REGULATORY IMPACT ANALYSIS FOR THE PROPOSED
GROUND WATER RULE
APRIL 5,2000
PREPARED FOR:
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF GROUND WATER AND DRINKING WATER
PREPARED BY:
THE CADMUS GROUP, INC.
1901 NORTH FORT MYER DRIVE
SUITE No. 1016
ARLINGTON, VA 22209
WITH
SCIENCE APPLICATIONS INTERNATIONAL CORPORATION
&
ABT ASSOCIATES INC.
EPA CONTRACT No. 68-C-99-206, WORK ASSIGNMENT 1-18
EPA CONTRACT No. 68-C-99-245, WORK ASSIGNMENT 0-16
-------
TABLE OF CONTENTS
1. Executive Summary 1-1
1.1 Introduction 1-1
1.2 Need for the Rule 1-1
1.3 Regulatory Options Considered 1-3
1.4 Baseline Analysis 1-4
1.5 Benefits of the Regulatory Options 1-5
1.6 Cost of the Regulatory Options 1-6
1.7 Economic Impact Analysis 1-6
1.8 Weighing the Benefits and Costs 1-8
2. Need for the Proposal 2-1
2.1 Introduction 2-1
2.2 Public Health Concerns 2-1
2.2.1 Contaminants and Their Health Effects 2-1
2.2.2 Sources of Contaminants 2-5
2.2.2.1 Sources of Ground Water Contamination 2-6
2.2.2.2 Factors Affecting Virus and Bacterial Transport in the
Subsurface 2-6
2.2.2.3 Other Factors that Contribute to the Contamination of
Drinking Water 2-7
2.2.2.4 Contamination of Drinking Water in Distribution Systems 2-8
2.2.2.5 Outbreak Data for Sources and Causes of Contamination 2-9
2.2.3 Exposure to the Contaminants 2-10
2.2.4 Sensitive Subpopulations 2-13
2.3 Current Control and Potential for Improvement 2-14
2.4 Regulatory History 2-16
2.5 Economic Rationale 2-20
2.5.1 Statutory Authority for Promulgating the Rule 2-20
2.5.2 The Economic Rationale for Regulation 2-21
2.6 References 2-23
3. Consideration of Regulatory Options 3-1
3.1 Introduction 3-1
3.2 Option Development Process 3-1
3.3 Summary of Alternatives Considered 3-2
3.3.1 Option 1: Sanitary Survey Only 3-2
3.3.2 Option 2: Sanitary Survey and Triggered Monitoring 3-2
3.3.3 Option 3: Multi-Barrier Approach 3-3
3.3.4 Option4: Across-the-BoardDisinfection 3-4
April 5, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis li
-------
3.4 References 3-5
4. Baseline Analysis 4-1
4.1 Introduction 4-1
4.2 Baseline Profile of Public Groundwater Systems 4-1
4.2.1 Number of Ground Water Systems 4-2
4.2.2 Population Served by Ground Water Systems 4-3
4.2.3 Treatment Profile 4-4
4.3 Baseline Health Effects 4-5
4.3.1 Exposure Assessment 4-8
4.3.2 Hazard Identification 4-13
4.3.3 Sensitive Subgroups 4-14
4.3.4 Risk Characterization 4-15
4.4 References 4-26
5. Benefits Analysis 5-1
5.1 Introduction 5-1
5.2 Structure of the Benefits Analysis 5-1
5.2.1 Human Health Benefits 5-1
5.2.2 Non-Health Benefits Assessment 5-3
5.2.3 Potential Health Risk Associated with Other Contaminants 5-3
5.3 Value of Health Effects With Rule (Acute Impacts) 5-4
5.3.1 Assumptions for Health Effects Modeling of Regulatory Scenarios 5-4
5.3.1.1 Option 1: Sanitary Survey Option 5-5
5.3.1.2 Option 2: Sanitary Survey and Triggered Monitoring Option 5-6
5.3.1.3 Option 3: Multi-Barrier Option 5-7
5.3.1.4 Option 4: Across-the-Board Disinfection Option 5-9
5.3.2 Results of Risk Calculations 5-9
5.3.3 Assumptions for Monetization of Health Benefits (Acute Illnesses) 5-11
5.3.3.1 Unit Cost-of-Illness 5-12
5.3.3.2 Unit Value of a Statistical Life 5-15
5.3.3.3 Reduction in Bacterial-Related Illnesses 5-16
5.3.4 Results of Monetization of Health Benefits (Acute Illnesses) 5-18
5.4 Other Benefits 5-21
5.4.1 Reduced Pain and Suffering 5-21
5.4.2 Reduced Chronic Illness 5-22
5.4.2.1 Type I Diabetes 5-22
5.4.2.2 Chronic Myocarditis 5-24
5.4.3 Non-Health Benefits 5-25
5.4.3.1 Reduced Uncertainty 5-26
5.4.3.2 Costs to Households to Avert Infection 5-26
April 5, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis lil
-------
5.4.3.3 Outbreak Response Costs Avoided 5-27
5.4.4 Benefits From the Reduction of Co-Occurring Contaminants 5-28
5.5 References 5-29
6. Cost Analysis 6-1
6.1 Introduction 6-1
6.2 Costing Methodology 6-1
6.2.1 CostModel Inputs 6-1
6.2.1.1 State Agency Costs 6-2
6.2.1.2 Public Water System (PWS) Costs 6-4
6.2.2 General Structure of the Cost Model 6-9
6.2.2.1 PWS Configuration and Occurrence 6-9
6.2.2.2 Discounting and the Cost of Capital 6-10
6.2.2.3 Calculating Household Costs 6-11
6.3 National Costs 6-11
6.3.1 Comparison of Annual Compliance Costs Across Regulatory Options 6-12
6.3.2 Option 1: Sanitary Survey Only 6-13
6.3.2.1 Total National Costs 6-13
6.3.2.2 C ost of Rule Components 6-13
6.3.3 Option 2: Sanitary Survey and Triggered Monitoring 6-15
6.3.3.1 Total National Costs 6-15
6.3.3.2 Cost of Rule Components 6-16
6.3.4 Option 3: Multi-Barrier Approach 6-17
6.3.4.1 Total National Costs 6-17
6.3.4.2 Cost of Rule Components 6-18
6.3.5 Option 4: Across-the-BoardDisinfection
6.3.5.1 Total National Costs 6-19
6.3.5.2 Cost of Rule Components 6-21
6.4 Household Costs 6-21
6.5 References 6-23
7. Economic Impact Analysis 7-1
7.1 Introduction 7-1
1.1 Regulatory Flexibility Act and Small Business Regulatory Enforcement
Fairness Act 7-1
7.2.1 Definition of Small Entity for the GWR 7-2
1.22 Requirements for the Initial Regulatory Flexibility Analysis 7-2
7.2.3 Small Entity Impacts 7-4
7.2.3.1 Numberof Small Entities Affected 7-4
7.2.3.2 Reporting and Recordkeeping 7-4
7.2.3.3 Small Entity Compliance Costs 7-6
April 5, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis iv
7.
7.
7.
-------
7.2.4 Coordination With Other Federal Rules 7-9
7.2.5 Minimization of Economic Burden 7-10
7.3 Unfunded Mandates Reform Act 7-11
7.3.1 Social Costs and Benefits 7-12
7.3.2 Disproportionate Impacts 7-13
7.3.3 Macro Economic Effects 7-14
7.3.4 Consultations with State, Local, and Tribal Governments 7-15
7.4 Paperwork Reduction Act 7-16
7.5 Protecting Children From Environmental Health Risks and Safety Risks 7-17
7.6 Environmental Justice 7-19
8. Summary of Costs and Benefits 8-1
8.1 Review of Regulatory Options, Costs and Benefits 8-1
8.1.1 Review of Regulatory Options 8-1
8.1.2 Review of National Cost Estimates 8-2
8.1.3 Review of National Benefits Estimates 8-3
8.2 Comparison of Benefits and Costs 8-3
8.2.1 National Benefit-Cost Comparison 8-3
8.2.2 Cost-Effectiveness 8-6
8.3 Uncertainty in Benefit and Cost Estimates 8-7
LIST OF EXHIBITS
Exhibit 1-1. Illnesses Caused by Waterborne Pathogens 1-2
Exhibit 1-2. Regulatory Options and Basic Provisions 1-3
Exhibit 1-3. Ground Water Systems and Population Served 1-4
Exhibit 1-4. Baseline Viral Illness/Deaths in Ground Water Systems 1-5
Exhibit 1-5. Viral Illnesses/Deaths Avoided 1-5
Exhibit 1-6. Annual Quantified Benefit of Avoided Illness and Deaths
(Millions of Dollars, 1999) 1-6
Exhibit 1-7. Comparison of Annual Compliance Costs Across Regulatory
Options 1-8
Exhibit 1-8. Average Annual Household Cost for GWR Options for CWS taking
Corrective Action or Fixing a Significant Defect 1-8
Exhibit 1-9. Summary of Monetized National Benefits and Costs (3% Discount Rate,
million $) 1-10
Exhibit 1-10. Summary of Monetized National Benefits and Costs (7% Discount Rate,
million $) 1-10
Exhibit 1-11. Incremental cost per Case Avoided Across GWR Options 1-11
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
-------
Exhibit 2-1. Illnesses Caused by Waterborne Fecal Viral Pathogens 2-3
Exhibit 2-2. Illnesses Caused by Major Waterborne Bacterial Pathogens 2-4
Exhibit 2-3. Etiology of Waterborne Outbreaks in Ground Water Systems, Community
and Noncommunity Water Systems, 1971-96 2-5
Exhibit 2-4. Waterborne Disease Outbreaks, Ground Water Sources, 1971-96 2-9
Exhibit 2-5. Preliminary Results of the AWWARF Study 2-12
Exhibit 2-6. Sensitive Populations in the United States 2-14
Exhibit 3-1. Regulatory Options and Basic Provisions 3-2
Exhibit 41. Total Number of Ground Water Systems by
System Type and Service Population Category 4-3
Exhibit 4-2. Total Population of Ground Water Systems
by System Type and Size 4-4
Exhibit 43. Percent of Community Ground Water Treatment Facilities Disinfecting by
Service Population Category (Flow-Weighted) 4-5
Exhibit 4-4. Percent of Noncommunity Ground Water
Systems Disinfecting by State 4-6
Exhibit 45. Risk Assessment Process for Pathogens in Drinking Water 4-7
Exhibit 46. Viral Occurrence and Concentration in Source Water 4-11
Exhibit 4-7. Populations Served by Undisinfected Ground Water Systems 4-12
Exhibit 48. EPA Estimates for Exposure Days 4-13
Exhibit 49. Hazard Identification of Viral Pathogens for the GWR Risk Assessment 4-14
Exhibit 4-10. Summary Table of Risk Calculation Factors Used and the Distribution Category
(Variability, Uncertainty, Constant) Used in the
Simulation Analysis 4-19
Exhibit 411. Summary of GWR Baseline Risk Calculations for Undisinfected Systems 4-21
Exhibit 412. Estimates of Baseline Type A Viral Illness and Death 4-23
Exhibit 413. Estimates of Baseline Type B Viral Illness and Death 4-23
Exhibit 414. Health Effects in Sensitive Subgroups as a Percent
of All Illnesses and Deaths 4-24
Exhibit 415. Distribution of Annual Individual Risks of Illness by Type of System 4-25
Exhibit 5-1. Overview of GWR Benefits 5-1
Exhibit 5-2. GWR Health Benefits Assessment Framework 5-2
Exhibit 5-3. Estimated Contamination Reductions for GWR Options 5-5
Exhibit 5-4. Remaining Viral Illnesses/Deaths for Regulatory Scenarios 5-10
Exhibit 5-5. Reduction in Illnesses/Deaths for Regulatory Scenarios 5-10
Exhibit 5-6. Monetization of Health Effects 5-12
Exhibit 5-7. Classification for Clinical Severity in Type B (Echovirus) Illnesses 5-13
Exhibit 5-8. Weighting for Clinical Severity in Type B (Echovirus) Illnesses 5-14
Exhibit 5-9. Type A (Rotavirus) Unit Cost-of-Illness
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
VI
-------
Estimates by Victim Age and Health Status 5-14
Exhibit 5-10. Type B (Enterovirus) Unit Cost-of-fllness Estimates by
Victim Age and Illness Severity 5-15
Exhibit 5-11. Summary of Clinical Characteristics of
Selected Bacterial Waterborne Pathogens 5-17
Exhibit 5-12. Health Benefits for Regulatory Scenarios ($Millions) 5-19
Exhibit 5-13. Distribution of Mean Total Benefits: By Pathogen Type 5-19
Exhibit 5-14. Distribution of Mean Total Benefits: By Health Status 5-20
Exhibit 5-15. Distribution of Mean Total Benefits (Viral Only): By System Size 5-20
Exhibit 5-16. Distribution of Mean Total Benefits: By System Type 5-21
Exhibit 5-17. Annual Number of People with Selected Disease 5-22
Exhibit 5-18. Case Study of Outbreak Costs-1984 Lucerne County Outbreak 5-27
Exhibit 6-1. Components Included in Each Regulatory Options 6-1
Exhibit 6-2. Assignment of Components' Costs 6-2
Exhibit 6-3. Examples of State Administrative Activities 6-3
Exhibit 6-4. Comparison of National Compliance Costs
Across Regulatory Options 6-12
Exhibit 6-5. Comparison of Mean Annual Compliance Costs Across Regulatory
Options by System Size Category 6-12
Exhibit 6-6. Comparison of Mean Annual Compliance Costs Across Regulatory
Options by System Type 6-13
Exhibit 6-7. Number of Affected Systems by Rule Component
(High Corrective Action/Low Significant Defect Scenario)
Option 1: Sanitary Survey Only 6-14
Exhibit 6-8. Total National Costs: Option 1-Sanitary Survey Only 6-14
Exhibit 6-9. National PWS Compliance Costs of the GWR by Rule Component
Option 1: Sanitary Survey Only 6-15
Exhibit 6-10. Number of Affected PWSs by Rule Component
(High Corrective Action/Low Significant Defect Scenario)
Option 2: Sanitary Survey and Triggered Monitoring 6-16
Exhibit 6-11. Total National Compliance Costs 6-16
Exhibit 6-12. National PWS Compliance Costs of the GWR by Rule Component
(High Corrective Action/Low Significant Defect Scenario)
Option 2: Sanitary Survey and Triggered Monitoring 6-17
Exhibit 6-13. Number of Affected Systems by Rule Component
(High Corrective Action/Low Significant Defect Scenario)
Option 3: Multi-Barrier Approach 6-18
Exhibit 6-14. Total National Compliance Costs Option 3: Multi-Barrier Approach 6-18
Exhibit 6-15. National PWS Compliance Costs of the GWR by Rule Component
(High Corrective Action/Low Significant Defect Scenario)
Option 3: Multi-Barrier Approach 6-19
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
VU
-------
Exhibit 6-16. Number of Affected Systems by Rule Component
(High Corrective Action/Low Significant Defect Scenario)
Option 4: Across-the-Board Disinfection 6-20
Exhibit 6-17. Total National Compliance Costs Option 4: Across-the-Board
Disinfection 6-20
Exhibit 6-18. National PWS Compliance Costs of the GWR by Rule Component
(High Corrective Action/Low Significant Defect Scenario)
Option 4: Across-the-Board Disinfection 6-21
Exhibit 6-19. Mean Annual Household Costs of the GWR Across Regulatory Options 6-22
Exhibit 6-20. Mean Annual Household Costs of the GWR Across Regulatory Options
for Public and Private CWSs Taking Corrective Action
or Fixing Significant Defects 6-22
Exhibit 7-1. Number of Small Systems Effected by the GWR 7-5
Exhibit 7-2. Comparison of Mean Baseline and After-Rule CWS Expenses 7-6
Exhibit 7-3. Comparison of CWS Baseline and Post-Compliance Expenses
(Systems Serving <100 People) 7-7
Exhibit 7-4. Comparison of CWS Baseline and Post-Compliance Expenses
(Systems Serving 101-500 People) 7-8
Exhibit 7-5. Comparison of CWS Baseline and Post-Compliance Expenses
(Systems Serving 501-1,000 People) 7-8
Exhibit 7-6. Comparison of CWS Baseline and Post-Compliance Expenses
(Systems Serving 1,001-3,300 People) 7-9
Exhibit 7-7. Comparison of CWS Baseline and Post-Compliance Expenses
(Systems Serving 3,301-10,000 People) 7-10
Exhibit 7-8. Annual Compliance Cost Impacts by PWS Ownership 7-14
Exhibit 7-9. Summary of the Ground Water Rule Total Respondents, Responses,
Burden, and Costs for PWSs and States 7-17
Exhibit 7-10. Viral Illnesses and Deaths Avoided In Children
Across Regulatory Alternatives 7-18
Exhibit 7-11. Viral Illnesses Avoided in Immuno-Compromised Persons
Across Regulatory Alternatives 7-19
Exhibit 8-1. GWR Regulatory Options and Main Requirements 8-1
Exhibit 8-2. Summary of National Benefits and Costs
(Using 3 Percent Discount Rate) 8-2
Exhibit 8-3. Summary of National Benefits and Costs
(Using 7 Percent Discount Rate) 8-2
Exhibit 8^. Incremental Comparison of National Costs and Benefits (7% discount rate) 8-4
Exhibit 8-5. Net Benefits of Each Regulatory Option
by System Size Category (Using 3 Percent Discount Rate (million$) 8-5
Exhibit 8-6. Net Benefits of Each Regulatory Option
by System Size Category (Using 7 Percent Discount Rate (million$) 8-5
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
Vlll
-------
Exhibit 8-7. Cost per Case Avoided Across GWR Options 8-6
LIST OF APPENDICES
Appendix A. Risk Assessment Inputs and Results
Appendix B-l. Results from Benefits Valuation; Upper Bound Drinking Water Consumption
Distribution (All Sources, Consumers Only)
Appendix B-2. Results from Benefits Valuation; Lower Bound Drinking Water Consumption
Distribution (Community Water Systems, All Respondents)
Appendix C. Inputs to Cost Model
Appendix D. Results of Cost Analysis
April 5, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis
-------
1. Executive Summary
1.1 Introduction
This document presents the analysis of the impacts of the proposed Ground Water Rule (GWR).
The proposed GWR has been developed by the Environmental Protection Agency (EPA) working with
States and other interested stakeholders. The primary goal of the proposed GWR is to improve public
health by identifying public ground water systems that are now, or are likely to become, fecally
contaminated, and to insure adequate measures are taken to remove or inactivate pathogens in drinking
water provided to the public by these systems. This document provides: a description of the need for
the rule, a description of the regulatory options, baseline information on ground water systems,
estimates of the monetized benefits and costs of the proposed rule, a description of additional
unquantified and nonmonetized benefits, analysis of the economic impact of the rule, and a comparison
of the overall benefits and costs of the rule alternatives.
1.2 Need for the Rule
EPA has developed the proposed GWR in fulfillment of its responsibility established under Section
1412(b)(8) that EPA develop regulations specifying the use of disinfectants for ground water systems
as necessary.
EPA believes that there is a substantial likelihood that fecal contamination of ground water supplies
is occurring at frequencies and levels that present a public health concern. Fecal contamination refers to
the contaminants, particularly the microorganisms, contained in human or animal feces. These
microorganisms may include bacterial and viral pathogens that can cause illnesses, and in some cases
death, in the individuals that consume them.
Fecal contamination is introduced to ground water from a number of sources including, septic
systems, leaking sewer pipes, landfills, sewage lagoons, cesspools, and storm water runoff.
Microorganisms can be transported with the ground water as it moves through an aquifer. The distance
that the microorganisms can be transported through a ground water aquifer depends on a number of
factors including, the nature of the microorganism, temperature, and soil properties. For example,
protozoan organisms are much larger in size than bacteria and viruses and are therefore much less
likely to be able to move through the soil matrix. The transport of microorganisms to wells or other
ground water system sources can also be affected by poor well construction (e.g., improper well seals)
that can result in large, open conduits for fecal contamination to pass unimpeded into the water supply.
Recent studies of public water system supply wells show that there are a number of ground water
supplies that contain fecal contamination. The American Water Works Association Research
Foundation (AWWARF) Study (Abbaszadegan, et. al, 1999) collected data from more than 400
public water supply wells located in 35 States, and is perhaps the most representative study of public
ground water supplies to date. This study found that almost 5 percent of the wells contained infectious
12 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
enteroviruses, almost 10 percent of the wells contained bacteria and almost 15 percent of the wells
contained rotavirus fragments that may or may not be capable of causing infections.
Waterborne pathogens contained in fecally contaminated water can result in a variety of illnesses
that range in the severity of their outcomes from mild diarrhea to kidney failure or heart disease. Exhibit
1-1 presents a list of the illnesses that are caused by pathogenic viruses and bacteria in fecally
contaminated ground water. The populations that are particularly sensitive to waterborne and other
pathogens include, infants, young children, pregnant and lactating women, the elderly and the chronically
ill. These individuals may be more likely to become ill as a result of exposure to the pathogens, and are
likely to have a more severe illness.
Exhibit 1-1. Illnesses Caused by Waterborne Pathogens
Viral Waterborne Illnesses
Bacterial Waterborne Illnesses
Gastroenteritis (diarrhea, stomach cramps etc.)
Myocarditis (heart disease)
Meningitis
Diabetes
Hepatitis
Paralysis
Gastroenteritis (diarrhea, stomach cramps etc.
Hemolytic uremic syndrome(kidney failure)
Cholera
Legionnaires Disease
Many ground water systems currently practice disinfection to inactivate or remove the pathogens in
ground water prior to distributing the water to their customers. However, data collected by the Centers
for Disease Control and Prevention (CDC) and EPA indicate that almost as many waterborne disease
outbreaks were reported between 1971 and 1996 in systems with disinfection treatment that was
inadequate or interrupted (134 outbreaks) as were reported in the same period among ground systems
that did not disinfect (163 outbreaks). The CDC outbreak data also indicate that fecal contamination
may be introduced into a public water system by the distribution system itself. Between 1971 and
1996, 49 reported outbreaks of the waterborne disease occurred as a result of distribution system
contamination. The reported outbreaks probably represent a small fraction of the total number of
waterborne disease outbreaks because reporting of outbreaks is voluntary, and not all States have
outbreak surveillance systems.
Currently the Total Coliform Rule is the only federal drinking water regulation that directly governs
the presence of microbes in public ground water systems. The rule applies to all public water systems,
and requires systems to collect samples from their distribution systems and test for the presence of
coliform bacteria. Total coliform monitoring is used to screen for fecal contamination, determine the
effectiveness of treatment, and determine the integrity of the distribution system. The frequency of total
coliform sampling depends upon the number of people served by the system and the system type.
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
1-3
-------
1.3 Regulatory Options Considered
EPA has been working with a regulatory workgroup, stakeholders and other interested parties to
develop regulatory options to address fecal contamination of ground water systems. Four options have
been developed through this process. The first three regulatory options Sanitary Survey, Sanitary
Survey and Triggered Monitoring, and the Multi-Barrier options build successively upon one another
adding mechanisms to detect and address ground water systems at risk of fecal contamination. The
fourth option, Across-the-Board Disinfection, does not include a component to target the systems that
are at risk, but requires all systems to install treatment to remove or inactivate microbial contamination.
Exhibit 1-2 lists the regulatory options and their components.
Exhibit 1-2. Regulatory Options and Basic Provisions
Provisions
Sanitary Survey
Triggered source water (microbial) monitoring1
Hydrogeologic sensitivity assessment and routine
source water monitoring2
All systems install/upgrade and maintain treatment
Regulatory Options
1
Sanitary Survey
/
2
Sanitary Survey and
Triggered Monitoring
/
/
3
Multi-Barrier
/
/
/
4
Across-the-
Board
Disinfection
/
/
1 Triggered by a total coliform-positive sample in the distribution system
2 For those systems determined to be hydrogeologically sensitive
The first regulatory option would require States and other primacy agencies to conduct a sanitary
survey of community ground water systems once every three years and, noncommunity water systems,
once every five years. Most States already perform sanitary surveys, but with wide variation in
frequency and stringency. This requirement would increase their frequency for most CWSs, specify
minimum sanitary survey elements, and ensure that systems correct significant deficiencies. The sanitary
survey reviews all aspects of the ground water system including the source, treatment, storage, pumps,
distribution system, monitoring records, operator certification, and management. States would require
systems to correct any significant deficiencies identified in the survey or to install disinfection treatment.
The second regulatory option would incorporate triggered monitoring of ground water systems
source water with the sanitary survey required under the first option. Source water monitoring would
be triggered by the detection of total coliform in the samples that systems collect for compliance with
the Total Coliform Rule. When a TCR sample in the distribution system is positive for total coliform,
the ground water system would be required to sample its sources within 24 hours and analyze the
sample for the presence of one of three fecal indicator organisms. Systems that find fecal indicators in
1-4
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
their source water would be required to eliminate the contamination from the well, obtain a new source
water, or provide 4-log inactivation or removal of viruses in the source water.
The third regulatory option combines the components of the first two options with routine
monitoring of sources that are sensitive to fecal contamination. The sensitivity of a well or other ground
water source to contamination would be determined by the State or other primacy agency based upon
hydrogeologic information which the State may have already compiled in its source water assessment or
from its well construction approval process. States would determine if the well is drawing water from
aquifers that are sensitive. At a minimum, States would consider wells in karst, fractured bedrock or
gravel cobble aquifers to be sensitive unless there was a hydrogeologic barrier present which prevents
the movement of microbial contamination. Systems determined to be sensitive would collect monthly
samples of their source water, and test these samples for the presence of one of three fecal indicator
organisms. If any of these tests are positive for the fecal indicator organism, the system would be
required to eliminate the contamination, obtain a new source of water, or provide disinfection treatment
that can achieve 4-log inactivation or removal of viruses in the source water.
The fourth regulatory option requires all ground water systems to disinfect their source waters
regardless of the potential risk of fecal contamination. The systems would be required to achieve 4-log
inactivation or removal of viruses. States would be required to conduct sanitary surveys as required
under the first three options. However, additional emphasis in the survey would be placed on ensuring
that the disinfection treatment is properly operated and maintained.
1.4 Baseline Analysis
There are approximately 156,000 public ground water systems in the United States that include
community water systems (systems serving year-round residents), nontransient noncommunity water
systems (e.g., systems serving factories, schools, office buildings, etc.) and transient noncommunity
water systems (e.g., systems serving restaurants, rest stops, etc.). Exhibit 1-3 lists the total number of
ground water systems and populations served by each system type. Ninety-nine percent of the ground
water systems are considered to be small, because they each serve fewer than 10,000 people.
Exhibit 1-3. Ground Water Systems and Population Served
Number of Systems
Population Served
Community
43,906
88.7 million
Nontransient
Noncommunity
19,322
5.3 million
Transient
Noncommunity
93,618
14.9 million
Total
156,846
108.9 million
EPA has prepared a risk assessment to estimate the number of viral illnesses and deaths resulting
from fecal contamination of ground water systems. The risk assessment estimates the number of
illnesses and deaths from rotavirus and enterovirus viruses (Type A and Type B viruses, respectively).
Type A viruses are highly infective but produce mild health effects, while Type B viruses are moderately
infective but with moderate to severe health effects. Exhibit 1-4 presents the estimated number of
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
1-5
-------
illnesses and deaths attributable to the presence of these viruses in public ground water systems with
current levels of treatment. There are uncertainties associated with the key assumptions made to
prepare the estimates including the number of systems with viral contamination of their source water or
distribution system, and the concentration of the viral pathogens in the contaminated water. The
uncertainty in these estimates is indicated by the 10th and 90th percentile values shown in parentheses
(indicating a 10 percent chance the estimate falls below the 10th percentile and a 10 percent chance it
falls above the 90th percentile).
Exhibit 1-4. Baseline Viral Illness/Deaths in Ground Water Systems
Mean annual illnesses
[10th and 90*1 percentile estimates]
Mean annual number of deaths
[10th and 90*1 percentile estimates]
Type A Viruses
133,498
(132,879-134,133)
1
(1-1)
Type B Viruses
34,157
(33,062-35,227)
14
(14-15)
These estimates also do not include the deaths and illnesses associated with bacterial contamination
of ground water systems. Data reported to CDC for waterborne disease outbreaks indicate that for
every five outbreak illnesses caused by virus (or of unknown etiology thought to be viruses) there is one
bacterial outbreak illness. Thus the number of baseline illnesses shown in Exhibit 1-4 are increased by
20 percent to account for bacterial pathogens in ground water systems.
1.5 Benefits of the Regulatory Options
The regulatory options under consideration for the GWR are expected to reduce viral illness and
deaths by reducing the public's exposure to the pathogens. Exhibit 1-5 presents the estimated
reductions expected for each rule alternative considered.
Exhibit 1-5. Viral Illnesses/Deaths Avoided
Viral Illnesses
Avoided
Viral Deaths
Avoided
Sanitary
Survey
13,596
1
Sanitary Surveys
Triggered
Monitoring
83,502
8
Multi-Barrier
96,305
9
Across-the-Board
Disinfection
132,129
12
The monetized benefit of the avoided illnesses is estimated using cost-of-illness estimates for
rotavirus (Type A) and enterovirus (Type B) of $158 per illness to $19,711 per illness, depending upon
the age of the patient, immune status, and severity of illness. The monetized benefit from viral deaths
avoided is estimated using a "value of a statistical life" estimate of $6.3 million (1999 dollars). The
monetized benefits of reduced bacterial illnesses and deaths are estimated by employing a simple ratio
assumption in which the benefits estimated for reduced viral infections were increased by an additional
1-6
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
20 percent to account for bacterial infection reduction benefits. This ratio is based on CDC data that
suggest that the ratio between waterborne bacterial illness and viral illness in ground water systems is
0.2. Exhibit 1-6 presents the estimated benefits associated with reduced viral and bacterial illnesses
and deaths in each regulatory option.
Exhibit 1-6. Annual Monetized Benefit of Avoided Illness and Deaths
(Millions of Dollars, 1999)
Illness Avoided
Mean
[10th and 90*1
percentile estimates]
Deaths Avoided
Mean
[10th and 90*1
percentile estimates]
Total Quantified
Benefit Mean
[10th and 90th
percentile
estimates]
Sanitary Survey
$22
[$7 to $38]
$11
[$2 to $20]
$33
[$9 to $58]
Sanitary Survey
and Triggered
Monitoring
$120
[$101 to $140]
$58
[$47 to $68]
$178
[$147 to $209]
Multi-Barrier
$139
[$11 5 to $163}
$66
[$54 to $79]
$205
[$169 to $242]
Across-the-Board
Disinfection
$192
[174 to $210]
$91
[$81 to $101]
$283
[$255 to $311]
EPA recognizes that, in addition to the benefits associated with reductions in acute illness and death
from viral and bacterial infection, the proposed GWR would provide chronic heath benefits as well as
non-health benefits. For example, medical and epidemiological literature identifies several potential
chronic diseases resulting from illnesses caused by enteroviruses (e.g., heart disease, diabetes, post-
viral fatigue syndrome, and pancreatitis). The strongest evidence for an association between enteroviral
infection and chronic decease appears to exist for the development of diabetes and myocarditis
(inflammation of the muscular walls of the heart). In addition, non-health benefits may result from
overall system improvements (e.g., upgrades to distribution systems, increased efficiencies, increased
frequency/intensity of process surveillance), from improved risk perception of drinking water quality, or
from avoided outbreak response costs. EPA was able to quantify neither the chronic health benefits
nor the non-health benefits in dollar terms. The benefits gained, however, are not inconsequential,
merely unquantifiable.
1.6 Cost of the Regulatory Options
To estimate the cost of the four regulatory options, the impact on both public water systems and on
States was considered. With all rule options, a greater proportion of the regulatory burden is placed on
those systems that do not currently disinfect to a 4-log virus inactivation. Other system costs vary with
the rule option and may include costs for monitoring, correcting significant deficiencies, or installing
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
1-7
-------
treatment. Depending on the option, States will incur costs for an increase in sanitary survey
requirements and frequency, for conducting sensitivity assessments, and for follow-up inspections.
Both systems and States will incur implementation costs.
The annual cost of the four rule options range from $76 million to $866 million using a 7 percent
discount rate (Exhibit 1-7). Using a 3 percent discount rate, the costs range from $73 million to $777
million. For the first three options, the costs increase as more methods are added for identifying fecally
contaminated wells and wells sensitive to fecal contamination. However, the costs of these methods
(e.g., hydrogeologic assessment, triggered and routine monitoring) are minor compared to the costs of
correcting fecally contaminated wells. The fourth option of Across-the-Board Disinfection is the most
costly because it requires all systems to install treatment regardless of actual or potential fecal
contamination.
Exhibit 1-7. Comparison of Annual Compliance Costs
Across Regulatory Options
Option/Regulatory Scenario
Option 1 : Sanitary Survey Only
Option 2: Sanitary Survey and
Triggered Monitoring
OptionS: Multi-Barrier Approach
Option 4: Across-the-Board
Disinfection Option
Mean Compliance Costs ($Millions)
[10th and 90th percentile estimates]
At 3%
$72.7
($71.1 to $74.4)
$157.6
($152.8 to $162. 4)
$182.7
($177.0 to $188. 4)
$777.1
($743.9 to $810. 3)
At 7%
$76
($74.3 to $77.7)
$168.5
($163.0 to $174.0)
$198.6
($191. 7 to $205. 5)
$866.0
($822.7 to $909. 4)
In addition to the corrective action costs and the costs to address significant deficiencies, EPA
estimated system monitoring costs and start-up costs. All options have additional monitoring
requirements, although they vary depending on rule option. The Agency also accounted for a system's
start-up cost to comply with the various rule options. These costs include time to read and understand
the rule; mobilization and planning; and staff training. The Agency also estimated system costs for
reporting and recordkeeping of any positive source water samples.
Depending on the option, States would face increased costs from the incremental difference in
sanitary survey requirements and frequency, from conducting one-time sensitivity assessments, and from
tracking monitoring information. States would also incur start-up costs and annual costs for data
management and training.
Household costs for systems that take corrective action to fix a significant defect or to address fecal
contamination are presented in Exhibit 1-8. The average increase in annual household costs for these
systems is between $2.45 to $3.86 for the first three options. The Across-the-Board Disinfection
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
results in the highest average annual household cost of $19.37. However, household costs increase
disproportionately across all options for those households served by the smallest sized systems. This
occurs because they serve fewer households to share the system's fixed costs. For example, under the
Multi-Barrier option, household costs would increase by approximately $5 per month for those served
by the smallest size systems (<100 households) while those served by the largest size systems
(>100,000 households) would face only a $0.02 increase in monthly household costs.
Although EPA estimated the cost of all the rule's components for drinking water systems and
States, there are some costs that the Agency did not monetize. These nonmonetized costs result from
uncertainties surrounding rule assumptions and from modeling assumptions.
Exhibit 1-8. Average Annual Household Cost for GWR Options for CWS taking
Corrective Action or Fixing a Significant Defect
SIZE CATEGORIES
<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
TOTAL
Sanitary Survey
Option
$29.86
$11.23
$5.72
$2.99
$1.39
$0.62
$0.30
$0.32
$2.45
Sanitary Survey and
Triqqered Monitoring Option
$67.19
$15.02
$6.29
$2.91
$1.45
$0.59
$0.70
$0.20
$3.34
Multi-Barrier Option
$62.48
$18.95
$625
$3.39
$2.74
$0.62
$1.01
$027
$3.86
Across-the-Board
Disinfection Option
$191.87
$81.38
$38.79
$23.45
$16.78
$4.87
$10.37
$1.66
$19.37
1.7 Economic Impact Analysis
As part of the rule promulgation process, EPA is required to perform a series of distributional
analyses that address the potential regulatory burden placed on entities that are affected by the various
requirements of this proposed rule. EPA analyzed potential GWR impacts including those to small
businesses, States, Tribes, local governments, and the private sector. Impacts on small business were
analyzed as part of the requirements outlined in the Regulatory Flexibility Act as amended by The Small
Business Regulatory Enforcement Fairness Act. The Agency also conducted an Unfunded Mandate
Reform Act analysis because the proposed rule is expected to have an annual impact of at least $100
million on State, local, and Tribal governments in aggregate and on the private sector. As required by
the Safe Drinking Water Act, a preliminary analysis of how this regulation would affect each system's
capacity was also completed.
The Agency also estimated the effects of this proposed rule on children's health and environmental
justice considerations. This rule is expected to disproportionally protect children from illness and death
that result from ingestion of fecally contaminated ground water. For example, children less than five
years of age make up only 7.2 percent of the U.S. population, while they receive 13 percent of the
benefits from the Multi-Barrier option's reduction in Type B viral illness (lower infectivity viruses with
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 1-9
-------
higher costs-of-illness). As required in Executive Order 12898 regarding environmental justice, the
proposed GWR will equally protect the health of all people served by public ground water systems,
regardless of income or minority status.
1.8 Weighing the Benefits and Costs
Both costs and benefits associated with the proposed GWR rise with the successively more
stringent regulatory options. With regards to monetized costs and benefits, only the Sanitary Survey
and Triggered Monitoring option and the Multi-Barrier option have positive net benefits. For the
Sanitary Survey and Triggered Monitoring Option, the monetized net benefits are estimated at $20.3
million using a 3 percent discount rate $9.4 million using a 7 percent discount rate. For the Multi-
Barrier Option, the monetized benefits are estimated at $22.3 million using a 3-percent discount rate
and $6.4 million using a 7 percent discount rate.
Of the remaining two options, the Across-the-Board Disinfection option has the largest negative net
benefit at negative $494.0 million (3 percent) and negative $583 million (7 percent). The net benefits
for the Sanitary Survey option are negative $40.2 million (3 percent) and negative $43.5 (7 percent).
Exhibit 1-9. Summary of Monetized National Benefits and Costs
(3% Discount Rate, million $)
Regulatory Option
Sanitary Survey Option
Sanitary Survey and Triggered Monitoring
Option
Multi-Barrier Option
Across-the-Board Disinfection Option
Benefit
$32.5
$177.9
$205.0
$283.1
Cost
$72.7
$157.6
$182.7
$777.1
Net Benefit
($40.2)
$20.3
$22.3
($494.0)
Exhibit 1-10. Summary of Monetized National Benefits and Costs
(7% Discount Rate, million $)
Regulatory Option
Sanitary Survey Option
Sanitary Survey and Triggered Monitoring
Option
Multi-Barrier Option
Across-the-Board Disinfection Option
Benefit
$32.5
$177.9
$205.0
$283.1
Cost
$76.0
$168.5
$198.6
$866.0
Net Benefit
($43.5)
$9.4
$6.4
($582.9)
It is important to remember that there are costs and benefits from the proposed GWR
that are not included in the monetized benefits presented above. For example, the
proposed GWR may provide benefits from associated with reductions in chronic illnesses
1-10
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
caused by enteroviruses (e.g., heart disease, diabetes, post-viral fatigue syndrome, and
pancreatitis). There are also non-health benefits that could not be monetized (e.g.,
upgrades to distribution systems, increased efficiencies, increased frequency/intensity of
process surveillance). Some costs, such as land acquisition, are not included in this RIA.
EPA does not have the data needed to quantify these costs benefits, but if they were, net
benefits of this rule would be greater than those listed in Exhibits 1-9 and 1-10.
In addition to the benefit cost analysis, the Agency examined the cost-effectiveness of
each option. As shown in Exhibit 1-11, the Sanitary Survey and Triggered Monitoring
option achieves the lowest incremental cost per case of illness avoided while the Across-
the-Board Disinfection option costs almost an additional $12,000 per case avoided
(relative to the Multi-Barrier option). The Multi-Barrier option has the second lowest
incremental cost per case of illness avoided with a cost of $1,954 per case that is only
Exhibit 1-11
Incremental Cost per Case Avoided Across GWR Options
(Relative to Next Least Stringent Option)
*lb,UUU
VI
(/)
0)
"o
0)
*2 $10,000
CO
o
5
1i>
o
o
c
Ol
£! $4,000
o
c
i- -I i i i
Sanitary Survey Sanitary Survey and Multi-Barrier Across-the-Board
Triggered Monitoring Disinfection
Regulatory Option
slightly higher than the Sanitary Survey and Triggered Monitoring option incremental cost
of $1,123.
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
1-11
-------
2. Need for the Proposal
2.1 Introduction
This document analyzes the impacts of the proposed Ground Water Rule (GWR). EPA intends
the GWR to address microbial contamination of ground water-supplied drinking water systems in
accordance with the Safe Drinking Water Act (SDWA) of 1974, as amended in 1986 and again in
1996. This regulatory impact analysis (RIA) provides background information on the rule, summarizes
the key components, discusses options to the rule, and estimates costs and benefits to the public and to
State governments. The RIA will be made available in conjunction with the proposed GWR.
The 1986 SDWA amendments directed EPA to establish national primary drinking water
regulations requiring disinfection as treatment for the inactivation of microbiological contaminants for all
public water systems, including systems supplied by ground water sources. The 1996 SDWA
amendments changed the mandate to require disinfection for ground water sources "as necessary." The
1996 amendments establish a statutory deadline of May 2002. EPA, however, intends to finalize the
GWR in the year 2000 to match implementation of other drinking water regulations and programs, such
as the Stage 1 Disinfection Byproducts Rule, the Filter Backwash Recycling Rule (FBRR), the Radon
Rule, and the Source Water Assessment Program (SWAP).
2.2 Public Health Concerns
This section describes the public health concerns to which the proposed GWR is directed. The
contaminants, both bacterial and viral, and their health effects are explained first. The following sub-
sections address sources, means of exposure, and effects of that exposure on sensitive populations.
Finally, this section ends with a discussion of current controls used to address these concerns.
2.2.1 Contaminants and Their Health Effects
EPA is concerned about any fecally-contaminated ground water supply as well as any ground
water system at risk for contamination. Fecal contamination is a general term that includes all of the
bacteria and viruses found in feces. These bacteria and viruses may be non-pathogenic, which do not
cause disease but serve as indicators of other bacteria or viruses, or pathogenic, which are disease-
causing. The types of non-pathogenic bacterial and viral micro-organisms found in feces include many
strains of Escherichia coli, other coliform bacteria, and the male-specific and somatic coliphage, which
are viruses that infect coliform bacteria. Because of their widespread presence in fecal material, the
coliform bacteria and coliphage viruses sometimes are used as indicators of fecal contamination. Total
coliforms (TC) include many coliform bacteria that are free living in the environment as well as fecal
bacteria. Fecal coliforms are bacteria more commonly found in human feces. Other bacteria that are
used as indicators of fecal contamination include the fecal streptococci (enterococci) and Clostridium
perfringens, a spore forming anaerobic organism that can persist for long periods of time in the
environment.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 2-1
-------
Examples of common fecal pathogens include the enteroviruses (e.g., echoviruses and
coxsackieviruses), rotavirus, hepatitis A virus (HAV), and bacteria such as Salmonella, Shigella, and
Campylobacter. Unlike bacterial pathogens, viruses cannot reproduce outside the host, although they
can survive and remain infectious. Also, with a few exceptions, viruses that can infect human cells
typically cannot infect the cells of other animals and vice versa. This contrasts with the bacterial
pathogens that sometimes can infect more than one host.
Enteric viral and bacterial microorganisms are excreted in the feces of humans and animals.
The word enteric (relating to the intestines, or more specifically, the human gut) indicates that the natural
habitat of these microorganisms is the intestinal tract of animals and humans (Domingue, 1983). The
enteric microorganisms, sometimes referred to as intestinal microflora, can survive in sewage and
leachate derived from septic tanks (septage). Therefore, when sewage and septage are released into
the environment, they are sources of intestinal microflora and potential sources of viral and bacterial
pathogens. Some human bacterial pathogens also are shed in the feces of infected animals.
Some enteric viruses may infect cells in tissues outside the gut, causing mild or serious
secondary effects ("sequela") such as myocarditis, conjunctivitis, meningitis or hepatitis. There is also
increasing evidence that the human body reacts to foreign invasion by viruses in ways that may also be
detrimental. For example, one hypothesis for the cause adult onset (Type 1) diabetes is that the human
body, responding to coxsackie B5 virus infection, attacks pathogens cells in an autoimmune reaction as
a result of similarities between certain pancreas cells and the viruses (Solimena and De Camilli, 1995).
Once enteric pathogens are ingested, the likelihood of infection varies depending on the
pathogenicity of the organism, since some pathogens are more infective at low doses than others. Once
a person becomes infected, the likelihood and severity of symptomatic illness also varies with the type
of pathogen, and with the level of acquired immunity and general resistance of the person.
When humans are infected by viruses that infect gut cells, these viruses become capable of
reproducing. As a result, humans shed viruses in stool, typically for a period of a few weeks to a few
months. Regardless of whether individuals infected by the waterborne pathogen have actual symptoms
of illness, such as diarrhea, they are still shedding the virus and this may result in the infection of other
people. This is called secondary spread and it can result from person-to-person contact or contact
with contaminated surfaces. As a result, waterborne viral pathogens may infect others via a variety of
routes. Examples of illnesses caused by known or suspected waterborne viral pathogens are shown in
Exhibit 2-1.
2-2 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
Exhibit 2-1. Illnesses Caused by Waterborne Fecal Viral Pathogens
Enteric Virus
Poliovirus
Coxsackievirus A
Coxsackievirus B
Echovirus
Norwalk virus and other caliciviruses
Hepatitis A virus
Hepatitis E virus
Small round structured viruses (probably
caliciviruses)
Rotavirus
Enteric Adenovirus
Astrovirus
Illness
Paralysis
Meningitis, fever, respiratory disease
Myocarditis, congenital heart disease, rash, fever,
meningitis, encephalitis, pleurodynia, diabetes
melitis, eye infections
Meningitis, encephalitis, rash, fever, gastroenteritis
Gastroenteritis
Hepatitis
Hepatitis
Gastroenteritis
Gastroenteritis
Respiratory disease, eye infections, gastroenteritis
Gastroenteritis
Bold highlights indicate diseases directly caused by the enteric virus; other illnesses are secondarily associated
with the virus.
Source: 1994 Encyclopedia of Microbiology
Some waterborne bacterial pathogens cause disease by rapid growth and dissemination (e.g.,
Salmonella) while others primarily cause disease via toxin production (e.g., Shigella, E. coll 0157,
Campylobacter jejuni). Campy lobacter, E. coll and Salmonella have a host range that includes
both animals and humans; Shigella is associated only with humans (Geldreich, 1996).
Most of the waterborne bacterial pathogens cause gastrointestinal illness, but some can cause
other severe illnesses as well. For example, Legionella causes Legionnaires Disease, a form of
pneumonia that has a fatality rate of about 15 percent. It can also cause Pontiac Fever, which is much
less severe than Legionnaires Disease, but which causes illness in almost everyone exposed. Several
strains of E. coli can cause severe disease, including kidney failure.
Secondary, or opportunistic pathogens such as Pseudomonas, usually cause illness only in
immunocompromised persons, or in other sensitive subpopulations, such as the very young or the
elderly. Some of the opportunistic pathogens can cause symptoms other than gastrointestinal illness,
e.g., meningitis, septicemia, pneumonia (Rusin et al, 1997). Other diseases such as bacterial enteritis
caused by Salmonella, Shigella, Campy lobacter jejuni, and Clostridium dificile occur with greater
frequency and severity in immunocompromised persons, e.g., AIDs patients (Framm and Soave,
1997). Some opportunistic bacterial pathogens can colonize and grow in the biofilm growth in water
system distribution lines. Examples of illnesses caused by waterborne bacterial pathogens are shown in
Exhibit 2-2.
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
2-3
-------
Exhibit 2-2. Illnesses Caused by Major Waterborne Bacterial Pathogens
Bacterial pathogen
Campylobacterjejuni
Shi gel la species
Salmonella species
Vibrio cholerae
Escherichia coli (several species, including E. coli
O157:H7)
Yersinia enterocolitica
Legionella species
Illnesses
Gastroenteritis, meningitis, associated with
reactive arthritis and Guillain-Barre paralysis
Gastroenteritis, dysentery, hemolytic uremic
syndrome, convulsions in young children,
associated with Reiters Disease (reactive
arthropathy)
Gastroenteritis, septicemia, anorexia, arthritis,
cholecystitis, meningitis, pericarditis, pneumonia,
typhoid fever
Cholera (dehydration and kidney failure)
Gastroenteritis, hemolytic uremic syndrome
(kidney failure)
Gastroenteritis, acute mesenteric lymphadenitis,
joint pain
Legionnaires Disease, Pontiac Fever
Source: Craun, 1999.
EPA and the Centers for Disease Control (CDC) data provide an indication of the types of
pathogens that have lead to waterborne disease outbreaks. Exhibit 2-3 presents the viral and bacterial
agents implicated as the cause of waterborne outbreaks in ground water systems reported to the CDC
from 1971 through 1996 (Craun, 1999). The 7 percent of outbreaks caused by protozoa (Giardia,
Cryptosporidiuni) indicate that those ground water systems were under the influence of surface water.
Fifteen percent of outbreaks were identified as bacterial, 9 percent were viral, and 6 percent were
chemical. The majority of outbreaks are caused by unknown microbial agents. The microbial agent is
difficult to determine because of the unavailability of analytical tools to identify different virus types.
Protozoan pathogens such as Cryptosporidia and Giardia are shed like bacteria and viruses in
the feces of infected individuals but, because of their larger size, protozoans are not normally
transported to ground water. Ground water sources found to be contaminated with either
Cryptosporidia or Giardia are considered ground water under the direct influence (GWUDI) of
surface water. GWUDI systems and, surface water sources in general, are regulated under the Surface
Water Treatment Rule (SWTR) and Interim Enhanced Surface Water Treatment Rule (IESWTR) (for
systems serving 10,000 people or more) and by the upcoming Long Term 1 Enhanced Surface Water
Treatment Rule (LT1ESWTR) (for systems serving fewer than 10,000 people).
2-4
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit 2-3. Etiology of Waterborne Outbreaks in Ground Water Systems,
Community and Noncommunity Water Systems, 1971 -96
Etiologic Agent
Bacterial
Shigella
Campylobacter
Salmonella, non-typhoid
£. co//
S. typhi
Yersinia
Plesiomonas shigelloides
Viral
Hepatitis A
Norwalk Agent
Protozoa
Giardia
Cryptosporidium
E. histolytica
Chemical
Undetermined microbial
Total
Percent of Outbreaks
15%
8%
3%
3%
1%
<1%
<1%
<1%
9%
-5%
-5%
7%
6%
1%
<1%
6%
63%
100%
Source: Craun, 1999.
2.2.2 Sources of Contaminants
Water obtained from ground water sources can contain microbial contaminants.
These contaminants can originate from the aquifer, wellhead, or within the distribution
system. This section presents a discussion of the potential sources of viral and bacterial
fecal contamination of ground water supplies. It describes sources of contamination in
ground water, factors that affect the survival and transport of fecal contaminants, and
contamination of drinking water in distribution systems and presents outbreak data for
sources and causes of ground water and drinking water contamination. Because
pathogens are associated with human and animal waste, the following discussions do not
necessarily focus on the specific types of microbe associated with each source, but fecal
contamination in general.
In addition to the sources and characteristics of fecal contamination, the occurrence
of contamination in ground water sources is highly variable due to local hydrologic,
hydrogeologic, and hydraulic conditions, soil characteristics, and land use patterns. The
study of how a microbe reaches a ground water source is generally termed the "fate and
transport" of a contaminant. Fate and transport factors must be known when attempting to
characterize the occurrence of microbial contaminants in a ground water source.
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
2-5
-------
2.2.2.1 Sources of Ground Water Contamination
Human and animal fecal matter contribute to the transmission of microbial
contamination. Fecal contamination of ground water can occur by several routes and at
several points in the process of providing public drinking water from ground water sources.
Fecal contamination from failed septic systems or sewage lagoons, leaking sewer lines,
land discharge, overflowing cesspools, and animal feedlots can reach the ground water
source through soils and fissures. Canter and Knox (1984) estimated the volume of septic
tank waste, alone, that is released into the subsurface to be one trillion gallons per year.
Other possible sources of fecal contamination include improperly treated wastewater used
to recharge the ground water or irrigate crop land and improper land application of raw or
treated sewage or sewage sludge.
Furthermore, solid wastes contaminated with human bacteria and viruses or animal
bacteria and viruses may contaminate ground water through individual waste disposal
practices, open dumping practices, and landfills (Washington State Department of Health,
1995). Improper land application of wastewaters associated with food processing or
animal slaughter may also contribute to the contamination of ground water sources of
drinking water. Animal wastes also carry microbial pathogens that can infect humans.
Such waste may enter ground water from unlined or leaky manure lagoons, spread
manure, and concentrated animal feeding operations (EPA, 1993; Washington State
Department of Health, 1995).
Storm water and surface water contaminated with human or animal pathogens may
transmit contamination to ground water through infiltration or direct injection. Storm water
may enter improperly constructed wells or improperly abandoned wells. Likewise,
contaminated surface waters carrying microbial pathogens may enter improperly
constructed or abandoned wells, causing ground water contamination (Washington State
Department of Health, 1995).
These sources of contamination may eventually reach the intake zone of a drinking
water well. Microbial contamination can also occur at the wellhead when wells are
improperly constructed, protected, and maintained. Furthermore, microbial contamination
can also occur in the distribution system when cross connection controls fail or when
leaking pipes allow infiltration of contaminants.
2.2.2.2 Factors Affecting Virus and Bacterial Transport in the
Subsurface
Many factors apparently control the removal and persistence of viruses and bacteria
in subsurface media. Because these factors are often interlinked and interrelated, defining
the processes involved in the survival and migration of viruses and bacteria is a complex
task. Factors such as pH, ionic strength, soil types, and type of virus, affect pathogenic
adsorption to soils. In addition, these factors are likely to have a direct or indirect effect on
pathogen survival. Factors that promote pathogen attachment to soils will also enhance
their survival (Vaughn and Landry, 1983). Two primary groups of factors influencing
26 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
microbial contaminant levels in ground water are: 1) the transport and survival of
microorganisms in the subsurface; and 2) the conditions of and near ground water intake
points. This section describes these factors for viruses and bacteria.
Factors Affecting the Transportation and Survival of Viruses in the Subsurface
The transport and persistence of a virus in the subsurface (i.e., the soil, unsaturated
zone and saturated zone) are important aspects of ground water contamination. Locally,
climatic changes, as well as agriculture and land use practices, influence and may alter the
complex soil environment. For example, wetter climatic conditions may result in high water
tables thereby potentially reducing the distance and time required by viruses to enter the
now shallower aquifers. In addition, sewage and sludge application to land may alter the
physical and chemical properties of soils and affect their capacity to impact virus migration
and survival (Bitton and Gerba, 1984).
Often factors affecting virus survival are complex, interrelated, and poorly
understood. Assessing virus survival is difficult because of the variety of factors influencing
survival and the temporal variations within factors (Keswick and Gerba, 1980). However,
according to Yates et al. (1985), most enteric viruses are stable between a pH range of 3
to 9. Yates et al. (1985) also believed that a low pH favors virus adsorption and a high pH
favors virus desorption from soil particles. Other factors affecting viral transport and
survival include light, temperature, hydrogeologic conditions, soil properties, inorganic
ions, the presence of organic matter in the soil, the type of virus, the presence or absence
of microbial activity, the iron content of the soil, and the soil moisture content.
Factors Affecting Bacterial Migration and Survival in Water and Soil
Bacterial survival varies for different types of bacteria and is dependent on a variety
of factors, such as temperature, hydrogeologic conditions, soil properties, pH, inorganic
ions content, organic matter content, bacterium type, microbial activity, and moisture
content. How these factors influence inactivation is often unknown (Yates and Yates,
1988). Generally, it takes two to three months to reduce pathogens to negligible numbers
after their application to soil; a survival time of five years, however, has been reported in
literature (Gerba and Bitton, 1984).
2.2.2.3 Other Factors that Contribute to the Contamination of Drinking
Water
Conditions at or near water-supply wells may contribute to the occurrence of ground
water contamination. Ground water contamination may occur at the wellhead in several
ways. The main causes are poor well location and/or construction, improperly abandoned
wells, the presence of testholes or exploratory wells, and well location within an area of
ground water development.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 2-7
-------
Poor Well Location and/or Construction
There are several ways fecal contamination can enter wells if they are
inappropriately located or poorly constructed. A water-supply well located in a low-lying
area is susceptible to flooding. An improperly constructed water-supply well may allow
surface runoff or surface waters to enter the well through a non-existent or broken well seal.
A well may be particularly vulnerable to surface water contamination if it is not adequately
cased and grouted. Ground water contamination may also result from water infiltrating into
the well through a contaminated gravelpack or the fill surrounding the well intake point.
In some cases, systems do not adhere to existing guidelines for the construction of
water-supply wells and wells may be drilled in or near potential sources of contamination. If
these wells penetrate the same aquifer as nearby domestic wells, their poor construction
may allow contaminants to enter the ground water source, thereby contaminating the water
used by these wells. Furthermore, since many old wells were constructed before the
institution of strict well construction guidelines, these wells are often not constructed in a
manner which prevents contamination.
Abandoned Wells
Historically, well abandonment and plugging have generally not been properly
planned, designed and executed (EPA, 1990; Canter et al., 1987). In many cases, the well
casing was pulled out if it was not too worn or corroded, thereby allowing contamination to
spread to different aquifers. Other wells may not have been adequately plugged, thus
providing a pathway for contamination. Occasionally, abandoned wells have also been
used as disposal sites for a variety of wastes. Such wells would then serve as conduits for
contaminated ground water to spread to other zones within an aquifer more rapidly or allow
contaminants to enter adjacent aquifers at lower hydraulic pressures (EPA, 1990).
Test Holes, Exploratory Wells and Monitoring Wells
Many test holes and exploratory wells have been dug or drilled into the subsurface
over time to search for substances such as oil, gas, coal, minerals, and water. In addition,
other holes have been drilled for testing and include soil boreholes and seismic shot holes
for geologic testing. Also, monitoring wells are often drilled to sample ground water
quality. When these holes are not backfilled, or when monitoring wells are not properly
constructed or abandoned, they provide potential conduits for contamination to enter
uncontaminated ground water sources.
2.2.2.4 Contamination of Drinking Water in Distribution Systems
Contamination within the distribution system may occur whether the source water is
ground water or surface water. Numerous incidents have been reported in literature of
contamination of ground water sources of drinking water in the distribution system. In
some cases, the ground water had not been disinfected, while in other cases waters had
been treated. In a rural Missouri township, an untreated ground water distribution system
28 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
became contaminated following the replacement of 45 water meters and the rupture and
replacement of two water mains (Swerdlow, etal., 1992; Geldreich, etal., 1992). This
system was not chlorinated after the maintenance and 243 cases of E. co//-induced
diarrhea resulted (Swerdlow, et al., 1992).
Inadequately disinfected distribution systems, including storage towers, can
develop microbial mats or biofilms. Initially biofilms may function as a filter, adsorbing
pathogens (Seunghyun, et al., 1997), but the pathogens may ultimately be shed (sloughed)
from the system, potentially contaminating the drinking water at the tap.
2.2.2.5
Outbreak Data for Sources and Causes of Contamination
Exhibit 2-4 summarizes EPA and CDC's Waterborne Disease Surveillance Data
of the total number of reported disease outbreaks attributed to consumption of
contaminated ground water from community and noncommunity systems. Exhibit 2-4
includes outbreaks caused by chemical or protozoan sources.
Exhibit 2-4. Waterborne Disease Outbreaks, Ground Water Sources, 1971-96
Type of Contamination
Source Contamination
Untreated Ground Water
Disinfected Ground Water
Filtered Ground Water
Distribution System Contamination
Inadequate Control of Chemical Feed
Miscellaneous, Unknown Cause
Total
Outbreaks in
Noncommunity
Systems
228
132
96
0
15
3
11
257
Outbreaks in
Community
Systems
72
31
38
3
34
5
3
114
Total Number of
Outbreaks
300
163
134
3
49
8
14
371
Source: EPA and CDC, 1 998. Unpublished report on Waterborne Disease Outbreaks in the U.S. from
1971-1996.
Between 1971 and 1996, 300 of the 371 reported waterborne outbreaks (81 percent) were
attributed to contaminated source water including untreated, interrupted, or inadequately disinfected
and inadequately filtered ground water. The second most common cause for disease outbreaks is
contamination of the distribution system, accounting for 49 (13 percent) of the 371 reported
waterborne outbreaks. Accounting for the balance of the outbreaks are inadequate chemical feed
resulting in chemical-related illnesses and those classified as miscellaneous. Cases reported as
miscellaneous are outbreaks where insufficient data exist to accurately categorize the source of the
contamination. The number of outbreaks reported to the CDC are believed to be an underestimate of
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
2-9
-------
the total number of waterborne outbreaks that actually occur (National Research Council, 1997).
Some of the reasons for the lack of recognition and reporting of outbreaks are as follows:
Some States do not have active disease surveillance systems. Thus, States that report the most
outbreaks may not be those in which the most outbreaks occur.
Health officials may not recognize the occurrence of small outbreaks, even in States with
effective disease surveillance systems. In cities, large outbreaks are more likely to be
recognized than sporadic cases or small outbreaks in which ill persons may consult different
physicians.
Some States do not always report identified waterborne disease outbreaks to the CDC.
Reporting outbreaks is voluntary.
Most cases of waterborne disease are characterized by general symptoms (diarrhea, vomiting,
etc.) that cannot be distinguished from other sources.
Only a small fraction of people who develop diarrheal illness seek medical assistance.
Many public health care providers may not have sufficient information to request the
appropriate clinical test.
