Batteiie
                                                          The Business of Innovation
                                                                505 King Avenue
                                                            Columbus, Ohio 43201 -2693
                                                          (614) 424-6424  Fax (614) 424-5263
March 2, 2007
Ms. Sineta Wooten
Project Officer
Program Assessment and Outreach Branch
National Programs Chemical Division
Office of Pollution Prevention and Toxics
U.S Environmental Protection Agency
1200 Pennsylvania Avenue NW (7404T)
Washington, D.C. 20460

Dear Sineta:

Reference: Contract No. EP-W-04-021, Work Assignment 2-13: Final Deliverables on the
Pilot Study of Targeting Elevated Blood-Lead Levels in Children

Enclosed for your comments and suggestions are the following four work products:

   1. Draft Final Report for the Pilot Study of Targeting Elevated Blood-Lead Levels in
      Children,
   2. A DVD containing the documented datasets corresponding to the series of Broad-Based
      National Models for Risk of Childhood Lead Poisoning at the county-level based on
      quarterly aggregate summary data provided by CDC, as well as the visualization tool for
      these models,
   3. A DVD containing the documented datasets corresponding to the series of High-
      Resolution Models for Risk of Childhood Lead Poisoning at the census-tract level based
      on surveillance data from the Massachusetts Department of Public Health, as well as the
      visualization tool for these models.
   4. A DVD containing a Microsoft Word version of the draft final report and the PowerPoint
      Presentation that Batteiie provided to EPA technical staff on March 2, 2007.

We very much enjoyed our collaboration with the technical staff at EPA's Office of
Environmental Information on this important and challenging project. We hope to have the
opportunity to continue this work with EPA in the near future.

-------
Ms. Sineta Woolen
March 2, 2007
Page 2
If you have any questions concerning these deliverables, please call me at 614/424-4547 or
Warren Strauss at 614/424-4275.

Sincerely,
Bruce E. Buxton
Senior Program Manager
Statistics and Information Analysis

BEB:lnl
Enclosures (4)

cc. MargarerConprnosn(Work Assignment Manager, U.S. EPA - All Four Work Products)
   Barry Nussbaum  "(Technical Advisor, U.S. EPA - All Four Work Products)
   Heather Case       (U.S. EPA - All Four Work Products)
   Ron Morony       (U.S. EPA - All Four Work Products)
   Paul Hunter        (Massachusetts Department of Public Health - Work Products 1 & 3)

-------
                                                            MARCH 2, 2007
              DRAFT FINAL REPORT FOR THE PILOT STUDY OF
         TARGETING ELEVATED BLOOD-LEAD LEVELS IN CHILDREN
                       EPA Contract No. EP-W-04-021
                                Prepared by
                                BATTELLE
                              505 King Avenue
                           Columbus, Ohio 43201
                                Prepared for
                        Sineta Wooteh, Project Officer
                Margaret Conomos, Work Assignment Manager
                     Barry Nussbaum, Technical Advisor
                  Program Assessment and Outreach Branch
                    National Program Chemicals Division
                   Office of Pollution Prevention and Toxics
                    U.S. Environmental Protection Agency."..
                    1200 Pennsylvania Avenue NW (7404T)
                          Washington, D.C. 20460
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines.' It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
                              BATTELLE DISCLAIMER

 This report is a work prepared for the United States government by Battelle. In no event shall
   either the United States government or Battelle have any responsibility or liability for any
  consequences of any. use, misuse, inability to use, or reliance upon the information contained
   herein, nor does either warrant or otherwise represent in any way the accuracy, adequacy,
                      efficacy, or applicability of the contents hereof.
                              ACKNOWELDGEMENTS

The EPA and the authors thank the organizations whose contributions made this report possible
    including the Lead Poisoning Prevention Branch at the Centers for Disease Control and
     Prevention, the Childhood Lead Poisoning Prevention Program at the Massachusetts
Department of Public Health, and the Office of Healthy Homes and Lead Hazard Control at the
                    U.S. Department of Housing and Urban Development.

    This report was based on work conducted by Battelle, with significant contributions from
  Warren Strauss, Tim Pivetz, Elizabeth Slone, Jyothi Nagaraja, Rona Boehm, Michael Schlatt,
           Darlene Wells, Jennifer Zewatsky, Michele Morora, and Bruce Buxton.
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
                                   TABLE OF CONTENTS
EXECUTIVE SUMMARY	v

1.0    INTRODUCTION	1
       1.1     Background and Purpose of Study	1
       1.2     Study Objectives	2
              1.2.1   Objective 1 — Combine and Manage Multiple Data Sources	2
              1.2.2   Objective 2 - Conduct Analyses to Identify Predictive Variables and
                     Model Children's Blood Lead Levels	2
              1.2.3   Objective 3 - Develop Visualization Tool to Graphically Model Predicted
                     Blood Lead Levels	2

2.0    STUDY METHODOLOGY	3
       2.1     General Approach	:	3
       ,2.2     Data Management	3
       2.4     Development of Multivariate Statistical Models	9
              2.4.1   Statistical Models for the Broad Coverage - Lower Resolution Model	9
              2.4.2   Statistical Models for the High Resolution Model within Massachusetts	11

3.0    DATA SOURCES AND DATABASE DEVELOPMENT	13
       3.1     Children's Blood Lead Measurements	 13
       3.2     Demographic Data	'.	15
       3.3     Environmental Data	20
              3.3.1   Concentrations of Lead in Air	21
              3.3.2   Toxics Release Inventory Data	22
              3.3.3   Water Quality Data	23
       3.4     Programmatic Data	23
              3.4.1   Programmatic Funding Variables	24
              3.4.2   EPA Region	24
              3.4.3   Housing Inspection Data (Massachusetts)	24
       3.5     Data Linkages	26

4.0    EXPLORATORY DATA ANALYSES	28
       4.1     Relationship between National Blood Lead Data and Explanatory Variables	28
       4.2     Relationship between Local Blood Lead Data and Explanatory Variables	32

5.0    STATISTICAL MODELING RESULTS	36
       5.1     Low Resolution Modeling Results	36
       5.2     High Resolution Modeling Results	52

6.0    GRAPHICAL PRESENTATION OF MODELING RESULTS..	64
       6.1     Maps of Observed and Predicted Blood Lead Outcomes	64
       6.2     Visualization Tool Development	69

7.0    DISCUSSION AND FUTURE WORK	72
       7.1     Major Findings	72
       7.2     Comparison of National Results and NHANES	73
       7.3     Data Issues	'.	75
              7.3.1   Biases from Geocoding	75
              7.3.2   Reporting Limits in Surveillance Data	75
              7.3.3   Selection Bias in Surveillance Data	76
              7.3.4   Limitations of Ecological Models for Predicting Within-Area Relationships	77
                                             i
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
              7.3.5   Use of 2000 Census Data and Other Time Invariant Data as Predictors	77
       7.4    Model Validation Issues	77
       7.5    Other Recommendations for Immediate Future Work	80

8.0    REFERENCES 	82


Appendix A    Exploratory Analysis Summary Pages	A-1

Appendix B    Massachusetts Data: Exploratory Analysis Summary Pages	B-1

Appendix C    Detailed Exploratory Analyses of 95th and 99th Percentile Variables
              In National Models	C-1

Appendix D    Detailed Discussion of National Exploratory Analyses	D-1

Appendix E    Detailed Discussion of Massachusetts Exploratory Analyses	E-1

Appendix F    U.S. Counties ad Massachusetts Census Tracts with Highest Predicted BLLs	F-1

Appendix G    Detailed Maps of National and State Model Outputs	G-1

Appendix H    Data Dictionaries for National and Massachusetts Databases	H-1


                                       LIST OF TABLES

Table 3-1      Initial Variables for Analysis Created From the 2000 Census	16
Table 4-1      Summary of Log-likelihood Ratios from each Model Fit to all Potential Explanatory
              Variables, National Data	30
Table 4-2      Summary of Log-likelihood Ratios from each Model Fit to all Potential Explanatory
              Variables, Massachusetts Data	33
Table,5-1      Summary of Variables Included in Final National Multivariate Model	39
Table 5-2      Model 1 (Unweigted GM) Parameter Estimates for Multivariate National Model	40
Table 5-3      Model 2 (Weighted GM) Parameter Estimates for Multivariate National Model	42
Table 5-4      Model 3 (Proportion >5 ug/dL) Parameter Estimates for Multivariate National Model	44
Table 5-5      Model 4 (Proportion >10 ug/dL) Parameter Estimates for Multivariate National
              Model	!	46
Table 5-6      Model 5 (Proportion >150 ug/dL) Parameter Estimates for Multivariate
              National Model	48
Table 5-7      Model 6 (Proportion >25 ug/dL) Parameter Estimates for Multivariate
              National Model	50
Table 5-8      Summary of Variables Included in Final Massachusetts Multivariate Model	54
Table 5-9      Massachusetts Multivariate Model Estimates	55


                                       LIST OF FIGURES

Figure 5-1     Histograms of Residuals from Fitted National Multivariate Model 1	41
Figure 5-2     Plot of National Multivariate Model Predicted Values versus Observed with Fitted
              Regression Line and 45ฐ Reference Line for Unweighted Geometric Mean Response.. 41
Figure 5-3     Histograms of Residuals from Fitted National Multivariate Model 2	43

                                              ii
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
Figure 5-4      Plot of National Multivariate Model Predicted Values versus Observed with Fitted
               Regression Line and 45ฐ Reference Line for Weighted Geometric Mean Response	43
Figure 5-5      Histograms of Residuals from Fitted National Multivariate Model 3	45
Figure 5-6a    Plot of National Multivariate Model Predicted Values versus Observed with Fitted
               Regression Line and 45ฐ Reference Line for Proportion of Children with
               BLL>5ug/dL.:	45
Figure 5-6b    Plot of National Multivariate Model Predicted Values versus Observed with Fitted
               Regression Line and 45ฐ Reference Line for Proportion of Children with
               BLL > 5 ug/dL (Logit Scale)	45
Figure 5-7      Histograms of Residuals from Fitted National Multivariate Model 4	47
Figure 5-8a    Plot of National Multivariate Model Predicted Values versus Observed with Fitted
               Regression Line and 45ฐ Reference Line for Proportion of Children with
               BLL > 10 ug/dL	;	47
Figure 5-8b    Plot of National Multivariate Model Predicted Values versus Observed with Fitted
               Regression Line and 45ฐ Reference Line for Proportion of Children with
               BLL > 10 ug/dL (Logit Scale)	47
Figure 5-9      Histograms of Residuals from Fitted National Multivariate Model 5	49
Figure 5-1 Oa   Plot of National Multivariate Model Predicted Values versus Observed with Fitted
               Regression Line and 45ฐ Reference Line for Proportion of Children with
               BLL>15ug/dL	49
Figure 5-1 Ob   Plot of-National Multivariate Model Predicted Values versus Observed with Fitted
               Regression Line and 45ฐ Reference Line for Proportion of Children with
               BLL > 15 ug/dL (Logit Scale)	49
Figure 5-11     Histograms of Residuals from Fitted National Multivariate Model 6	51
Figure 5-12a   Plot of National Multivariate Model Predicted Values versus Observed with Fitted
               Regression Line and 45ฐ Reference Line for Proportion of Children with
               BLL > 25 ug/dL	51
Figure 5-12b   Plot of National Multivariate Model Predicted Values versus Observed with Fitted
               Regression Line and 45ฐ Reference Line for Proportion of Children with
               BLL > 25 ug/dL (Logit Scale)	51
Figure 5-13    Histograms of Residuals from Fitted Massachusetts Multivariate Model 1	58
Figure 5-14    Plots for Predicted versus Observed Values with 45ฐ line from Fitted Massachusetts
               Multivariate Model 1	58
Figure 5-15    Histograms of Residuals from Fitted Massachusetts Multivariate Model 2	59
Figure 5-16    Plots for Predicted versus Observed Values with 45ฐ line from Fitted Massachusetts
               Multivariate Model 2	59
Figure 5-17    Histograms of Residuals from Fitted Massachusetts Multivariate Model 3	60
Figure 5-18a   Plot of Massachusetts Multivariate Model Predicted Values versus Observed with
               Fitted Regression Line and 45ฐ Reference Line for Proportion of Children with
               BLL>5ug/dL	:	60
Figure 5-18b   Plot of Massachusetts Multivariate Model Predicted Values versus Observed with
               Fitted Regression Line and 45ฐ Reference Line for Proportion of Children with
               BLL > 5 ug/dL (Logit Scale)	60
Figure 5-19    Histograms of Residuals from Fitted Massachusetts Multivariate Model 4	61
Figure 5-20a   Plot of Massachusetts Multivariate Model Predicted Values versus Observed with
               Fitted Regression Line and 45ฐ Reference Line for Proportion of Children with
               BLL>10ug/dL.	;	61
Figure 5-20b   Plot of Massachusetts Multivariate Model Predicted Values versus Observed with
               Fitted Regression Line and 45ฐ Reference Line for Proportion of Children with
               BLL > 10 ug/dL (Logit Scale)	61
Figure 5-21     Histograms of Residuals from Fitted Massachusetts Multivariate Model 5	62
                                               m
This information-is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
Figure 5-22a    Plot of Massachusetts Multivariate Model Predicted Values versus Observed with
               Fitted Regression Line and 45ฐ Reference Line for Proportion of Children with
               BLL>15ug/dL'.	:	62
Figure 5-22b    Plot of Massachusetts Multivariate Model Predicted Values versus Observed with
               Fitted Regression Line and 45ฐ Reference Line for Proportion of Children with
               BLL > 15 ug/dL (Logit Scale)	62
Figure 6-1      Observed and Predicted GM Blood Lead Levels in the U.S. by County, 2000 and
               2005	65
Figure 6-2      Observed and Predicted GM Blood Lead Levels in Region V by County, 2000 and
               2005	66
Figure 6-3.      Observed and Predicted Proportion of Children with Blood Lead Levels > 10 ug/dL
               in Region V by County, 2000 and 2005	67
Figure 6-4      Observed and Predicted Proportion of Children with Blood Lead Levels > 10 ug/dL
               in Massachusetts by Census Tract, 2000 and 2005	68
Figure 6-5      Response Surface of Predicted Geometric Mean Blood-Lead Concentration Across
               the State of Illinois from the Visualization Tool	70
Figure 6-6      Time Series Plot of Observed and Predicted Geometric Mean Blood-Lead
               Concentration in Cook County Illinois from the Visualization Tool	71
Figure 7-1      Comparison of National Surveillance Data to NHANES Data	74
                                               IV
This information is distributed solely for the purpose ofpre-dissenu'nation peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
                                EXECUTIVE SUMMARY

