reductions in TTHM and HAA5 concentration, which supplements the Surface Water Analytical Tool
(SWAT) predictions. EPA has also quantified uncertainty in the potential impacts of the Initial
Final Economic Analysis for the Stage 2 DBPR 5-1 December 2005
-------
Distribution System Evaluation (IDSE). Uncertainties are characterized using Monte Carlo simulation in
the cost and benefits models.
Predictive Tools Used to Develop the Compliance Forecast
EPA uses different methods for different system sizes and source water types to develop the
compliance forecasts, as shown in Exhibit 5.1. Because extensive data were available from the
Information Collection Request (ICR), detailed analysis tools were used to develop compliance forecasts
for large surface and ground water systems. For large surface water systems, EPA used two different
methodologies, both drawing from ICR data: the Surface Water Analytical Tool (SWAT) and the ICR
Matrix Method. The ICR Matrix Method uses TTHM and HAAS distribution system data from the ICR
to predict how many plants will need to change technology for a specific regulatory alternative. SWAT
uses a series of decision rules and algorithms to predict (1) which surface water plants need to change
treatment technology to meet a specific regulatory alternative, (2) which treatment technology those
plants will select based on a least cost decision tree. The ICR Matrix Method and SWAT produce
different results; thus, both are incorporated into a Monte Carlo simulation model to account for
uncertainties in both methods. The forecast for large ground water systems was generated using the ICR
Ground Water Delphi process, which convened a group of experts to evaluate plant configurations and
predict treatment technology selections for ground water plants that did not meet rule requirements.
Compliance forecasts for large surface and ground water systems were used to generate forecasts for
medium and small systems, making adjustments to account for operational and water quality
characteristics in medium and small systems that differ from those in large systems.
Exhibit 5.1 Tools Used to Develop the Stage 2 DBPR Compliance Forecasts
System Size
(Population Served)
Large (> 100,000 people)
Medium (10,000 to 99,999
people)
Small (<1 0,000 people)
Source Water Category
Surface Water
The Surface
Water Analytical
Tool (SWAT)
(Appendix A)
Extrapolation from
SWAT
(Appendix A)
Extrapolation from
SWAT, adjusted
to deal with small
system-specific
issues
(Appendix A)
ICR Matrix
Method
(Section 5.5)
Extrapolation
from ICR
Matrix Method
(Section 5.5)
Extrapolation
from ICR
Matrix Method
(Section 5.5)
Disinfecting Ground Water
ICR Ground Water Delphi
Group
(Appendix B)
Extrapolation from large
ground water system results
(Appendix B)
Extrapolation from large
ground water system results,
adjusted to deal with small
system-specific issues
(Appendix B)
Final Economic Analysis for the Stage 2 DBPR
5-2
December 2005
-------
Tools Used to Predict Changes in DBF Levels
For the benefits analysis, EPA needs information on the changes in both average and peak
TTHM and HAAS levels that result from implementation of the Stage 2 DBPR. Estimates of bladder
cancer cases avoided are based on reductions in average levels, while the illustrative analysis of potential
developmental and reproductive health benefits is based on reductions in occurrences of peak
concentrations.
To predict changes in average TTHM and HAAS levels for surface water systems, EPA uses two
methods: the ICR Matrix Method and SWAT. As noted in the previous section, the ICR Matrix Method
evaluates distribution system data to identify plants that would need to make treatment technology
changes to meet a specific regulatory alternative. To predict average DBP concentrations occurring after
treatment technology changes, EPA used TTHM and HAAS occurrence data for those surface water
plants already using chloramines and/or advanced technologies at the time of the ICR. The predicted
average TTHM and HAAS levels for all surface water plants is a weighted average for plants that do and
do not change treatment technology.
SWAT is a model that uses a series of decision rules and algorithms to predict (1) which surface
water plants need to change treatment technology to meet a specific regulatory alternative, (2) which
treatment technology those plants will select based on a least cost decision tree, and (3) resulting changes
in the national average TTHM and HAAS levels in distribution systems. As with the compliance forecast,
SWAT and the ICR Matrix Method produce different results; thus, both are incorporated into a Monte
Carlo simulation model to account for uncertainties in both methods.
ICR ground water plant data were not robust enough to develop a ground water model similar to
SWAT; therefore, the ICR Matrix Method is the only approach used to predict reductions in average
TTHM and HAAS levels for these systems.
To predict changes in the occurrence of peak TTHM and HAAS concentrations, EPA used only
the ICR Matrix Method1. Similar to the way in which it is used to evaluate changes in average TTHM
and HAAS concentrations, the ICR Matrix Method evaluates distribution system data to identify plants
that would need to make treatment technology changes to meet a specific regulatory alternative. To
predict occurrence of peaks after treatment technology changes, EPA analyzed TTHM and HAAS
occurrence data for those surface water plants already using chloramines and/or advanced technologies at
the time of the ICR. The predicted occurrence of peaks for all plants is a weighted average for plants that
do and do not make treatment technology changes.
Accounting for the Stage 1 DBPR
For cost and benefit analyses, the compliance forecast and consequent reduction in DBFs needs
to represent treatment technology changes from the pre-Stage 2 baseline (i.e., after implementation of the
Stage 1 DBPR). The best data available to characterize large plants are from the ICR, which were
1 Although the SWAT model was calibrated to national average TTHM and HAAS concentrations in
distribution systems and validated against industry treatment technology predictions, it was not calibrated to plant-
level DBP predictions and, thus, could not be used to assess changes in occurrence of peak levels. See Appendix A
for more information on SWAT.
Final Economic Analysis for the Stage 2 DBPR 5-3 December 2005
-------
collected before the Stage 1 DBPR compliance deadlines and likely represent pre-Stage 1 conditions2.
The compliance forecast, therefore, needs to account for treatment technology changes as a result of the
Stage 1 DBPR before predicting changes that are needed for the Stage 2 DBPR. Similarly, the post-Stage
2 TTHM and HAAS predictions need to take into account changes as a result of the Stage 1 DBPR before
predicting reductions that will occur as a result of the Stage 2 DBPR.
EPA uses a "delta" compliance forecast method that was developed by the Microbial /
Disinfection Byproducts (M-DBP) Technical Working Group (TWO). The method has four steps. First,
EPA characterizes treatment technologies and TTHM and HAAS occurrence for the Pre-Stage 1 baseline.
Second, EPA predicts treatment technology changes and subsequent reduction in TTHM and HAAS
levels from Pre-Stage 1 baseline to post-Stage 1 DBPR conditions. Third, treatment technology changes
and subsequent reductions in TTHM and HAAS levels are predicted from pre-Stage 1 baseline to post-
Stage 2 DBPR conditions. Lastly, results from Step 2 are subtracted from Step 3 to calculate the
incremental treatment technology change and TTHM/HAA5 reduction from post-Stage 1 DBPR to post-
Stage 2 DBPR conditions.
This delta method was selected by the M-DBP TWG over a more direct, two step approach (i.e.,
predict pre-Stage 2 conditions and then use the pre-Stage 2 conditions to predict impacts for Stage 2)
because modeling tools are not able to predict the treatment technology selection or TTHM, HAAS,
bromate, and chlorite levels at the plant level. The delta approach allows for potential errors in treatment
technology selection and TTHM and HAAS levels to be cancelled out for national level estimates. The
TWG believed that using the delta approach reduces the impact of uncertainty in SWAT predictive
equations for TTHM and HAAS. The delta approach is used with both the ICR Matrix Method and
SWAT analyses.
Accounting for the IDSE
Because the purpose of the IDSE is to identify Stage 2 compliance monitoring locations with high
DBP levels, it is possible that systems may measure higher DBP levels at Stage 2 compliance monitoring
sites than were measured under the ICR. This suggests that the number of plants predicted to make
treatment technology changes, the level of treatment they select, and the resulting reductions in TTHM
and HAAS levels, all based on ICR data, could be underestimated.
There is uncertainty in using the ICR data to represent pre-Stage 1 conditions because some plants may
have begun making changes prior to the ICR in anticipation of the Stage 1 DBPR (McGuire et al., 2002). See
Section 3.8 for a full discussion of uncertainties in ICR data.
Final Economic Analysis for the Stage 2 DBPR 5-4 December 2005
-------
The M-DBP TWG recommended that the Stage 1 and Stage 2 compliance forecast methodology
incorporate an operational safety margin of 20 percent to represent the operational level (i.e., 80 percent
of the MCL) at which systems typically take some action to ensure consistent compliance with a new
drinking water standard and the level at which systems target new treatment technologies to meet the
standard. EPA believes that this safety margin already accounts for the impacts of the IDSE for some
systems, including small systems, ground water systems, and those using chloramines3. EPA believes,
however, that the 20 percent safety margin is not sufficient to account for the potential impacts of the
IDSE on large and medium surface water systems because spatial variability of DBP levels and
distribution system complexity are greatest in these systems. Since the proposal, EPA developed a
methodology that analyzed ICR data from surface water plants to assess the extent of spatial variability of
TTHM and HAAS levels and used this as a basis for quantifying the impacts of the IDSE for large and
medium surface water systems.
5.3 Compliance Forecast Methodology
This section summarizes the tools used for the compliance forecast and provides details on how
EPA accounted for the Stage 1 DBPR and potential impacts of the IDSE on the compliance forecast.
Section 5.4 presents the results of the compliance forecast.
5.3.1 Tools for Surface and Ground Water Systems
EPA uses several tools to predict changes in treatment technology that will result from the Stage
2 DBPR. For surface water systems, EPA used results from both SWAT and the ICR Matrix Method.
For ground water systems, predictions were made using the ICR Ground Water Delphi Process. Exhibit
5.1 summarizes the tools used to develop the compliance forecasts. Detailed information on these
methodologies can be found in the referenced sections and appendices.
Surface Water Systems
The two tools used to predict changes in treatment technology and the resulting reductions in
DBP levels for surface water systems are SWAT and the ICR Matrix Method. SWAT is a modeling tool
developed by EPA during the M-DBP Federal Advisory Committee Act (FACA) process to evaluate
regulatory alternatives. SWAT uses source water and treatment data from the ICR along with a series of
empirical equations developed by researchers to model TTHM and HAA5 levels in distribution systems.
For each plant, SWAT predicts TTHM, HAA5, chlorite, and bromate levels and compares them to MCLs
for a given regulatory alternative. If the plant does not meet the MCLs, SWAT modifies the plant's
treatment until it can achieve compliance. SWAT uses a decision tree, arranged from lowest-cost to the
highest-cost treatment technology, to determine which new technology is selected for the plant. A total of
3EPA believes that the 20 percent safety margin accounts for potential impacts of the IDSE for small
systems because their distributions are not as complex when compared to large systems. EPA also believes that the
safety margin accounts for the IDSE for ground water systems because the year-to-year variability in source water
quality (and thus, TTHM and HAAS formation) is low. Chloramine systems generally observe lower spatial and
temporal variability in TTHM and HAAS distribution system levels (USEPA 2005k); thus, EPA believes the 20
percent safety margin accounts for potential impacts of the IDSE for these systems.
Final Economic Analysis for the Stage 2 DBPR 5-5 December 2005
-------
273 of the 350 ICR surface water plants had sufficient data to allow for modeling in SWAT. (For
characterization of the 273 modeled plants, see Appendix A.)
The ICR Matrix Method evaluates TTHM and HAAS distribution system data from the ICR to
identify plants that would need to make treatment technology changes to meet specific regulatory
alternatives. ICR plants are first screened to ensure that there are enough TTHM and HAAS distribution
system data so as not to skew the analysis. (See Chapter 3 for a discussion of the screening process,
including a discussion of data representativeness.) The method then places the screened plants into "bins"
based on their running annual average (RAA) and locational running annual average (LRAA) TTHM and
HAAS concentrations. Plants in bins that are non-compliant with Stage 1 are moved into compliant bins4.
The remaining plants that are non-compliant with Stage 2 regulatory alternatives are moved into
compliant bins. The ICR Matrix Method is limited in that it does not predict the specific technologies
plants will select; this information is only available from SWAT.
Results from both SWAT and the ICR Matrix Method are used in the primary analysis for all
regulatory alternatives. Section 5.3.6 compares the results and explains how they are incorporated into
the cost and benefits model.
SWAT and ICR Matrix Method compliance forecast results for large surface water systems were
applied directly to medium surface water systems, as the two size categories share similar source types
and operational capabilities. Adjustments were made to the forecasts for small surface water systems to
account for differences in water quality and operational constraints.
Ground Water Systems
The compliance forecasts for large ground water systems were generated using the ICR Ground
Water Delphi Process, which convened a group of experts to determine the treatment technology changes
that would be needed by systems that were not in compliance with the Stage 1 and Stage 2 DBPR. The
results were stratified based on plant location (Florida or Non-Florida) and extrapolated to national levels.
Because large and medium ground water systems are similar with respect to treatment configurations and
the well fields from which they draw, the compliance forecasts for large ground water systems were also
used for medium ground water systems (the ratio of Florida to Non-Florida ground water systems was
considered similar enough to allow for this extrapolation; see Appendix B for more details). The
compliance forecast for small ground water plants was based on results of the ICR Ground Water Delphi
process, but was adjusted to account for differences in total organic carbon (TOC), softening use, and the
ratio of Florida to Non-Florida ground water systems.
5.3.2 Accounting for the Stage 1 DBPR
For cost and benefit analyses, the compliance forecast needs to represent treatment technology
changes from the pre-Stage 2 baseline (i.e., after implementation of the Stage 1 DBPR). The best data
available to characterize large plants are from the ICR, which were collected before the Stage 1 DBPR
4As will be explained in Sections 5.5 and 5.6, revised TTHM and HAA5 levels for these plants are based on
an analysis of ICR data for plants that are Stage 2 compliant and already used chloramines and/or advanced
technologies prior to the ICR.
Final Economic Analysis for the Stage 2 DBPR 5-6 December 2005
-------
compliance deadlines and likely represent pre-Stage 1 conditions5. The compliance forecast, therefore,
needs to account for treatment technology changes as a result of the Stage 1 DBPR before predicting
changes that are needed for the Stage 2 DBPR.
The Stage 1 Regulatory Impact Analysis (RIA) (USEPA 1998a) includes a prediction of
chloramine and advanced technology use that will result from the Stage 1 DBPR. EPA did not use this
prediction as the post-Stage 1 baseline in this EA because new data and tools became available since the
Stage 1 RIA was developed (namely ICR and SWAT) that provide better characterization of plants and
allow for better prediction of treatment technology changes.
For surface water systems, a straightforward, 2-step approach for generating the Stage 2 DBPR
compliance forecast was originally considered during the M-DBP FACA deliberations. Under this
approach, SWAT would be used to predict treatment technology use and TTHM or HAAS levels for each
plant for post-Stage 1 conditions (i.e., after plants make changes to comply with Stage 1). Then SWAT
would evaluate post-Stage 1 conditions for each plant to predict treatment technology changes needed to
comply with the Stage 2 DBPR. In other words, SWAT would assess each plant for compliance with
Stage 1, make a treatment technology change if needed to comply with Stage 1, then evaluate post-Stage
1 TTHM and HAAS levels to determine if further changes are needed for Stage 2.
The M-DBP TWO identified a problem with this approach. SWAT predictive equations for
TTHM and HAAS were calibrated by comparing the ICR-observed values to the SWAT-predicted values
for the 273 plants used by the SWAT model. While the national average TTHM and HAAS levels
predicted by SWAT and observed by the ICR had good agreement, differences between SWAT and ICR
data were large for some plants. Thus, although national predictions were considered dependable,
SWAT-predicted TTHM and HAAS data for individual plants were uncertain. Because of this
uncertainty, the M-DBP TWO decided that it was inappropriate to evaluate SWAT-predicted post-Stage 1
TTHM and HAAS occurrence for each plant separately to assess compliance with the Stage 2 DBPR.
To minimize the impacts of uncertainty in SWAT plant-level predictions on the Stage 2 DBPR
compliance forecast, the M-DBP FACA developed a four-step, or "delta," approach for SWAT:
Step 1: Model TTHM and HAAS occurrence for the pre-Stage 1 baseline conditions (for
SWAT, predict using the model; for the ICR Matrix Method, use observed data).
• Step 2: Predict treatment technology changes from the pre-Stage 1 DBPR baseline to post-
Stage 1 DBPR conditions.
Step 3: Predict treatment technology changes from the pre-Stage 1 DBPR baseline to post-
Stage 2 DBPR conditions.
• Step 4: Take the difference, or delta, between the results from Steps 2 and 3 to calculate the
incremental treatment technology selection forecast and TTHM and HAAS reduction for
Stage 2 (from post-Stage 1 conditions).
5 There is uncertainty in using the ICR data to represent pre-Stage 1 conditions because some plants may
have begun making changes prior to the ICR in anticipation of the Stage 1 DBPR (McGuire 2003). See Section 3.8
for a full discussion of uncertainties in ICR data.
Final Economic Analysis for the Stage 2 DBPR 5-7 December 2005
-------
Because the same pre-Stage 1 baseline was used to evaluate compliance with both Stage 1 and
Stage 2, potential errors in selection of treatment technology and changes in TTHM and HAAS levels are
cancelled out for national level estimates in Step 4. The TWG believed that using the delta approach
reduces the impact of uncertainty in SWAT's predictive equations for TTHM and HAAS. The delta
approach was used for both SWAT and the ICR Matrix Method.
A similar delta approach was developed and used for ground water plants. Because ICR data for
ground water plants were not robust enough to allow for modeling in SWAT, ICR observed data were
used as the pre-Stage 1 baseline. Using the ICR Ground Water Delphi results, EPA calculated the percent
of plants exceeding Stage 1 and Stage 2 MCLs from the Pre-Stage 1 baseline and took the difference as
the percent of plants needing to change treatment technologies from post-Stage 1 to Stage 2.
An uncertainty in the delta approach is the implicit assumption that plants making treatment
technology changes to comply with the Stage 1 DBPR also meet the Stage 2 DBPR MCLs (Stage 1
forecasts are subtracted from Stage 2; only the delta is considered for costs and benefits). To illustrate the
mechanisms of the delta approach and this uncertainty, consider the plots of the maximum locational
running annual average (LRAA) and running annual average (RAA) for each plant in Exhibit 5.2.
Quadrant I contains those plants in compliance with both Stage 1 and Stage 2 considering a 20 percent
safety margin. Quadrant II contains plants that are in compliance with Stage 1, but not Stage 2. Plants in
Quadrant III are those that exceed the MCLs for both Stage 1 and Stage 2. With the delta approach,
plants in Quadrant III that make a treatment technology change to meet the Stage 1 DBPR move to
Quadrant I (i.e., they are Stage 2-compliant). Although EPA recognizes this uncertainty, the Agency
believes this the delta approach is reasonable for the following reasons:
• The Stage 2 DBPR is a required rule in the Safe Drinking Water Act (SDWA) Amendments
of 1996. Details of the Stage 2 DBPR were published in the Agreement in Principle, which
includes the Stage 2 MCLs, in December 2000, which is well before the Stage 1 compliance
deadlines. It is less costly and, therefore, in a water system's best interest to develop a
comprehensive treatment strategy to achieve simultaneous compliance with both Stage 1 and
Stage 2.
A large portion of systems use chloramines to achieve compliance with the Stage 1 DBPR.
Chloramines generally result in lower spatial and temporal variability of TTHM and HAAS
concentrations in distribution systems compared to chlorine, as discussed in Section 5.3.6 and
shown in Exhibit 5.8a. Therefore, systems that have switched to chloramines to comply with
Stage 1 DBPR will likely have LRAA values already below 80 (ig/L for TTHM and 60 (ig/L
for HAAS and will not need to make a second treatment technology change to comply with
the Stage 2 DBPR.
Final Economic Analysis for the Stage 2 DBPR 5-8 December 2005
-------
Exhibit 5.2 Compliance Evaluation of Screened ICR
Surface and Ground Water Plants
140
Ł 80-
140 -,
60 64 80 100
TTHM RAA, ug/L
48 60
HAAS RAA, ug/L
Note: Each point on the graph represents one plant
MCL (with 20% Safety Factor)
RAA = LRAA representation
Hants under the Stage 1 DBPR, but above the Stage 2
DBPR compliance targets
Source: analysis of ICR screened ground and surface water plants (N = 311)
Final Economic Analysis for the Stage 2 DBPR
5-9
December 2005
-------
One disadvantage of the delta approach is that compliance forecasts for the Stage 2 DBPR do not
take into account the specific advanced treatment technologies predicted for Stage 1. In some cases, a
more advanced treatment technology can be predicted for a plant to meet Stage 1 requirements as
compared to Stage 2. There are two reasons this can happen:
A plant is allowed to use chloramines in the Stage 2 model run, but not for the Stage 1 run.
For all SWAT model runs, 77 percent of plants were allowed to convert from free chlorine to
chloramines to comply with DBF rules. The 77 percent cap on chloramine conversion was
developed during the M-DBP FACA deliberations to represent site-specific circumstances
and other local factors that would preclude chloramine usage for reasons other than technical
suitability. The percentage is applied to each plant randomly by a Monte Carlo simulation
model. Because the number is assigned randomly, the situation can (and does) occur where a
plant is not allowed to use chloramines for Stage 1 compliance, but is allowed to use it for
Stage 2 compliance.
EPA considered ultraviolet (UV) an available technology for meeting Stage 2 requirements,
but not Stage 1 requirements, when developing the compliance forecasts. UV is an emerging
technology that has just recently been shown to be an effective disinfectant for many
microorganisms of concern in drinking water. Since the model starts with the same pre-Stage
1 baseline, some plants that are predicted to use more expensive technologies, such as ozone,
microfiltration/ultrafiltration (MF/UF), or granular activated carbon (GAC), to comply with
Stage 1 can achieve compliance with UV once it is included in the Stage 2 runs.
The first factor (randomness in chloramine use) has a relatively small impact on the compliance
forecast. The second factor, assumptions for UV availability, however, causes the delta from Stage 1 to
Stage 2 to be negative for some advanced technologies. EPA developed an approach to adjust the
technology selection forecast to correct for the negatives (called the "adjustment for negatives" step). In
summary, plants selecting UV for the Stage 2 DBPR were reallocated to advanced technologies (ozone,
MF/UF, and GAC-10-Minute Contact Time (GAC 10)) when overall predictions were lower for Stage 2
than for Stage 1. Details on the adjustment-for-negatives step, including flow charts and sample
calculations, are provided in Appendices A and B.
5.3.3 Operational Safety Margins
The M-DBP TWG recommended that a 20 percent operational safety margin be used for DBP
MCLs (TTHM, HAAS, bromate, and chlorite) when evaluating Stage 1 and all Stage 2 regulatory
alternatives. This safety margin is intended to represent the level (i.e., 80 percent of the MCL) at which
systems typically take some action to ensure consistent compliance with a new drinking water standard
and the level at which systems target new treatment technologies to meet the standard. In addition to
representing industry practices, the safety margin also is intended to account for year-to-year fluctuations
in DBP levels. (ICR data are limited to 1 year and might not represent the highest DBP concentrations
that occur in a system.) Individual systems may use higher or lower safety margins based on system-
specific conditions. Use of a safety margin is consistent with prior DBP regulatory development efforts.
Final Economic Analysis for the Stage 2 DBPR 5-10 December 2005
-------
5.3.4 Accounting for the IDSE
For most systems, compliance monitoring for the Stage 2 DBPR is preceded by an IDSE. The
purpose of the IDSE is to identify Stage 2 DBPR compliance monitoring sites that represent high TTHM
and HAAS concentrations in the distribution system. Not all systems must perform an IDSE.
Nontransient noncommunity water system (NTNCWSs) serving fewer than 10,000 people are not
required to conduct an IDSE. Other systems may not need to perform the IDSE if they demonstrate low
historic DBF distribution system concentrations or if they serve fewer than 500 people. Most systems
that conduct an IDSE are expected to monitor for one year.
There are several reasons why Stage 2 compliance monitoring sites identified by the IDSE may
have higher TTHM and HAAS levels than data collected under the ICR. First, monitoring under the
IDSE includes more sites, and samples are taken more frequently than the ICR. In addition, sites are
selected to represent both high TTHM and high HAAS levels. Whether an individual system finds sites
with higher TTHM and HAAS levels than those reported in the ICR depends on a number of factors,
including but not limited to:
• Spatial variability in TTHM and HAAS levels. The more variability in a system, the more
likely the system will find higher LRAAs during the IDSE. Spatial variability is influenced
by residual disinfectant type (free chlorine versus chloramines).
• Temporal variability in TTHM and HAAS levels. Seasonal variability related to increased
temperature and changes in source water quality may be better characterized during the IDSE
because they take more frequent samples (for large systems, 6 per year for the IDSE
compared to 4 per year required for the Stage 1 DBPR). Temporal variability is a much
larger factor for surface water systems than for ground water systems.
• Number of plants per system. Systems with a higher than average number of plants are
already sampling at many locations under the Stage 1 DBPR and thus have most likely
already captured much of the spatial variability of DBP levels in their distribution systems.
Conversely, systems with a lower than average number of plants have collected fewer
samples under Stage 1 and are thus more likely to find higher LRAAs when monitoring for
the IDSE.
System configuration. Systems having more complicated distribution systems (e.g., well-
looped systems with several large users, systems with multiple storage facilities, and systems
with pumping stations) are more likely to find higher LRAAs during their IDSEs.
Complexity of the system generally decreases with system size.
• Technical resources used to select ICR and/or Stage 1 sites. Systems with extensive residual
data, extensive DBP data, hydraulic models, and those that have performed tracer studies
should already have well-defined maximum residence time sites. These systems are less
likely to see significantly higher LRAAs as a result of their IDSEs.
The IDSE can potentially affect the compliance forecast in two ways. First, systems that appear
to be in compliance with the Stage 2 DBPR based on an evaluation of their ICR data might find increased
DBP concentrations at their new Stage 2 DBPR monitoring locations (post-ID SE), high enough to cause
them to make treatment technology changes. Second, systems that expect to be out of compliance with
Final Economic Analysis for the Stage 2 DBPR 5-11 December 2005
-------
the Stage 2 DBPR based on an evaluation of their ICR data may need to use more advanced technology
changes to meet rule requirements at new sites identified during the IDSE.
One limitation of EPA's compliance forecast methodology is the use of ICR data as model inputs;
ICR data are the best available data, but they may not represent Stage 2 DBPR compliance monitoring
results due to systems conducting the IDSE. This limitation is, in part, accounted for by the use of a 20
percent safety margin. EPA believes that the 20 percent safety margin accounts for potential impacts of
the IDSE for small surface water systems because their distribution systems are not usually as complex as
large systems. For ground water systems, EPA believes that a 20 percent safety margin already accounts
for the IDSE because the year-to-year variability in source water quality (and TTHM and HAAS levels) is
low. Similarly, chloramine systems generally observe lower spatial and temporal variability in TTHM
and HAAS distribution system levels (USEPA 2005k); thus, EPA believes the 20 percent safety margin
accounts for potential impacts of the IDSE for these systems.
EPA believes, however, that the 20 percent safety margin is not sufficient to account for the
potential impacts of the IDSE on large and medium surface water systems because spatial variability of
DBP levels and distribution system complexity are greatest in these systems. Since the proposal, EPA
developed a methodology to quantify the potential impacts of the IDSE on these systems.
5.3.4.1 Analysis of Spatial Variability in Large and Medium Surface Water Systems
ICR screened data, consisting of data from 213 surface water plants, were used to assess spatial
variability of DBFs in distribution systems of large and medium surface water systems. See section 3.7.1
for a description of the ICR data set and the screening method. (Only those plants with 3 of 4 quarters of
data that have TTHM and HAAS data for at least 3 of 4 distribution system locations are considered in the
analysis.)
EPA began with the simplifying assumption that the spatial variability in the ICR data represents
the variability that systems can find through IDSE monitoring. That is, all plants can find post-IDSE
maximum LRAA values according to the following formula6:
post-IDSE LRAAmax = ICR LRAAmax + (ICR LRAAmax - ICR LRAA2ndHl) (Equation 5.1)
Where:
post-IDSE LRAAmax = the maximum LRAA value found after the system has conduced the IDSE
ICR LRAAmax = the maximum LRAA value as reported for the last four quarters of the ICR
ICR LRAA2ndHl = the second highest LRAA as reported for the last four quarters of the ICR
Exhibit 5.3 predicts the potential increase in the percent of plants making treatment technology
changes from Stage 1 to Stage 2 by assuming that the Stage 2-compliant ICR surface water plants have
post-IDSE LRAAs according to Equation 5.1. Exhibit 5.4a characterizes the difference between the ICR
LRAAmax and ICR LRAA2ndHl for those surface water plants that are predicted to make treatment
6This method was selected over other alternatives since the highest location in the ICR data was not always
at the maximum residence time location. In addition, this approach more conservatively predicts spatial variability,
as it represents the highest two locations, and thus the smallest difference between two ICR-observed locations.
Final Economic Analysis for the Stage 2 DBPR 5-12 December 2005
-------
technology changes for Stage 1 and Stage 2, including those plants that are predicted to make treatment
technology changes if Equation 5.1 was applied. Note that the average difference between the ICR
LRAAmax and ICR LRAA2ndHl is 12.7 |ig/L for TTHM and 4.4 |ig/L for HAAS. The cumulative
distribution for ICR LRAAmax - LRAA2ndHl for TTHM and HAAS are shown in Exhibits 5.4b and 5.4c,
respectively.
Exhibit 5.3 Predicted Increase in Percent Making Treatment Technology Changes
based on Spatial Variability Analysis
Disinfectant
Type
CL2
CLJVI
All
Number of
Screened ICR
Surface Water
Rants
A
133
80
213
Si-Compliant Rants that are
already S2 Non-Compliant
Number
B
29
7
36
Percent of All
Rants
C = B/A
21.8%
8.8%
16.9%
S2-Compliant Rants that will
be S2 Non-Compliant after the
IDSE based on IRAA^ +
(LRAAww-LRAA2ndHI)
Number
D
11
4
15
Percent of All
Rants
E = D/A
8.3%
5.0%
7.0%
Notes: CL2 = free chlorine, CLM = chloramines
= the maximum LRAA value for each plant.
i = the second highest LRAA value for each plant.
Sources: A) See Section 3.7.1 for a detailed description of the ICR data set and screening
method.
B) & D) Stage 1 and Stage 2 compliance is based on an assessment of ICR TTHM and
HAAS occurrence data applying a 20 percent safety margin.
Exhibit 5.4a Analysis of Variability for Stage 2 Non-Compliant Plants
Number of Screened Stage 2 non-compliant Plants 1
Average of LRAAM/\x
Average of LRAA2ndm
Delta of Average of LRAAMAx and Average of LRAA2ndm
TTHM
CL2
40
64.50
49.79
14.71
CLM
11
57.63
52.31
5.31
All
51
63.02
50.33
12.68
HAAS
CL2
40
37.57
32.93
4.64
CLM
11
38.97
35.58
3.39
All
51
37.87
33.50
4.37
Notes
'Represents all screened ICR SW plants that are in compliance with Stage 1 but have Post-IDSE LRAAmax values > 64 ug/L for TTHM or
48 ug/L for HAAS.
See Section 3.7.1 for a detailed description of the ICR data set and screening method.
CL2 = free chlorine, CLM = chloramines
LRAAMAX = the maximum LRAA value for each plant.
LRAA2ndHi = the second highest LRAA value for each plant.
Final Economic Analysis for the Stage 2 DBPR
5-13
December 2005
-------
Exhibit 5.4b Cumulative Distribution of ICR LRAAMAX - ICR LRAA2ndH, for Stage 2
Non-Compliant Plants (TTHM data)
100%
90%
5 70%
ii
| 60%
'c
01
u
01
Q.
01
JS
3
E
3
o
20%
0%
12 16 20 24 28 32 36 40
- LRAA2ndH, for TTHM ftjg/L)
44
48
52
56
60
F;'/?a/ Economic Analysis for the Stage 2 DBPR
5-14
December 2005
-------
Exhibit 5.4c Cumulative Distribution of ICR LRAAMAX - ICR LRAA2ndH, for Stage 2
Non-Compliant Plants (HAAS Data)
100%
90%
70%
II
0)
i)
^
-------
5.3.4.2 Modifying the Operational Safety Margin
The next step in the process of quantifying the potential impacts of the IDSE is applying the
results in the previous section to the large and medium surface water compliance forecast methodology.
Changes to the compliance forecast must take into account:
The potential changes in the number of systems making treatment technology changes
The potential treatment technology changes selected by all systems
• The resulting reduction in TTHM and HAAS concentrations
EPA determined that the most effective way to revise the compliance forecast was to modify the
operational safety margin. A larger safety margin would affect the compliance forecasts of both SWAT
and the ICR Matrix Method (i.e., cause more plants to make treatment technology changes). A larger
safety margin would also impact the distribution of treatment technologies predicted by SWAT. (SWAT
re-evaluates compliance with the Stage 2 DBPR by considering the safety margin after evaluating each
treatment type in the decision tree.) Lastly, an increased safety margin could result in plants predicted to
install higher-cost treatment technologies to meet compliance with a lower numeric MCL value. The
decrease in TTHM and HAAS levels resulting from the additional plants making treatment technology
changes, and more advanced treatment technologies being selected would be automatically calculated by
SWAT.
To identify the most appropriate safety margin, EPA compared the results in Exhibits 5.3 and 5.4
to the compliance assessment in Exhibit 5.5. The variability analysis in Exhibit 5.3 predicts that an
additional 7 percent of all plants could make treatment technology changes as a result of the IDSE. This
prediction falls in between the compliance analysis of ICR for a 25 and a 30 percent safety margin
(compare to column E in Exhibit 5.5). Because EPA believes the analysis of spatial variability produces
an overestimate of potential impacts as discussed previously, a safety margin of 25 percent was chosen to
model the impacts of the IDSE. Although the true magnitude of the influence of the IDSE is unknown,
EPA believes that the compliance forecast based on analysis of spatial variability provides a plausible
prediction.
Final Economic Analysis for the Stage 2 DBPR 5-16 December 2005
-------
Exhibit 5.5 Compliance Analysis of ICR Screened Plants at Different Operational
Safety Margins
Plant Subset
All Screened Plants
Stage 1 non-compliant at 20% SM
Stage 2 non-compliant at 20% SM
Stage 2 non-compliant at 25% SM
Stage 2 non-compliant at 30% SM
TTHM/HAA5
LRAA Values
A
64/48
60/45
56/42
Number
of Plants
B
213
41
77
88
95
Percent of
All Plants
C = B/213
19.2%
36.2%
41 .3%
44.6%
Delta from
Stage 1 to
Stage 2
D = C-19.2%
16.9%
22.1%
25.4%
Incremental Percent
More Plants
Compared to 20
Percent SM for
Stage 2
E = D - 16.9%
5.2%
8.5%
Ratio of
Additional
Plants to the 20
Percent SM for
Stage 2
F = D/16.9%
1.3
1.5
Sources:
A. MCL * (1 - safety margin)
B. Assessment of ICR screened DBP dataset for surface water systems. See section 3.7.1 for a description of the dataset and
screening process
5.3.4.3 Incorporating Potential Impacts of the IDSE into the Compliance Forecast
EPA believes that the 20 percent operational safety margin already accounts for the impacts of
the IDSE for some systems, including small systems, ground water systems, and those using chloramines.
As discussed earlier, EPA believes that the 20 percent safety margin accounts for potential impacts of the
IDSE for small systems because their distribution systems are less complex than those of large systems.
EPA also believes that the safety margin accounts for the IDSE for ground water systems because the
year-to-year variability in source water quality (and thus, TTHM and HAAS formation) is low.
Chloramine systems generally observe lower spatial and temporal variability in TTHM and HAAS
distribution system levels (USEPA 2005k); thus, EPA believes the 20 percent safety margin accounts for
potential impacts of the IDSE for these systems.
EPA believes that the compliance forecast for some large and medium surface water systems
using a 20 percent safety margin is already conservative because the treatment technology decision tree
(see Appendix A) does not include distribution system operational improvements that systems are more
likely to use for compliance with Stage 2 as compared to Stage 1. However, for the reasons given in
Section 5.3.4, EPA believes that a compliance forecast based on a safety margin of 25 percent also
provides a plausible prediction for some large and medium surface water systems. Because both
operational safety margins are considered plausible, EPA's compliance forecast for large and medium
surface water systems assigns equal probability to the 20 and 25 percent safety margins. Compliance
forecast results in this chapter reflect this assumption.
Although the magnitude of the impacts of the IDSE is uncertain, it is important to note that the
IDSE will affect costs and benefits in the same direction. If higher DBFs are identified through the IDSE,
more systems make treatment technology changes and more advanced treatment technologies are
selected, increasing both costs and benefits. A sensitivity analysis in Section 5.7 shows that the
magnitude of the impact of the IDSE on the benefits analysis is expected to be higher than the magnitude
of the impact on the cost analysis.
Final Economic Analysis for the Stage 2 DBPR
5-17
December 2005
-------
5.3.5 Methodology for Incorporating SWAT and ICR Matrix Method Results into the
Compliance Forecast
As described in Section 5.3.1, EPA uses two tools to predict changes in treatment technology and
resulting reductions in DBF levels: SWAT and the ICR Matrix Method. Exhibit 5.6 compares the SWAT
and ICR Matrix Method predictions of plants changing treatment technology for Stage 1 and Stage 2.
(Comparisons of the predicted reductions in TTHM and HAA5 concentrations are presented in Sections
5.6 and 5.7.) Exhibit 5.6 shows that although the total predicted percentages of plants making treatment
technology changes for Stage 2 are similar for both methods, the percent of plants making treatment
technology changes to comply with Stage 1 differ between the ICR Matrix Method and SWAT by more
than 10 percentage points. Such differences are expected given the inherent differences in the two
methods and uncertainties associated with each7.
Because both SWAT and the ICR Matrix Method have associated uncertainty, results from both
are used to generate the compliance forecast for surface water systems. The ICR Matrix Method does not
predict the specific treatment technologies that plants will install. Thus, results from the ICR matrix
method are incorporated by comparing the predicted percent of plants making treatment technology
changes with SWAT results to create a ICR Matrix Method-to-SWAT multiplier, shown in Exhibit 5.6,
column G. EPA generated a uniform distribution with 1.0 as the 5th percentile value and the ICR Matrix
Method-to-SWAT multiplier for plants making treatment technology changes as the 95th percentile value.
Two separate distributions were used, one for the 20 percent safety margin and one for the 25 percent
safety margin, as shown in Exhibit 5.6. Exhibit 5.7 provides a graphical depiction of the two uniform
distributions for the Preferred Alternative.
7One possible explanation for this difference is that the maximum residence time reported in the ICR is not
consistent with the maximum TTHM and HAAS value reported for the system. As shown in Section A.6.4, the
maximum TTHM and HAAS LRAAs often occur at locations other than the one designated as the maximum
residence time location (MAX). Uncertainty, and especially underestimates, of the maximum residence time may
result in a lower percent of plants predicted by SWAT to make treatment technology changes for Stage 2, but not
Stage 1. The average residence time in SWAT is deemed to be more certain because it is based on the mean of the
four distribution system residence times reported in the ICR (for the distribution system equivalent sample point
(DSE), average sample point number 1 (AVE1), average sample point number 2 (AVE2), and MAX locations).
Final Economic Analysis for the Stage 2 DBPR 5-18 December 2005
-------
Exhibit 5.6 Predicted Percent of Plants Making Treatment Technology Changes to Meet Stage 1 and Stage 2
Regulatory Alternatives for the ICR Matrix Method and SWAT
Regulatory Alternative
Preferred Reg. Alternative (80/60 LRAA, IDSE), 20% SM
Preferred Reg. Alternative (80/60 LRAA, IDSE), 25% SM
Reg. Alternative 1 (Bromate = 5), 20% SM
Reg. Alternative 2 (80/60 Single High), 20% SM
Reg. Alternative 3 (40/30 RAA), 20% SM
ICR Matrix Method
% Changing
from pre-S1
to pre-S2
A
19.2%
19.2%
19.2%
19.2%
19.2%
% Changing
fom pre-S1
to post-32
B
36.2%
41.3%
36.2%
58.2%
72.8%
% Changing
from pre-S2
to post-32
C = B-A
16.9%
22.1%
16.9%
39.0%
53.5%
SWAT
% Changing
from pre-S1
to pre-32
D
32.5%
32.5%
32.5%
32.5%
32.5%
% Changing
fom pre-S1
to post-32
E
39.1%
45.4%
39.4%
63.1%
72.1%
% Changing
from pre-S2
to post-32
F = E-D
6.6%
12.9%
6.8%
30.6%
39.6%
ICR Matrix
Method -to-
SWAT
Multiplier
G = C/F
2.57
1.71
2.47
1.27
1.35
Notes: 1) The operational safety margin for Stage 1 is 20 percent for all analyses.
Sources: SWAT run summaries (USEPA 2001 b), ICR Matrix Method results (USEPA 2005a).
(A) Percent of plants changing from bin B2 from Exhibit 5.16a. ICR Matrix Method results for Reg. Alternative 1 are the same as for the Preferred
Alternative, assuming a 20 percent safety margin.
(B) The sum of percent of plants changing in bins B2 and A2 from Exhibits 5.16b through 5.16d. ICR Matrix Method results for Reg. Alternative 1 are
the same as for the Preferred Alternative, 20 percent safety margin.
(D) and (E) SWAT run summaries (USEPA 2001 b).
Final Economic Analysis for the Stage 2 DBPR
5-19
August 2005
-------
Exhibit 5.7 Uniform Distributions for Incorporating Results from SWAT and the
ICR Matrix Method into the Compliance Forecast for Surface Water Systems
.0
ro
.a
o
Uniform Distribution for a 20%
Safety Margin
_
ro
_a
o
Uniform Distribution for a 25%
Safety Margin
2 r 3
(2.57)
Multiplier
(1.71)
Multiplier
Note: The uniform distributions are used when generating the Stage 2 DBPR compliance forecasts for surface water
systems for the Preferred Alternative. The same method is applied to the other regulatory alternatives using the
multipliers in Exhibit 5.6.
Source: Exhibit 5.6.
5.3.6 Compliance Forecast Simulation Model
To include results from two surface water compliance forecast tools and to account for
uncertainty in the potential impacts of the IDSE in the compliance forecast, EPA developed a Monte
Carlo simulation model. The model follows three basic steps:
• Step 1: For large and medium surface water systems, the model randomly selects the SWAT-
predicted treatment technology selection delta for either the 20 or 25 percent safety margin
runs. Each safety margin has an equal (50 percent) chance of being selected. For small
surface water systems and all ground water systems, the treatment technology selection delta
for the 20 percent safety margin is always selected.
• Step 2: For large, medium, and small surface water systems, the model randomly selects the
ICR-to-SWAT multiplier from the appropriate uniform distribution from Exhibit 5.7 for the
safety margin selected in Step 1.
• Step 3: For large, medium, and small surface water systems, the model multiplies the result
from Step 2 by the treatment technology selection delta results identified in Step 1 to
calculate the percent and number of plants making treatment technology changes from Stage
1 to Stage 2. For ground water systems, the treatment technology selection delta for a 20
percent safety margin is always used to calculate the percent of plants changing from Stage 1
to Stage 2.
Final Economic Analysis for the Stage 2 DBPR
5-20
December 2005
-------
The process is repeated 10,000 times to produce a distribution of plants making treatment
technology changes from Stage 1 to Stage 2 (results from Step 3). This distribution is carried through the
cost model, as described in Chapter 7. Note that for large and medium surface water systems, only the
treatment technology selection delta for the 20 percent safety margin is used in Step 1 for Regulatory
Alternatives 1, 2, and 3 as the IDSE is not a component of these alternatives.
5.4 Compliance Forecast Results
Three types of compliance forecasts are presented in this EA:
1) Treatment Technologies-in-Place forecasts show the number and percent of plants that are
using a given treatment technology either before or after rule implementation. For pre-Stage
2 and post-Stage 2 treatment technologies-in-place, the calculated number and percent of
plants represent plants predicted to make treatment technology changes to comply with the
rule, added to the number and percent of plants already using the treatment technology before
rule promulgation.
2) Treatment Technology Selection forecasts show the number and percent of plants that are
predicted to add a given treatment technology to comply with the Stage 1 DBPR or Stage 2
DBPR regulatory alternatives. These results include only the number of plants that exceed
the rule's MCLs and that therefore must make treatment technology changes to comply with
the rule. Treatment technology selection results are always based on pre-Stage 1 conditions.
3) Treatment Technology Selection Delta forecasts show the incremental number and percent
of plants that must add a given treatment technology following the Stage 1 DBPR to meet
Stage 2 DBPR MCLs. Treatment technology selection deltas are calculated by subtracting
the Stage 1 DBPR treatment technology selection from the Stage 2 DBPR treatment
technology selection. These treatment technology selection delta are used for costing.
Note that all forecast results in subsequent sections have the same formatting, with system size
down the left hand column and treatment technology across the top. Exhibit 5.8 provides the exhibit
numbers for the compliance forecasts for the Stage 2 DBPR Preferred Alternative. Exhibit 5.9
summarizes compliance forecast results for community water systems (CWSs) and NTNCWSs for all
regulatory alternatives. Surface water system results in Exhibit 5.9 represent the mean values from the
Monte Carlo simulation model. These results with 90 percent confidence intervals and further population
breakouts are shown in Exhibit 7.3.
Final Economic Analysis for the Stage 2 DBPR 5-21 December 2005
-------
Exhibit 5.8 Compliance Forecast Exhibits for the
Stage 2 DBPR Preferred Alternative
Compliance Forecast Type
Pre-Stage 1 Treatment Technologies-in-Place
Stage 1 DBPR Treatment Technology Selection
Pre-Stage 2 Treatment Technologies-in-Place
Stage 2 DBPR Treatment Technology Selection Delta
Post-Stage 2 Treatment Technologies-in-Place
Exhibit Containing Results for the
Preferred Alternative 1
Exhibits 3. 13 and 3. 14
Exhibits C.1 and C.2
Exhibit 5. 10 and 5. 13
Exhibits 5. 11 and 5.1 42
Exhibits 5. 12 and 5. 15
Notes:
1. The first exhibit contains results for surface water plants, the second for ground water plants.
2. Treatment technology selection delta tables are used for costing.
Surface Water Systems
Section 3.6.2 describes how predictions of pre-Stage 1 treatment technologies-in-place were
developed. The pre-Stage 1 baseline for surface water systems is shown in Exhibit 3.13.
Exhibit 5.10 shows the predicted treatment technologies-in-place for surface water plants
following the Stage 1 DBPR (i.e., the pre-Stage 2 DBPR baseline). The treatment technologies-in-place
for small surface and ground water systems can be derived by adding the treatment technology selection
for the Stage 1 DBPR (Exhibit C. la for CWS, C. Ib for NTNCWS) to the treatment technologies-in-place
for the pre-Stage 1 baseline (Exhibit 3.13). This is not true, however, for plants in large and medium
systems. EPA assumes that pre-Stage 2, only medium and large systems employ advanced treatment
technologies. Therefore, the SWAT program produces a different type of result, called "ending
technologies," that accounts for these plants that already have an advanced treatment technology but must
select another to meet the rule alternative.
Exhibits 5.1 la through 5.1 Id present the treatment technology selection deltas, as a percentage
(5.1 la for CWS, 5.1 Ic for NTNCWS) and as the number of plants (5.1 Ib for CWS, 5.1 Id for NTNCWS),
for the Stage 2 DBPR Preferred Alternative for surface water systems. These exhibits are used to predict
costs for surface water systems. As described in Section 5.3, technology selection deltas for surface water
systems incorporate two compliance forecast methods: SWAT and the ICR Matrix Method. For the
Preferred Alternative, the Stage 2 treatment technology selection deltas for large and medium surface
water systems represent equal probability of a 20 and 25 percent safety margin to model the potential
impact of the IDSE.
Exhibits 5.12a through 5.12d present the final post-Stage 2 treatment technologies-in-place for
surface water systems under the Stage 2 DBPR Preferred Alternative.
Final Economic Analysis for the Stage 2 DBPR 5-22 December 2005
-------
Ground Water Systems
Exhibit 3.14 summarizes the pre-Stage 1 DBPRbaseline treatment technologies-in-place for
ground water treatment plants. The derivation of this baseline is discussed in Section 3.6.2. For plants in
large and medium ground water systems, ICR treatment data were used to derive the predicted percent of
plants using each treatment technology. The percentage of small ground water plants using each
treatment technology is based on evaluation of CWS data; EPA assumed that NTNCWSs use similar
treatment technologies for the size categories shown.
Exhibit 5.13 shows the predicted treatment technologies-in-place for ground water plants
following the Stage 1 DBPR (pre-Stage 2 DBPR baseline). The treatment technologies-in-place for the
pre-Stage 2 baseline can be derived by adding the treatment technology selection for the Stage 1 DBPR
(Exhibit C.2a for CWS, C.2b for NTNCWS) to the treatment technologies-in-place for the pre-Stage 1
baseline (Exhibit 3.14a for CWS, 3.14b for NTNCWS).
Exhibits 5.14a through 5.14d present the treatment technology selection deltas for ground water
systems. EPA used these deltas to predict the costs to ground water systems of complying with the Stage
2 DBPR. Exhibits 5.15a through 5.15d present EPA's prediction of treatment technologies that will be
employed after systems comply with the Stage 2 DBPR. For all plants, post-Stage 2 treatment
technologies-in-place can be derived by adding the predicted pre-Stage 2 treatment technologies-in-place
(Exhibit 5.13) to the treatment technology selections in Exhibit 5.14.
Final Economic Analysis for the Stage 2 DBPR 5-23 December 2005
-------
Exhibit 5.9 Plants in CWSs and NTNCWSs Making Treatment Technology
Changes From Stage 1 For Stage 2 DBPR Regulatory Alternatives
System Size
and Type
Baseline Number
of CWS and
NTNCWS Plants
A
Mean Estimate of Plants Changing Treatment
Technology from Stage 1 to Stage 2 (including CLM)
Number
B
Percent
C = B/A
Preferred Regulatory Alternative
SW> 10K
SW< 10K
GW> 10K
GW<10K
All Plants
Alternative 1
SW> 10K
SW< 10K
GW>10K
GW< 10K
All Plants
2,561
4,757
7,048
45,855
60,221
373
420
145
1,323
2,261
14.6%
8.8%
2.1%
2.9%
3.8%
(BR5)
2,561
4,757
7,048
45,855
60,221
305
420
145
1,327
2,197
1 1 .9%
8.8%
2.1%
2.9%
3.6%
Alternative 2 (80/60 SH)
SW> 10K
SW< 10K
GW> 10K
GW<10K
All Plants
Alternative 3
SW> 10K
SW< 10K
GW>10K
GW< 10K
All Plants
2,561
4,757
7,048
45,855
60,221
892
1,052
488
2,377
4,810
34.8%
22.1%
6.9%
5.2%
8.0%
(40/30 RAA)
2,561
4,757
7,048
45,855
60,221
1,193
1,306
334
1,613
4,446
46.6%
27.4%
4.7%
3.5%
7.4%
Notes: Uncertainty in the impacts of the IDSE for SW systems serving >10,000
people are reflected in estimates for the Preferred Regulatory Alternative
only. Estimates for all SW systems are the averages of the results of two
methods: SWAT and the ICR Matrix Method.
Sources: (A) Baseline Number of CWS and NTNCWS plants from Chapter 3,
Exhibit 3.2.
(B) Exhibits 5.11 and 5.14 for the Preferred Regulatory Alternative,
Exhibits C.3 and C.4 for Alternative 1, Exhibits C.7 and C.8 for Alternative
2, and Exhibits C.11 and C.12 for Alternative 3
Final Economic Analysis for the Stage 2 DBPR
5-24
December 2005
-------
Exhibit 5.10a Pre-Stage 2 DBPR Treatment Technologies-in-Place for CWS Surface Water Plants (Percent and Number of Plants by Residual Disinfection Type)
System Size
(Population Served)
<100
100-499
500-999
1,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1,000,000
Total Plants, %
System Size
(Population Served)
<100
100-499
500-999
1,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1,000,000
Total Plants, %
No Advanced Treatment Technologies1
CL2
A
% #
41.8% 150
35.6% 273
172
33.4% 378
421
35.0% 452
203
35.0% 213
26
34.9% 2,287
CLM
B
% #
29.7% 107
35.4% 272
171
41.3% 467
520
39.0% 504
226
39.0% 238
29
38.7% 2,534
GAC10 + AD
CL2
M
% #
0.5% 7
3
0.5% 3
0
0.2% 13
CLM
N
% #
0.6% 7
3
0.6% 4
0
0.2% 15
Chlorine Dioxide
CL2
C
% #
1.0% 7
5
1 .9% 22
24
3.3% 42
19
3.3% 20
2
2.2% 141
CLM
D
% #
0.9% 7
4
2.1% 24
27
3.7% 47
21
3.7% 22
3
2.4% 155
GAC20
CL2
O
% #
2.0% 7
1.1% 8
5
1.0% 12
13
0.2% 2
1
0.2% 1
0
0.8% 49
CLM
P
% #
1.3% 5
1.0% 7
5
1.2% 13
15
0.2% 2
1
0.2% 1
0
0.8% 49
UV
CL2
E
% #
CLM
F
% #
GAC20 + AD
CL2
Q
% #
0.0% 0
0.5% 4
2
0.5% 6
7
0.0% 0
0
0.0% 0
0
0.3% 18
CLM
R
% #
0.0% 0
0.4% 3
2
0.6% 7
7
0.0% 0
0
0.0% 0
0
0.3% 19
Ozone
CL2
G
% #
5.1% 39
24
4.0% 45
50
6.1% 78
35
6.1% 37
4
4.8% 314
CLM
H
% #
4.6% 35
22
4.5% 51
56
6.8% 87
39
6.8% 41
5
5.1% 337
Membranes
CL2
S
% #
2.1% 8
0.5% 3
2
0.2% 2
2
0.3% 4
2
0.3% 2
0
0.4% 26
CLM
T
% #
1.4% 5
0.4% 3
2
0.2% 2
2
0.4% 5
2
0.4% 2
0
0.4% 25
MF/UF
CL2
I
% #
14.5% 52
8.9% 68
43
6.2% 70
78
0.9% 1 1
5
0.9% 5
1
5.1% 333
CLM
J
% #
7.1% 26
4.8% 37
23
2.9% 32
36
1.0% 12
6
1.0% 6
1
2.7% 179
GAC10
CL2
K
% #
1.0% 13
6
1.0% 6
1
0.4% 27
CLM
L
% #
1.2% 15
7
1.2% 7
1
0.5% 30
TOTAL
CL2
U = A+C+E+G+I+K+M+O+Q+S
% #
60.4% 217
52.5% 403
254
47.3% 534
595
47.3% 610
274
47.3% 288
35
49.0% 3,209
CLM
V= B+D+F+H+J+L+N+P+R+T
% #
39.6% 142
47.5% 364
229
52.7% 596
664
52.7% 681
306
52.7% 322
39
51 .0% 3,342
Note: Detail may not add to totals due to independent rounding
'"No Adv" includes conventional, non-conventional, and softening plants.
Source: Surface water systems serving <10,000 people: Add Treatment Technologies-in-Place
Surface water systems serving 10,000 or more people: Use ending Treatment Technology predi
for the Pre-Stage 1 DBPR Baseline (Exhibit 3.13a) to Stage 1 Treatment Technology Selection (Exhibit C.1 a).
ictions from SWAT (FACA Screen SeriesS v3.0 Database) (USEPA, 2001 b).
Exhibit 5.10b Pre-Stage 2 DBPR Treatment Technologies-in-Place for NTNCWS Surface Water Plants (Percent and Number of Plants by Residual Disinfection Type)
System Size
(Population Served)
<100
100-499
500-999
1,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1,000,000
Total Plants, %
System Size
(Population Served)
<100
100-499
500-999
1,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1,000,000
Total Plants, %
No Advanced Treatment Technologies1
CL2
A
% #
41.8% 95
35.6% 111
38
33.4% 31
8
35.0% 2
0
35.0% 0
0
37.1% 285
CLM
B
% #
29.7% 67
35.4% 1 1 1
38
41.3% 38
10
39.0% 2
0
39.0% 0
0
34.7% 266
GAC10 + AD
CL2
M
% #
0.5% 0
0
0.5% 0
0
0.0% 0
CLM
N
% #
0.6% 0
0
0.6% 0
0
0.0% 0
Chlorine Dioxide
CL2
C
% #
1.0% 3
1
1.9% 2
0
3.3% 0
0
3.3% 0
0
0.8% 6
CLM
D
% #
0.9% 3
1
2.1% 2
1
3.7% 0
0
3.7% 0
0
0.8% 6
GAC20
CL2
O
% #
2.0% 4
1.1% 3
1
1.0% 1
0
0.2% 0
0
0.2% 0
0
1.3% 10
CLM
P
% #
1.3% 3
1.0% 3
1
1.2% 1
0
0.2% 0
0
0.2% 0
0
1.1% 8
UV
CL2
E
% #
,-'
CLM
F
% #
GAC20 + AD
CL2
Q
% #
0.0% 0
0.5% 2
1
0.5% 0
0
0.0% 0
0
0.0% 0
0
0.3% 3
CLM
R
% #
0.0% 0
0.4% 1
0
0.6% 1
0
0.0% 0
0
0.0% 0
0
0.3% 3
Ozone
CL2
G
% #
5.1% 16
5
4.0% 4
1
6.1% 0
0
6.1% 0
0
3.4% 26
CLM
H
% #
4.6% 14
5
4.5% 4
1
6.8% 0
0
6.8% 0
0
3.2% 25
Membranes
CL2
S
% #
2.1% 5
0.5% 1
0
0.2% 0
0
0.3% 0
0
0.3% 0
0
0.9% 7
CLM
T
% #
1.4% 3
0.4% 1
0
0.2% 0
0
0.4% 0
0
0.4% 0
0
0.7% 5
MF/UF
CL2
I
% #
14.5% 33
8.9% 28
9
6.2% 6
2
0.9% 0
0
0.9% 0
0
10.1% 77
CLM
J
% #
7.1% 16
4.8% 15
5
2.9% 3
1
1.0% 0
0
1.0% 0
0
5.2% 40
GAC10
CL2
K
% #
,-'
/
1.0% 0
0
1.0% 0
0
0.0% 0
CLM
L
% #
1.2% 0
0
1.2% 0
0
0.0% 0
TOTAL
CL2
U = A+C+E+G+I+K+M+O+Q+S
% #
60.4% 137
52.5% 164
56
47.3% 43
12
47.3% 2
0
47.3% 0
0
54.0% 414
CLM
V= B+D+F+H+J+L+N+P+R+T
% #
39.6% 89
47.5% 148
50
52.7% 49
13
52.7% 3
0
52.7% 1
0
46.0% 353
Note: Detail may not add to totals due to independent rounding
'"No Adv" includes conventional, non-conventional, and softening plants.
Source: Surface water systems serving <10,000 people: Add Treatment Technologies-in-Place for the Pre-Stage 1 DBPR Baseline (Exhibit 3.13b) to Stage 1 Treatment Technology Selection (Exhibit C.1 b).
Surface water systems serving 10,000 or more people: Use ending Treatment Technology predictions from SWAT (FACA Screen SeriesS v3.0 Database) (USEPA, 2001 b).
Final Economic Analysis for the Stage 2 DBPR
December 2005
-------
Exhibit 5.11 a
Stage 2 DBPR Treatment Technology Selection Deltas for CWS Surface Water Plants (Percent of Plants by Residual Disinfection Type)
Preferred Alternative
System Size
(Population
Served)
<100
100-499
500-999
1,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1, 000,000
Total %
System Size
(Population
Served)
<100
100-499
500-999
1,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1, 000,000
Total %
Converting to CLM Only
Mean 5th 95th
A
1.9% 1.0% 2.7%
4.1% 2.3% 5.9%
4.1% 2.3% 5.9%
4.2% 2.4% 6.1%
4.2% 2.4% 6.1%
8.6% 4.8% 12.0%
8.6% 4.8% 12.0%
8.6% 4.8% 12.0%
8.6% 4.8% 12.0%
5.8% 3.2% 8.2%
Chlorine Dioxide
CL2
Mean 5th 95th
B
0.1% 0.1% 0.2%
0.1% 0.1% 0.2%
0.2% 0.1% 0.2%
0.2% 0.1% 0.2%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.1% 0.0% 0.1%
GAC10 -(-Advanced Disinfectants
CL2
Mean 5th 95th
CLM
Mean 5th 95th
L M
,-'
/
1.2% 0.6% 1.7%
1.2% 0.6% 1.7%
1.2% 0.6% 1.7%
1.2% 0.6% 1.7%
0.5% 0.2% 0.7%
0.5% 0.2% 0.7%
0.5% 0.2% 0.7%
0.5% 0.2% 0.7%
0.5% 0.2% 0.7%
0.2% 0.1% 0.3%
CLM
Mean 5th 95th
C
0.4% 0.2% 0.5%
0.4% 0.2% 0.5%
0.9% 0.5% 1.3%
0.9% 0.5% 1.3%
0.3% 0.4% 0.1%
0.3% 0.4% 0.1%
0.3% 0.4% 0.1%
0.3% 0.4% 0.1%
0.5% 0.4% 0.6%
UV
CL2
Mean 5th 95th
D
4.1% 2.3% 5.9%
1.2% 0.7% 1.8%
1.2% 0.7% 1.8%
1.0% 0.5% 1.4%
1.0% 0.5% 1.4%
2.8% 0.6% 5.7%
2.8% 0.6% 5.7%
2.8% 0.6% 5.7%
2.8% 0.6% 5.7%
1.9% 0.7% 3.4%
GAC20
CL2
Mean 5th 95th
N
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.2% 0.0% 0.7%
0.2% 0.0% 0.7%
0.2% 0.0% 0.7%
0.2% 0.0% 0.7%
0.1% 0.0% 0.3%
CLM
Mean 5th 95th
O
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.1% 0.0% 0.2%
0.1% 0.0% 0.2%
0.1% 0.0% 0.2%
0.1% 0.0% 0.2%
0.0% 0.0% 0.1%
CLM
Mean 5th 95th
E
3.0% 1.7% 4.4%
1.3% 0.7% 1.8%
1.3% 0.7% 1.8%
1.3% 0.7% 1.8%
1.3% 0.7% 1.8%
0.8% 0.1% 1.6%
0.8% 0.1% 1.6%
0.8% 0.1% 1.6%
0.8% 0.1% 1.6%
1.2% 0.5% 1.9%
Ozone
CL2
Mean 5th 95th
F
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
GAC20 + Advanced Disinfectants
CL2
Mean 5th 95th
P
0.7% 0.4% 1.0%
0.6% 0.3% 0.8%
0.6% 0.3% 0.8%
0.5% 0.3% 0.7%
0.5% 0.3% 0.7%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.3% 0.2% 0.5%
CLM
Mean 5th 95th
Q
0.5% 0.3% 0.7%
0.7% 0.4% 1.0%
0.7% 0.4% 1.0%
0.8% 0.5% 1.2%
0.8% 0.5% 1.2%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.5% 0.3% 0.7%
CLM
Mean 5th 95th
G
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
MF/UF
CL2
Mean 5th 95th
H
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
Membranes
CL2
CLM
Mean 5th 95th Mean 5th 95th
R S
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.1% 0.0% 0.1%
0.1% 0.0% 0.1%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
CLM
Mean 5th 95th
I
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
Total Converting to CLM
Mean 5th 95th
T=A+C+E+G-H+K+M-i-O+
Q+S
5.4% 3.0% 7.8%
6.5% 3.6% 9.3%
6.5% 3.6% 9.3%
7.2% 4.0% 10.4%
7.2% 4.0% 10.4%
10.3% 5.6% 14.6%
10.3% 5.6% 14.6%
10.3% 5.6% 14.6%
10.3% 5.6% 14.6%
8.2% 4.5% 11.7%
GAC10
CL2
Mean 5th 95th
J
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
CLM
Mean 5th 95th
K
/
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
Plants Making Treatment Technology Changes
Mean 5th 95th
Mean 5th 95th
L=SUM(A:S)
10.2% 5.7% 14.6%
8.4% 4.7% 12.1%
8.4% 4.7% 12.1%
8.8% 4.9% 12.7%
8.8% 4.9% 12.7%
14.6% 6.8% 22.8%
14.6% 6.8% 22.8%
14.6% 6.8% 22.8%
14.6% 6.8% 22.8%
11.1% 5.7% 16.6%
8.8% 4.9% 12.7%
14.6% 6.8% 22.8%
11.1% 5.7% 16.6%
Note: Detail may not add to totals due to independent rounding
Source: Treatment Technology Selection for the Preferred Alternative minus the Stage 1 Treatment Technology Selection from Appendix C, Exhibit C. 1a.
Exhibit 5.11 b
Stage 2 DBPR Treatment Technology Selection Deltas for CWS Surface Water Plants (Number of Plants by Residual Disinfection Type)
Preferred Alternative
System Size
(Population
Served)
<100
100-499
500-999
1,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1, 000,000
Total Plants
System Size
(Population
Served)
<100
100-499
500-999
1,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1, 000,000
Total Plants
Converting to CLM Only
Mean 5th 95th
A
7 4 10
31 17 45
20 11 28
48 27 69
53 30 77
112 62 155
50 28 69
53 29 73
649
379 21 1 534
Chlorine Dioxide
CL2
Mean 5th 95th
B
1 1 1
1 0 1
2 1 3
2 1 3
000
000
000
000
638
GAC10 -(-Advanced Disinfectants
CL2
CLM
Mean 5th 95th Mean 5th 95th
L M
15 7 23
7 3 10
7411
1 0 1
30 15 45
639
3 1 4
3 1 4
0 0 1
12 5 18
CLM
Mean 5th 95th
C
324
2 1 3
10 6 15
11 6 16
4 6 1
2 3 1
2 3 1
000
34 26 40
UV
CL2
Mean 5th 95th
D
15 8 21
9 5 13
638
11 6 15
12 7 17
37 8 74
17 4 33
17 4 35
204
125 46 221
GAC20
CL2
Mean 5th 95th
N
000
000
000
000
000
309
1 0 4
1 0 4
0 0 1
6 0 18
CLM
Mean 5th 95th
O
000
000
000
000
000
1 0 3
1 0 1
1 0 1
000
3 1 6
CLM
Mean 5th 95th
E
11 6 16
10 5 14
639
14 8 20
16 9 23
10 2 21
5 1 9
5 1 10
1 0 1
77 35 123
Ozone
CL2
Mean 5th 95th
F
000
000
000
000
000
000
000
000
000
GAC20 + Advanced Disinfectants
CL2
Mean 5th 95th
P
2 1 3
426
3 1 4
538
639
000
000
000
000
21 12 30
CLM
Mean 5th 95th
Q
2 1 3
538
325
9 5 13
10 6 15
000
000
000
000
30 17 43
CLM
Mean 5th 95th
G
000
000
000
000
000
000
000
000
000
MF/UF
CL2
Mean 5th 95th
H
000
000
000
000
000
000
000
000
000
000
Membranes
CL2
CLM
Mean 5th 95th Mean 5th 95th
R S
000
000
000
000
000
000
000
000
000
000
000
0 0 1
000
000
000
000
000
000
000
1 0 1
CLM
Mean 5th 95th
I
000
000
000
000
000
000
000
000
000
000
Total Converting to CLM
Mean 5th 95th
T=A+C+E+G-H+K+M-i-O+
Q+S
19 11 28
50 28 72
31 17 45
81 45 117
91 51 131
133 72 189
60 33 85
63 34 89
8411
536 296 766
GAC10
CL2
Mean 5th 95th
J
CLM
Mean 5th 95th
K
000
000
000
000
000
000
000
000
000000
Plants Making Treatment Technology Changes
Mean 5th 95th
Mean 5th 95th
L=SUM(A:S)
36 20 53
64 36 92
40 23 58
100 56 143
111 62 160
188 88 294
84 40 132
89 42 139
11 5 17
724 371 1,088
352 196 506
373 174 582
724 371 1,088
Note: Detail may not add to totals due to independent rounding
Source: Above table with Treatment Technologies switching from an advanced Treatment Technology with CI2 to the same advanced Treatment Technology with CLM being moved into the CLM only column
Final Economic Analysis for the Stage 2 DBPR
5-26
December 2005
-------
Exhibit 5.11c
Stage 2 DBPR Treatment Technology Selection Deltas for NTNCWS Surface Water Plants (Percent of Plants by Residual Disinfection Type)
Preferred Alternative
System Size
(Population
Served)
<100
100-499
500-999
1,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1, 000,000
Total %
System Size
(Population
Served)
<100
100-499
500-999
1,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1, 000,000
Total %
Converting to CLM Only
Mean 5th 95th
A
1.9% 1.0% 2.7%
4.1% 2.3% 5.9%
4.1% 2.3% 5.9%
4.2% 2.4% 6.1%
4.2% 2.4% 6.1%
8.6% 4.8% 12.0%
0.0% 0.0% 0.0%
8.6% 4.8% 12.0%
0.0% 0.0% 0.0%
3.5% 1.9% 5.0%
Chlorine Dioxide
CL2
Mean 5th 95th
B
0.1% 0.1% 0.2%
0.1% 0.1% 0.2%
0.2% 0.1% 0.2%
0.2% 0.1% 0.2%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.1% 0.1% 0.1%
GAC10 -(-Advanced Disinfectants
CL2
Mean 5th 95th
CLM
Mean 5th 95th
L M
,-'
/
1.2% 0.6% 1.7%
0.0% 0.0% 0.0%
1.2% 0.6% 1.7%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.5% 0.2% 0.7%
0.0% 0.0% 0.0%
0.5% 0.2% 0.7%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
CLM
Mean 5th 95th
C
0.4% 0.2% 0.5%
0.4% 0.2% 0.5%
0.9% 0.5% 1.3%
0.9% 0.5% 1.3%
0.3% 0.4% 0.1%
0.0% 0.0% 0.0%
0.3% 0.4% 0.1%
0.0% 0.0% 0.0%
0.3% 0.2% 0.5%
uv
CL2
Mean 5th 95th
D
4.1% 2.3% 5.9%
1.2% 0.7% 1.8%
1.2% 0.7% 1.8%
1.0% 0.5% 1.4%
1.0% 0.5% 1.4%
2.8% 0.6% 5.7%
0.0% 0.0% 0.0%
2.8% 0.6% 5.7%
0.0% 0.0% 0.0%
2.0% 1.1% 2.9%
GAC20
CL2
Mean 5th 95th
N
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.2% 0.0% 0.7%
0.0% 0.0% 0.0%
0.2% 0.0% 0.7%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
CLM
Mean 5th 95th
O
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.1% 0.0% 0.2%
0.0% 0.0% 0.0%
0.1% 0.0% 0.2%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
CLM
Mean 5th 95th
E
3.0% 1.7% 4.4%
1.3% 0.7% 1.8%
1.3% 0.7% 1.8%
1.3% 0.7% 1.8%
1.3% 0.7% 1.8%
0.8% 0.1% 1.6%
0.0% 0.0% 0.0%
0.8% 0.1% 1.6%
0.0% 0.0% 0.0%
1.8% 1.0% 2.6%
Ozone
CL2
Mean 5th 95th
F
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
GAC20 + Advanced Disinfectants
CL2
Mean 5th 95th
P
0.7% 0.4% 1.0%
0.6% 0.3% 0.8%
0.6% 0.3% 0.8%
0.5% 0.3% 0.7%
0.5% 0.3% 0.7%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.6% 0.3% 0.8%
CLM
Mean 5th 95th
Q
0.5% 0.3% 0.7%
0.7% 0.4% 1.0%
0.7% 0.4% 1.0%
0.8% 0.5% 1.2%
0.8% 0.5% 1.2%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.7% 0.4% 0.9%
CLM
Mean 5th 95th
G
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
MF/UF
CL2
Mean 5th 95th
H
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
Membranes
CL2
CLM
Mean 5th 95th Mean 5th 95th
R S
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.1% 0.0% 0.1%
0.1% 0.0% 0.1%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
CLM
Mean 5th 95th
I
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
Total Converting to CLM
Mean 5th 95th
T=A+C+E+G-H+K+M-i-O+
Q+S
5.4% 3.0% 7.8%
6.5% 3.6% 9.3%
6.5% 3.6% 9.3%
7.2% 4.0% 10.4%
7.2% 4.0% 10.4%
10.3% 5.6% 14.6%
0.0% 0.0% 0.0%
10.3% 5.6% 14.6%
0.0% 0.0% 0.0%
6.3% 3.5% 9.1%
GAC10
CL2
Mean 5th 95th
J
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
CLM
Mean 5th 95th
K
/
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
Plants Making Treatment Technology Changes
Mean 5th 95th
Mean 5th 95th
L=SUM(A:S)
10.2% 5.7% 14.6%
8.4% 4.7% 12.1%
8.4% 4.7% 12.1%
8.8% 4.9% 12.7%
8.8% 4.9% 12.7%
14.6% 6.8% 22.8%
0.0% 0.0% 0.0%
14.6% 6.8% 22.8%
0.0% 0.0% 0.0%
9.0% 5.0% 13.0%
9.0% 5.0% 12.9%
14.6% 6.8% 22.8%
9.0% 5.0% 13.0%
Note: Detail may not add to totals due to independent rounding
Source: Treatment Technology Selection for the Preferred Alternative minus the Stage 1 Treatment Technology Selection from Appendix C, Exhibit C. 1b.
Exhibit 5.11d
Stage 2 DBPR Treatment Technology Selection Deltas for NTNCWS Surface Water Plants (Number of Plants by Residual Disinfection Type)
Preferred Alternative
System Size
(Population
Served)
<100
100-499
500-999
1,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1, 000,000
Total Plants
System Size
(Population
Served)
<100
100-499
500-999
1,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1, 000,000
Total Plants
Converting to CLM Only
Mean 5th 95th
A
426
13 7 18
426
426
1 1 2
0 0 1
000
000
000
27 15 39
Chlorine Dioxide
CL2
Mean 5th 95th
B
0 0 1
000
000
000
000
000
000
000
1 0 1
GAC10 -(-Advanced Disinfectants
CL2
CLM
Mean 5th 95th Mean 5th 95th
L M
000
000
000
000
000
000
000
000
000
000
CLM
Mean 5th 95th
C
1 1 2
0 0 1
1 0 1
000
000
000
000
000
3 1 4
uv
CL2
Mean 5th 95th
D
9 5 13
425
1 1 2
1 0 1
000
000
000
000
000
16 9 23
GAC20
CL2
Mean 5th 95th
N
000
000
000
000
000
000
000
000
000
000
CLM
Mean 5th 95th
O
000
000
000
000
000
000
000
000
000
000
CLM
Mean 5th 95th
E
7 4 10
426
1 1 2
1 1 2
000
000
000
000
000
14 8 20
Ozone
CL2
Mean 5th 95th
F
000
000
000
000
000
000
000
000
000
GAC20 + Advanced Disinfectants
CL2
Mean 5th 95th
P
2 1 2
2 1 2
1 0 1
0 0 1
000
000
000
000
000
426
CLM
Mean 5th 95th
Q
1 1 2
2 1 3
1 0 1
1 0 1
000
000
000
000
000
537
CLM
Mean 5th 95th
G
000
000
000
000
000
000
000
000
000
MF/UF
CL2
Mean 5th 95th
H
000
000
000
000
000
000
000
000
000
000
Membranes
CL2
CLM
Mean 5th 95th Mean 5th 95th
R S
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
CLM
Mean 5th 95th
I
000
000
000
000
000
000
000
000
000
000
Total Converting to CLM
Mean 5th 95th
T=A+C+E+G-H+K+M-i-O+
Q+S
12 7 18
20 11 29
7 4 10
7 4 10
2 1 3
1 0 1
000
000
000
48 27 70
GAC10
CL2
Mean 5th 95th
J
CLM
Mean 5th 95th
K
000
000
000
000
000
000
000
000
000000
Plants Making Treatment Technology Changes
Mean 5th 95th
Mean 5th 95th
L=SUM(A:S)
23 13 33
26 15 38
9 5 13
8 5 12
2 1 3
1 0 1
000
000
000
69 39 100
68 38 98
1 0 1
69 39 100
Note: Detail may not add to totals due to independent rounding
Source: Above table with Treatment Technologies switching from an advanced Treatment Technology with CI2 to the same advanced Treatment Technology with CLM being moved into the CLM only column
Final Economic Analysis for the Stage 2 DBPR
5-27
December 2005
-------
Exhibit 5.12,1
Post-Stage 2 DBPR Treatment Technologies-in-Place for CWS Surface Water Plants (Percent of Plants by Residual Disinfection Type|
Preferred Alternative
System Size
(Population Served)
<100
100-499
500-999
1,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1, 000,000
Total %
System Size
(Population Served)
<100
100-499
500-999
1,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1, 000,000
Total %
No Advanced Treatment
Technologies CL21
Mean 5th 95th
A
31.7% 27.2% 36.2%
27.2% 23.5% 30.9%
27.2% 23.5% 30.9%
24.6% 20.7% 28.5%
24.6% 20.7% 28.5%
29.3% 25.5% 33.2%
29.3% 25.5% 33.2%
29.3% 25.5% 33.2%
29.3% 25.5% 33.2%
27.3% 23.5% 31.2%
GAC10 + ADCL2
Mean 5th 95th
M
0.9% 0.8% 1.0%
0.9% 0.8% 1.0%
0.9% 0.8% 1.0%
0.9% 0.8% 1.0%
0.4% 0.3% 0.4%
No Advanced Treatment
Technologies CLM1
Mean 5th 95th
B
31.6% 30.8% 32.4%
39.5% 37.7% 41.3%
39.5% 37.7% 41.3%
45.6% 43.7% 47.4%
45.6% 43.7% 47.4%
41.2% 41.0% 41.4%
41.2% 41.0% 41.4%
41.2% 41.0% 41.4%
41.2% 41.0% 41.4%
41.9% 40.8% 43.1%
GAC10 + ADCLM
Mean 5th 95th
N
1.3% 1.0% 1.5%
1.3% 1.0% 1.5%
1.3% 1.0% 1.5%
1.3% 1.0% 1.5%
0.5% 0.4% 0.6%
Chlorine Dioxide
CL2
Mean 5th 95th
C
1.1% 1.0% 1.1%
1.1% 1.0% 1.1%
2.1% 2.0% 2.2%
2.1% 2.0% 2.2%
2.7% 2.3% 3.0%
2.7% 2.3% 3.0%
2.7% 2.3% 3.0%
2.7% 2.3% 3.0%
2.0% 1.8% 2.2%
GAC20 CL2
Mean 5th 95th
0
2.0% 2.0% 2.0%
1.1% 1.1% 1.1%
1.1% 1.1% 1.1%
1.0% 1.0% 1.0%
1.0% 1.0% 1.0%
0.2% 0.1% 0.3%
0.2% 0.1% 0.3%
0.2% 0.1% 0.3%
0.2% 0.1% 0.3%
0.8% 0.7% 0.8%
Chlorine Dioxide
CLM
Mean 5th 95th
D
1.2% 1.1% 1.4%
1.2% 1.1% 1.4%
3.0% 2.6% 3.4%
3.0% 2.6% 3.4%
3.7% 3.5% 4.0%
3.7% 3.5% 4.0%
3.7% 3.5% 4.0%
3.7% 3.5% 4.0%
2.8% 2.5% 3.1%
GAC20 CLM
Mean 5th 95th
P
1.3% 1.3% 1.3%
1.0% 1.0% 1.0%
1.0% 1.0% 1.0%
1.2% 1.2% 1.2%
1.2% 1.2% 1.2%
0.3% 0.2% 0.4%
0.3% 0.2% 0.4%
0.3% 0.2% 0.4%
0.3% 0.2% 0.4%
0.8% 0.8% 0.8%
UVCL2
Mean 5th 95th
E
4.1% 2.3% 5.9%
1.2% 0.7% 1.8%
1.2% 0.7% 1.8%
1.0% 0.5% 1.4%
1.0% 0.5% 1.4%
1.1% 0.3% 1.9%
1.1% 0.3% 1.9%
1.1% 0.3% 1.9%
1.1% 0.3% 1.9%
1.2% 0.6% 1.9%
GAC20 + AD CL2
Mean 5th 95th
Q
0.7% 0.4% 1 .0%
1.0% 0.8% 1.3%
1.0% 0.8% 1.3%
1.0% 0.8% 1.2%
1.0% 0.8% 1.2%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.6% 0.5% 0.7%
UVCLM
Mean 5th 95th
F
3.0% 1.7% 4.4%
1.3% 0.7% 1.8%
1.3% 0.7% 1.8%
1.3% 0.7% 1.8%
1.3% 0.7% 1.8%
1.6% 0.4% 2.9%
1.6% 0.4% 2.9%
1.6% 0.4% 2.9%
1.6% 0.4% 2.9%
1.5% 0.6% 2.4%
GAC20 + AD CLM
Mean 5th 95th
R
0.5% 0.3% 0.7%
1.1% 0.8% 1.4%
1.1% 0.8% 1.4%
1.4% 1.0% 1.8%
1.4% 1.0% 1.8%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.8% 0.6% 1.0%
Ozone CL2
Mean 5th 95th
G
5.1% 5.1% 5.1%
5.1% 5.1% 5.1%
4.0% 4.0% 4.0%
4.0% 4.0% 4.0%
5.3% 5.1% 5.5%
5.3% 5.1% 5.5%
5.3% 5.1% 5.5%
5.3% 5.1% 5.5%
4.5% 4.4% 4.6%
Membranes CL2
Mean 5th 95th
S
2.1% 2.1% 2.1%
0.5% 0.5% 0.5%
0.5% 0.5% 0.5%
0.2% 0.2% 0.2%
0.2% 0.2% 0.2%
0.3% 0.3% 0.3%
0.3% 0.3% 0.3%
0.3% 0.3% 0.3%
0.3% 0.3% 0.3%
0.4% 0.4% 0.4%
Ozone CLM
Mean 5th 95th
H
4.6% 4.6% 4.6%
4.6% 4.6% 4.6%
4.5% 4.5% 4.5%
4.5% 4.5% 4.5%
7.5% 7.3% 7.7%
7.5% 7.3% 7.7%
7.5% 7.3% 7.7%
7.5% 7.3% 7.7%
5.4% 5.3% 5.5%
Membranes CLM
Mean 5th 95th
T
1.4% 1.4% 1.4%
0.5% 0.4% 0.5%
0.5% 0.4% 0.5%
0.2% 0.2% 0.2%
0.2% 0.2% 0.2%
0.4% 0.4% 0.4%
0.4% 0.4% 0.4%
0.4% 0.4% 0.4%
0.4% 0.4% 0.4%
0.4% 0.4% 0.4%
MF/UF CL2
Mean 5th 95th
I
14.5% 14.5% 14.5%
8.9% 8.9% 8.9%
8.9% 8.9% 8.9%
6.2% 6.2% 6.2%
6.2% 6.2% 6.2%
0.8% 0.7% 0.8%
0.8% 0.7% 0.8%
0.8% 0.7% 0.8%
0.8% 0.7% 0.8%
5.0% 5.0% 5.1%
MF/UF CLM
Mean 5th 95th
J
7.1% 7.1% 7.1%
4.8% 4.8% 4.8%
4.8% 4.8% 4.8%
2.9% 2.9% 2.9%
2.9% 2.9% 2.9%
1.1% 1.0% 1.1%
1.1% 1.0% 1.1%
1.1% 1.0% 1.1%
1.1% 1.0% 1.1%
2.8% 2.8% 2.8%
TOTAL CL2
Mean 5th 95th
U = A+C+E+G+I+K+M+O+Q+S
55.0% 48.5% 61.6%
46.0% 41.5% 50.6%
46.0% 41.5% 50.6%
40.0% 35.5% 44.7%
40.0% 35.5% 44.7%
41.6% 36.1% 47.1%
41.6% 36.1% 47.1%
41.6% 36.1% 47.1%
41.6% 36.1% 47.1%
42.6% 37.6% 47.7%
GAC10CL2
Mean 5th 95th
K
0.9% 0.9% 0.9%
0.9% 0.9% 0.9%
0.9% 0.9% 0.9%
0.9% 0.9% 0.9%
0.4% 0.3% 0.4%
GAC10CLM
Mean 5th 95th
L
,-'
1.3% 1.2% 1.3%
1.3% 1.2% 1.3%
1.3% 1.2% 1.3%
1.3% 1.2% 1.3%
0.5% 0.5% 0.5%
TOTAL CLM
Mean 5th 95th
V = B+D+F+H+J+L+N+P+R+T
45.0% 42.6% 47.3%
54.0% 51.1% 56.8%
54.0% 51.1% 56.8%
60.0% 56.8% 63.1%
60.0% 56.8% 63.1%
58.4% 56.1% 60.7%
58.4% 56.1% 60.7%
58.4% 56.1% 60.7%
58.4% 56.1% 60.7%
57.4% 54.7% 60.1%
Note: Detail may not add to totals due to independent rounding
'No Advanced Treatment Technologies includes conventional, non-conventional, and softening plants.
Source: Surface water systems serving <10,000 people: Add Treatment Technologies-in-Place for the Pre-Stage 2 Baseline (Exhibit 5.10a) to the Treatment Technology Selection Delta for the Preferred Alternative. Surface water systems serving 10,000 or more people: Use ending Treatment
Technology predictions from SWAT (FACA Screen SeriesS v3.0 Database) for the Preferred Alternative.
Exhibit 5.12b
Post-Stage 2 DBPR Treatment Technologies-in-Place for CWS Surface Water Plants (Number of Plants by Residual Disinfection Type)
Preferred Alternative
System Size
(Population Served)
<100
100-499
500-999
1,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1, 000,000
Total Plants
System Size
(Population Served)
<100
100-499
500-999
1,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1, 000,000
Total Plants
No Advanced Treatment
Technologies CL21
Mean 5th 95th
A
114 98 130
209 180 237
131 114 149
278 234 322
310 261 359
379 329 429
170 147 193
179 155 203
22 19 24
1,792 1,538 2,046
GAC10 + ADCL2
Mean 5th 95th
M
12 10 13
556
656
1 1 1
23 20 26
No Advanced Treatment
Technologies CLM1
Mean 5th 95th
B
114 111 117
303 289 317
191 182 199
515 494 536
574 550 597
532 529 534
239 237 240
251 250 253
30 30 30
2,747 2,672 2,822
GAC10 + ADCLM
Mean 5th 95th
N
17 13 20
769
869
1 1 1
33 27 39
Chlorine Dioxide
CL2
Mean 5th 95th
C
889
555
23 23 24
26 25 27
35 30 39
15 14 17
16 14 18
222
131 120 142
GAC20 CL2
Mean 5th 95th
0
111
888
555
12 12 12
13 13 13
324
1 1 2
1 1 2
000
51 49 53
Chlorine Dioxide
CLM
Mean 5th 95th
D
9 8 11
657
34 30 39
38 33 43
48 45 51
22 20 23
23 21 24
333
183 166 200
GAC20 CLM
Mean 5th 95th
P
555
111
555
13 13 13
15 15 15
435
2 1 2
2 1 3
000
52 50 55
UVCL2
Mean 5th 95th
E
15 8 21
9 5 13
638
11 6 15
12 7 17
14 4 25
6 2 11
7 2 12
1 0 1
81 38 124
GAC20 + AD CL2
Mean 5th 95th
Q
2 1 3
8 6 10
546
11 9 14
13 10 15
000
000
000
000
39 30 49
UVCLM
Mean 5th 95th
F
11 6 16
10 5 14
639
14 8 20
16 9 23
21 5 37
9 2 17
10 3 17
1 0 2
99 42 155
GAC20 + AD CLM
Mean 5th 95th
R
2 1 3
9 6 11
547
16 12 20
18 13 22
000
000
000
000
49 36 63
Ozone CL2
Mean 5th 95th
G
39 39 39
24 24 24
45 45 45
50 50 50
69 66 72
31 30 32
33 31 34
444
295 290 301
Membranes CL2
Mean 5th 95th
S
888
333
222
222
222
444
222
222
000
25 25 26
Ozone CLM
Mean 5th 95th
H
35 35 35
22 22 22
51 51 51
56 56 56
97 94 99
43 42 45
46 44 47
656
356 350 361
Membranes CLM
Mean 5th 95th
T
555
434
222
222
222
656
223
333
000
26 26 27
MF/UF CL2
Mean 5th 95th
I
52 52 52
68 68 68
43 43 43
70 70 70
78 78 78
10 9 10
445
545
1 1 1
330 329 331
MF/UF CLM
Mean 5th 95th
J
26 26 26
37 37 37
23 23 23
32 32 32
36 36 36
14 13 14
666
767
1 1 1
181 181 182
TOTAL CL2
Mean 5th 95th
U = A+C+E+G+I+K+M+O+Q+S
198 174 221
353 318 388
222 201 244
452 400 504
504 446 562
537 466 608
241 209 273
254 220 287
31 27 35
2,792 2,461 3,122
GAC10CL2
Mean 5th 95th
K
12 11 12
556
656
1 1 1
23 22 24
GAC10CLM
Mean 5th 95th
L
,-'
17 16 17
778
888
1 1 1
33 32 34
TOTAL CLM
Mean 5th 95th
V = B+D+F+H+J+L+N+P+R+T
162 153 170
414 392 436
261 247 274
677 641 713
754 714 794
755 725 784
339 325 352
357 343 371
43 41 45
3,760 3,582 3,938
Note: Detail may not add to totals due to independent rounding
'No advanced Treatment Technologies includes conventional, non-conventional, and softening plants.
Source: Surface water systems serving <10,000 people: Add Treatment Technologies-in-Place for the Pre-Stage 2 Baseline (Exhibit 5.10a) to the Treatment Technology Selection Delta for the Preferred Alternative. Surface water systems serving 10,000 or more people: Use ending Treatment
Technology predictions from SWAT (FACA Screen SeriesS v3.0 Database) for the Preferred Alternative.
December 2005
Final Economic Analysis for the Stage 2 DBPR
5-28
-------
Exhibit 5.12c
Post-Stage 2 DBPR Treatment Technologies-in-Place for NTNCWS Surface Water Plants (Percent of Plants by Residual Disinfection Type|
Preferred Alternative
System Size
(Population Served)
<100
100-499
500-999
1,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1, 000,000
Total %
System Size
(Population Served)
<100
100-499
500-999
1,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1, 000,000
Total %
No Advanced Treatment
Technologies CL21
Mean 5th 95th
A
31.7% 27.2% 36.2%
27.2% 23.5% 30.9%
27.2% 23.5% 30.9%
24.6% 20.7% 28.5%
24.6% 20.7% 28.5%
29.3% 25.5% 33.2%
0.0% 0.0% 0.0%
29.3% 25.5% 33.2%
0.0% 0.0% 0.0%
28.2% 24.2% 32.1%
GAC10 + ADCL2
Mean 5th 95th
M
0.9% 0.8% 1.0%
0.0% 0.0% 0.0%
0.9% 0.8% 1.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
No Advanced Treatment
Technologies CLM1
Mean 5th 95th
B
31.6% 30.8% 32.4%
39.5% 37.7% 41.3%
39.5% 37.7% 41.3%
45.6% 43.7% 47.4%
45.6% 43.7% 47.4%
41.2% 41.0% 41.4%
0.0% 0.0% 0.0%
41.2% 41.0% 41.4%
0.0% 0.0% 0.0%
38.1% 36.6% 39.6%
GAC10 + ADCLM
Mean 5th 95th
N
1.3% 1.0% 1.5%
0.0% 0.0% 0.0%
1.3% 1.0% 1.5%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
Chlorine Dioxide
CL2
Mean 5th 95th
C
1.1% 1.0% 1.1%
1.1% 1.0% 1.1%
2.1% 2.0% 2.2%
2.1% 2.0% 2.2%
2.7% 2.3% 3.0%
0.0% 0.0% 0.0%
2.7% 2.3% 3.0%
0.0% 0.0% 0.0%
0.9% 0.9% 1.0%
GAC20 CL2
Mean 5th 95th
0
2.0% 2.0% 2.0%
1.1% 1.1% 1.1%
1.1% 1.1% 1.1%
1.0% 1.0% 1.0%
1.0% 1.0% 1.0%
0.2% 0.1% 0.3%
0.0% 0.0% 0.0%
0.2% 0.1% 0.3%
0.0% 0.0% 0.0%
1.3% 1.3% 1.3%
Chlorine Dioxide
CLM
Mean 5th 95th
D
1.2% 1.1% 1.4%
1.2% 1.1% 1.4%
3.0% 2.6% 3.4%
3.0% 2.6% 3.4%
3.7% 3.5% 4.0%
0.0% 0.0% 0.0%
3.7% 3.5% 4.0%
0.0% 0.0% 0.0%
1.2% 1.0% 1.3%
GAC20 CLM
Mean 5th 95th
P
1.3% 1.3% 1.3%
1.0% 1.0% 1.0%
1.0% 1.0% 1.0%
1.2% 1.2% 1.2%
1.2% 1.2% 1.2%
0.3% 0.2% 0.4%
0.0% 0.0% 0.0%
0.3% 0.2% 0.4%
0.0% 0.0% 0.0%
1.1% 1.1% 1.1%
UVCL2
Mean 5th 95th
E
4.1% 2.3% 5.9%
1.2% 0.7% 1.8%
1.2% 0.7% 1.8%
1.0% 0.5% 1.4%
1.0% 0.5% 1.4%
1.1% 0.3% 1.9%
0.0% 0.0% 0.0%
1.1% 0.3% 1.9%
0.0% 0.0% 0.0%
2.0% 1.1% 2.9%
GAC20 + AD CL2
Mean 5th 95th
Q
0.7% 0.4% 1 .0%
1.0% 0.8% 1.3%
1.0% 0.8% 1.3%
1.0% 0.8% 1.2%
1.0% 0.8% 1.2%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.9% 0.7% 1 .2%
UVCLM
Mean 5th 95th
F
3.0% 1.7% 4.4%
1.3% 0.7% 1.8%
1.3% 0.7% 1.8%
1.3% 0.7% 1.8%
1.3% 0.7% 1.8%
1.6% 0.4% 2.9%
0.0% 0.0% 0.0%
1.6% 0.4% 2.9%
0.0% 0.0% 0.0%
1.8% 1.0% 2.6%
GAC20 + AD CLM
Mean 5th 95th
R
0.5% 0.3% 0.7%
1.1% 0.8% 1.4%
1.1% 0.8% 1.4%
1.4% 1.0% 1.8%
1.4% 1.0% 1.8%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
1.0% 0.7% 1.3%
Ozone CL2
Mean 5th 95th
G
5.1% 5.1% 5.1%
5.1% 5.1% 5.1%
4.0% 4.0% 4.0%
4.0% 4.0% 4.0%
5.3% 5.1% 5.5%
0.0% 0.0% 0.0%
5.3% 5.1% 5.5%
0.0% 0.0% 0.0%
3.4% 3.4% 3.4%
Membranes CL2
Mean 5th 95th
S
2.1% 2.1% 2.1%
0.5% 0.5% 0.5%
0.5% 0.5% 0.5%
0.2% 0.2% 0.2%
0.2% 0.2% 0.2%
0.3% 0.3% 0.3%
0.0% 0.0% 0.0%
0.3% 0.3% 0.3%
0.0% 0.0% 0.0%
0.9% 0.9% 0.9%
Ozone CLM
Mean 5th 95th
H
4.6% 4.6% 4.6%
4.6% 4.6% 4.6%
4.5% 4.5% 4.5%
4.5% 4.5% 4.5%
7.5% 7.3% 7.7%
0.0% 0.0% 0.0%
7.5% 7.3% 7.7%
0.0% 0.0% 0.0%
3.2% 3.2% 3.2%
Membranes CLM
Mean 5th 95th
T
1.4% 1.4% 1.4%
0.5% 0.4% 0.5%
0.5% 0.4% 0.5%
0.2% 0.2% 0.2%
0.2% 0.2% 0.2%
0.4% 0.4% 0.4%
0.0% 0.0% 0.0%
0.4% 0.4% 0.4%
0.0% 0.0% 0.0%
0.7% 0.7% 0.7%
MF/UF CL2
Mean 5th 95th
I
14.5% 14.5% 14.5%
8.9% 8.9% 8.9%
8.9% 8.9% 8.9%
6.2% 6.2% 6.2%
6.2% 6.2% 6.2%
0.8% 0.7% 0.8%
0.0% 0.0% 0.0%
0.8% 0.7% 0.8%
0.0% 0.0% 0.0%
10.1% 10.1% 10.1%
MF/UF CLM
Mean 5th 95th
J
7.1% 7.1% 7.1%
4.8% 4.8% 4.8%
4.8% 4.8% 4.8%
2.9% 2.9% 2.9%
2.9% 2.9% 2.9%
1.1% 1.0% 1.1%
0.0% 0.0% 0.0%
1.1% 1.0% 1.1%
0.0% 0.0% 0.0%
5.2% 5.2% 5.2%
TOTAL CL2
Mean 5th 95th
U = A+C+E+G+I+K+M+O+Q+S
55.0% 48.5% 61.6%
46.0% 41.5% 50.6%
46.0% 41.5% 50.6%
40.0% 35.5% 44.7%
40.0% 35.5% 44.7%
41.6% 36.1% 47.1%
0.0% 0.0% 0.0%
41.6% 36.1% 47.1%
0.0% 0.0% 0.0%
47.7% 42.6% 52.9%
GAC10CL2
Mean 5th 95th
K
0.9% 0.9% 0.9%
0.0% 0.0% 0.0%
0.9% 0.9% 0.9%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
GAC10CLM
Mean 5th 95th
L
,-'
1.3% 1.2% 1.3%
0.0% 0.0% 0.0%
1.3% 1.2% 1.3%
0.0% 0.0% 0.0%
0.0% 0.0% 0.0%
TOTAL CLM
Mean 5th 95th
V = B+D+F+H+J+L+N+P+R+T
45.0% 42.6% 47.3%
54.0% 51.1% 56.8%
54.0% 51.1% 56.8%
60.0% 56.8% 63.1%
60.0% 56.8% 63.1%
58.4% 56.1% 60.7%
0.0% 0.0% 0.0%
58.4% 56.1% 60.7%
0.0% 0.0% 0.0%
52.3% 49.5% 55.0%
Note: Detail may not add to totals due to independent rounding
'No Advanced Treatment Technologies includes conventional, non-conventional, and softening plants.
Source: Surface water systems serving <10,000 people: Add Treatment Technologies-in-Place for the Pre-Stage 2 Baseline (Exhibit 5.1 Ob) to the Treatment Technology Selection Delta for the Preferred Alternative. Surface water systems serving 10,000 or more people: Use ending Treatment
Technology predictions from SWAT (FACA Screen SeriesS v3.0 Database) for the Preferred Alternative.
Exhibit 5.12d
Post-Stage 2 DBPR Treatment Technologies-in-Place for NTNCWS Surface Water Plants (Number of Plants by Residual Disinfection Type)
Preferred Alternative
System Size
(Population Served)
<100
100-499
500-999
1,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1, 000,000
Total Plants
System Size
(Population Served)
<100
100-499
500-999
1,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1, 000,000
Total Plants
No Advanced Treatment
Technologies CL21
Mean 5th 95th
A
72 62 82
85 73 96
29 25 33
23 19 26
657
1 1 2
000
000
000
216 186 246
GAC10 + ADCL2
Mean 5th 95th
M
000
000
000
000
000
No Advanced Treatment
Technologies CLM1
Mean 5th 95th
B
71 70 73
123 118 129
42 40 44
42 40 44
11 11 12
222
000
000
000
292 281 304
GAC10 + ADCLM
Mean 5th 95th
N
000
000
000
000
000
Chlorine Dioxide
CL2
Mean 5th 95th
C
1 1 1
222
1 1 1
000
000
000
000
777
GAC20 CL2
Mean 5th 95th
0
444
333
1 1 1
1 1 1
000
000
000
000
000
10 10 10
Chlorine Dioxide
CLM
Mean 5th 95th
D
1 1 1
323
1 1 1
000
000
000
000
9 8 10
GAC20 CLM
Mean 5th 95th
P
333
333
1 1 1
1 1 1
000
000
000
000
000
888
UVCL2
Mean 5th 95th
E
9 5 13
425
1 1 2
1 0 1
000
000
000
000
000
16 9 22
GAC20 + AD CL2
Mean 5th 95th
Q
2 1 2
324
1 1 1
1 1 1
000
000
000
000
000
759
UVCLM
Mean 5th 95th
F
7 4 10
426
1 1 2
1 1 2
000
000
000
000
000
14 8 20
GAC20 + AD CLM
Mean 5th 95th
R
1 1 2
435
1 1 2
1 1 2
000
000
000
000
000
8 5 10
Ozone CL2
Mean 5th 95th
G
16 16 16
555
1 1 1
000
000
000
000
26 26 26
Membranes CL2
Mean 5th 95th
S
555
1 1 1
000
000
000
000
000
000
000
777
Ozone CLM
Mean 5th 95th
H
14 14 14
555
1 1 1
000
000
000
000
25 25 25
Membranes CLM
Mean 5th 95th
T
333
1 1 2
0 0 1
000
000
000
000
000
000
555
MF/UF CL2
Mean 5th 95th
I
33 33 33
28 28 28
999
666
222
000
000
000
000
77 77 77
MF/UF CLM
Mean 5th 95th
J
16 16 16
15 15 15
555
333
1 1 1
000
000
000
000
40 40 40
TOTAL CL2
Mean 5th 95th
U = A+C+E+G+I+K+M+O+Q+S
124 110 139
144 130 158
49 44 54
37 33 41
10 9 11
222
000
000
000
366 327 406
GAC10CL2
Mean 5th 95th
K
000
000
000
000
000
GAC10CLM
Mean 5th 95th
L
,-'
000
000
000
000
000
TOTAL CLM
Mean 5th 95th
V = B+D+F+H+J+L+N+P+R+T
102 96 107
168 159 177
57 54 60
55 52 58
15 14 16
333
000
1 1 1
000
401 380 422
Note: Detail may not add to totals due to independent rounding
'No Advanced Treatment Technologies includes conventional, non-conventional, and softening plants.
Source: Surface water systems serving <10,000 people: Add Treatment Technologies-in-Place for the Pre-Stage 2 Baseline (Exhibit 5.1 Ob) to the Treatment Technology Selection Delta for the Preferred Alternative. Surface water systems serving 10,000 or more people: Use ending Treatment
Technology predictions from SWAT (FACA Screen SeriesS v3.0 Database) for the Preferred Alternative.
December 2005
Final Economic Analysis for the Stage 2 DBPR
5-29
-------
Exhibit 5.13a Pre-Stage 2 DBPR Treatment Technologies-in-Place for CWS Ground Water Plants (Percent and Number of Plants, by Residual Disinfectant Type)
Preferred Alternative
System Size
(Population Served)
<100
100-499
500-999
1,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1, 000,000
Total Plants
No Advanced
Treatment
Technologies CL21
A
# #
95.9% 6,161
95.3% 14,522
5,806
95.7% 7,264
4,816
89.2% 4,801
639
89.5% 821
24
94.6% 44,854
No Advanced
Treatment
Technologies CLM1
B
% #
2.4% 155
2.8% 426
170
2.5% 192
127
7.2% 389
52
7.1% 65
2
3.3% 1 ,578
UVCL2
C
% #
0.0% 0
0.0% 0
0
0.0% 0
0
UVCLM
D
% #
0.0% 0
0.0% 0
0
0.0% 0
0
0.0% 0
0.0% 0
Ozone CL2
E
% #
0.0% 0
0.2% 25
10
0.3% 22
15
0.8% 46
6
0.8% 8
0
0.3% 130
Ozone CLM
F
% #
0.0% 0
0.5% 74
29
0.9% 66
44
0.8% 42
6
0.7% 6
0
0.6% 267
GAC20
CL2
G
% #
0.0% 0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
GAC20 CLM
H
% #
0.9% 56
0.6% 96
39
0.1% 4
3
0.0% 2
0
0.0% 0
0
0.4% 200
Membranes
CL2
I
% #
0.3% 22
0.1% 20
8
0.1% 4
3
1.7% 90
12
1.7% 15
0
0.4% 1 73
Membranes
CLM
J
% #
0.5% 29
0.5% 79
32
0.5% 36
24
0.3% 14
2
0.2% 2
0
0.5% 21 7
TOTAL USING
CL2
K = A+C+E+G+I
% #
96.3% 6,183
95.6% 14,567
5,823
96.1% 7,290
4,833
91.7% 4,936
657
92.0% 844
25
95.2% 45,157
TOTAL USING
CLM
L = B+D+F+H+J
% #
3.7% 240
4.4% 676
270
3.9% 297
197
8.3% 446
59
8.0% 74
2
4.8% 2,262
Note: Detail may not add to totals due to independent rounding
1No Advanced Treatment Technologies includes conventional, non-conventional, and softening plants.
Source: Add Treatment Technologies-in-Place for the Pre-Stage 1 DBPR Baseline (Exhibit 3.14a) to Stage 1 Treatment Technology Selection (Exhibit C.2a).
Exhibit 5.13b Pre-Stage 2 DBPR Treatment Technologies-in-Place for NTNCWS Ground Water Plants (Percent and Number of Plants, by Residual Disinfectant Type]
Preferred Alternative
System Size
(Population Served)
<100
100-499
500-999
1 ,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1 ,000,000
Total Plants
No Advanced
Treatment
Technologies CL21
A
# #
95.9% 2,391
95.3% 2,028
561
95.7% 237
21
89.2% 3
0
89.5% 0
0
95.6% 5,241
No Advanced
Treatment
Technologies CLM1
B
% #
2.4% 60
2.8% 60
16
2.5% 6
1
7.2% 0
0
7.1% 0
0
2.6% 143
UVCL2
C
% #
0.0% 0
0.0% 0
0
0.0% 0
0
UVCLM
D
% #
0.0% 0
0.0% 0
0
0.0% 0
0
/
0.0% 0
0.0% 0
Ozone CL2
E
% #
0.0% 0
0.2% 3
1
0.3% 1
0
0.8% 0
0
0.8% 0
0
0.1% 5
Ozone CLM
F
% #
0.0% 0
0.5% 10
3
0.9% 2
0
0.8% 0
0
0.7% 0
0
0.3% 15
GAC20
CL2
G
% #
0.0% 0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
0
0.0% 0
GAC20 CLM
H
% #
0.9% 22
0.6% 13
4
0.1% 0
0
0.0% 0
0
0.0% 0
0
0.7% 39
Membranes
CL2
I
% #
0.3% 8
0.1% 3
1
0.1% 0
0
1.7% 0
0
1.7% 0
0
0.2% 12
Membranes
CLM
J
% #
0.5% 1 1
0.5% 1 1
3
0.5% 1
0
0.3% 0
0
0.2% 0
0
0.5% 27
TOTAL USING
CL2
K = A+C+E+G+I
% #
96.3% 2,400
95.6% 2,035
563
96.1% 237
21
91.7% 3
0
92.0% 0
0
95.9% 5,259
TOTAL USING
CLM
L = B+D+F+H+J
% #
3.7% 93
4.4% 94
26
3.9% 10
1
8.3% 0
0
8.0% 0
0
4.1% 225
Note: Detail may not add to totals due to independent rounding
1No Advanced Treatment Technologies includes conventional, non-conventional, and softening plants.
Source: Add Treatment Technologies-in-Place for the Pre-Stage 1 DBPR Baseline (Exhibit 3.14b) to Stage 1 Treatment Technology Selection (Exhibit C.2b).
Final Economic Analysis for the Stage 2 DBPR
5-30
December 2005
-------
Exhibit 5.14a
Stage 2 DBPR Treatment Technology Selection Deltas for CWS Ground Water Plants (Percent of Plants, by Residual Disinfectant Type)
Preferred Alternative
System Size
(Population Served)
<100
100-499
500-999
1 ,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1, 000,000
Total %
CLM
Only
A
1 .0%
1 .4%
1 .4%
1.1%
1.1%
1 .4%
1 .4%
1 .3%
1 .4%
1 .3%
UVCL2
B
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
UVCLM
C
1.1%
1 .6%
1 .6%
1 .6%
1 .6%
1 .3%
Ozone
CL2
D
0.0%
0.0%
0.0%
0.0%
0.0%
0.1%
0.1%
0.1%
0.1%
0.0%
Ozone
CLM
E
0.0%
0.0%
0.0%
0.0%
0.0%
0.2%
0.2%
0.2%
0.2%
0.0%
GAC20
CL2
F
0.4%
0.2%
0.2%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.1%
GAC20
CLM
G
0.0%
0.0%
0.0%
0.0%
0.0%
0.2%
0.2%
0.1%
0.1%
0.0%
Membranes
CL2
H
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
Membranes
CLM
I
0.0%
0.0%
0.0%
0.0%
0.0%
0.2%
0.2%
0.2%
0.2%
0.0%
Total Converting
to CLM
J = A+C+E+G+I
2.1%
3.0%
3.0%
2.7%
2.7%
2.0%
2.0%
1 .9%
2.0%
2.6%
Plants Making
Treatment Technology
Changes
K=SUM(A:I)
2.4%
3.2%
3.2%
2.7%
2.7%
2.1%
2.1%
2.0%
2.1%
2.8%
2.9%
2.1%
2.8%
Note: Detail may not add to totals due to independent rounding.
Source: Treatment Technology Selection for the Preferred Alternative minus the Stage 1 Treatment Technology Selection from Appendix C, Exhibit C.2a.