If a clinical test is ordered, the patient must comply, a laboratory must be available and be
proficient, and a positive result must be reported in a timely manner to the health agency.
Not all outbreaks are effectively investigated. Outbreaks are included in the CDC database
only if water quality and/or epidemiological data are collected to document that drinking water
was the route of disease transmission. Monitoring after the recognition of an outbreak may be
too late in detecting intermittent or a one-time contamination event.
The vast majority of ground water systems are noncommunity water systems (NCWSs).
Outbreaks associated with NCWSs are less likely to be recognized than those in community
water systems because NCWSs generally serve nonresidential areas and transient populations.
Although waterborne disease outbreaks have been linked to ground water sources, the cause
and population affected may vary for each outbreak. These outbreaks, however, demonstrate that
ground water sources are not free of pathogenic contaminants and thus support the need for the GWR.
True incidence of waterborne outbreaks and associated illness is unknown; in addition, persistent low
to moderate levels of endemic waterborne illness often go undetected by routine disease surveillance
programs. This lack of knowledge stems from inadequate surveillance of disease outbreaks, insufficient
outbreak detection methods, lack of epidemiologic investigation, and lack of microbial monitoring.
2-10 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
2.2.3 Exposure to the Contaminants
EPA has reviewed data from 13 recent or on-going studies of pathogen and fecal indicator
occurrence in ground waters that supply public water systems. Each study was conducted
independently and with a different objective and scope. Well selection, the number of samples
collected from each well, and the sample volumes that affect interpretation of results, also varied among
the studies. These data are, however, important to GWR development because they provide insight on:
1) the extent to which ground water may be contaminated; 2) possible fecal indicators for source water
monitoring under the GWR; 3) a national estimate of ground water pathogen occurrence; and 4) the
hydrogeologic sensitivity criteria which may influence source water vulnerability. In addition,
determining the occurrence of microbial contaminants in ground water sources of drinking water can be
used to yield a national estimate of public health risk. The Occurrence and Monitoring document for
the GWR contains a brief summary of each of the examined occurrence studies. This chapter discusses
the two studies most relevant to the economic analysis.
Each occurrence study investigated a combination of different pathogenic and/or indicator
viruses and bacteria. The researchers tested the samples analyzed in each study for viral pathogens
such as enteroviruses (also called "total cultureable viruses") and/or bacterial pathogens such as
Legionella and Aeromonas. Several viruses and bacteria were identified using the polymerase chain
reaction technique (PCR). PCR amplifies the DNA sequences so that they are detected more easily or
at more sensitive levels. Although PCR detections do not necessarily indicate the presence of viable,
infectious viruses, they do suggest that sources of contamination and pathways for the transmission of
fecal material exist. The studies screened for bacterial indicators of fecal contamination, including
enterococci (or fecal streptococci, which are closely related), fecal coliforms (or E. coli, which is
closely related), and C. perfringens. Samples were also examined for bacteriophage, which are
viruses that infect bacteria and serve as viral indicators of fecal contamination. Among the
bacteriophage identified were somatic coliphage and/or male-specific coliphage, both of which infect
the bacterium E. coli.
American Water Works Association Research Foundation (AWWARF^ Study (Abbaszadegan et al..
1998. American Water Works Service Company^
The objectives of the joint AWWARF, EPA, and American Waterworks Service Company
study, or "AWWARF Study," include determining the occurrence of virus contamination in source
water of public ground water systems, investigating water quality parameters and occurrence of
microbial indicators in ground water and possible correlation with human viruses, and developing a
statistically-based screening method to identify wells at risk of fecal contamination.
The AWWARF Study analyzed samples collected at 448 sites in 35 States. The researchers
excluded sites known to be under the influence of surface water, sites where well records were not
available, or sites where the well was poorly constructed. Preliminary results of the study show that
approximately 64 percent of the wells are located in unconsolidated aquifers, 27 percent are located in
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 2-11
-------
consolidated aquifers, and 9 percent are located in unknown geology. Most of the sites serve a
population greater than 3,300 people and most of these systems practice some form of disinfection.
The source water samples were analyzed using a variety of methods to detect pathogens and
indicators. Samples were analyzed to determine the occurrence of enteroviruses, total colform and
enterococci bacteria, rotavirus, hepatitis A virus, and both male-specific and somatic coliphage in
ground waters of the United States. In order for researchers to use the information gathered from this
study to generate national risk estimates, samples were collected from different geographical locations
with a variety of physical and chemical characteristics to closely match the actual national geologic
profile of ground water sources (Abbaszadegan et al, 1998). Exhibit 2-5 presents a summary of
preliminary AWWARF results.
Exhibit 2-5. Preliminary Results of the AWWARF Study
Assay
Enterovirus (cell culture)
Total coliform
Enterococci
Clostridium spores
Salmonella WG-49
Somatic Coliphage (E. coli C host)
Somatic and Male Specific Coliphage (E. coli C-3000 host)
Norwalk virus (PCR)
Enterovirus (PCR)
Rotavirus (PCR)
Hepatitis A Virus (PCR)
Percent Positive per Site
(number positive/samples analyzed)
4.8% (21/442)
9.90/0 (44/445)
8.7% (31/355)
1.8% (1/57)
9.5% (42/440)
4.1% (18/444)
10.8% (48/444)
0.96% (3/3 12)
15.9% (68/427)
14.6% (62/425)
7.2% (31/429)
Source: Abbaszadegan etal., 1998.
EPA/AWWARF Study (Lieberman et al.. (1994.1999}}
The study objectives included the following: 1) develop and evaluate a molecular biology
(PCR) monitoring method; 2) obtain occurrence data for human enteric viruses and Legionellct in
ground water; and 3) assess the microbial indicators of fecal contamination. This was accomplished by
sampling vulnerable wells nominated by States to confirm the presence of fecal indicators (Phase I) and
then choosing a subset of these for monthly sampling for one year (Phase n).
In Phase I, 96 of the 180 potentially vulnerable wells were selected for additional consideration.
Well vulnerability was established using historical microbial occurrence data and waterborne disease
outbreak history, known sources of human fecal contamination in close proximity to the well, and
sensitive hydrogeologic features (e.g., karst). Selected wells were located in 22 States and 2 U.S.
2-12
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
territories. Additional water quality information was then successfully obtained for 93 of the wells
through use of a single one liter grab sample that was subsequently tested for several microbial
indicators (See Table II-6). The wells from Phase I served as the well selection pool for Phase II
sampling.
In Phase n, 23 of the Phase I wells were selected for monthly sampling for one year. Seven
additional wells were selected from a list of State-nominated wells for a total of 30 wells, located in 17
States and 2 U.S. territories. The additional seven wells were based on other criteria, including
historical water quality data, known contaminant sources in close proximity to the well, hydrogeologic
character or to replace wells that were no longer available for sampling. Samples were analyzed for
enteroviruses, Legionella, enterococci, E. coli, Clostridium perfringens, total coliforms, somatic
coliphage, male-specific coliphage, and Bacteriodes phage. For each sample analyzed for enteric
viruses and bacteriophages, an average of approximately 6,000 liters of water were filtered and
analyzed by cell culture.
Enteroviruses were recovered in 20 samples from seven wells, and were speciated by
serotyping. Coxsackievirus and echovirus, as well as reovirus, were identified. The range in virus
concentration in enterovirus-positive samples was 0.9-212 MPN/100 liters (MPN, or most probable
number, is an estimate of concentration).
The hydrogeologic settings for the seven enterovirus-positive wells were karst (3), a gravel
aquifer (1), fractured bedrock (2), and a sandy soil and alluvial aquifer (1). The karst wells were all
positive more than once. The gravel aquifer was also enterovirus-positive more than once, with 4 of 12
monthly samples positive.
2.2.4 Sensitive Subpopulations
In assessing the potential impact of waterborne disease it is important to recognize that certain
sensitive individuals may be at a greater risk of serious illness than the general population. These
sensitive subgroups of the population include pregnant and lactating women, the very young, the elderly,
the immunocompromised, and the chronically ill. In total, these subgroups represent almost 20 percent
of the current population of the United States.
Pregnant and lactating women may be at an increased risk from enteric viruses as well as act as
a source of infection for neonates. Infection during pregnancy may also result in the transmission of
infection from the mother to the child in utero, during birth, or shortly thereafter. Since very young
children do not have fully developed immune systems, they are at increased risk and are particularly
difficult to treat.
Infectious diseases are also a major problem for the elderly because the immune function
declines with age. As a result, outbreaks of waterborne diseases can be devastating on the elderly
community (e.g., nursing homes) and may increase the possibility of significantly higher mortality rates in
the elderly than in the general population.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 2-13
-------
Immunocompromised individuals are an ever growing proportion of the population with the
relatively new and severe problem magnified by the AIDS epidemic and the escalation in organ and
tissue transplantations. Enteric pathogens take advantage of the impaired immune systems of these
individuals and set up generalized and persistent infections in the immunocompromised host. These
infections are particularly difficult to treat and can result in a significantly higher mortality than
immunocompetent persons.
Exhibit 2-6 presents the estimates of some of the sensitive populations in the United States who
are at increased risk of infection from, and the effects of, waterborne microorganisms and pathogens.
Exhibit 2-6. Sensitive Populations in the United States
Sensitive Population
Individuals
Pregnant/Lactating Women and Neonates
Pregnancies
Lactating Women
Neonates
5,657,000
2,247,635
4,002,000
Elderly
Elderly (over 65)
Residences in Nursing Homes or
Related Care Facilities
29,400,000
1,553,000
Chronically III
AIDS
Cancer Treatment Patients
Organ Transplant Recipients
581,429
1,853,795
22,736
Source: Department of Commerce, 1991; National Health Interview Survey
(NHIS), Bureau of the Census' 1996 Statistical Abstract of the U.S.
2.3 Current Control and Potential for Improvement
The underlying objective of the GWR is to build upon successful State requirements
and practices. EPA intends to strengthen what is in place, not replace it with new practices. While
protective practices (e.g., wellhead protection, disinfection, well siting, construction requirements, and
distribution system safeguards such as cross-connection control) are used by many States, EPA
recognizes the potential for improvement since few of these measures are used by all States.
Moreover, the States appear to employ a variety of interpretations of the same practice (EPA, 1996).
Source Protection
Protecting the ground water source is an important component in assuring that the wellhead
areas of all public water systems (PWSs) are free from contaminants that have adverse health affects.
Source protection includes implementing measures such as wellhead protection programs and
complying with well construction code requirements.
2-14
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Currently, 44 States have approved wellhead protection programs. A wellhead protection
program involves provisions for delineation of wellhead protection areas (WHPAs) for each public well
or well field, identification of all potential anthropogenic sources within the protection area, technical
and financial assistance, control measure implementation, education, training and demonstration projects
to protect wellhead areas from contaminants, contingency plans for alternative water supplies in
contamination cases, siting considerations for new wells, and public participation.
Recently, States have begun implementing the Source Water Assessment Program (SWAP).
In accordance with the 1996 SDWA Amendments, States were required to submit a program to EPA
by February 1999 and to implement the SWAP by November 2001. Intending to focus prevention
resources on drinking water protection, provisions for a State SWAP include: delineating the source
water protection area, conducting a contaminant source inventory, determining the susceptibility of the
public water supply to contamination from the inventoried sources, and releasing the results of the
assessments to the public.
Another source water protection method involves the use of well construction codes.
Currently, 48 States employ some form of construction code. The standards and procedures listed in
many construction codes are designed to protect a ground water source from contamination. These
standards and procedures regulate the entire process from initial penetration or excavation of the
ground, development of the well, equipment installation and disinfection, to final approval of the well for
use as a potable water supply. Many States also designate setback distances to ensure that the well is
not constructed at a site subject to current or potential contamination. Hydrogeologic data may also be
used to evaluate the adequacy of the site to provide a safe and healthful supply of water to the public.
Disinfection/Treatment
Disinfection is very important in reducing waterborne illnesses in the United States. With the
exception of Connecticut, all States require some form of disinfection for designated systems. The
criteria for determining which systems must disinfect varies from State to State. Fourteen States require
all systems to disinfect; of these States, however, at least nine have provisions allowing systems to
waive their disinfection requirements. Some States base the disinfection requirements upon Total
Coliform Rule (TCR) compliance; others base the requirement upon the date of construction or upon
the results of a sanitary survey.
Disinfection can consist of a variety of treatment technologies or disinfectants and is used to
inactivate, remove, or kill disease-causing microorganisms. Ground water disinfection usually involves
the use of chlorine disinfection through chlorine gas injection or hypochlorination. Other treatment
technologies include: chloramines, chlorine dioxide, mixed oxidants, ultraviolet light, ozone, reverse
osmosis, or nanofiltration.
The effectiveness of these treatment technologies varies widely by technology type and by
system operations. Technology effectiveness also may be dependent on whether the system specifies a
minimum level of treatment (i.e., a minimum disinfectant residual), a minimum CT value, or a microbial
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 2-15
-------
kill reduction value. Systems also may use disinfectant residuals to protect the distribution system from
re-contamination. CT refers to the product of the residual disinfectant concentration, C (in milligrams
per liter [mg/L]), and the disinfectant contact time, T (in minutes). EPA considers CT a primary
method for determining the level of inactivation for several treatment technologies.
System Integrity
Most States require system integrity measures, such as sanitary surveys and cross-connection
control, as additional measures to prevent microbial contamination and to protect the health of the
customer. Although these measures are required by almost every State, their requirements vary widely
between each State. Due to this variability and difference in program strengths, waterborne disease
outbreaks have occurred as a result of lapses in system integrity. EPA and CDC Waterborne Disease
Outbreak data show that 137 outbreaks occurred between 1971 and 1996 in ground water systems
that had treatment that was either inadequate or had failed altogether. That same data show that there
were 49 distribution system-related waterborne disease outbreaks between 1971 and 1996 (Craun,
1999).
Sanitary surveys are on-site inspections of the source water, treatment facilities, distribution
system, finished water storage tanks, the pumps and pump facilities, monitoring records, management
and operation, and operator compliance with State requirements of a PWS. Sanitary surveys allow the
PWS to identify existing or potential sources of contamination. With the exception of the State of
Washington, all States currently require sanitary surveys to be performed on ground water systems.
EPA found that many of these surveys are general in nature and that they differ in the types of systems
surveyed, the content of the survey, and who is designated to conduct the survey. EPA also found that
46 States do not specifically require systems to correct deficiencies and that a number of States do not
appear to have legal authority to require correction of defects.
With the exception of Delaware, all States currently implement some form of cross-connection
control. Cross-connection control involves the inspection of service connections with respect to the
risk of backflow and the consequences of backflow, provisions for eliminating cross-connections, the
installment of backflow prevention devices, and provisions for violations.
2.4 Regulatory History
This section briefly describes the existing regulations applicable to ground water systems.
These rules serve as the regulatory baseline for the GWR. The regulations that are discussed include
the Total Coliform Rule (TCR), Surface Water Treatment Rule (SWTR) and Interim Enhanced Surface
Water Treatment Rule (TfiSWTR), Information Collection Rule (ICR), Stage 1 Disinfection/Disinfection
By-Products Rule (Stage 1 DBPR), and Underground Injection Control.
Total Coliform Rule (TCR)
The TCR, promulgated in June of 1989, is applicable to all public water systems. The rule was
designed to protect public water supplies from adverse health effects associated with disease-causing
216 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
organisms (pathogens) and is the only current federal regulation addressing microbial contamination of
ground water systems.
Total coliforms are a group of closely-related bacteria that are generally free-living in
the environment but are also normally present in water contaminated with human and animal feces. As
shown in section 2.2.3, total coliforms are often associated with waterborne disease outbreaks.
Generally, total coliforms do not cause disease but are indicators that other harmful organisms may be
present. Specifically, coliform measurements are used to determine the efficiency of treatment, the
integrity of the water distribution system, and as a screen for fecal contamination. Their presence in
drinking water indicates that the system is either fecally-contaminated or potentially vulnerable to fecal
contamination.
The TCR requires systems to monitor their distribution system for total coliforms at a frequency
dependent upon two factors: 1) the number of people served; and 2) whether the system is a
community water system or noncommunity water system. The monitoring frequency ranges from 480
samples per month for the largest systems to once annually for some of the smallest systems. If a
system has a total coliform-positive sample, operators must: 1) test that sample for the presence of
fecal coliform or E. coli; 2) collect a set of repeat samples within 24 hours and analyze them for total
coliforms (and fecal coliform or E. coli, if positive); and 3) collect at least five routine samples in the
next month of sampling.
Under the TCR, a system that collects 40 or more samples per month (generally systems that
serve more than 3,300 people) violates the maximum contaminant level (MCL) if more than 5 percent
of the samples (routine + repeat) are total coliform-positive. A system that collects fewer than 40
samples per month violates the MCL if two samples (routine or repeat samples) are total coliform-
positive. For either size system, if two consecutive total coliform-positive samples occur at a site, and
one is fecal coliform/E1. co//'-positive, the system has an acute violation of the MCL, and must report to
the public immediately. The presence of fecal coliforms or E. coli indicates that recent fecal
contamination is present in the drinking water.
The rule also requires a sanitary survey every five years for community and every 10 years for
noncommunity ground water systems sampling fewer than five samples per month (about 97 percent of
the systems serve 3,300 or fewer). Other provisions of the TCR include criteria for invalidating a
positive or negative sample and a sample siting plan to ensure that all parts of the distribution system are
monitored over time.
Surface Water Treatment Rule (SWTR) and Interim Enhanced Surface Water Treatment Rule
OESWTR)
The SWTR, promulgated in June of 1989, covers all systems that use surface water or ground
water under the direct influence of surface water. It is intended to protect against the adverse health
effects of exposure to Giardia lamblia, viruses, and Legionella, as well as many other pathogens.
The rule requires all such systems to reduce the level of Giardia by 99.9 percent (3 logs) and viruses
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 2-17
-------
by 99.99 percent (4-logs). To accomplish this reduction, a system must filter its source water, unless it
can meet certain EPA-specified criteria on source water quality, and disinfect. More specifically, the
SWTR requires: 1) a 0.2 mg/L disinfectant dose at the treatment facility; 2) maintenance of a
detectable disinfectant residual in all parts of the distribution system; 3) combined filter effluent
performance standard for turbidity (e.g., for rapid filters, 5.0 nephelometric turbidity units (NTU) as a
maximum and 0.5 NTU in 95 percent of the sample readings during a month); and 4) watershed
protection and other requirements for unfiltered systems. The SWTR set a maximum contaminant level
goal (MCLG) of zero for Giardia, viruses, and Legionella. (The MCLG is a non-enforceable level
based only on health effects.)
In December 1998, EPA promulgated the IESWTR, which covers all systems that use surface
water or ground water under the direct influence of surface water that serve 10,000 or more people.
Key provisions include: a 2 log Cryptosporidium removal requirement for filtered systems;
strengthened combined filter effluent turbidity performance standards; individual filter turbidity
provisions; disinfection benchmark provisions to assure continued levels of microbial protection while
facilities take the necessary steps to comply with new disinfection byproduct standards; inclusion of
Cryptosporidium in the definition of ground water under the direct influence of surface water and in the
watershed control requirements for unfiltered public water systems; requirements for covers on new
finished water reservoirs; and sanitary surveys for all surface water systems regardless of size. The rule
set an MCLG of zero for Cryptosporidium. EPA plans to propose a companion microbial rule for
surface water systems serving less than 10,000.
Since the SWTR and IESWTR apply to ground water systems under the direct influence of
surface water, the GWR will not address these systems.
Information Collection Rule
The ICR is a monitoring and data reporting rule that was promulgated in May 1996. The data
and information provided by this rule is needed to support development of two regulations that EPA is
in the early process of developing, the Stage n Disinfection Byproducts Rule and a related microbial
rule, the Stage 2 Long Term Enhanced Surface Water Treatment Rule.
The ICR covers large PWSs serving populations over 100,000; a more limited set of ICR
requirements pertain to ground water systems serving between 50,000 and 100,000 people. About
300 PWSs operating 500 treatment plants are involved with the extensive ICR data collection. The
ICR requires systems to collect source water samples (and in some cases finished water samples)
monthly for 18 months and test the samples for the following organisms: Giardia, Cryptosporidium.,
viruses, total coliforms, and fecal coliforms or E. coli. The ICR also requires systems to determine the
concentrations of a host of disinfectant and disinfection byproduct concentrations in different parts of
the system. These disinfection byproducts form when disinfectants used for pathogen control react with
naturally occurring organic compounds already present in source water. Some of these byproducts are
toxic or carcinogenic. The rule also requires systems to provide specified operating and engineering
data to EPA. The required 18 months of monitoring under the ICR ended in December 1998.
218 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
As noted above, the only ground water systems affected by the ICR were those that serve at
least 50,000 people. These systems had to conduct treatment study applicability monitoring (by
measuring the level of total organic carbon) and, if necessary, treatment studies. In addition, ground
water systems serving at least 100,000 people had to obtain disinfectant and disinfection byproduct
occurrence and treatment data.
Stage 1 Disinfectants and Disinfection Byproducts Rule (DBPR)
The Stage 1 DBPR (63 FR 69389; December 16, 1998) (EPA, 1998) sets maximum residual
disinfection level limits for chlorine, chloramines, and chlorine dioxide and MCLs for chlorite, bromate,
and two groups of disinfection byproductstotal trihalomethanes (TTHMs) and haloacetic acids
(HAAS). TTHMs consist of the sum of chloroform, bromodichloromethane, dibromochloromethane,
and bromoform. HAAS consist of the sum of mono-, di-, and trichloroacetic acids, and mono- and
dibromoacetic acids. The rule requires water systems that use surface water or ground water to
remove specified percentages of organic materials, measured as total organic carbon, that may react
with disinfectants to form DBFs. Under the rule, precursor removal will be achieved through a
treatment technique (enhanced coagulation or enhanced softening) unless a system meets alternative
criteria.
The Stage 1 DBPR applies to all community water systems (CWSs) and nontransient
noncommunity water system (NTNCWS), whether surface water systems or ground water systems,
that treat their water with a chemical disinfectant for either primary or residual treatment. In addition,
certain requirements for chlorine dioxide apply to transient water systems.
A ground water system that disinfects with chlorine or other chemical disinfectants must comply
with the Stage 1 DBPR by December 2003. For ground water systems not under the direct influence
of surface water, sampling frequency will depend upon the number of people served.
Systems that serve 10,000 people or greater must take one sample per quarter per treatment
plant and analyze for TTHMs and HAAS.
Systems that serve fewer than 10,000 people must take one sample per year per treatment
plant during the month of warmest water temperature and analyze it for the same chemicals.
Systems must monitor for chlorine or chloramines at the same location and time that they
monitor for total coliforms. Additional monitoring for other chemicals is required for systems
that use ozone or chlorine dioxide.
Underground Injection Control Program
The EPA's Underground Injection Control program was established to protect aquifers that
are, or might be, sources of drinking water from underground injection of fluids through wells. Owners
and operators of injection wells are prohibited from allowing the movement of fluid containing any
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 2-19
-------
contaminant into underground sources of drinking water if the presence of that contaminant may cause a
violation of any primary drinking water regulation or may otherwise adversely affect human health. To
prevent such fluid movement, EPA or the appropriate State regulatory agency may, for any injection
well, impose requirements for construction, corrective action, operation, monitoring, reporting, and
plugging and abandonment. These regulations are designed to recognize varying geologic, hydrological
or historical conditions among different States or areas within a State.
The regulations included in 40 CFR 144.6 define five classes of wells. These wells may be
injected with fluids that are associated with hazardous waste or radioactive waste sites, natural gas or
oil production, extraction of minerals, or other purposes. Class V wells are those most often associated
with relevant ground water contamination, and include: 1) untreated sewage waste disposal wells; 2)
cesspools; 3) septic systems (undifferentiated disposal method); 4) septic systems (well disposal
method); 5) septic systems (drainfield disposal method); and 6) domestic wastewater treatment plant
effluent disposal wells. EPA regulates only multiple dwelling, community or regional septic systems, as
opposed to individual or single family residential septic systems, as Class V wells (40 CFR §
On July 29, 1998, EPA proposed changes to the Class V UIC regulations that would add new
requirements for three categories of Class V wells that pose a high risk when located in ground
water-based source water protection areas being delineated by States under the 1996 SDWA
Amendments (EPA, 1996b). Class V motor vehicle sewage waste disposal wells in such areas would
either be banned or would have to get a permit that requires fluids released in those wells to meet the
drinking water maximum contaminant levels (MCLs) at the point of injection. Class V industrial waste
disposal wells in ground water-based source water protection areas also would be required to meet the
MCLs at the point of injection, and large-capacity cesspools in such areas would be banned. EPA
proposed these new requirements to address three categories of wells identified as posing a high risk of
ground water contamination based on available information. These include motor vehicle waste
disposal wells, industrial waste disposal wells, and cesspools in ground water-based source water
protection areas. EPA expects to achieve substantial protection of underground sources of drinking
water by targeting the requirements to these particular wells.
2.5 Economic Rationale
This section of the RIA discusses the statutory authority and the economic rationale for
choosing a regulatory approach to protect public health from drinking water contamination. The
economic rationale is provided in response to Executive Order 12866, Regulatory Planning and
Review, which States,
[EJach agency shall identify the problem that it intends to address
(including, where applicable, the failures of the private market or public
institutions that warrant new agency action) as well as assess the
significance of that problem (Sect. 1 b(l)).
2-20 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
In addition, 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 (p. 3)." Therefore, the economic rationale laid out in this section should not be interpreted as the
Agency's approach to implementing SDWA. Rather, it is the Agency's economic analysis, as required
by the Executive Order, to support a regulatory approach to the public health issue at hand.
2.5.1 Statutory Authority for Promulgating the Rule
Section 1412(b)(8) of the Safe Drinking Water Act requires that EPA develop regulations
specifying the use of disinfectants for ground water systems, as necessary. Under is provision, EPA has
the responsibility to develop a ground water rule that specifies the appropriate use of disinfection. As
mentioned previously, the 1996 SDWA amendments establish a statutory deadline of 2002; EPA,
however, intends to promulgate the final GWR in 2000.
2.5.2 The Economic Rationale for Regulation
In addition to the statutory directive to regulate microbial contaminants in ground water, there is
also a strong economic rationale for government regulation. The need for regulation is a direct result of
the structure of the market for publically-provided drinking water. Economic theory suggests that
society's well being is maximized when goods are produced and sold in well functioning competitive
markets. A perfectly competitive market is said to exist when there are many producers of a product
selling to many buyers, and both producers and consumers have complete knowledge regarding the
products of each firm. In this perfectly competitive market, there must also be no barriers to entry in
the industry, meaning that firms in the industry must not have any advantage over potential new
producers. Two major factors in the public water supply industry do not satisfy the requirements for a
competitive market and lead to market failures that require regulation.
First, the public water market has monopolistic tendencies. These monopolies tend to exist
because it is not economically efficient to have multiple suppliers competing to build multiple systems of
pipelines, reservoirs, wells, and other facilities. Instead, a single firm or government entity performs
these functions under public control. Under monopolistic conditions, consumers are provided only one
level of service with respect to the quality attribute of the product, in this case drinking water quality.
Since water purveyors often operate in such a monopolistic environment they may not respond to the
usual market incentive to satisfy their consumers' desire for high drinking water quality.
Second, high information and transaction costs impede public understanding of the health and
safety issues concerning drinking water quality. The type of health risks potentially posed by trace
quantities of drinking water contaminants involve analysis and distillation of complex lexicological data
and health sciences. EPA recently promulgated the Consumer Confidence Report (CCR) Rule that
makes water quality information more easily available to consumers. The CCR Rule requires
community water systems to mail or make available an annual report on local drinking water quality to
their customers. Consumers, however, still have to understand the information for its health risk
implications. Furthermore, even if informed consumers are able to engage utilities regarding these health
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 2-21
-------
issues, the costs of such engagement-transaction costs (measured in personal time and commitment)
present another significant impediment to consumer expression of risk preference.
SDWA regulations are intended to provide a level of protection from exposure to drinking
water contaminants that would not otherwise occur in the existing market environment for public water
supply. The regulations set minimum performance requirements for all public water supplies in order to
protect all consumers from exposures to contaminants. SDWA regulations are not intended to
restructure flawed market mechanisms or to establish competition in supply, but rather, to regulate the
"product" produced within these markets. In other words, SDWA standards establish the level of
service to be provided in order to better reflect public preferences for safety. Also, the federal
regulations remove the high information and transaction costs that would be required for consumers to
make informed purchasing decisions by acting on behalf of all consumers in balancing the risk reduction
and the social costs of achieving this reduction.
2-22 Proposed Ground Water Rule-Regulatory Impact Analysis April 5, 2000
-------
2.6 References
Abbaszadegan, M., P.W. Stewart, M.W. LeChevallier, J.S. Rosen, and C.P. Gerba. 1998/1999.
Occurrence of Viruses in Ground Water in the United States. American Water Works
Association Research Foundation. Denver, CO, 157 p.
Abbaszadegan, M., P.W. Stewart, and M.W. LeChevallier. 1999. "A Strategy for Detection of
Viruses in Groundwater by PCR." Applied and Environmental Microbiology. Vol.
65(2):444-449.
Bitton, G., and C.P. Gerba. 1984. Ground Water Pollution Microbiology. John Wiley & Sons,
New York, NY.
Bureau of the Census. 1996. National Health History Interview Study; Statistical Abstract of the U.S.
Canter L. and R.C. Knox. 1984. Evaluation of Septic Tank System Effects on Ground Water.
U.S. Environmental Protection Agency, Washington D.C. EPA Publication No. EPA-300/2-
84-107.
Canter, L.W., R.C. Knox, and D.M. Fairchild. 1987. Ground Water Quality Protection. Lewis
Publishers, Inc., Chelsea, MI.
Craun, G. 1999. Personal Communication with SAIC. January.
Craun, G. 1991. "Causes of waterborne outbreaks in the United States." Water Sci. Technol.
24:17-20
Department of Commerce. 1991. Washington, DC.
Domingue. 1993.
EPA. 1998. National Primary Drinking Water Regulations; Stage 1 Disinfectants and Disinfection
Byproducts Rule; Final Rule. 63 FR 69389.
EPA. 1996. Ground Water Disinfection and Protective Practices in the United States. Prepared
by the U.S. Environmental Protection Agency and Science Applications International
Corporation.
EPA. 1996b. Amendments to the Safe Drinking Water Act. 63 FR 40585.
EPA. 1993. Wellhead Protection: A Guide for Small Communities. Seminar Publication. EPA
Office of Research and Development. EPA/625/R-93/002.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 2-23
-------
EPA. 1990. Ground Water Volume 1: Ground Water and Contamination. EPA Office of
Research and Development. EPA/625/6-90/016a.
Framm and Soave. 1997.
Geldreich, E.E. 1996. Microbial Quality of Water Supply in Distribution Systems. Lewis
Publishers, Boca Raton, FL.
Geldreich, E.E., K.R Fox, J.A. Goodrich, E.W. Rick, R.M. Clark, and D.L. Swerdolw. 1992.
"Searching for a water supply connection in the Cabool, Missouri disease outbreak of
Escherichia coli 0157:H7." Water Research. 26(8): 1127-1137.
Gerba C.P., and G. Bitton, editors. 1984. "Microbial pollutants: Their survival and transport pattern
to ground." Ground Water Pollution Microbiology. John Wiley and Sons, New York, NY.
Gerba, C.P., J.B. Rose, and C.N. Haas. 1996. "Sensitive populations: who is at the greatest risk?"
International J. of Food Microbiology. 30:113-123.
Keswick, B.H., and C.P. Gerba. 1980. "Viruses in ground water." ES&T. 14(11): 1290-1297.
Lieberman, R.J., L.C. Shadix, C.P. Newport, M.W.N. Frebis, S.E. Moyer, R.S. Safferman, R.E.
Stetler, D. Lye, G.S. Fout, and D. Dahling. 1999. "Source water microbial quality of some
vulnerable public ground water supplies." Unpublished report in preparation.
Lieberman, R.J., L.C. Shadix, B.S. Newport, S.R Crout, S.E. Buescher, RS. Safferman, R.E. Stetler,
D. Lye, G.S. Fout, and D. Dahling. 1994. "Source water microbial quality of some vulnerable
public ground water supplies." In Proceedings, Water Quality Technology Conference, San
Francisco, CA, October.
National Research Council. 1997. Safe Water From Every Tap, Improving Water Service to
Small Communities. National Academy Press, Washington, DC.
Rusin, P. A., J.B. Rose, C.N. Haas, C.P. Gerba. 1997. "Risk Assessment of Opportunistic Bacterial
Pathogens in Drinking Water." Water Rev. Environ. Contain. Toxicol. 152:57-85.
Seunghyun, K., and M.Y. Corapcioglu. 1997. "The role of biofilm growth in bacterial-facilitated
contaminant transport in porous media." Transport in Porous Media. 26:161-181. Kluwer
Academic Publishers, the Netherlands.
Solimena and De Camilli, 1995.
2-24 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
Swerdlow; D.L, B.A. Woodruff, R.C. Brady, P.M. Griffin, S. Tippen, H. Donnel Jr., E. Geldreich,
BJ. Payne, A. Meyer Jr., J.G. Wells, K.D. Greene, M. Bright, N.H. Bean, and P.A. Blake.
1992. "A waterborne outbreak in Missouri of Escherichia coli O157:H7 associated with
bloody diarrhea and death." Annals of Internal Medicine. 117(10): 812-819.
Vaughn, J.M., and E.F. Landry. 1983. "Viruses in soils and ground water." Viral Pollution of the
Environment. G. Berg, editor. CRC Press, Boca Raton, FL.
Washington State Department of Health. 1995. "Wellhead Protection Program Guidance Document."
DOH Publication No. 331-018. April.
Yates, M.V., and S.R. Yates. 1988. "Modeling microbial fate in the subsurface environment." CRC
Crit. Rev. Environ. Control. 17:307-344.
Yates, M.V., S.R. Yates, A.W. Warrick, and C.P. Gerba. 1985. "Preventing viral contamination of
drinking water." Groundwater Contamination and Reclamation. J. A WWA. 117-121.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 2-25
-------
3. Consideration of Regulatory Options
3.1 Introduction
In order to address the public health concerns presented in Section 2.2 and through
consultation with interested stakeholders, EPA has developed a number of regulatory options for
consideration. This chapter summarizes the option development process (Section 3.2) and provides a
brief description of each regulatory option considered by EPA in this economic analysis (Section 3.3).
3.2 Option Development Process
In 1992, EPA circulated a "straw man" proposal for review and comment, which began the
process of developing regulatory alternatives for addressing microbial contamination in ground water
systems. In 1993, EPA published a preliminary draft of the Ground Water Disinfection Rule (later
renamed the Ground Water Rule). After review of the public comments, EPA recognized that
additional information needed to be gathered and, in 1995, convened a GWR regulatory workgroup.
EPA used the workgroup as a vehicle to obtain comments and additional information regarding the
GWR. In 1996, EPA published a report on ground water- related statutes, regulations, guidance, and
disinfection practices gathered from 50 State drinking water programs (EPA, 1996a). In 1997, EPA
formally initiated another workgroup including members from EPA, other federal agencies, and State
agencies to cooperate in the development of a proposed GWR.
In December 1997, EPA initiated stakeholder meetings. EPA invited the general public to
attend via published notices in the Federal Register. EPA made a special effort to reach out to local
citizens, environmental groups, small businesses and water suppliers. Meetings conducted in
Washington, D.C.; Portland, Oregon; Madison, Wisconsin; and Dallas, Texas provided development
updates and allowed the stakeholders to provide the EPA with their comments.
In addition to the public meetings with stakeholders, EPA, as part of the consultation process
required by the Small Business Regulatory Enforcement Fairness Act (SBREFA), met with
representatives of small ground water systems (i.e., considered to be those serving less than 10,000
people) in March and April 1998. EPA presented possible regulatory requirements and requested
comments from the representatives during these meetings.
In January 1999, EPA published a preliminary draft preamble for the GWR and solicited
comment. The preliminary draft preamble described regulatory alternatives and requested public
comment on a number of potential modifications. EPA received 80 comments on the preliminary draft
preamble.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 3-1
-------
3.3 Summary of Alternatives Considered
As a result of the input received from stakeholders, the EPA workgroup, and other interested
parties, EPA constructed four regulatory options: (1) the sanitary survey option; (2) the sanitary
survey and triggered monitoring option; (3) the multi-barrier option; and the (4) across-the-board
disinfection option. Exhibit 3-1 summarizes the basic provisions of these options; the following sections
describe each option.
Exhibit 3-1. Regulatory Options and Basic Provisions
PROVISIONS
Sanitary Survey
Triggered1 source water (microbial)
monitoring
Hydrogeologic sensitivity
assessment and routine source
water monitoring2
All systems install/upgrade and
maintain treatment
Regulatory Options
1
Sanitary
Survey
y
2
Sanitary Survey
and Triggered
Monitoring
y
y
3
Multi-
Barrier
Approach
y
y
y
4
Across-the-
Board
Disinfection
y
y
triggered by a total coliform-positive sample in the distribution system
2for those systems determined to be hydrogeologically sensitive
3.3.1 Option 1: Sanitary Survey Only
Sanitary surveys are on-site inspections of the source water, treatment facilities, distribution
systems, finished water storage tanks, monitoring records, and the management and operation of a
public water system (PWS). This option would require sanitary surveys to be conducted by the State
at least once every three years for Community Water Systems (CWSs) and every five years for
Noncommunity Water Systems (NCWSs). In addition, the surveys would address eight elements: (1)
source; (2) treatment; (3) distribution system; (4) finished water storage; (5) pumps, pump facilities, and
controls; (6) monitoring, reporting, and data verification; (7) system management and operation; and (8)
operator compliance with State requirements. Operators would be required to correct any significant
deficiencies within 90 days of receiving the State sanitary survey report or have a State-approved
schedule for correcting these deficiencies.
3.3.2 Option 2: Sanitary Survey and Triggered Monitoring
The second regulatory option includes, in addition to all the sanitary survey components of the
first option, a microbial monitoring requirement for certain ground water systems. Ground water
systems that do not already treat to 4-log removal/inactivation of viruses and which have a total
3-2
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
coliform positive for any sample taken under the Total Coliform Rule must subsequently sample and
analyze the source water for either enterococci, E. coli, or coliphage as determined by the State or
primacy agent.
EPA believes source water monitoring enhances a targeted risk-based regulatory strategy by
addressing those systems with a high possibility of fecal contamination (i.e., those systems with proven
distribution contamination). These systems would be required to collect a source water sample within
24 hours of receiving notification of a total coliform positive in the distribution system and test the
sample for the presence of one of the fecal indicators. Positive source water samples would require the
system to treat the source water, eliminate the source of contamination, or provide an alternative source
of safe water no later than 90 days (or longer with at State-approved plan) from the date the
contamination is detected. Systems meeting the requirements through treatment could select among a
number of technologies including gas chlorination, hypochlorination, chlorine dioxide, ozone, mixed
oxidants, ultraviolet radiation, or chloramination. The State may waive the treatment technique
requirement for positive microbial monitoring samples based on five negative samples taken within 24
hours of notification of the first source water sample positive if there has been no total coliform positives
in the previous five years of system operation.
Systems that treat would also have to monitor to ensure that the treatment was effective. The
type of monitoring would vary based on the treatment technology selected by the system. For example
a system that selects hypochlorination would be required to monitor their chlorine residual (i.e., the
concentration of the chlorine compounds available to inactivate viruses or bacteria) at the point of entry
into the distribution systems and at points throughout the distribution system. Systems serving more
than 3,300 people would have to monitor the chlorine residual continuously, while systems under that
threshold would have to monitor one grab sample daily.
3.3.3 Option 3: Multi-Barrier Approach
The Multi-Barrier Approach builds on the sanitary survey and triggered monitoring
requirements of the first two options by including a hydrogeologic sensitivity assessment. This
assessment targets those systems where water can move quickly though the subsurface thereby
increasing the possibility of fecal contamination. The State or primacy agent will assess the sensitivity of
a system based on the well's hydrogeologic setting. EPA is currently proposing three sensitive
hydrogeologieskarst, gravel, and fractured bedrock.
EPA regards the hydrogeologic sensitivity assessment as equal in importance to sanitary
surveys in preventing or reducing microbial contamination; the Agency is, therefore, considering
requiring States to complete the sensitivity assessment within three years of the GWR effective date for
CWSs and within five years for NCWSs (i.e., six and eight years from the promulgation date). As part
of these requirements, EPA is considering requiring each State to conduct a one-time sensitivity
assessment since the hydrogeology (unlike sanitary survey components) should change little following
the initial assessment. This assessment is proposed for all existing and new systems that do not treat to
4-log virus inactivation/removal.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 3-3
-------
As a result of this assessment, a sensitive system must monitor its source water for a fecal
indicator monthly for a 12-month period. The monitoring requirement is the same as described under
triggered monitoring except that 12 routine samples are proposed rather than the one-time triggered
sample. EPA considers more routine sampling as necessary to identify episodic source water
contamination. For example, excessive rain in one month may wash contamination quickly into a karst
aquifer although it may not remain long enough to trigger a positive total coliform sample. If all samples
are negative during twelve months of monitoring, this requirement allows the State to reduce the
monitoring frequency to every quarter the system serves water to the public. The system may also
discontinue routine source water monitoring after twelve samples if the State determines that
contamination is highly unlikely or if the system switches to an alternative source that is not located in
sensitive hydrogeology.
Under the Multi-Barrier Approach, systems that are found to have source water contamination
(either through routine or triggered monitoring) would be required to treat to 4-log removal/inactivation,
eliminate the source of contamination, or provide an alternative water supply within 90 days (or longer if
the State or primacy agent approves a schedule). As discussed under the sanitary survey and triggered
monitoring option, if a system chooses treatment, it would have to ensure that it was effective by
monitoring for a disinfectant residual. This option also includes the one-time waiver for the treatment
technique as described above.
3.3.4 Option 4: Across-the-Board Disinfection
The fourth option considered by EPA would require all public ground water systems to install
and/or operate disinfection treatment processes capable of achieving a 4-log inactivation/removal of
viruses. Systems disinfecting to less than 4-log removal/inactivation of viruses would be required to
upgrade their treatment. Unlike the other options, the across-the-board disinfection option does not
consider the quality of a system's source water or potential for contamination. Similar to options 2 and
3, systems would have to monitor their treatment practices to ensure they are effective. Also, States
would be required to perform sanitary surveys of ground water systems to ensure the treatment systems
are being properly operated and to ensure there are no potential sources of contamination that cannot
be addressed by treatment.
34 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
3.4 References
EPA. 1996a. Ground Water Disinfection and Protective Practices in the United States. Office of
Ground Water and Drinking Water, Washington, D.C.
EPA. 1996b. Safe Drinking Water Act Amendments.
EPA and ASDWA. 1995. EPA/State Joint Guidance on Sanitary Surveys.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 3-5
-------
4. Baseline Analysis
4.1 Introduction
To develop forecasts of the benefits of the GWR, as well as forecasts of the economic and
financial impacts of the GWR's regulatory options on the ground water supply industry and ultimately
on customers, EPA had to develop a baseline before considering the effect of any single regulatory
option. A baseline is defined as a characterization of the industry and its operations prior to the
rulemaking.
The purpose of this chapter is twofold. First, Section 4.2 provides baseline information relative
to the GWR including the number of public water supplies affected, the population affected, and current
treatment practices. Second, Section 4.3 introduces the risk assessment modeling used to estimate
baseline health effects from contamination of ground water systems (GWS) potentially affected by the
GWR.
4.2 Baseline Profile of Public Ground Water Systems
EPA analyzed data on the number of ground water systems and the resources available to the
systems. Data inputs included the total number of affected systems, the households and populations
served by these systems, average and maximum system flow rates, operator expenses, and system
revenues and expenses. This analysis involved input from knowledgeable stakeholders and
incorporated the latest available research.
Prior to presenting baseline information for public ground water systems, it is necessary to first
define some terms used to describe water systems. EPA uses the following classifications: A public
water system is classified as either a community water system or noncommunity water system, the latter
of which is further classified as either transient or nontransient. These are defined as follows:
A public water system (PWS) is one that serves 25 or more people or has 15 or more
service connections and operates at least 60 days per year. A PWS can be publically
owned or privately owned.
A community water system (CWS) is one that serves at least 15 service connections
used by year-round residents or regularly serves at least 25 year-round residents.
A noncommunity water system (NCWS) does not serve year-round residents, but
serves at least 15 service connections used by travelers or intermittent users for at least
60 days each year, or serves an average of 25 individuals for at least 60 days a year.
NCWSs are further sub-classified into nontransient and transient systems.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 4-1
-------
S A nontransient noncommunity water system (NTNC) serves at least 25 of the same
persons over six months per year (e.g., factories, schools, office buildings, and hospitals
with their own water source).
S Transient noncommunity water systems (TNC) do not serve at least 25 of the same
persons over six months per year (e.g., many restaurants, rest stops, parks).
Public water systems are also classified by the source water they use as being either surface
water (e.g., drawn from lakes, streams, rivers, etc.) or ground water (e.g., drawn from wells or
springs). Some ground water sources (e.g., riverbank infiltration/galleries) are directly impacted by
adjacent surface water bodies and are referred to as ground water under the direct influence of surface
water (GWUDI). As noted in Section 2, the GWR does not address GWUDI systems, which are
instead subject to the requirements of the Surface Water Treatment Rule and the Interim Enhanced
Surface Water Treatment Rule.
Sources of Industry Profile Information
EPA uses the following as the two primary sources of information to characterize the universe
of ground water systems:
Safe Drinking Water Information System fSDWIS^)EPA's SDWIS contains data on all
PWSs, as reported by States and EPA regions. These data reflect both mandatory and
optional reporting components. States must report the system location, system type (CWS,
NTNC, or TNC), primary raw water source (ground water, surface water, or GWUDI), and
violations. Optional reporting fields include type of treatment and ownership type. Because
providing optional data is discretionary, EPA does not have complete data on every system for
these parameters; this is particularly common for noncommunity systems.
Community Water System Survey (CWSS^The second source of information, CWSS, is a
detailed survey of surface and ground water community water systems conducted by EPA in
1995 and published in 1997 (EPA, 1997). The CWSS is stratified to represent CWSs across
the U.S. The CWSS includes information such as the number of system operators, system
revenues, expenses, treatment practices, source water protection measures, and capacity (i.e.,
the amount of water the system is designed to deliver). The CWSS contains data from 1,980
water systems, of which 1,020 are ground water systems, 510 are surface water systems, and
450 represent purchased water systems (systems that purchase water from another PWS and
distribute this water to their customers).
4.2.1 Number of Ground Water Systems
Nationally, SDWIS indicates that there are over 156,000 public water systems that use ground
water as their primary source. The majority of ground water systems are NCWSs, with 60 percent
(93,000) transient and 12 percent (19,000) nontransient. CWSs make up the remaining 28 percent
4-2 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
(44,000) of all ground water systems. Although there are far more NCWSs, CWSs serve larger
numbers of people.
According to SDWIS (1997), 97 percent of the 44,000 CWSs, and nearly all of the NCWSs
that use ground water, serve fewer than 10,000 persons, which EPA defines as a small system.
Collectively, 99 percent of drinking water systems serve fewer than 10,000 people. About 97 percent
of the systems serve 3,300 people or fewer (for a total of 33.2 million
people). For CWSs, 78 percent serve fewer than 1,000 people (for a total of 8.1 million people), 67
percent serve fewer than 500 people (for a total of 4.6 million people), and 33 percent serve fewer
than 100 people (which is nearly 900,000 people). Exhibit 4-1 presents the number of ground water
systems in the United States by system type and population served.
Exhibit 4-1. Total Number of Ground Water Systems by
System Type and Service Population Category
Service Population
Category
100 or Less
101-500
501-1,000
1,001-3,300
3,301-10,000
10,001-50,000
50,001-100,000
>100,000
Total
Community
14,390
15,069
4,739
5,726
2,489
1,282
139
72
43,906
Nontransient
Noncommunity
9,714
6,925
1,927
686
59
11
0
0
19,322
Transient
Noncommunity
72,343
18,576
1,849
611
151
66
12
10
93,618
Total
96,447
40,570
8,515
7,023
2,699
1,359
151
82
156,846
Source: EPA, 1999a.
4.2.2 Population Served by Ground Water Systems
System population characteristics are important to this analysis for several reasons. It is
important to know the total population served by ground water systems so that the distribution of costs
and benefits of the GWR can be addressed. As presented in Exhibit 4-2, ground water CWSs serve
more than 88.7 million people, while ground water NCWSs serve about 20.2 million people. Overlaps
do occur, as individuals may be served by both types of systems, as well as systems providing a
combination of ground and surface water. For example, a person may be served by a surface water
CWS at home and by a ground water NCWS at work or at a restaurant. It should be noted that there
does not appear to be a consistent reporting standard for populations served by transient systems. In
addition, some States may report the total population served by a system over a year, while others may
report the average population served each day.
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
4-3
-------
Exhibit 4-2. Total Population of Ground Water Systems
uy o
Service Population
Category
100 or Less
101-500
501-1,000
1,001-3,300
3,301-10,000
10,001-50,000
50,001-100,000
>100,000
Total
Community
864,784
3,737,617
3,491,873
10,542,514
13,956,242
26,653,078
10,799,952
18,668,871
88,708,460
raiem i ype emu oi£e
Nontransient
Noncommunity
507,200
1,783,574
1,387,818
1,118,678
338,457
161,827
0
0
5,297,554
Transient
Noncommunity
3,160,753
4,109,429
1,410,457
1,075,745
829,854
1,675,640
891,000
1,782,667
14,935,545
Total
4,526,266
9,630,620
6,290,148
12,736,937
15,124,553
28,490,545
11,690,952
20,451,538
108,941,559
Source: EPA, 1999a.
4.2.3 Treatment Profile
This section presents information regarding the disinfection treatment technologies that are
currently implemented by ground water systems. This information is used to develop an assessment of
the number of systems that may need to install or modify treatment to meet new GWR requirements.
These data are also used in the RIA's cost analysis and in the determination of the compliance
monitoring burden. An analysis of current treatment effectiveness is necessary to establish a baseline of
finished water exposure and as a baseline to determine how much additional corrective treatment is
needed to meet GWR requirements.
Because ground water systems may employ more than one water supply source, they may have
more than one treatment facility. Therefore, the analysis of types of treatment in place must be
performed on an entry point/treatment facility level rather than on a system level. To determine the
number and percentage of entry points to community ground water systems, EPA analyzed responses
to the CWSS regarding current treatment practices. EPA considered a ground water entry point with
one or more disinfection treatments (e.g., chlorine gas, reverse osmosis) to be a disinfected facility.
Once EPA made this determination, the Agency calculated the percentage of ground water that is
disinfected for each population category. Exhibit 4-3 displays the percent of ground water treatment
facilities disinfecting, flow-weighted by service population category.
4-4
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit 4-3. Percent of Community Ground Water Treatment Facilities
Disinfecting by Service Population Category (Flow-Weighted)
Service Population Category
100 or less
101-500
501-1,000
1,001-3,300
3,301-10,000
10,001-50,000
50,001-100,000
100,000-1,000,000
% Facilities Disinfecting,
Flow-Weighted
45
73
76
75
74
91
64
81
Note: Flow-weighted calculation of the percentage of community water system flow which receives
disinfection treatment.
Source: EPA, 1997.
Because there currently are no surveys equivalent to the CWSS that define treatment practices
in noncommunity systems, the EPA compiled estimates from State public drinking water officials on the
percentage of systems in their State that currently disinfect. Exhibit 4-4 presents the estimates for
noncommunity systems. These estimates vary in their methods and levels of accuracy.
4.3 Baseline Health Effects
EPA has developed a risk assessment model to estimate the baseline number of illnesses and
deaths associated with ingesting pathogenic viruses in public ground water systems. This GWR risk
assessment follows the standard methodology developed by the National Research Council (NRC,
1983). The NRC defined three steps for risk assessment of contaminants in drinking water:
1) exposure assessment,
2) hazard identification, and
3) health effects assessment (risk characterization).
The method of calculating illnesses and deaths from exposures to waterborne pathogens in
ground water is presented in Exhibit 4-5. As shown in the figure, both exposure factors and the
pathogenicity of each organism are factored into the estimate of health effects in the exposed
population.
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
4-5
-------
Exhibit 4-4. Percent of Noncommunity Ground Water
Systems Disinfecting by State
State
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgia
Hawaii
Idaho
Illinois
ndiana
Iowa
Kansas
Kentucky
_ouisiana
Vlaine
Maryland
Massachusetts
Vlichigan
Vlinnesota
Mississippi
Missouri
NTNC
100
171
18
-100
8
952
(unknown)
17
100
66
100
o
~o
n2
<5
-15
100
100
23
4
10-15
3
-6
-3
46
17
TNC
82
4
10
-70
7
952
(unknown)
9
100
46
83
-S-O
""^
112
<5
__c
"*"O
100
100
9
6
-0
1
-3
-1
26
25
State
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
NTNC
8
8
14
4
-S-R
<50
23
0
31
2
19
1
16
35
70
100
33
__c
"*"O
14
41
95
5
17
TNC
4
1
-0
2
__O
""O
3720
<50
2
10
12
7
17
5026
1
6
19
52
100
17
-0
9
24
-50
<1
13
1 State combined the Community and NTNC numbers.
2 State combined the NTNC and TNC numbers.
Source: EPA, 1996.
4-6
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit 4-5. Risk Assessment Process for Pathogens in Drinking Water
EXPOSURE ASSESSMENT
Pathogen Occurrence in Source
Water (Hit Rate))
Concentration of Pathogen in Source
Water
Concentration of Pathogen Removed
or Inactivated During Treatment
Concentration of Pathogen in
Finished Water, Available for
Consumption
Potentially Exposed Population
(Includes Sensitive Subgroups)
Daily Intake
(Liters per Day Consumption)
Daily Dose
Annual Intake
(Days per Year Exposure)
(adapted from NRC, 1983)
^ HAZARD IDENTIFICATION «^
RISK CHARACTERIZATION
Infectivity: Dose-Response
Relationship (Probability of
Infection given Exposure)
Individual Daily Probability of
Infection
Individual Annual Probability
of Infection
Morbidity [Probability of Illness
Given Infection; Includes
Secondary Spread]
Annual Number of Illnesses
(Individual Risks x Population)
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
4-7
-------
Mortality
[Probability of Death Given
Illness]
Annual Number of Deaths
(Individual Risks x Population)
Exposure
Pathogenicity
Health Effect
The proposed GWR is expected to reduce the current incidence of illness caused by a wide
variety of viral and bacterial pathogens that are associated with fecal contamination of ground water.
Different pathogens cause different illnesses and each pathogen has a different probability of causing
illness or death. Medical research has not isolated all waterborne pathogens, nor has it thoroughly
characterized the pathogenicity or ability of these organisms to infect humans and cause them to
become ill or to die. EPA has selected two well-characterized viral pathogens having different
infectivity, morbidity, and mortality rates to represent a wide range of waterborne viral pathogens for
the GWR risk assessment. The selected representative viruses, rotavirus and echovirus, are described
as follows.
Rotavirus is a highly infectious virus that is a common cause of vomiting and diarrhea,
especially in children, but does not frequently result in life-threatening illness in the general
population of industrialized countries. For the purposes of this analysis, rotavirus represents a
large group of viruses referred to hereafter as Type A viruses. These viruses are suspected to
cause outbreaks of gastroenteritis in PWS drinking water supplies. These viruses include:
Norwalk, Norwalk-like small round structured viruses, caliciviruses, adenovirus, astrovirus, and
other enteric viruses.
Echovirus is a small RNA enterovirus that represents the group of waterborne viruses
referred to hereafter as Type B viruses that, although not highly infectious, may result in severe
health effects when illness occurs. Coxsackie viruses and Hepatitis A virus (HAV) are other
examples of Type B viruses.
The following sections present EPA's assumptions and methodology used in: 1) estimating
exposure, 2) characterizing hazards (i.e., pathogenicity), and 3) calculating the health effects of Type A
and Type B viruses under current baseline conditions.
4.3.1 Exposure Assessment
Exposure assessment is the first step in the risk assessment process. The exposure pathway
addressed in this risk assessment is ingestion of drinking water from public ground water supplies that
are contaminated with microbial pathogens from fecal pollution.
CDC surveillance data show that pathogens in ground water systems have often been the cause
of waterborne disease outbreaks (WBDOs), clusters of cases of acute gastroenteritis or other illness
4-8
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
that are reported to CDC through public health agencies. Low levels of pathogens also may cause
endemic (unreported, acute or chronic) disease in the exposed population. EPA believes that there are
three key exposure scenarios whereby drinking water delivered to consumers in public ground water
systems can be contaminated with fecal pathogens:
1) source water contamination in untreated systems;
2) source water contamination in disinfecting systems with treatment failures; and
3) distribution system contamination.
The remainder of this discussion of exposure assessment for the GWR addresses only the first
contamination scenario. Though not considered as significant a threat as fecal contamination of source
water for untreated systems, EPA's modeling of this first scenario also includes exposure from systems
that do treat (disinfect), taking into account the reduction in microbial levels from that treatment.
EPA believes there is insufficient information on the frequency and severity of drinking water
treatment failures and distribution system contamination events in GWSs to directly model the latter two
scenarios. However, the proportions of reported WBDOs caused by the three contamination
scenarios, as shown previously in Exhibit 2-A, serve as an indicator of the relative frequency of ground
water system contamination events that cause disease in exposed populations. These outbreak
proportions are factored into the estimates of total annual illnesses and deaths due to ingestion of
drinking water from public GWSs.
EPA distinguishes two types of ground water systems or points of entry with respect to the
likelihood and severity of source water contamination:
1) those with wells constructed in accordance with State requirements (hereinafter referred to
as properly-constructed wells). EPA estimates that properly-constructed wells comprise
83 percent of systems/points of entry, based upon data from ASDWA's Survey of Best
Management Practices for Community Ground Water Systems (ASDWA, 1997); and
2) those with poorly-constructed wells, which are assumed to comprise the remaining 17
percent of ground water systems/points of entry.
Whether the systems are properly or poorly constructed, the assessment of exposures to
pathogens from ground water sources requires that the following factors be quantified based on survey
data or best professional judgment:
the occurrence (presence/absence) of pathogens in source water;
the concentration of pathogens in source water when it is contaminated;
the level of pathogen inactivation in the system and resulting pathogen concentration
in tap water;
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 4-9
-------
the size of the exposed population, including sensitive subgroups; and
the volume of water ingested daily and how many days per year it is ingested.
EPA evaluated available occurrence and exposure data and developed assumptions regarding
these exposure factors, each of which is discussed briefly below.
Concentration of Pathogen in Source Water
Virus occurrence and virus concentration assumptions for this exposure assessment are based
on occurrence data from two recent U.S. ground water source surveys:
1) the AWWARF study (Abbaszadegan et al., 1998 American Water Works Service Co.): a
survey of mainly properly-constructed wells; and
2) the EPA/AWWA study (Lieberman et al., 1995): a survey of known contaminated wells,
assumed to be representative of poorly-constructed wells.
Both studies used a number of complementary analytical methods to detect and enumerate
microbial pathogens, including: enterovirus cell-culture, nucleic acid detection by reverse transcription-
polymerase chain reaction (RT-PCR), and standard fecal indicator methods. Viral occurrence is
estimated as the percent of wells that were found contaminated with virus during one or more sampling
events. Both studies found viral pathogens in a small fraction of the wells tested. The number of wells
identified as virus-positive in the surveys tended to vary with the specificity and sensitivity of the
methods used. For example, in the AWWARF study, 4.8 percent of wells were enterovirus-positive
by the cell-culture assay, whereas 15.9 percent of wells were enterovirus-positive by the RT-PCR
assay (final RT-PCR results are not available from the EPA/AWWA study).
Because the cell-culture method detects only intact, infective viruses, EPA used the fraction of
wells positive by the enterovirus cell-culture method as the preferred measure of viral pathogen
occurrence. However, because of the method's virus-specificity (i.e., only Type B viruses are typically
detected by the cell-culture method used in both studies), the ratio of cell-culture to RT-PCR-positive
wells was used to calculate Type A virus occurrence.
Viral occurrence estimates and supporting assumptions are summarized in Exhibit 4-6. The
exhibit also presents the concentration distribution of viral pathogens in contaminated source wells as
estimated using enterovirus data from the AWWARF study (for properly-constructed wells) and the
EPA/AWWA study (for poorly-constructed wells). As shown in the exhibit, contaminated, poorly-
constructed wells are assumed to have a mean concentration expressed in most probable number of
infectious units of virus per 100 liters of water of 29.41 ± 55.7 MPNIU/100 L, in comparison with a
mean concentration of 0.356 ± 0.297 MPNIU/100L in contaminated, properly-constructed wells.
EPA recognizes that there may be reasonable alternative approaches to interpreting the above
data and to using these data in the occurrence assessment to support the risk analysis. For example,
4-10 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
there may be merit to making more use of the male-specific and/or somatic coliphage results from the
AWWARF and EPAVAWWA studies than is currently done in the risk analysis. Also, additional data
may be available from other sources to enhance the data and analysis used here. EPA requests that
interested parties provide any additional information or suggestions for alternative methods for using the
existing information during the comment period following the proposal of the GWR.
Pathogen Inactivation in Ground Water Systems and Resulting Tap Water Concentrations
For the purposes of this exposure assessment, EPA assumed that the pathogen concentration in
tap water from undisinfected ground water systems is the same as the pathogen concentration in source
water. In contrast, properly-operating disinfecting systems are assumed to inactivate 99.99 percent, or
4-logs, of viral pathogens. Therefore, the concentration of pathogens in tap water from properly-
operating disinfecting systems is assumed to be 0.01 percent of the concentration in source water.