This pilot study seeks to develop statistical models to predict risk of childhood lead poisoning
within specified geographic areas based on a combination of demographic, environmental, and
programmatic information sources. Exposure factors associated with childhood lead poisoning
were investigated within census tracts for a community-focused set of models in Massachusetts,
as well as within counties across the U.S. in a series of national models. Aggregated summary
measures of childhood lead poisoning within defined geographic areas (census tracts and
counties) were used as the response variable in the statistical models, including geometric mean
blood-lead concentration, as well as proportion of children screened at or~above 5,10,15 and
25 ug/dL. These summary measures were constructed at 3-month (quarterly) intervals from
1995 through 2006 in counties across the nation using data from CDC's National Surveillance
Database, and from 2000-2006 in census tracts within the Commonwealth of Massachusetts
based on data provided by the Massachusetts Department of Public Health.

The results of this study suggest that longitudinal predictive  models can be developed  at the
county-level across the nation based on the use of quarterly summary information from CDC's
National Surveillance Database, and at the census-tract level within states that have a long
history of universal screening and reporting such as Massachusetts. These models  can be used to
descrAeJip^jrisk^childhopd lead ppj|oning^hangesj)ver  time within diffe^negio^~6f the
country, as well as within" small geographic a^aFwirninrstale^~(e7gTfc61im^
geographic areas within counties (e.g., census tracts). They can be used to predict the risk of
childhood lead poisoning in counties (or census tracts) with little or no surveillance data, and can
also be used to identify those counties (or census tracts) that are at highest risk at the end of the
period of observation.

The sta^sjti^;aJLmodeLchqsenj(a random effects^model_witii separatejntercepts and slopes
estimated within each county orcensus tract) also allowsTanking of geographiclareasjjased on
thejrate^fcdecline^gyej time jfter^accounting for tiiejixed effects variables of the model^
(although only among those areaTthaTprovided adequate surveillance data). These random
effects models were fit to the observed geometric mean blood-lead concentrations, as well as the
exceedance proportions within the context of a logistic regression model. Within the context of
the broad^based^nati^aljriodeMhesejandom_effi?cts a^lowjEP^^jdjntijyJhpse^cQunties that
are exp^riencingjjmore rapid reductionlnriskxrfdchildhoodlead poisoning over time  (to identify
best practices) and those co^ritieTthlrareexperiencing a significantly less rapid decline over
time (to identify areas in need of additional attention and resources for combating lead
poisoning), after already accounting for the demographic, programmatic and environmental
factors included in the multivariate model.

Within the series of national models at the county-level of geographic specificity, the data
suggest that there are significant differences in the distribution of childhood blood-lead
concentrations among the different regions of the country, and that the manner in which these
distributions change over time and are impacted by seasonality is also regionally specific. The
risk of childhood lead poisoning had a statistically significant downward trend over time in all
areas of the country with the exception of EPA Region 1.

                                           v
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
 After accounting for these regional differences, a number of demographic, environmental and
 programmatic variables were found to be highly predictive of childhood blood-lead
 concentrations among the different response variables modeled within this project. The specific
 variables that were found to be predictive within the multivariate models varied based on the
 response variable - however, there were certainly some variables that were found to be
 predictive in multiple models.  Iiuiddition tq_yarious censusjlempgraphicjyariables that were
 identified in previous riskjnod^igg^ejfi^^e^gy^^fji^^^g^p^r^nt-sin^le-parent families,
 race/ethTTici^^alOn^^
 models,~variables constrycted~from EPA's Safe Drinking Water Information System were
 identified in five'Sf-six-multi-variate models, and programmatic funding information from HUD
 was identified in all six multivariate models.  Variables indicating amount of lead emissions
i within counties from EPA's Toxics Release Inventory were only selected in one of , the six
Ymodels, likely due to the fact that these data are collinear with the air modeling data from EPA's
 1599 National Air Toxics Assessment.

 Within the context of the high resolution model developed using data from the Commonwealth
 of Massachusetts, a highly significant downward trend in the risk of childhood lead poisoning
 was also identified among the five models developed. Due to a very small number of children
 observed at or above 25 iig/dL within Massachusetts over the 2000-2006 period of observation,
 we were unable to fit this sixth model. After accounting for the long-term reduction over time
 and seasonality using similar methods that were employed in the broad-based national model,
 only the demographic and programmatic variables were included in the multivariate models for
 risk of childhood lead poisoning at the census tract level. Of particular interest were the
 variables that described the proportion of housing units within each census tract that were found
 to be in compliance and out of compliance with the Massachusetts Standard of Care.  In all five
 of the multivariate models, the risk of childhood lead poisoning was significantly-reduced as the
 proportion pj^h^sing-imi^                              census tract, fifaddition, for the
 last twomodels (which predicted proportion of children at orabove 10 and 15 ug/dL), the risk of
 childhood lead poisoning increased significantly as the proportion housing units out of
 compliance increased within a census tract.

 The observed and predicted values from the multivariate models (including predicted values
 where there were no observed surveillance data) were used to generate static maps using Arc-
 View software, and were loaded into a customized dynamic visualization tool that allows users
 to interact with the modeling results to assess how risk of childhood lead poisoning changes over
 time within specific regions of the country.  This tool will help EPA and others identify areas
 that remain at risk for childhood lead poisoning as we approach the 2010 goal of elimination of
 this preventable adverse health outcome.
                                            VI
 This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
 quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
 to represent any Agency determination or policy.

-------
1.0    INTRODUCTION

1.1    Background and Purpose of Study

Over the past 15 years, various childhood lead poisoning prevention programs (CLPPPs)
throughout the U.S. have conducted analyses of their screening data to develop "risk indices," or
mathematical models for predicting the prevalence of childhood lead poisoning in different
geographic areas within their regions of concern. These modeling efforts are generally intended
to both characterize the extent of the prevalence of childhood lead poisoning within their
geographic areas and support the development of targeted screening and outreach plans in order
to reach the 2010 goal of eliminating childhood lead poisoning throughout the U.S.

To date, the majority of modeling efforts have focused on combining screening information and
demographic data available from the U.S. Census. Previous studies have combined childhood
surveillance data (aggregated at the zip-code or census tract level) with demographic predictor
variables from the Census Bureau for the purposes of targeting geographic areas at higher risk of
childhood lead poisoning (Miranda, Dolinoy, and Overstreet 2002; Miranda et al. 2005rStrauss,
Nahhas et al. 2001). These studies have led to recommendations for using age of housing and
percent of population below the poverty line for targeting neighborhoods that may be of
increased risk for childhood lead poisoning (CDC 1997). Numerous studies have also been used
to document the relationship between children's blood-lead concentrations and measures of lead
in residential environmental media (dust, soil, air, water, and food) (HUD 1995; Lanphear et al.
1998; Strauss, Carroll et al. 2001). These studies have contributed to EPA and HUD regulations
and policies for identifying and reducing residential childhood lead exposures (24 CFR Part 35;
40 CFR Part 745; 40 CFR Part 745; U.S. Department of Housing and Urban Development
September.15, .1999).  Other studies have combined blood-lead surveillance data with
programmatic information on housing units treated to determine the positive impact of housing-
based intervention programs (Strauss et al. 2006).

The goal of this study is to explore models based on a hierarchical combination of demographic,
environmental, and programmatic information sources in order to predict the number of children
at risk of elevated blood lead levels for a given geographic area. While the models are highly
dependent on available data, this study provides a statistical methodology that combines each
data source in an appropriate manner, adjusting for global and local trends over time.  In doing
so, the models build upon concepts of hierarchical modeling and longitudinal data analysis.

As EPA, CDC, and other federal and state agencies prepare to meet the 2010 goal of eliminating
childhood lead poisoning, this pilot study of integrating several different types of data sources
hopefully improves the predictive power of models that rely on a single information source. This
allows for more efficient targeting of those geographic areas that need the most help in
eliminating childhood lead poisoning.
                                           1
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
1.2    Study Objectives

1.2.1   Objective 1 - Combine and Manage Multiple Data Sources

The first objective of the study was to combine multiple sources of information in order to assess
the impacts of various factors on children's blood lead levels. The study had to obtain and
manage data relating to blood lead levels, environmental exposure, demographic characteristics,
and programmatic support to state and local childhood lead-poisoning prevention efforts.
Missing, incomplete, or error-prone data were identified for each data source and steps were
taken to resolve data problems. Databases were developed to store and later combine each data
source in a manner that supported the development of predictive models. Master databases that
integrated multiple data sources were developed to enable efficient access to data required for
statistical analyses.  A data dictionary was prepared to document the various study databases.

1.2.2   Objective 2 - Conduct Analyses to Identify Predictive Variables and Model
       Children's Blood Lead Levels

The second study objective was toj^duct_statisricaLaj^y^es-inrorder-tO:developjnodels that
are predictive of risk of childhood leadppisoningjwi thin defined^ geographic areas as a function
of various^iffel^feTwironmentalf^^                                   As part of this
obje^tivera=National-model-was-de^lopedrf6r^re^ictin^sk^rthe county-level based on
surveillance data from the U.S. Centers for Disease Control and Prevention (CDC), and a local
model  was developed at the census tract level using blood-lead surveillance data from within the
Commonwealth of Massachusetts. As part of the model building process at both the National
and Local levels, the various data sources underwent exploratory analyses to investigate data
distributions, identify relationships between variables, and determine appropriate variables to
include in subsequent statistical models. Part of the exploratory analyses included an effort to
identify which environmental, programmatic, and demographic factors were most predictive of
risk of childhood lead poisoning.  Multivariate statistical models were then developed using
appropriate statistical software to combine the various data sources within a single model-that
accounted for trends in risk of childhood lead poisoning over time within defined geographic
areas.  Model diagnostics were reviewed, and models with the best fit were identified.

1.2.3   Objective 3 - Develop Visualization Tool to Graphically Model Predicted Blood
       Lead Levels

The third stodyxo^e^vejwasJo-dejfejop,an=appj^r^itevisuajizationtopl that allows users to
interact with thereluTteof^ejtetisticalrmodel-predi^tin^hildreTi's~blood lead levels across the
United^StatesFTfii^tool^roVides the user with the flexibility to visually compare the predicted
blood lead levels across areas of the country and also to drill down into individual counties or
census tracts to assess the input data that generated the predicted value.
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
2.0    STUDY METHODOLOGY

2.1    General Approach

This pilot study sought to develop models to predict the number of children at risk of elevated
blood lead levels for a given geographic area based on a hierarchical combination of
demographic, environmental, and programmatic information sources. Doing so required looking
at both the mechanisms of childhood lead risk assessment and control activities at the local level
as well as at broad trends across the U.S.  The two main analysis goals correspond to developing
predictive models at two different levels of geographic specificity, and appear as follows:

    1.     Broad Coverage (Lower Resolution) Model: This type of model is intended to be
          able to characterize broad trends over time in the prevalence of childhood lead
          poisoning at the county level across the entire  U.S.  This model was based on
          quarterly county-level aggregated surveillance data from the CDC and augmented
          with environmental, demographic, and programmatic (level of financial spending)
          information.