Exhibit 5.14b
Stage 2 DBPR Treatment Technology Selection Deltas for CWS Ground Water Plants (Number of Plants, by Residual Disinfectant Type)
Preferred Alternative
System Size
(Population Served)
<100
100-499
500-999
1 ,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1, 000,000
Total Plants
CLM
Only
A
62
213
85
82
54
75
10
12
0
595
UVCL2
B
0
0
0
0
0
/
0
UVCLM
C
70
242
97
118
78
606
Ozone
CL2
D
0
0
0
0
0
3
0
0
0
4
Ozone
CLM
E
0
0
0
0
0
12
2
2
0
15
GAC20
CL2
F
23
27
11
0
0
0
0
0
0
61
GAC20
CLM
G
0
0
0
4
2
8
1
1
0
17
Membranes
CL2
H
0
0
0
0
0
2
0
0
0
2
Membranes
CLM
I
0
0
0
0
0
11
2
2
0
15
Total Converting
to CLM
J = A+C+E+G+I
132
456
182
204
135
107
14
17
1
1,247
Plants Making
Treatment Technology
Changes
K=SUM(A:I)
155
483
193
204
135
111
15
18
1
1,314
1,170
145
1,314
Note: Detail may not add to totals due to independent rounding.
Source: Above table with Treatment Technologies switching from an advanced Treatment Technology with CI2 to the same advanced Treatment Technology with CLM being
moved into the CLM only column.
December 2005
Final Economic Analysis for the Stage 2 DBPR
5-31
-------
Exhibit 5.14c
Stage 2 DBPR Treatment Technology Selection Deltas for NTNCWS Ground Water Plants (Percent of Plants, by Residual Disinfectant Type)
Preferred Alternative
System Size
(Population Served)
<100
100-499
500-999
1 ,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1 ,000,000
Total %
CLM
Only
A
1 .0%
1 .4%
1 .4%
1.1%
1.1%
1 .4%
1 .4%
1 .3%
0.0%
1 .2%
UVCL2
B
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
UVCLM
C
1.1%
1 .6%
1 .6%
1 .6%
1 .6%
/
1 .4%
Ozone
CL2
D
0.0%
0.0%
0.0%
0.0%
0.0%
0.1%
0.1%
0.1%
0.0%
0.0%
Ozone
CLM
E
0.0%
0.0%
0.0%
0.0%
0.0%
0.2%
0.2%
0.2%
0.0%
0.0%
GAC20
CL2
F
0.4%
0.2%
0.2%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.3%
GAC20
CLM
G
0.0%
0.0%
0.0%
0.0%
0.0%
0.2%
0.2%
0.1%
0.0%
0.0%
Membranes
CL2
H
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
Membranes
CLM
I
0.0%
0.0%
0.0%
0.0%
0.0%
0.2%
0.2%
0.2%
0.0%
0.0%
Total Converting
to CLM
J = A+C+E+G+I
2.1%
3.0%
3.0%
2.7%
2.7%
2.0%
2.0%
1 .9%
0.0%
2.5%
Plants Making
Treatment Technology
Changes
K=SUM(A:I)
2.4%
3.2%
3.2%
2.7%
2.7%
2.1%
2.1%
2.0%
0.0%
2.8%
2.8%
2.1%
2.8%
Note: Detail may not add to totals due to independent rounding.
Source: Treatment Technology Selection for the Preferred Alternative minus the Stage 1 Treatment Technology Selection from Appendix C, Exhibit C.2b.
Exhibit 5.14d
Stage 2 DBPR Treatment Technology Selection Deltas for NTNCWS Ground Water Plants (Number of Plants, by Residual Disinfectant Type)
Preferred Alternative
System Size
(Population Served)
<100
100-499
500-999
1 ,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1 ,000,000
Total Plants
CLM
Only
A
24
30
8
3
0
0
0
0
0
65
UVCL2
B
0
0
0
0
0
0
UVCLM
C
27
34
9
4
0
75
Ozone
CL2
D
0
0
0
0
0
0
0
0
0
0
Ozone
CLM
E
0
0
0
0
0
0
0
0
0
0
GAC20
CL2
F
9
4
1
0
0
0
0
0
0
14
GAC20
CLM
G
0
0
0
0
0
0
0
0
0
0
Membranes
CL2
H
0
0
0
0
0
0
0
0
0
0
Membranes
CLM
I
0
0
0
0
0
0
0
0
0
0
Total Converting
to CLM
J = A+C+E+G+I
51
64
18
7
1
0
0
0
0
140
Plants Making
Treatment Technology
Changes
K=SUM(A:I)
60
67
19
7
1
0
0
0
0
154
153
0
154
Note: Detail may not add to totals due to independent rounding.
Source: Above table with Treatment Technologies switching from an advanced Treatment Technology with CI2 to the same advanced Treatment Technology with CLM being
moved into the CLM only column.
December 2005
Final Economic Analysis for the Stage 2 DBPR
5-32
-------
Exhibit 5.15a
Post-Stage 2 DBPR Treatment Technologies-in-Place for CWS Ground Water Plants (Percent of Plants, by Residual Disinfectant Type)
Preferred Alternative
System Size
(Population Served)
<100
100-499
500-999
1 ,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1, 000,000
Total %
No Advanced
Treatment
Technologies
CL21
A
93.5%
92.1%
92.1%
93.1%
93.1%
87.1%
87.1%
87.5%
87.4%
91 .8%
No Advanced
Treatment
Technologies
CLM1
B
3.4%
4.2%
4.2%
3.6%
3.6%
8.6%
8.6%
8.4%
8.5%
4.6%
UVCL2
C
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
UVCLM
D
1.1%
1 .6%
1 .6%
1 .6%
1 .6%
1 .3%
Ozone
CL2
E
0.0%
0.2%
0.2%
0.3%
0.3%
0.9%
0.9%
0.9%
0.9%
Ozone
CLM
F
0.0%
0.5%
0.5%
0.9%
0.9%
1 .0%
1 .0%
0.9%
0.9%
0.3% 0.6%
GAC20
CL2
G
0.4%
0.2%
0.2%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.1%
GAC20
CLM
H
0.9%
0.6%
0.6%
0.1%
0.1%
0.2%
0.2%
0.2%
0.2%
0.5%
Membranes
CL2
I
0.3%
0.1%
0.1%
0.1%
0.1%
1 .7%
1 .7%
1 .7%
1 .7%
0.4%
Membranes
CLM
J
0.5%
0.5%
0.5%
0.5%
0.5%
0.5%
0.5%
0.4%
0.4%
0.5%
Total Using CL2
K = A+C+E+G+I
94.2%
92.6%
92.6%
93.4%
93.4%
89.7%
89.7%
90.1%
90.0%
92.6%
Total Using CLM
L = B+D+F+H+J
5.8%
7.4%
7.4%
6.6%
6.6%
10.3%
10.3%
9.9%
10.0%
7.4%
Note: Detail may not add to totals due to independent rounding.
'No Advanced Treatment Technologies includes conventional, non-conventional, and softening plants.
Source: Add Treatment Technologies-in-Place for the Pre-Stage 2 Baseline (Exhibit 5.13a) to the Treatment Technology Selection Delta for the Preferred Alternative.
Exhibit 5.15b
Post-Stage 2 DBPR Treatment Technologies-in-Place for CWS Ground Water Plants (Number of Plants, by Residual Disinfectant Type)
Preferred Alternative
System Size
(Population Served)
<100
100-499
500-999
1 ,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1, 000,000
Total Plants
No Advanced
Treatment
Technologies
CL21
A
6,006
14,040
5,613
7,060
4,681
4,690
624
803
24
43,539
No Advanced
Treatment
Technologies
CLM1
B
217
640
256
274
181
464
62
77
2
2,173
UVCL2
C
0
0
0
0
0
/"
/"
0
UVCLM
D
70
242
97
118
78
606
Ozone
CL2
E
0
25
10
22
15
48
6
8
0
134
Ozone
CLM
F
0
74
29
66
44
53
7
8
0
282
GAC20
CL2
G
23
27
11
0
0
0
0
0
0
61
GAC20
CLM
H
56
96
39
8
5
10
1
2
0
217
Membranes
CL2
I
22
20
8
4
3
91
12
15
0
175
Membranes
CLM
J
29
79
32
36
24
25
3
4
0
232
Total Using CL2
K = A+C+E+G+I
6,051
14,111
5,641
7,086
4,698
4,829
642
827
25
43,910
Total Using CLM
L = B+D+F+H+J
372
1,131
452
501
332
553
74
91
3
3,510
Note: Detail may not add to totals due to independent rounding.
'No Advanced Treatment Technologies includes conventional, non-conventional, and softening plants.
Source: Add Treatment Technologies-in-Place for the Pre-Stage 2 Baseline (Exhibit 5.13a) to the Treatment Technology Selection Delta for the Preferred Alternative.
December 2005
Final Economic Analysis for the Stage 2 DBPR
5-33
-------
Exhibit 5.15c
Post-Stage 2 DBPR Treatment Technologies-in-Place for NTNCWS Ground Water Plants (Percent of Plants, by Residual Disinfectant Type)
Preferred Alternative
System Size
(Population Served)
<100
100-499
500-999
1 ,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1, 000,000
Total %
No Advanced
Treatment
Technologies
CL21
A
93.5%
92.1%
92.1%
93.1%
93.1%
87.1%
87.1%
87.5%
0.0%
92.8%
No Advanced
Treatment
Technologies
CLM1
B
3.4%
4.2%
4.2%
3.6%
3.6%
8.6%
8.6%
8.4%
0.0%
3.8%
UVCL2
C
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
UVCLM
D
1.1%
1 .6%
1 .6%
1 .6%
1 .6%
1 .4%
Ozone
CL2
E
0.0%
0.2%
0.2%
0.3%
0.3%
0.9%
0.9%
0.9%
0.0%
Ozone
CLM
F
0.0%
0.5%
0.5%
0.9%
0.9%
1 .0%
1 .0%
0.9%
0.0%
0.1% 0.3%
GAC20
CL2
G
0.4%
0.2%
0.2%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.3%
GAC20
CLM
H
0.9%
0.6%
0.6%
0.1%
0.1%
0.2%
0.2%
0.2%
0.0%
0.7%
Membranes
CL2
I
0.3%
0.1%
0.1%
0.1%
0.1%
1 .7%
1 .7%
1 .7%
0.0%
0.2%
Membranes
CLM
J
0.5%
0.5%
0.5%
0.5%
0.5%
0.5%
0.5%
0.4%
0.0%
0.5%
Total Using CL2
K = A+C+E+G+I
94.2%
92.6%
92.6%
93.4%
93.4%
89.7%
89.7%
90.1%
0.0%
93.4%
Total Using CLM
L = B+D+F+H+J
5.8%
7.4%
7.4%
6.6%
6.6%
10.3%
10.3%
9.9%
0.0%
6.6%
Note: Detail may not add to totals due to independent rounding.
'No Advanced Treatment Technologies includes conventional, non-conventional, and softening plants.
Source: Add Treatment Technologies-in-Place for the Pre-Stage 2 Baseline (Exhibit 5.13b) to the Treatment Technology Selection Delta for the Preferred Alternative.
Exhibit 5.15d
Post-Stage 2 DBPR Treatment Technologies-in-Place for NTNCWS Ground Water Plants (Number of Plants, by Residual Disinfectant Type)
Preferred Alternative
System Size
(Population Served)
<100
100-499
500-999
1 ,000-3,300
3,301-9,999
10,000-49,999
50,000-99,999
100,000-999,999
>=1, 000,000
Total Plants
No Advanced
Treatment
Technologies
CL21
A
2,331
1,961
543
230
20
3
0
0
0
5,088
No Advanced
Treatment
Technologies
CLM1
B
84
89
25
9
1
0
0
0
0
208
UVCL2
C
0
0
0
0
0
/"
/"
0
UVCLM
D
27
34
9
4
0
75
Ozone
CL2
E
0
3
1
1
0
0
0
0
0
5
Ozone
CLM
F
0
10
3
2
0
0
0
0
0
15
GAC20
CL2
G
9
4
1
0
0
0
0
0
0
14
GAC20
CLM
H
22
13
4
0
0
0
0
0
0
39
Membranes
CL2
I
8
3
1
0
0
0
0
0
0
12
Membranes
CLM
J
11
11
3
1
0
0
0
0
0
27
Total Using CL2
K = A+C+E+G+I
2,348
1,971
546
231
20
3
0
0
0
5,119
Total Using CLM
L = B+D+F+H+J
144
158
44
16
1
0
0
0
0
364
Note: Detail may not add to totals due to independent rounding.
'No Advanced Treatment Technologies includes conventional, non-conventional, and softening plants.
Source: Add Treatment Technologies-in-Place for the Pre-Stage 2 Baseline (Exhibit 5.13b) to the Treatment Technology Selection Delta for the Preferred Alternative.
December 2005
Final Economic Analysis for the Stage 2 DBPR
5-34
-------
5.5 Reduction in National Average TTHM and HAAS Levels
This section presents the predicted reductions in average TTHM and HAAS levels in distribution
systems as a result of the Stage 2 DBPR. The reductions in average levels are used in Chapter 6 to
approximate the reduction in bladder cancer cases as a result of the Stage 2 DBPR. This section presents
an overview of the methodology first, followed by detailed derivations for large and medium surface
water systems, small surface water systems, large and medium ground water systems, and small ground
water systems. Results are summarized in Section 5.5.6.
5.5.1 Overview of Methodology
The methodology for estimating the percent reduction in average TTHM and HAA5 values
resulting from the Stage 2 DBPR is as follows:
1) Predict TTHM and HAA5 levels for the pre-Stage 1 baseline (presented in Section 3.7.1)
2) Predict the percent reduction resulting from the Stage 1 DBPR and the resulting TTHM and
HAA5 levels for the pre-Stage 2 baseline
3) Predict the percent reduction resulting from the Stage 2 DBPR
Reductions are predicted separately for ground and surface water systems and for large and small
systems (4 predictions). A population-weighted percent reduction is then calculated for all systems.
5.5.2 Reductions for Large and Medium Surface Water Systems
The methodology for surface water systems is based on SWAT model output and the ICR Matrix
Method. Results from both methods are used to predict percent reduction in average DBFs. The first
section summarizes the methodology using SWAT. Section 5.5.2.2 shows how the ICR Matrix Method is
used to predict percent reductions. Section 5.5.2.3 compares results for the individual methods and
explains how they are combined using a Monte Carlo simulation model for the primary analysis.
5.5.2.1 SWAT Methodology
For each model run, SWAT predicted monthly TTHM and HAA5 levels for each of 273 plants
evaluated (273 plants out of the possible 350 ICR plants were used in the SWAT model). See Appendix
A for more information on SWAT, including a discussion of plant representativeness. Monthly data were
averaged for each plant to produce plant-mean data. All plant-means were averaged together to produce a
"mean of plant-means" value. The means of plant-means for Stage 1 and Stage 2 SWAT model runs were
compared to the SWAT initial plant run (simulating pre-Stage 1 conditions) to compute percent reduction.
SWAT-predicted TTHM and HAA5 results were used instead of available ICR-observed data for
the pre-Stage 1 DBPR baseline to allow for consistent comparison of pre-Stage 1 data to modeled pre-
Stage 2 and post-Stage 2 TTHM and HAA5 results. If observed data were used for pre-Stage 1
predictions, differences between pre-Stage 1 and pre-Stage 2 or post-Stage 2 results would reflect
Final Economic Analysis for the Stage 2 DBPR 5-35 December 2005
-------
potential inconsistencies between observed and predicted data sets, not just the expected treatment
technology change from pre-Stage 1 to pre-Stage 2 or post-Stage 2.
5.5.2.2 The ICR Matrix Method
Exhibits 5.16a through 5.16d show the derivation of percent reduction in national average TTHM
and HAAS levels using the ICR Matrix Method. The ICR Matrix Method for the Preferred Regulatory
Alternative with safety margins of 20 and 25 percent is presented in Exhibits 5.16a and 5.16b (the 25
percent safety margin is used to account for the potential impacts of the IDSE). Exhibits 5.16c and 5.16d
show the ICR Matrix Method for Regulatory Alternatives 2 and 3. Regulatory Alternative 1 was not
analyzed separately, as it contains the same MCLs for TTHM and HAA5 and therefore has the same
matrix as the Preferred Regulatory Alternative.
To illustrate how the ICR matrix method works, consider the data presented in Exhibit 5.16a for
the Preferred Alternative for a 20 percent safety margin. The left side of the exhibit contains three tables
or matrices that are divided into different "bins." The bins are cells defined by ranges of RAA values for
TTHM and HAA5 across the top, and maximum LRAA values for TTHM and HAA5 down the left-hand
side. The method works by moving plants from non-compliant bins (Bins B2 and A2) into the compliant
bin (Bin Al) in the second and third tables, representing their actions to comply with Stage 1 and Stage 2,
respectively.
The number and percent of plants in each bin under pre-Stage 1 conditions is shown in the tables
on the right-hand side of Exhibit 5.16a. Plants are assigned to a bin based on their RAA and LRAA
observations as calculated from the ICR data. Note that a plant is considered in one of the non-compliant
bins if it exceeds either the TTHM or HAA5 MCL. EPA assumes that plants making treatment
technology changes to comply with Stage 1 will also meet Stage 2 MCLs (e.g., be moved into the Stage
1- AND Stage 2-compliant Bin Al). This is consistent with the SWAT methodology, which considers
only the delta or additional plants that need to make changes to comply with the Stage 2 DBPR. EPA
recognizes there is uncertainty in this assumption (see the discussion in the compliance forecast section,
Section 5.3.2) but believes it is a reasonable approximation.
Unlike SWAT, the ICR Matrix Method does not use a model to predict which treatment
technologies will be used by plants for compliance, or to calculate the resulting changes in average
TTHM and HAA5 concentrations. EPA developed an alternative methodology to predict these effects.
Based on comparisons of ICR and historical DBP databases, researchers suspect that plants changed their
treatment technology in anticipation of the Stage 1 DBPR prior to the ICR data collection period
(McGuire et al. 2002). It follows, then, that at least some portion of the plants reporting advanced
treatment technology and/or chloramine used in the ICR had installed these treatment technologies to
reduce DBFs. Therefore, the TTHM and HAA5 levels of the ICR plants using advanced treatment
technologies and/or chloramines can provide an indication of the final TTHM and HAA5 levels for plants
that add these treatment technologies to comply with the Stage 2 DBPR.
The results of the analysis of TTHM and HAA5 levels for Stage 2-compliant plants that use
advanced treatment technologies and/or chloramines during the ICR are summarized in Exhibit 5.17. The
average of the plant-average TTHM and HAA5 concentrations for all four analyses are assumed to
represent the average TTHM and HAA5 concentrations for plants that will make treatment technology
changes to meet the Stage 1 and Stage 2 rules. The resulting change in the national average TTHM and
HAA5 concentrations is calculated as the weighted average for the Stage I/Stage 2 compliant plants and
Final Economic Analysis for the Stage 2 DBPR 5-36 December 2005
-------
the non-compliant changers. EPA recognizes that there is uncertainty in using the subset of ICR plants
using advanced technology and/or chloramines to model future changes in DBF occurrence but believes it
provides a plausible result.
Exhibit 5.16a ICR Matrix Method for the Stage 2 DBPR Preferred Alternative
(80/60 LRAA, IDSE)- 20 Percent Safety Margin
Bin
Assignment
A1
A2
B2
All Plants
Number
of Plants
136
36
41
213
Percent of
Plants
64%
17%
19%
100%
Average of Plant Averages
(ug/L)
TTHM
31.64
51.64
69.34
42.28
HAAS
20.67
33.12
53.36
29.07
P re-Stage 2
Bin
Assignment
A1
A2
/>\ /
All Plants
Number
of Plants
136
36
4/"
213
Percent of
Plants
64%
17%
19%
'•>'
100%
Average of Plant Averages
(ug/L)
TTHM
31.64
51.64
xr
34.99
HAAS
20.67
33.12
19,14
22.48
Post-Stage 2
Bin
Assignment
A1
/>\ /
/>\ /
All Plants
Number
of Plants
136
S
4/
213
Percent of
Plants
64%
11%
*">'
19%
'•>'
100%
Average of Plant Averages
(ug/L)
TTHM
31.64
xr
xr
31.58
HAAS
20.67
19,14
19,14
20.11
Notes: 1) In the first table on the left, A1 through B2 are the number of ICR plants that meet the criteria for each bin
under pre-Stage 1 conditions. Their calculated average TTHM and HAAS values based on the averages of
all plant-averages are shown in the first table on the right. A total of 213 ICR plants were evaluated.
2) Each cell (bin) represents a range of the TTHM and HAAS RAA concentrations and maximum LRAA
concentrations in |jg/L (i.e., RAA <64/48 means the plant needs to have its TTHM RAA level below 64 ug/L
and its HAAS RAA level below 48 ug/L to be placed into the bin). The maximum TTHM or HAAS result
determines a plant's bin placement.
3) Crossed-out bins represent plants that have moved from out of compliance bins to in compliance bins.
4) The gray bins on the right-hand side represent bins that have moved into compliance with pre-Stage 2
and post-Stage 2. The TTHM and HAAS concentrations for these plants are the averages of the values for
those ICR plants that are compliant with Stage 1 and Stage 2 and that use either an advanced treatment
technology, chloramines, or both (64 plants) from Exhibit 5.17.
Final Economic Analysis for the Stage 2 DBPR
5-37
December 2005
-------
Exhibit 5.16b ICR Matrix Method for a Stage 2 DBPR Preferred Alternative (80/60
LRAA, IDSE)- 25 Percent Safety Margin
Pre-Stage 1
Bin
A1
A2
B2
All Plants
Number
of Plants
125
47
41
213
Percent of
Plants
59%
22%
19%
100%
Average of Plant Averages
(ug/L)
TTHM
30.14
50.95
69.34
42.28
HAAS
19.39
33.60
53.36
29.07
Pre-Stage 2
Max
LRAA
<60/45
>= 60/45
(S2 non-
compliant)
RAA
<64/48 >=64/48 (S1
"ant)
A1+B2
A2
Bin
A1
A2
Ep
All Plants
Number
of Plants
125
47
41
213
Percent of
Plants
59%
22%
18%
100%
Average of Plant Averages
(ug/L)
TTHM
30.14
50.95
/29.33
34.58
HAAS
19.39
33.60
/1 7.16
22.17
Post-Stage 2
Bin
A1
/?
7
All Plants
Number
of Plants
125
47
41
213
Percent of
Plants
59%
22%
18%
100%
Average of Plant Averages
(ug/L)
TTHM
30.14
^9,33
/28,33
29.80
HAAS
19.39
/17.5S
/I7;S8
18.63
Notes:
1) In the first table on the left, A1 through B2 are the number of ICR plants that meet the criteria for each bin
under pre-Stage 1 conditions. Their calculated average TTHM and HAAS values based on the averages of
all plant-averages are shown in the first table on the right. A total of 213 ICR plants were evaluated.
2) Each cell (bin) represents a range of the TTHM and HAAS RAA concentrations and maximum LRAA
concentrations in |jg/L (i.e., RAA <64/48 means the plant needs to have its TTHM RAA level below 64 ug/L
and its HAAS RAA level below 48 ug/L to be placed into the bin). The maximum TTHM or HAAS result
determines a plant's bin placement.
3) Crossed-out bins represent plants that have moved from out of compliance bins to in compliance bins.
4) The gray bins on the right-hand side represent bins that have moved into compliance with pre-Stage 2
and post-Stage 2. The TTHM and HAAS concentrations for these plants are the averages of the values for
those ICR plants that are compliant with Stage 1 and Stage 2 and that use either an advanced treatment
technology, chloramines, or both (64 plants) from Exhibit 5.17.
Final Economic Analysis for the Stage 2 DBPR
5-38
December 2005
-------
Exhibit 5.16c ICR Matrix Method for Regulatory Alternative 2 (80/60 SH)
Pre-Stage 1
Single
High
<64/48
>= 64/48
(S2 non-
compliant)
RAA
<64/48 >=64/48(S1
non-compliant)
A1
A2
Bin
A1
A2
B2
All Plants
Number
of Plants
89
83
41
213
Percent of
Plants
42%
39%
19%
100%
Average of P
(ug
TTHM
25.18
47.25
69.34
42.28
ant Averages
/L)
HAAS
16.52
30.51
53.36
29.07
Pre-Stage 2
Bin
A1
A2
B2
All Plants
Number
of Plants
89
83
/"
213
Percent of
Plants
42%
39%
ts%
100%
Average of Plant Averages
(ug/L)
TTHM
25.18
47.25
24A9
33.74
HAAS
16.52
30.51
15,1:2
21.70
Post-Stage 2
RAA
<64/48 >=64/48(S1
non-compliant)
Bin
A1
/*
8,2'
All Plants
Number
of Plants
89
89 ..
/"
213
Percent of
Plants
42%
38%
j»
100%
Average of Plant Averages
(ug/L)
TTHM
25.18
24,88
24,88
25.07
HAAS
16.52
15,f2
/1 5,12 •'
15.71
Notes: 1) In the first table on the left, A1 through B2 are the number of ICR plants that meet the criteria for each bin
under pre-Stage 1 conditions. Their calculated average TTHM and HAAS values based on the averages of
all plant-averages are shown in the first table on the right. A total of 213 ICR plants were evaluated.
2) Each cell (bin) represents a range of the TTHM and HAAS RAA concentrations and maximum LRAA
concentrations in |jg/L (i.e., RAA <64/48 means the plant needs to have its TTHM RAA level below 64 ug/L
and its HAAS RAA level below 48 ug/L to be placed into the bin). The maximum TTHM or HAAS result
determines a plant's bin placement.
3) Crossed-out bins represent plants that have moved from out of compliance bins to in compliance bins.
4) The gray bins on the right-hand side represent bins that have moved into compliance with pre-Stage 2
and post-Stage 2. The TTHM and HAAS concentrations for these plants are the averages of the values for
those ICR plants that are compliant with Stage 1 and Stage 2 and that use either an advanced treatment
technology, chloramines, or both (64 plants) from Exhibit 5.17.
Final Economic Analysis for the Stage 2 DBPR
5-39
December 2005
-------
Exhibit 5.16d ICR Matrix Method for Regulatory Alternative 3 (40/30 RAA)
Pre-Stage 1
Bin
A1
A2
B2
All Plants
Number
of Plants
58
114
41
213
Percent of
Plants
27%
54%
19%
100%
Average of P
(up
TTHM
19.71
44.03
69.34
42.28
ant Averages
/L)
HAAS
12.91
28.55
53.36
29.07
Pre-Stage 2
RAA
<32/24
>= 32/24
(S2 non-
compliant)
RAA
<64/48 >=64/48(Sl
non-compliant)
A1+B2
A2
Bin
A1
A2
m
All Plants
Number
of Plants
58
114
43
213
Percent of
Plants
27%
54%
W%
100%
Average of Plant Averages
(ug/L)
TTHM
19.71
44.03
18,73
32.73
HAAS
12.91
28.55
13,04
21.30
Post-Stage 2
RAA
<64/48 >=64/48
-------
Exhibit 5.17 TTHM and HAAS Levels for Stage 2-Compliant Plants Using
Chloramines and/or an Advanced Treatment Technology
Subset of Stage 2
Compliant Plants
CLM only
ADVtech only
CLM & Adv. tech
Total
Preferred Regulatory Alternative (20 Percent
Safety Margin)
Number of
Plants
47
5
12
64
Mean TTHM
(M9/L)
34.50
32.20
19.33
31.48
Mean HAAS
(M9/L)
20.24
23.19
13.14
19.14
Subset of Stage 2
Compliant Plants
CLM only
ADVtech only
CLM & Adv. tech
Total
Preferred Regulatory Alternative (25 Percent
Safety Margin)
Number of
Plants
43
4
12
59
Mean TTHM
(M9/L)
32.44
31.07
19.33
29.68
Mean HAA5
(M9/L)
18.97
19.06
13.14
17.79
Subset of Stage 2
Compliant Plants
CLM only
ADVtech only
CLM & Adv. tech
Total
Regulatory Alternative 2
Number of
Plants
42
4
12
58
Mean TTHM
(M9/L)
32.02
31.07
19.33
29.33
Mean HAA5
(M9/L)
18.68
19.06
13.14
17.56
Subset of Stage 2
Compliant Plants
CLM only
ADVtech only
CLM & Adv. tech
Total
Regulatory Alternative 3
Number of
Plants
28
4
11
43
Mean TTHM
(M9/L)
27.37
31.07
16.74
24.99
Mean HAA5
(M9/L)
16.10
19.06
11.20
15.12
Notes: All TTHM and HAAS values represent the mean of plant-means
CLM = chloramine
Source: ICR Aux 1 database (USEPA 2000h), analysis of ICR screened data (213 surface water plants)
Final Economic Analysis for the Stage 2 DBPR
5-41
December 2005
-------
5.5.2.3 Combining SWAT and ICR Matrix Method Results
Exhibit 5.18 presents the results of SWAT and the ICR Matrix Method for post-Stage 2 national
average TTHM and HAAS levels. Note that predictions for a 20 and 25 percent safety margin are shown
separately for the Preferred Alternative. Both predictions are used in a Monte Carlo simulation model to
account for uncertainty in the impacts of the IDSE. (See Section 5.3.4 for a detailed description of the
methodology used to quantify uncertainty in the potential impacts of the IDSE on large and medium
surface water systems.) As shown in Exhibit 5.18, the ICR Matrix Method predicts a greater reduction in
TTHM and HAA5 levels compared to SWAT for the Preferred Alternative. Such differences are
expected given the inherent differences in the two methods and uncertainties associated with each one.
(Possible reasons for the differences between the two methods are noted in Section 5.3.6.) Because both
SWAT and the ICR Matrix Method have associated uncertainty, results from both are used to generate the
estimated percent reduction in TTHM and HAA5 concentrations for large and medium surface water
systems.
Similar to the compliance forecast model, EPA developed two uniform distributions based on
calculated the ICR Matrix Method-to-SWAT multipliers. The distributions use 1.0 as the 5th percentile
value and the ICR Matrix Method-to-SWAT multiplier as the 95th percentile value. There are two
distributions, one for the 20 percent safety margin and one for the 25 percent safety margin. Exhibit 5.19
provides a graphical depiction of the two uniform distributions for the Preferred Alternative.
To produce final estimates of percent reduction in TTHM and HAA5 for large surface water
systems, EPA developed a Monte Carlo simulation model, similar to the compliance forecast model
described in Section 5.3.6, with three basic steps. Note that results for TTHM and HAA5 are generated
independently.
Step 1: The model randomly selects the predicted TTHM or HAA5 reduction from SWAT, as
shown in column F of Exhibit 5.18 for either the 20 or 25 percent safety margin runs. Results for
each of the two safety margins have an equal (50 percent) chance of being selected.
Step 2: The model randomly selects the ICR-to-SWAT multiplier from the appropriate uniform
distribution from Exhibit 5.19 for the safety margin selected in Step 1.
Step 3: The model applies the multipliers from Step 2 to the TTHM or HAA5 reductions
identified in Step 1 to calculate the percent reduction from Stage 1 to Stage 2 for that iteration.
The process is repeated 10,000 times to produce a distribution of TTHM or HAA5 reductions
from Stage 1 to Stage 2 for large surface water systems. This distribution is carried through the benefits
model, as described in Chapter 6. Note that only TTHM or HAA5 reduction for the 20 percent safety
margin is used for Regulatory Alternatives 1, 2, and 3 as the IDSE is not a component of these
alternatives. Final estimates of predicted reduction in average TTHM and HAA5 concentrations are
presented in Section 5.5.5 for all system types and sizes.
Final Economic Analysis for the Stage 2 DBPR 5-42 December 2005
-------
Exhibit 5.18 Inputs to Monte Carlo Simulation Model: Estimated DBF Reduction from SWAT and ICR Matrix
Method
Regulatory Alternative
ICR Matrix Method
Mean of Plant
Means pre-
32 (ug/L)
A
Mean of Plant
Means post-32
(ug/L)
B
% Reduction
from pre-S2
to post-32
C = (B - A)/A
SWAT
Mean of Plant
Means pre-
32 (ug/L)
D
Mean of Plant
Means post-
32 (ug/L)
E
% Reduction
from pre-S2
to post-32
F = (E - D)/D
ICR Matrix
Method-to-
SWAT
Multiplier
G = C/F
TTHM
Preferred Reg. Alternative (80/60 LRAA, IDSE), 20% SM
Preferred Reg. Alternative (80/60 LRAA, IDSE), 25% SM
Reg. Alternative 1 (Bromate = 5), 20% SM
Reg. Alternative 2 (80/60 Single High), 20% SM
Reg. Alternative 3 (40/30 RAA), 20% SM
35.0
34.6
35.0
33.7
32.7
31.6
29.8
31.6
25.1
19.7
9.7%
13.8%
9.7%
25.7%
39.7%
35.5
35.5
35.5
35.5
35.5
33.8
32.5
33.0
23.7
21.0
4.7%
8.4%
6.9%
33.2%
40.8%
2.06
1.65
1.42
0.77
0.97
HAAS
Preferred Reg. Alternative (80/60 LRAA, IDSE), 20% SM
Preferred Reg. Alternative (80/60 LRAA, IDSE), 25% SM
Reg. Alternative 1 (BR = 5), 20% SM
Reg. Alternative 2 (80/60 SH), 20% SM
Reg. Alternative 3 (40/30 RAA), 20% SM
22.5
22.2
22.5
21.7
21.3
20.1
18.6
20.1
15.7
13.0
10.5%
16.0%
10.5%
27.6%
39.0%
25.0
25.0
25.0
25.0
25.0
23.8
22.9
23.6
16.5
13.9
4.7%
8.3%
5.6%
33.8%
44.3%
2.23
1.92
1.87
0.82
0.88
Sources: SWAT run summaries (USEPA 2001 b), ICR Matrix Method Results (USEPA 2005a)
(A) Average of plant-average values for all plants, pre-Stage 2 conditions from Exhibit 5.16a. ICR Matrix Method results for Reg. Alternative 1 are the
same as for the Preferred Alternative, Unadjusted.
(B) Average of plant-average values for all plants, post-stage 2 conditions from Exhibits 5.16b through 5.16d. ICR Matrix Method results for Reg.
Alternative 1 are the same as for the Preferred Alternative, Unadjusted.
(D) and (E) SWAT run summaries (USEPA 2001 b).
Economic Analysis for the Stage 2 DBPR
5-43
December 2005
-------
Exhibit 5.19 Inputs to the Monte Carlo Simulation Model: Uniform Distributions
Based on ICR Matrix Method-to-SWAT Multiplier
CO
O
CL
Uniform Distribution for a 20%
Safety Margin
CO
2
Q_
Uniform Distribution for a 25%
Safety Margin
(2.06)
Multiplier
(1.65)
Multiplier
Note: The uniform distributions are used when generating the predicted DBP reductions for surface water systems
for the Preferred Alternative. The same method is applied to the other regulatory alternatives using the
multipliers in Exhibit 5.18.
Source: Exhibit 5.18.
5.5.3 Reductions for Small Surface Water Systems
National Rural Water Association (NRWA) survey data were the basis for estimating baseline
TTHM and HAAS levels in small surface water systems (USEPA 200la). Pre-Stage 2 and Post-Stage 2
TTHM and HAAS levels for small surface water systems were assumed to be similar to large system
levels. Thus, both methods used for large and medium surface water systems (SWAT and ICR Matrix
Method) were used to calculate percent reduction in TTHM and HAAS levels for small surface water
systems. Note that EPA believes that the 20 percent safety margin already accounts for potential impacts
of the IDSE for small surface water systems because their distribution systems as not as complex
compared to large ground water systems. Therefore, an alternative percent reduction for a 25 percent
safety margin was not used for small surface water systems, as it was for large and medium surface water
systems.
Final estimated percent reductions in average TTHM and HAAS concentrations are presented in
Section 5.5.5 for all system types and sizes.
Final Economic Analysis for the Stage 2 DBPR
5-44
December 2005
-------
5.5.4 Reductions for Large and Medium Ground Water Systems
As described in Section 3.7, ICR data were used to characterize TTHM and HAAS levels for the
ground water system pre-Stage 1 baseline. EPA used the ICR Matrix Method to predict changes in
average TTHM and HAAS levels for large ground water systems following the Stage 1 and Stage 2 rules.
A detailed description of the method can be found in Section 5.5.2.2
The analysis of Stage 2-compliant, screened ground water plants using chloramines and/or an
advanced treatment technology at the time of the ICR data collection is shown in Exhibit 5.20. TTHM
and HAA5 levels for these plants are used to estimate the TTHM and HAA5 levels for those plants
changing treatment technology to meet Stage 1 and Stage 2 rules. The number of ICR GW plants that use
chloramines and/or advanced disinfectants and comply with the Stage 2 DBPR is low: 12 plants for the
Preferred Alternative (considering a 20 percent safety margin on compliance). This is roughly 15 percent
of the total number of screened ground water plants. EPA compared TOC levels for the Stage 2-compliant
ground water plants that use chloramines and/or an advanced treatment technology to levels for the Stage
2 non-compliant plants and found them to be similar.
Exhibit 5.21a shows the ICR Matrix Method for the Stage 2 DBPR Preferred Alternative.
Exhibits 5.21b and 5.21c show the method for Alternatives 2 and 3. Note that the ICR Matrix Method for
ground water plants is not performed for the 25 percent safety margin. EPA believes that for ground
water systems, the 20 percent safety margin analysis already accounts for the impacts of the IDSE since
these systems do not typically observe high year-to-year or seasonal variability in water quality (see
Section 5.3.4. for additional information). Section 5.5.5 summarizes percent reduction in TTHM and
HAA5 for all system sizes and source water types.
5.5.5 Reductions for Small Ground Water Systems
Data from seven states were used to characterize TTHM and HAA5 levels for small ground water
systems (USEPA 2000d). To derive the percent reduction in average TTHM and HAA5 levels for small
ground water systems, EPA compared the predicted percent of plants making treatment technology
changes for small ground water systems to large ground water systems. The percent reduction in DBP
concentrations as predicted by the ICR Matrix Method for large ground water systems (GWL) is
multiplied by the ratio of small ground water plants changing treatment technology (GCTS) to large
ground water plants changing treatment technology (GCTL), or
GWS = GWL * (GCTS/GCTL)
where,
GWL and GWS = predicted percent DBP reduction for large and small ground water systems,
respectively.
GCTL and GCTS = the percent of plants changing treatment technology for large and small ground
water plants (weighted average across large or small system size categories), respectively.
Final Economic Analysis for the Stage 2 DBPR 5-45 December 2005
-------
Exhibit 5.20 TTHM and HAAS Levels for Stage 2-Compliant Ground Water Plants
Using Chloramines and/or an Advanced Treatment Technology
Subset of Stage 2 Compliant
Plants
CLM only
Adv. tech with CLM
Adv. tech w/o CLM
Total
Preferred Regulatory Alternative
No. of
Plants
10
2
0
12
Mean TTHM
(ug/L)
29.0
19.9
0.0
27.5
Mean HAAS
(ug/L)
19.4
16.5
0.0
18.9
Subset of Stage 2 Compliant
Plants
CLM only
Adv. tech with CLM
Adv. tech w/o CLM
Total
Regulatory Alternative 2
No. of
Plants
6
2
0
8
Mean TTHM
(ug/L)
21.7
19.9
0.0
21.2
Mean HAAS
(ug/L)
14.2
16.5
0.0
14.8
Subset of Stage 2 Compliant
Plants
CLM only
Adv. tech with CLM
Adv. tech w/o CLM
Total
Regulatory Alternative 3
No. of
Plants
6
2
0
8
Mean TTHM
(ug/L)
18.5
19.9
0.0
18.8
Mean HAA5
(ug/L)
11.3
16.5
0.0
12.6
Notes: All TTHM and HAAS values represent the mean of plant-means.
ICR Matrix Method results for Reg. Alternative 1 are the same as for the Preferred Alternative, assuming a
20 percent safety margin.
While the number of plants is identical for Regulatory Alternatives 2 and 3, they represent a different subset
of plants.
Source: ICR Aux 1 database (USEPA 2000h), analysis of ICR screened data (82 ground water plants)
Final Economic Analysis for the Stage 2 DBPR
5-46
December 2005
-------
Exhibit 5.21 a ICR Matrix Method for Ground Water Plants for the Stage 2 DBPR
Preferred Alternative
P re-Stage 1
Bin
A1
A2
B2
All Plants
Number
of Plants
75
2
5
82
Percent of
Plants
91%
2%
6%
100%
Average of Plant Averages
(ug/L)
TTHM
11.62
35.29
63.54
15.36
HAAS
5.50
31.71
43.37
8.45
P re-Stage 2
Bin
A1
A2
82
All Plants
Number
of Plants
75
2
5
82
Percent of
Plants
91%
2%
8%
100%
Average of P
(ug
TTHM
11.62
35.29
27.50
/•
13.16
ant Averages
/L)
HAAS
5.50
31.71
18,95
„•••••
6.96
Post-Stage 2
Bin
A1
/*'
32
All Plants
Number
of Plants
75
/"
S
82
Percent of
Plants
91%
/2%
6%
100%
Average of P
(ug
TTHM
11.62
/27.SO /
27,50
12.97
ant Averages
/L)
HAAS
5.50
1<Ł».'
1%9f
6.64
Notes: 1) In the first table on the left, A1 through B2 are the number of ICR plants that meet the criteria for each bin
under pre-Stage 1 conditions. Their calculated average TTHM and HAAS values based on the averages of
all plant-averages are shown in the first table on the right. A total of 82 ICR plants were evaluated.
2) Each cell (bin) represents a range of the TTHM and HAAS RAA concentrations and maximum LRAA
concentrations in |jg/L (i.e., RAA <64/48 means the plant needs to have its TTHM RAA level below 64 ug/L
and its HAAS RAA level below 48 ug/L to be placed into the bin). The maximum TTHM or HAAS result
determines the bin placement.
3) Crossed-out bins represent plants that have moved from out of compliance bins to in compliance bins.
4) The gray bins on the right-hand side represent bins that have moved into compliance with pre-Stage 2
and post-Stage 2. The TTHM and HAAS concentrations for these plants are the averages of the values for
those ICR plants that are compliant with Stage 1 and Stage 2 and use an advanced treatment technology
and/or chloramines (12 plants), from Exhibit 5.20.
Final Economic Analysis for the Stage 2 DBPR
5-47
December 2005
-------
Exhibit 5.21 b ICR Matrix Method for Ground Water Plants for Regulatory
Alternative 2
P re-Stage 1
Bin
A1
A2
B2
All Plants
Number
of Plants
67
10
5
82
Percent of
Plants
82%
12%
6%
1 00%
Average of Plant Averages
(ug/L)
TTHM
8.76
35.47
63.54
15.36
HAAS
3.89
21.51
43.37
8.45
P re-Stage 2
Bin
A1
A2
B2
All Plants
Number
of Plants
67
10
§
82
Percent of
Plants
82%
12%
e%
100%
Average of Plant Averages
(ug/L)
TTHM
8.76
35.47
21.23
12.78
HAAS
3.89
21.51
1,4.80
6.70
Post-Stage 2
Bin
A1
/*
B2
'/
All Plants
Number
of Plants
67
to .
/*
82
Percent of
Plants
82%
,-'1'??& '
/7
1 00%
Average of Plant Averages
(ug/L)
TTHM
8.76
r™
/^^
11.04
HAAS
3.89
/? /
/T°
5.88
Notes: 1) In the first table on the left, A1 through B2 are the number of ICR plants that meet the criteria for each bin
under pre-Stage 1 conditions. Their calculated average TTHM and HAAS values based on the averages of
all plant-averages are shown in the first table on the right. A total of 82 ICR plants were evaluated.
2) Each cell (bin) represents a range of the TTHM and HAAS RAA concentrations and maximum LRAA
concentrations in |jg/L (i.e., RAA <64/48 means the plant needs to have its TTHM RAA level below 64 ug/L
and its HAAS RAA level below 48 ug/L to be placed into the bin). The maximum TTHM or HAAS result
determines the bin placement.
3) Crossed-out bins represent plants that have moved from out of compliance bins to in compliance bins.
4) The gray bins on the right-hand side represent bins that have moved into compliance with pre-Stage 2
and post-Stage 2. The TTHM and HAAS concentrations for these plants are the averages of the values for
those ICR plants that are compliant with Stage 1 and Stage 2 and use an advanced treatment technology
and/or chloramines (8 plants), from Exhibit 5.20.
Final Economic Analysis for the Stage 2 DBPR
5-48
December 2005
-------
Exhibit 5.21 c ICR Matrix Method for Ground Water Plants for Regulatory
Alternative 3
Pre-Stage 1
Bin
A1
A2
B2
All Plants
Number
of Plants
67
10
5
82
Percent of
Plants
82%
12%
6%
100%
Average of P
(ug
TTHM
7.90
41.27
63.54
15.36
ant Averages
/L)
HAAS
3.51
24.03
43.37
8.45
Pre-Stage 2
RAA
<32/24
>= 32/24
(S2 non-
compliant)
RAA
<64/48 >=64/48 (S1
non-compliant)
A1+B2
A2
Bin
A1
A2
m
All Plants
Number
of Plants
67
10
S
82
Percent of
Plants
82%
12%
/6%
100%
Average of Plant Averages
(ug/L)
TTHM
7.90
41.27
/I8.84
12.63
HAAS
3.51
24.03
/1 2,5?
6.57
Post-Stage 2
RAA
<64/48 >=64/48(S1
non-compliant)
Bin
A1
/*
82
All Plants
Number
of Plants
67
10
/•
82
Percent of
Plants
82%
/12%
f%
100%
Average of P
(ug
TTHM
7.90
18,94
/
18,84
/
9.90
ant Averages
/L)
HAAS
3.51
/12.57
1:2,57 •
5.17
Notes: 1) In the first table on the left, A1 through B2 are the number of ICR plants that meet the criteria for each bin
under pre-Stage 1 conditions. Their calculated average TTHM and HAAS values based on the averages of
all plant-averages are shown in the first table on the right. A total of 82 ICR plants were evaluated.
2) Each cell (bin) represents a range of the TTHM and HAAS RAA concentrations and maximum LRAA
concentrations in |jg/L (i.e., RAA <64/48 means the plant needs to have its TTHM RAA level below 64 ug/L
and its HAAS RAA level below 48 ug/L to be placed into the bin). The maximum TTHM or HAAS result
determines the bin placement.
3) Crossed-out bins represent plants that have moved from out of compliance bins to in compliance bins.
4) The gray bins on the right-hand side represent bins that have moved into compliance with pre-Stage 2
and post-Stage 2. The TTHM and HAAS concentrations for these plants are the averages of the values for
those ICR plants that are compliant with Stage 1 and Stage 2 and use an advanced treatment technology
and/or chloramines (8 plants), from Exhibit 5.20.
Final Economic Analysis for the Stage 2 DBPR
5-49
December 2005
-------
5.5.6 Results for All Systems
Exhibit 5.22 presents predicted pre-Stage 1 and pre-Stage 2 TTHM and HAAS mean distribution
system concentrations for surface and ground water plants. This analysis shows that the largest percent
reduction in DBF concentrations is for surface water plants from pre-Stage 1 to pre-Stage 2 conditions
(reductions range from approximately 27 to 57 percent for all plants). The reduction in TTHM and
HAAS levels for ground water plants is less, ranging from 11 to 18 percent. The percent reduction for
small surface water plants is greater than the percent reduction for large surface water plants because
plants in small systems did not have to meet the TTHM rule MCL of 100 i-ig/L prior to the Stage 1 DBPR.
Exhibit 5.23 shows the predicted percent reduction in TTHM and HAAS concentrations from pre-
Stage 2 to post-Stage 2 DBPR conditions for the Preferred Alternative. Exhibit 5.24a, 5.24b, and 5.24c
show similar information for regulatory alternatives 1, 2, and 3, respectively. Note that the combined
results for "All Systems" are generated using a Monte Carlo analysis. Also note that the percent
reduction for Regulatory Alternative 1 is slightly less than the percent reduction for the Preferred
alternative because none of the Regulatory Alternatives except the preferred include the IDSE component.
It is important to note that the calculation of the percent reduction in all plant-mean DBP levels
includes plants that make minor process changes as well as those that do not make any treatment
technology changes to meet the Stage 2 DBPR. The predicted percent reduction for the subset of plants
that add advanced treatment technologies or chloramines, therefore, is much higher. Among those plants
reducing DBP levels from Stage 1 to Stage 2, it is predicted that the average reductions in DBP levels will
be approximately 30 percent, and may range from less than 5 up to 60 percent.
Final Economic Analysis for the Stage 2 DBPR 5-50 December 2005
-------
Exhibit 5.22 Reduction in Average TTHM and HAAS Concentrations
from Pre-Stage 1 to Pre-Stage 2
Source Water
Type
SW
GW
System Size
(Population Served)
Large (=10, 000)
Small (< 10,000)
Large (=10, 000)
Small (< 10,000)
All Systems*
Population
A
160,935,736
8,422,403
65,152,168
28,514,211
263,024,518
TTHM
Pre-Stage 1
(ug/L)
B
48.7
82.8
15.4
16.5
38.05
Pre-Stage 2
(ug/L)
u
35.5
35.5
13.2
14.7
27.69
Percent
Reduction
D = (B-C)/B
27.2%
57.2%
14.3%
11.1%
27.2%
Source Water
Type
SW
GW
System Size
(Population Served)
Large (=10, 000)
Small (< 10,000)
Large (=10, 000)
Small (< 10,000)
All Systems*
Population
A
160,935,736
8,422,403
65,152,168
28,514,211
263,024,518
HAAS
Pre-Stage 1
(ug/L)
E
35.5
45.3
8.4
9.1
26.24
Pre-Stage 2
(ug/L)
F
25.0
25.0
7.0
7.8
18.67
Percent
Reduction
G = (E-F)/E
29.5%
44.8%
17.6%
13.7%
28.8%
Notes:
All TTHM and HAAS concentrations represent the mean of all plant-mean concentrations.
* TTHM and HAAS concentrations for all systems are the population-weighted values.
Sources:
(A) SDWIS 2003 3rd quarter freeze, community water system population (Exhibit 3.3, CWSs only).
(B) Large SW: Exhibit 3.15, SWAT plant-mean data. Small SW: Exhibit 3.21, NRWA plant-mean data.
Large GW: Exhibit 3.20, ICR plant-mean data. Small GW(B): Exhibit 3.21, State plant-mean data.Small
GW.
(E) Large SW: Exhibit 3.15, SWAT plant-mean data. Small SW: Exhibit 3.21, NRWA plant-mean data.
Large GW: Exhibit 3.20, ICR plant-mean data. Small GW: large GW HAAS * (small GW TTHM / large
GWTTHM).
(C) and (F) For SW, Pre-Stage 2 runs from SWAT (USEPA 2001 e). For large GW, pre-Stage 2 based on
the ICR matrix method. For small GW, pre-Stage 2 based on percent reduction in large GW and
comparison of percent changing technology (see methodology in Section 5.5).
Final Economic Analysis for the Stage 2 DBPR
5-51
December 2005
-------
Exhibit 5.23 Reduction in Average TTHM and HAAS Concentrations from
Pre-Stage 2 to Post-Stage 2, Preferred Alternative
Source Water
Type
SW
GW
System Size
(Population Served)
Large (> 10,000)
Small (< 10,000)
Large (> 10,000)
Small (< 10,000)
All Systems*
Population
A
160,935,736
8,422,403
65,152,168
28,514,211
263,024,518
TTHM
Pre-Stage 2
Level (ug/L)
B
35.5
35.5
13.2
14.7
27.69
Post-Stage 2 Level (ug/L)
Mean
5th
95th
C
32.2
32.9
13.0
14.4
25.53
33.7
33.8
/
26.44
30.7
32.0
/
24.60
Percent Reduction
Mean
5th
95th
D = (B - C) / B
9.2%
7.2%
1 .4%
2.0%
7.8%
5.1%
4.7%
4.5%
13.5%
9.7%
/
11.2%
Source Water
Type
SW
GW
System Size
(Population Served)
Large (> 10,000)
Small (< 10,000)
Large (> 10,000)
Small (< 10,000)
All Systems*
Population
A
160,935,736
8,422,403
65,152,168
28,514,211
263,024,518
HAAS
Pre-Stage 2
Level (ug/L)
E
25.0
25.0
7.0
7.8
18.67
Post-Stage 2 Level (ug/L)
Mean
5th
95th
F
22.5
23.1
6.6
7.4
16.96
23.7
23.8
17.71
21.1
22.4
16.12
Percent Reduction
Mean
5th
95th
G = (E - F) / E
9.9%
7.6%
4.5%
6.3%
9.2%
5.2%
4.7%
5.2%
15.5%
10.5%
13.7%
Notes: Detail may not add due to independent rounding.
All TTHM and HAAS concentrations represent the mean of all plant-mean concentrations. Total for "All Systems" is calculated using a Monte Carlo analysis.
Results for large SW systems for the preferred alternative represent the combined results for the 20% and 25% safety margins.
* TTHM and HAAS concentrations for all systems are the population-weighted values
Sources:
(A) SDWIS 2003 3rd quarter freeze, community water system population (Exhibit 3.3, CWSs only).
(B) and(E) Exhibit 5.22
(C) and (F) Outputs from the benefits Monte Carlo simulation model. Confidence bounds for large and medium SW systems account for uncertainties in compliance
forecast methodologies and potential impacts of the IDSE. Confidence bounds for small SW systems account for uncertainties in compliance forecast methodologies.
Final Economic Analysis for the Stage 2 DBPR
5-52
December 2005
-------
Exhibit 5.24a Reduction in Average TTHM and HAAS Concentrations from Pre-Stage 2 to Post-Stage 2,
Regulatory Alternative 1
Source Water
Type
SW
GW
System Size
(Population Served)
Large (=10,000)
Small (< 10,000)
Large (=10,000)
Small (< 10,000)
All Systems*
Population
A
160,935,736
8,422,403
65,152,168
28,514,211
263,024,518
TTHM
Pre-Stage 2
Level (ug/L)
b
35.5
35.5
13.2
14.7
27.69
Post-Stage 2 Level (ug/L)
Mean
5th
95th
C
32.5
32.5
13.0
14.4
25.71
33.0
33.0
26.05
32.0
32.0
25.38
Percent Reduction
Mean
8.3%
8.3%
1 .4%
2.0%
7.1%
5th
95th
D = (B - C) / B
6.9%
6.9%
5.9%
9.7%
9.7%
8.4%
Source Water
Type
SW
GW
System Size
(Population Served)
Large (=10,000)
Small (< 10,000)
Large (=10,000)
Small (< 10,000)
All Systems*
Population
A
160,935,736
8,422,403
65,152,168
28,514,211
263,024,518
HAAS
Pre-Stage 2
Level (ug/L)
E
25.0
25.0
7.0
7.8
18.67
Post-Stage 2 Level (ug/L)
Mean
5th
95th
F
23.0
23.0
6.6
7.4
17.23
23.6
23.6
17.62
22.4
22.4
16.85
Percent Reduction
Mean
5th
95th
G = (E - F) / E
8.1%
8.1%
4.5%
6.3%
7.7%
5.6%
5.6%
5.7%
10.5%
10.5%
9.8%
Notes: Detail may not add due to independent rounding.
All TTHM and HAAS concentrations represent the mean of all plant-mean concentrations. Total for "All Systems" is calculated using a Monte Carlo analysis.
* TTHM and HAAS concentrations for all systems are the population-weighted values.
Sources:
(A) SDWIS 2003 3rd quarter freeze, community water system population (Exhibit 3.3, CWSs only).
(B) and(E) Exhibit 5.22
(C) and (F) Outputs from the benefits Monte Carlo simulation model. Confidence bounds for SW systems account for uncertainties in compliance forecast
methodologies.
Final Economic Analysis for the Stage 2 DBPR
5-53
December 2005
-------
Exhibit 5.24b Reduction in Average TTHM and HAAS Concentrations from Pre-Stage 2 to Post-Stage 2,
Regulatory Alternative 2
Source Water
Type
SW
GW
System Size
(Population Served)
Large (=10,000)
Small (< 10,000)
Large (=10,000)
Small (< 10,000)
All Systems*
Population
A
160,935,736
8,422,403
65,152,168
28,514,211
263,024,518
TTHM
Pre-Stage 2
Level (ug/L)
b
35.5
35.5
13.2
14.7
27.69
Post-Stage 2 Level (ug/L)
Mean
5th
95th
C
25.0
25.0
11.0
12.9
20.24
26.4
26.4
21.16
23.7
23.7
19.33
Percent Reduction
Mean
5th
95th
D = (B - C) / B
29.4%
29.4%
16.1%
12.0%
26.9%
25.7%
25.7%
23.6%
33.2%
33.2%
/
30.2%
Source Water
Type
SW
GW
System Size
(Population Served)
Large (=10,000)
Small (< 10,000)
Large (=10,000)
Small (< 10,000)
All Systems*
Population
A
160,935,736
8,422,403
65,152,168
28,514,211
263,024,518
HAAS
Pre-Stage 2
Level (ug/L)
E
25.0
25.0
7.0
7.8
18.67
Post-Stage 2 Level (ug/L)
Mean
5th
95th
F
17.3
17.3
5.9
6.9
13.36
18.1
18.1
/
13.92
16.5
16.5
/
12.80
Percent Reduction
Mean
5th
95th
G = (E - F) / E
30.7%
30.7%
15.4%
1 1 .5%
28.5%
27.6%
27.6%
25.5%
33.8%
33.8%
31.4%
Notes: Detail may not add due to independent rounding.
All TTHM and HAAS concentrations represent the mean of all plant-mean concentrations. Total for "All Systems" is calculated using a Monte Carlo analysis.
* TTHM and HAAS concentrations for all systems are the population-weighted values.
Sources:
(A) SDWIS 2003 3rd quarter freeze, community water system population (Exhibit 3.3, CWSs only).
(B) and(E) Exhibit 5.22
(C) and (F) Outputs from the benefits Monte Carlo simulation model. Confidence bounds for SW systems account for uncertainties in compliance forecast
methodologies.
Final Economic Analysis for the Stage 2 DBPR
5-54
December 2005
-------
Exhibit 5.24c Reduction in Average TTHM and HAAS Concentrations from Pre-Stage 2 to Post-Stage 2,
Regulatory Alternative 3
Source Water
Type
SW
GW
System Size
(Population Served)
Large (=10,000)
Small (< 10,000)
Large (=10,000)
Small (< 10,000)
All Systems*
Population
A
160,935,736
8,422,403
65,152,168
28,514,211
263,024,518
TTHM
Pre-Stage 2
Level (ug/L)
B
35.5
35.5
13.2
14.7
27.69
Post-Stage 2 Level (ug/L)
Mean
5th
95th
C
21.2
21.2
9.9
12.0
17.41
21.4
21.4
17.72
21.0
21.0
17.10
Percent Reduction
Mean
5th
95th
D = (B - C) / B
40.2%
40.2%
24.8%
18.4%
37.1%
39.7%
39.7%
36.0%
40.8%
40.8%
38.2%
Source Water
Type
SW
GW
System Size
(Population Served)
Large (=10,000)
Small (< 10,000)
Large (=10,000)
Small (< 10,000)
All Systems*
Population
A
160,935,736
8,422,403
65,152,168
28,514,211
263,024,518
HAAS
Pre-Stage 2
Level (ug/L)
E
25.0
25.0
7.0
7.8
18.67
Post-Stage 2 Level (ug/L)
Mean
5th
95th
F
14.6
14.6
5.2
6.4
11.37
15.3
15.3
11.90
13.9
13.9
10.83
Percent Reduction
Mean
5th
95th
G = (E - F) / E
41 .6%
41 .6%
25.7%
19.0%
39.1%
39.0%
39.0%
36.2%
44.3%
44.3%
/
42.0%
Notes: Detail may not add due to independent rounding.
All TTHM and HAAS concentrations represent the mean of all plant-mean concentrations. Total for "All Systems" is calculated using a Monte Carlo analysis.
* TTHM and HAAS concentrations for all systems are the population-weighted values.
Sources:
(A) SDWIS 2003 3rd quarter freeze, community water system population (Exhibit 3.3, CWSs only)
(B) and(E) Exhibit 5.22
(C) and (F) Outputs from the benefits Monte Carlo simulation model. Confidence bounds for SW systems account for uncertainties in compliance forecast
methodologies.
Final Economic Analysis for the Stage 2 DBPR
5-55
December 2005
-------
5.6 Reduction in Frequency of Peak TTHM and HAAS Concentrations
Treatment technology changes to meet the Stage 1 and Stage 2 DBPRs can reduce the frequency
of peak TTHM and HAAS values and reduce all levels of TTHM and HAAS concentrations. Both effects
have potential health benefits, which are discussed in detail in Chapter 6. This section summarizes the
methodology for estimating reduced frequency of percent locations with peak TTHM and HAAS
concentrations as a result of the Stage 1 and Stage 2 Preferred Alternative. Appendix G contains an
analysis of percent reductions of peaks considered as individual measurements.
5.6.1 Methodology and Assumptions
For the purposes of this section, a "peak" TTHM or HAAS is defined as any individual
measurement greater than a specified threshold concentration. The level does not have to be sustained
over any period of time to be considered a peak measurement, and it can be a measurement taken at any
time in the year. As discussed in Chapter 6, the health data do not conclusively identify a peak TTHM or
HAAS level of concern. Therefore, the analyses in this section use the following alternative threshold
concentrations (or "study levels") for the purposes of defining peaks: 60, 75, 80, and 100 i-ig/L for TTHM
and 45, 60, and 75 |^g/L for HAAS. The analyses in this section predict the number of locations with at
least one TTHM and HAAS observation greater than each study level for pre-Stage 1, simulated pre-Stage
2 and simulated post-Stage 2 conditions. This information is used in the exposure assessment in Chapter
6.
EPA evaluated TTHM and HAAS data from both ground and surface water plants for the four
ICR distribution system sampling locations (DSE, AVG1, AVG2, and DS Maximum) to predict how the
frequency of peak TTHM and HAAS concentrations changes as a result of the Stage 1 and Stage 2 rules.
For surface water plants, ICR-observed data is used instead of SWAT-predicted data because, as
explained in Appendix A, SWAT-predicted TTHM and HAAS concentrations are valid only when
considering national averages, not at the plant level.
The method used to predict reduction in locations is the ICR Matrix Method. Section 5.5.2.2
provides a detailed description of how the matrix method predicts reductions in average DBP
concentrations. The method works in the same way to predict reduction in locations with peaks. Plants
are assigned to a bin (bins are defined in the tables in the left-hand side of the exhibit) based on their
RAA and LRAA observations, as calculated from the ICR data. The method works by moving plants
from non-compliant bins (Bins B2 and A2) into the compliant bin (Bin Al) in the second and third tables,
representing their actions to comply with Stage 1 and Stage 2, respectively.
Final Economic Analysis for the Stage 2 DBPR 5-56 December 2005
-------
Exhibit 5.25a ICR Matrix Method for Peak Locations for the Stage 2 DBPR, 20 Percent Safety Margin, Large
Surface and Ground Water Plants
Pre-Stage 1
Bin
A1
A2
B2
All Plants
Number of
Locations
880
155
195
1,230
Locations with
TTHM >60 ug/L
122
96
173
391
Locations with
TTHM>75ug/L
31
65
151
247
Locations with
TTHM >80 ug/L
19
54
142
215
Locations with
TTHM >100 ug/L
4
17
83
104
Locations with
HAAS >45 ug/L
80
82
136
298
Locations with
HAAS >60 ug/L
13
32
112
157
Locations with
HAAS >75 ug/L
3
9
72
84
Pre-Stage 2
Bin
A1
A2
/p ;
All Plants
Number of
Locations
880
155
V® /
1,230
Locations with
TTHM >60 ug/L
122
96
,40
258
Locations with
TTHM>75ug/L
31
65
^8
101
Locations with
TTHM >80 ug/L
19
54
A
74
Locations with
TTHM >100 ug/L
4
17
,-fO
21
Locations with
HAAS >45 ug/L
80
82
31
183
Locations with
HAAS >60 ug/L
13
32
3
48
Locations with
HAAS >75 ug/L
3
9
/O
12
Post-Stage 2
Bin
A1
Ki
$
32
All Plants
Number of
Locations
880
165
/IflfS
1,230
Locations with
TTHM >60 ug/L
122
32
40
193
Locations with
TTHM>75ug/L
31
4
^
40
Locations with
TTHM >80 ug/L
19
1
\
21
Locations with
TTHM >100 ug/L
4
8
0
4
Locations with
HAAS >45 ug/L
80
17
21
117
Locations with
HAAS >60 ug/L
13
/
3
18
Locations with
HAAS >75 ug/L
3
8
0
3
Notes: 1) In the Pre-Stage 1 tables, A1 through B2 are the number of locations for ICR plants that meet the criteria for each bin under pre-Stage 1 conditions. A
total of 1,230 locations for 311 screened ICR surface and ground water plants were evaluated.
2) Each cell (bin) represents a range of the TTHM and HAAS RAA concentrations and Maximum LRAA concentrations in |jg/L (i.e., RAA <64/48 means
the plant needs to have its TTHM RAA level below 64 ug/L and its HAAS RAA level below 48 ug/L to be placed into the bin). The maximum TTHM or
HAAS result determines the bin placement. Note that bins are based on a 20 percent safety margin on the Stage 1 and Stage 2 MCLs.
3) Crossed-out bins represent plants that have moved from out of compliance bins to in compliance bins.
4) The gray bins on the right-hand side represent bins that have moved into compliance with pre-Stage 2 and post-Stage 2. The percent locations with
TTHM and HAAS concentrations above each study level is the percent of locations above the study level for those ICR plants that are compliant with
Stage 1 and Stage 2 and use an advanced treatment technology and/or chloramines (as shown in Exhibit 5.26).
Final Economic Analysis for the Stage 2 DBPR
5-57
December 2005
-------
Exhibit 5.25b ICR Matrix Method for Peak Locations for the Stage 2 DBPR, 25 Percent Safety Margin, Large
Surface and Ground Water Plants
Pre-Stage 1
Bin
A1
A2
B2
All Plants
Number of
Locations
832
203
195
1,230
Locations with
TTHM >60 ug/L
90
128
173
391
Locations with
TTHM >75 ug/L
18
78
151
247
Locations with
TTHM >80 ug/L
11
62
142
215
Locations with
TTHM >100 ug/L
2
19
83
104
Locations with
HAAS >45 ug/L
52
110
136
298
Locations with
HAAS >60 ug/L
4
41
112
157
Locations with
HAAS >75 ug/L
1
11
72
84
Pre-Stage 2
Max
LRAA
<60/45
>= 60/45 (S2
non-
compliant)
RAA
<64/48
>=64/48 (S1 non
compliant)
A1 + B2
Bin
A1
A2
B2
All Plants
Number of
Locations
832
203
196
1,230
Locations with
TTHM >60 ug/L
90
128
40
258
Locations with
TTHM >75 ug/L
18
78
5
101
Locations with
TTHM >80 ug/L
11
62
A
74
Locations with
TTHM >100 ug/L
2
19
0
21
Locations with
HAAS >45 ug/L
52
110
21
183
Locations with
HAAS >60 ug/L
4
41
3
48
Locations with
HAAS >75 ug/L
1
11
0
12
Post-Stage 2
Bin
A1
A2
^
m
All Plants
Number of
Locations
832
J03
196
1,230
Locations with
TTHM >60 ug/L
90
x^1
40
171
Locations with
TTHM >75 ug/L
18
6
,-'5
28
Locations with
TTHM >80 ug/L
11
1
A
14
Locations with
TTHM >100 ug/L
2
8
0
2
Locations with
HAAS >45 ug/L
52
22
,,'21
94
Locations with
HAAS >60 ug/L
4
3
jfl
9
Locations with
HAAS >75 ug/L
1
0
0
1
Notes: 1) In the Pre-Stage 1 tables, A1 through B2 are the number of locations for ICR plants that meet the criteria for each bin under pre-Stage 1 conditions. A
total of 1,230 locations for 311 screened ICR surface and ground water plants were evaluated.
2) Each cell (bin) represents a range of the TTHM and HAAS RAA concentrations and Maximum LRAA concentrations in |jg/L (i.e., RAA <64/48 means
the plant needs to have its TTHM RAA level below 64 ug/L and its HAAS RAA level below 48 ug/L to be placed into the bin). The maximum TTHM or
HAAS result determines the bin placement. Note that bins are based on a 20 percent safety margin on the Stage 1 and on a 25 percent safety margin on
the Stage 2 MCLs.
3) Crossed-out bins represent plants that have moved from out of compliance bins to in compliance bins.
4) The gray bins on the right-hand side represent bins that have moved into compliance with pre-Stage 2 and post-Stage 2. The percent locations with
TTHM and HAAS concentrations above each study level is the percent of locations above the study level for those ICR plants that are compliant with
Stage 1 and Stage 2 and use an advanced treatment technology and/or chloramines (as shown in Exhibit 5.26).
Final Economic Analysis for the Stage 2 DBPR
5-58
December 2005
-------
Characterization of peak TTHM and HAAS levels for plants in each bin is shown in the tables on
the right-hand side of Exhibit 5.25a for the 20 percent safety margin and 5.25b for the 25 percent safety
margin. The first column shows the total number of locations in the bin. Subsequent columns show the
percent of the locations that have at least one TTHM sampling result above study levels of 60, 75, 80, and
100 (ig/L followed by the percent of locations with at least one HAA5 result above 45, 60, and 75 (ig/L.
Shaded rows represent those sampling locations associated with non-compliant plants that are expected to
make treatment technology changes to meet Stage 1, then Stage 2 compliance.
Similar to the method explained in Section 5.5.1.2, EPA used information on the occurrence of
peaks for ICR plants using advanced treatment technologies and/or chloramines at the time of the ICR to
estimate the occurrence of peaks for plants predicted to change treatment technology to comply with the
Stage 2 DBPR. The results of the analysis of TTHM and HAA5 peaks for this subset of plants is
summarized in Exhibit 5.26. The frequency of peak TTHM and HAA5 concentrations in Exhibit 5.26 are
assumed to represent the frequency of peak TTHM and HAA5 concentrations for plants that will make
treatment technology changes to meet the Stage 1 and Stage 2 rules (identified as shaded rows for the pre-
Stage 2 and post-Stage 2 tables in Exhibits 5.25a and 5.25b).
Exhibit 5.26 Frequency of Occurrence of Peak Locations for ICR Surface and
Ground Water Plants Using Chloramines and/or Advanced Treatment
Technologies
Technology
Category
Number of
Locations
A
percent or Locations witn i i MM
Peaks Above
60 |jg/L
C
75 |jg/L
D
80 |jg/L
E
100|jg/L
F
percent or Locations witn
HAAS Peaks Above
45 |jg/L
I
60 |jg/L
J
75|jg/L
K
Stage 2, (torn pi ia nee Based on a 20 Percent Safety M a rg in
CLM only
Adv. tech with CLM
Adv. tech w/o CLM
All plants
235
55
20
310
24.7%
7.3%
5.0%
20.3%
3.0%
1.8%
0.0%
2.6%
0.9%
0.0%
0.0%
0.6%
0.0%
0.0%
0.0%
0.0%
11.9%
1.8%
20.0%
10.6%
1.7%
0.0%
0.0%
1.3%
0.0%
0.0%
0.0%
0.0%
Stage;2 Compliance Based on a -25 Percent Safety Margin
CLM only
Adv. tech with CLM
Adv. tech w/o CLM
All plants
219
55
16
290
19.2%
7.3%
0.0%
15.9%
1.8%
1.8%
0.0%
1.7%
0.9%
0.0%
0.0%
0.7%
0.0%
0.0%
0.0%
0.0%
9.1%
1.8%
0.0%
7.2%
0.9%
0.0%
0.0%
0.7%
0.0%
0.0%
0.0%
0.0%
Notes:
Source:
Advanced technologies include chlorine dioxide,ozone, GAC, and membranes. Advanced technologies
DO NOT consider enhanced coagulation or enhanced softening.
The 25 percent safety margin results include ground water systems.
ICR database (USEPA 2000h), analysis of 311 screened ICR surface and ground water plants.
Final Economic Analysis for the Stage 2 DBPR
5-59
December 2005
-------
5.6.2 Results
Exhibit 5.27 summarize the results for each of four TTHM study levels assuming a 20 percent
safety margin on compliance. Using the ICR Matrix Approach, the predicted percent of locations with at
least one peak observation declines from 20.1 percent for pre-Stage 1 to 8.2 percent for pre-Stage 2 to 3.3
percent for post-Stage 2 DBPR conditions at a TTHM study level of 75 i-ig/L. EPA believes that results
based on a 20 percent safety margin are conservatively low because they do not consider the potential
impacts of the IDSE on large surface water plants (i.e., more plants may need to make treatment changes
as a result of the IDSE, causing additional reduction in occurrence of peaks). Results based on a 25
percent safety margin, however, are conservatively high because the analysis includes ground water
systems.
Exhibit 5.27 Predicted Percent of Distribution System Sampling Locations with
Peaks for Pre-Stage 1, Pre-Stage 2, and Post-Stage 2 Conditions,
20 Percent safety Margin
TTHM Study
Level
Evaluated
60 |jg/L
75 |jg/L
80 |jg/L
100 |jg/L
Pre-Stage 1 Conditions
No. of
Locations
Evaluated
A
1,230
1,230
1,230
1,230
No. of
Locations
with Peaks
B
391
247
215
104
Percent of
Locations
with Peaks
C = B/A
31 .8%
20.1%
17.5%
8.5%
Pre-Stage 2 Conditions
No. of
Locations
Evaluated
D
1,230
1,230
1,230
1,230
No. of
Locations
with Peaks
E
258
101
74
21
Percent of
Locations
with Peaks
F = E/D
20.9%
8.2%
6.0%
1 .7%
Post-Stage 2 Conditions
No. of
Locations
Evaluated
G
1,230
1,230
1,230
1,230
No. of
Locations
with Peaks
H
193
40
21
4
Percent of
Locations
with Peaks
I = H/G
15.7%
3.3%
1 .7%
0.3%
Sources: (A), (D), and (G) are the number of distribution system locations evaluated for 311 screened ICR surface
and ground water plants.
(B), (E), and (H) are number of locations with at least one TTHM observation over the TTHM study level as
derived in Exhibits 5.25.
5.7 Uncertainties in the Compliance Forecast and Subsequent DBF Reduction
There are numerous sources of uncertainty in the compliance forecast, as discussed previously in
this chapter and in detail in Appendices A and B. Exhibit 5.28 summarizes the key uncertainties in the
compliance forecast, with the exception of uncertainties in baseline data inputs (e.g., the ICR data), which
are discussed in detail in Section 3.8.
EPA believes that two of these uncertainties could have a potentially large impact on Stage 2
DBPR cost and benefit analysis:
• Uncertainty in the impacts of the Initial Distribution System Evaluation (IDSE) on the
Compliance Forecast for large and medium surface water sytems.
Uncertainty in compliance forecast tools.
EPA has adjusted the compliance forecast methodology to quantify and incorporate both of these
uncertainties into the compliance forecast results and the cost and benefits models. Section 5.3 discusses
the methods used to quantify these uncertainties. To help inform the reader of the potential magnitude of
Final Economic Analysis for the Stage 2 DBPR
5-60
December 2005
-------
these uncertainties, EPA has conducted several sensitivity analyses. Exhibit 5.29 shows the predicted
percent of all plants changing technology and resulting decrease in average TTHM concentration for the
primary analysis and four sensitivity analyses. Note that results reflect combined surface water and
ground water system analyses, with the ground water system inputs staying the same for each estimate.
Analyses 1 and 2 (beneath the primary analysis) show results assuming a 20 percent safety margin and 25
percent safety margin for large and medium surface water systems, respectively. The impact of the safety
margin is more pronounced for DBF reduction than for percent changing technology. This implies that
the uncertainty in the IDSE will have a larger impact on the national benefits estimates compared to the
national cost estimates.
Analyses 3 and 4 show results using the SWAT compliance forecast methodology for all surface
water systems, and the ICR Matrix Method compliance forecast methodology for all surface water
systems, respectively. The compliance forecast tool has a greater impact on the compliance forecast
results compared to the operational safety margin. The percent of all plants changing technology for the
ICR Matrix Method is approximately 15 percent higher than the primary analysis. The percent DBF
reduction for the ICR Matrix Method is about 30 percent higher than the primary analysis. EPA carried
this analysis through the entire benefits and cost models and found that overall, the benefits estimates
using the SWAT model are about 30 percent lower than the midpoint estimates, while those using the ICR
Matrix Method are about 30 percent higher. For national costs, estimates using the SWAT model are
about 25 percent lower then the midpoint estimate, while those using the ICR Matrix Method are about 25
percent higher. (Additional surface water systems changing technology have a disproportionately high
impact on total costs because surface water systems are typically much larger than ground water systems
and therefore, have higher unit costs.)
The remaining uncertainties listed in Exhibit 5.28 fall into two categories—uncertainty in the
DBF data and uncertainty in the models. These two categories are discussed in Sections 5.7.1 and 5.7.2,
respectively.
It is important to note that any biases in the compliance forecasts affect cost and benefits
similarly. If more plants make treatment technology changes than predicted, costs for the treatment
technology changes would be higher and there would also be a higher overall reduction in DBF levels.
Conversely, if fewer plants make treatment technology changes than predicted, treatment costs and
benefits from DBF reduction would be less.
Final Economic Analysis for the Stage 2 DBPR 5-61 December 2005
-------
Exhibit 5.28 Summary of Uncertainties in the Compliance Forecast
Source
Water
Type
Surface
Ground
All
All
All
Surface
Uncertainty
Uncertainty in
tools used to
derive compliance
forecast and DBP
reductions for
large surface
water systems
Uncertainty in
Ground Water
Delphi results
Uncertainty in
extrapolating
compliance
forecasts from
large to small
systems
Uncertainty in
using the delta
approach
(effectively
assuming that
plants changing
for Stage 1 meet
Stage 2 MCLs)
Operational safety
margin of 20
percent for Stage
1 and Stage 2
Impacts of the
IDSE on the
compliance
forecast for the
preferred
regulatory
alternative
Section With
Additional
Discussion
of
Uncertainty
A.6
B.2.2
A.9.1
&
B.3.1
5.3.2
5.3.3
5.3.4
Effect on Benefit Predictions
Under-
estimate
Over-
estimate
Unknown
Impact
Quantified in the primary
analysis (addresses potential
underestimate or overestimate)
X
X
X
X
Quantified in the primary
analysis (addresses potential
underestimate)
Effect on Cost Predictions
Under-
estimate
Over-
estimate
Unknown
Impact
Quantified in the primary
analysis (addresses potential
underestimate or overestimate)
X
X
X
X
Quantified in the primary
analysis (addresses potential
underestimate)
Final Economic Analysis for the Stage 2 DBPR
5-62
December 2005
-------
Exhibit 5.29 Potential Impact of Uncertainties on the Compliance Forecast and
DBF Reduction Analysis
Sensitivity Analysis for Surface
Water Systems
Equally weighted between 20%
and 25% Safety Margin for large
SW systems, SWAT and ICR MM
for all SW Systems
(Primary Analysis)
1) 20% Safety Margin for
Large SW systems
2) 25% Safety Margin for
Large SW systems
3) ICR Matrix Method for
all SW systems
4) SWAT Method for
all SW systems
Total Percent Changing
Technology from Stage 1 to
Stage 2 (all plants)
A
3.8%
3.6%
3.9%
4.3%
3.2%
Percent Reduction in
Average TTHM from Stage
1 to Stage 2 (all plants)
B
7.8%
6.3%
9.3%
9.9%
5.7%
Notes: Costs and benefits shown are mean estimates for all systems.
Percent results are indicated for all systems. Sensitivity analysis
criteria pertain only to the surface water system portion of the analysis.
The ground water system portion of the analysis is the same for each
sensitivity analysis. For each sensitivity analysis, only one factor was
changed (e.g., for the 20% safety margin analysis, both SWAT and the
ICR Matrix Method are considered).
Sources: (A) Stage 2 DBPR Cost Model (USEPA 2005I)
(B) Stage 2 DBPR Benefits Model (USEPA 2005h)
5.7.1 Uncertainty in DBF data
One factor that influences the compliance forecast results is the representativeness of the ICR
data. For example, ICR observed data are used as the basis for the pre-Stage 1 baseline, although EPA
recognizes that some plants had already made changes to comply with Stage 1 by the time the ICR was
conducted. Additionally, there are limitations associated with the subsets of the ICR data that were used
for the analysis with both SWAT and the ICR Matrix Method. Limitations of the ICR data are discussed
in Section 3.8. Data limitations related to SWAT are discussed in greater detail in Appendix A, and
limitations related to the ICR Matrix Method are discussed in this section.
5.7.1.1 Representativeness of the ICR data
There are uncertainties regarding the ICR TTHM and HAAS data in general, as explained in
detail in Section 3.8. EPA examined the climate conditions during the ICR sampling period to determine
the representativeness of DBP levels over time. On a nationwide basis, 1998 was hotter and wetter than
normal. Increased rainfall may have biased the results by increasing levels of contaminants from runoff,
Final Economic Analysis for the Stage 2 DBPR
5-63
December 2005
-------
such as TOC. Other constituents may have been lower than normal, such as bromide, which tends to rise
during droughts. Higher temperatures in 1998 could have caused the DBF levels in the ICR data to be
unusually high as compared to an average year.
In addition, the representativeness of the DBF sample results are uncertain due to the nature of
distribution system monitoring. Research has shown that TTHM and HAAS levels can vary as much as
20 percent over the course of a day at locations in the distribution system (Pereira et al. 2004). One grab
sample collected at a discrete point in time for the ICR does not represent this potential variability. In
addition to hourly variations, the ICR data were not required to be collected at evenly spaced intervals.
Thus, there is uncertainty in assuming that a single data point represents the typical level over the entire
quarter.
5.7.1.2 Uncertainty in the subset of ICR data used for the ICR Matrix Method
The data set used for the ICR Matrix Method contains 213 of the 353 surface water plants in the
ICR database, or roughly 60 percent. To evaluate the representativeness of the ICR Matrix Method
subset, EPA evaluated source water TOC and distribution system DBP data. Plants in the screened data
set have a mean influent TOC level of 3.21, and the plants excluded from this data set have mean influent
TOC level of 3.19, indicating relatively little bias from source water quality.
TTHM and HAAS data for plants that passed the data quality screen were compared to data for
the plants that were excluded from the analyses. For the excluded plants, only those that had matching
TTHM and HAAS samples were considered, leaving 75 of the 140 SW plants that were excluded. The
TTHM levels were higher in the excluded plants, with a plant mean of 50.5, as opposed to 42.3 for the
screened plants. The excluded plants may have higher overall DBP levels, and their inclusion would raise
the percentage of plants out of compliance with Stage 1 and Stage 2. However, these plants were
excluded due to missing data, so it is possible that the complete data set would have DBP levels more
similar to the screened plants.
5.7.2 Uncertainty in the Delta Approach
This section briefly describes some of the larger uncertainties in the compliance forecast.
Uncertainties associated with SWAT are discussed in greater detail in Appendix A.
Both SWAT and the ICR Matrix Method are limited in allowing systems to make multiple
treatment technology changes. In both models, when systems make treatment technology changes to
come into compliance with Stage 1, they will simultaneously come into compliance with Stage 2 MCLs.
It is possible that some plants may need to make a second treatment technology change after achieving
compliance with Stage 1 to achieve compliance with Stage 2, although EPA believes this is unlikely for
most systems. The Agency believes this uncertainty is small and the delta approach is reasonable for
reasons presented in Section 5.3.2.
There is also uncertainty associated with the 20 percent safety margin. Individual systems may
use higher or lower safety margins based on system-specific conditions. The M-DBP TWG
recommended that a 20 percent operational safety margin be used for DBP MCLs (TTHM, HAA5,
bromate, and chlorite) when evaluating all regulatory alternatives. This safety margin is consistent with
practices in prior DBP regulatory development efforts. It is intended to represent the level at which
Final Economic Analysis for the Stage 2 DBPR 5-64 December 2005
-------
systems typically take some action to ensure consistent compliance with a new drinking water standard
and the level at which systems target new treatment to meet the standard. In addition to representing
industry practices, the safety margin also is intended to account for year-to-year fluctuations in DBF data.
(ICR data are limited to 1 year and must not represent the highest DBF concentrations that occur in a
system.)
Final Economic Analysis for the Stage 2 DBPR 5-65 December 2005
-------
6. Benefits Analysis
6.1 Introduction
The mission of the Environmental Protection Agency (EPA) is to protect human health and to
safeguard the natural environment (USEPA 2000m). The Safe Drinking Water Act (SDWA) provides
that the
Administrator shall... publish a maximum contaminant level goal and promulgate a national
primary drinking water regulation for a contaminant... if the Administrator determines that -
(i) the contaminant may have an adverse effect on the health of persons; (ii) the contaminant
is known to occur or there is a substantial likelihood that the contaminant will occur in public
water systems with a frequency and at levels of public health concern; and (iii) in the sole
judgment of the Administrator, regulation of such contaminant presents a meaningful
opportunity for health risk reduction for persons served by public water systems. (42 USC
§300g-l(b)(l)(A))
When carrying out its statutory mandate, EPA must often make regulatory decisions using
incomplete or uncertain information. EPA believes it is appropriate and prudent to act to protect public
health when there are indications that exposure to a contaminant could present significant risks to the
public, rather than take no action until risks are unequivocally proven. Evidence from both human
epidemiology and animal toxicology studies indicate that the consumption of drinking water containing
disinfection byproducts (DBFs) may result in adverse health effects. The two main categories of such
potential effects that have been associated with DBFs are reproductive and developmental effects and
cancer (particularly bladder cancer). EPA has concluded that DBFs occur at levels that are a public health
concern in some public water systems (PWSs) that apply a chemical disinfectant, and that the Stage 2
Disinfectants and Disinfection Byproducts Rule (DBPR) presents a meaningful opportunity for a
reduction in the risk of adverse health effects.
Under Executive Order 12866, EPA must conduct an Economic Analysis (EA) for rules costing
over $100 million annually. The benefits analyses presented in this chapter follow the requirements of
the executive order and related Office of Management and Budget (OMB) and EPA guidance, and
provide a reasonable basis for estimating potential health benefits using the best available science.
EPA has quantified the benefits associated with expected reductions in the incidence of bladder
cancer. EPA also includes a sensitivity analysis for potential benefits from avoiding colon and rectal
cancers. Scientific knowledge about the association of reproductive and developmental health effects
with DBP exposure is not adequate to quantify these risks or the benefits of reduced DBP exposure in the
primary analysis. Nevertheless, although the results from different studies do not support a conclusion at
this time as to whether exposure to chlorinated drinking water or disinfection byproducts cause adverse
developmental or reproductive health effects, they do support a potential health concern. EPA believes
additional benefits from the Stage 2 DBPR could come from reducing potential reproductive and
developmental risks and that it is important to provide some quantitative indication of the potential risk.
To do this, EPA completed an illustrative calculation of potential benefits for one specific reproductive
effects endpoint (fetal loss).
Section 6.1.1 provides an overview of the methodology and key assumptions used to estimate the
benefits that may be attributed to the Stage 2 DBPR (including the illustrative calculations for a
developmental and reproductive health endpoint). Section 6.1.2 summarizes results.
Section 6.2 presents the problem identification and the assessment of potential hazard.
Carcinogenic and non-carcinogenic (e.g., reproductive and developmental) risks are presented with
Final Economic Analysis for the Stage 2 DBPR 6-1 December 2005
-------
toxicological and epidemiological evidence. Section 6.3 follows with an assessment of exposure. The
rule's benefits, including cancer cases avoided and the associated value of those benefits, are addressed in
Sections 6.4 and 6.5. Section 6.6 summarizes uncertainties of national benefits estimates. Section 6.7
contains a sensitivity analysis for benefits from avoiding colon and rectal cancers. Potential benefits from
reductions in one reproductive and developmental endpoint—fetal loss—are evaluated through an
illustrative calculation in Section 6.8.
The following provide additional details in support of this chapter:
Chapter 5 estimates reduction in TTHM and HAAS occurrence as a result of DBF risks in
support of the exposure assessment in Section 6.3.
• Appendix D shows the schedule for all rule activities (this information is used as input for the
quantified benefits calculation).
• Appendix E provides general background information on population attributable risk (PAR)
and a detailed description of the derivation of PAR values used to quantify benefits
associated with reduction in cancer cases. The development and modeling of the cessation
lag equations are described. Lastly, Appendix E shows detailed calculations for estimating
the number of bladder cancer cases avoided as a result of the Stage 2 DBPR.
• Appendix E2 provides calculations for the colon and rectal cancer sensitivity analysis.
• Appendix F describes the valuation of Stage 2 DBPR benefits and presents results for all
regulatory alternatives and sensitivity analyses.
• Appendix G provides detailed calculations for the illustrative calculation of potential
reproductive and developmental health impacts.
6.1.1 Overview of Methodology for Quantifying Stage 2 DBPR Benefits
Three categories of potential benefits are addressed in this EA: those associated with reductions
in the incidence of bladder cancer, those associated with reductions in the incidence of colon and rectal
cancers, and those associated with reductions in the incidence of adverse reproductive and developmental
health effects. The primary benefits analysis in this EA is based on reductions in bladder cancer cases.
Potential benefits associated with reduced incidence of colon and rectal cancers are quantified in a
sensitivity analysis, while potential benefits associated with decreasing adverse developmental and
reproductive health effects (specifically, fetal losses) are presented as an illustrative calculation.
EPA used similar approaches to estimate the number of bladder cancer cases avoided (the
primary benefits analysis), the number of colon and rectal cancer avoided (the sensitivity analysis), and
the number of avoided incidence of fetal loss (the illustrative calculation). The major steps in deriving
and characterizing cases avoided are:
• Estimate the current and future annual cases of illness from all causes
• Estimate how many cases can be attributed to current DBP occurrence and exposure
Estimate the reduction in future cases corresponding to anticipated reductions in DBP
occurrence and exposure due to the Stage 2 DBPR
Final Economic Analysis for the Stage 2 DBPR 6-2 December 2005
-------
All benefit calculations were performed using the Stage 2 DBPR Benefits Model (USEPA 2005h).
For bladder cancer, EPA computed the monetized benefits of the Stage 2 DBPR by multiplying
the estimated number of bladder cases avoided by the estimated monetary value associated with avoiding
both fatal and non-fatal cases of bladder cancer. The value of a statistical life (VSL) was used for fatal
bladder cancers, while two alternate estimates of willingness-to-pay to avoid non-fatal bladder cancer are
used (one based on avoiding a case of curable lymphoma and the other based on avoiding a case of
chronic bronchitis). EPA also computed the benefits for the reduction in colon and rectal cancer by using
the same VSL and willingness to pay (WTP) estimates. EPA recognizes that there could be a additional
value associated with the number of avoided fetal losses estimated in the illustrative calculation.
However, the Agency is unable at this time either to develop a specific estimate of this value or to use a
benefit transfer method to estimate the value from studies that address other endpoints (see Section 6.8 for
a full discussion of this issue).
Bladder Cancer
To calculate potential benefits from reduced incidence of bladder cancer cases, EPA began by
estimating the number of new bladder cancer cases occurring per year from all causes. The National
Cancer Institute's Surveillance, Epidemiology, and End Results (SEER, 2004) program provides data on
cancer rates (new cases per 100,000 population per year) as a function of age. EPA used this information
in conjunction with population-by-age data from the 2000 U.S. Census to estimate the number of new
cases of bladder cancer. Results show that the number of new bladder cancer cases per year starts to
increase at about age 35 and peaks at 1,500 to 2,000 cases from about age 66 to 85. Although the annual
rate of bladder cancer does not decline much after age 85, the number of new bladder cancer cases does,
which represents the overall decline in the number of individuals alive after that age. The resulting total
number of new bladder cancer cases per year, 56,506, is slightly lower than that currently estimated by
the American Cancer Society (ACS).1 This likely represents EPA's use of the census population data
from 2000.
To estimate the baseline number of cases attributable to DBP exposure, EPA used three different
approaches:
Using the range of Population Attributable Risk (PAR) values derived from consideration of
5 individual epidemiology studies used for the Stage 1 EA and the Stage 2 proposal EA
(yields apre-Stage 1 range of best estimates for PAR of 2% to 17%).
• Using the Odds Ratio (OR) of 1.2 from the Villanueva et al. (2003) meta-analysis that reflects
both sexes, ever exposed population from the studies considered (yields a pre-Stage 1 best
estimate for PAR of-16%)
Using the Villanueva et al. (2004) pooled data analysis to develop a dose-response
relationship for OR as a function of Average TTHM. The dose-response relationship was
modeled as linear with an intercept of OR = 1.0 at TTHM exposure level = 0 (yields a pre-
Stage 1 best estimate for PAR of-17%)
Taken together, the three approaches provide a reasonable estimate of the range of potential risks.
For the sake of simplicity, EPA carried only one of these approaches, that based on Villanueva et al.
(2003), through the entire benefits model.
lrThe American Cancer Society estimated in 2004 that 60,240 new cases of bladder cancer would occur in
the U.S. population that year (ACS website, 2004).
Final Economic Analysis for the Stage 2 DBPR 6-3 December 2005
-------
To quantify the reduction in cases, EPA assumes that there is a linear relationship between
average DBF concentration and relative risk of bladder cancer. Thus, percent reductions in national
average DBFs are used to determine the percent reductions in bladder cancer cases attributable to DBFs.
Predicted reductions in national average TTHM and HAAS levels resulting from predicted treatment
technology changes to comply with Stage 2 were used as indicators of overall chlorination DBF
reductions. The baseline cases attributable to DBFs multiplied by the percent reductions in TTHM or
HAAS concentrations result in the estimated annual bladder cancer cases "ultimately avoidable" for the
Stage 1 and Stage 2 rules.
Over the long run, the cases ultimately avoidable (derived as described above) will be attained.
They will not be achieved instantaneously, however. Research shows that a lag period (referred to as
"cessation lag") exists between the point in time when reduction in exposure to a carcinogen occurs and
the point in time when the full risk reduction benefit of that exposure reduction is realized by affected
individuals. Because there is no epidemiological or other empirical data available that specifically
address the rate of achieving bladder cancer benefits resulting from DBF reductions, EPA uses data from
three epidemiological studies that address the rate of risk reduction following exposure reduction to other
carcinogens (namely cigarette smoke and arsenic) to generate three possible cessation lag functions for
bladder cancer and DBFs.
The cessation lag functions are used to project the number of bladder cancer cases avoided each
year after implementation as a result of the Stage 2 DBPR over a 100-year period. A 100-year period was
selected as the timeframe after which effectively all of the exposed population is composed of individuals
exposed only to post-Stage 2 levels for their entire lifetime. At that time (and from that point forward)
the annual bladder cancer cases ultimately avoidable is achieved for the exposed population. The
projected number of cases avoided each year is also adjusted to reflect when systems are expected to
install new treatment to reduce DBFs based on the rule implementation schedule. Although a 100-year
cessation lag period is modeled, annual avoided cases of bladder cancer are calculated primarily for the
first 25 years after rule promulgation. A 25-year time period was used to coincide with the estimated life
span of capital equipment and a time lag of five to ten years for technology installation after rule
promulgation.
The final step in the benefit calculation is to monetize the average annual cases avoided. This is
done by applying economic values for avoided illnesses and deaths. EPA has estimated that 74 percent of
bladder cancer cases are non-fatal (USEPA, 1999a). The value of avoiding non-fatal bladder cancer cases
is based on people's WTP for incremental reductions in the risk they face of contracting cancer. The
metric of WTP to avoid an increased risk includes the desire to avoid treatment costs, pain and
discomfort, productivity losses, and any other adverse consequences related to a non-fatal case of bladder
cancer. Because specific estimates of WTP for avoiding non-fatal bladder cancer are not available, EPA
estimated values from two other non-fatal illnesses: curable lymphoma and chronic bronchitis. Both are
considered valid estimates of WTP for non-fatal cancer.
For fatal bladder cancer cases, VSL is used to capture the value of benefits. The VSL represents
an estimate of the monetary value of reducing risks of premature death from cancer. Therefore, it is not
an estimate of the value of saving a particular individual's life. Rather, it represents the sum of the values
placed on small individual risk reductions across an exposed population. Other economic factors are
taken into consideration when calculating benefits over time, such as income growth and social discount
rates.
Final Economic Analysis for the Stage 2 DBPR 6-4 December 2005
-------
There are several areas of uncertainty with respect to quantified benefits for bladder cancer.
Many are described qualitatively in the analysis, while other are incorporated explicitly as follows:
There is uncertainty in the percent reduction in TTHM and HAAS concentrations resulting
from predicted treatment technology changes (i.e., the compliance forecasts). Uncertainty in
SWAT and potential impacts of the IDSE are quantified in the primary analysis.
• Three approaches were used to estimate the baseline number of bladder cancer cases
attributable to DBF exposure. For the sake of simplicity, one approach using data from
Villanueva et al. (2003) was carried through the full benefits model.
The estimated PAR values from the Villanueva et al. (2003) meta-analysis include confidence
bounds resulting from statistical uncertainty in the odds ratio underlying the PAR calculation.
The confidence bounds from Villanueva et al. (2003) capture a significant portion of the
confidence intervals of the other two approaches.
Three independent cessation lag models derived from three different epidemiological studies
are used in the model. Also, two functional forms are used for each of these data sets and
uncertainty in the parameters of those functions is included in the analysis.
• EPA uses two alternatives for valuing non-fatal bladder cancer.
Colon and Rectal Cancers
Human epidemiology studies on chlorinated surface water have reported associations with colon
and rectal cancers. Colon and rectal cancers combined are the third most common type of new cancer
cases and deaths in both men and women in the U.S., excluding skin cancers. Therefore, any benefit from
reducing the potential incidence of colon and rectal cancers could be significant. EPA is including a
quantitative sensitivity analysis for benefits accrued from the Stage 2 DBPR from avoiding colon and
rectal cancers.
EPA estimated the reduction in colon and rectal cancer in a similar manner to bladder cancer
cases. Background incidence data were available from the SEER cancer registry and two quality studies
were chosen to estimate a PAR value. Using the percent reductions in DBFs and the smoking and lung
cancer cessation lag model, the number of colon and rectal cancer cases avoided annually was estimated
and monetized with the same VSL and WTP estimates as for bladder cancer.
Developmental And Reproductive Health Effects
As noted previously, EPA believes that additional benefits of this rule could come from
reductions in potential developmental and reproductive health effects, although the relationship of these
effects to DBP exposure is not known well enough to quantify risks or benefits in the main analysis. EPA
prepared an illustrative calculation of the benefits of reducing the potential risk of fetal loss, the non-
cancer effect for which the most epidemiological data exist in relation to DBP exposure. Because
approximately one million of the six million pregnancies that occur each year in the United States end in
a miscarriage or stillbirth (Ventura et al. 2000), avoiding even a small risk attributable to DBP exposure
by reducing DBP levels may potentially result in a significant number of avoided fetal losses.
EPA estimated the potential reduction in fetal losses in a similar manner to bladder cancer cases.
A range of possible PAR values for relating annual fetal losses to DBP exposure was obtained from
available epidemiological studies. Reductions in the number of peak DBP events due to the Stage 1
DBPR and the Stage 2 DBPR were estimated. Reductions in exposure to peak DBFs were assumed to be
Final Economic Analysis for the Stage 2 DBPR 6-5 December 2005
-------
proportional to reductions in peak DBF events. Like the analysis of bladder cancer, there is uncertainty in
fetal loss PAR values, which is reflected in the range of values used in the analysis. There are other
important uncertainties in this illustrative calculation, including the assumed proportional relationship
between reduction in fetal losses and reduction in exposure to peak levels due to the Stage 2 DBPR.
6.1.2 Summary of National Benefits of the Stage 2 DBPR
Exhibit 6.1 summarizes the estimated number of bladder cancer cases avoided as a result of the
Stage 2 DBPR and the monetized value of those cases. The benefits in Exhibit 6.1 are for the Preferred
Regulatory Alternative (which includes a requirement for the IDSE) for the Stage 2 DBPR. Benefit
estimates for the other regulatory alternatives were derived using the same methods as for the Preferred
Regulatory Alternative and are presented in Section 6.5.4.
The confidence bounds shown for the results in Exhibit 6.1 incorporate uncertainty in the PAR,
uncertainty in the compliance forecast and resulting reduction in DBP concentrations, and uncertainty in
cessation lag. The confidence bounds of the monetized benefits also incorporate uncertainty in the
valuation parameters. An estimated 26 percent of bladder cancer cases avoided are fatal, and 74 percent
are non-fatal. The monetized benefits therefore represent the estimate of avoiding both fatal and non-fatal
cancers in those proportions.
In addition to bladder cancer cases avoided, EPA provides an illustrative calculation of the
potential number of fetal losses that might be avoided per year, ranging from 0 to 3,700. The value of
other health benefits, including the potential reduction in other types of cancer such as colon or rectal,
could be significant, and is calculated in a sensitivity analysis. Also, the value of non-health benefits,
such as improved taste and odor of water, are expected to be positive. These are discussed further in
Section 6.4.
Exhibit 6.1 Summary of Quantified Benefits for the Stage 2 DBPR
Cessation Lag Model
used to estimate Annual
Bladder Cancer Cases
Avoided
Smoking/Lung
Cancer Model
Smoking/Bladder
Cancer Model
Arsenic/Bladder
Cancer Model
Annual Average Bladder
Cancer Cases Avoided for the
first 25 years
Mean
279
188
333
5th
103
61
138
95th
541
399
610
Discount Rate, WTP
for
Non-Fatal Bladder
Cancer Cases
3 %, Lymphoma
7 % Lymphoma
3 % Bronchitis
7 % Bronchitis
3 %, Lymphoma
7 % Lymphoma
3 % Bronchitis
7 % Bronchitis
3 %, Lymphoma
7 % Lymphoma
3 % Bronchitis
7 % Bronchitis
Annualized Benefits of Bladder
Cancer Cases Avoided
($Millions)
Mean
$1,531
$1,246
$763
$621
$1,032
$845
$514
$420
$1,852
$1,545
$922
$769
5th
$233
$190
$165
$135
$157
$129
$111
$91
$282
$235
$200
$167
95th
$ 3,536
$ 2,878
$1,692
$1,376
$ 2,384
$1,950
$1,141
$932
$ 4,276
$ 3,566
$ 2,045
$1,704
Note: Based on TTHM as an indicator, benefits were calculated using the Villanueva et al. (2003) PAR. Assumes 24
percent of cases are fatal, 76 percent are non-fatal (USEPA 1999a). The 90 percent confidence interval for cases
incorporates uncertainty in PAR, reduction in average TTHM and HAAS concentrations, and cessation lag. The 90
percent confidence bounds for monetized benefits reflect uncertainty in monetization inputs relative to mean cases.
Values are discounted and annualized in 2003$. EPA recognizes that benefits may be as low as zero since causality
has not yet been established between exposure to chlorinated water and bladder cancer.
Sources: Summarized from detailed figures presented in Appendix E (Exhibits E.38d, E.38h and E.38I) and F (Exhibits F.2v
and F.2w, F.3v and F.3w).
Final Economic Analysis for the Stage 2 DBPR
6-6
December 2005
-------
6.2 Problem Identification and Assessment of Potential Hazard
This section provides detailed information from the toxicological and epidemiological literature
for the key adverse health effects that have been associated with exposure to DBFs. In addition to the
studies and reviews presented here, EPA has addressed reproductive and developmental effects,
carcinogenicity, and other adverse health effects at length in several Health Criteria Documents.
Specifically, EPA has developed Drinking Water Criteria Documents for the following DBFs: brominated
trihalomethanes (USEPA 2005b), brominated haloacetic acids (USEPA 2005c), trichloroacetic acid (2005
d), and monochloroacetic acid (USEPA 2005e). EPA has also completed toxicological reviews of
dichloroacetic acid (IRIS 2003), bromate (IRIS 2001a), chloroform (IRIS 2001b), chlorine dioxide and
chlorite (IRIS 2000), and an addendum for dichloroacetic acid (USEPA 2005J). A similar document
exists for 3-Chloro-4-(dichloromethyl)-5-hydroxy-2(5H)-furanone (MX) (USEPA 2000f).
EPA's weight of evidence evaluation of the best available science on carcinogenicity and
reproductive and developmental effects, in conjunction with the widespread exposure to DBFs, supports
the incremental regulatory changes in today's rule that target lowering DBFs and providing equitable
public health protection.
6.2.1 Cancer
Several DBFs have been identified by EPA as probable or possible human carcinogens. EPA
believes that the cancer epidemiology and toxicology literature provides important information that
contributes to the weight of evidence for potential health risks from exposure to chlorinated drinking
water. At this time, the cancer epidemiology studies support a potential association between exposure to
chlorinated drinking water and cancer, but evidence is insufficient to establish a causal relationship. The
epidemiological evidence for an association between DBP exposure and colon and rectal cancers is not as
consistent as it is for bladder cancer, although similarity of effects reported in animal toxicity and human
epidemiology studies strengthens the evidence for an association with colon and rectal cancers. EPA
believes that the overall cancer epidemiology and toxicology data support the decision to pursue
additional DBP control measures as reflected in the Stage 2 DBPR. The following sections provide an
overview of the epidemiological and toxicological evidence for the carcinogenicity of key DBFs.
6.2.1.1 Epidemiological Evidence of DBP Carcinogenicity
A number of epidemiological studies have been conducted to investigate the relationship between
exposure to chlorinated drinking water and various cancers. These studies contribute to the overall
evidence on potential human health hazards from exposure to chlorinated drinking water.
Epidemiology studies provide useful health effects information because they reflect human
exposure to a drinking water DBP mixture through multiple routes of intake such as ingestion, inhalation
and dermal absorption. The greatest difficulty with conducting cancer epidemiology studies is the length
of time between exposure and effect. Higher quality studies have adequately controlled for confounding
and have limited the potential for exposure misclassification, for example, using DBP levels in drinking
water as the exposure metric as opposed to type of source water. Study design considerations for
interpreting cancer epidemiology data include sufficient follow-up time to detect disease occurrence,
adequate sample size, valid ascertainment of cause of the cancer, and reduction of potential selection bias
in case-control and cohort studies (by having comparable cases and controls and by limiting loss to
follow-up). Epidemiology studies provide extremely useful information on human exposure to
chlorinated water, which complement single chemical, high dose animal data.
Final Economic Analysis for the Stage 2 DBPR 6-7 December 2005
-------
Bladder Cancer - Causes and Risk Factors
The National Cancer Institute (NCI) lists the following primary risk factors for bladder cancer:
age, tobacco, occupation, certain infections, certain drug treatments, race, being male, family history and
personal history (NCI 2002). The American Cancer Society (ACS) estimates that there will be about
60,240 new cases of bladder cancer diagnosed in the United States (about 44,460 men and 15,600
women) in 2004 (ACS 2004). The cancer is more common in men than in women, although women who
smoke have twice the risk of men who smoke (Mayo Clinic 2004). Whites are about two times more
likely to develop bladder cancer than African Americans or Hispanics (ACS 2004).
The literature on bladder cancer and its causes and risk factors describes several well-known
etiologic agents. Two useful review articles on bladder cancer in humans are Cohen et al. (2000) and
Fukushima and Wanibuchi (2000). The following is derived largely from these reviews.
Bladder cancer involves a heterogenous group of tumors, but is often categorized into one of two
main types: squamous cell carcinomas (which are mainly seen as secondary to schistosomiasis
infections) and transitional (urothelial) cell carcinomas.
Schistosomiasis is caused by the parasite Schistosoma haematobium, which is most prevalent in
Egypt and other Nile River countries, but is found in other parts of Africa, the Middle East and India.
Generally, bladder cancers in areas that have endemic schistosomiasis involve squamous cell carcinomas
rather than transitional cell carcinomas, and also tend to occur in somewhat younger individuals (40s or
50s) than those typically experiencing transitional cell carcinomas. Cohen et al. (2000) indicated that the
incidence and mortality rates for bladder tumors of this etiology are not well characterized. Bladder
cancer related to Schistosoma infections does not appear to be directly relevant to that associated with
DBFs or other chemical agents, and is therefore not considered further here.
Bladder cancers not associated with Schistosoma infections occur throughout the developed
world, and these are almost all of the transitional cell carcinoma type, and include low-grade, recurrent
papillary tumors and high-grade invasive malignancies. In countries without schistosomiasis, over 95
percent of bladder cancers involve transitional cell tumors. There is a relatively long history of
association of these bladder cancers with specific environmental and occupational factors. For example,
an association of bladder cancer with workers in the aniline dyes industry dates back to a 1895 German
study by Rehn, with similar observations between exposure to certain aromatic amines and related
compounds in the dye industry throughout the twentieth century. A number of specific aromatic amine
compounds considered to be bladder carcinogens have been identified through epidemiological studies,
occupational studies, and animal toxicity studies. Well-recognized among these are:
• 2-Naphthylamine
• 4-Aminobiphenyl
Benzidine (and some benzidine-related "azo" dyes)
4,4-Methylenebis(2-chloraoaniline) (MBOCA)
• o-Toluidine
4-Chloro-o-toluidine
Methylenedianiline (MDA)
Other specific chemical compounds that have been associated with bladder cancer include the
analgesic phenacetin and some chemotherapeutic agents, such as cyclophosphamide and chlornaphazine
(N,N-bis(2-chloroethyl)-2-naphthylamine).
Although there has been some indication of a relationship between certain artificial sweeteners
(saccharin and cyclamates) with bladder cancer, these are now generally considered very weak
Final Economic Analysis for the Stage 2 DBPR 6-8 December 2005
-------
associations at best (Mayo Clinic 2004). Also, there do not appear to be any dietary factors associated
with bladder cancer beyond the associations made between cancer in general and diets high in fats, red
meats and fried foods. In a systematic literature review, Zeegers et al. (2004) concluded that coffee and
tea consumption are probably not associated with bladder cancer. While this study found some
convincing evidence of an association of alcohol consumption and bladder cancer, this risk was not found
to be statistically significant.