It is possible that viruses may be naturally attenuated to some extent while passing through
untreated ground water distribution systems. It is also known that treatment failures and distribution
system contamination events occur from time to time in disinfecting systems. However, EPA believes
there are insufficient data to directly quantify and model these events in the GWR model. Section 4.3.4
discusses the methods for indirectly estimating illness due to treatment and distribution system
contamination.
Exhibit 4-6. Viral Occurrence and Concentration in Source Water
Well Quality
Properly-Constructed Wells
(83%ofallGWSs)
Poorly-Constructed Wells
(17%ofallGWSs)
Virus Type
Type A virus
Type B virus
Type A virus
Type B virus
Percent of Wells
Contaminated
4.4 percent2
4.8 percent3
5.5 percent4
6.0 percent5
Mean Virus Concentration when
Contaminated(MPNIU1/100 L)
0.356 ±0.2976'7
0.356 ±0.2977
29.41 ±55.76'8
29.41 ±55.78
1 Most probable number of infectious units of virus.
2 AWWARF study: The RT-PCR methods detected the presence of rotavirus nucleic acids in 14.6 percent of
wells tested to which the ratio of enterovirus cell-culture to RT-PCR positive wells (0.3) was applied.
3 AWWARF study: The AWWARF study found that 4.8 percent of wells tested were positive forthe presence of
enteroviruses using the Buffalo Green Monkey (BGM) cell culture assay.
4 EPA/AWWA study: Because there are no rotavirus data available from the EPA/AWWA study at this time, it is
assumed that rotavirus (Type A virus) and echovirus (Type B virus) occur in poorly constructed wells in the same
ratio as calculated for properly constructed wells (4.4/4.8 = .92; .92 x 6.0 = 5.5).
5 EPA/AWWA study: Calculated by dividing the total number of positive BGM cell culture assays by the total
number of assays performed.
6 Because there are no concentration data for rotavirus available from either study, it is assumed that the mean
concentration of Type A virus in properly-constructed wells is the same as for Type B virus.
7 AWWARF study: Range of enterovirus (Type B virus ) concentrations in cell-culture isolates was 0.123 to 1.86
MPNIU/100 L; data are fitted to a lognormal distribution from which the mean and standard deviation are
calculated.
8 EPA/AWWA study: Range of enterovirus concentrations in cell-culture isolates was 0.9 to 212 MPNIU/100 L;
data are fitted to a lognormal distribution from which the mean and standard deviation are calculated.
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
4-11
-------
Potentially-Exposed Populations. Including Sensitive Subgroups
The populations served by all types of systems were presented previously in Exhibit 42;
Exhibit 4-7 presents the numbers of persons in the general population served by undisinfected ground
water systems (the subject of the GWR risk model calculations).
Exhibit 4-7. Populations Served by Undisinfected Ground Water Systems
Service
Population
Category
<100
101-500
501-1,000
1001-3,300
3,301-10,000
10,001-50,000
50,001-100,000
>100,000
Totals
Estimated Population Served Undisinfected Ground Water1
Community
Water
Systems (CWS)
471,214
1,005,419
852,017
2,667,256
3,670,492
2,309,365
3,898,783
3,472,410
18,346,956
Nontransient
Noncommunity
(NTNC) Systems
364,978
1,283,450
998,666
804,994
243,552
116,450
0
0
3,812,090
Transient
Noncommunity
(TNC) Systems
2,616,086
3,401,284
1,167,404
890,370
686,852
1,386,890
737,461
1,475,474
12,361,821
' Source: GWR model calculation (EPA 1999a).
Within the general population, there are sensitive subgroups, that is, groups of individuals who
may suffer more serious symptoms when they become ill as a result of exposure to pathogens in ground
water. Sensitive subgroups also may have a higher probability of mortality when they are ill, thus
suffering a disproportionate burden of the health risks from exposures to contaminated ground water.
Criteria distinguishing sensitive subgroups in this exposure assessment include:
Age: very young children (< 5 years) and elderly adults (> 65 years) are sensitive to many
viral pathogens; the exposure assessment uses statistics from the Bureau of Labor, 1990
census to distribute by age the populations in each type and size of ground water system;
and
Immunocompromised health status: AIDs patients, organ transplant patients,
nonhospitalized persons receiving cancer therapy, and nursing home patients are considered
sensitive to viruses based on compromised immune status. The first three groups comprise
approximately 1.0 percent of the general population, assumed to be divided equally among
all age groups. Nursing home residents constitute another 0.6 percent of the general
population, and are assumed to be older than 16 years.
Drinking Water Consumption Factors
4-12
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
The amount of drinking water consumed daily by individuals is a key input to the risk analyses
supporting all EPA drinking water MCL and MCLG regulations and determinations. The higher the
average volume of contaminated water consumed, the higher the average daily dose of pathogens.
Daily intake assumptions for this exposure assessment are age-based; that is, the amount of
water consumed daily varies by age group. Custom daily intake distributions developed for the age
groups used in this analysis use data from the 1994 1996 USDA, Continuing Survey of Food
Intakes by Individuals. (EPA, 2000) In this analysis, EPA selected the consumption distribution
identified as "all sources, consumers only" having an overall mean of 1.24 L/day (0.30 at the 10th
percentile and 2.35 L/day at the 90th percentile) to represent an upper bound estimate of daily
consumption values. Results of alternative model calculations using USDA consumption data for
"community water supply, all respondents" (mean of 0.927 L/day) are presented in Appendix A as a
lower bound estimate.
In addition to daily consumption, EPA included estimates of the number of days per year tap
water is consumed by users of different types of water systems. EPA's estimates for exposure days are
presented in Exhibit 4-8.
Exhibit 4-8. EPA Estimates for Exposure Days1
Type of System
Community Water Systems
Noncommunity Nontransient
Transient Noncommunity
Exposure Days Per Year
350
250
15
1 Number of days in which tapwater is consumed.
4.3.2 Hazard Identification
Hazard identification is the second step of the risk assessment process. Hazard identification
addresses the pathogenicity of each drinking water pathogen to the potentially exposed population.
Factors considered in this assessment include:
infectivity (the ability of a microorganism to colonize the body of the host),
morbidity (the probability of illness given infection), and
mortality (the probability of death given illness).
Infectivity assumptions are based on dose-response curves generated from challenge studies in
healthy adult volunteers. Morbidity rates are based on epidemiological studies, and mortality is
estimated on observed case-fatality ratios. Morbidity and mortality may vary with the age and overall
sensitivity of the receptor population. Pathogen hazard assumptions for Type A and B viruses when
ingested in ground water are summarized in Exhibit 4-9.
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
4-13
-------
Exhibit 4-9. Hazard Identification of Viral Pathogens for the
GWR Risk Assessment
Pathogen
Hazards
Definition
Model
Type A virus
Type B virus
Infectivity
Infectivity is the ability of
the pathogen to colonize
the host; it is defined by
dose-response
relationship.
General model of dose-
response (beta-Poisson):
P(I)= l-(l+N/P)-a
where:
P (I)= probability of
infection
N = number of pathogenic
viruses ingested
a, P= pathogen-specific
rate constants.1
Highly infective virus;
a = 0.26, P=0.422
Moderately infective
virus;
a = 0.374, P=1873
Morbidity
Primary morbidity is
the probability of
illness given infection;
can vary in sensitive
subgroups.
<2yrsold=0.88"
>2yrs = 0.15
<5yrsold =0.5'
> 5 to 16 years = 0.57'
> 16 years = 0.33'
Secondary spread is
the probability of
illness given contact
with a (primary) ill
person.
<2yrsold =0.55"
> 2 yrs old = 0 "
Triangular
distribution (all age
groups), from 0.11 to
0.55; mode = 0.35s
Mortality
Mortality is the
probability of
death as a
result of
illness.
7.3 xlO'66
< 1 month =
0.0092 ''
> 1 month =
0.0004110
1 Regli et al., 1991; 2 Ward et al., 1986; 3 Schiff et al., 1984; 4 Kapikian and Chanock, 1996; 5 Wenman et al., 1979
and Foster et al., 1980; 6 CEOH 1998; 7 Hall 1980; 8 Morens et al., 1991; 9 rate of mortality given infection; 10
Stedge, 1998.
For Type A viruses in children < 2 years and for Type B viruses in all age groups, the hazard
identification part of the model also includes a factor for secondary spread. Secondary spread refers to
contracting the illness through exposure to a person who became ill after exposure via ingestion of
contaminated ground water. No secondary spread data are available for viral infections explicitly
acquired via the ground water pathway. Nevertheless, secondary spread of waterborne illnesses is a
reasonable assumption because the pathogens of concern for the GWR are also commonly transmitted
by respiratory or direct contact (fecal-oral) pathways.
4.3.3 Sensitive Subgroups
Although it is generally believed that most persons are equally vulnerable to repeated infection
(i.e., colonization) by viruses and other microorganisms during their lifetime, factors such as being very
4-14
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
young, very old, or immunocompromised can determine whether severe illness follows infection and
whether death is the outcome of severe illnesses. The hazard identification factors presented in Exhibit
410 and incorporated in the GWR model include higher morbidity and/or mortality rates for some age
groups, typically for neonates (Type B viruses) and children (both Type A and Type B viruses).
The Very Young
The very young (e.g., infants less than one month old) are generally considered to be relatively
more sensitive to severe symptomatic illness and death from gastroenteritis and other waterborne viral
and bacterial illnesses. Some viral pathogens such as coxsackie virus B (a Type B virus) can be
transmitted transplacentally from an infected mother to her child in utero, during birth, or shortly
thereafter. This type of transmission places the infected newborn infant at risk of severe symptomatic
illness from meningitis or myocarditis, for which the case fatality rates are high (Gerba et al, 1996a).
Viral gastroenteritis, caused mainly by Type A viruses, is prevalent among U.S. children.
Primary or secondary transmission by the fecal-oral route contributes to high rates of illness in group
settings such as day-care centers that include diapered children. CDC has determined that the
incidence of rotavirus diarrhea can reach 0.30 episodes/child/year by age two, with a cumulative
incidence approaching 0.80 episodes/child by age five (Glass et al., 1996). Hospitalizations for
rotavirus diarrhea are most common in children six months to three years of age (Parashar et al., 1998),
while self-limiting Norwalk-like virus infections are prevalent in school-age children (LeBaron et al.,
1990). Although deaths from infectious diarrhea have generally declined among U.S. children since
1965 because of re-hydration therapy, newborn children, especially infants born prematurely, remain at
risk of death from severe diarrheal illness (Kilgore et al., 1995).
The Elderly
The elderly (persons over 65 years of age) are also at greater risk than the general population
of experiencing severe health effects from rotavirus diarrhea, hepatitis and other viral infections.
Sensitivity among persons in this age group is due to declining immunity and poorer general health
(Gerba et al., 1996a and b; Lew et al., 1991). Conditions such as cardiovascular disease make the
elderly more susceptible to complications of diarrhea such as electrolyte imbalance, dehydration, and
shock (Maasdam and Anuras, 1981, cited in Lew et al., 1991). More than half of the diarrheal deaths
that occur in the U.S. are among persons older than 74 years of age, and the risk of death from
diarrhea is generally higher among elderly persons confined to nursing homes and other care facilities
(Lew et al., 1991; Gerba et al., 1996 b). Thirty percent of diarrheal deaths among the elderly occur in
nonhospitalized patient care settings (e.g., nursing homes), although only 10 percent of persons in this
age group live in such settings (Lew et al., 1991).
The Immunocompromised
Immunocompromised and immuno-suppressed persons comprise a non-age-based population
sub-group who are sensitive to serious health effects from viral and bacterial infections. Although some
viral pathogens (e.g., hepatitis A and Norwalk virus) do not cause more severe illness or risk of death
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 4-15
-------
in immunocompromised persons, chronic diarrhea is a serious complication of AIDs, and rotavirus and
adenovirus are commonly isolated from stool samples from AIDs patients with diarrhea (Gerba et al,
1996a).
A limited number of available studies suggest that viral infections can contribute to deaths in
immunocompromised persons. Hierholzer (1992) reported case fatality ratios of 53 to 69 percent in
cancer-immuno-suppressed and bone-marrow transplant patients infected with adenovirus. Enteric
rotavirus, coxsackie virus and adenovirus infections were reported as the cause of similar case fatality
ratios among bone-marrow transplant patients in an earlier study (Yolken et al., 1982, cited in Gerba et
al., 1996a).
At this time, the morbidity and mortality factors assigned to immunocompromised subgroups for
Type A and B virus infections are the same as those used for the general population. EPA believes
there are insufficient data available for these subgroups to assume higher morbidity or mortality from
waterborne infections based on immune status. However, because there is a higher cost-of-illness
among severely immunocompromised persons having lengthy viral illnesses, the with-rule reductions in
numbers of illnesses and deaths in this subgroup are calculated by the risk model and are monitored
separately in the benefits calculations.
4.3.4 Risk Characterization
The third step of the risk assessment process is risk characterization. This section summarizes
the methods of calculation used to derive baseline estimates of health effects in the exposed
populations. Model calculations used to estimate annual illnesses and deaths due to source
contamination in undisinfected systems are presented first, followed by an explanation of the factors
used to estimate additional illnesses and deaths due to treatment failures and distribution contamination.
Finally, the results of the baseline (without rule) risk calculations are presented.
Risk Assessment Methodology
The modeling of baseline risks was performed using a two-step Monte Carlo simulation analysis
designed to provide both a "best" estimate of the number of illnesses and deaths due to viruses in
ground water, and an estimate of the uncertainty bounds around those values.
The simulation analysis was structured to incorporate two steps to separately address the
variability and uncertainty aspects of the input parameters in the algorithms used to calculate risk.
The first step of the simulation analysis used information characterizing the variability of viruses
in drinking water from ground water supplies together with information characterizing the variability in
drinking water consumption in the exposed population to arrive at an estimate of the distribution of
individual annual risks of infection. This distribution of individual risks of infection also provided an
estimate of the average annual risk of infection for the overall population and of the uncertainty in that
average risk.
416 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
The second step of the simulation analysis used the results of the first step along with factors
characterizing morbidity (illness given infection) and mortality (death given illness), together with the
number of people exposed from various types of ground water systems to arrive at the total national
estimate of illnesses and deaths due to viruses in ground water. The second step combined the average
annual risk of infection and its associated uncertainty with other factors to provide an estimate of the
total number of annual illnesses and deaths in the exposed population, as well as an estimate of the
uncertainty in those total numbers.
The following provides further detail for each of the two steps.
Step 1. Estimation of the Distribution of Individual Annual Risks of Infectioa
Average Individual Annual Risk of Infection, and Uncertainty (Standard
Error) in the Average Individual Annual Risk.
The basic risk assessment algorithm used in this analysis to compute individual annual risk is
shown below:
s ^.*\
(4-1)
Specifically, this algorithm provides an estimate of the probability that an individual will become
infected during a one-year period given an average daily risk of infection ofPDaily and D days of
exposure during the year.
As discussed in the preceding section on exposure, the number of days of exposure (drinking
water consumption) differs for exposure between community water systems (350 days), nontransient
noncommunity systems (250) days and transient noncommunity systems (10) days.
The individual daily risk of infection, PDai}y, incorporated in Equation (4-1) is calculated as:
(4-2)
The individual daily risk of infection is the probability that an individual will become infected
from consuming N viruses in water that day. The variables &, and P are pathogen-specific
dose-response model parameters. The values for a and P for the Type A and Type B viruses model in
this analysis are shown in Exhibit 4-9.
The daily ingestion of viruses for any individual is the product of the concentration of viruses in
drinking water and amount of drinking water consumed. There is, of course, variability in the expected
level of viruses in ground water from one system to the next, as well as variability in drinking water
consumption from one individual to the next. The Monte Carlo simulation performed in the first step of
the analysis incorporated both of these exposure factors as distributions to reflect that variability.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 4-17
-------
The variability in virus concentrations in ground water systems was characterized by lognormal
distributions using the mean and standard deviations for Type A and Type B viruses in properly and
poorly constructed wells as shown in Exhibit 4-6 as the parameters for those distributions.
The variability in drinking water consumption was characterized by age-specific custom
distributions derived from the information provided in the CSFn data discussed in the exposure section.
In addition to the occurrence distribution and consumption factors, the calculation of N also
includes assumptions concerning hit rate (portion of wells with viruses present), the fractions of
disinfected and undisinfected wells, the fraction of properly and improperly constructed wells, treatment
effectiveness in disinfected wells, and the fraction of viable viruses (assumed in this analysis to be 1.0).
There are three key outputs from this first step of the analysis
The distribution of individual risks, used for the risk characterization of the portion of the
population expected to experience risks at various levels (including risks to sensitive
subpopulations);
The mean or average individual risk, used in the Step 2 of the analysis as discussed below to
scale up the individual risks to the entire population in order to estimate the number of cases of
illness;
The standard error in the estimated mean individual risk, also used in Step 2 as a contributor to
the uncertainty in the estimated number of cases of illness.
Step 2. Estimation of Cases of Illness and Death in the Exposed Population and
Uncertainty Bounds on the Estimated Cases.
In the second step of the Monte Carlo analysis, the number of cases of illness and death in the
exposed population are calculated from the average individual risk of infection obtained in Step 1, the
morbidity factors that characterize the probability of becoming ill given an infection, secondary spread
factors, the mortality factors that characterize the probability of death given an illness, and the total
population exposed to whom the annual individual risk of infection applies.
The basic algorithm for calculating the number of illnesses in Step 2 is:
Illlness = Pop x PAnnAvg x MP x (l + Ms) x (l + Mo«) (4.3)
In Equation 4-3, Pop is the population exposed, PAnnAvg is me average annual indivdual risk of
infection obtained in Step 1, MP is the morbidity factor for primary illness given infection, Ms is the
secondary spread of illness factor, andMDis is the factor for computing additional cases of illness due to
treatment failure in disinfected systems and due to other distribution system sources (see further
discussion later in this section).
4-18 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
The algorithm for calcualting deaths in Step 2 is:
JLDeath = ^Illness x MM
(4-4)
In Equation 4-4, MM is the mortality factor reflecting the number of deaths expected among
those who become ill from viral infections. These factors are also provided in Exhibit 4-9.
In addition to providing "best" estimates of total illnesses and deaths, this second step of the
analysis also provides an estimate of uncertainty in those estimates that reflect the uncertainty estimated
in the average individual risk of infection obtained in Step 1, and uncertainty in one of the morbidity
factors used to calculate illnesses. Specifically, the secondary spread factor for Type B viruses for all
age groups was included as an uncertainty distribution (triangular distribution, see Exhibit 4-9).
Exhibit 4-10 provides an additional summary of the risk calculation factors and indicates
whether they were incorporated as a variability distribution, uncertainty distribution, or constant in the
Monte Carlo simulation.
Exhibit 4-10. Summary Table of Risk Calculation Factors Used & the Distribution
Category (Variability, Uncertainty, Constant) Used in the Simulation Analysis
Risk Calculation
Factor
Occurrence Hit Rate
Occurrence
Distribution
Fraction of
Disinfecting and
Undisinfecting
Sytems
Log Removal for
Disinfecting
Systems
Viability
Drinking Water
Consumption
Description and Use in Calculations
Used in Step 1 to characterize the
fraction of systems (and therefore
of population) having viruses
present in the source water.
Used in Step 1 to characterize the
concentrations of viruses in source
water.
Used in Step 1 to separate those
systems current practicing
disinfection from those that do not.
Used in Step 1 , an assumed 4-log
removal of virus concentration in
source water for those systems
practicing disinfection.
Used in Step 1 to indicate the
fraction of viruses in water
considered to be infectious
(assumed here to be 1 .0).
Used in Step 1 to characterize the
daily water consumption by various
age groups in the exposed
population.
Distribution -
Variability
X
X
Distribution -
Uncertainty
Constant
X
X
X
X
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
4-19
-------
Dose-response
Equation Parameters
Days of
Consumption
Average Annual
Individual Risk of
Infection
Population Served
Primary Morbidity
Factors
Secondary
Morbidity Factors
Disinfection Failure
Illness Factors
Distribution System
Illness Factors
Used in Step 1, empirically derived
parameters in equations used to
calculate daily and annual risk of
infection.
Used in Step 1 to indicate the
number of days per year an exposed
individual consumes water from
ground water sources.
Product of the Step 1 calculation
used in Step 2 to calculate cases of
illness and death.
Used in Step 2 to scale up the
annual individual infection risks to
total cases of infection, and
ultimately to total cases of illness
and death in the exposed
population.
Used in Step 2 to estimate the
number of illnesses per infection in
the exposed population.
Used in Step 2 to estimate the
number of additional illnesses
resulting from contact with
indivdiuals becoming ill through
primary consumption of drinking
water.
Used in Step 2 to estimate an
additional number of illnesses due
to treatment failure in disinfecting
systems added to the primary and
secondary illnesses from source.
Used in Step 1 to estimate an
additional number of illnesses due
to distribution systems sources
added to the primary and secondary
illnesses from source water
contamination.
X
X
(for Type B
virus)
X
X
X
X
X
(for Type A
virus)
X
X
It is important to recognize that the two-step procedure for calculating the number of cases of
illness and death in the population from exposure to viruses in ground water was carried out separately
for the Type A and Type B virus categories (reflecting different occurrence distributions and
dose-response relationships), different age groups (reflecting different morbidity and mortality factors),
different water system sizes (reflecting different numbers of people served), and different water system
types (reflecting different exposure days of consumption per year for community and noncommunity
systems). The results of these many separate estimates of risk and cases of illness and death were then
aggregated to obtain the overall estimates presented in Exhibits 4-12 (for Type A viruses) and 4-13
(for Type B viruses).
4-20
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
In presenting the results of this two-step procedure for computing the baseline illnesses and
deaths as shown in Exhibits 4-12 and 4-13, the best estimate is the mean of the iterations run in the
second step. The uncertainty in that estimate is characterized by the 10th percentile and 90th percentile
values obtained from those iterations. These imply that, given the uncertainty factors explicitly included
in the second step, there is a 10 percent chance that the actual number of cases falls below the 10th
percentile value, and a 10 percent chance that it falls above the 90th percentile value.
Illnesses and Deaths in Undisinfected and Disinfected Systems (exclusive of treatment failures)
The GWR model described above calculates annual numbers of illnesses and deaths due to
source contamination in undisinfected systems and in disinfected systems (assuming 4-log removal)
exclusive of outbreaks due to treatment failure. Exhibit 4-11 summarizes the model calculations, which
incorporate the model assumptions regarding drinking water exposure to pathogens from contaminated
sources (Section 4.3.1) and health hazards from Type A and Type B viral pathogens (Section 4.3.2).
Exhibit 4-11. Summary of GWR Baseline Risk
Calculations for Undisinfected Systems
Health
Effect
Calculation
Summary
Infection
Mean Individual
Daily Probability of
Infection
The model calculates the mean individual daily probability of infection using: the fraction of contaminated wells
(virus hit rate); the potentially exposed population; variable distributions of virus concentration in undisinfected
drinking water (same as source water concentration for untreated systems) and daily intake; and the rotavirus
dose-response rate constants for Type A virus or the echovirus rate constants for Type B. The probability of
infection given a dose of one of these pathogens is:
where: P(l) = probability of infection, N = numbers of pathogenic viruses ingested, and and = pathogen-
specific rate constants. The mean and standard deviation of the mean are calculated.
Mean Annual
Probability of
Infection
The model calculates the annual probability that an individual in the population category will be infected at least
once:
P (I *) = 1-[1-P (I)] d's
where: P (I Ann) = the annual probability of infection, P (I) = the mean daily probability of infection. The
cumulative geometric function incorporates the annual number of days of exposure (i.e., 350 days in CWSs,
250 days in NTNC systems, and 15 days in TNC systems).The mean and standard deviation of the mean
are calculated for each age group and type and size of ground water system.
Morbidity
Annual Number of
Illnesses
The annual number of illnesses is the annual number of infections multiplied by the fraction of infections causing
disease (i.e., morbidity rate), calculated for each age group (see Exhibit 4-10) and type of system. This
calculation incorporates a factor for secondary spread, as appropriate. The model applies the secondary
spread factor by age group as follows: secondary illnesses = (the age-specific number of primary illnesses) x
(rate of secondary spread).
Mortality
Annual Number of
Deaths
Deaths due to Type A virus in all age groups are calculated by multiplying the annual number of primary and
secondary illnesses by the case fatality rate of 7.3 per million cases of illnesses. For Type B virus, deaths in
the neonate (< 1 month) population are calculated by multiplying the annual number of primary and secondary
infections by 0.92 percent. Deaths in all other age groups are calculated by multiplying the annual numbers of
primary and secondary illnesses by the composite case fatality rate of 0.041 percent.
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
4-21
-------
Additional Illnesses and Deaths Resulting from Treatment Failures and Distribution System
Contamination
The number of illnesses and deaths resulting from treatment failure and distribution system
contamination is calculated using Waterborne Disease Outbreak Data (Craun, 1998) and the
estimates from the GWR risk assessment of illness in undisinfected ground water systems. Of the viral,
bacterial and unknown agent outbreak-related illnesses reported in ground water systems during
1991-1996, 2,924 occurred in populations drinking untreated tap water from nondisinfecting systems,
1,260 occurred in populations drinking from systems experiencing a treatment deficiency, and 944
were in populations drinking water from systems with distribution system contamination.
EPA believes that reported causes of contamination that result in reported outbreaks during this
time period reflect the relative proportions of the causes of contamination in public ground water
systems since implementation of the Total Coliform Rule in 1989. Therefore, it is estimated that for
every baseline waterborne illness in undisinfected CWS, NTNC and TNC ground water systems with
source contamination, there is an additional 0.43 illness in a system experiencing source contamination
with treatment failure. In addition, it is estimated that for every baseline waterborne illness in
undisinfected CWS or NTNC ground water systems, an additional 0.32 illness occurs due to
distribution system contamination. No additional illnesses due to distribution system contamination are
estimated for TNC systems because TNCs are typically connected directly to the water supply, and
therefore, do not have distribution systems.
Results of the Baseline Risk Calculations
Estimated annual numbers of illnesses from ingestion of Type A and Type B viruses in PWS
ground water systems are summarized in Exhibits 4-12 and 4-13. These tables present the calculated
mean, as well as the 10th and 90th percentile estimates of annual illness and deaths for Type A and Type
B viruses from the Monte Carlo simulation. Results of alternate model calculations using all the same
assumptions, but with the lower drinking water consumption distribution, are presented in Appendix A.
The total number of baseline annual illnesses calculated using the upper bound intake
distribution for Type A viruses (Exhibit 4-12) is about 12 percent higher than the estimate using the
lower bound intake distribution. For Type B viruses, the upperbound estimate (Exhibit 4-13) is 21
percent higher than the lower bound estimate.
4-22 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
Exhibit 4-12. Estimates of Baseline Type A Viral Illness and Death1
Cause/Source of
Contamination
Source contamination in
undisinfected GWSs
Source contamination in
GWSs with failed disinfection
Contamination of distribution
systems of GWSs
Total
Illnesses per Year
Mean
78,172
33,614
21,712
133,498
10th -90th
Percent! les
77,794-78,562
33,452-33,781
21,615-21,812
132,879-134,133
Deaths per Year
Mean
1
0
0
1
10th -90th
Percent! les
1 -1
0-0
0-0
1 -1
1 1llnesses and deaths per year are rounded to the nearest whole number.
Exhibit 4-13. Estimates of Baseline Type B Viral Illness and Death1
Cause/Source of Contamination
Source contamination in
undisinfected GWSs
Source contamination in GWSs
with failed disinfection
Contamination of distribution
systems of GWSs
Total
Illnesses per Year
Mean
19,642
8,446
6,069
34,157
10th -90th
Percent! les
19,019-20,253
8,178-8,709
5,869-6,265
33,062-35,227
Deaths per Year
Mean
8
4
3
14
10th -90th
Percentiles
8-8
3-4
2-3
14-15
1 1llnesses and deaths per year are rounded to the nearest whole number.
Summing the estimates of illness for both types of viruses gives a combined estimate of nearly
168,000 illnesses each year, the majority of which are attributable to the highly infective, but less lethal,
Type A viruses. This estimate is about 14 percent higher overall than the total number of illnesses
estimated using consumption distributions generated from the lower bound water consumption data.
The estimated combined number of deaths per year is 15, the majority of those being due to the more
lethal, but less infectious, Type B viruses.
Baseline Illnesses and Deaths in Sensitive Subgroups
Exhibits 4-12 and 4-13 above summarized the total estimated numbers of illnesses and deaths
each year from ingestion of virally-contaminated ground water under baseline exposure conditions. The
fractions of those illnesses and deaths that are estimated to occur among sensitive subgroups served by
ground water systems are presented in Exhibit 4-14. The sensitive subgroups included in this analysis
include:
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
4-23
-------
Immunocompromised persons in all age groups: AIDs patients, organ transplant patients,
nonhospitalized cancer therapy recipients, and nursing home residents (all of which are assumed
tobe> 16 years old);
Infants and young children < 5 years old; and
Elderly adults > 65 years old
Exhibit 4-14. Health Effects in Sensitive Subgroups as a Percent
of All Illnesses and Deaths
Virus Type
Type A Virus
Type B Virus
Heath
Effect
Illness
Death
Illness
Death
Immunocompromised
1.6%
1.6%
1.6%
0.5%
Infants and
Young Children
< 5 years old
18.1%
18.1%
5.1%
9.4%
Elderly
Adults
> 65 years old
11.5%
11.5%
12.8%
12.4%
Total
Sensitive
Subgroups
31.2%
31.2%
19.5%
22.3%
These sensitive subgroups comprise about 21 percent of the total exposed population, but
account for 31 percent of the Type A illnesses due to ingestion of contaminated ground water. These
results reflect the high morbidity rate (0.88) and the potential for secondary spread of Type A viruses in
children < 2 years old. Sensitive subgroups also account for >22 percent of the Type B deaths.
Although the observed morbidity rate of Type B illness in children < 5 is slightly lower than the
morbidity rate in older children (0.5 vs. 0.57), the high mortality rate of Type B viruses among neonates
(infants < 1 month old) contributes to a higher proportion of deaths in this subgroup and in sensitive
subgroups overall.
Distribution of Annual Individual Risk of Illness
The model results were also analyzed to determine the risk of becoming ill as a result of
ingesting contaminated water from a public ground water system. Risk of illness was estimated for a
period of one year using daily consumption distributions based on the USDA intake data for all
sources, and annual exposure periods as described previously (i.e., 350 days/yr. exposure in CWSs,
250 days/yr. in NTNCs, and 15 days/yr. in TNC systems) (see Section 4.3.1). The results of this
analysis are presented in Exhibit 4-15.
4-24
Proposed Ground Water Rule -Regulatory Impact Analysis
April 5, 2000
-------
Exhibit 4-15. Distribution of Annual Individual Risks of
Illness by Type of System
Virus
Type
Type A
Type B
System
Type
cws
TNC
NTNC
All Systems
CWS
TNC
NTNC
All Systems
(Percent of Population
with Risk in Range)
< 10 -4
98.3%
95.9%
96.2%
97.9%
99.0%
98.8%
96.6%
98.9%
> 1C'4
1.7%
4.1%
3.8%
2.1%
1.0%
1.2%
3.4%
1.1%
Population Total
(Millions)
88.9
14.9
5.3
109.1
88.9
14.9
5.3
109.1
The 10"4 risk level is commonly used by EPA as a criterion of risk in potable water supplies.
The results indicate that 2.1 percent of the exposed population served by public ground water systems
is at a 10"4 or greater risk of becoming ill with Type A viral illness, and 1.1 percent of the population is
at a 10"4 or greater risk of becoming ill with Type B illness under baseline exposure conditions.
Risk to a Highly Exposed Individual
The annual risk of illness was also estimated for a highly exposed individual. For this calculation,
it is assumed that a typical, highly-exposed individual would ingest drinking water from an undisinfected
PWS ground water system having a properly-constructed well. The source water from such a system is
assumed to be contaminated with viral pathogens at a concentration of 3.56 x 10 "3 viruses/L (the mean
concentration of viral pathogens in contaminated, properly-constructed wells). Because the source
water is not disinfected, there is no inactivation of viral pathogens in the system. It is also assumed that
the daily intake of drinking water is 2.345 L/day, the 90th percentile of the daily consumption
distribution for all ages. Annual exposure under this scenario is 350 days/year, the same rate assumed
for CWSs.
Using these assumptions for drinking water exposure, the calculated annual probability of Type
A viral illness for a highly exposed individual is 0.73 for a child < 2 years old and 0.083 for an
individual > 2 years old. For Type B illness, the annual probability of illness is 0.003 for a child < 16
years old and 0.002 for an individual > 16 years old. These results reflect the higher infectivity,
morbidity and secondary spread of Type A viruses in children in comparison with the moderately
infective Type B viruses.
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
4-25
-------
4.4 References
Abbaszadegan, M., P.W. Stewart, M.L. LeChevallier, J.S. Rosen, and C.P. Gerba. 1998.
Occurrence of 'viruses in ground water in the United States. Interim Report. March.
Association of State Drinking Water Administrators (ASDWA). 1997. "Survey of Best Management
Practices for Community Ground Water Systems."
LeBaron, C.W., M.D. Furutan, J.F. Lew, J.R Allen, D.V. Gouvea, D.C. Moe, D.S. Monroe; (1990) .
"Viral Agebts of Gastroenteritis; Public Health Importance and Outbreak management"; Emory
University.
CEOH. 1998. Preliminary draft report. Untitled. July.
Craun, GF. 1998 personal communication.
EPA. 1996. Ground Water Disinfection and Protective Practices in the United States. Office of
Ground Water and Drinking Water, Washington, D.C.
EPA. 1997. Community Water System Survey (CWSS), Volumes I and II. Office of Water,
Washington, D.C. EPA/815-R-97-001a and -OOlb.
EPA. 1999a. Drinking Water Baseline Handbook, Draft 1st Edition. EPA, Washington D.C.
EPA. 1999b. Baseline Profile Document for the Ground Water Rule. EPA.
EPA. 1999c. Model Systems for Public Water Systems. Draft.
EPA. 2000. 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). EPA Office of Water, Office of science and Technology. February, 2000
Gerba, C. P., J. B. Rose, and C. N. Haas. 1996a. "Sensitive populations: who is at the greatest risk?"
Int. J. Food Micro. 30:113-123.
Gerba, C.P., J.B. Rose, C.N. Haas and K.D. Crabtree. 1996b. "Waterborne rotavirus: a risk
assessment." Wat. Res. 30 (12): 2929-2940.
Glass, R.I., P.E. Kilgore, R.C. Holman, S. Jin, J.C. Smith, P.A. Woods, M.J. Clarke, M.S. Ho, and
J.R. Gentsch. 1996. "The epidemiology of rotavirus diarrhea in the United States: surveillance
and estimates of disease burden." J. Inf. Dis. 174 (Suppl 1): S5-S11.
4-26 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
Hall, C. E., M. K. Cooney, and J. P. Fox. 1980. "The Seattle virus watch program. I. Infection and
illness experience of virus watch families during a community-wide epidemic of echovirus type
30 aseptic meningitis." American J. of Public Health. 60:1456-1465.
Hierholzer, J.C. 1992. "Adenoviruses in the immunocompromised host." Clinical Microbiological
Reviews. 5 (3): 262-274.
Kapikian, A. Z. and R. M. Chanock. 1996. "Rotaviruses." Virology, second edition. B.N. Fields et
al., editors. Volumes 1 and 2, New York: Raven Press.
Kilgore, P.E., RC. Holman, M.J. Clarke, andRI. Glass. 1995. "Trends of diarrheal
diseaseassociated mortality in U.S. children, 1968 through 1991" JAMA 274(14):
1143-1148.
Le Baron et al. 1990
Lew, J.F., R.I. Glass, RE. Gangarosa, IP. Cohen, C. Bern, and C. Moe. 1991. "Diarrheal deaths in
the United States, 1979 through 1987, a special problem for the elderly" JAMA 265(24):
3280-3284.
Lieberman, R.J. et al. 1995. Viral andmicrobial methods for groundwater. Interim Report.
USEPA and AWWARF.
Maasdam, C.F. and S. Anuras. 1981. "Are you overlooking GI infections in your elderly patients?"
Geriatrics. 36:127-134.
Morens, D. M., M. A. Pallansch, and M. Moore. 1991. "Polioviruses and other enteroviruses."
Textbook of Human Virology, second edition. Robert B. Belshe, editor. St Louis: Mosby
Year Book.
NRC (National Research Council). 1983. "Risk Assessment in the Federal Government: Managing
the Process." National Academy Press, Washington, DC.
Parashar, U.D., RC. Holman, MJ. Clarke, J.S. Bresee, and R.I. Glass. 1998. "Hospitalizations
associated with rotavirus diarrhea in the United States, 1993 through 1995: surveillance based
on the new ICD-9-CM rotavirus-specific diagnostic code." J. Inf. Dis. 177:13-17.
Regli, S, J. B. Rose, C. N. Haas, and C. P. Gerba. 1991. "Modeling the risk from Giardia and
viruses in drinking water." JAWWA. p. 76-84.
Schiff, G. M., G. M. Stefanonic, E. C. Young, D. S. Sander, J. K. Pennecamp, and R. L. Ward. 1984.
Studies of echovirus-12 in volunteers: determination of minimal infectious dose and the effect of
previous infection on infectious dose." J. Inf. Dis. 150(6):858-866.
April5, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 4-27
-------
SDWIS (Safe Drinking Water Information System). 1997. EPA.
Stedge, Gerald. 1998. Personal communication with Gerald Stedge, Abt Associates. June.
Tucker, A. W., A. C. Haddix, J. S. Bresee et al. 1998. "Cost effectiveness analysis of a rotavirus
immunization program for the United States." JAMA. 279:1371-1376.
Ward, R. L. 1986. "Human rotavirus studies in volunteers: determination of infectious dose and
serological response to infection." J. Inf. Dis. 154(5):871.
Yolken, R.H., C.A. Bishop, T.R. Townsend, E.A. Bolyard, J. Bartlett, G.W. Santos, and R. Sabal.
1982. "Infectious gastroenteritis in bone-marrow transplant recipients." N. Engl. J. Med. 306
(17): 1010-2.
4-28 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
5. Benefits Analysis
5.1 Introduction
By reducing public exposure to waterborne viral and bacterial pathogens of concern, the
proposed GWR reduces the public's risks of acute illness and death from fecal contamination of
drinking water. The health-related benefits of the GWR are largely the result of avoided acute and
chronic illnesses and deaths attributable to reduced endemic and outbreak risks. This chapter presents
results of analysis of these health-based benefits including the monetized value of the avoided illnesses
and deaths estimated for rule options. In addition, this chapter discusses non-monetized health benefits
and other non-health benefits of the rule options.
5.2 Structure of the Benefits Analysis
Several recent reviews provide a general discussion of a wide range of possible ground water
disinfection benefits and ground water protection benefits (NWRI1997; NRC 1997; RTI1997; EPA
1995). These reviews, in combination with information on ground water services in general, serve to
frame the benefit categories relevant to the GWR. Two major categories, health benefits and non-
health benefits, as well as components within each, are presented in Exhibit 5-1.
Exhibit 5-1. Overview of GWR Benefits
Health Benefits
Reduction in acute illness
incidence
Reduction in chronic illness
incidence
* viral risk reduction (morbidity and mortality)
> bacterial risk reduction (morbidity and mortality)
* viral risk reduction (morbidity and mortality)
> bacterial risk reduction (morbidity and mortality)
Non-Health Benefits
Reduced uncertainty
Avoided costs of averting
behavior
Outbreak responses avoided
> improved perception of drinking water quality
* bottled water, point-of-use (POU) devices, etc.
* time spent on averting behavior: hauling/boiling water, etc.
* avoided costs to affected water systems and local governments
(provision of alternative water, issuing warnings and alerts)
5.2.1 Human Health Benefits
The quantified benefits identified in this chapter are predominantly from the economic value of
avoided human health risks (i.e, morbidity and mortality). This is predicated on the concept that the
health benefits associated with the promulgation of the GWR are equal to the value of the reduced risks
of morbidity and mortality from microbial pathogens. Assessing the economic value of the health
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
5-1
-------
benefits involves a two-step process, which includes a risk assessment to quantify a health endpoint
followed by a benefits valuation procedure. The process is illustrated in Exhibit 5-2.
Exhibit 5-2.
GWR Health Benefits Assessment Framework
HEALTH
ENDPOINT
Reduced
Acute
Morbidity
EFFECTS
VALUED
Reduced
Chronic
Morbidity
Reduced
Mortality
Lost Leisure
Time
Avoided
Medical
Costs
Avoided
Pain and
Discomfort
Avoided
VALUATION
APPROACH
Cost of Illness Analysis
To Estimate Average Cost
per Illness Avoided
(by Degree of Severity)
Not Valued:
Qualitatively Discussed
Benefits Transfer of Value
of Statistical Life or Life
Year from Studies of
Individuals' WTP to Avoid
Mortality Risks
Risk assessment-This assessment is used to determine and quantify the health endpoints
identified for the GWR as shown in Exhibit 5-2. These endpoints include reductions in both
acute and chronic illnesses (i.e., reduced morbidity) and reductions in the number of deaths
each year (i.e. reduced mortality). Health endpoints are quantified using a risk assessment to
estimate the number of illnesses and deaths each year for both baseline and post-regulatory
scenarios. The difference between these estimates is the avoided cases of morbidity or
mortality. For this RIA, the risk assessment focuses on acute illnesses and associated deaths
caused by viral pathogens ingested in tap water from ground water sources. (Chronic effects
and bacterial pathogen infections, while not quantified directly using the risk assessment model,
are discussed below.)
5-2
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
The risk assessment forecasting necessarily requires a number of assumptions to be made
regarding exposures to viral pathogens in drinking water. These assumptions are presented in
Section 4.3, Baseline Risk Assessment Health Effects. Additional assumptions required to
assess the health effects of the four regulatory options for the GWR are presented in Section
5.3.1. The results of the risk analysis for each of the four regulatory options (i.e., in terms of
illnesses and deaths avoided) are presented in Section 5.3.2.
Benefits valuation-This process applies the appropriate unit values to the results of the risk
assessment to estimate the value of the specific illness and clinical endpoint identified in exposed
populations using the risk assessment. The assumptions and inputs used to determine the
appropriate unit values are discussed in Section 5.3.3. The results of this analysis, the
monetized benefits of avoided morbidity and mortality for each GWR option, are presented in
section 5.3.4.
The benefits of the reduced exposure to bacteria in drinking water are also discussed in this
chapter. The health effects of reduced bacterial infections are not assessed using the risk assessment
model. Rather these health benefits are valued by employing a simple ratio assumption in which the
monetized benefits estimated for reduced viral infections were increased by an additional 20 percent to
account for bacterial infection reduction benefits.
The other potential health benefits associated with the proposed rule are the reduced risks of
chronic morbidity and corresponding mortality associated with viral and bacterial contamination. These
are not quantified within this RIA. As discussed in Section 5.4.2, the benefit of avoiding these chronic
cases may be significant, as affected individuals incur significant costs in medical care and losses in
productivity and quality of life in such instances. The reader is reminded not to discount the value
associated with reducing chronic viral and bacterial illness simply because they are not quantified in this
RIA, as EPA's inability to quantify them due to data limitations does not suggest they are not significant.
5.2.2 Non-Health Benefits Assessment
In addition to the health-based benefits introduced above, there are a number of non-health
benefits that also arise from promulgation of the rule. Non-health benefits may result from overall
system improvements (e.g., upgrades to distribution systems, increased efficiencies, increased
frequency/intensity of process surveillance), from improved risk perception of drinking water quality, or
from avoided outbreak response costs. While these costs are not quantified for this RIA, these
potential benefits are discussed qualitatively in Section 5.4.3.
5.2.3 Potential Health Risk Associated with Other Contaminants
The Agency is aware that the proposed GWR has the potential to increase health risks in some
circumstances; these risks however, can be controlled. The increased risks that may result from this
rule stem from the installation of disinfection equipment by systems currently not treating. Risks may
stem from either or both of the following problems.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 5-3
-------
Start-up ContaminationWhen disinfection is first introduced into a previously undisinfected
system, the disinfectant can react with pipe scale causing increased risk from some
contaminants and water quality problems. Contaminants that may be released include lead,
copper, and arsenic. It could also lead to a temporary discoloration of the water as the scale is
loosened from the pipe. These risks can be reduced by gradually phasing in disinfection to the
system, by routine flushing of distribution system mains and by maintaining a proper corrosion
control program.
Disinfection ByproductsFor some ground water systems, using a chlorine-based disinfectant
or ozone could result in an increased risk from disinfection byproducts. Risk from disinfecting
systems (including an estimate of the ground water systems, which will commence disinfection
as a result of the GWR) has already been addressed in the Stage 1 Disinfection Byproducts
Rule. Overall, only a small number of ground water systems with high source water organic
carbon precursors are expected to have high levels of disinfection byproducts from using
chlorine. However, those that do can avoid this problem by choosing an alternative disinfectant
or precursor control technology (e.g., chloramination, membranes, or ultraviolet).
5.3 Value of Health Effects With Rule (Acute Impacts)
The only benefit EPA has estimated and monetized for this RIA is acute health effects of viral
and bacterial infections. This chapter explains how health effects were estimated (Section 5.3.1), how
they were monetized (Section 5.3.3), and presents results for each effort (Sections 5.3.2 and 5.3.4,
respectively).
5.3.1 Assumptions for Health Effects Modeling of Regulatory Scenarios
In estimating the health effects of the GWR, EPA performed risk calculations for the four GWR
options described in Chapter 3.: Option 1Sanitary Surveys only; Option 2Sanitary Surveys with
Triggered Monitoring; Option 3Multi-Barrier Approach; and Option 4
Across-the-Board-Disinfection. The estimation procedure was two-fold as follows:
First, using assumptions regarding reductions in viral exposure from source contamination for
each regulatory option, the model was used to calculate annual numbers of illnesses and deaths
in ground water systems (GWSs).
Secondly, using CDC ratios applied to the results of the first step and assumptions for each
option regarding post-regulatory reductions in rates of disinfection failure and distribution
contamination, additional outbreak-related illnesses and deaths were estimated.
To model the reduction in exposure from source contamination that would result from
implementation of the four regulatory options, EPA assumed reductions in the number of undisinfected
ground water systems/points of entry that are potentially contaminated with viral pathogens under
baseline conditions. The reduction varies with expectations regarding the effectiveness of each option in
identifying and correcting significant defects at the source. Reductions in treatment failure rate and in
5-4 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
distribution system contamination are also addressed for each option. The relevant exposure
assumptions for each option and type of system are summarized in Exhibit 5-3 and are discussed in the
following sections.
Exhibit 5-3. Estimated Contamination Reductions for GWR Options
Regulatory Option
Option 1. Sanitary Survey
Only
Option 2. Sanitary Survey
and Triggered Monitoring
Option 3. Multi-Barrier
Option 4. Across-the-
Board Disinfection
Estimated Reduction in
Viral Source Contamination
of Undisinfected Grou nd
Water Sources
Properly
Constructed
0%
30-54%
38-77%
100%
Improperly
Constructed
40-60%
58-82%
63-91
1 00%
Estimated
Reduction in rate
of Disinfection
Failure for GWSs
with viral
contamination of
the source
0-26% (CWS)
0-43% (NCWS)1
77-1 00%
77-1 00%
77-1 00%
Estimated
Reduction in
Distribution
System
Contamination
with Virus of
GWSs
0-25%
(NA for TNC)2
0-25%
(NA for TNC)2
0-25%
(NA for TNC)2
0-25%
(NA for TNC)2
1 Non-community water systems (NCWS), both transient and non-transient, have an estimated reduced risk of
contamination of 0-43%; community water systems (CWS) reduced risk is 0-26%.
2 Reduction of risk in transient non-community (TNC) systems was not considered.
5.3.1.1 Option 1: Sanitary Survey Option
Because EPA would, under this option, require the State (or primacy agent) to conduct
periodic sanitary surveys for all ground water systems (i.e., at least every three years for CWSs, and at
least every five years for NCWSs), this option is expected to identify significant defects in wells, which
could lead to source contamination (e.g., an improperly cased well) in the treatment process (e.g.,
inadequate disinfectant feed rate) and in the distribution system (e.g., uncovered storage tank). Under
this option, systems would be required to correct these significant defects. Based upon data from a
recent survey of ground water systems (ASDWA, 1997), EPA estimates that between 11 and 13
percent of systems will correct significant defects as a result of implementation of this option (resulting in
the elimination of over 22,000 significant defects). Based on these assumptions, EPA made the
following estimates of risk reduction.
Properly constructed wellsImplementing sanitary surveys alone will result in no significant
reductions in source contamination in systems with properly constructed wells.
Improperly constructed wellsSanitary survey inspectors will identify significant defects that
are apparent from a visual inspection of the well, well construction records, or the surrounding
area. Correction of these significant defects will eliminate pathways for viral contamination to
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
5-5
-------
reach source waters in 40 to 60 percent of the poorly constructed wells. Defects that are
underground or that are not reflected in well construction records will not be detected.
Disinfection failure EPA estimates that correction of significant deficiencies in disinfection
treatment processes will result in between 0 and 26 percent reduction in treatment failures in
community water systems and between a 0 and 43 percent reduction in non-community water
systems. Interrupted treatment caused by failed systems or operator error accounts for 26
percent of the reported outbreaks in community ground water systems and 43 percent of the
reported outbreaks in non-community systems (Craun, 1999). EPA estimates that correction
of significant defects could prevent some or all of the interruptions in disinfection. The
estimated range accounts for the uncertainty as to the exact percentage.
Distribution system contamination EPA estimates that correction of significant defects in the
distribution system identified by sanitary surveys will result in a 0 to 25 percent reduction of
fecal contamination of distribution systems in both community and non-transient non-community
water systems. Half of the reported ground water system outbreaks caused by distribution
system contamination were specifically caused by cross connections or storage tank
deficiencies (Craun, 1999). EPA estimates that as many as half of these defects will be found
and eliminated as a result of sanitary surveys. The estimated range accounts for the uncertainty
as to the exact percentage.
5.3.1.2 Option 2: Sanitary Survey and Triggered Monitoring Option
The sanitary survey and triggered monitoring option combines the sanitary survey and
correction of significant defects with triggered monitoring of source water to identify improperly
constructed systems for corrective action. Because the sanitary survey component of this option
imposes identical requirements to the sanitary survey option, EPA expects the sanitary survey portion of
this option to achieve the same reductions in source contamination of poorly constructed wells, of
treatment failure, and of distribution systems contamination.
The additional triggered monitoring component of this option will identify systems with fecal
contamination of their sources that could not otherwise be identified through a sanitary survey of the
system. Under the triggered monitoring requirements, systems are required to sample their source
water for the presence of a fecal indicator following the detection of total coliform in their distribution
system. EPA estimated the effectiveness of the triggered monitoring requirements in identifying and
eliminating source water contamination by evaluating the ability of the triggered monitoring to identify the
pathogen contaminated wells. Assumptions made by EPA include the following:
An estimated 30-54 percent of wells that are contaminated with viral pathogens will be
detected through triggered monitoring. EPA made this assumption by using monitoring results
of samples taken from wells that were found to contain viral pathogens in the EPA/AWWARF
study (Lieberman et al, 1997, 1999). Seven wells in this study were found to contain viral
pathogens in at least one of the twelve samples taken over the course of a year. Each of these
wells was also tested for the presence of the three fecal indicators (E. coli, enterococci
56 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
bacteria, or male-specific coliphage). Of the 84 samples from these seven virally contaminated
wells, 42 (50 percent) tested positive for E. coli, 45 (54 percent) tested positive for
enterococci, and 25 (30 percent) tested positive for the presence of male-specific coliphage.
States are provided flexibility in determining the most appropriate of the three methods,
therefore EPA has assumed a range of effectiveness in detecting fecally contaminated wells
between the low and high percentages.
Corrective action will be required for all identified contaminated sources; this will entail
eliminating the source of the contamination, finding an alternative source, or installing
disinfection.
Based on these assumptions and relevant assumptions from the sanitary survey option, EPA
made the following estimates for risk reduction.
Properly constructed wellsIn wells considered to be properly constructed, EPA assumed 30
to 54 percent of the systems with pathogen contamination will be identified through triggered
monitoring. As with the sanitary survey option, no source contamination reduction is expected
in the properly constructed wells due to sanitary surveys.
Improperly constructed wellsIn improperly constructed systems, EPA assumed reduction of
source water contamination by 58 to 82 percent of the baseline. This is based on the Agency's
assumption that 40 to 60 percent of the contaminated poorly constructed wells will be identified
through sanitary surveys, and an additional 30 to 54 percent of the remaining undetected
contaminated wells will be identified through triggered monitoring.
Disinfection failureEPA estimates a reduction of 77 to 100 percent of incidences in which
systems with pathogen contaminated source water inadequately disinfect or remove the
pathogens. The Agency made this assumption based on the option's requirement both that
systems that disinfect must achieve a 4 log removal or inactivation of pathogenic viruses and
that systems ensure compliance with this inactivation level by routinely monitoring the
disinfectant residual. EPA estimates that this provision will eliminate one half to all of the
interruptions in treatment and will eliminate three fourths to all of the instances in which systems
do not adequately disinfect their source water.
Distribution system contaminationA 0 to 25 percent reduction of fecal contamination of
distribution systems and the associated illnesses/deaths in community and non-transient non-
community water systems (same as aforementioned sanitary survey option).
5.3.1.3 OptionS: Multi-Barrier Option
The Multi-Barrier option builds on the sanitary survey and the triggered monitoring components
described above by adding routine source water monitoring to provide an effective means of identifying
wells in the most sensitive hydrogeologic conditions. If the States or primacy agents determine a
system's source to be hydrogeologically sensitive to microbial contamination, the system is required to
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 5-7
-------
perform routine source water monitoring. Routine monitoring requires the collection of source water
samples monthly (at least for the first year of monitoring) to ensure that periodic fecal contamination has
a greater likelihood of being detected than it would with a single sample. Specific assumptions for
modeling this option include the following.
Fifteen percent of the ground water sources will be found to be sensitive and, therefore, subject
to routine source water monitoring.
Between 20 and 50 percent of the wells with pathogen contamination are located in conditions
that States will determine to be sensitive, and would therefore be subject to routine monitoring.
Between 71 and 100 percent of pathogen contaminated ground water sources that are subject
to routine source water monitoring will be found to be fecally contaminated and required to
take corrective action. EPA estimated the effectiveness of routine monitoring in detecting viral
pathogen contamination based on EPA/AWWARF sampling data for seven wells that were
found to contain viable pathogenic viruses. EPA reviewed the test results for the three fecal
indicators from which the States will choose for routine source water monitoring. Of the seven
wells that were virally contaminated over the course of the year, five (71 percent) tested
positive at least once for the presence of E. coli, all seven (100 percent) tested positive for the
presence of enterococci and five (71 percent) tested positive for the presence of male specific
coliphage. For the purposes of this analysis, EPA assumes that States will select E. coli as the
fecal indicator for routine monitoring; this gave EPA the lower bound of 71 percent.
Triggered monitoring will identify 30 to 54 percent of the source-contaminated wells as
described in the sanitary survey and triggered monitoring option above.
Based on these assumptions and relevant assumptions from Options 1 and 2 above, EPA made
the following estimates of risk reduction.
Properly constructed wellsBetween 38 and 77 percent of the properly constructed wells with
pathogen contamination will be identified as a result of this rule option.
Improperly constructed wellsBetween 63 and 91 percent of the improperly constructed wells
with pathogen contamination will be identified as a result of this rule option.
Disinfection failureAs with the sanitary survey and triggered monitoring option, this option
requires systems that practice disinfection to achieve a 4 log removal or inactivation of virus. It
also requires systems to ensure compliance with this inactivation level by routinely monitoring
the disinfectant residual. EPA anticipates that these requirements will significantly reduce the
incidences in which systems with viral pathogen-contaminated source water inadequately
disinfect or remove the pathogens. EPA estimated a reduction of these incidences by 77 to
100 percent (see Option 2 above).
Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
Distribution system contaminationA 0 to 25 percent reduction of fecal contamination of
distribution systems and the associated illnesses/deaths in community and non-transient non-
community water systems (see Option 1 above).
5.3.1.4 Option 4: Across-the-Board Disinfection Option
The across-the-board disinfection option will reduce viral pathogens in ground water system
through three mechanisms.
Properly and improperly constructed wellsFirst, because all systems would be required to
install disinfection treatment, EPA assumes that 99.99 percent of the pathogenic viruses in
source water would be inactivated (4 log removal) before they reach the customer's tap.
Therefore, there would be a complete elimination of source contaminated, undisinfected
systems.
Disinfection failureSecond, systems would be required to monitor the disinfectant residual
concentration. This will insure a consistent level of disinfection treatment, which is adequate to
remove 99.99 percent of the pathogenic viruses. The Agency assumes the same level of
reductions in failure of disinfection treatment systems equivalent with the previous two options
(see Option 2 and 3 above), 77 to 100 percent.
Distribution system contaminationThird, systems would be required to correct any significant
deficiencies identified in sanitary surveys. This would result in reductions in contamination of
distribution systems that are similar to the two previous options (see Option 2 above), i.e., 0 to
25 percent reductions.
5.3.2 Results of Risk Calculations
The results of the risk assessment for the baseline and the four regulatory scenarios are
presented in Exhibits 5^1 and 5-5. The first table presents the mean numbers of illnesses and deaths
estimated annually under each scenario. Because there are uncertainties in the values assigned to some
model parameters (e.g, the uncertainties in the percent reductions of source contamination, disinfection
failure, and distribution contamination anticipated with each option [see Exhibit 5-3]), the risk
assessment model generates a distribution of estimates of annual illnesses and deaths. The calculated
mean as well as the 10th and 90th percentile estimates of the number of annual illnesses and deaths were
obtained from the distribution of results for each type of virus for both the baseline conditions and for
each regulatory option. These outputs are summarized in Exhibit 5^1. Exhibit 5-5 presents the
calculated incremental reductions in illnesses and deaths from the current baseline estimates (see
chapter 4). The results presented below are based upon the USDA estimate of daily direct and indirect
drinking water consumption having a mean of 1.24 L/day. Detailed results may be reviewed in
Appendix A along with the results for the water consumption distribution with a mean of 0.927 L/day.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 5-9
-------
Exhibit 5-4. Remaining Viral Illnesses/Deaths for Regulatory Scenarios1
Scenario
Baseline
Option 1: Sanitary
Survey Only
Option 2: Sanitary
Survey and Triggered
Monitoring
Option 3: Multi-
Barrier Option
Option 4: Across the
Board Disinfection
and Sanitary Survey
Virus
Type
Type A
Type B
Total
Type A
Type B
Total
Type A
Type B
Total
Type A
Type B
Total
Type A
Type B
Total
Illnesses per Year
Mean
133,498
34,157
167,655
122,941
31,143
154,084
67,200
17,115
84,315
56,953
14,462
71,415
28,467
7,111
35,578
10th -90th
Percentiles
132,879-134,133
33,062-35,227
165,941 -169,360
118,601 -127,194
29,843-32,440
148,444-159,634
59,621 -74,630
15,154- 19,010
74,775-93,640
45,971 -67,492
11,830-17,135
57,801 -84,627
21,575-35,536
5,631 -8,570
27,206-44,106
Deaths Per Year
Mean
1
14
15
1
13
14
0
7
7
0
6
6
0
3
3
10th -90th
Percentiles
1 -1
14-15
15-16
1 -1
13-14
14-15
0-1
6-8
6-9
0
5-7
5-7
0
2-4
2-4
1 Using Age-Based Consumption Distributions for All Sources, Consumers Only
Exhibit 5-5. Reduction1 in Illnesses/Deaths for Regulatory Scenarios
Scenario
Option 1 : Sanitary Survey Only
Option 2: Sanitary Survey and Triggered Monitoring
Option 3: Multi- Barrier Approach
Option 4: Across-the-Board Disinfection
Net Reduction in
Viral Illnesses per
Year (Mean)
13,596
83,502
96,305
132,129
Net Reduction in
Viral Deaths Per
Year (Mean)
1
8
9
12
1 Reductions for each scenario are measured as incremental difference from baseline values; value may not
match differences calculated from Exhibit 5-4 due to rounding.
Option 1: Sanitary Survey Alternative-This option is estimated to reduce the number
of waterborne viral illnesses in public GWSs by over 13,600 illnesses each year in comparison with the
baseline (an 8 percent reduction in illnesses). The sanitary survey option is also estimated to reduce by
at least one per year the number of deaths that result from waterborne illness.
5-10
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Option 2: Sanitary Survey and Triggered Monitoring Alternative-This option is
estimated to reduce the number of waterborne viral illnesses by approximately 83,500 illnesses each
year in comparison with the baseline (about a 50 percent reduction in illnesses). This option is also
estimated to reduce the number of deaths that result from waterborne illness by about eight each year, a
reduction of over half the baseline rate.
The difference between the health effects estimates for this option and those for Option lisa
net reduction of nearly 70,000 illnesses and seven deaths. This is the expected reduction that would
result from actions taken in response to the results of triggered monitoring and including disinfection
monitoring requirements.
Option 3: Multi-Barrier Option-The Multi-Barrier option is estimated to reduce the
number of waterborne viral illnesses by just over 96,300 illnesses each year from the current baseline
estimate (a 57 percent reduction in total illnesses). The Multi-Barrier option is also estimated to reduce
the number of deaths that result from waterborne illness by about nine each year.