    2.     High Resolution Model: This type of model represents the effort to assess the relative
          contribution of various exposure sources associated with elevated blood-lead
          concentrations within select communities.  This type of model certainly pays homage
          to the concept that exposure that contribute to  childhood lead-poisoning are likely to
          be community specific. This analysis goal was met through modeling Census-tract
          level surveillance data within Massachusetts as well as housing unit lead assessment
          and/or control activities. These data sources were augmented with all of the
          environmental, demographic, and programmatic information used in the national
          model with the addition of state programmatic funding levels.

The primary objective of this pilot study was to utilize combined information from different
sources at various levels of geographic and temporal specificity to more accurately target
geographic areas at high risk for not meeting the 2010 goal of eliminating childhood
lead-poisoning. As such, the study required careful integration of a variety of data sources with
various characteristics and documentation. Data to support this study were gathered from
multiple  sources, including federal, state and local lead poisoning prevention programs, as well
as publicly available data that were downloaded from the internet (e.g., Census data, EPA's
Toxic Release Inventory, etc.).

2.2    Data Management

When each data source was received, the data and supporting documentation were reviewed to
gain knowledge on the structure, relationship and quality  of the data. Battelle database managers
worked with the project team to determine the final format for each database; desired uses of the
databases; as well as the requirements for maintaining the databases. Based on this information,
master databases were constructed in SQL server for both the national low-resolution model and
for the high-resolution model based on Massachusetts data that integrated the various
environmental, demographic, and programmatic variables, and facilitated statistical analyses of
                                           3
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
the combined data. These datasets were translated directly to SAS datasets for statistical
analysis, and were also transferred to Microsoft Access for delivery to EPA. The Microsoft
Access database includes a compact version of each database utilized in the statistical analysis,
with any extraneous variables removed. In addition, the Microsoft Access database also includes
a copy of the integrated longitudinal dataset used to support the final multivariate models
developed within this project.

Throughout the development process, Battelle conducted checks for completeness on all study
databases, and worked with data-sharing collaborators and EPA to attempt to complete missing
data as necessary to complete the proposed statistical analyses. Any changes to the databases
(corrections, additions, deletions, etc.) were documented in appropriate meta-data files, and
reported to EPA within the data dictionary attached to this report as Appendix H. As part of
constructing and maintaining these databases, Battelle will develop appropriate documentation of
the combined master databases.

Battelle followed Standard Operating Procedures (SOPs) to ensure the proper storage, backup,
and retrieval of datasets created and analyzed for this study. Additional details of these SOPs
can be found in the Quality Management Plan prepared for this project (Battelle  2007).

2.3    Descriptive Data Analyses

The analysis began with an assessment of the study sample, i.e., the proportion of counties and
census tracts in the sample with complete data for both die response variable and the explanatory
variables.  Prior to the fitting of any descriptive statistics to assess the predictive ability of any of
the explanatory variables, the blood-lead response variables needed to be constructed based on
the CDC and Massachusetts blood-lead surveillance data.  These data sources contain
information on individual blood-lead testing results on children, and were aggregated into
quarterly summary statistics (number of children observed, arithmetic and geometric mean, and
number of children observed at or above 5,10,15, and 25 ug/dL) at the county-level (for the
CDC data) and the .census tract level (for the Massachusetts data). An executable was developed
to extract these quarterly summary statistics from each county from CDC's SQL server database
for children aged 6-36 months, and a similar executable was deployed to create parallel summary
statistics at the census tract level for the Massachusetts surveillance data.  Because, of
confidentiality restrictions, county/quarter (or census tract/quarter) combinations with fewer than
5 observations were automatically  eliminated from the dataset. Battelle also eliminated data
reported prior to 1995 from the analysis database prior to statistical analysis.

Once the aggregated summary datasets were constructed, they were reviewed for possible
problems associated with childhood lead poisoning prevention programs not following universal
reporting protocols (for some localities, data was only transmitted to the CDC National-
Surveillance Database for children with elevated blood-lead concentrations over certain periods
of time). A screening algorithm was developed to remove these suspect data from the analysis
dataset - resulting in the elimination of less than 3 percent of the aggregate summary records
from the National database. The screening algorithm was also applied to the Massachusetts data
- however no records were eliminated, as Massachusetts was following universal screening and
reporting guidelines over the 2000-2006 time period for which they provided data. Additional
                                           4
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
detail on the manner in which the blood-lead response variables were constructed can be found
in Section 3.1.

In preparation for developing longitudinal statistical models, univariate summaries of each
variable as a function of time were generated and comparisons made of these distributions using
side-by-side box-plots for continuous data or bar-charts for categorical data.  This helps verify
that the data are clean and ready for analysis and identify cells with sparse data.  Such descriptive
analyses  were conducted on each database, to characterize the distributions of all observed
variables using  frequency distributions for categorical variables, and simple summary statistics
(mean, median, mode, minimum, maximum, and select percentiles) for continuous variables.
Distributional assumptions may also be explored for certain variables, as appropriate, in
preparation for more sophisticated models. For example, some environmental concentration data
may depart from normality, and follow a log-normal distribution.  In these cases, we may
additionally report the geometric mean and geometric standard deviation as part of the simple
descriptive summary.

The univariate descriptions were then followed by fitting a series of cross-sectional bivariate
relationships between the blood-lead response variable(s) and each candidate explanatory
variable.  These cross-sectional relationships were explored as a function of time to better
understand the stability of these relationships, and whether they change over time, so that they
can be modeled appropriately in the more sophisticated longitudinal analyses. These analyses
will also  help identify which explanatory variables are most predictive of the blood-lead
response variable.

In preparation for more sophisticated statistical analyses, such as the Generalized Linear Mixed
Logistical Regression Model outlined below, we also performed relevant stratified analyses to•
investigate interactions discussed in the data analysis plan. For examplerwe investigated the
population_density-variable in this mannerras density may serve as a surrogate to differentiate
between rural and urban geographic areas in the analyses - and exposure variables may be
different  in these types of areas. Similarly, we investigated EPA regions as a potential
stratification variable.  If variation in the measure of effect is not observed (e.g., odds ratios)
across the levels of a third variable, however,  we know we can probably treat that third variable
as a potential confounder in the multivariate model, rather than as an effect modifier. If the odds
ratios differ markedly—e.g.,  the effect appears to be protective in one subgroup and hazardous in
another subgroup—we know we must consider the third variable as an effect modifier.

Specific variables within each type were explored using four general approaches - (1)
histograms or side-by-side boxplots of the candidate explanatory variable, (2) simple linear
regression line plots exploring the relationship between predicted GM blood lead levels and the
explanatory variable for each of the three specified time periods, (3) distributional summaries of
the explanatory variable across the three time periods, and (4) statistical modeling of the
relationship between the explanatory variable and various blood-lead response variables after
adjusting for the effect of time.
                                           5
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
Appendices A and B presents these four analyses for each explanatory variable investigated for
the national and Massachusetts data, respectively. The four main analyses are described in more
detail below.

Histograms or Side-by-Side Boxplots of Potential Explanatory Variables
Using one record for each quarterly county- or Census tract-level data point, a histogram
illustrating the distribution of the explanatory variable is presented. A fitted line assists with
assessing the distribution of each potential explanatory variable (e.g., whether the data are
approximately normally distributed). Histograms were plotted for potential predictor variables
that were time invariant.  For predictor variables that varied over time within the analysis dataset,
side-by-side boxplots were used to characterize the distribution over the time-periods, using an
average of the predictor variable across the quarters in which blood-lead concentration were
observed within each time-period and area.

Plots of Predicted GM Blood Lead Levels and Explanatory Variables
The county- and Census tract-level quarterly blood lead data were fit to each explanatory
variable to generate predicted GM blood lead levels across the range of the explanatory variable
for each of the time periods. A simple linear regression line plot summarizes this analysis with
one line for each time period. This analysis allows comparison of the relationship between the
explanatory variable and predicted blood lead level trends across time periods. If the
relationship is stable across time, roughly parallel lines are evident. If the effect of the variable
on blood-lead varies over time, non-parallel (and perhaps intersecting) lines are observed. In this
case, separate slopes may need to be fit for these variables over different periods of time in the
more sophisticated longitudinal analyses.

Distributional Summaries
The first table presented for each explanatory variable contains a series of summary statistics for
each of the time periods including sample size, number missing, mean, and standard error.  The
sample size is relative to the number of quarters represented in the analysis dataset - and
therefore, these distributions correspond to the analysis dataset (and not necessarily to the
distribution of the variable across the nation or state).  The distribution of the data for each time
period is also presented (minimum, median, and maximum and 10th, 25th, 75th, and 90th
percentiles).  Comparing the summary data across time allows assessment of changes in the
explanatory variable over time for the groups of tracts included in the analysis for each time
period. Generally, the mix of counties and Massachusetts Census tracts included in each of the
time periods is similar, so that the distribution of the data from each period is also similar.

Statistical Modeling of Relationship between Explanatory Variables and GM Blood Lead
Levels and Exceedance of Blood Lead Thresholds
For each explanatory variable,  six models were run and are presented, using a mixed models
analysis of variance approach and a generalized linear mixed models approach as detailed below:

o A mixed models analysis of variance (i.e., a random effects model for continuous data) was
   usecftomodellhe geometric mean (GM) as a function of predictor variables as follows:
                                            6
 This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
 quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
 to represent any Agency determination or policy.

-------
          Where (i indexes county or Census tract, j indexes time), the beta parameters (P)
          represent a vector of fixed effects, and the delta parameters (5) represent random-
          effects that allow each county or Census tract to have their own trend over time. The
          Xy variable is typically mean-centered in this series of models, allowing the intercept
          term to be relatively stable across the multiple predictor variables being investigated.
          In this model it can be assumed that 801 and 8u jointly follow a multivariate normal
                                                              [cr2   cr21
                                                                y    " , and the residual

          error is also assumed to follow a normal distribution with mean zero and variance
       •  Model 1 follows the above approach - where the responses are weighted equally
       •  Model 2 follows the above approach - where the responses (GM) are weighted by the
          number of children observed (screened) within each record (county or Census
          tract/quarter).

o  .Generalized Linear Mixed Models (i.e., a random effects model for binomial data) that
   models-me'propoTtibn of children exceeding certain thresholds as a function of predictor
   variables:
          Where (i indexes county or Census tract, j indexes time), the beta parameters (P)
          represent a vector of fixed effects, and the delta parameters (5) represent random-
          effects that allow each county or Census tract to have its own trend over time.
          Similar to the above mixed modeling approach, the Xy variable is also mean-centered
          in this series of models, allowing the intercept term to be relatively stable across the
          multiple predictor variables being investigated. In this model, it can be assumed that
          5oi and 81, jointly follow a multivariate normal distribution with mean zero and
                               Fo?i
          covariance matrix 2 =   1
       •  Model 3 will follow the above approach - where Yy represents the number of
          children observed with blood-lead concentrations at or above 5 ug/dL, and ny
          represents the total number of children screened within each record (county or Census
          tract/quarter).
       •  Model 4 will follow the above approach - where Yy represents the number of
          children observed with blood-lead concentrations at or above 10 ug/dL, and ny
          represents the total number of children screened within each record (county or Census
          tract/quarter).

                                            7
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
       •  Model 5 will follow the above approach - where Yy represents the number of
          children observed with blood-lead concentrations at or above 15 ug/dL, and n,j
          represents the total number of children screened within each record (county or Census
          tract/quarter).
       •  Model 6 will follow the above approach - where Yy represents the number of
          children observed with blood-lead concentrations at or above 25 ug/dL, and n,j
          represents the total number of children screened within each record (county or Census
          tract/quarter).

To allow comparison of the different variables explored within each variable type, tables are
included in Section 4 that present the log-likelihood statistic from each model run and presented
in Appendices A and B. Within each variable category, the variable that provided the best fit for
each of the six models is highlighted in yellow. To ensure compatibility in the likelihood based
statistics being used to make comparisons among the different candidate predictor variables,
missing values for predictor variables were imputed using the mean of the distribution. The
number of imputed values that were necessary is provided by the nmiss column in the table of
distributional summaries described above.  Those variables highlighted in yellow have the
largest likelihood statistic, and will likely become candidate predictors for the multivariate
statistical models.