The most significant environmental factor that is associated with bladder cancer is tobacco smoke
(ACS 2004). The causative agent(s) in tobacco smoke is not known, but it should be noted that cigarette
smoke contains aromatic amine compounds, including some of those that have been specifically linked
with bladder cancer in other studies as noted above. Ingestion of arsenic, notably as a drinking water
contaminant, has been associated with bladder cancer. Consumption of chlorinated drinking water has
also been associated with bladder cancer.
It has been reported that smoking is the attributable factor for about 50 percent of bladder cancers
in men and 30 percent in women (NCI 2004). The NCI also indicates that up to 25 percent of bladder
cancers may be attributable to occupational factors, again notably those involving exposure to certain
aromatic amine compounds (Mayo Clinic 2004). Other studies have estimated that occupational exposure
may be the attributable factor in 20 percent of cases (Silverman et al. 1989a, Silverman et al. 1989b, and
Silverman et al. 1990). Less than half a percent of bladder cancers can be attributed to a rare bladder birth
defect, and hereditary factors may account for 1 percent (ACS 2004).
In the benefits analysis supporting the January 2001 arsenic rule, EPA did not derive an estimate
of the bladder cancer attributable factor for arsenic in drinking water, and no other estimate was found in
the literature. However, based on estimates of annual bladder cancer cases avoided by the various arsenic
regulations considered, including the most stringent of them at 3 • g/L (approximately 30 to 80 cases per
year), it would appear that the total cases attributable to arsenic in drinking water is a small fraction of the
approximately 60,000 new cases reported each year.
For both the Stage 1 Final Rule (November 1998) and the proposed Stage 2 Rule (August 2003),
EPA provided estimates that chlorinated drinking water consumed prior to these rules maybe responsible
for between 2 and 17 percent of bladder cancers.
It is not appropriate simply to add the estimates of percent contribution to total bladder cancer
estimated separately for each of the individual causes described above, as there may be overlap among
them. The Venn diagram shown in Exhibit 6.2 is intended to provide a schematic depiction of possible
overlaps among the major attributable causes of bladder cancer noted above: smoking, occupational
exposures, and drinking water.
Final Economic Analysis for the Stage 2 DBPR 6-9 December 2005
-------
Exhibit 6.2 Venn Diagram of Bladder Cancer in the U.S. Population
H
The total bladder cancers each year is represented by the space in the rectangle, and is the sum of
A+B+C+D+E+F+G+H =100 percent. Smoking, noted previously as being associated with
approximately 50 percent of bladder cancers, is represented by the circle with areas A+B+C+D.
Occupational sources, estimated as the principal factor for 25 percent of bladder cancer cases, is the circle
composed of areas B+C+E+F. Drinking water is represented by the circle with areas C+D+F+G.
The overlap areas among these circles (that is, areas B, C, D and F) represent circumstances
where contributions from multiple sources may not be fully accounted for in the epidemiological data. To
the extent this is the case, the sum of the two major individual sources noted above - smoking at 50
percent, occupational at 25 percent - would be less than the 75 percent implied by adding those two
values.
Bladder Cancer - Studies Supporting EPA 's PAR Analysis
More evidence is available to support a possible association between bladder cancer and
chlorinated water or DBF exposure than for other cancers. The Stage 1 DBPR Regulatory Impact
Assessment (USEPA 1998a) presents EPA's review of the large body of epidemiology literature for
bladder cancer and its association with DBFs in drinking water. From this review, EPA concluded that
although causality has not been established, the data support a potential association, which is a concern.
Particular gaps in EPA's understanding include the reason for inconsistent results across subpopulations
in the different studies, especially for males versus females and smokers versus nonsmokers.
Final Economic Analysis for the Stage 2 DBPR
6-10
December 2005
-------
For both the Stage 1 DBPR EA and the Stage 2 proposal EA, EPA used five epidemiological
studies conducted in the 1980s and 1990s to calculate a range of PAR values for bladder cancer
associated with exposure to chlorinated drinking water. The five epidemiological studies used by EPA
are as follows (note that Cantor et al. 1985 and Cantor et al. 1987 use the same epidemiological data):
• Cantor etal. (1985; 1987)
• McGeehinetal. (1993)
• King and Marrett (1996)
Freedman et al. (1997)
• Cantor etal. (1998)
Exhibit 6.3 provides relevant summary information for each of these studies and the PAR values
calculated from them by EPA. Appendix E provides additional information on the derivation and use of
PAR values in general, as well as additional details on the PAR values derived from these studies by
EPA.
All of these studies include adjustments in their analyses to account for possible confounding by
other factors that may contribute to bladder cancer, notably sex, age, and smoking. Cantor et al. and
McGeehin et al. also included adjustments for occupational exposure.
As shown in Exhibit 6.3, the estimated PAR percentages from the studies range from 2 percent to
17 percent. Those values are "best estimates" derived from the study data as described in Appendix E.
EPA has also estimated 95 percent confidence intervals for those PAR values using a Monte Carlo
simulation procedure, which is also described in Appendix E. In most cases, the lower confidence bound
has been truncated at 0 percent based on biological plausibility considerations. That is, notwithstanding
statistical indications of PAR values < 0 percent implied by odds ratios < 1.0, there is no toxicological or
epidemiological data to support a conclusion that increased DBP exposure would reduce bladder cancer.
Final Economic Analysis for the Stage 2 DBPR 6-11 December 2005
-------
Exhibit 6.3 Summary of Epidemiology Studies for Bladder Cancer Associated
with Chlorinated Drinking Water and EPA Calculated PAR Values
Study
Cantor et al.
(1985)
Cantor et al.
(1987)2
McGeehin et
al. (1993)
King and
Marrett(1996)
Freedman et
al. (1997)
Cantor et al.
(1998)
Description
Case-control study of
association between
bladder cancer and
consumption of
chlorinated surface
water
Case-control study of
association between
bladder cancer and
consumption of
chlorinated surface
water
Case-control study of
association between
bladder cancer and
consumption of
chlorinated surface
water
Case-control study of
association between
bladder cancer and
consumption of
chlorinated surface
water
Nested case-control
study of association
between bladder
cancer and
consumption of
chlorinated drinking
water
Case-control study of
association between
bladder cancer and
consumption of
chlorinated surface
water
Summary of Results
- Odds ratio for all whites with over 59 years of
exposure is 1.1 (95% Confidence Interval: 0.8-
1.5)
- Odds ratio for nonsmokers is 2.3 (95%
Confidence Interval: 1.3-4.2)
- Odds ratio for current smokers is 0.6 (95%
Confidence Interval: 0.3-1.2)
- Odds ratio for both sexes with over 59 years
of exposure to tap water is 1 .4 (95%
Confidence Interval: 0.9-2.3)
- Odds ratio for nonsmokers with over 59
years of exposure to tap water is 3.1 (95%
Confidence Interval: 1.3-7.3)
- Odds ratio for bladder cancer with over 30
years of exposure is 1.8 (95% Confidence
Interval: 1.1-2.9)
- Odds ratio for cases consuming over 5
glasses of tap water per day is 2.0 (95%
Confidence Interval: 1.1-2.8)
- Odds ratio for bladder cancer for 35 years of
exposure compared to 10 years is 1.42 (95%
Confidence Interval: 1.10-1.81)
- Bladder cancer risk increased with years of
exposure
- Risk increases by 1 1 percent with each
1,OOOug/LTHM-year3
- Odds ratio for bladder cancer using 1975
measure of exposure is 1.2 (95% Confidence
Interval: 0.9-1.6)
- Slight gradient of increasing risk with
increasing duration noted only among
smokers
- Little overall association between bladder
cancer risk and exposure to chlorination
byproducts
- Bladder cancer risk increased with exposure
duration
Comments
Majority of water
systems contained
less than 20 pg/L
THMs.
Results were
statistically significant
for non-smokers only
Level of total THMs,
residual chlorine, or
nitrates not associated
with bladder cancer
risk controlling for
years of exposure.
Statistically significant
only for lengthy
exposures. Results
provide no support for
an interaction between
volume of water
consumed and years
of exposure to THMs
level > 49 ug/L.
Further stratification
by gender showed
elevated odds ratios to
be restricted to male
smokers.
Opposite trends were
found in males and
females. Total lifetime
and average lifetime
TTHM levels show all
risk increases are
apparently restricted
to male smokers.
PAR
(95% Cl)1
2%
(0%-15%)
15%
(0%-31%)
17%
(0% - 33%)
17%
(1%-28%)
3%
(0% - 22%)
3%
(0% - 8%)
1 Confidence intervals truncated at zero to reflect biological plausibility. The actual lower confidence level is often negative.
2 The Cantor et al. 1987 study is based upon the same data set as the Cantor et al. 1985 study. OR and PAR values for
Cantor 1987 reflect modifications to the inclusion criteria and adjustments for confounders relative to the analysis
performed in the 1985 study.
3 THM-years are the product of the continuous estimate of a given THM level (in ug/L) and years at that level, analogous to
pack-years of cigarette smoking.
Final Economic Analysis for the Stage 2 DBPR
6-12
December 2005
-------
Just prior to completion of the Stage 2 DBPR proposal, Villanueva et al. (2003) published a
meta-analysis of epidemiological studies addressing bladder cancer related to exposure to chlorinated
drinking water. The Villanueva et al. (2003) meta-analysis included most of the studies that EPA had
used for Stage 1 and the Stage 2 proposal (see Exhibit 6.3). The specific studies considered by these
authors are as follows:
• Cantor etal. (1998)
• Koivusalo etal. (1998)
King and Marrett (1996)
• McGeehinetal. (1993)
• Vena etal. (1993)
• Cantor etal. (1987)
• Wilkins and Comstock (1981)
• Doyle etal. (1997)
The first six of these are case-control studies, the latter two are cohort studies.
Villanueva et al. (2003) developed several sets of ORs reflecting different exposure conditions
(mid-term, long-term, and ever-exposed), separately for men, women as well as for both sexes combined.
In addition to estimating overall OR values for those several populations and exposure conditions,
Villanueva et al. (2003) presented data showing an increase in OR with increased duration of exposure.
The authors present a dose-response analysis quantifying that relationship.
For the purposes of supporting the final Stage 2 analysis, EPA has chosen to use the estimated
OR from Villanueva et al. (2003) for the "ever-exposed, both sexes" category. Exhibit 6.4 provides a
summary of the 6 studies that were used by Villanueva et al. (2003) for this exposure group; Exhibit 6.5
summarized the weights, the OR values for those individual studies and the combined OR obtained by the
authors.
Final Economic Analysis for the Stage 2 DBPR 6-13 December 2005
-------
Exhibit 6.4 Summary of Epidemiology Studies from Villanueva et al. (2003) for
Bladder Cancer Associated with Chlorinated Drinking Water used in
Developments of the PAR Analysis
Study
Description
Summary of Results
Comments
Cantor et
al. (1998)
Case-control study of
association between
bladder cancer and
consumption of
chlorinated surface
water.
- Little overall association between
bladder cancer risk and exposure to
chlorination byproducts
- Bladder cancer risk increased with
exposure duration
Opposite trends were found in males
and females. Total lifetime and
average lifetime TTHM levels show
all risk increases are apparently
restricted to male smokers.
Koivusalo
etal.
(1998)
Case-control study of
association between
estimated historical
exposure to drinking
water mutagenicity and
kidney and bladder
cancers.
- Non-significant excess risk of
bladder cancer associated with
mutagenic drinking water for men and
women
- Statistically significant OR (2.59,
95% Cl 1.13-5.94) fora 3,000 net
revertants/L increase in average
exposure for nonsmoking men with
• 80 years estimable exposure history
- Authors claim their study (at
publication) was first to report an
exposure-response relationship
between the quantitative level of
drinking water chlorination
by-products and kidney cancer.
- Authors acknowledge that higher
OR for those with • 80 years in
highest exposure category (in this
study, 3,000 net revertants/L) could
indicate that exposure period used
was too short to appropriately study
the relationship w/bladder cancer,
and risk may be underestimated in
this study.
King and
Marrett
(1996)
Case-control study of
association between
bladder cancer and
consumption of
chlorinated surface
water.
- Odds ratio for bladder cancer for 35
years of exposure compared to 10
years is 1.42 (95% Confidence
Interval: 1.10-1.81)
- Bladder cancer risk increased with
years of exposure
- Risk increases by 11 percent with
each 1,000 ug/L THM-year
Statistically significant only for
lengthy exposures. Results provide
no support for an interaction between
volume of water consumed and years
of exposure to THMs levels > 49
M9/L.
McGeehin
etal.
(1993)
Case-control study of
association between
bladder cancer and
consumption of
chlorinated surface
water.
- Odds ratio for bladder cancer with
over 30 years of exposure is 1.8 (95%
Confidence Interval: 1.1-2.9)
- Odds ratio for cases consuming over
5 glasses of tap water per day is 2.0
(95% Confidence Interval: 1.1-2.8)
Level of total THMs, residual
chlorine, or nitrates are not
associated with bladder cancer risk,
controlling for years of exposure.
Cantor et
al. (1987)1
Case-control study of
association between
bladder cancer and
consumption of
chlorinated surface
water.
- Odds ratio for both sexes with over
59 years of exposure to tap water is
1.4 (95% Confidence Interval: 0.9-2.3)
- Odds ratio for nonsmokers with over
59 years of exposure to tap water is
3.1 (95% Confidence Interval: 1.3-7.3)
Long-term bladder cancer risks are
more prominent in nonsmokers.
Wilkins and
Comstock
(1981)2
Cohort study of
association between
bladder cancer and
chlorinated surface
water.
- Relative Risk of 1.8 for men (95%
Confidence Interval: 0.8-4.75)
- Relative Risk of 1.6 for women (95%
Confidence Interval: 0.54-6.32)
Results not statistically significant.
1The Cantor et al. 1987 study is based upon the same data set as the Cantor et al. 1985 study. OR and PAR values for Cantor
1987 represent modifications to the inclusion criteria and adjustments for confounders relative to the analysis performed in the 1985
study.
2 Freedman et al. (1997) is a subset of the Wilkins and Comstock 1981 study and was used in Stage 1 and the Stage 2 proposal for
calculation of bladder cancer PAR range values.
Final Economic Analysis for the Stage 2 DBPR
6-14
December 2005
-------
Exhibit 6.5 Estimated OR for Ever-Exposed, Both Sexes Category from
Villanueva et al. (2003) Meta-Analysis
Study
Cantor et al. (1998)
Koivusalo et al. (1998)
King and Marrett (1 996)
McGeeehin et al. (1993)
Cantor et al. (1987)
Wilkins&Comstock(1981)
Weight
34.9
6.6
19.2
8.5
28.7
2.2
Combined
OR
1.1
1.4
1.4
1.3
1.2
1.7
1.2
95% Cl
0.9-1.3
0.9-2.1
1.1 - 1.8
0.9-1.9
1.0-1.5
0.8-3.6
1.1 -1.4
Based on the combined OR value of 1.2 and the corresponding 95 percent CI values of 1.1 and
1.4, respectively, EPA calculated a PAR from the Villanueva et al. (2003) study of 15.8 percent (95
percent CI = 8.5 percent - 27.2 percent). See Appendix E for details of the PAR calculation.
Subsequent to the Stage 2 DBPR proposal, Villanueva et al. (2004) published a pooled-analysis
of studies addressing the potential association between DBFs and bladder cancer. This pooled analysis
included the six studies summarized in Exhibit 6.6.
Exhibit 6.6 Summary of Epidemiology Studies from Villanueva et al. (2004) for
Bladder Cancer Associated with Chlorinated Drinking Water used in
Developments of the PAR Analysis
Study
Lynch et al.
(1989)
Cordieret al.
(1993)
Cantor et al.
(1998)
Description
Case control study of
association between years
of exposure to chlorinated
drinking water and bladder
cancer.
Hospital-based case control
study of occupational risks
of bladder cancer. TTHM
data previously
unpublished.
Case-control study of
association between
bladder cancer and
consumption of chlorinated
surface water.
Summary of Results
- Adjusted OR for both
sexes =1.52(95%CI 1.10-
2.10) for average exposure
more than 1« g/L THM
compared with • «1 • g/L THM
- Adjusted OR for both
sexes = 1 .02 (95% CI 0.66-
1 .57) for average exposure
more than 1« g/L THM
compared with • «1 • g/L THM
- Little overall association
between bladder cancer risk
and exposure to chlorination
byproducts
- Bladder cancer risk
increased with exposure
duration
Comments
Statistically significant
for both sexes
combined, and also
increased for men and
women separately.
TTHM data was
previously unpublished.
Opposite trends were
found in males and
females. Total lifetime
and average lifetime
TTHM levels show all
risk increases are
apparently restricted to
male smokers.
Final Economic Analysis for the Stage 2 DBPR
6-15
December 2005
-------
Study
Description
Summary of Results
Comments
Koivusalo et
al. (1998)
Case-control study of
association between
estimated historical
exposure to drinking water
mutagenicity and kidney
and bladder cancers.
- Non-significant excess risk
of bladder cancer associated
with mutagenic drinking
water for men and women
- Statistically significant OR
(2.59, 95% Cl 1.13-5.94) for
a 3,000 net revertants/L
increase in average
exposure for nonsmoking
men with • 80 years
estimable exposure history
- Authors claim their
study (at publication)
was first to report an
exposure-response
relationship between
the quantitative level of
drinking water
chlorination
by-products and kidney
cancer.
- Authors acknowledge
that higher OR for
those with • 80 years in
highest exposure
category (in this study,
3,000 net revertants/L)
could indicate that
exposure period used
was too short to
appropriately study the
relationship w/bladder
cancer, and risk may
be underestimated in
this study.
King and
Marrett(1996)
Case-control study of
association between
bladder cancer and
consumption of chlorinated
surface water.
- Odds ratio for bladder
cancer for 35 years of
exposure compared to 10
years is 1.42 (95%
Confidence Interval: 1.10-
1.81)
- Bladder cancer risk
increased with years of
exposure
- Risk increases by 11
percent with each 1,000
ug/L THM-year
Statistically significant
only for lengthy
exposures. Results
provide no support for
an interaction between
volume of water
consumed and years of
exposure to THMs
levels > 49 ug/L.
Porru (2003)
Hospital-based case control
study of association
between years of exposure
to chlorinated drinking
water and bladder cancer.
TTHM data previously
unpublished.
- Adjusted OR for men =
4.74 (95% Cl 0.76-29.6) for
average exposure more than
1 • g/L THM compared with • •
1-g/LTHM
TTHM data was
previously unpublished.
This pooled analysis (Villanueva et al. 2004) focused on TTHM exposure specifically and
presented OR estimates (adjusted for age, smoking, occupation, coffee consumption and education) for
men and women separately, as well as for both sexes combined for any exposure to TTHMs and as a
function of average TTHM exposure and cumulative TTHM exposure. The authors also evaluated the
relationship between OR for bladder cancer and duration of exposure to chlorinated water.
For support of the Stage 2 DBPR, EPA is using information based on the relationship between
OR and average TTHM exposure from the Villanueva et al. (2004) study for both sexes combined. These
estimates as presented in Villanueva et al. (2004) are shown in Exhibit 6.7.
Final Economic Analysis for the Stage 2 DBPR
6-16
December 2005
-------
Exhibit 6.7 Summary of Estimated OR Values Associated with Average TTHM
Exposures for Both Sexes from Villanueva et al. (2004)
Average TTHM (M9/L)
0
>0
0-1
> 1
0-1
>1 -5
>5-25
> 25 - 50
>50
OR
1.0
1.2
1.0
1.2
1.0
1.1
1.2
1.2
1.3
95% Cl
NA
1.0-1.4
NA
1.1 -1.3
NA
0.9-1.3
1.0-1.4
1.0-1.4
1.1 -1.5
EPA also obtained additional detailed data on the relationship between average TTHM levels and
OR from the authors of the Villanueva et al. (2004) study (Kogevinas and Villanueva 2005). These
additional data are presented in Exhibit 6.8. Using a linear relationship fit to the data in Exhibit 6.8
provided by Kogevinas and Villanueva (2005), and an estimate of a national average Pre-Stage 1 TTHM
concentration of 38.05 (ig/L, EPA has estimated a pre-Stage 1 PAR value of 17.1 percent (95 percent CI
= 2.5 percent -33.1 percent). See Appendix E for details of the PAR calculation.
Exhibit 6.9 provides a summary of the estimated annual bladder cancer cases attributable to DBFs
reflecting pre-Stage 1 occurrence and exposure levels for the three approaches to estimating risk
described above:
• Range of PAR values from five individual studies
• Villanueva et al. (2003), and
• Villanueva et al. (2004).
Taken together, the three approaches provide a reasonable estimate of the range of potential risks.
These estimates, summarized in Exhibit 6.9, assume an annual total of 56,506 new cases of bladder
cancer from all causes. These values are obtained by multiplying the 56,506 total cases by the
appropriate Pre-Stage 1 PAR values presented.
Final Economic Analysis for the Stage 2 DBPR
6-17
December 2005
-------
Exhibit 6.8 Detailed Data on OR as a Function of Average TTHM Exposure Level
Provided by Kogevinas and Villanueva (2005)
Average
TTHM (|jg/L)
0
10
20
30
40
50
60
70
80
90
100
110
120
130
Odds Ratio
1.0
1.1
1.2
1.2
1.2
1.2
1.3
1.3
1.4
1.5
1.6
1.7
1.8
1.9
Lower 95% Cl
-
1.0
1.0
1.0
1.0
1.0
1.1
1.1
1.1
1.1
1.1
1.1
1.0
1.0
Upper 95% Cl
-
1.3
1.4
1.4
1.4
1.4
1.5
1.6
1.7
1.9
2.2
2.6
3.1
3.7
Exhibit 6.9 Estimates of Pre-Stage 1 Annual Bladder Cancer Cases
Attributable to DBPs
Study
Five Studies Used in Stage 1
and Stage 2 Proposal1
Villanueva et al. (2003)
Villanueva et al. (2004)
Lower 95% Cl
0
4,830
1,412
Best Estimate of
Annual Cases
Attributable to DBPs
1,130-9,606
8,899
9,670
Upper 95% Cl
18,647
15,376
18,716
1Forthe estimates for the five studies used in Stage 1 and the Stage 2 proposal, the range shown for the "Best
Estimate" reflects the 2 percent and 17 percent PAR values; the lower 95 percent Cl reflects 0 percent PAR and the
upper 95 percent Cl reflects 33 percent PAR (the highest of the upper 95 percent Cl estimates for the individual
studies).
The details of the calculations of the Pre-Stage 1 attributable cases shown in Exhibit 6.9 are
provided in Appendix E.
Final Economic Analysis for the Stage 2 DBPR
6-18
December 2005
-------
Other New Cancer Studies
In the Stage 1 DBPR, EPA concluded that the epidemiological evidence suggested a potential
increased risk for bladder cancer due to DBF exposure. Some key studies EPA considered for Stage 1
include Cantor et al. (1998), Doyle et al. (1997), Freedman et al. (1997), King and Marrett (1996),
McGeehin et al. (1993), Cantor et al. (1987), and Cantor et al. (1985). Several studies published since the
Stage 1 DBPR continue to support an association between increased risk of bladder cancer and exposure
to chlorinated surface water (Chevrier et al. 2004, Koivusalo et al. 1998, Yang et al. 1998). One study
found no effects on a biomarker of genotoxicity in urinary bladder cells from TTHM exposure
(Ranmuthugala et al. 2003). Epidemiological reviews and meta-analyses generally support the possibility
of an association between chlorinated water or THMs and bladder cancer (Villanueva et al. 2004,
Villanueva et al. 2003, Villanueva et al. 2001, Mills et al. 1998). The World Health Organization (WHO
2000) found data inconclusive or insufficient to determine causality between chlorinated water and any
health endpoint, although they concluded that the evidence is better for bladder cancer than for other
cancers.
In the Stage 1 DBPR, EPA concluded that early studies suggested a small possible increase in
rectal and colon cancers from exposure to chlorinated surface waters. The database of studies on colon
and rectal cancers continues to support a possible association, but evidence remains mixed. For colon
cancer, one newer study supports the evidence of an association (King et al. 2000a) while others showed
inconsistent findings (Hildesheim et al. 1998, Yang et al. 1998). Rectal cancer studies are also mixed.
Hildesheim et al. (1998) and Yang et al. (1998) support an association with rectal cancer while King et al.
(2000a) did not. A review of colon and rectal cancer concluded evidence was inconclusive but that there
was a stronger association for rectal cancer and chlorination DBFs than for colon cancer (Mills et al.
1998). The WHO (2000) review reported that studies showed weak to moderate associations with colon
and rectal cancers and chlorinated surface water or THMs but that evidence is inadequate to evaluate
these associations.
Recent studies on kidney, brain, and lung cancers and DBP exposure support a possible
association (kidney: Yang et al. 1998, Koivusalo et al. 1998; brain: Cantor et al. 1999; lung: Yang et al.
1998). However, so few studies have examined these endpoints that definitive conclusions cannot be
made. Studies on leukemia found little or no association with DBFs (Infante-Rivard et al. 2002,
Infante-Rivard et al. 2001). A recent study did not find an association between pancreatic cancer and
DBFs (Do et al. 2005). A study researching multiple cancer endpoints found an association between
THM exposure and all cancers when grouped together (Vinceti et al. 2004). In the development of the
Stage 2 DBPR, EPA has evaluated several key bladder cancer studies. Summary information on these
studies can be found in Exhibit 6.10.
Final Economic Analysis for the Stage 2 DBPR 6-19 December 2005
-------
Exhibit 6.10 Summary of Bladder Cancer Epidemiology Studies and Review/Meta-analysis Studies
Reviewed for Stage 2 DBPR
Author(s)
Study Type
Exposure(s)
Studied
Outcome(s)
Measured
Findings
Do et al. 2005
Case-control
study in
Canada, 1994-
1997.
Estimated chlorinated
DBFs, chloroform,
BDCM concentrations.
Pancreatic
cancer.
No association was found between pancreatic cancer and
exposure to chlorinated DBFs, chloroform, or BDCM.
Chevrier et al.
2004
Case-control
study in
France, 1985-
1987.
Compared THM levels,
duration of exposure,
and 3 types of water
treatment (ozonation,
chlorination,
ozonation/chlorination)
Bladder
cancer.
A statistically significant decreased risk of bladder cancer was
found as duration of exposure to ozonated water increased. This
was evident with and without adjustment for other exposure
measures. A small association was detected for increased bladder
cancer risk and duration of exposure to chlorinated surface water
and with the estimated THM content of the water, achieving
statistical significance only when adjusted for duration of ozonated
water exposures. Effect modification by gender was noted in the
adjusted analyses.
Vinceti et al.
2004
Retrospective
cohort study in
Italy, 1987-
1999.
Standardized mortality
ratios from all causes
vs. cancer for
consumers drinking
water with high THMs.
15 cancers
including
colon,
rectum, and
bladder.
Mortality ratio from all cancers showed a statistically significant
small increase for males consuming drinking water with high
THMs. For females, an increased mortality ratio for all cancers
was seen but was not statistically significant. Stomach cancer in
men was the only individual cancer in which a statistically
significant excess in mortality was detected for consumption of
drinking water with high THMs.
Ranmuthugala
et al. 2003
Cohort study in
3 Australian
communities,
1997.
Estimated dose of
TTHM, chloroform,
and bromoform from
routinely-collected
THM measurements
and fluid intake diary.
Frequency of
micronuclei in
urinary
bladder
epithelial
cells.
Relative risk estimates for DMA damage to bladder cells for THM
dose metrics were near 1.0. The study provides no evidence that
THMs are associated with DMA damage to bladder epithelial cells,
and dose-response patterns were not detected.
Infante-Rivard
et al. 2002
Population-
based case-
control study in
Quebec, 1980-
1993.
Estimated prenatal
and postnatal
exposure to THMs and
polymorphisms in two
genes.
Acute
lymphoblastic
leukemia.
Data are suggestive, but imprecise, linking DMA variants with risk
of acute lymphoblastic leukemia associated with drinking water
DBPs. The number of genotyped subjects for GSTT1 and
CYP2E1 genes was too small to be conclusive.
Final Economic Analysis for the Stage 2 DBPR
6-20
December 2005
-------
Author(s)
Infante-Rivard
etal. 2001
King et al.
2000a
Cantor et al.
1999
Cantor et al.
1998
Hildesheim et
al. 1998
Study Type
Population-
based case-
control study in
Quebec, 1980-
1993.
Population-
based case-
control study in
southern
Ontario, 1992-
1994.
Population-
based case-
control study in
Iowa, 1984-
1987.
Population-
based case-
control study in
Iowa, 1986-
1989.
Population-
based case-
control study in
Iowa, 1986-
1989.
Exposure(s)
Studied
Compared water
chlorination (never,
sometimes, always)
and exposure to
TTHMs, metals, and
nitrates.
Compared source of
drinking water and
chlorination status.
Estimated TTHM
levels, duration of
exposure, and tap
water consumption.
Compared level and
duration of THM
exposure (cumulative
and average), source
of water, chlorination,
and water
consumption.
Compared level and
duration of THM
exposure (cumulative
and average), source
of water, chlorination,
and water
consumption.
Compared level and
duration of THM
exposure (cumulative
and average), source
of water, chlorination,
and water
consumption.
Outcome(s)
Measured
Acute
lymphoblastic
leukemia.
Colon and
rectal cancer.
Brain cancer.
Bladder
cancer.
Colon and
rectal cancer.
Findings
No increased risk for lymphoblastic leukemia was observed for
prenatal exposure at average levels of TTHMs, metals or nitrates.
However, a non-statistically significant, small increased risk was
seen for postnatal cumulative exposure to TTHMs and chloroform
(both at above the 95th exposure percentile of the distribution for
cases and controls), for zinc, cadmium, and arsenic, but not other
metals or nitrates.
Colon cancer risk was statistically associated with cumulative long
term exposure to THMs, chlorinated surface water, and tap water
consumption metrics among males only. Exposure-response
relationships were evident for exposure measures combining
duration and THM levels. Associations between the exposure
measures and rectal cancer were not observed for either gender.
Among males, a statistically significant increased risk of brain
cancer was detected for duration of chlorinated versus non-
chlorinated source water, especially among high-level consumers
of tap water. An increased risk of brain cancer for high water
intake level was found in men. No associations were found for
women for any of the exposure metrics examined.
A statistically significant positive association between risk of
bladder cancer and exposure to chlorinated groundwater or
surface water reported for men and for smokers, but no
association found for male/female non-smokers, or for women
overall. Limited evidence was found for an association between
tapwater consumption and bladder cancer risk. Suggestive
evidence existed for exposure-response effects of chlorinated
water and lifetime THM measures on bladder cancer risk.
Increased risks of rectal cancer was associated with duration of
exposure to chlorinated surface water and any chlorinated water,
with evidence of an exposure-response relationship. Risk of rectal
cancer is statistically significant increased with >60 years lifetime
exposure to THMs in drinking water, and risk increased for
individuals with low dietary fiber intake. Risks were similar for men
and women and no effects were observed for tapwater measures.
No associations were detected for water exposure measures and
risk of colon cancer.
Final Economic Analysis for the Stage 2 DBPR
6-21
December 2005
-------
Author(s)
Koivusalo et
al. 1998
Yang et al.
1998
Doyle et al.
1997
Freedman et
al. 1997
King and
Marrett 1 996
Study Type
Population-
based case-
control study in
Finland, 1991-
1992.
Cross-sectional
study in
Taiwan, 1982-
1991.
Prospective
cohort study in
Iowa, 1987-
1993.
Population-
based case-
control study in
Maryland,
1975-1992.
Case-control
study in
Ontario,
Canada, 1992-
1994.
Exposure(s)
Studied
Estimated residential
duration of exposure
and level of drinking
water mutagenicity.
Examined residence in
chlorinated (mainly
surface water sources)
relative to non-
chlorinated (mainly
private well) water.
Examined chloroform
levels and source of
drinking water.
Estimated duration of
exposure to
chlorinated water.
Compared exposure to
chlorinated municipal
water (yes/no).
Compared source of
drinking water and
chlorination status.
Estimated TTHM
levels, duration of
exposure, and tap
water consumption.
Outcome(s)
Measured
Bladder and
kidney
cancer.
Cancer of
rectum, lung,
bladder,
kidney, colon,
and 11
others.
Colon,
rectum,
bladder, and
8 other
cancers in
women.
Bladder
cancer.
Bladder
cancer.
Findings
Drinking water mutagenicity was associated with a small,
statistically significant, exposure-related excess risk for kidney and
bladder cancers among men; weaker associations were detected
for mutagenic water and bladder or kidney cancer among women.
The effect of mutagenicity on bladder cancer was modified by
smoking status, with an increased risk among non-smokers.
Residence in chlorinating municipalities (vs. non-chlorinating) was
statistically significantly associated with the following types of
cancer in both males and females: rectal, lung, bladder, and kidney
cancer. Liver cancer and all cancers were also statistically
significantly elevated in chlorinated towns for males only. Mortality
rates for cancers of the esophagus, stomach, colon, pancreas,
prostate, brain, breast, cervix uteri and uterus, and ovary were
comparable for chlorinated and non-chlorinated residence.
Statistically significant increased risk of colon cancer, breast
cancer and all cancers combined was observed for women
exposed to chloroform in drinking water, with evidence of
exposure-response effects. No associations were detected
between chloroform and bladder, rectum, kidney, upper digestive
organs, lung, ovary, endometrium, or breast cancers, or for
melanomas or non-Hodgkin's lymphoma. Surface water exposure
(compared to ground water users) was also a significant predictor
of colon and breast cancer risk.
There was a weak association between bladder cancer risk and
duration of exposure to municipal water for male cigarette
smokers, as well as an exposure-response relationship. No
association was seen for those with no history of smoking,
suggesting that smoking may modify a possible effect of
chlorinated surface water on the risk of bladder cancer.
Statistically significant associations were detected for bladder
cancer and chlorinated surface water, duration or concentration of
THM levels and tap water consumption metrics. Population
attributable risks were estimated at 14 to 16 percent. An exposure-
response relationship was observed for estimated duration of high
THM exposures and risk of bladder cancer.
Final Economic Analysis for the Stage 2 DBPR
6-22
December 2005
-------
Author(s)
Cordieref al.
1993
McGeehin et
al. 1993
Vena et al.
1993
Lynch et al.
1989
Cantor et al.
1987 (and
Cantor et al.
1985)
Wilkins and
Comstock
1981
Study Type
Hospital-based
case-control
study in
France, 1984-
1987.
Population-
based case-
control study in
Colorado,
1990-1991.
Case-control
study in
western NY,
1979-1985.
Case-control
study in Iowa,
1894-1979.
Population-
based case-
control study in
10 areas of the
U.S., 1977-
1978.
Cohort study in
Washington
County, MD,
1963-1975
Exposure(s)
Studied
Estimated duration of
exposure to TTHMs.
Compared source of
drinking water, water
treatment, and tap
water versus bottled
water. Estimated
duration of exposure to
TTHMs and levels of
TTHMs, nitrates, and
residual chlorine.
Compared
consumption of fluids,
including chlorinated
tap water.
Compared source of
drinking water.
Estimated chlorinated
water consumption
and duration of
exposure.
Compared source of
drinking water.
Estimated total
beverage and tap
water consumption
and duration of
exposure.
Compared chlorinated
surface water.
Outcome(s)
Measured
Bladder
cancer.
Bladder
cancer.
Bladder
cancer.
Bladder
cancer.
Bladder
cancer.
Bladder
cancer.
Findings
No associations were not detected for bladder cancer and
exposure to TTHMs.
Statistically significant associations were detected for bladder
cancer and duration of exposure to chlorinated surface water. The
risk was similar for males and females and among nonsmokers
and smokers. The attributable risk was estimated at 14.9 percent.
High tap water intake was associated with risk of bladder cancer in
a exposure-response fashion. No associations were detected
between bladder cancer and levels of TTHMs, nitrates, and
residual chlorine.
Bladder cancer and consumption of total fluids and tap water alone
showed a significant finding for both age categories (under 65
years and over 65 years). A dose-response relationship was
observed for consumption of tap water and total fluid intake. Risks
associated with tap water consumption were higher for
nonsmokers.
Bladder cancer was statistically associated with duration of
exposure to chlorinated drinking water sources for both sexes.
Bladder cancer was statistically associated with duration of
exposure to chlorinated surface water for women and nonsmokers
of both sexes. The largest risks were seen when both exposure
duration and level of tap water ingestion were combined. No
association was seen for total beverage consumption.
Incidence rates for bladder cancer among men were nearly twofold
higher in the chlorinated surface water group than in the referent
group (results not statistically significant).
Final Economic Analysis for the Stage 2 DBPR
6-23
December 2005
-------
Author(s)
Study Type
Exposure(s)
Studied
Outcome(s)
Measured
Findings
Reviews/Meta-analyses
Villanueva et
al. 2004
Review and
meta-analysis
of 6 case-
control studies.
Individual-based
exposure estimates to
THMs and water
consumption over a
40-year period.
Bladder
cancer.
The meta-analysis suggests that risk of bladder cancer in men
increases with long-term exposure to TTHMs. An exposure-
response pattern was observed among men exposed to TTHMs,
with statistically significant risk seen at exposures higher than 50
ug/L. No association between TTHMs and bladder cancer was
seen for women.
Villanueva et
al. 2003 (and
Goebell et al.
2004)
Review and
meta-analysis
of 6 case-
control studies
and 2 cohort
studies.
Compared source of
water and estimated
duration of exposure to
chlorinated drinking
water.
Bladder
cancer.
The meta-analysis findings showed a moderate excess risk of
bladder cancer attributable to long-term consumption of chlorinated
drinking water for both genders, particularly in men. Statistically
significance seen with men and combined both sexes. The risk
was higher when exposure exceeded 40 years.
Villanueva et
al. 2001
Qualitative
review of 31
cancer studies.
Compared exposure to
TTHM levels,
mutagenic drinking
water, water
consumption, source
water, types of
disinfection
(chlorination and
chloramination), and
residence times.
Cancer of
bladder,
colon,
rectum, and 5
other
cancers.
Review found that although results for cancer studies varied and
were not always statistically significant, evidence for bladder
cancer is strongest, and all 10 of the bladder cancer studies
showed increased cancer risks with ingestion of chlorinated water.
The authors felt associations with chlorinated water and cancer of
the colon, rectum, pancreas, esophagus, brain, and other cancers
were inconsistent.
WHO 2000
Qualitative
reviews of
various studies
in Finland,
U.S., and
Canada.
Various exposures to
THMs.
Various
cancers.
Studies reviewed reported weak to moderate increased
relative risks of bladder, colon, rectal, pancreatic, breast, brain or
lung cancer associated with long-term exposure to chlorinated
drinking water. The authors felt evidence is inconclusive for an
association between colon cancer and long-term exposure to
THMs, that evidence is insufficient to evaluate a causal
relationship between THMs and rectal, bladder, and other cancers.
They found no association between THMs and increased risk of
cardiovascular disease.
Final Economic Analysis for the Stage 2 DBPR
6-24
December 2005
-------
Author(s)
Mills et al.
1998
Study Type
Qualitative
review of 22
studies.
Exposure(s)
Studied
Examined TTHM
levels and water
consumption.
Compared source of
water and 2 types of
water treatment
(chlorination and
chloramination).
Outcome(s)
Measured
Cancer of
colon,
rectum, and
bladder.
Findings
Review suggests possible increases in risks of bladder cancer with
exposure to chlorinated drinking water. The authors felt evidence
for increased risk of colon and rectal cancers is inconclusive,
though evidence is stronger for rectal cancer.
Final Economic Analysis for the Stage 2 DBPR
6-25
December 2005
-------
6.2.1.2 Toxicological Evidence of DBF Carcinogenicity
Toxicological studies provide important information on the potential carcinogenicity of DBFs in
humans. EPA's Integrated Risk Information System (IRIS), which is accessible at
http://www.epa.gov/iris, provides detailed descriptions of cancer risk assessments that EPA has
performed for seven DBFs. Included on IRIS are weight-of-evidence characterizations of the
carcinogenic potential of those seven DBFs and lifetime unit cancer risk factors for five of the seven,
based primarily on animal toxicological data. As with all risk evaluations based on animal toxicological
studies, several extrapolations were required to establish lifetime unit cancer risks for humans (e.g., from
high to low doses, from nonhuman species to humans, and for DBFs, from gavage to ingestion of water).
Exhibit 6.11 provides a summary of the cancer risk assessments for those seven DBFs as presented on the
IRIS database.
Analyses done for the Stage 2 DBPR follow the 1999 EPA Proposed Guidelines for Carcinogenic
Risk Assessment (USEPA 1999b). In March 2005, EPA updated and finalized the Cancer Guidelines and
a Supplementary Children's Guidance, which include new considerations on mode of action for cancer
risk determination and additional potential risks due to early childhood exposure (USEPA 2005f, USEPA
2005g). Conducting the cancer evaluation using the 2005 Cancer Guidelines would not result in any
change from the existing analysis. With the exception of chloroform, no mode of action has been
established for other specific regulated DBFs. Although some of the DBFs have given mixed
mutagenicity and genotoxicity results, having a positive mutagenicity study does not necessarily mean
that a chemical has a mutagenic mode of action. The extra factor of safety for children's health protection
does not apply, because the new Supplementary Children's Guidance requires application of the
children's factor only when a mutagenic mode of action has been identified.
The lifetime unit risk factors shown in Exhibit 6.11 for bromoform, bromodichloromethane, and
dibromochloromethane were included in the cancer risk assessment and benefit analyses performed by
EPA in support of the Stage 1 DBPR promulgated in 1998. Since the Stage 1 DBPR, EPA has updated
the quantitative risk assessments for these three DBFs in order to represent the methodology proposed in
the 1996/1999 draft cancer guidelines (USEPA 1996c and 1999b), resulting in revisions to the unit risk
factors. Also, a new study of dichloroacetic acid (DCAA) tumorigenicity in mice by DeAngelo et al.
(1999) examined doses lower than those used in previously published studies and has been judged by
EPA to be suitable for quantification of risk, also using the newer methodology.
Except for DCAA, these updated cancer risk assessments do not yet appear on the IRIS database.
A toxicological review for DCAA exists on IRIS. EPA is completing a new brominated THM Criteria
Document for bromoform, bromodichloromethane, and dibromochloromethane that supports the Stage 2
Rule, and this document provides details on the animal toxicological data used to derive the new cancer
unit risk factors for these DBFs.
The updated cancer risk factors for these four DBFs are presented in Exhibit 6.12. They have
been used to estimate the pre-Stage 2 baseline cancer cases, the pre-Stage 1 concentrations of these
compounds, the changes in those concentrations following Stage 1, and the estimated number of people
exposed. As described in the Criteria Documents, cancer risk values were developed by fitting the key
animal toxicological data to linearized multistage models using the Maximum Likelihood Estimation
(MLE) method. MLE method is a standard statistical procedure used to estimate model parameter values
(in this case, parameters for the linearized multistage model) that have the highest likelihood from among
all possible parameters of generating the observed data. The risk factors shown in Exhibit 6.12 are then
computed from the linearized multistage dose-response model based on the MLE parameters obtained
using the study data.
Final Economic Analysis for the Stage 2 DBPR 6-26 December 2005
-------
The first risk factor is based on the estimated dose that the model predicts will result in a
carcinogenic response in 10 percent of the subjects (referred to as the Effective Dose for 10 percent
response, or ED10). (Note: This unit risk factor is also sometimes referred to as the MLE estimate since it
represents the dose taken directly from the curve fit by the MLE method.)
The second risk factor, which represents a more conservative estimate of the risk (and which
corresponds more directly to the lifetime unit risks shown in Exhibit 6.11 from the IRIS database), is
based on the lower 95 percent confidence bound on the dose that the model predicts will result in a
carcinogenic response in 10 percent of those exposed to the chemical, relative to control (referred to as
the Lower Bound on the Effective Dose for 10 percent response, or LED10).
Exhibit 6.11 Summary of EPA's Cancer Risk Assessments as currently presented
on IRIS for Specific DBFs
Chemical
Bromoform
Bromodichloromethane
Chloroform
Dibromochloromethane
Dichloroacetic Acid
Trichloroacetic Acid
Bromate
EPA's Human
Carcinogen Assessment
Probable1
Probable1
Probable1
Likely human carcinogen
under high-exposure
conditions that lead to
cytotoxicity and regenerative
hyperplasia in susceptible
tissues. Not likely without
cytotoxicity and cell
regeneration.2
Possible1
Likely2
Possible1
Probable1
Likely to be carcinogenic via
oral route of exposure2
Lifetime Unit
Cancer Risk
Factor
2.3X10'7('g/L)-1
1.8X10-6(-g/L)-1
Not Available
2.4X10-6(-g/L)-1
1 .4 X 1 0-6 (• g/L)'1
Not Available
2X10'5('g/L)-1
Date and Source
1993 (IRIS)
1993 (IRIS)
2001 (IRIS)
1992 (IRIS)
2003 (IRIS)
1996 (IRIS)
2001 (IRIS)
EPA's Human Carcinogen Assessment reported, as classified under EPA 1986 Cancer Risk Assessment Guidelines
(USEPA1986).
EPA's Human Carcinogen Assessment reported, as classified under EPA 1996 and 1999 Proposed Cancer Risk
Assessment Guidelines (USEPA 1996c and 1999b).
In both cases, EPA derives unit risk values assuming low-dose linearity and no threshold to
estimate risk. As shown in Exhibit 6.12, the baseline number of annual Pre-Stage 2 cancer cases
calculated from the risk factors for these four DBFs are 39 cases for the ED10 risk factors and 91 cases for
the LED10 risk factors. Assuming that DBF risk reductions for Stage 2 for the entire population average
7.76 percent, corresponding to the reduction in average TTHM levels (see Exhibit 6.19), Stage 2 cancer
cases avoided based on the toxicological data range from 1.7 to 4.0 cases per year.
Final Economic Analysis for the Stage 2 DBPR
6-27
December 2005
-------
Several limitations must be considered in conjunction with the interpretation and use of these
cancer risk estimates. There are only seven DBFs (those shown in Exhibit 6.11) for which EPA has
determined that adequate toxicology studies are available to support an assessment of their potential for
carcinogenicity in humans. As discussed elsewhere in this document, a large number of DBFs are present
in drinking water that has been disinfected, including many substances that have not yet been specifically
identified. It must also be recognized that these highly controlled toxicology studies involve exposure to
each respective DBF separately, while actual exposure to humans is to a mixture that includes many other
DBFs in a wide array of relative proportions. Lastly, these toxicology studies limit exposure to the oral
route only, whereas humans are generally exposed to DBFs in drinking water not only by the oral route
but by dermal exposure and inhalation as well.
Final Economic Analysis for the Stage 2 DBPR 6-28 December 2005
-------
Exhibit 6.12 Quantification of Cancer Risk for BDCM, Bromoform, DBCM, and DCAA, Pre-Stage 2 Baseline
Source Water
Type
Pre-Stage 2
Cone (ug/L),
Mean of Plant
Means,
DS Average
A
Population
B
Derivation of cases using ED10
Lifetime unit
risk (cases/
person)/
(mg/kg-day)
C
Lifetime unit risk
cone, (cases/
person)/ (ug/L)
D=C*(1/1000)*
(2L/day)*(1/70 kD)
Annual unit risk
cone, (cases/
person)/ (ug/L)
E=D/70 years
Baseline
Cases
F=A*B*E
Derivation of cases using LED10
Lifetime unit
risk (cases/
person)/
(mg/kg-day)
G
Lifetime unit risk
cone, (cases/
person)/ (ug/L)
H=G*(1/1000)*
(2L/day)*(1/70 kD)
Annual unit risk
cone, (cases/
person)/ (ug/L)
I=H/70 years
Baseline
Cases
J=A*B*I
BDCM
SW
GW
Total
8.20
3.05
169,358,139
93,666,379
263,024,518
2.2E-02
2.2E-02
6.29E-07
6.29E-07
8.98E-09
8.98E-09
12.5
2.6
15.0
3.4E-02
3.4E-02
9.71 E-07
9.71 E-07
1 .39 E-08
1 .39 E-08
19.3
4.0
23.2
Bromoform
SW
GW
Total
2.69
2.22
169,358,139
93,666,379
263,024,518
3.4E-03
3.4E-03
9.71 E-08
9.71 E-08
1 .39E-09
1 .39E-09
0.6
0.3
0.9
4.5E-03
4.5E-03
1 .29E-07
1 .29E-07
1 .84E-09
1 .84E-09
0.8
0.4
1.2
DBCM
SW
GW
Total
5.50
3.13
169,358,139
93,666,379
263,024,518
1.7E-02
1.7E-02
4.86E-07
4.86E-07
6.94E-09
6.94E-09
6.5
2.0
8.5
4.0E-02
4.0E-02
1.14E-06
1.14E-06
1 .63 E-08
1 .63 E-08
15.2
4.8
20.0
DCAA
SW
GW
Total
11.98
4.28
169,358,139
93,666,379
263,024,518
1.5E-02
1.5E-02
4.29E-07
4.29E-07
6.12E-09
6.12E-09
Grand Total
12.4
2.5
14.9
39.3
4.8E-02
4.8E-02
1 .36E-06
1 .36E-06
1 .94 E-08
1 .94 E-08
39.4
7.8
47.2
91.7
Note: Unit risk factors are different from Exhibit 6.5 - see text for discussion.
Sources: A) SW: SWAT DBP Summary Statistics, Run 300 (Pre-Stage 2); GW: Stage 2 Benefits Model
B) Stage 2 Population Baseline: Exhibit 3.3
C) The unit risk factor is based on the ED10 (effective dose for 10 percent response) based on the Maximum Likelihood Estimation (MLE) method
(provided by Nancy Chiu of EPA's Health and Ecological Criteria Division in email (5/22/03)).
D) This calculation assumes a 70 kg person with the average drinking water consumption rate of 2L/day. kD = kilogram-Day.
G) The unit risk factor is based on the LED10 (lower 95 percent confidence bound on effective dose for 10 percent response) based on the MLE
method (provided by Nancy Chiu in email (5/22/03)).
Final Economic Analysis for the Stage 2 DBPR
6-29
December 2005
-------
More research on DBFs is underway at EPA and other research institutions. Summaries of on-
going studies may be found on EPA's DRINK website (http://www.epa.gov/safewater/drink/intro.html).
Two-year bioassays by the National Toxicology Program (NTP) released in abstract form have recently
been completed on BDCM and chlorate. The draft abstract on BDCM reported no evidence of
carcinogenicity when BDCM was administered via drinking water (NTP 2005a). The results of this draft
report do not affect the Stage 2 benefits analysis, because the quantified benefits are based on data from
epidemiological studies as presented in Section 6.4 and Appendix E. Another recent study, a modified
two-year bioassay on BDCM in the drinking water, reported little evidence of carcinogenicity (George et
al. 2002). In a previous NTP study, tumors were observed, including an increased incidence of kidney,
liver, and colon tumors, when BDCM was administered at higher doses by gavage in corn oil (NTP
1987). EPA will examine new information on BDCM as it becomes available. In the chlorate draft
abstract, NTP found some evidence that it may be a carcinogen (NTP 2004). Chlorate is a byproduct of
hypochlorite and chlorine dioxide systems. A long-term, two-year bioassay NTP study on DBA is also
complete but has not yet undergone peer review (NTP 2005b).
Another significant advancement beyond the Stage 1 DBPR was the evaluation of the chloroform
tumorigenicity data on the basis of its nonlinear mode of action following the draft 1999 proposed
Guidelines for Carcinogen Risk Assessment (USEPA 1999b). The new chloroform assessment became
available on IRIS in October 2001.
An International Life Sciences Institute (ILSI) Expert Panel recommended that DBP risk cannot
be assessed by single-chemical testing approaches alone (ILSI and RSI 1998). The report suggested the
use of modern approaches (e.g., studies relating chemical structure to toxicity, application of molecular
biology techniques, studies of mechanism of action), the use of a 3-tiered testing approach (i.e., in-vitro
tests; short-term screening tests or 90-day animal studies; long-term chronic bioassays). It also
recommended a focus on three scenarios: (1) defined (simple) mixtures of less than 10 DBFs; (2) whole
mixtures produced by simulating disinfection scenarios; and (3) real drinking water samples or their
extracts.
Other byproducts with carcinogenic potential
Along with the reduction in DBFs from chlorination such as TTHM and HAAS as a result of the
Stage 2 DBPR, there may be increases in other DBFs as systems switch from chlorine to alternative
disinfectants. For all disinfectants, many DBFs are not regulated and many others have not yet been
identified. EPA will continue to review new studies on DBFs and their occurrence levels to determine if
they pose possible health risks. EPA continues to support regulation of TTHM and HAAS as indicators
for chlorination DBP occurrence and believes that operational and treatment technology changes made
because of the Stage 2 DBPR will result in an overall decrease in risk.
Emerging DBFs
lodo-DBPs and nitrogenous DBFs including halonitromethanes are DBFs that have recently been
reported (Richardson et al. 2002, Richardson 2003). One recent occurrence study sampled quarterly at
twelve plants using different disinfectants across the U.S. for several iodo-THMs and halonitromethane
species (Weinberg et al. 2002). The concentrations of iodo-THMs and halonitromethane in the majority
of samples in this study were less than the analytical minimum reporting levels; plant-average
concentrations of iodo-THM and halonitromethane species were typically less than 0.002 mg/L, which is
an order of magnitude lower than the corresponding average concentrations of TTHM and HAAS at those
same plants. Chloropicrin, a halonitromethane species, was also measured in the ICR with a median
concentration of 0.0002 mg/L across all surface water samples. No occurrence data exist for the
iodoacids due to the lack of a quantitative method and standards. Further work on chemical formation of
iodo-DBPs and halonitromethanes is needed.
Final Economic Analysis for the Stage 2 DBPR 6-30 December 2005
-------
lodoacetic acid was found to be cytotoxic and genotoxic in Salmonella and mammalian cells
(Plewa et al. 2004a) as were some of the halonitromethanes (Kundu et al. 2004; Plewa et al. 2004b).
Although potent in these in vitro screening studies, further research is needed to determine if these DBFs
are active in living systems. No conclusions on human health risk can be drawn from such preliminary
studies.
N-nitrosamines
Another group of nitrogenous DBFs are the N-nitrosamines. A number of N-nitrosamines exist,
and N-nitrosodimethylamine (NDMA), a probable human carcinogen (IRIS 1993), has been identified as
a potential health risk in drinking water. NDMA is a contaminant from industrial sources and a potential
disinfection byproduct from reactions of chlorine or chloramine with nitrogen containing organic matter
and from some polymers used as coagulant aids. Studies have produced new information on the
mechanism of formation of NDMA, but there is not enough information at this time to draw conclusions
regarding a potential increase in NDMA occurrence as systems change treatment. Although there are
studies that examined the occurrence of NDMA in some water systems, there are no systematic
evaluations of the occurrence of NDMA and other nitrosamines in US waters. Recent studies have
provided new occurrence information that shows NDMA forms in both chlorinated and chloraminated
systems. Barrett et al. (2003) reported median concentrations of less than 2 ng/L for the seven chlorine
systems studied and less than 3 ng/L for 13 chloramine systems. Another study demonstrated that factors
other than disinfectant type may play an important role in the formation of NDMA (Schreiber and Mitch
2005). More research is underway to determine the extent of NDMA occurrence in drinking water
systems. EPA has proposed monitoring for NDMA under Unregulated Contaminant Monitoring Rule 2
(USEPA 2005q).
Risk assessments have estimated that the 10"6 lifetime cancer risk level is 7 ng/L based on
induction of tumors at multiple sites. NDMA is also present in food, tobacco smoke, and industrial
emissions, and additional research is underway to determine the relative exposure of NDMA in drinking
water to these other sources.
Other DBFs
Some systems, depending on bromide and organic precursor levels in the source water and
treatment technology selection, may experience a shift to higher ratios, or concentrations, of brominated
DBFs while the overall TTHM or HAAS concentration may decrease. In some instances where
alternative disinfectants are used, levels of chlorite and bromate may increase as a result of systems
switching to chlorine dioxide or ozone, respectively. However, EPA anticipates that changes in chlorite
and bromate concentration as a result of the Stage 2 DBPR will be minimal (see Section 6.4). For most
systems, overall levels of DBFs, as well as brominated DBF species, should decrease as a result of this
rule. EPA continues to believe that precursor removal is a highly effective strategy to reduce levels of
DBFs.
Other toxicological effects
The Agency has modified the reference dose (RfD) values of the chlorinated acetic acids since the
Stage 1 DBPR. Under the Stage 1 DBPR there was no established RfD for monochloroacetic acid
(MCAA). Data from a drinking water exposure study of MCAA in rats by DeAngelo et al. (1997) were
used to establish an RfD of 0.01 mg/kg-day based on observed increases in spleen weights. Data from
DeAngelo et al. (1997) were also used to calculate anew RfD of 0.03 mg/kg-day for trichloroacetic acid
based on observed effects on body weight and liver effects.
Final Economic Analysis for the Stage 2 DBPR 6-31 December 2005
-------
WHO review of toxicology literature (2000)
The IPCS report on Disinfectants and Disinfection Byproducts (WHO 2000) emphasizes that the
bulk of the toxicology data focus primarily on carcinogenesis. The Task Group found BDCM to be of
particular interest because it produces tumors in both rats and mice at several sites. Although the HAAs
appear to be without significant genotoxic activity, the brominated HAAs appear to induce oxidative
damage to deoxyribose nucleic acid (DNA), leading to tumor formation.
6.2.1.3 Issues with Human and Animal Cancer Data Concordance
According to the Guidelines for Carcinogen Risk Assessment (USEPA 2005f), tumor site
concordance between human and test animal is not necessary to determine carcinogenic potential;
mechanistic considerations should only be applied when there is sufficient data to support a mode of
action. The guidelines state that "Target organ concordance is not a prerequisite for evaluating the
implications of animal study results for humans." Although concordance of effects between the test
species and humans is highly desirable and lends credence and support in the analysis of mode of action,
it is not necessary to cite animal and human tumor site concordance in order to justify a quantitative
cancer risk assessment. In addition, there is insufficient data to support potential mode(s) of action for
DBFs, with the exception of chloroform. Therefore, consideration of site concordance across species for
cancer is not appropriate at this time.2
It is important to consider some key similarities and differences across species when conducting
such a cross-species concordance evaluation for bladder cancer. The mechanisms controlling cell growth
and differentiation are similar across species, but there are marked differences in the way these
mechanisms are managed in various tissues within a given species. In addition, it is important to consider
the differences in exposure routes for the animal studies (ingestion and gavage) and the epidemiological
studies, which may include inhalation and dermal exposure.
Disinfection byproducts may be associated with bladder cancer in humans, but may appear as
kidney or liver cancers in laboratory animal toxicology studies in rodents; laboratory animal toxicology
studies of individual DBFs have reported cancer of the liver and kidney. Although this appears to
indicate a lack of concordance in target organs, concordance on the general tissue type may be present in
different organs. Kidney cancer in the renal pelvis, which has been reported for a few DBFs, may be
linked with bladder cancer, as the renal pelvis is lined with the same transitional cell epithelium as found
in the bladder (Cohen et al. 1988). Further, the pathogenesis of transitional cell carcinomas in the urinary
bladder appears to be similar throughout the renal pelvis, ureter, and urinary bladder (Cohen et al. 1988).
For example, the DBF MX has been reported to cause individual transitional cell hypertrophy with
karyomegaly in the urinary bladder of rats (Komulainen et al. 1997). A review of occupational cancer of
the urinary tract reported that approximately 15 percent of kidney neoplasms are in the renal pelvis and
appear to be caused by the same carcinogens as bladder neoplasms (Schulte et al. 1987). Finally, over 90
percent of human bladder cancers involve the transitional cell epithelium or urothelium (Silverman et al.
1996), suggesting that a component of the kidney and bladder can be considered part of the same target
tissue, known as the urinary tract.
DBFs may not be unique with respect to an apparent lack of tissue-specific concordance across
species. A majority of the non-DBF agents that clearly act through genotoxic mechanisms, including
benzidine and benzidine-derived azo dyes (Cohen and Johansson 1992), are known to cause urinary
bladder carcinoma in humans and to cause cancer in rodents at various sites that do not always include the
urinary bladder (Rice et al. 1999). Other examples of this include inorganic arsenicals, which are human
2Site concordance has been found for colon cancer and for reproductive and toxicological endpoints.
Final Economic Analysis for the Stage 2 DBPR 6-32 December 2005
-------
bladder carcinogens but are negative in animals studies (NRC 1990, USEPA 1994a). Conversely,
nitrosamines (i.e., N-methyl-N-nitrosourea) are bladder carcinogens in laboratory animals, but not in
humans (Cohen and Johansson 1992). Similar to disinfection byproducts, caffeine is a risk factor for
bladder cancer in humans, but there is no evidence of increased risk reported in laboratory animals
(Cohen and Johansson 1992). Although 40 percent of the NTP chemicals that cause bladder tumors are
not mutagenic, EPA concludes that a majority of the agents that clearly act through genotoxic
mechanisms are known to cause urinary bladder carcinoma in humans and rodents at various sites that do
not always include the urinary bladder.
Another potential influence on the difference in observations in animals and humans is the
difference in exposure route in animal studies versus human studies. The animal studies all use ingestion
or gavage as the route of administration, whereas human drinking water exposure includes inhalation and
dermal exposure. For example, tumor responses from chloroform exposure in the liver and kidney of rats
varied by route of exposure, sex and strain (ILSI 1997).
EPA has completed a comprehensive review of the cancer data on disinfection byproducts; and
while there is evidence from human cancer epidemiology studies that lifetime consumption of the DBP
mixture within chlorinated surface water poses a bladder cancer risk, the specific causative constituents
have not been identified. Since there is no definitive support provided by studies to date for or against the
generalization of adverse effects across different organs or between or within species, EPA concludes that
target organ concordance for cancer is not a prerequisite for evaluating the implications of animal study
results for humans at this time. EPA will reevaluate this issue as new data become available to support a
mode-of-action.
6.2.1.4 Conclusions
EPA concludes that the epidemiological and toxicological studies support a weight-of-evidence
conclusion that there may be an association between DBFs and cancer. The evidence is insufficient to
establish a causal relationship. The following are the key factors used to support EPA's weight-of-
evidence conclusion:
• There is some evidence from animal studies for the carcinogenicity of individual DBFs
included in this rule. Exhibit 6.11 summarizes the Agency findings on the
carcinogenicity of seven DBFs. They have all been characterized on IMS as either
"possible" or "probable" carcinogens under EPA's 1986 guidelines, and in some cases
also as "likely" carcinogens under EPA's 1996/1999 guidelines. One of these
(chloroform) has been evaluated based on its mode of action, with the finding that it is
likely to be carcinogenic only under high-exposure conditions that lead to cytotoxicity
and regenerative hyperplasia.
Epidemiological data from individual investigations and from a meta-analysis
(Villanueva et al. 2003) links exposure to chlorinated water with an increased risk for
bladder cancer in some population subgroups.
• An analysis of pooled data by Villanueva et al. (2004) links exposure to TTHMs with an
increased risk of bladder cancer.
• The epidemiological data cannot link specific DBFs with cancer risk because of
difficulties in characterizing the exposure. Exposure in some epidemiology studies was
monitored purely in terms of chlorinated water, which contains a mixture of DBFs, some
of which have not yet been identified, as well as a variety of other drinking water
contaminants. In other studies the DBP exposure was monitored in terms of
Final Economic Analysis for the Stage 2 DBPR 6-33 December 2005
-------
trihalomethane concentrations (a variable mixture of four individual DBFs). Thus, the
Agency must rely on both the bioassays for individual chemicals as well as the
epidemiology data in making a weight-of-evidence determination.
Associations of chlorinated water to cancer of the colon, rectum, and kidney were found
in some cases but the data are less robust than the data for bladder cancer.
EPA has a research program that continues to examine the relationship between exposure to
DBFs and carcinogenicity. Additional data needs include information on modes of action, the reasons for
inconsistencies in findings between men and women, inconsistencies across studies in the role of
smoking, and carcinogenicity testing for selected brominated and chlorinated DBFs administered in
drinking water. New studies are under way or planned that would help provide this type of data.
6.2.2 Reproductive and Developmental Health Effects
Both human epidemiology studies and animal toxicology studies have examined associations
between chlorinated drinking water or DBFs and reproductive and developmental health effects. Based
on the weight-of-evidence evaluation of the reproductive and developmental epidemiology data EPA
concludes that a causal link between adverse reproductive or developmental health effects and exposure
to chlorinated drinking water or DBFs has not been established, but that there is a potential association.
Despite inconsistent findings across studies, some recent studies continue to suggest associations between
DBF exposure and various adverse reproductive and developmental effects. In addition, data from a
number of toxicology studies, although the majority of them were conducted using high doses,
demonstrate biological plausibility for some of the effects observed in epidemiology studies. EPA
concludes that no dose-response relationship or causal link has been established between exposure to
chlorinated drinking water or disinfection byproducts and adverse reproductive or developmental health
effects. EPA's evaluation of the best available studies, particularly epidemiology studies, is that they do
not support a conclusion at this time as to whether exposure to chlorinated drinking water or disinfection
byproducts cause adverse reproductive and developmental health effects, but do provide an indication of a
potential health hazard concern that warrants incremental regulatory action beyond the Stage 1 DBPR.
The Centers for Disease Control (CDC) reports that, for the 10 year period between 1986 and
1996, spontaneous fetal losses were estimated to be between 0.8 million and 1.0 million per year. For
births in the United States in the year 2002 (as reported by the CDC, 2005), 1.4 percent of births are
considered very low birth weight (defined by CDC as below 1,500 g) and 7.8 percent are considered low
birth weight (defined by the CDC as below 2,500 g). Birth defects are reported to occur in approximately
1 in 33 live births per year in the United States (CDC 2005). Although research has identified some risk
factors for these adverse birth outcomes, including nutritional factors (e.g., lack of folic acid
supplementation) and fetal exposure to tobacco smoke and alcohol, the causes of most such outcomes are
unknown.
A variety of research is underway to examine the potential role that maternal exposure to specific
contaminants might play in these adverse outcomes. The following sections provide a review of the
literature addressing the potential relationship between DBFs and adverse reproductive and
developmental outcomes.
Final Economic Analysis for the Stage 2 DBPR 6-34 December 2005
-------
6.2.2.1 Epidemiological Evidence of Adverse Reproductive and Developmental Health Effects
As discussed previously, epidemiology studies have the strength of relating human exposure to
DBF mixtures through multiple intake routes. Although the critical exposure window for reproductive
and developmental effects is much smaller than that for cancer (generally weeks versus years), exposure
assessment is also a main limitation of reproductive and developmental epidemiology studies. Exposure
assessment uncertainties and possible exposure classification errors arise from limited data on DBF
concentrations and maternal water usage and source over the course of the pregnancy. However,
classification errors typically push the true risk estimate towards the null value (Vineis 2004). According
to Bove et al. (2002), "Difficulties in assessing exposure may result in exposure misclassification biases
that would most likely produce substantial underestimates of risk as well as distorted or attenuated
exposure-response trends." Studies of rare outcomes (e.g., individual birth defects) often have limited
statistical power because of the small number of cases being examined. This limits the ability to detect
statistically significant associations for small to moderate relative risk estimates. Small sample sizes also
result in imprecision around risk estimates reflected by wide confidence intervals. In addition to the
limitations of individual studies, evaluating reproductive and developmental epidemiology studies
collectively is difficult because of the methodological differences between studies and the wide variety of
endpoints examined. These factors may contribute to inconsistencies in the scientific body of literature as
noted below.
More recent studies tend to be of higher quality because of improved exposure assessments and
other methodological advancements. For example, studies that use THM levels to estimate exposure tend
to be higher quality than studies that define exposure by source or treatment. These factors were taken
into account by EPA when comparing and making conclusions on the reproductive and developmental
epidemiology literature. What follows is a summary of available epidemiology literature on reproductive
and developmental endpoints such as spontaneous abortion, stillbirth, neural tube and other birth defects,
low birth weight, and intrauterine growth retardation. Information is grouped, where appropriate, into
three categories on fetal growth, viability, and malformations, and reviews are described separately
afterward.
Epidemiology reports and reviews
Fetal growth
Many studies looked for an association between fetal growth (mainly small for gestational age,
low birth weight, and pre-term delivery) and chlorinated water or DBFs. The results from the collection
of studies as a whole are inconsistent. A number of studies support the possibility that exposure to
chlorinated water or DBFs are associated with adverse fetal growth effects (Infante-Rivard 2004, Wright
et al. 2004, Wright et al. 2003, Kallen and Robert 2000, Gallagher et al. 1998, Kanitz et al. 1996, Bove et
al. 1995, Kramer et al. 1992). Other studies showed mixed results (Porter et al. 2005, Savitz et al. 2005,
Yang 2004) or did not provide evidence of an association (Toledano et al. 2005, Jaakkola et al. 2001,
Dodds et al. 1999, Savitz et al. 1995) between DBF exposure and fetal growth. EPA notes that recent,
higher quality studies provide some evidence of an increased risk of small for gestational age and low
birth weight.
Fetal viability
While the database of epidemiology studies for fetal loss endpoints (spontaneous abortion or
stillbirth) remains inconsistent as a whole, there is suggestive evidence of an association between fetal
loss and chlorinated water or DBP exposure. Various studies support the possibility that exposure to
chlorinated water or DBFs is associated with decreased fetal viability (Toledano 2005, Dodds et al. 2004,
King et al. 2000b, Dodds et al. 1999, Waller et al. 1998, Aschengrau et al. 1993, Aschengrau et al. 1989).
Other studies did not support an association (Bove et al. 1995) or reported inconclusive results (Savitz et
Final Economic Analysis for the Stage 2 DBPR 6-35 December 2005
-------
al. 2005, Swan et al. 1998, Savitz 1995) between fetal viability and exposure to THMs or tapwater. A
recent study by King et al. (2005) found little evidence of an association between stillbirths and haloacetic
acids after controlling for trihalomethane exposures, though non-statistically significant increases in
stillbirths were seen across various exposure levels.
Fetal malformations
A number of epidemiology studies have examined the relationship between fetal malformations
(such as neural tube, oral cleft, cardiac, or urinary defects, and chromosomal abnormalities) and
chlorinated water or DBFs. It is difficult to assess fetal malformations in aggregate due to inconsistent
findings and disparate endpoints being examined in the available studies. Some studies support the
possibility that exposure to chlorinated water or DBFs is associated with various fetal malformations
(Cedergren et al. 2002, Hwang et al. 2002, Dodds and King 2001, Klotz and Pyrch 1999, Bove et al.
1995, Aschengrau et al. 1993). Other studies found little evidence (Shaw et al. 2003, Kallen and Robert
2000, Dodds et al. 1999, Shaw et al. 1991) or inconclusive results (Magnus et al. 1999) between
chlorinated water or DBF exposure and fetal malformations. Birth defects most consistently identified as
being associated with DBFs include neural tube defects and urinary tract malformations.
Other endpoints have also been examined in recent epidemiology studies. One study suggests an
association between DBFs and decreased menstrual cycle length (Windham et al. 2003), which, if
corroborated, could be linked to the biological basis of other reproductive endpoints observed. No
association between THM exposure and semen quality was found (Fenster et al. 2003). More work is
needed in both areas to support these results.
Reviews
An early review supported an association between measures of fetal viability and tap water (Swan
et al. 1992). Three other reviews found data inadequate to support an association between reproductive
and developmental health effects and THM exposure (Reif et al. 1996, Craun 1998, WHO 2000). Mills et
al. (1998) examined data on and found support for an association between fetal viability and
malformations and THMs. Another review presented to the Stage 2 MDBP FACA found some evidence
for an association with fetal viability and some fetal malformations and exposure to DBFs but reported
that the evidence was inconsistent for these endpoints as well as for fetal growth (Reif et al. 2000). Reif
et al. (2000) concluded that the weight of evidence from epidemiology studies suggests that "DBFs are
likely to be reproductive toxicants in humans under appropriate exposure conditions," but from a risk
assessment perspective, data are primarily at the hazard identification stage. Nieuwenhuijsen et al. (2000)
found some evidence for an association between fetal growth and THM exposure and concluded evidence
for associations with other fetal endpoints is weak but gaining weight. A qualitative review by
Villanueva et al. (2001) found evidence generally supports a possible association between reproductive
effects and drinking chlorinated water. Graves et al. (2001) supports a possible association for fetal
growth but not fetal viability or malformations. More recently, Bove et al. (2002) examined and
supported an association between small for gestational age, neural tube defects and spontaneous abortion
endpoints and DBFs. Following a meta-analysis on five malformation studies, Hwang and Jaakkola
(2003) concluded that there was evidence which supported associations between DBFs and risk of birth
defects, especially neural tube defects and urinary tract defects. More detail on some of these critical
reviews is presented later in this section. Exhibit 6.13 provides summary information for key
epidemiological reports and reviews to the Stage 2 DBPR.
Final Economic Analysis for the Stage 2 DBPR 6-36 December 2005
-------
Exhibit 6.13 Summary of Reproductive/Developmental Epidemiology Studies
Author(s)
Study Type
Exposure(s) Studied
Outcome(s)
Measured
Findings
Porter et al.
2005
Cross-
sectional study
in Maryland,
1998-2002.
Estimated trimester-
specific and
pregnancy-average
exposures to THMs
and HAAs, including
individual DBFs.
Intrauterine
growth
retardation.
No consistent association or dose-response relationship was
found between exposure to either TTHM or HAAS and
intrauterine growth retardation. Results suggest an increased
risk of intrauterine growth retardation associated with TTHM
and HAAS exposure in the third trimester, although only HAAS
results were statistically significant.
Savitz et al.
2005
Population-
based
prospective
cohort study in
three
communities
around the
U.S., 2000-
2004.
Estimated TTHM,
HAA9, and TOX
exposures during
pregnancy. Individual
brominated THMs and
HAA species were
examined. Indices
examined included
concentration,
ingested amount,
exposure from
showering and
bathing, and an
integration of all
exposures combined.
Early and late
pregnancy loss,
preterm birth,
small for
gestational age,
and term birth
weight.
No association with pregnancy loss was seen when high TTHM
exposures were compared to low exposures. When examining
individual THMs, a statistically significant association was found
between bromodichloromethane (BDCM) and pregnancy loss.
Although non-statistically significant, an increased risk similar in
magnitude was seen between dibromochloromethane (DBCM)
and pregnancy loss. Some increased risks were seen for
losses at greater than 12 weeks' gestation for TTHM, BDCM,
and TOX (total organic halide), but most results generally did
not provide support for an association. Preterm birth showed a
small inverse relationship with DBP exposure (i.e. higher
exposures were less likely to have a preterm birth), but this
association was weak. TTHM exposure of 80 ug/L was
significantly associated with twice the risk for small for
gestational age during the third trimester.
Toledano et al.
2005
Large cross-
sectional study
in England,
1992-1998.
Linked mother's
residence at time of
delivery to modeled
estimates of TTHM
levels in water zones.
Stillbirth, low
birth weight.
A significant association between TTHM and risk of stillbirth,
low birth weight, and very low birth weight was observed in one
of the three regions. When all three regions were combined,
small, but non-significant, excess risks were found between all
three outcomes and TTHM and chloroform. No associations
were observed between reproductive risks and BDCM or total
brominated THMs.
Final Economic Analysis for the Stage 2 DBPR
6-37
December 2005
-------
Author(s)
Study Type
Exposure(s) Studied
Outcome(s)
Measured
Findings
Dodds et al.
2004 (and King
et al. 2005)
Population-
based case-
control study
in Nova Scotia
and Eastern
Ontario, 1999-
2001.
Estimated THM and
HAA exposure at
residence during
pregnancy. Linked
water consumption
and showering/bathing
to THM exposure.
Stillbirth.
A statistically significant association was observed between
stillbirths and exposure to total THM, BDCM, and chloroform.
Associations were also detected for metrics which incorporated
water consumption, showering and bathing habits. Elevated
relative risks were observed for intermediate exposures for total
HAA and DCAA measures; TCAA and brominated HAA
exposures showed no association. No statistically significant
associations or dose-response relationships between any HAAs
and stillbirth were detected after controlling for THM exposure.
Infante-Rivard
2004
Case-control
study of
newborns in
Montreal,
1998-2000.
Estimated THM levels
and water
consumption during
pregnancy. Exposure
from showering and
presence of two
genetic
polymorphisms.
Intra uterine
growth
retardation.
No associations were found between exposure to THMs and
intrauterine growth retardation. However, a significant effect
was observed between THM exposure and intrauterine growth
retardation for newborns with the CYP2E1 gene variant.
Findings suggest that exposure to THMs at the highest levels
can affect fetal growth but only in genetically susceptible
newborns.
Wrightefa/.
2004
Large cross-
sectional
study:
Massachusett
s, 1995-1998.
Estimated maternal
third-trimester
exposures to TTHMs,
chloroform, BDCM,
total HAAs, DCA, TCA,
MX and mutagenicity
in drinking water.
Birth weight,
small for
gestational age,
preterm
delivery,
gestational age.
Statistically significant reductions in mean birth weight were
observed for BDCM, chloroform, and mutagenic activity. An
exposure-response relationship was found between THM
exposure and reductions in mean birth weight and risk of small
for gestational age. There was no association between preterm
delivery and elevated levels of HAAs, MX, or mutagenicity. A
reduced risk of preterm delivery was observed with high THM
exposures. Gestational age was associated with exposure to
THMs and mutagenicity.
Yang 2004 (and
Yang et al.
2000)
Large cross-
sectional
studies in
Taiwan, 1994-
1996.
Compared maternal
consumption of
chlorinated drinking
water (yes/no).
Low birth
weight, preterm
delivery.
Residence in area supplied with chlorinated drinking water
showed a statistically significant association with preterm
delivery. No association was seen between chlorinated
drinking water and low birth weight.
Fenster et al.
2003
Small
prospective
study in
California,
1990-1991.
Examined TTHM
levels within the 90
days preceding semen
collection.
Sperm motility,
sperm
morphology.
No association between TTHM level and sperm mobility or
morphology. BDCM was inversely associated with linearity of
sperm motion. There was some suggestion that water
consumption and other ingestion metrics may be associated
with different indicators of semen quality.
Final Economic Analysis for the Stage 2 DBPR
6-38
December 2005
-------
Author(s)
Shaw et al. 2003
Windham et al.
2003
Wrightefa/.
2003
Cedergren et al.
2002
Hwang et al.
2002
Study Type
2 case-control
maternal
interview
studies: CA,
1987-1991.
Prospective
study: CA,
1990-1991.
Cross-
sectional
study:
Massachusett
s, 1990.
Retrospective
case-control
study:
Sweden,
1982-1997.
Large cross-
sectional study
in Norway,
1993-1998.
Exposure(s) Studied
Estimated THM levels
for mothers'
residences from before
conception through
early pregnancy.
Estimated exposure to
THMs through
showering and
ingestion over average
of 5.6 menstrual cycles
per woman.
Estimated TTHM
exposure in women
during pregnancy
(average for
pregnancy and during
each trimester).
Examined maternal
periconceptional DBP
levels and used CIS to
assign water supplies.
Compared exposure to
chlorination (yes/no)
and water color levels
for mother's residence
during pregnancy.
Outcome(s)
Measured
Neural tube
defects, oral
clefts, selected
heart defects.
Menstrual
cycle, follicular
phase length
(in days).
Birth weight,
small for
gestational age,
preterm
delivery,
gestational age.
Cardiac
defects.
Birth defects
(neural tube
defects,
cardiac,
respiratory
system, oral
cleft, urinary
tract).
Findings
No associations or exposure-response relation were observed
between malformations and TTHMs in either study.
Findings suggest that THM exposure may affect ovarian
function. All brominated THM compounds were associated with
significantly shorter menstrual cycles with the strongest finding
for chlorodibromomethane. There was little association
between TTHM exposure and luteal phase length, menses
length, or cycle variability.
Statistically significant associations between 2nd trimester and
pregnancy average TTHM exposure and small for gestational
age and fetal birth weight were detected. Small, statistically
significant increases in gestational duration/age were observed
at increased TTHM levels, but there was little evidence of an
association between TTHM and preterm delivery or low birth
weight.
Exposure to chlorine dioxide in drinking water showed statistical
significance for cardiac defects. THM concentrations of 1 0 • g/L
and higher were significantly associated with cardiac defects.
No excess risk for cardiac defect and nitrate were seen.
Risk of any birth defect, cardiac, respiratory system, and urinary
tract defects were significantly associated with water
chlorination. Exposure to chlorinated drinking water was
statistically significantly associated with risk of ventricular septal
defects, and an exposure-response pattern was seen. No other
specific defects were associated with the exposures that were
examined.
Final Economic Analysis for the Stage 2 DBPR
6-39
December 2005
-------
Author(s)
Dodds and King
2001
Jaakkola et al.
2001
Ka'llen and
Robert 2000
Dodds et al.
1999 (and King
et al. 2000b)
Klotz and Pyrch
1999 (and Klotz
and Pyrch 1 998)
Study Type
Population-
based
retrospective
cohort in Nova
Scotia, 1988-
1995.
Large cross-
sectional study
in Norway,
1993-1995.
Large cross-
sectional
cohort study in
Sweden,
1985-
1994.
Population-
based
retrospective
cohort study in
Nova Scotia,
1988-1995.
Population-
based case-
control study
in New Jersey,
1993-1994.
Exposure(s) Studied
Estimated THM,
chloroform, and
bromodichloromethane
(BDCM) exposure.
Compared chlorination
(yes/no) and water
color (high/low) for
mother during
pregnancy.
Linked prenatal
exposure to drinking
water disinfected with
various methods (no
chlorine, chlorine
dioxide only, sodium
hypochlorite only).
Estimated TTHM level
for women during
pregnancy.
Estimated exposure of
pregnant mothers to
TTHMs and HAAs,
and compared source
of water.
Outcome(s)
Measured
Neural tube
defects,
cardiovascular
defects, cleft
defects,
chromosomal
abnormalities.
Low birth
weight, small
for gestational
age, preterm
delivery.
Gestational
duration, birth
weight,
intra uterine
growth,
mortality,
congenital
malformations,
and other birth
outcomes.
Low birth
weight, preterm
birth, small for
gestational age,
stillbirth,
chromosomal
abnormalities,
neural tube
defects, cleft
defects, major
cardiac defects.
Neural tube
defects.
Findings
Exposure to BDCM was associated with increased risk of
neural tube defects, cardiovascular anomalies. Chloroform was
not associated with neural tube defects, but was associated
with chromosomal abnormalities. No association between THM
and cleft defects were detected.
No evidence found for association between prenatal exposure
to chlorinated drinking water and low birth weight or small for
gestational age. A reduced risk of preterm delivery was noted
for exposure to chlorinated water with high color content.
A statistically significant difference was found for short
gestational duration and low birth weight among infants whose
mother resided in areas using sodium hypochlorite, but not for
chlorine dioxide. Sodium hypochlorite was also associated with
other indices of fetal development but not with congenital
defects. No other effects were observed for intrauterine growth,
childhood cancer, infant mortality, low Apgar score, neonatal
jaundice, or neonatal hypothyroidism in relation to either
disinfection method.
A statistically significant increased risk for stillbirths and high
total THMs and specific THMs during pregnancy was detected,
with higher risks observed among asphyxia-related stillbirths.
Bromodichloromethane had the strongest association and
exhibited an exposure-response pattern. There was limited
evidence of an association between THM level and other
reproductive outcomes. No congenital anomalies were
associated with THM exposure, except for a non-statistically
significant association with chromosomal abnormalities.
A significant association was seen between exposure to THMs
and neural tube defects. No associations were observed for
neural tube defects and haloacetic acids or haloacetonitriles.
Final Economic Analysis for the Stage 2 DBPR
6-40
December 2005
-------
Author(s)
Magnus etal.
1999
Gallagher et al.
1998
Swan et al. 1 998
Waller et al.
1998 (and
Waller et al.
2001)
Kanitz et al.
1996
Study Type
Large cross-
sectional study
in Norway,
1993-1995.
Retrospective
cohort study of
newborns in
Colorado,
1990-1993.
Prospective
study in
alifornia, 1990-
1991.
Prospective
cohort in
California,
1989-1991.
Cross-
sectional study
in Italy, 1988-
1989.
Exposure(s) Studied
Compared chlorination
(yes/no) and water
color (high/low) at
mothers' residences at
time of birth.
Estimated THM levels
in drinking water
during third trimester
of pregnancy.
Compared
consumption of cold
tap water to bottled
water during early
pregnancy.
Estimated TTHM
levels during first
trimester of pregnancy
via ingestion and
showering.
Compared 3 types of
water treatment
(chlorine dioxide,
sodium hypochlorite,
and chlorine
dioxide/sodium
hypochlorite).
Outcome(s)
Measured
Birth defects
(neural tube
defects, major
cardiac,
respiratory,
urinary, oral
cleft).
Low birth
weight, term
low birthweight,
and preterm
delivery.
Spontaneous
abortion.
Spontaneous
abortion.
Low birth
weight, body
length, cranial
circumference,
preterm
delivery, and
other effects.
Findings
Statistically significant associations were seen between urinary
tract defects and chlorination and high water color (high content
of organic compounds). No associations were detected for
other outcomes or all birth defects combined. A non-statistically
significant, overall excess risk of birth defects was seen within
municipalities with chlorination and high water color compared
to municipalities with no chlorination and low color.
Weak, non-statistically significant association with low birth
weight and TTHM exposure during the third trimester. Large
statistically significant increase for term low birthweight at
highest THM exposure levels. No association between preterm
delivery and THM exposure.
Pregnant women who drank cold tap water compared to those
who consumed no cold tap water showed a significant finding
for spontaneous abortion at one of three sites.
Statistically significant increased risk between high intake of
TTHMs and spontaneous abortion compared to low intake.
BDCM statistically associated with increased spontaneous
abortion; other THMs not. Reanalysis of exposure yielded less
exposure misclassification and relative risks similar in
magnitude to earlier study. An exposure-response relationship
was seen between spontaneous abortion and ingestion
exposure to TTHMs.
Smaller body length and small cranial circumference showed
statistical significant association with maternal exposure to
chlorinated drinking water. Neonatal jaundice linked statistically
to prenatal exposure to drinking water treated with chlorine
dioxide. Length of pregnancy, type of delivery, and birthweight
showed no association.
Final Economic Analysis for the Stage 2 DBPR
6-41
December 2005
-------
Author(s)
Bove et al. 1 995
(and Bove et al.
1992a&1992b)
Savitz et al.
1995
Aschengrau et
al. 1993
Kramer et al.
1992
Study Type
Large cohort
cross-
sectional study
in New Jersey,
1985-1988.
Population-
based case-
control study:
North
Carolina,
1988-1991.
Case-control
study in
Massachusett
s, 1977-1980.
Population-
based case-
control study
in Iowa, 1989-
1990.
Exposure(s) Studied
Examined maternal
exposure to TTHM and
various other
contaminants.
Examined TTHM
concentration at
residences and water
consumption (during
first and third
trimesters).
Source of water and 2
types of water
treatment (chlorination,
chloramination).
Examined chloroform,
DCBM, DBCM, and
bromoform levels and
compared type of
water source (surface,
shallow well, deep
well).
Outcome(s)
Measured
Low birth
weight, fetal
deaths, small
for gestational
age, birth
defects (neural
tube defects,
oral cleft,
central nervous
system, major
cardiac).
Spontaneous
abortion,
preterm
delivery, low
birth weight.
Neonatal death,
stillbirth,
congenital
anomalies.
Low birth
weight,
prematurity,
intra uterine
growth
retardation.
Findings
Weak, statistically significant increased risk found for higher
TTHM levels with small for gestational age, neural tube defects,
central nervous system defects, oral cleft defects, and major
cardiac defects. Some association with higher TTHM exposure
and low birth weight. No effect seen for preterm birth, very low
birth weight, or fetal deaths.
There was a statistically significant increased miscarriage risk
with high THM concentration, but THM intake (based on
concentration times consumption level) was not related to
pregnancy outcome. No associations were seen for preterm
delivery or low birth weight. Water source was not related to
pregnancy outcome either, with the exception of a non-
significant, increased risk of spontaneous abortion for bottled
water users. There was a non-statistically significant pattern of
reduced risk with increased consumption of water for all three
outcomes.
There was a non-significant, increased association between
frequency of stillbirths and maternal exposure to chlorinated
versus chloraminated surface water. An increased risk of
urinary track and respiratory track defects and chlorinated water
was detected. Neonatal death and other major malformations
showed no association. No increased risk seen for any adverse
pregnancy outcomes for surface water versus ground and
mixed water use.
Statistically significant increased risk for intrauterine growth
retardation effects from chloroform exposure were observed.
Non-significant increased risks were observed for low birth
weight and chloroform and for intrauterine growth retardation
and DCBM. No intrauterine growth retardation or low birth
weight effects were seen for the other THMs, and no effects on
prematurity were observed for any of the THMs.
Final Economic Analysis for the Stage 2 DBPR
6-42
December 2005
-------
Author(s)
Shawef a/. 1991
(and Shaw et al.
1990)
Aschengrau et
al. 1989
Study Type
Small case-
control study:
Santa Clara
County, CA,
1981-1983.
Case-control
study in
Massachusett
s, 1976-1978.
Exposure(s) Studied
Estimated chlorinated
tap water
consumption, mean
maternal TTHM level,
showering/bathing
exposure at residence
during first trimester.
Source of water and
exposure to metals
and other
contaminants.
Outcome(s)
Measured
Congenital
cardiac
anomalies.
Spontaneous
abortion.
Findings
Following reanalysis, no association between cardiac
anomalies and TTHM level were observed.
A statistically significantly association was detected between
surface water source and frequency of spontaneous abortion.
Reviews/Meta-analyses
Hwang and
Jakkola 2003
Bove et al. 2002
Review and
meta-analysis
of 5 studies.
Qualitative
review of 1 4
studies.
Compared DBP levels,
source of water,
chlorine residual, color
(high/low), and 2 types
of disinfection:
chlorination and
chloramination.
Examined THM levels.
Compared drinking
water source and type
of water treatment.
Birth defects
(respiratory
system, urinary
system, neural
tube defects,
cardiac, oral
cleft).
Birth defects,
small for
gestational age,
low birth
weight, preterm
delivery,
spontaneous
abortion, fetal
death.
The meta-analysis supports an association between exposure
to chlorination by-products and the risk of any birth defect,
particularly the risk of neural tube defects and urinary system
defects.
Review found the studies of THMs and adverse birth outcomes
provide moderate evidence for associations with small for
gestational age, neural tube defects, and spontaneous
abortions. Authors felt risks may have been underestimated
and exposure-response relationships distorted due to exposure
misclassification.
Final Economic Analysis for the Stage 2 DBPR
6-43
December 2005
-------
Author(s)
Graves et al.
2001
Villanueva et al.
2001
Nieuwenhuijsen
et al. 2000
Study Type
Review of
toxicological
and
epidemiologic
al studies
using a weight
of evidence
approach.
Qualitative
review of 1 4
reproductive
and
developmental
health effect
studies.
Qualitative
review of
numerous
toxicological
and
epidemiologic
al studies.
Exposure(s) Studied
Examined water
consumption, duration
of exposure, THM
levels, HAA levels, and
other contaminants.
Compared source of
water, water treatment,
water color (high/low),
etc.
Compared exposure to
TTHM levels,
mutagenic drinking
water, water
consumption, source
water, types of
disinfection
(chlorination and
chloramination), and
residence times.
Examined levels of
various DBFs, water
consumption, and
duration of exposure.
Compared water color,
water treatment,
source of water, etc.
Outcome(s)
Measured
Low birth
weight, preterm
delivery, small
for gestational
age,
intra uterine
growth
retardation,
specific birth
defects,
neonatal death,
decreased
fertility, fetal
resorption, and
other effects.
Spontaneous
abortion, low
birth weight,
small for
gestational age,
neural tube
defects, other
reproductive
and
developmental
outcomes.
Low birth
weight, preterm
delivery,
spontaneous
abortions,
stillbirth, birth
defects, etc.
Findings
Weight of evidence suggested positive association with DBP
exposure for growth retardation such as small for gestational
age or intrauterine growth retardation and urinary tract defects.
Review found no support for an association between DBP
exposure and low birth weight, preterm delivery, some specific
birth defects, and neonatal death. The review resulted in
inconsistent findings for all birth defects, all central nervous
system defects, neural tube defects, spontaneous abortion, and
stillbirth.
Review found positive associations between increased
spontaneous abortion, low birth weight, small for gestational
age, and neural tube defects and drinking chlorinated water in
most studies although not always with statistical significance.
The review supports some evidence of association between
THMs and low birth weight, but inconclusive. Review found no
evidence of association between THMs and preterm delivery,
and that associations for other outcomes (spontaneous
abortions, stillbirth, and birth defects) were weak but gaining
weight.
Final Economic Analysis for the Stage 2 DBPR
6-44
December 2005
-------
Author(s)
Reif et al. 2000
WHO 2000
Craun, ed. 1998
Study Type
Qualitative
reviews of
numerous
epidemiologic
al studies.
Qualitative
reviews of
various
studies in
Finland, U.S.,
and Canada.
Qualitative
review of 1 0
studies, focus
on California
cohort study.
Exposure(s) Studied
Compared source of
water supply and
methods of
disinfection.
Estimated TTHM
levels.
Various exposures to
THMs.
Examined THM levels
and water
consumption, and
compared source of
water and water
treatment (chlorine,
chloramines, chlorine
dioxide).
Outcome(s)
Measured
Birth weight,
low birth
weight,
intra uterine
growth
retardation,
small for
gestational age,
preterm deliver,
somatic
parameters,
neonatal
jaundice,
spontaneous
abortion,
stillbirth,
developmental
anomalies.
Various
reproductive
and
developmental
effects.
Stillbirth,
neonatal death,
spontaneous
abortion, low
birth weight,
preterm
delivery,
intra uterine
growth
retardation,
neonatal
jaundice, birth
defects.
Findings
Weight of evidence suggested DBFs are reproductive toxicants
in humans under appropriate exposure conditions. The review
reports findings between TTHMs and effects on fetal growth,
fetal viability, and congenital anomalies as inconsistent.
Reviewers felt data are at the stage of hazard identification and
did not suggest a dose-response pattern of increasing risk with
increasing TTHM concentration.
Review found some support for an association between
increased risks of neural tube defects and miscarriage and
THM exposure. Other associations have been observed, but
the authors believed insufficient data exists to assess any of
these associations.
Associations between DBFs and various reproductive effects
were seen in some epidemiological studies, but the authors felt
these results do not provide convincing evidence for a causal
relationship between DBFs and reproductive effects.
Final Economic Analysis for the Stage 2 DBPR
6-45
December 2005
-------
Author(s)
Mills et al. 1998
Relief al. 1996
Study Type
Qualitative
review of 22
studies.
Review of 3
case-control
studies and 1
cross-
sectional
study.
Exposure(s) Studied
Examined TTHM
levels and water
consumption.
Compared source of
water and 2 types of
water treatment
(chlorination and
chloramination).
Examined THM levels
at residences, dose
consumption,
chloroform. Compared
source of waters and 2
types of water
treatment (chlorination
and chloramination).
Outcome(s)
Measured
Various
reproductive
and
developmental
effects.
Birth defects
(central
nervous
system, neural
tube defects,
cardiac, oral
cleft,
respiratory,
urinary tract),
spontaneous
abortion, low
birth weight,
growth
retardation,
preterm
delivery,
intra uterine
growth
retardation,
stillbirth,
neonatal death.
Findings
Review found studies suggest possible increases in adverse
reproductive and developmental effects, such as increased
spontaneous abortion rates, small for gestational age, and fetal
anomalies, but that insufficient evidence exists to establish a
causal relationship.
Studies reviewed suggest that exposure to DBFs may increase
intrauterine growth retardation, neural tube defects, major heart
defects, and oral cleft defects. Review found epidemiologic
evidence supporting associations between exposure to DBFs
and adverse pregnancy outcomes to be sparse and to provide
an inadequate basis to identify DBFs as a reproductive or
developmental hazard.
Economic Analysis for the Stage 2 DBPR
6-46
December 2005
-------
Author(s)
Swan et al.
1992
Study Type
Qualitative
review of 5
studies in
Santa Clara
County, CA
(Deane et al.
1992,
Wrensch et al.
1992, Hertz-
Picciotto et al.
1992,
Windham et
al. 1992,
Fenster et al.
1992).
Exposure(s) Studied
Compared maternal
consumption of
residence tap water to
bottled water.
Outcome(s)
Measured
Spontaneous
abortion.
Findings
Four of the studies reviewed suggest that women drinking
bottled water during the first trimester of pregnancy may have
reduced risk of spontaneous abortion relative to drinking tap
water. No association seen in the fifth study. Review
concluded that if findings are causal and not due to chance or
bias, data suggest a 10-50% increase in spontaneous abortion
risk for pregnant women drinking tap water over bottled water.
Economic Analysis for the Stage 2 DBPR
6-47
December 2005
-------
Critical review of epidemiology literature by Reifet al. (2000)
Reif et al. (2000) conducted a critical review of the epidemiology literature pertaining to potential
reproductive and developmental effects of exposure to DBFs in drinking water. Reif presented much of
this data during the FACA process and the critical review of the literature was important in this process.
The review included 16 peer-reviewed scientific manuscripts and published reports of which 10 were
previously discussed in the Stage 1 DBPR. The authors evaluated associations between DBFs and
outcomes grouped as effects on (1) fetal growth (birth weight [as a continuous variable]; low birth weight
[defined as <2,500 grams]; term low birth weight [defined as <2,500 grams]; very low birth weight
[defined as <1,500 grams]; preterm delivery [defined as <37 weeks of gestation] and intrauterine growth
retardation [or decreased rate of growth of the fetus]); (2) fetal viability (spontaneous abortion and
stillbirth); and, (3) risk of fetal malformations (all malformations, oral cleft defects, major cardiac defects,
neural tube defects, and chromosomal abnormalities).
Reifet al. (2000) found mixed evidence in the epidemiological literature they reviewed for
associations between DBFs and effects on fetal growth. Studies using TTHM concentrations reached
differing conclusions. Some studies found weak but statistically significant associations (Gallagher et al.
1998, Bove et al. 1992b, Bove et al. 1995), but two found none (Dodds et al. 1999, Savitz et al. 1995).
Studies with qualitative exposure assessment designs are similarly variable in their findings (Kanitz et al.
1996, Kallen and Robert 2000, Yang et al. 2000).
For effects on fetal viability, the authors reported that some evidence exists for an increased risk
of spontaneous abortion and stillbirth. Increased rates of spontaneous abortion associated with TTHM
levels of 75 (ig/L or more were reported by Waller et al. (1998). Aschengrau et al. (1989) reported a
doubling of risk of spontaneous abortion for exposure to surface water compared to ground and mixed
water. Although Savitz et al. (1995) found an association between high levels of THMs and spontaneous
abortion, no relationship with dose or water source was discovered. As discussed previously, an
increased risk of stillbirth was found to be associated with THM and BDCM exposure (Dodds et al. 1999,
King et al. 2000b). Aschengrau et al. (1993) found an association between stillbirth and the use of
chlorinated versus chloraminated water systems. A weak association was found for the use of surface
water systems and risk of stillbirth, but these authors found little evidence for an association between
TTHM and risk of stillbirth (Bove et al. 1992a, Bove et al. 1995).
For congenital abnormalities related to DBF exposure, Reifet al. (2000) reported that the
relatively few studies available in the technical literature provide an inconsistent pattern both in terms of
associating exposure with the occurrence of anomalies in general, and with respect to identifying specific
anomalies that result from exposure. The authors conceded that an assessment of congenital anomalies is
difficult due to the small number of cases available for evaluation and possible selection bias due to
elective terminations of pregnancy. In addition, the authors stated that (1) categorizing defects may yield
etiologically dissimilar aggregations and may dilute the estimated risk; (2) at higher DBF concentrations,
multiple or lethal defects may be induced and the outcomes may be expressed as spontaneous abortion or
stillbirth, or cause unrecognized fetal loss; and (3) cases with recognized, single anomalies may represent
only a portion of the full range of the potential effects.
Reifet al. provide several possible explanations for the discrepancies and inconsistencies between
the epidemiologic studies: (1) substantial differences existed between methods of exposure assessment
and, in some cases, definition of the outcome; (2) referent groups varied across studies; (3) the
composition of DBF mixtures may have varied across locales and studies; and (4) other classes of DBFs
may be the causal agents and THMs may or may not be an appropriate exposure indicator for those DBFs.
Exposure misclassification in the studies may either hide a true effect or, in rare circumstances, create an
artificial effect.
Final Economic Analysis for the Stage 2 DBPR 6-48 December 2005
-------
Reif et al. also reviewed the epidemiology literature for dose-response relationships. The
researchers did not find a continuous pattern of increasing risk with increasing concentration of TTHM,
but they did observe a general trend of small increases in risk for concentrations of TTHM greater than
100 • g/L.
Based on information provided in the literature, Reif et al. estimated PAR for each outcome in
each study. Appendix E provides a detailed discussion of the derivation of PAR values from
epidemiological studies and their use in risk and benefits assessments.
Reif et al. explored the difference in potential health risk across TTHM thresholds of 80 and 60
(ig/L. ORs with 95 percent confidence intervals from various studies are compared in Exhibit 6.14 and
PAR values with 95 percent confidence intervals (truncated at zero to represent biological relevance)
from various studies are compared in Exhibit 6.15. The distribution of exposure levels differed among
the studies, but when normalized in this way the studies appear to provide some support for establishing a
threshold level for TTHM. The point estimates of PAR in Exhibit 6.15 are generally higher when 60
• g/L is used as the cut-point rather than 80 • g/L. This seems to indicate that an important reduction in
disease occurrence may be obtained by eliminating not only TTHM exposure levels above 80 • g/L, but
also levels between 60 • g/L and 80 • g/L. The authors note that this conclusion is tentative because
many of the 95 percent confidence intervals on the ORs were very wide and extended to values of one
and lower. Moreover, they suggest caution when interpreting the PAR values and note that "[s]ince a
number of assumptions regarding attributable fraction do not appear to hold, population attributable risks
are unlikely to be useful with the current data set."
The findings for low birth weight are varied and do not strongly support a threshold of 80 or 60
(ig/L. Reif et al. noted that the higher outcomes in the Gallagher et al. (1998) study may result from
decreased non-differential misclassification (a type of bias) by taking spatial variability into account.
Also, the DBP mixture may have been different from the mixtures used in other studies. There does not
appear to be an increased association between TTHM and intrauterine growth retardation or preterm birth
above the thresholds in question, based on the findings presented in Exhibit 6.14. The Waller et al.
(1998) study presents higher ORs for spontaneous abortions than Savitz et al. (1995). The ORs for
spontaneous abortion varied from region to region, possibly due to a difference in concentrations of
BDCM and other byproducts (Waller et al. 1998). The ORs for neural tube defects were generally higher
than for other defects above both thresholds. There was no strong evidence of increased risk, however,
for oral cleft defects or major cardiac defects (the ORs for both defects, based on Bove et al. (1995), were
high). Overall, the ORs across the 60 and 80 (ig/L thresholds were similar, but tended to be slightly
higher for 80 (ig/L.
Final Economic Analysis for the Stage 2 DBPR 6-49 December 2005
-------
Exhibit 6.14 Odds Ratios (and 95 Percent Confidence Intervals 1) Calculated by
Reif et al. (2000) for Reproductive and Developmental Health Endpoints at TTHM
Levels of > 80 ug/L versus < 80 ug/L and > 60 ug/L versus < 60 ug/L
Health
Endpoint
> 80 ug/L versus <
Low Birth Weight
Intrauterine Growth
Retardation
Preterm Birth
Spontaneous
Abortion
Stillbirths
Neural Tube Defects
Oral Cleft Defects
Major Cardiac
Defects
Dodds et
al. (1999)
Bove et al.
(1995)
Klotz and
Pyrch
(1998)
Savitz et
al. (1995)
Waller et
al. (1998)
Gallagher
etal. (1998)
80 ug/l_ TTHM
1.09
(0.99,1.19)
1.05
(0.98,1.12)
1.01
(0.92,1.10)
N/A
1.59
(1.21,2.10)
1.37
(0.88,2.15)
1.01
(0.63,1.63)
0.87
(0.70,1.08)
1.20
(1.02,1.41)
1.12
(1.03,1.22)
1.09
(0.99,1.19)
N/A
0.65
(0.45,0.95)
2.12
(1.00,4.49)
1.95
(0.87,2.90)
1.59
(0.87,2.90)
N/A
N/A
N/A
N/A
N/A
1.35
(0.65,2.79)
N/A
N/A
1.01
(0.69,1.50)
N/A
0.74
(0.51,1.07)
1.06
(0.63,1.78)
N/A
N/A
N/A
N/A
N/A
N/A
N/A
1.29
(0.98,1.69)
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
> 60 ug/L versus < 60 ug/L TTHM
Low Birth Weight
Intrauterine Growth
Retardation
Preterm Birth
Spontaneous
Abortion
Stillbirths
Neural Tube Defects
Oral Cleft Defects
Major Cardiac
Defects
1.06
(0.98,1.16)
1.05
(0.99,1.12)
0.98
(0.91,1.06)
N/A
1.56
(1.18,2.06)
1.01
(0.66,1.56)
0.91
(0.59,1.40)
0.94
(0.78,1.14)
1.07
(0.96,1.18)
1.04
(0.99,1.09)
0.96
(0.91,1.02)
N/A
0.80
(0.65,0.97)
1.34
(0.76,2.38)
1.25
(0.78,2.02)
0.93
(0.59,1.45)
N/A
N/A
N/A
N/A
N/A
1.79
(1.08,2.95)
N/A
N/A
1.35
(0.90,2.01)
N/A
1.01
(0.71,1.44)
0.97
(0.56,1.67)
N/A
N/A
N/A
N/A
N/A
N/A
N/A
1.22
(0.98,1.53)
N/A
N/A
N/A
N/A
2.24
(1.03,4.88)
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Notes: 1Lower confidence limit is truncated at zero
N/A indicates that data for that health endpoint was not presented in the study.
Source: Adapted from Reif et al. (2000).
Final Economic Analysis for the Stage 2 DBPR
6-50
December 2005
-------
Exhibit 6.15 PAR Values (and 95 Percent Confidence Intervals 1) Calculated by
Reif et al. (2000) for Reproductive and Developmental Health Endpoints at TTHM
Levels of > 80 ug/L versus < 80 ug/L and > 60 ug/L versus < 60 ug/L (Values are
Percentages)
Health
Endpoint
> 80 ug/L versus
Low Birth Weight
Intrauterine Growth
Retardation
Preterm Birth
Spontaneous
Abortion
Stillbirths
Neural Tube
Defects
Oral Cleft Defects
Major Cardiac
Defects
Dodds et
al. (1999)
Bove et al.
(1995)
Klotz and
Pyrch
(1998)
Savitz et al.
(1995)
Waller et al.
(1998)
Gallagher
etal.
(1998)
< 80 ug/l_ TTHM
2.4% (0,4.9 )
1.3% (0,3.1)
0.2% (0,2.2)
N/A
14.1%
(4.6,22.7)
9.8% (0,23.3)
0.4% (0,1 3.1)
N/A
1.5% (0.1, 2.9)
0.9% (0.2, 1.6)
0.7% (0,1. 5)
N/A
N/A
7.6% (0,1 7.0)
6.4% (0,1 3. 9)
4.1% (0,1 0.2)
N/A
N/A
N/A
N/A
N/A
3.0%(0,10.2)
N/A
N/A
0.5% (0,1 3.1)
N/A
N/A
1.9% (0,1 8.2)
N/A
N/A
N/A
N/A
N/A
N/A
N/A
4.5% (0,9.5)
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
> 60 ug/L versus < 60 ug/L TTHM
Low Birth Weight
Intrauterine Growth
Retardation
Preterm Birth
Spontaneous
Abortion
Stillbirths
Neural Tube
Defects
Oral Cleft Defects
Major Cardiac
Defects
3.2% (0,7.3)
2.7% (0,5.8)
N/A
N/A
22.5%
(7.7,34.9)
0.5% (0,20.0)
N/A
N/A
1 .8% (0,4.6)
1.0% (0,2.4)
N/A
N/A
N/A
7.8% (0,22.5)
5.9% (0,1 8.0)
N/A
N/A
N/A
N/A
N/A
N/A
14.1% (1,25.5)
N/A
N/A
18.8% (0,39.0)
N/A
0.7% (0,20.4)
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
6.5% (0,1 3.6)
N/A
N/A
N/A
N/A
6.4%
(0,14.1)
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Notes: 1Lower confidence limit is truncated at zero
N/A indicates that data for that health endpoint were not presented in the study. Note that Reif et al. (2000)
did not present PAR values for effects where the OR as shown in Exhibit 6.14 was < 1 (PAR considered by
the authors in those cases as undefined).
Source: Adapted from Reif et al. (2000).
Critical review of epidemiology literature by Bove et al. (2002)
Bove et al. (2002) conducted a qualitative review of 14 reproductive and developmental
epidemiology studies on exposure to chlorination byproducts in drinking water (many studies are the
same as those reviewed by Reif et al. [2000]). Endpoints reviewed include small size for gestational age,
birth defects (e.g., neural tube defects, cleft defects, and cardiac effects), and spontaneous abortions.