The difference between the estimates for this option and those for Option 2 is a net estimated
reduction of nearly 13,000 illnesses and one death; this is the expected reduction that would result from
source water monitoring and resulting corrective actions.
Option 4: Across-the-Board Disinfection Alternative-This alternative is estimated
to reduce the number of waterborne viral illnesses by approximately 132,000 illnesses each year (a 79
percent reduction in illnesses). Across-the-board disinfection is also estimated to reduce the number of
deaths which result from waterborne illness by about 12 each year.
The difference between the estimates for this option and those for Option 3 is an estimated net
reduction of approximately 35,000 illnesses and three deaths; this is the expected reduction that would
result from 100 percent of public GWSs using disinfection. Although all GWSs would treat ground
water under this option, a few, less frequent treatment failure and distribution system contamination
events each year would continue to cause a residual number of illnesses and deaths in the population
served by ground water systems.
5.3.3 Assumptions for Monetization of Health Benefits (Acute Illnesses)
Having estimated reduced illness and mortality from the GWR's four scenarios, EPA then
monetized the health benefits. Using estimates of the number of avoided illnesses and deaths expected
to result from promulgation of any of the options EPA applied unit estimates of "cost-of-illness" and
"value of a statistical life," respectively, to estimate the benefit of the avoided illnesses and deaths (see
Exhibit 5-6). The unit costs and the bacterial infection ratio are explained in the following sections, after
which the monetized results for each of the four regulatory options are presented.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 5-11
-------
Exhibit 5-6. Monetization of Health Effects
Illnesses Avoided from
Reduction in Viral Infection
(From Risk Model)
Deaths Avoided from
Reduction in Viral Infection
(From Risk Model)
Unit "Cost-of-lllness"
(Introducedin Section 5.3.3.1)
Unit "Value of Statistical Life"
(Introduced in Section 5.3.3.2)
Bacterial Infection Ratio
(Introducedin Section 5.3.3.3)
Bacterial Infection Ratio
(Introduced in Section 5.3.3.3)
Total Monetized Value of Health Benefits
(Results presented in Section 5.3.4)
5.3.3.1 Unit Cost-of-lllness
EPA chose to use cost-of-illness
(COI) as the best available means of valuing
illnesses avoided by application of the
GWR. In theory, the cost of an
illness is the present discounted value of the
lifetime stream of costs that result from the
illness. The COI estimates described in this
section also consider the associated direct
and indirect costs incurred due to illness.
Direct costs describe the cost burden of
medical care to affected individuals. Indirect
costs describe the opportunity costs
associated with illness, such as productivity
and leisure losses. Note, however, that COI
underestimates total willingness-to-pay (WTP)
associated with the illness.
An appropriate value for a reduction in risk of an
illness experienced by all exposed individuals in a
population is the sum of these individuals' WTP to
avoid the illness before it occurs. Conversely, one
could use an "ex post" or damage function approach
to value reduction in risk. The damage function
approach multiplies the mean WTP to avoid a case
of the illness by the expected number of cases
avoided.) Estimates of WTP for specific risk
reductions or estimates of WTP to avoid a case of
certain illnesses are, however, often unavailable.
This is true with regard to the waterborne illnesses
that may be avoided as a result of this proposed
rule. Cost-of-illness was, therefore, used as a proxy
for WTP.
because it does not address the pain and suffering
Apogee/Hagler Bailly (1998) describes the general method used to calculate COI, and provides the
derivations of direct and indirect costs for all of the COI values used in this analysis. The discussion
below explains why the values differ across three factorsvictim age, illness severity, and immune
status.
5-12
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Cost-of-illness: Factors Affecting Estimates
For the viruses of concern under the GWR, a review of the medical and epidemiological
literature revealed that the nature and extent of the acute health effects varied by severity and by
population subgroup for each virus. Due to data limitations in the risk assessment, these viruses of
concern were categorized as Type A (represented by Rotavirus) and Type B (represented by
Echovirus). The former describes highly infective pathogens with less expensive costs of illness (such as
rotavirus), while the latter describes less infective pathogens (such as echovirus) that are associated with
more costly illnesses.
Victim Age: In general, the annual health benefits for the GWR were calculated by
multiplying the annual number of acute illnesses avoided by the COI per case. Age categories were
created based on the different clinical manifestations of disease or where differences in indirect costs of
illness could be identified. For Type A pathogens, the age groups involved were: less than two, two to
five, five to 16, and over 16. Because of the nature of illnesses associated with Type B enteroviruses
(e.g., sepsis-like illness in neonates), the under-five-year-old age groups for Type B pathogens had to
be further segregated into the following categories: less than one month, one month to one year, and
one to five years.
Severity of Illness: The COIs for the enteroviruses varied not only by age, but also by
illness severity. Since unit COIs vary widely between severity categories, they had to be segregated
into three severity classifications of illness prior to valuation. These severity classes (i.e., mild,
moderate, and severe) were used only for Type B viruses (See Exhibit 5-7).
Exhibit 5-7. Classification for Clinical Severity in Type B (Echovirus) Illnesses
Severity of
Type B Viral
Infection
Mild
Moderate
Severe
Conditions Affecting
all Ages for each
Severity Level
non-specific febrile illness,
respiratory illness,
gastrointestinal illness
aseptic meningitis
myopericarditis
Age Specific Clinical Severity
Neonates (<1 month)
exanthum (skin eruptions)
encephalitis
sepsis-like illness (with
hepatitis)
Children (1-5 years)
herpangina (throat lesions),
pleurodynia (affection of
thoracic tendons/muscles)
(none)
(none)
Source: Dirckx (1997)
The aggregate number of avoided illnesses were divided into three categories, each describing a
different illness severity level that is associated with different direct and indirect cost components.
Severity was assigned using weights derived from two studies (Morens 1978; Melnick, 1996) in which
the distributions of various clinical symptoms were described among a group of affected individuals (see
Exhibit 5-8).
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
5-13
-------
Exhibit 5-8. Weighting for Clinical Severity in Type B (Echovirus) Illnesses
Age Group
Neonates (<1 month)
Newborns(1 month-1year)
Children (1-5 years)
Others (>5years)
Mild
26%
52%
52%
52%
Moderate
50%
46%
46%
46%
Severe
24%
2%
2%
2%
(%) indicates the portion/weighting of the total number of illnesses assigned between the severity categories of
each age category.
Victim Immune Status: Certain segments of the population are more likely to develop
more severe symptoms due to their compromised immune systems. Therefore, for each
immunocompromised individual who becomes ill, a higher unit COI estimate is assumed. Unit COI
estimates for Type A viruses were specifically developed for immunocompromised individuals also
using rotavirus. Due to data limitations concerning the effect of rotavirus infection on the
immunocompromised population, these unit COI estimates were derived using a modified version of the
COI framework. Most of the same inputs were used as for healthy individuals, except that the
percentage of ill individuals seeking inpatient and outpatient care was increased and assumed to be 100
percent. Due to similar data limitations, Type B viral illnesses among immunocompromised
subpopulations were assigned the "severe enterovirus" COI for the appropriate age categories.
Cost-of-Illness: Estimates for Unit Costs
Unit COIs were developed for each virus of concern by age, level of severity, and health status;
these are presented in the Exhibits 5-9 and 5-10. The unit costs are in May 1999 dollars, direct costs
having been updated using the "medical care services" expenditure category of the Consumer Price
Index among all urban consumers and indirect costs using the CPI-U for all items (BLS 1999).
Exhibit 5-9. Type A (Rotavirus) Unit Cost-of-Illness
Estimates by Victim Age and Health Status
Victim Age
Age<2
2 < Age < 5
5 < Age < 16
Age > 16
Cost-of-Illness (Direct & Indirect)1
Healthy
$921
$507
$212
$349
Immuncompromised
$4,666
$4,666
$4,637
$4,912
1Costs do not include pain and suffering.
Source: Apogee/Hagler Bailly (1998)
As indicated in Exhibit 5-9, the Type A viral infections have cost impacts to
immunocompromised populations that are from five to 20 times more severe compared to healthy
populations. Also of note, impacts to healthy infants, including in this case neonates, new borns, and
one to two year olds, have a cost impact of two to four times more than other age groups.
5-14
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit 5-10. Type B (Enterovirus) Unit Cost-of-lllness Estimates by
Victim Age and Illness Severity
Victim Age
Age < 1 mo.
1 mo. < Age < 5
5 < Age < 16
Age > 16
Cost-of-lllness (Direct & Indirect)1
Mild
$347
$311
$ 158
$285
Moderate
$11,283
$ 8,856
$ 7,283
$ 7,626
Severe
$19,711
$ 8,742
$ 8,742
$ 9,703
1Costs do not include pain and suffering.
Source: Apogee/Hagler Bailly (1998).
Review of the Type B viral cost-of-illness, as presented in Exhibit 5-10, indicates even more
pronounced impacts on one select subpopulation, that of neonates. In the case of severe cases of
enteroviral infection, costs of illness to treat neonates is more than double that of all other age groups.
5.3.3.2 Unit Value of a Statistical Life
EPA chose to use "value of a statistical
life" as the best available means of valuing deaths
avoided by application of the GWR.
Conceptually, the value of mortality is measured
as an individual's WTP to reduce mortality risk,
aggregated across all affected individuals. It also
reflects the value of morbidity that precedes
death. The dollar amount a person would be
willing to pay for mortality risk reduction does
not indicate the value that he places on his life.
Rather, it reflects an individual's value of small
reductions in the probability of death distributed
over a large population, referred to as the "value
of a statistical life" (VSL).
Determining VSL
To better understand VSL, consider a drinking water
regulation that reduces, for a population of 10,000,
the mortality rate associated with contamination by
microorganisms from ten out of 10,000 to five out of
10,000. This regulation, therefore, would save, on
average, five "statistical lives" (so-called because
there is no way to predict which members of the total
population would be saved by the regulation). If
each of the 10,000 individuals in the "population" is
willing to pay $500 for this reduction in the
probability of death, then the "willingness to pay" for
the population as a whole is $5 million. Since an
average of five lives are saved by implementing the
regulation, the VSL per life saved equals $1 million.
Valuation of avoided mortalities due to the GWR involves identifying VSL estimates that
represent similar types of mortality risks as those associated with waterborne risks and possibly
adjusting VSL estimates to better fit the waterborne risk context. For purposes of this RIA, EPA
reviewed several studies. One, a 1997 EPA study, was based on a best-fit distribution of 26 "policy-
relevant value-of-life studies." The VSL in 1990 dollars was characterized as a Weibull distribution
with a mean of $4.8 million per life and a standard deviation of $3.24 million; these results were
updated to 1999 dollars. For this RIA, the mortality benefits of the GWR were calculated by
multiplying the number of avoided deaths by the updated VSL, which was estimated at $6.3 million.
5.3.3.3 Reduction in Bacterial-Related Illnesses
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
5-15
-------
Impacts of Bacterial Outbreaks
Serious bacterial health effects are common in
outbreak situations and pose significant burdens on
community health and non-health resources. For
example, in a 1993 Salmonella typhimurium
waterborne outbreak in Gideon, MO, the health and
non-health effects were widespread. Approximately
44 percent of individuals were ill, and of those
surveyed, 29 percent sought medical attention and
four percent were hospitalized (Angulo et al., 1997).
Absenteeism from school increased by 250 percent,
and the sales of anti-diarrheal medicines increased
In addition to the expected benefits
from reducing viral infections, the GWR is
expected to provide benefits related to the
potential reduction of illnesses and deaths due
to waterborne bacteria. The avoided cost-of-
illnesses may be as substantial as those seen for
viruses.
Bacterial Infections: Background
The extent to which a bacterial illness
develops depends on various pathogenic
characteristics of the organism, including the
strain and virulence of the organism, and on various host characteristics such as age or immune status.
Such factors help to explain, for example, why Campylobacter cases are among the most common, yet
cause the least number of hospitalizations in outbreak situations. Understanding the nature and severity
of the illness helps to frame the economic impact of waterborne outbreaks, illnesses and deaths caused
by bacteria.
Craun (1999) conducted a study of microbial waterborne outbreaks reported to the Center for
Disease Control for ground water systems in the United States during the 26-year period from
1971-1996. Of the outbreak illnesses in ground water systems caused by bacteria observed during the
1971-1996 period, the majority (53 percent) were due to Shigella, followed by Salmonella (23
percent), Campylobacter (11 percent), and Escherichia coli O6:H16 (10 percent). Highlighted
below in Exhibit 5-11 are the typical clinical characteristics as well as any associated complications and
the annual incidence of these four bacteria.
Most frequently, these bacteria cause
acute gastroenteritis, although some have been
shown to cause chronic conditions. For
example, E. coli O157:H7 infection is the
leading cause of hemolytic uremic syndrome
(HUS), the most common cause of acute
kidney failure in children, Salmonella infections
may lead to reactive arthritis, and
Campylobacter infections can lead to Guillain-
Barre syndrome, one of the most common
causes of paralysis.
Impacts on Sensitive Populations
Young children are more likely to develop HUS from
£. co//O157:H7 hemorrhagic colitis (approximately
eight percent of children). When diagnosed, they
would consistently require blood transfusions or
kidney dialysis during their prolonged hospital stays.
Similar requirements exist for AIDS patients who
tend to suffer more severe salmonellosis not only
about twenty times more often than the general
population, but also suffer more recurrent episodes
(Altekruse et al., 1997). This indicates that unlike
viral and other bacterial infections, rates of
salmonellosis are reportedly higher in sensitive
populations than in the general population, and
would therefore be associated with higher costs.
5-16
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit 5-11. Summary of Clinical Characteristics
Selected Bacterial Waterborne Pathogens
of
Organism Campylobacter spp.
Common types
Acute Disease
watery diarrhea
bloody diarrhea
vomiting
fever
Illness Duration
Treatment
Annual Cases of
Illness from All
Exposure
Pathways
Complications
C.jejuni, C. fetus
gastroenteritis
F
»
»
7-10 days
self-limiting antibiotics in
severe cases
2 to 4 million
> relapse in 20% of
cases
> 0.1 case-fatality
> arthritis, hemolytic
uremic syndrome
(HUS), meningitis,
recurrent colitis,
Guillain-Barre
syndrome
Escherichia coli
0157:H7
hemorrhagic colitis
»
F
5-10 days
oral rehydration; avoid anti-
diarrheal agents; antibiotics
not recommended
10,000-20,000
> up to 50% mortality seen in
the elderly, 3-5% in HUS
patients
> HUS, renal failure, coma
Salmonellaspp.
S. typhi, S. paratyphi, S.
typhimurium, S. enteritidis
gastroenteritis, typhoid fever
»
2-7 days
self-limiting; oral rehydration;
antibiotics not recommended
40,000-50,000
> 10% fatality in typhoid fever
vs. 1%in other cases
> reactive arthritis, Reiter's
syndrome
Shigellaspp.
S. sonnei, S. flexneri, S.
boydii, S. dysenteriae
bacillary dysentery
»
4-7 days
self-limiting; oral rehydration
or antibiotics in severe
cases
25,000-30,000
* 10-1 5% fatality
> Reiter's disease,
reactive arthritis, HUS
= typical feature 1 = common finding F = occasionally reported
Source:Rivera-Matos(1996); Fauci (1998); NFID(1996); US FDA (1998)
Estimates for Unit Costs of Illness: Bacterial Infections
The clinical profiles of the bacteria highlighted above suggest that unit COI estimates for
bacterial illnesses in the general population would primarily be driven by the indirect cost component, as
few direct medical costs are associated with the favorable prognoses of acute bacterial illnesses
described. Medical attention is usually not required for simple oral rehydration treatment, and
antibiotics are usually not recommended due in part to the emerging drug-resistant nature of many
bacterial strains. Given the self-limiting nature of bacterial disease, the direct cost component of a single
case is expected to be relatively low. In contrast, the indirect cost component for a typical case is high
due to the number of caretaker and work loss days the victim accumulates until the symptoms subside.
In the relatively few cases in which symptoms are severe, however, the direct costs will be
higher due to the medical attention and the specific medical treatment (e.g., antibiotics, dialysis,
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
5-17
-------
colectomy, hospital costs) sought by victims suffering from illnesses lasting over a week, or from
complications that develop into chronic conditions. For example, because of such costs associated
with the bloody nature of hemorrhagic colitis and the likelihood of HUS complications as a result of E.
coli O157:H7 illness (about 2 to 7 percent of infections lead to this complication), the base unit COI
estimate for E. coli O157:H7 is likely to be higher than that of Campy lobacter, Salmonella, or
Shigella, mainly as a result of the higher percentage of victims that would seek medical attention and
treatment. Similarly, the indirect costs would be higher in severe cases as a result of the prolonged
duration of illnesses.
These bacterial profiles provide a summary of the clinical nature of illness, as well as of the
likely treatment course for an affected individual. This information was useful in defining the potential
magnitude of a case of bacterial illness. However, due to data limitations in both the risk assessment
and the COI estimates, a simplified assumption was used to monetize these health benefits from avoided
bacterial illnesses under the GWR.
In the aforementioned study of microbial waterborne outbreaks from 1971-1996 (Craun,
1999), outbreaks caused by bacterial agents caused 12,860 reported illnesses during this period,
compared to 69,572 illnesses which were attributed to viral or unknown etiologic agents. Assuming
that all illnesses classified as unknown etiologies are viral in nature, EPA estimates the ratio between
waterborne bacterial illness and viral illness in ground water systems to be 0.2. EPA therefore
assumed, for this analysis, that the additional health benefits associated with avoided bacterial illnesses
are proportionally equal to 20 percent of the benefits due to reduced viral risk.
5.3.4 Results of Monetization of Health Benefits (Acute Illnesses)
The results of the monetization of health benefits for each of the four regulatory scenarios are
presented in the following exhibit. Following that summary are four tables that present a comparison of
the distribution of acute health benefits across the four regulatory options. The distribution of benefits
are presented by pathogen type, by the health status of the consumer, by the size of the system, and by
the type of system. More detailed results (e.g., including percentiles, breakouts among pathogens, and
health effects types) may be reviewed in Appendix B.
As shown in Exhibit 5-12, total benefits for the three risk-based regulatory scenarios range
from $32.5 million for Option 1, sanitary survey alone, to $205.0 million for Option 3, the Multi-Barrier
approach. The fourth option, in which all GWSs are subject to disinfection, results in benefits of
$283.1 million. The largest increase in benefits is between Options 1 and Option 2 with a total
difference of $145.4 million. Option 2 and 3 differ by just over $27 million, and Options 3 and 4 differ
by just under $78 million.
518 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
Exhibit 5-12. Health Benefits for Regulatory Scenarios ($Millions)
GWR Regulatory Options
Option 1 : Sanitary Survey Only
Option 2: Sanitary Survey and
Triggered Monitoring
Option 3: Multi-Barrier
Approach
Option 4: Across-the-Board
Disinfection
Mean Benefits
(10th to 90th percentile)
Morbidity
$21.9
($6. 8 to $37. 9)
$120.4
($100.6to$140.4)
$138.8
($114.7to$163.2)
$192.0
($173.7to$210.2)
Mortality
$10.6
($2.0 to $19.7)
$57.5
($46.8 to $68.2)
$66.2
($53. 8 to $78. 7)
$91.1
($81.3to$100.9)
Total
$32.5
($8. 8 to $57. 6)
$177.9
($147 to $208. 6)
$205.0
($168.5to$241.9)
$283.1
($255.1 to $31 1.1)
In all cases the majority of the total benefits are from morbidity benefits. For all four options,
the proportion of the benefits that is attributable to reduced mortality is approximately one-third of the
total benefit.
Exhibit 5-13 presents the breakout of total benefits among the control of the three types of
pathogens, Type A viral, Type B viral, and bacterial infections. The benefits received from controlling
Type B infections make up the majority of the total benefits, as may be expected given the severity of
the health impacts from this agent.
Exhibit 5-13. Distribution of Mean Total Benefits: By Pathogen Type
Option/Regulatory Scenario
Option 1 : Sanitary Survey Only
Option 2: Sanitary Survey and Triggered
Monitoring
Option 3: Multi-Barrier Approach
Option 4: Across-the-Board Disinfection
Percent of Total, by Pathogen
Type A
19%
20%
20%
20%
Type B Bacterial
64%
63%
63%
63%
17%
17%
17%
17%
Mean Total
Benefits
($Millions)
$32.5
$177.9
$205.0
$283.1
Exhibit 5-14 presents the breakout of total benefits between healthy population and the
immunocompromised populations. The immunocompromised populations make up approximately 5
percent of the total benefits for all options.
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
5-19
-------
Exhibit 5-14. Distribution of Mean Total Benefits: By Health Status
Option/Regulatory Scenario
Option 1 : Sanitary Survey Only
Option 2: Sanitary Survey and Triggered
Monitoring
Option 3: Multi-Barrier Approach
Option 4: Across-the-Board Disinfection
Percent of Total,
by Subpopulations
Healthy
95%
95%
95%
95%
Immunocompromised
5%
5%
5%
5%
Mean Total
Benefits
($Millions)
$32.5
$177.9
$205.0
$283.1
Exhibit 5-15 presents the breakout of total benefits among the eight sizes of systems, as defined
by number of individuals served by each system. The benefits accrued are roughly proportional to the
population served, although for all options, the three smaller categories tend to receive a somewhat
lesser portion of the benefits (i.e., given that 12.5 percent of benefits to each size category would be
proportionally even) than four of the five categories representing larger populations systems.
Exhibit 5-15. Distribution of Mean Total Benefits (Viral Only): By System Size
Option/Regulatory Scenario
Option 1 : Sanitary Survey Only
Option 2: Sanitary Survey and
Triggered Monitoring
Option 3: Multi-Barrier Approach
Option 4: Across-the-Board
Disinfection
Percent of Total, By System Size
0
0
T-
V
3%
4%
4%
4%
0
i §
IO T-
I I
O O
0 0
T- 10
9%
10%
10%
10%
7%
8%
8%
8%
*: o
CO T-
<"> o
3 5
^ CO
T- CO
15%
15%
15%
15%
18%
18%
18%
18%
10Kto50K
50Kto100K
12%
11%
11%
11%
18%
18%
18%
18%
100Kto1M
16%
16%
16%
16%
Mean Total
Benefits
($Millions)
$27.1
$148.2
$170.8
$235.9
Exhibit 5-16 presents the breakout of total benefits among the three types of systems (i.e.,
community versus non-community, and of the latter, transient versus non-transient). The benefits
accrue in the majority to the community water systems and least of all to the non-community, non-
transient systems.
5-20
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit 5-16. Distribution of Mean Total Benefits: By System Type
Option/Regulatory Scenario
Option 1 : Sanitary Survey Only
Option 2: Sanitary Survey and Triggered Monitoring
Option 3: Multi-Barrier Approach
Option 4: Across-the-Board Disinfection
Percent of total,
by System Type
CWS
86%
82%
82%
82%
NTNC TNC
12%
13%
13%
13%
1%
5%
5%
5%
Mean Total
Benefits
($Millions)
$32.5
$177.9
$205.0
$283.1
5.4 Other Benefits (Unquantified)
EPA recognizes that, in addition to the benefits associated with reductions in acute illness and
death from viral and bacterial infection, the GWR will provide other benefits. As illustrated in Exhibit
5-1, total benefits also include chronic heath benefits as well as non-health benefits. EPA was not able
to monetize either of these. The benefits gained, however, are not inconsequential, merely unable to be
assigned a dollar value. The following discussions are presented to provide the reader with some
understanding of the potential magnitude of these benefits.
5.4.1 Reduced Pain and Suffering
As mentioned earlier in this chapter, the true value of reducing acute morbidity is an individual's
willingness-to-pay (WTP) to avoid a case of illness. Like the cost-of-illness estimates used in this RIA,
WTP estimates include the direct medical costs and indirect productivity losses associated with an
illness. However, WTP estimates also include some costs that the cost-of-illness approach is unable to
capture; including the value of avoiding the pain and suffering associated with acute microbial illness.
Cost-of-illness estimates do not include the value of reduced pain and suffering because the disutility of
illness is not associated with a market cost. Since pain and suffering is not associated with a market
cost, placing a value on it is not possible without resorting to primary survey based research, such as a
contingent valuation study.
When considering the self-limiting nature of viral and bacterial illness, especially in healthy
adults, it is reasonable to assume that the value of reducing the pain and suffering associated with acute
microbial illness is a significant portion of the total WTP to avoid the illness. Therefore, the agency
recognizes that this analysis may significantly underestimate the true value of reduced acute morbidity
resulting from the proposed GWR.
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
5-21
-------
5.4.2 Reduced Chronic Illness
While chronic illnesses were not included in the monetized benefits summarized in Section 5.3,
a review of related health impacts reveals that the potential benefits from avoiding these chronic impacts
may in fact be substantial. While this RIA does not quantify in dollar terms the benefit of avoiding
chronic illness and deaths, this section discusses the potential benefits qualitatively and illustrates the
significance of these secondary benefits.
Although a review of the medical and epidemiological literature identified several potential
chronic diseases resulting from illnesses caused by enteroviruses (e.g., heart disease, diabetes, post-
viral fatigue syndrome, and pancreatitis), the strongest evidence for a viral role appears to exist for the
development of diabetes and myocarditis (inflammation of the muscular walls of the heart).
Because the causal relationship is not well established and the number of cases associated with
drinking water is unknown, the Agency was not able to quantify benefits from the GWR on reducing
these diseases. Nonetheless, as illustrated in Exhibit 5-17, the total number of these conditions from all
pathways in the United States is substantial. Additionally, 3.5 percent of heart disease deaths (study
was for 1993) were due to cardiomyopathy (NHLBI, 1996). The potential benefits of avoiding some
of these health effects cannot be overlooked, and may be significant.
Exhibit 5-17. Annual Number of People with Selected Disease
Selected Disease
Diabetes (of all kinds)
Chronic Heart Disease (including myocarditis and cardiomyopathy)
Number of people1
6,962,000
4,148,000
1 Average annual number of people with disease forthree year period from 1990-1992.
Source: Collins, 1997
Although enteroviruses are suspected to play a role in the development of chronic illnesses as
discussed, sufficient data are not available to forecast the number of avoided chronic cases resulting
from the proposed GWR. An extensive literature review proved, however, that costs of a single case
of diabetes or heart disease are significant. Cost estimates for a case of diabetes and a case of chronic
myocarditis (using the cost per case of an "average case of heart disease" as a proxy for chronic
myocarditis) are presented below to demonstrate the magnitude of potential benefits per avoided case
of chronic illness.
5.4.2.1 Type I Diabetes
The potential involvement of enteroviral infection in Type 1 insulin-dependent diabetes mellitus
(IDDM) has previously been suggested (Nakao, 1971) and is being researched by the American
Diabetes Association (ADA, 1996). In some people, autoimmune reactions to enterovirus have been
observed to destroy the pancreatic cells, which produce insulin. This autoimmunity appears to develop
5-22 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
in less than 5 percent of the general population and progresses to diabetes in less than 1 percent of the
general population.
Enteroviral infection during pregnancy of the mother has also been observed as a risk factor for
childhood-onset diabetes (Dahlquist et al., 1995). It is suggested that an enterovirus infection initiates
and accelerates the autoimmune process typical of IDDM cases, rather than actually causing the clinical
illness (Hyoty et al., 1995).
Costs of Illness: Diabetes
The most comprehensive work regarding the economic burden of diabetes in the United States
was conducted for the American Diabetes Association. In their report "Economic Consequences of
Diabetes Mellitus in the United States in 1997," Fox et al. (1998) presented the direct medical and
indirect costs attributable to diabetes, as well as a total and per capita estimate of expenditures of
people with and without diabetes. Improving on their estimates and methodology from their 1992 effort
(Fox et al., 1993), this national prevalence-based COI study also compares the health care
expenditures of diabetics in 1997 to non-diabetics.
The authors created a holistic estimate of the health care expenditures attributable to diabetes in
1997, by including: 1) medical expenditures attributable to diabetes (i.e., the cost due to the excess of
prevalence of diabetes related chronic complications and general medical conditions in people with
diabetes), and 2) total medical expenditures incurred among people with diabetes (i.e., the cost for all
services for people with diabetes). Annual per capita expenditure estimates were also calculated and
defined as the sum of the expenditures among diabetics in 1997, divided by the 1997 diabetic
population. The estimates do not, however, include pain and suffering nor do they include lost
productive and leisure time
The per capita medical expenditures for people with diabetes is $10,825 for people with
diabetes versus $2,869 among people without diabetes. Therefore the annual cost of diabetic care is
therefore $7,956 per person.1 The net productivity loss for each person due to diabetes totaled $1,567
for 18-64 year olds and $502 for those 65 and older.2 These are sums of costs attributable to diabetes
from productivity loss from work, from restricted-activity and from bed-disability.
According to the 1980-1987 Hospital Cost and Utilization Project (HCUP), a national sample
of more than 500 hospitals, which represent an unweighted 20 percent sample of discharges, the mean
age of diagnosis for a case of diabetes mellitus within the study was 53 (Elixhauser et al., 1993).
Assuming that a patient incurs treatment for diabetes each year throughout the duration of his expected
Costs updated from January 1995 dollars to May 1999 dollars using the CPI-U for "medical care services"
(=254.0-236.3=1.0749).
2 Costs updated from January 1997 dollars to May 1999 dollars using the CPI-U for "all items" (= 166.2 -
159.1 = 1.0446).
AprilS, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 5-23
-------
life from age 53,3 the present value of the direct medical costs and indirect costs of illnesses would be
$101,775 (7 percent discount rate). This figure could be even greater if the cost of premature death or
pain and suffering were incorporated. While this is a simple approximation of the magnitude of a COI
value for this illness, it captures the lifetime costs of diabetes in those who survive the first year through
their life expectancy period from the age of diagnosis.
5.4.2.2 Chronic Myocarditis
The enteroviruses are reportedly responsible for approximately 50 percent of the myocarditis
cases in North America (Luppi et al., 1998). Coxsackievirus B infections are increasingly the primary
cause of myocardial disease in both adult and children populations. Melnick (1996) reported that up to
39 percent of persons infected with coxsackievirus develop cardiac abnormalities. Furthermore, about
5 percent of all symptomatic coxsackievirus patients induce heart disease, affecting the myocardium
(the heart muscle), the pericardium (the membranous sac around the heart), the endocardium (the
interior lining of the heart), or all three.
While many of these cases resolve without complication, it is believed that some acute cases
resurface as chronic conditions (Sainani et al., 1968, Abelmann, 1978, Archard et al., 1987, Luppi et
al., 1998, Hufnagel, 1998). This may occur in infected individuals, depending on viral or host factors
such as the virulence of the virus strain, the character of the virus, or the immunity of the patient (Okuni
etal., 1975).
Clinically, it is difficult to differentiate between cardiomyopalhy and chronic myocarditis (Okuni
et al., 1975). Therefore, it is thought that viral myocarditis actually may be responsible for some of
these cases that are diagnosed as cardiomyopathy (Abelmann, 1978). For example, the Idiopathic
Cardiomyopathy Research Committee of Japan reported that 40 percent of patients suffered
myocardial sequelae and about 4 percent of them showed dilated cardiomyopathy-like features (Kawai
et al., 1987). In fact, Archard et al. (1987) suggested that group B coxsackieviruses may actually play
a role in the development of dilated cardiomyopathy. Therefore, it is not uncommon for a patient to be
suffering chronic myocarditis and to only be diagnosed with it after death (Kline and Saphir; 1960,
Smith, 1966; Morimoto et al., 1992). They will have received treatment throughout life for conditions
such as congestive heart failure or dilated cardiomyopathy, and not for chronic myocarditis (Smith,
1966; Morimoto et al., 1992; Luppi et al., 1998).
Costs-of-Illness: Chronic Myocarditis
The annual direct cost-of-illness associated with an "average case of heart disease" was
estimated to be $4,559.4 This estimate was derived from data originally computed by Thomas
26.9 years: Life Tables. Table 6-3. "Expectation of Life at Single Years of Age, by Race and Sex: United
States, 1995."(NCHS, 1998).
Cost updated from January 1995 dollars to May 1999 dollars using the CPI-U for "medical care services" (=
254.0-219.8 = 1.1556).
5-24 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
Hodgson5 (Hodgson, 1984; Hodgson, 1998) of the National Center for Health Statistics (NCHS) for
"heart disease" (which includes International Class of Diseases, 9th Revision (ICD-9) codes 391-398,
402, 404, 410^16, 420-429). Since no specific cost data were available for chronic myocarditis, or
cardiomyopathy (ICD-9 code 425), annual per capita costs for an "average case of heart disease"
were computed using his data on "heart disease" in conjunction with prevalence numbers from the 1995
National Health Interview Survey.6'7
Indirect costs of "other heart disease" were estimated by Cropper and Krupnick (1990) who
used information from the 1978 Social Security Survey of Disabled and Work to model the effects of
disease on labor participation and earnings. Cropper and Krupnick found that the annual indirect cost
ranged from $3,264 to $6,699 depending on the age of the individual and the age of illness onset.8
Again, it is important to note that these costs do not include pain and suffering.
According to the 1980-1987 HCUP study of 500 hospitals, the mean age of diagnosis for
cardiomyopathy was 60 (Elixhauser et al., 1993). Using this diagnostic category as a proxy for chronic
myocarditis,9 the lifetime cost-of-illness could be substantial. For example, the present value of both
direct and indirect costs for a patient with the condition would be $52,971 given an average life
expectancy of 21.1 years (7 percent discount rate). This figure could be even greater if the costs of lost
earnings and of premature death were incorporated.
5.4.3 Non-Health Benefits
In addition to the quantified and unquantified health-based benefits discussed above, there are a
number of non-health benefits that also arise from promulgation of the rule. Non-health benefits may
result from overall system improvements (e.g., upgrades to distribution systems, increased efficiencies,
increased frequency/intensity of process surveillance), from improved risk perception of drinking water
quality, or from avoided outbreak response costs.
Chief economist and acting director, Division of Health and Utilization Analysis, NCHS, CDC.
Chronic illness prevalence rates (cases per 1,000 individuals) for "heart disease" were multiplied by the
total United States population to obtain the total number of heart disease cases in 1995. The "average case of heart
disease" per person in 1995 was subsequently calculated by dividing the total cost of heart disease in 1995 by the
total number of heart disease cases in 1995. Prevalence figures were from Current Estimates of the National Health
Interview Survey, 1995 (Benson and Marano, 1998), and the total United States population as of January 1, 1996 was
obtained from the Census Bureau.
Without more detailed information, this simplified method assumes that the cost of any heart disease,
whether ischemic or other, would be the same within this major disease group. This is a major limitation of these
estimates, as hospital costs for coronary heart disease may not be the same for hypertensive disease, for example.
o
Costs updated from January 1977 dollars to May 1999 dollars using the CPI-U for "all items: (= 166.2 -=-
58.5 = 2.8410).
o
As previously mentioned, group B coxsackieviruses are suspected to play a role in the development of
dilated cardiomyopathy.
AprilS, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 5-25
-------
These non-health benefits are not quantified for this RIA. The Agency has considered these
benefits, however, and presents the following discussion to illustrate their potential magnitude.
5.4.3.1 Reduced Uncertainty
To the extent that the GWR decreases consumers' uncertainty about expected health outcomes
from consumption of drinking water, the rule should provide direct benefits independent of risk
reduction benefits. In other words, drinking water consumers may be willing to pay a risk premium for
regulatory action if it reduces their uncertainty about whether they will become ill or not (Moore, 1990).
Conceptually, whether or not consumers would be willing to pay something extra to reduce
uncertainty in the GWR context depends on several complicated factors, including consumers' degree
of risk aversion, their perceptions about drinking water quality, and the expected probability and
severity of human health effects associated with microbial contamination of drinking water. For
example, risk premiums would be expected only for consumers who are risk averse. Further, the
magnitude of any premium would be expected to be positively related to the probability and severity of
expected health outcomes, and the degree to which consumers' perceive them to be affected by
regulatory action.
5.4.3.2 Costs to Households to A vert Infection
To the extent that the GWR can be expected to
reduce a household's perceptions of the health
risks associated with drinking water, regulatory
action should reduce household averting actions
i , A i , 11 potential microbial contamination of publicly-
and costs. Any such cost savings would Jupp|jed drinkjng ^^ inc|uding; ^secujng
represent a regulatory benefit. A number of
factors, however, limit the relevance of this
Examples of Household Avoidance
Individual households often take steps to avoid
drinking water from alternative sources (e.g., bottled
water), 2) installation of home treatment systems
actions can involve significant cash outlays and
implicit costs (e.g., time costs).
^ ^ i u c± ^1 /^iim ^ ^ r\ >e.g., point-of-use and point-of-entry treatment), and
potential benefit in the GWR context. One is ^ ^ tap water ^ for consuymption Th;ese
the possibility that regulatory action may not
affect household perceptions of health risks
enough to motivate them to forego averting
actions. A related factor is that many
households that undertake averting action for health reasons may be especially risk averse (e.g.,
households with infants or immuncompromised persons). These households might be expected to
pursue averting actions regardless of the level of regulatory control if they believe such actions may
provide added protection against microbial risks. Should this be the case, households would also be
excluded from the quantified benefit analysis.
5-26 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
5.4.3.3 Outbreak Response Costs A voided
Examples of Outbreak Response Costs
Affected water systems and local governments can
incur costs associated with providing alternative
water supplies and issuing customer water use
warnings and health alerts. Commercial
establishments (e.g., restaurants) and their
customers can incur costs due to interrupted and
lost service (i.e., lost producer and consumer
surplus). Local businesses, institutions (e.g.,
schools), and households can all incur costs
While outbreak prevention generates the
health benefits discussed above, it also results in
significant non-health benefits. To the extent that
the GWR reduces the likelihood of illness
outbreaks, these avoided response costs are
potentially numerous and varied. A literature
review identified five studies that use the averting
cost approach to estimate household and other
costs resulting from short-term contamination
episodes of drinking water supplies (Abdalla,
1990; Abdalla et al., 1992; Harrington et al.,
1985; Sun et al., 1992; RTI, 1997). The most relevant of these for the GWR analysis is a study by
Harrington et al. (1985) in which the resulting costs of drinking water contamination by Giardia in
Luzerne County, Pennsylvania, were evaluated (see Exhibit 5-18). The outbreak provides a theoretical
and empirical example of how outbreak costs are incurred by individuals, businesses, and local
governments, in an overall exposed population of 75,000.
Exhibit 5-18. Case Study of Outbreak Costs-1984 Lucerne County Outbreak1
Explanation
Losses due to actions taken by individuals to avoid the contaminated water. The
predominant cost was due to time losses involved in boiling water.
Cost of providing alternative water for restaurants, bars, schools and other
businesses during the outbreak.
The burden for government agencies.
The burden for the water supply utility.
Costs (Millions)2
$21.11 to $62. 813
$1.06
$0.37
$3.0
1 Source: Harrington etal. (1985)
2 All costs were updated to May 1999 dollars using the CPI-U for"all items," from 1984 dollars (= 166.2 * 101.9 =
1.6310
3 Depending upon the assumed method of estimating the implicit after-tax wage rate of the unemployed,
homemakers, and retirees
The primary concern of the study was on the total losses resulting from the outbreak including
the value of lost work time and the value of reduced leisure time activities due to illness, the cost of
medical care, the costs of actions taken to avoid drinking contaminated water, such as the cost of
bottled water and boiling water, the costs of epidemiological and water system surveys, and the costs of
temporary measures taken by the water utilities to ensure safe water supplies. Unfortunately, the study
was not able to address any losses associated with pain and suffering, or "with anxiety over the
possibility of contracting giardiasis, and with diminished intrinsic value, resulting from the loss of a 'pure'
water supply for drinking." (Harrington, 1985). Outbreak effects on businesses regarding lost sales or
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
5-27
-------
lost productivity were also not investigated. These additional factors could add significantly to the cost
of waterborne disease outbreaks, therefore the benefit from avoiding outbreaks could be even greater.
5.4.4 Benefits From the Reduction of Co-Occurring Contaminants
If a system chooses to install treatment, it may choose a technology that would also address
other drinking water contaminants. For example, when using packed tower aeration to treat radon, it is
the accepted engineering practice, and in some States an existing requirement, to also install disinfection
treatment for removal of microbial contaminants introduced in the aeration treatment process.
Depending on the dosage and contact time, the routine disinfection would also address possible or
actual fecal contamination in the source water. If systems had an iron or manganese problem, the
addition of an oxidant and filtration can treat this problem as well as fecal contamination. Also, some
membrane technologies installed to remove bacteria or viruses can reduce or eliminate many other
drinking water contaminants including arsenic. EPA is currently proposing rules to address both radon
and arsenic. Because of the difficulties in establishing which systems would have all three problems of
fecal contamination, radon, and arsenic or any combination of the three, no estimate was made of the
benefits from the overlap of these rules.
5-28 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
5.5 References
Abdalla, C.W. 1990. Measuring economic losses from ground water contamination: An investigation of
household avoidance costs. Water Resources Bulletin 26:451-63.
Abdalla, C.W. et al. 1992. Valuing environmental quality changes using averting expenditures: An
application to groundwater contamination. Land Economics 68(2): 163-9.
Abelmann, W.H. 1978. Viral myocarditis and its sequelae. Annual Review of Medicine 24:145-52.
Altekruse, S.F. et al. 1997. Emerging foodborne diseases. Emerging Infectious Diseases
3(3):285-93.
American Diabetes Association (ADA). 1996. "Making a difference: Type I Diabetes Research.
Nationwide Research Program." .
Angulo, F.J. et al. 1997. A community waterborne outbreak of salmonellosis and the effectiveness of a
boil water order. Am JPublic Health 87(4):580^.
Apogee/Hagler Bailly, Inc. 1998. Potential Benefits of the Ground Water Disinfection Rule.
Prepared for the U.S. Environmental Protection Agency, Office of Ground Water and Drinking
Water by Apogee/Hagler Bailly, Inc. under subcontract to International Consultants, Inc. March.
Archard, L.E. et al. 1987. The role of coxsackie B viruses in the pathogenesis of myocarditis, dilated
cardiomyopathy and inflammatory muscle disease. Biochem Soc Symp 53:51-62.
Association of State Drinking Water Administrators (ASDWA). 1997. Analysis of Best Management
Practices for Community Ground Water Systems Survey Data
Benson, V. and M.A. Marano. 1998. Current Estimates of the National Health Interview Survey,
1995. National Center for Health Statistics. Vital Health Stat 10(199).
Bureau of Labor Statistics (BLS). "Current Series" of BLS Data, Consumer Price IndexAll Urban
Consumers, not seasonally adjusted. Data extracted on 21 June 1999.
Collins, J.G. 1997. Prevalence of Selected Chronic Conditions: United States, 1990-1992.
Hyattsville, MD: National Center for Health Statistics. Vital Health Statistics 10(194). DHHS
Publication No. (PHS) 97-1522.
Craun, G.F. 1986. Statistics of waterborne outbreaks in the U.S. (1920-1980). In Craun, GF. et al.
Waterborne Diseases in the United States. Boca Raton, FL: CRC Press, Inc., pp. 73-158.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 5-29
-------
Craun, G.F. and R. Calderon. 1997. Microbial risks in groundwater systems: Epidemiology of
waterborne outbreaks. Under the Microscope: Examining Microbes in Groundwater.
AWWARF.
Craun, G.F. 1998 personal communication.
Craun, G.F. 1999 personal communication.
Cropper, M.L. and AJ. Krupnick. June 1990. The Social Costs of Chronic Heart and Lung
Disease. Resources for the Future Discussion Paper QE89-16-REV.
Dahlquist, G.G. et al. 1995. Maternal enteroviral infection during pregnancy as a risk factor for
childhood IDDM. Diabetes 44:408-13.
Dirckx, J.H., ed. 1997. Stedma's Concise Medical Dictionary for the Health Professions, 3rd ed.
Baltimore, MD: Williams & Wilkins.
Elixhauser, A. et al. 1993. Clinical Classifications for Health Policy Research: Discharge Statistics
by Principal Diagnosis and Procedure. Division of Provider Studies Research Note 17, Agency
for Health Care Policy and Research, Rockville, MD: Public Health Services. AHCPR Publ. No.
93-0043.
Fauci, A.S. et al., eds. 1998. Harrison's Principles of Internal; Medicine, 14th ed. NY: McGraw-Hill
Companies, Inc., Health; Professional's Division, Part 7 "Infectious Diseases."
Fox, N.R. et al. 1993. Direct and Indirect Costs of Diabetes in the United States in 1992.
American Diabetes Association. Alexandria, VA.
Fox N.R., et al. 1998. Economic consequences of diabetes mellitus in the U.S. in 1997. Diabetes
Care 21(2): 296-3 09.
Harrington, W. et al. 1985. The Benefits of Preventing an Outbreak ofGiardiasis due to Drinking
Water Contamination. Draft final report prepared for the U.S. EPA by Resources for the Future.
Washington, DC. September.
Hodgson, T.A. 12 November 1998. Personal communication.
Hodgson, T.A. and A.N. Kopstein. 1984. Health care expenditures for major diseases in 1980.
Health Care Financing Review. 5(4): 1-12.
Hufnagel, G 1998. Symptoms, diagnosis and treatment of myocarditis and dilated cardiomyopathy
(DCM). .
5-30 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
Hyoty, H. et al. 1995. A prospective study of the role of coxsackie B and other enterovirus infections in
the pathogenesis of IDDM. Diabetes 44:652-7.
International Classification of Diseases, 9th Revision, Clinical Modification. 5th edition, volume 1.
1994. Washington, DC: U.S. Government Printing Office. DHHS Publ. No. PHS 94-1260.
December.
Kawai, S. et al. 1987. A morphological analysis of chronic myocarditis. Japanese Circulation Journal
51:1385-91.
Kline, I.K. and O. Saphir. 1960. Chronic pernicious myocarditis. Am Heart J59(5):681-97'.
Lieberman, R.J., Shadix, L.C., Newport, C.P. Frebis, M.W.N. Moyer, S.E., Safferman, R.S., Stetler,
R.E., Lye, D., Fout, G.S., and Dahling, D. 1999. "Source water microbial quality of some
vulnerable public ground water supplies." Unpublished report in preparation.
Lieberman, R.J., L.C. Shadix, B.S. Newport, S.R Crout, S.E. Buescher, R.S. Safferman, R.E.
Stetler, D. Lye, G.S. Fout, and D. Dahling. 1994. "Source water microbial quality of some
vulnerable public ground water supplies." in Proceedings, Water Quality Technology Conference,
San Francisco, CA, October, 1994.
Luppi, P. et al. 1998. Idiopathic dilated cardiomyopathy: A superantigen-driven autoimmune disease.
Circulation 98:777-85.
Melnick, J.L. 1996. Enteroviruses: Polioviruses, coxsackieviruses, echoviruses, and newer
enteroviruses. In Fields, B.N. et al., eds. Virology, 3rd ed., vol. 2. Philadelphia: Lippincott-Raven
Publishers, pp. 655-712.
Moore, MJ. and W. Kip Viscusi. 1990. Compensation Mechanisms for Job Risks: Wages,
Workers' Compensation, and Product Liability. Princeton, NJ: Princeton University Press.
Morens, D.M. 1978. Enteroviral diseases in early infancy. JPediatr 92(3):374-7.
Morimoto, S-I. et al. 1992. Clinical and pathological features of chronic myocarditis: Four autopsy
cases presenting as dilated cardiomyopathy. Am J Cardiovascular Pathology 4(2): 181-91.
Nakao, T. 1971. Coxsackie viruses and diabetes. Lancet 2:1423.
NCHS (National Center for Health Statistics). 1998. Vital Statistics of the United States, 1995,
Preprint of Volume II, Mortality, Part A Sec 6 Life Tables. Hyattsville, MD.
NFID (National Foundation for Infectious Diseases). 1996. Bethesda, MD.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 5-31
-------
National Health, Lung and Blood Institute (NHLBI), National Institutes of Health. 1996. Morbidity
and Mortality: 1996 Chartbookon Cardiovascular, Lung, and Blood Diseases. May.
National Research Council. 1997. Valuing Ground Water: Economic Concepts and Approaches.
Washington, DC: National Academy Press.
National Water Research Institute. 1997. Groundwater Disinfection Regulation Workshop. January
Okuni, M. et al. 1975. Studies on myocarditis in childhood, with special reference to the possible role
of immunological process and the thymus in the chronicity of the disease. Japanese Circulation
Journal 39:463-70.
Research Triangle Institute (RTI). 1997. Valuing Drinking Water Quality: Theory, Methods, and
Research Needs. Draft report prepared for the USEPA. April.
Rivera-Matos, I.R. and T.G. Cleary. 1996. Foodborne and waterborne illness in children. In Advances
in Pediatric Infectious Diseases. MosbyYear Book, Inc. Volume 11, pp. 101-34.
Sainani, G.S. et al. 1968. Adult heart disease due to the coxsackie virus B infection. Medicine
47(2): 133^17.
Smith, W.G. 1966. Adult heart disease due to the coxsackie virus group B. Brit Heart .728:204-20.
Sun, J. et al. 1992. Estimating the benefits of groundwater contamination control. Southern Journal of
Agricultural Economics 24(2): 63-71.
Tolley, G. etal., eds. 1994. Valuing Health for Policy: An Economic Approach. Chicago: University
of Chicago Press.
U.S. Bureau of the Census. Census, residential population estimates by month and single year of age.
U.S. EPA. 1995. A Framework for Measuring the Economic Benefits of Ground Water. U.S.
EPA Office of Water and Office of Policy, Planning, and Evaluation. October.
U.S. FDA. 1998. Center for Food Safety & Applied Nutrition. Foodborne Pathogenic
Microorganisms and Natural Toxins Handbook. Washington, DC.
5-32 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
6. Cost Analysis
6.1 Introduction
This chapter presents the national cost estimates for the proposed Ground Water Rule (GWR). It
presents the unit costs, identifies the underlying assumptions used to prepare the cost estimates, and
describes the methodology used to compile these assumptions to estimate the cost of the four GWR
options.
Section 6.2 presents an introduction to the unit costs and costing assumptions that EPA has made
for each of the rule components (Section 6.2.1) and the methodology used by the cost model to
develop national estimates (Section 6.2.2). Section 6.3 presents the national cost estimates and
Section 6.4 presents the average household costs.
6.2 Costing Methodology
This section presents a summary of EPA's assumptions made to prepare estimates of the national
costs of the proposed GWR and other regulatory options. It contains a description of the estimates of
unit costs (the cost that would be incurred by each State, individual treatment facility or system) and the
predicted actions that systems and States will make to comply with the proposed GWR.
6.2.1 Cost Model Inputs
The proposed GWR and other rule options are composed of rule components that identify and/or
correct conditions that permit fecal contamination to reach ground water system consumers' taps.
Exhibit 6-1 identifies the components of each GWR option. Several of these components are included
in more than one of the GWR options.
Exhibit 6-1. Components Included in Each Regulatory Options
Rule Scenario Components
Option 1 :
Sanitary Survey
Only
Option 2:
San. Survey
and Triggered
Monitoring
Monitoring and Assessment
Sanitary Survey
Triggered Monitoring
Sensitivity Assessment
Routine Monitoring
Corrective Action
For Significant Defects
For Fecal Contamination
Compliance Monitoring
Option 3:
Multi Barrier
Approach
Option 4:
Across-the-
Board
Disinfection
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
6-1
-------
The unit costs applicable to each may be either costs incurred by the State in establishing the
component's requirements or by the water system to comply with the regulations; for some components
both entities incur costs. Exhibit 6-2 presents, for each of the components, where the costs are
incurred. Following that are brief discussions regarding the components' unit costs or cost
assumptions. Greater detail regarding these inputs may be found in Appendix C, Inputs to Cost
Modeling. Only summary unit costs are presented in this document; more detailed descriptions of the
assumptions and methodologies used to develop these cost estimates are presented in the Cost and
Technology Document for the Ground Water Rule (EPA, 1999a).
Exhibit 6-2. Assignment of Components' Costs
Rule Scenario Components
Costs to State
Costs to System
Monitoring and Assessment
Sanitary Survey
Triggered Monitoring
Sensitivity Assessment
Routine Monitoring
(A)
(A)
(N)
Corrective Action
For Significant Defects
For Fecal Contamination
Compliance Monitoring
(A)
(A)
(A)
(A) Administrative Costs only
(N) No costs, States are expected to make assessments without system involvement
Although EPA estimated the cost of all the rule's components for drinking water systems and
States, there are some costs that the Agency did not monetize. These nonmonetized costs result from
uncertainties surrounding rule assumptions and from modeling assumptions. For example, EPA did not
estimate a cost for systems to acquire land if they needed to build a treatment facility or drill a new well.
This was not estimated because many systems will be able to construct new wells or treatment facilities
on land already owned by the utility. In addition, if the cost of land was prohibitive, a system may
chose another lower cost alternative such as connecting to another source. In addition, the Agency did
not develop cost estimates for all conceivable corrective actions or significant deficiencies that a system
may encounter. Instead, a representative sample was chosen, as discussed below under corrective
action.
6.2.1.1 State Agency Costs
As indicated above in Exhibit 6-2, States incur costs for all components; for all but two of the
components the costs are strictly administrative costs. In addition to the administrative costs the
following discussion addresses costs incurred by required provisions for sanitary surveys and
hydrogeologic sensitivity assessments.
6-2
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
State CostsAdministrative Costs
States will incur administrative costs of implementing the GWR. These administrative costs are
not directly required by specific provisions of GWR options, but are necessary for States to ensure the
provisions of the GWR are properly carried out. States will need to allocate time for their staff to
establish and then maintain the programs necessary to comply with the GWR. Staff time resources are
estimated in terms of full-time equivalents (FTEs). EPA assumed a cost of $64,480 for one FTE,
including overhead and fringe benefits. Time requirements for a variety of State agency activities and
responses are estimated for this RIA. Exhibit 6-3 lists activities required to start the program following
promulgation of the GWR as well as the annual activities that a State will require to continue
implementation of the GWR.
Exhibit 6-3. Examples of State Administrative Activities
Start Up Activities
Public Notification
Regulation Adoption and Program Development
Upgrade Data Management Systems
Initial Lab Certification and Training
System Training and Technical Assistance
Staff Training
Annual Activities
Coordination with EPA
Lab Certification
On-Going Technical Assistance
SDWIS Reporting
Clerical
Supervision
In addition to the administrative costs of developing and maintaining a program for GWR
compliance, States will be required to spend time responding to ground water sources that are found to
be fecally contaminated. These costs include time to review plans and specifications, prepare violation
letters, and conduct data entry.
State CostsSanitary Survey
The GWR options increase both the scope and the frequency of sanitary surveys. EPA
estimated that on average, States currently conduct sanitary surveys of community ground water
systems once every five years and noncommunity ground water systems once every 10 years. EPA
assumed that, under the GWR, sanitary surveys will be conducted by the State (or primacy agent) on
all noncommunity ground water systems once every five years. For community ground water systems,
EPA assumed that all of the community ground water systems that achieve a 4-log inactivation or
removal of virus will have sanitary surveys conducted every five years, and the remaining community
ground water systems will have sanitary surveys conducted once every three years.
The scope of sanitary surveys are expanded to address eight specific components of a PWS.
EPA estimated the incremental increase in cost for performing and preparing a sanitary survey. These
incremental costs range from as low $30 per survey for systems serving under 100 individuals to $700
per survey per system for the largest systems.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 6-3
-------
State CostsHydrogeologic Sensitivity Assessment
The hydrogeologic sensitivity assessment, a component only of the multiple-barrier option, will
be performed by States (or other primacy agent) on each ground water source to determine if the
source is sensitive to microbial contamination and requires monitoring to ensure there is no fecal
contamination. EPA assumes hydrogeologic sensitivity assessments would be performed on all ground
water sources that are not providing 4-log inactivation or removal of virus. EPA estimates that 15
percent of the systems that are assessed will be determined to be sensitive, based upon data collected
for the AWWA Study (Abbaszadegan et al., 1998, 1999).
EPA estimated the time for States to locate existing hydrogeologic data, such as a well
construction record, and for a State assessor to inspect and review these data. These average costs
increase in relative proportion to the system size, ranging from as low as $62 per assessment for
systems serving under 100 individuals to $3,224 per survey per system for the largest systems.
6.2.1.2 Public Water System (PWS) Costs
PWS CostsSanitary Survey
As discussed under State costs above, sanitary surveys will increase both in scope and in
frequency. The only incremental increase in costs that a system will incur as a result of this component
is the additional time it will take to accompany the State inspector conducting the survey.
Noncommunity water systems' sanitary survey frequency will be once every five years versus once
every 10 years under current requirements. Community water systems which do not achieve a 4-log
inactivation of virus will assist in sanitary surveys once every three years instead of once every five
years. There is no change in frequency for community ground water systems which achieve 4-log
inactivation of viruses. In addition to the increased frequency, the scope of the sanitary survey is
increased to address eight specific components of a PWS. EPA estimated the cost increase to PWSs
to meet the requirements of the GWR for sanitary survey as ranging from as low $110 per survey for
systems serving under 100 individuals to $1,900 per survey per system for the largest systems.
PWS CostsTriggered Source Water Monitoring
Triggered source water monitoring is a component of two regulatory options: Options 2 and 3.
Only systems that do not achieve 4-log inactivation of virus will be subject to this provision. Triggered
monitoring requires collection and analysis of samples at the ground water systems source, following the
detection of total coliform in one or more samples collected for compliance with the Total Coliform
Rule. While States have the option of requiring the triggered source water samples to be tested for the
presence of one of three fecal indicators, for the purpose of this cost analysis, EPA assumed that States
will select E. coli as the indicators of contamination for analysis. EPA estimated the cost for triggered
monitoring to be $53 per sample, including laboratory analysis ($25) and one hour of the system
operator's time (at an estimated cost of $28 per hour) to collect the sample, arrange for delivery to the
laboratory and to review the results of the analysis. No additional costs are assumed for installation of
a tap or re-piping of wells to permit sampling, as EPA assumed all wells are equipped with existing taps
for sampling.
64 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
If a system detects the fecal indicator at its source, then the system must take a corrective
action to either eliminate the contamination, obtain a new uncontaminated source, or install treatment to
achieve a 4-log removal or inactivation of virus. Several compliance estimates are necessary to
develop estimates of the cost associated with triggered monitoring: the frequency with which systems
will have to perform triggered monitoring; the frequency that a waiver will be granted; the duration of
the triggered monitoring; and the number of systems that are expected to test positive for the fecal
indicator. These factors are discussed in greater detail below.
Frequency of Performing Triggered Monitoring: EPA estimated the probability of a ground water
system's total coliform sample testing positive, which would therefore, trigger the monitoring
requirements, as a part of its regulatory impact analysis for the Total Coliform Rule (EPA, 1989). EPA
calculated the frequency of total coliform positives per year per system by multiplying the number of TC
samples required per year by the probability of a TC positive. All community water systems serving
under 3,000 individuals and all noncommunity water sources are estimated to have zero to three
triggered source water samples per year. Larger community water systems vary, with the estimate for
largest system at 7 to 22 triggered sources water samples per year.
Waiver from Triggered Monitoring: The option allows States to waive the triggered monitoring
requirements if a PWS demonstrates that the total coliform contamination is not source water related.
EPA assumed that the probability of a PWS receiving this waiver for a single entry point is 10 percent.
Also, a one-time repeat sampling waiver exists for both triggered monitoring and routine monitoring.
Once a PWS finds a single positive sample, they may take five repeat samples, and if all five repeat
samples are negative, the source water is considered not to be contaminated. For the purposes of this
analysis, it is assumed that all PWSs that have a positive source water sample will make use of this one-
time sampling waiver.
Duration of Triggered Monitoring: For the purposes of the cost model, EPA assumed that all
contaminated entry points will be discovered in the first year. Therefore, the entry points with no
source water positive samples, or those with a single positive sample and five negative follow-up
samples in the first year, will continue to undertake triggered monitoring sampling for the remainder of
the period of analysis. The number of samples they will take each year is assigned using a uniform
distribution based on the number of expected total coliform violations they might have per year.
Number of Systems Testing Positive: EPA estimated the number of systems that will test positive in
triggered monitoring by reviewing indicator occurrence data. The AWWARF study (Abbazadegan et
al, 1998, 1999) found enteroccocci bacteria in 9 percent of the wells sampled. Wells were selected for
the study to be representative of hydrogeologic conditions for public water supply wells in the United
States, and most wells in the study were only sampled once. EPA determined that the enterococci
occurrence from the AWWARF study provides the best estimate of the percentage of wells that will be
found to test positive for the presence of a fecal indicator in triggered source water sampling and,
therefore, assumed that 9 percent of the systems tested will be found to contain fecal contamination.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 6-5
-------
PWS CostsRoutine Source Water Monitoring
Routine source water monitoring, a component only of Option 3, the Multi-Barrier Option,
involves monthly sampling of those ground water sources that are determined to be at high risk for the
presence of fecal contamination. The hydrogeologic sensitivity assessment performed by the State is
used to determine which wells are at high risk of contamination. Only systems that do not achieve 4-log
inactivation of virus will be subject to this provision. Similar to triggered monitoring, several compliance
assumptions were needed to model the cost of routine monitoring and they are described below.
Frequency of Performing Routine Monitoring: High risk wells would be sampled routinely each month
and tested for the presence of fecal contamination using one of three possible indicators as selected by
the State. EPA assumed that States will select E. coli as the indicator of contamination for analysis.
States may reduce the frequency of monitoring for high risk ground water sources after one year of
monitoring if there are no samples that test positive. States may also waive source water monitoring
altogether after the first year if the State determines that fecal contamination of the well is highly unlikely.