Due to the iterative nature and complexity of the Mixed Models Analysis of Variance and
Generalized Linear Mixed Modeling Approaches, these models did not always converge.
Models that failed to converge for a particular predictor variable are discussed in the results
sections within Appendices D and E, and are also indicated in Tables 4-1 and 4-2 as well as the
summary pages of Appendices A and B by blank cells. Cases in which model convergence is not
attained will likely translate to exclusion of that particular variable  when building the
multivariate model. Note that because of the sparseness of data for children with blood lead
levels at or above 25 ug/dL within the Massachusetts data, Model 6 failed to converge across all
variables. Thus, Model 6 results are not presented or discussed for the Massachusetts models.
                                            8
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
2.4 (   Development of Multivariate Statistical Models

2.4.1  Statistical Models for the Broad Coverage - Lower Resolution Model

This model is being used to characterize broad trends over time in the prevalence of childhood
lead poisoning across the entire U.S.  The various surveillance, environmental sources,
demographic characteristics, and programmatic support data sources were aggregated to the
county level for all localities with universal screening and reporting. Quarterly estimates of each
candidate predictor variable were created for each county within the U.S., including those
county/quarter combinations that did not include observed blood-lead response variable
information (allowing us to extrapolate the model predictions to geographic areas and time-
points that were not represented within CDC's National Surveillance Database).

For the purposes of discussion, we assume that the modeling approach will focus on a logistic
regression model for the proportion of children that have elevated blood-lead concentrations
(>10 ug/dL).  The temporal nature of declining childhood lead poisoning will be addressed via
classic concepts of longitudinal data modeling of the low resolution data.  Let

   Yy represent the number of children that were detected with blood-lead concentration above
    10 ug/dL from the i* county and j* point in time (quarter),

   n\j represent the number of children that had their blood-lead concentration tested from within
   the i  county and j* point in time (quarter),
   Please note that we expect that %<#,;/, where Afy represents the total population of children
   in the i"1 county and j point in time.

   t,j represent time (in years) corresponding to the Yy response variable, and

   Xy represent a series of predictor variables associated with the Yy response variable. These
   predictor variables may represent air monitoring data, drinking water data, census
   demographic data, programmatic data on federal financial support for lead poisoning
   prevention, and other related information as detailed above that can potentially help predict
   the prevalence of lead poisoning at the county level.

We introduce the following as a potential baseline model:
Where the beta parameters (P) represent a vector of fixed effects, and the delta parameters (5)
represent random effects that allow each county to have their own trend over time. Li this1 model
it can be assumed that 60i and 61 i jointly follow a multivariate normal distribution with mean zero
                           2     2
                            2     2 "1
and covariance matrix ฃ =    "     " |  .
                         _<72,   <722J
This.informatipn is distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
Counties with larger 5i parameters estimates represent areas where lead-poisoning has not
significantly decreased over time.  Similarly, the parameter estimates can be used to identify
those counties with the highest predicted prevalence of childhood lead poisoning at various time
points in the future.

In building the multivariate statistical model for the Broad-Based Modeling Objective, Battelle
first evaluated the predictive ability of each candidate predictor variable that was considered
within the exploratory analyses. For the environmental predictor variables in particular
(information from EPA's 1999 National Air Toxics Assessment, Safe Drinking Water
Information System, and Toxics Release Inventory), the data were largely concentrated at zero.
We therefore also investigated a series of zero/one indicator variables that represent
county/quarter combinations at or above the 95th and 99th percentile of observed values of these
environmental predictor variables within the analysis dataset.

Once the predictive ability of each candidate variable was established within the exploratory
analyses described earlier, candidate predictor variables were classified into groups (e.g.,
housing age, income, education, air modeling, programmatic financial support, etc.) and then the
single best predictor variable within each group was selected for possible inclusion within each
of the six multivariate statistical models being developed.  If the selected variable demonstrated a
relationship with risk of lead poisoning that changes over time (as evidenced by intersecting lines
in the plots generated in the exploratory analyses), then this interaction was taken into
consideration within the evaluation of the predictive ability of the candidate variable(s).

In addition to investigating the predictive ability of each potential environmental, programmatic
and demographic variable, various different stratification variables (region of the country,
population density)  and covariates (time trend and seasonally) were investigated. As a result, all
six multivariate statistical models adjust for a categorical variable that differentiates among the
risk of childhood lead poisoning within  the 10 EPA regions.  Within each EPA region, a separate
intercept, trend over time, and seasonally terms (based on fitting intercepts for each quarter of
time) was included in the multivariate statistical model.

Our approach to determining which environmental, programmatic, and demographic variables
were included in the model followed a backwards elimination process - in which each variable
group's best predictor variable identified earlier was included in the first model - with variables
being eliminated from the model when they were not deemed to be highly significant. This
model building process was also aided by investigation of the selected environmental,
programmatic, and demographic variables for issues of potential colinearity via investigation of
correlation matrices and principal components analysis. The resulting multivariate statistical
models were parsimonious - and in most cases only included variables that were highly
statistically significant.  In a few cases,  a variable was left in the model without being highly
significant - because its elimination caused a large drop in the model log-likelihood (suggesting
that the model is significantly improved with the addition of a variable whose slope is not
significantly different from zero).
                                            10
This, information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
Once the'multivariate statistical models were developed, model fit diagnostics were evaluated
and documented.

The parameter estimates for the six National Multivariate Statistical Models are provided in the
results section.  The results of these models are also explored in multiple ways.  Maps were
generated to demonstrate observed and predicted geometric mean blood-lead concentration and
proportion of children at.or above 10 ug/dL within each EPA Region for the Years 2000 and
2005   (data were appropriately averaged across the four quarters in each of these years prior to
mapping). Lists were also generated to identify the 150 highest risk counties across the U.S. at
the end of the observation period (2006) as predicted by each of the six models, as well as the 10
highest risk counties within each state.

Finally, the predicted-values from these multivariate statistical models (extrapolated to
county/quarter combinations not represented in the CDC surveillance database) were integrated
in a unique data visualization tool.  The product of this effort will be a time-series of maps (or a
movie) mat spatially interpolates risk of childhood lead poisoning as a function of various
appropriate predictor variables The visualization tool (being delivered to EPA as an executable)
will allow users to interact with the modeling results at different levels of temporal and
geographic specificity.  The tool will allow the user to select an appropriate response variable
(e.g. proportion of children with blood-lead concentrations above 5 ug/dL) and play a movie that
displays a time-series of maps that displays how the predicted (or observed) risk changes over
time across the  various counties within a selected state.  The user will also be able to zoom in on
a rectangular area, to see these results with a higher degree of geographic specificity.  The user
will also be able to stop the movie (or rewind, fast-forward) to isolate specific points in time.  By
using the mouse, the user will also be able to select a specific county and the tool will then
display the observed and predicted data for that particular county in a separate window.  The
visualization tool was written in C++, and was built in a manner that will allow EPA to modify
the model and for Battelle to quickly import the resulting data from a modification into the tool.

2.4.2   Statistical Models for the High Resolution Model within Massachusetts

High-resolution models will be utilized to identify the relative contribution of various types of
exposure sources in elevated risk for childhood lead-poisoning within select communities within
the Commonwealth of Massachusetts.  These types of sources include housing factors, broader
environmental exposure, demographic composition, and  programmatic resources.  While this
type of model pays homage to the concept that exposures contributing to childhood lead-
poisoning are likely community-specific, analysis of the  high-resolution models may have
certain limitations including selection bias and generalizability to other geographic areas.

The Massachusetts Department of Public Health (MDPH) entered into a limited use data sharing
agreement with Battelle, allowing them to provide blood-lead testing results on individual
children (aged 6-36 months) and housing inspection data in a format that preserves linkages
through a housing unit identification variable.  These data will be utilized in two different
modeling approaches. The first modeling approach will seek to develop census tract quarterly
summary measures similar to the National Model for blood-lead (e.g. exceedance proportions

                                           11
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
and geometric means), as well as summary measures for the proportion of housing units in each
census tract that are known to be in (or out of) compliance with the Massachusetts standard of
care (for use as a potential explanatory variable).  MDPH has also provided Battelle with
summary information regarding HUD and State funding of residential housing interventions
(lead hazard control and abatement) - which will be used to develop a longitudinal summary of
current and cumulative per-capita spending on residential intervention within each census tract
(using various assumptions on the allocation of such dollars). Other explanatory variables, such
as the census, toxic release inventory, 1999 National Air Toxics Assessment, and water quality
data will be available for use in these models.

These census-tract level summary data (both response variable and explanatory variables) were
modeled using a similar approach as what is being proposed for the National (Low Resolution)
Model - only the unit of clustering was census tract rather than county.
                                            12
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed
to represent any Agency determination or policy.

-------
3.0    DATA SOURCES AND DATABASE DEVELOPMENT

The main two goals of the statistical analysis were to develop a predictive model(s) to implement
targeted interventions based on a better understanding of (1) the relative importance of various
exposure sources in addition to leaded paint in housing and (2) the geographic areas across the
U.S. that remain at increased risk for childhood lead- poisoning. To do so, blood-lead data were
combined with various environmental, demographic, and programmatic datasets at different
levels of geographic specificity and coverage. A description of each of these data sources as
well as discussion of how. they were combined is included in this section.

3.1    Children's Blood Lead Measurements

The statistical models are based upon blood-lead levels of children corresponding to the various
geographic areas studied. To enable national analyses, CDC's Lead Poisoning Prevention
Branch provided quarterly summary data from their national surveillance database for children
aged 6-36 months within each county that had submitted data. These summary measures
included the number of children screened, percentage of children who exceeded certain blood-
lead thresholds, and arithmetic and geometric mean blood-lead concentration for state/local
grantees with a history of universal screening and reporting.

The intention was to have the models reflect the annual prevalence of childhood lead poisoning
over time. Thus, the data were summarized so that each child could only be reported once a
year.  An algorithm was developed to select representative screening test(s) for children with
multiple results with an objective of having children represented in the analysis dataset
maximally once a year. For a given patient, the algorithm preferentially selected tests
confirming elevated blood lead levels and then selected follow-up tests taken beyond nine
months of the previously selected test.  Screening tests were selected when no confirming record
was available.

The response variable consists of quarterly summary statistics from 1995-2006 on the
distribution of observed blood-lead concentrations in counties across the nation, based on
information from CDC's national surveillance database. After summarizing the test-level data
by year, quarter, and county, counties that contained less than five test records in a quarter were
excluded for confidentiality reasons. The time series of summary statistics within select counties
were initially investigated to determine appropriate exclusion criteria to ensure that the data
retained for analysis represented blood-lead concentrations that were universally reported (i.e.,
there were periods of time in which some state or local childhood lead poisoning prevention
programs only reported elevated blood-lead concentrations - and these data needed to be
eliminated from the analysis).  Thus, the number of quarterly summary statistics varied from
county to county within our analysis dataset.

As a prelude to developing the screening algorithm for elimination of data from counties that
were not following universal reporting protocols, a subset of data from counties with obvious
non-universal reporting was identified from within the National  quarterly aggregate summary
database. The algorithm was developed based on application to this subset of data prior to being

                                          13
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
utilized on the remainder of the National Surveillance database.  The algorithm is based on the
following:

Let
   •   n,j represent the number of children observed in the i* county during the j* quarter,
   •   P90(nO represent the 90th percentile of observed n,j within the i* county,
   •   GMij represent the geometric mean blood-lead concentration observed in the 1th county
       during the j"1 quarter, and
   •   P50(GMO represent the 50th percentile of observed OMy within the i* county,
   •   P10,j represent the proportion of children with blood-lead concentration observed at or
       above 10 ug/dL in the i* county during the j* quarter

Then the following 3 exclusion/inclusion criteria are applied sequentially:

       Criteria #1: If ny < Max(P90(nO/5,15) and (GMy > 2* P50(GMj) or P10y> 0.75) then
       exclude the data from the i* county during the j* quarter. This exclusion criteria
       essentially eliminates county/quarter combinations with relatively lower screening
       penetration (compared to when peak screening was achieved) and high blood-lead
       concentrations. The rationale for this exclusion criteria is that the periods of time in
       which a lead poisoning prevention program is not conducting universal reporting will
       involve fewer reported testing results that have higher blood-lead concentrations.

       Criteria #2: If n,j > 100 and GMy <7 then include the data from the i* county during the
       j* quarter.  This criteria was added to include a small number of county/quarter
       combinations within the testing subset of data that were eliminated by the first exclusion
       criteria but did hot appear to be inconsistent with the remainder of data that would be
       included in the analyses. This second criteria was inspected carefully upon application to
       the entire National dataset, to ensure that it was reintroducing data into the analysis in a
       manner consistent with the data analysis goals.

       Criteria #3: If ny < 100 and GMy >10 then exclude the data from the i* county during the
       jm quarter.  This third criteria was established to exclude a small amount of data that was
       not captured by the first exclusion criteria (mostly representing counties with a median
       observed blood-lead concentration slightly above 5 ug/dL)

Within the current CDC national surveillance quarterly summary dataset at the county-level,
there were 69,165 records.  Criteria #1 excluded 1,681 records (2.43%), Criteria #2 re-inserted 6
records, and Criteria #3 eliminated an additional 189 records from the final analysis dataset.