Studies that evaluated the end point of "small for gestational age" were limited due to lack of adequate
Final Economic Analysis for the Stage 2 DBPR
6-51
December 2005
-------
exposure information and low study participation rate. Studies conducted in Denver (Gallagher et al.
1998), northern New Jersey (Bove et al. 1995), central North Carolina (Savitz et al. 1995), and Nova
Scotia (Dodds et al. 1999) based their exposure estimates on tap water samples of THMs taken
concurrently with the pregnancy period. However, only one study (Gallagher et al. 1998) involved
modeling the distribution system characteristics and matching the residence with the appropriate sample
location; this likely minimized exposure misclassification, and further strengthened the relationship found
in the Denver study between TTHM and small for gestational age (OR = 5.9; no confidence intervals
reported).
Bove et al. (2002) believed that there was some consistency in the findings for neural tube defects
and oral cleft defects, but not for cardiac defects. In the two studies that evaluated neural tube defects and
individual THMs, one obtained similar associations for neural tube defects for chloroform and BDCM
(Klotz and Pyrch 1998), while the other study found a much stronger association with BDCM (Dodds and
King 2001). Of the two studies that evaluated oral cleft defects and levels of THMs, one found an
association with TTHM (Bove et al. 1995) and the other found an association with chloroform, but not
TTHM (Dodds and King 2001).
The California prospective cohort study (Waller et al. 1998) found a correlation between
spontaneous abortion and specific THMs, especially BDCM. An association was also reported for
spontaneous abortion when TTHM levels were evaluated, but this relationship disappeared when water
consumption habits were taken into account. Bove et al. (1995) noted that this study's low participation
rate was a notable weakness. In addition, because the maternal interviews were conducted after the loss
had occurred, the potential for recall bias in water consumption habits during pregnancy was introduced.
A Massachusetts study (Aschengrau et al. 1993) found no excess spontaneous abortion correlating with
treatment type (i.e., chlorination vs. chloramination), but significant effects were found for water source
(surface water, ground water).
Bove et al. (2002) evaluated three studies on the incidence of fetal deaths and THM levels that
had very different results. The Nova Scotia study (Dodds and King 2001) found a strong association,
especially with BDCM levels. In contrast, the northern New Jersey study (Bove et al. 1995) could not
evaluate the individual THM levels or information on the cause of death. Therefore, its finding of no
excess could be the result of misclassification biases due to the failure to evaluate individual THMs and
specific causes of death. The Massachusetts study (Aschengrau et al. 1993) found an association between
stillbirths and chlorinated surface water when compared with chloraminated surface water.
Critical review of epidemiology literature by Nieuwenhuijsen et al. (2000)
Nieuwenhuijsen et al. (2000) reviewed the toxicological and epidemiological literature and
evaluated the potential risk of chlorination DBFs on human reproductive health. The authors reviewed 10
epidemiological studies according to the exposure measure used: water source and water treatment,
routinely collected measurements of THMs, and routinely collected THM measurements and estimation
of individual THM ingestion. The authors commented that assessment of exposure is one of the weakest
aspects of the available epidemiological studies.
The authors concluded that the evidence from a small number of studies suggests a weak
association for spontaneous abortions, stillbirths, and birth defects, and the weight of evidence for a
stronger association is increasing as more quality studies are completed. Nieuwenhuijsen et al. concluded
that, "although studies report small risks that are difficult to interpret, the large number of people exposed
to chlorinated water supplies constitutes a public health concern."
Final Economic Analysis for the Stage 2 DBPR 6-52 December 2005
-------
Critical review of epidemiology literature by Graves et al. (2001)
Graves et al. (2001) considered the toxicological and epidemiological evidence for various
reproductive and developmental effects based on outcome, using a weight-of-evidence procedure. The
studies included in the review examined water consumption, duration of exposure, THM levels, HAA
levels, and presence of other contaminants. Many compared source water type, water treatment method,
water color (high/low), and other physical properties. Endpoints which the authors found that the weight
of evidence showed no association included low birth weight (5 articles), very low birth weight (2
articles), preterm delivery (7 articles), cesarean delivery (1 article), congenital anomalies by severity (1
article), spina bifida (1 article), cleft lip and palate (4 articles), cardiac anomalies (4 articles),
gastrointestinal anomalies (1 article), genital anomalies (1 article), integument anomalies (1 article),
musculoskeletal anomalies (1 article), chromosomal abnormalities (1 article), and neonatal death (1
article). Endpoints which the authors found that the weight of evidence showed mixed, inconsistent or
weak results included neonatal jaundice (1 article), all congenital anomalies/birth defects (3 articles), all
CNS anomalies (2 articles), neural tube defects (4 articles), respiratory anomalies (2 articles),
SAB/miscarriage (3 articles), and stillbirth/fetal death (4 articles). Those endpoints that were suggestive
of an association included growth retardation including term low birth weight (3 articles), IUGR or SGA
(3 articles), and small body length and cranial circumference (1 article), and urinary tract defects (2
articles). The authors note that the exposure characterization in the epidemiological studies to may not be
adequate to show an association of small magnitude. The authors also caution the use of quarterly or
routine monitoring of THMs matched to maternal residence as a representation of exposure.
Critical review of the epidemiology literature by Hwang and Jaakkola (2003)
Hwang and Jaakkola (2003) reviewed epidemiological studies for birth defects and performed a
meta-analysis of the studies which provided estimates of exposure on one or more birth defects. The
review presented studies including the following endpoints: any birth defect (3 studies), neural tube defect
(4 studies), major cardiac effect (3 studies), respiratory defect (2 studies), oral cleft defect (3 studies), and
urinary system defect (2 studies). The meta-analysis supports an association between exposure to
chlorination by-products and the risk of any birth defect, particularly the risk of neural tube defects and
urinary system defects. Results for cardiac defects, respiratory defects, and oral clefts were inconsistent.
EPA 's epidemiology research program
EPA's epidemiology research program continues to examine the potential relationship between
exposure to DBFs and adverse developmental and reproductive effects. The Agency is supporting several
studies using improved study designs to provide better information for characterizing potential risks.
6.2.2.2 Toxicological Evidence of Adverse Reproductive and Developmental Health Effects
EPA has evaluated published studies of the potential adverse effects of DBFs on the reproductive
and developmental health of laboratory animals. Especially pertinent information comes from reviews of
the toxicology literature by Dr. Roche lie Tyl (2000): "Review of Animal Studies for Reproductive and
Developmental Toxicity Assessment of Drinking Water Contaminants: DBFs" and by the World Health
Organization (2000): "Environmental Health Criteria 216: Disinfectants and Disinfection Byproducts."
Review of Tyl (2000)
Tyl evaluated the literature using the EPA developmental (USEPA 1991b) and reproductive
(USEPA 1996d) toxicity risk assessment guidelines. Tyl presented this critical review during the FACA
process, and the analysis was important in deliberations. Tyl focused her analysis on making
determinations regarding hazard identification (that is, identifying the specific types of adverse effects
Final Economic Analysis for the Stage 2 DBPR 6-53 December 2005
-------
caused by these substances) and the adequacy of data from the available studies to support the
development of dose-response assessments.
Exhibit 6.16, adapted and updated from Tyl (2000), lists the types of reproductive and
developmental toxicology studies that have been performed for various disinfectants and specific DBFs.
In Exhibit 6.16, the study types are classified as either screening studies or as dose-response studies.
Tyl concluded, based upon a weight-of-evidence approach to the analysis of the available,
relevant literature, that "some of the DBFs have the intrinsic capacity to do harm, specifically to the
developing conceptus and the male (and possibly the female) reproductive system." Specific reproductive
and developmental hazards that have been identified and associated with exposure to various DBFs are
summarized in Exhibit 6.17.
Notwithstanding the evidence supporting the identification of developmental and reproductive
effects from DBF exposure, Tyl also concluded that the weight of evidence does not support a dose-
response evaluation based on existing studies. (She notes as "one possible exception" the 1996 Chemical
Manufacturers Association's two-generation rat study on chlorite.)
Tyl also noted in her summary and conclusions that in a review of animal literature for the
purpose of risk assessment, "biological plausibility" is a major concern. Tyl pointed to several aspects of
both the in vitro and in vivo studies that support the biological plausibility that DBFs can cause adverse
reproductive and developmental effects. In particular, there was an observed temporal relationship
between the exposures in toxicological studies and the occurrence of the developmental (e.g., embryonic
neural tube, embryonic heart) or reproductive process (e.g., spermatogenesis). The observed effects were
reproducible in the same or similar study designs. The effects were consistent across study designs. The
effects observed in animal toxicological studies were comparable to those observed in some human
epidemiological studies (e.g., embryonic heart and neural tube defects, full litter resorption/miscarriage,
spontaneous abortion, or stillbirth).
Final Economic Analysis for the Stage 2 DBPR 6-54 December 2005
-------
Exhibit 6.16 Availability of Reproductive and Developmental Toxicology Studies
for Specific DBFs
Disinfectant or DBF
Screens - Hazard Identification
WEC
NTP35
Dav
CKA
CKA++
Male
Reoro
Dose Response
Sea II
Multi-
GEN
DISINFECTANTS
Chlorine
Chlorine Dioxide
Chloramine
X
X
X
X
TRIHALOMETHANES
Chloroform
Bromoform
Bromodichloromethane
Dibromochloromejfjgfle^^^^
X
X
X
X
X
X
X
X
X
X
X
X
X
X
HALOACETIC ACIDS
Monochloroacetic acid
Dichloroacetic acid
Trichloroacetic acid
Monobromoacetic acid
Dibromoacetic acid
Tribromoacetic acid
Bromochloroacetic acid
Bromodichloroacetic acid
DibromochloroacŁjjc^cjd^^_
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
P
HALOACETONITRILES
Chloroacetonitrile
Dichloroacetonitrile
Trichloroacetonitrile
Bromoacetonitrile
Dibromoacetonitrile
Tribromoacetonitrile
Bromochloroacel^jjjjrjje^^^^
X
X
X
X
X
X
X
X
X
X
X
ALDEHYDES
Formaldehyde
Acetaldehyde
Prooanal
X
X
X
X
X
X
X
X
MISCELLANEOUS
1 ,1 -Dichloropropanone
Hexachloropropanone
Dichloromethane
Dibromomethane
MX
Bromate
Chlorite
X
X
X
X
X
X
X
X
X
X
Notes: X = Completed and published in the literature; P = In planning stage; WEC = Whole embryo culture; NTP 35
Day = NTP 35-day reproductive/ developmental toxicity screen; CKA = Chernoff-Kavlock Assay; CKA (++) =
Chernoff-Kavlock Assay (modified); Male Repro. = Short-term adult male reproductive toxicity screen;
Seg II = Segment II developmental toxicity study; Multi-GEN = Multigeneration reproductive toxicity study.
Source: Adapted and updated from Tyl (2000).
Final Economic Analysis for the Stage 2 DBPR
6-55
December 2005
-------
Exhibit 6.17 Reproductive and Developmental Health Effects
Associated with DBFs in Toxicological Studies
Type of Effect
Developmental defects
Whole litter resorption
(miscarriage/spontaneous abortion)
Fetotoxicity (reduced fetal body
weights, increased anomalies like
chromosomal defects)
Male reproductive defects
DBF
Trichloroacetic acid (TCAA), dichloroacetic acid (DCAA), and
monochloroacetic acid (MCAA)
Chloroform, bromoform, bromodichloromethane (BDCM),
dibromochloromethane (DBCM), DCAA, TCAA,
dichloroacetonitrile (DCAN), and trichloroacetonitrile (TCAN)
Chloroform, BDCM, DBCM, DCAA, TCAA, DCAN, TCAN,
dibromoacetonitrile (DBAN), bromochloroacetonitrile (BCAN),
monochloroacetonitrile (MCAN) acetaldehyde, formaldehyde
DCAA, dibromoacetic acid (DBAA), BDCM, formaldehyde
Source: Adapted from Tyl (2000).
Critical Review of Toxicological Literature by Graves et al. (2001)
As described in the prior Section 6.2.2.1 on epidemiological evidence, Graves, et al. conducted a
weight of evidence analysis on epidemiologic and toxicologic studies of the association between DBFs
and reproductive or developmental effects. Study results for the epidemiological and lexicological
weight of evidence analyses were combined and are explained in summary form in section 6.2.2.1.
The authors noted that the toxicological data support that normal exposure to DBFs through tap
water would generally not cause "adverse effects," however, for the effects listed as suggestive of positive
associations or producing mixed, inconsistent, or weak results, further epidemiological research is
warranted. One problem highlighted by the authors is that the current literature lacks accurate or detailed
exposure assessment data. Accurate measurement of consumption (and exposure through dermal contact
and inhalation) is needed, along with information on the DBF components of the drinking (or bathing)
water. Alternatively, the authors indicate that when biomarkers are developed, they will potentially
provide information on individual exposure.
Review of WHO (2000)
The International Programme on Chemical Safety (IPCS) of the World Health Organization
(WHO) published an evaluation of Disinfectants and DBFs in its Environmental Health Criteria
monograph series (WHO 2000). In this review of the toxicology data on reproductive and developmental
effects from DBF exposure, the WHO concludes that although the data on these effects are not as robust
as the cancer database, these effects are of potential health concern. They also conclude that reproductive
effects in females have been principally embryolethality and fetal resorptions associated with the
haloacetonitriles (HANs) and the dihaloacetates, while DCAA and DBAA have both been associated with
adverse effects on male reproduction, including testicular toxicity and spermatoxic effects.
New Toxicology Data Since the Stage 1 DBPR
Since promulgating the Stage 1 DBPR, more research on DBFs is underway at EPA and other
research institutions. For more information, on-going studies may be found on EPA's DRINK website
(http://www.epa.gov/safewater/drink/intro.html). Summaries of new studies are provided below.
Final Economic Analysis for the Stage 2 DBPR
6-56
December 2005
-------
Chen et al. (2003) studied the in vitro effect of bromodichloromethane on chorionic gonadotropin
(CG) secretion by human placental trophoblast cultures. Exposure to bromodichloromethane caused a
significant dose-dependent decrease in the secretion of immunoreactive and bioactive CG. The lowest
concentration that produced a statistically significant response was 0.02 • M, the lowest concentration
tested. Chen et al. (2004) also reported that addition of 0.02 to 2 mM of bromodichloromethane inhibited
morphological differentiation of human mononucleated cytotrophoblast cells to multinucleated
syncytiotrophoblast-like colonies. The significance of the findings reported by Chen et al. (2003, 2004)
for human health is that placental trophoblasts are the sole source of CG during normal human pregnancy
and play a major role in the maintenance of the conceptus.
Christian et al. (2001) conducted a developmental toxicity study with pregnant New Zealand
White rabbits exposed to BDCM in drinking water at concentrations of 0, 15, 150, 450, and 900 parts per
million (ppm) in drinking water on gestation days 6-29. The No-Observed-Adverse-Effect-Level
(NOAEL) and Lowest-Observed-Adverse-Effect-Level (LOAEL) identified for maternal toxicity in this
study were 13.4 milligrams per kilogram per day (mg/kg-day) (150 ppm) and 35.6 mg/kg-day (450 ppm),
respectively, based on decreased body weight gain. The developmental NOAEL was 55.3 mg/kg-day
(900 ppm) based on the absence of statistically significant, dose-related effects at any tested
concentration. Christian et al. also conducted a developmental study of BDCM in a second species,
Sprague-Dawley rats. Rats were exposed to BDCM in the drinking water at concentrations of 0, 50, 150,
450, and 900 ppm on gestation days 6 to 21. The concentration-based maternal NOAEL and LOAEL for
this study were 150 ppm and 450 ppm, respectively, based on statistically significant, persistent
reductions in maternal body weight and body weight gains. Based on the mean consumed dosage of
bromodichloromethane, these concentrations correspond to doses of 18.4 mg/kg-day and 45.0 mg/kg-day,
respectively. The concentration-based developmental NOAEL and LOAEL were 450 ppm and 900 ppm,
respectively, based on a significantly decreased number of ossification sites per fetus for the fore limb
phalanges (bones of the hand) and the hindlimb metatarsals and phalanges. These concentrations
correspond to mean consumed doses of 45.0 mg/kg-day and 82.0 mg/kg-day, respectively.
Christian et al. (2002a) summarized the results of a two-generation reproductive toxicity study on
bromodichloromethane conducted in Sprague-Dawley (SD) rats. Bromodichloromethane was
continuously provided to test animals in the drinking water at concentrations of 0, 50, 150, or 450 ppm.
Average daily doses estimated for the 50, 150 and 450 ppm concentrations were reportedly 4.1 to 12.6,
11.6 to 40.2, and 29.5 to 109 mg/kg-day, respectively. The parental NOAEL and LOAEL were 50 and
150 ppm, respectively, based on statistically significant reduced body weight and body weight gain; Fl
and F2 generation pup body weights were reduced in the 150 and 450 ppm groups during the lactation
period after the pups began to drink the water provided to the dams. Body weight and body weight gain
were also reduced in the 150 and 450 ppm Fl generation males and females. A marginal effect on estrous
cyclicity was observed in Fl females in the 450 ppm exposure group. Small (• 6 percent), but statistically
significant, delays in Fl generation sexual maturation occurred at 150 ppm (males) and 450 ppm (males
and females) as determined by timing of vaginal patency or preputial separation. The study authors
considered these effects to be a secondary response associated with reduced body weights cause appears
to be dehydration brought about by taste aversion to the compound. The results of this study identify
NOAEL and LOAEL values for reproductive effects of 50 ppm (4.1 to 12.6 mg/kg-day) and 150 ppm
(11.6 to 40.2 mg/kg-day), respectively, based on delayed sexual maturation.
Bielmeier et al. (2001) conducted a series of experiments to investigate the mode of action in
bromodichloromethane-induced full litter resorption (FLR). The study included a strain comparison of
F344 and SD rats. In the strain comparison experiment, female SD rats (13 to 14/dose group) were dosed
with 0, 75, or 100 mg/kg-day by aqueous gavage in 10 percent Emulphor® on gestation day (GD) 6 to 10.
F344 rats (12 to 14/dose group) were dosed with 0 or 75 mg/kg-day administered in the same vehicle.
The incidence of FLR in the bromodichloromethane-treated F344 rats was 62 percent, while the incidence
of FLR in SD rats treated with 75 or 100 mg/kg-day of bromodichloromethane was 0 percent. Both
strains of rats showed similar signs of maternal toxicity, and the percent body weight loss after the first
Final Economic Analysis for the Stage 2 DBPR 6-57 December 2005
-------
day of dosing was comparable for SD rats and the F344 rats that resorbed their litters. The rats were
allowed to deliver and pups were examined on postnatal days 1 and 6. Surviving litters appeared normal
and no effect on post-natal survival, litter size, or pup weight was observed. The series of experiments
conducted by Bielmeier et al. identified a LOAEL of 75 mg/kg-day (the lowest dose tested) based on FLR
in F344 rats. A NOAEL was not identified. Mechanistic studies reported by Bielmeier et al. indicate that
BDCM-induced pregnancy loss is likely to be luteinizing hormone (LH)-mediated (Bielmeier et al. 2004).
In this more recent study, Bielmeier et al. hypothesizes that BDCM alters LH levels by disrupting the
hypothalamic-pituitary-gonadal axis or by altering the responsiveness of the corpora lutea to LH. These
possible mechanisms are potentially relevant to pregnancy maintenance in humans. EPA believes the
finding of BDCM-induced pregnancy loss in F344 rats is re levant to risk assessment, and may provide
insight into the epidemiological finding of increased risk of spontaneous abortion associated with
consumption of BDCM (Waller et al. 1998, 2001).
Christian et al. (2002b) performed a two-generation drinking water study of DBAA in rats. Male
and female Sprague-Dawley rats (30/sex/exposure group) were administered DBAA in drinking water at
concentrations of 0, 50, 250, or 650 ppm continuously from initiation of exposure of the parental (P)
generation male and female rats through weaning of the F2 offspring. Based on testicular
histomorphology indicative of abnormal spermatogenesis in P and Fl males, the parental and
reproductive/developmental toxicity LOAEL and NOAEL are 250 and 50 ppm, respectively.
Previous studies by EPA have reported adverse effects of DBAA, administered via oral gavage,
on spermatogenesis that impacted male fertility (Linder et al. 1994, 1995, 1997) at doses comparable to
those achieved in the Christian et al. (2002b) study. Based on these studies collectively, DBAA is
spermatotoxic. Moreover, Veeramachaneni et al. (2000) reported in an abstract that sperm from male
rabbits exposed to DBA in utero from gestation days 15 and throughout life reduced the fertility of
artificially inseminated females as evidenced by reduced conceptions. When published, this study may
support the evidence that DBA is a male reproductive system toxicant.
In addition, research on DBAA by Klinefelter et al. (2001) has demonstrated statistically
significant delays in both vaginal opening and preputial separation using the body weight on the day of
acquisition (at postnatal day 45) as the co-variant. This was not found by Christian et al. (2002b) using
the body weight at weaning as the statistical covariant. However, the authors analyzed the data for
preputial separation and vaginal opening with body weight on the day of weaning as a co-variant rather
than body weight on the day of acquisition, i.e. the day that the prepuce separates or the day the vagina
opens. It is likely that there was an increase in body weight from postnatal day 21 (weaning) until
preputial separation (day 45) that was independent of the delay in sexual maturation. A more recent study
by Klinefelter et al. (2004) found that exposure to either 4, 40, or 400 ppm of DBAA from GD 15 through
post-natal day (PND) 21 did not result in any significant reproductive alterations, but effects were seen
with continuous exposure until adulthood. Males and females exposed to 400 ppm DBAA were reported
to have delayed preputial separation and vaginal openings by 4 and 3 days, and also an increased
responsiveness of both the testis ad ovary to human choriogonadotropin (hCG). At 4 ppm DBAA and
higher, hCG-stimulated testosterone production by testicular parenchyma on PND 56 was increased.
Also, continuous exposure to DBAA compromised the quality of proximal cauda epididymal sperm.
Kaydos et al. (2004) administered DBA and BCA, individually and in combination to investigate
the effect on fertility and spermatogenesis in the rat. Since humans are exposed to a complex mixture of
DBFs in disinfected drinking water, it is important to study the potential effects of mixtures of DBFs that
may elicit similar reproductive effects. The authors of this study were able to find dose and effect
additivity, and in some cases synergism, for haloacid-induced decrease in fertility.
Although the Christian et al. (2002b) study was conducted in accordance with EPA's 1998 testing
guidelines, EPA has incorporated newer, more sophisticated measures into recent intramural and
Final Economic Analysis for the Stage 2 DBPR 6-58 December 2005
-------
extramural studies that have not yet been incorporated into the testing guidelines. Such measures include
changes in specific proteins in the sperm membrane proteome and fertility assessments via in utero
insemination. EPA believes that additional research is needed, using these newer toxicological measures,
to clarify the extent to which DBAA poses human reproductive or developmental risk. The database on
male reproductive effects from exposure to DBAA is incomplete and is not suitable for quantitative risk
assessment at this time. It does identify reproductive effects as an area of concern.
In addition, EPA has prepared individual supporting documents that provide detailed summaries
of the relevant new information, as well as an overall characterization of the human health risks from
exposure to these DBFs (USEPA 2000f, 2005b-e, 2005J; IRIS 2000, 2001a-b, 2003). Overall,
reproductive and developmental toxicology studies indicate a possible reproductive/developmental health
hazard although they are preliminary in nature for the majority of DBFs, and the dose-response
characteristics of most DBFs have not been quantified. Some of the reproductive effects of DCAA were
quantified as part of the RfD development process, and impacts of DCAA on testicular structure are one
of the critical effects in the study that is the basis of the RfD (IRIS 2003).
Biological plausibility for the effects observed in reproductive and developmental
epidemiological studies has been demonstrated through various toxicological studies on some individual
DBFs (e.g., Bielmeier et al. 2001, Bielmeier et al. 2004, Narotsky et al. 1992, Chen et al. 2003, Chen et
al. 2004). Some of these studies were conducted at high doses, but similarity of effects observed between
toxicology studies and epidemiology studies strengthens the weight of evidence for a possible association
between adverse reproductive and developmental health effects and exposure to chlorinated surface
water.
6.2.2.3 Conclusions
EPA believes that toxicology and epidemiology data do not support a conclusion at this time as to
whether exposure to chlorinated drinking water causes adverse reproductive or developmental effects, but
do support a potential health concern. Although scientific knowledge about the association of
reproductive and developmental health effects with DBF exposure is not known well enough to quantify
these risks or the benefits of reduced DBF exposure, EPA concludes that the data are sufficient to
determine that a concern exists that warrants additional regulatory action. The following are the specific
key factors used to support EPA's weight-of-evidence conclusion:
• The results of several studies performed by different researchers with different methods
at different research sites show similar trends.
• Some health effects observed in animal toxicological studies are comparable to those
observed in some human epidemiological studies (e.g., embryonic heart and neural tube
defects, full litter resorption/miscarriage, spontaneous abortion, or stillbirth) showing
similarity of effects between animal toxicity and human epidemiology studies.
• Difficulties in assessing exposure to DBFs, resulting in exposure misclassification, may
underestimate reproductive and developmental risks associated with DBFs. It is possible
that some of the inconsistencies reported in epidemiological and toxicological study
results are due to these misclassifications, and the true effects may be greater than
demonstrated. A spurious effect would be produced only in rare cases, and is unlikely as
described in Reif et al. (2000) and Bove et al. (2002).
EPA's epidemiology and toxicology research programs continue to examine the relationship
between exposure to DBFs and potential adverse reproductive and developmental health effects. EPA is
Final Economic Analysis for the Stage 2 DBPR 6-59 December 2005
-------
also supporting several studies using improved study designs to provide better information for
characterizing potential risks.
6.3 Exposure Assessment
6.3.1 Population Exposed
Because DBFs are formed when disinfectants combine with organic compounds, the population
at risk is identified as the population served by drinking water systems that disinfect. A very large portion
of the United States population—approximately 94 percent—is potentially exposed to DBFs in
disinfected drinking water. Exhibit 6.18 contains EPA's estimates of the population potentially exposed.
Nearly 260 million people in the United States are served by community water systems (CWSs) that
apply a disinfectant to water to protect against microbial contaminants. In addition to those served by
CWSs, just over 2 million individuals are served regularly by nontransient noncommunity water systems
(NTNCWSs). (See Exhibit 3.3 for population served by different system types.)
Two population subgroups of concern are women of child-bearing age and developing fetuses.
Women of child-bearing age are generally considered to be those in the age range of 15 to 45. The
estimated U.S. population for the year 2000 is 281 million, of which approximately 64 million (23
percent) are females between the ages of 15 and 45. Because approximately 94 percent of the population
is served by PWSs that disinfect, it can be estimated that about 60 million women of child-bearing age are
served by these water supplies. Currently, there are approximately 4 million live births each year. Again,
using the factors above, it can roughly be estimated that more than 3.8 million infants are born each year
to mothers served by a disinfecting water supply.
Exhibit 6.18 Estimated Population Exposed to DBFs
in Drinking Water
System Size
(Population Served)
<100
101-500
501-1,000
1,001-3,300
3,301-10,000
10,001-50,000
50,001-100,000
100,001-1,000,000
1,000,000 +
Total
Population Served
by Disinfecting
Systems
387,001
2,898,196
3,210,348
10,776,884
19,664,185
51,311,073
26,218,703
88,996,647
59,561,481
263,024,518
Percent of
Total
Population
Served by
Disinfecting
Systems
0.15%
1.10%
1 .22%
4.10%
7.48%
19.51%
9.97%
33.84%
22.64%
100.00%
Source: Derived from the Stage 2 DBPR population baseline
for surface and disinfecting ground water CWSs in Exhibit 3.3.
Final Economic Analysis for the Stage 2 DBPR
6-60
December 2005
-------
6.3.2 Routes of Exposure
An important route of exposure to DBFs is from the direct ingestion of drinking water and from
the consumption of food that has been cleaned, processed or prepared with drinking water. EPA has
examined drinking water consumption data from the 1994-1996 USDA Continuing Survey of Food
Intakes by Individuals (USDA 1997) and determined that mean daily drinking water consumption across
all ages, sexes, and regions in the United States ranges from 0.9 to 1.2 liters per day, with an upper 95th
percentile range of 2.5 to 2.9 liters per day (USEPA 2000i).
People also can be exposed to some contaminants in drinking water by routes of exposure other
than ingestion, particularly by inhalation and dermal contact from showering, bathing, washing dishes,
washing clothes, or swimming. An international work group was convened in 2000 to assess the
challenges involved in assessing exposure to DBFs in epidemiological studies. The workgroup concluded
that accurate exposure characterization (from all routes) is extremely important for a valid risk assessment
(Arbuckle et al. 2002). The remainder of this section focuses on routes of exposure other than direct
ingestion.
Some studies have found that exposure due to inhalation and skin absorption during showering
may actually be higher than ingestion-related exposure (Kuo et al. 1998, Backer et al. 2000, Miles et al.
2002). Backer et al. (2000) found that when volunteers either showered with tap water for 10 minutes,
bathed in tap water for 10 minutes, or drank 1 liter of tap water over 10 minutes, the highest levels of
THMs were found in the blood samples from people who took 10 minute showers, whereas the lowest
levels were found in the blood samples from people who drank 1 L of water in 10 minutes. The results
from this study indicate that household activities such as bathing and showering are important routes for
human exposure to THMs.
Singer et al. (2003) examined the extent to which everyday uses of disinfected drinking water
activities in the household effect levels of THMs in the blood. It was concluded that the average ratio of
chloroform in blood to chloroform in water is greatest for activities where there is a potential for
significant inhalation or dermal exposure (i.e. showering, bathing, and washing dishes by hand).
The route of exposure may depend on the volatility and chemical phase of the DBP in question
(Weisel et al. 1999, Xu et al. 2002, Xu and Weisel 2003, 2004, 2005). THMs are more volatile than
HAAs and exist in the air at background levels (Weisel et al. 1999). Xu and Weisel (2003) found that the
estimated dose from inhalation of particulate-phase DBFs represented less than 1 percent of the ingestion
dose, and the vapor-phase DBFs can contribute to over 10 percent of the ingestion dose during a shower.
Weisel et al. (1999) found a significant correlation between breath concentrations of chloroform after
showering and water concentration. They were unable to show a relationship between breath
concentration and overall exposure due to variability in when breath samples were taken and the fast rate
of THM metabolization. However, almost all breath concentrations, except for chloroform, were below
the detection limit, most likely due to low water concentrations. Weisel et al. also found a link between
urinary trichloroacetic acid (TCAA) excretion rates and TCAA exposure, calculated as the water's TCAA
concentration multiplied by the volume of water consumed by volunteers over a 48 hour period, adjusted
for home filters and boiling. TCAA exposure generally fell below 10 ng over 48 hours, but ranged up to
50 ng. In addition, Batterman et al. (2000) quantified volatization rates for several TTHMs when
preparing, storing and serving tap water and found that heating up tap water (as in hot beverages) reduces
the amount of exposure upon ingestion.
Exposure appears to differ between males and females, with males absorbing more chloroform
than females (Corley et al. 2000). Most significantly, dermal absorption, especially in women, is
significantly affected by water temperature, since at higher temperatures blood flow to the skin is higher,
allowing the blood to take in more DBFs (Corley et al. 2000, Gordon et al. 1998). Measuring the
chloroform breath concentrations of test subjects immersed in bath water at concentrations of-90 fig/L at
Final Economic Analysis for the Stage 2 DBPR 6-61 December 2005
-------
35°C for 30 minutes, Corley et al. calculated that men would absorb 42 (ig and women would absorb 12
(ig of chloroform. These exposures were based on models taking blood flow to the skin and skin
permeability into consideration. At 40°C, this increased to 44 (ig for men and 40 (ig/L for women. (This
includes chloroform exhaled, metabolized, and maintained in the body.) At 30°C, on the other hand,
most subjects had chloroform breath levels below detection limits. For example in comparison to dermal
absorption, if 2 liters of water with the same chloroform concentration were ingested, the exposure from
ingestion would range from 79 to 194 (ig.
6.3.2.1 Special Exposure Issues for Pregnant Women
Because of the potential reproductive and developmental effects of DBFs, pregnant women
represent a population subset of special concern with respect to the intake of DBFs from drinking water.
Because the kidneys work harder in pregnancy to expel waste material from the body, drinking an
adequate volume of water is extremely important. Pregnant women can become dehydrated easily, which
can lead to fetotoxicity. Thus, women who are pregnant are encouraged to drink a minimum of eight 8-
ounce glasses of water a day to ensure proper hydration (March of Dimes 1999).
Pregnant women cannot avoid tap water completely. As discussed in the previous section, DBP
exposure can occur through inhalation and dermal contact from a variety of activities, including
showering and bathing. Zender et al. (2001) found that pregnant women bathed more often and for a
longer duration than non-pregnant women but showering patterns were similar. Bottled water is not
necessarily safer than tap water and is more expensive. Bottled water, even if safer, may not be an option
for economically disadvantaged pregnant women. Because pregnant women cannot avoid tap water, the
expected reduction in DBP exposure that is estimated for the Stage 2 DBPR is especially important in
providing public health protection to them and their developing children.
6.3.3 Exposure Reduction
It is well recognized that DBP concentrations can vary greatly throughout a distribution system
and over time at the same location in a distribution system. The Stage 1 DBPR requires systems to meet
the maximum contaminant level (MCL) standards and associated compliance monitoring requirements as
running annual averages (RAAs) of 80 • g/L for TTHM and 60 • g/L for HAAS as averaged across all
monitoring locations. It is possible that some systems can achieve the average concentration targeted by
the Stage 1 DBPR, and yet still have some locations in the distribution system where average DBP levels
are far in excess of the system-wide target at some, or even at all, times. The peak exposures resulting
from these high concentrations are of particular concern in regard to potential adverse reproductive and
developmental health effects. Exposure at locations having repeatedly high sample concentrations are of
particular concern for pregnant women, who are encouraged to drink more water than the average person
and who may be especially sensitive to the potential effects of DBFs, as explained in the previous section.
Under the Stage 2 DBPR preferred regulatory alternative (which includes a requirement for the
IDSE), TTHM and HAAS MCLs will remain at 80 (ig/L and 60 (ig/L, respectively, but compliance will
be based on the locational running annual average (LRAA). Exhibits 1.3 and 4.1 illustrate how the
LRAA and RAA are calculated. This revised compliance calculation requirement will reduce average
DBP levels in the entire distribution system as well as avoid having average DBP levels at any sampling
location exceed 80 and 60 • g/L for TTHM and HAAS, respectively. Systems are required to meet the
Stage 2 DBPR MCLs at revised sampling locations that will be identified through the IDSE to further
ensure that peak occurrence events are captured and controlled.
Final Economic Analysis for the Stage 2 DBPR 6-62 December 2005
-------
The Stage 2 DBPR Preferred Regulatory Alternative is expected to yield health benefits by
achieving the following effects in those systems subject to the rule:
1) Reducing exposures to all DBFs levels.
2) Reducing exposures to single peak occurrences or repeated peak occurrence at location
that consistently exceed TTHM and HAAS MCL levels.
The next sections discuss these two ways that the Stage 2 DBPR reduces exposures to DBFs.
6.3.3.1 Reducing Exposure to All Levels of DBFs
In Chapter 5, EPA estimates the reduction in the national average TTHM and HAAS
concentrations occurring in drinking water distribution systems as a result of new treatment to meet Stage
2 DBPR requirements. Results from this analysis for TTHM are summarized below in Exhibit 6.19.
The average reduction in plant-mean TTHM and HAAS concentrations is assumed to represent
the range of reductions for all chlorination DBFs. Using these two DBP classes as "indicators" for all
chlorination DBFs may overestimate or underestimate the true concentration reduction (see Section 6.6
for a summary of uncertainties). However, because measurable halogen-substituted DBP concentrations,
comprised primarily of TTHM and HAAS, are estimated to make up 30 to 60 percent of the measured
total organic halide (TOX) concentration (Singer 1999), TTHM and HAAS reductions are assumed to be
reasonable indicators of the overall chlorination DBP reductions. Separate evaluations for TTHM and
HAAS are carried throughout the analyses.
The average reduction in TTHM and HAAS concentration is a key input in the estimation of
benefits of the Stage 2 DBP Rule. The reduction in concentration leads to a reduction in exposure, which
in turn reduces the incidence of disease. Section 6.4 details the estimation of the benefits of the Stage 2
DBP Rule.
Exhibit 6.19 National Average TTHM1 Reduction Estimates
Preferred Alternative
Alternative 1
Alternative 2
Alternative 3
Mean and 90% Confidence Bounds on
Percent TTHM Reduction from Pre-
Stage 2 to Post-Stage 2
Mean
7.81%
7.15%
26.91%
37.12%
5th %ile
Lower CB
4.53%
5.93%
23.61%
36.00%
95th %ile
Upper CB
11.18%
8.36%
30.20%
38.25%
Note: Estimates of mean and 90% confidence bounds (CB) incorporate uncertainty in
compliance forecast methodologies.
1 Reductions in HAASs are very similar. See Section 5.5 for more detail.
Source: Stage 2 Benefits Model (USEPA 2005h)
Final Economic Analysis for the Stage 2 DBPR
6-63
December 2005
-------
6.3.3.2 Reducing Exposure to Peak DBF Occurrences
EPA used distribution system data from the Information Collection Rule (ICR) to estimate the
reduction in occurrences of peak DBF concentrations that result from the Stage 2 DBPR. Section 5.6.1
provides a detailed explanation of the methodology used to generate these estimates and presents results
in Exhibit 5.27. Section G.3 presents an analysis of the reduction in peak observations as well as
reduction in three-quarter averages that are greater than 75 • g/L. Note that since the developmental and
reproductive health data described in Section 6.2 do not conclusively identify the peak level of concern,
Section 5.6 and G.3 provide an analysis for several possible peak TTHM and HAA5 concentrations, or
study levels.
Exhibit 5.27 shows that, at a TTHM study level of 75 • g/L, the percent of distribution system
sampling locations with at least one peak observation declines from 20.1 percent for pre-Stage 1 to 8.2
percent for pre-Stage 2 to 3.3 percent for post-Stage 2 DBPR conditions. To translate estimated changes
in peak DBF occurrence as a result of the Stage 2 DBPR to changes in peak DBF exposure, the following
assumptions are used:
1) Each ICR sampling-location (DSE, AVG1, AVG2, and DS Maximum) represents an equal
portion (25 percent) of the total population served by the plant.
2) Peak DBF occurrence for the 311 large ICR surface and ground water plants evaluated in
Section 5.5 is representative of the peak DBF occurrence for all plants (large and small).
The first assumption may overestimate the population represented by the DS Maximum location
(i.e., 25 percent may be too high) and thus, may overestimate the population exposed to peaks. This
potential overestimate, however, is minimized because ICR data showed that the peak TTHM level
occurred somewhere other than the DS maximum location approximately 52 percent of the time (see
Chapter 3 of the Stage 2 DBPR Occurrence Document (USEPA 2005k)). The rationale for the second
assumption is provided in the next three paragraphs.
ICR data pertains to all systems serving 100,000 or more people. The 311 plants evaluated
represent 62 percent (311/500) of all plants in the ICR. Systems serving 100,000 or more people serve
approximately 149 million people, or 56 percent (149 million/264 million) of the total population served
by disinfecting systems (Exhibit 6.18 provides a summary of the population served by each disinfecting
system size category). Thus, the 311 plants encompasses approximately 35percent (56 percent x 62
percent) of the total population served by disinfecting systems.
Because medium-sized systems serving 10,000 to 99,999 people are expected to have treatment
technologies and source water quality very similar to large systems serving 100,000 or more people, EPA
believes that ICR large-system data is adequate for characterizing peak DBP occurrence for medium
systems. (See Appendices A and B for comparisons of source water quality data and treatment
technologies in place for medium and large systems.)
For small systems serving fewer than 10,000 people, using ICR data to characterize pre-Stage 1
peak occurrence may bias the results of this analysis for two reasons. First, small systems serving fewer
than 10,000 people were not required to comply with the 1979 TTHM standard of 100 • g/L and may
have higher DBP levels than indicated by ICR data. Alternately, small systems may have lower DBP
levels than indicated by ICR data since they are made up of a higher proportion (more than 75 percent) of
ground-water-only systems compared to large systems. It is expected that these two biases offset each
other to some extent in the analysis of pre-Stage 1 data. It is important to note that biases in
characterization of DBP peaks nationally that are caused by differences in small-system occurrence are
minimized because systems serving fewer than 10,000 people represent only 14.8 percent of the total
population served.
Final Economic Analysis for the Stage 2 DBPR 6-64 December 2005
-------
TTHM and HAAS concentrations are highly variable in distribution systems; it is probable that
this analysis does not capture the true variability in exposure to peaks. Uncertainties with interpretation
of ICR data for the purposes of this exposure assessment include:
• The extent to which small-system occurrence is represented;
• Year-to-year variability of DBF occurrence data that might be affected by changes in source
water quality (e.g., drought years versus non-drought years);
The extent to which each ICR sampling point represents an equal fraction of the population
served; and
• The extent to which ICR sampling locations represent compliance monitoring locations when
trying to estimate reductions in exposure resulting from compliance with Stage 1 and Stage 2
DBPRs.
The assumptions in this section are necessary, however, for predicting exposure changes given the limited
data on DBF occurrence in small systems and in distribution systems in general. Using the two
assumptions listed above, the reduction in plant-locations with peaks as a result of the Stage 2 DBPR
(shown in Exhibit 5.27) can be taken to represent the reduction in exposure to peaks nationally as a result
of the Stage 2 DBPR. For example, for a TTHM study level of 80 • g/L, the percent of the population
exposed to peak DBFs is predicted to decline from 6.0 to 1.7 percent (a 70 percent reduction) as a result
of the Stage 2 DBPR.
6.4 Benefits of the Stage 2 DBPR: Reduced Incidence of Adverse Effects
6.4.1 Reduced Incidence of Bladder Cancer Cases
This section presents EPA's estimates of the expected reduction in the incidence rate of new
bladder cancer cases as a result of the Stage 2 DBPR. The methodology used to obtain these estimates is
also discussed in this section. Additional details on the methodology are provided in Appendix E. Also,
Section 6.2.1 presents information on the annual incidence of new bladder cancer cases attributable to all
sources and, in particular, to DBFs prior to Stage 1, that is used as a key input to the estimation of the
reduction in annual cases due to the Stage 2 DBPR. The estimates of Pre-Stage 1 bladder cancer cases
attributable to DBFs are presented in Section 6.2.1 for PAR values derived from three data sources: five
studies used for Stage 1 and Stage 2 Proposal, Villanueva et al. (2003), and Villanueva et al. (2004).
Similarly, this section provides separate estimates of the avoidable cases based on the attributable cases
estimated from those three sources.
Several key assumptions underlie the calculation of bladder cancer cases avoided by the Stage 2
DBPR. The most important one is that a causal relationship exists between exposure to chlorinated
surface water and bladder cancer. However, EPA and the international bodies (e.g. WHO) that classify
risk recognize that such causality has not yet been established.
Other important assumptions regarding the bladder cancer risk from DBFs, and the risk reduction
from lower DBP levels, are:
• That there is no threshold below which there is no risk,
That the risk is linearly related to DBP exposure levels resulting from the range of DBP
levels in drinking water, and
Final Economic Analysis for the Stage 2 DBPR 6-65 December 2005
-------
• That reduction in bladder cancer risks can be estimated from reduction in levels of TTHMs
and HAASs acting as indicators of chlorination DBFs in drinking water.
Taken together, these assumptions provide the basis for calculating the reductions in expected
annual cases of bladder cancer in the population exposed to DBFs in drinking water resulting from the
Stage 2 DBPR. So, for example, if the annual incidence of bladder cancer attributable to DBFs remaining
after Stage 1 is X cases, and the Stage 2 DBPR is determined to reduce the overall average concentrations
of TTHMs or HAASs by 5 percent, then it would be estimated that, over time, there would be 0.05X
fewer new bladder cancer cases occurring each year as a result of the Stage 2 rule.
The number of cases avoided (and the resulting monetized benefits discussed in subsequent
sections) was calculated using TTHM and HAAS as indicators for exposure to all chlorination DBFs (see
Section 6.3.3.2 for discussion of the use of TTHM and HAAS as indicators). However, for analyses
presented in the rest of this chapter, only the results of calculations using TTHM are presented, to
simplify the discussion. Benefits calculated using TTHM as an indicator are similar to those calculated
using HAAS. Detailed results for all analyses using both TTHM and HAAS as indicators are presented in
Appendices E and F.
Another key set of assumptions used to estimate the reduction of bladder cancer incidence relates
to when the expected reductions in new cancer cases begin to occur. Individual cancer risks at any point
in time generally represent lifetime exposure levels and not just current or very recent exposure levels.
Therefore, it would not be appropriate to assume that individuals exposed to some level of DBFs for a
substantial portion of their lifetime would immediately attain a reduced risk of bladder cancer when the
DBF levels in their water system are reduced as a result of compliance with the Stage 2 DBPR. A
transition from pre-Stage 2 risks to the post-Stage 2 risks—referred to here as the "cessation lag"—has
therefore been included in the calculation of cancer cases avoided each year to account for this factor.
Three subsections describing the estimation of cancer cases avoided follow. In the first, the
annual cancer cases avoided are calculated without taking the cessation lag transition into account. This
establishes the "annual cases ultimately avoidable" that will be achieved once the full effect of the
reduced DBF exposure is realized. The second subsection takes into account the effect of the cessation
lag transition period between current and post-regulatory risk levels. A final subsection further accounts
for the timing of the avoided cancer cases by considering the implementation schedule of the rule (that is,
not all affected systems will implement the rule simultaneously).
As noted previously, EPA has developed three approaches to estimating the number of bladder
cancer cases attributable to DBFs (five studies; Villanueva et al. (2003); and Villanueva et al. (2004)).
This section presents the estimates of the bladder cancer cases ultimately avoidable using all three
methods for estimating the baseline population attributable risk (PAR). However, for the sake of
simplicity, one method for estimating PAR based on Villanueva et al. (2003) is carried through the full
benefits analysis to account for cessation lag, implementation schedule, and monetization. A perspective
on how using one of the other two approaches would impact the benefits is provided in Section 6.5.
6.4.1.1 Annual Cancer Cases Ultimately Avoidable
Once the Stage 2 DBPR has been fully implemented, the incidence of bladder cancer cases
annually is anticipated to decline to a new, lower value representing the lower average DBF exposure
levels. That new value will be achieved over time as lifetime risks for individuals who currently consume
water at the high, pre-Stage 2 DBF levels become more influenced by the lower, post-Stage 2 levels.
Over the long-term, the lower incidence of new bladder cancer cases will represent the difference in the
risk of new generations of individuals who are exposed largely or solely to the post-Stage 2 levels for
their lifetimes instead of the higher pre-Stage 2 levels.
Final Economic Analysis for the Stage 2 DBPR 6-66 December 2005
-------
This long-term, steady-state difference between the annual bladder cancer cases attributable to
DBFs before and after the Stage 2 rule is referred to as the "annual cases ultimately avoidable." Note that
this term "ultimately avoidable" is different from the "total attributable" cases. The total attributable
refers to all the annual bladder cancer cases due to DBFs as presented in Section 6.2; the cases ultimately
avoidable refers to that portion of the total attributable cases that a specific reduction in average DBF
concentrations would eliminate.
To calculate the post-Stage 2 ultimately-avoidable cancer cases, it is necessary to begin with the
pre-Stage 1 cancer incidence from all causes, determine the total cases attributable to DBFs, determine the
cases ultimately avoidable by Stage 1, and then determine the cases ultimately avoidable by Stage 2. This
is done as follows:
Estimate of the total cases attributable to DBFs (under pre-Stage 1 conditions) in drinking
water by applying the age-based PAR values to each of the age-based cases per year
attributable to all causes.
• Estimate the maximum number of those total attributable cases that are avoided by the Stage
1 rule based on the percent reduction in average DBF levels due to Stage 1.
• Subtract the maximum number avoidable by Stage 1 from the total pre-Stage 1 attributable
cases to obtain the post-Stage 1 (pre-Stage 2) attributable cases.
• Estimate the maximum number of the remaining pre-Stage 2 attributable cases that are
avoided by the Stage 2 rule based on the percent reduction in average DBF levels due to
Stage 2.
As discussed in Section 6.2.1, EPA has developed three approaches for estimating the total
attributable pre-Stage 1 cases:
• Using the range of Population Attributable Risk (PAR) values derived from consideration of
5 individual epidemiology studies used for the Stage 1 EA and the Stage 2 proposal EA
(yields a pre-Stage 1 range of best estimates for PAR of 2% to 17%).
Using the OR of 1.2 from the Villanueva et al. (2003) meta-analysis that reflects both sexes,
ever exposed population from the studies considered (yields a pre-Stage 1 best estimate for
PAR of-16%)
• Using the Villanueva et al. (2004) pooled data analysis to develop a dose-response
relationship for OR as a function of Average TTHM. The dose-response relationship was
modeled as linear with an intercept of OR = 1.0 at TTHM exposure level = 0 (yields a pre-
Stage 1 best estimate for PAR of-17%)
These three approaches were then used to gauge the percentage of cases attributable to DBP
exposure (i.e., PAR). Taken together, the three approaches provide a reasonable estimate of the range of
potential risks. EPA has long recognized that while the several epidemiology studies described in this
chapter indicate a potential association between exposure to DBFs in drinking water and bladder cancer
incidence, uncertainty remains with respect to quantifying the number of new bladder cases that occur
each year that can be attributed to that exposure. EPA notes that existing epidemiological evidence has
not conclusively established causality between DBP exposure and any health risk endpoints, so the lower
bound of potential bladder cancer cases may be as low as zero.
Two basic methodologies for using the epidemiology data are represented in the three
approaches. The first is to consider multiple studies separately rather than combining the information into
Final Economic Analysis for the Stage 2 DBPR 6-67 December 2005
-------
a single estimate of the attributable risk. The second is to combine the information provided by multiple
epidemiology studies using either a meta-analysis or a pooled data analysis. Each methodology has
advantages and disadvantages.
One advantage to keeping estimates of individual studies separate and presenting them as a full
range of plausible results is that an explicit depiction of the extent of uncertainty that exists in the
quantitative risk estimate is retained. EPA chose to consider studies separately in the economic analyses
for both the Stage 1 DBF rule and the proposal for the Stage 2 DBF rule. EPA relied upon a range of risk
estimates derived separately from 5 key studies that were published in the 1980's and 1990's. The
individual estimates of the fraction of bladder cancer cases attributable to DBF exposure (or more
specifically to chlorinated water exposure) obtained from each of these five studies covered a wide range:
2% to 17%. Further, as EPA noted, consideration of uncertainty for each of the individual estimates leads
a wider range of values and, on the low end, includes the possibility of 0%.
One criterion to consider when deciding whether to combine multiple studies is the heterogeneity
of the data. In developing the Stage 1 rule, EPA evaluated two meta-analyses available at that time
(Poole et al. 1997, Morris et al. 1992) and concluded that the existing studies were too heterogeneous to
be combined in any way.
Meta-analyses and pooled data analyses are two approaches that are used to combine the
information provided by multiple epidemiology studies. In a meta-analysis, the measures of an effect size
obtained in the individual studies (such as the OR) are weighted, typically by the inverse of the variance
of the effect size, and the weighted values combined to obtain the overall estimate of that effect. In a
pooled data analysis, the underlying data of the multiple studies are combined together, typically without
weighting, and an estimate of the effect is made from the combined data as though it were obtained from
a single study.
Meta-analysis is more commonly used for combining multiple epidemiology studies than is
pooled data analysis. If heterogeneity is not properly controlled for across the studies used, pooled data
analysis can be subject to outcomes that are greater, less, and often opposite that of the outcomes
observed in the individual studies (Bravata and Olkin 2001). Although the results of meta-analysis can
also be affected by heterogeneity across the studies used, it is not as subject to these same effects.
Meta-analysis can also combine data by weighting certain studies more than others, while pooled data
analysis cannot do this. However, whereas meta-analysis is limited to consideration of the specific effect
measures studied by the author's of the underlying studies, pooled data analysis can provide an
opportunity to evaluate an effect that was not specifically considered in some or all of the underlying
studies.
EPA determined that the meta-analysis published by Villanueva et al. (2003) and the pooled data
analysis published by Villanueva et al. (2004), both of which combine the results of multiple select
studies, offer reasonable approaches to arriving at a single, overall estimate of attributable risk while still
retaining an appropriate characterization of the uncertainty in that risk estimate.
The Villanueva et al. (2003) meta-analysis, which considered four of five of the same studies that
EPA has used historically for its PAR analyses in addition to two other lower weighted studies, obtained
results that are consistent with the five study estimates. The meta-analysis found a relationship between
duration of exposure to DBFs (or chlorinated water) and risk of bladder cancer, which EPA used to
inform the relationship between exposure and risk. With this approach to estimating risk, EPA assumes
that the exposure of the study populations is characteristic of the national pre-Stage 1 exposure without
knowing the exposure levels explicitly.
The Villanueva et al. (2004) pooled data analysis produced results that are consistent with the
other approaches. The Villanueva et al. (2004) paper provided a dose response relationship between OR
Final Economic Analysis for the Stage 2 DBPR 6-68 December 2005
-------
and TTHM concentrations that allowed EPA to estimate PAR values based specifically on the estimated
average concentrations of TTHMs before and after implementation of the Stage 2 rule, a unique feature
not possible with the other two approaches. A variety of methods, including modeling, were used to
estimate TTHM concentrations. In using the Villanueva et al. (2004) analysis to estimate risk, EPA
assumes that these estimated exposures represent the exposure of the study populations and that the study
population exposures are characteristic of the national pre-Stage 1 exposure. In addition, the Villanueva
et al. (2004) paper used different studies, one of which is unpublished, than the other approaches. In
using the analysis, EPA assumes that the relationship found between exposure and risk is valid for the US
population although the study populations in the pooled analysis are from Italy, Canada, France, and
Finland as well as the U.S.
Exhibit 6.20 summarizes the estimates of ultimately avoidable cases based on these three sets of
PAR value estimates. The details for the calculations resulting in the estimates provided in Exhibit 6.20
are presented in Appendix E, Section E.4.2.1. It is important to note that in quantifying the reduction in
cases, EPA assumed a linear relationship between average DBP concentration and bladder cancer risks.
Because of this, EPA considers these estimates to be an upper bound on the annual reduction in bladder
cancer cases due to the rule.
There are two major sources of uncertainty reflected in the estimates presented in Exhibit 6.20 for
each of the three PAR estimate sources.
First, each shows three "rows" indicating the uncertainty in the percent DBP reductions that are
predicted to occur between Stage 1 and Stage 2. These three estimates provided for each of the PAR
estimate sources correspond to the Best Estimate of the Stage 2 reduction in DBFs, along with the lower
and upper 95 percent CI bounds on those percent DBP reductions.
The second source of uncertainty reflected in each estimate corresponds to uncertainty in the OR
values provided in each of the data sets and the PAR values derived from them. The low and high
estimates shown in each row correspond to the lower and upper 95 percent confidence intervals on the
OR/PAR estimates, with the best estimates reflected in the numbers circled between them. Note that for
the five studies used for Stage 1 and the Stage 2 proposal, the "best estimates" include numbers based on
both the 2 percent and 17 percent PAR values that underlie these calculations.
Focusing on the best estimate of percent DBP reductions, the range of estimates of Stage 2 annual
bladder cancer cases ultimately avoidable derived from the "five studies" set of PARs extends from 0
cases to 1,060 cases per year, with best estimates of 64 (for the 2 percent PAR estimate) and 546 for the
17 percent PAR estimate. For the estimates derived from the Villanueva et al. (2003) PAR values, the
annual cases ultimately avoidable for the best estimate of percent DBP reductions ranges from 275 to 874,
with a best estimate of 506. For the estimates derived from the Villanueva et al. (2004) PAR values, the
annual cases ultimately avoidable for the best estimate of percent DBP reductions ranges from 80 to
1,064, with a best estimate of 550.
As noted previously, the estimates of ultimately avoidable cases from Stage 2 shown here set an
upper bound on the annual reduction in bladder cancer cases due to the DBP reductions from this rule.
As a result of cessation lag and the phasing in of reductions over time in accordance with the
implementation schedule, these annual reductions are not expected to be realized for a substantial number
of years after the rule is promulgated. The sections that follow account for both of these factors in
estimating the annual benefits of the Stage 2 rule.
For the calculation of the annual benefits that include consideration of cessation lag,
implementation schedule and monetization, EPA is using as its starting point the annual cases ultimately
avoidable derived using the OR and PAR values obtained from the Villanueva et al. (2003) study. The
estimates of total attributable cases and cases ultimately avoidable for Stage 2 based on Villanueva et al.
Final Economic Analysis for the Stage 2 DBPR 6-69 December 2005
-------
(2003) fall in the middle of all of the estimates and capture a substantial portion of the overall range
reflecting the uncertainty in both the underlying OR and PAR values as well as the range of uncertainty in
DBF reductions for Stage 2.
It is important to note that in running the full benefits simulation model using the Villanueva et
al. (2003) inputs for PAR and incorporating all of the uncertainties for DBF concentration changes, the
mean estimate of annual cases ultimately avoidable produced by the full model is 581 (CB=234-1,083) as
compared to the "best estimate" of 506 (CB=275-874) calculated above.
Final Economic Analysis for the Stage 2 DBPR 6-70 December 2005
-------
Exhibit 6.20 Comparison of Range of Estimates of Stage 2 Cases Ultimately Avoidable
for Three PAR Approaches and DBP Reductions
Approach 1
"5 Studies"
(PAR = 2
Approach 2
Villanueva etal.
(2003)
(PAR= 15.7%)
Approach 3
Villanueva et al.
(2004)
(PAR= 17.1%)
DBP Red. = 4.5%
(lower CB)
DBP Red. = 7.8%
(best estimate)
DBP Red. = 11.2%
(upper CB)
DBP Red. = 4.5%
(lower CB)
DBP Red. = 7.8%
(best estimate)
DBP Red. = 11.2%
(upper CB)
DBP Red. = 4.5%
(lower CB)
DBP Red. = 7.8%
(best estimate)
DBP Red. = 11.2%
(upper CB)
37
317
615
64
546
1,060
92
782
1,518
1 59 293
507
275 506
874
393
724
1 ,252
47
319
617
80
550
1,064
115
787
1,523
0
1,000
2,000
Annual Cases Ultimately Avoidable
Abbreviation: PAR = Population Attributable Risk (values shown are best estimates). CB = Confidence Bound
Notes: Estimated annual cases ultimately avoidable are based on predicted DBP reduction from Stage 1 to Stage 2. Results shown assume that percent reduction in average TTHM
concentrations is an indicator of percent reduction in concentrations of all DBPs. Three contributions to the uncertainty in the estimate of the annual cases ultimately avoidable by Stage 2 are
displayed in this exhibit: (1) uncertainty in the approach used to estimate PAR; (2) uncertainty in the underlying data used to derive the PAR estimates for each approach, represented by the
95 percent confidence intervals displayed in each horizontal bar; and (3) uncertainty in the percent reduction in the national average DBP levels achieved by Stage 2 is represented by the lower
90 percent CB, best estimate, and upper 90 percent CB values shown for each approach. For Approach 1, the hatched boxes represent the 2 percent to 17 percent range of best estimates
from the five separate studies considered.in the Odds Ratios underlying the PAR values. The estimates in boxes are the overall mean (best) estimates for each approach.
Final Economic Analysis for the Stage 2 DBPR
6-71
December 2005
-------
6.4.1.2 Annual Cancer Cases Avoided Accounting for Cessation Lag
Recently, the Arsenic Rule Benefits Review Panel (ARBRP) of EPA's Science Advisory Board
(SAB) addressed cessation lag in detail and provided guidance to account for the transition period between
higher and lower steady-state risks (USEPA 200Id). The ARBRP coined the term "cessation-lag" to
emphasize the focus on the timing of the attenuation of risk after reduction in exposures. They did this to
avoid confusion with the more traditional term of "latency," which represents the time period from when
initial exposure occurs to when an increase in risk from a carcinogen occurs.3
Although the focus of the cessation lag discussion in the SAB review was on reducing levels of
arsenic in drinking water, much of their consideration of this issue had more general applications beyond
the arsenic issue they were considering at that time. In particular, the SAB noted that:
• The same model should be used to estimate the time pattern of exposure and response as is
used to estimate the potency of the carcinogen.
• If possible, information about the mechanism by which cancer occurs should be used in
estimating the cessation lag (noting that late-stage mechanisms in cancer formation imply a
shorter cessation lag than early-stage mechanisms). The cessation lag tends to be shorter (i.e.
the curve steeper) in late-stage mechanisms such as cancer because a small reduction in
exposure leads to a large decrease in risk.
• If specific data are not available for characterizing the cessation lag, an upper bound for
benefits can be provided based on the assumption of immediately attaining steady-state results.
In the absence of specific cessation lag data, other models should be considered to examine the
influence of the lag.
Following the release of the SAB's report, EPA began to explore approaches for including the
cessation lag in risk reduction models and for calculating benefits for the arsenic regulation. EPA
recognized, however, that the concept of cessation lag may be applicable not only to arsenic, but also to
other drinking water contaminants having a cancer endpoint.
In response to the SAB cessation lag recommendations, EPA has:
• Conducted a study that considered peer review comments and resulted in the 2003 final report
Arsenic in Drinking Water: Cessation Lag Model (USEPA 2003r).
• Presented a poster of the above report at the Society of Toxicology Conference (Schulman et
al. 2004).
Initiated development of general criteria for incorporating cessation lag modeling in benefits
analyses for other drinking water regulations.
In the effort to develop a cessation lag model specific to DBFs, EPA reviewed the available
epidemiological literature for information relating to the timing of exposure and response, but could not
3 The SAB included the following statement in its report on arsenic, to emphasize this difference: "An
important point is that the time to benefits from reducing arsenic in drinking water may not equal the estimated time
since first exposure to an adverse effect. A good example is cigarette smoking: the latency between initiation of
exposure and an increase in lung cancer risk is approximately 20 years. However, after cessation of exposure, risk
for lung cancer begins to decline rather quickly. A benefits analysis of smoking cessation programs based on the
observed latency would greatly underestimate the actual benefits."
Final Economic Analysis for the Stage 2 DBPR 6-72 December 2005
-------
identify any studies that were adequate, alone or in combination, to support a model specifically for DBFs
in drinking water. Thus, in keeping with the SAB recommendation to consider other models in the
absence of substance-specific cessation lag information, EPA explored the use of information on other
carcinogens that could be used as an indicator to characterize the influence of cessation lag in calculating
benefits. EPA investigated three different models and determined that using all three as separate
alternatives would better characterize the uncertainty than selecting one or combining the three together
into a single cessation lag model.
The carcinogen for which the most extensive database was available for characterizing cessation
lag was for cigarette smoking. EPA examined several extensive epidemiological studies on the risks of
adverse health effects, including lung cancer and bladder cancer, for smokers and former smokers. In
addition, EPA included data for arsenic in drinking water and bladder cancer. The three studies used, and
the cancer end-point and risk factor they each consider, are:
• Hrubec and McLaughlin (1997a): Smoking and Lung Cancer
• Hartge et al. (1987): Smoking and Bladder Cancer
• Chen and Gibb (2003): Arsenic and Bladder Cancer
Each of these studies provides information on how the cancer risk for individuals having some
high level of exposure to the risk factor for a substantial portion of their lives changes over time toward the
risk observed for other individuals who have experienced some lower level of exposure. The first two
involve a change from smoking to non-smoking (complete cessation) while the third involves a change
from a high arsenic exposure level of 50 i-ig/L in drinking water to a lower, but non-zero, level of 10 i-ig/L.
The Hrubec and McLaughlin (1997a) report is a comprehensive study involving a 26-year follow-up of
almost 300,000 U.S. male veterans. Hartge et al. (1987) used data from 8,764 subjects in the National
Bladder Cancer Study and their cigarette smoking histories. In addition, EPA selected Chen and Gibb
(2003) to develop a cessation lag model based on arsenic exposure in drinking water and bladder cancer.
Aside from chloroform, no mode of action has been established either for specific DBFs or for
chlorinated water in general related to bladder cancer. Thus, EPA assumes that the mode of action may be
adequately represented by the mixed initiator and promoter aspects of the cigarette smoking model. As
discussed in the SAB report and the EPA Cessation Lag report (USEPA 200Id and USEPA 2003r),
carcinogens that act solely or primarily as initiators would tend to show a longer cessation lag (lower rate
of risk reduction following reductions in exposure) than carcinogens that act solely or primarily as
promoters. The available information on tobacco smoke and lung cancer suggests that it involves both
initiators and promoters, so the cessation lag derived from smoking data is expected to represent the
combined influence of these different mechanisms.
The smoking/bladder cancer data is appropriate to use because the target organ is the same.
"Portal of entry" differences are particularly important in differentiating between smoking/lung cancer and
DBPs/bladder cancer. The route of exposure for smoking/bladder cancer is similar to DBP ingestion in
that there is expected to be a first pass through the liver followed by entry into the bloodstream prior to
reaching the bladder.
In the case of dermal exposure, which is considered a major route, DBFs enter the bloodstream
directly and can reach the bladder without that "first pass" through the liver. This could argue for a
similarity with the smoking/lung cancer model because it involves a more direct "portal of entry".
While there are some reasons to favor the smoking/bladder cancer data, the largest area of
uncertainty is the appropriateness of using any smoking/cancer data to derive a cessation lag model for
DBFs. In light of the overarching uncertainties in the use of smoking data, developing separate cessation
lag models based on smoking/lung cancer data and smoking/bladder cancer data is a reasonable approach.
EPA has therefore judged that these two models, plus the arsenic/bladder cancer model, are equally
Final Economic Analysis for the Stage 2 DBPR 6-73 December 2005
-------
plausible for characterizing cessation lag in order to estimate benefits attributable to DBF reduction. The
equations representing the three cessation lag models as well as details regarding their derivation and
implementation can be found in Appendix E.
Another important consideration is that two out of three cessation lag models (those based on
smoking) involve complete cessation of exposure, whereas in the case of DBFs, the exposure is only
reduced. In some water systems the reduction is only 10 percent, whereas in others it may be as high as 60
percent, with an average of approximately 30 percent. This moderate reduction in exposure may prevent
full DNA repair, which some scientists interpret as the basis for the short cessation lag associated with full
cessation of smoking. However, the third cessation lag model considers only a partial reduction in
exposure.
Exhibit 6.2 la illustrates the effect of the cessation lag model. The results from the cessation lag
models show that the majority of the potential cases avoided occur within the first fifteen years after initial
reduced exposure to DBFs. For example, fifteen years after the exposure reduction has occurred, the
annual cases avoided will be 489 for the smoking / lung cancer cessation lag model, 329 for the smoking /
bladder cancer cessation lag model, and 534 cases for the arsenic / bladder cancer cessation lag model.
These represent approximately 84%, 57%, and 92%, respectively, of the estimated 581 annual cases
ultimately avoidable by the Stage 2 DBPR based on the Villanueva et al. (2003) PAR estimates. Note that
the estimate of 581 annual cases ultimately avoidable differs slightly from the 506 cases ultimately
avoidable shown for the Villanueva et al. (2003) PAR estimates in the preceding section. The 581 number
is generated by the full Monte Carlo simulation benefits model taking uncertainty in the PAR and DBP
reductions into account, whereas the 506 number is a direct calculation based on the individual "best"
estimates of PAR and DBP reduction values without accounting for those uncertainties.
6.4.1.3 Adjustments in Annual Cancer Cases Avoided to Account for the Rule Implementation
Schedule
In addition to the delay in reaching a steady-state level of risk reduction as a result of cessation
lag, there is a delay in attaining maximum exposure reduction across the entire affected population that
results from the Stage 2 DBPR implementation schedule. For example, large surface water PWSs have 6
years from rule promulgation to meet the new Stage 2 MCLs, with an additional 2-year extension possible
for capital improvements. For the benefit and cost analysis, EPA estimates that some percentage will
make treatment technology changes in year 4, some will make treatment technology changes in year 5, and
so on. Appendix D shows the assumptions regarding the schedule for installation of treatment
technologies to meet Stage 2 DBPR requirements. In general, EPA assumes that a fairly constant
increment of systems will complete installation of new treatment technologies each year, with the last
systems installing treatment by 2016.
The delay in exposure reduction resulting from the rule implementation schedule is incorporated
into the benefits model by adjusting the cases avoided for the given year. For example, if 10 percent of
systems install treatment equipment (and start realizing reductions in cancer cases) in year 1, only that
portion of the cases will begin the cessation lag equilibrium process in that year. Exhibit 6.2 Ib provides
ths estimated cases avoided each of the first 25 years after the rule promulgation considering both
cessation lag and implementation schedule.
EPA analyses of available data indicate that 26 percent of bladder cancers are fatal and 74 percent
are non-fatal (USEPA 1999a). Annual cases avoided were apportioned to each category proportionately.
Final Economic Analysis for the Stage 2 DBPR 6-74 December 2005
-------
Exhibit 6.21 a Comparison of Alternative Cessation Lag Models: Estimates of
Annual Cases Avoided by Year Following Exposure Reduction
(TTHM as an indicator)
600
1!
Arsenic / Bladder Cancerr Model
Smoking / Lung Cancer Model
Smoking / Bladder Cancer Model
20
30 40 50 60 70
Years After DBP Exposure Reduction Begins
80
90
100
Final Economic Analysis for the Stage 2 DBPR
6-75
December 2005
-------
Exhibit 6.21 b Estimates of Annual Cases Avoided by Year For Three Cessation
Lag Models, Considering Rule Implementation Schedule (TTHM as an Indicator)
Note.
Year
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Smoking/Lung
Cancer Cessation
Lag Model
Total
0
0
0
0
0
24
62
111
170
220
265
305
341
371
396
416
433
448
460
471
481
489
496
503
509
Percent1
0%
0%
0%
0%
0%
4%
11%
19%
29%
38%
46%
53%
59%
64%
68%
72%
75%
77%
79%
81%
83%
84%
86%
87%
88%
Smoking/Bladder
Cancer Cessation
Lag Model
Total
0
0
0
0
0
23
54
90
132
161
184
204
221
237
251
265
278
289
301
311
321
330
339
347
355
Percent1
0%
0%
0%
0%
0%
4%
9%
16%
23%
28%
32%
35%
38%
41%
43%
46%
48%
50%
52%
54%
55%
57%
59%
60%
61%
Arsenic /Bladder
Cancer Cessation
Lag Model
Total
0
0
0
0
0
45
110
187
275
334
379
412
438
458
475
488
499
509
516
523
528
533
537
541
544
Percent1
0%
0%
0%
0%
0%
8%
19%
32%
48%
58%
65%
71%
76%
79%
82%
84%
86%
88%
89%
90%
91%
92%
93%
93%
94%
1Rercent of annual cases ultimately avoidable achieved during each of the first 25 years. The benefits model
estimates 581 (90% CB = 234 - 1,083) annual cases ultimatefy avoidable using the Villanueva et al. (2003)
PAR inputs and including uncertainty in these and DBP reductions.
EPA recognizes that cases may be as low as zero since causality has not yet been established between
exposure to chlorinated water and bladder cancer.
Source: Exhibits E.38a, E.38e and E.38L
6.4.2 Reduced Incidence of Reproductive and Developmental Effects
As discussed earlier, both epidemiological and toxicological evidence suggest the potential for
increased health risk for pregnant women and their fetuses exposed to DBFs in drinking water. Although
the levels of DBFs associated with specific potential adverse reproductive and developmental effects is not
known and no causal link has been established, EPA believes that lowering the overall levels of DBFs in
distribution systems and in particular by reducing the incidence of peak levels is important from a public
health perspective.
EPA believes that the current scientific knowledge on reproductive and developmental health
effects is not strong enough to quantify risk in the primary benefits analysis. However, an illustrative
calculation considering the range of possible benefits for one specific effect in this category—avoided
cases of fetal loss—is presented in Section 6.8. The discussion in Section 6.2 and the illustrative
calculation in Section 6.8 suggest that the benefits from reduced DBP exposure in terms of both avoided
incidence of potential reproductive and developmental effects and in terms of the potential monetized
value of those avoided cases could be significant.
Final Economic Analysis for the Stage 2 DBPR
6-76
December 2005
-------
6.4.3 Other Health-Related Benefits
The scientific literature indicates that exposure to DBFs may be related to health effects other than
reproductive, developmental, and bladder cancer effects. Some studies have indicated an association
between consumption of chlorinated drinking water and colon and rectal cancer, while other studies have
shown no association. Since 1998, several new studies have been published that contribute to the weight
of evidence relating DBF exposure with colon and rectal cancer. As TTHM and HAAS levels are reduced
under the Stage 2 DBPR, other potentially carcinogenic chlorination DBFs (both known and unknown)
will be reduced as well. These collateral effects may further reduce the number of colon and rectal cancer
cases. Both toxicology and epidemiology studies indicate that other cancers may be associated with DBF
exposure but currently there is not enough data to quantify or monetize these risks. However, EPA
believes that the association between exposure to DBFs and colon and rectal cancers is possibly
significant, so an analysis of benefits is presented as a sensitivity analysis (see Section 6.7).
6.4.4 Non-Health-Related Benefits
The Stage 2 DBPR may increase consumer confidence in the quality of drinking water. Drinking
water consumers may be willing to pay a premium for regulatory action if it reduces their risk of becoming
ill. Consumers' WTP depends on several factors, including their degree of risk aversion, their perceptions
about drinking water quality, and the expected probability and severity of potential human health effects
associated with DBFs.
Most people who switch to bottled water or use filtration devices do so because of taste and odor
problems and health-related concerns. Chlorine dioxide, ozone, and chloramines have historically been
used to address taste and odor. To the extent that the Stage 2 DBPR changes perceptions of the health
risks associated with drinking water and improves taste and odor, it may reduce actions such as buying
bottled water or installing filtration devices. Any resulting cost savings would be a regulatory benefit.
As PWSs move away from conventional treatment to more advanced treatment technologies, other
non-health benefits are anticipated besides better-tasting and smelling water. Installation of certain
advanced treatment technologies can remove many contaminants in addition to those specifically targeted
by the Stage 2 DBPR, including those that EPA may regulate in the future. For example, membrane
technology (depending on pore size), can be used to lower DBP formation, but it can also remove many
other contaminants (e.g. bacteria and protozoans) that EPA is in the process of regulating. GAC lowers
nutrient availability for bacterial growth, produces a biologically more stable finished water, and facilitates
management of water quality in the distribution system. Since GAC also removes synthetic organic
chemicals( SOCs), it provides additional protection from exposure to chemicals associated with accidental
spills or environmental runoff. Removal of any contaminants that may face regulation could result in
future cost savings to a water system.
6.4.5 Potential Increases in Health Risks
It is important to maintain a balance between the risks from DBFs and those from microbial
pathogens in drinking water. The Microbial-Disinfectants/Disinfection Byproducts (M-DBP) Advisory
Committee considered the impact of DBP control on microbial protection when they recommended the
MCLs in the Stage 2 DBPR. For example, as described in Chapter 4 of this EA, the M-DBP Advisory
Committee debated whether the bromate MCL should be lowered. The Stage 1 DBPR set the MCL for
bromate at 10 (ig/L, partly because that was the limit of EPA's measuring capability at that time. Methods
now exist to measure lower concentrations of bromate, which would allow a lower limit to be set.
However, the committee was concerned that a lower bromate MCL might discourage systems from
switching to (or continuing to use) ozone to increase microbial protection. Unlike chlorine, ozone can
Final Economic Analysis for the Stage 2 DBPR 6-77 December 2005
-------
inactivate Cryptosporidium, a focus of the Long Term 2 Enhanced Surface Water Treatment Rule
(LT2ESWTR). Therefore, to encourage the use of ozone, M-DBP Advisory Committee recommended that
EPA not change the bromate MCL.
Along with the reduction in chlorination DBFs such as TTHM and HAAS, there may be increases
in other DBFs as systems change from chlorine to other disinfectants. Exhibits 6.22 and 6.23 compare the
SWAT-predicted monthly average and plant-mean average concentrations for chlorite (a potential
byproduct of chlorine dioxide disinfection) and bromate (a potential byproduct of ozone disinfection) for
pre-Stage 2 and post-Stage 2 conditions. These exhibits show that the predicted changes in bromate and
chlorite concentration as a result of the Stage 2 DBPR are expected to be minimal. These changes are
minimal in part because bromate and chlorite MCLs already exist.
Another potential increase in health risks is due to increases in N-nitrosodimethylamine (NDMA),
which is formed during both the chlorination and the chloramination process (see Section 6.2.1.2).
Chapter 5 shows that many systems that do not currently meet the Stage 2 requirements will do so by
switching to chloramines.
In addition, the baseline of disinfecting systems will increase as ground water systems begin
adding disinfection by complying with the Ground Water Rule. In these circumstances, it is assumed that
when a system adds disinfection, it will ensure compliance with the Stage 2 DBPR. Essentially, these
systems will move from a position of little to no risk (non-disinfecting) to a Post-Stage 2 risk. This
increase in risk due to the Ground Water Rule will be small on a national level because of the small
population served by most non-disinfecting systems. More detail on this analysis can be found in
Appendix M.
Final Economic Analysis for the Stage 2 DBPR 6-78 December 2005
-------
Exhibit 6.22a Predicted Chlorite Plant-Mean Concentration for Pre-Stage 2 and
Post-Stage 2
100%
90%
-------
Exhibit 6.23a Predicted Bromate Plant-Mean Concentrations for Pre-Stage 2 and
Post-Stage 2
100%
90%
10%
0%
— Post-Stage 2
^Pre-Stage 2
15
20
25
Bromate (ug/L)
Exhibit 6.23b Predicted Bromate Monthly Average Concentrations for Pre-Stage 2
and Post-Stage 2
— Post-Stage 2
— Pre-Stage 2
o
345
Bromate (ug/L)
Source: DS Average data from SWAT runs 300 and 303 (USEPA 2001 b).
Final Economic Analysis for the Stage 2 DBPR
6-80
December 2005
-------
6.5 Valuation of Health Benefits for the Stage 2 DBPR
Once the benefits of implementing the Stage 2 DBPR have been identified, a monetary value must
be assigned to allow comparison with the costs of the regulation. The following sections draw on the
valuation literature to attribute the most appropriate values to each type of benefit derived from the rule.
Where the available information is not sufficient to quantify monetary benefits, a qualitative discussion of
potential value is presented.
To augment the valuation data from the literature, EPA incorporated into the Stage 2 DBPR
analyses recommendations from reviews of previous regulations. In particular, recommendations made by
EPA's SAB with regard to the benefits analysis of the recently promulgated Arsenic Rule (66 FR 6976,
January 22, 2001) were incorporated into the Stage 2 DBPR analyses as appropriate. Even though these
recommendations were made in the context of the arsenic regulation (USEPA 200 Id), certain
recommendations regarding methodology (e.g., incorporation of cessation lag, calculation and presentation
of uncertainties, etc.) are applicable to other impact analyses, including the Stage 2 DBPR. This section is
organized as follows:
Section 6.5.1 Presents a qualitative discussion of the value of reductions in potential adverse
reproductive developmental health effects derived from the Stage 2 DBPR.
Section 6.5.2 Explains the methodology used for quantitative valuation of reductions in bladder
cancer cases attributable to the Stage 2 DBPR.
Section 6.5.3 Summarizes the total potential benefits attributable to the Stage 2 DBPR from
implementation of the preferred regulatory alternative (which includes the
requirement for the IDSE).
Section 6.5.4 Compares estimated benefits attributable to the preferred regulatory alternative
(which includes a requirement for the IDSE) to those estimated for other
alternatives.
6.5.1 Value of Reductions in Potential Adverse Reproductive and Developmental Health Effects
Potential adverse reproductive and developmental effects may impose a large economic cost on the
nation. Although many miscarriages do not have associated medical costs, some do, with costs ranging
from $5,000 to $11,000 depending on the conditions of the miscarriage and the length of stay in the
hospital (HCUPnet 2000). The full economic benefit of avoiding a miscarriage would also include the
monetary value of forgoing the associated pain, suffering, and loss. Another potential benefit is that the
life of the fetus is saved. Avoiding these costs represents one potential economic benefit of the rule.
Low birth weight also imposes costs on society. A report from The Future of Children (Lewit et
al. 1995) estimates that approximately 40,000 infants die each year as a result of low birth weight and that
$5.4 billion is spent each year on the additional services that low-birth-weight children require for health
care and, eventually, special education and child care. The Lewit et al. (1995) study also estimates that
$5.5 to $6 billion are spent each year caring for low-birth-weight infants and children. It estimates that
such children are almost 50 percent more likely than normal-birth-weight children to require special
education. Additionally, the costs of caring for low-birth-weight infants is increasing as their chances for
survival increase.
Final Economic Analysis for the Stage 2 DBPR 6-81 December 2005
-------
The cost for treating and caring for those with birth defects is high. For example, the lifetime cost
for a case of spina bifida is estimated at $300,000. Estimates for the lifetime cost of a heart defect range
from $100,000 to $400,000 (CDC 1995). These costs account only for the estimated medical,
developmental, and special education services attributed to each case. They do not include the pain and
suffering of the children with these conditions or the emotional strain on their parents and other family
members.
If even a small proportion of fetal losses, low birth weights, premature births, congenital
anomalies, and infertility problems are attributable to DBFs, the associated monetary value of benefits
from the Stage 2 DBPR would still be great. Given the high cost of medical care, the many ways to value a
pregnancy saved, the unknown pain and suffering as a result of potential adverse reproductive and
developmental effects, the high percentage of the U.S. population exposed to DBFs, and the large number
of pregnant women, the Stage 2 DBPR presents a potential for substantial savings to society. Although
uncertainties in the estimation of potentially avoided adverse reproductive and developmental health
effects preclude a definitive evaluation of associated benefits in the primary analysis, EPA has conducted
an illustrative calculation to estimate a range of possible benefits associated with fetal losses (see Section
6.8).
6.5.2 Value of Reductions in Bladder Cancer Cases
EPA analyses of available data indicate that 26 percent of bladder cancers are fatal and 74 percent
are non-fatal (USEPA 1999a). Annual cases avoided were apportioned to each category proportionately.
Valuation data for fatal and non-fatal bladder cancer cases are summarized below. This is
followed by an explanation of how these data are adjusted to current price levels and for income elasticity
effects, allowing proper incorporation into the Stage 2 DBPR benefits model. The next section (6.5.3)
describes how these values are combined with the bladder cancer case reductions to yield the total
estimated benefits resulting from the Stage 2 DBPR.
Value of Avoiding a Fatal Case of Bladder Cancer
For fatal bladder cancer cases, the value of a statistical life (VSL) is used to calculate the value of
benefits. The VSL represents an estimate of the monetary value of reducing risks of premature death from
cancer. The VSL does not represent the value of saving a particular individual's life; it represents the sum
of the values placed on small individual risk reductions across an exposed population. For example, if a
regulation were to reduce the risk of premature death from cancer by 1/1,000,000 for 1 million exposed
individuals, the regulation would "save" one statistical life (1,000,000 x 1/1,000,000). If each of the
1,000,000 people were (on average) willing to pay $5 to achieve their risk reduction of 1/1,000,000
anticipated from the regulation, the VSL would be $5 million ($5 x 1,000,000).
An EPA study characterized the range of possible VSL values as a Weibull distribution with a
mean of $4.8 million (1990 price level), based on 26 studies (USEPA 1997a). This is the value
recommended for use in EPA's Guidelines for Preparing Economic Analyses (USEPA 2000J) and
endorsed by the SAB (USEPA 200 Id). For purposes of the Stage 2 DBPR benefits analysis, the VSL
Weibull distribution (with parameters of location = 0, scale = 5.32, shape = 1.51) was incorporated into the
benefits model using a Monte Carlo simulation. This allows quantification of the uncertainty surrounding
the benefit estimates.
Final Economic Analysis for the Stage 2 DBPR 6-82 December 2005
-------
Value of Avoiding the Morbidity Increment of a Fatal Case of Bladder Cancer
The VSL represents the value of avoiding a premature death. This valuation, however, does not
explicitly take into account the medical costs associated with the period of illness (morbidity increment)
leading up to death. In its review of the Arsenic Rule, the SAB suggested that the appropriate measure to
use in valuing the avoidance of the morbidity increment is the medical cost attributable to a cancer case
(USEPA 200Id). Based on available data, EPA estimates the medical costs for a fatal bladder cancer case
to be $93,927 at 1996 price levels (USEPA 1999a). This cost (updated to 2003 price levels) is applied as a
point estimate to each fatal case of bladder cancer in the benefits model.
Value of Avoiding a Non-fatal Case of Bladder Cancer
For a case of non-fatal bladder cancer, a willingness to pay (WTP) measure is used to estimate the
value a person would place on reducing the risk of a case of non-fatal bladder cancer. It accounts for the
desire to avoid treatment costs, pain and discomfort, productivity losses, and any other adverse
consequences related to contraction of a non-fatal case of bladder cancer. As is the case with VSL
valuation, the cumulative WTP for this risk reduction across an exposed population can be used to
represent the statistical value of avoiding the illness itself. WTP is a more comprehensive measure of the
total value that a person would place on avoiding a cancer case than the much simpler cost of illness (COI)
measure.
A review of the available literature did not reveal any studies that specifically measured the WTP
to avoid the risk of non-fatal bladder cancer. Instead, two surrogate estimates are used: one based on
avoiding a case of curable lymph cancer (lymphoma) and the other based on avoiding a case of chronic
bronchitis4. Results using both WTP estimates are presented throughout the remainder of the analyses.
The WTP to avoid the risk of contracting curable lymphoma is derived from a survey by Magat et
al. (1996) that evaluates the risk-risk trade-off between curable lymphoma and death using a reference
lottery metric. A reference lottery is a methodology that educates survey respondents of the health
consequences of a particular disease (in this case curable lymphoma) and, based on this information,
presents them with choices related to health outcomes. The choices in health outcomes made by the
respondents can be further evaluated to derive quantitative measures of relative risk aversion. Based on
the outcomes of the Magat et al. study, it was determined that the median risk-risk trade-off (relative risk
aversion) for contracting a curable case of lymphoma was equivalent to 58.3 percent of the risk attributed
to reducing the chances of sudden death (i.e., the average person would pay 58.3 percent of what they
would pay to reduce the risk of sudden death to achieve an equal risk reduction for contracting curable
lymphoma). Based on the Magat et al. study results, EPA calculated a WTP distribution for non-fatal
bladder cancer as a percentage of the VSL distribution, resulting in a mean WTP value of $2.8 million
($4.8 million x 58.3 percent) at 1990 price levels (see Appendix F, Section F.I for additional information
on the derivation of this WTP estimate).
The WTP values for avoiding the risk of chronic bronchitis are consistent with those defined for
the Stage 1 DBPR. They are best represented by a lognormal distribution with a mean of $587,500,
standard deviation of $264,826, and a maximum value of $1.5 million at 1998 price values (USEPA
1998a,Viscusietal. 1991).
4 Previous EPA analyses (Stage 1 DBPR and Arsenic Rule) used the WTP value for avoiding a case of
chronic bronchitis for benefits transfer calculations. The SAB review of the Arsenic benefits analysis identified the
curable lymphoma WTP value as another metric that could be used in benefits valuation because ".. .the endpoint
being valued more nearly corresponds to nonfatal bladder cancer..." (USEPA 2001d). The SAB suggested,
however, that calculations using the WTP for chronic bronchitis also be presented. This analysis follows the SAB's
recommendation.
Final Economic Analysis for the Stage 2 DBPR 6-83 December 2005
-------
Although the WTP to avoid curable lymphoma or chronic bronchitis is not a perfect substitute for
the WTP to avoid a case of non-fatal bladder cancer, it is a reasonable value to use in a benefits transfer
methodology. Non-fatal internal cancers, regardless of type, generally present patients with very similar
treatment, health, and long-term quality of life implications, including surgery, radiation or chemotherapy
treatments (with attendant side effects), and generally diminished vitality over the duration of the illness.
In the absence of more specific studies, the WTP values for avoiding a case of curable lymphoma or a case
of chronic bronchitis provide a reasonable, though not definitive, substitute for the value of avoiding non-
fatal bladder cancer.
Updating Price Levels
All valuation parameters must be updated to the same price level so comparisons can be made in
real terms. Values for VSL, WTP, and the morbidity increment used in the model are updated based on
adjustment factors derived from Bureau of Labor Statistics (BLS) consumer price index (CPI) data so that
each represents a year 2003 price level. Exhibit 6.24 presents these updates.
Exhibit 6.24 VSL, WTP, and Morbidity Increment Price Level Updates
Valuation Parameter
Morbidity Increment
VSL
WTP - Non-Fatal Lymphoma
WTP - Chronic Bronchitis
Base
Year
1996
1990
1990
1998
Mean Value
in Base
Year
(Millions)
$ 0.1
$ 4.8
$ 2.8
$ 0.6
CPI Update
Factor
1.30
1.41
1.41
1.13
Values at Year 2003 Price Level
(Millions)
Mean
$ 0.1
$ 7.8
$ 4.4
$ 0.8
5th %tile
N/A
$ 1.2
$ 0.7
$ 0.4
95th %tile
N/A
$ 17.9
$ 10.1
$ 1.4
Note: Morbidity increment value is presented as a point estimate.
Source: Derived from Appendix F (Exhibits F.1a, F.1b, and F.1f).
Adjustments for Real Income Growth and Elasticity
Although the price level (year 2003) is held constant throughout the benefits model, projections of
benefits in future years are subject to income elasticity adjustments. These represent changes in valuation
in relation to changes in real income. For example, if, for every 1 percent increase in real income, a
particular consumer's WTP for a particular item increases by 1 percent, this would be represented by an
income elasticity of 1. For most items, income elasticity values are actually less than 1, indicating that
valuation of most items does not increase as fast as real income levels.
Based on an evaluation of the income elasticity literature, Kleckner and Neuman (2000) identified
published studies from which elasticity values could be derived for both fatal and non-fatal potential health
effects. For fatal cancers, they identified a triangular distribution with a central estimate of 0.40 (low end:
0.08; high end: 1.00) to represent the uncertainty of that income elasticity value. For non-fatal cancers, a
triangular distribution with a central estimate of 0.45 (low end: 0.25; high end: 0.60) best represents the
value. These distributions are used as assumptions in the Monte Carlo simulation to further characterize
uncertainty in benefits estimates.
To apply the income elasticity values in the model, they must be combined with projections of real
income growth over the time frame for analysis. Population and real gross domestic product (GDP)
Final Economic Analysis for the Stage 2 DBPR
6-84
December 2005
-------
projections are combined to calculate per-capita real GDP values.5 Percent changes in these values over
time can then be combined with income elasticity figures to derive a single adjustment factor.6 Given any
two time periods, this factor can be calculated as follows:
Income elasticity adjustment factor = (elj - eI2 -12 - Ij) / (eI2 - elj -12 - Ij)
where: e = income elasticity
Ij = real income (per-capita GDP) in the base year
I2 = real income (per-capita GDP) in the year of analysis
Income elasticity adjustment factors are calculated from the same base year as the values subject to
adjustment. For example, income elasticity factors for fatal cancers are calculated from a 1990 base year
because that is the base year used in the study from which VSL estimates are derived.7 The mean values of
the income adjustment factors calculated for the Stage 2 benefits model range from 1.160 to 1.488 for fatal
cancer valuation and 1.063 to 1.400 for non-fatal valuation over the 25-year analysis time frame
(Appendix F presents detailed spreadsheets of these calculations). The adjusted yearly values for the VSL
and WTP (at a 2003 price level) are then calculated by multiplying the base value of each by the
appropriate income elasticity adjustment factor.8 Exhibit 6.25 presents the results of the income elasticity
adjustments for the 25-year analysis time frame.9
In the Stage 2 benefits analysis the income-adjusted VSL estimates are applied to the year in
which cases have been avoided. An alternative approach supported by some economists, and used in other
EPA analyses, is for the income adjustments to be applied only up to the time that exposures are reduced
rather than over the cessation lag. Because of the shorter time period over which income growth would be
calculated the alternative would result in smaller income adjustment. To use the alternative EPA would
need to link the year cancers are avoided to a specific year of exposure reduction. This cannot be done
with the risk assessment and cessation lag application in the Stage 2 analysis, where estimated cases
avoided are based on a transition from one steady state to another. The VSL income adjustment approach
used in this EA will tend to overstate benefits somewhat relative to the alternative described above. EPA
recognizes this potential bias, but notes that is small in comparison to other uncertainties in valuation, as
well as uncertainties in the risk assessment and estimates of cases avoided.
5 Ideally, income elasticity adjustments would be calculated using real per capita personal income growth.
However, real per capita GDP is used as a proxy for real per capita personal income growth due to lack of
appropriate projections for real personal income growth. Historical data suggests that GDP and personal income
grow at similar rates (i.e., Table B-31 of the 2002 Economic Report of the President shows that both real per capita
GDP and disposable personal income grew at an average annual rate of 2.3 percent between 1959 and 2000).
6 See Appendix A of Kleckner and Neuman (2000) for additional information on the derivation and
application of income elasticity adjustments.
7 The distribution of VSL values used in this EA is derived from a meta-analysis of 26 different VSL
studies, all representing different years price levels. These price levels were updated to a common 1990 price level
as part of the analysis in "The Benefits and Costs of the Clean Air Act, 1970-1990" (USEPA 1997b), from which the
distribution used in this EA is taken.
8 Because the morbidity increment represents a point estimate of direct medical costs, and income elasticity
figures used in this analysis are based on WTP values, income elasticity adjustments were not applied to the
projected morbidity increment values (only the CPI update factor is applied).
9 A 25-year analysis time frame was chosen to represent the period before which most systems would need
to reinvest in capital equipment replacement (a 20-year useful life is assumed for the analysis). Since the benefits, as
derived in this analysis, are a result of installing treatment equipment, this time frame was also applied to benefit
projections.
Final Economic Analysis for the Stage 2 DBPR 6-85 December 2005
-------
Exhibit 6.25 Value of Morbidity Increment, VSL, and WTP by Year, Adjusted for Income Elasticity
Year
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
Fatal Cancer Cases
Morbidity
Increment
Point
Estimate
0.12
0.12
0.12
0.12
0.12
0.12
0.12
0.12
0.12
0.12
0.12
0.12
0.12
0.12
0.12
0.12
0.12
0.12
0.12
0.12
0.12
0.12
0.12
0.12
0.12
VSL
Mean
Value
7.76
7.85
7.95
8.03
8.12
8.22
8.31
8.40
8.49
8.59
8.68
8.78
8.88
8.98
9.08
9.18
9.28
9.38
9.49
9.59
9.70
9.80
9.91
9.87
9.95
Median
Value
6.73
6.81
6.88
6.95
7.03
7.10
7.18
7.25
7.33
7.41
7.48
7.57
7.64
7.72
7.81
7.89
7.97
8.05
8.15
8.23
8.32
8.41
8.49
8.46
8.52
90% Confidence Interval
5th %tile
1.19
1.20
1.21
1.23
1.24
1.25
1.26
1.28
1.29
1.30
1.31
1.32
1.34
1.35
1.36
1.37
1.38
1.40
1.41
1.42
1.43
1.45
1.46
1.46
1.47
95th %tile
17.91
18.12
18.33
18.55
18.76
18.99
19.24
19.43
19.64
19.87
20.13
20.34
20.62
20.86
21.15
21.40
21.67
21.95
22.20
22.45
22.71
22.96
23.28
23.17
23.40
Non-Fatal Cancer Cases
WTP - Non-Fatal Lymphoma
Mean
Value
4.43
4.48
4.53
4.57
4.61
4.66
4.70
4.74
4.79
4.83
4.88
4.92
4.97
5.02
5.06
5.11
5.16
5.20
5.25
5.30
5.34
5.39
5.44
5.42
5.46
Median
Value
3.85
3.89
3.93
3.97
4.00
4.04
4.08
4.12
4.16
4.20
4.24
4.28
4.32
4.36
4.40
4.44
4.48
4.52
4.56
4.60
4.64
4.69
4.73
4.71
4.74
90% Confidence Interval
5th %tile
0.68
0.69
0.70
0.70
0.71
0.71
0.72
0.73
0.74
0.74
0.75
0.76
0.77
0.77
0.78
0.79
0.79
0.80
0.81
0.82
0.82
0.83
0.84
0.84
0.84
95th %tile
10.13
10.24
10.36
10.45
10.57
10.67
10.76
10.85
10.95
11.07
11.16
11.27
11.38
11.49
11.61
11.72
11.82
11.94
12.06
12.17
12.28
12.40
12.52
12.48
12.56
WTP - Bronchitis
Mean
Value
0.80
0.81
0.82
0.82
0.83
0.84
0.85
0.86
0.86
0.87
0.88
0.89
0.90
0.91
0.91
0.92
0.93
0.94
0.95
0.96
0.97
0.98
0.99
0.98
0.99
Median
Value
0.74
0.75
0.75
0.76
0.77
0.78
0.78
0.79
0.80
0.81
0.81
0.82
0.83
0.84
0.85
0.85
0.86
0.87
0.88
0.89
0.89
0.90
0.91
0.91
0.91
90% Confidence Interval
5th %tile
0.36
0.37
0.37
0.37
0.38
0.38
0.38
0.39
0.39
0.40
0.40
0.40
0.41
0.41
0.41
0.42
0.42
0.43
0.43
0.43
0.44
0.44
0.45
0.44
0.45
95th %tile
1.45
1.46
1.48
1.49
1.51
1.52
1.54
1.55
1.56
1.58
1.60
1.61
1.63
1.64
1.66
1.67
1.69
1.71
1.72
1.74
1.76
1.78
1.79
1.79
1.80
Notes: All values in millions of year 2003 dollars. Detail may not add exactly due to independent rounding.
Source: Exhibit F.1f
Final Economic Analysis for the Stage 2 DBPR
6-86
December 2005
-------
6.5.3 Value of Benefits Resulting from the Stage 2 DBPR for the Preferred Alternative
To assess the total value of benefits resulting from the Stage 2 DBPR, both the qualitative and
quantitative benefits must be considered. Although information is not sufficient to quantify the value of
preventing potential adverse reproductive and developmental health effects in the primary benefits
analysis, the number of cases avoided and associated value could be significant (see Section 6.8).
Likewise, the value of other health and non-health benefits could be substantial, for example reduction in
risk of other cancers and reductions in other regulated contaminants. Thus, the primary, quantitative
benefits analysis is a conservative estimate of the total benefits of this regulation.
To calculate the total value of benefits derived from reductions in bladder cancer cases as a result
of the Stage 2 DBPR, a stream of monetary benefits is calculated by combining the annual cases avoided
(Exhibit 6.21) with valuation inputs (Exhibit 6.25) using a Monte Carlo simulation. The Monte Carlo
simulation allows the characterization of uncertainty around the modeling outputs based on the uncertainty
in the various inputs. The benefits model uses distributions of VSL, WTP, and income elasticity values to
attribute monetary values (with uncertainty bounds) to the mean number of bladder cancer cases avoided.
The values for cancer cases avoided for the three cessation lag models and for both WTP estimates (for
curable lymphoma and chronic bronchitis) were calculated and carried through the Stage 2 DBPR benefits
model. The results for fatal, non-fatal, and total benefits are presented in Exhibits 6.26 (note that the
alternative analysis assumptions, e.g. Arsenic/Bladder Cancer Cessation Lag model and WTP for
Lymphoma, are shown in the table headings).
Calculating and Discounting the Stream of Benefits
To allow comparison of future streams of costs and benefits, it is common practice to adjust both
streams to a present value (PV) using a discount rate. This process takes into account the time preference
that society places on expenditures and benefits and allows comparison of cost and benefit streams that
vary over a given time period.10 A present value for any future period can be calculated using the
following equation:
PV = V(t)/(l+R)t
Where: t = The number of years from the reference period (year 0 of the benefits stream)
R = Discount rate
V(t) = The benefits occurring t years from the reference period
There is much discussion among economists of the proper discount rate to use for policy analysis.
Therefore, for Stage 2 DBPR benefits analyses, PV calculations are made using two rates thought to best
represent current policy evaluation methodologies, 3 and 7 percent. Historically, the use of 3 percent is
based on rates of return on relatively risk-free investments, as described in the Guidelines for Preparing
Economic Analyses (USEPA 2000J). The rate of 7 percent is recommended by OMB as an estimate of
"before-tax rate of return to incremental private investment" (USEPA 1996b). To allow evaluation on an
annual basis, the PV of benefits is annualized using the same discount rates. Exhibit 6.28 presents the
annualized present value of estimated benefits for the Stage 2 DBPR over the 25-year time frame for
analysis.
10 See EPA's Guidelines for Preparing Economic Analyses (USEPA 2000J) for a full discussion of the use
of discount rates in the evaluation of policy decisions.
Final Economic Analysis for the Stage 2 DBPR 6-87 December 2005
-------
Exhibit 6.26a Non-Discounted Stream of Benefits from the Stage 2 DBPR
Preferred Regulatory Alternative, All Systems, WTP Curable Lymphoma,
TTHM as Indicator
Year
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
Ann.
Avg.
Smoking/Lung Cancer Cessation Lag Model
Mean Value
$
$
$
-
$
$ 135
$ 349
$ 631
$ 978
$ 1 ,277
$ 1 ,557
$ 1,810
$ 2,040
$ 2,242
$ 2,416
$ 2,566
$ 2,698
$ 2,817
$ 2,924
$ 3,023
$ 3,113
$ 3,198
$ 3,277
$ 3,309
$ 3,371
$ 1 ,749
90% Confidence Interval
5th %tile
$
$
$
$
$
$ 21
$ 53
$ 97
$ 150
$ 195
$ 238
$ 277
$ 312
$ 342
$ 368
$ 391
$ 410
$ 428
$ 445
$ 459
$ 473
$ 485
$ 496
$ 501
$ 510
$ 266
95th %tile
$
$
$
$
$
$ 310
$ 803
$ 1 ,450
$ 2,246
$ 2,935
$ 3,581
$ 4,164
$ 4,697
$ 5,166
$ 5,574
$ 5,924
$ 6,230
$ 6,515
$ 6,765
$ 6,995
$ 7,205
$ 7,407
$ 7,603
$ 7,672
$ 7,823
$ 4,043
Smoking/Bladder Cancer Cessation Lag Model
Mean Value
$
$
$
-
$
$ 127
$ 302
$ 515
$ 762
$ 933
$ 1 ,080
$ 1 ,206
$ 1 ,323
$ 1 ,432
$ 1 ,536
$ 1 ,635
$ 1,729
$ 1,821
$ 1 ,909
$ 1 ,994
$ 2,078
$ 2,158
$ 2,237
$ 2,284
$ 2,353
$ 1,177
90% Confidence Interval
5th %tile
$
$
$
-
$
$ 20
$ 46
$ 79
$ 117
$ 143
$ 165
$ 184
$ 202
$ 218
$ 234
$ 249
$ 263
$ 277
$ 290
$ 303
$ 315
$ 327
$ 339
$ 346
$ 356
$ 179
95th %tile
$
$
$
$
$
$ 293
$ 694
$ 1,182
$ 1,749
$ 2,144
$ 2,483
$ 2,774
$ 3,045
$ 3,299
$ 3,544
$ 3,774
$ 3,993
$ 4,211
$ 4,416
$ 4,616
$ 4,808
$ 5,000
$ 5,190
$ 5,297
$ 5,461
$ 2,719
Arsenic/Bladder Cancer Cessation Lag Model
Mean Value
$
$
$
-
$
$ 252
$ 618
$ 1 ,068
$ 1 ,583
$ 1 ,942
$ 2,224
$ 2,442
$ 2,621
$ 2,771
$ 2,899
$ 3,011
$ 3,110
$ 3,199
$ 3,279
$ 3,353
$ 3,422
$ 3,487
$ 3,548
$ 3,560
$ 3,606
$ 2,080
90% Confidence Interval
5th %tile
$
$
$
-
$
$ 39
$ 95
$ 163
$ 242
$ 297
$ 340
$ 373
$ 400
$ 422
$ 442
$ 458
$ 473
$ 486
$ 499
$ 510
$ 519
$ 529
$ 537
$ 539
$ 546
$ 316
95th %tile
$
$
$
$
$
$ 580
$ 1,422
$ 2,453
$ 3,637
$ 4,464
$ 5,116
$ 5,616
$ 6,032
$ 6,383
$ 6,690
$ 6,951
$ 7,181
$ 7,398
$ 7,586
$ 7,761
$ 7,920
$ 8,077
$ 8,231
$ 8,253
$ 8,369
$ 4,805
Notes: All values in millions of year 2003 dollars. Detail may not add to totals due to independent rounding.
EPA recognizes that benefits may be as low as zero since causality has not yet been established between
exposure to chlorinated water and bladder cancer.
The 90 percent confidence bounds for cases reflect uncertainty in PAR, reduction in average TTHM and
HAAS concentrations, and cessation lag.
Benefits were calculated using the Villanueva et al. (2003) PAR.
Source: Derived from Exhibit F.2u.
Final Economic Analysis for the Stage 2 DBPR
6-88
December 2005
-------
Exhibit 6.26b Non-Discounted Stream of Benefits from the Stage 2 DBPR
Preferred Regulatory Alternative, All Systems, WTP Chronic Bronchitis, TTHM as
Indicator
Year
2005
2006
2007
2008
200S
201 C
2011
2012
2013
2014
2015
2016
2017
2018
201 Ł
202C
2021
2022
2023
2024
2025
2026
2027
2028
2029
Ann.
Avg.
Smoking/Lung Cancer Cessation Lag Model
Mean Value
$
$
$
$
$
$ 67
$ 172
$ 312
$ 484
$ 632
$ 772
$ 898
$ 1,013
$ 1,114
$ 1,201
$ 1,277
$ 1,344
$ 1,404
$ 1,459
$ 1,509
$ 1,556
$ 1,600
$ 1,641
$ 1,656
$ 1,688
$ 872
90% Confidence Interval
5th %tile
$
$
$
$
$
$ 15
$ 38
$ 69
$ 106
$ 139
$ 169
$ 196
$ 221
$ 243
$ 261
$ 277
$ 291
$ 304
$ 315
$ 325
$ 335
$ 344
$ 352
$ 355
$ 362
$ 189
95th %tile
$
$
$
$
$
$ 146
$ 380
$ 687
$ 1 ,065
$ 1 ,392
$ 1,702
$ 1 ,980
$ 2,238
$ 2,462
$ 2,660
$ 2,830
$ 2,983
$ 3,122
$ 3,244
$ 3,358
$ 3,464
$ 3,562
$ 3,663
$ 3,693
$ 3,772
$ 1 ,936
Smoking/Bladder Cancer Cessation Lag Model
Mean Value
$
$
$
$
$
$ 63
$ 149
$ 255
$ 377
$ 462
$ 535
$ 598
$ 657
$ 712
$ 764
$ 814
$ 861
$ 908
$ 952
$ 996
$ 1,038
$ 1,080
$ 1,120
$ 1,143
$ 1,179
$ 587
90% Confidence Interval
5th %tile
$
$
$
$
$
$ 14
$ 33
$ 56
$ 83
$ 101
$ 117
$ 131
$ 143
$ 155
$ 166
$ 177
$ 187
$ 196
$ 206
$ 215
$ 223
$ 232
$ 240
$ 245
$ 253
$ 127
95th %tile
$
$
$
$
$
$ 138
$ 329
$ 560
$ 829
$ 1,017
$ 1,180
$ 1,319
$ 1,451
$ 1,572
$ 1,691
$ 1,803
$ 1,911
$ 2,018
$ 2,118
$ 2,216
$ 2,311
$ 2,404
$ 2,501
$ 2,550
$ 2,633
$ 1,302
Arsenic/Bladder Cancer Cessation Lag Model
Mean Value
$
$
$
$
$
$ 125
$ 306
$ 528
$ 784
$ 962
$ 1,103
$ 1,212
$ 1 ,301
$ 1 ,377
$ 1 ,442
$ 1 ,499
$ 1 ,549
$ 1 ,595
$ 1 ,636
$ 1 ,675
$ 1,711
$ 1 ,744
$ 1 ,776
$ 1 ,782
$ 1 ,806
$ 1 ,036
90% Confidence Interval
5th %tile
$
$
$
$
$
$ 27
$ 67
$ 116
$ 172
$ 211
$ 241
$ 265
$ 284
$ 300
$ 313
$ 325
$ 336
$ 345
$ 353
$ 361
$ 368
$ 375
$ 381
$ 382
$ 387
$ 224
95th %tile
$
$
$
$
$
$ 274
$ 673
$ 1,162
$ 1,725
$ 2,118
$ 2,432
$ 2,671
$ 2,874
$ 3,042
$ 3,193
$ 3,321
$ 3,437
$ 3,545
$ 3,638
$ 3,725
$ 3,808
$ 3,884
$ 3,965
$ 3,974
$ 4,035
$ 2,300
Notes: All values in millions of year 2003 dollars. Detail may not add to totals due to independent rounding.
EPA recognizes that benefits may be as low as zero since causality has not yet been established between
exposure to chlorinated water and bladder cancer.
The 90 percent confidence bounds for cases reflect uncertainty in PAR, reduction in average TTHM and
HAAS concentrations, and cessation lag.
Benefits were calculated using the Villanueva et al. (2003) PAR.
Source: Derived from Exhibit F.3u.