Duration of Routine Monitoring: EPA assumed that all contaminated entry points will be discovered in
the first year of routine monitoring. EPA also assumed that the entry points either with no source water
positive samples or with a single positive sample and five negative follow-up samples in the first year
will continue to undertake routine sampling once a quarter for the remainder of the period of analysis.
Waiver from Corrective Action: A waiver could be granted by the State if the system collects five
repeat samples from the well that tested positive within 24 hours and does not find the fecal indicator in
any of five samples. This waiver can only be granted once per source. Because of the high costs
associated with corrective actions, EPA assumed that all systems with a routine source water positive
sample will resample their source within 24 hours of detecting the fecal contamination. EPA estimated
that 20 percent of the systems that perform the repeat sampling will not find fecal indicators in any of
the repeat samples and will receive waivers from the State.
Number of Systems Testing Positive: Using EPA/AWWARF study data (Abbazadegan et. al., 1998,
1999), EPA estimated that 15 percent of the sources for ground water systems will be determined to
be sensitive and therefore subject to routine monitoring (See Appendix C for more details). Of these
wells, EPA estimates that 50 percent will test positive for the presence of a fecal indicator based upon
anE. coli occurrence in wells vulnerable to contamination (Lieberman et. al., 1994, 1999). Ground
water systems with wells that test positive for the presence of a fecal indicator would be required to
take action to correct the contamination unless the system were able to obtain a one-time waiver from
the State. As for triggered monitoring analysis (see above), EPA estimated the cost for triggered
monitoring to be $53 per sample and assumed that all wells are equipped with existing taps for
sampling.
PWS CostsCorrective Action For Significant Defects
The primary purpose of conducting sanitary surveys is to identify significant defects in public
water systems for correction. The costs for correction of significant deficiencies are dependent almost
entirely upon the nature of the deficiency. Because States have the authority to define significant
deficiencies under the proposed GWR and options, EPA must predict the types of deficiencies that will
6-6 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
be found and corrected as a result of the rule. EPA consulted with experts from within the Agency and
from States to develop a list of corrective actions to address deficiencies that are likely to be identified
in sanitary surveys of ground water systems (EPA, 1996a). These are as follows:
Correction of Significant Deficiencies at the Source
Replace a sanitary well seal,
Rehabilitate an existing well, and
Drill a new well
Correction of Significant Deficiencies in Treatment Systems
Adjust disinfection chemical feed rate
Increase contact time prior to first customer
Correction of Significant Deficiencies in Distributions System
Install backflow prevention device
Replace/repair cover on a storage tank
Install security measures at storage tank site
Install a redundant booster pump
Costs were developed for each of these (See Appendix C for unit costs and details); these
costs are one-time expenditures that occur in the year the significant deficiency is found, except for
adjustments to the disinfection feed rate, which are ongoing costs.
Each of the regulatory options requires each PWS to correct any significant defect found during
a sanitary survey. The assignment of any significant deficiency is done as a two-step process within the
cost analysis model, and is done independently for each sanitary survey over the 20-year period of
analysis. First, a PWS is designated as having or not having the potential to have one or more
significant defects resulting from a single sanitary survey based on a probability estimate.
Second, each PWS that is predicted to have a significant defect, in a single sanitary survey, may
be assigned one or more of the six potential significant defects according to a probability distribution
(See Appendix C for the probability distribution). Because the corrections of significant defects are
dependent upon the defects defined as significant by States and the conditions at the facilities, both of
which are unknown, EPA used a high scenario/low scenario estimating procedure to bound the
uncertainty. The low cost scenario assumes a greater percentage of the systems with significant defects
will have defects which are less expensive to correct (e.g., more systems will have to replace their
sanitary well seal than will have to perform a complete rehabilitation of their well). This high/low
bounding should provide a reasonable estimate of the uncertainty with respect to the types of defects
that will have to be corrected.
This two-step process is repeated for each sanitary survey the PWS undertakes over the 20-
year period of analysis. Since the timing of the sanitary surveys are not known, an average annual PWS
cost of correcting significant defects is calculated by summing the cost of correcting all significant
defects over the 20-year period of analysis and then dividing by 20. The average annual PWS cost of
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 6-7
-------
correcting significant defects includes the cost of developing engineering plans for submission to the
State (See Appendix C for details).
PWS CostsCorrective Action For Fecal Contamination
Detection of fecal indicators in the source of undisinfected ground water systems requires
corrective action under Options 2 and 3. Corrective action includes eliminating the contamination from
the source, obtaining an alternative source of water, or providing disinfection treatment that achieves 4-
log inactivation or removal of viruses. Because costs are based on the corrective action which may
vary (i.e., the corrective action is selected by the system after consultation with the State and based
upon the size of the system), EPA assumed that a variety of corrective actions could be implemented by
ground water systems that detect fecal contamination within their source waters. The corrective actions
include eliminating the contamination from the source water (address contamination source or replace
source) or treating the water to achieve a 4-log inactivation/removal of virus.
Eliminate Contamination: EPA developed unit cost estimates for four corrective actions to eliminate
contamination from the system's source of water (detailed unit costs are presented in Appendix C).
Depending on the corrective actions, there may up to three different cost estimates: capital cost (the
cost of constructing/installing the equipment), replacement cost (cost of replacing significant components
of the system after several years operation, and operation and maintenance costs (or O&M) (annual
cost of operating equipment and performing routine maintenance). The four options EPA considered
for eliminating contamination from the source are:
rehabilitate an existing well;
remove/relocate existing source of contamination (septic tank);
construct a new well; and
purchase water from a nearby system.
Disinfect Source Water to Achieve a 4-log Inactivation/Removal of Virus: Additionally, EPA
developed costs for installing and operating eight types of disinfection systems to achieve required
standards (detailed unit costs are presented in Appendix C). Included in this analysis are costs for
capital, annual operation and maintenance, and year 10 replacement cost. Year 10 replacement costs
are estimated for the systems that will require replacement of a significant component halfway through
the design life of the system. The Agency developed costs for these eight disinfection technologies:
chlorine gas disinfection,
hypochlorination,
chloramination,
chlorine dioxide disinfection,
mixed oxidants disinfection,
ozonation,
reverse osmosis filtration, and
ultraviolet disinfection.
EPA developed estimates of corrective actions that ground water systems with fecal
contamination would undertake to eliminate or treat their contamination. These estimates considered
the current implementation of treatment types, the cost of the corrective action, and the need for
systems to comply with provisions of the Disinfection Byproducts Rule (DBPR).
6-8
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Current implementation of treatment types: EPA assumed that the portion of systems that will choose
treatment versus nontreatment corrective actions is proportional to the percentage of systems in each
service population category that currently perform disinfection treatment. Because of the uncertainty
inherent in projecting the number of systems that would undertake each corrective action, EPA
assumed varying percentages of the nontreatment corrective actions to provide upper and lower cost
bounds.
Distribution of Corrective Action: Each entry point that is predicted to require a corrective action is
assigned one of 13 potential corrective actions according to a probability distribution (See Appendix C
for the probability distribution). In order to determine the sensitivity of the cost estimates to these
probabilities, two scenarios were considered. Under the first scenario, entry points were assigned to
low cost corrective actions with greater probability (known as the Low CA scenario), while in the
second scenario, entry points were assigned to high cost corrective actions with greater probability
(known as the High CA scenario). After the model assigns each entry point a specific corrective
action, the costs, both capital and operations & maintenance (O&M) costs, are determined using the
aforementioned unit costs (details of these costs are in Appendix C).
6.2.2 General Structure of the Cost Model
In order to calculate the national and system-level costs of compliance, the agency used the
following information: the technology unit cost information and compliance forecast, both described
above; information on the inventory of drinking water systems; national occurrence information; and
various baseline characteristics of PWSs, such as technology in-place.
The national cost of compliance was estimated using a Monte-Carlo simulation model
specifically developed for the GWR. The GWR cost model was developed in Microsoft Excel© using
the Crystal Ball© Monte-Carlo simulation add-in. The main advantage to this modeling approach is
that, in addition to providing average compliance costs, it also estimates the range of costs within each
PWS size and type category. Hence, the GWR cost model allows for variability in PWS configuration,
current treatment in-place, and source water quality to be captured in the compliance cost estimates.
This information is ideal for examining PWS level impacts and technology affordability.
6.2.2.1 PWS Configuration and Occurrence
Each PWS is defined in the GWR cost model by the population it serves and the number of
entry points to the distribution system it has, as entry points are used as a proxy for potential or actual
points of treatment. The simulation was conducted using a sample of 3,000 PWS populations for each
of the 62 PWS size/type/ownership categories (up to nine size categories; three PWS typesCWS,
NTNC and TNC; matrixed against three ownership typespublic, private, and ancillary) taken from
the Safe Drinking Water Information System (SDWIS). The Agency developed distribution of the
number of entry points for each size category using information from the Community Water System
Survey (EPA, pending) (see Exhibit 4-4). A limitation of these data is that they were developed from
data collected from community water systems only.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 6-9
-------
Each Community Water System's design flow (DF) and average daily flow (ADF) were
calculated as a function of population served by the PWS using the following flow equations
(Geometries and Characteristics of Community Water Systems. EPA, 1999):
Publicly Owned CWS:
DF(kgpd)=0.5499xPop09554
ADF(kgpd) = 0.0858 xPopl'05S4
Privately Owned CWS:
DF(kgpd)=OA\68xPop°-960*
ADF(kgpd) = 0.0667 xPop1'062*
Average flow for NCWS was estimated based upon average flow rates of 27.1 gal/day per
person for TNC systems and 30.5 gal/day per person for NTNC systems. Utilizing data from SDWIS,
EPA prepared estimates of the number of noncommunity water systems by their service area types
(e.g., the number of noncommunity water systems serving restaurants), and an estimate of the average
flow per person (See Appendix C). Design flow rates were calculated based upon the ratio of design
to average flow rates from the above equations for community water systems.
Both system design flow and average daily flow are assumed to be divided equally among all of
a PWS's entry points, therefore, the design flow and average daily flow per entry point are easily
calculated. Each entry point is then designated as currently providing inactivation treatment or not
currently providing inactivation treatment according to the percentage of systems achieving 4-1 og
inactivation. This is done independently for each entry point within a given PWS. For example, a given
PWS can have one entry point that currently treats, while having two entry points that currently have no
treatment in place.
6.2.2.2 Discounting and the Cost of Capital
For the purposes of this analysis, PWS and State implementation costs are tracked over a 20-
year period. This time frame was chosen because most of the capital equipment included in the analysis
has a 20-year useful life, and PWSs often finance their capital improvements over a 20-year period.
However, the capital and O&M costs of each rule option are incurred at different points in time over
the course of the period of analysis.
Two different adjustments are made in this analysis in order to render future costs comparable
with current costs, reflecting the fact that a cost outlay today is a greater burden than an equivalent cost
outlay sometime in the future. The first adjustment is made when the cost estimates that are derived are
being used as an input in benefit-cost analysis. In this instance, costs are annualized using a discount
rate so that the costs of each regulatory option can be directly compared with the annual benefits of the
corresponding regulatory option. Annualization is the same process as calculating a mortgage payment;
the result is that we have a constant annual cost to compare with our constant annual benefits.
610 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
The choice of an appropriate discount rate is a very complex and controversial issue among
economists and policy makers alike. Therefore, the Agency compares costs and benefits using two
alternative discount rates, in part to determine the effect the choice of discount rate has on the analysis.
The annualized costs of each regulatory option are calculated and displayed using both a seven percent
discount rate required by the Office of Management and Budget (OMB) and a three percent discount
rate which the Agency believes more closely approximates the social discount rate.
The second adjustment is made when the cost estimates that are derived are used as an input
into an economic impact analysis, such as an affordability analysis, an analysis of PWS-level costs, or
household-level costs. In these cases, rather than use a discount rate when determining the annualized
costs, an after tax cost-of-capital rate is used. This rate should reflect the true after-tax cost of capital
PWSs face, net of any government grants or subsidies.
6.2.2.3 Calculating Household Costs
Household level costs are considered a good proxy for the affordability of rule compliance on
CWSs, since affordability at the household level is necessary for CWS cost recovery through increased
water rates. This of course assumes that nonresidential customers of CWSs, such as businesses, can
pass along any increase in water costs to their customers through increased prices on their goods or
services. Household costs are calculated for each CWS that is either publically or privately owned.
In order to calculate the average household-level cost of compliance for a single CWS, the
CWS's annual compliance cost is first divided by the CWS's average daily flow (1,000 gallons per
day), and then multiplied by 365 days, to determine the CWS's cost of compliance per 1,000 gallons
produced. Finally, the CWS's cost of compliance per 1,000 gallons (kg) is multiplied by the average
annual consumption per residential connection (kg) to arrive at the average annual cost of compliance
per household for the CWS. The estimates of average annual consumption per residential connection
used in this analysis are provided in Appendix C.
6.3 National Costs
This section details the results of the national compliance cost modeling. For each regulatory
option considered, the following information is provided:
The number of PWS undertaking each rule component;
The national annual PWS compliance costs;
The national annual State implementation costs.
Appendix D provides further detail on the distribution of national compliance costs across
system sizes and types.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 6-11
-------
6.3.1 Comparison of Annual Compliance Costs Across Regulatory Options
Exhibit 6-4 provides a comparison of the total annual cost of compliance across the four
regulatory options. The costs steadily increase as one moves from the sanitary survey option, to the
sanitary survey and triggered monitoring option, to the multi-barrier option. However, the costs
increase by almost a factor of five from Option 3, the Multi-Barrier Approach, to Option 4, Across-
the-board disinfection. This increase in costs results from the fact that the first three regulatory options
are targeted, to differing degrees, at PWSs that have a demonstrated potential, either through sanitary
surveys or through source water monitoring, to provide their customers contaminated drinking water.
Option 4, Across-the-board Disinfection option, as the name implies requires all PWSs to treat their
source water, even if there is no demonstrated potential or actual contamination. This means that costs
are being incurred by many more PWSs. Exhibit 6-5 and Exhibit 6-6 show the comparison of total
annual costs across the four regulatory options, by system size category and system type respectively.
Exhibit 6-4. Comparison of National Annual Compliance Costs
Across Regulatory Options (millions of dollars, 1998)
Option/Regulatory Scenario
Option 1 : Sanitary Survey Only
Option 2: Sanitary Survey and
Triggered Monitoring
Option 3: Multi Barrier Approach
Option 4: Across-the-Board
Disinfection Option
Mean Compliance Costs ($Millions)
At 3%
$72.7
($71.1 to $74.4)
$157.6
($152. 8 to $162.4)
$182.7
($177.0 to $188. 4)
$777.1
($743. 9 to $810. 3)
At 7%
$76
($74.3 to $77.7)
$168.5
($163.0 to $174.0)
$198.6
($191. 7 to $205. 5)
$866.0
($822.7 to $909. 4)
Exhibit 6-5. Comparison of Mean Annual Compliance Costs Across Regulatory
Options by System Size Category (millions of dollars, 1998)
Option/Regulatory Scenario
Option 1: Sanitary Survey
Only
Option 2: Sanitary Survey
and Triggered Monitoring
Option 3: Multi-Barrier
Approach
Option 4: Across-the-Board
Disinfection
Mean Cost by System Size
<100
$23.5
$73.5
$79.5
$198.0
100-
500
$14.6
$34.5
$42.7
$197.0
500-
1,000
$5.9
$10.2
$12.7
$76.6
1K- 3.3K-
3.3K 10K
$7.7
$11.9
$14.9
$106.1
$4.7
$7.5
$11.8
$89.9
10K- 50K- 100K-
50K 100K 1M
$3.8
$6.1
$6.1
$44.4
$1.0
$3.0
$3.7
$44.0
$1.8
$1.1
$1.6
$11.3
Mean Total
Costs
($Millions)
$72.7
$157.6
$182.7
$777.1
* 3% discount rate
6-12
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit 6-6. Comparison of Mean Annual Compliance Costs Across Regulatory
Options by System Type (millions of dollars, 1998)
Option/Regulatory Scenario
Option 1 : Sanitary Survey Only
Option 2: Sanitary Survey and Triggered
Monitoring
OptionS: Multi-Barrier Approach
Option 4: Across-the-Board Disinfection
Percent of Total Cost by System
Type*
CWS
$35.9
$56.5
$66.8
$391.5
NTNC
$5.1
$14.6
$17.3
$85.8
TNC
$22.1
$76.7
$88.8
$290.0
Mean Total
Costs
($Millions)
$72.7
$157.6
$182.7
$777.1
* 3% discount rate
6.3.2 Option 1: Sanitary Survey Only
6.3.2.1 Total National Costs
Under all of the regulatory alternatives under consideration, all PWSs must perform the
minimum requirement of conducting sanitary surveys and correcting significant defects discovered
through the sanitary survey. Exhibit 6-7 below shows that all of the PWSs are expected to conduct
sanitary surveys under this option, and approximately 60 percent of the PWSs will be required to
correct significant defects over the 20-year period of the cost model simulation. As shown in Exhibit
6-8, this regulatory option produced a range of total national compliance costs among PWSs from $71
to $74 million at a three percent discount rate, and from $74 to $77 million at a seven percent discount
rate.
Exhibit 6-8 also shows that unlike system costs, the costs remain fairly constant between
compliance scenarios because of annual fixed State costs of approximately $10 million, for annual costs
ranging from $17.5 to $18 million, depending on the discount rate. Note, that these values do not
include State costs incurred for lab certification as no lab certification is required under this regulatory
option.
6.3.2.2 Cost of Rule Components
Given the range of total annual national costs of PWSs described above, the High Corrective
Action/Low Significant Defect scenario presents an approximate mid-range value of the expected
compliance costs. Exhibit 6-9 demonstrates that 89 percent of the overall system costs are due to
correction of significant defects, while the remaining 11 percent of PWS costs are from other
compliance activities, such as monitoring, record-keeping or conducting surveys.
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
6-13
-------
Exhibit 6-7. Number of Affected Systems by Rule Component
(High Corrective Action/Low Significant Defect Scenario)
Option 1: Sanitary Survey Only
180,0001
160,000"
Number of
Systems
Sanitary
Survey
Significant
Defects
Triggered
Monitoring
Compliance
Monitoring
Routine
Monitoring
Corrective
Actions
Exhibit 6-8. Total National Costs: Option 1--Sanitary Survey Only
Cost Type
System Costs
State Costs:
Total Costs
Mean Compliance Costs ($Millions)
At 3 %
$55.2
($53. 7 to $56. 8)
$17.5
($17.5 to $17.6)
$72.7
($71.1 to $74.4)
At 7%
$57.9
($56. 2 to $59. 5)
$18.1
($18.1 to $18.2)
$76.0
($74.3 to $77.7)
6-14
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
ffeftvuJ PWS Comjtiance Costs oTtoe GWRtoy Bute Component
Optrm 1: Sanitary Survey Only
Otfw System
6.3.3 Option 2: Sanitary Survey and Triggered Monitoring
6.3.3.1 Total National Costs
The Sanitary Survey and Triggered Monitoring Option builds significantly upon the Sanitary
Survey Option. The Exhibit 6-10 addresses components included in this regulatory alternative
including source water monitoring, corrective action, (including system modifications to currently
treating entry points that do not achieve 4-log removal) and compliance monitoring. Over 120,000
PWSs are expected to undergo triggered monitoring under Option 2, Sanitary Survey and Triggered
Monitoring. As shown in Exhibit 6-11, these additional requirements increase the total national cost of
compliance for PWS to $153 million to $162 million at a three percent discount rate, and $163 million
to $174 million at a discount rate of seven percent.
Exhibit 6-11 also shows that the overall annual State costs are also increased under this
regulatory option, ranging from $19 to $20 million depending on the discount rate, despite the same
annual fixed State costs previously described. This is, in part, a result of the lab certification costs
previously omitted, as well as the cost to the States associated with tracking PWS monitoring and
reviewing corrective action and system modification engineering plans and permits.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 6-15
-------
Exhibit 6-10.
Number of Affected PWSs by Rule Component
(High Corrective Action/Low Significant Defect Scenario)
Option 2: Sanitary Survey and Triggered Monitoring
160,000 '
20,000
Number of Sanitary Significant Triggered Compliance Routine Corrective
Systems Survey Defects Monitoring Monitoring Monitoring Actions
Exhibit 6-11. Total National Compliance Costs
Option 2: Sanitary Survey and Triggered Monitoring (million $)
Cost Type
System Costs
State Costs:
Total Costs
Mean Compliance Costs ($Millions)
At 3%
$138.7
($133.9 to $143. 5)
$18.9
($18.8to$18.9)
$157.6
($152.8 to $162. 4)
At 7%
$148.7
($143.2 to $154. 3)
$19.8
($19.7to$19.8)
$168.5
($163.0 to $174.0)
6.3.3.2 Cost of Rule Components
Given the High Corrective Action/Low Significant Defect scenario example, the national PWS
compliance costs are composed mostly of monitoring, recordkeeping and survey costs (55 percent)
(Exhibit 6-12) (note that system modifications indicated in the exhibit refer to systems which must
modify their existing disinfection practices to achieve 4-log inactivation of virus as a corrective action).
6-16
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
The remainder is distributed between corrective actions (23 percent) and correction of significant
defects (20 percent). This is in sharp contrast to the Sanitary Survey Option in which 89 percent of the
PWS's cost was the result of significant defect correction.
Naftvul PWS Compliance Cists ftttheGW&byRule Conpmatt
(High Corrective Action/Low Significant Defect Scenario)
Option 2: Sanitary Survey and TriaerefMenitamq
Mwfffcafcn* A Atffe*
6.3.4 Option 3: Multi-Barrier Approach
6.3.4.1 Total National Costs
The regulatory components of the Sanitary Survey and Triggered Monitoring Option and the
Multi-Barrier Option are very similar, with one notable exception. Under the Multi-Barrier Option, a
hydrogeological sensitivity assessment is included in the compliance scenario, a regulatory component
which provides more targeted source water monitoring. Rather than requiring all entry points to
undergo triggered monitoring, those deemed sensitive by the hydrogeological assessment must practice
a more stringent routine monitoring regime. From Exhibit 6-13 below, most of the PWSs affected
under the Multi-Barrier Option are expected to conduct triggered monitoring, as under the Sanitary
Survey and Triggered Monitoring Option, but approximately 20,000 PWSs are expected to conduct
routine monitoring. This results in total annual national PWS compliance costs ranging from $177 to
$188 million at a three percent discount rate, and from $192 to $206 million at a seven percent
discount rate (Exhibit 6-14 below).
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 6-17
-------
Exhibit 6-13. Number of Affected Systems by Rule Component
(High Corrective Action/Low Significant Defect Scenario)
Option 3: Multi-Barrier Approach
Number of
Systems
Sanitary
Survey
Significant
Defects
Triggered
Monitoring
Compliance
Monitoring
Routine
Monitoring
Corrective
Actions
Exhibit 6-14. Total National Compliance Costs
Option 3: Multi-Barrier Approach (million $)
Cost Type
System Costs
State Costs:
Total Costs
Mean Compliance Costs ($Millions)
At 3 % At 7%
$162.2
($156.4 to $167. 9)
$20.6
($20. 6 to $20. 6)
$182.7
($177.0 to $188. 4)
$176.5
($169.6 to $183.4)
$22.1
($22.1 to $22.1)
$198.6
($191. 7 to $205.5)
Also shown in Exhibit 6-14, unlike PWS costs, few cost components change between
scenarios at the State level under the Multi-Barrier Option. This is largely due to the $10 million annual
fixed State costs that remain consistent between compliance scenarios. The annual State costs are
approximately $21 million at a three percent discount rate, and $22 million at a seven percent discount
rate.
6.3.4.2 Cost of Rule Components
Using the High Corrective Action/Low Significant Defect scenario as a mid-range example of
the overall distribution of PWS costs described above, Exhibit 6-15 shows that greatest portion of the
6-18
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
costs result from monitoring, record keeping, and sanitary surveys (note that system modifications
indicated in the exhibit refer to systems which must modify their existing disinfection practices to achieve
4-log inactivation of virus as a corrective action). The second largest component of the cost burden is
attributed to correction actions taken by water systems (31 percent). Correction of significant defects
accounted for only 17 percent of the overall PWS costs of this compliance scenario.
Etftffit fi- 15. Nrftttjf PWS Conjtiartce Costs dTffte GWRby Rule Canpon&tt
(High Corrective Action/Low Significant Defect Scenario)
Option 3:
£3htr System C*s& (J/oniftawf.,
3-fi
6.3.5 Option 4: Across-the-Board Disinfection
6.3.5.1 Total National Costs
Under the Across-the-Board Disinfection Option, all entry points not currently achieving a 4-
log removal standard will be required to either 1) undertake a corrective action (which may not
necessarily include fixing an existing well or drilling a new well), or 2) modify their treatment technique
to achieve 4-log. Since all PWSs are expected to achieve the 4-log standard, there is no source water
monitoring component of this alternative. Exhibit 6-16 below shows the distribution of PWSs expected
to undertake the various rule components.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 6-19
-------
Exhibit 6-16.
Number of Affected Systems by Rule Component
(High Corrective Action/Low Significant Defect Scenario)
Option 4: Across-the-Board Disinfection
20,000 '
Number of Sanitary Significant Triggered Compliance Routine Corrective
Systems Survey Defects Monitoring Monitoring Monitoring Actions
As Exhibit 6-17 below demonstrates, this option results in the highest compliance costs at both
the PWS and the State levels. Total annual PWS costs across the nation range from $744 million to
$810 million at a discount rate of three percent, and from $823 million to $909 million at a seven
percent discount rate. Annual State costs, including the same annual fixed costs as previously
discussed, amount to over $25 million (three percent discount rate) or over $28 million (seven percent
discount rate).
Exhibit 6-17. Total National Compliance Costs
Option 4: Across-the-Board Disinfection (million $)
Cost Type
System Costs
State Costs:
Total Costs
Mean Compliance Costs ($Millions)
At 3 %
$751.8
(718. 7 to $785.0)
$25.2
($25.2 to $25.2)
$777.1
(743. 9 to $810.3)
At 7%
$837.4
($794.1 to $880.7)
$28.6
($28.6 to $28.6)
$866
($822. 7 to $909.4)
6-20
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
6.3.5.2 Cost of Rule Components
The high PWS costs are largely due to corrective actions (66 percent) and other PWS costs,
such as conducting surveys (26 percent), as shown in Exhibit 6-18 below (note that system
modifications as indicated in the exhibit refer to systems which must modify their existing disinfection
practices to achieve 4-1 og inactivation of virus as a corrective action.). Five percent of PWS costs are
due to system modifications to achieve 4-log removal. Of the four regulatory options discussed, the
percentage of PWS costs from correction of significant defects are the lowest under the Across-the-
Board Disinfection Option (only 3 percent).
ExhOut & 7ft ttaftvuf PWS Cmtpffiance Castr atttttt GWffby Rule Component
(High Corrective Action/Low Significant Defect Scenario)
Option*
tithtr SfOtu CaOs (Mmitotif,
System «/(rffcj«ins»A5f*/t -Huf 5
6.4 Household Costs
Exhibits 6-19 and 6-20 show the range of household costs for each of the four regulatory
options by PWS size category. Exhibit 6-19 presents average household costs for all community water
systems, including those systems which do not have to take corrective action for significant defects or
fecal contamination. Exhibit 6-20 presents average household costs only among systems which must
take corrective action. Household costs tend to decrease as system size increases, due mainly to the
economies of scale for the corrective actions. Cumulative distribution of household costs for each
option are in Appendix C.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 6-21
-------
Exhibit 6-19.
Mean Annual Household Costs of the GWR Across Regulatory Options
All Public and Private CWSs
SIZE CATEGORIES
<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
TOTAL
Sanitary Survey
Option
$16.79
$5.59
$2.87
$1.50
$0.72
$0.30
$0.17
$0.18
$1.26
Sanitary Survey and
Triggered Monitoring Option
$51.88
$10.88
$4.04
$1.93
$0.97
$0.43
$0.49
$0.15
$2.40
Multi-Barrier Option
$46.30
$12.84
$4.15
$2.27
$1.74
$0.46
$0.73
$0.20
$2.67
Across-the-Board
Disinfection Option
$156.20
$54.45
$26.22
$16.52
$12.30
$3.34
$9.12
$1.44
$13.98
Exhibit 6-20.
Mean Annual Household Costs of the GWR Across Regulatory Options
for Public and Private CWSs Taking Corrective Action
or Fixing Significant Defects
SIZE CATEGORIES
<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
TOTAL
Sanitary Survey
Option
$29.86
$11.23
$5.72
$2.99
$1.39
$0.62
$0.30
$0.32
$2.45
Sanitary Survey and
Triggered Monitoring Option
$67.19
$15.02
$6.29
$2.91
$1.46
$0.59
$0.70
$0.20
$3.34
Multi-Barrier Option
$62.48
$18.95
$6.25
$3.39
$2.74
$0.62
$1.01
$0.27
$3.86
Across-the-Board
Disinfection Option
$191.87
$81.38
$38.79
$23.45
$16.78
$4.87
$10.37
$1.66
$19.37
6-22
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
6.5 References
Abbaszadegan, M., P.W. Stewart, M.W. LeChevallier, Rosen, Jeffery S. and C.P. Gerba.
1998/1999. Occurrence of viruses in ground water in the United States. American Water
Works Association Research Foundation. Denver, CO, 157 p.
Abbaszadegan, M., P.W. Stewart, and M.W. LeChevallier. 1999. "A Strategy for Detection of
Viruses in Groundwater by PCR." Applied and Environmental Microbiology, Vol.
65(2):444-449..
EPA Office of Ground Water and Drinking Water, 1996. Workshop on PredictingMicrobial
Contamination of Groundwater Systems, July 10 11, 1996, Proceedings Report. US
EPA, Washington DC, September.
EPA. 1996a. Ground Water Disinfection and Protective Practices in the United States. Office of
Ground Water and Drinking Water, Washington, D.C.
EPA. 1999a. Drinking Water Baseline Handbook, Draft 1st Edition. EPA, Washington D.C.
EPA. 1999b. Baseline Profile Document for the Ground Water Rule. EPA.
EPA. 1999c. Model Systems for Public Water Systems. Draft.
Lieberman, R.J., Shadix, L.C., Newport, C.P. Frebis, M.W.N. Moyer, S.E., Safferman, R.S., Stetler,
R.E., Lye, D., Fout, G.S., and Dahling, D. 1999. "Source water microbial quality of some
vulnerable public ground water supplies." Unpublished report in preparation.
Lieberman, R.J., L.C. Shadix, B.S. Newport, S.R Crout, S.E. Buescher, R.S. Safferman, R.E.
Stetler, D. Lye, G.S. Fout, and D. Dahling. 1994. "Source water microbial quality of some
vulnerable public ground water supplies." in Proceedings, Water Quality Technology
Conference, San Francisco, CA, October.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 6-23
-------
7. Economic Impact Analysis
7.1 Introduction
As part of the rule promulgation process, EPA is required to perform a series of
distributional analyses that address the potential regulatory burden placed on entities that are affected
by the various rule requirements. This chapter contains EPA's analysis and statements with regard to
five federal mandates: Executive Order 12886 (Regulatory Planning and Review); the Regulatory
Flexibility Act (RFA) of 1980, as amended by the Small Business Regulatory Enforcement Fairness
Act (SBREFA) of 1996; the Unfunded Mandates Reform Act (UMRA) of 1995; Executive Order
13045 (Protection of Children From Environmental Health Risks and Safety Risks); and Executive
Order 12989 (Federal Actions to Address Environmental Justice in Minority Populations and Low-
Income Populations). A summary of an additional analysis, conducted to fulfill requirements set forth
by the Paperwork Reduction Act, is addressed within this chapter, but the actual analysis is contained
in a separate document, Information Collection Request for the Ground Water Rule.
Several of these directives contain provisions requiring an explanation of why the rule is
necessary, the statutory authority upon which it is based, and the primary objectives it is intended to
achieve. A complete discussion of the background information and the statutory authority for this
rulemaking is located in Chapter 2, "Need for Proposal." Specifically, section 2.5.1 addresses the
statutory authority for promulgating this rule. The RFA and SBREFA analyses are contained in section
7.2 while the UMRA analysis is in Section 7.3. Issues such as the paperwork burden of this proposed
rule, children's health and safety, and environmental justice are addressed in sections 7.4, 7.5, and 7.6,
respectively.
7.2 Regulatory Flexibility Act and Small Business Regulatory
Enforcement Fairness Act
Under the Regulatory Flexibility Act (RFA), 5 U.S.C. 601 et seq.. as amended by the
Small Business Regulatory Enforcement Fairness Act (SBREFA), EPA is required to prepare a
regulatory flexibility analysis unless the Agency certifies that the rule will not have "a significant
economic impact on a substantial number of small entities." A regulatory flexibility analysis describes
the impact of the regulatory action on small entities as part of the rule promulgation process. The
Agency must also consult with small entity representatives (SERs) and convene a Small Business
Advocacy Review (SBAR) Panel prior to publication of the proposed rule if the Agency is unable to
certify that the rule will not have a significant economic impact on a substantial number of small entities.
The SBAR Panel has 60 days to consult with SERs likely to be impacted by the rule and to make
recommendations designed to reduce the impact of the proposed rule on small entities. The Agency
must consider these recommendations when drafting the proposed rule. Because the Agency was
unable to certify the GWR, the Agency convened a SBAR Panel.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 7-1
-------
The SBAR Panel members for the GWR were the Small Business Advocacy Chair of the
Environmental Protection Agency, the Director of the Standards and Risk Management Division in the
Office of Ground Water and Drinking Water (OGWDW) within EPA's Office of Water, the
Administrator for the Office of Information and Regulatory Affairs of the Office of Management and
Budget (OMB), and the Chief Counsel for Advocacy of the Small Business Administration (SB A).
The Panel convened on April 10, 1998 and met seven times before the end of the 60-day period on
June 8, 1998. The culmination of these meetings was the SBAR Panel's report, Final Report of the
SBREFA Small Business Advocacy Review Panel on EPA 's Planned Proposed Rule for National
Primary Drinking Water Regulations: Ground Water. The SER comments on components of the
GWR, and the background information provided to the SBAR Panel and the SERs are available for
review in the water docket. This information and the Agency's response to the Panel's
recommendations in developing the proposed GWR are summarized below.
7.2.1 Definition of Small Entity for the GWR
The Agency has taken comment on and finalized its intent to define "small entity" as a
public water system that serves 10,000 or fewer persons for purposes of its regulatory flexibility
assessments under the RFA for all future drinking water regulations. See Consumer Confidence
Reports (CCR) Final Rule, 63 FR 44511, Aug. 19, 1998 and Proposed Rule, 63 FR 7620 Feb. 13,
1998. The Agency discussed at length in the preamble to the proposed rule, the basis for its decision to
use this definition and to use the single definition of small public water system whether the system was a
"small business," "small nonprofit organization," or "small governmental jurisdiction." EPA also
consulted with the Small Business Administration on the use of this definition as it relates to small
businesses. Subsequently, the Agency has used this definition in developing its regulations under the
Safe Drinking Water Act. In defining small entities in this manner, EPA recognizes that baseline
conditions in source water and treatment and operational practices may differ for systems serving fewer
than 10,000 people when compared to systems serving 10,000 or more persons.
7.2.2 Requirements for the Initial Regulatory Flexibility Analysis
The Regulatory Flexibility Act requires EPA to complete an Initial Regulatory Flexibility
Analysis (IRFA) addressing the following:
The need for the rule;
The objectives of and legal basis for the proposed rule;
A description of, and where feasible, an estimate of the number of small entities to which
the rule will apply;
A description of the proposed reporting, record keeping, and other compliance
requirements of the rule, including an estimate of the types of small entities, which will be
7-2 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
subject to the requirements and the type of professional skills necessary for preparation of
reports or records;
An identification, to the extent practicable, of all relevant federal rules that may duplicate,
overlap, or conflict with the proposed rule; and
A description of "any significant regulatory alternatives" to the proposed rule that
accomplish the stated objectives of the applicable statutes, and that minimize any significant
economic impact of the proposed rule on small entities; the analysis is to discuss significant
regulatory alternatives such as:
- Establishing different compliance or reporting requirements or timetables that take into
account the resources of small entities;
- Clarifying, consolidating, or simplifying compliance and reporting requirements under
the rule for small entities;
- Using performance rather than design standards; and
- Exempting small entities from coverage of the rule or any part of the rule
To assist the SERs and the SBAR Panel in their deliberations, OGWDW prepared an
IRFA to provide some background on the need for the proposed rule and the possible components of
a rule. Prior to convening the SBAR Panel, OGWDW consulted with a group of 22 SERs likely to be
impacted by a GWR. The SERs included small system operators, local government officials, small
business owners (e.g., a bed and breakfast with its own water supply), and small nonprofit organization
(e.g., a church with its own water supply for the congregation). The SERs were provided with
background information on the rule, on the need for the rule and the potential requirements. The SERs
were asked to provide input on the potential impacts of the rule from their perspective. All 22 SERs
commented on the information provided in the IRFA. These comments were provided to the SBAR
Panel when the Panel convened. After a teleconference between the SERs and the Panel, the SERs
were invited to provide additional comments on the information provided. Three SERs provided
additional comments on the rule components after the teleconference.
In general, the SERs consulted on the GWR were concerned about the impact of the rule
on small water systems (because of their small staff and limited budgets), the additional monitoring that
might be required, and the data and resources necessary to conduct a hydrogeologic sensitivity
assessment or sanitary survey.
The SBAR Panel suggested that, given the number of systems that could be affected by the
rule, EPA consider focusing compliance requirements on those systems most at risk of fecal
contamination. From this perspective, the panel suggested that EPA evaluate whether it would be
appropriate to establish different rule requirements for systems based on system type, size or location.
The Panel also suggested providing States with maximum flexibility, consistent with ensuring an
appropriate minimum level of public health protection, to tailor specific requirements to individual
system needs and resources.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 7-3
-------
The SBAR Panel's recommendations to address the SERs Ground Water Rule concerns were
considered in developing the regulatory options analyzed in this rulemaking. The results of an updated analysis of
the impact of the preferred regulatory option on small water systems options is presented below.
7.2.3 Small Entity Impacts
For purposes of this regulatory flexibility analysis, the results of the economic impact
analysis for small water systems under the proposed Multi-Barrier option are summarized. Estimates of
the number of small entities affected and the cost of complying with each component of the Multi-
Barrier approach are presented. The estimated impacts for this preferred option are based on the
national mean compliance cost across the four compliance scenarios. Since taking an arithmetic mean
of the system-level impacts across compliance scenarios is not possible, system-level impacts are
investigated using various corrective action and significant defects scenarios. The high correction
action/low significant defect scenario is considered a middle-of-the-road cost scenario in the following
discussion.
7.2.3.1 Number of Small Entities A ffected
According to the December 1997 data from EPA's Safe Drinking Water Information System
(SDWIS), there are 156,846 community water systems and noncommunity water supplies providing
potable ground water to the public, of which 155,254 (99 percent) are classified by EPA as small
entities. These are presented in Exhibit 4-1 and 4-2. EPA estimates that these small ground water
systems served a population of more than 48 million. Roughly one-quarter of these systems were
estimated to be community water systems serving fixed populations on a year-round basis.
Under the proposed option, all community and noncommunity water systems are affected by at
least one requirement of the GWR, namely, the sanitary survey provision. The other GWR components
are estimated to affect different numbers of small systems. Exhibit 7-1 shows the estimates of affected
systems for each component of the proposed Multi-Barrier GWR option.
7-4 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
-------
Only systems determined to be sensitive that do not already treat to 4-log inactivation or
removal are required to conduct the additional routine monitoring. These systems must test their source
water monthly for a year. If no fecal indicators are found after 12 months of monitoring, the State may
reduce the monitoring frequency for that system. Similarly, if a nonsensitive system does not have a
distribution system, any sample taken for TCR compliance is effectively a source water sample so an
additional triggered source water sample would not be required. In both cases, however, if the system
has a positive sample forE.coli, coliphage, or fecal coliform, the system is required to conduct the
necessary follow-up actions.
The estimated record keeping and reporting burden associated with these provisions of the rule
are presented in Section 7.5 (Paperwork Reduction Act) below.
7.2.3.3 Small Entity Compliance Costs
When determining the costs and benefits of this proposed rule, EPA considered the full range of
both potential costs and benefits for the rule. The flexibility of the risk-based targeted approach of the
rule aims to reduce the cost of compliance with the rule. Small systems, in particular, benefit from this
design. Estimates of compliance costs of the Multi-Barrier approach to these systems are presented
below in Exhibit 7-2.
Given the available data, EPA determined that an expenditure test was the most reliable method
to gauge the potential economic impact of the proposed GWR on small systems. Using information on
current, or baseline, CWS expenses from the 1995 CWSS and the estimated cost impacts of the
proposed GWR, EPA developed a Monte Carlo simulation model to estimate the effect of GWR
compliance expenditures on total small water system expenses. Exhibit 7-2 provides the basic results
of this analysis. For each CWS size category, the mean system-level baseline expenses, mean system-
level GWR compliance costs, the mean system-level after-rule total expenses, and the mean percentage
increase in system-level total expenses is shown. This analysis shows that the smallest water systems,
those serving populations of fewer than 100 persons, will experience the greatest adverse impact,
relative to current expense levels (12%increase). Meanwhile, on average, systems serving over 1,000
people will see a very modest increase in total expenses (1% increase).
Of course, examining average impacts masks the distributional impacts of the proposed GWR.
Exhibits 7-3 through 7-7 show the detailed results of the simulation model, which estimated the
distribution of baseline expenses and projected expenses after compliance with the Multi-Barrier option
(preferred rule option). The percentage increase in expenses is also provided. These charts illustrate
two important points. First, in the smallest CWS size category, half of the systems will see their total
expenses rise by more than 10 percent as a result of the GWR. Ten percent of these smallest of
systems will see their annual expenses double after compliance with the proposed rule. On the other
hand, in a somewhat larger CWS size category (i.e., those serving 1,000 to 3,300 people), half of the
systems will have an increase in total expenses of about one percent, while only two percent of systems
will see their annual expenses double.
76 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
Exhibit 7-4. Comparison of CWS Baseline and Post-
Compliance Expenses
(Systems Serving 101-500 People)
$300,000
$250,000' '
$200,000-
UJ $150,000
$100,000-
$50,000- '
1000.00%
10% 20% 30% 40% 50% 60% 70% 80% 90% 95% 98% 100%
Percentage of Systems
~Baseline Expenses With Rule Expenses - - - -Percentage Increase in Expenses
Exhibit 7-5. Comparison of CWS Baseline and Post-Compliance
Expenses
(Systems Serving 501-1,000 People)
1 1 1 1 1 1 1 1 1 1 1 1- 0.01%
10% 20% 30% 40% 50% 60% 70% 80% 90% 95% 98% 100%
Percentage of Systems
-Baseline Expenses With Rule Expenses - - - -Percentage Increase in Expenses
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
7-7
-------
Exhibit 7-6. Comparison of CWS Baseline and
Post-Compliance Expenses
(Systems Serving 1,001-3,300 People)
Expenses
ro
3
c
c
1 I.UUU.UUU
$900,000 -
$800,000 -
$700,000
$600,000
$500,000
$400,000
$300,000 '
$200,000 -
$100,000 '
(t-n -
4
."
* I
t' i '
/ 1
t
1
,..-' f
- '" /
--" ^
--""" f
L^^
20% 30% 40% 50% 60% 70% 80%
Percentage of Systems
' 10.00% §
' 1.00% e
0.01%
98% 100%
Baseline Expenses
With Rule Expenses -Percentage Increase in Expenses
7.2.4 Coordination With Other Federal Rules
To avoid duplication of effort, the proposed GWR encourages States to use their source
water assessments, which are being developed by each State, when the assessment provides data
relevant to the sensitivity assessment of a system. The schedule for the sensitivity assessment (within
three years for CWS and five years for NCWS) should allow States to complete the assessment and
the first round of sanitary surveys concurrently if they choose to do so.
EPA has structured this GWR proposal as a targeted, risk-based approach to reducing
fecal contamination. The only regulatory requirement that applies to all ground water systems is the
sanitary survey. The required frequency for community systems is once every three years, which may
be changed by the State to once every five years if the system either treats to 4-1 og inactivation for
removal of virus or has an outstanding performance record documented in previous inspections and has
no history of total coliform MCL or monitoring violations since the last sanitary survey under current
ownership. The required frequency for sanitary surveys is every five years for noncommunity systems.
The majority of the small systems are noncommunity systems so the majority of systems will only have a
sanitary survey once every five years. At this frequency, EPA believes that the requirements will not be
burdensome for even the smallest systems. Similarly, the only additional monitoring requirements in
today's proposal are for undisinfected systems that are either located in sensitive hydrogeologic settings
7-8
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
or have a total coliform positive sample in the distribution system. The monitoring required for a total
coliform positive sample under the TCR would be a one-time event while the monitoring for sensitive
systems would be on a routine monthly basis for at least a year.
Finally, the SBAR Panel noted that disinfection of public water supplies may result in an
increase in other contaminants of concern, depending on the characteristics of the source water and the
distribution system. Of particular concern were disinfection byproducts, lead, copper, and arsenic.
EPA believes that these issues, when they occur will be very localized and may be addressed through
selection of the appropriate corrective action. EPA has provided States and systems with the flexibility
to select among a variety of corrective actions. These include options such as UV disinfection, or
purchasing water from another source, which should avoid problems with other disinfection.
7.2.5 Minimization of Economic Burden
On an annual basis, the cost of the preferred alternative is $182.7 million, assuming an
interest rate (i.e., cost of capital) of three percent (or $198.6 million, assuming an interest rate of seven
percent). In developing this proposal, however, EPA considered the recommendations of the SBAR to
minimize the cost impact to small systems. The proposed multi-barrier, risk-based approach was
designed to achieve maximum public health protection while avoiding excessive compliance costs
associated with Across-the-Board Disinfection regulatory compliance requirements.
To mitigate the associated compliance cost increases across water systems, the proposed
GWR also provides States with considerable flexibility when implementing the rule. This flexibility,
recommended also by the SBAR, will allow States to work within their existing program, but give
systems and the general public a clear understanding of what constitutes a significant deficiency.
Similarly, the rule allows States to consider the characteristics of individual systems when determining
an appropriate corrective action. States have the flexibility to use any disinfection treatment technology,
provided it achieves 4-1 og inactivation or removal of pathogens.
To determine the costs and benefits of this proposed rule, EPA considered the full range of
potential costs and benefits for the rule. The flexibility in the rule is designed to reduce the cost of
compliance with the rule, particularly for small systems. While determining the costs of the various
technologies, EPA estimated the percentage of systems in consultation with the States that will choose
between the different technologies, in part based on system size. EPA also considered a range of
benefits from reduction in illness and mortality to avoided cost of averting behavior, reduced
uncertainty, and avoided outbreak costs. However, only reductions in acute viral and bacterial illnesses
and mortality from virus are monetized. More detailed information is included in Chapter 5 ("Benefits
Analysis") and Chapter 6 ("Cost Analysis") of this RIA.
EPA further recognized that the operational characteristics of water systems' are highly
variable. In this proposal, State's have considerable flexibility when working with systems to address
significant deficiencies and take corrective action.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 7-9
-------
7.3 Unfunded Mandates Reform Act
Title H of the Unfunded Mandates Reform Act of 1995 (UMRA), P.L. 104-4, establishes
requirements for 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 to State, local, and Tribal governments, in the aggregate, or
to the private sector, of $100 million or more in any one year. Before promulgating an EPA rule, for
which a written statement is needed, section 205 of the UMRA 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. 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 or least burdensome alternative if the
Administrator publishes with the final rule an explanation on why that alternative was not adopted.
Before EPA establishes any regulatory requirements that may significantly or uniquely affect
small governments, including Tribal governments, it must have developed, under section 203 of the
UMRA, a small government agency plan. The plan must provide for notification to potentially affected
small governments, enabling officials of affected small governments to have meaningful and timely input
in the development of EPA regulatory proposals with significant Federal intergovernmental mandates;
and informing, educating, and advising small governments on compliance with the regulatory
requirements.
EPA has determined that this rule contains a Federal mandate that may result in expenditures of
$100 million or more for State, local, and Tribal governments, in the aggregate, and the private sector in
any one year. Accordingly, under Section 202 of the UMRA, EPA is obligated to prepare a written
statement addressing:
The authorizing legislation;
Cost-benefit 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;
Macro-economic 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; and
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.
7-10 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
The authorizing legislation, item one, is described in section 2.5.1. Items two through five are
addressed below, with the exclusion of future compliance costs, which are discussed in Chapter 6.
Regulatory alternatives, item six, are addressed in Chapter 3 and Chapter 6. Chapter 8 shows the cost
effectiveness analysis for each option.
7.3.1 Social Costs and Benefits
The social benefits are those that primarily accrue to the public through an increased level of
protection from viral and bacterial illness due to exposure to microbial pathogens in drinking water. To
assign a monetary value to the illness, EPA used cost-of-illness by several age categories to estimate
the benefits from the reduction in viral illness that result from this rule. This is considered a lower-bound
estimate of actual benefits because it does not include the pain and discomfort associated with the
illness. Mortalities were valued using a value of statistical life estimate consistent with EPA policy.
Chapter 5 presents the benefit analysis, which includes both qualitative and monetized benefits of
improvements to health and safety. The estimated annual benefit of the proposed GWR is $205 million
under the Multi-Barrier Option.
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 6 of this document details the cost analysis performed for the
GWR. Under the preferred option of the GWR, the likely compliance scenario is expected to result in
a total annualized cost of approximately $182.7 million using a three percent discount rate (or $198.6
million using a seven percent discount rate).
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 Grants Program. Additional funding is available from other programs administered either
by EPA or other Federal agencies. These include EPA's Drinking Water State Revolving Fund
(DWSRF), U.S. Department of Agriculture's Rural Utilities' Loan and Grant Program, and Housing
and Urban Development's Community Development Block Grant Program.
For example, SDWA authorizes the Administrator of the EPA to award capitalization grants to
States, which in turn can provide low cost loans and other types of assistance to eligible public water
systems. The DWSRF assists public water systems with financing the costs of infrastructure needed to
achieve or maintain compliance with SDWA requirements. Each State has considerable flexibility in
determining the design of its DWSRF Program and to direct funding toward its most pressing
compliance and public health protection needs. States may also, on a matching basis, use up to 10
percent of their DWSRF allotments for each fiscal year to assist in running the State drinking water
program. In addition, States have the flexibility to transfer a portion of funds to the Drinking Water
State Revolving Fund from the Clean Water State Revolving Fund.
Furthermore, a State can use the financial resources of the DWSRF to assist small systems the
majority of which are ground water systems. In fact, a minimum of 15 percent of a State's DWSRF
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 7-11
-------
grant must be used to provide infrastructure loans to small systems. Two percent of the State's grant
may be used to provide technical assistance to small systems. For small systems that are
disadvantaged, up to 30 percent of a State's DWSRF may be used for increased loan subsidies. Under
the DWSRF, tribes have a separate set-aside which they can use.
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 up to 10,000 people. In fiscal year
1997, the RUS had over $1.3 billion in available funds. Also, three sources of funding exist under the
CDBG program to finance building and improvements of public faculties such as water systems. The
three sources of funding include: 1) direct grants to communities with populations over 200,000; 2)
direct grants to States, which they in turn award to smaller communities, rural areas, and colonies in
Arizona, California, New Mexico, and Texas; and 3) direct grants to U.S. Territories and Trusts. The
CDBG budget for fiscal year 1997 totaled over $4 billion.
7.3.2 Disproportionate Impacts
This analysis examines disproportionate impacts upon geographic or social segments of the
nation. In general, the costs that a public water system, whether publicly or privately owned, will incur
to comply with this rule will depend on many factors that are not generally based on location.
However, the data needed to confirm this assessment and to analyze other impacts of this problem are
not available; therefore, EPA looked at three other factors:
The impacts of small versus large systems and the impacts within the five small system size
categories;
The costs to public versus private water systems; and
The costs to households (See Section 6.4).
The first measure of disproportionate impact considers the cost incurred by small and large
systems. Small systems will experience a greater impact than large systems under the GWR. The
higher cost to the small ground water systems is mostly attributable to the large number of these types
of systems (i.e., 99 percent of ground water systems serve <10,000 people). Other reasons for the
disparity include: 1) large systems are more likely to already disinfect their ground water (disinfection
exempts a system from triggered and routine monitoring), 2) they typically have greater technical and
operational expertise, and 3) they are more likely to engage in source protection programs. The
potential economic impact among the small systems will be the greatest for systems serving less than
100 persons, as shown in Exhibits 6-19 and 6-20.
The second measure of impact is the relative total cost to privately owned water systems
compared to that incurred by publicly owned water systems. Exhibit 7-8 reveals that 28 percent of the
system compliance costs are borne by publicly owned PWSs, while 61 percent is borne by privately
7-12 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
owned PWSs. This is a result of the fact that 73 percent of PWSs are owned by private entities. EPA
has no basis for expecting cost per system to differ systematically with ownership.
The costs to households has been examined and is summarized in Section 6.4 of this document.
Exhibit 7-8. Annual Compliance Cost
Impacts by PWS Ownership
3 Percent Discount Rate
SYSTEM TYPE
Public System Cost
State Cost
Total Public Cost
Private System Cost
Ancilliary System Cost
Total Private Cost
Cost
(million $)
$52.0
$20.6
$72.6
$103.8
$6.4
$110.2
% of Total
Cost
28%
11%
39%
57%
4%
61%
7 Percent Discount Rate
SYSTEM TYPE
Public System Cost
State Cost
Total Public Cost
Private System Cost
Ancilliary System Cost
Total Private Cost
Cost
(million $)
$56.5
$22.1
$78.7
$113.0
$7.0
$119.9
% of Total
Cost
28%
11%
39%
57%
4%
61%
7.3.3 Macro Economic Effects
Under UMRA Section 202, EPA is required to estimate the potential macro-economic
effects of the regulation. Macro-economic 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 Gross
Domestic Product (GDP). In 1998, real GDP was $7,552 billion so a rule would have to cost at least
$19 billion 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.
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
7-13
-------
The macro-economic effects on the national economy from the proposed GWR should be negligible
based on the fact that the total expected annual costs of the preferred regulatory option for this rule are
estimated to be $182.7 million using a three percent discount rate (or $198.6 million using a seven
percent discount rate).
7.3.4 Consultations with State, Local, and Tribal Governments
Consistent with the intergovernmental consultation provisions of section 204 of UMRA
section 204 of the UMRA and Executive Order 12875 "Enhancing the Intergovernmental Partnership,"
EPA has initiated consultations with the governmental entities affected by this rule. EPA held four
public meetings for all stakeholders and two Association of State Drinking Water Administrators early
involvement meetings. Because of the Rule's impact on small entities, the Agency convened a Small
Business Advocacy Review (SBAR) Panel in accordance with the Regulatory Flexibility Act (RFA) as
amended by the Small Business Regulatory Enforcement Fairness Act (SBREFA) to address small
entity concerns, including small local governments specifically. EPA consulted with small entity
representatives prior to convening the Panel to get their input on the GWR. Of the 22 small entity
participants, five represented small governments. EPA also made presentations on the GWR to the
national and local chapters of the American Water Works Association, the Ground Water Foundation,
the National Ground Water Association, the Rural Water Association, and the National League of
Cities. Twelve State drinking water representatives also participated in the Agency's GWR
workgroup.
In addition to these consultations, EPA circulated a draft of this proposed rule and
requested comment from the public through an informal process. Specifically, on February 3, 1999,
EPA posted on the EPA's Internet web page and mailed out over 300 copies of the draft to people
who had attended the 1997 and 1998 public stakeholder meetings as well as people on the EPA
workgroup. EPA received 79 letters or electronic responses to this draft: 34 from State government
(representing 30 different States), 25 from local governments, 10 from trade associations, six from
Federal government agencies, and four from other people/organizations. No comments were received
from Tribal governments. EPA reviewed the comments and carefully considered their merit. The
proposed GWR reflects many of the commentors' points and suggestions.
To inform and involve Tribal governments in the rulemaking process, EPA presented the
GWR at the 16th Annual Consumer Conference of the National Indian Health Board, at the annual
conference of the National Tribal Environmental Council, and at an Office of Ground Water and
Drinking Water (OGWDW)/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 over 100 tribes were represented at the annual conference of the National
Tribal Environmental Council. At both conferences, an OGWDW representative conducted two
workshops on EPA's drinking water program and upcoming regulations, including the GWR.
Comments received from Tribal governments regarding the GWR focused on concerns and
some opposition to mandatory disinfection for ground water systems. They also suggested that any
waiver process be adequately characterized by guidance and simple to implement. The proposed
7-14 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
GWR was designed so that a majority of systems will not be required to disinfect. Systems will have
the opportunity to correct significant deficiencies, find a new source, and in some cases monitor for
fecal contamination. Disinfection is only required under the proposed GWR if these other measures do
not work. However, some systems in coordination with the primacy agent or State, might choose
disinfection over these other options because it may be the least costly alternative.
At the OGWDW/Inter Tribal Council of Arizona meeting, representatives from 15 tribes
participated. In addition, over 500 tribes and Tribal organizations were sent the presentation materials
and meeting summary. Because many tribes have ground water systems, participants expressed
concerns over some elements of the rule. Specifically, they had concerns about how the primacy agent
would determine significant deficiencies identified in a sanitary survey and how the sensitivity assessment
would be conducted. Because no tribes currently have primacy, EPA is the primacy agent and will
identify significant deficiencies as part of sanitary surveys and conduct the sensitivity assessments.
7.4 Paperwork Reduction Act
The information collected as a result of this rule will allow the State and EPA to evaluate
PWS compliance with the rule. For the first three years after promulgation of this rule, the major
information requirements pertain to start up costs for States to satisfy primacy requirements and
systems to become familiar with the rule. Responses to the request for information are mandatory (Part
141). The information collected is not confidential.
EPA is required to estimate the burden on PWS for complying with the GWR. Burden
means the total time, effort, or financial resources expended by persons to generate, maintain, retain, or
disclose or provide information to or for a Federal agency. This includes the time needed to review
instructions; develop, acquire, install, and utilize technology and systems for the purposes of collecting,
validating, and verifying information, processing and maintaining information, and disclosing and
providing 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; and transmit or otherwise disclose the information.
Exhibit 7-9 presents EPA's estimates of the annual burden on PWS and States for
reporting and record keeping from the first three years after promulgation of the preferred multi-
barrier option. It should be noted that the majority of the monitoring, record keeping and reporting
burden occurs beyond the three-year period of the estimate. The Information Collection Request
prepared by EPA, includes a estimate for 10-year time frame to show the costs and burdens
beyond the initial period covered by the ICR, to reflect the reality of full rule implementation.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 7-15
-------
Exhibit 7-9. Summary of the Ground Water Rule
Total Respondents, Responses, Burden, and Costs for PWSs and States
PWSs
States and
Territories
Total
Number
Respondents
Annually
52,331
56
52,387
Number
Responses
Annually
99,821
168
99,989
Total
Annual
Burden
(hrs)
263,238
88,107
351,345
Total
Annual
Labor Cost
$7,819,882
$2,332,979
$10,152,861
Total
Annual
Capital
Cost
$1,376,302
$0
$1,376,302
Total
Annual
O&M Cost
$0
$0
$0
7.5 Protecting Children From Environmental Health
Risks and Safety Risks
Executive Order (EO) 13045 (62 FR 19885, April 23, 1997) applies to any rule initiated
after April 21, 1997, or proposed after April 21, 1998, that (1) is determined to be "economically
significant" as defined under E.O. 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 EPA.
As discussed in Chapter 5 of this RIA, in developing the risk and benefits analysis for the
GWR, the effects on children, both in terms of unique risk and cost-of-illness estimates, were explicitly
taken into consideration. This analysis suggests that the proposed rule provides a greater per capita
health benefit to children than to adults, mostly due to the high cost-of-illness associated with viral
illnesses avoided in young children. In other words, the analysis suggests that the viral and bacterial
illnesses of concern to the GWR disproportionately effect children, and therefore, the benefits of the
proposed rule accrue disproportionately to children.
As can be seen in Exhibit 7-10, the proposed Multi-Barrier option results in the second
most number of cases of illness avoided-second only to the Across-the-Board Disinfection option.
Given the extraordinary costs of the Across-the-Board Disinfection option, EPA believes that the
proposed Multi-Barrier option is the most protective of children's health of all reasonably feasible
alternatives.
7-16
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit 7-10. Viral Illnesses and Deaths Avoided In Children
Across Regulatory Alternatives
GWR Option
Sanitary Survey
Sanitary Survey
and Triggered
Monitoring
Multi-Barrier
Across-the-
Board
Disinfection
Number of Viral Illnesses Avoided per Year
< 5 years old
5-16 years old
2,292
1,773
13,044
9,974
15,058
11,508
21,125
16,059
Number of Viral Deaths Avoided per Year
< 5 years old
5-16 years old
0
0
1
1
1
1
1
2
Annual Cost (million $)
3% discount rate
7% discount rate
$72.7
$76.0
$157.6
$168.5
$182.7
$198.6
$777.1
$866.0
With regard to sensitive sub-populations, EPA explicitly examined the effects of the
proposed rule both on young children and immuno-compromised individuals. As discussed above,
Exhibit 7-10, illustrates that the proposed Multi-Barrier option is the most protective of children's
health of all reasonably feasible alternatives. Similarly, Exhibit 7-11, below shows that the
proposed Multi-Barrier option results in the second most number of cases of illness avoided among
immuno-compromised individuals-second only to the Across-the-Board Disinfection option.
Given the extraordinary costs of the Across-the-Board Disinfection option, EPA believes that the
proposed Multi-Barrier option is the most protective of immuno-compromised individuals' health
of all the reasonably feasible alternatives.