To enable analyses at a finer level of geographic detail than the county level, the Massachusetts
Department of Public Health provided blood-lead surveillance data on specific testing results for
individual children (with confidential identification information excluded) so that data could be
summarized and reported by Census tract. The Massachusetts blood-lead surveillance data
represents all children aged 6-36 months tested from the period 2000-2006.  As with the national
data, quarterly Census-tract level records were created for analysis.

                                            14
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
Due to selection bias, it is expected that the CDC National Surveillance dataset as well as the
Massachusetts surveillance data may show higher proportions of elevated blood-lead
concentrations than found in the general population. For this reason, the proportion of children
with elevated blood-lead concentrations as well as the distribution of the potential continuous
summary measure derived from the surveillance data were compared with those reported by the
most recent six-years of available CDC National Health and Nutrition Examination Survey
(NHANES). Results of this comparison are presented in Section 7.2. In the future, to account
for differences between the surveillance and NHANES data, modifications could be made to the
models to calibrate the surveillance data to better match the national distribution of childhood
blood-lead  concentrations as appropriate (Strauss, 2001).

3.2    Demographic Data

Demographic information from the 2000 U.S. Census were utilized in both the high and low
resolution models, with data being acquired at the county level for the entire nation and at the
Census tract level for Massachusetts. The Census 2000 data gathered by the Census Bureau
includes over 1,000 variables. To narrow the scope of the project, 43 variables within 9 general
categories were selected and explored,  most of which had been used previously by Battelle in a
CDC-sponsored study to predict risk of elevated blood-lead concentrations at the Census tract
level (Strauss, 2001).  Li many cases, the Census variables are constructed from counts or
summary statistics published in the detailed Census 2000 tables.  For example, within each
geographic area, the Census Bureau reported the number of houses that were built before 1950
and the median income of all households.  In order for the analysis to draw comparisons from
tract to tract and/or county to county, however, the  Census variables needed to be manipulated in
a fashion that depended upon the format of the variable. For example, count variables, such as
the number of housing units built before 1950, were changed to percentages. Summary statistic
variables describing income on the other hand, may be standardized within state to adjust for
between-state differences in the cost of living. Table 3-1 supplies the list of the variables
investigated within the nine categories  notes on how they were calculated.
                                           15.
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
Table 3-1. Initial Variables for Analysis Created From the 2000 Census
Variable
Group
Density
Race
Age
Family
Structure
Education
Census Variable1*1
Persons
Housing units
White population
Black population
Indian, Eskimo, and Aleut
population
Asian Pacific population
Other Race population
Native Hawaiian and Other
Pacific Islander population
Multiple Race population
Hispanic population
Children Less than or Equal
to 6 Years Old
Median Age*
Median Age of Children Less
than or Equal to 6 Years Old*
Single Parent9" = Single Male
with Children + Single
Female with Children
Less than a 9th grade
Education
Less than high school1" = #13
+ persons with 9th to 12th
grade education without
obtaining a high school
diploma
Less than college1" = #14 +
Persons with high school
diploma, but no college
experience
Less than college degree1" =
#15 + Persons that attended
college without obtaining a
college diploma
Format
Count
Count
Count
Count
Count
Count
Count
Count
Count
Count
Count
Statistic
Statistic
Count
Count
Count
Count
Count
Calculation
Land Area
(Units = .001 km2)
Land Area
(Units = .001 km2
Persons
Persons
Persons
Persons
Persons
Persons
Persons
Persons
Persons


Household with
Children Less than or
equal to 18 years old =
Married Couple with
children + Single Male
with Children + Single
Female with Children
Persons 18 years old
and over
Persons 18 years old
and over
Persons 18 years old
and over
Persons 18 years old
and over
Analyzed Variable
Population Density
Housing Density
Pet White
Pet Black .
Pet American Indian and
Alaskan Native
Pet Asian
Pet Other Race
Pet Native Hawaiian and
Other Pacific Islander
Pet Multiple Race
Pet Hispanic
Pet le 6 years
Median age of persons
Median age of persons le 6
years
Pet Single Parent
Pet less than 9th grade
Pet no HS degree
Pet no college
Pet no college degree
                                                 16
 This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
 quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
 to represent any Agency determination or policy.

-------
Table 3-1. (continued)
Variable
Group
Income
Poverty
Level
Housing
Units
Census Variable*
Household Median Income
Family Median Income
Per Capita Income
Households without earnings
Households without wages
Households that obtain public
assistance
Persons below poverty level
Persons who are less then or
equal to five years old that are
below poverty level*
Families with total income
below the poverty level
Families with total income
below the poverty level that
have children under 5 years
old.
Vacant
Housing Units Built before
1940
Housing Units Built before
1950
Housing Units Built before
1960
Housing Units Built before
1970
Housing Units Built before
1980
Median Year that Housing
Units were Built
Median Year that Housing
Units were Built - Calculated
by Battelle
Format
Statistic
Statistic
Statistic
Count
Count
Count
Count
Count
Count
Count
Count
Count
Count
Count
Count
Count
Statistic
Statistic
Calculation



Households
Households
Households
Persons for whom
poverty status is
determined
Persons who are less
then or equal to five
years old for whom
poverty status is
determined.
;
Families
Families with children
under five years old.
Housing Units
Housing Units
Housing Units
Housing Units
Housing Units
Housing Units


Analyzed Variable
Standardized Median
Income for Households
Standardized Median
Income of Families
Standardized per capita
income of persons
Pet No Earnings
Pet No Wage or Salary
Pet With Public Assistance
Pet Persons Below Poverty
Pet Persons Below Poverty
of Age LES Below
Pet Families Below Poverty
Pet Poverty of Families
with Children LT 5
Pet Vacant
Pet Pre 1940 Housing
Pet Pre 1950 Housing
Pet Pre 1960 Housing
Pet Pre 1970 Housing
Pet Pre 1980 Housing
Median Year Built
Calculated Median Year
Built
                                                   17
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
Table 3-1. (continued)
Variable
Group
Occupied
Housing
Units
Housing
Value
Census Variable*
Housing Units that are rented
Occupied Housing Units Built
before 1940
Occupied Housing Units Built
before 1950
Occupied Housing Units Built
before 1960
Occupied Housing Units Built
before 1970
Occupied Housing Units Built
before 1980
Median Year that Occupied
Housing Units were Built
Median Rent
Value of Owner Occupied
Housing Units
Format
Count
Count
Count
Count
Count
Count
Statistic
Statistic
Statistic
Calculation
Occupied Housing Units
Occupied Housing Units
Occupied Housing Units
Occupied Housing Units
Occupied Housing Units
Occupied Housing Units



Analyzed Variable
Pet Renter Occupied
Pet Pre 1940 Occupied
Housing
Pet Pre 1950 Occupied
Housing
Pet Pre 1960 Occupied
Housing
Pet Pre 1970 Occupied
Housing
Pet Pre 1980 Occupied
Housing
Median Year Built -
Occupied Only
Standardized Median
Gross-Rent
Standardized Median
Housing Unit Value
* Variables that were created by combining different pieces of information from the 2000 Census
Income and Poverty
Median income per household, family, and person were calculated. Additionally, the proportion
of households that do not receive any wages, do not receive any earnings and do receive public
assistance were investigated. The Census defines earnings and wages as follows:

    •   "Earnings" represent the amount of income received regularly before deductions for
       personal income taxes, Social Security, bond purchases, union dues, Medicare
       deductions, etc.
    •   "Wages" include total money earnings received for work performed as an employee
       during the calendar year 1999. It includes wages, salary, Armed Forces pay,
       commissions, tips, piece-rate payments, and cash bonuses earned before deductions were
       made for taxes, bonds, pensions, union dues, etc.

Similar to the income variables described above, the poverty level of individuals and families
within each county were summarized as the variables Percent Persons and Percent Families
Below the Poverty Level. In order to focus on the poverty level of the children within each
county, however, we created the variables Percent Persons Five Years and Under and Percent
Families with Children Under Five Years Below Poverty Level. Note that in calculating the
various percentages for each of the variables, the denominator changes.  Also note that for some
of the multivariate models presented later in the report, some of the income variables may have
been rescaled to represent income in thousands  of dollars, to allow the parameter estimates for
the regression models to be discemable within the first 3 significant digits.
                                           18
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
Race
The Census Bureau presents five general race groups; (1) White, (2) Black,.(3) Indian, Eskimo,
and Aleut, (4) Asian Pacific and (5) Other, each of which was included and explored separately.
Additional variables were included on percent of Native Hawaiians and Other Pacific Islanders,
percent of the population reporting multiple races (Percent Multiple Races), and percent of the.
population reporting that they are Hispanic (Percent Hispanic).

Housing Cost
Two variables were constructed, to investigate housing cost - Median Rent and Median Housing
Value. Median Housing Value includes the value of all housing units (owned and rented).  Both
of these variables were standardized to account for state-to-state differences in the cost of living.
Note that for some of the multivariate models, presented later in .the report, some of the housing
cost variables may have been reseated to represent housing cost in thousands of dollars, to allow
the parameter estimates for the regression models to be discemable within the first 3 significant
digits.

Occupancy
Occupied housing units are more likely to have lead paint removed than vacant homes.  Thus, the
percent of housing units that are vacant potentially indicates the level of care taken to maintain
buildings within the area. Buildings that are not occupied are more likely to accumulate dust or
debris to which the children of an area may be exposed upon reoccupancy. Percent of vacant
housing units was explored for those reasons. Similarly, the standard of care could be different
between rental properties and owner-occupied properties. Thus, the percent of rental units in an
area was also explored.  The percent of occupied housing units that are rented, rather than
owned, was calculated by dividing the number of rented occupied housing units within an area
by the total number of occupied housing units.

Family Structure
The Census Bureau does not supply a unique variable that indicates the number of single parent
households within an area.  Therefore, this variable was created by combining Census variables
as follows:
       M = Number of Households with a male householder (no wife present) whose own
           children are under 18 years old
       F = Number of Households with a female householder (no husband present) whose own
           children are under 18 years old
       T = M + F + Number of married couples with own children under 18 years.

The Percent of Single  Parent Households variable used represented (M+F)/T.

Housing Age
During the 1950's, as the United States started to become aware of the consequences associated
with the exposure of lead in paint, the use of lead paint within homes began to decrease.  In
1977, however, the use of lead paint in homes became illegal. Thus, the years during which the
housing units were built within each area is important to characterize; older homes are more
likely to contain lead paint than newer homes.  A number of variables related to housing age by

                                           19
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
county were investigated to identify those that best predict children's blood lead levels.  Census
data on the full population of housing units as well as the population of occupied housing units
were investigated.  Note that for some of the multivariate models presented later in the report, the
median age of house variable may was centered at 1950 to provide stability to the intercept term
in the models.

Children's Age
The Census Bureau does not report all data by single years of age.  More typically the agency
reports the total number of people that fall into various age categories. The variable, "Pet  le 6
years" was created to identify the number of children within each geographic area less than or
equal to six years of age at the time of the 2000 Census.  {Additionally, we calculated the median
age of the total population and of those less than or equal to six years old by taking a weighted
average of the midpoint of each age category (the counts are used as the weights).

Education
A series of variables pertaining to the proportion of adults with various levels of education were
created as follows:
       L9   = Number of people older than 18, that have less than a 9th grade education
       L12  = Number of people older than 18, that have 9th though 12th grade experience, but
              do not have a high school diploma
       12   = Number of people older than 18, that obtained a high school diploma or GED
       C    = Number of people older than 18, that have some college experience but did not
              receive a college degree
       T    = Number of People that are over than 18 years old

Percentage variables were created from the L9 through C variables by dividing them by the total
number of people over 18 years old.  Exploratory analyses were conducted upon the four
percentage variables.

Population Variables
Since both counties and Census tracts vary with respect to spatial area and population, and
previous work suggests that risk of childhood lead poisoning differs between rural and urban
areas, we utilized a population density variable as a potential explanatory variable or effect
modifier in the statistical models. Population density was explored in two ways. The first
divides the number of people within the tract by the amount of land area measured in .001 square
kilometers.  The second divides the number of housing units by the amount of land area
measured in .001 square kilometers.  Housing units include the following:  a house, apartment,
mobile home, group of rooms or single room that is occupied as separate living quarters.

3.3    Environmental Data

Environmental data acquired for this project include air and groundwater monitoring data
aggregated at the county level for the low resolution model and at higher resolutions for the
Massachusetts analyses. In cases where the data were available for a limited number of air-
monitoring stations or drinking water samples available for the region(s) being investigated, geo-

                                           20
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed
to represent any Agency determination or policy.