Final Economic Analysis for the Stage 2 DBPR
6-89
December 2005
-------
Exhibit 6.27 Benefits Summary for the Stage 2 DBPR,
Preferred Regulatory Alternative (Millions, 2003$)
Approach (PAR Source)
Approach 1 : "5 Studies"
Approach 2: Villaneuva 2003
Approach 3: Villaneuva 2004
Best Estimate
Pre-Stage 1
PAR
A
2% or 17%
(0%, 33%)
15.7% (8.5%,
27.2%)
17.1% (2.5%,
33.1%)
Best Estimate
Annual Cases
Ultimately
Avoided1
B
64 or 546 (0,
1060)
506 (275, 874)
550(80, 1064)
Average
Annualized
Cases Avoided
for 25 years
C
35 or 301 (0, 656)
279(103,541)
303 (30, 658)
Annualized
Expected Benefits
3%2
(Millions, 2003$)
D
$194or$1,652($0,
$4,288)
$1 ,531 ($233,
$3,536)
$1 ,664 ($068,
$4,304)
Annualized
Expected Benefits
7%2
(Millions, 2003$)
E
$158or$1,346($0,
$3,490)
$1,246 ($190,
$2,878)
$1 ,355 ($055,
$3,503)
Notes: Detail may not add to totals due to independent rounding.
Smoking-lung cessation lag model and chronic lymphoma WTP used to calculate benefits.
as indicator.
_., ..... EPA recognizes that benefits may be as low as zero since causality has not
ed between exposure to chlorinated water and bladder cancer. The estimate of annual
1 Based on TTHM ;
yet been estabhsne
cases ultimately avoided shown here are those calculated from the PAR values as described in 6.4.1.1. The
full benefits simulation model, which incorporates uncertainty in the impacts of the IDSE and uncertainty in
SWAT predictive equations, and in the estimates of PAR, produces slightly higher mean estimates.
Specifically, the model estimates 581 cases versus the 506 cases shown here for Approach 2. Cases
avoided have not yet been discounted.
2 The 90 percent confidence bounds for benefits incorporates uncertainty in the VSL, WTP, and income
elasticity adjustment relative to mean cases.
Source: (A) Exhibit 6.20.
(B) Exhibit E. 19.
(C, D) The cases avoided from column B multiplied by the annualized benefits per case, derived from Exhibit
6.28.
Final Economic Analysis for the Stage 2 DBPR
6-90
December 2005
-------
Exhibit 6.28 Benefits Summary for the Stage 2 DBPR,
Preferred Regulatory Alternative (Millions, 2003$)
Adverse Reproductive and Developmental Health Effects Avoided
Causality has not been established, and numbers and types of cases avoided, as well as the value of such cases, were not quantified
in the primary benefits analysis. Given the numbers of women of child-bearing age exposed (58 million), the evidence indicates that
the number of cases and the value of preventing those cases could be significant. See results of the illustrative calculation in Section
6.8.
Number and Value of Estimated Bladder Cancer Cases Avoided 1
Causality has not been established; however, the weight of evidence supports PAR estimates of potential benefits.
Cessation Lag Model
used to estimate
Annual Bladder Cancer
Cases Avoided
Smoking/Lung
Cancer Model
Smoking/Bladder
Cancer Model
Arsenic/Bladder
Cancer Model
Annual Average Bladder Cancer
Cases Avoided for the first 25
years 2
Mean
279
188
333
5th
103
61
138
95th
541
399
610
Discount Rate, WTP for
Non-Fatal Bladder
Cancer Cases
3 %, Lymphoma
7 % Lymphoma
3 % Bronchitis
7 % Bronchitis
3 %, Lymphoma
7 % Lymphoma
3 % Bronchitis
7 % Bronchitis
3 %, Lymphoma
7 % Lymphoma
3 % Bronchitis
7 % Bronchitis
Annualized Benefits of Bladder
Cancer Cases Avoided
(Millions, 2003$) 3
Mean
$1,531
$1,246
$763
$621
$1,032
$845
$514
$420
$1,852
$1,545
$922
$769
5th
$233
$190
$165
$135
$157
$129
$111
$91
$282
$235
$200
$167
95th
$ 3,536
$ 2,878
$ 1,692
$ 1,376
$ 2,384
$ 1,950
$ 1,141
$932
$ 4,276
$ 3,566
$ 2,045
$1,704
Other Health Benefits
Qualitative assessment indicates that the value of other health benefits could be positive and significant.
Non-Health Benefits
Qualitative assessment indicates that the value of non-health benefits could be positive.
Notes: Detail may not add to totals due to independen
1 Based on TTHM as indicator. Villanueva et a
as zero since causality has not yet been estab
t rounding.
. (2003) for PAR. EPA recognizes that benefits may be as low
isned between exposure to chlorinated water and bladder
cancer.
2 The 90 percent confidence interval for cases incorporates uncertainty in PAR, reduction in average TTHM
and HAA5 concentrations, and cessation lag.
3. The 90 percent confidence bounds for monetized benefits reflect uncertainty in monetization inputs relative
to mean cases. Assumes 24 percent of cases are fatal, 76 percent are non-fatal (USEPA 1999a).
Source: Summarized from detailed figures presented in Appendix E (Exhibits E.38d, E.38h, E.38I) and F (Exhibits
F.2v and F.2w, F.3v and F.3w).
6.5.4 Comparison of the Value of Benefits for Regulatory Alternatives
This section compares the benefits of decreasing DBF occurrence under the Stage 2 DBPR
preferred alternative with the three evaluated alternatives. The alternatives are summarized below (see
Chapter 4 for a more detailed discussion):
Preferred Alternative:
Alternative 1:
80 (ig/L TTHM and 60 (ig/L HAAS as an LRAA; bromate MCL of 10
(ig/L as an RAA based on monthly samples taken at the finished water
point (no change from the Stage 1 DBPR for bromate); compliance
monitoring proceeded by the IDSE.
80 (ig/L TTHM and 60 (ig/L HAAS as an LRAA; bromate MCL of S^g/L
as an RAA based on monthly samples taken at the finished water point.
Final Economic Analysis for the Stage 2 DBPR
6-91
December 2005
-------
Alternative 2: 80 (ig/L TTHM and 60 (ig/L HAAS as the single maximum value for any
sample taken during the year; bromate MCL of 10 (ig/L as an RAA based
on monthly samples taken at the finished water point (no change from the
Stage 1 DBPR for bromate).
Alternative 3: 40 (ig/L TTHM and 30 (ig/L HAAS as an RAA of all distribution samples
taken; bromate MCL of 10 (ig/L as an RAA based on monthly samples
taken at the finished water point (no change from the Stage 1 DBPR for
bromate).
Exhibit 6.29 compares the value of benefits for all Stage 2 DBPR regulatory alternatives, using
TTHM as an indicator for reduction in all chlorination DBFs. Although the primary objective of the Stage
2 DBPR is to reduce the non-quantified risks of adverse potential reproductive and developmental health
effects, the exhibit also shows that the quantified benefits of reducing cancer cases are considerable
regardless of which alternative is chosen. Chapter 7 examines the costs of these alternatives, and Chapter
9 compares their benefits and costs. An additional analysis in Appendix N considers the cost effectiveness
of the alternatives in terms of quality-adjusted life years saved.
Exhibit 6.29 Number and Annualized Value of Estimated Bladder Cancer Cases
Avoided for All Stage 2 DBPR Regulatory Alternatives, Villanueva et al. (2003) for
Baseline Risk (Millions, 2003$)1
Average Annual
Number of Cases
Avoided 2
Annualized
Benefits of Cases
Avoided3
Discount Rate,
WTP for Non-
Fatal Cases
-
3%, Lymphoma
7%, Lymphoma
3%, Bronchitis
7%, Bronchitis
Preferred Alternative
Mean
279
$ 1 ,531
$ 1 ,246
$763
$621
5th
103
$233
$190
$165
$135
95th
541
$ 3,536
$ 2,878
$ 1 ,692
$ 1 ,376
Alternative 1
Mean
250
$ 1 ,377
$1,126
$686
$561
5th
127
$209
$172
$149
$122
95th
397
$3,180
$ 2,600
$ 1 ,522
$ 1 ,243
Alternative 2
Mean
939
$5,167
$ 4,227
$ 2,575
$2,105
5th
483
$786
$644
$558
$457
95th
1,466
$ 1 1 ,936
$ 9,758
$5,712
$ 4,665
Alternative 3
Mean
1,296
$7,130
$ 5,832
$ 3,552
$ 2,904
5th
675
$ 1 ,085
$888
$769
$630
95th
1,988
$ 16,468
$ 13,464
$ 7,880
$ 6,436
Notes: 1Based on TTHM as indicator, and the smoking and lung cancer cessation lag model. EPA recognizes that
benefits may be as low as zero since causality has not yet been established between exposure to chlorinated
water and bladder cancer.
2 The 90 percent confidence interval for cases incorporates uncertainty in PAR, reduction in average TTHM
and HAAS concentrations, and cessation lag.
3 The 90 percent confidence bounds for monetized benefits reflect uncertainty in monetization inputs relative
to mean cases.
"The Preferred Alternative avoids more cases of bladder cancer than the next least expensive
alternative-Alternative 1-because it is the only alternative for which the IDSE is considered, resulting in
increased benefits. However, Alternative 1 is the only alternative requiring a lower bromate standard, whose
benefits are not considered in this analysis. Should potential cancer cases averted by lowering the bromate
standard be quantified, Alternative 1 would capture more benefits.
Assumes 24 percent of cases are fatal, 76 percent are non-fatal (USEPA 1999a).
Source: Summarized from detailed figures presented in Appendix E (Exhibits E.38d, E.40d, E.41d, E.42d) and
Appendix F (Exhibits F.2v and w, F.Svandw, F.6b, F.7b, F.8b, F.9b, F.10b, F.11b).
Final Economic Analysis for the Stage 2 DBPR
6-92
December 2005
-------
6.6 Uncertainties
Many factors contribute to uncertainty in national benefits estimates. Uncertainty exists in model
inputs such as the estimated PAR values and the cessation lag models. To assess uncertainty in the
approach used to estimate the number of bladder cancer cases in the baseline that can be attributed to DBF
occurrence and exposure, and the number of cases that can be avoided by implementation of the Stage 2
DBPR, three approaches were used to estimate Pre-Stage 1 PAR (see Appendix E for more detail). To
quantify uncertainty in cessation lag, three independent cessation lag models derived from three different
epidemiological studies are used. Also, two functional forms are used for each of these data sets and
uncertainty in the parameters of those functions is included in the analysis (see Appendix E for more
detail). For monetization of benefits, EPA uses two alternatives for valuing non-fatal bladder cancer.
Other uncertainties, such as the linear relationship between DBP reductions and reductions in bladder
cancer cases avoided, are discussed qualitatively. A summary of the key uncertainties and the effects of
uncertainty in those assumptions on the benefits and cost analyses are presented in Exhibit 6.29.
EPA believes that uncertainty in the compliance forecast has a potentially large influence on
benefit (and cost) estimates in this EA. Thus, the Agency has attempted to quantify the uncertainty by
giving equal weight to two different compliance forecast approaches. One compliance forecast approach is
based on the SWAT predictions, and the other is based on the ICR matrix method. The ICR Matrix
Method uses the same basic approach as SWAT, but uses TTHM and HAAS data from the ICR directly to
estimate the percent of plants changing technology to comply with the Stage 2 DBPR and the resulting
DBP reduction (see Chapter 5 for more information on these approaches). To characterize the uncertainty
of the compliance forecast results, EPA assumes a uniform distribution between SWAT and ICR Matrix
Method results. That is, the national benefits estimates presented in this chapter represent the midpoint
between benefits estimated using the SWAT model, and those estimated using the ICR Matrix Method.
Benefits estimates using the SWAT model are about 30% lower than the midpoint estimates while those
using the ICR Matrix Method are about 30% higher.
Two of the greatest uncertainties affecting the benefits of the Stage 2 DBPR are related to non-
quantified benefits estimates. Both of these factors result in an underestimation of quantified Stage 2
DBPR benefits. To inform the reader of the potential magnitude of these benefits, Section 6.7 provides
results from a sensitivity analysis for colon and rectal cancers combined. An illustrative analysis of
potential developmental and reproductive benefits of the Stage 2 DBPR is discussed in Section 6.8.
In addition to the uncertainties listed in Exhibit 6.30, the potential costs or benefits of a possible
interactive effect from the promulgation of more than one rule in a short period of time are also not
quantified. EPA has taken into account compliance with the Stage 1 DBPR and considered the potential
impacts of the Ground Water Rule on non-treatment costs in Appendix H, and considered potential
impacts of the Arsenic Rule, and the LT2ESWTR. EPA addresses potential increased risk due to ground
water systems adding disinfection under the Ground Water Rule in Appendix M.
Final Economic Analysis for the Stage 2 DBPR 6-93 December 2005
-------
Exhibit 6.30 Uncertainties and Possible Effect on Estimate of Benefits
Uncertainty
Uncertainty in DBP reductions for
surface water systems
Uncertainty in DBP reductions for
ground water systems
Analysis of reduction in DBP
occurrence does not include results of
IDSE
Uncertainty in valuation inputs (WTP
and VSL)
Uncertainty in the bladder cancer
PAR value1
Analysis of exposure reduction
assumes TTHM and HAAS to be
proxies for all chlorination DBPs
DBPs have a linear no-threshold
dose-response relationship for
bladder cancer effects
Uncertainty in cessation lag function
Benefits of reduced cancers other
than bladder cancer are not included
in the quantitative analysis
Potential reproductive and
developmental health effects avoided
are not quantified in the primary
analysis
Section with
Full Discussion
of Uncertainty
Chapter 5
Chapter 5
5.3
6.5.2
6.1.1
Appendix E
6.3.3
6.2.1
6.4.2.2
Appendix E
6.7
6.8
Effect on Benefit Estimate
Under-
estimate
Over-
estimate
Unknown
Impact
Quantified in the primary analysis
(addresses potential underestimate or
overestimate)
X
Quantified in the primary analysis
(addresses for potential underestimate)
Quantified in the primary analysis
(addresses for potential underestimate or
overestimate)
Quantified in the primary analysis
(addresses range of potential effects, but
true values could lie outside the range)
X
X
Quantified in the primary analysis
(addresses potential underestimate or
overestimate)
Quantified in a sensitivity analysis
(addresses potential underestimate)
X
To assess uncertainty in PAR estimates, three approaches were used to estimate Pre-Stage 1 PAR, as shown in
Appendix E of the EA. PAR value average estimate of 16 percent used to calculate the number of bladder cancer
cases avoided is not absolute. EPA recognizes that the number of cases may be as low as zero since causality has
not yet been established between exposure to chlorinated water and bladder cancer.
6.7 Sensitivity Analysis for Other Cancers
Colon and rectal cancers combined are the third most common site (excluding skin) of new cancer cases
and deaths in both men and women in the U.S. The American Cancer Society (ACS) estimated that
104,950 new colon and rectal cancer cases will be diagnosed in 2005, with 56,290 resulting in deaths
(ACS 2005). Human epidemiology studies on chlorinated surface water have reported associations with
colon and rectal cancers.
Final Economic Analysis for the Stage 2 DBPR
6-94
December 2005
-------
In the development of the Stage 1 DBPR, EPA investigated estimating a PAR for colon and rectal
cancers and concluded that the data was not conclusive enough to calculate a PAR. Prior to the Stage 1
DBPR, several population-based case control and prospective cohort studies had been published that
evaluated the association between consumption of chlorinated drinking water and colon or rectal cancer
(Wilkins and Comstock 1981, Bean et al. 1982, Cragle et al. 1985, Young 1987, Doyle 1997, Koivusalo et
al. 1997, Hildesheim et al. 1998).
Wilkins and Comstock (1981) examined annual mortality rates in Washington County, MD from
1963-1975 comparing municipal residents with deep well users. For rectal cancer, a relative risk (RR) of
1.42 was reported, but this was not statistically significant. No difference was observed for colon cancer.
Bean et al. (1982) examined cancer incidence data for the state of Iowa for the years 1969-1978
considering differences for surface and ground water supplies (and well depth for ground water) and for
different population-size municipalities. They observed higher incidence of rectal cancer for both males
and females in surface water versus ground water for all size municipalities. No difference was observed
for colon cancer. Cragle et al. (1985), a hospital-based case-control study, and Young et al. (1987), a case-
control study, were excluded due to inadequate exposure assessment. Doyle et al. (1997), a prospective
cohort study, indicated that in comparison with women who used municipal ground water sources, women
with municipal surface water sources were at an increased risk of colon cancer and all cancers combined.
Doyle (1997) did not report an association between consumption of chlorinated surface water and cancer
of the rectum and anus. Koivusalo et al. (1997) conducted an historical cohort study of 621,431 individuals
in 56 towns in Finland and compared cancer incidence between locations based on a measure of raw water
and chlorination treatment. Comparing locations representing exposure to chlorinated surface water and
no exposure to chlorinated surface water, they observed a statistically significant increased risk of rectal
cancer for women (RR = 1.38), but no increased risk for rectal cancer for men nor for colon cancer for
either sex. Based on these studies for colon and for rectal cancer, it was not possible to estimate a PAR
range at the time of the Stage 1DBPR. Hildesheim et al. (1998) indicated an association between rectal
cancer and exposure to chlorinated water, but did not observe a significant increase in risk of colon cancer
associated with chlorinated surface water use.
Since the Stage 1 DBPR, additional human epidemiology studies have been published that
investigated the potential relationship between colon and rectal cancers and exposure to chlorinated
surface water (King et al. 2000a and Yang et al. 1998). The database of studies on colon and rectal
cancers continues to support a possible association, but evidence remains mixed. King et al. (2000a), a
population-based case-control study, found evidence of an increased colon cancer risk for males with
cumulative exposure to THMs and duration of exposure to chlorinated surface water. No associations
were observed between exposure measures and rectal cancer. Yang et al. (1998) conducted a cross-
sectional mortality study in Taiwan which compared cancer mortality rates from chlorinating
municipalities and non-chlorinating municipalities. The resulting Ratios of Age-adjusted Mortality Rates
(SRR) were significantly higher for males and for females for rectal cancer but not for colon cancer. This
study did not control for diet or smoking. Also, because this study is based on mortality, not incidence, it
makes this study problematic for doing risk analysis for bladder cancer incidence.
Because the area of the colon and rectum is the third most common site (excluding skin) of new
cancer cases and deaths in both men and women in the U.S., and human epidemiology studies on
chlorinated surface water have reported potential associations with colon and rectal cancers, EPA chose to
perform a sensitivity analysis on potential benefits from avoiding colon and rectal cancers. Of the
available studies, EPA has identified Hildesheim et al. (1998) and King et al. (2000a) as providing
adequate data to estimate PAR values for colon and rectal cancers. Additional details on these studies are
presented below.
Hildesheim et al. (1998) conducted a population-based case-control study and found an association
between duration of chlorinated surface water use and rectal cancer in Iowa residents in 1986-1989. It
should be noted that this study is essentially the same research group and study setting as the Cantor et al.
Final Economic Analysis for the Stage 2 DBPR 6-95 December 2005
-------
(1998) study on bladder cancer that was used in the main benefits analysis. No important association was
found for colon cancer and related sites for either duration of exposure or TTHM estimates. ORs were
presented for several characterizations of exposure such as whether the study group was exposed to
chlorinated surface, ground water, or any chlorinated water, duration of exposure, total lifetime THM
level, and lifetime average THM concentration. This study was used to estimate a PAR for rectal cancer.
King et al. (2000a) conducted a population-based case-control study in southern Ontario, Canada.
This study is essentially the same research group and study setting as the King and Marrett (1996) study
that was used in the Villanueva et al. (2003) meta-analysis. ORs were presented for duration of exposure to
chlorinated water and for exposure to various levels of TTHMs. An association was found among males
for colon cancer risk with cumulative exposure to TTHMs, duration of exposure to chlorinated surface
water, and duration of exposure to • 50 • g/L and to 75 • g/L of TTHM. These relationships were not
observed in females. No association was found between rectal cancer risk by any of the measures of
exposure to disinfection byproducts in this study. This study was used to estimate a PAR for colon cancer
(for males only).
PAR for colon cancer
EPA estimated a PAR value for colon cancer for males only using data from the King et al. (2000)
study. PAR was calculated using the method described in Appendix E for calculating PAR for bladder
cancer for the five epidemiological studies used in Approach 1 for bladder cancer. Using the data from
Table 2 of the King et al. (2000) study reflecting years of exposure to chlorinated water resulted in a PAR
value estimate of 24.5%. Note again that this only applies to males since the authors did not detect
associations for females.
PAR for rectal cancer
EPA estimated a PAR value for rectal cancer for both sexes only using data from the Hildesheim
et al. (1998) study. As for colon cancer, PAR for rectal was calculated using the method described in
Appendix E for calculating PAR for bladder cancer for the five epidemiological studies used in Approach
1 for bladder cancer. Using the data from Table 1 of the Hildesheim et al. (1998) study, reflecting years of
exposure to any chlorinated water, resulted in a PAR value estimate of 11.8%.
Estimated Benefits for Reductions in Colon and Rectal Cancers
Using the above PAR values with data from SEER and the American Cancer Society for colon and
rectal cancer incidence, EPA estimated the annualized value of the cases of colon cancer avoided from
Stage 2 for males only and of cases of rectal cancer avoided for both sexes. The results are presented in
Exhibit 6.31, which also shows the annualized value of the bladder cancer cases avoided from Stage 2.
These results are based on TTHM reductions and include the smoking / lung cancer cessation lag model.
Additionally, a sensitivity analysis in Appendix N to this document incorporates into a cost effectiveness
analysis the quality-adjusted life years that are potentially saved for avoided colon and rectal cancer cases
EPA recognizes that actual risks and PAR values could be zero due to uncertainties in the
scientific evidence. Detailed benefits information can be found in Appendix E2.
Final Economic Analysis for the Stage 2 DBPR 6-96 December 2005
-------
Exhibit 6.31 Annualized Value1 of Estimated Bladder Cancer Cases Avoided for
the Primary Analysis, and Estimated Colon and Rectal Cancer Cases Avoided for
the Sensitivity Analysis (Millions, 2003$)2
Smoking/Lung Cancer Cessation Lag Model
Discount Rate, WTP
for Non-Fatal Cases
3 %, Lymphoma
7 % Lymphoma
3 % Bronchitis
7 % Bronchitis
Bladder Cancer - Primary
Analysis
Mean
$1,531
$1,246
$763
$621
5th
$233
$190
$165
$135
95th
$3,536
$2,878
$1,692
$1,376
Colon Cancer Sensitivity
Analysis
Mean
$2,395
$1,991
$1,193
$991
5th
$364
$303
$259
$215
95th
$5,530
$4,594
$2,645
$2,195
Rectal Cancer Sensitivity
Analysis
Mean
$819
$680
$408
$339
5th
$125
$104
$88
$74
95th
$1,890
$1,570
$904
$750
Notes: 1The 90 percent confidence bounds shown in the exhibit incorporate uncertainty in the VSL, WTP, and
income elasticity adjustment.
2 Based on TTHM as indicator for the Preferred Regulatory Alternative. EPA recognizes that the benefits may
be as low as zero since causality has not yet been established between exposure to chlorinated water and
bladder, colon, or rectal cancer.
Source: Quantitative values are summarized from detailed figures presented in Appendix F (Exhibits F.2ab, F.2ac, F.3
ab, F.Sac, F.12c and d, F.13c and d, F.14c and d, F.15c and d).
Caveats to analysis
The uncertainties that pertain to the PAR analysis for bladder cancer (see Section 6.4 - 6.6) also
pertain to the PAR analysis for colon and rectal cancer. However, additional uncertainties pertain to the
risk estimates for colon and rectal cancer. Exposure levels and associated risk from limited
epidemiological studies may not be representative of the general population; and inconsistency of the
relative risk for cancer type and gender (the King et al study shows an association for males only while the
Hildesheim et al study reported an association for rectal but not for colon cancer) are significant and
difficult to explain. Therefore, the estimates from this sensitivity analysis should only be interpreted as
indication of possible benefits from this rule while also recognizing that the benefits may be as low as
zero.
6.8 Potential Fetal Losses Avoided
EPA believes that additional benefits from this rule could come from reduction in developmental
and reproductive health effects. EPA does not believe the available evidence provides an adequate basis
for quantifying potential reproductive or developmental risks in the main benefits analysis. Furthermore,
no causal link between DBFs and these risks has yet been established. (See Section 6.2.2 for a complete
discussion of the reproductive and developmental health effects of DBFs.) Nevertheless, given the
widespread exposure to DBFs, the importance society places on reproductive and developmental health,
and the nearly 1 million fetal losses each year in the U.S., the Agency believes that it is appropriate to
provide some quantitative indication of the potential risk in these categories. To do this, PAR calculations
from several studies on the relationship between chlorinated water exposure and fetal loss have been
adapted and applied to national statistics on the annual incidence of fetal loss (shown in Section 6.8.1).
Section 6.8.2 discusses valuation of these potential fetal losses avoided.
Final Economic Analysis for the Stage 2 DBPR
6-97
December 2005
-------
6.8.1 Reproductive Effects Illustrative Calculation
EPA has calculated the unadjusted ORs or RRs associated with each of four population-based
epidemiological studies of fetal loss: Waller et al. 2001, King et al. 2000b, Savitz et al. 1995, and Savitz et
al. 2005. All are high-quality studies that have sufficient sample sizes and high study response rates,
adjustments for known confounders11, have exposure assessment information from water treatment data,
residential histories, and THM measurements. These are summarized in Exhibit 6.32. Because the
populations in these four studies appear to have TTHM exposures significantly greater than those of the
general U.S. population, EPA has chosen to scale the results using ICR data to derive PAR values that are
more relevant to the general population (Appendix G).
The four studies (using unadjusted data to allow for comparability, and scaled to the TTHM levels
reported in the ICR data base) yield median PARs of 0, 0.4, 1.7, and 1.9 percent12. Using the annual
incidence of fetal loss reported by CDC, the median PARs for these four studies suggest that the incidence
of fetal loss attributable to exposure to chlorinated drinking water could range from 0 to 18,700 annually.
EPA assumed that this potential risk range pertained to pre-Stage 1 exposure conditions. Thus, to evaluate
potential reduction in fetal loss for the Stage 2 DBPR, EPA had to first estimate the potential reductions
for the Stage 1 DBPR and then the subsequent percent reductions for Stage 2 DBPR. For this analysis,
EPA assumed that reduction in risk is proportional to the percent reductions in the number of observations
having quarterly TTHM concentration measurements above the study population cut offs. This analysis
would imply that the Stage 1 DBPR would reduce potential fetal loss by 73 percent and that the Stage 2
DBPR would further reduce remaining potential fetal loss by 75 percent. This analysis would further
imply that a range of 0 to 3,700 fetal losses could be avoided per year as a result of the Stage 2 DBPR.
Using the three-quarter average, the number of fetal losses attributable to DBFs is reduced by 90 percent
for Stage 1, with the remaining fetal losses eliminated by Stage 2. Refer to Appendix G for derivation of
PARs and detailed calculation of potential fetal losses avoided.
Caution is required in interpreting the numbers derived above because there may be significant
differences in exposure patterns of the study populations and the national population (e.g., different types
of DBP mixtures having similar TTHM levels). The estimates presented here are not part of EPA's
primary benefits analysis, and the ranges are not meant to suggest upper and lower bounds. Rather, they
are intended to illustrate quantitatively the potential risk implications of some of the published results.
EPA reiterates that causality has not yet been established and that each of the studies has a lower 95
percent confidence bound for the estimated attributable risk below zero.
1 :Use of unadjusted OR or RR estimates has the effect of including possible biases from known
confounders; however, EPA believes the unadjusted estimates are adequate for purposes of the illustrative
calculations presented here. It was not possible to calculate ORs adjusted for the same confounders from the four
studies.
12The calculated lower 95 percent confidence intervals on PAR for all four studies were less than zero. This
means the possibility of no effect cannot be ruled out with 95 percent confidence on the basis of any single study.
The upper 95 percent confidence intervals for the PAR estimates were 3, 4, 4, and 6 percent.
Final Economic Analysis for the Stage 2 DBPR 6-98 December 2005
-------
Exhibit 6.32 Summary of the Fetal Loss Human Epidemiology Studies
Study
Waller et
al. 2001
King et
al. 2000b
Savitz et
al. 1995
Savitz et
al. 2005 2
Population
Prospective
cohort of 4,209
pregnant women
in prepaid health
plan in CA 1989-
91
Population-
based
retrospective
cohort of 47,275
births in Nova
Scotia, Canada
1988-1995
Population-
based case-
control study of
126 cases and
122 controls in
NC 1988-91
Prospective
cohort of 2,41 3
pregnant women
from 3 water
systems in the
U.S., 2000-2004
Exposure
Assessment
Used utility total
trihalomethane
(TTHM) data to
estimate exposure via
ingestion and
showering during first
trimester of
pregnancy.
Linked mother's
residence at time of
delivery to the levels
of specific TTH Ms
monitored in the
distribution system of
the utility and
averaged predicted
values of TTHM levels
for all the months
covering the
pregnancy.
Linked existing
distribution system
TTHM concentration
data to maternal
residence and water
consumption data.
The fourth week of
pregnancy used to
assign the reported
quarterly average
TTHM.
Weekly or biweekly
distribution system
DBP concentration
data were collected
and linked with
maternal residence
and water
consumption data
(during first and
second trimesters).
Periconceptual, early
and late gestational
exposure windows
were examined.
Outcome
Spontaneous
abortion
(• 20 weeks
of gestation)
Stillbirth
Spontaneous
abortion
Spontaneous
abortion,
including
early (<1 2
wks) and late
(> 12 wks)
fetal losses
Results1
Recalculated 80
• g/L compared to
<80 • g/L for
unweighted utility-
wide average
RR=1.25
(0.99,1.6)
Study based
prevalence of
exposure = 15%
Recalculated 75
• g/L compared to
<75 • g/L RR=1 .28
(95% Cl 0.98, 1 .7)
Study based
prevalence of
exposure = 32%
Recalculated 81
• g/L compared to
<81 'g/LOR=1.06
(0.6,1.8)
Study based
prevalence of
exposure = 35%
Utilized reported
statistic of 75 • g/L
compared to <75
• g/L OR=0.96
(0.68-1 .35) (Table
6.1, pg. 75)
Study based
prevalence of
exposure = 18%
Potential
Confounders
Evaluated
Gestational age at
interview, maternal
age, cigarette
smoking, history of
pregnancy loss,
maternal race,
employment during
pregnancy
Smoking, maternal
age
Maternal age, race,
education, marital
status, poverty level,
smoking, alcohol use,
nausea, employment
Maternal age,
tobacco use, race,
ethnicity, education,
marital status,
income, alcohol use,
caffeine
consumption, body
mass index, age at
menarche,
employment,
diabetes, pregnancy
history, prior fetal
loss, induced
abortion history,
vitamin use
1 EPA recalculated OR and RR values using crude Odds Ratios for the fetal loss sensitivity analysis.
This study was added to the analysis since the proposal.
Final Economic Analysis for the Stage 2 DBPR
6-99
December 2005
-------
6.8.2 Value of Potential Reductions in Fetal Losses Avoided
EPA has not monetized the value of potentially avoided fetal loss but recognizes the significant
value of improvement in developmental and reproductive health in general. See Section 6.5.1 for a
discussion of the value of reducing reproductive and developmental health risks.
The agency is considering further work specific to fetal loss valuation. One possible area of
further research is the value that prospective parents have for reducing risks during pregnancy. In this
regard, the substantial lifestyle changes that prospective parents often undertake during pregnancy suggests
that reducing these kinds of risks are of value. A second possible area would be benefit transfer
methodologies that address how existing studies can inform the estimation of the benefits of reduced fetal
loss.
When valuation studies specific to the health endpoints of a regulation are lacking, the Agency
typically draws upon existing studies of similar health endpoints to estimate benefits. The "transfer" of the
results of these studies to value similar health endpoints must be done carefully and methodically,
controlling for differences in the health endpoints and in the relevant populations. Some researchers have
attempted to transfer values using sophisticated analytical techniques such as preference calibration
methods (e.g., Smith et al. 2002). Regardless of the approach used, "benefit transfer" requires systematic
comparison of the similarities and differences in the health effects in the studies and those resulting from
the regulation. Application of benefit transfer leads to a detailed qualitative examination of the
implications of using those studies and potentially to empirical adjustments to the results of the existing
studies.
Until more information on these subjects is available, EPA cannot estimate a monetized value for
avoiding fetal loss. However, research on valuation and benefit transfer continues to progress. The
Agency anticipates that new research will support further efforts to value reproductive and developmental
endpoints.
Final Economic Analysis for the Stage 2 DBPR 6-100 December 2005
-------
7. Cost Analysis
7.1 Introduction
This chapter estimates the national costs of the Stage 2 Disinfectants and Disinfection Byproducts
Rule (DBPR). National costs include treatment technology changes to comply with the rule as well as
non-treatment costs, such as Initial Distribution System Evaluations (IDSEs), additional routine
monitoring, and operational evaluations.
The data presented in this chapter are derived from analyses of Information Collection Rule (ICR)
data, ICR Supplemental Survey data, National Rural Water Survey data, Surface Water Analytical Tool
(SWAT) model results, the American Water Works Association Water Utility Database (Water:\STATS)
(AWWA 2000), State/Primacy Agency data, and the results of two expert opinion processes for small
systems. For a complete explanation of these data sources, see Chapter 3 and Appendices A and B.
Section 7.1.1 of this chapter summarizes the methodology and data inputs used to estimate
national costs. Sections 7.2 through 7.4 discuss the inputs that are used in the cost model. These are:
labor rates and laboratory fees (7.2), non-treatment costs (7.3), technology unit costs (7.4), and
compliance forecasts (derived in Chapter 5). Section 7.5 discusses the cost model itself, including how it
accounts for uncertainty in some inputs, the calculation of the costs, and projection and discounting of
those costs. The results of the cost model are discussed in Section 7.6. Section 7.7 discusses
unqualified costs and Section 7.8 gives a summary of the uncertainties in the cost calculations. Section
7.9 presents a comparison of the costs of the final rule option compared to other regulatory alternatives
that were analyzed.
In support of this chapter:
Appendix D contains the rule implementation schedule for different system types and rule
activities (used in projecting and annualizing costs over a 25-year period).
• Appendix H offers a more complete explanation of the laboratory costs and labor hours for
implementation, IDSE, monitoring plans, additional routine monitoring, as well as the
assumptions and calculations for the operational evaluation costs.
• Appendix I presents additional detail for technology unit costs.
• Appendix J presents cost projections, present value estimates, and annualization of costs for
all Stage 2 DBPR regulatory alternatives and sensitivity analyses.
Appendix K contains documentation for the Stage 2 DBPR cost model, including a detailed
file list and flow charts.
7.1.1 Overview of Methodology for Quantifying Stage 2 DBPR Costs
To estimate the national costs of the Stage 2 DBPR, the Environmental Protection Agency (EPA)
calculated the incremental costs that public water systems (PWSs) and their States/Primacy Agencies are
expected to incur from the Stage 1 DBPR to the Stage 2 DBPR. Cost analyses for PWSs include an
identification of treatment process improvements that systems may make, as well as estimates of the costs
Final Economic Analysis for the Stage 2 DBPR 7-1 December 2005
-------
to implement the rule1, conduct IDSEs, prepare monitoring plans, perform additional routine monitoring,
and perform operational evaluations. System costs were estimated for different system types (community
water systems (CWS) or nontransient noncommunity water systems (NTNCWS)), source water types
(ground or surface), and size categories (nine categories based on population served, consistent with the
Drinking Water Baseline Handbook). Estimates of costs to State/Primacy Agencies represent estimated
labor burdens that States/Primacy Agencies would face, such as training employees on the requirements
of the Stage 2 DBPR, responding to PWS reports, and recordkeeping.
Costs were calculated using a cost model programmed in SAS version 9.1. Exhibit 7.1
summarizes the model inputs and outputs. The model combines baseline data discussed in Chapter 3 with
inputs for labor rates, laboratory fees, non-treatment costs, technology unit costs, and compliance
forecasts from Chapter 5 to calculate nominal costs for the rule. All costs are calculated in 2003 dollars.
Costs are then distributed in the years in which they are expected to be incurred and are discounted to
give a net present value of costs using both a 3 and 7 percent social discount rate. The present value costs
are then annualized using the same social discount rate2.
Household costs are also generated by the cost model using assumptions regarding household
water usage and total households served per plant.
Exhibit 7.1 Stage 2 DBPR Cost Model Inputs and Outputs
Inputs
- Industry Baseline
- Compliance Forecasts
- Labor Rates and Lab Fees
- Technology Unit Costs
- Non-treatment costs
- Stage 2 rule
Implementation Schedule
SAS Cost
Model
Outputs
- Nominal Costs (one time
and annual)
- Annualized Costs,
Present Value in 2003$
- Household Costs
'For the purposes of this Economic Analysis (EA), rule implementation activities are assumed to include
activities such as reading the rule and training personnel. The activities are summarized in Section 7.3.1 and
discussed in more detail in Appendix H.
2 For more information on discounting costs see EPA's Guidelines for Preparing Economic Analyses
(USEPA 2000J).
Final Economic Analysis for the Stage 2 DBPR
7-2
December 2005
-------
7.1.2 Cost Summary
Number of systems performing non-treatment activities
Exhibit 7.2a shows the baseline number of systems subject to the Stage 2 DBPR and the
estimated number of those systems performing various rule activities (implementation, IDSE monitoring,
monitoring plans, additional routine monitoring, and operational evaluations). Appendix H provides the
derivation of these values.
As shown in columns B and C, EPA estimates that all disinfecting CWSs and NTNCWSs will
have to perform at least minimal implementation activities (reading and understanding the rule, training,
etc.). As shown in column F, all large systems and most small systems are required to prepare a Stage 2
compliance monitoring plan. The number of systems performing IDSE monitoring (shown in column D)
is only a fraction of all systems because some will choose to perform studies, meet the criteria for a 40/30
certification, or receive waivers from IDSE requirements.
EPA has established a population-based monitoring approach for the Stage 2 DBPR, where
monitoring requirements are no longer based on number of plants per system as under the Stage 1 DBPR.
Depending on the number of plants in a given system, the number of Stage 2 compliance samples
required per year may stay the same, decrease, or increase from Stage 1 requirements. See Exhibit H.8
for information on changes in monitoring burden.
EPA expects that some number of Stage 2-compliant systems will find TTHM and HAAS levels
high enough to trigger the requirement for an operational evaluation. Column H shows the estimated
number of systems that may require operational evaluations.
In addition to those ground water systems that currently disinfect, EPA predicts that some
systems will install disinfection to comply with the anticipated Ground Water Rule (GWR). Exhibit 7.2b
shows the number of systems predicted to install chemical disinfection as a result of the Ground Water
Rule, based on the Regulatory Impact Analysis (RIA) prepared for the May 2000 GWR proposal and
current system information available. Because the GWR is expected to be promulgated within 8 months
after the Stage 2 DBPR is promulgated, EPA expects new systems adding disinfection to meet GWR
requirements to simultaneously achieve compliance with Stage 2 MCLs. Therefore, these systems are not
included in the treatment baseline. The IDSE will likely not apply to these systems because they are
expected to install disinfection after the IDSE requirement is complete. Systems installing disinfection
for the GWR will, however, be required to prepare monitoring plans and monitor DBFs for the first time
under Stage 2. Exhibit 7.2b shows that all newly disinfecting ground water systems will prepare
monitoring plans and will be required to conduct monitoring for the first time. It should be noted that
actual monitoring-related costs for newly disinfecting ground water systems may be different from those
estimated here depending upon the details of the final GWR.
Final Economic Analysis for the Stage 2 DBPR 7-3 December 2005
-------
Exhibit 7.2a Baseline Systems Subject to Non-Treatment-Related Rule Activities
System Size
(Population Served)
Stage 2
DBPR
System
Baseline
A
Number and Percent of Systems Performing Various Rule Activities
Implementation
B C=B/A*100
IDSE Monitoring
D E=D/A*100
Stage 2 Monitoring Plans
F G=F/A*100
Operational
Evaluations
H I=H/A*100
Surface Water and Mixed CWSs
<100
100-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1 ,000,000+
National Totals
1,085
2,212
1,470
2,588
2,042
1,773
334
281
18
11,803
1,085 100%
2,212 100%
1,470 100%
2,588 100%
2,042 100%
1,773 100%
334 100%
281 100%
18 100%
11,803 100%
678 62%
1 ,382 62%
1 ,385 94%
2,438 94%
1 ,888 92%
1 ,524 86%
273 82%
226 81 %
15 83%
9,809 83%
678 62%
1 ,382 62%
1,470 100%
2,588 100%
2,042 100%
1,773 100%
334 100%
281 100%
18 100%
10,566 90%
4 0%
8 0%
10 1%
18 1%
57 3%
189 11%
68 20%
64 23%
6 33%
424 4%
Ground Water Only CWSs
<100
100-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1 ,000,000+
National Totals
7,935
9,821
3,998
4,852
2,200
1,222
136
63
3
30,229
7,935 100%
9,821 100%
3,998 100%
4,852 100%
2,200 100%
1,222 100%
136 100%
63 100%
3 100%
30,229 100%
336 4%
416 4%
708 18%
859 18%
389 18%
216 18%
24 18%
18 29%
0 15%
2,966 10%
336 4%
416 4%
3,998 100%
4,852 100%
2,200 100%
1,222 100%
136 100%
63 100%
3 100%
13,225 44%
0 0%
0 0%
0 0%
0 0%
0 0%
0 0%
0 0%
0 0%
0 0%
0 0%
Surface Water and Mixed NTNCWSs
<100
100-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1 ,000,000+
National Totals
231
317
106
93
24
5
0
1
0
777
231 100%
317 100%
106 100%
93 100%
24 100%
5 100%
0
1 100%
0
777 100%
0 0%
0 0%
0 0%
0 0%
0 0%
4 80%
0
1 100%
0
5 1%
0 0%
0 0%
0 0%
0 0%
0 0%
5 100%
0
1 100%
0
6 1%
0 0%
0 0%
0 0%
0 0%
0 0%
0 0%
0
0 0%
0
0 0%
Ground Water Only NTNCWSs
<100
100-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1 ,000,000+
National Totals
GRAND TOTAL
2,493
2,129
589
247
21
3
0
0
0
5,483
48,293
2,493 100%
2,129 100%
589 100%
247 100%
21 100%
3 100%
0 100%
0 100%
0
5,483 100%
48,293 100%
0 0%
0 0%
0 0%
0 0%
0 0%
1 18%
0 18%
0 100%
0
1 0%
12,780 26%
0 0%
0 0%
0 0%
0 0%
0 0%
3 100%
0 100%
0 100%
0
4 0%
23,800 49%
0 0%
0 0%
0 0%
0 0%
0 0%
0 0%
0 0%
0 0%
0
0 0%
424 1%
Notes:
Detail may not add to totals due to independent rounding.
Non-treatment-related rule activities, in addition to those shown in the table, also include routine compliance monitoring. Some systems
are expected to take more samples and some less from Stage 1 to Stage 2 depending on the number of plants in their systems. Overall
the Stage 2 DBPR results in an increase in the total number of compliance samples taken from the Stage 1 DBPR. See Exhibit H.8a,
column H, for the change in total samples for the different size categories.
Sources:
(A), (B), (D), (F), and (H): Appendix H, Exhibit H.15a.
Final Economic Analysis for the Stage 2 DBPR
7-4
December 2005
-------
Exhibit 7.2b Non-Treatment-Related Rule Activities for Systems Installing
Disinfection to Comply with the Ground Water Rule
System Size
(Population Served)
Baseline No. of
Systems Adding
Disinfection for the
GWR
A
Stage 2 Monitoring Plans
B
C=B/A*100
Surface Water and Mixed CWSs
<100
100-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1 ,000,000+
National Totals
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
-
-
-
-
-
-
-
-
-
-
-
Ground Water Only CWSs
<100
100-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1 ,000,000+
National Totals
354
439
86
104
47
10
1
2
0
1,042
354
439
86
104
47
10
1
2
0
1,042
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
Surface Water and Mixed NTNCWSs
<100
100-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1 ,000,000+
National Totals
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
-
-
-
-
-
-
-
-
-
-
-
Ground Water Only NTNCWSs
<100
100-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1 ,000,000+
National Totals
GRAND TOTAL
669
572
184
77
7
1
0
0
0
1,510
2,552
669
572
184
77
7
1
0
0
0
1,510
2,552
100%
100%
100%
100%
100%
100%
100%
100%
-
100%
100%
Notes:
Detail may not add to totals due to independent rounding.
Non-Treatment-Related Rule Activities, in addition to those shown in
the table, include routine compliance monitoring for all systems.
Sources:
(A), (B): Appendix H, Exhibit H.12b.
Final Economic Analysis for the Stage 2 DBPR
7-5
December 2005
-------
Number of plants making treatment technology changes
Exhibit 7.3 shows the baseline number of plants and the estimated percentage of those plants that
are predicted to make treatment technology changes. The mean estimated percentage of plants making
treatment technology changes is approximately 4 percent for all systems, with a much higher percentage
for surface water plants. The baseline number of ground water plants is larger than that of surface water
plants, however, there is a larger absolute number of ground water plants that are predicted to make
treatment technology changes.
The 90-percent confidence interval around the mean for the surface water plant estimate accounts
for alternative compliance forecast methodologies (SWAT and the ICR Matrix Method) and uncertainty
in the potential impacts of the IDSE. Derivation of the compliance forecast is discussed in detail in
Chapter 5.
One-time costs
One-time costs for systems include initial capital, implementation, IDSE, and monitoring plan
costs. State/Primacy Agency costs include those associated with implementation, IDSEs, and monitoring
plans. Exhibit 7.4 summarizes estimated total initial capital and other one-time costs of the Stage 2
DBPR for systems and States/Primacy Agencies.
Annualized Costs
Exhibit 7.5a and b summarize the average annualized costs for the Stage 2 DBPR at 3 and 7
percent discount rates, respectively. System costs range from approximately $56 to $102 million
annually at a 3 percent discount rate, with a mean estimate of approximately $79 million per year. At a 7
percent discount rate, system costs range from approximately $55 to $99 million annually, with a mean
estimate of approximately $77 million per year. State costs are estimated to be between $1.70 and $1.71
million per year, depending on the discount rate.
Final Economic Analysis for the Stage 2 DBPR 7-6 December 2005
-------
Exhibit 7.3 Number and Percent of Plants Making Treatment Technology
Changes for the Stage 2 DBPR
System Size
(Population Served)
Stage 2 DBPR
Plant Baseline
A
Number of Plants Making Treatment
Technology Changes
Mean
B
5th Percentile
C
95th Percentile
D
Percentage of Plants Making Treatment
Technology Changes
Mean
E=B/A
5th Percentile
F=C/A
95th Percentile
G = D/A
Primarily Surface Water CWSs
<100
100-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1 ,000,000+
National Totals
359
767
483
1,129
1,258
1,292
579
610
74
6,552
36
64
40
100
111
188
84
89
11
724
20
36
23
56
62
88
40
42
5
371
53
92
58
143
160
294
132
139
17
1,088
10.2%
8.4%
8.4%
8.8%
8.8%
14.6%
14.6%
14.6%
14.6%
11.1%
5.7%
4.7%
4.7%
4.9%
4.9%
6.8%
6.8%
6.8%
6.8%
5.7%
14.6%
12.1%
12.1%
12.7%
12.7%
22.8%
22.8%
22.8%
22.8%
16.6%
Primarily Ground Water CWSs
<100
100-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1 ,000,000+
National Totals
6,423
15,242
6,093
7,587
5,030
5,382
716
918
27
47,419
155
483
193
204
135
111
15
18
1
1,314
-
2.4%
3.2%
3.2%
2.7%
2.7%
2.1%
2.1%
2.0%
2.1%
/
/
2.8%
Primarily Suface Water NTNCWSs
<100
100-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1 ,000,000+
National Totals
226
312
106
91
25
5
0
1
0
766
23
26
9
8
2
1
0
0
0
69
13
15
5
5
1
0
0
0
0
39
33
38
13
12
3
1
0
0
0
100
10.2%
8.4%
8.4%
8.9%
8.8%
14.6%
-
14.6%
-
9.0%
5.7%
4.7%
4.7%
5.0%
4.9%
6.8%
-
6.8%
-
5.0%
14.6%
12.1%
12.1%
12.8%
12.7%
22.8%
-
22.8%
-
13.0%
Primarily Ground Water NTNCWSs
<100
100-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1 ,000,000+
National Totals
Grand Total All Plants
2,493
2,129
589
247
21
3
0
0
0
5,483
60,220
60
67
19
7
1
0
0
0
0
154
2,261
1,877
2,655
2.4%
3.2%
3.2%
2.7%
2.7%
2.1%
2.1%
2.0%
-
2.8%
3.8%
3.1%
/
/
4.4%
Note:
Sources:
Detail may not add to totals due to independent rounding. Estimated from the pre-Stage 2 Baseline.
(A) Exhibit 3.2, column AB.
(B)-(D) Compliance Forecast Inputs in Chapter 5, and the Stage 2 DBPR Cost Model. The 90 percent confidence
intervals for surface water systems represent alternative compliance forecast methodologies (SWAT and the ICR Matrix
Method) and uncertainty in the potential impacts of the IDSE.
Final Economic Analysis for the Stage 2 DBPR
7-7
December 2005
-------
Exhibit 7.4 Initial Capital and One-Time Costs for the Stage 2 DBPR ($Millions)
Total Initial Capital Costs for the Rule
(90% Confidence Bounds)
CWS Total Initial Capital
(90% Confidence Bounds)
NTNCWS Total Initial Capital
(90% Confidence Bounds)
CWS One-Time Costs
Implementation
IDSE
Monitoring Plans
NTNCWS One-Time Costs
Implementation
IDSE
Monitoring Plans
State/Primacy Agency One-Time Costs
Implementation
IDSE
Monitoring Plans
Surface Water Systems
Serving < 10,000
$ 100.9
(55.51 - 149.42)
$ 94.83
(52.22-140.40)
$ 6.03
(3.29 - 9.02)
$ 21.8
$ 2.4
$ 18.4
$ 1.0
$ 0.2
$ 0.2
$
$ 0.0
Serving > 10,000
$ 451.57
(246.01 - 630.04)
$ 450.6
(245.48-628.71)
$ 1.0
(0.53-1.33)
$ 33.5
$ 1.7
$ 31.1
$ 0.7
$ 0.1
$ 0.0
$ 0.1
$ 0.0
Disinfecting Ground Water Systems
Serving < 10,000
$ 179.8
(148.85-210.88)
$ 167.1
(138.11 -196.13)
$ 12.8
(10.74-14.75)
$ 13.5
$ 6.0
$ 5.8
$ 1.6
$ 1.3
$ 1.1
$
$ 0.2
Serving > 10,000
$ 107.3
(96.31 -118.23)
$ 107.1
(96.18-118.06)
$ 0.1
(0.13-0.16)
$ 3.3
$ 0.9
$ 2.0
$ 0.4
$ 0.0
$ 0.0
$ 0.0
$ 0.0
Total
$ 839.5
(546.68-1,108.57)
$ 819.6
(531.99-1,083.30)
$ 19.9
(14.69-25.27)
$ 72.0
$ 11.0
$ 57.4
$ 3.6
$ 1.6
$ 1.3
$ 0.1
$ 0.2
$ 10.9
$ 7.8
$ 2.2
$ 0.9
Note: Detail may not add due to independent rounding. 90-percent confidence bounds reflect uncertainty in unit treatment costs.
Sources: Initial Capital Costs from Exhibit 7.13. Implementation, IDSE, and Monitoring Plan costs from Exhibit 7.7.
Stage/Primacy Agency One-Time Costs from Exhibit H.21
Final Economic Analysis for the Stage 2 DBPR
7-8
December 2005
-------
Exhibit 7.5a Total Annualized Costs for Stage 2 DBPR Rule Activities ($Millions/Year, 3 Percent Discount Rate)
System Costs
System Size
(Population
Served)
Capital Costs
Mean
Value
90 Percent
Confidence Bound
Lower
(5th %tile
Upper
(95th
%tile)
O&M Costs
Mean
Value
90 Percent
Confidence Bound
Lower
(5th %tile
Upper
(95th
%tile)
Non-Treatment Costs
(Point Estimate)
Implement-
ation
IDSE
Monitoring
Plans
Moni-
toring
Operational
Evalations
Total System Costs
Mean
Value
90 Percent
Confidence Bound
Lower
(5th %tile
Upper
(95th
%tile)
Surface Water CWSs
< 10,000
> 10,000
$4.21
$20.60
$2.32
$11.22
$6.23
$28.75
$6.10
$14.33
$3.41
$9.03
$8.83
$21.55
$0.12
$0.09
$0.93
$1.59
$0.05
$0.03
-$0.07
-$1.14
$0.02
$0.11
$11.34
$35.61
$6.76
$20.93
$16.10
$50.97
Surface Water NTNCWSs
< 10,000
> 10,000
$0.27
$0.04
$0.15
$0.02
$0.40
$0.06
$0.57
$0.03
$0.32
$0.02
$0.82
$0.04
$0.01
$0.00
$0.00
$0.00
$0.00
$0.00
$0.02
$0.00
$0.00
$0.00
$0.86
$0.08
$0.49
$0.05
$1.25
$0.11
Ground Water CWSs
< 10,000
> 10,000
$7.41
$4.87
$6.13
$4.37
$8.70
$5.36
$7.20
$6.00
$6.60
$5.64
$7.79
$6.37
$0.30
$0.05
$0.29
$0.10
$0.08
$0.02
$1.05
$0.00
$0.00
$0.00
$16.33
$11.04
$14.45
$10.18
$18.21
$11.90
Ground Water NTNCWSs
< 10,000
> 10,000
TOTAL
$0.57
$0.01
$37.97
$0.48
$0.01
$24.69
$0.65
$0.01
$50.17
$0.75
$0.01
$34.98
$0.69
$0.01
$25.72
$0.81
$0.01
$46.22
$0.06
$0.00
$0.62
$0.00
$0.00
$2.91
$0.01
$0.00
$0.19
$0.42
$0.01
$0.28
$0.00
$0.00
$0.12
$1.80
$0.03
$77.08
$1.65
$0.02
$54.53
$1.95
$0.03
$100.51
State
Costs
$1.71
Total Costs of the Rule
90 Percent
Confidence Bound
Upper
Mean Lower (95th
Value (5th %tile) %tile)
$78.80 $56.24 $102.22
Notes: Detail may not add due to independent rounding. 90 percent confidence bounds reflect uncertainty in technology compliance forecast and unit treatment costs.
Estimates are discounted to 2003 and given in 2003 dollars.
Sources Capital Costs: SW CWS, Exhibit J.2bb; SW NTNCWS, Exhibit J.2bf; GW CWS, Exhibit J.2bj; GW NTNCWS, Exhibit J.2bn.
O&M Costs: SW CWS, Exhibit J.2bc; SW NTNCWS, Exhibit J.2bg; GW CWS, Exhibit J.2bk; GW NTNCWS, Exhibit J.2bo.
Non-Treatment Costs: SW CWS, Exhibit J.2bd; SW NTNCWS, Exhibit J.2bh; GW CWS, Exhibit J.2bl; GW NTNCWS, Exhibit J.2bp.
State Costs; Appendix J, Exhibit J.2as
Final Economic Analysis for the Stage 2 DBPR
7-9
December 2005
-------
Exhibit 7.5b Total Annualized Costs for Stage 2 DBPR Rule Activities ($Millions/Year, 7 Percent Discount Rate)
System Costs
System Size
(Population
Served)
Capital Costs
Mean
Value
90 Percent
Confidence Bound
Lower
(5th %tile
Upper
(95th
%tile)
O&M Costs
Mean
Value
90 Percent
Confidence Bound
Lower
(5th %tile
Upper
(95th
%tile)
Non-Treatment Costs
(Point Estimate)
Implement-
ation
IDSE
Monitoring
Plans
Moni-
toring
Operational
Evaluations
Total System Costs
Mean
Value
90 Percent
Confidence Bound
Lower
(5th %tile
Upper
(95th
%tile)
Surface Water CWSs
< 10,000
> 10,000
$4.53
$23.00
$2.50
$12.53
$6.71
$32.10
$4.86
$11.66
$2.72
$7.35
$7.04
$17.54
$0.15
$0.11
$1.16
$2.06
$0.06
$0.04
-$0.06
-$0.90
$0.01 1 $10.72
$0.08| $36.06
$6.54
$21.27
$15.08
$51.03
Surface Water NTNCWSs
< 10,000
> 10,000
$0.29
$0.05
$0.16
$0.03
$0.43
$0.07
$0.45
$0.02
$0.25
$0.01
$0.66| $0.01
$o.os| $0.00
$0.00
$0.00
$0.00
$0.00
$0.01
$0.00
$0.00| $0.76
$0.0o| $0.08
$0.43
$0.05
$1.11
$0.11
Ground Water CWSs
< 10,000
> 10,000
$7.98
$5.39
$6.60
$4.84
$9.37
$5.94
$5.74
$4.87
$5.26
$4.57
$6.21
$5.16
$0.38
$0.06
$0.36
$0.13
$0.09
$0.02
$0.84
$0.00
$0.00| $15.38
$0.0o| $10.46
$13.53
$9.62
$17.24
$11.31
Ground Water NTNCWSs
< 10,000
> 10,000
TOTAL
$0.61
$0.01
$41.86
$0.51
$0.01
$27.16
$0.70
$0.01
$55.33
$0.60
$0.01
$28.21
$0.55
$0.01
$20.73
$0.65
$0.01
$37.29
$0.07
$0.00
$0.78
$0.00
$0.00
$3.71
$0.01
$0.00
$0.23
$0.33
$0.01
$0.23
$0.00
$0.00
$0.10
$1.62
$0.02
$75.11
$1.48
$0.02
$52.94
$1.77
$0.02
$97.67
State
Costs
$1.70
Total Costs of the Rule
90 Percent
Confidence Bound
Upper
Mean Lower (95th
Value (5th %tile) %tile)
$76.81 $54.64 $99.36
Notes: Detail may not add due to independent rounding. 90 percent confidence bounds reflect uncertainty in technology compliance forecast and unit treatment costs.
Estimates are discounted to 2003 and given in 2003 dollars.
Sources Capital Costs: SW CWS, Exhibit J.2br; SW NTNCWS, Exhibit J.2bv; GW CWS, Exhibit J.2bz; GW NTNCWS, Exhibit J.2cd.
O&M Costs: SW CWS, Exhibit J.2bs; SW NTNCWS, Exhibit J.2bw; GW CWS, Exhibit J.2ca; GW NTNCWS, Exhibit J.2ce.
Non-Treatment Costs: SW CWS, Exhibit J.2bt; SW NTNCWS, Exhibit J.2bx; GW CWS, Exhibit J.2cb; GW NTNCWS, Exhibit J.2cf.
State Costs: Appendix J, Exhibit J.2aw
Final Economic Analysis for the Stage 2 DBPR
7-10
December 2005
-------
7.2 Labor Rates and Laboratory Fees
Labor Rates
Labor costs to PWSs are estimated using hourly labor rates for technical and managerial labor
categories. Labor rates representative of national averages, based on Bureau of Labor Statistics (BLS)
figures as reported in Labor Costs for National Drinking Water Rules (USEPA 2003e), are used in all
analyses. The technical and managerial wage rates vary with system size and include fringe benefits.
Labor rates do not differ between surface and ground water systems.
Exhibit 7.6a shows the technical and managerial rates for each of the nine standard system size
categories used in this EA. Exhibit 7.6b shows the technical and managerial rates according to the system
size categories used to specify Stage 2 monitoring requirements. All rates have been adjusted to 2003
dollars using the Employment Cost Index (ECI) (BLS 2003).
To account for the composition of staff at PWSs of varying sizes, EPA uses only the technical
rate for systems serving fewer than 3,300 people. For systems serving 3,300 or more people, EPA uses a
ratio of 80 percent technical labor to 20 percent managerial labor to arrive at a weighted labor rate. The
final labor rates used for the cost analysis are shown in the combined column (column C) in Exhibits 7.6a
and b.
Exhibit 7.6a System Wage Rates by Standard Size Categories
System Size (Population
Served)
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Labor Rate (2003$/hr)
Technical
A
$21 .44
$23.09
$24.74
$24.74
$25.34
$26.05
$26.05
$31 .26
$31 .26
Managerial Combined
B
$44.36
$47.78
$51 .20
$51 .20
$51 .20
$51 .20
$51 .20
$51 .20
$51 .20
C
$21 .44
$23.09
$24.74
$24.74
$30.51
$31 .08
$31 .08
$35.25
$35.25
Notes: EPA estimates that systems with populations greater than 3,300 use a combination of
operators (technical) and engineers (managerial), with an 80/20 ratio between the two,
respectively.
Source: Labor Costs for National Drinking Water Rules, Exhibits 20 and 21 (USEPA, 2003s)
Final Economic Analysis for the Stage 2 DBPR
7-11
December 2005
-------
Exhibit 7.6b System Wage Rates by Monitoring Size Categories
System Size (Population Served)
Labor Rate (2003$/hr)
Technical
A
Managerial
B
Combined
C
Surface Water and Mixed CWSs
<500
500-3,299
3,300-9,999
10,000-49,999
50,000-249,999
250,000-999,999
1 ,000,000-4,999,999
5M+
$22.55
$24.74
$25.34
$26.05
$28.00
$31.26
$31.26
$31.26
$46.65
$51.20
$51.20
$51.20
$51.20
$51.20
$51.20
$51.20
$22.55
$24.74
$30.51
$31.08
$32.64
$35.25
$35.25
$35.25
Disinfecting Ground Water Only CWSs
<500
500-9,999
10,000-99,999
100,000-499,999
500,000+
$22.35
$24.86
$26.05
$31.26
$31.26
$46.25
$51.20
$51.20
$51.20
$51.20
$22.35
$24.86
$31.08
$35.25
$35.25
Surface Water and Mixed NTNCWSs
<500
500-3,299
3,300-9,999
10,000-49,999
50,000-249,999
250,000-999,999
1 ,000,000-4,999,999
5M+
$22.39
$24.74
$25.34
$26.05
$31.26
N/A
N/A
N/A
$38.84
$51.20
$51.20
$51.20
$51.20
N/A
N/A
N/A
$22.39
$24.74
$30.51
$31.08
$35.25
N/A
N/A
N/A
Disinfecting Ground Water Only NTNCWSs
<500
500-9,999
10,000-99,999
100,000-499,999
500,000+
$22.20
$24.76
$26.05
$31.26
N/A
$45.94
$51.20
$51.20
$51.20
N/A
$22.20
$24.76
$31.08
$35.25
N/A
Notes: EPA estimates that systems with populations greater than 3,300 use a combination of
operators (technical) and engineers (managerial), with an 80/20 ratio between the two,
respectively.
Source: Labor Costs for National Drinking Water Rules, Exhibits 20 and 21 (USEPA, 2003s)
EPA recognizes that there may be significant variation in labor rates across all PWSs. However,
data are not currently available that would allow assignment of labor rates to specific PWSs based on
characteristics such as size, classification, or geographical region. In the absence of such data and
because analyses in this EA are performed on a national level, the data from Labor Costs for National
Drinking Water Rules are used for all systems.
Labor costs attributable to States for administrative tasks are estimated based on an average
annual full-time-equivalent (FTE) labor cost, including overhead and fringe benefits, of $65,255 (2001$).
This rate was established based on data from the 2001 State Drinking Water Needs Analysis (ASDWA
2001). For use in the Stage 2 EA analyses, the $65,255 annual rate was updated to 2003 dollars
($70,132) using the ECI (BLS 2003) and converted to an hourly basis (1 FTE = 2,080 hours) to establish
a State rate of $33.60 per hour.
Final Economic Analysis for the Stage 2 DBPR
7-12
December 2005
-------
Laboratory Fees
A laboratory fee, expressed as a cost per sample, is associated with TTHM and HAAS monitoring
costs for IDSE and additional routine monitoring. Based on laboratory costs reported in the 1996 ICR,
EPA estimated the laboratory fee at $200 per sample for all size categories. This estimate does not
include shipping costs. For systems serving 10,000 people or more, a shipping cost of $10 was added to
account for the fact that larger systems often have in-house laboratory facilities and can take advantage of
bulk discounts. For systems serving fewer than 10,000 people, a shipping cost of $40 is added because
very few small systems have in-house laboratory facilities and they have less opportunity to take
advantage of bulk discounts. Laboratory fees are not expected to differ substantially between disinfecting
ground water systems and surface water systems.
7.3 Non-Treatment Costs for Systems and States/Primacy Agencies
This section presents the estimated national costs for systems and States/Primacy Agencies to
perform Stage 2 DBPR activities that are not related to treatment. These activities have been described in
Chapter 1 and are described in detail in Appendix H. The following subsections provide a brief summary
of each activity and key assumptions used to estimate costs for each:
7.3.1 Rule Implementation
7.3.2 Initial Distribution System Evaluations
7.3.3 Monitoring Plans
7.3.4 Additional Routine Monitoring
7.3.5 Operational Evaluations
7.3.6. Results (One-Time and Yearly Costs)
Appendix H provides the methodology and calculations for all non-treatment-related costs. Note
that cost calculations in Appendix H are performed using system inventory data broken out by system
size categories that are different from the standard nine system size categories used elsewhere in this EA.
EPA believes these alternate categories to be more appropriate for establishing the number of samples
required per system. Section 7.3.6 summarizes the one-time and yearly costs for all non-treatment-related
Stage 2 DBPR activities.
Monitoring requirements for the Stage 1 DBPR were dependent on the system type (NTNCWS or
CWS), the source water type (surface or ground) and the number of plants per system. EPA has
identified several potential issues when requirements are based on the number of plants per system.
These include:
• The number of required sample sites may be either excessive or insufficient to represent
TTHM and HAAS occurrence throughout the system, particularly in situations where a small
system uses multiple plants, or where a very large system uses a small number of plants.
• Plant-based sampling requirements for mixed systems (i.e., those receiving disinfected
surface water and ground water in their distribution system) may be excessive, depending
upon the system's characteristics.
• Plant-based monitoring requirements pose unique implementation issues for systems that use
temporary supplies during the year.
Final Economic Analysis for the Stage 2 DBPR 7-13 December 2005
-------
An alternative population-based approach was included in the proposed Stage 2 DBPR, and many
positive comments were received. For this reason and those listed above, a population-based monitoring
approach was adopted for the final rule whereby IDSE and Stage 2 monitoring requirements are based
only on system type, source water type, and retail population served.
7.3.1 Rule Implementation
Public Water Systems
All systems subject to the Stage 2 DBPR will incur one-time costs for staff to read the rule and
become familiar with its provisions and to be trained on its requirements. The technical and managerial
labor rates presented in Section 7.2 were used along with estimates of labor hours to generate
implementation costs for all systems. The mix of labor rates used to estimate implementation costs varies
by activity and system size as summarized in Appendix H.
States/Primacy Agencies
State/Primacy Agency implementation activities include:
• Public notification
Regulation adoption and program development
• Training State/Primacy Agency staff
• Training PWS staff
• Technical assistance
Updating the data management system
The number of FTEs required per activity was estimated by EPA based on previous experience
with other rules. State/Primacy Agency activities include public notification (0.1 FTEs), regulation
adoption and program implementation (0.50 FTEs), training State/Primacy Agency staff (0.25 FTEs),
training PWS staff and technical assistance (1.00 FTE), and updating the management system (0.10
FTEs). The labor rates used to estimate State/Primacy Agency costs are presented in Section 7.2. The
number of States and territories included the 50 States (or EPA regions where States do not have
primacy), 6 territories, and 1 tribal government.
7.3.2 Initial Distribution System Evaluations
Public Water Systems
The purpose of the IDSE is to identify compliance monitoring sites that are representative of the
highest TTHM and FŁAA5 levels in the distribution system. IDSEs can be performed by either (1)
conducting standard monitoring or (2) completing an System Specific Study (SSS) that may include
existing monitoring data or hydraulic modeling results. NTNCWSs serving fewer than 10,000 people are
not subject to the IDSE requirements of the Stage 2 DBPR. A CWS or NTNCWS does not have to
perform the IDSE if: (1) all Stage 1 DBPR compliance samples are less than or equal to 40 • g/L for
TTHM and 30 • g/L for HAA5, or (2) the system serves fewer than 500 people and qualifies for the very
small system waiver.
Systems performing an IDSE will incur costs for evaluating their distribution systems to identify
sampling sites, preparing an IDSE monitoring plan, sampling, and reporting results. Systems electing to
Final Economic Analysis for the Stage 2 DBPR 7-14 December 2005
-------
complete an SSS may not incur sampling costs; however, they will still incur labor costs for preparing an
IDSE study plan and preparing the report. Also, some systems that do not perform the IDSE may have
more sampling sites required under the Stage 2 DBPR than under the Stage 1 DBPR and, thus, will incur
a small labor cost for selecting new Stage 2 sites. A detailed description of the process that EPA used to
estimate the total national IDSE costs, as well as detailed calculation tables, are presented in Appendix H.
States/Primacy Agencies
States/Primacy Agencies also will incur costs as a result of the IDSEs. The activities they will
conduct include analyzing PWS IDSE reports, determining which systems cannot receive a very small
system waiver, consulting with PWSs, and IDSE recordkeeping. The estimates of burden and costs
depend on the IDSE option (standard monitoring plan or SSS) and system size. Total costs for
States/Primacy Agencies for the IDSE are estimated to be $2.2 million.
7.3.3 Monitoring Plans
Public Water Systems
Most systems are required to prepare a monitoring plan for routine Stage 2 DBPR monitoring.
Many base the new monitoring plan on the IDSE results.
As described in Section 7.1.2, some ground water systems may choose to add disinfection to meet
the requirements of the GWR after it becomes final. Because the GWR is expected to be promulgated
within 8 months after the Stage 2 DBPR is promulgated, EPA expects new systems adding disinfection to
meet GWR requirements to simultaneously achieve compliance with Stage 2 MCLs. Therefore, as
discussed in Chapter 3 of this EA, these systems are not included in the treatment baseline. The IDSE
will likely not apply to these systems because they are expected to add disinfection after the IDSE
requirement is complete. These systems will, however, need to prepare Stage 2 monitoring plans.
Estimates of labor hours for systems to complete the monitoring plans are in Appendix H.
States/Primacy Agencies
States/Primacy Agencies will also incur costs to review and approve monitoring plans submitted
by systems. Estimates of burden and cost for monitoring plans are found in Appendix H. The estimated
costs for State/Primacy Agency review and approval of monitoring plans is $926,016.
Final Economic Analysis for the Stage 2 DBPR 7-15 December 2005
-------
7.3.4 Additional Routine Monitoring
Public Water Systems
EPA has established a population-based monitoring approach for the Stage 2 DBPR, where
monitoring requirements are no longer based on number of plants per system as under the Stage 1 DBPR.
As a result, systems may have the same, fewer, or more monitoring sites for the Stage 2 DBPR compared
to Stage 1, depending on number of plants and how the State/Primacy Agency determined their Stage 1
monitoring requirements. Some systems will have the same monitoring requirements under the Stage 2
DBPR as they did under the Stage 1 DBPR, and will, therefore, incur no additional costs for this activity.
In other cases, incremental costs will be negative (that is, costs will be reduced) under the Stage 2 DBPR
for systems that have fewer sites than the Stage 1 DBPR (see Exhibit H.8 in Appendix H). Systems with
more monitoring sites under the Stage 2 DBPR than under the Stage 1 DBPR will incur costs. Changes in
numbers of samples from Stage 1 to Stage 2 and total costs for additional routine monitoring are provided
in Appendix H.
As described in Section 7.1.2, some ground water systems may choose to add disinfection to meet
the requirements of the GWR after it becomes final. These are systems that do not disinfect currently that
will elect to install disinfection and monitor for DBFs for the first time to comply with the GWR
requirements. Because the GWR is expected to be promulgated within 8 months after the Stage 2 DBPR
is promulgated, EPA expects new systems adding disinfection to meet GWR requirements to
simultaneously achieve compliance with Stage 2 MCLs. Therefore, these systems are not included in the
treatment baseline. The IDSE will likely not apply to these systems because they are expected to add
disinfection after the IDSE requirement is complete. Systems adding disinfection for the GWR, however,
will need to monitor DBFs for the first time under Stage 2. The derivation of routine monitoring costs for
systems that add disinfection to achieve compliance with the GWR is provided in Appendix H. The
actual monitoring costs for newly disinfecting ground water systems may be different from those
estimated here depending upon the details of the final GWR.
Costs for additional routine monitoring include laboratory analysis and labor for taking the
sample.
States/Primacy Agencies
States/Primacy Agencies will incur costs related to review and evaluation of monitoring data
submitted by systems. EPA estimates states will require 0.4 FTEs to track compliance data, update data
management systems, and file records. These costs are estimated to total $1.6 million per year.
Final Economic Analysis for the Stage 2 DBPR 7-16 December 2005
-------
7.3.5 Operational Evaluations
Public Water Systems
To address excess DBF levels that may occasionally occur (but do not cause rule violations), the
Stage 2 DBPR contains a provision for operational evaluations. An operational evaluation level is
exceeded when a sample result when multiplied by two, added to the sum of the previous two quarters'
samples, and divided by four would result in a concentration greater than 80 • g/L for TTHM or 60 • g/L
for HAAS. The equation for calculating the concentration for an operational evaluation (COL ) is:
where Qj is the concentration of the of the DBF two quarters ago, Q2is the concentration of the DBF in
the previous quarter, and Q3 is the concentration of the DBF in the current quarter. If COL is greater than
the maximum contaminant level (MCL) for the DBF measured, then the operational evaluation level is
exceeded. For example, if a system had first-quarter and second-quarter results of 75 |o,g/L for TTHM and
had a third-quarter result of 90 |o,g/L, then the calculation would yield:
(75 ng/L +75 ng/L + 2*(90 jig/L))/4 = 82.5
meaning the operational evaluation level is exceeded.
If an operational evaluation level is exceeded, systems must conduct an "operational evaluation"
to investigate and document the cause. Appendix H provides estimates of the number of systems
expected to exceed an operational evaluation level and the associated costs.
States/Primacy Agencies
States will incur some costs to review operational evaluations submitted by systems. Appendix H
estimates the time and costs that will be spent by States to review these operational evaluations. Costs for
States/Primacy Agencies to review the operational evaluations are estimated to be $114,173 per year.
7.3.6 Results (One-Time and Yearly Costs)
Exhibit 7.7 summarizes the one-time system costs for implementation, IDSEs, and development
of Stage 2 monitoring plans, along with the yearly costs for additional routine monitoring and operational
evaluations. Note that IDSE costs make up the majority of the one-time costs. Total one-time system
costs for implementation, IDSEs, and Stage 2 monitoring plans are approximately $73.6 million
(including $12.3 million for implementation activities, $57.4 million for IDSE activities, and $3.8 million
for monitoring plans). Total annual system costs for additional routine monitoring and operational
evaluations are approximately $0.6 million (including $0.4 million for additional routine monitoring, $0.2
million for operational evaluations). As shown in Exhibit 7.7, additional routine monitoring costs are
positive for some systems and negative for others (because of the redefined monitoring requirements from
a plant-based to a population-based approach). One-time and annual costs for States/Primacy Agencies
are summarized in Appendix H, Exhibit H.21.
Final Economic Analysis for the Stage 2 DBPR 7-17 December 2005
-------
Exhibit 7.7 Summary of System Costs for Non-Treatment Related Stage 2 DBPR
Rule Activities (One-Time and Yearly)
System Size
(Population Served)
One Time-Costs ($)
Implementation
IDSE
Stage 2
Monitoring
Plans
Total
Annual Costs ($)
Additional Routine
Monitoring
Operational Evaluations
Total
ABC D=A+B+C E F G=E+F
Surface Water and Mixed CWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
National Totals
$ 244,635
$ 498,740
$ 363,678
$ 640,272
$ 623,055
$ 1,212,306
$ 239,846
$ 212,143
$ 15,227
$ 4,049,902
$ 447,582
$ 912,489
$ 3,140,721
$ 5,529,388
$ 8,379,826
$ 17,851,398
$ 6,426,075
$ 6,113,574
$ 731,365
$ 49,532,418
$ 76,408
$ 155,774
$ 181,839
$ 320,136
$ 258,721
$ 461,867
$ 93,524
$ 93,984
$ 11,566
$ 1,653,819
$ 768,625
$ 1,567,003
$ 3,686,237
$ 6,489,795
$ 9,261,602
$ 19,525,571
$ 6,759,446
$ 6,419,700
$ 758,158
$ 55,236,138
$ (52,103)
$ (106,222)
$ (331,698)
$ (583,969)
$ 953,611
$ (2,477,619)
$ 216,219
$ 276,429
$ 36,517
$ (2,068,834)
$ 534
$ 1,089
$ 3,011
$ 5,301
$ 20,870
$ 98,959
$ 39,199
$ 38,699
$ 5,076
$ 212,739
$ (51,568)
$ (105,133)
$ (328,686)
$ (578,668)
$ 974,481
$ (2,378,660)
$ 255,418
$ 315,128
$ 41,593
$ (1,856,095)
100% Ground Water CWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
National Totals
$ 1,596,365
$ 1,975,736
$ 894,469
$ 1,085,531
$ 492,179
$ 797,681
$ 88,492
$ 46,421
$ 2,180
$ 6,979,054
$ 221,266
$ 273,849
$ 1,931,945
$ 2,344,617
$ 1,063,047
$ 1,642,671
$ 182,233
$ 166,938
$ 5,964
$ 7,832,529
$ 77,140
$ 95,472
$ 507,572
$ 615,991
$ 279,290
$ 320,895
$ 35,599
$ 30,689
$ 1,868
$ 1,964,515
$ 1,894,770
$ 2,345,057
$ 3,333,986
$ 4,046,139
$ 1,834,515
$ 2,761,247
$ 306,325
$ 244,048
$ 10,013
$ 16,776,099
$ 96,758
$ 119,753
$ 553,626
$ 671,883
$ 304,631
$ 117,333
$ 13,017
$ (92,140)
$ (25,261)
$ 1,759,600
$
$
$
$
$
$
$
$
$
$
$ 96,758
$ 119,753
$ 553,626
$ 671,883
$ 304,631
$ 117,333
$ 13,017
$ (92,140)
$ (25,261)
$ 1,759,600
Surface Water and Mixed NTNCWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
National Totals
$ 46,558
$ 63,891
$ 23,602
$ 20,707
$ 6,591
$ 3,263
$
$ 740
$
$ 165,353
$
$
$
$
$
$ 46,876
$
$ 23,725
$
$ 70,601
$
$
$ 5,245
$ 4,602
$ 1,216
$ 1,303
$
$ 313
$
$ 12,678
$ 46,558
$ 63,891
$ 28,847
$ 25,309
$ 7,807
$ 51,442
$
$ 24,778
$
$ 248,632
$
$
$
$
$ 25,473
$
$
$ 3,860
$
$ 29,333
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$ 25,473
$
$
$ 3,860
$
$ 29,333
100% Ground Water NTNCWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
National Totals
Grand Total All Systems
$ 498,070
$ 425,353
$ 131,289
$ 55,048
$ 4,781
$ 2,082
$ 189
$ 215
$
$ 1,117,027
$ 12,311,336
$
$
$
$
$
$ 855
$ 78
$
$
$ 932
$ 57,436,480
$ 74,304
$ 63,456
$ 51,924
$ 21,771
$ 1,891
$ 1,143
$ 104
$ 192
$
$ 214,785
$ 3,845,797
$ 572,375
$ 488,809
$ 183,212
$ 76,819
$ 6,672
$ 4,080
$ 371
$ 406
$
$ 1,332,744
$ 73,593,613
$ 175,519
$ 149,893
$ 253,333
$ 106,220
$ 9,226
$ 13,563
$ 1,233
$ 2,085
$
$ 711,072
$ 431,171
$
$
$
$
$
$
$
$
$
$
$ 212,739
$ 175,519
$ 149,893
$ 253,333
$ 106,220
$ 9,226
$ 13,563
$ 1,233
$ 2,085
$
$ 711,072
$ 643,910
Notes: Detail may not add due to independent rounding.
Source: Appendix H, Exhibit H.16. Costs for Stage 2 monitoring plans and additional routine monitoring include those costs for systems that are projected
to add disinfection to comply with the Ground Water Rule.
7.4 Technology Unit Costs
Available treatment technologies for reducing DBFs were identified during the Stage 2
Microbial-Disinfectants/Disinfection Byproducts (M-DBP) Federal Advisory Committees Act (FACA)
deliberations (USEPA 2000n). They include alternative disinfectants such as ozone, ultraviolet light
Final Economic Analysis for the Stage 2 DBPR
7-18
December 2005
-------
(UV), and chlorine dioxide, as well as DBF precursor removal treatment technologies such as
microfiltration or ultrafiltration. Converting to chloramines for residual disinfection was also identified as
a relatively inexpensive treatment technology that can limit DBF formation in many distribution systems.
Unit cost estimates for these treatment technologies are in units of dollars per plant for initial
capital and yearly O&M activities. Household unit costs are in units of dollars per household per year.
Derivation of unit costs for a wide range of plant sizes, represented by different design and average daily
flow rates, are provided in the document, Technologies and Costs for Control ofMicrobial Contaminants
and Disinfection Byproducts (USEPA 2005n). EPA uses mean design flow and average daily flow for
each of the nine system size categories (shown in Exhibit 3.4) to estimate unit costs for each treatment
technology for each system type, source water type, and size category.
Section 7.4.1 describes the treatment technologies and operating conditions that are used to
predict national treatment costs of the Stage 2 DBPR. Section 7.4.2 discusses alternatives to treatment
identified by EPA and others during the FACA process and explains why these processes are not included
in the cost analysis for the Stage 2 DBPR. Section 7.4.3 discusses uncertainty in the unit costs and
explains how the uncertainties were explicitly accounted for in the Stage 2 cost model.
7.4.1 Treatment Technologies Used to Estimate Costs
This section discusses the treatment technologies available for surface water plants first, followed
by the treatment technologies used for ground water plants. A discussion of uncertainties in unit costs
follows. A summary of unit treatment costs is presented at the end of the section. Appendix I supports
the data in Section 7.4.1, showing the detailed derivation of unit costs over the entire range of expected
plant flows.
Treatment Technologies for Surface Water Plants
Exhibit 7.8a lists the treatment technologies that are available to surface water plants for
complying with Stage 1 DBPR and Stage 2 DBPR regulatory alternatives. The SWAT decision tree, as
noted in Appendix A, includes both installation of treatment technologies and changes to operational
practices. Although changes in operational practices may result in small increases in chemical costs or
minor capital improvements, EPA assumes their costs are negligible as compared to the costs of the
advanced treatment technologies (e.g., UV, ozone, granular activated carbon, microfiltration/ultra-
filtration). Therefore, the predicted costs for the Stage 2 DBPR do not include costs for operational
changes (Section 7.8 summarizes uncertainties in national cost estimates). The treatment technologies
presented in the SWAT Decision Tree (Appendix A) that are not costed in this EA because they are
assumed to incur negligible costs are:
• Adjusting disinfection dose
• Moving the point of chlorination
• Enhanced coagulation/enhanced softening (required for the Stage 1 DBPR)
• Turbo coagulation
Although EPA assumes that large surface water plants serving 100,000 people or more can select
any of the treatment technologies presented in Exhibit 7.8a, small plants may not be able to use a
particular treatment technology because of operational constraints or other reasons. Limitations on the
Final Economic Analysis for the Stage 2 DBPR 7-19 December 2005
-------
use of treatment technologies by small systems, summarized in the second column in Exhibit 7.8a, were
identified during the small surface water expert review process (see Appendix A for details).
The last column in Exhibit 7.8a identifies the design criteria and operating conditions for each
treatment technology in this EA for which costs are calculated. To capture the range of costs, the
Technology and Cost (T&C) document evaluated treatment technologies over a range of possible influent
water qualities and operating conditions (USEPA 2005n). For the purposes of estimating the costs of the
Stage 2 DBPR, the Technical Workgroup (TWO) selected water quality and operating parameters that
would capture the typical circumstances under which plants may use the treatment technology. EPA does
not propose that all systems would operate under these conditions, but they suffice to generate capital and
O&M costs typical of the range of system types and sizes. While these assumptions simplify the true
variety of operating conditions, EPA believes they capture reasonable estimates of national costs. The
uncertainties associated with selecting these operational parameters and conditions are summarized in
Section 7.8.
Appendix A presents additional information on how each treatment technology was modeled in
SWAT, including log removal and disinfection credits for Giardia and viruses. Note that, for UV, the
disinfection credit for viruses was 2.0 logs for all runs. The primary disinfectant (usually chlorine) dose
was then modified to meet the remaining disinfection requirements.
Some advanced treatment technologies were considered in combination for surface water systems
to meet the Stage 2 DBPR (as modeled in SWAT). When treatment technologies are a combination of
two or more unit processes (e.g., Granular Activated Carbon—20-Minute Contact Time (GAC20) with
Advanced Disinfectants such as Chlorine Dioxide or Ozone), the technology unit costs are assumed to be
the sum of the costs for each unit process. For example, the cost for implementing GAC20 with an
Advanced Disinfectant (e.g., Ozone) is equal to the sum of GAC20 and Ozone unit costs. This may over-
estimate unit costs since some economies of scale are expected when two or more treatment technologies
are installed at the same time.
Final Economic Analysis for the Stage 2 DBPR 7-20 December 2005
-------
Exhibit 7.8a Treatment Technologies for Surface Water Plants
Technology
Switching to
chloramines (CLM) as a
residual disinfectant
Chlorine dioxide (CLO2)
Ultraviolet light
disinfection (UV)1
Ozone
Microfiltration /
Ultrafiltration (MF/UF)
Granular activated
carbon filtration, empty-
bed contact time of 1 0
minutes (GAC10)
GAC10 + Advanced
Disinfectants
Granular activated
carbon filtration, empty-
bed contact time of 20
minutes (GAC20)
GAC20 + Advanced
Disinfectants
Membranes (MF/UF +
nanofiltration [IMF])
Constraints
Can be used alone or in
conjunction with all other
treatment technologies in
this exhibit
Assumed not practical for
systems fewer than 1 00
people
None
Assumed not practical for
systems serving fewer
than 100 people
None
Assumed not practical for
small systems serving
fewer than 10,000 people
Assumed not practical for
small systems serving
fewer than 10,000 people
None
None
None
Design Criteria and Operating Conditions
Ammonia dose = 0.55 mg/L (output from SWAT)
EPA assumed that plants will not build a new contact
basin for chlorine dioxide
CIO2 dose = 1 .25 mg/L
Median water quality parameters —
UV254 = 0.051 cm'1, turbidity = 0.1 NTU, alkalinity = 60
mg/L as CaCO3, hardness = 100 mg/L as CaCO3
Dose = 40 mJ/cm2
Design dose = 3.2 mg/L, contact time = 12 minutes2
Median water quality parameters —
Temperature=10°C, disposal to sewer
Reactivation frequency = 360 days3
On-site regeneration
Chlorine dioxide as the advanced disinfectant
Reactivation frequency = 360 days3
On-site regeneration
Reactivation frequency = 90 days3
Onsite regeneration used for systems serving > 10,000
Media replacement used for systems serving < 10,000
Systems serving > 10,000 (GAC20 + chlorine dioxide)
Systems serving 100 - 9,999 (GAC20 + ozone)
Systems serving < 1 00 (GAC20 + UV)
On-site media regeneration used for systems serving >
10,000
Media replacement used for systems serving < 10,000
Median water quality parameters —
MF/UF: 10°C, disposal to sewer
NF: 10°C, ocean discharge
Notes:
1 Available for Stage 2 DBPR regulatory alternatives only; not considered available for the Stage 1 DBPR. UV was
assumed to be used as a supplement to chlorine to achieve some of the required Giardia and virus inactivation,
thereby reducing chlorine dosages.
2 Dose does not consider Cryptosporidium inactivation, and, therefore, may not represent what systems would do
to meet Long Term 2 Enhanced Surface Water Treatment Rule (LT2ESWTR) requirements. However, the higher
dose is accounted for in the LT2ESWTR EA.
3 Median reactivation frequency generated by SWAT.
Source: T&C document (USEPA 2005n), FACA deliberations for Stage 2 treatment technologies (USEPA 2000n).
SWAT Decision Tree (Appendix A), and Small Surface Water Delphi Groups (Appendix A).
Final Economic Analysis for the Stage 2 DBPR
7-21
December 2005
-------
Treatment Technologies for Ground Water Plants
Exhibit 7.8b lists the treatment technologies used to estimate costs of the Stage 1 DBPR and
Stage 2 DBPR regulatory alternatives for ground water plants. Fewer treatment technologies are
applicable to disinfecting ground water plants than to surface water plants. As noted in Appendix B, the
ICR Ground Water Delphi process concluded that disinfecting ground water systems would choose
primarily from four treatment technologies—conversion to chloramines, ozone, GAC20, and
nanofiltration. Limitations on treatment technology use by small systems, as identified during the small
ground water system expert review process, are provided in the second column.
Because UV was still very much an emerging treatment technology when the Ground Water
Delphi process was conducted (Spring, 2000), UV was not considered as a treatment option for large
ground water plants for either the Stage 1 or Stage 2 DBPRs. UV was, however, considered an available
treatment technology for small ground water systems to meet Stage 2 DBPR requirements. The
small-system expert reviewers assumed UV would be used instead of chlorine to achieve 4.0 logs of virus
inactivation in all circumstances. Current UV validation techniques as outlined in the UV Disinfection
Guidance Manual (USEPA 2003g) are not currently feasible for validation of 4-log virus inactivation.
Therefore it will be necessary to use two UV reactors in series, each achieving 2-log virus inactivation, to
achieve the desired 4-log virus removal. A dose of 200 millijoules per centimeter square (mJ/cm2) is
assumed necessary to achieve 2-log virus inactivation, much higher than the UV dose of 100 mJ/cm2
required by the LT2 ESWTR for 2 log viral inactivation (USEPA 2005n). The unit cost estimates shown
in Exhibit 7.11 for UV ($/plant) for small ground water plants represent new estimates for two reactors in
series with a higher dose (200 mJ/cm2).3 The cost assumptions for ground water systems to achieve
compliance with the Stage 2 DBPR are therefore probably overstated.
3Note, EPA updated the 40 mJ/cm2 UV unit costs based on data obtained for recent installations of this
technology. Similar data for 200 mJ/cm2 UV systems were not available within the time frame required to include in
this analysis
Final Economic Analysis for the Stage 2 DBPR 7-22 December 2005
-------
Exhibit 7.8b Treatment Technologies for Disinfecting Ground Water Plants
Treatment
Technology
Switching to
CLM as a
Residual
Disinfectant
UV2
Ozone
GAC20
Nanofiltration
Constraints
This advanced treatment technology
can be used alone or in conjunction
with all of the following treatment
technologies
Was considered only for small
systems with populations of fewer
than 10,000 people. Requires two
reactors in series.
Assumed not practical for systems
serving fewer than 100 people
None
None
Design Criteria and Operating Conditions
Ammonia dose = 0.15 mg/L1
Median water quality parameters —
UV254 = 0.051 cm'1, turbidity = 0.1 NTU, alkalinity =
60 mg/L as CaCO3, hardness = 100 mg/L as CaCO3
Dose = approximately 200 mJ/cm
Design dose = approximately 3.2 mg/L, contact time
= 12 minutes
Reactivation frequency = 240 days3
On-site regeneration used for systems serving >
10,000
Media replacement used for systems < 10,000
Median water quality parameters —
Temperature=10°C, ocean discharge
Notes:
1 Dose based on decisions from the ICR Ground Water Delphi Group.
2 Available for Stage 2 DBPR regulatory alternatives only; not considered available for the Stage 1 DBPR.
3 Reactivation frequency based on decisions from the ICR Ground Water Delphi Group.
Source: T&C document (USEPA 2005n), FACA deliberations for Stage 2 treatment technologies (USEPA 2000n),
and ICR and Small Ground Water Delphi Groups (Appendix B).
Unit Costs for the Stage 2 DBPR
Capital and O&M unit costs for each treatment technology are derived from the T&C document
(USEPA 2005n). The T&C document contains between 16 and 20 point estimates of capital and O&M
costs over the range of expected design and average flow rates. Appendix I displays these point estimates
for the treatment technologies, design criteria, and operating conditions listed in Exhibits 7.8a and b. In
previous T&C drafts, non-linear cost curves were generated for specific flow ranges based on a more
limited set of point estimates. Because the number of point estimates for the unit costs was increased to
better represent the full range of expected flows, EPA believes that direct straight-line interpolation
between the point values is adequate for characterizing the changes in unit costs as flow increases or
decreases. Along with the 16 to 20 point estimates, Appendix I graphically shows the relationship of unit
cost and flow (i.e., the point estimates connected by straight lines).
Unit treatment costs for each system type and size category are estimated using (1) the capital and
O&M cost data in Appendix I and (2) the mean design and average daily flow values presented in Exhibit
3.4. For example, the design flow for UV for surface water plants in CWSs serving between 10,000 and
49,999 people is estimated to be 5.324565 millions of gallons per day (MGD) (the value in Exhibit 3.4 is
rounded to 5.325 MGD). Exhibit 1.5 shows that the capital cost for UV surface water plants is $362,965
for a design flow of 3.5 MGD and $544,728 for a design flow of 7.0 MGD. The cost for a 5.325 MGD
plant can be calculated by linear interpolation as:
Unit Cost
= $362,965 + ($544,728 - $362,965) * (5.324565 MGD - 3.5 MGD)/(7.0
MGD-3.5 MGD)
Final Economic Analysis for the Stage 2 DBPR
7-23
December 2005
-------
Unit Cost =$457,719
(Note, detail may not exactly equal value in Exhibit 7. lOa due to independent rounding.)
Household unit treatment costs ($/HH/year) are estimated in order to provide a potential and
approximate measure of the increase in water bills that is expected to result from the Stage 2 DBPR.
Steps for deriving household unit costs are outlined below.
• Step 1: Capital costs of treatment are amortized over a 20 year period using the cost-of-
capital rates summarized in Exhibit 7.9. These rates are derived from Development of Cost of
Capital Estimates for Public Water Systems, Final Report (USEPA 200 le).4 For each
treatment technology, amortized capital costs are added to annual O&M costs to produce
annual treatment costs in units of dollars per plant per year ($/plant/yr).
• Step 2: Results from step 1 ($/plant/yr) are converted to dollars per 1000 gallons per year
($/kgal/yr) using the following formula:
CHH = CP/(ADF*365* 1000)
where: Cjjjj = unit cost per household in $/kgal
CP = unit cost per plant $/plant/yr
ADF = average daily flow in mgd
Step 3: Estimated household usage rate (in units of 1,000 gallons per year, or kgal/yr) is used
to convert results from step 2 ($/kgal/yr) to dollars per household per year ($/FiFi/yr)
according to the formula:
^AHH = (-HH QHH
where: CAKK = annual household unit cost ($/FiFi/yr)
QHH = annual household usage rate (kgal/yr)
Exhibit 7.9 summarizes the annual household usage rates used to estimate household costs in this
chapter. These are from the Baseline Handbook, as derived from the 1995 CWSS5. EPA recognizes that
there may be significant variation in household water usage between specific PWSs (table C.4.2.3 of the
handbook presents confidence intervals around these estimates), but believes that mean usage rate values
are adequate for characterizing household costs.
4 Cost-of-capital estimates are used to account for interest payments that PWSs may incur and pass along to
customers in the form of water bill increases. These rates may be different than social discount rates (3 and 7
percent) used elsewhere in the economic analysis. Social discount rates are more appropriate for estimating
economic impacts on a national level. See EPA's Guidelines for Preparing Economic Analyses (USEPA 2000J) for
a full discussion of the use of social discount rates in the evaluation of policy decisions.
5 Note that household usage rates for the affordability analysis in Chapter 8 are median values from 1995
CWSS data, whereas the values in Exhibit 7.13 are mean values.
Final Economic Analysis for the Stage 2 DBPR 7-24 December 2005
-------
Exhibit 7.9 Household Cost Inputs
System Size
(Population
Served)
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Mean Annual
Water Usage per
Household
(kgal/year)
A
83
83
104
87
97
109
119
125
125
Public
Cost of
Capital
Rate
B
5.31%
5.31%
5.58%
5.58%
5.58%
5.20%
5.24%
5.24%
5.24%
Private
Cost of
Capital
Rate
C
6.22%
6.22%
6.22%
6.22%
6.22%
5.66%
6.27%
6.27%
6.27%
Note: Data derived from 1995 CWSS
Sources:
(A) Derived from the Third Edition of the Baseline Handbook (USEPA,
2001 c) Table C.4.2.3, all systems. Rates for systems serving < 500
people revised based on further analysis by EPA.
(B) and (C) Development of Cost of Capital Estimates for Public Water
Systems, Final Report (USEPA, 2001 d).
As an example, consider the derivation of the household unit costs for UV for a CWS surface
water system serving between 10,000 and 49,999 people. The capital cost is $457,719 as calculated
above. The O&M cost from Exhibit 7.1 Ob is $15,274/plant/year. In that size category 90.03 percent of
households are served by public systems and 9.97 percent by private systems (see Exhibit 3.3). So the
discount rate to be used for amortizing the capital is then:
i = .9003*(0.052) + .0997*(0.0566) = 0.0525
Amortizing the capital at a 5.25 percent discount rates gives:
^cap-arm
= $457,719*0.0525*(1 + 0.0525)20/((1 + 0.0525)20 - 1) = $37,511.12/plant/year
Cp = $37,511.12 + $15,274.16 = $52,785.28/plant/year
This is the result of step 1 listed above. Step 2 involves multiplying the cost by the average daily flow to
obtain a unit cost per thousand gallons of water.
CHH = ($52,785.28/plant/year)/[(2.34Mgal/plant/d)*(365d/yr)*(1000kgal/Mgal)] = $0.0618/kgal/yr.
Step 3 involves multiplying by the household usage rate, which from Exhibit 7.9 is 109 kgal/yr
CAHH = $0.06/kgal/yr* 109kgal/HH/yr = $6.73
This is the value shown in Exhibit 7. lOc.
Exhibits 7.10a-b and 7.11a-b summarize annual O&M costs ($/plant/year) and capital costs
($/plant) for surface and ground water treatment technologies, respectively. Note that unit costs are
Final Economic Analysis for the Stage 2 DBPR
7-25
December 2005
-------
different for surface and ground water plants in each size category because they are based on different
mean design and average daily flows per plant as shown in Exhibit 3.4. Costs are provided for each
treatment technology and for each of the nine population size categories for plants in CWSs. Unit costs
for NTNCWSs are not presented, but can be derived from the data in Appendix I using NTNCWS flows,
as summarized in Exhibit 3.4.
Exhibits 7. lOc and 7.1 Ic summarize household unit treatment costs ($/HH/year) for surface and
ground water treatment technologies, respectively. The household unit costs in Exhibits 7. lOc and 7.1 Ic
represent only the household costs for installation and operation of the treatment technology and do not
include costs for other items such as the IDSE or monitoring. A description of the process used to
generate distributions of possible household treatment costs is provided in Section 7.5.3.
Final Economic Analysis for the Stage 2 DBPR 7-26 December 2005
-------
Exhibit 7.1 Oa Capital Unit Costs ($/Plant) for CWS Surface Water Plants
System Size
(Population
Served)
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Plant Design
Flow (MOD)
0.02
0.08
0.21
0.57
1.42
5.32
8.26
29.89
200.95
CLM
$29,104.44
$29,104.44
$33,047.01
$40,954.40
$83,772.20
$83,772.20
$85,666.74
$172,356.04
$485,633.73
CLO2
$38,637.25
$41,950.61
$80,980.80
$201,714.84
$218,640.92
$341,382.88
$877,596.16
UV
$12,389.73
$23,031.80
$45,188.94
$87,869.72
$318,958.94
$457,719.08
$645,428.85
$2,142,573.70
$7,756,654.16
Ozone
$401 ,494.03
$610,467.18
$989,688.20
$1,992,626.96
$2,558,175.82
$5,726,786.46
$25,393,578.52
MF/UF
$195,605.59
$373,422.89
$668,228.42
$1,032,468.83
$2,002,992.32
$5,831,476.20
$8,469,691.08
$26,059,504.89
$146,969,927.83
GAC10
$2,662,884.51
$3,622,545.65
$8,993,844.94
$36,851,865.10
GAC10+AD
/
/
/
/
/
$2,864,599.35
$3,841,186.57
$9,335,227.82
$37,729,461.26
GAC20
$49,239.18
$120,214.53
$274,674.40
$630,258.56
$1,726,876.12
$4,388,911.73
$6,043,134.98
$15,370,784.57
$64,692,915.84
GAC20+AD
$49,239.18
$120,214.53
$319,863.34
$1,240,725.74
$2,716,564.32
$4,590,626.57
$6,261 ,775.90
$15,712,167.45
$65,570,511.99
Membranes
$260,909.86
$511,246.10
$922,064.41
$1,585,978.03
$3,336,150.82
$10,977,349.65
$16,315,154.73
$51,427,679.24
$271,760,784.19
Source: Design flows from Exhibit 3.4, and unit costs from Appendix I.
Exhibit 7.1 Ob Annual O&M Unit Costs ($/Plant/Year) for CWS Surface Water Plants
System Size
(Population
Served)
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
Plant-Average
Daily Flow
(MGD)
0.01
0.03
0.08
0.22
0.59
2.34
3.80
14.61
107.80
CLM
$1 ,366.97
$1,486.13
$2,786.70
$3,075.12
$6,367.05
$9,532.26
$1 1 ,365.79
$20,636.42
$73,080.99
CL02
$14,200.27
$16,491.53
$17,767.16
$20,288.25
$23,993.48
$27,157.89
$48,315.25
$197,514.28
UV
$3,465.33
$4,567.48
$5,906.07
$8,060.73
$10,462.98
$15,274.16
$17,309.16
$32,774.25
$170,152.92
Ozone
$55,555.11
$58,860.42
$61,074.17
$65,448.37
$88,075.83
$103,776.82
$216,130.20
$1,241,600.91
MF/UF
$6,788.93
$10,560.66
$25,696.45
$40,850.24
$90,783.00
$258,314.24
$401,754.12
$1 ,337,846.76
$8,908,719.65
GAC10
$103,214.63
$138,118.16
$337,930.01
$1,767,266.63
GAC10+AD
$127,208.11
$165,276.06
$386,245.26
$1,964,780.90
GAC20
$19,849.99
$47,768.56
$59,595.71
$120,988.58
$188,170.95
$292,402.22
$385,208.70
$1 ,074,985.73
$6,091,016.57
GAC20+AD
$19,849.99
$61,968.83
$65,501.78
$182,062.75
$253,619.32
$316,395.70
$412,366.60
$1,123,300.98
$6,288,530.85
Membranes
$14,964.21
$25,798.47
$62,334.66
$110,067.20
$256,285.82
$817,602.69
$1,291,782.17
$4,500,839.35
$30,384,366.43
Source: Average daily flows from Exhibit 3.4, and unit costs from Appendix I.
Final Economic Analysis for the Stage 2 DBPR
7-27
December 2005
-------
Exhibit 7.1 Oc Household Unit Treatment Costs ($/Household/Year) for CWS Surface Water Plants
System Size
(Population
Served)
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1 ,000,000+
Mean Annual
Water Usage
per HH (kgal)
83
83
104
87
97
109
119
125
125
CLM
$139.50
$32.32
$20.10
$6.99
$6.09
$2.09
$1.58
$0.82
$0.36
CLO2
$116.11
$70.95
$22.76
$12.29
$5.17
$3.88
$1.80
$0.86
UV
$163.48
$53.30
$34.99
$16.56
$16.97
$6.73
$6.07
$4.93
$2.58
Ozone
$454.26
$333.82
$120.46
$67.58
$32.07
$27.07
$16.18
$10.62
MF/UF
$851 .00
$345.06
$296.45
$137.12
$117.89
$93.93
$94.61
$81 .94
$66.95
GAC10
$41.01
$37.58
$25.38
$15.31
GAC10+AD
$46.18
$41.46
$27.17
$16.16
GAC20
$869.14
$473.88
$297.68
$186.18
$151.41
$83.20
$75.95
$55.03
$36.36
GAC20+AD
$869.14
$589.99
$332.68
$306.64
$218.99
$88.37
$79.84
$56.83
$37.22
Membranes
$1,348.43
$565.14
$505.44
$261.11
$243.95
$219.10
$226.64
$205.32
$167.99
Source: Mean water usage per household derived from the Third Edition of the Baseline Handbook (USEPA, 2001c)
Final Economic Analysis for the Stage 2 DBPR
7-28
December 2005
-------
Exhibit 7.11 a Capital Cost ($/Plant) for CWS Disinfecting Ground Water Plants
System Size
(Population Served)
<100
100-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1 ,000,000+
Plant Design
Flow (MGD)
0.02
0.06
0.16
0.38
0.89
1.64
4.12
7.68
41.35
CLM
$29,104.44
$29,104.44
$30,298.78
$39,102.18
$49,613.33
$83,772.20
$83,772.20
$84,799.17
$98,772.20
UV
$47,108.15
$70,701.54
$126,977.04
$252,548.67
$724,535.96
Ozone
$0.00
$0.00
$370,630.22
$507,022.28
$761 ,428.90
$1 ,081 ,364.49
$1 ,730,048.55
$2,466,239.92
$7,443,344.25
GAC20
$51 ,560.52
$102,166.90
$220,628.98
$457,866.77
$1,064,410.99
$1 ,671 ,400.95
$3,239,129.48
$5,162,317.80
$18,067,167.43
NF
$67,676.53
$119,226.85
$209,226.96
$379,952.47
$828,831 .05
$1 ,598,457.82
$3,994,609.23
$7,341 ,646.08
$33,373,795.86
Source: Design flows from Exhibit 3.4, and unit costs from Appendix I.
Exhibit 7.11b Annual O&M Costs ($/Plant/Year) for CWS Disinfecting Ground
Water Plants
System Size
(Population Served)
<100
100-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1 ,000,000+
Plant-Average
Daily Flow
(MGD)
0.01
0.02
0.05
0.13
0.34
0.72
2.01
4.26
27.22
CLM
$1 ,361 .57
$1,414.02
$1 ,470.58
$2,958.23
$4,184.39
$6,186.52
$7,788.54
$9,383.04
$19,308.42
UV
$7,837.50
$10,317.16
$13,703.26
$18,540.93
$22,261 .81
/"
/""
/""
/
Ozone
$0.00
$0.00
$55,839.34
$59,990.54
$62,469.10
$66,975.28
$84,465.80
$108,576.27
$350,166.93
GAC20
$10,760.70
$19,848.70
$33,800.85
$54,578.39
$97,058.06
$108,465.68
$157,248.60
$239,037.74
$965,369.32
NF
$7,835.50
$11,411.69
$27,593.43
$43,144.08
$109,460.54
$193,835.38
$484,005.34
$992,000.90
$5,723,540.30
Source: Average daily flows from Exhibit 3.4, and unit costs from Appendix I.
Final Economic Analysis for the Stage 2 DBPR
7-29
December 2005
-------
Exhibit 7.11c Household Unit Treatment Costs ($/Household/Year) for CWS
Disinfecting Ground Water Plants
System Size
(Population Served)
<100
100-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1 ,000,000+
Mean Annual
Water Usage
per HH (kgal)
83
83
104
87
97
109
119
125
125
CLM
$177.77
$51 .26
$22.99
$1 1 .40
$6.53
$5.45
$2.39
$1.32
$0.34
UV
$543.11
$214.42
$138.94
$72.63
$65.17
Ozone
$0.00
$0.00
$495.03
$187.08
$98.87
$65.00
$37.03
$25.16
$12.08
GAC20
$693.37
$374.38
$297.85
$169.68
$145.77
$102.54
$69.19
$53.61
$30.76
NF
$625.03
$283.38
$257.26
$145.98
$139.86
$135.66
$132.32
$128.64
$106.41
Source: Mean water usage per household derived from the Third Edition of the Baseline Handbook (USEPA, 2001c)
7.4.2 Alternatives to Treatment
During the FACA process, the M-DBP TWO identified many ways systems could reduce the
residence time of the water in their distribution systems, thereby reducing TTHM and HAAS
concentrations. These included:
Flushing more frequently, or looping sections of the distribution system to eliminate dead
ends.
• Modifying portions of the distribution system with problematic DBF levels.
• Optimizing storage to minimize retention time in the distribution system.
The costs for these activities could range from close to zero (e.g., changing tank operations
without making capital improvements) to more substantial costs for reconfiguring storage facilities or
looping distribution system networks. The benefits from distribution system activities that reduce DBF
concentrations also vary widely and are dependent on system-specific conditions. Therefore, these
activities were not evaluated in this EA (see Section 7.7 for a discussion of unquantifiable costs).
Other alternatives to treatment that could reduce TTHM and HAAS levels in the distribution
system include connecting to a nearby water system or identifying another water source that has lower
DBF precursor levels. While the latter may not be feasible for some remote systems, EPA estimates that
more than 22 percent of all small systems are located within metropolitan regions where distances
between neighboring utilities will not present a prohibitive barrier. To estimate this percentage, EPA
used the April 2000 Safe Drinking Water Information System (SDWIS) database to compare the ZIP
codes for CWSs that serve 100 or fewer people and medium and large CWSs. EPA then determined that
out of the 446 surface water CWSs that serve 100 or fewer people, 98 (22 percent) have ZIP codes that
are identical to those for medium and large CWSs (however, size of the ZIP code zone was not
considered). Consolidation with another water system may represent the least-cost alternative for many
small systems.