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
7-17
-------
Exhibit 7-11. Viral Illnesses Avoided in Immuno-Compromised
Persons Across Regulatory Alternatives
GWR Option
Sanitary Survey
Sanitary Survey
and Triggered
Monitoring
Multi-Barrier
Across-the-
Board
Disinfection
Number of Viral Illnesses Avoided per Year
< 5 years old
5-1 6 years old
> 16 Years Old
23
18
191
130
100
1,086
151
115
1,253
211
161
1,746
Number of Viral Deaths Avoided per Year
< 5 years old
5-1 6 years old
> 16 Years Old
0.00
0.00
0.01
0.01
0.00
0.03
0.01
0.00
0.04
0.01
0.01
0.05
Annual Cost (million $)
3% discount rate
7% discount rate
$72.7
$76.0
$157.6
$168.5
$182.7
$198.6
$777.1
$866.0
7.6 Environmental Justice
Executive Order 12898 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 has consulted with minority and low-income stakeholders.
The Environmental Justice Executive Order requires the Agency to consider environmental
justice issues in the rulemaking and to consult with Environmental Justice (EJ) stakeholders. There are
two aspects of the proposed GWR that relate specifically to this policy, the overall nature of the rule,
and the convening of a stakeholder meeting specifically to address environmental justice issues.
As part of EPA's responsibilities to comply with Executive Order 12898, the Agency held
a stakeholder meeting on March 12, 1998 to address various components of pending drinking water
regulations; and how they may impact sensitive sub-populations, minority populations, and low-income
populations. Topics discussed included treatment techniques, costs and benefits, data quality, health
effects, and the regulatory process. Participants included national, State, Tribal, municipal, and
individual stakeholders. EPA conducted the meetings by video conference call with participants in
7-18
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
eleven cities. This meeting was a continuation of stakeholder meetings that started in 1995 to obtain
input on the Agency's Drinking Water Programs. The major objectives for the March 12, 1998
meeting were:
Solicit ideas from Environmental Justice stakeholders on known issues concerning current
drinking water regulatory efforts;
Identify key issues of concern to EJ stakeholders; and
Receive suggestions from EJ stakeholders concerning ways to increase representation of
Environmental Justice communities in EPA regulatory efforts.
In addition, EPA developed a plain-English guide specifically for this meeting to assist
stakeholders in understanding the multiple and sometimes complex drinking water issues.
The GWR applies to all public water systems that use ground water as their source water,
including community water systems, nontransient noncommunity water systems, and transient
noncommunity water systems. Consequently, the health protection benefits provided by this proposed
rule are equal across all of the income and minority groups served by these systems. Existing
regulations such as the Surface Water Treatment Rule and Interim Enhanced Surface Water Treatment
Rule provide similar health benefit protection to communities that use surface water or ground water
under the influence of surface water. Therefore, EPA believes this rule will equally protect the health of
all minority and low-income populations served by systems regulated under this rule from exposure to
microbial contamination.
7-19
-------
8. Summary of Costs and Benefits
8.1 Review of Regulatory Options, Costs and Benefits
The proposed Ground Water Rule (GWR) must specify the appropriate use of disinfection
treatment and simultaneously addresses other components of ground water system operation and
maintenance to assure public health protection. In this RIA, EPA has analyzed the cost and health
benefit impacts associated with four regulatory options. Since the basic provisions for the four options
build upon one another, the associated costs and benefits are also expected to increase from the less
stringent to the more stringent options. Chapter 5 ("Benefits Analysis") described in detail the
estimated national health benefits of the GWR regulatory options, while Chapter 6 ("Cost Analysis")
described the projected national compliance cost estimates. This chapter presents a summary and
comparison of the national benefits and costs for each of these four regulatory options.
8.1.1 Review of Regulatory Options
The four regulatory options that EPA evaluated for this proposed rulemaking capture a range of
benefits and costs based on the number of PWSs estimated to be affected by the specific compliance
requirements. Each option, for example, assumes that a different proportion of PWSs will be required
to take action, depending on the requirement set forth under the rule option. In general, these four
options may be characterized by the main regulatory requirements summarized in Exhibit 8-1 below.
Exhibit 8-1. GWR Regulatory Options and Main Requirements
Regulatory
Compliance
Requirement
Sanitary survey
Triggered monitoring
Hydrogeologic sensitivity assessment
and routine monitoring
All systems must install or upgrade and
maintain treatment
GWR Option
Option: 1
Sanitary
Survey Only
/
Option: 2
San. Survey
& Triggered
Monitoring
/
/
Option 3:
Multi-Barrier
Approach1
/
/
/
Option 4:
Across-the-
Board
Disinfection
/
/
1 Preferred Option
EPA has selected the Multi-Barrier Option as its preferred option. Based upon the information
presented in this document, EPA believes that monetized net benefits may be maximized under the
Multi-Barrier option, that this option provides additional benefits that EPA did not monetize, and that it
is best achieves the rule's objective to reduce the risk of illness and death from microbial contamination
in PWSs relying on ground water.
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
8-1
-------
As shown in Exhibit 8-1 above, the regulatory requirements build incrementally upon one
another for a more targeted approach to identify and correct any systems not in compliance. The
exception to this trend is the Across-the-Board Disinfection option, which requires that all ground
water systems achieve and demonstrate 4-log treatment (inactivation and/or removal). The
components and differences between these regulatory scenarios are discussed in more detail in
Chapter 3 ("Consideration of Regulatory Options").
8.1.2 Review of National Cost Estimates
Exhibit 8-2 presents EPA's estimates for the total national cost for the four GWR regulatory
scenarios. Using a 3 percent discount rate, costs on an annualized basis range from $72.7 million for
Option 1, Sanitary Survey only, to $777.1 million for Option 4, Across-the-Board Disinfection.
Assuming a 7 percent discount rate, as shown in Exhibit 8-3, the range of total annualized national cost
increases to $76.0 million annually under Option 1, Sanitary Survey only, to $866.0 million for Option
4, Across-the-Board Disinfection.
Exhibit 8-2. Summary of National Benefits and Costs
(Using 3 Percent Discount Rate)
Regulatory Option
Option 1 : Sanitary Survey Only
Option 2: Sanitary Survey and Triggered Monitoring
OptionS: Multi-Barrier Approach1
Option 4: Across-the-Board Disinfection
Millions ($)
Benefit
$32.5
$177.9
$205.0
$283.1
Cost
$72.7
$157.6
$182.7
$777.1
Net Benefit
($40.2)
$20.3
$22.3
($494.0)
1 Preferred Option
Exhibit 8-3. Summary of National Benefits and Costs
(Using 7 Percent Discount Rate)
Regulatory Option
Option 1 : Sanitary Survey Only
Option 2: Sanitary Survey and Triggered Monitoring
Option 3: Multi-Barrier Approach1
Option 4: Across-the-Board Disinfection
Millions ($)
Benefit
$32.5
$177.9
$205.0
$283.1
Cost
$76.0
$168.5
$198.6
$866.0
Net Benefit
($43.5)
$9.4
$6.4
($582.90)
1 Preferred Option
The four options considered in this proposed GWR reflect increasing levels of protection
against outbreaks from microbial contamination employing a variety of control measures. The total
annual cost of compliance steadily increases across Options 1, 2 and 3the Sanitary Survey, Sanitary
Survey and Triggered Monitoring, and Multi-Barrier options, respectively. The total annual cost of
8-2
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
compliance for Option 4, Across-the-Board Disinfection ($866 million at 7 percent discount rate) is,
however, over four times the cost of compliance for Option 3, Multi-Barrier approach ($198.6 million
at a 7 percent discount rate). This trend is due largely to the requirement that all GWSs treat their
water source under the Across-the-Board Disinfection option, regardless of the source water quality
and potential for fecal contamination.
8.1.3 Review of National Benefits Estimates
The monetized health benefits associated with each of the four GWR options is also shown in
Exhibit 8-2 and Exhibit 8-3. As noted above, these four rule options provide increasing levels of
protection against microbial contamination, as reflected in the estimated health benefits. Annual national
benefits range from $32.5 million for Option 1, Sanitary Survey only, to $283.1 million for Option 4,
Across-the-Board Disinfection. The national benefits estimates do not show the four-fold increase that
was observed in national costs between Option 3, the Multi-Barrier Approach, and Option 4, Across-
the-Board Disinfection. The value of national benefits under these two regulatory options differ by
approximately $78.1 million annually.
8.2 Comparison of Benefits and Costs
This section presents a comparison of total benefits and costs for each of the four GWR
options. Three separate analyses are considered, including a direct comparison of aggregate national
cost and benefits, the presentation of net benefits, and the results of a cost-effectiveness analysis of each
regulatory option.
8.2.1 National Benefit-Cost Comparison
Exhibits 8-2 and 8-3 present monetized net benefits (i.e., the absolute difference between the
total value of national costs and benefits for each rule option). Both Option 2, Sanitary Survey and
Triggered Monitoring, and Option 3, Multi-Barrier Approach, show positive net benefits whether
capital costs are annualized at 3 or 7 percent. Under both discount rate scenarios, both the Sanitary
Survey option and the Across-the-Board Disinfection have negative net benefits (see also Exhibit 8-4).
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 8-3
-------
Exhibit 8-4.
Comparison of National Costs and Benefits (7% Discount Rate)
$1,000.0
$900.0
$800.0
$700.0
$600.0
$500.0
$400.0
$300.0
$200.0 --
$100.0
Sanitary Survey
Sanitary Survey and
Triggered Monitoring
Multi-Barrier
Across-the-Board
Disinfection
D Costs (million $)
$76.0
$168.5
$198.6
E3 Benefits (million $)
$32.5
$177.9
$205.0
$283.1
Exhibit 8-5 and Exhibit 8-6 present the estimated monetized net benefits of the four rule
options by systems size, under discount rates of 3 and 7 percent, respectively. As the system size
category increases, the net benefits of each rule option increase. This reflects the fact that the health
benefits of each option are a linear function of population, while the per capita cost of compliance
drops, as system size increases, reflecting economies of scale in the production of clean drinking water.
EPA also expects nonmonetized net benefits to increase as system sizes increase because a larger
number of people would be affected by benefits such as decreased chronic illness, improved
distribution systems, and greater confidence in public water supplies.
8-4
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit 8-5. Net Benefits of Each Regulatory Option
by System Size Category (Using 3 Percent Discount Rate (million$))
Option/Regulatory Scenario
Option 1: Sanitary
Survey Only
Option 2: Sanitary Survey
and Triggered Monitoring
OptionS: Multi-
Barrier Approach1
Option 4: Across-the-Board
Disinfection
$ Millions; By System Size
o
o
T-
V
($22.4)
($65.8)
($70.6.)
($185.8)
0
0 0
0 0
m T-
I I
O T-
o o
T- m
($11.6.)
($16.4)
($21.8)
($167.9)
($3.5)
$3.4
$2.9
($54.7)
CO T-
-------
8.2.2 Cost-Effectiveness
Cost-effectiveness analysis is another commonly used measure of the economic efficiency. This
analysis compares how well the regulatory options are meeting the intended regulatory objectives. For
the proposed GWR, the cost-effectiveness can be measured as the cost per case of illness avoided.
Exhibit 8-7 shows the incremental cost per case avoided for each GWR option. Specifically,
the bar graph in Exhibit 8-7 shows that the additional cases avoided under Option 4, Across-the-
Board Disinfection, will cost approximately $12,000 more per case than under Option 3, Multi-Barrier
Approach. The Sanitary Survey and Triggered Monitoring option achieves the lowest incremental cost
per case of illness avoided at $1,123 per case while the Multi-Barrier option is only slightly larger at
$1,954 per case.
8.3 Uncertainty in Benefit and Cost Estimates
Exhibit 8-7 Incremental Cost per Case Avoided Across GWR Options
$1b,UUU
tfl
-------
First, there is uncertainty about the baseline number of systems and population because of data
limitations in SDWIS. For example, some systems use both ground and surface water, but because of
other regulatory requirements, are labeled in SDWIS as surface water systems. Therefore, EPA does
not have a reliable estimate of how many of these mixed systems exist. The SDWIS data on
noncommunity water systems does not have a consistent reporting convention for population served.
Some noncommunity systems may report the population served over the course of a year, while others
may report the population served on an average day. Also, SDWIS does not require noncommunity
systems to provide information on current disinfection practices and, in some cases, it may overestimate
the population served. For example, a park may report the population served yearly instead of daily.
Second, the risk calculations concerning the baseline number of illnesses and the reduction of
illnesses that result from the various rule options contain some uncertainty. For example, a nationally
representative study of baseline microbial occurrence in ground water does not exist. EPA chose the
AWWARF study to represent properly constructed wells described in Chapter 4, because it is the
most geologically representative of the thirteen studies that were available. EPA also relied on data
from the EPA/AWWARF study to represent improperly constructed wells because this study targeted
wells vulnerable to contamination and wells tested monthly for a year. Additionally, EPA had to rely on
CDC outbreak data to characterize benefits associated with treatment failures and distribution system
contamination. The Agency also assumed that the occurrence of fecal contamination will remain
constant throughout the rule. However, this might not be the case if increased development results in
fecal contamination of a larger number of aquifers in areas served by ground water systems.
Also, EPA did not have dose-response data for all viruses and bacteria associated with
previous ground water disease outbreaks. For viral illness, the Agency used echovirus and rotavirus as
surrogates for all pathogenic viruses from fecal contamination that can be found in ground water. By
using these two viruses, the Agency captured the effects of low-to-medium infectivity viruses that cause
severe illness, and high infectivity viruses that cause more mild illness. Another source of uncertainty is
the number of baseline bacterial illnesses caused by ground water contamination. The bacterial risk
could not be modeled because of lack of occurrence and dose-response data. Estimates of bacterial
illness were made based on a ratio of bacterial to viral outbreaks as documented by the CDC and
applied to the viral risk estimate discussed above.
Third, some uncertainty exists regarding the costs of today's rule because of the diverse nature
of possible significant deficiencies systems would need to address. In addition, the rule's flexibility
leads to some uncertainty in the estimates of who will be affected by each rule component and how
nonts and systems will respond to significant deficiencies. These uncertainties could either under or
overestimate the costs of the rule.
Fourth, EPA intends to propose regulations for radon and arsenic in drinking water that will
affect a number of ground water systems and their disinfection practices. EPA also intends to finalize
the Stage 2 Disinfection Byproducts Rule by the statutory deadline of May 2002. It is extremely
difficult to estimate the combined effects of these possible regulations on ground water systems because
of various combinations of contaminants that some systems may need to address. However, it is
possible for a system to choose treatment technologies that would deal with multiple problems. The
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis 8-7
-------
combined total cost impact of these drinking water rules is uncertain; however, it would likely be less
than the cost of the simple sum of the estimated individual rules.
Finally, there are costs and benefits that are not monetized in this RIA. For example, a review
of the medical and epidemiological literature identified several potential chronic diseases resulting from
illnesses caused by enteroviruses (e.g., heart disease, diabetes, post-viral fatigue syndrome, and
pancreatitis)the strongest evidence for a viral role appears to exist for the development of diabetes
and myocarditis (inflammation of the muscular walls of the heart).
Because the causal relationship is not well established and the number of cases associated with
drinking water is unknown, the Agency was not able to quantify benefits from the GWR on reducing
these diseases. Nonetheless the total number of these conditions from all pathways in the United States
is substantial; it is estimated that nearly 7 million people have one form of diabetes and approximately
4.1 million have chronic heart disease (including myocarditis and cardiomyopathy).
In addition, the RIA does not include the value o f reduced pain and suffering because the
disutility of illness is not associated with a market cost.
There are also non-health benefits of the rule that could not be monetized, such as, the value of
upgrades to distribution systems, increased efficiencies, and increased frequency/intensity of process
surveillance.
Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
Appendix A
Risk Assessment Inputs and Results
-------
TABLE OF CONTENTS
Appendix A. Risk Assessment Inputs and Results A-l
A.I Inputs to the Risk Assessment A-2
A. 1.1 Potentially-Exposed Populations, Including Sensitive Subgroups A-2
A. 1.2 Viral Pathogen Occurrence and Concentration in Source Water A-3
A. 1.3 Pathogen Inactivation in Ground Water Systems and Resulting Baseline Tap
Water Concentrations A-5
A. 1.4 Drinking Water Consumption Factors A-5
A. 1.5 Hazard Identification Parameters A-6
A. 1.6 Illnesses and Deaths in Undisinfected and Disinfected (4-log removal)
Systems A-7
A. 1.7 Additional Illnesses and Deaths Resulting from Treatment
Failures and Distribution System Contamination A-7
A.2. Results of the Risk Calculations A-10
A.2.1 Results of the Risk Calculations: Baseline A-10
A.2.2 Results of the Risk Calculations: Sanitary Survey-Only Option A-l 3
A.2.3 Results of the Risk Calculations: Sanitary Survey and
Triggered Monitoring Option A-l 5
A.2.4 Results of the Risk Calculations: Option 3Multiple Barrier Option . . A-l 7
A.2.5 Results of the Risk Calculations: Across-the-Board Disinfection and
Sanitary Survey Option A-l9
LIST OF EXHIBITS
Exhibit A-l. Populations Served by Non-Disinfecting Ground Water Systems A-2
Exhibit A-2. Fractions of General Population having Age-Based and Health-Based
Sensitivity to Viral Pathogens in Ground Water A-2
Exhibit A-3. Viral Occurrence and Concentration in Source Water A-4
Exhibit A-4. Hazard Identification of Viral Pathogens for the GWR Risk Assessment A-8
Exhibit A-5. Summary of GWR Baseline Risk Calculations for Undisinfected SystemsA-9
Exhibit A-6. Estimates of Baseline Type A Viral Illness and Death
("All Sources, Consumers Only" Age-Based Consumption Distributions)\-10
Exhibit A-7. Estimates of Baseline Type B Viral Illness and Death
("All Sources, Consumers Only" Age-Based Consumption Distributions)\-l 1
Exhibit A-8. Estimates of Baseline Type A Viral Illness and Death
("Community Water Supply, All Respondents" Age-Based ConsumptionAHJtHbutions)
Exhibit A-9. Estimates of Baseline Type B Viral Illness and Death
("Community Water Supply, All Respondents" Age-Based ConsumptionA)isll2ibutions)
-------
Exhibit A-10. Estimates of Type A Viral Illness and Death after Option 1Sanitary
Survey Only ("All Sources, Consumers Only" Age-Based Consumption LM-stBbutions)
Exhibit A-l 1.Estimates of Type B Viral Illness and Death after Option 1Sanitary
Survey Only ("All Sources, Consumers Only" Age-Based Consumption Dii-stlrSbutions)
Exhibit A-l2.Estimates of Type A Viral Illness and Death after Option 1Sanitary
Survey Only ("Community Water Supply, All Respondents" Age-Based Consumption
Distributions) A-14
Exhibit A-l3.Estimates of Type B Viral Illness and Death after Option 1Sanitary
Survey Only ("Community Water Supply, All Respondents" Age-Based Consumption
Distributions) A-14
Exhibit A-l4.Estimates of Type A Viral Illness and Death after Option 2Sanitary
Survey and Triggered Monitoring ("All Sources, Consumers Only" Age-Based
Consumption Distributions) A-l 5
Exhibit A-l 5. Estimates of Type B Viral Illness and Death after Option 2Sanitary
Survey and Triggered Monitoring ("All Sources, Consumers Only" Age-Based
Consumption Distributions) A-l6
Exhibit A-l 6.Estimates of Type A Viral Illness and Death after Option 2Sanitary
Survey and Triggered Monitoring ("Community Water Supply, All Respondents" Age-
Based Consumption Distributions) A-l6
Exhibit A-l 7.Estimates of Type B Viral Illness and Death after Option 2Sanitary
Survey and Triggered Monitoring ("Community Water Supply, All Respondents" Age-
Based Consumption Distributions) A-l7
Exhibit A-l 8.Estimates of Type A Viral Illness and Death after Option 3Multi-Barrier
Approach ("All Sources, Consumers Only" Age-Based Consumption DisMtiulions)
Exhibit A-l 9.Estimates of Type B Viral Illness and Death after Option 3Multi-Barrier
Approach ("All Sources, Consumers Only" Age-Based Consumption DisMtiiSions)
Exhibit A-20.Estimates of Type A Viral Illness and Death after Option 3Multi-Barrier
Approach ("Community Water Supply, All Respondents" Age-Based Consumption
Distributions) A-l 8
Exhibit A-21. Estimates of Type B Viral Illness and Death after Option 3Multi-Barrier
Approach ("Community Water Supply, All Respondents" Age-Based Consumption
Distributions) A-l 9
Exhibit A-22.Estimates of Type A Viral Illness and Death from Ingestion of Ground Water
after Option 4Across-the-Board Disinfection ("All Sources, Consumers Only" Age-
Based Consumption Distributions) A-20
Exhibit A-23.Estimates of Type B Viral Illness and Death from Ingestion of Ground Water
after Option 4Across-the-Board Disinfection ("All Sources, Consumers Only" Age-
Based Consumption Distributions) A-20
Exhibit A-24.Estimates of Type A Viral Illness and Death from Ingestion of Ground Water after
Option 4Across-the-Board Disinfection ("Community Water Supply, All
Respondents" Age-Based Consumption Distributions) A-21
Exhibit A-25.Estimates of Type B Viral Illness and Death from Ingestion of Ground
Water after Option 4Across-the-Board Disinfection (Community Water
System Age-Based Consumption Distributions) A-21
-------
Included in this appendix are the inputs and results from the risk assessment performed for the
Ground Water Rule (GWR) benefits determination. Inputs for the risk assessment are organized under
Section A.I. Results of the risk assessment are organized under Section A.2 and include:
Results of the GWR baseline analysis (current conditions), as well as results for each of the fou
regulatory options modeled;
For each set of the five outputs, two sets of tables, each of which presents results for the two
types of modeled viral pathogens (Type A and Type B). Each table includes the mean annual
morbidity (illness) and mortality (death) estimates for PWS ground water systems; the 10th and
90th percentile estimates are also included to characterize the uncertainty associated with each
mean value. All numbers (mean and the 10th and 90th percentile values) are rounded to the
nearest whole number. The mean number represents the "best" estimate of the number of
illnesses and deaths (remaining and avoided) based on the input assumptions and calculations
used in the Monte Carlo simulation analysis. Given the uncertainties quantified in the model,
there is a 10% probability that the actual values are below the 10th percentile and a 10%
probability that they are above the 90th percentile.
The first set of tables in each section presents the results of calculations using daily, age-based
consumption distributions incorporating US Department of Agriculture (USD A) 1994-1996
daily individual intake data for community water, all drinking water sources ("all sources,
consumers only"). The "all sources, consumers only" results represent the upper bound estimat*
of annual illnesses and deaths for the baseline scenario and each regulatory option. The second
set of tables presents results generated using the USD A daily consumption data for "community
water supply, all respondents". The "community water supply, all respondents" results represent
the lower bound consumption for estimates of annual illnesses and deaths for each scenario.
April5, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis A-l
-------
A.1 Inputs to the Risk Assessment
Inputs to the GWR risk assessment are summarized in this section of the appendix.
A.1.1 Potentially-Exposed Populations, Including Sensitive Subgroups
Exhibit A-l lists the numbers of persons in the potentially exposed populations served by
undisinfected ground water systems. The populations are broken down by type and size of system. The
fraction of the population in sensitive subgroups is shown in Exhibit A-2.
Exhibit A-1. Populations Served by Non-Disinfecting Ground Water Systems
Service
Population
Category
<100
101-500
501-1,000
1001-3,300
3,301-10,000
10,001-50,000
50,001-100,000
>100,000
Totals
Estimated Population Served Undisinfected Ground Water1
Community Water
Systems (CWS)
471,214
1,005,419
852,017
2,667,256
3,670,492
2,309,365
3,898,783
3,472,410
18,346,956
Nontransient
Noncommunity
(NTNC) Systems
364,978
1,283,450
998,666
804,994
243,552
116,450
0
0
3,812,090
Transient
Noncommunity
(TNC) Systems
2,616,086
3,401,284
1,167,404
890,370
686,852
1,386,890
737,461
1,475,474
12,361,821
Source: GWR model calculation. Reference for total population served in each service population category is the Drinking Water Baseline
Handbook, draft, First Edition (EPA 1999J.
A-2
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit A-2. Fractions of General Population having Age-Based and
Health-Based Sensitivity to Viral Pathogens in Ground Water
Age Group
Neonate ( birth to 1 month)
Toddler ( >1 month to 2 years)
Young Child ( >2 to 5 years)
Child ( >5 yrs. to 16 years )
Adult (>1 6 to 65 years)
Elderly (>65 years)
Age
Fraction of
General
Population
0.005
0.0235
0.0442
0.1591
0.642
0.126
Immunocompromised
AIDs Patients;
Organ Transplant
Sensitive Patients; Non-
Based on Hospitalized Cancer
Age Only Therapy Patients
0.99
0.99
0.99
0.942
0.01
0.01
0.01
0.01
0.01
0.01
Nursing
Home
(age-
adjusted
factor)1
0.0081
1 Assumes all nursing home patients are > 16 years old.
A.1.2 Viral Pathogen Occurrence and Concentration in Source Water
Virus occurrence and virus concentration assumptions for this exposure assessment are
discussed together because both are based on occurrence data from the AWWARF study
(Abbaszadegan et al., 1998), a survey of mainly properly constructed wells; and the EPA/AWWA
study (Lieberman et al., 1995), a survey of known contaminated wells and therefore, assumed to be
representative of poorly constructed wells. Viral occurrence estimates and supporting assumptions are
summarized in Exhibit A-3. EPA estimates that among properly constructed drinking water source
wells, 4.4 percent are contaminated with Type A virus and 4.8 percent are contaminated with Type B
virus. Among poorly constructed wells, it is estimated that 5.5 percent are contaminated with Type A
virus and 6.0 percent are contaminated with Type B virus. Contaminated, poorly-constructed wells are
assumed to have a mean concentration of 29.41 ± 55.7 MPNIU/ 100 L, in comparison with a mean
concentration of 0.356 ± 0.297 MPNIU/ 100L. in contaminated properly-constructed wells. These
occurrence distributions were incorporated as lognormal distributions in the Monte Carlo simulation t
reflect the variability in virus concentration among contaminated systems.
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
A-2
-------
Exhibit A-3. Viral Occurrence and Concentration in Source Water
Well Quality
Properly-Constructed Wells
(83%ofallGWSs)
Poorly-Constructed Wells
(17%ofallGWSs)
Virus Type
Type A virus
Type B virus
Type A virus
Type B virus
Percent of Wells
Contaminated
4.4 percent2
4.8 percent3
5.5 percent4
6.0 percent5
Mean Virus
Concentration
when Contaminated
(MPNIU100 L)1
0.356 ±0.2976'7
0.356±0.2977
29.41 ±55.78'9
29.41 ±55.79
1 Most probable number of infectious units of virus.
2 AWWARF study: The RT-PCR methods detected the presence of rotavirus nucleic acids in 14.6 percent of
wells tested to which the ratio of enterovirus cell-culture to RT-PCR positive wells (0.3) was applied.
3 AWWARF study: The AWWARF study found that 4.8 percent of wells tested were positive forthe presence of
enteroviruses using the Buffalo Green Monkey (BGM) cell culture assay.
4 EPA/AWWA study: Because there are no rotavirus data available from the EPA/AWWA study at this time, it is
assumed that rotavirus (Type A virus) and echovirus (Type B virus) occur in poorly constructed wells in the same
ratio as calculated for properly constructed wells (0.92).
5 EPA/AWWA study: Calculated by dividing the total number of positive BGM cell culture assays by the total
number of assays performed.
6 Because there are no concentration data for rotavirus available from either study, it is assumed that the mean
concentration of Type A virus in properly-constructed wells is the same as for Type B virus.
7 AWWARF study: Range of enterovirus (Type B virus ) concentrations in cell-culture isolates was 0.123 to 1.86
MPNIU/100 L; data are fitted to a lognormal distribution from which the mean and standard deviation are
calculated.
8 Because there are no concentration data for rotavirus available from either study, it is assumed that the mean
concentration of Type A virus in poorly-constructed wells is the same as for Type B virus.
9 EPA/AWWA study: Range of enterovirus concentrations in cell-culture isolates was 0.9 to 212 MPNIU/100 L;
data are fitted to a lognormal distribution from which the mean and standard deviation are calculated.
A-4
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
A.1.3 Pathogen Inactivation in Ground Water Systems and Resulting
Baseline Tap Water Concentrations
Undisinfected Systems
The pathogen concentration in tap water from undisinfected systems is assumed to be the same
as the pathogen concentration in source water.
Disinfecting Systems
Properly operating disinfecting systems are assumed to inactivate 99.99 percent (4 logs) of vira
pathogens, and the concentration of pathogens in tap water from properly operating disinfecting
systems is assumed to be 0.01 percent of the concentration in source water. The model does not
include assumptions regarding pathogen inactivation during treatment failure or distribution system
contamination events because of insufficient data on these events.
A.1.4 Drinking Water Consumption Factors
Daily Intake
Custom daily intake distributions were developed for this analysis using Grapher for Windows
software. These distributions correspond to the age-bins for which morbidity data are available. The
distributions incorporate age-based intake data reported in the 1994 1996 USDA, Continuing
Survey of Food Intakes by Individuals for "all sources, consumers only" and for "community water
supply, all respondents" (cited in EPA 1999a). The USDA "all sources, consumers only" data set
represents the upper bound and the "community water supply, all respondents" data set the lower
bound of daily consumption values considered for this risk assessment. Shown below are the key
characteristics of the two drinking water consumption distributions for the overall population (all ages^
obtained by EPA from the CSFII data.
Mean
lst%-tile
5th %-tile
10th%-tile
25th %-tile
5 0th %-tile
"All Sources, Consumers
Only" (L/day)
1.241
0.047
0.184
0.294
0.584
1.045
Community Water Supply,
All Respondents (L/day)
0.927
0
0
0.032
0.264
0.710
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
A-5
-------
75th%-tile
90th %-tile
95th%-tile
99th %-tile
1.640
"All Sources, Consumers
Only" (L/day)
2.345
2.922
4.808
1.313
Community Water Supply,
All Respondents (L/day)
2.016
2.544
4.242
Annual Exposure
For each type of ground water system (GWS) (i.e., CWS, NTNC, and TNC), the number of
exposure days per year (i.e., the number of days in which tap water is consumed) is estimated. It is
assumed that consumers in CWSs ingest drinking water from those sources 350 days/year while it is
assumed that consumers in NTNC systems ingest drinking water from those sources 250 days/year,
and that drinking water from TNC systems is consumed 15 days/year.
A.1.5 Hazard Identification Parameters
Hazard identification parameters included in the GWR model include: infectivity (the ability of
microorganism to colonize the body of the host); morbidity (the probability of illness given infection);
and mortality (the probability of death given illness). Pathogen hazard assumptions for Type A and B
viruses when ingested in ground water are summarized in Exhibit A-4.
Epidemiological data on viral illness explicitly acquired via the ground water ingestion pathway
are limited to a few CDC surveillance studies of waterborne disease outbreaks in groundwater systems
For this risk assessment, the assumed rates of morbidity and mortality for Type A and Type B viruses
are based on national disease surveillance reports or on reported observations during viral epidemics.
Typically these studies address several routes of transmission (i.e, waterborne, direct contact, and/or
respiratory transmission).
For Type A viruses, the morbidity rate in children < 2 years old is estimated to be 0.88, based
on epidemiological data from national studies summarized by Kapikian and Chanock (1996). Older
children and adults are more likely to be asymptomatic or to experience mild symptoms when infected
with Type A virus because of acquired immunity. The morbidity rate for Type A viral illness among
persons > 2 years old is therefore assumed to be 0.1, a conservative estimate based on several
community studies (Wenman et al., 1979; Foster et al., 1980). For Type B virus, the morbidity rate
also varies by age and is based on a community-wide study by Hall (1980). The results of this study
were consistent with reported morbidity rates from the New York Viral Watch (Kogon, 1969), a large,
multi-year study of viral disease (CEOH 1998).
A-6
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Secondary transmission of Type A and Type B viruses also has been reported. For Type A
viruses, the review by Kapikian and Chanock (1996) suggests that young children regularly transmit
Type A virus to other children and to their adult care givers by direct contact transmission; older
children and adults are not assumed to transmit the virus by this secondary route. For the less infection
Type B virus, a triangular distribution of the secondary morbidity rate for all age groups is based on
epidemiological studies reviewed by Morens et al. (1991).
Mortality due to Type A and Type B viral illnesses is not well characterized. The Type A virus
mortality rate assumed for this risk assessment is based on surveillance of a birth cohort of 3.9 millior
U.S. children, followed for 5 years (Tucker et al. 1998). The observed mortality due to Type A viral
illness in this cohort was 0.00073 percent, i.e., 20 deaths among 2.7 million cases of illness (CEOH,
1998). In the absence of adult Type A mortality data, this rate is assumed for all age groups. For Type
B viruses, a mortality rate of 0.92 percent of infected children < 1 month old is assumed. This rate is
based on studies of infected infants during epidemics of Type B illness in newborn nurseries (CEOH
1998). A mortality rate of 0.041 percent is assumed for all other persons based on calculations by
Stedge et al. (1998) that 2 percent of Type B illnesses are severe and that 2 percent of those seriously
ill will die as a result of illness.
A.1.6 Illnesses and Deaths in Undisinfected and Disinfected (4-log removal)
Systems
The GWR model calculates annual numbers of illnesses and deaths due to source contamination
in undisinfected systems. Exhibit A-5 summarizes the model calculations, which incorporate the mod
assumptions regarding drinking water exposures to pathogens from contaminated sources and health
hazards from Type A and Type B viral pathogens.
A.1.7 Additional Illnesses and Deaths Resulting from Treatment
Failures and Distribution System Contamination
For every baseline waterborne illness in an undisinfected CWS, NTNC, or TNC ground water
system with source contamination, it is estimated that there is an additional 0.43 illness in a ground
water system experiencing source contamination with treatment failure. Also, for every baseline
waterborne illness in an undisinfected CWS or NTNC ground water system with source contamination
(TNC systems do not have distribution systems), an additional 0.32 illness is estimated due to
distribution system contamination.
April5, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis A-7
-------
Exhibit A-4. Hazard Identification of Viral Pathogens for the
GWR Risk Assessment
Pathogen
hazards
Definition
Model
Type A
virus
TypeB
virus
Infectivity
Infectivity is the ability of
the pathogen to colonize
the host; it is defined by
dose-response
relationship.
General model of dose-
response (beta-Poisson):
where:
P(l)= probability of
infection
N = number of
pathogenic viruses
ingested
, = pathogen-specific
rate constants.1
Highly infective virus;
=0.26, »=0.422
Moderately infective
virus;
=0.374, =1873
Morbidity
Primary morbidity is
the probability of
illness given infection;
can vary in sensitive
subgroups.
<2yrsold = 0.884
>2yrs = 0.1 5
< 5yrs old = 0.57
> 5 to 16 years =
0.577
> 16 years = 0.337
Secondary spread
is the probability of
illness given
contact with a
(primary) ill person.
<2yrsold =0.554
>2yrsold = 04
Triangular
distribution (all age
groups), from 0.11
to 0.55; mode -
0.358
Mortality
Mortality is
the probability
of death as a
result of
illness.
7.3 x10-66
< 1 month =
0.00926'9
> 1 month =
0.0004110
1 Regli etal., 1991; 2 Ward etal., 1986; 3 Schiff etal., 1984; 4 Kapikian and Chanock, 1996; 5 Wenman etal.,
1979 and Foster etal., 1980; 6 CEOH 1998; 7 Hall 1980; 8 Morens etal., 1991; 9 rate of mortality given
infection; lOStedge, 1998.
A-8
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit A-5. Summary of GWR Baseline Risk Calculations for
Undisinfected and Disinfected (4-log removal) Systems
Health
Effect
Calculation
Summary
Infection
Mean Individual
Daily
Probability of
Infection
The model calculates the mean individual daily probability of infection
using: the fraction of contaminated wells (virus hit rate); the potentially
exposed population; variable distributions of virus concentration in
undisinfected drinking water (same as source water concentration for
untreated systems) and daily intake; and the rotavirus dose-response rate
constants for Type A virus or the echovirus rate constants for Type B. The
probability of infection given a dose of one of these pathogens:
where: P(l) = probability of infection, N = numbers of pathogenic viruses
ingested, and and = pathogen-specific rate constants. The mean
and standard deviation of the mean are calculated.
Mean Annual
Probability of
Infection
The model calculates the annual probability that an individual in the
population category will be infected at least once:
where: P (I Ann) = the annual probability of infection, P (I) = the mean daily
probability of infection. The cumulative geometric function incorporates the
annual number of days of exposure (i.e., 350 days in CWSs, 250 days in
NTNC systems, and 15 days in TNC systems). The mean and standard
deviation of the mean are calculated for each age group and type and size
of ground water system .
Morbidity
Annual Number
of Illnesses
The annual number of illnesses is the annual number of infections
multiplied by the fraction of infections causing disease (i.e., morbidity
rate), calculated for each age group and type of system. This calculation
incorporates a factor for secondary spread as appropriate. The model
applies the secondary spread factor by age group as follows: secondary
illnesses = (the age-specific number of primary illnesses) * (rate of
secondary spread).
Mortality
Annual Number
of Deaths
Deaths due to Type A virus in all age groups are calculated by multiplying
the annual number of primary and secondary illnesses by the case fatality
rate of 7.3 per million cases of illnesses. For Type B virus, deaths in the
neonate (< 1 month) population are calculated by multiplying the annual
number of primary and secondary infections by 0.92 percent. Deaths in
all other age groups are calculated by multiplying the annual numbers of
primary and secondary illnesses by the composite case fatality rate of
0.041 percent.
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
A-9
-------
A.2. Results of the Risk Calculations
In addition to calculating risks from ingestion of ground water under baseline (i.e., current)
conditions, this assessment performs risk calculations for four GWR options: (1) sanitary surveys only
(2) sanitary surveys with triggered monitoring; (3) a multiple barrier approach; and (4) across the boarc
disinfection treatment. For each option, the model is used to calculate annual numbers of illnesses and
deaths in undisinfected ground water systems. The CDC ratios of outbreak-related illnesses due to
treatment failure of disinfecting systems and distribution system contamination (Exhibit A-6) are used
estimate additional illnesses from those types of contamination. Each GWR option is discussed briefly
below.
A.2.1 Results of the Risk Calculations: Baseline
Estimated annual numbers of illness from ingestion of Type A Virus and Type B virus in public
ground water systems are summarized in Exhibits A-6 through A-9. These exhibits present the
calculated mean and the 10th and 90th percentile estimates of annual illness and deaths for Type A
virus and Type B virus, respectively. Summing the estimates of illness for both types of viruses gives a
combined mean estimate of approximately 168,000 illnesses each year for the model runs using the "A
Sources, Consumers Only" daily intake distributions. The vast majority of these illnesses are
attributable to the highly infective, but less lethal Type A viruses. The combined mean estimated
number of deaths per year given the "All Sources, Consumers Only" daily intake distributions is 15, the
majority of those being due to the more lethal but less infectious Type B viruses.
Exhibit A-6. Estimates of Baseline Type A Viral Illness and Death
("All Sources, Consumers Only" Age-Based Consumption Distributions)
Cause/Source of
Contamination
Source contamination in
undisinfected ground
water systems
Source contamination in
ground water systems
with failed disinfection
Contamination of
distribution systems of
ground water systems
Total
Illnesses per Year
10th
percentile
77,794
33,452
21,615
132,879
Mean
78,172
33,614
21,712
133,498
90th
percentile
78,562
33,781
21,812
134,133
Deaths per Year
10th
percentile
1
0
0
1
Mean
1
0
0
1
90th
percentile
1
0
0
1
A-10
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit A-7. Estimates of Baseline Type B Viral Illness and Death
("All Sources, Consumers Only" Age-Based Consumption Distributions)
Cause/Source of
Contamination
Source contamination in
undisinfected ground
water systems
Source contamination in
ground water systems
with failed disinfection
Contamination of
distribution systems of
ground water systems
Total
Illnesses per Year
10th
percentile
19,019
8,178
5,869
33,062
Mean
19,642
8,446
6,069
34,157
90th
percentile
20,253
8,709
6,265
35,227
Deaths per Year
10th
percentile
8
3
2
14
Mean
8
4
3
14
90th
percentile
8
4
3
15
Exhibit A-8. Estimates of Baseline Type A Viral Illness and Death
("Community Water Supply, All Respondents" Age-Based Consumption
Distributions)
Cause/Source of
Contamination
Source contamination in
undisinfected ground
water systems
Source contamination in
ground water systems
with failed disinfection
Contamination of
distribution systems of
ground water systems
Total
Illnesses per Year
10th
percentile
65,422
28,132
18,220
111,777
Mean
65,878
28,328
18,360
112,566
90th
percentile
66,324
28,519
18,497
113,329
Deaths per Year
10th
percentile
0
0
0
1
Mean
0
0
0
1
90th
percentile
0
0
0
1
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
A-ll
-------
Exhibit A-9. Estimates of Baseline Type B Viral Illness and Death
("Community Water Supply, All Respondents" Age-Based Consumption
Distributions)
Cause/Source of
Contamination
Source contamination in
undisinfected ground
water systems
Source contamination in
ground water systems
with failed disinfection
Contamination of
distribution systems of
ground water systems
Total
Illnesses per Year
10th
percentile
14,622
6,287
4,516
25,425
Mean
15,107
6,496
4,671
26,273
90th
percentile
15,587
6,703
4,825
27,112
Deaths per Year
10th
percentile
6
3
2
11
Mean
6
3
2
11
90th
percentile
6
3
2
11
A.2.2 Results of the Risk Calculations: Option 1-- Sanitary Survey-Only
Exhibits A-l 0 through A-l 3 summarize the estimated annual numbers of illness from ingestion of
Type A virus and Type B virus remaining after implementation of the GWR sanitary survey requirement.
The difference between these estimates and those for the baseline is the expected reduction in death and
illness that would result from implementation of sanitary surveysalone. Given the "All Sources, Consumers
Only" consumption distributions, this option is estimated to reduce the mean number of illnesses by over
13,500 illnesses eachyear in comparison with thebaseline. This option is also estimated to reduce the
mean number of deaths resulting from waterborne illness by one a year.
A-12
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit A-10. Estimates of Type A Viral Illness and Death after
Option 1Sanitary Survey Only
("All Sources, Consumers Only" Age-Based Consumption Distributions)
Cause/Source of
Contamination
Source contamination in
undisinfected ground
water systems
Source contamination in
ground water systems
with failed disinfection
Contamination of
distribution systems of
ground water systems
Total
Illnesses per Year
10th
percentile
70,503
28,451
17,156
118,601
Mean
71,600
32,352
18,989
122,941
90th
percentile
72,686
36,168
20,798
127,194
Deaths per Year
10th
percentile
1
0
0
1
Mean
1
0
0
1
90th
percentile
1
0
0
1
Exhibit A-11. Estimates of Type B Viral Illness and Death after
Option 1Sanitary Survey Only
("All Sources, Consumers Only" Age-Based Consumption Distributions)
Cause/Source of
Contamination
Source contamination in
undisinfected ground
water systems
Source contamination in
ground water systems
with failed disinfection
Contamination of
distribution systems of
ground water systems
Total
Illnesses per Year
10th
percentile
17,357
6,898
4,775
29,843
Mean
17,977
7,855
5,311
31,143
90th
percentile
18,612
8,816
5,859
32,440
Deaths per Year
10th
percentile
7
3
2
13
Mean
8
3
2
13
90th
percentile
8
4
2
14
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
A-I:
-------
Exhibit A-12. Estimates of Type A Viral Illness and Death after
Option 1Sanitary Survey Only
("Community Water Supply, All Respondents" Age-Based Consumption
Distributions)
Cause/Source of
Contamination
Source contamination in
undisinfected ground
water systems
Source contamination in
ground water systems
with failed disinfection
Contamination of
distribution systems of
ground water systems
Total
Illnesses per Year
10th
percentile
59,381
24,009
14,512
100,023
Mean
60,321
27,255
16,057
103,633
90th
percentile
61,259
30,507
17,632
107,185
Deaths per Year
10th
percentile
0
0
0
1
Mean
0
0
0
1
90th
percentile
0
0
0
1
Exhibit A-13. Estimates of Type B Viral Illness and Death after
Option 1Sanitary Survey Only
("Community Water Supply, All Respondents" Age-Based Consumption
Distributions)
Cause/Source of
Contamination
Source contamination in
undisinfected ground
water systems
Source contamination in
ground water systems
with failed disinfection
Contamination of
distribution systems of
ground water systems
Total
Illnesses per Year
10th
percentile
13,342
5,312
3,671
22,951
Mean
13,819
6,037
4,086
23,942
90th
percentile
14,304
6,780
4,512
24,937
Deaths per Year
10th
percentile
6
2
2
10
Mean
6
3
2
10
90th
percentile
6
3
2
10
A-14
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
A.2.3 Results of the Risk Calculations: Option 2: Sanitary Survey and
Triggered Monitoring
Estimated annual illnesses from ingestion of Type A virus and Type B virus in public ground
water systems given implementation of the Sanitary Survey and Triggered Monitoring Option are
summarized in Exhibits A-14 through A-17. Given the "All Sources, Consumers Only" consumption
distributions, this option is estimated to reduce the mean number of waterborne illnesses by over
83,000 illnesses annually, in comparison with the baseline. The Sanitary Survey and Triggered
Monitoring Option is also estimated to reduce the mean number of deaths resulting from waterborne
illness by about 8 per year, a greater than 50 percent reduction in the baseline.
Exhibit A-14. Estimates of Type A Viral Illness and Death after
Option 2Sanitary Survey and Triggered Monitoring
("All Sources, Consumers Only" Age-Based Consumption Distributions)
Cause/Source of
Contamination
Source contamination in
undisinfected ground
water systems
Source contamination in
ground water systems
with failed disinfection
Contamination of
distribution systems of
ground water systems
Total
Illnesses per Year
10th
percentile
35,860
1,314
17,181
59,621
Mean
42,066
6,121
19,013
67,200
90th
percentile
48,144
10,999
20,843
74,630
Deaths per Year
10th
percentile
0
0
0
0
Mean
0
0
0
0
90th
percentile
0
0
0
1
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
A-15
-------
Exhibit A-15. Estimates of Type B Viral Illness and Death after
Option 2Sanitary Survey and Triggered Monitoring
("All Sources, Consumers Only" Age-Based Consumption Distributions)
Cause/Source of
Contamination
Source contamination in
undisinfected ground
water systems
Source contamination in
ground water systems
with failed disinfection
Contamination of
distribution systems of
ground water systems
Total
Illnesses per Year
10th
percentile
8,866
270
4,769
15,154
Mean
10,465
1,340
5,311
17,115
90th
percentile
12,053
2,397
5,863
19,010
Deaths per Year
10th
percentile
4
0
2
6
Mean
4
1
2
7
90th
percentile
5
1
2
8
Exhibit A-16. Estimates of Type A Viral Illness and Death after
Option 2Sanitary Survey and Triggered Monitoring
("Community Water Supply, All Respondents" Age-Based Consumption
Distributions)
Cause/Source of
Contamination
Source contamination in
undisinfected ground
water systems
Source contamination in
ground water systems with
failed disinfection
Contamination of
distribution systems of
ground water systems
Total
llnesses per Year
10th
percentile
30,146
1,068
14,497
49,970
Mean
35,338
5,147
16,050
56,535
90th
percentile
40,584
9,207
17,597
62,818
Deaths per Year
10th
percentile
0
0
0
0
Mean
0
0
0
0
90th
percentile
0
0
0
0
A-16
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit A-17. Estimates of Type B Viral Illness and Death after
Option 2Sanitary Survey and Triggered Monitoring
("Community Water Supply, All Respondents" Age-Based Consumption
Distributions)
Cause/Source of
Contamination
Source contamination in
undisinfected ground
water systems
Source contamination in
ground water systems
with failed disinfection
Contamination of
distribution systems of
ground water systems
Total
Illnesses per Year
10th
percentile
6,821
205
3,667
11,685
Mean
8,046
1,027
4,086
13,159
90th
percentile
9,285
1840
4,500
14,630
Deaths per Year
10th
percentile
3
0
2
5
Mean
3
0
2
5
90th
percentile
4
1
2
6
A.2.4 Results of the Risk Calculations: Option 3Multi-Barrier Approach
Estimated annual illnesses from ingestion of Type A virus and Type B virus in public ground
water systems after implementation of the Multiple Barrier Option are summarized in Exhibits A-18
through A-21. Given "All Sources, Consumers Only" consumption distibutions, the Multiple Barrier
Option is estimated to reduce the mean number of waterborne viral illnesses by over 96,000 illnesses
each year in comparison with the baseline. This option is also estimated to reduce the mean number of
deaths from waterborne illness by about nine each year.
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
A-17
-------
Exhibit A-18. Estimates of Type A Viral Illness and Death after
Option 3Multi-Barrier Option
("All Sources, Consumers Only" Age-Based Consumption Distributions)
Cause/Source of
Contamination
Source contamination in
undisinfected ground
water systems
Source contamination in
ground water systems
with failed disinfection
Contamination of
distribution systems of
ground water systems
Total
Illnesses per Year
10th
percentile
21,023
1,410
17,177
45,971
Mean
31,161
6,777
19,015
56,953
90th
percentile
41,306
12,241
20,847
67,492
Deaths per Year
10th
percentile
0
0
0
0
Mean
0
0
0
0
90th
percentile
0
0
0
0
Exhibit A-19. Estimates of Type B Viral Illness and Death after
Option 3Multi-Barrier Option
("All Sources, Consumers Only" Age-Based Consumption Distributions)
Cause/Source of
Contamination
Source contamination in
undisinfected ground
water systems
Source contamination in
ground water systems
with failed disinfection
Contamination of
distribution systems of
ground water systems
Total
Illnesses per Year
10th
percentile
5,152
301
4,770
11,830
Mean
7,677
1,484
5,301
14,462
90th
percentile
10,223
2,631
5,848
17,135
Deaths per Year
10th
percentile
2
0
2
5
Mean
3
1
2
6
90th
percentile
4
1
2
7
A-18
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit A-20. Estimates of Type A Viral Illness and Death after
Option 3Multi-Barrier Option
("Community Water Supply, All Respondents" Age-Based Consumption
Distributions)
Cause/Source of
Contamination
Source contamination in
undisinfected ground
water systems
Source contamination in
ground water systems
with failed disinfection
Contamination of
distribution systems of
ground water systems
Total
Illnesses per Year
10th
percentile
17,712
1,139
14,504
38,591
Mean
26,189
5,689
16,065
47,943
90th
percentile
34,666
10,173
17,630
56,995
Deaths per Year
10th
percentile
0
0
0
0
Mean
0
0
0
0
90th
percentile
0
0
0
0
Exhibit A-21. Estimates of Type B Viral Illness and Death after
Option 3Multi-Barrier Option
("Community Water Supply, All Respondents" Age-Based Consumption
Distributions)
Cause/Source of
Contamination
Source contamination in
undisinfected ground
water systems
Source contamination in
ground water systems
with failed disinfection
Contamination of
distribution systems of
ground water systems
Total
Illnesses per Year
10th
percentile
3,933
231
3,665
9,059
Mean
5,907
1,117
4,086
11,110
90th
percentile
7,877
1,989
4,510
13,138
Deaths per Year
10th
percentile
2
0
2
4
Mean
2
0
2
5
90th
percentile
3
1
2
5
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
A-19
-------
A.2.5 Results of the Risk Calculations: Option 4-Across-the-Board Disinfection
and Sanitary Survey
Estimated annual illnesses from ingestion of Type A virus and Type B virus in public ground
water systems after implementation of the Across-the-Board Disinfection and Sanitary Survey Option
are summarized in Exhibits A-22 through A-25. Although all systems would treat ground water under
this option, a few, less frequent disinfection failure and distribution system contamination events each
year would continue to cause a few residual illnesses and deaths in populations served by ground water
systems. Given "All Sources, Consumers Only" consumption distributions, this option is estimated to
reduce the mean number of waterborne viral illnesses by greater than 132,000 per year and the mean
number of deaths by about 12 per year.
Exhibit A-22. Estimates of Type A Viral Illness and Death from Ingestion of
Ground Water after Option 4Across-the-Board Disinfection and Sanitary Survey
("All Sources, Consumers Only" Age-Based Consumption Distributions)
Cause/Source of
Contamination
Source contamination in
undisinfected ground
water systems
Source contamination in
ground water systems
with failed disinfection
Contamination of
distribution systems of
ground water systems
Total
Illnesses per Year
10th
percentile
778
1,769
17,177
21,575
Mean
790
8,662
19,015
28,467
90th
percentile
801
15,655
20,847
35,536
Deaths per Year
10th
percentile
0
0
0
0
Mean
0
0
0
0
90th
percentile
0
0
0
0
A-20
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit A-23. Estimates of Type B Viral Illness and Death from Ingestion of
Ground Water after Option 4Across the Board Disinfection and Sanitary Survey
("All Sources, Consumers Only" Age-Based Consumption Distributions)
Cause/Source of
Contamination
Source contamination in
undisinfected ground
water systems
Source contamination in
ground water systems
with failed disinfection
Contamination of
distribution systems of
ground water systems
Total
Illnesses per Year
10th
percentile
12
367
4,770
5,631
Mean
12
1,797
5,301
7,111
90th
percentile
13
3,204
5,848
8,570
Deaths per Year
10th
percentile
0
0
2
2
Mean
0
1
2
3
90th
percentile
0
1
2
4
Exhibit A-24. Estimates of Type A Viral Illness and Death from Ingestion of
Ground Water after Option 4Across-the-Board Disinfection and Sanitary Survey
("Community Water Supply, All Respondents" Age-Based Consumption
Distributions)
Cause/Source of
Contamination
Source contamination in
undisinfected ground
water systems
Source contamination in
ground water systems
with failed disinfection
Contamination of
distribution systems of
ground water systems
Total
Illnesses per Year
10th
percentile
606
1,452
14,504
18,109
Mean
615
7,261
16,065
23,941
90th
percentile
623
13,081
17,631
29,830
Deaths per Year
10th
percentile
0
0
0
0
Mean
0
0
0
0
90th
percentile
0
0
0
0
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
A-21
-------
Exhibit A-25. Estimates of Type B Viral Illness and Death from Ingestion of
Ground Water after Option 4Across-the-Board Disinfection
("Community Water Supply, All Respondents" Age-Based Consumption
Distributions)
Cause/Source of
Contamination
Source contamination in
undisinfected ground
water systems
Source contamination in
ground water systems
with failed disinfection
Contamination of
distribution systems of
ground water systems
Total
Illnesses per Year
10th
percentile
9
279
3,665
4,340
Mean
9
1,352
4,086
5,447
90th
percentile
10
2,395
4,510
6,554
Deaths per Year
10th
percentile
0
0
2
2
Mean
0
1
2
2
90th
percentile
0
1
2
3
A-22
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Appendix B-1
Results from Benefits Valuation
Using the Upper Bound Drinking Water
Consumption Distribution
(All Sources, Consumers Only)
-------
TABLE OF CONTENTS
B.I Introduction B-l
B.2 Monetization of Health Benefits B-2
B.2.1 Option 1: Sanitary Survey Only B-2
B.2.2 Option 1: Sanitary Survey Only B-3
B.2.3 Option 2: Sanitary Survey and Triggered Monitoring B^
B.2.4 Option 2: Sanitary Survey and Triggered Monitoring B-5
B.2.5 Option 3: Multi-Barrier Approach B-6
B.2.6 Option 3: Multi-Barrier Approach B-7
B.2.7 Option 4: Across-the-Board Disinfection B-8
B.2.8 Option 4: Across-the-Board Disinfection B-9
B.3 Distribution of Health Benefits B-ll
B.3.1 System Size B-ll
B.3.2 System Type B-12
B.3.3 Multi-Barrier Morbidity Distribution by Age B-13
B.3.4 Multi-Barrier Mortality Distribution by Age B-14
B.3.5 Multi-Barrier Age Type A Benefits Distribution by Age B-15
B.3.6 Multi-Barrier Age Type B Benefits Distribution by Age B-16
B.3.7 Multi-Barrier Age Benefits Distribution by Age B-17
B-24 GWRRIA DraftDo Not Cite, Quote, or Distribute March 31, 2000
-------
LIST OF EXHIBITS
Exhibit B-l. Health Benefits from Sanitary Survey Option B-2
Exhibit B-2. Health Effects for Sanitary Survey Option B-3
Exhibit B-3. Health Benefits from Sanitary Survey and Triggered Monitoring Option (million$) B-4
Exhibit B-4. Health Effects of Sanitary Survey and Triggered Monitoring Option B-5
Exhibit B-5. Health Benefits from Multi-Barrier Approach (million$) B-6
Exhibit B-6. Health Effects for Multi-Barrier Approach B-7
Exhibit B-7. Health Benefits from Across-the-Board Disinfection Option (million$) B-8
Exhibit B-9. Acute Health Benefits of the GWR Regulatory Options, by System Size (millionS) B-9
Exhibit B-8. Health Effects of Across-the-Board Disinfection Option B-10
Exhibit B-9. Acute Health Benefits of the GWR Regulatory Options,
by System Size (million$) B-10
Exhibit B-10. Acute Health Benefits of the GWR Regulatory Options,
by System Type (million$) B-12
Exhibit B-ll. Viral Related Health Benefits from Reduced Morbidity
by Illness, System Type, and Victim Age Multi-Barrier Approach (million$) B-13
Exhibit B-12. Viral Related Health Benefits from Reduced Mortality by Illness,
System Type, and Victim Age Multi-Barrier Approach (million$) B-14
Exhibit B-13. Comparison of Selected Age Categories in the U.S. Population to Their
Relative Roles in Type A Morbidity and Mortality Health Benefits
Multi-Barrier Approach B-15
Exhibit B-14. Comparison of Selected Age Categories in the U.S. Population to
Their Relative Roles in Type B Morbidity and Mortality Health Benefits
Multi-Barrier Approach B-16
Exhibit B-15. Total GWR Morbidity and Mortality Health Benefits by Illness Type,
and Victim Age Multi-Barrier Approach B-17
-------
Appendix B-l: Upper Bound Drinking Water Consumption Distribution; All Sources, Consumers Only
B.1 Introduction
Included in this appendix are the results from the Ground Water Rule (GWR) benefits
valuation and include: 1) results of the valuation of the output of the risk modeling and 2) selected
distribution analysis.
Results of the valuation exercise are organized under section B.I and include:
Results for each of the four regulatory scenarios;
For each of the four sets of outputs,
One detailed summary table with cost details for health and immunocompromised populations,
for Type A and B viral and bacterial morbidity and mortality impacts; and
A pie chart indicating the relative contribution of each factor to the total benefits.
Distribution analysis is included in Section B.3 and includes analysis by size, type, and, for the
Multi-Barrier Approach, by age.
April 5, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis B-l
-------
Appendix B-l: Upper Bound Drinking Water Consumption Distribution; All Sources, Consumers Only
B.2 Monetization of Health Benefits
B.2.1 Option 1: Sanitary Survey Only
Exhibit B-l below shows the range of potential health benefits under the Sanitary Survey
Option given the uncertainty in the risk assessment results. Details of the mean, 10th percentile, and
90th percentile of the annual expected value of reduced morbidity and mortality for this regulatory
option is provided according to the immune status of the affected victim population. The overall annua'
health benefits of the Sanitary Survey Option range from a low (10th percentile) of $8.8 million to a hig
(90th percentile) of $57.6 million, with a mean of $32.5 million.
Exhibit B-1.
Health Benefits from Sanitary Survey Option
(million$)
Pathogen Type
Morbidity Benefits
Healthy
Immuno-
compromised
Mortality Benefits
Healthy
Immuno-
compromised
Type A virus
1 Oth percentile
mean
90th percentile
$ 3.0
$ 4.9
$ 7.1
$ 0.5
$ 0.9
$ 1.3
$ 0.3
$ 0.5
$ 0.8
$ 0.0
$ 0.0
$ 0.0
Type B virus
1 Oth percentile
mean
90th percentile
$ 2.1
$ 12.0
$ 22.4
$ 0.1
$ 0.5
$ 0.9
$ 1.4
$ 8.3
$ 15.5
$ 0.0
$ 0.0
$ 0.1
Viral Subtotal
1 Oth percentile
mean
90th percentile
$ 5.0
$ 16.9
$ 29.4
$ 0.6
$ 1.4
$ 2.2
$ 1.7
$ 8.8
$ 16.2
$ 0.0
$ 0.1
$ 0.2
Bacterial Subtotal
1 Oth percentile
mean
90th percentile
$ 1.0
$ 3.4
$ 5.9
$ 0.1
$ 0.3
$ 0.4
$ 0.3
$ 1.8
$ 3.2
$ 0.0
$ 0.0
$ 0.0
GWR Total
1 Oth percentile
TOTAL (mean)
90th percentile
$ 6.8
$ 21.9
$ 37.9
$ 2.0
$ 10.6
$ 19.7
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
B-2
-------
Appendix B-l: Upper Bound Drinking Water Consumption Distribution; All Sources, Consumers Only
B.2.2 Option 1: Sanitary Survey Only
The percentage distribution of these mean health benefits by pathogen type is displayed in
Exhibit B-2 below.
Exhibit B-2.
Health Effects for Sanitary Survey Option
Type A Mortality
2%
Bacterial Morbidity.