-------
spatial modeling techniques might be used as appropriate to develop predictions across the entire
region. Existence of industrial sources of lead within each county and Census tract as indicated
by the Toxics Release Inventory (TRI) were also included as an environmental data source.
Each of theses data sources are discussed in further detail below.

3.3.1  Concentrations of Lead in Air

EPA maintains a number of ongoing air monitoring programs that collect data over time on.
concentrations of various criteria air pollutants, air toxics, constituents of paniculate matter, and
other airborne chemicals. Each of these monitoring programs have multiple air monitoring
stations that are deployed throughout the country to meet various goals associated with the Clean
Air Act and other federal and state regulations and programs.  For example, some of the
monitoring stations are placed in close proximity to industrial sources of pollution and major
populations centers, while other stations are place in remote areas to assess background chemical
concentrations. While many of these monitoring sites provide information on the concentration
of lead in air over time, a quick assessment of the spatial coverage of these monitoring networks
suggested that making use of these data would be problematic for this study due to time and
resource constraints.  Lead concentrations in air from the monitoring networks are not available
in the majority of counties that will be covered in the low resolution model, or the census tracts
that will be covered in the high resolution models - as shown at the following EPA Website
(http://www.epa.gov/airtrends/lead.html).

Instead of utilizing air monitoring data as described above, the study utilized modeled
predictions of concentrations of lead in air from EPA's 1999 National Scale Air Toxics
Assessment - in which county and Census tract-level predictions are available throughout the
entire country based on the use of predictive models. Documentation for the 1999 National
Scale Air Toxics Assessment, as well as the predicted air concentration data can be found at
http://www.epa.gov/ttn/atw/natal999/tables.ntml.  the predictions were generated using the
Assessment System for Population Exposure Nationwide, or ASPEN.  This model is based on
the EPA's Industrial Source Complex Long Term model (ISCLT) which simulates the behavior
of the pollutants after they are emitted into the atmosphere.  ASPEN uses estimates of toxic air
pollutant emissions and meteorological data from National Weather Service Stations to estimate
air toxics concentrations  nationwide.

The ASPEN model takes into account important determinants of pollutant concentrations, such
as:

    •   rate of release
    •   location of release
    •   the height from which the pollutants are released
    •   wind speeds and directions from the meteorological stations nearest to the release
    •   breakdown of the pollutants in the atmosphere after  being released (i.e.,  reactive decay)
    •   settling of pollutants out of the atmosphere (i.e., deposition)
    •   transformation of one pollutant into another (i.e., secondary formation)
                                          21
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
The model estimates toxic air pollutant concentrations for .every county and Census tract in the
continental United States, however - these data are only available for 1999. Both the Broad-
Based National Model and the High-Resolution Model within Massachusetts considered the
integration of information from the ASPEN Model.  The National Model investigated the
median, average, and 95th percentile predicted air lead concentration within each county, while
the High Resolution Model only considered the average predicted air lead concentration within
each census tract. Within the National Model, the median, average and 95th percentile predicted
air-lead concentrations were mostly distributed near zero. For this reason, zero/one indicator
variables were created to indicate that the observed value of these ASPEN Model predictions
were observed at or above the 95th and 99th percentile within the analysis dataset for potential use
within the predictive models. In addition, EPA collaborators identified a subset of 20 counties
with observed elevated air-lead concentrations, and an indicator variable was used to assess
whether these 20 counties had higher risk of childhood lead poisoning in the predictive models.

The second air lead variable investigated is based on predictions from the HAPEM5 (Hazardous
Air Pollutants Exposure Model, Version 5) model. According to the EPA website, "the
HAPEM5 model has been designed to predict the 'apparent' inhalation exposure for specified
population groups and air toxics. Through a series of calculation routines, the model makes use
of census data, human activity patterns, ambient air quality levels, climate data, and
indoor/outdoor concentration relationships to estimate an expected range of 'apparent' inhalation
exposure-concentrations for groups of individuals."1  Because air quality concentrations in
indoor environments can be quite different than those in the outdoor environment, an exposure
model is generally employed to predict the apparent inhalation exposure. The Air Exposure
(HAPEM5) Model variable captures the predicted exposure data from this model.

The third air lead variable considered, Air Hazard Quotient (HQ),  is derived from the 1999
National Scale Air Toxics Assessment data. This variable represents lifetime exposure for
children at the centroids of each Census tract or county. Lifetime exposure is calculated based
on considering annual exposures and yearly activity patterns.   The HAPEM5 and HQ air lead
variables were only considered within the context of the High Resolution Model within
Massachusetts.

3.3.2  Toxic Release Inventory Variables

EPA's Toxic Release Inventory catalogs various sources of lead, based on information provided
by industrial facilities. This data source was used,to generate county- and Census tract-level
estimates of the total amount of lead and/or lead-containing compounds that are released by
industrial facilities into the environment via air, surface water, or underwater injection.
Although the above described ASPEN modeling results are based on the (airborne) emissions
data and how they would theoretically translate into average ambient air-lead concentrations, the
data from the TRI are available for multiple years and for other types of emissions (such as
surface water). Thus, this information has potential to add predictive power to the models.
 1httD://ePa.gov/ttn/atw/natal999/tBd/teddraft.html
                                           22
 This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
 quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
 to represent any Agency determination or policy.

-------
Three types of TRI variables were utilized - total compounds, lead only, and total lead. Within
each types, five pollution variables were explored - total lead in the air, lead in fugitive air, lead
from smokestacks, lead in surface water, and lead in water by injection. Thus, 15 total TRI data
variables were evaluated.

Within the National Model, the distributions of the TRI emissions variables were mostly
concentrated near zero. For this reason, additional zero/one indicator variables were created to
indicate that the observed value of these TRI emissions were observed at or above the 95th and
99* percentile within the analysis dataset for potential use within the predictive models.

3.3.3   Water Quality Data

The plumbing system inside a home and from the street to the home may contribute to drinking
water contamination. To address this potential source, EPA obtained data from their Drinking
Water Information System that includes tap water lead levels for public water systems.  Public
water treatment plants are monitored over a period of years via random sampling of tap water
within their jurisdictions. The length of monitoring varies depending on the water quality, i.e.,
systems found to have higher water lead levels undergo monitoring for longer periods of time.
Data available from this monitoring program include 90th percentile water lead values for each
public water treatment facility, the population size served by each facility, the start and end date
for the monitoring period, and the county in which the facility is located. These data were used
to construct a population-size weighted average 90th percentile water lead concentration variable
within each county/quarter combination.

Since there were some county/quarter combinations with no observed data from EPA's Safe
Drinking Water Information System, an indicator variable was developed to indicate whether the
county/quarter included a monitored  facility (or not) - allowing an intercept to be fit among
those county/quarters with no drinking water monitoring, and a slope estimate  to be fit for the
effect of the weighted average 90* percentile drinking water lead concentration among reported
facilities.

EPA's Safe Drinking Water Information System data was not geocoded to the  census-tract level,
and therefore  these data were only available for use in supporting the Broad Based National
Model at this  time.

3.4    Programmatic Data

Most of the explanatory variables being explored in this project are considered risk factors for
childhood lead poisoning. Among factors that might mitigate these risks, it was anticipated that
the level and characteristics of programmatic support from either federal, state, or local sponsors
may contribute towards meaningful reductions in the prevalence of childhood lead poisoning.
The level of financial support available within each county served as a proxy for programmatic
support in the low resolution (National) models. In the high resolution models run for
Massachusetts, information from housing inspections were also explored within the statistical
                                          23
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
models. The following sections detail the specific characteristics of the variables used within the
models.

3.4.1   Programmatic Funding Variables

The goal of this variable is to construct a longitudinal history of current and cumulative
per-capita dollars allocated to each county and Census tract to combat childhood lead poisoning.
For use in both the national and state models, Battelle obtained data from HUD's Office of
Healthy Homes and Lead Hazard Control on grants funded since the inception of the Lead-Based
Paint Hazard Control Grant Program in 1992. Data were also obtained from CDC's Lead
Poisoning Prevention Branch on their program's grant funding approximately three weeks prior
to the end of this project, and therefore these data were only able to be integrated into the
Massachusetts models because of time constraints.

Four variables were generated from these data and analyzed - current and cumulative funding
allocated to each county or Census tract to combat childhood lead poisoning, both Standardized
by number of children per tract and Not Standardized. The Standardized variable is a funding
per child variable while the Not Standardized versions are funding for geographic area variables.
For the high-resolution model in the Commonwealth of Massachusetts, information on within-
state funding levels was obtained and analyzed. Within-state funding data were available down
to the township level.  The state, HUD, and CDC funding data also were combined to create
Total Funding variables, including both current and cumulative levels and both standardized and
not standardized versions. The total funding variables also were only investigated as part of the
Massachusetts analyses.

3.4.2  EPA Region

EPA Region was investigated as a potential predictor of children's blood lead levels to determine
if that high-level geographic indicator should be included as a stratification  variable in the
national multivariate models.

3.4.3  Housing Inspection Data (Massachusetts)

The Commonwealth of Massachusetts maintains an extensive database on all lead-based paint
inspections conducted over time (dating back to the early 1990's). The Massachusetts
Department of Public Health provided a database  that contains a single record for each
inspection, with the following information: housing-unit id, census tract, date of inspection,
result of inspection (whether the housing unit was found to be in compliance with Massachusetts
standards).  The database contains records on over 200,000  housing units - with many housing
units having multiple inspections over time. Note that for units with multiple records, .time
periods in which the units were both in and out of compliance with the Massachusetts standards
were identified.

These data can be used in the Massachusetts high resolution models in two  ways.  First, a
longitudinal summary measure of the proportion of housing within each census tract that was

                                            24
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
known to be in compliance with the Massachusetts standards was developed.  It was anticipated
that within a census tract, Las this-proportion increasggjwer-time. the risk of childhood lead
poisoning:will:decrease. Second, due to the fact that individual blood-lead records from
Massachusetts with linkable housing-unit identification variables were available, a determination
could be made regarding whether a housing unit was in compliance at the time of the blood-lead
test for each child in the database (with potential outcomes of the determination being yes, no,
and unknown).

The first approach described above is consistent with the methods for exploring aggregated
summary blood-lead information over time within each Census tract. The second approach
allows utilization of some predictive information at the individual child level. This information
may help improve prediction, and also help assess what information might be lost when
transitioning from individual-level data to aggregate summary data in the analyses.
Unfortunately, due to time and resource constraints, only the first method was able to be
explored within this project.  Thus, the three measures listed below were calculated using four
different methods.  The three measures are:  .

   •   P - represents the Proportion of Housing Units within a Census tract that are assumed to
       Meet the Massachusetts Standard of Care at any given time;
   •   F - represents the Proportion of Housing Units within a Census tract that are assumed to
       Not Meet the Massachusetts Standard of Care at any given time; and
   •   N - represents the Proportion of Housing Units within a Census tract with Housing
       Inspection Information at any given time.

As noted, the measures were generated in four different ways, each handling the longitudinal
information in a slightly different manner. The four measures, numbered in the model results
from 1 to 4, are listed below.

       1.  Naive Method 1 - Create a longitudinal history for each housing unit inspected, and
          treat the first inspection observation as being representative for time periods
          preceding that inspection.
       2.  Naive Method 2 - Create a longitudinal history for each housing unit inspected, and
          assume missing information for time period preceding the first test on each unit.
       3.  Naive Method 3 - Create a longitudinal history for each housing unit inspected, and
          treat the first inspection observation as being representative for time periods
          preceding that inspection if the housing unit failed, and assume missing information
          for time period preceding the first test if the unit passed.
       4.  MDPH Approved Method - Create a longitudinal history for each housing unit, with
          different rules for the treatment of the time-period preceding the first test based on (a)
          the housing inspection result and (b) the reason for ordering the inspection.

Note; that for housing units with multiple inspections, each housing inspection result is assumed
to be representative of the house (either pass or fail) until the next result. The last result is
carried forward over time (e.g. if the last observed inspection on a house passed in November of
1998, - that particular house is  assumed to be meeting the Massachusetts standard of care over all

                                            25
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
qualify guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
subsequent time periods in the dataset).  If multiple inspections occur on the same house within a
particular quarter (3-month interval), we take the maximum result (with pass being coded as a 1,
and fail being coded as zero) to represent the house. The 0/1 results are then summed across all
observed housing units within each census tract over time (quarters). The summed results are
then divided by the number of housing units reported within each census tract from the 2000
Census.

While all of the above described housing inspection variables were investigated in the
exploratory data analyses, only the P4, F4, and N4 variables associated with the MDPH-
approved method of constructing the longitudinal history within each housing unit observed was
considered within the context of the multivariate models.