Final Economic Analysis for the Stage 2 DBPR
7-30
December 2005
-------
7.4.3 Uncertainty in Unit Costs
In developing the unit costs used in this EA, the design criteria for the compliance treatment
technologies were selected to represent typical, or average, conditions for the universe of systems. As a
result, there is uncertainty in these unit cost estimates, as they are based on designs and quotes assuming
average conditions, rather than on an detailed aggregation of State, regional, or local estimates based on
actual field conditions. To model the uncertainty around unit costs in this EA, the national average unit
cost factors are characterized as triangular distributions with minimum and maximum values set at the
following percentages relative to the best estimate:
• Capital costs: ± 30%
• O&M costs: ±15%
These percentages were developed by EPA based on input from engineering professionals and
represent recommendations from the National Drinking Water Advisory Council (NDWAC) (2001) in
their review of the national cost estimation methodology for the Arsenic Rule. EPA believes that the
uncertainties in capital and O&M costs for a given treatment technology are independent of one another
and that uncertainties across all treatment technologies are independent.
7.5 The Stage 2 DBPR Cost Model
The cost model combines compliance forecasts, as described in Chapter 5, with the technology
unit costs, as described in Section 7.4, to obtain national treatment costs. The model accounts for
uncertainty in the national costs in several ways, as explained in Section 7.5.1. Once the treatment costs
and the uncertainty surrounding them are determined, they are combined with the non-treatment costs
from Section 7.3. The costs are then projected over time and discounted to obtain present value and
annualized costs for the rule as described in Section 7.6.2. Lastly, household costs are determined as
described in Section 7.6.3. A detailed list of all cost model files and flow charts is in Appendix K.
7.5.1 Probability Analysis to Estimate Nominal Treatment Costs
Inputs to Incorporate Uncertainty in National Costs
EPA recognizes that there is uncertainty in the inputs used to determine the national costs of this
rule. Ideally, the model would quantify each uncertainty. Data regarding the quantitative impact of each
element of uncertainty, however, are not available. EPA has developed an approach to quantifying the
more significant areas of uncertainty that potentially have a large influence on national costs:
Uncertainty in potential impact of the IDSE on the compliance forecast for large and
medium surface water systems, discussed in Chapter 5.
Uncertainty in methods used to develop the compliance forecast (i.e., SWAT and the ICR
Matrix Method), as discussed in Chapter 5 and Appendix A.
Uncertainty in average unit costs (capital and O&M), as discussed in Section 7.4.3.
Final Economic Analysis for the Stage 2 DBPR 7-31 December 2005
-------
As described in Chapter 5, it is possible that systems may measure higher DBF levels at Stage 2
compliance monitoring sites than were measured under the ICR. EPA believes that the 20 percent safety
margin on compliance (used throughout the primary cost analysis) already accounts for the potential
impacts of the IDSE for small surface water systems, ground water systems, and chloramine systems.
The Agency believes, however, that the 20 percent safety margin is not sufficient to account for the
potential impacts of the IDSE on all large and medium surface water systems because spatial variability
of DBF levels and distribution system complexity are greatest in these systems. Based on the analysis of
spatial variability in DBF distribution system data and other factors, EPA developed an approach to
quantify the potential impacts of the IDSE by preparing an alternative compliance forecast using a 25
percent safety margin (see Section 5.3.4 for details). Because a 20 percent safety margin may already
account for the IDSE for some large a medium surface water systems, results based on both a 20 and 25
percent safety margin are used in the cost model.
Since the proposal, EPA developed an second method, called the ICR Matrix Method, to estimate
the percent of plants making treatment technology changes to meet the Stage 2 DBPR. The percent of
plants making treatment technology changes from Stage 1 to Stage 2 are different for the ICR Matrix
Method and SWAT. Because both SWAT and the ICR Matrix Method have associated uncertainty,
results from both are used to generate the compliance forecast for surface water systems.
The ICR Matrix Method does not predict the specific treatment technologies that plants will
install. Thus, results from the ICR matrix method are incorporated by comparing the predicted percent of
plants making treatment technology changes with SWAT results to create a ICR Matrix Method-to-
SWAT multiplier. EPA generated a uniform distribution with 1.0 as the 5th percentile value and the ICR
Matrix Method-to-SWAT multiplier for plants making treatment technology changes as the 95th percentile
value. Two separate distributions were used for large and medium surface water systems, one for the 20
percent safety margin and one for the 25 percent safety margin. Exhibit 7.12 provides a graphical
depiction of the two uniform distributions for the Preferred Alternative. The implications of using one
method as opposed to combining results from both is discussed in Section 7.8.
Uncertainty in technology unit costs is described in Section 7.4.3. In summary, EPA developed
triangular distributions for both capital (±30 percent) and O&M costs (±15 percent) to incorporate the
uncertainty in unit treatment costs.
Final Economic Analysis for the Stage 2 DBPR 7-32 December 2005
-------
Exhibit 7.12 Uniform Distributions for Incorporating the ICR Matrix Method-to-
SWAT Multiplier into the Compliance Forecasts for Surface Water Systems
CO
O
i
Uniform Distribution for a 20%
Safety Margin
CO
O
i
Uniform Distribution for a 25%
Safety Margin
(2.91)
Multiplier
(1.98)
Multiplier
Source: Exhibit 5.6
Probability Analysis in the Cost Model
The cost model incorporates uncertainty using a Monte Carlo simulation. The model follows four
basic steps, the first three of which are consistent with the probability analysis to generate the compliance
forecast as described in Section 5.3.6:
Step 1: For large and medium surface water systems, the model randomly selects the
SWAT-predicted treatment technology selection delta for either the 20 or 25 percent
safety margin runs. Each safety margin has an equal (50 percent) chance of being
selected. For small surface water systems and all ground water systems, the treatment
technology selection delta for the 20 percent safety margin is always selected.
Step 2: For large, medium, and small surface water systems, the model randomly selects
the ICR-to-SWAT multiplier from the appropriate uniform distribution from Exhibit 7.12
for the safety margin selected in Step 1.
Step 3: For large, medium, and small surface water systems, the model multiplies the
result from Step 2 by the treatment technology selection delta results identified in Step 1
to calculate the percent and number of plants making treatment technology changes from
Stage 1 to Stage 2. For ground water systems, the treatment technology selection delta
for a 20 percent safety margin is always used to calculate the percent of plants changing
from Stage 1 to Stage 2.
• Step 4: The model randomly selects a number from 0.7 to 1.3 (± 30%) to represent
uncertainty in national capital costs for all technologies. Then it randomly selects a
number from 0.85 to 1.15 (± 15%) to represent uncertainty in national O&M costs for all
technologies. The model multiplies the treatment technology selection delta (step 3) by
Final Economic Analysis for the Stage 2 DBPR
7-33
December 2005
-------
the technology unit costs and the two uncertainty factors to estimate national treatment
costs (O&M)
The process is repeated 10,000 times to produce a distribution of plants making treatment
technology changes from Stage 1 to Stage 2 (results from Step 3, same as reported in Chapter 5) and
national treatment costs (results from Step 4). For more detail on this simulation, see the flow charts and
file description in Appendix K.
7.5.2 Projections and Discounting to Produce Annualized Costs
There are two kinds of nominal cost estimates for both treatment and non-treatment activities: (1)
one-time costs that occur near the beginning of the rule implementation period, and (2) yearly costs that
systems and States/Primacy Agencies will incur after systems have made necessary changes to treatment
and/or monitoring to comply with the Stage 2 DBPR. For the purposes of this EA, one-time and yearly
costs were projected over a 25-year time period to coincide with the estimated life span of capital
equipment and a time lag of 5 to 10 years for treatment technology installation after rule promulgation.
The projected schedules for all rule activities are summarized in Appendix D.
As described previously in this chapter and in the discussion of benefits in Chapter 6, it is
common practice to adjust benefits and costs to a present value6 using a social discount rate so that they
can be compared to one another. This process takes into account the time preference that society places
on expenditures and allows comparison of cost and benefit streams that are variable over a given time
period.7 Similar to calculating the present value of benefits (see Section 6.5), the present value of costs
for any future period can be calculated using the following equation:
PV = V(t)/(l+R)t
Where: t = The number of years from the reference period
R = Social discount rate
V(t) = The cost occurring t years from the reference period
There is much discussion among economists of the proper social discount rate to use for policy
analysis. For Stage 2 DBPR cost analyses, present value calculations are made using two social discount
rates thought to best represent current policy evaluation methodologies, 3 and 7 percent. Historically, the
use of 3 percent is based on rates of return on relatively risk-free investments, as described in the
Guidelines for Preparing Economic Analyses (USEPA 2000J). The rate of 7 percent is a recommendation
of the Office of Management and Budget (OMB) as an estimate of "before-tax rate of return to
incremental private investment" (USEPA 1996b). For any future cost, the higher the discount rate, the
lower the present value. Specifically, a future cost (or stream of costs) evaluated at a 7 percent social
discount rate will always result in a lower total present value cost than the same future cost evaluated at a
3 percent rate.
6 For purposes of analyses in this EA, all present value figures are presented at a year 2003 price level.
Present value calculations are performed to the expected year rule implementation (2005).
7 See EPA's Guidelines for Preparing Economic Analyses (USEPA 2000J) for a full discussion of the use of
social discount rates in the evaluation of policy decisions.
Final Economic Analysis for the Stage 2 DBPR 7-34 December 2005
-------
To allow evaluation of alternatives on an annual basis, their total present value costs are
annualized using the same social discount rates (3 and 7 percent) over 25 years. When applying social
discount rates to annualize costs, the higher the discount rate, the higher the annualized cost. Thus, the
magnitudes of the discount rates influence costs in the opposite direction (i.e., a present value cost
annualized at a 7 percent rate will always result in higher values than the same present value cost
annualized at a 3 percent rate). The final relationship between annualized costs at 3 and 7 percent is
dependent on the time frame for annualization, as well as when the costs are incurred (as set forth in the
Rule Activity Schedule in Appendix D). Given a long enough time frame, the 7 percent annualized value
will eventually be greater than the 3 percent annualized value.
In summary, the methodology for projecting and discounting costs is as follows:
Project all nominal costs (treatment, non-treatment, and State) over a 25-year time horizon
based on the rule implementation schedule in Appendix D.
• Calculate total present value costs using social discount rates. The same rates were used as
for the benefits calculation: 3 and 7 percent (see Section 6.5.3).
Annualize the costs over 25 years using the same social discount rates.
Values derived using this methodology are presented in subsequent sections of Chapter 7.
Detailed spreadsheets of all cost calculations are provided in Appendix J.
7.5.3 Methodology for Estimating Household Costs
EPA assumes that generally systems will pass some or all of the costs of a new regulation onto
their customers in the form of rate increases. As noted in Section 7.4.1, household costs, which are in
units of dollars per household per year ($/HH/yr), are estimated in this chapter to provide a measure of
potential increases in water bills as a result of the Stage 2 DBPR, if all costs are passed onto consumers.
A distribution of possible household costs is developed for each system type and population size
category. The distribution reflects that some households are served by systems that will incur only non-
treatment costs (e.g., implementation, IDSE, monitoring plans, additional routine monitoring, and
operational evaluation costs) as a result of the Stage 2 DBPR, while others are served by systems that
incur treatment costs as well. Treatment activities can range from converting to chloramines, which has a
relatively low cost, to installing membranes, which has a relatively high cost.
Data inputs specific to household cost calculations are shown in Exhibit 7.9, and a description of
the household cost methodology is provided below. Section 7.4.1 provides a detailed description of the
derivation of household unit treatment costs, and Exhibits 7. lOc and 7.1 Ic present household unit
treatment costs for surface and ground water treatment technologies. Appendix K provides a detailed list
of files and flow charts for the Stage 2 DBPR household cost model.
• Step 1: The average number of households served per plant is calculated by dividing the total
households served (Exhibit 3.5) by the total number of plants (Exhibit 3.2) in each system
size category.
• Step 2: The number of households incurring different types of costs is based on the number
of plants incurring costs, as derived in Appendix H and Section 3.5, respectively. The
percent of systems performing non-treatment related rule activities is derived in Appendix H
Final Economic Analysis for the Stage 2 DBPR 7-35 December 2005
-------
and summarized in Exhibit 7.2. EPA assumes that the percent of systems performing these
activities is equivalent to the percent of plants performing activities. The number of plants
making treatment technology changes to meet Stage 2 is shown in Exhibits 5.14a and 5.14c8.
The number of plants incurring costs is multiplied by the average number of households
served per plant to estimate the total number of households incurring different types of costs.
Step 3: For each type of cost (treatment and non-treatment), a household unit cost is
computed using the method described in Section 7.4.
Step 4: The annual household unit costs ($/HH/yr) are combined with the number of
households incurring each type of cost (results from step 2) to generate distributions of
possible household treatment costs. The different types of costs are combined assuming that
treatment and non-treatment costs are independent of one another. This might not always be
the case, but EPA believes it is a realistic approximation.
7.6 Results
This section presents the results of the cost model described in the preceding sections. Section
7.6.1 shows the number of plants performing various rule activities. Section 7.6.2 displays the one-time
capital and non-treatment costs as calculated by the model. Section 7.6.3 displays the results after the
cost projections and discounting are performed. Section 7.6.4 displays the household cost distributions.
7.6.1 Number of Plants Making Treatment Technology Changes
The number of plants making treatment technology changes is determined from the compliance
forecasts as described in Chapter 5. The model produces a distribution of the number of plants making
treatment technology changes to represent uncertainty in the compliance forecast for surface water
systems. As shown in Exhibit 7.3, the mean estimated number of plants making treatment technology
changes is 2,261 plants.
7.6.2 One-Time Costs
Exhibit 7.13 summarizes the estimated initial capital investment and yearly O&M costs. This
exhibit is broken out by system type, source water type, and system size category. Appendix J and
Exhibits J. Ib through J. Id provide similar cost information (total initial capital and yearly O&M costs)
for the Stage 2 DBPR regulatory alternatives.
7.6.3 Total Annual Costs
Appendix J contains results from each step of the cost projection and discounting process
described in Section 7.6.2 for each regulatory alternative. For the Preferred Alternative, Exhibits J.2a
through J.2ar show the nominal costs projected over the rule schedule, and Exhibits J.2as through J.2cf
8Only plants in CWSs are used to generate HH cost distributions. NTNCWSs do not typically provide
water to households.
Final Economic Analysis for the Stage 2 DBPR 7-36 December 2005
-------
show the present value of each cost calculated to the expected year of rule implementation (2005). The
annualization step is shown at the bottom of the present value exhibits.
Exhibits 7.14a and 7.14b present the stream of present value costs and the total annualized costs
for the Stage 2 DBPR Preferred Alternative at 3 and 7 percent discount rates, respectively. These tables
are equivalent to Exhibits J.2as and J.2aw in Appendix J.
Final Economic Analysis for the Stage 2 DBPR 7-3 7 December 2005
-------
Exhibit 7.13 Total Initial Capital Costs ($Millions) and Yearly O&M Costs
($Millions/Year)
Source
Surface
Water
Ground
Water
System
Classification
CWSs
NTNCWSs
CWSs
NTNCWSs
System
Size
(population
served)
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
All Sizes
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
All Sizes
Subtotal
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
All Sizes
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
1,000,000+
All Sizes
Subtotal
Total
Capital Costs
Mean
Value
$ 1.09
$ 3.27
$ 3.86
$ 24.39
$ 62.23
$ 113.20
$ 67.40
$ 183.98
$ 86.04
$ 545.44
$ 0.67
$ 1.32
$ 0.85
$ 1.89
$ 1.29
$ 0.55
$
$ 0.41
$
$ 6.99
$ 552.43
$ 8.34
$ 33.19
$ 20.18
$ 39.43
$ 65.91
$ 59.09
$ 14.96
$ 29.70
$ 3.38
$ 274.18
$ 3.17
$ 5.04
$ 2.47
$ 1.61
$ 0.46
$ 0.10
$ 0.02
$ 0.03
$
$ 12.90
$ 287.08
$ 839.51
Median
Value
$ 1.07
$ 3.22
$ 3.78
$ 24.27
$ 61.92
$ 113.98
$ 68.08
$ 186.24
$ 86.46
$ 549.03
$ 0.66
$ 1.31
$ 0.84
$ 1 .88
$ 1.28
$ 0.55
$
$ 0.41
$
$ 6.95
$ 555.97
$ 8.34
$ 33.18
$ 20.18
$ 39.42
$ 65.86
$ 59.08
$ 14.96
$ 29.71
$ 3.38
$ 274.11
$ 3.17
$ 5.04
$ 2.47
$ 1.61
$ 0.46
$ 0.10
$ 0.02
$ 0.03
$
$ 12.90
$ 287.01
$ 842.98
90 Percent
Confidence Bound
Lower
(5th %tile)
$ 0.58
$ 1.77
$ 2.08
$ 13.37
$ 34.42
$ 62.72
$ 37.41
$ 98.21
$ 47.14
$ 297.70
$ 0.36
$ 0.72
$ 0.46
$ 1.04
$ 0.71
$ 0.30
$
$ 0.22
$
$ 3.82
$ 301.52
$ 7.19
$ 28.04
$ 17.00
$ 32.35
$ 53.53
$ 53.39
$ 13.38
$ 26.43
$ 2.97
$ 234.29
$ 2.73
$ 4.25
$ 2.07
$ 1.32
$ 0.38
$ 0.09
$ 0.02
$ 0.03
$
$ 10.87
$ 245.16
$ 546.68
Upper
(95th %tile)
$ 1.68
$ 4.94
$ 5.89
$ 36.07
$ 91.81
$ 157.05
$ 93.50
$ 257.75
$ 120.41
$ 769.10
$ 1.03
$ 2.00
$ 1.30
$ 2.80
$ 1.90
$ 0.76
$
$ 0.57
$
$ 10.36
$ 779.46
$ 9.53
$ 38.38
$ 23.34
$ 46.54
$ 78.34
$ 64.79
$ 16.53
$ 32.95
$ 3.79
$ 314.20
$ 3.62
$ 5.81
$ 2.87
$ 1.90
$ 0.55
$ 0.11
$ 0.02
$ 0.03
$
$ 14.91
$ 329.11
$ 1,108.57
O&M Costs
Mean
Value
$ 0.20
$ 0.82
$ 0.61
$ 3.36
$ 5.32
$ 6.04
$ 3.41
$ 8.17
$ 4.91
$ 32.84
$ 0.12
$ 0.33
$ 0.13
$ 0.26
$ 0.11
$ 0.03
$
$ 0.02
$
$ 1.00
$ 33.85
$ 0.98
$ 3.68
$ 1.96
$ 3.00
$ 2.55
$ 5.03
$ 1.28
$ 2.83
$ 0.43
$ 21.73
$ 0.37
$ 0.55
$ 0.23
$ 0.10
$ 0.01
$ 0.01
$ 0.00
$ 0.00
$
$ 1.29
$ 23.02
$ 56.86
Median
Value
$ 0.20
$ 0.82
$ 0.61
$ 3.36
$ 5.34
$ 6.00
$ 3.36
$ 7.87
$ 4.65
$ 32.21
$ 0.12
$ 0.33
$ 0.13
$ 0.26
$ 0.11
$ 0.03
$
$ 0.02
$
$ 1.00
$ 33.22
$ 0.98
$ 3.68
$ 1.96
$ 3.00
$ 2.55
$ 5.03
$ 1.28
$ 2.83
$ 0.43
$ 21.73
$ 0.37
$ 0.55
$ 0.23
$ 0.10
$ 0.01
$ 0.01
$ 0.00
$ 0.00
$
$ 1.29
$ 23.02
$ 56.23
90 Percent
Confidence Bound
Lower
(5th %tile)
$ 0.11
$ 0.46
$ 0.34
$ 1 .88
$ 2.97
$ 3.74
$ 2.13
$ 5.21
$ 3.11
$ 19.95
$ 0.07
$ 0.19
$ 0.07
$ 0.15
$ 0.06
$ 0.02
$
$ 0.01
$
$ 0.56
$ 20.52
$ 0.91
$ 3.38
$ 1.80
$ 2.73
$ 2.33
$ 4.76
$ 1.20
$ 2.64
$ 0.40
$ 20.16
$ 0.35
$ 0.51
$ 0.21
$ 0.09
$ 0.01
$ 0.01
$ 0.00
$ 0.00
$
$ 1.18
$ 21.34
$ 41.86
Upper
(95th %tile)
$ 0.29
$ 1.19
$ 0.88
$ 4.86
$ 7.70
$ 8.66
$ 4.95
$ 12.52
$ 7.73
$ 48.78
$ 0.17
$ 0.48
$ 0.20
$ 0.38
$ 0.16
$ 0.04
$
$ 0.03
$
$ 1.46
$ 50.24
$ 1.05
$ 3.98
$ 2.12
$ 3.26
$ 2.76
$ 5.30
$ 1.36
$ 3.02
$ 0.46
$ 23.31
$ 0.40
$ 0.60
$ 0.25
$ 0.11
$ 0.02
$ 0.01
$ 0.00
$ 0.00
$
$ 1.39
$ 24.70
$ 74.94
Notes: All values in millions of year 2003 dollars.
Detail may not add exactly to totals due to independent rounding.
Source: This exhibit is identical to Exhibit J.1a
Final Economic Analysis for the Stage 2 DBPR
7-38
December 2005
-------
Exhibit 7.14a Total Annualized Costs at 3 Percent Social Discount Rate ($Millions)
Preferred Alternative
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
Total
Ann.
Surface Water CWS
Mean
Value
$ 0.8
$ 10.8
$ 20.6
$ 15.0
$ 99.7
$ 101.6
$ 103.8
$ 105.1
$ 50.3
$ 38.2
$ 26.3
$ 21.1
$ 20.5
$ 19.9
$ 19.3
$ 18.7
$ 18.2
$ 17.7
$ 17.2
$ 16.7
$ 16.2
$ 15.7
$ 15.2
$ 14.8
$ 14.4
$ 817.6
$ 47.0
90 Percent
Confidence Bound
Lower
(5th %tile)
$ 0.8
$ 10.8
$ 20.6
$ 15.0
$ 54.8
$ 55.9
$ 57.4
$ 58.0
$ 28.4
$ 21.8
$ 15.2
$ 12.3
$ 12.0
$ 11.6
$ 11.3
$ 10.9
$ 10.6
$ 10.3
$ 10.0
$ 9.7
$ 9.4
$ 9.2
$ 8.9
$ 8.6
$ 8.4
$ 482.2
$ 27.7
Upper
(95th %tile)
$ 0.8
$ 10.8
$ 20.6
$ 15.0
$ 139.8
$ 143.2
$ 146.8
$ 149.3
$ 73.5
$ 56.8
$ 39.7
$ 32.0
$ 31.0
$ 30.1
$ 29.2
$ 28.4
$ 27.6
$ 26.8
$ 26.0
$ 25.2
$ 24.5
$ 23.8
$ 23.1
$ 22.4
$ 21.8
$ 1,167.9
$ 67.1
Surface Water NTNCWS
Mean
Value
$ 0.0
$ 0.1
$ 0.0
$ 0.0
$ 1.0
$ 1.1
$ 1.1
$ 1.2
$ 1.3
$ 1.3
$ 1.0
$ 0.7
$ 0.7
$ 0.7
$ 0.6
$ 0.6
$ 0.6
$ 0.6
$ 0.6
$ 0.6
$ 0.5
$ 0.5
$ 0.5
$ 0.5
$ 0.5
$ 16.3
$ 0.9
90 Percent
Confidence Bound
Lower
(5th %tile)
$ 0.0
$ 0.1
$ 0.0
$ 0.0
$ 0.6
$ 0.6
$ 0.6
$ 0.7
$ 0.7
$ 0.7
$ 0.6
$ 0.4
$ 0.4
$ 0.4
$ 0.4
$ 0.4
$ 0.3
$ 0.3
$ 0.3
$ 0.3
$ 0.3
$ 0.3
$ 0.3
$ 0.3
$ 0.3
$ 9.3
$ 0.5
Upper
(95th %tile)
$ 0.0
$ 0.1
$ 0.0
$ 0.0
$ 1.4
$ 1.6
$ 1.7
$ 1.8
$ 1.8
$ 1.9
$ 1.5
$ 1.0
$ 1.0
$ 1.0
$ 0.9
$ 0.9
$ 0.9
$ 0.8
$ 0.8
$ 0.8
$ 0.8
$ 0.8
$ 0.7
$ 0.7
$ 0.7
$ 23.6
$ 1.4
Disinfecting Ground Water CWS
Mean
Value
$ 0.2
$ 3.1
$ 1.0
$ 6.0
$ 43.1
$ 43.6
$ 44.0
$ 46.4
$ 41.2
$ 37.1
$ 24.8
$ 16.0
$ 15.5
$ 15.1
$ 14.6
$ 14.2
$ 13.8
$ 13.4
$ 13.0
$ 12.6
$ 12.3
$ 11.9
$ 11.6
$ 11.2
$ 10.9
$ 476.5
$ 27.4
90 Percent
Confidence Bound
Lower
(5th %tile)
$ 0.2
$ 3.1
$ 1.0
$ 6.0
$ 37.4
$ 37.8
$ 38.2
$ 40.6
$ 36.2
$ 32.5
$ 22.2
$ 14.9
$ 14.5
$ 14.1
$ 13.7
$ 13.3
$ 12.9
$ 12.5
$ 12.1
$ 11.8
$ 11.4
$ 11.1
$ 10.8
$ 10.5
$ 10.2
$ 428.8
$ 24.6
Upper
(95th %tile)
$ 0.2
$ 3.1
$ 1.0
$ 6.0
$ 48.8
$ 49.3
$ 49.7
$ 52.2
$ 46.3
$ 41.6
$ 27.4
$ 17.1
$ 16.6
$ 16.1
$ 15.6
$ 15.2
$ 14.7
$ 14.3
$ 13.9
$ 13.5
$ 13.1
$ 12.7
$ 12.3
$ 12.0
$ 11.6
$ 524.2
$ 30.1
Disinfecting Ground Water NTNCWS
Mean
Value
$ 0.0
$ 0.5
$ 0.0
$ 0.0
$ 2.1
$ 2.0
$ 1.9
$ 2.3
$ 2.6
$ 2.7
$ 2.0
$ 1.4
$ 1.3
$ 1.3
$ 1.2
$ 1.2
$ 1.2
$ 1.1
$ 1.1
$ 1.1
$ 1.0
$ 1.0
$ 1.0
$ 1.0
$ 0.9
$ 31.8
$ 1.8
90 Percent
Confidence Bound
Lower
(5th %tile)
$ 0.0
$ 0.5
$ 0.0
$ 0.0
$ 1.8
$ 1.7
$ 1.6
$ 2.0
$ 2.3
$ 2.4
$ 1.8
$ 1.3
$ 1.3
$ 1.2
$ 1.2
$ 1.1
$ 1.1
$ 1.1
$ 1.0
$ 1.0
$ 1.0
$ 1.0
$ 0.9
$ 0.9
$ 0.9
$ 29.2
$ 1.7
Upper
(95th%tile)
$ 0.0
$ 0.5
$ 0.0
$ 0.0
$ 2.3
$ 2.3
$ 2.2
$ 2.5
$ 2.9
$ 2.9
$ 2.2
$ 1.4
$ 1.4
$ 1.3
$ 1.3
$ 1.3
$ 1.2
$ 1.2
$ 1.2
$ 1.1
$ 1.1
$ 1.1
$ 1.0
$ 1.0
$ 1.0
$ 34.4
$ 2.0
Primacy Agencies
Point Estimate
$ 3.7
$ 3.6
$ 0.1
$ 1.8
$ 0.7
$
$ 1.3
$ 1.3
$ 1.3
$ 1.2
$ 1.2
$ 1.2
$ 1.1
$ 1.1
$ 1.1
$ 1.0
$ 1.0
$ 1.0
$ 0.9
$ 0.9
$ 0.9
$ 0.9
$ 0.8
$ 0.8
$ 0.8
$ 29.8
$ 1.7
Total
Mean
Value
$ 4.7
$ 18.1
$ 21.8
$ 22.9
$ 146.6
$ 148.2
$ 152.1
$ 156.2
$ 96.6
$ 80.5
$ 55.3
$ 40.3
$ 39.1
$ 38.0
$ 36.9
$ 35.8
$ 34.8
$ 33.8
$ 32.8
$ 31.8
$ 30.9
$ 30.0
$ 29.1
$ 28.3
$ 27.5
$ 1,372.1
$ 78.8
90 Percent
Confidence Bound
Lower
(5th %tile)
$ 4.7
$ 18.1
$ 21.8
$ 22.9
$ 95.4
$ 96.1
$ 99.2
$ 102.6
$ 68.9
$ 58.6
$ 41.0
$ 30.1
$ 29.2
$ 28.4
$ 27.5
$ 26.7
$ 26.0
$ 25.2
$ 24.5
$ 23.8
$ 23.1
$ 22.4
$ 21.7
$ 21.1
$ 20.5
$ 979.4
$ 56.2
Upper
(95th %tile)
$ 4.7
$ 18.1
$ 21.8
$ 22.9
$ 193.0
$ 196.3
$ 201.7
$ 207.2
$ 125.7
$ 104.4
$ 72.0
$ 52.6
$ 51.1
$ 49.6
$ 48.2
$ 46.8
$ 45.4
$ 44.1
$ 42.8
$ 41.5
$ 40.3
$ 39.2
$ 38.0
$ 36.9
$ 35.8
$ 1,780.0
$ 102.2
Present values in millions of 2003 dollars. Estimates are discounted to 2005.
Detail may not add exactly to totals due to independent rounding.
Ann = value of total annualized at discount rate.
This exhibit is identical to Exhibit J.2as, which is derived from Exhibits J.2a through r
Final Economic Analysis for the Stage 2 DBPR
7-39
December 2005
-------
Exhibit 7.14b Total Annualized Costs at 7 Percent Social Discount Rate ($Millions)
Preferred Alternative
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
Total
Ann.
Surface Water CWS
Mean
Value
$ 0.8
$ 9.6
$ 17.7
$ 12.4
$ 79.3
$ 77.8
$ 76.5
$ 74.6
$ 34.4
$ 25.1
$ 16.6
$ 12.9
$ 12.0
$ 11.2
$ 10.5
$ 9.8
$ 9.2
$ 8.6
$ 8.0
$ 7.5
$ 7.0
$ 6.5
$ 6.1
$ 5.7
$ 5.3
$ 545.1
$ 46.8
90 Percent
Confidence Bound
Lower
(5th %tile)
$ 0.8
$ 9.6
$ 17.7
$ 12.4
$ 43.6
$ 42.8
$ 42.3
$ 41.2
$ 19.4
$ 14.3
$ 9.6
$ 7.5
$ 7.0
$ 6.6
$ 6.1
$ 5.7
$ 5.4
$ 5.0
$ 4.7
$ 4.4
$ 4.1
$ 3.8
$ 3.6
$ 3.3
$ 3.1
$ 324.1
$ 27.8
Upper
(95th %tlle)
$ 0.8
$ 9.6
$ 17.7
$ 12.4
$ 111.2
$ 109.6
$ 108.2
$ 106.0
$ 50.2
$ 37.3
$ 25.1
$ 19.5
$ 18.2
$ 17.0
$ 15.9
$ 14.9
$ 13.9
$ 13.0
$ 12.1
$ 11.3
$ 10.6
$ 9.9
$ 9.3
$ 8.6
$ 8.1
$ 770.4
$ 66.1
Surface Water NTNCWS
Mean
Value
$ 0.0
$ 0.1
$ 0.0
$ 0.0
$ 0.8
$ 0.8
$ 0.8
$ 0.9
$ 0.9
$ 0.8
$ 0.6
$ 0.4
$ 0.4
$ 0.4
$ 0.4
$ 0.3
$ 0.3
$ 0.3
$ 0.3
$ 0.2
$ 0.2
$ 0.2
$ 0.2
$ 0.2
$ 0.2
$ 9.8
$ 0.8
90 Percent
Confidence Bound
Lower
(5th %tile)
$ 0.0
$ 0.1
$ 0.0
$ 0.0
$ 0.4
$ 0.5
$ 0.5
$ 0.5
$ 0.5
$ 0.5
$ 0.4
$ 0.2
$ 0.2
$ 0.2
$ 0.2
$ 0.2
$ 0.2
$ 0.2
$ 0.2
$ 0.1
$ 0.1
$ 0.1
$ 0.1
$ 0.1
$ 0.1
$ 5.6
$ 0.5
Upper
(95th %tile)
$ 0.0
$ 0.1
$ 0.0
$ 0.0
$ 1.1
$ 1.2
$ 1.2
$ 1.3
$ 1.3
$ 1.2
$ 0.9
$ 0.6
$ 0.6
$ 0.5
$ 0.5
$ 0.5
$ 0.4
$ 0.4
$ 0.4
$ 0.4
$ 0.3
$ 0.3
$ 0.3
$ 0.3
$ 0.3
$ 14.2
$ 1.2
Disinfecting Ground Water CWS
Mean
Value
$ 0.2
$ 2.8
$ 0.8
$ 5.0
$ 34.3
$ 33.4
$ 32.4
$ 32.9
$ 28.2
$ 24.4
$ 15.7
$ 9.7
$ 9.1
$ 8.5
$ 8.0
$ 7.4
$ 6.9
$ 6.5
$ 6.1
$ 5.7
$ 5.3
$ 5.0
$ 4.6
$ 4.3
$ 4.0
$ 301.2
$ 25.8
90 Percent
Confidence Bound
Lower
(5th %tile)
$ 0.2
$ 2.8
$ 0.8
$ 5.0
$ 29.8
$ 29.0
$ 28.2
$ 28.8
$ 24.7
$ 21.4
$ 14.1
$ 9.1
$ 8.5
$ 7.9
$ 7.4
$ 6.9
$ 6.5
$ 6.1
$ 5.7
$ 5.3
$ 4.9
$ 4.6
$ 4.3
$ 4.0
$ 3.8
$ 269.7
$ 23.1
Upper
(95th %tile)
$ 0.2
$ 2.8
$ 0.8
$ 5.0
$ 38.8
$ 37.8
$ 36.7
$ 37.1
$ 31.6
$ 27.4
$ 17.4
$ 10.4
$ 9.7
$ 9.1
$ 8.5
$ 7.9
$ 7.4
$ 6.9
$ 6.5
$ 6.1
$ 5.7
$ 5.3
$ 4.9
$ 4.6
$ 4.3
$ 332.7
$ 28.6
Disinfecting Ground Water NTNCWS
Mean
Value
$ 0.0
$ 0.5
$ 0.0
$ 0.0
$ 1.7
$ 1.5
$ 1.4
$ 1.6
$ 1.8
$ 1.7
$ 1.3
$ 0.8
$ 0.8
$ 0.7
$ 0.7
$ 0.6
$ 0.6
$ 0.6
$ 0.5
$ 0.5
$ 0.5
$ 0.4
$ 0.4
$ 0.4
$ 0.3
$ 19.2
$ 1.6
90 Percent
Confidence Bound
Lower
(5th %tile)
$ 0.0
$ 0.5
$ 0.0
$ 0.0
$ 1.4
$ 1.3
$ 1.2
$ 1.4
$ 1.6
$ 1.6
$ 1.2
$ 0.8
$ 0.7
$ 0.7
$ 0.6
$ 0.6
$ 0.6
$ 0.5
$ 0.5
$ 0.5
$ 0.4
$ 0.4
$ 0.4
$ 0.3
$ 0.3
$ 17.5
$ 1.5
Upper
(95th %tile)
$ 0.0
$ 0.5
$ 0.0
$ 0.0
$ 1.9
$ 1.7
$ 1.6
$ 1.8
$ 2.0
$ 1.9
$ 1.4
$ 0.9
$ 0.8
$ 0.8
$ 0.7
$ 0.7
$ 0.6
$ 0.6
$ 0.5
$ 0.5
$ 0.5
$ 0.4
$ 0.4
$ 0.4
$ 0.4
$ 20.9
$ 1.8
Primacy Agencies
Point Estimate
$ 3.4
$ 3.2
$ 0.1
$ 1.5
$ 0.6
$
$ 1.0
$ 0.9
$ 0.9
$ 0.8
$ 0.8
$ 0.7
$ 0.7
$ 0.6
$ 0.6
$ 0.5
$ 0.5
$ 0.5
$ 0.4
$ 0.4
$ 0.4
$ 0.4
$ 0.3
$ 0.3
$ 0.3
$ 19.8
$ 1.7
Total
Mean
Value
$ 4.3
$ 16.1
$ 18.7
$ 18.9
$ 116.6
$ 113.5
$ 112.2
$ 110.9
$ 66.0
$ 52.9
$ 35.0
$ 24.6
$ 23.0
$ 21.5
$ 20.1
$ 18.7
$ 17.5
$ 16.4
$ 15.3
$ 14.3
$ 13.4
$ 12.5
$ 11.7
$ 10.9
$ 10.2
$ 895.1
$ 76.8
90 Percent
Confidence Bound
Lower
(5th %tile)
$ 4.3
$ 16.1
$ 18.7
$ 18.9
$ 75.9
$ 73.6
$ 73.1
$ 72.8
$ 47.0
$ 38.6
$ 26.0
$ 18.3
$ 17.1
$ 16.0
$ 15.0
$ 14.0
$ 13.1
$ 12.2
$ 11.4
$ 10.7
$ 10.0
$ 9.3
$ 8.7
$ 8.1
$ 7.6
$ 636.7
$ 54.6
Upper
(95th %tile)
$ 4.3
$ 16.1
$ 18.7
$ 18.9
$ 153.6
$ 150.3
$ 148.7
$ 147.0
$ 85.9
$ 68.7
$ 45.6
$ 32.1
$ 30.0
$ 28.0
$ 26.2
$ 24.5
$ 22.9
$ 21.4
$ 20.0
$ 18.7
$ 17.4
$ 16.3
$ 15.2
$ 14.2
$ 13.3
$ 1,157.9
$ 99.4
Present values in millions of 2003 dollars. Estimates are discounted to 2005.
Detail may not add exactly to totals due to independent rounding.
Ann = value of total annualized at discount rate.
: This exhibit is identical to Exhibit J.2aw, which is derived from Exhibits J.2a through rr.
Final Economic Analysis for the Stage 2 DBPR
7-40
December 2005
-------
7.6.4 Household Cost Results
Exhibit 7.15 presents the mean, median, 90th percentile, and 95th percentile of expected rate
increases, along with the percent of households that are expected to face an increase of $1 or less per
month ($12 or less per year) or $10 or less each month ($120 or less per year). Data are provided for all
systems subject to the rule and for only the subset of systems making treatment technology changes.
Note that household cost increases in Exhibit 7.15 include costs for non-treatment-related rule activities
(implementation, IDSE, additional routine monitoring, and operational evaluations).
Exhibits 7.16a and b show the cumulative distribution of household cost increases for all surface
water and ground water systems, and Exhibits 7.17a and b shows the distribution of household cost
increases for only those systems making treatment technology changes. Additionally, Exhibit 7.17c
shows the cumulative distributions in systems making treatment technology changes for the five small
system size categories.
As shown in Exhibit 7.15, the mean, median, and 90th percentile household costs increase for all
systems (including those that do not make treatment technology changes) are $0.62, $0.03, and $0.36,
respectively. The mean, median, and 90th percentile household cost increases for systems that install new
treatment technologies are $5.53, $0.80 and $10.04, respectively. Note that the number of households
affected by plants installing treatment could be greater than shown in Exhibit 7.15 because an entire
system would most likely incur costs even if only some of the plants for that system make treatment
technology changes (this would result in lower household costs, however). EPA assumes that systems
will pass some or all of the costs of a new regulation on to their customers in the form of rate increases. It
should also be noted that these are very conservative estimates, assuming that cheaper compliance
alternatives are not used, and that systems do not receive any financial assistance.
EPA estimates that, as a whole, households subject to the Stage 2 DBPR face minimal increases
in their annual costs. Approximately 86 percent of the households potentially subject to the rule are
served by systems serving at least 10,000 people; these systems experience the lowest increases in costs
due to significant economies of scale. Households served by small systems that make treatment
technology changes will face the greatest increases in annual costs. Exhibit 7.17c provides additional
detail regarding the distribution of household costs for small systems.
Although cost model results predict that a few very small systems will experience large
household cost increases as a result of adding advanced treatment technology for the Stage 2 DBPR, these
predictions are probably not realistic because small systems have other alternatives available to them
besides making treatment technology changes. For example, some of these systems currently may be
operated on a part-time basis; therefore, they may be able to modify the current operational schedule or
use excess capacity to avoid installing a costly treatment technology to comply with the Stage 2 DBPR.
The system also may identify another water source that has lower TTHM and HAA5 precursor levels.
Systems that can identify such an alternate water source may not have to treat that water as intensely as
their current source, resulting in lower treatment costs. Systems may elect to connect to a neighboring
water system. While this may not be feasible for some remote systems, EPA estimates that more than 22
percent of all small water systems are located within metropolitan regions (USEPA 2003t) where
distances between neighboring systems will not present a prohibitive barrier.
Final Economic Analysis for the Stage 2 DBPR 7-41 December 2005
-------
Exhibit 7.15 Annual Household Cost Increases
Households Served by All Plants
All Systems
All Small Systems
SW< 10,000
SW> 10,000
GW< 10,000
GW> 10,000
Total Number of
Households Served
101,553,868
14,261,241
3,251,893
62,137,350
1 1 ,009,348
25,155,277
Mean Annual
Household
Cost Increase
$ 0.62
$ 2.20
$ 4.58
$ 0.46
$ 1.49
$ 0.13
Median Annual
Household
Cost Increase
$ 0.03
$ 0.10
$ 0.79
$ 0.02
$ 0.02
$ 0.00
90th Percentile
Annual
Household Cost
Increase
$ 0.36
$ 0.79
$ 2.69
$ 0.35
$ 0.39
$ 0.03
95th Percentile
Annual
Household Cost
Increase
$ 0.98
$ 2.57
$ 7.24
$ 1.81
$ 0.99
$ 0.08
Percentage of
Annual
Household Cost
Increase < $12
99%
97%
95%
99%
98%
100%
Percentage of
Annual
Household Cost
Increase < $120
100%
100%
99%
100%
100%
100%
Households Served by Plants Adding Treatment
All Systems
All Small Systems
SW< 10,000
SW> 10,000
GW< 10,000
GW> 10,000
Total Number of
Households Served
10,161,304
591 ,623
285,911
9,060,119
305,712
509,562
Mean Annual
Household
Cost Increase
$ 5.53
$ 46.48
$ 43.05
$ 2.83
$ 49.69
$ 5.97
Median Annual
Household
Cost Increase
$ 0.80
$ 18.47
$ 13.79
$ 0.80
$ 16.65
$ 1.37
90th Percentile
Annual
Household Cost
Increase
$ 10.04
$ 168.85
$ 173.53
$ 6.98
$ 109.86
$ 26.82
95th Percentile
Annual
Household Cost
Increase
$ 22.40
$ 197.62
$ 177.93
$ 11.31
$ 197.62
$ 33.84
Percentage of
Household Cost
Increase < $12
92%
38%
47%
96%
31%
79%
Percentage of
Household Cost
Increase < $120
99%
89%
85%
100%
92%
100%
Notes: Detail may not add to total due to independent rounding. Number of households served by systems adding treatment will be higher than households
served by plants adding treatment because an entire system will incur costs even if only some of the plants for that system add treatment (this would result in
lower household costs, however).
Source: Results represent the sum of treatment and non-treatment costs. Household costs for treatment are derived from household unit costs in Exhibits
7.10c and 7.11c combined with technology selection deltas, shown in Chapters. Household costs for non-treatment-related rule activites are derived from
mean costs for each system size category for implementation, IDSE, monitoring plans, additional routine monitoring, and significant excursion (as derived in
Appendix H). See section 7.5.3 for additional information on the derivation of household costs.
Final Economic Analysis for the Stage 2 DBPR
7-42
December 2005
-------
Exhibit 7.16a Household Cost Distributions,
All Surface Water Systems Subject to the Rule
1200
1000
800
HH/Y
o
O
2 600
o
.c
01
w>
3
O
I
15
c 400
200
1200
1000
800
OTC
400
200
^
99.5% 99.6% 99.7% 99.8% 99.9% 100.0%
1
1
0 -I—++-
0%
• ••
-•*-
-•-•-
20%
40% 60% 80%
Cumulative Percent of Households (N=65,389,243)
1000/
Final Economic Analysis for the Stage 2 DBPR
7-43
December 2005
-------
Exhibit 7.16b Household Cost Distributions,
All Ground Water Systems Subject to the Rule
onn ^
700 -
600
I
to
o
0
73 400 -
"o
.c
(A
3
O
1 300
CO
3
C
C
200
-inn
Q
I
800
700
600 -
500 -
400 -
300
200
100 -
0 -
1
*
/* *
« » 4«* *»* •• *
1
A
t
•
0% 20% 40% 60% 80% 100°/
Cumulative Percent of Households (N=36, 164,625)
Final Economic Analysis for the Stage 2 DBPR
7-44
December 2005
-------
Exhibit 7.17a Household Cost Distributions,
Surface Water Systems Making Treatment Technology Changes
1200 -,
1000
I
1.
to
o
o
;o
o
3
O
15
3
C
800
600
400
200
1200
1000
800
400
200
99.5% 99.6% 99.7% 99.8% 99.9% 100.0%
0 +-••
20%
40% 60% 80%
Cumulative Percent of Households (N=9,346,030)
100°X
Final Economic Analysis for the Stage 2 DBPR
7-45
December 2005
-------
Exhibit 7.17b Household Cost Distributions,
Ground Water Systems Making Treatment Technology Changes
800
700 -
600
S1"
I 500 -
to
o
O
•o 400 -
o
.c
CD
w>
3
O
CO
3
C
9nn
100
o
1
800
700
600
300
100
0 -
«
* • *
»
*
1
***
.*•+ *****
0% 20% 40% 60% 80% 100°X
Cumulative Percent of Households (N=815,274)
_F/'wa/ Economic Analysis for the Stage 2 DBPR
7-46
December 2005
-------
Exhibit 7.17c Household Cost Distributions, Small Systems Making Treatment
Technology Changes (Surface and Ground)
$1 200 -|
g
$800 -
0
o
"o
8!
3
$0 -I
0
45,637 HHs (500-999) /!
160,205 HHs (1,000-3,299) il
342,525 HHs (3,300 - 9,999) ^ jj
1 I
: I
; I
: A
• 1
' !y' ii
' -"~~ — __^^^^^
^ \ 1
% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Cumulative Percent Affected (N=591,623)
7.7 Non-Quantified Costs
All significant costs that EPA has identified have been quantified. In some instances, EPA did
not include a potential cost element because its effects are relatively minor and difficult to estimate. For
example, it may be less costly for a small system to merge with neighboring systems than to add
advanced treatment. Such changes have both costs (legal fees and connecting infrastructure) and benefits
(economies of scale). Likewise, procuring a new source of water would have costs for new infrastructure,
but could result in lower treatment costs. Operational costs such as changing storage tank operation were
not considered as alternatives to treatment. These might be options for systems with a single problem
area with a long residence time. In the absence of detailed information needed to evaluate situations such
as these, EPA has included a discussion of possible effects where appropriate. In general, however, the
expected net effect of such situations is lower costs to PWSs. Thus, the EA tends to present
conservatively high estimates of costs in relation to non-quantified costs.
7.8 Uncertainty Analysis
Many factors contribute to uncertainty in national cost estimates. Uncertainty in baseline data
inputs, such as the total number of disinfecting plants and their typical average and design flow rates, are
described in detail in Section 3.8. Other cost model inputs such as labor rates and laboratory fees also
contain uncertainties. In these cases, EPA has evaluated available data and estimated a cost input value to
Final Economic Analysis for the Stage 2 DBPR
7-47
December 2005
-------
represent the average of all water systems nationally. EPA recognizes that there is uncertainty in this
average, and variability in the characteristics of individual systems. The influence of these uncertainties
on national cost estimates is expected to be minor.
Key areas of uncertainty and their potential effects on the estimate of national costs are presented
in Exhibit 7.18. EPA believes that uncertainty in the compliance forecast has a potentially large influence
on cost (and benefit) estimates in this EA. Thus, the Agency has attempted to quantify the uncertainty by
giving equal weight to two different compliance forecast approaches. One compliance forecast approach
is based on the SWAT predictions, and the other is based on the ICR matrix method. The ICR Matrix
Method uses the same basic approach as SWAT, but uses TTHM and HAAS data from the ICR directly to
estimate the percent of plants changing technology to comply with the Stage 2 DBPR and the resulting
DBP reduction (see Chapter 5 for more information on these approaches). To characterize the uncertainty
of the compliance forecast results, EPA assumes a uniform distribution between SWAT and ICR Matrix
Method results. That is, the national cost estimates presented in this chapter represent the midpoint
between costs estimated using the SWAT model, and those estimated using the ICR Matrix Method. Cost
estimates using the SWAT model are about 25% lower than the midpoint estimates while those using the
ICR Matrix Method are about 25% higher.
In addition to quantifying some uncertainties in the compliance forecasts, EPA has explicitly
accounted for uncertainty in estimated treatment technology unit costs. Treatment costs are modeled
using a triangular distribution of ± 30 percent for Capital, and ±15 percent for O&M costs to recognize
uncertainty in the assumptions used to produce the national average unit costs.
Final Economic Analysis for the Stage 2 DBPR 7-48 December 2005
-------
Exhibit 7.18 Cost Uncertainty Summary
Uncertainty
Uncertainty in the industry
baseline (SDWIS and 1995
CWSS data)
Uncertainty in observed data and
predictive tools used to
characterize DBP occurrence and
advanced treatment technology
use for the pre-Stage 1 baseline
Uncertainty in predictive tools
used to develop the compliance
forecast for surface water
systems (SWAT and ICR Matrix
Method)
Uncertainty in ground water and
small surface water compliance
forecast methodologies
Uncertainty in the potential impact
of the IDSE on the compliance
forecast for large and medium
surface water systems
Treatment costs do not include
costs for minor operational
changes predicted by SWAT
Median operational and water
quality parameters considered for
treatment technology unit costs
Economies of scale for
combination treatment
technologies not considered
UV dose assumptions needed for
viral inactivation
Potential low-cost alternatives to
treatment not considered
Uncertainties in unit costs
Section With
Full Discussion
of Uncertainty
3.4
3
Chapters,
Appendix A
Appendices A
and B
Chapter 5
7.4.1
7.4.1
7.4.1
7.4.1
7.4.2
7.4.3
Effect on Estimate of National Costs
Underestimate
Overestimate
Unknown
Impact
X
X
Quantified in primary analysis
(addresses potential overestimate or
underestimate)
X
Quantified in primary analysis
(addresses potential underestimate)
X
X
X
X
X
Quantified in primary analysis
(addresses potential overestimate or
underestimate)
Final Economic Analysis for the Stage 2 DBPR
7-49
December 2005
-------
7.9 Comparison of Regulatory Alternatives
During the development of the Stage 2 DBPR, many regulatory alternatives were considered. Of
these alternatives, four (including the Preferred Alternative analyzed in this chapter) were chosen for
further, in-depth analysis. Chapter 4 provides a description of each alternative. Chapter 5 present
compliance forecasts for the Preferred Alternative, and Appendix C presents the compliance forecasts for
the other three alternatives, including the treatment technology selection deltas.
The same process used for developing costs for the Preferred Alternative was used to develop
costs for the other alternatives (see section 7.5 for a description of the Stage 2 DBPR cost model).
Exhibit 7.19 presents the summary of annualized costs for each alternative (detailed costs for each
alternative are presented in Appendix J). While the total annualized costs in Exhibit 7.19 include costs
for non-treatment-related rule activities (implementation, IDSE, monitoring plans, additional routine
monitoring, and operational evaluations), any increase in cost for the regulatory alternatives is fully
attributable to treatment costs.
Exhibit 7.19 Total Annualized Cost for the Stage 2 DBPR Regulatory Alternatives
($Millions)
Rule Alternative
Preferred
Alt. 1
Alt. 2
Alt. 3
Total Annualized Cost (SMillions)
3 Percent Discount Rate
Mean Estimate
$ 78.8
$ 254.1
$ 421.7
$ 634.2
7 Percent Discount Rate
Mean Estimate
$ 76.8
$ 241 .8
$ 406.4
$ 613.1
Source: Appendix J.
For the Preferred Alternative, see Exhibit J.2as for 3% and J.2aw for 7%.
For Alternative 1, see Exhibit J.3i for 3% and J.3m for 7%.
For Alternative 2, see Exhibit J.4i for 3% and J.4m for 7%.
For Alternative 3, see Exhibit J.5i for 3% and J.5m for 7%.
Final Economic Analysis for the Stage 2 DBPR
7-50
December 2005
-------
8. Economic Impact Analysis
8.1 Introduction
As part of the rulemaking process, the Environmental Protection Agency (EPA) is required to
address the potential direct and indirect burdens that the Stage 2 Disinfectants and Disinfection
Byproducts Rule (DBPR) may place on certain types of governments, businesses, and populations. This
chapter presents the analyses EPA has performed in accordance with the following 12 Federal
requirements:
1) The Regulatory Flexibility Act (RFA) of 1980, as amended by the Small Business Regulatory
Enforcement Fairness Act (SBREFA) of 1996.
2) An analysis of small system affordability to determine variance treatment technologies in
accordance with Section 1415(e)(l) of the 1996 Safe Drinking Water Act (SDWA)
Amendments.
3) Feasible treatment technologies available to all systems as required by Section 1412(b)(4)(E)
of the 1996 SDWA Amendments.
4) A Technical, Financial, and Managerial Capacity Assessment as required by Section
1420(d)(3) of the 1996 SDWA Amendments.
5) The Paperwork Reduction Act (A separate Information Collection Request document
contains the complete analysis).
6) The Unfunded Mandates Reform Act (UMRA) of 1995.
7) Executive Order 13175 (Consultation and Coordination with Indian Tribal Governments).
8) Impacts on sensitive subpopulations as required by Section 1412(b)(3)(c)(i) of the 1996
SDWA Amendments.
9) Executive Order 13045 (Protection of Children from Environmental Health Risks and Safety
Risks).
10) Executive Order 12898 (Federal Actions to Address Environmental Justice in Minority
Populations and Low-Income Populations).
11) Executive Order 13132 (Federalism).
12) Executive Order 13211 (Actions Concerning Regulations That Significantly Affect Energy
Supply, Distribution, or Use).
Many of the requirements and executive orders listed above call for an explanation of why the
rule is necessary, the statutory authority for the rule, and the primary objectives that the rule is intended to
achieve (refer to Chapter 2 for more information regarding the objectives of the rule). More specifically,
they are designed to assess the financial and health effects of the rule on sensitive, low-income, and Tribal
populations as well as on small systems. The chapter also examines how much additional capacity
systems will need to meet Stage 2 DBPR requirements and whether there are existing, feasible treatment
technologies and treatment techniques available to meet rule requirements.
8.2 Regulatory Flexibility Act and Small Business Regulatory Enforcement Fairness
Act
The Regulatory Flexibility Act (RFA) generally requires an agency to prepare a regulatory
flexibility analysis for any rule subject to notice and comment rulemaking requirements under the
Administrative Procedure Act or other statute, unless the Agency certifies that the rule will not have a
significant economic impact on a substantial number of small entities (5 U.S.C. 603(a)). Small entities
include small businesses, small organizations, and small governmental jurisdictions.
Final Economic Analysis for the Stage 2 DBPR 8-1 December 2005
-------
8.2.1 Determining Significant Impacts on Small Entities
EPA conducted a screening analysis to determine if the Stage 2 DBPR would have a significant
economic impact on a substantial number of small entities. In this analysis, EPA evaluated the potential
economic impact of the rule on small entities by comparing annualized compliance cost as a percentage of
annual revenues1 for different small-entity classifications. Chapter 3 of this Economic Analysis (EA)
provides data on the small entities potentially subject to the Stage 2 DBPR, and Chapter 7 discusses
changes systems would need to make, as well as the likely costs.2 Using information from these two
chapters, along with additional information from the Safe Drinking Water Information System (SDWIS),
the Community Water System Survey (CWSS), and the U.S. Census, EPA conducted a quantitative
analysis of small-system impacts resulting from the rule.
Defining Small Entities Affected by the Rule
The RFA provides default definitions for each type of small entity. Small entities are defined as:
(1) a small business as defined by the Small Business Administration (SBA) regulations at 13 CFR
121.201; (2) a small governmental jurisdiction that is a government of a city, county, town, school district
or special district with a population of less than 50,000; or (3) a small organization that is any "not-for-
profit enterprise which is independently owned and operated and is not dominant in its field." However,
the RFA authorizes an agency to adopt alternative definitions that "are appropriate to the activities of the
Agency after proposing the alternative defmition(s) in the Federal Register and taking comment" (5
U.S.C. §601(3)-(5)) for each category of small entity. In addition, to establish an alternative small
business definition, agencies must consult with the SBA's Chief Council for Advocacy.
The RFA references the definition of "small business" found in the Small Business Act, which
authorizes the SBA to define "small business" further by regulation. The SBA defines small businesses
by category using the North American Industry Classification System (NAICS). The NAICS code for
public water supplies (PWSs) is 22131 (Water Supply and Irrigation Systems), and State agencies that
include drinking water programs are classified as 92411 (Administration of Air and Water Resource and
Solid Waste Management Programs) or 923120 (Administration of Public Health Programs). Ancillary
systems (i.e., those that supplement the function of other establishments like factories, power plants,
mobile home parks, etc.) cannot be categorized in a single NAICS code. For ancillary systems, the
NAICS code is that of the primary establishment or industry. Examples of small businesses include
small, privately owned PWSs and for-profit businesses where provision of water may be ancillary, such
as mobile home parks or day-care centers. Examples of small organizations include churches, schools,
and homeowner associations.
The Stage 2 DBPR applies to all community water systems (CWSs) and nontransient
noncommunity water systems (NTNCWSs) that add a disinfectant other than ultraviolet light (UV) or that
deliver water that has been treated with a disinfectant other than UV. A small CWS can be a business,
government, or organization, since all entity types may supply water to the same population year round.
A NTNCWS can be a business (other than a water utility) or an organization, since both types of entities
may regularly supply water to at least 25 of the same people at least 6 months per year, but not year
1 Revenue information was used whenever available. When it was not available, different measures, such
as sales or annual operating expenditures, were used.
2 System information in this chapter is from the system baseline presented in Chapter 3 (Exhibit 3.2). For
discussions of plants making treatment technology changes, refer to the Stage 2 DBPR treatment plants baseline
(Exhibit 3.2). Systems conducting rule activities are presented in Exhibit 7.2, and plants adding treatment are
presented in Exhibit 7.3.
Final Economic Analysis for the Stage 2 DBPR 8-2 December 2005
-------
round. Although the SBA and the RFA provide clear definitions for small businesses, organizations, and
governmental jurisdictions, small entities are not necessarily small water systems. The size of the entity
has no relation to the number of people it serves as a water supply. Furthermore, data are not collected on
businesses, organizations, and governmental jurisdictions in terms of the number of water customers they
serve. Therefore, EPA chose to use an alternative definition for small entities.
For purposes of assessing the impacts of the Stage 2 DBPR on small entities, EPA considered
small entities to be PWSs serving 10,000 or fewer people, which is the cut-off level Congress specified in
the 1996 Amendments to SDWA for small system flexibility provisions. Because this definition does not
correspond to the RFA's definitions for small businesses, governments, and nonprofit organizations, EPA
requested comment on an alternative definition of a small entity in the preamble to the proposed
Consumer Confidence Report (CCR) regulation (63 FR 7620 February 13, 1998). In the preamble to the
final CCR regulation (63 FR 44511 August 19, 1998), EPA stated its intent to establish this alternative
definition for regulatory flexibility assessments under the RFA for all drinking water regulations and has,
therefore, used it for the Stage 2 DBPR.
Obtaining Data on the Number of Small Entities and Their Revenues or Expenditures
EPA obtained data on the number of small entities in each category, which are presented in
Column A of Exhibit 8.1. The numbers of entities and their distribution among categories are derived
from EPA's Baseline Handbook (USEPA 2001c) and the 1995 CWSS (USEPA 1997c), respectively.
Approximately 43.0 percent of small entities are owned by governments, 37.3 percent are owned by
businesses and ancillary systems, and 19.7 percent are owned by organizations.
EPA also estimated the annual revenues or expenditures of small entities, presented in Column B
of Exhibit 8.1. PWS inventories, managed by EPA and other organizations, have traditionally been
categorized by size and by the characteristics of the population served (i.e., CWSs, NTNCWSs, and
transient noncommunity water systems (TNCWSs)) rather than by NAICS code. Revenues by NAICS
code are not readily applicable to EPA's categorization of systems. Therefore, alternative methods for
determining revenue were developed, as discussed below.
The estimated revenues for small entities in Exhibit 8.1 are from the Bureau of the Census (U.S.
Department of Commerce 1992), the SDWIS, and additional data on independent privately owned CWSs,
special districts, and authorities from the 1995 CWSS (USEPA 1997c). Column A of Exhibit 8.1 shows
the numbers of entities classified as small businesses, governments, and organizations, obtained using
information from the Third Edition of the Baseline Handbook (USEPA 200 Ic). These numbers were
used to determine the weighted averages of estimated revenue. Column C shows the estimated revenues.
Small-government entities include municipal, county, state, federal, military, and special district
systems. Data on all revenue for townships and municipalities were obtained from the 7992 Census of
Governments (U.S. Department of Commerce 1992) and converted to 2003 dollars by applying a
conversion factor calculated from the national income and product account tables of the U.S. Bureau of
Economic Analysis.3 Specifically, the price deflators for 1992 and 2003 were obtained from Chain-Type
Price Indexes for State and Local Governments (U.S. Department of Commerce BEA 2004a). The
average revenue for all small governments with PWSs was calculated at $2,649,186.
Small-business entities in Exhibit 8.1 include both CWSs and NTNCWSs, such as privately
owned CWSs, mobile home parks, country clubs, hotels, manufacturers, hospitals, and other
establishments. For this analysis, all hospitals and day care centers are assumed to be businesses, as are
3Methodology recommended by Bruce E. Baker, State and Local Governments, Government Division, U.S.
Bureau of Economic Analysis.
Final Economic Analysis for the Stage 2 DBPR 8-3 December 2005
-------
50 percent of systems classified as "other."4 Estimated average revenue for the small businesses affected
by the Stage 2 DBPR is $2,555,888.
Small organizations include primarily nonprofit NTNCWS such as schools and homeowner
associations. The revenue estimates for small nonprofit organizations serving 500 or more people are
actually higher than those for small businesses because the total number of such systems is small, and a
large proportion of these organizations are schools and colleges with large budgets. This category also
includes 50 percent of systems classified as "other." The average estimated revenue for small
organizations affected by the Stage 2 DBPR is $4,750,838.
EPA also calculated the average estimated revenue for all small entities. This estimate is
weighted to account for the number of small entities in each category (government, business, and
organization) affected by the Stage 2 DBPR. This overall average is $2,981,331.
Measuring Significant Impacts
To evaluate the impact that a small entity is expected to incur as a result of the rule, this analysis
calculates the entity's annualized compliance cost as a percentage of sales (for privately owned entities)
or the entity's annualized compliance cost as a percentage of annual governmental revenue or
expenditures (for publicly owned entities). The Interim Guidance for EPA Rulewriters for the RFA as
amended by the SBREFA (March 1999) suggests using 1 percent as a threshold for determining
significance, although additional factors may be considered. If compliance costs are less than 1 percent of
sales or revenues for fewer than 1,000 entities, which represent less than 20 percent of all affected small
entities, then in most cases there is no significant impact. In addition, the guidance suggests that if fewer
than 100 entities experience economic impacts of 3 percent of their revenues or greater, then in most
cases there is no significant impact.
Exhibit 8.1 presents the data that EPA used for the screening analysis. The numbers of entities
expected to incur costs of more than 1 and 3 percent of their revenues are presented in Columns D and F,
respectively. The numbers of entities experiencing impacts of more than 1 and 3 percent of their revenues
were compared to the total number of entities in each size category to calculate percentages, shown in
Columns E and G.
4The "other" category contains systems that do not yet have a specific function identified.
Final Economic Analysis for the Stage 2 DBPR 8-4 December 2005
-------
Exhibit 8.1 Annualized Compliance Cost as a Percentage of Revenues for All
Small Entities
Small Systems by
Source of Water and
Type of Entity
Number of
Small
Systems
A
Percent of
Small
Systems
B
Average
Annual
Estimated
Revenues1 per
System ($)
C
Systems Experiencing
Costs of >1% of their
Revenues 2|3
Number of
Systems
D=A*E
Percent of
Systems
E
Systems Experiencing
Costs of >3% of their
Revenues 2|4
Number of
Systems
F=A*G
Percent of
Systems
G
Primarily Surface Water and GWUDI Systems
Small Governments
Small Businesses
Small Organizations
All Small Entities Using
Primarily Surface Water or
GWUDI
1,827
1,584
838
4,250
43%
37%
20%
100%
$2,649,186
$2,555,888
$4,750,838
$2,981,331
41
36
15
92
2.24%
2.25%
1.84%
2.16%
18
16
6
40
1 .00%
1 .03%
0.66%
0.94%
Primarily Ground Water Systems
Small Governments
Small Businesses
Small Organizations
All Small Entities Using
Primarily Ground Water
14,865
12,888
6,817
34,570
43%
37%
20%
100%
$2,649,186
$2,555,888
$4,750,838
$2,981,331
176
153
25
354
1.18%
1.18%
0.37%
1.02%
18
18
8
45
0.12%
0.14%
0.12%
0.13%
Note: Detail may not add due to independent rounding.
1 Revenue information was used whenever available. When it was not available, other measures such as sales or
annual operating expenditures were used. Data were not available to differentiate revenue for small entities by
system sizes or by source water type. The revenue estimates reflect ground water as well as surface water systems.
2 Compliance costs were compared to average annual revenue to determine whether 20 percent of small entities
would incur costs exceeding 1 percent of their average annual revenues. Thresholds to determine whether a rule has
a significant impact on a substantial number of small entities are taken from the Interim Guidance for EPA Rulewriters
for the RFA as amended by the SBREFA (March 1999).
3 Compliance costs incurred by each entity were compared to 1 percent of average annual revenues to determine
whether 1,000 or more entities will experience an impact of 1 percent or greater of average annual revenues.
4 Compliance costs incurred by each entity were compared to 3 percent of average annual revenues to determine
whether 100 or more entities will experience an impact of 3 percent or greater of average annual revenues.
Sources:
(A) Number of disinfecting CWSs and NTNCWSs serving fewer than 10,000 people from the system baseline in
Exhibit 3.2 (Column V), multiplied by 43%, 37.3%, and 19.7% to obtain number of small government, small
businesses, and small organizations, respectively.
(B) Percent of small governments, businesses, and organizations derived from the 1995 CWSS (USEPA 1997c).
(C) Small Governments: Revenues from 1992 Census of Governments, GC92(4)-4: Finances of Municipal and
Township Governments, U.S. Dept. of Commerce, Bureau of the Census; price deflators from Table 8.11, Chain-Type
Quantity and Price Indexes for Government. All other price adjustments were calculated using the Consumer Price
Index.
(E, G) Derived from the Stage 2 DBPR Cost Model (USEPA 2005i).
To consider whether 1,000 or more entities will experience an impact of 1 percent or greater of
average annual revenues, EPA compared compliance costs incurred by each entity to 1 percent of average
annual revenue. EPA estimated that a total of 92 small entities using surface water or ground water under
the influence of surface water (GWUDI) and 354 small entities using ground water, representing 2.16 and
1.02 percent of all small entities affected by the Stage 2 DBPR, respectively, will experience an impact of
1 percent or greater of average annual revenues. This is less than the criteria of 1,000 entities used to
determine significant impact.
Using a similar methodology, EPA also considered whether 100 or more entities will experience
an impact of 3 percent or greater of average annual revenues. The Agency determined that 40 small
Final Economic Analysis for the Stage 2 DBPR
8-5
December 2005
-------
entities using surface water or GWUDI and 45 small entities using ground water, representing 0.94 and
0.13 percent of all small entities subject to the Stage 2 DBPR, respectively, will experience an impact of 3
percent or greater of average annual revenues. This is less than the criterion of 100 entities used to
determine significant impact.
Based on the large number of small entities, the nature of the economics for community water
systems, and the information presented in Exhibits 8.1, EPA is certifying that the Stage 2 DBPR will not
lead to significant economic impacts for a substantial number of small entities. CWSs have many
resources available to them that other industries do not have. For example, financial assistance to small
systems may be available from programs administered by EPA or other Federal agencies, as described in
sections 8.2.2 and 8.3.3.
Because EPA is certifying that the Stage 2 DBPR will not lead to significant economic impacts
for a substantial number of small entities, EPA is not required by the RFA, as amended by SBREFA, to
conduct a final regulatory flexibility analysis (FRFA). Nevertheless, EPA has tried to reduce the impact
of this rule on small systems.
8.2.2 Summary of the SBREFA Process
The RFA, as amended by SBREFA, and Section 203 of UMRA require EPA to provide small
governments with an opportunity for timely and meaningful participation in the regulatory development
process. EPA provided stakeholders, including small governments, with several opportunities to provide
input on the Stage 2 DBPR. For example, EPA conducted three conference calls in Washington, DC to
solicit feedback and information from the Small Entity Representatives (SERs) on Stage 2 DBPR impacts
on small systems. SERs included small-system operators, local government officials, and small nonprofit
organizations.
During the first call, held on January 28, 2000, EPA presented an overview of SDWA, as
amended in 1996, and SBREFA. Issues and schedules for the Stage 2 DBPR rules were also discussed.
The second call was held on February 25, 2000. EPA presented the stakeholders with an overview of the
EPA regulatory development process and background on the development of the Stage 2 Microbial-
Disinfectants/Disinfection Byproduct (M-DBP) Rules, particularly regarding health risks, issues and
options identified by the Federal Advisory Committees Act (FACA) Committee, and DBP and microbial
occurrence in small systems. The third meeting was held on April 7, 2000. EPA presented SERs with a
cost estimate and an impact analysis for selected regulatory options. In addition, EPA presented SERs
with schedules for the FACA and SBREFA processes.
These three conference calls generated a wide range of information, issues, and technical input
from SERs. In general, the SERs were concerned about the impact of these proposed rules on small water
systems (because of their small staff and limited budgets), small systems' ability to acquire the technical
and financial capability to implement requirements, maintaining the flexibility to tailor requirements to
their needs, and other limitations of small systems. The Agency used the feedback received during these
meetings in developing the Stage 2 DBPR. EPA also mailed a draft version of the rule's preamble to the
attendees of these meetings.
The Agency convened a Small Business Advocacy Review (SBAR) Panel to obtain advice and
recommendations of representatives of the regulated small entities, including those of small local
governments, in accordance with Section 609(b) of the RFA. The small entity stakeholders received
background on the need for the rule and the possible components of the rule to assist them with their
deliberations. EPA convened the SBAR Panel after completing the consultation meetings with SERs on
Final Economic Analysis for the Stage 2 DBPR 8-6 December 2005
-------
the Stage 2 DBPR. Eight of the small entities were governments. SER's concerns were provided to the
SBAR Panel when the Panel convened on April 25, 2000.
8.3 Small-System Affordability
Section 1415(e)(l) of SDWA allows States/Primacy Agencies to grant variances to small water
systems (i.e., those serving fewer than 10,000 people) in lieu of complying with a maximum contaminant
level (MCL) if EPA determines that no nationally affordable compliance treatment technologies exist for
that combination of system size and water quality. These variances may be granted only when EPA has
identified a variance treatment technology under Section 1412(b)(15) forthe contaminant, system size,
and source water quality in question. To list variance treatment technologies, three showings must be
made.
1) EPA must determine, on a national level, that there are no compliance treatment technologies
that are available and affordable for the given combination of system size and source water
quality.
2) If there is no nationally affordable compliance treatment technology, EPA must identify a
variance treatment technology that may not reach the MCL, but that will achieve the
maximum contaminant reduction affordable. This treatment technology must be listed as a
small-system variance treatment technology by EPA for small systems to be able to rely on it
for regulatory purposes.
3) EPA must make a finding, on a national level, that the use of the variance treatment
technology would be protective of public health.
The State/Primacy Agency must then make a determination for each system as to whether the
system can afford to meet the MCL based on affordability criteria developed by the State/Primacy
Agency. If the State/Primacy Agency determines that compliance is not affordable for the system, it may
grant a variance, but it must establish terms and conditions, as necessary, to ensure that the variance
adequately protects human health.
The 1996 SDWA Amendments identify three categories of small PWSs that need to be addressed:
(1) those serving a population of 3,301 to 10,000; (2) those serving a population of 501 to 3,300; and (3)
those serving a population of 25 to 500. SDWA requires EPA to make determinations of available
compliance treatment technologies and, if needed, variance treatment technologies for each size category.
A compliance treatment technology is a technology that is affordable and that achieves compliance with
the MCL and/or treatment technique. Compliance treatment technologies can include point-of-entry
(POE) or point-of-use (POU) treatment units. Variance treatment technologies are specified only for
those system size/source water quality combinations for which there are no listed compliance treatment
technologies.
EPA determined that affordable compliance treatment technologies for each of the three
categories of small systems are available for small systems forthe Stage 2 DBPR. Therefore, variance
treatment technologies are not required. The following sections show how small system affordability was
evaluated for the Stage 2 DBPR. The analysis is consistent with the current methodology used in the
document National-Level Affordability Criteria Under the 1996 Amendments to the Safe Drinking Water
Act (USEPA 1998c) and the Variance Technology Findings for Contaminants Regulated Before 1996
(USEPA 1998d).
Final Economic Analysis for the Stage 2 DBPR 8-7 December 2005
-------
8.3.1 Affordability Threshold
EPA discussed its draft national-level affordability criteria in the August 6, 1998, Federal
Register for the contaminants regulated before 1996. National-level affordability criteria were developed
by identifying an "affordability threshold" (i.e., the total annual household water bill that would be
considered affordable). In developing this threshold, EPA considered the percentage of median
household income (MHI) spent by an average household on comparable goods and services, including
housing (28 percent), transportation (16 percent), food (12 percent), energy and fuels (3.3 percent),
telephone (1.9 percent), water and other public services (0.7 percent), entertainment (4.4 percent), and
alcohol and tobacco (1.5 percent). Another key factor that EPA used to select an affordability threshold
was cost comparisons with other risk reduction activities for drinking water. Section 1412(b)(4)(E)(ii) of
SDWA identifies both POU and POE devices as compliance technologies for small systems. EPA
examined the projected costs of these options, and also investigated the costs associated with supplying
bottled water for drinking and cooking. The median income percentages associated with these risk-
reduction activities were more than 2.5 percent for POE devices and bottled water, and 2 percent for POU
devices. Based on the foregoing analysis, EPA developed an affordability criterion of 2.5 percent of MHI
for the affordability threshold (USEPA 1998c).
The median water bill for households in each small system size category was subtracted from the
affordability threshold to obtain the affordable level of expenditure per household for new treatment. This
difference is referred to as the "available expenditure margin." Based on EPA's 1995 CWSS, median
water bills were about $250 per year for small system customers. However, the available expenditure
margins are expected to change because water rates and MHI have increased. The 1995 MHI was
updated to 2003 dollars using the Consumer Price Index (CPI) (BLS 2004). The results are shown in
Exhibit 8.2.
The baseline for annual water bills (median water bill from the 1995 CWSS) also increased to
account for regulations promulgated after 1996, but before the Stage 2 DBPR is promulgated. For each
rule promulgated after 1996, the total national costs for each small size category were averaged over the
number of households within that size category. The mean costs per household for each size category
were added to the national median annual household water bills. This was done for the following rules:
the Arsenic Rule, Long Term 1 Enhanced Surface Water Treatment Rule (LT1ESWTR), Stage 1 DBPR,
Filter Backwash Recycling Rule, and Radionuclides Rule.5 The adjusted water bill may underestimate
costs, which were spread over the total number of households by size category, as opposed to only the
households affected by each rule. Exhibit 8.2 presents the analysis before and after the baseline water bill
was adjusted for regulations promulgated after 1996.
5 To adjust the baseline water bills to account for costs, EPA spread total annual or annualized costs
attributable to regulations promulgated after 1996 (at a 3 percent discount rate) over the total number of households
by size category.
Final Economic Analysis for the Stage 2 DBPR 8-8 December 2005
-------
Exhibit 8.2 Derivation of Available Expenditure Margin
Population
Served
0-500
501 -3,300
3,301 -10,000
Small System Available Expenditures
Baseline (2003$)
MHI (2003)
A
$ 40,734
$ 38,655
$ 39,536
Pre-1996
Median
Waterbills (2003)
B
$ 255
$ 222
$ 219
Adjusted
Waterbills
C
$ 285
$ 243
$ 239
%MHI for
Waterbills
D=C/A
0.70%
0.63%
0.60%
Affordability
Threshold
E=A*2.5%
$ 1,018
$ 966
$ 988
Available
Expenditure
Margin
F=E-C
$ 733
$ 724
$ 750
Sources:
A) 1995 CWSS and 2000 Census data, adjusted to 2003$ using the Consumer Price Index (BLS 2004).
B) 1995 CWSS (as reported in the document "Variance Technology Findings for Contaminants Regulated
Before 1996), adjusted to 2003$ using the Consumer Price Index (BLS 2004).
C) Pre-1996 Median Water Bills (2003) from Column B plus mean HH costs for drinking water rules since
1996.
8.3.2 Affordable Compliance Treatment Technologies
Section 1412(b)(4)(E)(ii) of SDWA, as amended in 1996, requires EPA to list treatment
technologies that achieve compliance with MCLs established under the Act that are affordable and
applicable to typical small drinking water systems. Owners and operators may choose any treatment
technology or technique that best suits their conditions, as long as the MCL is met.
This section presents household costs ($ per household per year) for various treatment
technologies. The methodology for generating household cost estimates is explained in detail in section
7.6.4. In general, the analysis in this section followed the methodology in section 7.6.4; however, some
inputs for household cost calculations in this section are different and, in some cases, are conservatively
high compared to data used to generate household cost distributions in Chapter 7. A conservatively high
estimate of household costs is used to more accurately represent the high-end variability in household
costs. This allows affordability of the Stage 2 DBPRto be more confidently assessed across the range of
all affected small systems.
The size categories specified in SDWA for affordable treatment technology determinations are
different than the nine standard size categories used in the majority of this EA, and subsequently, mean
design and average daily flows for each category are different. The values for design and average flows
(shown in Exhibit 8.2) are derived from the 1995 CWSS (USEPA 1997c).
Final Economic Analysis for the Stage 2 DBPR
8-9
December 2005
-------
Exhibit 8.3 Affordability Analysis Inputs
System Size
(Population
Served)
25-500
501-3,300
3,301-10,000
Flows (mgd)
Median
Average Daily
Flow
a
0.015
0.17
0.7
Design Flow
b
0.058
0.5
1.8
Median HH
Consumption
Rates (kgal/yr)
c
72
74
77
HH
Consumption
Rate Used for
Costing
d = 1.15*c
83
85
89
Sources:
A, B, and C: 1995 CWSS (USEPA 1997c).
D: Consumption rates were adjusted upward by 15 percent to account for distribution system leaks.
For each treatment technology, unit treatment costs ($ per 1,000 gallons) are estimated using the
flow rates shown in Exhibit 8.3 and the technology unit costs in Appendix I (see Exhibit 1.27 for details
on the derivation of unit treatment costs). As suggested in Variance Technology Findings for
Contaminants Regulated Before 1996 (USEPA 1998d), capital costs were annualized using a 7 percent
discount rate rather than the cost-of-capital rates used to generate the distribution of household costs in
Chapter 7. The unit treatment costs ($ per 1,000 gallons) were multiplied by annual household
consumption rates to determine the annual household cost increase ($ per household) for each treatment
technology. Annual consumption rates are shown in Exhibit 8.3 and represent median yearly
consumption derived from the 1995 CWSS (USEPA 1997c). (Note that mean yearly household
consumption are shown in Exhibit 7.9 and are used for the household cost estimates in Chapter 7.) The
values shown in Exhibit 8.3 were adjusted upward by 15 percent to account for water lost in the
distribution system due to leaks, as suggested in Variance Technology Findings for Contaminants
Regulated Before 1996 (USEPA 1998d).
Exhibits 8.4a and 8.4b show the compliance treatment technologies for the Stage 2 DBPR for
surface water and ground water systems along with their mean annual household costs for each of the
three size categories. Exhibit 8.4c presents annual household cost increases for all households served by
plants installing treatment to comply with the Stage 2 DBPR for systems serving 0 to 500, 501 to 3,300
and 3,301 to 10,000 people6. The mean, median, 90th percentile, and 95th percentile values are shown as
well as the available expenditure margin and the number of households and plants that will experience
annual cost increases above the available expenditure margin.
For a $733 affordability threshold in the 0 to 500 category, integrated membranes with
chloramines and granulated activated carbon (GAC) with advanced disinfectants are above the
affordability threshold. Fourteen plants are expected to install GAC20 with advanced disinfectants and
one plant is expected to install integrated membranes with chloramines to comply with the rule. In the
500 to 3,300 category and the 3,300 to 10,000 category, no treatment technologies are above the
affordability thresholds for these size categories.
Although the size categories specified by SDWA for the affordability analysis do not specifically include
systems serving fewer than 25 people (per SDWA), these systems are included in all other analyses in this EA and
are accounted for in Exhibit 8.4c. Thus, the estimate of the number of systems and households experiencing cost
increases in Exhibit 8.4c is conservatively high.
Final Economic Analysis for the Stage 2 DBPR
8-10
December 2005
-------
Exhibit 8.4a Affordable Compliance Treatment Technologies and Household Unit
Treatment Costs ($/HH/Year) for Surface Water Systems
Compliance Technologies
Chloramines (0.15 mg/L)
Chlorine Dioxide (1.25 mg/L) 1
UV (40mJ/cm2)
MF/UF1
GAC20 (EBCT=20 min, 90 day regeneration) 1
GAC20 + Advanced Disinfectants
Integrated Membranes
System Size (Population Served)
0-500
$ 62.88
$ 121.92
$ 89.21
$ 583.74
$ 644.69
$ 733.90
$ 907.52
501-3,300
$ 9.41
$ 29.22
$ 20.55
$ 174.74
$ 209.42
$ 366.53
$ 316.45
3,301-10,000
$ 5.03
$ 9.99
$ 14.59
$ 115.86
$ 134.39
$ 195.02
$ 240.22
Exhibit 8.4b Affordable Compliance Treatment Technologies and Household Unit
Treatment Costs ($/HH/Year) for Ground Water Systems
Compliance Technologies
Chloramines (0.15 mg/L)
UV (200mJ/cm2)
Ozone (0.5-log dose) 1
GAC20 (EBCT=20 min, 240 day regeneration) 1
Nanofiltration 1
System Size (Population Served)
0-500
$ 62.70
$ 236.32
$ 1,300.43
$ 414.74
$ 323.78
501-3,300
$ 9.28
$ 65.81
$ 157.11
$ 159.38
$ 141.71
3,301-10,000
$ 4.88
$ 50.38
$ 60.62
$ 95.81
$ 124.36
1 Zero percent of plants were predicted to select this treatment technology.
Source: Exhibit 1.27.
Final Economic Analysis for the Stage 2 DBPR
8-11
December 2005
-------
Exhibit 8.4c Distribution of Household Unit Treatment Costs for Plants Adding Treatment
Systems Size
(population
served)
0-500
501 - 3,300
3,301 - 10,000
Number of Households
Served by Plants
Adding Treatment
(Percent of all
Households Subject to
the Stage 2 DBPR)
A
43045 ( 3% )
205842 ( 4% )
342525 ( 5% )
Mean Annual
Household
Cost Increase
B
$201 .55
$58.41
$37.05
Median Annual
Household
Cost Increase
C
$299.01
$29.96
$14.59
90th Percentile
Annual
Household
Cost I ncrease
D
$299.01
$75.09
$55.25
95th Percentile
Annual
Household
Cost Increase
E
$414.74
$366.53
$200.05
Available
Expenditure
Margin
($/hh/yr)
F
$733
$724
$750
Number of
Households with
Annual Cost
Increases Greater
then the Available
Expenditure Margin
G
964
0
0
Number of Surface
Water Plants with
Annual Cost
Increases Greater
than the Available
Expenditure Margin
H
15
0
0
Number of
Groundwater Plants
with Annual Cost
Increases Greater
than the Available
Expenditure Margin
I
0
0
0
Total Number of
Plants with Annual
Cost Increases
Greater than the
Available
Expenditure Margin
J = H + l
15
0
0
Notes: Household unit costs represent treatment costs only, as presented in Exhibits 8.4a and 8.4b.
Source: Household unit costs in Exhibits 8.4a and 8.4b combined with treatment technology selection deltas in Exhibits 5.11 and 5.14.
Final Economic Analysis for the Stage 2 DBPR
8-12
December 2005
-------
EPA believes, however, that the number of plants in small systems predicted to add advanced
treatment technologies is overstated for two reasons: 1) distribution system modifications are not
considered in the compliance forecast, and 2) Stage 2 DBPR requirements for small systems are similar to
Stage 1 DBPR requirements and might not trigger compliance violations.
Very few households will experience cost increases above the available expenditure margin as a
result of adding advanced treatment technology to comply with the Stage 2 DBPR. However, these
predictions are likely overestimated because small systems have lower-cost alternatives available to them
besides adding treatment, such as operating on a part-time basis, identifying an alternative water source,
and connecting to a neighboring water system. Low-cost alternatives to reduce total trihalomethanes
(TTHM) and haloacetic acid (HAAS) levels also include distribution system modification such as
reducing average residence time include flushing distribution mains more frequently, eliminating loops
and dead ends, and optimizing storage to minimize retention time. Refer to Chapter 7 for a more detailed
discussion of household cost increases.
Under the Stage 1 DBPR, surface and ground water systems serving fewer than 500 people must
have one site for TTHM and HAAS monitoring. Under the Stage 2 DBPR, systems have to add a site
only if their highest TTHM and HAAS concentrations are at different locations. EPA estimates that
approximately 3/4 of systems serving fewer than 100 people will have one site representing both their
highest TTHM and HAAS concentrations for the Stage 2 DBPR. Systems with one site for Stage 1 and
one site for the Stage 2 DBPR would produce the same measure of compliance whether calculated using a
running annual average (RAA) (Stage 1 compliance) or a locational running annual average (LRAA)
(Stage 2 compliance).
In addition, it is anticipated that systems currently predicted by EPA to select higher-cost
technologies may be able to use less expensive treatment technologies by the time the Stage 2 DBPR is
implemented. This is because the compliance decision tree (summarized in Chapter 3 and described in
detail in Appendices A and B) represents current limitations on the use of inexpensive treatment
technologies (e.g., chloramines and UV) taking into account operational and constructability constraints.
These limitations may not exist by the time the rule is implemented due to advances in treatment and
innovations by manufacturers.
Under section 1416(a), EPA or a State may exempt a PWS from any requirements related to an
MCL or treatment technique of an NPDWR, if it finds that (1) due to compelling factors (which may
include economic factors such as qualification of the PWS as serving a disadvantaged community), the
PWS is unable to comply with the requirement or implement measure to develop an alternative source of
water supply; (2) the exemption will not result in an unreasonable risk to health; and; (3) the PWS was in
operation on the effective date of the NPDWR, or for a system that was not in operation by that date, only
if no reasonable alternative source of drinking water is available to the new system; and (4) management
or restructuring changes (or both) cannot reasonably result in compliance with the SDWA or improve the
quality of drinking water.
8.3.3 Funding Options for Disadvantaged Systems
EPA believes that there is another mechanism in SDWA to address cost impacts on small systems
that serve primarily low-income households. Systems that meet criteria established by the State/Primacy
Agency could be classified as disadvantaged communities under §1452(d) of SDWA. They can receive
additional subsidies through the Drinking Water State Revolving Fund (DWSRF), including forgiveness
of principal. Under DWSRF, States/Primacy Agencies must provide a minimum of 15 percent of the
available funds for infrastructure loans to systems serving fewer than 10,000 or fewer people. Two
percent of the State's/Primacy Agency's grant is set-aside funding that can only be used to provide
Final Economic Analysis for the Stage 2 DBPR 8-13 December 2005
-------
technical assistance to small systems. In addition, up to 14 percent of the State's/Primacy Agency's grant
may be used to provide technical, managerial, and financial assistance to all system sizes. For small
systems that are disadvantaged, as defined by the State/Primacy Agency, up to 30 percent of a
State's/Primacy Agency's DWSRF may be used for increased loan subsidies. This assistance can take the
form of lower interest rates, principal forgiveness, or negative interest rate loans. The State may also
extend repayment terms of loans for disadvantaged communities to up to 30 years.
Small systems will be encouraged to discuss their infrastructure needs for complying with the
Stage 2 DBPR with their State/Primacy Agency to determine their eligibility for DWSRF loans, and, if
eligible, to ask for assistance in applying for the loans.
In addition to the DWSRF, money is available from the Department of Agriculture's Rural Utility
Service (RUS) and Housing and Urban Development's Community Development Block Grant (CDBG)
Program. RUS provides loans, guaranteed loans, and grants to improve, repair, or construct water supply
and distribution systems in rural areas and towns with a population of up to 10,000 people. In fiscal year
2003, RUS had over $1.5 billion of available funds for water and environmental programs. Also, three
sources of funding exist under the CDBG program to finance building and improvements of public
faculties such as water systems. These include: 1) direct grants to communities with populations over
200,000; 2) direct grants to States/Primacy Agencies, which in turn are awarded to smaller communities,
rural areas, and colonas in Arizona, California, New Mexico, and Texas; and 3) direct grants to U.S.
territories and trusts. The CDBG budget for the formula program for fiscal year 2003 totaled over $4.4
billion.
8.4 Feasible Treatment Technologies for All Systems
In accordance with Section 1412(b)(4)(E) of the 1996 SDWA Amendments, EPA examined
whether there were existing, feasible treatment technologies and treatment techniques available that
would allow systems to meet the Stage 2 DBPR requirements. EPA examined alternatives for best
available technologies (BATs) using two methods: Information Collection Rule (ICR) treatment studies
and Surface Water Analytical Tool (SWAT) predictions. A discussion of the evaluation is provided in
sections 8.4.1 and 8.4.2. Results from these two evaluations show that all systems can meet the TTHM
and HAAS LRAA MCLs (80 micrograms per liter ((ig/L) and 60 (ig/L, respectively) using one of the
three following treatment technologies:
1) GAC adsorbers with at least 10 minutes of empty-bed contact time and an annual average
reactivation/replacement frequency no greater than 120 days, plus enhanced coagulation or
enhanced softening.
2) Nanofiltration using a membrane with a molecular weight cutoff of 1000 Daltons or less.
3) GAC adsorbers with at least 20 minutes of empty-bed contact time and an annual average
reactivation/replacement frequency no greater than 240 days.
Section 8.4.3 discusses BATs specifically for consecutive systems.
Final Economic Analysis for the Stage 2 DBPR 8-14 December 2005
-------
8.4.1 ICR Treatment Studies
The ICR treatment studies were designed to evaluate the technical feasibility of using GAC and
nanofiltration to remove DBF precursors prior to the addition of chlorine-based disinfectants (USEPA
2000o, Hooper and Allgeier 2002). EPA used TOC levels in the source or finished water to determine
whether the ICR treatment study requirement applied to plants. Specifically, surface water plants with
annual average source water TOC concentrations greater than 4 milligrams per liter (mg/L) and ground
water plants with annual average finished water TOC concentrations greater than 2 mg/L were required to
conduct treatment studies. Thus, the plants required to conduct treatment studies generally had waters
with organic DBF precursor levels that were significantly higher than the national means of 3.2 mg/L and
1.5 mg/L for ICR surface and ground water plants, respectively (USEPA 20051).
Plants that used GAC typically evaluated performance at two empty-bed contact times, 10 and 20
minutes, and over a range of operational times to evaluate the unsteady nature of TOC removal by GAC.
This allowed GAC performance to be assessed with respect to empty-bed contact time, as well as
reactivation/replacement frequency. Plants that conducted membrane treatment studies evaluated one or
two nanofiltration membranes with molecular weight cutoffs less than 1,000 Daltons. Regardless of the
treatment technology evaluated, all treatment studies evaluated post-treatment DBP formation under
distribution system conditions representative of the full-scale plant at the average residence time, using
free chlorine as the primary and residual disinfectant.
The results of the ICR treatment study suggest that GAC would be an appropriate treatment
technology for surface water systems and some ground water systems with influent TOC concentrations
below approximately 6 mg/L (USEPA 2000o). (The ICR and National Rural Water Association (NRWA)
data indicate that over 90 percent of plants have average influent TOC levels below 6 mg/L.) Larger
systems would likely realize an economic benefit from on-site reactivation, which could allow them to
use smaller, 10-minute empty-bed contact time contactors with more frequent reactivation (i.e., 120 days
or less). Most small utilities would not find it economically advantageous to install on-site carbon
reactivation facilities, and thus would opt for larger, 20-minute empty-bed contact time contactors, with
less frequent carbon replacement (i.e., 240 days or less). EPA recognizes that some small systems
attempting to implement GAC20 may face GAC supply challenges.
Theoretically, there is a linear relationship between empty-bed contact time and reactivation
interval. Assuming equivalent performance, a doubling of the empty-bed contact time would be expected
to result in a doubling of the reactivation interval. If this is the case, the 10-minute empty-bed contact
time contactor reactivated at 120 days should result in equivalent performance to a 20-minute empty-bed
contact time contactor reactivated at 240 days. However, the ICR treatment study data demonstrated that
the 20-minute contactors generally outperform the 10-minute contactors. On the other hand, larger
systems will typically operate with a larger number of parallel contactors compared to small systems,
resulting in improved performance. Thus, the benefit that small systems gain by using a larger empty-bed
contact time will be offset by use of a smaller number of parallel contactors. Based on these
considerations, the proposed reactivation/replacement interval for the 20-minute contactor is simply
double the reactivation/replacement interval for a 10-minute contactor.
The ICR treatment study demonstrated that approximately 70 percent of the surface water plants
that conducted GAC studies could meet the 80/60 (ig/L TTHM/HAA5 MCLs with a 20 percent safety
factor (i.e., 64 (ig/L and 48 (ig/L, respectively) using GAC with 10 minutes of empty-bed contact time
and a 120-day reactivation frequency. It also showed that 78 percent of the plants could meet the MCLs
using GAC with 20 minutes of empty-bed contact time and a 240-day reactivation frequency. As
discussed previously, the treatment studies were conducted at plants having poorer source and finished
water quality than the national average. Therefore, EPA believes that the percentages of plants in the
Final Economic Analysis for the Stage 2 DBPR 8-15 December 2005
-------
GAC studies that could meet the MCLs with the BATs translate to much higher percentages of plants nationwide.
The ICR treatment study results also demonstrated that nanofiltration was the better DBF control
treatment technology for ground water sources with high TOC concentrations (i.e., above approximately
6 mg/L). The results of the membrane treatment studies showed that all ground water plants could meet
the 80/60 (ig/L TTHM/HAA5 MCLs with a 20 percent safety factor (i.e., 64 (ig/L and 48 (ig/L,
respectively) at the average distribution system residence time using nanofiltration (USEPA 2000o,
Hooper and Allgeier 2002). Although nanofiltration is generally more expensive than GAC, it would be
less expensive than GAC for high TOC ground waters that require minimal pretreatment. Also,
nanofiltration is an accepted treatment technology for treatment of the high-TOC ground waters in areas
of the country, such as Florida and parts of the southwest.
8.4.2 BAT Evaluation Using SWAT
The second method that EPA used to examine alternatives for BAT was SWAT, which was
developed to compare alternative regulatory strategies as part of the Stage 1 and Stage 2 M-DBP
Advisory Committee deliberations (Seidel 2001). EPA considered the following BAT options:
Enhanced coagulation (EC)/softening with chlorine
• EC/softening with chlorine and no pre-disinfection
• EC and GAC10
• EC and GAC20
EC and chloramines
EC/softening is required under the Stage 1 DBPR for conventional plants. In the model, GAC 10
was defined as granular activated carbon with an empty-bed contact time of 10 minutes and a reactivation
frequency of no more than 90 days. GAC20 was defined as granular activated carbon with an empty-bed
contact time of 20 minutes and a reactivation or replacement frequency of no more than 90 days. EPA
assumed that systems would be operating to achieve both the Stage 2 MCLs of 80 (ig/L TTHM and 60
(ig/L HAAS as an LRAA and the Surface Water Treatment Rule (SWTR) removal and inactivation
requirements of 3-log for Giardia and 4-log for viruses. EPA also evaluated the BAT options under the
assumption that plants operate to achieve DBP levels 20 percent below the MCL (i.e., a 20 percent safety
factor). These assumptions along with other inputs for the SWAT runs are consistent with those specified
in Appendix A.
The compliance percentages forecasted by SWAT are indicated in Exhibit 8.5. EPA estimates
that over 97 percent of large systems will be able to achieve the Stage 2 MCLs, regardless of post-
disinfection choice, if they apply the BAT (i.e., EC and GAC10). As shown in the current Occurrence
Document (USEPA 2005k), the source water quality (e.g., DBP precursor levels) in medium and small
systems is comparable to or belter than that for large systems. Using the large-system estimate as a proxy
for medium and small systems, EPA believes it is conservative to assume that at least 90 percent of these
systems will be able to achieve the Stage 2 MCLs if they were to apply one of the GAC BATs. EPA
realizes that it may not be economically feasible for small systems to install and operate an on-site GAC
reactivation facility. Thus, it is assumed that small systems may adopt GAC20 (with 240 days of empty-
bed contact time) in a replacement mode over GAC 10. Some small systems may find that another
defined BAT, like nanofiltration, will be cheaper than the GAC20 in a replacement mode because their
specific geographic locations may make routine GAC shipment expensive.
Final Economic Analysis for the Stage 2 DBPR 8-16 December 2005
-------
Exhibit 8.5 SWAT Model Predictions of Percent of Large Plants in Compliance
with TTHM and HAAS Stage 2 MCLs after Application of Specified Treatment
Technologies
Treatment
Technology
(EC)
EC (no pre-disinfection)
EC&GAC10
EC & GAC20
EC & All Chloramines
Compliance with 80/60 LRAA
Residual Disinfectant
Chlorine
73.5%
73.4%
100%
100%
NA
Chloramine
76.9%
88.0%
97.1%
100%
83.9%
All
Systems
74.8%
78.4%
99.1%
100%
NA
Compliance with 64/48 LRAA
(20% Safety Factor)
Residual Disinfectant
Chlorine
57.2%
44.1%
100%
100%
NA
Chloramine
65.4%
62.7%
95.7%
100%
73.6%
All
Systems
60.4%
50.5%
98.6%
100%
NA
Source: Seidel (2001).
8.4.3 BATs for Consecutive Systems
EPA is also proposing a BAT for consecutive systems to meet the TTHM and HAAS MCLs of 80
and 60 (ig/L, respectively. Presumably, consecutive systems that may need to employ the BAT are
receiving water from their wholesaler(s) that barely meets or does not meet the MCLs. Removal of
TTHM and HAAS is difficult after they have formed. EPA believes that the best compliance strategy for
consecutive systems is to collaborate with wholesalers on the water quality they need. However, this is a
private agreement over which EPA does not have jurisdiction. There are expected to be wholesalers
treating water with whom consecutive systems cannot work out agreements to enable the consecutive
systems to meet the MCLs.
EPA is specifying chloramination with management of hydraulic flow and storage to minimize
residence time in the distribution system as a BAT for consecutive systems that serve at least 10,000
people. Chloramination has been used for residual disinfection for many years to minimize the formation
of chlorination DBFs, including TTHM and HAAS (USEPA 20051). EPA estimates that over 50 percent
of large subpart H systems serving at least 10,000 people use chloramination for Stage 1 DBPR. The
BAT provision to manage hydraulic flow and minimize residence time in the distribution system is
intended to facilitate the maintenance of the chloramine residual and minimize the likelihood of
nitrification. If consecutive systems receive chlorinated water that is close to, but lower than, the MCLs,
they should in most cases be able to use chloramination to stop the formation of TTHM and HAA5 in
their distribution system and thereby meet the MCL. If consecutive systems are already receiving
chloraminated water from the wholesaler that is meeting the MCLs, the consecutive system should also be
able to meet the MCL. In either of these situations, distribution system flow maintenance is important for
maintaining the chloramine residual.
For those consecutive systems serving fewer than 10,000 people, EPA believes that the best
compliance strategy for consecutive systems is to collaborate with wholesalers on the water quality they
need. For consecutive systems that are having difficulty meeting the MCLs, EPA is specifying a BAT of
chloramination with management of hydraulic flow and storage to minimize residence time in the
Final Economic Analysis for the Stage 2 DBPR
8-17
December 2005
-------
distribution system for systems serving at least 10,000 and management of hydraulic flow and storage to
minimize residence time in the distribution system for systems serving fewer than 10,000. EPA believes
that small consecutive systems can use this BAT to comply with the Stage 2 DBPR, but if they cannot,
then they can apply to the State for a variance. Chloramines are not included as a BAT for consecutive
systems serving fewer than 10,000 people due to concerns about the ability of systems to properly control
the process, given that many have no treatment capability or expertise. EPA is also concerned about such
systems having operational difficulties such as distribution system nitrification.
EPA believes that the various BATs for non-consecutive systems are not appropriate for
consecutive systems because their efficacy in controlling DBFs is based on precursor removal.
Consecutive systems face the unique challenge of receiving waters in which DBFs are already present if
the wholesale systems has used a residual disinfectant, which BATs for non-consecutive systems do not
effectively remove. GAC is not cost-effective for removing DBFs. Dioxin is a potent carcinogen and a
byproduct of GAC regeneration when GAC has been used to adsorb DBFs. Nanofiltration can be
moderately effective at removing TTHM and HAAS, but only with membranes that have a very low
molecular weight cutoff and very high cost of operation. Therefore, EPA believes that GAC and
nanofiltration are not appropriate BATs for consecutive systems.
8.5 Effect of Compliance with the Stage 2 DBPR on the Technical, Managerial, and
Financial Capacity of Public Water Systems
Section 1420(d)(3) of SDWA, as amended, requires that, in promulgating a National Primary
Drinking Water Regulation (NPDWR), the Administrator shall include an analysis of the likely effect of
compliance with the regulation on the technical, managerial, and financial (TMF) capacity of PWSs. The
following analysis fulfills this statutory obligation by identifying the incremental impact that the Stage 2
DBPR will have on the TMF of regulated water systems. Analyses presented in this document represent
only the impact of new or revised requirements, as established by the Stage 2 DBPR; the impacts of
previously established requirements on system capacity are not considered.
Overall water system capacity is defined in Guidance on Implementing the Capacity Development
Provisions of the Safe Drinking Water Act Amendments of 1996 (USEPA 1998e) as the ability to plan for,
achieve, and maintain compliance with applicable drinking water standards. Capacity encompasses three
components: technical, managerial, and financial. Technical capacity is the operational ability of a water
system to meet SDWA requirements. Key issues of technical capacity include:
• Source water adequacy—Does the system have a reliable source of water with adequate
quantity? Is the source generally of good quality and adequately protected?
Infrastructure adequacy—Can the system provide water that meets SDWA standards? What
is the condition of its infrastructure, including wells or source water intakes, treatment and
storage facilities, and distribution systems? What is the infrastructure's life expectancy?
Does the system have a capital improvement plan?
• Technical knowledge and implementation—Are the system's operators certified? Do the
operators have sufficient knowledge of applicable standards? Can the operators effectively
implement this technical knowledge? Do the operators understand the system's technical and
operational characteristics? Does the system have an effective Operations and Maintenance
(O&M) program?
Final Economic Analysis for the Stage 2 DBPR 8-18 December 2005
-------
Managerial capacity is the ability of a water system's managers to make financial, operating, and
staffing decisions that enable the system to achieve and maintain compliance with SDWA requirements.
Key issues include:
Ownership accountability—Are the owners clearly identified? Can they be held accountable
for the system?
• Staffing and organization—Are the operators and managers clearly identified? Is the system
properly organized and staffed? Do personnel understand the management aspects of
regulatory requirements and system operations? Do they have adequate expertise to manage
water system operations (i.e., to conduct implementation, IDSE, additional routine
monitoring, and operational evaluation activities to meet the Stage 2 DBPR requirements)?
Do personnel have the necessary licenses and certifications?
Effective external linkages—Does the system interact well with customers, regulators, and
other entities? Is the system aware of available external resources, such as technical and
financial assistance?
Financial capacity is a water system's ability to acquire and manage sufficient financial resources
to allow the system to achieve and maintain compliance with SDWA requirements. Key issues include:
• Revenue sufficiency—Do revenues cover costs?
• Creditworthiness—Is the system financially healthy? Does it have access to capital through
public or private sources?
Fiscal management and controls—Are adequate books and records maintained? Are
appropriate budgeting, accounting, and financial planning methods used? Does the system
manage its revenues effectively?
8.5.1 Requirements of the Stage 2 DPBR
The Stage 2 DBPR establishes five new requirements that may affect the TMF capacity of
affected PWSs:
1) Compliance with MCLs established for TTHM and HAAS: MCLs of SO^g/L and 60(ig/L for
TTHM and HAAS, respectively, measured as LRAAs at the monitoring sites identified as a
result of the IDSEs required under the Stage 2 DBPR.
2) Conducting an IDSE to identify sample locations for Stage 2 compliance monitoring that
represent distribution system sites with high TTHM and HAAS levels.
3) Preparing a monitoring plan, based on information in the IDSE and consultation with the
State/Primacy Agency, that details the sites and times for compliance sampling.
4) Additional routine monitoring for DBFs.
5) If the operational evaluation level is exceeded, systems must conduct an operational
evaluation and submit a report to the State/Primacy Agency no later than 90 days after being
notified of the analytical result that exceeded the operational evaluation level.
Final Economic Analysis for the Stage 2 DBPR 8-19 December 2005
-------
In addition, personnel from systems regulated under the Stage 2 DBPR will need to familiarize
themselves with the rule and its requirements.
8.5.2 Systems Subject to the Stage 2 DBPR
The Stage 2 DBPR will apply to all CWSs and NTNCWSs that add a primary or residual
disinfectant other than UV, or that deliver water that has been treated with a disinfectant other than UV.
The Stage 2 DBPR may affect 42,032 CWSs and 6,260 NTNCWSs—48,293 systems in all. While most
will not, some systems may require increased TMF capacity to comply with the new requirements, or will
need to tailor their choices to match their capacities.
8.5.3 Impact of the Stage 2 DBPR on System Capacity
The estimates presented in Exhibits 8.6 and 8.7 represent the anticipated impact of the Stage 2
DBPR on small and large system capacity as a result of the measures that systems are expected to adopt to
meet the requirements of the rule (e.g., selecting monitoring sites for the IDSE, installing/upgrading
treatment, operator training, communication with regulators and the service community, etc.). The extent
of the impact of a particular requirement on system capacity is estimated using a scale of 0-5, where 0
represents a requirement that is not anticipated to have any impact on system capacity, 1 represents a
requirement that is expected to have a minimal impact on system capacity, and 5 represents a requirement
that is anticipated to have a very significant impact on system capacity.
Criteria used to develop the scores and associated impacts are discussed further in section 8.5.5.
8.5.4 Rationale for Scores
The baseline assumed for the purposes of this analysis, which identifies the incremental impact of
the Stage 2 DBPR on the TMF capacity of systems, is complete implementation of the Stage 1 DBPR, the
Interim Enhanced Surface Water Treatment Rule (IESWTR), and the Long Term 1 Enhanced Surface
Water Treatment Rule (LT1ESWTR). As a result, it is anticipated that many of the systems facing the
most difficult DBP challenges will have made appropriate modifications to their treatment process (e.g.,
changed point of disinfection, installed membrane technologies, etc.) to achieve compliance with these
rules, and, therefore, will not need to install additional treatment technology to achieve compliance with
the Stage 2 DBPR. However, the revised methodology for measuring system compliance with the MCLs
for TTHM and HAA5 (i.e., LRAA) will require systems to reduce peak levels in DBP concentrations
within their distribution systems. Since an LRAA represents a more stringent testing standard than an
RAA, it is likely that some systems that previously met requirements established by the Stage 1 DBPR
will be required to make changes to their treatment processes to comply with the Stage 2 DBPR. The
derivation of the scores assigned in Exhibits 8.5 and 8.6 and the rationale behind them is described in the
next section.
Final Economic Analysis for the Stage 2 DBPR 8-20 December 2005
-------
Exhibit 8.6 Estimated Impact of the Stage 2 DBPR on Small System Capacity
(0 = no impact, 1 = minimal impact, and 5 = very significant impact)
Requirement
Familiarization with
requirements of the rule
Conducting IDSE monitoring
Plants with major treatment
technology changes
Stage 2 monitoring plan1
Additional routine monitoring1
Operational evaluations
Number of
Systems/
Plants
(Percent)
44,453 systems
(100%)
10,478 systems
(24%)
1,743 plants
(3%)
22,499 systems
(48%)
97 systems
(0.2%)
Technical Capacity
Source Water
Adequacy
0
0
2
0
0
0
Infrastructure
Adequacy
0
0
4
0
0
0
Technical
Knowledge &
Implementation
1
2
4
1
1
1
Managerial Capacity
Ownership
Accountability
0
1
2
1
0
1
Staffing &
Organization
1
0
2
0
0
1
Effective External
Linkages
0
2
3
1
0
1
Financial Capacity
Revenue
Sufficiency
0
3
5
1
3
0
±± !r=
0
0
5
0
0
0
o3
D)
ID 4=
0 Ł=
(/) o
LL O
0
1
3
0
0
0
1Some systems are expected to take more samples and some are expected to take less from Stage 1 to Stage 2, depending on the number of plants in their
systems. Overall, the Stage 2 DBPR results in an increase in the total number of compliance samples taken from the Stage 1 DBPR. Exhibit H.8a presents the
change in total samples for different system size categories.