11%
Type B Morbidity
38%
Type B Mortality
26%
Bacterial Mortality
""" 5%
Type A Morbidity
18%
Overall, the health benefits associated with the GWR under the Sanitary Survey Option are
attributed to reductions of Type B viral illnesses (38 percent) and deaths (26 percent), compared to
Type A health benefits (18 and 2 percent for Type A illness and death, respectively). Although there
are significantly fewer cases of Type B illness and deaths, it is the more costly illness as discussed in
SectionB.2.3.
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
B-3
-------
Appendix B-l: Upper Bound Drinking Water Consumption Distribution; All Sources, Consumers Only
B.2.3 Option 2: Sanitary Survey and Triggered Monitoring
Exhibit B-3 presents the 10th percentile and mean, and 90th percentiles of the potential health benefits
under the Sanitary Survey and Triggered Monitoring Option. The overall annual health benefits range
from a low (10th percentile) of $147.4 million to a high (90th percentile) of $208.6 million with a meai
of $177.9 million.
Exhibit B-3. Health Benefits from Sanitary Survey and Triggered
Monitoring Option (million$)
Pathogen Type
Morbidity Benefits
Healthy
Immuno-
compromised
Mortality Benefits
Healthy
Immuno-
compromised
Type A virus
1 Oth percentile
mean
90thpercentile
$ 24.6
$ 27.9
$ 31.3
$ 4.4
$ 5.0
$ 5.6
$ 2.6
$ 3.0
$ 3.3
$ 0.0
$ 0.0
$ 0.1
Type B virus
1 Oth percentile
mean
90th percentile
$ 52.6
$ 64.7
$ 76.9
$ 2.2
$ 2.7
$ 3.2
$ 36.2
$ 44.6
$ 53.1
$ 0.1
$ 0.2
$ 0.3
Viral Subtotal
1 Oth percentile
mean
90th percentile
$ 77.2
$ 92.6
$ 108.1
$ 6.6
$ 7.7
$ 8.8
$ 38.8
$ 47.6
$ 56.4
$ 0.2
$ 0.3
$ 0.4
Bacterial Subtotal
1 Oth percentile
mean
90th percentile
$ 15.4
$ 18.5
$ 21.6
$ 1.3
$ 1.5
$ 1.8
$ 7.8
$ 9.5
$ 11.3
$ 0.0
$ 0.1
$ 0.1
GWR Total
1 Oth percentile
TOTAL (mean)
90th percentile
$ 100.6
$ 120.4
$ 140.4
$ 46.8
$ 57.5
$ 68.2
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
B-4
-------
Appendix B-l: Upper Bound Drinking Water Consumption Distribution; All Sources, Consumers Only
B.2.4 Option 2: Sanitary Survey and Triggered Monitoring
The percentage distribution of these mean health benefits by pathogen type is displayed in Exhibit B-4.
As can be seen in this Sanitary Survey and Triggered Monitory Option, the greatest benefit results in a
reduced Type B Morbidity (38 percent) followed by a reduction in Type B Mortality (25 percent).
Exhibit B-4. Health Effects of Sanitary Survey and Triggered
Monitoring Option
Type A Mortality
Type B Mortality
25%
Bacterial Morbidity
11%
Type B Morbidity
38%
Bacterial Mortality
5%
Type A Morbidity
19%
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
B-5
-------
Appendix B-l: Upper Bound Drinking Water Consumption Distribution; All Sources, Consumers Only
B.2.5 Option 3: Multi-Barrier Approach
Exhibit B-5 presents the range of potential health benefits under the Multi-Barrier Approach
given the uncertainty in the risk assessment results, including the mean, 10th percentile, and 90th
percentile of the annual expected value of reduced morbidity and mortality for this regulatory option.
The overall annual health benefits of the Multi-Barrier Approach range from a low (10th percentile) of
$168.5 million to a high (90th percentile) of $241.9 million, with a mean of $205.0 million.
Exhibit B-5. Health Benefits from Multi-Barrier Approach (million$)
Pathogen Type
Morbidity Benefits
Healthy
Immuno-
compromised
Mortality Benefits
Healthy
Immuno-
compromised
Type A virus
1 Oth percentile
mean
90th percentile
$ 27.5
$ 32.2
$ 37.1
$ 5.0
$ 5.8
$ 6.7
$ 2.9
$ 3.4
$ 4.0
$ 0.0
$ 0.1
$ 0.1
Type B virus
1 Oth percentile
mean
90th percentile
$ 60.5
$ 74.5
$ 88.6
$ 2.5
$ 3.1
$ 3.7
$ 41.7
$ 51.4
$ 61.2
$ 0.1
$ 0.3
$ 0.4
Viral Subtotal
1 Oth percentile
mean
90th percentile
$ 88.1
$ 106.8
$ 125.6
$ 7.5
$ 8.9
$ 10.4
$ 44.7
$ 54.9
$ 65.1
$ 0.2
$ 0.3
$ 0.4
Bacterial Subtotal
1 Oth percentile
mean
90th percentile
$ 17.6
$ 21.4
$ 25.1
$ 1.5
$ 1.8
$ 2.1
$ 8.9
$ 11.0
$ 13.0
$ 0.0
$ 0.1
$ 0.1
GWR Total
1 Oth percentile
TOTAL (mean)
90th percentile
$ 114.7
$ 138.8
$ 163.2
$ 53.8
$ 66.2
$ 78.7
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
B-C
-------
Appendix B-l: Upper Bound Drinking Water Consumption Distribution; All Sources, Consumers Only
B.2.6 Option 3: Multi-Barrier Approach
The percentage distribution of these mean health benefits by pathogen type is displayed in
Exhibit B-6. As seen under the Sanitary Survey and Triggered Monitoring Option, the majority of the
overall health benefits of the GWR under the Multi-Barrier Approach are attributed to reductions of
Type A and B viral illnesses, 19 and 38 percent, respectively.
Exhibit B-6. Health Effects for Multi-Barrier Approach
Type A Mortality
2%
Type B Mortality
25%
Bacterial Morbidity
11%
Type B Morbidity
38%
Bacterial Mortality
5%
Type A Morbidity
19%
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
B-7
-------
Appendix B-l: Upper Bound Drinking Water Consumption Distribution; All Sources, Consumers Only
B.2.7 Option 4: Across-the-Board Disinfection
Exhibit B-7 presents the mean, 10th and 90th percentiles of the potential health benefits under
the Across-the-Board Disinfection Option. This regulatory alternative examines the effect of regulatic
should all systems be required to implement treatment practices, assuming that the treatment reduced
annual illnesses and deaths with greater than 99.9 percent effectiveness. Illnesses and deaths, however.
were assumed to still occur given the possibility of treatment failure or distribution system
contamination. Under the Across-the-Board Disinfection Option, the overall annual health benefits
range from a low (10th percentile) of $255.0 million to a high (90th percentile) of $311.1 million, with
mean of $283.1 million.
Exhibit B-7. Health Benefits from Across-the-Board Disinfection
Option (million$)
Pathogen Type
Morbidity Benefits
Healthy
Immuno-
compromised
Mortality Benefits
Healthy
Immuno-
compromised
Type A virus
1 Oth percentile
mean
90th percentile
$ 42.1
$ 45.2
$ 48.3
$ 7.6
$ 8.2
$ 8.7
$ 4.5
$ 4.8
$ 5.2
$ 0.1
$ 0.1
$ 0.1
Type B virus
1 Oth percentile
mean
90th percentile
$ 91.3
$ 102.4
$ 113.5
$ 3.8
$ 4.2
$ 4.7
$ 63.0
$ 70.7
$ 78.4
$ 0.2
$ 0.4
$ 0.5
Viral Subtotal
1 Oth percentile
mean
90th percentile
$ 133.4
$ 147.6
$ 161.8
$ 11.4
$ 12.4
$ 13.4
$ 67.4
$ 75.5
$ 83.5
$ 0.3
$ 0.4
$ 0.5
Bacterial Subtotal
1 Oth percentile
mean
90th percentile
$ 26.7
$ 29.5
$ 32.4
$ 2.3
$ 2.5
$ 2.7
$ 13.5
$ 15.1
$ 16.7
$ 0.1
$ 0.1
$ 0.1
GWR Total
1 Oth percentile
TOTAL (mean)
90th percentile
$ 173.7
$ 192.0
$ 210.2
$ 81.3
$ 91.1
$ 100.9
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
-------
Appendix B-l: Upper Bound Drinking Water Consumption Distribution; All Sources, Consumers Only
B.2.8 Option 4: Across-the-Board Disinfection
The distribution of the annual expected value of reducing morbidity and mortality is shown in
Exhibit B-8. The pattern is identical to that previously seen in the Multi-Barrier Approach: 38 percen!
of the health benefits were attributable to reduced Type B morbidity, 19 percent to reduced Type A
morbidity, 25 percent to reduced Type B mortality, 12 percent and 5 percent to reduced bacterial
morbidity and mortality, respectively, and 2 percent to reduced Type A mortality
Exhibit B-8. Health Effects of Across-the-Board Disinfection Option
Type A Mortality
2%
Type B Mortality
25%
Bacterial Morbidity
11%
Bacterial Mortality
Type B Morbidity
38%
Type A Morbidity
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
B-9
-------
Appendix B-l: Upper Bound Drinking Water Consumption Distribution; All Sources, Consumers Only
B.3 Distribution of Health Benefits
B.3.1 System Size
Exhibit B-9 presents the acute health benefits (i.e., reductions in morbidity and mortality), of
the GWR options, by system size. As can be seen in this exhibit, the greatest overall health benefit
results from the Across-the-Board Disinfection Option, and least with the Sanitary Survey Option.
B.3.2 System Type
Exhibit B-9. Acute Health Benefits of the GWR Regulatory Options,
by System Size (millionS)
MORBIDITY BENEFITS
Size Category
<100
100-500
500-1K
1K-3.3K
3.3K-10K
10K-50K
50K-100K
100K-1M
Total
Sanitary Survey
S 0.6
S 1.7
S 1.3
S 2.8
S 3.3
S 2.2
S 3.2
S 3.1
S 18.2
Sanitary Survey and
Triggered Monitoring
S 4.5
S 10.4
S 7.7
S 15.0
S 17.5
S 11.4
S 17.6
S 16.2
S 100.3
MORTALITY BEiNEFn
Size Category
<100
100-500
500-1K
1K-3.3K
3.3K-10K
10K-50K
50K-100K
100K-1M
Total
Sanitary Survey
$ 0.3
$ 0.8
$ 0.6
$ 1.4
$ 1.6
$ 1.1
$ 1.6
$ 1.5
S 8.9
Sanitary Survey and
Trissered Monitorins
$ 1.9
$ 4.7
$ 3.6
^ 79
3 1.2.
$o
-------
Appendix B-l: Upper Bound Drinking Water Consumption Distribution; All Sources, Consumers Only
Exhibit B-10 presents the acute health benefits of the four GWR regulatory options by system
type. The least health benefits will result from implementing the Sanitary Survey Option.
Exhibit B-10. Acute Health Benefits of the GWR Regulatory Options,
by System Type (million$)
VIRAL MORBIDITY BENEFITS
Size Category
CWS
NTNC
TNC
Total
Sanitary Survey
$ 15.7
$ 2.2
$ 0.3
$ 18.2
Sanitary Survey and
Triggered Monitoring
$ 81.7
$ 12.9
$ 5.7
$ 100.3
Multi-Barrier
$ 94.3
$ 14.8
$ 6.6
$ 115.7
Across-the-Board
Disinfection
$ 129.6
$ 21.5
$ 8.9
$ 160.0
VIRAL MORTALITY BENEFITS
Size Category
CWS
NTNC
TNC
Total
Sanitary Survey
$ 7.7
$ 1.1
$ 0.1
$ 8.9
TOT
Size Category
CWS
NTNC
TNC
Total
Sanitary Survey
$ 4.7
$ 0.7
$ 0.1
$ 5.4
Sanitary Survey and
Triggered Monitoring
$ 40.2
$ 6.1
$ 1.6
$ 47.9
Multi-Barrier
$ 46.3
$ 7.0
$ 1.9
$ 55.2
Across-the-Board
Disinfection
$ 63.8
$ 9.6
$ 2.5
$ 75.9
PAL BACTERIAL BENEFITS
Sanitary Survey and
Triggered Monitoring
$ 24.4
$ 3.8
$ 1.5
$ 29.6
Multi-Barrier
$ 28.1
$ 4.3
$ 1.7
$ 34.2
Across-the-Board
Disinfection
$ 38.7
$ 6.2
$ 2.3
$ 47.2
TOTAL GWR BENEFITS
Size Category
CWS
NTNC
TNC
Total
Sanitary Survey
$ 28.1
$ 3.9
$ 0.5
$ 32.5
Sanitary Survey and
Triggered Monitoring
$ 146.3
$ 22.8
$ 8.8
$ 177.9
Multi-Barrier
$ 168.8
$ 26.1
$ 10.1
$ 205.0
Across-the-Board
Disinfection
$ 232.0
$ 37.3
$ 13.7
$ 283.1
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
B-ll
-------
Appendix B-l: Upper Bound Drinking Water Consumption Distribution; All Sources, Consumers Only
B.3.3 Multi-Barrier Morbidity Distribution by Age
The resulting viral-related health benefits resulting form implementation of the Multi-Barrier
Approach is presented in Exhibit B-l 1. Overall, the reduced Type A virus morbidity will be $4.3
million for TNC systems.
Exhibit B-11. Viral Related Health Benefits from Reduced Morbidity by Illness,
System Type, and Victim Age Multi-Barrier Approach (million$)
Age Group
Less than 5 v.o.
Between 5-16 v.o.
Over 16 y.o.
TOTAL
TYPE A VIRUS
CWS
$ 9.6
$ 1.6
$ 17.6
S 28.8
NTNC
$ 1.6
$ 0.3
$ 3.0
S 4.9
TNC
$ 1.4
$ 0.2
$ 2.7
S 4.3
TYPE B VIRUS
CWS
$ 3.9
$ 9.8
$ 51.8
S 65.5
NTNC
$ 0.6
$ 1.4
$ 7.8
S 9.8
TNC
$ 0.1
$ 0.3
$ 1.8
S 2.3
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
B-12
-------
Appendix B-l: Upper Bound Drinking Water Consumption Distribution; All Sources, Consumers Only
B.3.4 Multi-Barrier Mortality Distribution by Age
The Reduced Mortality benefits as a result of implementing the Multi-Barrier Approach are
presented in Exhibit B-12. The greatest health benefit, $32.7 million, will occur with regard to the
Type B virus and those more than 16 years old.
Exhibit B-12. Viral Related Health Benefits from Reduced Mortality by Illness,
System Type, and Victim Age Multi-Barrier Approach (million$)
Age Group
Less than 5 y.o.
Between 5-16 y.o.
Over 16 y.o.
TOTAL
TYPE A VIRUS
CWS
$ 0.5
$ 0.3
$ 1.9
$ 2.6
NTNC
$ 0.1
$ 0.0
$ 0.3
$ 0.5
TNC
$ 0.1
$ 0.0
$ 0.3
$ 0.4
TYBEB VIRUS
CWS
$ 4.2
$ 6.8
$ 32.7
$ 43.7
NTNC
$ 0.6
$ 1.0
$ 4.9
$ 6.5
TNC
$ 0.1
$ 0.2
$ 1.2
$ 1.5
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
B-I:
-------
Appendix B-l: Upper Bound Drinking Water Consumption Distribution; All Sources, Consumers Only
B.3.5 Multi-Barrier Age Type A Benefits Distribution by Age
Currently, the costs of Type A illness fall heavily on those under two years of age. Therefore,
the benefit of reductions in Type A illness are disproportionately captured by those in this age group.
As demonstrated in Exhibit B-l 3, children under two years of age make up only 2.8 percent of the
U.S. population, while 28.9 percent of the reduction in Type A illness related costs are attributable to
these young children.
Exhibit B-13. Comparison of Selected Age Categories in the U.S.
Population to Their Relative Roles in Type A Morbidity and Mortality
Health Benefits Multi-Barrier Approach
Distribution of Total U.S. Population by Age
over 16 y.o..
76.8%
less than 2 y.o.
2.8%
2-5 y.o.
4.4%
5-16 y.o.
15.9%
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
B-14
-------
Appendix B-l: Upper Bound Drinking Water Consumption Distribution; All Sources, Consumers Only
Distribution of Total Type A Health Benefits by Age
less
over 16y.o.
62.0%
1han2y.o.
28.9%
5-16 y.o.
6.1%
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
B-15
-------
Appendix B-l: Upper Bound Drinking Water Consumption Distribution; All Sources, Consumers Only
B.3.6 Multi-Barrier Age Type B Benefits Distribution by Age
As demonstrated in Exhibit B-14, a similar situation exists with Type B viruses, although to a lesser
degree. Children under five years of age make up only 7.2 percent of the U.S. population, while 7.4
percent of the reduction in Type B illness related costs are attributable to these children.
Exhibit B-14. Comparison of Selected Age Categories in the
U.S. Population to Their Relative Roles in Type B Morbidity and
Mortality Health Benefits Multi-Barrier Approach
Distribution of Total U.S. Population by Age
less than 1 mo.
0.1%
1 mo.-l y.o.
1.3%
over 16 y.o.
76.8%
5-16 y.o.
15.9%
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
B-16
-------
B.3.7 Multi-Barrier Age Benefits Distribution by Age
Exhibit B-15 brings all of the virus and age-related information together to illustrate the
breakdown of health benefits associated with the Multi-Barrier Approach. Two important points are
made in this chart. As discussed earlier, the majority of health benefits are derived from reductions in
Type A and Type B virus morbidity. Also, across virus type, because over 76 percent of the
population is over 16 years of age, most of the benefit of reducing exposure to both Type A and Type
B viruses is captured by people in this age group.
Exhibit B-15. Total GWR Morbidity and Mortality Health Benefits by
Illness Type, and Victim Age Multi-Barrier Approach
-------
Appendix B-2
Results from Benefits Valuation
Using the Lower Bound Drinking Water
Consumption Distribution
(Community Water Supply, All Respondents)
-------
TABLE OF CONTENTS
B.I Introduction B-l
B.2 Monetization of Health Benefits B-2
B.2.1 Option 1: Sanitary Survey Only B-2
B.2.2 Option 1: Sanitary Survey Only B-3
B.2.3 Option 2: Sanitary Survey and Triggered Monitoring B-4
B.2.4 Option 2: Sanitary Survey and Triggered Monitoring B-5
B.2.5 Option 3: Multi-Barrier Approach B-6
B.2.6 Option 3: Multi-Barrier Approach B-7
B.2.7 Option 4: Across-the-Board Disinfection B-8
B.2.8 Option 4: Across-the-Board Disinfection B-9
B.3 Distribution of Health Benefits B-ll
B.3.1 System Size B-ll
B.3.2 System Type B-12
B.3.3 Multi-Barrier Morbidity Distribution by Age B-13
B.3.4 Multi-Barrier Mortality Distribution by Age B-14
B.3.5 Multi-Barrier Age Type A Benefits Distribution by Age B-15
B.3.6 Multi-Barrier Age Type B Benefits Distribution by Age B-16
B.3.7 Multi-Barrier Age Benefits Distribution by Age B-17
April 5, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis
-------
Appendix B-l: Upper Bound Drinking Water Consumption Distribution; All Sources, Consumers Only
LIST OF EXHIBITS
Exhibit B-l. Health Benefits from Sanitary Survey Option B-2
Exhibit B-2. Health Effects for Sanitary Survey Option B3
Exhibit B3. Health Benefits from Sanitary Survey
and Triggered Monitoring Option (million$) B^
Exhibit B-4. Health Effects of Sanitary Survey and Triggered Monitoring Option B-5
Exhibit B-5. Health Benefits from Multi-Barrier Approach (millionS) B-6
Exhibit B-6. Health Effects for Multi-Barrier Approach B-7
Exhibit B-7. Health Benefits from Across-the-Board Disinfection Option (million$) B-8
Exhibit B-9. Acute Health Benefits of the GWR Regulatory Options,
by System Size (million$) B-9
Exhibit B-8. Health Effects of Across-the-Board Disinfection Option B-10
Exhibit B-9. Acute Health Benefits of the GWR Regulatory Options,
by System Size (million$) B-10
Exhibit B-10. Acute Health Benefits of the GWR Regulatory Options,
by System Type (million$) B-12
Exhibit B-ll. Viral Related Health Benefits from Reduced Morbidity by Illness,
System Type, and Victim Age Multi-Barrier Approach (millionS) B-l 3
Exhibit B-12. Viral Related Health Benefits from Reduced Mortality by Illness,
System Type, and Victim Age Multi-Barrier Approach (million$) B-14
Exhibit B-13. Comparison of Selected Age Categories in the U.S. Population to Their
Relative Roles in Type A Morbidity and Mortality Health Benefits
Multi-Barrier Approach B-15
Exhibit B-14. Comparison of Selected Age Categories in the U.S. Population to
Their Relative Roles in Type B Morbidity and Mortality Health Benefits
Multi-Barrier Approach B-16
Exhibit B-15. Total GWR Morbidity and Mortality Health Benefits by Illness Type,
and Victim Age Multi-Barrier Approach B-l 7
Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
Appendix B-2: Lower Bound Drinking Water Consumption Distribution; Community Water Supply, All Respondents
B.1 Introduction
Included in this appendix are the results from the Ground Water Rule (GWR) benefits
valuation and include: 1) results of the valuation of the output of the risk modeling and 2) selected
distribution analysis.
Results of the valuation exercise are organized under section B.I and include:
Results for each of the four regulatory scenarios;
For each of the four sets of outputs,
One detailed summary table with cost details for health and immunocompromised populations,
for Type A and B viral and bacterial morbidity and mortality impacts; and
A pie chart indicating the relative contribution of each factor to the total benefits.
Distribution analysis is included in Section B.3 and includes analysis by size, type, and, for the
Multi-Barrier Approach, by age.
April5, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis B-3
-------
Appendix B-2: Lower Bound Drinking Water Consumption Distribution; Community Water Supply, All Respondents
B.2 Monetization of Health Benefits
B.2.1 Option 1: Sanitary Survey Only
Exhibit B-l below shows the range of potential health benefits under the Sanitary Survey
Option given the uncertainty in the risk assessment results. Details of the mean, 10th percentile, and
90th percentile of the annual expected value of reduced morbidity and mortality for this regulatory
option is provided according to the immune status of the affected victim population. The overall annual
health benefits of the Sanitary Survey Option range from a low (10th percentile) of $7.2 million to a hig
(90th percentile) of $45.0 million, with a mean of $25.6 million.
Exhibit B-1.
Health Benefits from Sanitary Survey Option
(million$)
Pathogen Type
Morbidity Benefits
Healthy
Immuno-
compromised
Mortality Benefits
Healthy
Immuno-
compromised
Type A virus
1 Oth percentile
mean
90th percentile
$ 2.5
$ 4.1
$ 5.9
$ 0.5
$ 0.7
$ 1.1
$ 0.3
$ 0.4
$ 0.6
$ 0.0
$ 0.0
$ 0.0
Type B virus
1 Oth percentile
mean
90th percentile
$ 1.6
$ 9.3
$ 17.2
$ 0.1
$ 0.4
$ 0.7
$ 1.1
$ 6.4
$ 11.9
$ 0.0
$ 0.0
$ 0.1
Viral Subtotal
1 Oth percentile
mean
90th percentile
$ 4.1
$ 13.4
$ 23.1
$ 0.5
$ 1.1
$ 1.8
$ 1.3
$ 6.8
$ 12.5
$ 0.0
$ 0.0
$ 0.1
Bacterial Subtotal
1 Oth percentile
mean
90th percentile
$ 0.8
$ 2.7
$ 4.6
$ 0.1
$ 0.2
$ 0.4
$ 0.3
$ 1.4
$ 2.5
$ 0.0
$ 0.0
$ 0.0
GWR Total
1 Oth percentile
TOTAL (mean)
90th percentile
$ 5.6
$ 17.4
$ 29.9
$ 1.6
$ 8.2
$ 15.1
B-4
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Appendix B-2: Lower Bound Drinking Water Consumption Distribution; Community Water Supply, All Respondents
B.2.2 Option 1: Sanitary Survey Only
The percentage distribution of these mean health benefits by pathogen type is displayed in
Exhibit B-2 below.
Exhibit B-2.
Health Effects for Sanitary Survey Option
Type A Mortality
2%
Bacterial Morbidity
11%
Type B Morbidity
38%
Type B Mortality
25%
Bacterial Mortality
5%
Type A Morbidity
19%
Overall, the health benefits associated with the GWR under the Sanitary Survey Option are
attributed to reductions of Type B viral illnesses (38 percent) and deaths (25 percent), compared to
Type A health benefits (nineteen and two percent for Type A illness and death, respectively). Although
there are significantly fewer cases of Type B illness and deaths, it is the more costly illness as discuss*
in Section B.2.3.
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
B-5
-------
Appendix B-2: Lower Bound Drinking Water Consumption Distribution; Community Water Supply, All Respondents
B.2.3 Option 2: Sanitary Survey and Triggered Monitoring
Exhibit B-3 presents the 10th percentile and mean, and 90th percentiles of the potential health benefits
under the Sanitary Survey and Triggered Monitoring Option. The overall annual health benefits range
from a low (10th percentile) of $115.9 million to a high (90th percentile) of $163.4 million with a meai
of $139.6 million.
Exhibit B-3. Health Benefits from Sanitary Survey and Triggered
Monitoring Option (million$)
Pathogen Type
Morbidity Benefits
Healthy
Immuno-
compromised
Mortality Benefits
Healthy
Immuno-
compromised
Type A virus
1 Oth percentile
mean
90th percentile
$ 20.7
$ 23.4
$ 26.2
$ 3.8
$ 4.3
$ 4.8
$ 2.2
$ 2.5
$ 2.8
$ 0.0
$ 0.0
$ 0.0
Type B virus
1 Oth percentile
mean
90th percentile
$ 40.4
$ 49.7
$ 59.1
$ 1.7
$ 2.1
$ 2.4
$ 27.7
$ 34.1
$ 40.6
$ 0.1
$ 0.2
$ 0.3
Viral Subtotal
1 Oth percentile
mean
90thpercentile
$ 61.1
$ 73.2
$ 85.3
$ 5.4
$ 6.3
$ 7.2
$ 29.9
$ 36.6
$ 43.4
$ 0.1
$ 0.2
$ 0.3
Bacterial Subtotal
1 Oth percentile
mean
90th percentile
$ 12.2
$ 14.6
$ 17.1
$ 1.1
$ 1.3
$ 1.4
$ 6.0
$ 7.3
$ 8.7
$ 0.0
$ 0.0
$ 0.1
GWR Total
1 Oth percentile
TOTAL (mean)
90th percentile
$ 79.9
$ 95.4
$ 111.0
$ 36.0
$ 44.2
$ 52.4
B-6
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Appendix B-2: Lower Bound Drinking Water Consumption Distribution; Community Water Supply, All Respondents
B.2.4 Option 2: Sanitary Survey and Triggered Monitoring
The percentage distribution of these mean health benefits by pathogen type is displayed in Exhibit B-4.
As can be seen in this Sanitary Survey and Triggered Monitory Option, the greatest benefit results in a
reduced Type B Morbidity (37 percent) followed by a reduction in Type B Mortality (25 percent).
Exhibit B-4. Health Effects of Sanitary Survey and Triggered
Monitoring Option
Type A Mortality
2%
Type B Mortality
Bacterial Morbidity
11%
Type B Morbidity.
37%
Bacterial Mortality
5%
Type A Morbidity
20%
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
B-7
-------
Appendix B-2: Lower Bound Drinking Water Consumption Distribution; Community Water Supply, All Respondents
B.2.5 Option 3: Multi-Barrier Approach
Exhibit B-5 presents the range of potential health benefits under the Multi-Barrier Approach
given the uncertainty in the risk assessment results, including the mean, 10th percentile, and 90th
percentile of the annual expected value of reduced morbidity and mortality for this regulatory option.
The overall annual health benefits of the Multi-Barrier Approach range from a low (10th percentile) of
$132.7 million to a high (90th percentile) of $189.4 million, with a mean of $161.0 million.
Exhibit B-5. Health Benefits from Multi-Barrier Approach (million$)
Pathogen Type
Morbidity Benefits
Healthy
Immuno-
compromised
Mortality Benefits
Healthy
Immuno-
compromised
Type A virus
1 Oth percentile
mean
90thpercentile
$ 23.2
$ 27.1
$ 31.1
$ 4.2
$ 4.9
$ 5.6
$ 2.5
$ 2.9
$ 3.3
$ 0.0
$ 0.0
$ 0.1
Type B virus
1 Oth percentile
mean
90th percentile
$ 46.6
$ 57.4
$ 68.0
$ 1.9
$ 2.4
$ 2.8
$ 32.0
$ 39.4
$ 46.7
$ 0.1
$ 0.2
$ 0.3
Viral Subtotal
1 Oth percentile
mean
90th percentile
$ 69.8
$ 84.4
$ 99.0
$ 6.1
$ 7.3
$ 8.5
$ 34.4
$ 42.3
$ 50.0
$ 0.1
$ 0.2
$ 0.3
Bacterial Subtotal
1 Oth percentile
mean
90th percentile
$ 14.0
$ 16.9
$ 19.8
$ 1.2
$ 1.5
$ 1.7
$ 6.9
$ 8.5
$ 10.0
$ 0.0
$ 0.0
$ 0.1
GWR Total
1 Oth percentile
TOTAL (mean)
90th percentile
$ 91.2
$ 110.0
$ 129.0
$ 41.5
$ 51.0
$ 60.4
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Appendix B-2: Lower Bound Drinking Water Consumption Distribution; Community Water Supply, All Respondents
B.2.6 Option 3: Multi-Barrier Approach
The percentage distribution of these mean health benefits by pathogen type is displayed in
Exhibit B-6. As seen under the Sanitary Survey and Triggered Monitoring Option, the majority of the
overall health benefits of the GWR under the Multi-Barrier Approach are attributed to reductions of
Type A and B viral illnesses, 20 and 37 percent, respectively.
Exhibit B-6. Health Effects for Multi-Barrier Approach
Type A Mortality
2%
Type B Mortality
25%
Bacterial Morbidity
11%
Type B Morbidity
37%
Bacterial Mortality
5%
Type A Morbidity
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
B-9
-------
Appendix B-2: Lower Bound Drinking Water Consumption Distribution; Community Water Supply, All Respondents
B.2.7 Option 4: Across-the-Board Disinfection
Exhibit B-7 presents the mean, 10th and 90th percentiles of the potential health benefits under
the Across-the-Board Disinfection Option. This regulatory alternative examines the effect of regulatic
should all systems be required to implement treatment practices, assuming that the treatment reduced
annual illnesses and deaths with greater than 99.9 percent effectiveness. Illnesses and deaths, however.
were assumed to still occur given the possibility of treatment failure or distribution system
contamination. Under the Across-the-Board Disinfection Option, the overall annual health benefits
range from alow (10th percentile) of $201.1 million to a high (90th percentile) of $244.4 million, with
mean of $222.7 million.
Exhibit B-7. Health Benefits from Across-the-Board Disinfection
Option (million$)
Pathogen Type
Morbidity Benefits
Healthy
Immuno-
compromised
Mortality Benefits
Healthy
Immuno-
compromised
Type A virus
1 Oth percentile
mean
90th percentile
$ 35.4
$ 37.9
$ 40.5
$ 6.4
$ 6.9
$ 7.4
$ 3.8
$ 4.1
$ 4.3
$ 0.1
$ 0.1
$ 0.1
Type B virus
1 Oth percentile
mean
90th percentile
$ 70.5
$ 78.9
$ 87.4
$ 2.9
$ 3.3
$ 3.6
$ 48.3
$ 54.2
$ 60.0
$ 0.2
$ 0.3
$ 0.4
Viral Subtotal
1 Oth percentile
mean
90thpercentile
$ 105.9
$ 116.9
$ 127.9
$ 9.3
$ 10.2
$ 11.0
$ 52.1
$ 58.2
$ 64.4
$ 0.2
$ 0.3
$ 0.4
Bacterial Subtotal
1 Oth percentile
mean
90th percentile
$ 21.2
$ 23.4
$ 25.6
$ 1.9
$ 2.0
$ 2.2
$ 10.4
$ 11.6
$ 12.9
$ 0.0
$ 0.1
$ 0.1
GWR Total
1 Oth percentile
TOTAL (mean)
90th percentile
$ 138.3
$ 152.4
$ 166.7
$ 62.8
$ 70.3
$ 77.7
B-10
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Appendix B-2: Lower Bound Drinking Water Consumption Distribution; Community Water Supply, All Respondents
B.2.8 Option 4: Across-the-Board Disinfection
The distribution of the annual expected value of reducing morbidity and mortality is shown in
Exhibit B-8. The pattern is similar to that previously seen in the Multi-Barrier Approach: 38 percent <
the health benefits were attributable to reduced Type B morbidity, 20 percent to reduced Type A
morbidity, 24 percent to reduced Type B mortality, eleven percent and five percent to reduced
bacterial morbidity and mortality, respectively, and two percent to reduced Type A mortality.
Exhibit B-8. Health Effects of Across-the-Board Disinfection Option
Type A Mortality
2%
Type B Mortality
24%
Bacterial Morbidity
11%
Type B Morbidity
38%
Bacterial Mortality
5%
Type A Morbidity
20%
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
B-ll
-------
Appendix B-2: Lower Bound Drinking Water Consumption Distribution; Community Water Supply, All Respondents
B.3 Distribution of Health Benefits
B.3.1 System Size
Exhibit B-9 presents the acute health benefits (i.e., reductions in morbidity and mortality), of
the GWR options, by system size. As can be seen on this exhibit, the greatest overall health benefit
results from the Across-the-Board Disinfection Option and least with the Sanitary Survey Option.
Exhibit B-9. Acute Health Benefits of the GWR Regulatory
Options, by System Size (million$)
MORBIDITY BENEFITS
Size Category
<100
100-500
500- IK
1K-3.3K
3.3K-10K
10K-50K
50K-100K
100K-1M
Total
Sanitary Survey
$ 0.5
$ 1.3
S 1.1
$ 2.2
S 2.6
$ 1.7
S 2.6
$ 2.4
S 14.5
Sanitary Survey and
Triggered Monitoring
$ 3.6
$ 8.2
S 6.1
$ 11.9
S 13.9
$ 9.0
S 14.0
$ 12.8
S 79.5
Multi-Barrier
$ 4.1
$ 9.5
S 7.0
$ 13.7
S 16.0
$ 10.4
S 16.1
$ 14.9
S 91.7
Across-the-Board
Disinfection
$ 5.7
$ 13.3
S 9.9
$ 19.1
S 22.1
$ 14.3
S 22.2
$ 20.4
S 127.0
MORTALITY BENEFITS
Size Category
<100
100-500
500- IK
1K-3.3K
3.3K-10K
10K-50K
50K-100K
100K-1M
Total
Sanitary Survey
$ 0.2
$ 0.6
$ 0.5
S 1.0
$ 1.2
S 0.8
$ 1.2
$ 1.2
S 6.8
Sanitary Survey and
Triggered Monitoring
$ 1.5
$ 3.6
$ 2.8
S 5.6
$ 6.6
S 4.2
$ 6.6
$ 6.0
S 36.8
Multi-Barrier
$ 1.7
$ 4.1
$ 3.2
S 6.4
$ 7.6
S 4.9
$ 7.6
$ 7.0
S 42.5
Across-the-Board
Disinfection
$ 2.3
$ 5.7
$ 4.4
S 8.9
$ 10.4
S 6.7
$ 10.5
$ 9.6
S 58.6
TOTAL BENEFITS
Size Category
<100
100-500
500- IK
1K-3.3K
3.3K-10K
10K-50K
50K-100K
100K-1M
Total
Sanitary Survey
$ 0.7
$ 2.0
$ 1.6
S 3.3
$ 3.9
S 2.5
$ 3.8
$ 3.6
S 21.3
Sanitary Survey and
Trissered Monitorins
$ 5.1
$ 11.8
$ 8.9
S 17.5
$ 20.4
S 13.2
$ 20.6
$ 18.9
S 116.3
Multi-Barrier
$ 5.8
$ 13.6
$ 10.2
S 20.2
$ 23.6
S 15.3
$ 23.7
$ 21.8
S 134.2
Across-the-Board
Disinfection
$ 8.0
$ 19.0
$ 14.3
S 28.0
$ 32.5
S 21.0
$ 32.7
$ 30.0
S 185.6
B-12
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Appendix B-2: Lower Bound Drinking Water Consumption Distribution; Community Water Supply, All Respondents
B.3.2 System Type
Exhibit B-10 presents the acute health benefits of the four GWR regulatory options by system
type. The least health benefits will result from implementing the Sanitary Survey Option.
Exhibit B-10. Acute Health Benefits of the GWR Regulatory Options,
by System Type (million$)
VIRAL MORBIDITY BENEFITS
Size Category
CWS
NTNC
TNC
Total
Sanitary Survey
$ 12.5
$ 1.8
$ 0.2
S 14.5
Sanitary Survey and
Triggered Monitoring
$ 64.8
$ 10.2
$ 4.5
S 79.5
Multi-Barrier
$ 74.9
$ 11.7
$ 5.2
S 91.7
Across-the-Board
Disinfection
$ 102.9
$ 17.1
$ 7.0
S 127.0
VIRAL MORTALITY BENEFITS
Size Category
CWS
NTNC
TNC
Total
Sanitary Survey
$ 6.0
$ 0.8
$ 0.1
S 6.8
Sanitary Survey and
Triggered Monitoring
$ 30.9
$ 4.7
$ 1.3
S 36.8
Multi-Barrier
$ 35.7
$ 5.4
$ 1.4
S 42.5
Across-the-Board
Disinfection
$ 49.2
$ 7.4
$ 1.9
S 58.6
TOTAL BACTERIAL BENEFITS
Size Category
CWS
NTNC
TNC
Total
Sanitary Survey
$ 3.7
$ 0.5
$ 0.1
S 4.3
Sanitary Survey and
Triggered Monitoring
$ 19.1
$ 3.0
$ 1.2
S 23.3
Multi-Barrier
$ 22.1
$ 3.4
$ 1.3
S 26.8
Across-the-Board
Disinfection
$ 30.4
$ 4.9
$ 1.8
S 37.1
TOTAL GWR BENEFITS
Size Category
CWS
NTNC
TNC
Total
Sanitary Survey
$ 22.1
$ 3.1
$ 0.4
S 25.6
Sanitary Survey and
Triggered Monitoring
$ 114.8
$ 17.8
$ 7.0
S 139.6
Multi-Barrier
$ 132.7
$ 20.4
$ 7.9
S 161.0
Across-the-Board
Disinfection
$ 182.6
$ 29.4
$ 10.7
S 222.7
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
B-I:
-------
Appendix B-2: Lower Bound Drinking Water Consumption Distribution; Community Water Supply, All Respondents
B.3.3 Multi-Barrier Morbidity Distribution by Age
The resulting viral-related health benefits resulting form implementation of the Multi-Barrier
Approach is presented in Exhibit B-l 1. Overall, the reduced Type A virus morbidity will be $3.5
million for TNC systems.
Exhibit B-11. Viral Related Health Benefits from Reduced Morbidity by Illness,
System Type, and Victim Age Multi-Barrier Approach (million$)
Age Group
Less than 5 y.o.
Between 5-16 y.o.
Over 16 y.o.
TOTAL
TYPE A VIRUS
CWS
$ 8.0
$ 1.4
$ 15.0
S 24.3
NTNC
$ 1.4
$ 0.2
$ 2.6
S 4.2
TNC
$ 1.1
$ 0.2
$ 2.2
S 3.5
TYPE B VIRUS
CWS
$ 2.9
$ 7.5
$ 40.1
S 50.5
NTNC
$ 0.4
$ 1.1
$ 6.0
S 7.5
TNC
$ 0.1
$ 0.2
$ 1.4
S 1.7
B-14
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Appendix B-2: Lower Bound Drinking Water Consumption Distribution; Community Water Supply, All Respondents
B.3.4 Multi-Barrier Mortality Distribution by Age
The Reduced Mortality benefits as a result of implementing the Multi-Barrier Approach are
presented in Exhibit B-12. The greatest health benefit, $25.3 million, will occur with regard to the
Type B virus and those more than 16 years old.
Exhibit B-12. Viral Related Health Benefits from Reduced Mortality by Illness,
System Type, and Victim Age Multi-Barrier Approach (million$)
Age Group
Less than 5 y.o.
Between 5-16 y.o.
Over 16 y.o.
TOTAL
TYPE A VIRUS
CWS
$ 0.4
$ 0.2
$ 1.6
$ 2.2
NTNC
$ 0.1
$ 0.0
$ 0.3
$ 0.4
TNC
$ 0.1
$ 0.0
$ 0.2
$ 0.3
TVBEB VIRUS
CWS
$ 3.0
$ 5.2
$ 25.3
$ 33.5
NTNC
$ 0.4
$ 0.8
$ 3.8
$ 5.0
TNC
$ 0.1
$ 0.2
$ 0.9
$ 1.1
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
B-15
-------
Appendix B-2: Lower Bound Drinking Water Consumption Distribution; Community Water Supply, All Respondents
B.3.5 Multi-Barrier Age Type A Benefits Distribution by Age
Currently, the costs of Type A illness fall heavily on those under two years of age. Therefore,
the benefit of reductions in Type A illness are disproportionately captured by those in this age group.
As demonstrated in Exhibit B-13, children under two years of age make up only 2.8 percent of the
U.S. population, while 28.4 percent of the reduction in Type A illness related costs are attributable to
these young children.
Exhibit B-13. Comparison of Selected Age Categories in the U.S.
Population to Their Relative Roles in Type A Morbidity and Mortality
Health Benefits Multi-Barrier Approach
Distribution of Total U.S. Population by Age
less than 2 y.o.
2.8%
2-5 y.o.
4.4%
5-16 y.o.
15.9%
B-16
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Appendix B-2: Lower Bound Drinking Water Consumption Distribution; Community Water Supply, All Respondents
B.3.6 Multi-Barrier Age Type B Benefits Distribution by Age
As demonstrated in Exhibit B-14, a similar situation exists with Type B viruses, although to a lesser
degree. Children under five years of age make up only 7.2 percent of the U.S. population, while 7.0
percent of the reduction in Type B illness related costs are attributable to these children.
Exhibit B-14. Comparison of Selected Age Categories in the
U.S. Population to Their Relative Roles in Type B Morbidity and
Mortality Health Benefits Multi-Barrier Approach
Distribution of Total U.S. Population by Age
less than 1 mo.
mo.-l y.o.
1.3%
over 16 y.o.
76.8%
5-16 y.o.
15.9%
Distribution of Total Type B Health Benefits by Age
Distributio:
over 16 y.o.
78.1%
less than 1 mo.
1.8%
5-16 y.o.
15.0%
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
B-17
-------
Appendix B-2: Lower Bound Drinking Water Consumption Distribution; Community Water Supply, All Respondents
B.3.7 Multi-Barrier Age Benefits Distribution by Age
Exhibit B-15 brings all of the virus and age-related information together to illustrate the
breakdown of health benefits associated with the Multi-Barrier Approach. Two important points are
made in this chart. As discussed earlier, the majority of health benefits are derived from reductions in
Type A and Type B virus morbidity. Also, across virus type, because over 76 percent of the
population is over 16 years of age, most of the benefit (74 percent) of reducing exposure to both Type
A and Type B viruses is captured by people in this age group.
Exhibit B-15. Total GWR Morbidity and Mortality Health Benefits by
Illness Type, and Victim Age Multi-Barrier Approach
B-18
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Appendix C
Inputs to Cost Model
-------
TABLE OF CONTENTS
C.I Introduction C-l
C.2 Unit Costs and Cost Assumptions C-l
C.2.1 Administrative Costs C-2
C.2.1.1 Unit Costs
C-2
C.2.2 Sanitary Survey C-4
C.2.2.1 Unit Costs C-4
C.2.2.2 Costing Assumptions C-5
C.2.3 Corrective Action for Significant Defects C-6
C.2.3.1 Unit Costs C-7
C.2.3.2 Compliance Forecast C-8
C.2.4 Hydrogeologic Sensitivity Assessment C-10
C.2.4.1 Unit Costs C-10
C.2.4.2 Costing Assumptions C-10
C.2.5 Triggered Source Water Monitoring C-l 1
C.2.5.1 Unit Costs C-ll
C.2.5.2 Costing Assumptions C-ll
C.2.6 Routine Source Water Monitoring C-12
C.2.6.1 Unit Costs C-13
C.2.6.2 Costing Assumptions C-13
C.2.7 Corrective Action for Fecal Contamination C-14
C.2.7.1 Unit Costs C-14
C.2.8 Non-Quantifiable Costs C-l8
C.3 Non-monetary Inputs to Cost Model C-l9
C.4 References C-21
LIST OF EXHIBITS
Exhibit C-l. Components for Risk-Based Regulatory Options C-l
Exhibit C-2. Estimated State Resources Required for GWR Administration C-3
Exhibit C-3. Estimated State Resources Required for GWR Administration C-3
Exhibit C-4. Estimated State Resources Required to
Respond to Source Water Contamination C-4
Exhibit C-5. Estimated Costs to State for Sanitary Surveys (1998$) C-5
Exhibit C-6. Estimated Costs to Systems for Sanitary Surveys (1998$) C-5
Exhibit C-7. Percentage of Significant Defects C-6
Exhibit C-8. Estimated Costs for Correction of Significant Sanitary Defects (1998$) C-8
Exhibit C-9. Estimated Distribution of Corrective
Actions Among Systems With a Significant Deficiency C-9
Exhibit C-10.Estimated Costs for States and
Systems to Perform Hydrogeologic Sensitivity Assessments (1998$) . C-10
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit C-l 1 .Estimated Number of Triggered Source Water Samples Per Year C-12
Exhibit C-12.Estimated Costs for Eliminating Cause of
Contamination or Obtaining Alternate Water Source (1998$) C-15
Exhibit C-14.Estimated Selection of Corrective Action by
Systems with Source Water Contamination C-l7
Exhibit C-15.Non Community System Flows C-19
April 5, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis
-------
C.1 Introduction
This Appendix presents the background national cost estimates for the proposed Ground
Water Rule (GWR). It provides a description of unit costs and underlying assumptions used to prepare
the cost estimates, and the methodology used to compile these assumptions to estimate the cost of the
four proposed GWR options. Additional details may be found in this Technical Background Document
on Cost Modeling for the GWR RIA.
The proposed GWR and other rule options incorporate of rule components that identify and/or
correct conditions that permit fecal contamination to reach ground water system consumer's taps.
Several of these components are included in more than one of the GWR options. The different rule
options include several components, as presented in Exhibit C-l.
Exhibit C-1. Components for Risk-Based Regulatory Options
Rule Scenario
Components
Option 1 :
Sanitary
Survey Only
Option 2: San.
Survey and
Triggered
Monitoring
Monitoring and Assessment
Sanitary Survey
Triggered Monitoring
Sensitivity Assessment
Routine Monitoring
Response and Compliance Monitoring (Treatment Assurance)
Significant Defects
Corrective Action
Compliance Monitoring
Option 3:
Multi Barrier
Approach
Option 4:
Across-the-
Board
Disinfection
Section C.2 presents the unit costs and costing assumptions that EPA has made for each of the
rule components. In addition to the components shown in the exhibit, unit cost and costing assumption
are also provided for administrative costs, costs expected to be incurred by both the States and the
regulated entities. Section C.3 presents additional non-monetary cost model inputs
C.2 Unit Costs and Cost Assumptions
This section presents a summary of EPA's assumptions used to prepare estimates of the
national costs of the proposed GWR and other regulatory options. It contains a description of the
estimates of unit costs (the cost that would be incurred by each State, individual treatment facility or
system) and the predicted actions that systems and States will make to comply with the proposed
GWR.
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
C-l
-------
Only summary unit costs are presented in this document, more detailed descriptions of the
assumptions and methodologies used to develop these cost estimates are presented in the Cost and
Technology Document for the Ground Water Rule (US EPA, 1999a).
C.2.1 Administrative Costs
States will incur administrative costs upon implementation of the GWR. These administrative
costs are not directly required by specific provisions of GWR options, but are necessary for States to
ensure the provisions of the GWR are properly carried out. States will need to allocate time for their
staff to establish and then maintain the programs necessary to comply with the GWR.
C.2.1.1 Unit Costs
Resources are estimated in terms of full-time equivalents (FTEs). EPA has assumed a cost of
$64,480 for one FTE, including overhead and fringe. Time requirements for a variety of State agency
activities and responses are estimated for this RIA.
Exhibit C-2 lists activities required for the State to start the program following promulgation ol
the GWR. Start-up activities include developing and adopting State regulations that meet the Federal
GWR requirements. States must also train their staff and the water system's staff on the new
requirements, and modify their data management systems to track any new information that must be
reported by systems to the State. For the GWR options that include monitoring with a laboratory
method not currently required by the State, the State must devote a portion of its staff time to certifyin
laboratories for the new method.
C2 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
Exhibit C-2. Estimated State Resources Required for GWR Administration
(One Time Start-up Activities)
Administrative Activity
Public Notification
Regulation Adoption and Program
Development
Upgrade Data Management Systems
Initial Lab Certification and Training
System Training and Technical Assistance
Staff Training
Estimated State Resources (FTE)
0.1
0.5
1.3
0.39
1.0
0.23
Estimated Cost
$6,500
$32,200
$83,800
$25,100
$64,500
$14,800
Exhibit C-3 lists the annual resources that a State will require to continue implementation of thi
GWR. On an annual basis, States must coordinate with their particular EPA Region to be certain the
State is consistent with Federal requirements. States must also continue to train State and drinking
water system staffs, maintain laboratories certifications and report system compliance information to 1
Safe Drinking Water Information System (SDWIS).
Exhibit C-3. Estimated State Resources Required for GWR Administration
(Annual Activities)
Administrative Activity
Coordination with EPA
Lab Certification
On-Going Technical Assistance
SDWIS Reporting
Clerical
Supervision
Staff Training
Estimated State Resources (FTE)
0.5
0.5
0.5
0.5
0.2
0.22
0.05
Estimated Cost
$32,200
$32,200
$32,200
$32,200
$12,900
$14,200
$3,200
In addition to the administrative costs of developing and maintaining a program for GWR
compliance, States will be required to spend time responding to ground water sources that are found to
be fecally contaminated. EPA's estimates of the average time required for a single source testing
positive for the presence of a fecal indicator are presented in Exhibit C-4.
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
C-3
-------
Exhibit C-4. Estimated State Resources Required to
Respond to Source Water Contamination
Activity
Review Plans and
Specification
Violation Letter
Data Entry
State Resources for Small (<10,000)
Ground Water System (hours)
16
4
4
State Resources for Large (>10,000)
Ground Water System (hours)
32
4
4
C.2.2 Sanitary Survey
In addition to the increase in scope of the sanitary survey, the GWR options also increase the
frequency that the surveys will be performed. Federal regulations under 40 CFR §141.121 permit
reduction in total coliform sampling for ground water systems serving less than 1,000 in certain cases.
To qualify, a community system must have had a sanitary survey that found the system to be free of
defects in the past five years. Noncommunity ground water systems that have had a sanitary survey
within the first 10 years of rule implementation also qualify.
Based upon these requirements and a review of State regulations, EPA has estimated that on
average, States currently conduct sanitary surveys of community ground water systems once every five
years and noncommunity ground water systems once every 10 years. The frequency of sanitary
surveys under the proposed GWR and options is once every three years for community ground water
systems (with the possible reduction to five years) and once every five years for noncommunity
systems.
C.2.2.1 Unit Costs
Most States already require sanitary surveys, therefore, EPA has estimated the incremental
increase in cost for performing and preparing a sanitary survey which addresses those components
specific to the GWR. Exhibit C-5 presents the cost estimates for State inspectors to perform sanitary
surveys and for system operators to provide the information to meet current sanitary survey
requirements and to meet the requirements of the options for the proposed GWR. These incremental
costs range from as low as $30 per survey for systems serving under 100 individuals to $700 per
survey per system for the largest systems.
C-4
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit C-5. Estimated Costs to State for Sanitary Surveys (1998$)
Cost for
Under current
requirements
With proposed
GWR requirements
Service Population Catec
<100
$190
$220
101-500
$190
$250
501-1,000
$250
$370
1,001-
3,300
$280
$430
3,301-
10,000
$370
$590
jory
10,001-
50,000
$530
$870
50,001-
100,000
$1,050
$1,500
>100,000
$2,200
$2,900
In addition to State costs, ground water systems will incur additional costs to meet the
requirements of the GWR for sanitary surveys. These incremental costs range from as low as $30 per
survey for systems serving under 100 individuals to as much as $700 per survey per system for the
largest systems. These costs are presented in Exhibit C-6.
Exhibit C-6. Estimated Costs to Systems for Sanitary Surveys (1998$)
Cost for
Undercurrent
requirements
With proposed
GWR requirements
Service Population Catec
<100
$80
$110
101-500
$80
$140
501-1,000
$140
$220
1,001-
3,300
$170
$280
3,301-
10,000
$220
$360
jory
10,001-
50,000
$310
$560
50,001-
100,000
$670
$980
>100,000
$1,460
$1,900
C. 2.2.2 Costing Assumptions
EPA assumes that sanitary surveys will be conducted by the State (or primacy
agent) on all noncommunity ground water systems within five years of promulgation of the
Rule. For community ground water systems, EPA has assumed that all of the community
ground water systems that achieve a 4 log inactivation or removal of virus will conduct
sanitary surveys conducted every five years, and the remaining community ground water
systems will conduct sanitary surveys once every three years.
Within the cost analysis model, it is necessary to calculate the incremental
difference in current sanitary surveys versus the proposed sanitary survey. For both the
public water system (PWS) and the State, a current average annual sanitary survey cost is
calculated by multiplying the baseline cost per survey by the number of surveys the PWS
would undergo over the 20-year period of analysis without the GWR (two for nontransient
noncommunity [NTNC] and transient noncommunity [TNC] systems and four for community
water systems [CWSs]), and then dividing by 20.
As shown in Exhibit C-7, EPA estimates that between 11 and 13 percent of the
systems surveyed will be found to have significant defects that will require correction. This
estimate is based upon data from a survey of best management practices in community
ground water systems conducted by the Association of State Drinking Water
Administrators (ASDWA, 1997). ASDWA's questionnaire asked the survey respondents
if the system has any uncorrected significant defects. Over 800 survey responses were
received from community ground water systems in three different Total Coliform Rule
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
C-5
-------
(TCR) compliance categories that included (1) systems with no TCR MCL violations in the
past two years, (2) systems with at least one non-acute TCR-MCL violations within the past
two years, and (3) systems with at least one acute TCR-MCL violation in the past two
years. EPA weighted the responses from systems in each TCR compliance category to
develop the estimates of the national percentage of systems with a significant defect.
Because no other data was available, EPA has assumed that the percentage of
noncommunity ground water systems with significant deficiencies is proportional to the
number obtained by the ASDWA survey for CWSs.
Exhibit C-7. Percentage of Significant Defects
System Type
cws
TNC
NTNC
Service Population Category
<100
13%
12%
12%
101-500
12%
12%
12%
501-1,000
12%
12%
12%
1,001-
3,300
12%
12%
12%
3,301-
10,000
13%
12%
12%
10,001-
50,000
13%
11%
50,001-
100,000
12%
12%
>100,000
12%
11%
C.2.3 Corrective Action for Significant Defects
The primary purpose of conducting sanitary surveys is to identify significant defects
in PWSs for correction. Currently, there are Federal Regulations that provide primacy
agencies with the authority to require that systems correct significant defects. Under the
proposed GWR and other options, a sanitary survey includes, but is not limited to:
a defect in design, operation, or maintenance, or a failure or malfunction of
the sources, treatment, storage, or distribution system that the State
determines to be causing, or has the potential for causing the introduction of
contamination into the water delivered to consumers.
States would be required to define what constitutes a significant deficiency as part
of their primacy package.
All of the GWR options require the primacy agent to notify systems of significant
deficiencies and require the systems to correct the significant defect within 90 days of
notification. Systems that cannot correct the significant defect within 90 days would be
required to submit a schedule to the State or primacy agent for their review and approval.
Systems in consultation with the State may correct the significant deficiency, switch to an
alternate source of water, or disinfect their source to 4 log inactivation. States or primacy
agencies would be required to confirm that a system has corrected its significant
deficiency either by receipt of notification from the system or by conducting an on-site
inspection.
C. 2.3.1 Unit Costs
The costs for correction of significant deficiencies depend almost entirely upon the
nature of the deficiency. Because States have the authority to define significant
C-6
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
deficiencies under the proposed GWR and options, EPA must predict the types of
deficiencies that will be found and corrected as a result of the rule. EPA consulted with
experts from within the Agency and from States to develop a list of deficiencies that are
likely to be identified in sanitary surveys of ground water systems (EPA, 1996 a). The list
indicated three general areas in which defects may be found: the source, the treatment
system, and the distribution system. From this list, EPA developed a representative list of
corrective actions from each of these general areas. The representative corrective actions
are listed below.
Correction of Significant Deficiencies at the Source
replacing a sanitary well seal,
rehabilitating an existing well, and
drilling a new well
Correction of Significant Deficiencies in Treatment Systems
adjusting disinfection chemical feed rate
increasing contact time prior to first customer
Correction of Significant Deficiencies in Distributions System
install backflow prevention device
replace/repair storage tank cover
install security measures at storage tank site
install a redundant booster pump
Exhibit C-8 presents the costs for correcting significant deficiencies. All costs are
one-time expenditures that occur in the year the significant deficiency is found and
corrected, except for adjusting the treatment chemical feed rate, which is an ongoing
operating cost incurred in each year of operation.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis C7
-------
Exhibit C-8. Estimated Costs for Correction of
Significant Sanitary Defects (1998$)
Cost for
Replacing a Sanitary
Well Seal (per well)
Rehabilitating an
Existing Well (per well)
Replace/Repair Cover
on Storage Tank(per
tank)
Install Security
Measures at Tank Site
[per tank)
Install Backflow
Drevention Device
[per connection)
nstall a Redundant
Booster Pump (per
Dump)
Service Population Category
<100
$3,300
$10,700
$2,700
$9,100
$2,100
$5,800
101-500
$3,300
$10,700
$6,000
$9,100
$2,100
$7,400
501-1,000
$3,300
$10,700
$45,000
$9,100
$2,100
$9,400
1,001- 3,300
$3,300
$10,700
$62,000
$9,100
$2,200
$11,000
3,301-
10,000
$3,300
$10,700
$138,000
$12,500
$2,200
$13,000
10,001-
50,000
$3,300
$10,700
$159,000
$12,500
$3,000
$14,900
50,001-
100,000
$3,300
$10,700
$159,000
$12,500
$4,700
$21,300
>1 00,000
$3,300
$10,700
$159,000
$12,500
$4,700
$27,600
C. 2.3.2 Compliance Forecast
The corrections of significant defects that will be undertaken by ground water
systems depend upon the defects defined as significant by States, and the conditions at
the facilities. Both of these factors are unknown. To provide a reasonable estimate of the
bounds of the uncertainty, with respect to the types of defects that will have to be corrected,
EPA has developed estimates of a low cost scenario and a high cost scenario for
correction of significant defects. The low cost scenario assumes a greater percentage of
the systems with significant defects will have defects which are less expensive to correct
(e.g., more systems will have to replace their sanitary well seal than will have to perform a
complete rehabilitation of their well) in comparison to the high cost scenario. The
percentages of systems with significant defects that were assumed to perform the
representative correction in the high cost and low cost scenarios are listed in Exhibit C-9.
C-8
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit C-9. Estimated Distribution of Corrective
Actions Among Systems With a Significant Deficiency
Replacing a Sanitary Well Seal
Rehabilitating an Existing Well
Replace/Repair Cover on Storage Tank
Install Security Measures at Tank Site
Install Backflow Prevention Device
Install a Redundant Booster Pump
Low Cost Scenario
50%
35%
2%
5%
5%
3%
High Cost Scenario
35%
50%
5%
2%
3%
5%
Each of the regulatory options requires each PWS to correct any significant defect
found during a sanitary survey. The assignment of any significant deficiency is done as a
two-step process within the cost analysis model, and is done independently for each
sanitary survey over the 20-year period of analysis:
First, a PWS is designated as having or not having the potential to have one or
more significant defects, resulting from a single sanitary survey based on a
probability estimate.
Second, each PWS that is predicted to have a significant defect, in a single
sanitary survey, may be assigned one or more of the six potential significant defects
according to a probability distribution. In order to determine the sensitivity of the
cost estimates to these probabilities, two scenarios are considered. Under the first
scenario, PWSs are assigned to low cost significant defects with greater
probability (known as the Low SD scenario), while in the second scenario, PWSs
are assigned to high cost significant defects with greater probability (known as the
High SD scenario).
This process is repeated for each sanitary survey the PWS undertakes over the 20-
year period of analysis.
Since the timing of the sanitary surveys are not known, an average annual PWS
cost of correcting significant defects is calculated by summing the cost of correcting all
significant defects over the 20-year period of analysis and then dividing by 20. The
average annual PWS cost of correcting significant defects includes the cost of developing
engineering plans for submission to the State.1
In addition to the PWS's cost of correcting each significant defect, the State incurs
a cost in reviewing the PWS's engineering plans. The State resources for plan review per
corrected significant defect is shown in Exhibit C-4. Since the timing of the sanitary
1 It is assumed that 97.1 percent of the average annual PWS cost of correcting significant
defects is for the actual correction of the defects, while 2.9 percent is for plan development and
submission to the State for review and approval.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis C-9
-------
surveys are not known, the average annual State cost of reviewing the PWS's engineering
plans is calculated by multiplying the unit cost for plan review by the number of significant
defects corrected over the 20-year period of analysis and then dividing by 20.