3.5    Data Linkages

The primary objective of this pilot study was to utilize combined information from different
sources at various levels of geographic and temporal specificity to more accurately target
geographic areas at high risk for not meeting the 2010 goal of eliminating childhood
lead-poisoning. As such, work on the study required careful integration of a variety of data
sources with various characteristics and documentation. Data to support this study was gathered
from a variety of sources, including federal, state and local lead poisoning prevention programs,
as well as publicly available data downloaded.from the internet (e.g., Census data, EPA's Toxic
Release Inventory, etc.), as detailed in the previous sections.

When Battelle first received ejeh:data:sgurce, the datarandrsupporting.dpcumentation was
reviewed to gain knowledgegn die^sfflc^re,-relation^prand:q^i^oTtiie-data\ Battelle
database managers worked with the project team (including collaborators providing data to the
project, as well  as EPA) to determine the final format for each database; desired uses of the
databases; as well as the requirements for maintaining the databases. Based on this information,
Battelle constructed separate master databases for the national model and for the high-resolution
Massachusetts model that integrate the various environmental, demographic, and programmatic
variables, and facilitate statistical analyses of the combined data. These databases were
constructed by combining data from a variety of formats including MS SQL Server, MS Access,
Excel, ACSII, Access, Arc View, and SASฎ electronic databases.  In order to combine the
various data sets, they were merged on key fields, including state, county, Census tract, and time
period. The data being used for analyses of a particular geographic level, e.g. county, are
comparable because they are representative of that geographic area.

Throughout the development process, Battelle conducted checks for completeness on all study
databases, and worked with data-sharing collaborators and EPA to attempt to complete missing
data as necessary to support the proposed statistical analyses. Any changes to the databases
(corrections, additions, deletions, etc.) were documented in appropriate metadata files.
Documentation of the combined master databases is included in Appendix H.

Battelle followed Standard Operating Procedures (SOPs) to ensure the proper storage, backup,
and retrieval of datasets created and analyzed for this study. The various databases were backed

                                           26
This information is distributed solely for the purpose of pre-dissemination peer review'under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
up to tape nightly via Battelle's automated backup routines, and were only accessible to
members of the project team. CD-ROM backups were made on a regular basis to serve as a
safeguard in case the backup system failed for any reason.

Microsoft Access and SQL server were the primary software tools used for data management,
although some data or results. The SASฎ System was the primary statistical data analysis tool
used on this project. Arc View software was used to translate results into maps, as seen in
Appendices F and G.

The data utilized for the study were maintained in a manner that preserved the confidentiality of
all the data and prevented its unauthorized release. As data files were received from EPA,
Battelle handled the original data (e.g. data with personal identifiers) as though the data were
classified as confidential business information (CBI) under the Toxic Substances Control Act
(TSCA), even though EPA may not specifically classify these data as "CBI." While in Battelle's
possession, the data files were not shared with anyone outside of the project team.
                                            27
This information is distributed solely for the purpose ofpre-dissenu'nation peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
4.0    EXPLORATORY DATA ANALYSES

Because the goal of this study was to develop a series of statistical models that predict the risk of
childhood lead poisoning at die geographic level across multiple response variables (geometric
mean blood-lead concentration as well as proportion of children screened at or above 5,10, IS
and 25 ug/dL), all potential predictor variables first were explored individually to determine their
predictive ability.  Results from these bivariate analyses were assessed to identify the set of
variables to include in the multivariate model that predicts how the risk of childhood lead
poisoning changes over time among the various Census tracts and counties included in the
analysis.

This section of the report provides the results of the series of exploratory analyses described in
Section 2.2, which were performed to assess the potential predictive power of various candidate
demographic, environmental, and programmatic variables for potential use in the multivariate
models. These exploratory analyses initiated with an assessment of the study sample, i.e., the
proportion of counties in the sample with complete and reliable data for both the response
variable and the explanatory variables.

Each candidate predictor variable was reviewed with particular attention focusing on the manner
in which the county-level predictor variables would be merged with the quarterly summary
blood-lead information prior to fitting the statistical models. In preparation for developing
longitudinal statistical models, univariate summaries of each predictor variable as a function of
time were produced.  Comparisons of these distributions were made using side-by-side box-plots
for continuous data or bar-charts for categorical data. This helps verify that the data are clean
and ready for analysis and identify cells with sparse data. Such descriptive analyses were
conducted on each predictor variable database to characterize the distributions of all observed
variables using frequency distributions for categorical variables, and simple summary statistics
(mean, median, mode, minimum, maximum, and select percentiles) for continuous variables.

The univariate descriptions were then followed by fitting a series of cross-sectional bivariate
relationships between the blood-lead response variable(s) and each candidate explanatory
variable. These cross-sectional relationships were explored as a function of time to better
understand the stability of these relationships, and whether they change over time, so that they
can be modeled appropriately in the more sophisticated longitudinal analyses. These analyses
also help identify which explanatory variables are most predictive of the blood-lead response
variable.

4.1    Relationship between National Blood Lead Data and Explanatory Variables

The response variable for the national data analysis consisted of quarterly summary statistics
from 1995-2006 on the distribution of observed blood-lead concentrations in counties across the
nation, based on information from CDC's national Childhood Lead Poisoning Surveillance
Database. The time series of summary statistics within select counties were initially investigated
to determine appropriate exclusion criteria to ensure that the data retained for analysis
represented blood-lead concentrations that were universally reported (i.e., there were periods of

                                          28
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
time in which some state or local childhood lead poisoning prevention programs only reported
elevated blood-lead concentrations - and these data needed to be eliminated from the analysis).
Thus, the number of quarterly summary statistics varied from county to county within our
analysis dataset.

The national blood lead data were categorized into four time periods - (1) 1995 to December 31,
1999; (2) January 1,2000 to December 31,2001; (3) January 1,2002 to December 31,2003, and
(4) January 1,2004 to present - so that change over time could be evaluated. Using the specified
four time periods split the dataset of quarterly county-level records into roughly similar sizes.
Presented below are the exploratory analysis results for the demographic, environmental, and
programmatic variables investigated.  Detailed figures and tables containing results are included
in Appendix A.  A detailed discussion of the results seen in Appendix A is contained in
Appendix D.

To allow comparison of the different variables explored within each variable type, Table 4-1
presents the log-likelihood statistic from each model run and presented in Appendix A. Within
each variable category, the variable that provided the best fit for each of the six models is
highlighted in yellow. For example, within the income category, the model including Percent No
Household Wage provided the best fit for both geometric mean models and the models of
proportion of children with blood lead levels above 10,15, and 25 ug/dL, while the model
including Median Household Income  provided the best fit when considering proportion of
children with blood lead levels above 5 ug/dL. Those variables highlighted in yellow were the
most likely to become candidate predictors for the multivariate statistical models.
                                           29
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
Table 4-1. Summary of Log-likelihood Ratios from each Model Fit to all Potential Explanatory Variables
Variable
Category
-Income
Race
Housing Costs
Occupancy
Single Parent
Variable
Median Family Income ($)
Median Household Income ($)
Median Per Capita Income ($)
Percent No Household Earnings
Percent No Household Wage
Percent Household on Public Assistance
Percent Below Poverty Line
Percent Family Income Below Poverty Line
Percent Less than 5 Years in Poverty
Percent Amer. Indian and Alaskan Native Alone
Percent Asian Alone
Percent Black Alone
Percent White Alone
Percent Native Hawaiian and Other Pacific Islander Alone
Percent Other Race Alone
Percent Multiple Races
Percent Hispanic
Median Rent ($)
Housing Value ($)
Percent Rented
Percent Vacant
Percent Single Parent
Model 1
153509.7
153495.9
153535.7
153502.9
153465.6
153566.5
153574.0
153571.0
153573.8
153603.3
153595.4
153607.3
153614.8
153587.5
153598.0
153577.6
153613.5
153451.2
153528.1
153611.1
153612.8
1536053
Model 2
196109.6.
196091.2
196139.9
196103.8
196062.1
196189.3
196202.0
196198.9
196203.0
196225.8
196186.4
196231.9
196231.1
196203.5
196212.4
196181.7
196225.9
196012.7
196103.0
196234.7
196232.7
196228.1
Model 3
250237.8
250202.1
250330.4
250366.0
250286.2
250628.3
250549.6
250537.4
250586.9
.
250584.7
250677.0
250732.2
250701.3
.
250674.6

250037.0
250298.4
250729.0
2506905
250699.0
Model 4
269439.6
269412.9
269484.1
.
269319.9
269718.0
269712.8
269709.0
.
269826.8
269722.2
269831.7
269777.2
269750.7
269735.6
269739.6
269762.1
269220.1
269449.0
269994.2
269872.6
2699365
National Data
Model 5
300097.7
.
300159.6
299925.5
2997663
300259.1
.
300335.7
300297.0
300456.3
300405.9
300306.7
300252.9
300302.9
300302.9
300289.9
300306.9
299864.1
300049.1
301038.3
300727.6
300558.7
Model 6
360822.5
360576.0
360986.6
360043.1
359818.9
360701.1
361059.6
361064.1
.
361009.1
361015.5
360634.1
360433.8
360766.7
360834.0
360789.4
360745.0
360483.1
.
362209.3
362162.7
•
                                                                        30
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality guidelines. It has not been formally
disseminated by EPA. It does not represent and should not be construed to represent any Agency determination or policy.

-------
Variable
Category
Housing Age
Children's Age
Education
Population
HUD Funding
AirLead
Variable
Year Built
Year Occupied Unit Built
Percent Built Before 1940 -
Percent Built Before 1950
Percent Built Before 1960
Percent Built Before 1970
Percent Built Before 1980
Percent Occupied Units Built Before 1940
Percent Occupied Units Built Before 1950
Percent Occupied Units Built Before 1960
Percent Occupied Units Built Before 1970
Percent Occupied Units Built Before 1980
Percent Less than 6 Years of Age
Number Less than 6 Years of Age
Percent Less than 9lh Grade
Percent without High School Degree •
Percent without any College
Percent without College Degree
Total Housing Units
Total Population
Housing Density
Current HUD Funding
Cumulative HUD Funding
Average Air Lead
Median Air Lead
95* Percentile Air Lead
Model 1
1533805
153390.5
153396.2
153395.1
153419.2
153432.3
153438.9
153410.1
153406.4
153425.0
153435.9
153445.2
1535903
153637.5
153585.7
153591.2
153543.5
153541.2
153640.8
153642.5 •
1536313
153549.8
153621.4
153605.1
1535985
153610.3
Model 2
1959533
195964.1
195945.0
195948.6
195993.2
196021.0
196032.5
195960.6
195960.8
195999.5
196024.9
196037.9
196196.4
196257.8
196215.8
196219.4
196145.1
196149.8
196260.1
196262.0
196252.5
195616.7
196227.3
196228.9
196224.6
1962323
Model 3
t
250433.1
250448.6
<
.
2504623
250467.9
250462.8
250429.5
.
2504723
250483.5
250651.9
.
m
250547.1
.
250344.5
.
250705.1
250722.5
2476454
250649.2
250722.2
2507213

Model 4
269400.5
269427.2
269296.9
269269.1
269403.5
269498.7
269472.9
269316.9
2692943
269427.6
269531.2
269497.2
269671.4
269857.9
269715.8
2696903
269428.4
269423.0
269875.0
269871.5
269791.0
269335.7
269690.9
269805.1
269808.9
.
Model 5
300399.5
300439.0
300081.1
300078.6
300407.5
300594.4
300453.7
300093.2
300103.5
300442.9
300652.8
300492.1
3001453
300609.0
3004233
. 300419.6
299930.1
300009.5
3006763
300653.6
.
312171.9
.
300468.5
300501.8
300441.6
Model 6
362406.8
362487.4
t
~B
362573.0
363033.7
362627.2
3613793
361606.8
362626.8
363133.5
36269.1.5
3603584
361515.6
3612113
.
360304.1
360728.8
3617193
361643.0
360996.7
397510.7
361195.8
3611174
361193.4
.
                                                                        31
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality guidelines. It has not been formally
disseminated by EPA. It does not represent and should not be construed to represent any Agency determination or policy.

-------
4.2    Relationship between Local Blood Lead Data and Explanatory Variables

Many of the variables investigated for the national model were also explored for the local
modeling using Massachusetts data.  All of the Census data were used in both models, although
at the Census tract level instead of at the county level. The various demographic, environmental,
and programmatic variables were explored using the same techniques as the national data, which
were described in Section 2.2. Detailed figures and tables containing exploratory results are
included in Appendix B. A detailed discussion of the results seen in Appendix B is contained in
Appendix E.  Table 4-2 presents the log-likelihood statistics that resulted from the bivariate
modeling. Variables presenting the best model fit within each variable category are highlighted
in yellow.
                                             32
 This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
 quality guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed
 to represent any Agency determination or policy.