Notes: (1) To analyze the impact of these requirements on system capacity, the requirements believed to have the most and the least impact on affected
systems (i.e., the installation of treatment to ensure compliance with the LRAA MCLs for TTHM and HAAS, and familiarization with the requirements of the
rule, respectively), were analyzed first. These initial analyses were then used as the bases against which the relative impact of the remaining
requirements were assessed. The impact estimates developed for each requirement were also compared to those developed for the Long Term 2
Enhanced Surface Water Treatment Rule (LT2ESWTR) to ensure cross-rule consistency and to enable cross-rule comparisons.
(2) The scores presented above represent the worst-case scenario; the requirements of this rule are expected to have less impact on the capacity of most
systems affected by each requirement.
Source: Number and percent of small systems subject to familiarization of rule, IDSE monitoring, and operational evaluations from Exhibit 7.2a Columns B, D, and
J (sum of surface water/GWUDI, and ground water CWSs and NTNCWSs serving fewer than 10,000 people), respectively. Number and percent of small
systems subject to Stage 2 monitoring plan is the sum of Exhibit 7.2a Column F and Exhibit 7.2b Column B. Number of small plants making treatment
technology changes from Exhibit 7.3 Column B (sum of surface water/GWUDI, and ground water CWSs and NTNCWSs serving fewer than 10,000
people). Impact on capacity is determined relative to previous regulations as a function of the cost and number of systems/plants that require additional
capacity to comply with each requirement, as described in section 8.5.5.
Final Economic Analysis for the Stage 2 DBPR
December 2005
-------
Exhibit 8.7 Estimated Impact of the Stage 2 DBPR on Large System Capacity
(0 = no impact, 1 = minimal impact, and 5 = very significant impact)
Requirement
Familiarization with
requirements of the rule
Conducting IDSE monitoring
Plants with major treatment
technology changes
Stage 2 monitoring plan1
Additional routine monitoring1
Operational evaluations
Number of
Systems/
Plants
(Percent)
3,839 systems
(100%)
2,302 systems
(60%)
518 plants
(5%)
3,853 systems
(100%)
327 systems
(9%)
Technical Capacity
Source Water
Adequacy
0
0
2
0
0
0
Infrastructure
Adequacy
0
0
3
0
0
0
Technical
Knowledge &
Implementation
1
2
3
1
0
0
Managerial Capacity
Ownership
Accountability
0
1
2
1
0
1
Staffing &
Organization
1
0
2
0
0
0
^ Ł= D)
III
LU LU _l
0
1
2
1
0
1
Financial Capacity
Revenue
Sufficiency
0
1
4
1
1
0
Credit
Worthiness
0
0
4
0
0
0
o3
D)
o "c
in o
il O
0
0
3
0
0
0
1Some systems are expected to take more samples and some are expected to take less from Stage 1 to Stage 2, depending on the number of plants in their
systems. Overall, the Stage 2 DBPR results in an increase in the total number of compliance samples taken from the Stage 1 DBPR. Exhibit H.8a presents the
change in total samples for different system size categories.
Notes: (1) To analyze the impact of these requirements on system capacity, the requirements believed to have the most and the least impact on affected
systems (i.e., the installation of treatment to ensure compliance with the LRAA MCLs for TTHM and HAAS, and familiarization with the requirements of the
rule, respectively), were analyzed first. These initial analyses were then used as the bases against which the relative impact of the remaining
requirements were assessed. The impact estimates developed for each requirement were also compared to those developed for the LT2ESWTR to
ensure cross-rule consistency and to enable cross-rule comparisons.
(2) The scores presented above represent the worst case scenario; the requirements of this rule are expected to have less impact on the capacity of most
systems affected by each requirement.
Source: Number and percent of large systems subject to familiarization of rule, IDSE monitoring, and operational evaluations from Exhibit 7.2a Columns B, D, and
J (sum of surface water/GWUDI, and ground water CWSs and NTNCWSs serving 10,000 or more people), respectively. Number and percent of large
systems subject to Stage 2 monitoring plan is the sum of Exhibit 7.2a Column F and Exhibit 7.2b Column B. Number of large plants making treatment
technology changes from Exhibit 7.3 Column B (sum of surface water/GWUDI, and ground water CWSs and NTNCWSs serving 10,000 or more people).
Impact on capacity is determined relative to previous regulations as a function of the cost and number of systems/plants that require additional capacity to
comply with each requirement, as described in section 8.5.5.
Final Economic Analysis for the Stage 2 DBPR
December 2005
-------
8.5.5 Derivation of Stage 2 DBPR Scores
EPA developed a 5-point scoring system to analyze the likely effect of compliance with an
NPDWR on the technical, managerial, and financial capacity of PWSs. For each regulation, it is
necessary to complete the following steps:
1) Determine the type and number of PWSs to which the regulation applies.
2) List all of the requirements of the regulation.
3) Determine the type and number of PWSs to which each requirement applies.
4) Evaluate the impact of each requirement on the capacity of affected PWSs.
The determination of the universe of affected systems and the evaluation of the capacity impact
of individual requirements requires the use of the cost and technical information contained in SDWIS,
EAs developed for other rules, information collection requests, and other supporting documentation for
the rule. These data sources are also used to develop a qualitative description of the expected response of
affected systems to each requirement.
The overall evaluation of the impact of a requirement on the affected systems, presented in
Exhibits 8.6 and 8.7, is determined by the impact of each requirement on nine sub-categories of
capacity—three sub-categories under each of the broader divisions of technical, managerial, and financial
capacity. Within these sub-categories, a professional engineer with extensive water system experience
reviewed the costs, number of systems affected, and complexity of each requirement. After estimating
the technical, managerial, and financial impacts within each sub-category, the professional engineer
assigned the scores using best professional judgment. Costs were considered cumulatively for each
requirement for small and large systems. This score represents the additional capacity that systems will
need to develop to comply with each requirement. Due to a lack of available information on operating
budgets, this analysis does not include a quantitative component.
To ensure the ability to make cross-rule comparisons, to standardize the assignment of numerical
scores, and to minimize the subjectivity of the scoring system, the professional engineer compared the
Stage 2 DBPR requirements to requirements of regulations for which capacity impact analyses have
already been conducted (e.g., Ground Water Rule, LT1ESWTR). Similar requirements are assigned
similar impact scores.
These group assignments are reviewed by the EPA Rule Manager and other EPA staff cognizant
of small system issues to ensure that they accurately represent the cumulative impact of the rule
requirements on system capacity. Any disagreements over the assignments are discussed. The EPA Rule
Manager and other EPA staff discuss the rationale for the disagreement and evaluate whether the
assignments need to be adjusted. EPA adjusts the assignments only after review of the rule support
documents and an analysis of the expected system response to the rule requirements.
8.5.5.1 Familiarization with the Stage 2 DBPR
The requirements established under the Stage 2 DBPR are straightforward (use of LRAA instead
of RAA to determine compliance with the MCLs for DBFs) and are grounded in requirements previously
established under the Stage 1 DBPR. As a result, EPA does not expect that small or large systems
regulated under this rule will face more than a minimal challenge to their technical or managerial capacity
as a result of efforts to familiarize themselves with the Stage 2 DBPR. Furthermore, familiarization with
the rule will not impact the financial capacity of either large or small systems.
Final Economic Analysis for the Stage 2 DBPR 8-23 December 2005
-------
8.5.5.2 Conducting an Initial Distribution System Evaluation
The IDSE was incorporated into the Stage 2 DBPRto ensure that systems monitor for DBFs at
the location where TTHM and HAAS values are highest. IDSEs are required for most PWSs under the
Stage 2 DBPR, but will not affect the capacity of all systems subject to the requirement to the same
extent. Some systems will be able to meet this requirement without conducting extensive additional
monitoring, through an IDSE waiver. CWSs serving fewer than 500 people that have TTHM and HAAS
data qualify for a very small system waiver. Systems for which every Stage 1 DBPR TTHM and HAAS
compliance sample is less than or equal to 40/30 (ig/L, respectively, and which did not have TTHM or
HAAS violations during this period may qualify for the 40/30 certification and not perform the IDSE.
NTNCWSs that serve fewer than 10,000 people are not subject to the IDSE requirement. It is expected
that large surface water systems will typically be required to conduct the most monitoring for the IDSE,
while small ground water systems will be required to conduct the least.
Before doing an IDSE, systems required to monitor for DBFs will need to select monitoring
locations. Identifying appropriate sampling locations is expected to require a modest improvement in the
technical and managerial capacity of many systems. This requirement will have a much smaller impact
on the capacity of systems that do not have to monitor than on those that do. While the former may need
to contact their regulatory agencies to obtain waivers for the IDSE requirement and to meet reporting
requirements, they will not be required to conduct as many new technical analyses of their distribution
system and its impact on finished water quality. Regardless of whether a system must conduct additional
monitoring as part of an IDSE, this requirement will have an impact on ownership accountability, since
all new or historic monitoring data must be logged and submitted to the appropriate regulatory agency.
It is expected that large surface water systems will typically be required to conduct the greatest
amount of monitoring for the IDSE, while small ground water systems will be required to conduct the
least. Lab analysis of samples for TTHM and HAAS are expensive (estimated at $200 per sample plus
$10 to $40 for shipping costs, depending on system size; see Chapter 7 for discussion of laboratory costs).
While these costs are not typically prohibitive for large systems, they may represent a significant
challenge to the financial capacity of small systems. Some of these systems will need to revisit their
current budgeting practices and fee structures to meet these additional costs.
8.5.5.3 Compliance with MCLs for TTHM and HAAS
The impact of the revised DBF MCLs on the managerial capacity of systems is not anticipated to
be as great as the technical and financial challenges. However, system managers will need to review the
implications of the revised method for measuring compliance with the MCLs for TTHM and HAAS, and
may need to hire more highly certified operators or provide additional training for existing operators to
ensure that system staff can safely and effectively operate all new elements of the system's treatment train
at all times. In addition, systems will need to rely on, and improve upon, their communication with
regulators, technical and financial assistance providers, and their service community.
Systems whose finished water does meet the MCLs for TTHM or HAAS calculated using LRAAs
will need to adjust, change, or enhance their treatment practices. The installation, operation, and
maintenance of new treatment technologies will require a substantial enhancement of these systems'
technical capacity, particularly for small systems. Specifically, source water adequacy may be reduced if
marginal sources are no longer viable. The system may also need to improve its infrastructure, and
system operators will require correspondingly greater technical expertise to operate new treatment
processes.
Final Economic Analysis for the Stage 2 DBPR 8-24 December 2005
-------
Note, however, that based on the recommendations of the Small Surface Water System Delphi
Group convened by the Agency, EPA assumed for the purposes of the capacity impact analysis that small
surface water systems would not install treatment technologies that were beyond their technical or
managerial capacity in order to meet the requirements of this rule. For example, very small systems
(those serving fewer than 500 people) would not install chlorine dioxide treatment because it requires a
system operator to be in the plant every day. Instead, EPA assumed these smallest systems would install
UV or membrane technologies, which do not require a system operator to be present each day. While the
operation and maintenance of membrane treatment elements may challenge the technical capacity of
small system operators, UV treatment is easy to implement, does not require the same level of technical
know-how, and is relatively low cost. As a result, only some of those small systems that must install
treatment to meet the MCLs for TTHM and HAAS are expected to experience the full impact detailed in
Exhibit 8.6.
While some small systems that must install new treatment to meet the revised MCL requirement
will face a substantial challenge to their capacity, it is expected that this requirement will not have as
dramatic an impact on large systems for the reasons described below.
Management for both large and small systems will need to work closely with regulatory agencies
to receive approval on proposed design/treatment modifications. Since larger systems tend to have more
developed relationships with regulatory personnel, as well as more established means of communicating
with their customers, the impact on the managerial capacity of small systems is expected to be greater
than the impact on the managerial capacity of large systems. Further, large system operators tend to have
a higher level of expertise than their small system counterparts. As a result, large system operators will
not require as much training to adequately operate and maintain any new treatment that must be installed.
This requirement will not pose as great of a technical or managerial challenge to large systems.
The impact of the Stage 2 DBPR on the financial capacity of regulated systems is closely tied to
the rule's impact on the technical capacity of these systems. Systems that must install additional
treatment processes or upgrade their current treatment processes will face high costs. These costs may
pose particular difficulties for many of the affected systems since the majority are relatively small (i.e.,
serving fewer than 3,300 customers), and therefore typically have a smaller revenue base and fewer
households over which they can distribute additional costs. In addition, large systems may take better
advantage of economies of scale than smaller systems because they buy larger quantities of chemicals and
equipment. However, it is anticipated that some systems may elect to develop an alternative source (e.g.,
one with lower levels of naturally-occurring organic material) or interconnect with a nearby system if
treatment costs prove prohibitive.
To obtain funding from either public or private sources, systems will need to demonstrate sound
financial accounting and budgeting practices, and the ability to repay their debts. As a result, many of the
smallest systems that do not currently charge explicitly for water service (e.g., mobile home parks, camp
grounds, etc.) may need to begin billing their customers. Those systems that already charge for water
service will likely need to increase their rates (frequently requiring approval of the local public utilities
commission or public services commission, board approval, or vote within the service boundary), and
improve their recordkeeping procedures. Again, this poses less of a challenge to large systems that have
established billing practices and have developed close relationships with public utilities commissions.
Therefore, on the basis of the TMF challenges posed by this requirement, it is anticipated that the
implementation of the revised monitoring methodology will have a substantial impact on the capacity of
the 1,743 small plants and 518 large plants that are expected to make treatment technology changes to
reduce DBP concentrations to comply with this rule (see Exhibit 7.3). The other systems are expected to
experience only minor TMF impacts.
Final Economic Analysis for the Stage 2 DBPR 8-25 December 2005
-------
8.5.5.4 Stage 2 Monitoring Plan
Most systems are required to prepare a Stage 2 DBPR monitoring plan that includes monitoring
locations, monitoring dates, and compliance calculation procedures. Most systems will have to base the
new monitoring plan on the IDSE results and Stage 1 compliance monitoring locations. Some systems
will only need to update existing monitoring plans. The Stage 2 DBPR monitoring plan will be similar to
monitoring plans already in existence. Therefore, the requirement is expected to have a minimal impact.
8.5.5.5 Additional Routine Monitoring
It is anticipated that the additional routine monitoring required for some systems will have a
relatively limited impact on TMF capacity. These systems already have experience sampling for DBFs
and only a small number of additional samples may be required. Some systems may take fewer or the
same number of samples under Stage 2 than Stage 1. Nonetheless, it is important to consider that the
monitoring costs may strain the financial capacity of some small systems, especially since the sampling
costs are high for TTHM and HAAS. Additional routine monitoring is expected to have minimal impact
on the capacity of large systems since costs will be spread across a large population base.
8.5.5.6 Operational Evaluations
The operational evaluation level is exceeded when a sample result (when multiplied by 2 and
added to the previous two quarters and then divided by 4) results in a concentration greater than 80 i-ig/L
for TTHM or 60 |^g/L for HAAS. If a system has exceeded the operational evaluation level, it must
conduct a operational evaluation and submit a report to the State/Primacy Agency no later than 90 days
after being notified of the analytical result that exceeded the operational evaluation level. The evaluation
involves an examination of system treatment and distribution operational practices, including storage tank
operations, excess storage capacity, distribution system flushing, changes in sources or source water
quality, and treatment technology changes or problems that may contribute to TTHM and HAAS
formation and what steps could be considered to minimize future exceedances. Based on the conclusions
of their evaluation, some systems may install additional treatment or modify their distribution system
(e.g., improve tank mixing by adding a recirculation system). However, since such modifications are not
required, they are not considered for the purposes of this analysis.
An operational evaluation will require system operators to have a thorough knowledge of the
components of their treatment train and distribution system. Further, some system operators, particularly
small system operators, may require additional training or input from State/Primacy Agency or local
extension agents. This requirement may also have a small impact on the managerial capacity of some
systems, since owners will be required to ensure that these evaluations are conducted. Note, however,
that this requirement does not require systems to notify the public or provide an explanation in their
CCRs—minimizing the need for them to further develop their public outreach efforts. Finally, given that
an operational evaluation is expected to take only 2 to 16 hours (depending on system size), this
requirement will probably not have any impact on the financial capacity of large or small systems. Large
systems with more complex distribution networks are expected to spend more time per exceedance than
smaller, simpler systems. The frequency of exceedances is also expected to decrease over time as systems
begin identifying the causes and working with their States/Primacy Agencies to reduce future
exceedances.
Final Economic Analysis for the Stage 2 DBPR 8-26 December 2005
-------
8.5.6 Summary
The Stage 2 DBPR may have a substantial impact on the capacity of the 1,743 plants in small
systems and 518 plants in large systems that must make changes to their treatment process to meet the
Stage 2 DBPR requirements. However, while the impact to these systems is potentially significant, only
3.8 percent of all plants regulated under the Stage 2 DBPR (2,261 of 60,220) will be affected by this
requirement. Since individual systems may employ more than one plant, it is likely that fewer than 1,620
systems (3.4 percent of systems) will be affected by this requirement. The new IDSE and monitoring
requirements are expected to have a small impact on the technical and managerial capacity of small
systems, a moderate impact on the financial capacity of some small systems, and a much smaller impact
on large systems. The capacity of systems that must conduct an operational evaluation will only be
impacted in a minor way, while those systems that must only familiarize themselves with the rule (the
large majority of systems) will not face any capacity impact as a result of the Stage 2 DBPR.
8.6 Paperwork Reduction Act
The information collected as a result of the Stage 2 DBPR allows the States/Primacy Agencies
and EPA to determine appropriate requirements for specific systems and to evaluate compliance with the
rule. The Paperwork Reduction Act requires EPA to estimate the burden of complying with the rule on
PWSs, States, and territories. Burden means the total time, effort, and financial resources required to
generate, maintain, retain, disclose, or provide information to or for a Federal agency. This burden
includes the time needed to:
• Review instructions
Develop, acquire, install, and utilize treatment technology and systems for the purposes of
collecting, validating, and verifying information
• Process and maintain information, and disclose and provide information
• Adjust the existing ways to comply with any previously applicable instructions and
requirements
Train personnel to be able to respond to a collection of information
Search data sources
• Complete and review the collection of information
• Transmit or otherwise disclose the information
For the first 3 years following promulgation of the final rule, the major information requirements
involve developing and submitting a monitoring plan, conducting the IDSE Standard Monitoring Program
or the System Specific Study, and submitting the IDSE report. The information collection requirements
are mandatory under Part 141, the NPDWRs, and the information collected is not confidential. This
information will allow the systems to determine appropriate treatment requirements and will allow
States/Primacy Agencies and EPA to evaluate systems' compliance with the rule. The calculation of
Stage 2 DBPR burden and costs can be found in Information Collection Request for the Stage 2
Disinfectants and Disinfection Byproducts Rule (USEPA 20051). Exhibit 8.8 provides a summary of the
results of the Information Collection Request calculations.
Final Economic Analysis for the Stage 2 DBPR 8-27 December 2005
-------
Exhibit 8.8 Summary of Average Annual Burden Hours and Labor Costs
NTNCWSs
CWSs
States and
Territories
Total
Average Annual
Burden (Hours)
9,506
139,802
79,221
228,529
Average
Annual Labor
Costs
($Millions)
$0.2
$3.7
$2.7
$6.6
Average
Annual O&M
Costs
($Millions)
$0.0
$9.8
$0.0
$9.8
Average Annual
Capital Costs
($Millions)
$0
$0
$0
$0.0
Average
Annual Costs
($Millions)
$0.2
$13.5
$2.7
$16.4
Note: Figures represent burden and cost for the 3-year Information Collection Request approval only. Detail may
not add due to independent rounding.
Source: Information Collection Request for the Stage 2 Disinfectants and Disinfection Byproducts Rule (USEPA
2005I).
The estimate of annual average burden for Stage 2 DBPR for States/Primacy Agencies and
systems is 228,529 hours. This estimate covers the first 3 years of the Stage 2 DBPR and includes
implementation of a portion of the IDSE (small system reports are not due until the fifth year). The
annual average aggregate cost estimate is $9.8 million for operation and maintenance as a purchase of
service for lab work, and $6.6 million is associated with labor. EPA assumes systems affected by the
Stage 2 DBPR have already purchased the basic equipment to record chlorine concentrations, disinfection
efficacy, and other parameters to comply with the Stage 1 DBPR. Therefore, there are no capital start-up
costs associated with information collection under this rule. The annual burden per response is 4.18
hours, and the frequency for response (average responses per respondent) is 7.59 annually. Respondents
include 57 States/Primacy Agencies and 21,549 PWSs. The estimated number of likely respondents is
7,202 per year (the product of burden hours per response, frequency, and respondents may not total the
annual average burden hours due to rounding).
8.7 Unfunded Mandates Reform Act Analysis
8.7.1 UMRA Requirements and their Impact on the Stage 2 DBPR
The UMRA of 1995, Public Law 104-4, consists of four Titles and numerous sections. Sections
201 through 205 of Title II, entitled "Regulatory Accountability and Reform," are relevant to the Stage 2
DBPR and are discussed in this section. Title II, Section 201 of the UMRA, requires Federal agencies to
assess the effects of their regulatory actions on State, local, and Tribal governments, and the private
sector. Under UMRA Section 202, EPA generally must prepare a written statement, including a cost-
benefit analysis, for proposed and final rules with "Federal mandates" that may result in expenditures by
State, local, and Tribal governments, in the aggregate, or by the private sector, of $100 million or more in
any 1 year. Section 203 requires the Agency to establish a small government agency plan before
establishing any regulatory requirements that may significantly or uniquely affect small governments.
Section 204 of the UMRA requires the Agency to develop an effective process to permit elected
officers of State, local, and Tribal governments to provide meaningful and timely input in the
development of regulatory proposals that contain significant Federal intergovernmental mandates.
Finally, Section 205 generally requires EPA to identify and consider a reasonable number of regulatory
alternatives and adopt the least costly, most cost-effective, or least burdensome alternative that achieves
the objectives of the rule before promulgating a rule for which a written statement is needed under
Section 202. The provisions of Section 205 do not apply when they are inconsistent with applicable law.
Moreover, Section 205 allows EPA to adopt an alternative other than the least costly, most cost-effective,
Final Economic Analysis for the Stage 2 DBPR
8-28
December 2005
-------
or least burdensome alternative if the Administrator publishes with the final rule an explanation why that
alternative was not adopted.
EPA has determined that this rule may contain a Federal mandate that results in expenditures of
$100 million or more for State, Local and Tribal governments, in the aggregate, or for the private sector
in any one year. While the annualized costs fall below the $100 million threshold (as shown in Exhibit
8.9), the costs in some future years may be above the $100 million mark as PWSs make capital
investments and finance these through bonds, loans, and other means. EPA's year by year cost tables do
not reflect that financed investments spread out these costs over many years. In addition, the cost analysis
does not consider that some systems may be eligible for financial assistance (e.g., low-interest loans and
grants) through such programs as EPA's DWSRF.
The Stage 2 DBPRis promulgated pursuant to Section 1412 (b)(l)(A) of the SDWA, as amended
in 1996, which directs EPA to promulgate a NPDWR for a contaminant if EPA determines that the
contaminant may have an adverse effect on the health of persons, occurs in PWSs with a frequency and at
levels of public health concern, and regulation presents a meaningful opportunity for health risk
reduction.
Pursuant to the requirements of Sections 202 and 205 of UMRA, EPA prepared a written
statement addressing the following items:
• The authorizing legislation
• Benefit-cost analysis, including an analysis of the extent to which the costs of State, local and
Tribal governments will be paid for by the Federal government
Estimates of future compliance costs and disproportionate budgetary effects
• Macroeconomic effects
A summary of EPA's consultation with State, local, and Tribal governments and their
concerns, including a summary of the Agency's evaluation of those comments and concerns
• Identification and consideration of regulatory alternatives and the selection of the least costly,
most cost-effective, or least burdensome alternative that achieves the objectives of the rule
The legislative authority for the Stage 2 DBPR is discussed in Chapter 2. The remaining items are
discussed below, but are also addressed in other chapters of this EA, such as Chapters 4 and 7.
Final Economic Analysis for the Stage 2 DBPR 8-29 December 2005
-------
Exhibit 8.9 Public and Private Costs for the Stage 2 DBPR
(Annualized at 3 and 7 Percent, $Millions)
Surface Water Systems Costs
Ground Water Systems Costs
State Costs
Tribal Costs
Total Public
Surface Water Systems Costs
Ground Water Systems Costs
Total Private
GRAND TOTAL
3% Discount
Rate
$ 41.4
$ 20.3
$ 1.7
$ 0.4
$ 63.8
$ 6.4
$ 8.5
$ 15.0
$ 78.8
7% Discount
Rate
$ 41.2
$ 19.2
$ 1.7
$ 0.4
$ 62.5
$ 6.3
$ 8.0
$ 14.3
$ 76.8
Percent of 3%
Grand Total
Costs
53%
26%
2%
1%
81%
8%
11%
19%
100%
Percent of 7%
Grand Total
Costs
54%
25%
2%
0%
81%
8%
10%
19%
100%
Note: Detail may not add due to independent rounding.
Source: Derived from Exhibit 7.5 (costs) and Exhibit 3.2 (public/private breakout).
8.7.2 Social Benefits and Costs
The social benefits are those that accrue primarily to the public through increased protection from
cancer and reproductive and developmental effects. To assign a monetary value to the reductions in fatal
and non-fatal cancer cases, EPA estimated the current and future annual cases of bladder cancer from all
causes, the number of cases attributed to DBF occurrence and exposure, and the reduction in future cases
corresponding to anticipated reductions in DBF occurrence and exposure due to the Stage 2 DBPR.
Mortalities from cancer were valued using a value of statistical life estimate consistent with EPA's policy.
EPA also used two alternate (but equally valid) estimates of willingness-to-pay to avoid non-fatal bladder
cancer (one based on avoiding a case of curable lymphoma and the other based on avoiding a case of
chronic bronchitis).
Chapter 6 presents the benefits analysis, which includes both qualitative and monetized benefits
of the rule. Although EPA estimated the number of avoided incidence of fetal loss, it is only presented as
an illustrative example. The potential nonquantifiable benefits may include reproductive health effects,
developmental health effects, reduction in other cancers, and benefits from reduction of other DBFs, co-
occurring contaminants, or emerging contaminants. In addition, certain non-health-related benefits may
exist, such as perceptions of drinking water quality, ecological, and other unknown effects.
Measuring the social costs of the rule requires identifying affected entities by ownership (public
or private), considering regulatory alternatives, calculating regulatory compliance costs, and estimating
any disproportionate impacts. Chapter 7 of this document details the cost analysis performed for the
Stage 2 DBPR. The likely compliance scenario is expected to result in total annualized costs of
approximately 78.8 million using a 3-percent discount rate (or 76.8 million using a 7-percent discount
rate). Exhibit 8.10 summarizes the annualized costs and benefits for each regulatory alternative.
Final Economic Analysis for the Stage 2 DBPR
8-30
December 2005
-------
Exhibit 8.10 Total Annualized Benefits and Costs of Regulatory Alternatives
(SMillions, 2003$)
Regulatory
Alternative
Preferred
Alternative
Alternative A11
Alternative A2
Alternative A3
Mean Annualized
Benefits,
Lymphoma (3%)
$ 1,530.79
$ 1,376.60
$ 5,167.40
$ 7,129.61
Mean Annualized
Benefits,
Lymphoma (7%)
$ 1 ,246.50
$ 1,126.39
$ 4,227.19
$ 5,832.37
Mean Annualized
Benefits,
Bronchitis (3%)
$ 762.76
$ 685.87
$ 2,574.59
$ 3,552.24
Mean Annualized
Benefits,
Bronchitis (7%)
$ 620.66
$ 560.80
$ 2,104.60
$ 2,903.78
Mean
Annualized
Costs (3%)
$ 78.80
$ 254.14
$ 421.71
$ 634.20
Mean
Annualized
Costs (7%)
$ 76.81
$ 241.81
$ 406.45
$ 613.07
Footnote 1: Alternative 1 appears to have fewer benefits (MILYs) than the Preferred Alternative because it does not incorporate the IDSE,
as explained in Chapter 4. Furthermore, this EA does not quantify the benefits of reducing the MCL for bromate (and potentially associated
cancer cases), a requirement that is included only in Alternative 1.
Source: Benefits from Appendix F. Costs from Appendix J: For the Preferred Alternative, see Exhibit J.2as for 3% and J.2aw for 7%. For
Alternative 1, see Exhibit J.3i for 3% and J.3m for 7%. For Alternative 2, see Exhibit J.4i for 3% and J.4m for 7%. For Alternative 3, see
Exhibit J.5i for 3% and J.5m for 7%.
Various Federal programs exist to provide financial assistance to State, local, and Tribal
governments in complying with this rule. The Federal government provides funding to States that have
primary enforcement responsibility for their drinking water programs through the Public Water Systems
Supervision (PWSS) Grants Program. States may use these funds to develop primacy programs or to
contract with other State agencies to assist in the development or implementation of their primacy
programs. However, they may not use these funds to contract with regulated entities (i.e., water systems).
States may use PWSS Grants to set up and administer a State program that includes such activities as
public education, testing, training, technical assistance, development and administration of a remediation
grant and loan or incentive program (excluding the actual grant or loan funds), or other regulatory or
nonregulatory measures.
Additional funding is available from other programs administered by EPA or other Federal
agencies. These include EPA's DWSRF, the U.S. Department of Agriculture's Rural Utilities' Loan and
Grant Program, and the Department of Housing and Urban Development's Community Development
Block Grant (CDBG) Program. Refer to section 8.3.3 for a more detailed discussion on funding.
8.7.3 Disproportionate Budgetary Effects
UMRA is intended to reduce the burden on State, local, and Tribal governments of Federal
mandates that are not accompanied by adequate Federal funding. Section 202 of UMRA requires an
analysis of possible disproportionate budgetary effects of certain classes of rules, in which Stage 2 DBPR
falls.7 Such an analysis is required if EPA determines that accurate estimates are reasonably feasible.
The specific concern is disproportionate budgetary effects of the Stage 2 DBPR upon certain areas or
industries:
Any particular regions of the United States
7 "...[T]he agency shall prepare a written statement containing. . . (3) estimates by the agency, if and to the
extent that the agency determines that accurate estimates are reasonably feasible, of.. . (B) any disproportionate
budgetary effects of the Federal mandate upon any particular regions of the nation or particular State, local, or Tribal
government, urban or rural or other types of communities, or particular segments of the private sector.
Final Economic Analysis for the Stage 2 DBPR 8-31 December 2005
-------
Any particular State, local, or Tribal government
Urban or rural or other types of communities
Any segment of the private sector
This EA has considered how best to interpret and comply with these requirements. The
remainder of this section describes ways to consider these requirements, whether meaningful data can be
provided, and whether accurate estimates are possible. The general conclusion for this section, however,
is that there are little basis and insufficient data to make accurate estimates of budgetary impacts that
differ among groups, governments, types of communities, or segments of the private sector.
Most of the following analyses begin with national data and then disaggregate those data, when
possible, using other measures. Because the data and estimates are national in scope, parameters tend to
be merely proportional extensions based on a characteristic, not "bottom-up" estimates of actual
differences among types of systems, communities, or economic sectors. Thus, the analyses may not
reveal true differences attributable to the impacts of the rule alternatives on various regions. Local
conditions at each regulated entity will drive the actual cost impacts of the rule (e.g., areas with high
levels of DBF precursors).
When considering disproportionate impacts, it is necessary to consider whom the Stage 2 DBPR
affects. The rule, by definition, covers some communities and a segment of the private sector. Most
CWSs and NTNCWSs that add a primary or residual disinfectant other than UV, or that deliver water
than has been treated with a disinfectant other than UV, will incur some costs. In an economic sense,
differences between communities and utilities do not disadvantage one group over the other because the
systems are not in a national market that allows for direct competition for customers. In general, those
systems are better considered local natural monopolies.
Regions
EPA determined that the Stage 2 DBPR may have disproportionate budgetary effects on certain
geographic regions. Higher TOC levels in source water present special challenges to some areas of the
country (for detailed analysis, see section 3.5.2 and Appendix B). Ground water systems in Florida, in
particular, will be heavily impacted by the Stage 2 DBPR due to the high levels of TOC in their source
water and the large number of ground water systems located in the State. Other areas with high levels of
precursors, such as TOC or bromide, will also be adversely affected. However, those systems with high
precursor and DBP levels are also the ones most likely to receive the greatest benefit from the rule.
State, Local, or Tribal Governments
There is no expectation that there will be disproportionate budgetary effects upon State, local, or
Tribal governments. Costs are expected to be proportional to the risk posed by DBFs, even if unevenly
distributed among systems and perhaps types of systems. Furthermore, there are no accurate estimates to
address the differing budgetary effects of the Stage 2 DBPR on State, local or Tribal governments.
There are few data available that bear on this issue. Exhibit 8.8 breaks out national-level costs for
PWSs, Tribal costs, and State costs, but only allocates costs to these categories rather than revealing any
disproportionate impacts on the budgets of these groups. Exhibits 8.1 la and 8.1 Ib imply that State
impacts may be larger to the extent that States contain a greater proportion of small disinfecting systems
(particularly Texas, Florida, and New York).
Final Economic Analysis for the Stage 2 DBPR 8-32 December 2005
-------
Exhibit 8.11 a Number of Small Disinfecting Systems by State
[211 American Samoa
\~9~\ Guam
|i2?| N. Mariana Islands
[Is] Palau H-u] Puerto Rico
p5o| Virgin Islands
Number of Small Disinfecting
CWSs and NTNCWSs
| 2,000 or more (4)
Q 800 to 1,999 (15)
Q 400 to 799 (14)
n 0 to 399 (24)
0 D.C.
Final Economic Analysis for the Stage 2 DBPR 8-33
December 2005
-------
Exhibit 8.11 b Percent of Small Disinfecting Systems by State
D.C.
^^ - m
\
-------
Exhibit 8.12a Total Annualized Cost of Compliance for CWSs
(3 and 7 Percent Discount Rates) ($Millions)
Source Water Category
Surface Water Systems
Ground Water Systems
Tribal Systems
Total
Total Annual Cost to
Systems Serving < 10,000
People ($ Millions)
3 Percent
$ 11.3
$ 16.1
$ 0.3
$ 27.7
7 Percent
$ 10.7
$ 15.1
$ 0.3
$ 26.1
Total Annual Cost to
Systems Serving > 10,000
People ($ Millions)
3 Percent
$ 35.6
$ 11.0
$ 0.0
$ 46.6
7 Percent
$ 36.1
$ 10.4
$ 0.0
$ 46.5
Exhibit 8.12b Annualized Cost of Compliance for NTNCWSs
(3 and 7 Percent Discount Rates) ($Millions)
Source Water Category
Surface Water Systems
Ground Water Systems
Tribal Systems
Total
Total Annual Cost to
Systems Serving < 10,000
People ($ Millions)
3 Percent
$ 0.9
$ 1.8
$ 0.1
$ 2.7
7 Percent
$ 0.8
$ 1.6
$ 0.0
$ 2.4
Total Annual Cost to
Systems Serving > 10,000
People ($ Millions)
3 Percent
$ 0.1
$ 0.0
$
$ 0.1
7 Percent
$ 0.1
$ 0.0
$
$ 0.1
Note: Detail may not add due to independent rounding (some data are rounded to zero if less than $0.05 million).
Source: Derived from Exhibit 7.5; for this exhibit, Tribal system costs are apportioned by the percent of Tribal systems
in each size category and source water type (see Exhibit 8.13).
Segments of the Private Sector
EPA performed an impact analysis for public and private systems (Exhibit 8.9 and 8.13). The
percent public and private systems shown in Exhibit 8.13 indicate that publically owned CWSs are
expected to incur greater costs than privately owned CWSs. However, costs to individual public and
private water systems will not differ substantially.
Discount rates for capital costs of private systems, as presented in Exhibit 7.9, are approximately
1 percent higher than public systems. Based on these discount rates, a private system may owe $7,000
more than a public system for a $1 million loan. Since the water industry is regulated, rates of public and
private systems are monitored and are not likely to fluctuate substantially over time. However, the rates
can be adjusted to help a system qualify for a low-cost bond. Overall, the Stage 2 DBPR is not expected
to have a disproportionate impact on public and private systems.
Final Economic Analysis for the Stage 2 DBPR
8-35
December 2005
-------
Exhibit 8.13 Percentages and Costs by Public and Private Sector
(Costs Annualized at 3 and 7 Percent)
% Public
% Private
Total Cost for
Public
Systems (3%)
Total Cost for
Private
Systems (3%)
Total Cost for
Public
Systems (7%)
Total Cost for
Private
Systems (7%)
Surface Water CWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
> 1 Million
45.7%
62.6%
77.0%
84.7%
90.7%
90.1%
85.8%
87.1%
87.3%
54.3%
37.4%
23.0%
15.3%
9.3%
9.9%
14.2%
12.9%
12.7%
Subtotal
$ 0.08
$ 0.40
$ 0.40
$ 2.58
$ 6.31
$ 7.53
$ 4.93
$ 12.45
$ 6.29
$ 40.98
$ 0.09
$ 0.24
$ 0.12
$ 0.47
$ 0.65
$ 0.83
$ 0.82
$ 1.84
$ 0.91
$ 5.98
$ 0.08
$ 0.37
$ 0.42
$ 2.44
$ 5.92
$ 7.83
$ 4.95
$ 12.60
$ 6.23
$ 40.84
$ 0.09
$ 0.22
$ 0.13
$ 0.44
$ 0.61
$ 0.86
$ 0.82
$ 1.86
$ 0.91
$ 5.94
Surface Water NTNCWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
> 1 Million
45.4%
48.5%
45.8%
48.4%
58.8%
56.2%
0.0%
100.0%
0.0%
54.6%
51.5%
54.2%
51 .6%
41 .2%
43.8%
0.0%
0.0%
0.0%
Subtotal
$ 0.05
$ 0.13
$ 0.05
$ 0.12
$ 0.08
$ 0.02
$
$ 0.03
$
$ 0.48
$ 0.06
$ 0.13
$ 0.06
$ 0.12
$ 0.06
$ 0.02
$
$
$
$ 0.45
$ 0.04
$ 0.11
$ 0.05
$ 0.10
$ 0.07
$ 0.02
$
$ 0.03
$
$ 0.44
$ 0.05
$ 0.12
$ 0.06
$ 0.11
$ 0.05
$ 0.02
$
$
$
$ 0.41
Ground Water CWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
> 1 Million
12.1%
39.0%
65.7%
78.3%
84.1%
84.4%
80.0%
85.6%
100.0%
87.9%
61.0%
34.3%
21 .7%
15.9%
15.6%
20.0%
14.4%
0.0%
Subtotal
$ 0.13
$ 1.50
$ 1.68
$ 3.23
$ 3.96
$ 5.02
$ 1.21
$ 2.69
$ 0.42
$ 19.85
$ 0.97
$ 2.34
$ 0.88
$ 0.89
$ 0.75
$ 0.93
$ 0.30
$ 0.45
$
$ 7.52
$ 0.12
$ 1.37
$ 1.55
$ 3.03
$ 3.87
$ 4.75
$ 1.16
$ 2.57
$ 0.39
$ 18.81
$ 0.90
$ 2.15
$ 0.81
$ 0.84
$ 0.73
$ 0.88
$ 0.29
$ 0.43
$
$ 7.03
Ground Water NTNCWSs
<100
100-499
500-999
1,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
> 1 Million
20.4%
48.9%
60.7%
50.8%
44.4%
84.2%
100.0%
0.0%
0.0%
79.6%
51.1%
39.3%
49.2%
55.6%
15.8%
0.0%
100.0%
0.0%
Subtotal
Grand Total
$ 0.10
$ 0.32
$ 0.25
$ 0.10
$ 0.02
$ 0.01
$ 0.00
$
$
$ 0.81
$ 62.12
$ 0.39
$ 0.34
$ 0.16
$ 0.10
$ 0.02
$ 0.00
$
$ 0.00
$
$ 1.02
$ 14.97
$ 0.09
$ 0.29
$ 0.22
$ 0.09
$ 0.01
$ 0.01
$ 0.00
$
$
$ 0.73
$ 60.81
$ 0.36
$ 0.31
$ 0.14
$ 0.09
$ 0.02
$ 0.00
$
$ 0.00
$
$ 0.92
$ 14.30
Source: Derived from Exhibit 3.2 (public/private) and Exhibit 7.5 (costs).
Final Economic Analysis for the Stage 2 DBPR
8-36
December 2005
-------
8.7.4 Macroeconomic Effects
Under UMRA Section 202, EPA is required to estimate the potential macroeconomic effects of
the regulation. These include effects on productivity, economic growth, full employment, and creation of
Gross Domestic Product (GDP) (USEPA 2000J). Macroeconomic effects tend to be measurable in
nationwide econometric models only if the economic impact of the regulation reaches 0.25 percent to 0.5
percent of GDP. In 2003, real GDP was $10,321 billion (U.S. Department of Commerce BEA 2004b);
thus, a rule would have to cost at least $26 billion annually to have a measurable effect. A regulation
with a smaller aggregate effect is unlikely to have any measurable impact, unless it is highly focused on a
particular geographic region or economic sector. The Stage 2 DBPR should not have a measurable effect
on the national economy; the total annualized costs for the rule range from $78.8 million to $76.8 million
using a 3 and 7 percent discount rate, respectively. Using these annualized figures as a measure, the
annual cost of the Stage 2 DBPR is an insignificant fraction of a $26 billion annual cost that would be
considered a measurable macroeconomic impact. Thus, annualized Stage 2 DBPR costs measured as a
percentage of the national GDP will only decline overtime as GDP grows.
8.7.5 Consultation with Small Governments
Before the Agency establishes any regulatory requirements that may significantly or uniquely
affect small governments, including Tribal governments, it must develop, under Section 203 of the
UMRA, a small government agency plan. The plan must provide notice of rule requirements to
potentially affected small governments, enabling their officials to have meaningful and timely input in the
development of EPA regulatory proposals with significant Federal intergovernmental mandates. The plan
must also inform, educate, and advise small governments on compliance with the regulatory
requirements.
EPA has determined that the Stage 2 DBPR does not contain regulatory requirements that would
significantly or uniquely affect a substantial number of small governments (see section 8.2).
Nevertheless, EPA consulted with small governments to address impacts of the rule that might uniquely
affect them. As described in section 8.2.2, a variety of stakeholders, including small governments, had
the opportunity for timely and meaningful participation in the regulatory development process through
the SBREFA process, public stakeholder meetings, and Tribal meetings.
8.7.6 Consultation with State, Local, and Tribal Governments
Section 204 of the UMRA requires the Agency to develop an effective process to permit elected
officers of State, local, and Tribal governments (or their designated authorized employees) to provide
meaningful and timely input in the development of regulatory proposals that contain significant Federal
intergovernmental mandates. Consistent with these provisions, EPA held consultations with the
governmental entities affected by this rule prior to proposal, as described in sections 8.2.2 and 8.8.
Representatives from State, local, and Tribal governments were involved in the development of
the Agreement in Principle, which was created early in the regulatory process. EPA provided the
Association of State Drinking Water Administrators (ASDWA) with an opportunity to comment before
officially proposing the Stage 2 DBPR. EPA accepted comments from ASDWA and other FACA
members, such as the National League of Cities (NLC), on a draft of the Stage 2 DBPR posted on their
Web site.
In addition to these efforts, EPA will educate, inform, and advise small systems, including those
run by small governments, about the Stage 2 DBPR requirements. The Agency is developing plain-
English guidance that will explain what actions a small entity must take to comply with the rule. Also,
Final Economic Analysis for the Stage 2 DBPR 8-37 December 2005
-------
the Agency has developed fact sheets that concisely describe various aspects and requirements of the
Stage 2 DBPR. Additional details on Tribal involvement in the rulemaking process can be found in
section 8.8.
8.7.7 Regulatory Alternatives Considered
As required under Section 205 of UMRA, EPA considered several regulatory alternatives and
numerous approaches to ensure safe levels of DBFs throughout a system's entire distribution system.
Chapter 4 provides a detailed discussion of these alternatives. EPA chose the Preferred Regulatory
Alternative because it provided substantial benefits at an acceptable level of costs. In addition, the FACA
Committee recommended the Preferred Regulatory Alternative in the Stage 2 M-DBP Agreement in
Principle.
8.7.8 Impacts on Small Governments
In developing this rule, EPA consulted with small governments pursuant to Section 203 of
UMRA to address impacts of regulatory requirements in the rule that might significantly or uniquely
affect small governments. In preparation for the Stage 2 DBPR, EPA conducted an analysis on small
government impacts and included small government officials or their designated representatives in the
rulemaking process. A variety of stakeholders, including small governments, had the opportunity for
timely and meaningful participation in the regulatory development process through the SBREFA process,
public stakeholder meetings, and Tribal meetings. Representatives of small governments took part in the
SBREFA process for this rulemaking and attended public stakeholder meetings. Through participation
and exchange in the SBREFA process and various meetings, EPA notified some potentially affected small
governments of requirements under consideration.
As previously stated, EPA has determined that this rule does not contain regulatory requirements
that would significantly or uniquely affect small governments. As described in section 8.2, EPA certified
that the Stage 2 DBPR will not have significant economic impact on a substantial number of small
entities. As shown in Exhibit 8.12, estimated annual expenditures per small system for the Stage 2 DBPR
are $27.7 for CWSs and $2.7 for NTNCWSs (at a 3 percent discount rate).
8.8 Indian Tribal Governments
Executive Order 13175, entitled "Consultation and Coordination with Indian Tribal
Governments" (65 FR 67249; November 9, 2000), requires EPA to develop "an accountable process to
ensure meaningful and timely input by Tribal officials in the development of regulatory policies that have
Tribal implications." The Executive Order defines "policies that have Tribal implications" to include
regulations that have "substantial direct effects on one or more Indian Tribes, on the relationship between
the Federal government and the Indian Tribes, or on the distribution of power and responsibilities
between the Federal government and Indian Tribes."
Under Executive Order 13175, EPA may not issue a regulation that has Tribal implications, that
imposes substantial direct compliance costs, and that is not required by statute, unless the Federal
government provides the funds necessary to pay the direct compliance costs incurred by Tribal
governments, or EPA consults with Tribal officials early in the process of developing the proposed
regulation and develops a Tribal summary impact statement. EPA has concluded that the Stage 2 DBPR
may have Tribal implications, because it may have substantial direct compliance costs on Tribal
governments, as specified in Executive Order 13175.
Final Economic Analysis for the Stage 2 DBPR 8-38 December 2005
-------
Total Tribal costs are estimated to be approximately $391,773 per year (at a 3 percent discount
rate), distributed across 755 water systems owned by Tribes. The costs for individual systems depend on
system size and source water type. Of the 755 Tribal water systems that may be affected by the Stage 2
DBPR, 654 use ground water as a source and 101 systems use surface water or GWUDI. Since the
majority of Tribal water systems are ground water systems serving fewer than 500 people, only 15.6
percent of all Tribal water systems will likely have to conduct an IDSE. As a result, the Stage 2 DBPR is
most likely to have an impact on Tribes using surface water or GWUDI serving more than 500 people.
Accordingly, EPA provides the following Tribal summary impact statement, as required by Section
5 (b)of Executive Order 13175. The results of the analysis conducted for the Tribal summary impact
statement are presented in Exhibit 8.14
EPA consulted with Tribal officials early in the development of the Stage 2 DBPR to permit them
to have meaningful and timely input. Tribes were able to have long-term input in the rule by participating
in the Federal Advisory Committee. During the Las Vegas EPA/Inter-Tribal Council of Arizona in
February 1999, a number of Tribal representatives requested that the All Indian Pueblo Council (AIPC)
representative be the FACA representative for Federal Tribes, given his knowledge of drinking water
systems. Approximately 20 Tribes are associated with the AIPC.
In addition to obtaining FACA Tribal input, EPA presented the Stage 2 DBPR at three
conferences: the 16th Annual Consumer Conference of the National Indian Health Board, the National
Tribal Environmental Council's Annual Conference in April 2000, and the EPA/Inter-Tribal Council of
Arizona, Inc. Tribal consultation meeting. Over 900 attendees representing Tribes from across the
country attended the National Indian Health Board's Consumer Conference, and representatives from
over 100 Tribes attended the annual conference of the National Tribal Environmental Council. Finally,
representatives from 15 Tribes participated at the EPA/Inter-Tribal Council of Arizona meeting. At the
first two conferences, an EPA representative conducted two workshops on their drinking water program
and upcoming regulations, including the Stage 2 DBPR. The presentation materials and meeting
summary were sent to over 500 Tribes and Tribal organizations.
EPA distributed fact sheets describing the requirements of the Stage 2 DBPR and requested
Tribal input at an annual EPA Tribal meeting in San Francisco and a Native American Water Works
Association meeting in Scottsdale, Arizona. EPA also worked through its Regional Indian Coordinators
and the National Tribal Operations Committee to mail fact sheets on the Stage 2 DBPR to all of the
Federally recognized Tribes in November 2000.
After reviewing the fact sheets, a few Tribes requested more information and expressed concern
about having to implement too many regulations. They were also concerned about infrastructure costs
and the lack of funding attached to the rule. In response to a Tribal representative's comments, EPA
explained the health protection benefits associated with the Stage 2 DBPR, which some members of the
Tribal Caucus also noted. EPA directed Tribes to the Agreement in Principle on the EPA Web site for
more information.
On January 24, 2002, EPA held a teleconference for Tribal representatives as another step in
Tribal consultation. Prior to the teleconference, EPA sent invitations to all Federally-recognized Tribes,
along with fact sheets explaining the rule. Twelve Tribal representatives and four regional Tribal
Program Coordinators attended the teleconference, requested further explanation of the rule, and
expressed concerns about funding sources. Tribes also called EPA after the teleconference to provide
additional feedback.
In the spirit of Executive Order 13175, and consistent with EPA's policy to promote
communications between EPA and Tribal governments, EPA specifically solicited comment on the
propose rule from Tribal officials. As required by section 7(a) of Executive Order 13175, when EPA sent
the draft of the final rule to Office of Management and Budget (OMB) for review pursuant to Executive
Final Economic Analysis for the Stage 2 DBPR 8-39 December 2005
-------
Order 12866, EPA included a certification from its tribal consultation official stating that EPA had met
the Executive Order's requirements in a meaningful and timely manner.
Final Economic Analysis for the Stage 2 DBPR 8-40 December 2005
-------
Exhibit 8.14 Annual Cost of Compliance for Tribal Systems by System Type and Size (Annualized at 3 Percent)
System Size/Type
Number of
Tribal
Systems
Affected by
the Stage 2
DBPR
A
Systems Conducting
Implementation
Activities
Percent
B
Number
C = B*A
Systems Conducting
IDSE Monitoring
Percent
D
Number
E = D*A
Systems Conducting
Additional Routine
Monitoring
Percent
F
Number
G = F*A
Systems Conducting
Significant Excursion
Activities
Percent
H
Number
I = H*A
Systems Conducting
Stage 2 Monitoring
Plans
Percent
J
Number
K = J*A
Mean
Annualized
Cost per
System
L
Estimated
Total Tribal
Costs
M=A*L
Primarily Surface Water CWSs
D100
100-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
> 1 Million
National Subtotal
19
46
13
14
5
0
0
0
0
97
100%
100%
100%
100%
100%
NA
NA
NA
NA
19
46
13
14
5
NA
NA
NA
NA
97
62%
62%
94%
94%
92%
NA
NA
NA
NA
12
29
12
13
5
NA
NA
NA
NA
71
0%
0%
0%
0%
66%
NA
NA
NA
NA
0
0
0
0
3
NA
NA
NA
NA
3
0%
0%
1%
1%
3%
NA
NA
NA
NA
0
0
0
0
NA
NA
NA
NA
0
62%
62%
100%
100%
100%
NA
NA
NA
NA
12
29
13
14
5
NA
NA
NA
NA
73
$ 160
$ 292
$ 353
$ 1,177
$ 3,409
$
-
$
$
$ 3,038
$ 13,423
$ 4,595
$ 16,476
$ 17,043
$
-
$
$
$ 54,575
Primarily Disinfecting Ground Water CWSs
D100
100-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
> 1 Million
National Subtotal
149
233
78
71
21
4
0
0
0
556
100%
100%
100%
100%
100%
100%
NA
NA
NA
149
233
78
71
21
4
NA
NA
NA
556
4%
4%
18%
18%
18%
18%
NA
NA
NA
6
10
14
13
4
1
NA
NA
NA
47
0%
0%
56%
56%
56%
57%
NA
NA
NA
0
0
44
40
12
2
NA
NA
NA
97
0%
0%
0%
0%
0%
0%
NA
NA
NA
0
0
0
0
0
0
NA
NA
NA
0
4%
4%
100%
100%
100%
100%
NA
NA
NA
6
10
78
71
21
4
NA
NA
NA
0
$ 139
$ 391
$ 639
$ 850
$ 2,139
$ 4,871
$
$
$
$ 20,712
$ 91 ,055
$ 49,813
$ 60,380
$ 44,916
$ 19,483
$
$
$
$ 286,358
Final Economic Analysis for the Stage 2 DBPR
December 2005
-------
Exhibit 8.14 Annual Cost of Compliance for Tribal Systems by System Type and Size (Annualized at 3 Percent)
(Continued)
System Size/Type
Number of
Tribal
Systems
Affected by
the Stage 2
DBPR
A
Systems Conducting
Implementation
Activities
Percent
B
Number
C= B*A
Systems Conducting
IDSE Monitoring
Percent
D
Number
E = D*A
Systems Conducting
Additional Routine
Monitoring
Percent
F
Number
G = F*A
Systems Conducting
Significant Excursion
Activities
Percent
H
Number
I = H*A
Systems Conducting
Stage 2 Monitoring
Plans
Percent
J
Number
K = J*A
Mean
Annualized
Cost per
System
L
Estimated
Total Tribal
Costs
M=A*L
Primarily Surface Water NTNCWSs
D100
100-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
> 1 Million
National Subtotal
2
1
0
1
0
0
0
0
0
4
100%
100%
NA
100%
NA
NA
NA
NA
NA
2
1
NA
1
NA
NA
NA
NA
NA
4
0%
0%
NA
0%
NA
NA
NA
NA
NA
0
0
NA
0
NA
NA
NA
NA
NA
0
0%
0%
NA
0%
NA
NA
NA
NA
NA
0
0
NA
0
NA
NA
NA
NA
NA
0
0%
0%
NA
0%
NA
NA
NA
NA
NA
0
0
NA
0
NA
NA
NA
NA
NA
0
0%
0%
NA
0%
NA
NA
NA
NA
NA
0
0
NA
0
NA
NA
NA
NA
NA
0
$ 448
$ 818
$
$ 2,590
$
$
$
$
$
$ 897
$ 818
$
$ 2,590
$
$
$
$
$
$ 4,304
Primarily Disinfecting Ground Water NTNCWSs
D100
100-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
> 1 Million
National Subtotal
TOTALS
31
28
13
24
2
0
0
0
0
98
755
100%
100%
100%
100%
100%
NA
NA
NA
NA
31
28
13
24
2
NA
NA
NA
NA
98
755
0%
0%
0%
0%
0%
NA
NA
NA
NA
0
0
0
0
0
NA
NA
NA
NA
0
118
0%
0%
100%
100%
100%
NA
NA
NA
NA
0
0
13
24
2
NA
NA
NA
NA
39
140
0%
0%
0%
0%
0%
NA
NA
NA
NA
0
0
0
0
0
NA
NA
NA
NA
0
0
0%
0%
0%
0%
0%
NA
NA
NA
NA
0
0
0
0
0
NA
NA
NA
NA
0
73
$ 199
$ 312
$ 692
$ 808
$ 1,619
$
$
-
-
$ 6,161
$ 8,746
$ 8,994
$ 19,397
$ 3,238
$
$
-
$
$ 46,536
$ 391,773
Sources: (A) Number of Indian Lands from SDWIS 4th Quarter FY2003 data.
(B, D, F, and H) Derived from Exhibit H.12.
(J) Mean costs are total annualized costs (at 3 percent) (Exhibits J.2ba, J.2be, J.2bi, J.2bm) divided by the number of primarily ground or primarily
surface water CWSs or NTNCWSs in the size category (Exhibit 3.2).
Final Economic Analysis for the Stage 2 DBPR
December 2005
-------
8.9 Impacts on Sensitive Subpopulations
EPA's Office of Water has historically considered risks to sensitive subpopulations (including
fetuses, infants, and children) when establishing drinking water assessments, advisories and other
guidance, and standards (USEPA 1989, USEPA 1991a). The disinfection of public drinking water
supplies to prevent waterborne disease is the most successful public health program in U.S. history
(USEPA 1991a). However, numerous DBFs that result from chemical disinfection may have potential
health risks. Thus, maximizing health protection for sensitive subpopulations requires balancing risks to
achieve the recognized benefits of controlling waterborne pathogens while minimizing risk of potential
DBP toxicity. Experience shows that waterborne disease from pathogens in drinking water is a major
concern for children and other subgroups (e.g., the elderly, immunocompromised, and pregnant women)
because of their greater vulnerabilities (Gerba et al. 1996). EPA believes that, based on animal studies,
DBFs may also potentially pose risks to fetuses and pregnant women (USEPA 1998f). In addition,
because the elderly population (age 65 and above) is naturally at a higher risk of developing bladder
cancer, their health risks may further increase as a result of long-term DBP exposure (National Cancer
Institute 2002).
In developing this rule, risks to sensitive subpopulations, including children, were taken into
account in the assessments of disinfectants and DBFs (see sections 6.2.1 and 6.3). For each of the DBFs
included in the Stage 2 DBPR, the MCLGs are derived using the most sensitive endpoint among all
available data and an intraspecies uncertainty factor of 10, which accounts for human variability,
including sensitive subpopulations like children. The Agency has evaluated alternative regulatory options
and selected the one that balances cost with significant benefits, including those for sensitive
subpopulations. The Stage 2 DBPR will result in a reduction in cancer risk and a potential reduction in
reproductive and developmental risk to fetuses and pregnant women. It should be noted that the
LT2ESWTR, which accompanies this rule, reduces pathogens in drinking water and further protects
sensitive subpopulations.
SDWA identifies pregnant women as a sensitive subpopulation. Epidemiological and
toxicological research suggests a potential association between exposure to DBFs and adverse
reproductive and developmental health effects such as spontaneous abortion, stillbirth, neural tube
defects, cardiovascular effects, and low birth weight. Stage 2 DBPR will help to protect pregnant women
and their fetuses from adverse health effects that may be caused by exposure to elevated DBP levels. In
this respect, any benefits derived from implementation of Stage 2 DBPR provisions should have a
positive health impact on this sensitive subpopulation. Research outlining the potential health benefits of
the Stage 2 DBPR to both sensitive subpopulations and the general public is discussed in greater detail in
Chapter 6 of this EA.
8.9.1 Protecting Children from Environmental Health Risks and Safety Risks
Executive Order 13045 (62 FR 19885 April 23, 1997) applies to any rule initiated after April 21,
1998, that (1) is determined to be "economically significant" as defined under Executive Order 12866;
and (2) concerns an environmental, health, or safety risk that EPA has reason to believe may have a
disproportionate effect on children. If the regulatory action meets both criteria, EPA must evaluate the
environmental health or safety effects of the planned rule on children, and explain why the planned
regulation is preferable to other potentially effective and reasonably feasible alternatives considered by
the Agency.
The Stage 2 DBPR is not subject to the Executive Order because it is not economically significant
as defined in Executive Order 12866. EPA has consistently and explicitly considered risks to infants and
children in all assessments developed for this rulemaking and presents the environmental health and
safety effects of DBFs on children in sections 6.2.1 and 6.5.1. For each of the DBFs included in the Stage
Final Economic Analysis for the Stage 2 DBPR 8-43 December 2005
-------
2 DBPR, EPA has compiled analyses of the available data used for deriving the maximum contaminant
level goal (MCLG) to determine if these values are protective for fetuses and children.
The Agency concluded that the Stage 2 DBPR will result in greater risk reduction for children
than for the general population. The MCLGs of all DBFs in the rule help protect fetuses, infants, and
children from potential adverse developmental/reproductive effects.
8.10 Environmental Justice
Executive Order 12898 (59 FR 7629) establishes a Federal policy for incorporating
environmental justice into Federal agency missions by directing agencies to identify and address
disproportionately high and adverse human health or environmental effects of its programs, policies, and
activities on minority and low-income populations. The Agency has considered environmental-justice-
related issues concerning the potential impacts of this action and consulted with minority and low-income
stakeholders.
Two aspects of the Stage 2 DBPR comply with the order that requires the Agency to consider
environmental justice issues in the rulemaking and to consult with stakeholders representing a variety of
economic and ethnic backgrounds. These include: (1) the overall nature of the rule, and (2) the convening
of a stakeholder meeting specifically to address environmental justice issues.
The Agency has built on the efforts conducted during the Stage 1 DBPR development to comply
with Executive Order 12898. On March 12, 1998, EPA held a stakeholder meeting to address various
components of pending drinking water regulations and how they might impact sensitive subpopulations,
minority populations, and low-income populations. This meeting was a continuation of stakeholder
meetings that started in 1995 to obtain input on the Agency's Drinking Water Programs. Topics
discussed included treatment techniques, costs and benefits, data quality, health effects, and the regulatory
process. Participants were national, State, Tribal, municipal, and individual stakeholders. EPA
conducted the meeting by video conference call between 11 cities. The major objectives for the March
12, 1998, meeting included the following:
• To solicit ideas from stakeholders on known issues concerning current drinking water
regulatory efforts.
To identify key areas of concern to stakeholders.
To receive suggestions from stakeholders concerning ways to increase representation of
communities in the Office of Ground Water and Drinking Water (OGWDW) regulatory
efforts.
In addition, EPA developed a plain-English guide for this meeting to assist stakeholders in
understanding the multiple, and sometimes complex, issues surrounding drinking water regulations.
The Stage 2 DBPR and other drinking water regulations promulgated or under development are
expected to have a positive effect on human health regardless of the social or economic status of a
specific population. The Stage 2 DBPR serves to provide a similar level of drinking water protection to
all groups. Where water systems have high DBP levels, they must reduce levels to meet the MCLs.
Further, to the extent that DBP levels in drinking water might be disproportionately high now among
minority or low-income populations (which is unknown), the Stage 2 DBPR will work to remove those
differences. Thus, the Stage 2 DBPR meets the intent of Federal policy requiring incorporation of
environmental justice into Federal agency missions.
Final Economic Analysis for the Stage 2 DBPR 8-44 December 2005
-------
The Stage 2 DBPR applies uniformly to CWSs and NTNCWSs that add a disinfectant other than
UV light or that deliver water that has been chemically disinfected. Consequently, the health protection
from DBF exposure that this rule provides is equal across all income and minority groups served by
systems regulated by this rule.
8.11 Federalism
Executive Order 13132, "Federalism" (64 FR 43255 August 10, 1999), requires EPA to develop
an accountable process to ensure "meaningful and timely input by State and local officials in the
development of regulatory policies that have federalism implications." "Policies that have federalism
implications" are defined in the executive order to include regulations that have "substantial direct effects
on the States, on the relationship between the national government and the States, or on the distribution of
power and responsibilities among the various levels of government."
Under Section 6(b) of Executive Order 13132, EPA may not issue a regulation that has federalism
implications, imposes substantial direct compliance costs, and is not required by statute, unless the
Federal government provides the funds necessary to pay the direct compliance costs incurred by State and
local governments, or consults with State and local officials early in the process of developing the
proposed regulation.
EPA has concluded that the Stage 2 DPBR will not have federalism implications. It will not
impose substantial direct effects on the States, on the relationship between the national government and
the States, or on the distribution of power and responsibilities among various levels of government, as
specified in Executive Order 13132. The Stage 2 DBPR has total annualized costs ranging from $78.8
million to $76.8 million using a 3 and 7 percent discount rate, respectively. Thus, the requirements of
Sections 6(b) and 6(c) of the executive order do not apply to this rule.
Although Executive Order 13132 does not apply to this rule, EPA did consult with State and local
officials early in the process of developing the Stage 2 DPBR to permit them to have meaningful and
timely input into its development. On February 20, 2001, EPA held a dialogue on both the Stage 2 DBPR
and LT2ESWTR with representatives of State and local governmental organizations including those that
represent elected officials. Representatives from the following organizations attended the meeting:
ASDWA, the National Governors' Association (NGA), the National Conference of State Legislatures
(NCSL), the International City/County Management Association (ICMA), NLC, the County Executives
of America, and health departments. Questions ranged from a basic inquiry into how Cryptosporidium
gets into water to more detailed queries about anticipated implementation guidance, procedures, and
schedules. No concerns were expressed. Some of the State and local organizations that attended this
meeting were also participants in the Stage 2 M-DBP Federal Advisory Committee and signed the
Agreement in Principle. In addition, EPA consulted with a mayor in the SBREFA consultation. EPA
considered all input from these consultations in the development of the Stage 2 DBPR.
In the spirit of Executive Order 13132 and consistent with EPA's policy to promote
communications between EPA and State and local governments, EPA specifically solicited comment on
the proposed rule from State and local officials.
8.12 Actions Concerning Regulations That Significantly Affect Energy Supply,
Distribution, or Use
Executive Order 13211, "Actions Concerning Regulations That Significantly Affect Energy
Supply, Distribution, or Use" (66 FR 28355 May 22, 2001), provides that agencies shall prepare and
submit to the Administrator of the Office of Information and Regulatory Affairs, OMB, a Statement of
Final Economic Analysis for the Stage 2 DBPR 8-45 December 2005
-------
Energy Effects for certain actions identified as "significant energy actions." Section 4(b) of Executive
Order 13211 defines "significant energy actions" as "any action by an agency (normally published in the
Federal Register) that promulgates or is expected to lead to the promulgation of a final rule or regulation,
including notices of inquiry, advance notices of proposed rulemaking, and notices of proposed
rulemaking: (l)(i) that is a significant regulatory action under Executive Order 12866 or any successor
order, and (ii) is likely to have a significant adverse effect on the supply, distribution, or use of energy; or
(2) that is designated by the Administrator of the Office of Information and Regulatory Affairs as a
significant energy action."
The Stage 2 DBPR has not been designated by the Administrator of the Office of Information and
Regulatory Affairs as a significant energy action because it is not likely to have a significant adverse
effect on the supply, distribution, or use of energy. This determination represents the analysis presented
below.
Energy Supply
The first consideration is whether the Stage 2 DBPR would adversely affect the supply of energy.
The Stage 2 DBPR does not regulate power generation, either directly or indirectly and the public and
private utilities that the Stage 2 DBPR regulates do not, as a rule, generate power. Further, the cost
increases borne by customers of water utilities as a result of the Stage 2 DBPR are a small percentage of
the total cost of water, except for a few small systems that will need to spread the cost of installing
advanced treatment technologies over a narrow customer base. Therefore, those customers that are power
generation utilities are unlikely to face any significant effects as a result of the Stage 2 DBPR. In
summary, the Stage 2 DBPR does not regulate the supply of energy, does not generally regulate the
utilities that supply energy, and is unlikely to significantly affect the customer base of energy suppliers.
Thus, the Stage 2 DBPR would not adversely affect the supply of energy.
In response to the Stage 2 DBPR, some water utilities are expected to increase their energy use,
and those impacts are discussed later in this section.
Energy Distribution
The second consideration is whether the Stage 2 DBPR would adversely affect the distribution of
energy. The Stage 2 DBPR does not regulate any aspect of energy distribution. PWSs that are regulated
by the Stage 2 DBPR already have electrical service. As derived later in this section, the Stage 2 DBPR is
projected to increase peak electricity demand at water utilities by only 0.008 percent. Therefore, EPA
estimates that the existing connections are adequate and that the Stage 2 DBPR has no discernable
adverse effect on energy distribution.
Energy Use
The third consideration is whether the Stage 2 DBPR would adversely affect the use of energy.
Because some PWSs are expected to add treatment technologies that use electrical power, this potential
impact of the Stage 2 DBPR on the use of energy requires further evaluation. The analyses that underlay
the estimation of costs in Chapter 7 for the Stage 2 DBPR are national in scope and do not identify
specific plants or utilities that may install treatment in response to the rule. As a result, no analysis of the
effect on specific energy suppliers is possible with the available data. The approach used to estimate the
impact of energy use, therefore, focuses on national-level impacts. The analysis estimates the additional
energy use due to the Stage 2 DBPR, and compares that to the national levels of power generation in
terms of average and peak loads.
The first step in the analysis is to estimate the energy used by the treatment technologies expected
to be installed as a result of the Stage 2 DBPR. Energy use is not directly stated in Technologies and
Final Economic Analysis for the Stage 2 DBPR 8-46 December 2005
-------
Costs for Control ofMicrobial Contaminants and Disinfection By-Products (USEPA 2005n), but the
annual cost of energy for each treatment technology addition or upgrade necessitated by the Stage 2
DBPR is provided. An estimate of plant-level energy use is derived by dividing the total energy cost per
plant for a range of flows by an average national cost of electricity of $0.076 per kilowatt hour per year
(kWh/y) (U.S. DOE EIA 2004a8). The energy use per plant for each flow range and treatment technology
is then multiplied by the number of plants predicted to install each treatment technology in a given flow
range (treatment technology selection forecasts are presented in Chapter 7). The energy requirements for
each flow range are then added to produce a national total. No electricity use is subtracted to account for
the treatment technologies that may be replaced by new treatment technologies, resulting in a
conservative estimate of the increase in energy use. Exhibit 8.14 shows the estimated energy use for each
Stage 2 DBPR compliance treatment technology in kWh/y. The incremental national annual energy
usage is approximately 0.12 million megawatt-hours (MWh). Although the energy usage after
implementing the Stage 2 DBPR is expected to be greater than before implementation (advanced
treatment technologies typically require more energy than conventional treatment technologies), the net
increase in energy usage is not expected to be significant.
Exhibit 8.16 provides a sample calculation for chloramines showing the increase in energy usage
as a result of the Stage 2 DBPR.
To determine if the additional energy required for systems to comply with the rule would have a
significant adverse effect on the use of energy, EPA compared the numbers in Exhibit 8.14 to the national
production figures for electricity. According to the U.S. Department of Energy's Energy Information
Administration, electricity producers generated 3,848 million MWh of electricity in 2003 (USDOE EIA
2004b9). Therefore, even using the highest assumed energy use for the Stage 2 DBPR (i.e., 116,302,140
kWh/y, or 116,302 MWh/y), the rule when fully implemented would result in only a 0.003 percent
increase in annual average energy use. This calculation is shown below:
116,302 MWh/y - 3,848,000,000 MWh/y * 100 = 0.003%
EPA is aware that DOE has updated its 2003 "average national cost of electricity per kilowatt hour per
year" from $0.076 to $0.074. However, EPA continues to use the $0.076 value to maintain consistency with the
Technologies and Cost Document (USEPA 2005n). EPA is aware that DOE has updated its 2003 average national
cost of electricity per kilowatt hour per year from $0.076 to $0.074. However, EPA continues to use the $0.076
value to maintain consistency with the Technologies and Cost Document.
9 EPA is aware that DOE has updated its estimate of total electricity produced in 2003 from 3,848 million
to 3,883 million. However, EPA continues to use the 3,848 million estimate to maintain consistency with related
electricity estimates used in this EA and the Technologies and Cost Document.
Final Economic Analysis for the Stage 2 DBPR 8-47 December 2005
-------
Exhibit 8.15 Increase in Energy Usage as a Result of the Stage 2 DBPR
Technology
Chloramines (with and without advanced tech.)
Chlorine Dioxide
UV
Ozone (0.5 log)
GAC10 + Adv. Disinfectants
GAC20
GAC20 + Adv. Disinfectants
Membranes
TOTAL
Number of Plants
Selecting the
Technology
(a)
1,971
43
912
19
42
100
61
18
3,167
Average Energy Usage per
Plant per Year for all
Plants Selecting the
Technology
(kWh/plant/yr)
1,764
3,578
34,588
98,182
835,330
162,411
132,168
1,103,656
36,727
Total Increase in Energy
Usage as a Result of the
Stage 2 DBPR
(kWh/yr)
(b)
3,476,461
154,809
31,545,906
1,890,156
35,321 ,879
16,264,624
8,001 ,027
19,647,278
116,302,140
Notes: Detail may not add due to independent rounding.
Sources: (a) Number of plants selecting each treatment technology is derived from Exhibits 5.11b, 5.11d, 5.14b,
5.14d. Note that the number of plants selecting chloramines is the number of plants selecting chloramines
only PLUS the number selecting chloramines with advanced treatment technology (making the total in this
exhibit higher than the total number of plants making treatment technology changes in Exhibit 7.3).
(b) Energy costs derived from the Technologies and Costs Document (USEPA 2005n) for the treatment
plant design conditions listed in Exhibit 7.8. Energy costs were converted to energy usage by dividing the
costs by the unit costs for energy listed in Table 4-3 of the Technologies and Costs Document. Energy
usage is different for different size categories; the average per plant is the weighted average for all plants
selecting the treatment technology.
Final Economic Analysis for the Stage 2 DBPR
8-48
December 2005
-------
Exhibit 8.16 Sample Calculation for Determining Increase in Energy Usage:
Chloramines
System Size
(Population
Served)
Average Daily
Flow per Plant
(mgd)
A
Total No.
of
Plants
B
Chloramines (Ground Water, Ammonia Dose = 0.15 mg/l; Surface Water,
Ammonia Dose = 0.55 mg/l)
Number of
Plants
Selecting
Chloramines
c
Annual Energy
Cost per Plant
($/plant/yr)
D
Annual Energy
Requirement
(kWhr/plant/yr)
E = D/$0.076 per
kWhr
Total Energy Usage
for Plants Selecting
Chloramines
(kWhr/year)
F = C*E
Pripijirily Suj%ee -Water -6WS»
<100
100-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
> 1 Million
0.01
0.03
0.08
0.22
0.59
2.34
3.80
14.61
107.80
359
767
483
1,129
1,258
1,292
579
610
74
19
50
31
81
91
133
60
63
8
67
67
115
124
200
200
200
300
4,550
876
876
1,518
1,633
2,632
2,632
2,632
3,947
59,875
16,999
43,541
47,538
132,846
238,544
350,559
157,247
248,503
454,475
Prippr% Qrtwn 1 Million
0.01
0.03
0.08
0.21
0.63
3.33
-
22.94
-
226
312
106
91
25
5
0
1
0
12
20
7
7
2
1
0
0
0
67
67
124
124
200
200
0
300
-
876
876
1,633
1,633
2,632
2,632
0
3,947
-
10,696
17,719
11,223
10,823
4,739
1,357
0
407
-
Pri|WK% @TOU.n4:W»tw,NTNOWS»
<100
100-499
500-999
1 ,000-3,299
3,300-9,999
10,000-49,999
50,000-99,999
100,000-999,999
> 1 Million
TOTALS
0.00
0.02
0.07
0.18
0.64
2.86
9.92
17.19
-
2,493
2,129
589
247
21
3
0
0
0
60,220
51
64
18
7
1
0
0
0
0
1,971
67
67
100
124
200
200
200
200
-
876
876
1,309
1,633
2,632
2,632
2,632
2,632
-
44,771
55,752
23,066
10,837
1,517
166
15
14
-
3,476,461
Notes: Detail may not add due to independent rounding.
Sources: (A) The flows are taken from Exhibit 3.4.
(B) The baseline numbers of plants are taken from Exhibit 3.2.
(C) Numbers of plants selecting Chloramines are taken from Exhibits 5.11b, 5.11d, 5.14b, 5.14d.
(D) The electricity cost per plant is taken from the Technologies and Costs Document (USEPA 2005n).
(E) Electricity cost is $0.076/kWh, as presented in the Technologies and Costs Document (USEPA 2005n).
In addition to average energy use, the impact at times of peak power demand is important. To
examine whether increased energy usage might significantly affect the capacity margins of energy
suppliers, their peak-season generating capacity reserve was compared to an estimate of peak incremental
Final Economic Analysis for the Stage 2 DBPR
8-49
December 2005
-------
power demand by water utilities. Both energy use and water use peak in the summer months, so the most
significant effects on supply would be seen then. In the summer of 2003, U.S. generation capacity
exceeded consumption by 15 percent, or approximately 160,000 megawatts (MW) (USDOE EIA
2004b10). Assuming around-the-clock operation of water treatment plants, the total energy requirement
for the Stage 2 DBPR (Exhibit 8.15) can be divided by 8,760 hours per year to obtain an average power
demand of 13.28 MW. This is only 0.008 percent of the capacity margin available at peak use. This
calculation is presented below:
1. 116,302,140 kWh/y * (y/8,760 hr) * (MW/1,000 kW) = 13.28 MW
2. 13.28 MW - 160,000 MW * 100 = 0.008%
Assuming that power demand is proportional to water flow through the plant and that peak flow
can be as high as twice the average daily flow during the summer months, about 26.55 MW (13.28 MW x
2) could be needed to operate the treatment technologies installed to comply with the Stage 2 DBPR.
This is still only a very small fraction (0.017 percent) of the U.S. capacity margin available at peak use
(160,000 MW).
Although EPA recognizes that not all regions have a 15 percent capacity margin and that this
margin varies across regions and over time, this analysis represents the effect of the rule on national
energy supply, distribution, and use. While certain areas have experienced shortfalls in generating
capacity in the recent past, a peak incremental power requirement of 26.55 MW nationwide is not likely
to significantly change the energy supply, distribution, or use in any given area.
Conclusion
The Stage 2 DBPR is not a "significant energy action" as defined in Executive Order 13211,
"Actions Concerning Regulations That Significantly Affect Energy Supply, Distribution, or Use" (66 FR
28355 May 22, 2001) because it is not likely to have a significant adverse effect on the supply,
distribution, or use of energy (as a function of annual average use and conditions of peak power demand).
The total increase in energy usage by water systems as a result of the Stage 2 DBPR is predicted
to approximately 116 million kWh/y, which is less than three one-thousandths of 1 percent of the total
energy produced in 2003. While the rule may have some adverse energy effects, EPA does not believe
that this constitutes a significant adverse effect on the energy supply.
10 EPA is aware that DOE has updated its estimate of capacity exceeding consumption in the summer of
2003 from 160,000 to 159,000 MW. However, EPA continues to use the estimate of 160,000 MW to maintain
consistency with related electricity estimates used in this EA and the Technologies and Cost Document (USEPA
2005n).
Final Economic Analysis for the Stage 2 DBPR 8-50 December 2005
-------
9. Comparison of Benefits and Costs of the Stage 2 DBPR
9.1 Introduction
This chapter presents a summary and comparison of the benefits and costs of the Stage 2
Disinfectants and Disinfection Byproducts Rule (DBPR). Evaluation on a national level shows that the
benefits derived from the Stage 2 DBPR are likely to exceed the costs. The following sections present
summary results from the body of the economic analysis (EA), followed by a discussion of the results.
The first sections focus on analysis of the Stage 2 DBPR Preferred Regulatory Alternative, followed by a
comparison of this alternative to the other alternatives considered.
For comparison purposes, this chapter sometimes presents only mean estimates of benefits and
costs. These estimates are discussed in Chapters 6 and 7, respectively. To avoid repetition, the following
discussion assumes the reader is familiar with those chapters. The remaining sections of this chapter are
organized as follows:
9.2 Summary of National Benefits, Costs and Net Benefits of the Stage 2 DBPR Preferred
Regulatory Alternative
9.2.1 National Benefits Summary
9.2.2 National Cost Summary
9.2.3 National Net Benefits
9.3 Comparison of Regulatory Alternatives
9.3.1 Comparison of Reductions in DBP Occurrence
9.3.2 Comparison of Benefits and Costs
9.3.3 Cost-Effectiveness
9.4 Effect of Uncertainties on the Estimation of Net National Benefits
9.5 Summary of Conclusions
As described in Chapter 6 and Appendix E, benefits for the Stage 2 DBPR are estimated using
three different cessation lag models, and either TTHMs or HAASs as an indicator of chlorination DBFs.
Note that because the maximum rate of risk reduction derived from the Smoking/Lung Cancer cessation
model falls in between the maximum rate of risk reduction derived from the Smoking/Bladder Cancer and
Arsenic/Bladder Cancer cessation lag models, only benefits derived from the Smoking/Lung Cancer
cessation lag model are presented as a representative comparison to costs. TTHM is used as a
chlorination DBP indicator unless otherwise noted.
9.2 Summary of National Benefits, Costs, and Net Benefits of the Stage 2 Preferred
Regulatory Alternative
This section summarizes national benefits, costs, and net benefits of the Stage 2 DBPR Preferred
Regulatory Alternative.
The rule will be implemented over time, not instantaneously, and therefore, the treatment costs
incurred and benefits realized by the affected systems and population they serve will vary by year.
Exhibits 9. la and 9. Ib summarize the undiscounted benefit and cost estimates incurred by systems,
according to the implementation schedule (see Appendix D), over the 25 year period analyzed in this EA.
Exhibit 9. la shows benefits calculated using a willingness to pay (WTP) surrogate of curable lymphoma
for valuing avoided non-fatal bladder cancer and 9. Ib presents estimates using chronic bronchitis. As
Final Economic Analysis for the Stage 2 DBPR 9-1 December 2005
-------
explained in Chapter 6 of this EA, WTP surrogates provide an estimate of the dollar amount society is
willing to pay to avoid a case of non-fatal bladder cancer.
Exhibit 9.1a Summary of Benefit and Cost Estimates by Year for the Stage 2
Preferred Regulatory Alternative Using Lymphoma WTP ($Millions)
Year
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
Systems <1 0,000
Benefits
A
$
$
$
$
$
$ 4.6
$ 12.1
$ 21.9
$ 33.9
$ 48.0
$ 64.3
$ 77.3
$ 89.0
$ 99.5
$ 108.9
$ 117.2
$ 124.3
$ 130.6
$ 136.3
$ 141.5
$ 146.2
$ 150.5
$ 154.6
$ 156.4
$ 159.6
Systems
Cost
B
$ 0.4
$ 6.4
$ 5.0
$ 18.0
$ 47.9
$ 49.2
$ 50.8
$ 55.7
$ 60.7
$ 64.5
$ 46.8
$ 27.1
$ 27.1
$ 27.1
$ 27.1
$ 27.1
$ 27.1
$ 27.1
$ 27.1
$ 27.1
$ 27.1
$ 27.1
$ 27.1
$ 27.1
$ 27.1
Systems > 10,000
Benefits
C
$
$
$
$
$
$ 130.2
$ 336.8
$ 609.5
$ 944.2
$ 1,228.6
$ 1,492.7
$ 1,733.1
$ 1,951.4
$ 2,143.0
$ 2,306.7
$ 2,448.8
$ 2,574.1
$ 2,686.4
$ 2,788.0
$ 2,881.1
$ 2,967.0
$ 3,047.2
$ 3,122.5
$ 3,152.3
$ 3,211.4
Systems
Cost
D
$ 0.7
$ 9.4
$ 19.4
$ 6.5
$ 126.3
$ 133.1
$ 140.3
$ 146.4
$ 67.4
$ 45.1
$ 30.4
$ 30.4
$ 30.4
$ 30.4
$ 30.4
$ 30.4
$ 30.4
$ 30.4
$ 30.4
$ 30.4
$ 30.4
$ 30.4
$ 30.4
$ 30.4
$ 30.4
State Costs
E
$ 3.9
$ 3.9
$ 0.2
$ 2.1
$ 0.8
$
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
All Systems & State
Benefits
F=A+C
$
$
$
$
-
$ 134.8
$ 348.9
$ 631.3
$ 978.0
$ 1,276.7
$ 1,557.0
$ 1,810.5
$ 2,040.4
$ 2,242.5
$ 2,415.6
$ 2,566.0
$ 2,698.5
$ 2,817.0
$ 2,924.3
$ 3,022.5
$ 3,113.2
$ 3,197.7
$ 3,277.1
$ 3,308.6
$ 3,370.9
Total Cost
G=B+D+E
$ 4.9
$ 19.7
$ 24.5
$ 26.5
$ 175.0
$ 182.3
$ 192.7
$ 203.9
$ 129.9
$ 111.4
$ 78.9
$ 59.2
$ 59.2
$ 59.2
$ 59.2
$ 59.2
$ 59.2
$ 59.2
$ 59.2
$ 59.2
$ 59.2
$ 59.2
$ 59.2
$ 59.2
$ 59.2
Notes: Values are discounted and annualized in 2003$. Based on TTHM as an indicator, Villanueva et
al. (2003) for baseline risk, and smoking/lung cancer cessation lag model. Some numbers may not add
correctly due to rounding. Assumes 26 percent of cases are fatal, 74 percent are non-fatal (USEPA
1999a). EPA recognizes that benefits may be as low as zero since causality has not yet been
established between exposure to chlorinated water and bladder cancer.
Sources: Benefits: Appendix F.2a - F.2i, F.2k- F2.s
Costs: Appendix J.2a - J.2i, J.2k - J.2s, J.2v - J.2ad, J.2af - J.2an, J.2ar
Final Economic Analysis for the Stage 2 DBPR
9-2
December 2005
-------
Exhibit 9.1 b Summary of Benefit and Cost Estimates by Year for the Stage 2
Preferred Regulatory Alternative Using Bronchitis WTP ($Millions)
Year
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
Systems <1 0,000
Benefits
A
$
$
$
$
$
$ 2.3
$ 6.0
$ 10.8
$ 16.8
$ 23.8
$ 31.9
$ 38.4
$ 44.2
$ 49.4
$ 54.2
$ 58.3
$ 61.9
$ 65.1
$ 68.0
$ 70.6
$ 73.1
$ 75.3
$ 77.4
$ 78.3
$ 79.9
Cost
B
$ 0.4
$ 6.4
$ 5.0
$ 18.0
$ 47.9
$ 49.2
$ 50.8
$ 55.7
$ 60.7
$ 64.5
$ 46.8
$ 27.1
$ 27.1
$ 27.1
$ 27.1
$ 27.1
$ 27.1
$ 27.1
$ 27.1
$ 27.1
$ 27.1
$ 27.1
$ 27.1
$ 27.1
$ 27.1
Systems
Benefits
C
$
$
-
$
$
$ 64.3
$ 166.5
$ 301.5
$ 467.4
$ 608.7
$ 740.0
$ 859.9
$ 968.9
$ 1,064.9
$ 1,147.2
$ 1,218.8
$ 1,282.3
$ 1,339.3
$ 1,391.1
$ 1,438.8
$ 1,483.0
$ 1,524.4
$ 1,563.4
$ 1,577.8
$ 1,608.5
> 10,000
Cost
D
$ 0.7
$ 9.4
$ 19.4
$ 6.5
$ 126.3
$ 133.1
$ 140.3
$ 146.4
$ 67.4
$ 45.1
$ 30.4
$ 30.4
$ 30.4
$ 30.4
$ 30.4
$ 30.4
$ 30.4
$ 30.4
$ 30.4
$ 30.4
$ 30.4
$ 30.4
$ 30.4
$ 30.4
$ 30.4
State Costs
E
$ 3.9
$ 3.9
$ 0.2
$ 2.1
$ 0.8
$
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
$ 1.7
All systems
Benefits
F=A+C
$
$
-
$
$
$ 66.6
$ 172.5
$ 312.3
$ 484.2
$ 632.5
$ 771.9
$ 898.3
$ 1,013.1
$ 1,114.4
$ 1,201.3
$ 1,277.2
$ 1,344.2
$ 1,404.4
$ 1,459.1
$ 1,509.4
$ 1,556.0
$ 1,599.7
$ 1,640.8
$ 1,656.1
$ 1,688.4
Cost
G=B+D+E
$ 4.9
$ 19.7
$ 24.5
$ 26.5
$ 175.0
$ 182.3
$ 192.7
$ 203.9
$ 129.9
$ 111.4
$ 78.9
$ 59.2
$ 59.2
$ 59.2
$ 59.2
$ 59.2
$ 59.2
$ 59.2
$ 59.2
$ 59.2
$ 59.2
$ 59.2
$ 59.2
$ 59.2
$ 59.2
Notes: Values are discounted and annualized in 2003$. Based on TTHM as an indicator,
Villanueva et al. (2003) for baseline risk, and smoking/lung cancer cessation lag model. Some
numbers may not add correctly due to rounding. Assumes 26 percent of cases are fatal, 74
percent are non-fatal (USEPA 1999a). EPA recognizes that benefits may be as low as zero
since causality has not yet been established between exposure to chlorinated water and
bladder cancer.
Sources: Benefits: Appendix F.2a - F.2i, F.2k - F2.s
Costs: Appendix J.2a - J.2i, J.2k - J.2s, J.2v - J.2ad, J.2af - J.2an, J.2ar
The analyses in this EA assume that implementation of this rule will begin in 2006. If
implementation of the rule were to begin a year or two later, the fundamental conclusions of the analysis
would not be significantly changed. In the first few years, before systems have installed treatment, no
benefits are realized, although some costs are incurred for rule implementation and the Initial Distribution
System Evaluation (IDSE). Some bladder cancer cases are projected to be avoided in the year following
installation of treatment and each year thereafter, adjusted to account for the cessation lag (see Section
6.4.2 for a description of adjustments made to projected cancer cases avoided to account for cessation
lag). By 2016, all treatment is projected to be installed, and yearly system costs thereafter are constant,
representing only operations and maintenance (O&M), monitoring, and yearly operational evaluation
costs. Using the rule implementation schedule presented in Appendix D of this EA, approximately 87%
of the cases ultimately avoidable will be avoided by 25 years after promulgation, and 100% of the cases
will be avoided by 100 years after promulgation.
Final Economic Analysis for the Stage 2 DBPR
9-3
December 2005
-------
9.2.1 National Benefits Summary
The Environmental Protection Agency (EPA) has determined from its analysis of the available
animal toxicological studies and human epidemiological studies that the Stage 2 DBPR could provide
benefits resulting from reduced incidence of cancer, particularly bladder cancer, and reduced incidence of
adverse reproductive and developmental effects.
Because of limitations in the available data, it is not possible to quantify all of the health benefits
of the Stage 2 DBPR. In particular, the science is not strong enough to quantify the risk of reproductive
and developmental health effects resulting from DBP exposure. Nevertheless, although the results from
different studies are mixed, a weight of evidence evaluation of the health effects data suggests a potential
association between DBP exposure and various adverse reproductive and developmental outcomes. To
help inform the assessment of the Stage 2 DBPR benefits, EPA has prepared an illustrative calculation for
one specific reproductive effects endpoint (fetal loss). Results from this analysis show that 0 to 3,700
fetal losses could potentially be avoided annually as a result of the Stage 2 DBPR. More detail on this
analysis can be found in Appendix G. Additionally, EPA reviewed the literature on colon and rectal
cancers, which combined are the third most common site (excluding skin) of new cases and deaths in both
men and women in the U.S. Human epidemiology studies on chlorinated surface water have reported
associations with colon and rectal cancers, although there still remains some conflicting evidence. Since
Stage 1, new studies have become available that provide for an estimation of population attributable risk
(PAR), explained further in Chapter 6 of this EA. Hence, EPA chose to perform a sensitivity analysis on
benefits from avoiding colon and rectal cancers, the results of which are shown in Exhibit 6.31 and
detailed in Appendix F. An additional analysis in Appendix N presents the results of a cost effectiveness
analysis in terms of quality-adjusted life years (QALYs) for avoided cases of bladder cancer; it also
incorporates QALYs that would be saved for potentially avoided cases of colon and rectal cancers in a
sensitivity analysis (section 3.0 of Appendix N).
Other nonqualified health and non-health benefits derived from rule implementation also could
contribute to the overall value of benefits. Nonqualified benefits are discussed in detail in Chapter 6 and
are summarized below in Exhibit 9.2.
Final Economic Analysis for the Stage 2 DBPR 9-4 December 2005
-------
Exhibit 9.2 Summary of Nonquantified National Benefits of the Stage 2 DBPR
Nonquantified Benefit
Adverse Reproductive
Health Effects Avoided
Developmental Health
Effects Avoided (e.g.,
congenital anomalies)
Other Adverse Health
Effects Avoided
(Reduction in other cancers,
including colon and rectal
cancers, and benefits from
reduction of other DBFs, co-
occurring contaminants, or
emerging contaminants)
Adverse Non-Health Effects
Avoided
(Perceptions of drinking water
quality, ecological, and other
unknown effects)
Group(s)
Affected
Women (and men) of
reproductive age
Pregnant women
Fetuses
Children/adults with birth
defects
All individuals exposed to
elevated levels of DBFs in
drinking water
All individuals
Type of Cost Avoided
Medical
Care
/
/
/
/
/
Pain&
Suffering
/
/
/
/
/
Lifetime
Care
/
/
Other1
/
/
/
/
Note: Discussions of these health effects are presented in Chapter 6.
Footnote 1: Includes costs such as potential changes to life plans due to fetal loss and lost opportunities for family
life. For patients of cancer or other diseases potentially induced by DBP exposure this includes opportunity costs in
reduced work opportunities and reduced participation in social and family life. Adverse non-health effect costs
potentially include reduced enjoyment of drinking water through perception of water as undesirable in taste or odor,
and costs for bottled water or home filters.
EPA has quantified the expected range of avoided new cases of bladder cancer each year,
including both fatal and non-fatal cases. : In addition, EPA has estimated the monetized value of
avoiding these cases using estimates of willingness to pay (WTP) for non-fatal cancer2 and the value of a
statistical life (VSL) for fatal cancer cases. Exhibits 9.3 and 9.4 summarize, respectively, these
quantified and monetized benefits estimates based on total trihalomethane (TTHM) and haloacetic acid
(HAAS) reduction as an indicator3 for reduced levels of all chlorination DBFs. Data are generally
summarized using an estimate of the mean and 90 percent confidence intervals, which address some
1 Causality has not been clearly established in the association between consumption of DBFs in drinking
water and bladder cancer and therefore the number of bladder cancer cases avoided could be as low as zero.
2Because specific estimates of WTP for avoiding non-fatal bladder cancer are not available, EPA estimated
the WTP from two other non-fatal illnesses: chronic bronchitis and curable lymphoma.
3 The number of cases avoided and the resulting benefits were calculated using both TTHM and HAAS as
indicators of exposure to all chlorination DBFs. However, because results for both indicators were similar, only the
results of calculations using TTHM as an indicator are presented, to simplify presentation. Detailed results for all
analyses using both TTHM and HAAS as indicators are presented in Appendix F.
Final Economic Analysis for the Stage 2 DBPR
9-5
December 2005
-------
elements of uncertainty and variability, such as the distribution of bladder cancer cases obtained from a
Monte Carlo simulation.
Exhibit 9.3 Summary of Annual Bladder Cancer Cases Ultimately Avoidable for
the Stage 2 DBPR Preferred Regulatory Alternative
DBF Indicator
Bladder Cancer Cases
Mean
5th
95th
Annual Average Ultimately Avoidable1
TTHM
HAAS
581
680
232
261
1084
1288
Annual Average Avoidable Over 25 Years2
TTHM
HAAS
Mean 5th 95th
279
325
103
115
541
642
Notes: Estimates are discounted and annualized. The 90 percent confidence
bounds around benefits reflect uncertainty in monetization inputs relative to mean
cases. Based on TTHM as an indicator, Vlllanueva et al. (2003) for baseline risk,
and smoking/lung cancer cessation lag model. EPA recognizes that benefits may
be as low as zero since causality has not yet been established between exposure to
chlorinated water and bladder cancer.
Footnotes: 1) Benefits (avoided cases) are estimated using Vlllanueva et al. (2003)
for baseline risk. Ultimately avoidable annual cases represents the number of
cases to be avoided annually following the cessation lag, and its presentation is
consistent with OMB recommendations to extend the horizon of analysis to include
significant benefits. The ultimately avoidable number is reached approximately 100
years from the start of implementation of the Stage 2 regulation, although about
61 % of the ultimate number of case avoided occurs by year 25.
2) Average annual avoided cases is based upon the 25-year period of analysis and
so is much lower than the ultimate number of cases to be avoided following the
cessation lag period. The cessation lag is explained in detail in Appendix E of the
Stage 2 EA.
Source: Appendix E, Exhibits E.12, E.38d, E.39d
Exhibit 9.3 shows two kinds of estimates. One is the annual average of bladder cancer cases that
will be avoided following full installation of any treatment technology changes and completion of the
cessation lag period; this is the maximum, or steady state number of cases that will be avoided. The
second set of numbers is lower because it is the average number of cases avoided annually over the 25
year period of analysis, incorporating the years prior to full installation of treatment technology changes
Final Economic Analysis for the Stage 2 DBPR
9-6
December 2005
-------
resulting from the Stage 2 DBPR and prior to completion of the cessation lag period. The cessation lag is
explained further in Appendix E and Chapter 6 of this EA.
Exhibit 9.4 (below) presents monetized estimates of the value of avoided non-fatal cases of
bladder cancer using lymphoma and bronchitis WTP estimates as surrogates for the costs of illness. The
VSL is applied to estimates of fatal cases avoided, and includes factors for income growth and income
elasticity that vary by year4. For fatal and non-fatal cases, the data in Exhibit 9.4 represent the monetized
values from each year, discounted to the year 2003 (to obtain present values) and annualized over the 25
year period. These figures represent the annualized value of the estimated annual number of illnesses and
deaths avoided according to the rule schedule; they are also the annualized values of the undiscounted
benefits data presented in Exhibit 9.1.
Exhibit 9.4 Estimated Annualized National Benefits for the Stage 2 Preferred
Regulatory Alternative ($Millions)
WTP for Non-Fatal Bladder Cancer
Cases, Lymphoma as Surrogate
Mean
90% Confidence Bound
Lower
(5th %ile)
upper
(95th %ile)
WTP for Non-Fatal Bladder Cancer
Cases, Bronchitis as Surrogate
Mean
90% Confidence Bound
Lower
(5th %ile)
upper
(95th %ile)
3% Discount Rate
$ 1,531
$ 233
$ 3,536
$ 763
$ 165
$ 1,692
7% Discount Rate
$ 1,246
$ 190
$ 2,878
$ 621
$ 135
$ 1,376
Notes: Values are discounted and annualized in 2003$. The 90 percent
confidence bounds around benefits reflect uncertainty in monetization inputs
relative to mean cases. Based on TTHM as an indicator, Villanueva et al. (2003) for
baseline risk, and smoking/lung cancer cessation lag model. Some numbers may
not add correctly due to rounding. Assumes 26 percent of cases are fatal, 74
percent are non-fatal (USEPA 1999a). EPA recognizes that benefits may be as low
as zero since causality has not yet been established between exposure to
chlorinated water and bladder cancer.
Sources: Appendix F.2v, 2w, 3v, 3w
Appendix J.2as and J.2aw
4In the Stage 2 benefits analysis the income-adjusted VSL estimates are applied to the year in which cases
have been avoided. An alternative approach supported by some economists, and used in other EPA analyses, is for
the income adjustments to be applied only up to the time that exposures are reduced rather than over the cessation
lag. Because of the shorter time period over which income growth would be calculated the alternative would result
in smaller income adjustment. To use the alternative EPA would need to link the year cancers are avoided to a
specific year of exposure reduction. This cannot be done with the risk assessment and cessation lag application in
the Stage 2 analysis, where estimated cases avoided are based on a transition from one steady state to another. The
VSL income adjustment approach used in this EA will tend to overstate benefits somewhat relative to the alternative
described above. EPA recognizes this potential bias, but notes that is small in comparison to other uncertainties in
valuation, as well as uncertainties in the risk assessment and estimates of cases avoided.
Final Economic Analysis for the Stage 2 DBPR
9-7
December 2005
-------
9.2.2 National Cost Summary
The national annual costs of the Stage 2 DBPR result from activities associated with rule
implementation, IDSEs, monitoring plans, additional routine monitoring, operational evaluations, and
changes in treatment technologies by some water systems. Cost analysis methodology is thoroughly
discussed in Chapter 7 and Appendix J. These costs are summarized below in Exhibits 9.5a and 9.5b.
As with the benefits, these estimates of cost are best characterized by distributions. Uncertainty
in the predictions of plants making treatment technology changes, together with uncertainty in the
estimates of capital and O&M unit costs, contribute to the uncertainty in estimates of national costs. See
Section 7.8 for further explanation of the distribution of cost estimates.
Exhibits 9.5a and 9.5b summarize the cost information by its two main components: capital and
one-time costs, and O&M. Both cost components occur over time, hence, the present value is calculated
at 3 and 7 percent discount rates and those amounts are annualized over the 25 year period of analysis.
The estimated mean annualized cost of the Stage 2 DBPR Preferred Regulatory Alternative is $76.8
million at a 3 percent discount rate and $78.8 million at a 7 percent discount rate.
Final Economic Analysis for the Stage 2 DBPR 9-8 December 2005
-------
Exhibit 9.5a Annualized Costs for Stage 2 DBPR Preferred Regulatory Alternative Rule Activities
($Millions/Year, 3% Discount Rate)
System Costs
System Size
(Population
Served)
Capital Costs
Mean
Value
90 Percent
Confidence Bound
Lower
(5th %tile)
Upper
(95th
%tile)
O&M Costs
Mean
Value
90 Percent
Confidence Bound
Lower
(5th %tile)
Upper
(95th
%tile)
Non-Treatment Costs
(Point Estimate)
Implement-
ation
IDSE
Monitoring
Plans
Moni-
toring
Significant
Excursion
Total System Costs
Mean
Value
90 Percent
Confidence Bound
Lower
(5th %tile)
Upper
(95th
%tile)
Surface Water CWSs
< 10,000
> 10,000
$4.21
$20.60
$2.32
$11.22
$6.23
$28.75
$6.10
$14.33
$3.41
$9.03
$8.83
$21.55
$0.12
$0.09
$0.93
$1.59
$0.05
$0.03
-$0.07
-$1.14
$0.02
$0.11
$11.34
$35.61
$6.76
$20.93
$16.10
$50.97
Surface Water NTNCWSs
< 10,000
> 10,000
$0.27
$0.04
$0.15
$0.02
$0.40
$0.06
$0.57
$0.03
$0.32
$0.02
$0.82
$0.04
$0.01
$0.00
$0.00
$0.00
$0.00
$0.00
$0.02
$0.00
$0.00
$0.00
$0.86
$0.08
$0.49
$0.05
$1.25
$0.11
Ground Water CWSs
< 10,000
> 10,000
$7.41
$4.87
$6.13
$4.37
$8.70
$5.36
$7.20
$6.00
$6.60
$5.64
$7.79
$6.37
$0.30
$0.05
$0.29
$0.10
$0.08
$0.02
$1.05
$0.00
$0.00
$0.00
$16.33
$11.04
$14.45
$10.18
$18.21
$11.90
Ground Water NTNCWSs
< 10,000
> 10,000
TOTAL
$0.57
$0.01
$37.97
$0.48
$0.01
$24.69
$0.65
$0.01
$50.17
$0.75
$0.01
$34.98
$0.69
$0.01
$25.72
$0.81
$0.01
$46.22
$0.06
$0.00
$0.62
$0.00
$0.00
$2.91
$0.01
$0.00
$0.19
$0.42
$0.01
$0.28
$0.00
$0.00
$0.12
$1.80
$0.03
$77.08
$1.65
$0.02
$54.53
$1.95
$0.03
$100.51
State
Costs
$1.71
Total Costs of the Rule
90 Percent
Confidence Bound
Upper
Mean Lower (95th
Value (5th %tile) %tile)
$78.80 $56.24 $102.22
Notes: Detail may not add due to independent rounding. 90 percent confidence bounds reflect uncertainty in technology compliance forecasts and unit treatment costs.
Notes: Values are discounted and annualized in 2003$. Based on TTHM as an indicator.
Sources: Exhibit 7.5a
Capital Costs: SW CWS, Exhibit J.2bb; SW NTNCWS, Exhibit J.2bf; GW CWS, Exhibit J.2bj; GW NTNCWS, Exhibit J.2bn.
O&M Costs: SW CWS, Exhibit J.2bc; SW NTNCWS, Exhibit J.2bg; GW CWS, Exhibit J.2bk; GW NTNCWS, Exhibit J.2bo.
Non-Treatment Costs: SW CWS, Exhibit J.2bd; SW NTNCWS, Exhibit J.2bh; GW CWS, Exhibit J.2bl; GW NTNCWS, Exhibit J.2bp.
State Costs; Appendix J, Exhibit J.2as
Final Economic Analysis for the Stage 2 DBPR
9-9
December 2005
-------
Exhibit 9.5b Annualized Costs for Stage 2 DBPR Preferred Regulatory Alternative Rule Activities
($Millions/Year, 7 Percent Discount Rate)
System Costs
System Size
(Population
Served)
Capital Costs
Mean
Value
90 Percent
Confidence Bound
Lower
(5th %tile)
Upper
(95th
%tile)
O&M Costs
Mean
Value
90 Percent
Confidence Bound
Lower
(5th %tile)
Upper
(95th
%tile)
Non-Treatment Costs
(Point Estimate)
Implement-
ation
IDSE
Monitoring
Plans
Moni-
toring
Significant
Excursion
Total System Costs
Mean
Value
90 Percent
Confidence Bound
Lower
(5th %tile)
Upper
(95th
%tile)
Surface Water CWSs
< 10,000
> 10,000
$4.53
$23.00
$2.50
$12.53
$6.71
$32.10
$4.86
$11.66
$2.72
$7.35
$7.04
$17.54
$0.15
$0.11
$1.16
$2.06
$0.06
$0.04
-$0.06
-$0.90
$0.01
$0.08
$10.72
$36.06
$6.54
$21.27
$15.08
$51.03
Surface Water NTNCWSs
< 10,000
> 10,000
$0.29
$0.05
$0.16
$0.03
$0.43
$0.07
$0.45
$0.02
$0.25
$0.01
$0.66
$0.03
$0.01
$0.00
$0.00
$0.00
$0.00
$0.00
$0.01
$0.00
$0.00
$0.00
$0.76
$0.08
$0.43
$0.05
$1.11
$0.11
Ground Water CWSs
< 10,000
> 10,000
$7.98
$5.39
$6.60
$4.84
$9.37
$5.94
$5.74
$4.87
$5.26
$4.57
$6.21
$5.16
$0.38
$0.06
$0.36
$0.13
$0.09
$0.02
$0.84
$0.00
$0.00
$0.00
$15.38
$10.46
$13.53
$9.62
$17.24
$11.31
Ground Water NTNCWSs
< 10,000
> 10,000
TOTAL
$0.61
$0.01
$41.86
$0.51
$0.01
$27.16
$0.70
$0.01
$55.33
$0.60
$0.01
$28.21
$0.55
$0.01
$20.73
$0.65
$0.01
$37.29
$0.07
$0.00
$0.78
$0.00
$0.00
$3.71
$0.01
$0.00
$0.23
$0.33
$0.01
$0.23
$0.00
$0.00
$0.10
$1.62
$0.02
$75.11
$1.48
$0.02
$52.94
$1.77
$0.02
$97.67
State
Costs
$1.70
Total Costs of the Rule
90 Percent
Confidence Bound
Upper
Mean Lower (95th
Value (5th %tile) %tile)
$76.81 $54.64 $99.36
Notes: Detail may not add due to independent rounding. 90 percent confidence bounds reflect uncertainty in technology compliance forecasts and unit treatment costs.
Notes: Values are discounted and annualized in 2003$. Based on TTHM as an indicator.
Sources: Exhibit 7.5b
Capital Costs: SW CWS, Exhibit J.2br; SW NTNCWS, Exhibit J.2bv; GW CWS, Exhibit J.2bz; GW NTNCWS, Exhibit J.2cd.
O&M Costs: SW CWS, Exhibit J.2bs; SW NTNCWS, Exhibit J.2bw; GW CWS, Exhibit J.2ca; GW NTNCWS, Exhibit J.2ce.
Non-Treatment Costs: SW CWS, Exhibit J.2bt; SW NTNCWS, Exhibit J.2bx; GW CWS, Exhibit J.2cb; GW NTNCWS, Exhibit J.2cf.
State Costs: Appendix J, Exhibit J.2aw
Final Economic Analysis for the Stage 2 DBPR
9-10
December 2005
-------
9.2.3 National Net Benefits
Net benefits are the difference between the estimated value of avoided cases of bladder cancer
resulting from the Stage 2 DBPR and the estimated costs of complying with the rule. The Stage 2 DBPR
will be implemented over time and, therefore, the treatment technology changes that PWSs implement
and the benefits realized by the populations they serve will vary year to year until implementation is
complete; the benefits will be at their maximum, steady state following the cessation lag period
(explained further in Chapter 6 and Appendix E). Exhibit 9.6 takes the present value of the costs and
benefits listed by year in Exhibit 9. la-b, and annualizes them over the 25 year period of analysis at 3 and
7 percent discount rates. Exhibit 9.6 shows that, in using either the 3 or 7 percent discount rates, net
benefits for the Stage 2 DBPR Preferred Regulatory Alternative are positive, indicating that the regulation
results in a net benefit to society.
When the magnitude of uncertainty around the costs and benefits estimates differs significantly,
another approach to evaluating net benefits is a "breakeven analysis." In Exhibit 9.7, costs and benefits
are adjusted to present value (based on 2003 dollars) and annualized over 25 years at a 3 and 7 percent
discount rate to generate the average annualized values. The upper bound on costs and lower bound on
benefits are presented as a percentage of the corresponding mean estimates, since these bounds will
determine if the alternative is acceptable at the lowest estimate of its net benefits. Exhibit 9.7 shows that
the amount of uncertainty around the upper bound on the cost estimate is approximately 30 percent larger
than the estimated mean cost; the lower bound on the benefits estimate is approximately 85 percent lower
than the estimated mean benefit. Furthermore, EPA recognizes that the quantified benefits, based on
reduced cases of bladder cancer as shown in Exhibit 9.3, could be zero for all alternatives since causality
has not yet been established between exposure to chlorinated water and bladder cancer.