C.2.4 Hydrogeologic Sensitivity Assessment
The hydrogeologic sensitivity assessment is a component of the multiple-barrier
option for the GWR. The hydrogeologic sensitivity assessment is performed by States, or
another primacy agent, on each ground water source to determine if the source is sensitive
to microbial contamination and therefore, requires monitoring to insure there is no fecal
contamination.
C.2.4.7 Unit Costs
Costs for the hydrogeologic sensitivity assessment include compiling existing data
reviewing that data in order to determine the sensitivity of an aquifer. EPA has estimated
the time for States to locate existing hydrogeologic data, such as a well construction
record or for a State assessor to inspect and review this data. EPA has assumed that the
hydrogeologic sensitivity assessment will be performed, based on existing data. EPA has
also assumed that the sensitivity assessment will be performed on-site, however this is not
a requirement under the GWR multiple barrier option. Cost estimates for States to
perform hydrogeologic assessment are presented in Exhibit C-10.
Exhibit C-10. Estimated Costs for States and
Systems to Perform Hydrogeologic Sensitivity Assessments (1998$)
Cost for
State cost for performing
hydrogeologic sensitivity
assessment (per system)
Service Population Category
<100
$62
101-500
$124
501-
1,000
$186
1,001-
3,300
$248
3,301-
10,000
$310
10,001-
50,000
$620
50,001-
100,000
$1,178
>100,000
$3,224
C. 2.4.2 Costing Assumptions
EPA assumes that hydrogeologic sensitivity assessments will be performed on all
ground water sources that are not providing 4 log inactivation or removal of virus. EPA
estimates that 15 percent of the systems that are assessed will be determined to be
sensitive. This estimate is based upon data collected for the AWWA Study
(Abbaszadegan et al., 1998; 1999). As part of this study, system operators were asked to
review well construction information and provide detailed information about the
hydrogeologic setting in which their well was located. Of the survey respondents that were
able to provide this information, 15 percent indicated that their wells were located in
unconfined aquifers that were karst, fractured bedrock, or unknown geology. EPA also
assumed that wells in unconfined conditions with unknown hydrogeology will be classified
as sensitive hydrogeology (i.e., sensitive to fecal contamination).
C-10
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
C.2.5 Triggered Source Water Monitoring
Triggered source water monitoring is a component of two regulatory options.
Triggered monitoring requires collection and analysis of samples at the sources of the
ground water systems following the detection of total coliform (TC) in one or more samples
collected for compliance with the Total Coliform Rule. States have the option of requiring
the triggered source water samples to be tested for the presence of one of three fecal
indicators; fecal coliform/E. coli, enterococci, or male specific coliphage. If a system
detects the fecal indicator at its source, then the system must take a corrective action to
either eliminate the contamination, obtain a new uncontaminated source, or install
treatment to achieve a 4 log removal or inactivation of virus.
C.2.5.7 Unit Costs
For the purpose of the cost analysis, EPA has assumed that States will select E.
coli as the indicator of contamination for analysis. States, systems, and laboratories have
much greater familiarity with this method in comparison to the other available methods.
Fecal coliform/E. coli analysis is already required under the Total Coliform Rule. EPA has
estimated that the cost for triggered monitoring is $53 per sample. This cost assumes a
$25 cost for performing laboratory analysis. This is based upon common commercial
laboratory costs and one hour of the system operators time (at an estimated cost of $28
per hour) to collect the sample, arrange for delivery to the laboratory, and to review the
results of the analysis. EPA has assumed that all wells are equipped with existing taps for
sampling prior to the application of any treatment chemicals. Therefore, no additional
costs are assumed for installation of a tap or re-piping of wells to permit sampling.
C. 2.5.2 Costing Assumptions
Two compliance estimates are necessary to develop cost estimates associated
with triggered monitoring; the frequency with which systems will have to perform triggered
monitoring and the number of systems that are expected to test positive for the fecal
indicator.
EPA estimated the probability of a ground water system's total coliform sample
testing positive as a part of its regulatory impact analysis for the Total Coliform Rule (EPA,
1989). These estimates varied based upon the size of the system. EPA believes that
these probabilities are a reasonable estimate of current conditions. EPA calculated the
frequency of total coliform positives per year per system by multiplying the number of TC
samples required per year by the probability of a TC positive. The results are the
estimated number of triggered monitoring samples per year, presented in Exhibit C-11.
The option allows States to waive the triggered monitoring requirements if a PWS
demonstrates that the total coliform contamination is not source water related. This
analysis assumes that the probability of a PWS receiving this waiver for a single entry
point is 10 percent. Also, a one-time repeat sampling waiver exists for both triggered
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis Cll
-------
monitoring and routine monitoring. Once a PWS finds a single positive sample, they may
take five repeat samples, and if all five repeat samples are negative, the source water is
considered not to be contaminated. For the purposes of this analysis, it is assumed that
all PWSs that have a positive source water sample will make use of this one-time
sampling waiver.
Exhibit C-11. Estimated Number of Triggered Source Water Samples Per Year
System Type
cws
TNC
NTNC
Service Population Category
<100
0-3
0-3
0-3
101-500
0-3
0-3
0-3
501-
1,000
0-3
0-3
0-3
1,001-
3,300
0-3
0-3
0-3
3,301-
10,000
0-4
0-3
0-3
10,001-
50,000
1-4
50,001-
100,000
4-7
>100,000
7-22
Since it is assumed that all contaminated entry points will be discovered in the first
year, the entry points with no source water positive samples, or those with a single positive
sample and five negative follow-up samples in the first year, will continue to undertake
triggered monitoring sampling for the remainder of the analysis period. The number of
samples they will take each year is assigned using a uniform distribution based on the
number of expected total coliform violations they might have per year (see Exhibit C-11).
It is assumed that 8.83 percent2 of these entry points will have a single source water
positive sample, and will take five repeat samples. None of these will have a positive
repeat sample.
EPA has estimated the number of systems that will test positive in triggered
monitoring by reviewing indicator occurrence data. The AWWARF study (Abbazadegan et
al., 1998; 1999) found enteroccocci bacteria in nine percent of the wells sampled. Wells
selected for the study were representative of hydrogeologic conditions for public water
supply wells in the United States, and most wells in the study were only sampled once.
EPA has determined that the enterococci occurrence from the AWWARF study provides
the best estimate of the percentage of wells that will be found to test positive for the
presence of a fecal indicator in triggered source water sampling, and therefore, has
assumed that nine percent of the systems tested will be found to contain fecal
contamination.
C.2.6 Routine Source Water Monitoring
Routine source water monitoring is a component of the multiple barrier option for
the GWR. Routine source water monitoring involves monthly sampling of those ground
2 (Probability of not having a single positive result in year l)/(Probability of not having a repeat
positive result in year one) x (Probability of having a single positive result in year one) = (Probability (
having a single positive result in years 2-20). ((1-0.09)7(1-0.09*0.8))*0.09 = 0.883.
C-12
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
water sources that are determined to be at high risk for the presence of fecal
contamination. The hydrogeologic sensitivity assessment is used to determine which wells
are at high risk of contamination. High risk wells would be sampled each month and
tested for the presence of fecal contamination using one of three possible indicators; E.
co//, Enteroccci, or male specific coliphage. The State will select the fecal indicator.
Ground water systems with wells that test positive for the presence of a fecal indicator will
be required to take action to correct the contamination (See Section C.2.7), unless the
system is able to obtain a one-time waiver from the State. A waiver could be granted by
the State if the system collects five repeat samples from the well testing positive within 24
hours and the system does not detect the fecal indicator in any of five samples. This
waiver can only be granted once per source.
States may reduce the frequency of monitoring for high risk ground water sources
after one year of monitoring if there are no samples that test positive. States may also
waive source water monitoring altogether after the first year if the State determines that
fecal contamination of the well is highly unlikely.
C.2.6.1 Unit Costs
As with triggered monitoring, EPA has assumed that States will select E. coli as the
indicator of contamination for analysis. States, systems, and laboratories have much
greater familiarity with this method in comparison to the other available methods. Fecal
coliform/E. coli are analysis already required under the Total Coliform Rule. EPA has
estimated that the cost for triggered monitoring is $53 per sample. This cost assumes a
$25 cost for performing laboratory analysis. This is based upon common commercial
laboratory costs and one hour of the system operators' time (at an estimated cost of $28
per hour) to collect the sample, arrange for delivery to the laboratory, and to review the
results of the analysis. EPA has assumed that all wells are equipped with existing taps for
sampling prior to the application of any treatment chemicals. Therefore, no additional
costs are assumed for installation of a tap or re-piping of wells to permit sampling.
C. 2.6.2 Costing Assumptions
EPA has estimated that 15 percent of the sources for ground water systems will be
determined to be sensitive and, therefore subject to routine monitoring. EPA has
determined that the EPA/AWWARF study data (Lieberman et al., 1994; 1999) provides
the best estimate of results which can be expected from routine monitoring. This study
sampled 30 wells that were determined to be vulnerable to contamination based upon
criteria that included hydrogeologic sensitivity. Each of the wells was sampled monthly
over the course of one year for a variety of fecal indicators including E. coli. Fifty percent
(15/30) of the wells in the EPA/AWWARF study tested positive for the presence of E. coli.
However, three of the 15 wells with positive samples only tested positive in one of twelve
samples. Using this data, EPA has estimated that 50 percent of the ground water sources
that are required to perform routine monitoring will detect the fecal indicator. Because of
the high costs associated with corrective actions, EPA has assumed that all systems with
a routine source water positive sample will resample within 24 hours of detecting the fecal
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis C-13
-------
contamination. EPA has estimated that 20 percent of the systems that perform the repeat
sampling will not find fecal indicators in any of the repeat samples and will receive waivers
from the State.
It is assumed that all contaminated entry points will be discovered in the first year,
the entry points with no source water positive samples, or those with a single positive
sample and five negative follow-up samples in the first year, will continue to undertake
routine sampling
once a quarter for the remainder of the period of analysis. It is assumed that 41.67
percent3 of these entry points will have a single source water positive sample, and will take
five repeat samples. None of these should have a positive repeat sample.
C.2.7 Corrective Action for Fecal Contamination
Detection of fecal indicators in the source of undisinfected ground water systems
requires corrective action under the two GWR options. Corrective action includes
eliminating the contamination from the source, obtaining an alternative source of water, or
providing disinfection treatment that achieves 4 log inactivation or removal of viruses.
C.2.7.1 Unit Costs
The costs of the corrective actions vary based upon the corrective action selected
by the system after consultation with the State and based upon the size of the system.
EPA has assumed that a variety of corrective actions could be implemented by ground
water systems that detect fecal contamination within their source waters. The corrective
actions include;
Eliminate Contamination
Rehabilitate an existing well
Remove/relocate existing source of contamination (septic tank)
Construct a new well
Purchase water from a nearby system
3 (Probability of not having a single positive result in year l)/(Probability of not having a repeat
positive result in year one) x (Probability of having a single positive result in year one) = (Probability (
having a single positive result in years 2-20). ((1-0.5)7(1-0.5*0.8))*0.5 = 0.4167.
C14 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
Disinfect Source Water to Achieve a 4 log Inactivation/Removal of Virus
Install and operate chlorine gas disinfection
Install and operate hypochlorination
Install and operate chloramination
Install and operate chlorine dioxide disinfection
Install and operate mixed oxidants disinfection
Install and operate ozonation
Install and operate reverse osmosis filtration
Install and operate ultraviolet disinfection
Costs for each of the corrective actions that eliminate the cause of contamination or
obtain water from an alternative source are presented in Exhibits C-12. Costs have been
developed for each of the non-treatment corrective actions for a range of sizes of systems.
For all of the non-treatment corrective actions, except purchasing water, all costs consist of
capital costs incurred during the first year in which the corrective action is undertaken.
Purchasing water includes both a capital cost for constructing a pipeline to connect the
system to the new supply and the net increased cost which the system will pay to purchase
water from the system. The cost presented subtracts the system's cost of producing its
own water that the system no longer has to incur from the estimated purchase price.
Depending on the corrective actions, there may up to three different cost estimates:
capital cost (the cost of constructing/installing the equipment), replacement cost (cost of
replacing significant components of the system after several years of operation), and
operation and maintenance costs (annual cost of operating equipment and performing
routine maintenance).
Exhibit C-12. Estimated Costs for Eliminating Cause of
Contamination or Obtaining Alternate Water Source (1998$)
Cost for
Rehabilitating an existing
well
[Capital cost)
Relocating an existing
source of contamination
[Capital cost)
Constructing a new well
[Capital cost)
Constructing a pipeline to
enable Purchasing water
(Capital Cost)
Met cost of purchasing
water
($ 71,000 gal)
Service Population Category
<100
$8,700
$15,100
$12,000
$156,500
$1.03
101-500
$10,700
$15,100
$26,900
$156,500
$1.08
501-1,000
$10,700
$15,100
$26,900
$179,400
$0.58
1,001- 3,300
$10,700
$15,100
$26,900
$179,400
$1.32
3,301-
10,000
$10,700
$15,100
$26,900
$219,200
$1.92
10,001-
50,000
$10,700
$15,100
$26,900
$219,200
$1.24
50,001-
100,000
$10,700
$15,100
$26,900
$319,600
$1.28
>1 00,000
$10,700
$15,100
$26,900
$353,300
$0.84
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
C-15
-------
Costs for disinfecting ground water in order to achieve a 4 log inactivation of virus
are presented in Exhibit C-13. Costs are presented by average daily and design flow
instead of by system population served. Exhibit C-13 presents capital costs and annual
operation and maintenance costs for all treatment types, and presents year 10
replacement cost for most treatment technologies. Year 10 replacement costs are
estimated for the systems that will require replacement of a significant component halfway
through the design life of the system.
C. 2.7.2 Compliance Forecast
Selection of the most appropriate corrective action will be made by the system with
the assistance/review of the State. EPA developed estimates of corrective actions that
ground water systems with fecal contamination would undertake to eliminate or treat their
contamination. These estimates considered the current implementation of treatment
types, the cost of the corrective action, and the need for systems to comply with provisions
of the Disinfection Byproducts Rule (DBPR). The portion of systems that will choose
treatment versus non-treatment corrective actions is assumed proportional to the
percentage of systems in each service
Exhibit C-13. Estimated Costs for Disinfection Treatments (1998$)
Cost for
Chlorine gas feed
capital cost
Chlorine gas feed
eplacement cost
/ear 10
Chlorine gas feed
annual O&M cost
Hypochlorite feed
capital cost
Hypochlorite feed
eplacement cost
/ear 10
Hypochlorite annual
O&M cost
Chloramination
system capital cost
Chloramination
system replacement
cost year 10
Chloramination
annual O&M cost
Chlorine Dioxide
System Capital Cost
Average Daily Flow/Design Flow (MGD)
0.007/0.03
$65,000
$11,200
$3,600
$7,000
$2,800
$4,000
$124,000
$22,500
$7,100
$94,000
0.026/0.10
$65,000
$11,200
$3,700
$7,000
$2,800
$5,000
$124,000
$22,500
$7,200
$149,000
0.09/0.30
$59,000
$17,000
$8,900
$45,000
$6,700
$6,000
$124,000
$22,500
$7,300
$169,000
0.21/0.75
$63,000
$18,600
$10,300
$45,000
$6,700
$10,000
$124,000
$22,500
$7,700
$189,000
0.82/2.2
$79,000
$23,200
$12,600
$45,000
$6,700
$27,000
$127,000
$23,100
$10,000
$233,000
3.25/7.8
$140,000
$41,000
$20,200
N/A
N/A
N/A
$202,000
$35,200
$23,100
$387,000
11.2/23.5
$300,000
$101,000
$45,800
N/A
N/A
N/A
$406,000
$115,700
$54,800
$806,000
45/81
$630,000
$230,000
$143,000
N/A
N/A
N/A
$838,000
$244,800
$165,000
$1,637,000
C-16
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit C-13. Estimated Costs for Disinfection Treatments (1998$)
Cost for
Chlorine Dioxide
System replacement
cost year 10
Chlorine Dioxide
annual O&M Cost
Mixed Oxidant
capital cost
Mixed Oxidant
eplacement costs
/ear 10
Mixed Oxidants
annual O&M costs
Ozonation Capital
Cost
Ozonation Annual
O&M cost
Reverse Osmosis
Capital Cost
Reverse Osmosis
Annual O&M Cost
Ultraviolet
Disinfection Capital
Cost
Ultraviolet
Disinfection Annual
O&M Cost
Average Daily Flow/Design Flow (MGD)
0.007/0.03
$21,000
$4,000
$155,000
$69,000
$3,800
$79,000
$3,000
$170,000
$15,000
$14,000
$800
0.026/0.10
$68,000
$9,700
$442,000
$220,000
$17,400
$110,000
$3,800
$360,000
$31,000
$31,000
$2,000
0.09/0.30
$79,000
$10,900
$944,000
$490,000
$35,900
$277,000
$12,200
$970,000
$66,000
$85,000
$5,500
0.21/0.75
$88,000
$12,800
N/A
N/A
N/A
$336,000
$14,100
$2,100,000
$130,000
$140,000
$9,400
0.82/2.2
$130,000
$18,700
$1,254,000
$640,000
$56,400
$569,000
$19,100
$4,800,000
$430,000
$358,000
$34,000
3.25/7.8
$170,000
$36,800
$1,604,000
$830,000
$185,700
$1,470,000
$43,200
$14,600,000
$1,420,000
N/A
N/A
11.2/23.5
$340,000
$89,500
$2,279,000
$1,200,000
$592,300
$3,290,000
$99,000
$36,500,000
$4,600,000
N/A
N/A
45/81
$630,000
$308,000
$2,256,000
$2,000,000
$2,255,600
$6,550,000
$336,000
$117,000,000
$18,000,000
N/A
N/A
1 Cost for ultraviolet disinfection are calculated at sliahtlv different flow rates than indicated in the table.
population category that currently perform disinfection treatment. Because of the
uncertainty inherent in projecting the number of systems that would undertake each
corrective action, EPA assumed varying percentages of the non-treatment corrective
actions to provide and upper and lower cost bounds. Estimates of the corrective actions
that will be undertaken by systems with source water contamination are presented in
Exhibit C-14.
Each entry point that is predicted to undertake a corrective action is assigned one
of 13 potential corrective actions according to a probability distribution. In order to
determine the sensitivity of the cost estimates to these probabilities, two scenarios were
considered. Under the first scenario, entry points were assigned to low cost corrective
actions with greater probability (known as the Low CA scenario), while in the second
scenario, entry points were assigned to high cost corrective actions with greater
probability (known as the High CA scenario).
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
C-17
-------
Exhibit C-14. Estimated Selection of Corrective Action by
Systems with Source Water Contamination
Corrective Action
Rehabilitating an
existing well
Constructing an New
Well
Durchasing water
Eliminating Source of
contamination
Chlorine Gas
Disinfection
Hypochlorination
Chloramination
Chlorine Dioxide
Disinfection
Mixed Oxidants
Ozonafon
Reverse Osmosis
Jltraviolet Disinfection
Service Population Category
<100
25-30%
10-15%
1-3%
11-13%
0%
41%
0%
1%
0%
1%
0%
3%
101-500
10-15%
6-12%
1-3%
5-8%
0%
63%
1%
1%
0%
1%
1%
3%
501-1,000
8-12%
3-6%
1 -3 %
7 - 8 %
0%
69%
0%
1%
1%
1%
1%
3%
1,001- 3,300
8-12%
5-6%
1-3%
7-8%
0%
69%
0%
1%
1%
1%
1%
3%
3,301-
10,000
7-10%
3-5%
1-3%
5-6%
31%
41%
1%
1%
1%
1%
1%
3%
10,001-
50,000
6-8%
2-4%
0%
3%
77%
5%
1%
1%
0%
1%
1%
1%
50,001-
100,000
21-25%
8-11%
0%
9-10%
51%
0%
3%
1%
0%
1%
1%
1%
>1 00,000
8-10%
4-6%
0%
2%
78%
0%
3%
1%
0%
1%
0%
1%
After each entry point, that will undertake a corrective action, is assigned to a type
of corrective action, the capital and operations & maintenance (O&M) costs for these
corrective actions are calculated. Since it is assumed that all corrective actions occur in
year one, all corrective actions require a capital expenditure in year one. Some also
require a replacement capital expenditure in year 10. For all technologies with an O&M
cost component, equal O&M expenditures are assumed to occur each year. Depending
on the unit cost equations for the respective corrective action, the capital and O&M costs
are either fixed parameters or a simple function of the entry point's design flow or average
daily flow.
It is assumed that the capital cost of each corrective action includes the PWS's cost
of developing engineering plans for submission to the State.4 In addition to the PWS's
cost of corrective action, the State incurs a cost in reviewing the PWS's engineering plans.
The cost of plan review per corrective action is shown in Exhibit C-1. Finally, the State is
required to perform an on-site investigation if a PWS chooses to meet the corrective
action requirement by rehabilitating a well or removing a known source of contamination.
C.2.8 Non-Quantifiable Costs
4 It is assumed that 97.1 percent of the capital cost is for the actual corrective action, while 2.9
percent is for plan development and submission to the State for review and approval.
C-18
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Although EPA has estimated the cost of all the rule's components on drinking water
systems and States, there are some costs that the Agency did not quantify. These non-
quantifiable costs result from uncertainties surrounding rule assumptions and from
modeling assumptions. For example, EPA did not estimate a cost for systems to acquire
land if they needed to build a treatment facility or drill a new well. This was not costed
because many systems will be able to construct new wells or treatment facilities on land
already owned by the utility. In addition, if the cost of land was prohibitive, a system may
chose another lower cost alternative such as connecting to another source. A cost for
systems choosing this alternative is quantified in our analysis.
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis C19
-------
C.3 Non-monetary Inputs to Cost Model
Exhibit C-15. Non Community System Flows
Service Area Type
Day Care Centers
Highway Rest Areas
Hotels/Motels
Interstate Carriers
Medical Facilities
Mobile Home Parks
Restaurants
Schools
Service Stations
Summer Camps
Water Wholesalers
Agricultural Prod/Services
Airparks
Bowling Centers
Construction Activities
Churches
Campgrounds/RV Parks
Fire Departments
Federal Parks
Forest Service
Golf and Country Clubs
Landfills
Libraries
Mines
Misc. Amusement Parks
Military Bases
Migrant Labor Camps
Misc. Recreation Areas
Museums
Nursing Homes
Office Parks
Prisons
Race Tracks
Retailers (excluding food)
Retailers (food)
State Parks
Utilities
Zoological Gardens
Avg.
Flow
gpd/
person
15
5
65
5
100
100
8.5
25
10
42.5
100
100
4
3
3
10
45
100
10
5
25
25
15
25
20
100
50
5
10
100
15
120
5
10
18.5
7.5
25
25
Population
Served by
TNC Systems
10,213
516,369
558,443
11,257
208,623
66,797
2,255,959
150,365
326,644
765,742
791,429
22,770
67,116
23,170
0
1,301,552
639,160
12,578
93,665
37,600
254,016
0
3,330
0
88,038
2,900
27,900
337,152
35,280
0
197,600
0
58,000
184,128
142,988
842,518
6,025
3,300
Average
Flow for all
TNC Systems
153,195
2,581,845
36,298,795
56,285
20,862,300
6,679,700
19,175,652
3,759,125
3,266,440
32,544,035
79,142,900
2,277,000
268,464
69,510
0
13,015,520
28,762,200
1,257,800
936,650
188,000
6,350,400
0
49,950
0
1,760,760
290,000
1,395,000
1,685,760
352,800
0
2,964,000
0
290,000
1,841,280
2,645,278
6,318,885
150,625
82,500
Population
Served by
NTNC
Systems
61,653
6,105
46,680
35,221
144,061
19,236
154,528
3,015,155
12,177
6,711
46,075
27,968
6,060
0
5,247
11,500
19,680
4,018
780
4,494
11,716
3,432
0
13,447
66,462
37,525
2,079
22,533
0
13,910
129,542
121,940
0
120,775
45,724
13,712
84,621
0
Average
Flow for all
NTNC
Systems
924,795
30,525
3,034,200
176,105
14,406,100
1,923,600
1,313,488
75,378,875
121,770
285,218
4,607,500
2,796,800
24,240
0
15,741
115,000
885,600
401,800
7,800
22,470
292,900
85,800
0
336,175
1,329,240
3,752,500
103,950
112,665
0
1,391,000
1,943,130
14,632,800
0
1,207,750
845,894
102,840
2,115,525
0
C-20
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit C-15. Non Community System Flows
Service Area Type
Mfg. (food)
Mfg. (Mach & Comp. Equip.)
Mfg. (Elec. Equip. & Comps)
Mfq. (Chemicals)
Mfg. (Furniture & Fixtures)
Mfg. (Misc. Manufacturing)
Mfg. (Fab. Metal Products)
Mfg. (Paper& Allied Prods)
Mfg. (Petroleum Refining)
Mfg. (Primary Metals)
Mfg. (Printing)
Mfg. (Rub. & Misc. Plastics)
Mfg. (Stone, Clay, Glass, etc)
Mfg. (Tobacco Products)
Mfq. (Transportation Equip.)
Mfg. (Textiles)
Mfg. (Lumbers Wood Prods)
Unknowns
Mixed Knowns
TOTALS
Avg.
Flow
gpd/
person
35
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
30
Population
Served by
TNC Systems
158,301
0
0
0
2,750
8,991
3,300
0
0
0
0
0
2,775
0
0
0
2,775
5,500
214,345
10,441,364
Average
Flow for all
TNC Systems
5,540,535
0
0
0
68,750
224,775
82,500
0
0
0
0
0
69,375
0
0
0
69,375
137,500
283,665,464
Population
Served by
NTNC
Systems
285,910
40,000
2,133
12,384
1,472
275,146
43,804
38,560
35,855
26,278
4,000
2,000
29,146
1,500
1,080
34,590
15,300
16,856
92,797
5,319,657
Average
Flow for all
NTNC
Systems
10,006,850
1,000,000
53,325
309,600
36,800
6,878,650
1,095,100
964,000
896,375
656,950
100,000
50,000
728,650
37,500
27,000
864,750
382,500
421,400
0
159,233,246
Source: Geometries and Characteristics of Public Water Systems, US EPA, 1 999.
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
C-21
-------
C.4 References
Abbaszadegan, M., P.W. Stewart, M.W. LeChevallier, Rosen, Jeffery S. and C.P. Gerba. 1998/1999. Occurrence of
viruses in ground water in the United States. American Water Works Association Research Foundation.
Denver, CO, 157 p.
ASDWA, 1997. Evaluation of Best Management Practices Survey Data for Community Ground Water Systems.
Association of State Drinking Water Administrators, Washington DC, December.
EPA. 1996a. Ground Water Disinfection and Protective Practices in the United States. Office of Ground Water and Drinking
Water, Washington, B.C.
EPA. 1999a. Drinking Water Baseline Handbook, Draft 1st Edition. EPA, Washington D.C.
Lieberman, R.J., Shadix, L.C., Newport, C.P. Frebis, M.W.N. Moyer, S.E., Safferman, R.S., Stetler, R.E., Lye, D.,
Fout, G.S., and Dahling, D. 1999. "Source water microbial quality of some vulnerable public ground water
supplies." unpublished report in preparation.
Lieberman, R.J., L.C. Shadix, B.S. Newport, S.R. Grout, S.E. Buescher, R.S. Safferman, R.E. Stetler, D. Lye, G.S.
Fout, and D. Dahling. 1994. "Source water microbial quality of some vulnerable public ground water
supplies." in Proceedings, Water Quality Technology Conference, San Francisco, CA, October, 1994.
-------
Appendix D
Results of Cost Analysis
-------
LIST OF EXHIBITS
Exhibit D-l. Total Costs for Option 1: Sanitary Survey Only D-3
Exhibit D-2. Total Costs by PWS Size for Option 1: Sanitary Survey Only D-4
Exhibit D-3. Total Costs by PWS Type for Option 1: Sanitary Survey Only D-6
Exhibit D-4. Total Costs for Option 2: Sanitary Survey and Triggered Monitoring . . . D-7
Exhibit D-5. Total Costs by PWS Size Option 2: Sanitary Survey and
Triggered Monitoring D-8
Exhibit D-6. Total Costs by PWS Type Option 2: Sanitary Survey and
Triggered Monitoring D-10
Exhibit D-7. Total Costs for Option 3: Multiple Barrier Approach D-l 1
Exhibit D-8. Total Costs by PWS Size for Option 3: Multiple Barrier Approach .... D-l2
Exhibit D-9. Total Costs by PWS Type for Option 3: Multiple Barrier Approach . . . D-l4
Exhibit D-10.Total Costs for Option 4: Across-the-Board Disinfection D-l5
Exhibit D-11.Total Costs by PWS Size for Option 4: Across-the-Board Disinfection D-16
Exhibit D-12.Total Costs by PWS Type for Option 4: Across-the-Board Disinfection D-l8
Exhibit D-13.Household Costs of the GWR (High Corrective Action/Low Significant
Defect Scenario) Option 1: Sanitary Survey Only D-l9
Exhibit D-14.Household Costs of the GWR
(High Corrective Action/Low Significant Defect Scenario)
Option 2: Sanitary Survey and Triggered Monitoring D-20
Exhibit D-l5.Household Costs of the GWR
(High Corrective Action/Low Significant Defect Scenario)
Option 3: Multi Barrier Approach D-21
Exhibit D-16.Household Costs of the GWR
(High Corrective Action/Low Significant Defect Scenario)
Option 4: Across-the-Board Disinfection D-22
Exhibit D-l 7.Comparative Household Costs of the GWR Across Regulatory Options
For Systems Serving 0-500 People D-23
Exhibit D-l 8.Comparative Household Costs of the GWR Across Regulatory Options
For All System Size Categories D-24
-------
This appendix contains exhibits summarizing the detailed results of running the GWR cost mod<
for each of the four regulatory options:
Option 1: Sanitary Survey Only
Option 2: Sanitary Survey and Triggered Monitoring
Option 3: Multiple Barrier Approach
Option 4: Across-the-Board Disinfection
For each option, this appendix presents three exhibits, covering the following topics:
Total National Compliance Costs
Costs, within this table, are reported as they impact two affected entities:
Systems
States
Using four scenarios:
Low Corrective Action Costs and Low Significant Defect Scenario
Low Corrective Action Costs and High Significant Defect Scenario
High Corrective Action Costs and Low Significant Defect Scenario
High Corrective Action Costs and High Significant Defect Scenario
Reported in two forms:
Annualized Costs
Present Value
Employing two discount rates (3 percent and 7 percent).
National Compliance Costs of the GWR by Public Water System Size
Costs, within this table, are reported as they impact two affected entities:
Systems
States
Using four scenarios:
Low Corrective Action Costs and Low Significant Defect Scenario
Low Corrective Action Costs and High Significant Defect Scenario
High Corrective Action Costs and Low Significant Defect Scenario
High Corrective Action Costs and High Significant Defect Scenario
Aprils, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis Dl
-------
With costs broken out by the following system sizes:
less than 100 persons,
101-500 persons,
501-1,000 persons,
1,001-3,000 persons,
3,001-10,000 persons,
10,001-50,000 persons,
50,001-100,000 persons,
100,001-1,000,000 persons, and
Reported in Annualized Costs
Employing two discount rates (3 percent and 7 percent).
National Compliance Costs of the GWR by Public Water System Type
Costs, within this table, are reported as they impact two affected entities:
Systems
States
Using four scenarios:
Low Corrective Action Costs and Low Significant Defect Scenario
Low Corrective Action Costs and High Significant Defect Scenario
High Corrective Action Costs and Low Significant Defect Scenario
High Corrective Action Costs and High Significant Defect Scenario
With costs broken out by the following system types:
Community Water Systems
Nontransient, Noncommunity Water Systems
Transient/Noncommunity Water Systems
Reported in Annualized Costs
Employing two discount rates (3 percent and 7 percent).
D2 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
Exhibit D-1. Total Costs for Option 1: Sanitary Survey Only ($million)
TOTAL NATIONAL COSTS
Annual System Costs:
Annual Cost (Low CA/Low SD)
Annual Cost (Low CA/High SD)
Annual Cost (High CA/Low SD)
Annual Cost (High CA/High SD)
Annual State Costs:
Annual Cost (Low CA/Low SD)
Annual Cost (Low CA/High SD)
Annual Cost (High CA/Low SD)
Annual Cost (High CA/High SD)
Present Value of System Costs:
PV Cost (Low CA/Low SD)
PV Cost (Low CA/High SD)
PV Cost (High CA/Low SD)
PV Cost (High CA/High SD)
Present Value of State Costs:
PV Cost (Low CA/Low SD)
PV Cost (Low CA/High SD)
PV Cost (High CA/Low SD)
PV Cost (High CA/High SD)
DISCOUNT RATE
3% 7%
$53.7
$56.8
$53.7
$56.8
$17.6
$17.5
$17.6
$17.5
$798.3
$845.5
$798.3
$845.5
$261 .2
$259.6
$261 .2
$259.6
$56.2
$59.5
$56.2
$59.5
$18.2
$18.1
$18.2
$18.1
$595.6
$630.5
$595.6
$630.5
$192.9
$191.7
$192.9
$191.7
Notes:
Low CA = Low Cost Corrective Action Scenario
High CA = High Cost Corrective Action Scenario
Low SD = Low Cost Significant Defect Scenario
Hiah SD = Hiah Cost Sianificant Defect Scenario
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
D-3
-------
Exhibit D-2. Total Costs by PWS Size for Option 1: Sanitary Survey Only
DISCOUNT
RATE
Low Corrective Action / Low
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / Low
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / High
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / High
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / Low
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / Low
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / High
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / High
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
SYSTEM SIZE CATEGORIES
<100 101-500 501-1,000 1,001-3,300 3,301-10,000
$18.1
$4.1
$22.3
$18.1
$4.1
$22.3
$20.6
$4.0
$24.6
$20.6
$4.0
$24.6
$19.1
$4.3
$23.4
$19.1
$4.3
$23.4
$21.7
$4.2
$25.9
$21.7
$4.2
$25.9
$11.4
$1.9
$13.3
$11.4
$1.9
$13.3
$13.9
$1.9
$15.8
$13.9
$1.9
$15.8
$12.0
$2.0
$14.0
$12.0
$2.0
$14.0
$14.5
$2.0
$16.5
$14.5
$2.0
$16.5
$5.4
$0.6
$6.0
$5.4
$0.6
$6.0
$5.3
$0.6
$5.8
$5.3
$0.6
$5.8
$5.6
$0.6
$6.2
$5.6
$0.6
$6.2
$5.5
$0.6
$6.1
$5.5
$0.6
$6.1
$7.9
$0.6
$8.5
$7.9
$0.6
$8.5
$6.4
$0.6
$7.0
$6.4
$0.6
$7.0
$8.2
$0.6
$8.8
$8.2
$0.6
$8.8
$6.7
$0.6
$7.3
$6.7
$0.6
$7.3
$5.0
$0.3
$5.3
$5.0
$0.3
$5.3
$3.9
$0.3
$4.2
$3.9
$0.3
$4.2
$5.2
$0.3
$5.5
$5.2
$0.3
$5.5
$4.1
$0.3
$4.4
$4.1
$0.3
$4.4
D-4
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit D-2. Total Costs by PWS Size for Option 1: Sanitary Survey Only
(continued)
SYSTEM SIZE CATEGORIES
10,001-50,000 50,001-100,000 100,001-1,000,000 TOTAL
DISCOUNT
RATE
Low Corrective Action / Low
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / Low
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / High
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / High
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / Low
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / Low
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / High
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / High
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
$3.6
$0.2
$3.8
$3.6
$0.2
$3.8
$3.6
$0.2
$3.8
$3.6
$0.2
$3.8
$3.8
$0.2
$4.0
$3.8
$0.2
$4.0
$3.7
$0.2
$4.0
$3.7
$0.2
$4.0
$0.8
$0.0
$0.8
$0.8
$0.0
$0.8
$1.1
$0.0
$1.1
$1.1
$0.0
$1.1
$0.8
$0.1
$0.9
$0.8
$0.1
$0.9
$1.1
$0.1
$1.2
$1.1
$0.1
$1.2
$1.4
$0.0
$1.4
$1.4
$0.0
$1.4
$2.0
$0.0
$2.1
$2.0
$0.0
$2.1
$1.5
$0.0
$1.5
$1.5
$0.0
$1.5
$2.1
$0.0
$2.2
$2.1
$0.0
$2.2
$53.7
$17.6
$71.2
$53.7
$17.6
$71.2
$56.8
$17.5
$74.3
$56.8
$17.5
$74.3
$56.2
$18.2
$74.4
$56.2
$18.2
$74.4
$59.5
$18.1
$77.6
$59.5
$18.1
$77.6
AprilS, 2000 Proposed Ground Water Rule - Regulatory Impact Analysis D-5
-------
Exhibit D-3. Total Costs by PWS Type for Option 1: Sanitary Survey Only
SYSTEM TYPE
Non-Transient/IMon- Transierrt/IMon-
Community Water Community Water Community
Systems Systems Water Systems
DISCOUNT
RATE
Low Corrective Action / Low
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / Low
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / High
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / High
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / Low
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / Low
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / High
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / High
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
$32.9
$2.9
$35.8
$32.9
$2.9
$35.8
$33.1
$2.8
$35.9
$33.1
$2.8
$35.9
$34.3
$3.0
$37.3
$34.3
$3.0
$37.3
$34.5
$2.9
$37.5
$34.5
$2.9
$37.47
$3.8
$0.9
$4.7
$3.8
$0.9
$4.7
$4.5
$0.9
$5.4
$4.5
$0.9
$5.4
$4.0
$0.9
$5.0
$4.0
$0.9
$5.0
$4.8
$0.9
$5.7
$4.8
$0.9
$5.71
$16.9
$4.0
$21.0
$16.9
$4.0
$21.0
$19.2
$4.0
$23.1
$19.2
$4.0
$23.1
$17.9
$4.2
$22.1
$17.9
$4.2
$22.1
$20.2
$4.1
$24.3
$20.2
$4.1
$24.32
TOTAL
$53.7
$17.6
$71.2
$53.7
$17.6
$71.2
$56.6
$17.£
$74.3
$56.6
$17.£
$74.3
$56.2
$18.2
$74.'
$56.2
$18.2
$74.'
$59.£
$18.1
$77.6
$59.£
$18.1
$77.61
D6 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
Exhibit D-4. Total Costs for Option 2: Sanitary Survey and Triggered Monitoring
TOTAL NATIONAL COSTS
Annual System Costs:
Annual Cost (Low CA/Low SD)
Annual Cost (Low CA/High SD)
Annual Cost (High CA/Low SD)
Annual Cost (High CA/High SD)
Annual State Costs:
Annual Cost (Low CA/Low SD)
Annual Cost (Low CA/High SD)
Annual Cost (High CA/Low SD)
Annual Cost (High CA/High SD)
Present Value of System Costs:
PV Cost (Low CA/Low SD)
PV Cost (Low CA/High SD)
PV Cost (High CA/Low SD)
PV Cost (High CA/High SD)
Present Value of State Costs:
PV Cost (Low CA/Low SD)
PV Cost (Low CA/High SD)
PV Cost (High CA/Low SD)
PV Cost (High CA/High SD)
DISCOUNT RATE
3% 7%
$133.9
$140.0
$137.4
$143.5
$18.9
$18.8
$18.9
$18.8
$1,992.8
$2,083.0
$2,044.3
$2,134.6
$280.8
$280.3
$280.8
$280.3
$143.2
$149.5
$148.0
$154.3
$19.8
$19.7
$19.8
$19.7
$1,517.5
$1,584.3
$1,567.4
$1,634.1
$209.5
$209.1
$209.5
$209.1
Notes:
Low CA = Low Cost Corrective Action Scenario
High CA = High Cost Corrective Action Scenario
Low SD = Low Cost Significant Defect Scenario
Hiah SD = Hiah Cost Sianificant Defect Scenario
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
D-7
-------
Exhibit D-5. Total Costs by PWS Size Option 2: Sanitary Survey and
Triggered Monitoring
DISCOUNT
RATE
Low Corrective Action / Low
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / Low
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / High
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / High
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / Low
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / Low
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / High
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / High
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
SYSTEM SIZE CATEGORIES
<100 101-500 501-1,000 1,001-3,300 3,301-10,000
$66.6
$4.6
$71.2
$68.6
$4.6
$73.2
$69.2
$4.6
$73.8
$71.2
$4.6
$75.7
$70.9
$4.9
$75.9
$73.8
$4.9
$78.7
$73.6
$4.9
$78.5
$76.4
$4.9
$81.3
$31.4
$2.3
$33.7
$31.7
$2.3
$34.0
$32.7
$2.3
$34.9
$33.0
$2.3
$35.3
$33.5
$2.4
$36.0
$34.1
$2.4
$36.5
$34.9
$2.4
$37.3
$35.4
$2.4
$37.8
$8.4
$0.7
$9.1
$9.9
$0.7
$10.6
$9.1
$0.7
$9.8
$10.6
$0.7
$11.3
$9.0
$0.7
$9.8
$10.7
$0.7
$11.4
$9.7
$0.7
$10.4
$11.4
$0.7
$12.1
$10.2
$0.7
$10.9
$11.3
$0.7
$12.0
$11.0
$0.7
$11.7
$12.1
$0.7
$12.9
$10.9
$0.8
$11.7
$12.2
$0.8
$12.9
$11.8
$0.8
$12.5
$13.0
$0.8
$13.8
$6.9
$0.4
$7.2
$6.9
$0.4
$7.3
$7.4
$0.4
$7.7
$7.4
$0.4
$7.8
$7.4
$0.4
$7.8
$7.5
$0.4
$7.9
$7.9
$0.4
$8.3
$8.0
$0.4
$8.4
D-8
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit D-5. Total Costs by PWS Size Option 2: Sanitary Survey and
Triggered Monitoring (continued)
SYSTEM SIZE CATEGORIES
10,001-50,000 50,001-100,000 100,001-1,000,000
DISCOUNT
RATE
Low Corrective Action / Low
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / Low
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / High
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / High
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / Low
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / Low
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / High
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / High
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
$5.7
$0.3
$6.1
$5.6
$0.3
$5.9
$6.0
$0.3
$6.3
$5.8
$0.3
$6.1
$6.2
$0.3
$6.6
$6.1
$0.3
$6.4
$6.5
$0.3
$6.8
$6.3
$0.3
$6.6
$3.7
$0.1
$3.7
$2.3
$0.1
$2.3
$3.7
$0.1
$3.8
$2.3
$0.1
$2.4
$4.0
$0.1
$4.0
$2.5
$0.1
$2.5
$4.0
$0.1
$4.1
$2.5
$0.1
$2.6
$1.0
$0.0
$1.1
$1.0
$0.0
$1.1
$1.1
$0.0
$1.1
$1.1
$0.0
$1.1
$1.1
$0.0
$1.2
$1.2
$0.0
$1.2
$1.2
$0.0
$1.2
$1.2
$0.0
$1.2
TOTAL
$133.9
$18.9
$152.8
$137.4
$18.9
$156.3
$140.0
$18.8
$158.9
$143.5
$18.8
$162.3
$143.2
$19.8
$163.0
$148.0
$19.8
$167.7
$149.5
$19.7
$169.3
$154.3
$19.7
$174.0
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
D-9
-------
Exhibit D-6. Total Costs by PWS Type Option 2: Sanitary Survey and
Triggered Monitoring
SYSTEM TYPE
Non-Transient/Non- Transient/Non-
Community Water Community Water Community
Systems Systems Water Systems
DISCOUNT
RATE
Low Corrective Action / Low
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / Low
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / High
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / High
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / Low
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / Low
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / High
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / High
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
$50.6
$3.6
$54.2
$52.3
$3.6
$55.9
$53.5
$3.6
$57.1
$55.2
$3.6
$58.8
$54.1
$3.8
$57.9
$56.1
$3.8
$59.9
$57.2
$3.8
$60.9
$59.2
$3.8
$62.9
$13.3
$1.0
$14.4
$13.1
$1.0
$14.1
$14.0
$1.0
$15.1
$13.8
$1.0
$14.8
$14.4
$1.1
$15.4
$14.2
$1.1
$15.3
$15.1
$1.1
$16.2
$14.9
$1.1
$16.0
$70.0
$4.5
$74.5
$72.0
$4.5
$76.5
$72.5
$4.5
$76.9
$74.4
$4.5
$78.9
$74.8
$4.8
$79.6
$77.6
$4.8
$82.5
$77.3
$4.8
$82.1
$80.2
$4.8
$85.0
TOTAL
$133.9
$18.9
$152.8
$137.4
$18.9
$156.3
$140.0
$18.8
$158.9
$143.5
$18.8
$162.3
$143.2
$19.8
$163.0
$148.0
$19.8
$167.7
$149.5
$19.7
$169.3
$154.3
$19.7
$174.0
D-10
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit D-7. Total Costs for Option 3: Multiple Barrier Approach
TOTAL NATIONAL COSTS
Annual System Costs:
Annual Cost (Low CA/Low SD)
Annual Cost (Low CA/High SD)
Annual Cost (High CA/Low SD)
Annual Cost (High CA/High SD)
Annual State Costs:
Annual Cost (Low CA/Low SD)
Annual Cost (Low CA/High SD)
Annual Cost (High CA/Low SD)
Annual Cost (High CA/High SD)
Present Value of System Costs:
PV Cost (Low CA/Low SD)
PV Cost (Low CA/High SD)
PV Cost (High CA/Low SD)
PV Cost (High CA/High SD)
Present Value of State Costs:
PV Cost (Low CA/Low SD)
PV Cost (Low CA/High SD)
PV Cost (High CA/Low SD)
PV Cost (High CA/High SD)
DISCOUNT RATE
3% 7%
$156.4
$162.7
$161.6
$167.9
$20.6
$20.6
$20.6
$20.6
$2,327.3
$2,421 .0
$2,403.8
$2,497.5
$306.0
$306.0
$306.0
$306.0
$169.6
$176.1
$176.8
$183.4
$22.1
$22.1
$22.1
$22.1
$1,796.3
$1,865.6
$1,873.3
$1,942.7
$234.4
$234.4
$234.4
$234.4
Notes:
Low CA = Low Cost Corrective Action Scenario
High CA = High Cost Corrective Action Scenario
Low SD = Low Cost Significant Defect Scenario
Hiah SD = Hiah Cost Sianificant Defect Scenario
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
D-ll
-------
Exhibit D-8. Total Costs by PWS Size for Option 3: Multiple Barrier Approach
DISCOUNT
RATE
Low Corrective Action / Low
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / Low
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / High
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / High
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / Low
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / Low
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / High
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / High
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
SYSTEM SIZE CATEGORIES
<100 101-500 501-1,000 1,001-3,300 3,301-10,000
$70.8
$5.6
$76.4
$74.3
$5.6
$80.0
$73.4
$5.6
$79.0
$76.9
$5.6
$82.5
$76.5
$6.3
$82.9
$81.6
$6.3
$87.9
$79.2
$6.3
$85.5
$84.2
$6.3
$90.6
$39.1
$2.7
$41.8
$39.2
$2.7
$41.9
$40.7
$2.7
$43.4
$40.8
$2.7
$43.5
$42.4
$3.0
$45.4
$42.9
$3.0
$45.9
$44.1
$3.0
$47.1
$44.6
$3.0
$47.6
$12.1
$0.8
$12.8
$11.2
$0.8
$12.0
$12.7
$0.8
$13.5
$11.9
$0.8
$12.6
$13.0
$0.9
$13.9
$12.2
$0.9
$13.1
$13.7
$0.9
$14.5
$12.9
$0.9
$13.7
$13.5
$0.8
$14.3
$13.8
$0.8
$14.6
$14.3
$0.8
$15.1
$14.6
$0.8
$15.4
$14.7
$0.9
$15.5
$15.0
$0.9
$15.9
$15.5
$0.9
$16.3
$15.8
$0.9
$16.7
$10.:
$0y
$10. £
$12.:
$0y
$12. £
$10. £
$0y
$11. C
$12. £
$0y
$13.C
$11.1
$0y
$11. i
$13.;
$0y
$13.7
$11. E
$0y
$12. C
$13.7
$0y
$14.:
D-12
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit D-8. Total Costs by PWS Size for Option 3: Multiple Barrier Approach
(continued)
SYSTEM SIZE CATEGORIES
10,001-50,000 50,001-100,000 100,001-1,000,000
DISCOUNT
RATE
Low Corrective Action / Low
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / Low
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / High
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / High
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / Low
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / Low
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / High
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / High
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
$5.3
$0.3
$5.7
$5.9
$0.3
$6.3
$5.6
$0.3
$5.9
$6.2
$0.3
$6.5
$5.8
$0.4
$6.2
$6.5
$0.4
$6.8
$6.1
$0.4
$6.5
$6.8
$0.4
$7.1
$3.9
$0.1
$3.9
$3.3
$0.1
$3.4
$3.9
$0.1
$4.0
$3.4
$0.1
$3.4
$4.2
$0.1
$4.3
$3.6
$0.1
$3.7
$4.3
$0.1
$4.3
$3.7
$0.1
$3.7
$1.6
$0.1
$1.6
$1.5
$0.1
$1.6
$1.6
$0.1
$1.7
$1.6
$0.1
$1.6
$1.8
$0.1
$1.8
$1.7
$0.1
$1.8
$1.8
$0.1
$1.9
$1.7
$0.1
$1.8
TOTAL
$156.4
$20.6
$177.0
$161.6
$20.6
$182.1
$162.7
$20.6
$183.3
$167.9
$20.6
$188.4
$169.6
$22.1
$191.7
$176.8
$22.1
$199.0
$176.1
$22.1
$198.2
$183.4
$22.1
$205.5
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
D-13
-------
Exhibit D-9. Total Costs by PWS Type for Option 3: Multiple Barrier Approach
SYSTEM TYPE
Non-Transient/Non- Transient/Non-
Community Water Community Water Community
Systems Systems Water Systems
DISCOUNT
RATE
Low Corrective Action / Low
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / Low
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / High
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / High
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / Low
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / Low
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / High
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / High
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
$60.6
$3.9
$64.6
$62.1
$3.9
$66.0
$63.7
$3.9
$67.7
$65.2
$3.9
$69.1
$65.6
$4.2
$69.8
$67.2
$4.2
$71.5
$68.8
$4.2
$73.0
$70.4
$4.2
$74.7
$15.6
$1.2
$16.8
$15.8
$1.2
$17.1
$16.3
$1.2
$17.6
$16.6
$1.2
$17.8
$17.0
$1.4
$18.4
$17.4
$1.4
$18.8
$17.8
$1.4
$19.1
$18.1
$1.4
$19.5
$80.2
$5.6
$85.8
$83.6
$5.6
$89.3
$82.7
$5.6
$88.3
$86.1
$5.6
$91.8
$87.0
$6.4
$93.4
$92.2
$6.4
$98.6
$89.6
$6.4
$96.0
$94.8
$6.4
$101.2
TOTAL
$156.4
$20.6
$177.0
$161.6
$20.6
$182.1
$162.7
$20.6
$183.3
$167.9
$20.6
$188.4
$169.6
$22.1
$191.7
$176.8
$22.1
$199.0
$176.1
$22.1
$198.2
$183.4
$22.1
$205.5
D14 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
Exhibit D-10. Total Costs for Option 4: Across-the-Board Disinfection
TOTAL NATIONAL COSTS
Annual System Costs:
Annual Cost (Low CA/Low SD)
Annual Cost (Low CA/High SD)
Annual Cost (High CA/Low SD)
Annual Cost (High CA/High SD)
Annual State Costs:
Annual Cost (Low CA/Low SD)
Annual Cost (Low CA/High SD)
Annual Cost (High CA/Low SD)
Annual Cost (High CA/High SD)
Present Value of System Costs:
PV Cost (Low CA/Low SD)
PV Cost (Low CA/High SD)
PV Cost (High CA/Low SD)
PV Cost (High CA/High SD)
Present Value of State Costs:
PV Cost (Low CA/Low SD)
PV Cost (Low CA/High SD)
PV Cost (High CA/Low SD)
PV Cost (High CA/High SD)
DISCOUNT RATE
3% 7%
$718.7
$723.9
$779.8
$785.0
$25.2
$25.2
$25.2
$25.2
$10,691.7
$10,769.9
$11,601.0
$11,679.2
$375.5
$375.0
$375.5
$375.0
$794.1
$799.6
$875.3
$880.7
$28.6
$28.6
$28.6
$28.6
$8,412.9
$8,470.7
$9,272.7
$9,330.5
$303.2
$302.9
$303.2
$302.9
Notes:
Low CA = Low Cost Corrective Action Scenario
High CA = High Cost Corrective Action Scenario
Low SD = Low Cost Significant Defect Scenario
Hiah SD = Hiah Cost Sianificant Defect Scenario
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
D-15
-------
Exhibit D-11. Total Costs by PWS Size for Option 4: Across-the-Board
Disinfection
DISCOUNT
RATE
Low Corrective Action / Low
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / Low
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / High
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / High
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / Low
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / Low
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / High
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / High
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
<100
$169.8
$9.0
$178.8
$206.1
$9.0
$215.1
$172.0
$9.0
$181.0
$208.2
$9.0
$217.2
$190.7
$10.8
$201.5
$241.7
$10.8
$252.5
$192.9
$10.8
$203.7
$243.9
$10.8
$254.7
SYSTEM SIZE CATEGORIES
101-500 501-1,000 1,001-3,300 3,301-10,000
$186.8
$4.0
$190.8
$198.2
$4.0
$202.2
$187.8
$4.0
$191.8
$199.3
$4.0
$203.3
$204.3
$4.8
$209.1
$220.4
$4.8
$225.2
$205.4
$4.8
$210.2
$221 .6
$4.8
$226.3
$74.1
$0.9
$75.0
$76.6
$0.9
$77.5
$74.8
$0.9
$75.7
$77.3
$0.9
$78.2
$81.4
$1.0
$82.4
$84.4
$1.0
$85.4
$82.0
$1.0
$83.1
$85.0
$1.0
$86.1
$103.4
$0.8
$104.2
$106.6
$0.8
$107.4
$104.0
$0.8
$104.8
$107.3
$0.8
$108.0
$113.7
$0.9
$114.6
$117.1
$0.9
$118.0
$114.3
$0.9
$115.2
$117.7
$0.9
$118.6
$84.7
$0.4
$85.0
$93.9
$0.4
$94.3
$85.1
$0.4
$85.5
$94.4
$0.4
$94.8
$93.3
$0.4
$93.7
$102.9
$0.4
$103.3
$93.8
$0.4
$94.2
$103.4
$0.4
$103.8
D16 Proposed Ground Water Rule - Regulatory Impact Analysis April 5, 2000
-------
Exhibit D-11. Total Costs by PWS Size for Option 4: Across-the-Board
Disinfection (continued)
SYSTEM SIZE CATEGORIES
10,001-50,000 50,001-100,000 100,001-1,000,000
DISCOUNT
RATE
Low Corrective Action / Low
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / Low
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / High
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / High
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / Low
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / Low
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / High
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / High
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
$45.2
$0.3
$45.5
$42.7
$0.3
$43.1
$45.4
$0.3
$45.7
$43.0
$0.3
$43.3
$50.2
$0.4
$50.6
$47.5
$0.4
$47.9
$50.4
$0.4
$50.8
$47.7
$0.4
$48.1
$43.7
$0.1
$43.8
$44.0
$0.1
$44.1
$43.8
$0.1
$43.8
$44.1
$0.1
$44.1
$48.0
$0.1
$48.1
$48.2
$0.1
$48.3
$48.1
$0.1
$48.2
$48.2
$0.1
$48.3
$10.9
$0.0
$11.0
$11.5
$0.0
$11.6
$11.0
$0.0
$11.0
$11.5
$0.0
$11.6
$12.6
$0.1
$12.7
$13.2
$0.1
$13.2
$12.6
$0.1
$12.7
$13.2
$0.1
$13.3
TOTAL
$718.7
$25.2
$743.£
$779.8
$25.2
$805.C
$723.£
$25.2
$749.1
$785.C
$25.2
$810.2
$794.1
$28.6
$822.7
$875.3
$28.6
$903.£
$799.6
$28.6
$828.2
$880.7
$28.6
$909.3
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
D-17
-------
Exhibit D-12. Total Costs by PWS Type for Option 4:
Across-the-Board Disinfection
SYSTEM TYPE
Non-Transient/Non- Transient/Non-
Community Water Community Water Community
Systems Systems Water Systems
DISCOUNT
RATE
Low Corrective Action / Low
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / Low
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / High
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / High
Significant Defect 3%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / Low
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / Low
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
Low Corrective Action / High
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
High Corrective Action / High
Significant Defect 7%
System Costs (Annual)
State Costs (Annual)
Total Costs (Annual)
$376.3
$3.6
$379.9
$397.0
$3.6
$400.6
$378.8
$3.6
$382.5
$399.5
$3.6
$403.2
$414.8
$4.3
$419.0
$438.1
$4.3
$442.4
$417.4
$4.3
$421.7
$440.8
$4.3
$445.0
$81.5
$2.1
$83.5
$85.5
$2.1
$87.5
$82.0
$2.1
$84.0
$86.0
$2.1
$88.1
$89.6
$2.5
$92.1
$96.1
$2.5
$98.6
$90.2
$2.5
$92.6
$96.7
$2.5
$99.1
$260.9
$9.8
$270.7
$297.3
$9.8
$307.1
$263.1
$9.8
$272.8
$299.5
$9.8
$309.2
$289.7
$11.8
$301 .5
$341 .0
$11.8
$352.8
$292.0
$11.8
$303.8
$343.3
$11.8
$355.1
TOTAL
$718.7
$25.2
$743.9
$779.8
$25.2
$805.0
$723.9
$25.2
$749.1
$785.0
$25.2
$810.2
$794.1
$28.6
$822.7
$875.3
$28.6
$903.9
$799.6
$28.6
$828.2
$880.7
$28.6
$909.3
D-18
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit D-13.
Household Costs of the GWR
(High Corrective Action/Low Significant Defect Scenario)
Option 1: Sanitary Survey Only
Percent of Households
IUU.U7o
OU.U /o
70.0% '
bU.U/o
en n% -
30.0% "
1 U.U/o
nno/ .
//r: ;<--:>'-"'
-------
Exhibit D-14.
Household Costs of the GWR
(High Corrective Action/Low Significant Defect Scenario)
Option 2: Sanitary Survey and Triggered Monitoring
in
o
o
.c
(1)
(A
3
O
T
n-
O
4-i
C
0)
o
0)
0.
iuu.u%
yu.uyo
oU.Uyo
/U.Uyo
OU.Uyo
OU.Uyo
4U.Uyo
oU.Uyo
zU.Uyo
1 U.Uyo
n no/_ .
/^ -"''*** '
7 /''.' /
-' /'' x
* . ' *
--' ^ /
* i *
/ /' i
' / 7
s /
i
*/ t
/ /
*
/
/
f
»
/
4
#
S
+ i~- -- " "^ . *"
H « ^^
«
r
0-500
501-3,300
- - - 0-10,000
10,000-1 Million
All Systems
$0.50 $1 $3 $5 $10 $25 $50 $75 $100 $200 $400 $600
Maximum Annual Household Cost
D-20
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit D-15.
Household Costs of the GWR
(High Corrective Action/Low Significant Defect Scenario)
Option 3: Mult! Barrier Approach
Percent of Systems
iuu.u%
yu.uyo
oU.Uyo
/U.Uyo
OU.Uyo
OU.Uyo
4U.Uyo
oU.Uyo
zU.Uyo
1 U.Uyo
n no/_ .
/r^" ,>/" ~^..'''"~
/' * r ^
/ / 4 * -r
/ / / /
/' 7 /
/' * * i
// : / /
/ ' /
£ //
//
.''/ /
7 /
/ /
/ .>'
/ s
/
f
4
*
/
»
^
S
- 0-500
501-3,300
-- 0-10,000
10,000-1 Million
All Systems
$0.50 $1 $3 $5 $10 $25 $50 $75 $100 $200 $400 $600
Maximum Annual Household Cost
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
D-21
-------
100.0%
0.0%
Exhibit D-16.
Household Costs of the GWR
(High Corrective Action/Low Significant Defect Scenario)
Option 4: Across-the-Board Disinfection
- 0-500
-501-3,300
- 0-10,000
$0.50 $1 $3 $5 $10 $25 $50 $75 $100 $200 $400 $600
Maximum Annual Household Cost
D-22
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
-------
Exhibit D-17.
Comparative Household Costs of the GWR Across Regulatory Options
For Systems Serving 0-500 People
Maximum Annual Household Costs
April 5, 2000
Proposed Ground Water Rule - Regulatory Impact Analysis
D-23
-------
Exhibit D-18.
Comparative Household Costs of the GWR Across Regulatory Options
For All System Size Categories
90%
80%
70%
60%
50%
/
o>
3
O
I
"o
4-i
C
O
0)
0.
30%
20%
10%
0%
Survey
Sanitary Survey and Triggered
Monitoring
Multi-Barrier
-- Across-the-Board Disinfection
$0.50 $1.00 $3.00 $5.00 $10.00 $25.00 $50.00 $75.00 $100.00 $200.00 $400.00 $600.00
Maximum Annual Household Costs
D-24
Proposed Ground Water Rule - Regulatory Impact Analysis
April 5, 2000
------- |