Using the estimate with less uncertainty, in this case, costs, a calculation is made of the minimum
level of benefits which, if achieved, would cause the rule to break even. Comparing this breakeven level
of benefits with the actual benefits estimate provides a measure of the likelihood that the Stage 2 DBPR
Preferred Regulatory Alternative will have positive net benefits. Exhibit 9.8 presents the breakeven
analysis for average annual cases of bladder cancer avoided for the 25 year period of analysis. Based on
the estimated cost of the rule ($78 million/year at a 3 percent discount rate) shown in Exhibit 9.7, the
number of bladder cancer cases that must be avoided annually to break even is a mean of 18 cases (range
of 13 to 23 cases based on a 90% confidence interval), using the WTP for non-fatal lymphoma as the
basis for valuing non-fatal cancer cases. Based on the WTP for avoiding chronic bronchitis, the rule
must avoid a mean of 99 cases to break even (range of 70 to 128 using a 90% confidence interval).
By comparison, Exhibit 9.8 shows that the estimated mean of bladder cancer cases avoided
through promulgation of the Stage 2 DBPR Preferred Regulatory Alternative is 279; the lower 5th
percentile and upper 95th percentile bounds are 103 and 541 cases, respectively. Using a 3 percent
discount rate, the mean annual number of bladder cancer cases avoided (over the 25 year analysis period)
is more than 14 times the mean break even estimate using lymphoma WTP, and more than 2.5 times the
mean break even estimate using chronic bronchitis WTP. Furthermore, the lower bound (5th percentile =
103) on the estimated mean number of cases avoided is greater than the upper bound on the breakeven
estimate using the non-fatal lymphoma WTP (95th percentile = 23), and greater than the mean estimate
using chronic bronchitis WTP (mean = 99). Results for the same calculations using a 7 percent discount
rate are also shown in Exhibit 9.8.
The break even analysis indicates that the Stage 2 DBPR is likely to have more benefits than
costs based upon implementation of the Preferred Regulatory Alternative. However, EPA recognizes that
causality has not yet been established between DBFs and bladder cancer.
Final Economic Analysis for the Stage 2 DBPR 9-11 December 2005
-------
It is also important to note that the non-quantified benefits (e.g., reduction in developmental and
reproductive risk) are not included in the primary benefits analysis but could be substantial. This means
that the number of bladder cancer cases that must be avoided to break even could potentially be less than
shown in Exhibit 9.8, if benefits are also achieved by a reduction in developmental and reproductive risks.
Exhibit 9.6 Annualized Mean Net Benefits for the Stage 2 Preferred Regulatory
Alternative ($Millions)
WTP for Non-
Fatal Bladder
Cancer Cases
Mean
Benefits
Mean
Costs
Mean Net
Benefits
3 Percent, 25 Years
Lymphoma
Bronchitis
$ 1,531
$ 763
$ 79
$ 79
$ 1,452
$ 684
7 Percent, 25 Years
Lymphoma
Bronchitis
$ 1,246
$ 621
$ 77
$ 77
$ 1,170
$ 544
Notes: Values are discounted and annualized in 2003$.
Based on TTHM as an indicator, Villanueva et al. (2003)
for baseline risk, and smoking/lung cancer cessation lag
model. Some numbers may not add correctly due to
rounding. Assumes 26 percent of cases are fatal, 74
percent are non-fatal (USEPA 1999a). EPA recognizes
that benefits may be as low as zero since causality has not
yet been established between exposure to chlorinated
water and bladder cancer.
Sources:
Costs - Appendix J, Exhibits J.2as, J.2aw
Benefits - Appendix F, Exhibits F.2v, F.2w, F.3v, F.3w
Final Economic Analysis for the Stage 2 DBPR
9-12
December 2005
-------
Exhibit 9.7 Estimated Annualized National Costs and Benefits for the Stage 2
Preferred Regulatory Alternative with Uncertainty Measured as a Percent of the
Mean ($Millions)
Benefits
Mean
90% Confidence Bound
Lower
(5th %ile)
Upper
(95th %ile)
Costs
Mean
90% Confidence Bound
Lower
(5th %ile)
Upper
(95th %ile)
Lower
Bound of
Benefits as
% of Mean
Benefits
Estimate
Upper
Bound of
Costs as %
of Mean
Cost
Estimate
3% Discount Rate
$ 1,531
$ 233
$ 3,536
$ 79
$ 56
$ 102
15%
130%
7% Discount Rate
$ 1,246
$ 190
$ 2,878
$ 77
$ 55
$ 99
15%
129%
Sources: Appendix F.2v, 2w, 3v, 3w
Appendix J.2as and J.2aw
Notes: Values are discounted and annualized in 2003$. 90 percent confidence bounds for benefits reflect
uncertainty in monetization inputs relative to mean cases; around costs reflect uncertainty in technology
compliance forecasts and unit treatment costs. Based on TTHM as an indicator, Villanueva et al. (2003) for
baseline risk, and smoking/lung cancer cessation lag model. Assumes 26 percent of cases are fatal, 74
percent are non-fatal (USEPA 1999a). EPA recognizes that benefits may be as low as zero since causality
has not yet been established between exposure to chlorinated water and bladder cancer.
Exhibit 9.8 Estimated Breakeven Points (Number of Bladder Cancer Cases
Avoided) for the Stage 2 Preferred Regulatory Alternative
Average Annual Avoided Cases
Mean
279
90 Percent
Confidence Bound
Lower
(5th %ile)
103
Upper
(95th %ile)
541
WTP for Non
Fatal
Bladder
Cancer
Cases
Lymphoma
Bronchitis
Breakeven Cases
(at 3 Percent, 25 Years)
Mean
18
99
90 Percent
Confidence Bound
Lower
(5th %ile)
13
70
Upper
(95th %ile)
23
128
Breakeven Cases
(at 7 Percent, 25 Years)
Mean
17
96
90 Percent
Confidence Bound
Lower
(5th %ile)
12
68
Upper
(95th %ile)
22
124
Notes: Breakeven cases are derived by dividing the regulation cost by the WTP estimates. Values are discounted and annualized in 2003$.
The 90 percent confidence bounds for cases avoided reflect uncertainty in PAR, in reduction in average TTHM and HAAS concentrations, and
in cessation lag. The 90 percent confidence bounds around the estimates of break even cases result from the 90 percent confidence bounds
around the cost estimates. Based on TTHM as an indicator, Villanueva et al. (2003) for baseline risk, and smoking/lung cancer cessation lag
model. Some numbers may not add correctly due to rounding. Assumes 26 percent of cases are fatal, 74 percent are non-fatal (USEPA
1999a). EPA recognizes that benefits may be as low as zero since causality has not yet been established between exposure to chlorinated
water and bladder cancer.
Sources:
Breakeven cases: Total regulation cost from Appendix J, Exhibits J.2as and J.2aw; WTP estimates from "Stage 2 Valuation Inputs Model,
Values by Year"
Cases avoided: from Appendix E, Exhibits E.24, E.38d
Final Economic Analysis for the Stage 2 DBPR
9-13
December 2005
-------
9.3 Comparison of Regulatory Alternatives
As discussed in Chapter 4, the development and evaluation of regulatory alternatives was
undertaken as part of a consultation process convened under the Federal Advisory Committees Act
(FACA). The FACA process narrowed hundreds of regulatory options down to four major alternatives
for further evaluation, one of which was designated, in an Agreement in Principle (65 FR 83015
December 2000), as the Stage 2 DBPR Preferred Regulatory Alternative. These four alternatives are
summarized below.
Preferred Alternative: 80 (ig/L TTHM and 60 (ig/L HAA5 as an LRAA; bromate MCL of 10
(ig/L as an RAA based on monthly samples taken at the finished water
point (no change from the Stage 1 DBPR for bromate). Compliance
monitoring preceded by IDSE..
Alternative 1: 80 (ig/L TTHM and 60 (ig/L HAA5 as an LRAA; bromate MCL of
5(ig/L as an RAA based on monthly samples taken at the finished water
point.
Alternative 2: 80 (ig/L TTHM and 60 (ig/L HAA5 as the single maximum value for any
sample taken during the year; bromate MCL of 10 (ig/L as an RAA
based on monthly samples taken at the finished water point (no change
from the Stage 1 DBPR for bromate).
Alternative 3: 40 (ig/L TTHM and 30 (ig/L HAA5 as an RAA of all distribution
samples taken; bromate MCL of 10 (ig/L as an RAA based on monthly
samples taken at the finished water point (no change from the Stage 1
DBPR for bromate).
Detailed benefit and cost analyses for each of the alternatives are presented in Chapters 6 and 7,
respectively, and are summarized in this chapter. The following sections present several analyses that
compare the Preferred Regulatory Alternative to the other three alternatives.
9.3.1 Comparison of Reductions in DBF Occurrence
Given that the goal of the Stage 2 DBPR is to reduce adverse health effects by reducing exposure
to DBFs in drinking water, a useful starting point for comparing regulatory alternatives is a comparison of
reduction in DBFs estimated for each alternative. Exhibit 9.9 presents percent reductions in DBFs for the
Stage 2 DBPR regulatory alternatives. These reductions are calculated as the average of annual plant
means at the point representing average residence time in the distribution system for each regulatory
alternative.
Although Exhibit 9.9 shows much greater percent reductions in DBFs for Alternatives 2 and 3,
these figures tell very little without further evaluation. Specifically, these values need to be evaluated in
the context of the costs and benefits associated with achieving them.
Final Economic Analysis for the Stage 2 DBPR 9-14 December 2005
-------
Exhibit 9.9 Comparison of DBF Reduction (of Annual Plant Mean TTHM Data)
Regulatory Alternative
Preferred
Alternative 1
Alternative 2
Alternative 3
Percent TTHM Reduction from Pre-
Stage 2 to Post-Stage 2
Mean
7.81%
7.15%
26.91%
37.12%
5th
4.53%
5.93%
23.61%
36.00%
95th
11.18%
8.36%
30.20%
38.25%
Note: The 90 percent confidence intervals around the mean estimate
represent alternative methodologies (SWAT and the ICR Matrix Method)
for surface water systems and the potential impacts of the IDSE.
Source: Exhibit 6.19
9.3.2 Comparison of Benefits and Costs
Exhibit 9.10 presents a summary of the quantified and monetized annualized benefits for each
alternative considered. The average annual number of cases avoided as shown in Exhibit 9.10 is based
upon the 25 year period of analysis; it represents a portion of the annual average number of cases
ultimately avoided, which will be achieved following completion of the cessation lag period. The
benefits derive from PWS treatment adjustments and a predicted subsequent reduction in DBF reductions,
along with a corresponding reduction in risk for bladder cancer. As stated earlier, the potential
nonqualified benefits in the form of reduced risk for developmental and reproductive health effects may
be significant, and would be in addition to the benefits quantified in Exhibit 9.10.
Exhibit 9.11 follows with a presentation of the quantified monetized costs of the regulatory
alternatives. The annualized total costs of the rule include costs for rule implementation, IDSEs,
monitoring plans, additional routine monitoring, operational evaluations, and changes in treatment
technologies by some water systems; these cost estimates include costs for PWSs and the State agencies.
The regulatory alternatives are arranged in Exhibits 9.10 and 9.11 in order of increasing cost from
left to right and top to bottom, respectively. Exhibit 9.10 shows that the Preferred Regulatory Alternative
avoids more cases of bladder cancer than the next least expensive alternative (Alternative 1). This is
because incorporating the IDSE, which increases benefits, is only considered under the Preferred
Alternative (explained further in Chapter 4). Additionally, Alternative 1 would capture more benefits if
potential cancer cases avoided by lowering the bromate standard (included only in Alternative 1) were
quantified. The Preferred Alternative avoids far fewer cases than the more expensive Alternatives 2 and
3. The Microbial-Disinfectants/Disinfection Byproducts (M-DBP) Advisory Committee did not favor
alternative Al because they were concerned that lowering the bromate level to 5 ug/L could have adverse
effects on microbial protection (see Chapter 4 for a full discussion). The committee also believed that the
current health effects data were not certain enough to warrant the drastic shifts in the Nation's drinking
water treatment practices likely to be caused by alternatives A2 and A3. Exhibit 9.11 shows that the costs
of the Preferred Regulatory Alternative are far less than costs for the three alternatives: approximately
1/3 the cost of Al; less than 1/5 the cost of A2; and less than 1/8 the cost of A3.
Final Economic Analysis for the Stage 2 DBPR
9-15
December 2005
-------
Exhibit 9.10 Comparison of Number and Annualized Value of Estimated Bladder
Cancer Cases Avoided for All Regulatory Alternatives ($Millions)
Average Annual Number of
Cases Avoided2
Annualized Mean Benefits
of Cases Avoided
(90% Confidence Bounds)
Discount Rate,
WTP for Non-
Fatal Cases
3%, Lymphoma
7%, Lymphoma
3%, Bronchitis
7%, Bronchitis
Regulatory Alternative
Preferred
279
(103-541)
$1,531
($233 - 3,536)
$1,246
($190-2,878)
$763
($165-1,692)
$621
($135-1,376)
Alternative 11
250
(127-397)
$1,377
($209-3,180)
$1,126
($172-2,600)
$686
($149-1,522)
$561
($122-1,243)
Alternative 2
939
(483-1,466)
$5,167
($786-11,936)
$4,227
($644 - 9,758)
$2,575
($558-5,712)
$2,105
($457 - 4,665)
Alternative 3
1296
(675-1,988)
$7,130
($1,085-
16,468)
$5,832
($888-13,464)
$3,552
($769-7,880)
$2,904
($630 - 6,436)
Notes: Values are discounted and annualized in 2003$. The 90 percent confidence bounds for cases avoided reflect uncertainty
in PAR, reduction in average TTHM and HAAS concentrations, and cessation lag estimates. The 90 percent confidence bounds for
benefits reflect uncertainty in monetization inputs relative to mean cases. Based on TTHM as an indicator, Villanueva et al. (2003)
for baseline risk, and smoking/lung cancer cessation lag model. Assumes 26 percent of cases are fatal, 74 percent are non-fatal
(USEPA 1999a). EPA recognizes that benefits may be as low as zero since causality has not yet been established between
exposure to chlorinated water and bladder cancer.
Footnotes: 1) Alternative 1 has lower benefits than the Preferred Alternative because it does not incorporate the IDSE.
Additionally, benefits of Alternative 1 would be higher if the value of reduced bromate was quantified. 2) The estimate of average
annual avoided cases is based upon the 25-year period of analysis and so is much lower than the ultimate number of cases to be
avoided following the cessation lag period. Estimates of ultimately avoidable cases are shown in Exhibit 9.3 of this chapter.
Sources: Appendix F.2v, 2w, 3v, 3w, 4d-e, 5d-e, 6b, 7b, 8b, 9b, 10b, 11b
Appendix E, Exhibits E.38d, 40d, 41 d, 42d
Final Economic Analysis for the Stage 2 DBPR
9-16
December 2005
-------
Exhibit 9.11 Comparison of Costs for All Regulatory Alternatives ($Millions)
Regulatory
Alternative
Preferred
Alternative 1
Alternative 2
Alternative 3
Undiscounted Costs at Full Implementation, 2003$
PWS Costs
Capital Costs
Mean
Value
$840
$2,476
$4,708
$7,248
90 Percent
Confidence
Bound
Lower
(5th
%tile)
$547
$1,572
$4,040
$6,047
Upper
(95th
%tile)
$1,109
$3,453
$5,411
$8,519
O&M Costs
Mean
Value
$57
$215
$326
$483
90 Percent
Confidence
Bound
Lower
(5th
%tile)
$42
$142
$288
$413
Upper
(95th
%tile)
$75
$290
$366
$554
Non-Treatment Costs (Point Estimate)
Implem
entatio
n
$12
$12
$12
$12
IDSE
$57
$57
$57
$57
Monit
oring
Plans
$4
$4
$4
$4
Monitor
ing
$0
$0
$0
$0
Significant
Excursion
$1,315.1
$1,315.1
$1,315.1
$1,315.1
State
Costs
$13
$13
$13
$13
Annualized Total Regulation Costs
Discounted at 3%, 25
Years
Mean
Value
$79
$254
$422
$634
90 Percent
Confidence
Bound
Lower
(5th
%tile)
$56
$166
$368
$536
Upper
(95th
%tile)
$102
$346
$478
$736
Discounted at 7%, 25
Years
Mean
Value
$77
$242
$406
$613
90 Percent
Confidence
Bound
Lower
(5th
%tile)
$55
$158
$354
$518
Upper
(95th
%tile)
$99
$330
$461
$713
Notes: Detail may not add due to independent rounding. The 90 percent confidence bounds reflect uncertainty in compliance forecasts and unit treatment
costs. Values are in 2003$.
Estimates are discounted to 2003, and shown in 2003 dollars.
Sources: Capital Costs: Appendix J, Exhibit J.1a-d
O&M Costs: Appendix J, Exhibit J.1a - d
Non-Treatment Costs: Appendix J, Exhibit J.2aq
State Costs: Appendix J, Exhibit J.1J
Total regulation costs: Appendix J, Exhibit J.2 as, aw, J.3-5(i and m)
Net Benefits
Net benefits are calculated as the difference between the monetized benefits and cost estimates.
Exhibit 9.12 presents net benefits based upon the annualized present value of quantified, monetized
benefits at 3 and 7 percent discount rates using lymphoma and bronchitis as a WTP surrogate for bladder
cancer. Accounting for the unquantified benefits that may be achieved through reduction of
developmental and reproductive health risks would raise the overall net benefits. Exhibit 9.12 shows
that, based upon the mean values for costs and benefits of each regulatory alternative, all alternatives
provide benefits greater than their costs. It should be noted that accrual of any of the non-quantified
benefits, described in Exhibit 9.2, would increase these net benefits.
Maximum Net Benefits
At both 3 and 7 percent discount rates, the Preferred Regulatory Alternative achieves more net
benefits than Alternative 1, while Alternatives 2 and 3 achieve more net benefits than the Preferred
Regulatory Alternative. However, The M-DBP Advisory Committee did not favor Alternatives 2 and 3
because it believed that the health effects data are not certain enough to warrant such a drastic shift in the
nation's drinking water treatment practices.
Incremental Net Benefits
The goal in comparing incremental net benefits is generally to identify the regulatory alternative,
in a series of increasingly stringent alternatives, having marginal net benefits that come the closest to
zero while still being positive. Protection to the level of this alternative captures the most benefits
possible among the alternatives without experiencing diminishing marginal returns. In Exhibit 9.13, each
additional regulatory alternative generally costs more per benefit dollar (monetized value of risk
reduction) than its predecessors; the exception is Alternative 1, which is dominated by the Preferred
Final Economic Analysis for the Stage 2 DBPR
9-17
December 2005
-------
Alternative in this analysis5 (costs more but produces fewer benefits) and hence is not included in the
comparison (Exhibit 9.13). Alternatives 2 and 3 again look worth pursuing because their incremental
benefits are positive. However, net benefits do not include the unqualified benefits. Because the
Preferred Alternative uses a risk targeting strategy, the effect of unquantified benefits on the Preferred
Alternative is expected to be greater in proportion to cost as compared to Alternatives 2 and 3.
Exhibit 9.12 Comparison of Mean Net Benefits for All Regulatory Alternatives
($Millions)
WTP for Non-Fatal
Bladder Cancer
Cases
Lymphoma
Bronchitis
Rule
Alternative
Preferred
A11
A2
A3
Preferred
A11
A2
A3
Mean Net Benefits
(Million$)
3%
$ 1,452
$ 1,122
$ 4,746
$ 6,495
$ 684
$ 432
$ 2,153
$ 2,918
7%
$ 1,170
$ 885
$ 3,821
$ 5,219
$ 544
$ 319
$ 1,698
$ 2,291
Notes: All values are discounted and annualized in
2003$. Based on TTHM as an indicator, Villanueva et al.
(2003) for baseline risk, and smoking/lung cancer
cessation lag model. Assumes 26 percent of cases are
fatal, 74 percent are non-fatal (USEPA 1999a). EPA
recognizes that benefits may be as low as zero since
causality has not yet been established between exposure
to chlorinated water and bladder cancer.
Footnote 1: Alternative 1 appears to have fewer benefits
than the Preferred Alternative because it does not
incorporate the IDSE, as explained in Chapter 4.
Furthermore, this EA does not quantify the benefits of
reducing the MCL for bromate (and potentially associated
cancer cases), a requirement that is included only in
Alternative 1.
Sources: Exhibits 9.10 and 9.11
Alternative 1 appears to have fewer benefits than the Preferred Alternative because it does not incorporate
the IDSE, as explained in Chapter 4. Furthermore, this EA does not quantify the benefits of reducing the MCL for
bromate (and potentially associated cancer cases), a requirement that is included only in Alternative 1.
Final Economic Analysis for the Stage 2 DBPR
9-18
December 2005
-------
Exhibit 9.13 Incremental Net Benefits for All Regulatory Alternatives ($ Millions)
WTP for Non-
Fatal Bladder
Cancer Cases
Rule
Alternative
Annual
Costs
A
Annual
Benefits
B
Incremental
Costs
C
Incremental
Benefits
D
Incremental
Net Benefits
E=D-C
3 Percent Discount Rate
Lymphoma
Bronchitis
Preferred
Alternative 11
Alternative 2
Alternative 3
Preferred
Alternative 11
Alternative 2
Alternative 3
$ 79
$ 254
$ 422
$ 634
$ 79
$ 254
$ 422
$ 634
$ 1,531
$ 1,377
$ 5,167
$ 7,130
$ 763
$ 686
$ 2,575
$ 3,552
$ 79| $ 1,531
$ 1,452
Footnote 1
$ 343
$ 212
$ 79
$ 3,637
$ 1,962
$ 763
$ 3,294
$ 1,750
$ 684
Footnote 1
$ 343
$ 212
$ 1,812
$ 978
$ 1,469
$ 765
7 Percent Discount Rate
Lymphoma
Bronchitis
Preferred
Alternative 11
Alternative 2
Alternative 3
Preferred
Alternative 11
Alternative 2
Alternative 3
$ 77
$ 242
$ 406
$ 613
$ 77
$ 242
$ 406
$ 613
$ 1,246
$ 1,126
$ 4,227
$ 5,832
$ 621
$ 561
$ 2,105
$ 2,904
$ 77| $ 1,246|$ 1,170
Footnote 1
$ 330
$ 207
$ 77
$ 2,981
$ 1,605
$ 621
$ 2,651
$ 1,399
$ 544
Footnote 1
$ 330
$ 207
$ 1,484
$ 799
$ 1,154
$ 593
Notes: Values are discounted and annualized in 2003$. Based on TTHM as an indicator, Villanueva et al. (2003)
for baseline risk, and smoking/lung cancer cessation lag model. Assumes 26 percent of cases are fatal, 74
percent are non-fatal (USEPA 1999a). EPA recognizes that benefits may be as low as zero since causality has
not yet been established between exposure to chlorinated water and bladder cancer.
Footnote 1: Alternative 1 appears to have fewer benefits than the Preferred Alternative because it does not
incorporate the IDSE, as explained in Chapter 4. Furthermore, this EA does not quantify the benefits of reducing
the MCL for bromate (and potentially associated cancer cases), a requirement that is included only in Alternative
1. This means that Alternative 1 is dominated by the Preferred Alternative in this analysis (having higher costs
than the Preferred Alternative but lower benefits), and so it is not included in the incremental comparison of
alternatives (Columns C - E). OMB states this in terms of comparing cost effectiveness ratios, but the same rule
applies to an incremental cost, benefits, or net benefits comparison: "When constructing and comparing
incremental cost-effectiveness ratios, [analysts] ... should make sure that inferior alternatives identified by the
principles of strong and weak dominance are eliminated from consideration." (OMB Circular A-4, p. 10)
Sources: Costs: Appendix J.2 (as,aw), J.3-5 (i,m)
Benefits: Appendix F.2 - 3 (v-w), F.6 -11 (c,d)
Final Economic Analysis for the Stage 2 DBPR
9-19
December 2005
-------
9.3.3 Cost-Effectiveness
Evaluation of the relative merits of one alternative over another is often made with regard to cost-
effectiveness. This concept can be defined simply as getting the greatest benefits for a given expenditure,
or imposing the least cost while achieving a given level of benefits. Although cost effectiveness does not
usually account fully for social impacts, it does provide a measure of technical efficiency and can
supplement a benefit cost analysis (BCA) or other analyses as part of a comprehensive EA. In a cost
effectiveness analysis (CEA), generally either the costs or the benefits will be monetized and the
nonmonetized factor will be quantified.
In this CEA, technical efficiency is evaluated in terms of the cost per case of bladder cancer
avoided (fatal and non-fatal) reviewed in a relative sense (by comparison among alternatives) and an
absolute sense (by comparison to thresholds). In calculating the cost per case avoided, the cases of
illnesses and fatalities are discounted to make the benefits comparable to the discounted costs. OMB
Circular A-4 provides that "there is professional consensus that future health effects, including both
benefits and costs, should be discounted at the same rate....This consensus applies to both BCA and
CEA"(p. 34). Costs and benefits are discounted based upon when they accrue in the analysis period,
using 3 and 7 percent annual discount rates for both. Although a similar analysis cannot be made for the
unqualified benefits of the rule, it is expected that the results would follow the same pattern, with more
stringent alternatives generally (with the exception of Alternative 1, See Footnote 4) reaping more
benefits at higher total costs than less stringent alternatives.
For the purpose of evaluating cost-effectiveness in an absolute sense, this CEA compares the cost
per case avoided, shown in the two left hand columns in Exhibit 9.14, to an established threshold. If an
alternative is cost effective, this cost ratio will be equal to or less than the value of the WTP estimate. In
this EA, the WTP values for avoidance of non-fatal lymphoma (mean of $4.49 million in 2003$) or
chronic bronchitis (mean of $0.80 million in 2003$) serve as surrogates for a measure of what society is
willing to pay to avoid a non-fatal case of bladder cancer. For the purpose of this comparison, the cost
ratios should be lower than these thresholds.6 The Preferred Alternative meets this criterion for both
thresholds: at discount rates of 3 and 7 percent, its cost ratios of $0.35 and $0.43 million, respectively, are
less than or equal to the lymphoma and bronchitis WTP values of $4.49 million and $0.58 million,
respectively. Alternatives 1, 2, and 3 meet this criterion with regard to the WTP for avoiding non-fatal
lymphoma, and Alternatives 2 and 3 also cost less than the WTP for chronic bronchitis. Alternative 1 is
not competitive for reasons explained previously.7
6The WTP values, shown here in 2003$, could be used to develop an annualized WTP value for the most
accurate comparison to the annualized cost per case avoided values presented in Exhibit ES. 11. First, they would be
increased over the period of analysis to reflect the elasticity of WTP in response to increases in real income. Second,
the WTP value would be weighted differentially over the period of analysis to reflect in the annualized value the
difference in the number of cases avoided, which varies on an annual basis. Third, because the cases avoided in the
CEA ratio include both fatal and non-fatal cases, the VSL would be incorporated into a weighted average (to reflect
that 26% of cases are fatal) with the WTP value after it, too, was increased over time to reflect the above 2
considerations. However, each of these factors would increase the threshold used in this analysis; therefore
annualized WTP values are not calculated because the Preferred Alternative, and most of the other alternatives, have
costs per case avoided that are already below the lowest of the thresholds ($0.80 million in 2003$).
7As mentioned previously, Alternative 1 appears to have fewer benefits than the Preferred Alternative
because it does not incorporate the IDSE, as explained in Chapter 4. Furthermore, this EA does not quantify the
benefits of reducing the MCL for bromate (and potentially associated cancer cases), a requirement that is included
only in Alternative 1.
Final Economic Analysis for the Stage 2 DBPR 9-20 December 2005
-------
A relative comparison determines the lowest cost per case avoided among the alternatives.
Exhibit 9.14 presents each regulatory alternative in increasing order of these cost ratios. The estimated
cost ratio for the Preferred Alternative is always lower than the cost ratios of the other alternatives,
indicating that the Preferred Alternative is the most cost-effective by this measure at both 3 and 7 percent
discount rates. Alternatives 2 and 3 have cost ratios closest to those of the Preferred Alternative and are
more cost effective than Alternative 1, which has the highest cost ratio based on current available
information (See Footnote 6).
An incremental CEA determines, for a series of increasingly stringent alternatives, the marginal
gain for the increase in expenditure from one alternative to the next more stringent alternative. As in the
previous paragraphs describing comparisons using the average cost per case avoided, the incremental
analysis can provide information in an absolute sense (by comparing to thresholds) or a relative sense (by
comparing among alternatives). If the incremental cost is less than the WTP values to avoid the risk
(mean of $4.49 million and $0.80 million for lymphoma and bronchitis, respectively), then the regulation
is cost effective. An incremental cost that is greater than the WTP value would indicate that an
alternative does not avoid enough additional cases beyond the less stringent alternative to justify the
additional cost. The two right hand columns in Exhibit 9.14 show that, setting aside Alternative 1 (See
Footnote 5), each alternative is cost effective in an absolute sense-relative to the WTP value for avoiding
non-fatal lymphoma. Additionally, the Preferred Alternative and Alternative 2 are also cost effective
compared to the lowest threshold (WTP to avoid chronic bronchitis), and Alternative 3 generally costs
less than the lowest threshold at a 3 percent discount rate (at a 7 percent discount rate, the ratio for
Alternative 3 slightly exceeds the threshold, with a unit cost ratio of $0.85 million). In a relative
comparison, the Preferred Alternative, capturing all of the benefits of implementing any reasonable
regulation over maintaining the status quo, captures the largest portion of benefits and has a
lower"incremental" cost than the other alternatives.8 As expected, Alternatives 2 and 3 show a pattern of
increasing incremental cost with increasing stringency.
An additional CEA is presented in terms of the quality-adjusted life years (QALYs) saved for
avoided cases of bladders cancers in Appendix N to this analysis; a sensitivity analysis in section 3.0 of
Appendix N incorporates the QALYs potentially would be saved from avoided cases of colon and rectal
cancers.
In summary, the average cost per case avoided compares favorably to both of the thresholds
(WTP surrogate values for avoiding bladder cancer) used in this EA, indicating that the alternatives are all
cost effective by this measure. In a relative comparison among the alternatives (setting aside Alternative
1, for reasons explained previously), the Preferred Alternative is the most cost effective.
The incremental gain of a first alternative (in a series of increasingly stringent alternatives) is equivalent to
the CEA ratio of that alternative and captures the large amount of benefits achieved by having a rule (compared to
the status quo). The differences between subsequent rule alternatives are quite narrow by comparison. Since
Alternative 1 is more expensive than the Preferred Alternative but this EA does not calculate additional benefits for
it, its incremental ratio would be negative, therefore it is excluded from the comparison in Exhibit 9.14. Alternative
1 appears to have fewer benefits than the Preferred Alternative because it does not incorporate the IDSE, as
explained in Chapter 4. Furthermore, this EA does not quantify the benefits of reducing the MCL for bromate (and
potentially associated cancer cases), a requirement that is included only in Alternative 1.
Final Economic Analysis for the Stage 2 DBPR 9-21 December 2005
-------
Exhibit 9.14 Incremental Cost Per Case Avoided1 for All Regulatory Alternatives
by Discount Rate ($Millions)
Rule
Alternative
Preferred
Alternative 1
Alternative 2
Alternative 3
Cost Per Case
Avoided
3%
$ 0.33
$ 1.18
$ 0.52
$ 0.57
7%
$ 0.41
$ 1.42
$ 0.63
$ 0.69
Incremental Cost
Per Case Avoided
3%
$ 0.33
7%
$ 0.41
Footnote 2
$ 0.60
$ 0.69
$ 0.73
$ 0.85
Notes: Values are discounted and annualized in 2003$. Based
on TTHM as an indicator, Villanueva et al. (2003) for baseline
risk, and smoking/lung cancer cessation lag model. Assumes 26
percent of cases are fatal, 74 percent are non-fatal (USEPA
1999a). EPA recognizes that benefits may be as low as zero
since causality has not yet been established between exposure
to chlorinated water and bladder cancer.
Footnotes: 1) The cost effectiveness ratios are a potentially a
high estimate in that the regulatory costs in the numerator are not
adjusted by subtracting the medical costs associated with cases
avoided to produce a net cost numerator. Subtraction of theses
costs would not be expected to alter the ranking of alternatives.
In the case where thresholds of maximum public expenditure per
case avoided are prescribed, defining the numerator more
precisely by making such adjustments would be appropriate.
2) In reference to conducting incremental CEA, OMB states that
the analyst should make sure that "When constructing and
comparing incremental cost-effectiveness ratios, [analysts] ...
should make sure that inferior alternatives identified by the
principles of strong and weak dominance are eliminated from
consideration." (OMB Circular A-4, p. 10) Alternative 1 is
dominated by the Preferred Alternative and is therefore not
included in the incremental analysis. The reason for this
domination is mainly that the Preferred Alternative includes IDSE
and Alternative 1 does not; and to a lesser degree because the
bromate control included in Alternative 1 increases the costs but
the benefits of this control are not quantified at this time.
Alternative 2 is compared directly to the Preferred Alternative
(skipping Alternative 1) in this analysis.
Sources:
Discounted cases avoided: Appendix E, Exhibits E.38a; E.40-
E.42,a
Discounted costs: Appendix J, Exhibits J.2as,aw; J.3-5, (i,m)
Final Economic Analysis for the Stage 2 DBPR
9-22
December 2005
-------
Benefit Cost Ratios
Benefit cost ratios can be used to rule out an alternative that does not have at least a ratio of
benefits to cost equal to or greater than 1. Exhibit 9.15 shows that each regulatory alternative of the Stage
2 DBPR has a benefit cost ratio greater than 1. However, as in the case of the CEA ratio, the information
provided by the benefit cost ratios does not allow for comparison across the regulations unless either the
benefits or costs are identical for all alternatives, which they are not in this case. A thorough analysis
must consider the net benefits, and the absolute costs and benefits in addition to the benefit cost ratio.
Exhibit 9.15 Benefit Cost Ratios for All Regulatory Alternatives
Rule Alternative
Preferred
Alternative 1 1
Alternative 2
Alternative 3
Benefit Cost Ratio
3 Percent
Discount Rate
19.4
5.4
12.3
11.2
7 Percent
Discount Rate
16.2
4.7
10.4
9.5
Notes: Based on TTHM as an indicator, Villanueva et al. (2003)
for baseline risk, smoking/lung cancer cessation lag model, and
lymphoma as WTP estimate. Assumes 26 percent of cases are
fatal, 74 percent are non-fatal (USEPA 1999a). EPA recognizes
that benefits may be as low as zero since causality has not yet
been established between exposure to chlorinated water and
bladder cancer.
Footnote 1: Alternative 1 appears to have fewer benefits than
the Preferred Alternative because it does not incorporate the
IDSE, as explained in Chapter 4. Furthermore, this EA does not
quantify the benefits of reducing the MCL for bromate (and
potentially associated cancer cases), a requirement that is
included only in Alternative 1.
Source: Exhibit 9.13
9.4 Effect of Uncertainties on the Estimation of Net National Benefits
Detailed discussions of the assumptions and uncertainties associated with national benefits and
costs are contained in Chapters 3, 5, 6 and 7. A summary of the key uncertainties and the effects of
uncertainty in those assumptions on the benefits and cost analyses are presented in Exhibit 9.16. See
Sections 3.8, 5.7, 6.6, and 7.8 for a full listing of assumptions for which there is uncertainty.
EPA is aware that there is uncertainty in the prediction of the net benefits. Where possible, this
uncertainty was incorporated into the cost and benefits models and is incorporated in the range of costs
and benefits shown in Exhibits 9.10 and 9.11. In some cases enough information was not available to
predict the magnitude of uncertainty in either costs or benefits. Uncertainties that are not quantified are
also listed in Exhibit 9.16.
EPA believes that uncertainty in the compliance forecast has a potentially large influence on cost
and benefits estimates in this EA. Thus, the Agency has attempted to quantify the uncertainty by giving
Final Economic Analysis for the Stage 2 DBPR
9-23
December 2005
-------
equal weight to two different compliance forecast approaches. One compliance forecast approach is
based on the SWAT predictions, and the other is based on the ICR matrix method. The ICR Matrix
Method uses the same basic approach as SWAT, but uses TTHM and HAAS data from the ICR directly to
estimate the percent of plants changing technology to comply with the Stage 2 DBPR and the resulting
DBF reduction (see Chapter 5 for more information on these approaches). To characterize the uncertainty
of the compliance forecast results, EPA assumes a uniform distribution between SWAT and ICR Matrix
Method results. That is, the national cost estimates presented in this chapter represent the midpoint
between costs estimated using the SWAT model, and those estimated using the ICR Matrix Method. Cost
estimates using the SWAT model are about 25% lower than the midpoint estimates while those using the
ICR Matrix Method are about 25% higher. Benefits estimates using the SWAT model are about 30
percent lower than the midpoint estimate, while those using the ICR Matrix Method are about 30 percent
higher.
A related source of uncertainty in producing net benefits is the potential impacts of the IDSE on
the compliance forecast, which will determine the treatment technology installed and the level of costs
incurred and benefits achieved by the rule. The primary economic analysis of the Stage 2 DBPR
incorporates the uncertainty of the IDSE effect by forecasting treatment technology change for large and
medium surface water systems based upon a safety margin that is equally weighted between 20 and 25
percent. To inform the reader of the potential magnitude of this uncertainty, EPA conducted a sensitivity
analysis to compare the predicted percentage of plants changing technology and related reduction in DBP
occurrence for the 20 and 25 percent safety margin. Results, presented in Exhibit 5.29, show that the
impact of the safety margin on the analysis is more pronounced for national DBP reduction estimates than
for estimates of total percent of plants changing treatment technology. This implies that uncertainty in the
IDSE will have a larger impact on the national benefits estimates compared to the national cost estimates.
Another source of uncertainty is the approach used to estimate the number of bladder cancer
cases in the baseline that can be attributed to DBP occurrence and exposure, and the number of cases that
can be avoided by implementation of the Stage 2 DBPR. EPA has developed three approaches to
estimating the number of bladder cancer cases attributable to DBFs. Taken together, the three approaches
provide a reasonable estimate of the range of potential risks. For simplicity's sake, one estimate, based on
a 2003 meta-analysis by Villanueva et al, is carried through the full benefits analysis.
Final Economic Analysis for the Stage 2 DBPR 9-24 December 2005
-------
Exhibit 9.16 Effects of Uncertainties on National Estimates
Assumptions for Which
There Is Uncertainty
Uncertainty in the industry
baseline (SDWIS and 1995
CWSS data)
Uncertainty in observed data
and predictive tools used to
characterize DBP occurrence
for the pre-Stage 1 baseline
Uncertainty in predictive tools
used to develop the
compliance forecast for
surface water systems
(SWAT and ICR Matrix
Method)
Uncertainty in ground water
compliance forecast
methodologies
Operational safety margin of
20%
Impacts of the IDSE on the
compliance forecast and
predicted DBP reduction for
the Preferred Regulatory
Alternative
Uncertainty in the PAR value
Reduction in TTHM and
HAAS used as proxies for all
chlorination DBPs
DBPs have a linear no-
threshold dose-response
relationship for bladder
cancer effects
Uncertainty in benefits
valuation inputs
Section
with Full
Discussion
of
Uncertainty
3.4
3.7
Chapter 5,
Appendix A
Chapter 5,
A and B
5.2
5.7
6.1.1
Appendix E
6.3.3
6.2.1
6.5.2
Potential Effect on Benefit
Estimate
Under-
estimate
Over-
estimate
Un-
known
Impact
X
X
Quantified in primary analysis
(addresses potential
underestimate or overestimate)
X
X
Quantified in the primary
analysis (addresses potential
underestimate)
Quantified in the primary
analysis (addresses range of
potential effects, but true values
could lie outside the range)
X
X
Quantified in the primary
analysis (addresses potential
underestimate or overestimate)
Potential Effect on Cost
Estimates
Under-
estimate
Over-
estimate
Un-
known
Impact
X
X
Quantified in primary analysis
(addresses potential
underestimate or overestimate)
X
X
Quantified in the primary
analysis (addresses potential
underestimate)
Final Economic Analysis for the Stage 2 DBPR
9-25
December 2005
-------
Assumptions for Which
There Is Uncertainty
Benefits of reduced cancers
other than bladder cancer are
not included in the
quantitative analysis
Value of potential
reproductive and
developmental health effects
avoided is not quantified in
the primary analysis
Treatment costs do not
include costs for minor
operational changes
predicted by SWAT
Median operational and water
quality parameters
considered for technology
unit costs
Economies of scale for
combination treatment
technologies not considered
UV dose assumptions
needed for viral inactivation
Potential low-cost alternatives
to treatment not considered
Uncertainties in unit costs
Section
with Full
Discussion
of
Uncertainty
6.7
6.8
7.4.2
7.4.1
7.4.1
7.4.1
7.4.2
7.4.3
Potential Effect on Benefit
Estimate
Under-
estimate
Over-
estimate
Un-
known
Impact
Quantified in a sensitivity
analysis (addresses potential
underestimate)
X
Potential Effect on Cost
Estimates
Under-
estimate
X
Over-
estimate
X
X
X
Un-
known
Impact
X
Quantified in primary analysis
(addresses potential
overestimate or underestimate)
Final Economic Analysis for the Stage 2 DBPR
9-26
December 2005
-------
9.5 Summary of Conclusions
The following is a summary of the important points that must be considered when weighing the
benefits and costs of the Stage 2 DBPR.
1) The quantified benefits estimate is potentially understated because it does not include the
benefits for reductions in adverse reproductive and developmental health effects, other health
effects, and non-health effects associated with DBF reduction (Exhibits 9.2 - 9.4). At the
same time, EPA recognizes that the lower bound of the quantified benefits estimates could be
as low as zero since causality has not yet been established between exposure to chlorinated
water and bladder cancer.
2) The mean cost estimates of $78.8 million (3 percent discount rate) and $76.8 million (7
percent discount rate) represent EPA's best estimates of the monetary impacts of the Stage 2
DBPR Preferred Regulatory Alternative.
3) The Stage 2 DBPR Preferred Regulatory Alternative provides the greatest benefits at a cost
level that is considered reasonable (i.e., is not cost prohibitive) (Exhibits 9.11 and 9.12) given
the uncertainties in health effects data. Alternatives 2 and 3 have a greater net benefit than
the Preferred Alternative. However, net benefits do not include the unqualified benefits.
Because the Preferred Alternative uses a risk targeting strategy, unquantified benefits effects
for the Preferred Alternative are expected to be higher in proportion to cost as compared to
Alternatives 2 and 3.
4) Evaluation of the cost-effectiveness of the Preferred Regulatory Alternative shows it to be the
most cost effective according to the measures evaluated (Exhibit 9.14).
5) The number of cases of cancer that must be avoided to break even with the cost of the
Preferred Alternative rule are approximately 18 and 17 cases using 3 and 7 percent discount
rates, respectively, and the WTP estimate for avoiding non-fatal lymphoma cases. Using the
WTP for avoided chronic bronchitis, the break even points are about 99 and 96 cases based
on 3 percent and 7 percent discount rates, respectively. The mean estimate of 279 bladder
cancer cases avoided annually is much higher than (achieves more than) both of these
thresholds (Exhibit 9.8).
As a result of all these considerations, EPA believes that the annualized benefits of the Stage 2
DBPR will likely exceed the annualized national costs and will be effective in reducing the risks to
consumers from exposure to DBFs in drinking water.
Final Economic Analysis for the Stage 2 DBPR 9-27 December 2005
-------
10. References
American Cancer Society (ACS) Website. 2004. What are the Key Statistics for Bladder Cancer? Cancer
Reference Information, http://www.cancer.org/. Accessed 2004.
American Cancer Society Website. 2005. Overview: Colon and Rectum Cancer. How Many People Get
Colorectal Cancer?
http://www.cancer.org/docroot/CRI/content/CRI_2_2_lX_How_Many_People_Get_Colorectal_
Cancer.asp?sitearea=
American Water Works Association (AWWA). 2000. WATER:\STATS. Database containing 1996
AWWA survey of water systems. AWWA: Denver, CO. June, 2000.
Amy, G., M. Siddiqui, K. Ozelcin, H.W. Zhu, C.Wang. 1998. Empirically based models for predicting
chlorination and ozonation by-products: haloacetic acids, chloralhydrate, and bromate. EPA
Report CX 8195 79.
Arbuckle, T.E., S.E. Hrudey, S.W. Krasner, J.R Nuckols, S.D. Richardson, P. Singer, P. Mendola, L.
Dodds, C. Weisel, D.L. Ashley, K.L. Froese, RA. Pegram, I.R Schultz, J. Reif, A.M. Bachand,
P.M. Benoit, M. Lynberg, C. Poole, and K. Waller. 2002. Assessing exposure in epidemiologic
studies to disinfection by-products in drinking water: report from an international workshop.
Environmental Health Perspectives. 110(Suppl. 1):53-60.
Aschengrau, A., S. Zierler, and A. Cohen. 1989. Quality of Community Drinking Water and the
Occurrence of Spontaneous Abortions. Archives of Environmental Health. 44:283-290.
Aschengrau, A, Zierler, S., and Cohen, A. 1993. Quality of Community Drinking Water and the
Occurrence of Late Adverse Pregnancy Outcomes. Archives of Environmental Health. 48:105-
113.
Association of State Drinking Water Administrators (ASDWA). 2001. ASDWA Needs Analysis.
Backer, L.C., D.L. Ashley, M.A. Bonin, F.L. Cardinali, S.M. Kieszak, and J.V. Wooten. 2000.
Household Exposures to Drinking Water Disinfection By-products: Whole Blood Trihalomethane
Levels. Journal of Exposure Analysis and Environmental Epidemiology. 10(4):321-326.
Ballester, N.A., and J.P. Malley. 2004. Journal AWWA - Sequential Disinfection of Adenovirus Type 2
with UV-Chlorine-Chloramine. American Water Works Association, October, 2004. p. 97.
Barrett S., C. Hwang, Y. Guo, S.A. Andrews, and R. Valentine. 2003. Occurrence of NDMA in
Drinking Water: A North American Survey, 2001- 2002. Proceedings of 2003 AWWA Annual
Conference, Anaheim, CA.
Batterman, S., A.T. Huang, S.G. Wang, and L. Zhang. 2000. Reduction of ingestion exposure to
trihalomethanes due to volatilization. Environmental Science and Technology. 34(20):4418-
4424.
Bean, J.A., P. Isacson, W.J. Hausler, Jr., and J. Kohler. 1982. Drinking water and cancer incidence in
Iowa. I. Trends and incidence by source of drinking water and size of municipality. American
Journal of Epidemiology. 116(6): 912-923.
Final Economic Analysis for the Stage 2 DBPR R-l December 2005
-------
Bielmeier, S.R., D.S. Best, D.L. Guidici, and M.G. Narotsky. 2001. Pregnancy Loss in the Rat Caused
by Bromodichloromethane. Toxicological Sciences. 59(2):309-315.
Bielmeier, S.R., D.S. Best, and M.G. Narotsky. 2004. Serum Hormone Characterization and Exogenous
Hormone Rescue of Bromodicholoromethane-Induced Pregnancy Loss in the F344 Rat.
Toxicological Sciences. 77(1): 101-108.
Boardman, Anthony E., David H. Greenberg, Aidan R. 1996. Vining and David L. Weimer, Cost-Benefit
Analysis Concepts and Practice, p. 405.
Bove, F.J., M.C. Fulcomer, J.B. Klotz, J. Esmart, E.M. Dufficy, R. Zagraniski, and J.E. Savrin. 1992a.
Public Drinking Water Contamination and Birthweight, Fetal Deaths, and Birth Defects: A Cross-
Sectional Study (Phase IV-A). New Jersey Department of Health. April 1992.
Bove, F.J., M.C. Fulcomer, J.B. Klotz, J. Esmart, E.M. Dufficy, R. Zagraniski, and J.E. Savrin. 1992b.
Public Drinking Water Contamination and Birthweight, and Selected Birth Defects: A Case-
Control Study (Phase IV-B). New Jersey Department of Health. May 1992.
Bove, F.J., M.C. Fulcomer, J.B. Klotz, J. Esmart, E.M. Dufficy, and J.E. Savrin. 1995. Public Drinking
Water Contamination and Birth Outcomes. American Journal of Epidemiology. 141(9):850-862.
Bove F.J., Y. Shim, and P. Zeitz. 2002. Drinking Water Contaminants and Adverse Pregnancy
Outcomes: A Review. Environmental Health Perspectives. 110(Suppl. l):61-74.
Bravata, D.M. and I. Olkin. 2001. Simple pooling versus combining in meta-analysis. Evaluation and
the Health Professions. 24(2):218-230.
Cantor, K.P., R. Hoover, and P. Hartge. 1985. Drinking Water Source and Bladder Cancer: A Case-
Control Study. In: Water Chlorination: Chemistry, Environmental Impact and Health Effects,
vol. 5, R.L. Jolley, RJ. Bull, and W.P. Davis (eds.). 1:145-152. Chelsea, MI: Lewis Publishers,
Inc.
Cantor, K.P., R. Hoover, P. Hartge, T.J. Mason, D.T. Silverman, R. Airman, D.F. Austin, M.A. Child,
C.R. Key, L.D. Marrett, M.H. Myers, A.S. Narayana, L.I. Levin, J.W. Sullivan, G.M. Swanson,
D.B. Thomas, and D.W. West. 1987. Bladder Cancer, Drinking Water Source, and Tap Water
Consumption: A Case-Control Study. Journal of the National Cancer Institute. 79(6): 1269-1279.
Cantor, K.P., C.F. Lunch, M. Hildesheim, M. Dosemeci, J. Lubin, M. Alavanja, and G.F. Craun. 1998.
Drinking Water Source and Chlorination Byproducts I. Risk of Bladder Cancer. Epidemiology.
9(l):21-28.
Cantor, K.P, C.F. Lynch, M.E. Hildesheim, M. Dosemeci, J. Lubin, M. Alavanja, and G. Craun. 1999.
Drinking water source and Chlorination byproducts in Iowa. III. Risk of brain cancer. American
Journal of Epidemiology. 150(6):552-560.
Cedergren, M.I., A.J. Selbing, O. Lofinan, and B.A.J. Kallen. 2002. Chlorination byproducts and nitrate
in drinking water and risk for congenital cardiac defects. Environmental Research.
89(2):124-130.
CDC. 1995. Economic Costs of Birth Defects and Cerebral Palsy-United States, 1992.
http://www.cdc.gov/mmwr/preview/mmwrhtml/00038946.htm.
Final Economic Analysis for the Stage 2 DBPR R-2 December 2005
-------
CDC. 2005. Birthweight and Gestation Fact Sheet for 2002.
http://www.cdc.gov/nchs/fastats/birthwt.htm
Chen, C.W., and H. Gibb. 2003. Procedures for calculating cessation lag. Regulatory Toxicology and
Pharmacology. 38(2): 157-65.
Chen, J., G.C. Douglas, T.L. Thirkill, P.N. Lohstroh, S.R. Bielmeir, M.G. Narotsky, D.S. Best, R.A.
Harrison, K. Natarajan, R.A. Pegram, J.W. Overstreet, and B.L. Lasley. 2003. Effect of
bromodichloromethane on chorionic gonadotropin secretion by human placental trophoblast
cultures. Toxicological Sciences. 76(l):75-82.
Chen, J., T.L. Thirkill, P.N. Lohstroh, S.R. Bielmeir, M.G. Narotsky, D.S. Best, RA. Harrison, K.
Natarajan, R.A. Pegram, J.W. Overstreet, B. L. Lasley, and G.C. Douglas. 2004.
Bromodichloromethane inhibits human placental trophoblast differentiation. Toxicological
Sciences. 78(1): 166-174.
Chevrier, C., B. Junod, and S. Cordier. 2004. Does ozonation of drinking water reduce the risk of
bladder cancer? Epidemiology. 15(5):605-614.
Christian, M.S., R.G. York, A.M. Hoberman, R.M. Diener, and L.C. Fisher. 2001. Oral (drinking water)
developmental toxicity studies of bromodichloromethane (BDCM) in rats and rabbits.
International Journal of Toxicology. 20(4):225-237.
Christian M.S., R.G. York, A.M. Hoberman, R.M. Diener, and L.C. Fisher. 2002a. Oral (drinking water)
Two Generation Reproductive Toxicity Study of Bromodichloromethane (BDCM) in Rats.
International Journal of Toxicology. 21(2): 115-146.
Christian, M.S., R.G. York, A.M. Hoberman, L.C. Frazee, L.C. Fisher, W.R. Brown, and D.M. Creasy.
2002b. Oral (drinking water) Two Generation Reproductive Toxicity Study of Dibromoacetic
Acid (DBA) in Rats. International Journal of Toxicology. 21(4):237-76.
Clark, R.M., J.Q. Adams, and B.W. Lykins. 1994. DBP Control in Drinking Water: Cost and
Performance. Journal of Environmental Engineering. 120(4):759-782.
Cohen, S.M., M. Cano, T. Sakata, and S.L. Johansson. 1988. Ultrastructural characteristics of the fetal
and neonatal rat urinary bladder. Journal of Scanning Micros. 2:2091-2104.
Cohen, S.M., and S.L. Johansson. 1992. Epidemiology and Etiology of Bladder Cancer. Urologic
Clinics of North America. 19(3): 421-428.
Cohen, S.M., T. Shirai, and G. Steineck. 2000. Epidemiology and etiology of premalignant and
malignant urothelial changes. Scandinavian Journal of Urology and Nephrology.
Suppl.(205): 105-15.
Cordier, S., J. Clavel, J.C. Limasset, L. Boccon-Gibod, N. Le Moual, L. Mandereau, and D. Hemon.
1993. Occupational risks of bladder cancer in France: a multicentre case-control study.
International Journal of Epidemiology. 22(3):403-11.
Corley, R.A., S.M. Gordon, and L.A. Wallace. 2000. Physiologically Based Pharmacokinetic Modeling
of the Temperature-dependent Dermal Absorption of Chloroform by Humans Following Bath
Water Exposures. Toxicological Science. 53(1): 13-23.
Final Economic Analysis for the Stage 2 DBPR R-3 December 2005
-------
Cragle, D.L., C.M. Shy, R.J. Struba, and E.J. Siff. 1985. A Case-Control Study of Colon Cancer and
Water Chlorination in North Carolina. In: Water Chlorination: Chemistry, Environmental Impact
and Health Effects, vol. 5. R.L. Jolley, R.J. Bull, and W.P. Davis (eds.). Chelsea, MI: Lewis
Publishers, Inc.
Craun GC, ed. 1998. EPA Panel Report and Recommendations for Conducting Epidemiological
Research on Possible Reproductive and Developmental Effects of Exposure to Disinfected
Drinking Water. U.S. EPA, NHEERL. Research Triangle Park, NC.
DeAngelo, A.B., F.B. Daniel, B.M. Most, and G.R.Olson. 1997. Failure of Monochloroacetic Acid and
Trichloroacetic Acid Administered in the Drinking Water to Produce Liver Cancer in Male
F344/N rats. Journal of Toxicology and Environmental Health. 52(5):425-445.
DeAngelo, A.B., M.H. George, and D.E. House. 1999. Hepatocarcinogenicity in the male B6C3F1
mouse following a lifetime exposure to dichloroacetic acid in the drinking water: dose-response
determination and modes of action. Journal of Toxicology and Environmental Health.
58(8):485-507.
Do, M.T., N.J. Birkett, K.C. Johnson, D. Krewski, P. Villeneuva, and the Canadian Cancer Registries
Epidemiology Research Group. 2005. Chlorination Disinfection By-products and Pancreatic
Cancer Risk. Environmental Health Perspectives. 113(4):418-424.
Dodds, L., W. King, C. Wolcott, and J. Pole. 1999. Trihalomethanes in public water supplies and
adverse birth outcomes. Epidemiology. 10(3): 233-237.
Dodds, L. and W.D. King. 2001. Relation between trihalomethane compounds and birth defects. Journal
for Occupational and Environmental Medicine. 58(7): 443-46.
Dodds, L., W. King, A.C. Allen, B.A. Armson, D.B. Deshayne, and C. Nimrod. 2004. Trihalomethanes
in public water supplies and risk of stillbirth. Epidemiology. 15(2): 179-186.
Dominitz, J.A. and Provenzale, D., "Patient Preferences and Quality of Life Associated with Colorectal
Cancer Screening," AmJGastroenterol, 92: 2171-2178, 1997.
Doyle, T.J., W. Sheng, J.R. Cerhan,C.P. Hong, T.A. Sellers, L.H. Kushi, and A.R Folsom. 1997. The
Association of Drinking Water Source and Chlorination By-Products with Cancer Incidence
Among Postmenopausal Women in Iowa: A Prospective Cohort Study. American Journal of
Public Health. 87(7).
Edwards, M. 1997. Predicting DOC removal during enhanced coagulation. Journal of the American
Waterworks Association. 89(5):78-89.
Fenster, L., K. Waller, G. Windham, T. Henneman, M. Anderson, P. Mendola, J.W. Overstreet and S.H.
Swan. 2003. Trihalomethane levels in home tap water and semen quality. Epidemiology.
14:650-658.
Freedman, M., K.P. Cantor, N.L. Lee, L.S. Chen, H.H. Lei, C.E. Ruhl, and S.S. Wang. 1997. Bladder
Cancer and Drinking Water: A Population-Based Case Control Study in Washington County,
Maryland. Cancer Causes and Control. 8(5):738-744.
Fukushima, S. and H. Wanibuchi. 2000. Prevention of urinary bladder cancer: The interface between
experimental and human studies. Asian Pacific Journal of Cancer Prevention. 1:15-33.
Final Economic Analysis for the Stage 2 DBPR R-4 December 2005
-------
Gallagher, M.D., J.R. Nuckols, L. Stallones, and D.A. Savitz. 1998. Exposure to trihalomethanes and
adverse pregnancy outcomes. Epidemiology. 9:484-489.
George, M.H., G.R. Olson, D. Doerfler, T. Moore, S. Kilburn, and A.B. DeAngelo. 2002.
Carcinogenicity of bromodichloromethane administered in drinking water to male F344/N rats
and B6C3F(1) mice. International Journal of Toxicology. 21(3):219-230.
Gerba, C.P., J.B. Rose, and C.N.Haas. 1996. Sensitive Populations: Who is at the Greatest Risk.
International Journal of Food and Microbiology. 30:113-123.
Goebell, P.J., C.M. Villanueva, and A.W. Rettenmeier. 2004. Environmental exposure, chlorinated
drinking water, and bladder cancer. World Journal of Urology. 21(6):424-432.
Goetzel, R.Z., S.R. Long, R.J. Ozminkowski, K. Hawkins, S. Wang, and W. Lynch. 2004. Health,
Absence, Disability, and Presenteeism Cost Estimtes of Certain Physical and Mental Health
Conditions Affecting U.S. Employers. J Occup Environ Med. 46 (4): 1-15. April 2004.
Gold, M.R., J.E. Siegel, L.B. Russel, and M.C. Weinstein. 1996. Cost-effectiveness in Health and
Medicine. New York: Oxford University Press.
Gold, M.R, D. Stevenson, and D.G. Fryback. 2002. HALYS and QALYS and DALYS, Oh My:
Similarities and Differences in Summary Measures of Population Health. Annual Review of
Public Health. 23:115-34.
Gordis, L. 2000. Epidemiology. Second Edition. W.B. Saunders Company. Philadelphia, PA.
Gordon, S.M., L.A. Wallace, P.J. Callahan, D.V. Kenny, and M.C. Brinkman. 1998. Effect of Water
Temperature on Dermal Exposure to Chloroform. Environ Health Perspectives. 106(6):337-345.
Graves, C.G., G.M. Matanoski and R.G. Tardiff 2001. Weight of evidence for an association between
adverse reproductive and developmental effects and exposure to disinfection by-products: a
critical review. Regulatory Toxicology and Pharmacology. 34:103-124.
Grunberg, S.M., Boutin, N., Ireland, A., Miner, S., Silvera, J., and Ashikaga, T. Impact of
Nausea/Vomiting on Quality of Life as a Visual Analogue Scale-Derived Utility Score. Support
Care Cancer. 4(6): 435-439. November 1996.
Grunberg, S.M., Srivastava, A., Grunberg, K.J., Weeks, J. 2002. Intensity of Chemotherapy-Induced
Emesis and Overall Survival as Determinants of a Global Utility Score. Support Care Cancer.
10(8): 624-629. November 2002.
Halpern, M.T., B.W. Gillespie, and K.E. Warner. 1993. Patterns of Absolute Risk of Lung Cancer
Mortality in Former Smokers. Journal of the National Cancer Institute. 85(6): 457-464.
Hartge, P., D. Silberman, R. Hoover, C. Schairer, R. Altaian, D. Austin, K. Cantor, M. Child, C. Key, and
L.D. Marrett. 1987. Changing Cigarette Habits and Bladder Cancer Risk: A Case-Control Study.
Journal of the National Cancer Institute. 78(6): 1119-1125.
Harvard School of Public Health, Cost-Effectiveness Analysis Database (accessed January 2005):
www.hcra.harvard.edu/pdf/preferencescores.pdf
Havelaar, A.H., de Hollander, A.E.M., Teunis, P.F.M., Evers, E.G., van Kranen, H.J., Versteegh, P.M.,
van Koten, J.E.M., and Slob, W. 2000. Balancing the Risks and Benefits of Drinking Water
Disinfection: Disability Adjusted Life-Years on the Scale. Environmental Health Perspectives,
108(4): 315-321. April 2000.
Final Economic Analysis for the Stage 2 DBPR R-5 December 2005
-------
Hildesheim, M.E., K.P. Canbor, C.F. Lynch, M. Dosemeci, J. Lubin, M. Alavanja, and G.F. Craun. 1998.
Drinking Water Source and Chlorination Byproducts: Risk of Colon and Rectal Cancers.
Epidemiology. 9(l):29-35.
Hooper, S.M. and S.C. Allgeier. 2002. GAC and Membrane Treatment Studies Data Analysis. In:
Information Collection Rule Data Analysis Report. McGuire, M.J., J. McLain, and A. Obolensky
(eds.). Denver, CO: AWWARF.
Hristova, L. and Hakama, M. 1997. Effect of Screening for Cancer in the Nordic Countries on Deaths,
Cost, and Quality of Life up to the Year 2017. Acta Oncology. 36S9: 1-60. 1997.
Hrubec, Z. and McLaughlin, J.K. 1997b. Former cigarette smoking and mortality among U.S. veterans: a
26-year follow-up. In: Changes in cigarette related disease risks and their implication for
prevention and control. D.M. Burns, L. Garfinkel, J.M. Samet (eds.). NIH Monograph No. 8,
National Institutes of Health. 501-530. Washington, DC: National Cancer Institute.
Hwang, B., P. Magnus, and J.K. Jaakkola. 2002. Risk of specific birth defects in relation to chlorination
and the amount of natural organic matter in the water supply. American Journal of
Epidemiology. 156(4):374-382.
Hwang, B.F. and J.J.K. Jaakkola. 2003. Water chlorination and birth defects: A systematic review and
meta-analysis. Archives of Environmental Health. 58(2):83-91.
Industrial Economics, Incorporated. 2002. Valuing Time Losses Due to Illness under the 1996
Amendments to the Safe Drinking Water Act (Draft Final). Prepared for the Office of Ground
Water and Drinking Water, U.S. Environmental Protection Agency. October 2002.
Infante-Rivard, C., E. Olson, L. Jacques, and P. Ayotte. 2001. Drinking Water Contaminants and
Childhood Leukemia. Epidemiology. 12(l):3-9.
Infante-Rivard, C., D. Amre, and D. Sinnett. 2002. GSTT1 and CYP2E1 polymorphisms and
trihalomethanes in drinking water: effect on childhood leukemia. Environmental Health
Perspective. 110(6): 591-593.
Infante-Rivard, C. 2004. Drinking water contaminants, gene polymorphisms, and fetal growth.
Environmental Health Perspectives. 112(11):1213-1216.
International Life Sciences Institutes (ILSI). 1997. An evaluation of EPA's proposed guidelines for
carcinogen risk assessment using chloroform and dichloroacetate as case studies: Report of an
expert panel. Washington, D.C., November.
International Life Science Institute and Risk Science Institute (ILSI and RSI). 1998. Assessing the
Toxicity of the Exposure of Mixture to Disinfection Byproducts. Published as part of the Research
Recommendations under a cooperative agreement with U.S. EPA's Office of Water, March.
IRIS. 1993. Integrated Risk Information System (IRIS). N-nitrosodimethylamine (NDMA).
Washington, DC: U.S. EPA. http://www.epa.gov/iris/subst/0045.htm
IRIS. 2000. Integrated Risk Information System (IRIS). Toxicological Review of Chlorine Dioxide and
Chlorite. Washington, DC: U.S. EPA. EPA/635/R-00-007.
http://www.epa.gov/iris/toxreviews/0496-tr.pdf
Final Economic Analysis for the Stage 2 DBPR R-6 December 2005
-------
IRIS. 2001a. Integrated Risk Information System (IRIS). Toxicological Review Bromate. Washington,
DC: U.S. EPA. EPA/635/R-01/002. http://www.epa.gov/iris/toxreviews/1002-tr.pdf
IRIS. 2001b. Integrated Risk Information System (IRIS). Toxicological Review of Chloroform.
Washington, DC: U.S. EPA. EPA/635/R-01/001.
http://www.epa.gov/iris/toxreviews/0025-tr.pdf
IRIS. 2003. Integrated Risk Information System (IRIS). Toxicological Review for Dichloroacetic Acid:
Consensus Review Draft. Washington, DC: U.S. EPA. EPA/635/R-03-007.
http://www.epa.gov/iris/subst/0654.htm
Jaakkola, J.J.K., P .Magnus, A. Skrondal, B.F. Hwang, G. Becher, and E. Dybing. 2001. Fetal growth
and duration of gestation relative to water chlorination. Journal of Occupational and
Environmental Medicine. 58(7):437-442.
Kallen, B.A.J .and E. Robert. 2000. Drinking water Chlorination and Delivery Outcome - a Registry
Based Study in Sweden. Reproductive Toxicology. 14:303-309.
Kanitz, S, Y. Franco, V. Patrone, M. Caltabellotta, E. Raffo, C. Riggi, D. Timitilli, and G. Ravera. 1996.
Association between drinking water disinfection and somatic parameters at birth. Environmental
Health Perspectives. 104(5):516-520.
Kaydos, E.H., J.D. Suarez, N.L.,Roberts, K. Bobseine, R. Zucker, J. Laskey, and G.R. Klinefelter. 2004.
Haloacid Induced Alterations in Fertility and the Sperm Biomarker SP22 in the Rat Are Additive:
Validation of an ELISA. Toxicological Sciences. 8:430-442.
King, W.D. and L.D. Marrett. 1996. Case-Control Study of Bladder Cancer and Chlorination By-
Products in Treated Water (Ontario, Canada). Cancer Causes Control. 7:596-604.
King, W.D., L.D. Marrett, and C.G. Woolcott. 2000a. Case-Control Study of Colon and Rectal Cancers
and Chlorination Byproducts in Treated Water. Cancer Epidemiology, Biomarkers and
Prevention. 9(8):813-818.
King, W., L. Dodds and A. Allen. 2000b. Relation between Stillbirth and Specific Chlorination By-
products in Public Water Supplies. Environmental Health Perspectives. 108(9):883-886.
King, W.D., L. Dodds, A.C. Allen, B.A. Armson, D. Fell, and C. Nimrod. 2005. Haloacetic acids in
drinking water and risk for stillbirth. Journal of Occupational and Environmental Medicine.
62(2): 124-127.
Kleckner, N. and J. Neuman. 2000. Update to Recommended Approach to Adjusting WTP Estimates to
Reflect Changes in Real Income. Industrial Economics, Incorporated. Memorandum to Jim
DeMocker. U.S. Environmental Protection Agency. September 30, 2000.
Klinefelter, G.R., E.S. Hunter, and M. Narotsky. 2001. Reproductive and Developmental Toxicity
Associated with Disinfection By-Products of Drinking Water. In: Microbial Pathogens and
Disinfection By-Products of Drinking Water. ILSI Press, 309-323.
Klinefelter, G.R., L.F. Strader, J.D. Suarez, N.L. Roberts, J.M. Goldman, and A.S. Murr. 2004.
Continuous Exposure to Dibromoacetic Acid Delays Pubertal Development and Compromises
Sperm Quality in the Rat. Toxicological Science. 81:49-429.
Final Economic Analysis for the Stage 2 DBPR R-7 December 2005
-------
Klotz, J.B. and L.A. Pyrch. 1998. A Case Control Study of Neural Tube Defects and Drinking Water
Contaminants. U.S. Department of Health and Human Services, Agency for Toxic Substances
and Disease Registry (ATSDR).
Klotz, J.B. and L.A. Pyrch. 1999. Neural tube defects and drinking water disinfection byproducts.
Epidemiology. 10(4):383-390.
Koechling, M.T., A.N. Rajbhandari, and R.S. Summers. 1998. Proceedings. American Waterworks
Association Annual Conference, Dallas, TX. June 1998, 363-373.
Kogevinas, M. 2005. The importance of cultural factors in the recognition of occupational disease.
Journal of Occupational and Environmental Medicine. 62(5):286.
Kogevinas, M., F. Fernandez, and C. Villanueva. 2005. Personal communication regarding Odds Ratios
as a function of exposure duration to DBFs in Villanueva et al (2003). February 14, 2005.
Kogevinas, M. and C. Villanueva.. 2005. Personal communication regarding THM exposure levels
presented in Villanueva et al (2004). July 15, 2005.
Koivusalo, M., Hakulinen, T., Vartiainen, T., Pukkala, E., Jaakkola, J.J., and Tuomisto, J. 1998.
Drinking water mutagenicity and urinary tract cancers: a population-based case-control study in
Finland. American Journal of Epidemiology. 148(7):704-12.
Koivusalo, M., E. Pukkala, T. Vartiainen, J. J. K. Jaakkola, and T. Hakulinen. 1997. Drinking water
chlorination and cancer-a historical cohort study in Finland. Cancer Causes and Control.
8(2): 192-200.
Komulainen, H., V.M. Kosma, S.L. Vaittinen, T. Vartiainen, E. Kaliste-Korhonen, S. Lotjonen, R.K.
Tuominen, and J. Tuomisto. 1997. Carcinogenicity of the drinking water mutagen 3-chloro-4-
(dichloromethyl)-5-hydroxy-2(5H)-furanone in the rat. Journal of the National Cancer Institute.
89(12):848-56.
Kramer M.D., C.F. Lynch, P. Isacson, and J.W. Hanson. 1992. The Association of waterborne
chloroform with intrauterine growth retardation. Epidemiology. 3(5):407-413.
Kundu, B., S.D. Richardson, P.O. Swartz, P.P. Matthews, A.M. Richard, and D.M. DeMarini. 2004.
Mutagenicity in Salmonella of halonitromethanes: a recently recognized class of disinfection by-
products in drinking water. Mutation Research. 562:39-65.
Kuo, H.W., T.F. Chiang, I.I. Lo, J.S. Lai, C.C. Chan, and J.D. Wang. 1998. Estimates of Cancer Risk
from Chloroform Exposure During Showering in Taiwan. Science of the Total Environment.
218(1): 1-7.
Lewitt, E.M., L.S. Baker, H. Gorman, and P.H. Shiono. 1995. The Direct Cost of Low Birth Weight.
The Future of Children. 5(1).
Linder, R.E., G.R. Klinefelter, L.F. Strader, J.D. Suarez, C.J. Dyer. 1994. Acute spermatogenic effects
of bromoacetic acids. Fundamental and Applied Toxicology. 22(3):422-30.
Linder, R.E., G.R. Klinefelter, L.F. Strader, M.G. Narotsky, J.D. Suarez, N.L. Roberts, and S.D.
Perreault. 1995. Dibromoacetic acid affects reproductive competence and sperm quality in the
male rat. Fundamental and Applied Toxicology. 28(1):9-17.
Final Economic Analysis for the Stage 2 DBPR R-8 December 2005
-------
Linder, R.E., G.R. Klinefelter, L.F. Strader, J.D. Suarez, andN.L. Roberts NL. 1997. Spermatotoxicity
of dichloroacetic acid. Reproductive Toxicology. ll(5):681-8.
Lynch, C.F., R.F. Woolson, T. O'Gorman, and K.P. Cantor. 1989. Chlorinated drinking water and
bladder cancer: effect of misclassification on risk estimates. Archives of Environmental Health.
44(4):252-9.
MacLehose, R. 2005. Personal communication on spontaneous abortion data in Savitz et al. (2005).
Novembers, 2005.
Magat, W.A., W.K. Viscusi, and J. Huber. 1996. A Reference Lottery Metric for Valuing Health.
Management Science. 42:1118-1130.
Magnus, P., J.J.K. Jaakkola, A. Skrondal, J. Alexander, G. Becher, T. Krogh, and E. Dybing. 1999.
Water chlorination and birth defects. Epidemiology. 10(5):513-517.
Malcolm Pirnie Inc. 1992. Water Treatment Plant Simulation Program Version 1.21 User's Manual.
Malcolm Pirnie Inc. June 1992.
March of Dimes Birth Defects Foundation. 1999. Deliver the Best, http://www.modimes.org.
Mauskopf, J.A. and French, M.T. 1991. Estimating the Value of Avoiding Morbidity and Mortality from
Foodborne Illnesses. Risk Analysis. 11 (4):619-631. December 1991.
Mayo Clinic Website. 2004. Bladder Cancer.
http://www.mayoclinic.com/invoke.cfm?id=DS00177&dsection=3. Accessed 2004.
McDowell, I. and Newell, C. 1996. Measuring Health: A Guide to Rating Scales and Questionnaires,
Oxford University Press: New York. 1996.
McGeehin, M.A., J.S. Reif, J.C. Becher, and E.J. Mangione. 1993. Case Control Study of Bladder
Cancer and Water Disinfection Methods in Colorado. American Journal of Epidemiology. 138.
McGuire, M.J., J.L. McLain, and A. Obolensky. 2002. Information Collection Rule Data Analysis.
Awwa Research Foundation and AWWA, Denver.
Miles, A.M., P.C. Singer, D.L. Ashley, M.C. Lynberg, P. Mendola, P.H. Langlois and J.R. Nuckols.
2002. Comparison of trihalomethanes in tap water and blood. Environmental Science and
Technology. 36(8): 1692-1698.
Mills, C.J., R.J. Bull, K.P. Cantor, J. Reif, S.E. Hrudey, and P. Huston. 1998. Workshop report. Health
risks of drinking water chlorination by-products: report of an expert working group. Chronic
Disease in Canada. 19(3):91-102.
Morris, R.D., Audet, A.M., Angelillo, I.F. Chalmers, T.C. Mosteller, F. 1992. Chlorination, chlorination
by-products, and cancer: A meta-analysis. American Journal of Public Health. 82:955-63.
Narotsky, M.G., R.A. Pegram, and R.J. Kavlock. 1992. Full-litter resorptions caused by low-molecular
weight halocarbons in F-344 rats. Teratology. 45:472-473.
Final Economic Analysis for the Stage 2 DBPR R-9 December 2005
-------
National Cancer Institute (NCI) website. 2002. What You Need to Know About Bladder Cancer.
http://www.cancer.gov/cancertopics/wyntk/bladder/page4. Posted 09/07/2001, Updated
09/16/2002. Accessed 2004.
National Drinking Water Advisory Council. 2001. Report of the Arsenic Cost Working Group to the
National Drinking Water Advisory Council. Final Report. August 14, 2001.
National Research Council (NRC). 1990 Arsenic in Drinking Water. National Research Council,
Committee on Toxicology. National Academy Press: Washington.
National Toxicology Program. 1987. Toxicity and carcinogenesis studies of bromodichloromethane
(CAS No. 75-27-4) in F344/N rats and B6C3F1 mice (gavage studies). Technical Report Series
No. 321. Research Triangle Park, NC: U.S. Department of Health and Human Services.
National Toxicology Program. 2004. Toxicology and Carcinogenesis Studies of Sodium Chlorate (CAS
No. 7775-09-9) in F344/N Rats and B6C3F1 Mice (Drinking Water Studies) - Draft Abstract.
TR-517.
http://ntp-server.niehs.nih.gov/index.cfm?objectid=00132319-FlF6-975E-778A4E6504EB9191
National Toxicology Program. 2005a. Toxicology and carcinogenesis studies of bromodichloromethane
(CAS No. 75-27-4) in male F344/N rats and female B6C3F1 mice (Drinking Water Studies) -
Draft Abstract. TR-532.
http://nrp.niehs.nih.gov/INDEX. CFM?OBJECTID=00271EF5-F1F6-975E-73E6FE7AEE1A1A31
National Toxicology Program (NTP). 2005b. Water disinfection byproducts (dibromoacetic acid).
CASNO: 631-64-1.
http://ntp.niehs.nih.gov/index.cfm?objectid=071A45CC-A74F-C13F-lAFDE911CEC2FBDC
(accessed April 1, 2005).
Ness, R.M., Holmes, A.M., Klein, R. and Dittos, R. 1999. Utility Valuations for Outcome States of
Colorectal Cancer. American Journal of Gastroenterology. 94 (6): 1650-1657. June 1999.
Nieuwenhuijsen, M.J., M.B. Toledano, N.E. Eaton, J. Fawell, and P. Elliott. 2000. Chlorination
disinfection by-products in water and their association with adverse reproductive outcomes: a
review. Journal of Occupational and Environmental Medicine. 57(2):73-85.
Norum, J. and Olsen, J.A.. 1997. A Cost-effectiveness Approach to the Norwegian Follow-up
Programme in Colorectal Cancer. Annals of Oncology, 8(11): 1081-1087.
Office of Management and Budget. 2003. Circular A-4. September 2003.
Pereira, V.J., H.S. Weinberg, and P.C. Singer. 2004. Temporal and spatial variability of DBFs in a
choraminated distribution system. Journal of the American Water Works Association.
96(11):91-102.
Plewa, M.J., S.D. Richardson and P. Jazwierska. 2004a. Halonitromethane drinking water disinfection
byproducts: chemical characterization and mammalian cell cytotoxicity and genotoxicity.
Environmental Science and Technology. 38(1): 62-68.
Plewa, M.J., E.D. Wagner, S.D. Richardson, A.D. Thruston Jr, Y.-T. Woo, and A.B. McKague. 2004b.
Chemical and biological characterization of newly discovered iodo-acid drinking water
disinfection by-products. Environmental Science and Technology. 38(18): 4713-4722.
Final Economic Analysis for the Stage 2 DBPR R-10 December 2005
-------
Poole, C.A. 1997. Analytical meta-analysis of epidemiological studies of chlorinated surface water and
cancer: Quantitative review and reanalysis of the work published by Morris et al., Am J Public
Health 1992. 82:955-963. Prepared for USEPA, National Center for Environmental Assessment,
Cincinnati.
Porru S, Scotto di Carlo A, Carta A, Placidi D. 2003. Bladder cancer and occupational activity. G Ital
Med Lav Ergon. 25(3):298-300.
Porter, C.K., S.D. Putnam, K.L. Hunting, and M.R. Riddle. 2005. The Effect of Trihalomethane and
Haloacetic Acid Exposure on Fetal Growth in a Maryland County. American Journal of
Epidemiology. 162(4):334-344.
Ramsey, S.D., Andersen, M.R., Etzioni, R., Moinpour, C., Peacock, S., Potosky, A., and Urban, N.,
Quality of Life in Survivors of Colorectal Carcinoma, Cancer, 88(6): 1294-1303, March 15, 2000.
Ranmuthugala, G., L. Pilotto, W. Smith, T. Vimalasiri, K. Dear, and R. Douglas. 2003. Chlorinated
drinking water and micronuclei in urinary bladder epithelial cells. Epidemiology.
14(5):617-622.
Reif, J.S., M.C. Hatch, M. Bracken, L. Holmes, B. Schwetz, and P.C. Singer. 1996. Reproductive and
developmental effects of disinfection byproducts in drinking water. Environmental Health
Perspectives. 104:1056-1061.
Reif, J.S., A. Bachand and M. Andersen. 2000. Reproductive and Developmental Effects of Disinfection
By-Products. Bureau of Reproductive and Child Health, Health Canada, Ottawa, Ontario,
Canada. Executive summary available at http://www.hc-sc.gc.ca/pphb-
dgspsp/publicat/reif/index.html.
Rice J.M., R.A. Baan, M. Blettner, C. Genevois-Charmeau, Y. Grosse, D.B. McGregor, C. Partensky, and
J.D. Wilbourn. 1999. Rodent tumors of urinary bladder, renal cortex, and thyroid gland in IARC
Monographs evaluations of carcinogenic risk to humans. Toxicological Sciences. 49(2): 166-71.
Richardson, S.D. 2003. Disinfection by-products and other emerging contaminants in drinking water.
Trends in Analytical Chemistry. 22(10):666-684.
Richardson, S.D., J.E. Simmons and G. Rice. 2002. Disinfection by-products: the next generation.
Environmental Science and Technology. 36(9):198A-205A
Rockhill, B., B. Newman, and C. Weinberg. 1998. Use and Misuse of Population Attributable Fractions.
American Journal of Public Health. 88(1): 15-19.
R.S. Means Company Inc. 1999. RSMeans Facilities Construction Cost Data. 14th annual ed. Kingston,
MA.
Savitz, D.A., K.W. Andrews, and L.M. Pastore. 1995. Drinking water and pregnancy outcome in central
North Carolina: Source, Amount, and Trihalomethane levels. Environmental Health
Perspectives. 103(6), 592-596.
Savitz, D.A., Singer, P.C., Hartmann, K.E., Herring, A.H., Weinberg, H.S., Makarushka, C., Hoffman, C.,
Chan, R. and Maclehose, R. 2005. Drinking Water Disinfection By-Products and Pregnancy
Outcome. Sponsored by Microbial/Disinfection By-Products Research Council. Jointly funded
by Awwa Research Foundation and U.S. Environmental Protection Agency.
Final Economic Analysis for the Stage 2 DBPR R-ll December 2005
-------
Schreiber, I.M. and W.A. Mitch. 2005. Influence of the order of reagent addition on NDMA formation
during chloramination. Environmental Science and Technology. 39(10):3811-8.
Schulman, A., M. Manibusan, C. Abernathy, M. Messner, J. Donohue, and D. Gaylor. 2004. Arsenic in
Drinking Water: Cessation Lag Model. Toxicological Sciences.
Schulte, P.A., K. Ringen, G.P. Hemstreet, E. Ward. 1987. Occupational cancer of the urinary tract.
Journal of Occupational Medicine. 2(1):85-107.
Seidel, C. 2001. BAT Memorandum on SWAT Runs for Stage 2 BAT Evaluation. June 25, 2001.
Shaw, G.M., S.H. Swan, J.A. Harris, and L.H. Malcoe. 1990. Maternal water consumption during
pregnancy and congenital cardiac anomalies. Epidemiology. 1(3):206-211.
Shaw, G.M., L.H. Malcoe, A. Milea, and S.H. Swan. 1991. Chlorinated water exposures and cardiac
anomalies. Epidemiology. 2:459-460.
Shaw, G.M., D. Ranatunga, T. Quach, E. Neri, A. Correa, and R.R. Neutra. 2003. Trihalomethane
exposure from municipal water supplies and selected congenital malformations. Epidemiology.
14(2):191-199.
Silverman, D.T., L.I. Levin, R.N. Hoover, and P. Hartge. 1989a. Occupational Risks of Bladder Cancer
in the United States: I. White Men. Journal of the National Cancer Institute. 81: 1472-1480.
Silverman, D.T., L.I. Levin, and R.N. Hoover. 1989b. Occupational Risks of Bladder Cancer in the
United States: II. Nonwhite Men. Journal of the National Cancer Institute. 81: 1480-1483.
Silverman, D.T., L.I. Levin, and R.N. Hoover. 1990. Occupational Risks of Bladder Cancer Among
White Women in the United States. American Journal of Epidemiology. 132(3): 453-461.
Silverman, D.T., A.S. Morrison, and S.S. Devesa. 1996. Bladder cancer. In: Cancer Epidemiology and
Prevention. 2nd edition. Schottenfeld, D. and J.F. Fraumeni Jr. (eds.). 1156-1179. New York:
Oxford University Press.
Singer, P.C. 1999. Formation and Control of Disinfection By-Products in Drinking Water. 13-18.
American Waterworks Association, Denver, Colorado.
Singer, P.C., E. DePaz, D.L. Ashley, B. Blount, J.R. Nuckols, C.R. Wilkes, D. Cade, C. Lyu, S. Gordon,
J. Masters, and M. Brinkman. 2003. Impact of Trihalomethane Concentrations in Tap Water and
Water Use Activities on Biological Levels of Trihalomethanes. AWWA WQTC Conference.
Smith, V.K., G. Van Houtven, and S.K. Pattanayak. 2002. Benefit transfer via preference calibration:
'Prudential algebra'for policy. Land Economics. 78(1): 132-152.
Solarik, G., R.S. Summers, J. Sohn, W.J. Swanson, Z.K. Chowdhury, and G.L. Amy. 2000. Extensions
and Verification of the Water Treatment Plant Model for DBP Formation. In: Natural Organic
Matter and Disinfection Byproducts Characterization and Control in Drinking Water. Amy, G.L.,
S. Huang, and S. Krasner (eds.). American Chemical Society Symposium Series, volume 761.
Washington, D.C.: American Chemical Society.
Final Economic Analysis for the Stage 2 DBPR R-12 December 2005
-------
Stiggelbout, A.M., Kiebert, G.M., Kievit, J., Leer, J.W.H., Habbema, J.D.F., and De Haes, J.D.J.M. 1995.
The 'Utility' of the Time Trade-off Method in Cancer Patients: Feasibility and Proportional
Trade-off, Journal of Clinical Epidemiology. 48(10): 1207-1214.
Summers, R.S., G. Solarik, V.A. Hatcher, R.S. Isabel, and J.F. Stile. 1998. Impact of Point of Chlorine
Addition and Coagulation. Final Project Report. USEPA Office of Ground Water and Drinking
Water: Cincinnati, OH.
Surveillance, Epidemiology, and End Results (SEER) Program (www.seer.cancer.gov). 2004.
SEER*Stat Databases: Incidence - SEER 11 Regs + AK Public-Use, Nov 2003 Sub for Expanded
Races (1992-2001) and Incidence - SEER 11 Regs Public-Use, Nov 2003 Sub for Hispanics
(1992-2001), National Cancer Institute, DCCPS, Surveillance Research Program, Cancer
Statistics Branch, released April 2004, based on the November 2003 submission.
Surveillance, Epidemiology, and End Results (SEER) Program (www.seer.cancer.gov). 2005. Ries,
L.A.G., M.P. Eisner, C.L. Kosary, B.F. Hankey, B.A. Miller, L. Clegg, A. Marietta, E.J. Feuer,
B.K. Edwards (eds.). SEER Cancer Statistics Review, 1975-2002. National Cancer Institute.
Bethesda, MD, released 2005, based on November 2004 submission.
Swan, S.H., R.R. Neutra, M. Wrensch, I. Hertz-Picciotto, G.C. Windham, L. Fenster, D.M. Epstein, and
M. Deane. 1992. Is drinking water related to spontaneous abortion? Reviewing the evidence
from the California Department of Health Services Studies. Epidemiology. 3:83-93.
Swan, S.H., K. Waller, B. Hopkins, G. Windham, L. Fenster, C. Schaefer, and R Neutra. 1998. A
prospective study of spontaneous abortion; relation to amount and source of drinking water
consumed in early pregnancy. Epidemiology. 9:126-133.
Toledano, M.B., M.J. Nieuwenhuijsen, N. Best, H. Whitaker, P. Hambly, C. de Hoogh, J. Fawell, L.
Jarup, and P. Elliott. 2005. Relation of trihalomethane concentrations in public water supplies to
stillbirth and birth weight in three water regions in England. Environmental Health Perspectives.
113(2):225-232.
Torrance, G.W., Zhang, Y., Feeny, D., Furlong, W., Barr, R.. Multi-attribute preference functions for a
comprehensive health status classification system. CHEPA Working Paper Series No. 92-18,
McMaster University, 1992.
Tseng, T., M. Edwards, and Z.K. Chowdhury. 1996. American Water Works Association National
Enhanced Coagulation and Softening Database.
Tyl, R.W. 2000. Review of Animal Studies for Reproductive and Developmental Toxicity Assessment
of Drinking Water Contaminants: Disinfection By-Products (DBFs). RTI Project No. 07639.
Research Triangle Institute.
U.S. Bureau of Labor Statistics. 2003. Employment Cost Index, http://www.bls.gov
U.S. Bureau of Labor Statistics. 2004. Consumer Price Index, http://www.bls.gov
U.S. Census Bureau. 2001. Households and Families: 2000. Census 2000 Brief. C2KBR/01-8.
USDA (U.S. Department of Agriculture). 1997. 1994-1996 USDA Continuing Survey of Food Intakes
by Individuals.
Final Economic Analysis for the Stage 2 DBPR R-13 December 2005
-------
U.S. Department of Commerce, Bureau of the Census. 1992. 1992 Census of Governments, GC92(4)-4:
Finances of Municipal and Township Governments.
U.S. Department of Commerce, Bureau of Economic Analysis. 2004a. Table 1.1.6. Real Gross
Domestic Product, Chained Dollars, Billions of chained (2000) dollars, http://www.bea.gov/
U.S. Department of Commerce, Bureau of Economic Analysis. 2004b. Table 3.9.4. Price Indexes for
Government Consumption Expenditures and Gross Investment, http://www.bea.gov/
USDOE, Energy Information Administration. 2004a. Table Id. Average Retail Price for Bundled and
Unbundled. http://www.eia.doe.gov/cneaf/electricity/esr/esr_tabs.html
USDOE, Energy Information Administration. 2004b. Table ES. Summary Statistics for the United
States, 1992 through 2003.
http://www.eia.doe.gov/cneaf/electricity/page/at_a^lance/gen_tabs.html
USEPA. 1983. Ground Water Supply Survey: Summary of Volatile Organic Contaminant Occurrence
Data. Office of Drinking Water, January 1983.
USEPA. 1986. Guidelines for Carcinogen Risk Assessment. Federal Register, 51(185):33992-34003.
Risk Assessment Forum. EPA 630-R-00-004. September 1986.
USEPA. 1989. Review of Environmental Contaminants and Toxicology. Office of Drinking Water
Health Advisories, 106:225.
USEPA. 1990. Guidance Manual for Compliance with the Filtration and Disinfection Requirements for
Public Water Systems Using Surface Water Sources. Science and Technology Branch Criteria
and Standards Division Office of Drinking Water.
USEPA. 199la. National Primary Drinking Water Regulations; Synthetic Organic Chemicals and
Inorganic Chemicals; Monitoring for Unregulated Contaminants; National Primary Drinking
Water Regulations Implementation; National Secondary Drinking Water Regulations; Final Rule.
Federal Register, 56:20:3526-3597. January 31, 1991.
USEPA. 1991b. Part V, Environmental Protection Agency: Guidelines for Developmental Toxicity Risk
Assessment; Notice. Federal Register, 56:234:63798-63826. Decembers, 1991.
USEPA. 1993. Executive Order 12866. Regulatory Planning and Review. Federal Register,
58:190:51735-51744. October 4, 1993.
USEPA. 1994a. Carcinogenicity peer review of cacodylic acid. Office of Pesticide Programs. U.S.
Environmental Protection Agency: Washington, D.C. 20 pp.
USEPA. 1994b. Comment Response Database for the Stage 1 Disinfection Byproducts Rule. Comment
IDM1.065-001.
USEPA. 1996a. Ground Water Disinfection and Protective Practices in the United States. Prepared by
the EPA and Science Applications International Corporation. November 1996.
USEPA. 1996b. Economic Analysis of Federal Regulations Under Executive Order 12866. Office of
Management and Budget. January 11, 1996.
Final Economic Analysis for the Stage 2 DBPR R-14 December 2005
-------
USEPA. 1996c. Proposed Guidelines for Carcinogen Risk Assessment. Office of Research and
Development. EPA 600-P-92-003C. Federal Register, 61:79:17960-18011. April 23, 1996.
USEPA. 1996d. Part II, Environmental Protection Agency: Reproductive Toxicity Risk Assessment
Guidelines; Notice. Federal Register, 61:212:56274-56322. October 31, 1996.
USEPA. 1997a. Summaries of New Health Effects Data. Office of Science and Technology, Office of
Water. Washington, D.C. October 1997.
USEPA. 1997b. The Benefits and Costs of the Clean Air Act, 1970-1990. Prepared for U.S. Congress.
USEPA. 1997c. Community Water Systems Survey (CWSS), Volumes I and II. Office of Water,
Washington, D.C. EPA 815-R-97-001a and -00Ib.
USEPA. 1997d. Comment Response Database for the Stage 1 Disinfection Byproducts Rule. Comment
ID Jl.057-002.
USEPA. 1998a. Regulatory Impact Analysis for the Stage 1 Disinfectants/Disinfection Byproducts Rule.
Washington, D.C. EPA-815-B-98-002. PB 99-111304. November 12, 1998.
USEPA. 1998b. Comment Response Database for the Stage 1 Disinfection Byproducts Rule. Comment
IDD.009-025.
USEPA. 1998c. National-Level Affordability Criteria Under the 1996 Amendments to the Safe Drinking
Water Act. Prepared by International Consultants, Inc., Hagler Bailly Services, Inc and Janice A.
Beecher, Ph.D. for the EPA. August, 1998.
USEPA. 1998d. Variance Technology Findings for Contaminants Regulated Before 1996. Office of
Water. EPA 815-R-98-003. September, 1998
USEPA. 1998e. Guidance on Implementing the Capacity Development Provisions of the Safe Drinking
Water Act Amendments of 1996. Office of Water. July, 1998. EPA 816-R-98-006.
USEPA. 1998f. National Primary Drinking Water Regulations; Disinfectants and Disinfection
Byproducts; Notice of Data Availability; Proposed Rule. Federal Register, 63:61:15606-15692.
March 31, 1998.
USEPA. 1998g. Quantification of Cancer Risk from Exposure to Chlorinated Water. Office of Science
and Technology, Office of Water. Washington, DC. March 13, 1998.
USEPA. 1999a. Cost of Illness Handbook. Office of Pollution Prevention and Toxics. Chapter 1 II. 8.
Cost of Bladder Cancer. September, 1999. 54pp.
USEPA. 1999b. Guidelines for Carcinogen Risk Assessment. July SAB Review draft. Office of
Research and Development, Washington, DC. USEPA NCEA-F-0644.
http://www.epa.gov/ncea/raf/crasab.htm.
USEPA. 1999c. Treatment Technologies. M-DBP Federal Advisory Committee (FACA2), Meeting #5.
http://www.epa.gov/safewater/
Final Economic Analysis for the Stage 2 DBPR R-15 December 2005
-------
USEPA. 2000a. Surface Water Analytical Tool (SWAT) Version 1.1- Program Design and
Assumptions. Prepared by Malcolm Pirnie, Inc.
USEPA. 2000b. ICR Supplemental Survey Database. Prepared by DynCorp, Inc.
USEPA. 2000c. Geometries and Characteristics of Water Systems Report. Office of Ground Water and
Drinking Water. EPA 815-R-00-024. December, 2000.
USEPA. 2000d. TTHM Plant-Mean Data from Seven States. EPA Office of Ground Water and
Drinking Water.
USEPA. 2000e. Data Reliability Analysis of the EPA Safe Drinking Water Information System/Federal
Version (SDWIS/FED). Office of Water. EPA 816-R-00-020. October, 2000.
USEPA. 2000f Quantitative Cancer Assessment for MX and Chlorohydroxyfuranones. Contract NO.
68-C-98-195. August 11, 2000. Office of Water, Office of Science and Technology, Health and
Ecological Criteria Division, Washington, DC.
USEPA. 2000g. Regulatory Impact Analysis for the Proposed Ground Water Rule. Office of Ground
Water and Drinking Water. Contract 68-C-99-206 and 245. April 5, 2000.
USEPA. 2000h. ICR AUX1 Database. April, 2000 version.
USEPA. 2000i. Estimated per Capita Water Ingestion in the United States. Based on Data Collected by
the United States Department of Agriculture's 1994-96 Continuing Survey of Food Intakes by
Individuals. Office of Science and Technology. EPA Contracts 68-C4-0046 and 68-C-99-233.
USEPA. 2000J. Guidelines for Preparing Economic Analysis. Office of the Administrator. EPA-240-R-
00-003. September, 2000.
USEPA. 2000m. Stage 2 Microbial and Disinfection Byproducts Federal Advisory Committee Agreement
in Principle. FR 65:251:83015-83024. December 29, 2000. http://www.epa.gov/fedrgstr/EPA-
WATER/2000/December/Day-29/w33306.htm and
http://www.epa.gov/safewater/disinfection/st2agreement.html .
USEPA. 2000n. Stage 2 M/DBP FACA Meeting Summaries. November 1997 to June 2000.
http://www.epa.gov/safewater/
USEPA. 2000o. ICR Treatment Study Database CD-ROM, Version 1.0. EPA 815-C-00-003.
USEPA. 200la. User Database. Prepared by The Cadmus Group, Inc.
USEPA. 200 Ib. SWAT run summaries for the Stage 2 DBPREA. Prepared by The Cadmus Group, Inc.
USEPA. 200Ic. Drinking Water Baseline Handbook, Third Edition. Draft. Prepared by International
Consultants, Inc. Contract 68-C6-0039. May, 2001.
USEPA. 200Id. Arsenic Rule Benefits Analysis: An SAB Review. Science Advisory Board. EPA-
SAB-RSAC-01 -005. May 2001.
USEPA. 2001e. Development of Cost of Capital Estimates for Public Water Systems, Final Report.
November, 2001.
USEPA. 2003g. Draft Ultraviolet Disinfection Guidance Manual, EPA 815-D-03-007, June 2003.
Final Economic Analysis for the Stage 2 DBPR R-16 December 2005
-------
USEPA. 2003r. Arsenic in Drinking Water: Cessation Lag Model. Prepared by Sciences International.
Contract No. 68-C-98-195. January 2003.
USEPA. 2003s. Labor Costs for National Drinking Water Rules.
USEPA. 2003t. SDWIS Database. 4th quarter SDWIS freeze data.USEpA 2005a ICR Matnx Method
for the Stage 2 DBPR EA. Prepared by The Cadmus Group, Inc.
USEPA. 2005b. Drinking Water Criteria Document for Brominated Trihalomethanes. Washington, DC.
EPA822-R-05-011.
USEPA. 2005c. Drinking Water Criteria Document for Brominated Acetic Acids. Washington, DC.
EPA 822-R-05-007.
USEPA. 2005d. Drinking Water Addendum to the Criteria Document for Trichloroacetic Acid.
Washington, DC. EPA 822-R-05-010.
USEPA. 2005e. Drinking Water Addendum to the Criteria Document for Monochloroacetic Acid.
Washington, DC. EPA 822-R-05-008.
USEPA. 2005f Guidelines for carcinogen risk assessment. Office of Research and Development,
Washington, DC. EPA/630/P-03/001F. Available online at http://cfpub.epa.gov/ncea/.
USEPA. 2005g. Supplemental guidance for assessing susceptibility from early-life exposure to
carcinogens. Office of Research and Development, Washington, DC. EPA/630/R-03/003F.
Available online at http://cfpub.epa.gov/ncea/.
USEPA. 2005h. Stage 2 Benefits Model. Prepared by the Cadmus Group, Inc.
USEPA. 2005i. Stage 2 Cost Model. Prepared by the Cadmus Group, Inc.
USEPA. 2005J. Drinking Water Addendum to the IRIS Toxicological Review of Dichloroacetic Acid.
Washington, D.C. EPA 822-R-05-009.
USEPA. 2005k. Occurrence Assessment for the Final Stage 2 Disinfectants and Disinfection Byproducts
(D/DBPs) Rule. Prepared by The Cadmus Group, Inc. Contract 68-C-99-206. December 2005.
EPA815-R-05-011.
USEPA. 20051. Information Collection Request for National Primary Drinking Water Regulations: Final
Stage 2 Disinfectants and Disinfection Byproducts Rule. Prepared by The Cadmus Group, Inc.
Contract 68-C-99-206. December 2005. EPA 815-Z-05-002.
USEPA. 2005m. Economic Analysis for Long Term 2 Enhanced Surface Water Treatment Rule.
Prepared by The Cadmus Group, Inc., Arlington, VA. Contract 68-C-02-026. December 2005.
USEPA. 2005n. Technologies and Costs for the Final Long term 2 Enhanced Surface Water Treatment
Rule and Final Stage 2 Disinfectants and Disinfection Byproducts Rule. Prepared by The
Cadmus Group, Inc. Contract 68-C-99-206. December 2005. EPA 815-R-05-012.
USEPA. 2005o. Regulatory Impact Analysis for the Final Clean Air Interstate Rule.
EPA-452/R-05-002. Appendix G, pp. 369 - 420. March 2005.
Final Economic Analysis for the Stage 2 DBPR R-17 December 2005
-------
USEPA. 2005p. The Cost of Illness Handbook. Accessed on-line in January 2005:
www.epa.gov/oppt/coi/mdex.htntljSEpA 2005q Unregulated Contaminant Monitoring
Regulation (UCMR) for Public Water Systems Revisions. Proposed Rule. 70 FR
49093-49138, August 22, 2005.
USFDA. Preliminary Regulatory Impact Analysis and Initial Regulatory Flexibility Analysis of the
Proposed Rules to Ensure the Safety of Juice and Juice Products; Proposed Rule. 63 FR 24253;
May 1, 1998.
USFDA. Hazard Analysis and Critical Control Point (HAACP); Procedures for the Safe and Sanitary
Processing and Importing of Juice; Final Rule. 66 FR 6137; January 19, 2001.
USFDA. Establishment and Maintenance of Records Under the Public Health and Security and
Bioterrorism Preparedness and Response Act of 2002; Establishment and Maintenance of
Records for Foods; Notice of Public Meeting; Availability of Draft Guidance for Records Access
Authority; Final Rules and Notice. 69 FR 71561; December 9, 2004.
Veeramachaneni, D.N.R., J.S. Palmer, C.M. Kane, and T.T. Higuchi. 2000. Dibromoacetic acid, a
disinfection by-product in drinking water, impairs sexual function and fertility in male rabbits.
Biology of Reproduction. 62:246.
Vena, J.E., S. Graham, J. Freudenheim, J. Marshall, M. Zielezny, M. Swanson, and G. Sufrin. 1993.
Drinking Water, Fluid Intake, and Bladder Cancer in Western New York. Archives of
Environmental Health. 48(3): 191-197.
Ventura, S.J., W.D. Mosher, S.C. Curtin, J.C. Abma, and S. Henshaw. 2000. Trends in Pregnancies and
Pregnancy Rates by Outcome: Estimates for the Unites States, 1976-96. National Center for
Health Statistics. Vital Health Statistics. 21(56).
Villanueva, C.M., M. Kogevinas and J.O. Grimalt. 2001a. Drinking water chlorination and adverse
health effects: a review of epidemiological studies. Medicina Clinica. 117(1): 27-35. [Spanish]
Villanueva, C.M., Fernandez, F., Malats, N., Grimalt, J.O., and M. Kogenvinas. 2003b. Meta-analysis of
Studies on Individual Consumption of Chlorinated Drinking Water and Bladder Cancer. Journal
of Epidemiology Community Health. 57:166-173.
Villanueva, C.M., K.P. Cantor, S. Cordier, J.J.K. Jaakkola, W.D. King, C.F. Lynch, S. Porru, and M.
Kogevinas. 2004. Disinfection byproducts and bladder cancer a pooled analysis. Epidemiology.
15(3):357-367.
Vinceti, M., G. Fantuzzi, L. Monici, M. Cassinadri, G. Predieri, and G. Aggazzotti. 2004. A
retrospective cohort study of trihalomethane exposure through drinking water and cancer
mortality in northern Italy. Science of the Total Environment. 330(l-3):47-53.
Vineis. 2004. A self-fulling prophecy: are we underestimating the role of the environment in the gene-
environment interaction research? International Journal of Epidemiology. 33:945-946.
Viscusi, W.K.,W.A. Magat, and J. Huber. 1991. Pricing Environmental Health Risks: Survey
Assessments of Risk-Risk and Risk-Dollar Trade-Offs for Chronic Bronchitis. Journal of
Environmental Economics and Management. 21.
Final Economic Analysis for the Stage 2 DBPR R-18 December 2005
-------
Waller, K., S.H. Swan, G. DeLorenze, and B. Hopkins. 1998. Trihalomethanes in Drinking Water and
Spontaneous Abortion. Epidemiology. 9(2): 134-140.
Waller, K., S.H. Swan, G.C. Windham, and L. Fenster. 2001. Influence of exposure assessment methods
on risk estimates in an epidemiologic study of total trihalomethane exposure and spontaneous
abortion. Journal of Exposure Analysis and Environmental Epidemiology. 11(6):522-31.
Weinberg, H.S., S.W. Krasner, S.D. Richardson, and A.D. Thruston, Jr. 2002. The Occurrence of
Disinfection By-Products (DBFs) of Health Concern in Drinking Water: Results of a Nationwide
DBF Occurrence Study, U.S. Environmental Protection Agency, National Exposure Research
Laboratory, Athens, GA. EPA/600/R-02/068. http://www.epa.gov/athens/publications/
reports/EPA_600_R02_068 .pdf
Weinstein, M.C., Siegel, J.E., Garber, A.M., Lipscomb, J., Luce, B.R., Manning, Jr., W.G., and Torrance,
G.W. Productivity Costs, Time Costs and Health-related Quality of Life: A Response to the
Erasmus Group. Health Economics. 6:505-510. 1997.
Weisel, C.P., H. Kim, P. Haltemeier, and J.B. Klotz. 1999. Exposure Estimates to Disinfection By-
products of Chlorinated Drinking Water. Environmental Health Perspectives. 107(2).
Whynes, O.K. and Neilson, A.R. 1993. Convergent Validity of Two Measures of the Quality of Life,
Health Economics, 2: 229-235.
Wilkins, J.R., III and G.W. Comstock. 1981. Source of drinking water at home and site-specific cancer
incidence in Washington County, Maryland. American Journal of Epidemiology. 114(2): 178-
190.
Windham, G.C., K. Waller, M. Anderson, L. Fenster, P. Mendola, and S. Swan. 2003. Chlorination
Byproducts in Drinking Water and Menstrual Cycle Function. Environmental Health
Perspectives, doi: 10.1289/ehp.5922.
WHO. 2000. World Health Organization, International Programme on Chemical Safety (IPCS).
Environmental Health Criteria 216: Disinfectants and Disinfectant By-products.
Wright, J.M., J. Schwartz and D.W. Dockery. 2003. Effect of trihalomethane exposure on fetal
development. Occupational and Environmental Medicine. 60(3): 173-180.
Wright, J.M., J. Schwartz and D.W. Dockery. 2004. The effect of disinfection by-products and
mutagenic activity on birth weight and gestational duration. Environmental Health Perspectives.
112(8):920-925.
Xu, X., T.M. Marino, J.D. Laskin, and C.P. Weisel. 2002. Pericutaneous absorption of trihalomethanes,
haloacetic acids, and haloketones. Toxicology and Applied Pharmacology. 184(1): 19-26.
Xu, X. and C.P. Weisel. 2003. Inhalation exposure to haloacetic acids and haloketones during
showering. Environmental Science and Technology. 37(3):569-576.
Xu, X. and C.P. Weisel. 2004. Dermal uptake of chloroform and haloketones during bathing. Journal of
Exposure Analysis and Environmental Epidemiology. 1-8.
Xu, X. and C.P. Weisel. 2005. Human respiratory uptake of chloroform and haloketones showering.
Journal of Exposure Analysis and Environmental Epidemiology. 15:6-16.
Final Economic Analysis for the Stage 2 DBPR R-19 December 2005
-------
Yang, C.Y., H.F. Chiu, M.F. Cheng, and S.S. Tsai. 1998. Chlorination of Drinking Water and Cancer
Mortality in Taiwan. Environmental Research. 78:1-6.
Yang, C.Y., B. Cheng, S. Tsai, T. Wu, M. Lin, and K. Lin. 2000. Association between Chlorination of
Drinking Water and Adverse Pregnancy Outcome in Taiwan. Environmental Health
Perspectives. 108(8).
Yang, CY. 2004. Drinking water Chlorination and adverse birth outcomes in Taiwan. Toxicology.
198:249-254.
Young, T.B., D.A. Wolf, and M.S. Kanarek. 1987. Case-Control Study of Colon Cancer and Drinking
Water Trihalomethanes in Wisconsin. International Journal of Epidemiology. 16(190).
Zeegers, M.P., E. Kellen, F. Buntinx, and P.A. van den Brandt. 2004. The association between smoking,
beverage consumption, diet and bladder cancer: a systematic literature review. World Journal of
Urology. 21(6):392-401.
Zender, R., A.M. Bachand, and J.S. Reif 2001. Exposure to tap water during pregnancy. Journal of
Exposure Analysis and Environmental Epidemiology. 11(3):224-230.
Final Economic Analysis for the Stage 2 DBPR R-20 December 2005
------- |