-------
Table 4-2. Summary of Log-likelihood Ratios from each Model Fit to-all Potential Explanatory Variables, Massachusetts Data
Variable Category
Income
Race
Housing Costs
Occupancy
Single Parent
Home Age
Variable
Median Family Income ($)
Median Household Income ($)
Median Per Capita Income ($)
Percent No Household Earnings
Percent No Household Wage
Percent Household on Public Assistance
Percent Below Poverty Line
Percent Family Income Below Poverty Line
Percent Less than 5 Years in Poverty
Percent Amer. Indian and Alaskan Native Alone
Percent Asian Alone
Percent Black Alone
Percent White Alone
Percent Native Hawaiian and Other Pacific Islander Alone
Percent Other Race Alone
Percent Multiple Races
Percent Hispanic
Median Rent ($)
Housing Value ($)
Percent Rented
Percent Vacant
Percent Single Parent
Year Built
Year Occupied Unit Built
Percent Built Before 1940
Percent Built Before 1950
Percent Built Before 1960
Percent Built Before 1970
Percent Built Before 1980
Percent Occupied Units Built Before 1940
Percent Occupied Units Built Before 1950
Percent Occupied Units Built Before 1960
Model 1
51727.2
51689.2
51917.2
52020.1
52039.6
51963.8
51974.1
51991.3
52025.2
52259.2
52264.2
52170.1
52123.5
52273.4
52202.6
52042.2
52181.3
52004.0
52094.5
52006.7
52186.5
51747.7
51949.7
51966.8
51923.9
51897.9
51959.8
52052.3
52073.4
51931.0
51910.8
51975.9
Model 2
48154.7
48114.6
48345.5
48445.8
48467.3
48385.7
48389.9
48412.1
48446.0
48692.5
48699.0
48599.1
48547.4
48706.6
48633.4
48470.6
48609.2
48440.1
48520.9
48421.1
48617.5
48155.6
48360.5
48377.9
48335.3
48308.1
48374.0
48469.4
48490.9
48342.4
48321.2
48389.9
Model 3
86501.8
86375.5
86812.3
86853.4
86858.3'
86854.4
86778.4
86847.5
86924.8
87120.6
87139.8
87051.0
86960.9
87135.6
87055.1
86889.1
87019.6
86942.8
87032.6
86681.5
87003.7
86542.8
86621.1
86641.4
86505.1
86483.1
86653.5
86806.3
86850.4
86512.2
86497.8
86675.0
Model 4
139627.0
139433.1
140036.2
139531.9
139459.8
139836.6
139558.4
139686.4

139739.5
.
139781.5
139784.2
139740.3
139688.4
1396613
139808.6
139972.2
140053.9
139426.7
139451.9
139654.1
139739.2
139748.6
139476.8
139547.1
139697.8
139775.8
139771.3
139475.5
139549.3
139713.0
Model 5
- 178071.4
177919.9
178467.2
177346.7
177184.8
178218.1
177838.6
177952.4
178034.4
177392.1
177384.8
177691.4
178066.6
177378.7
177606.3
178104.6
177754.4
177952.2
178202.0
177818.5
177021.2
178338.5
178275.2
178258.9
178110.9
178188.5
178061.5
177977.1
177896.0
178078.2
178149.3
178044.5
                                                                 33
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality guidelines. It has not been formally
disseminated by EPA. It does not represent and should not be construed to represent any Agency determination or policy.

-------
Variable Category

Children
Education
Population
Air
HUD Funding
TRI
Variable
Percent Occupied Units Built Before 1970
Percent Occupied Units Built Before 1980
Percent Less than 6 Years of Age
Number Less than 6 Years of Age
Percent Less than 9™ Grade
Percent without High School Degree
Percent without any College
Percent without College Degree
Total Housing Units
Total Population
Housing Density
Air Dispersion (ASPEN) Model
Air Exposure (HAPEM5) Model
Air Hazard Quotient (HQ)
Current HUD Funding ($ per Child)
Cumulative HUD Funding ($ per Child)
Current State Funding ($ per Child)
Cumulative State Funding ($ per Child)
Current CDC Funding ($ per Child)
Cumulative CDC Funding ($ per Child)
Current Total Funding ($ per Child)
Cumulative Total Funding ($ per Child)
Current HUD Funding ($ per Census Tract)
Cumulative HUD Funding ($ per Census Tract)
Current State Funding ($ per Census Tract)
Cumulative State Funding ($ per Census Tract)
Current CDC Funding ($ per Census Tract)
Cumulative CDC Funding ($ per Census Tract)
Current Total Funding ($ per Census Tract)
Cumulative Total Funding ($ per Census Tract)
TRI Compounds (Total Air)
TRI Compounds (Fugitive Air)
Model!
52070.4
52083.6
52280.4
52243.2
52006.5
51852.7
51787.2
51822.1
52286.9
52223.0
52272.6
52275.1
52273.1
522723
52290.9
52287.3
52162.4
52200.7
52288.3
52292.6
52282.8
52292.2
52302.9
52306.8
52232.2
52210.6
52297.5
52303.7
52303.1
52300.1
52295.5
52287.7
Model 2
48486.2
48501.0
48713.7
48678.2
48435.7
48279.9
48218.2
48251.2
48720.8
48660.6
48697.2
48708.3
48706.3
4870S.5
48722.4
48723.3
48582.7
48617.7
48720.8
48725.0
48711.9
48721.9
48737.1
48740.4
48659.5
48638.3
48729.5
48737.2
48735.6
48732.3
48728.8
48720.7
Model 3
86829.5
86860.0
87126.9
86913.2
86828.7
86709.8
86729.0
86759.6
87090.7
86909.7
87068.2
87136.9
87134.9
87134.0
87140.4
87163.4
87014.3
87044.0
87151.7
87136.0
.
87145.9
87167.5
87097.6
87178.1
87222.7
87162.7
87121.6
87173.9
87179.7
87153.5
87146.1
Model 4
139799.3
139776.7
.
138995.8
139726.7
139857.7
140104.0
140084.9
139461.7
138948.0
139579.8
139740.1
139737.8
139737.0
139755.6
139804.9
139706.6
139740.2
139760.5
139706.1

.
139723.5
.
140076.2
140117.0
.
139570.8
139790.6
139780.5
139761.4
139750.0
Model 5
177979.1
177867.1
177615.0
•
177859.2
178346.7
178689.0
178625.6
176701.5
.
177260.2
177377.1
177375.0
177374.2
177426.7
177444.3
177503.7
177459.9
177456.0
177330.9
177377.3
.
177172.3
177127.9
178005.1
178018.7
177462.6
176972.0
177370.8
177447.9
177395.8
177399.8
                                                                        34
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality guidelines.  It has not been formally
disseminated by EPA. It does not represent and should not be construed to represent any Agency determination or policy.

-------
Variable Category
TRI (continued)
Housing Inspection
Variable
TRI Compounds (Stacks)
TRI Compounds (Water Surface)
TRI Lead Only (Total Air)
TRI Lead Only (Fugitive Air)
TRI Lead Only (Stacks)
TRI Lead Only (Water Surface)
TRI Total Lead (Total Air)
' TRI Total Lead (Fugitive Air)
TRI Total Lead (Stacks)
TRI Total Lead (Water Surface)
PI: Proportion of Housing Units Passing MA Standard of Care:
Naive Method 1
Fl : Proportion of Housing Units Failing MA Standard of Care:
Naive Method 1
Nl: Proportion of Housing Units Assessed: Naive Method 1
P2: Proportion of Housing Units Passing MA Standard of Care:
Naive Method 2 . .
F2: Proportion of Housing Units Failing MA Standard of Care:
Naive Method 2
N2: Proportion of Housing Units Assessed: Nai've Method 2
P3: Proportion of Housing Units Passing MA Standard of Care:
Naive Method 3
F3: Proportion of Housing Units Failing MA Standard of Care:
Naive Method 3
N3: Proportion of Housing Units Assessed: Nai've Method 3
P4: Proportion of Housing Units Passing MA Standard of Care:
MDPH Method
F4: Proportion of Housing Units Failing MA Standard of Care:
MDPH Method
N4: Proportion of Housing Units Assessed: MDPH Method
Model 1
52295.6
52285.5
52291.0
52289.6
52291.7
52274.4
52295.8
52291.5
52295.6
52285.4
52214.0
52108.5
52131.7
52240.8
52199.5
-52208.4
52240.8
52108.5
52160.8
52240.5
52106.5
52160.3
Model 2
48728.9
48718.8
48724.0
48722.6
48725.0
48708.5
48729.0
. 48724.6
48728.9
48718.6
48643.0
48548.2
48554.8
48671.5
48645.5
48641.6
48671.5
48548.2
48586.2
48671.1
48545.8
48585.5
Model 3
87154.3
87147.0
87153:1
87151.4
87153.7
87138.1
87155.2
87154.0
87154.3
87147.4
87070.2
86916.6
86963.0
87101.3
87046.4
87066.2
87101.3
86916.6
86996.1
87098.8
86919.8
86994.8
Model 4
139761.8
139753.0
139762.9
139759.1
139757.2
139746.5
139759.7
139761.5
139760.6
139753.3
139774.6
139803.2
139849.2
139767.8
139861.5
139826.1
139767.8
139803.2
139865.4
139769.1
139808.9
•
Model 5
177395.5
177391.9
177405.1
177404.2
177390.5
177386.0
177392.4
177401.9
177393.4
177393.0
177816.1
178520.9
178224.8
177800.8
178375.2
178110.9
177800.8
178520.9
178257.4
177809.5
178518.4
178259.8
                                                                        35
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality guidelines.  It has not been formally
disseminated by EPA. It does not represent and should not be construed'to represent any Agency determination or policy.

-------
5.0    STATISTICAL MODELING RESULTS

As described in Section 2.3, for each statistical model within each of the two broad model types
(Low and High Resolution) the variables that led to the best model fits were initially included in
a multivariate statistical model and assessed jointly to determine which variables were predictive
of children's blood lead levels.  As results of each model run were reviewed, some variables
were dropped from the model if they were not significant predictors of the outcome variable and
were not improving the fit of the model by being included. Thus, each model was run and results
assessed multiple times until a final model was reached.  The sections below present the final
model results for the national risk models (Section 5.1) and the local risk models for
Massachusetts (Section 5.2). Maps of the predicted results are included in Section 6 and in
Appendix G.

As noted previously, not all potential data sources were able to be assessed for their potential
inclusion in the final multivariate models because they were received too close to the end of this
pilot project. It is possible that inclusion of other variables (e.g., CDC and combined funding in
the national model) could suggest changes in the final models and improve the models'
predictive abilities.

5.1     Low Resolution Modeling Results

Table 5-1 presents the full set of variables included in the final multivariate models for Models 1
through 6. Across all six models, the time and space variables were important predictors of the
various outcomes. The same three variables related to time and space were included in all six
models:

       •  EPA Region,
       •  the interaction of EPA Region and time period (1995-1999,2000-2001, 2002-2003,
          and 2004-2006), and
       •  the interaction of EPA Region and quarter of the year with the 3rd quarter (July-
          September) associated with the highest predicted lead levels.

Below we list the other variable types explored and summarize the set of variables included in
the final models.

   •   Income - The best-fitting income variable remained in Models 2, 3, and 6.
   •   Race - The best-fitting race variables were included in Models 1 through 5.
   •   Housing Cost - Median Rent was only included in Models  1 and 2.
   •   Occupancy - The best-fitting occupancy variable was included in Models 1,4,5, and 6.
   •   Single Parent Status - The percent of single parent households was included in all
       models.
   •   Housing Age - The best-fitting housing age variable was included in all models.
   •   Children's Age - The percent of children less than or equal to six years old was included
       in Models 4 and 6.
   •   Education Level - The education variables were removed from all models.
                                          36
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
    •   Population - The population variables were removed from all models.
    •   Air Lead - The best-fitting air lead variable was included in all models.
    •   TRI - A TRI-related variable was only included in Model 3.
    •   Drinking Water - Drinking water-related variables were included in all models except
       Model 3.
    •   HUD Funding - The interaction of HUD funding and the four time periods was included
       in all models.

Thus, race, family structure, housing age, air lead levels, drinking water lead levels, and HUD
funding are the variables that improved the predictive capability of at least majority of the
national models. There were differences among the six models in terms of the variables that
provided the best-fitting final model. Models 1  and 2 that modeled unweighted and weighted
geometric mean blood lead levels, respectively,  were similar. The differences were (1) that
Percent of Households with No Wages was included in Model 2 was dropped from Model 1, and
(2) that Percent Rented was included in Model 1 while the occupancy variables was dropped
from Model 2.

Among the four models of proportion of blood lead levels above 5,10,15, and 25 ug/dL
(Models 3 to 6), Model 3 contained four main differences from the others, (1) Model 3 did not
include an occupancy variable, (2) a TRI variable was included, (3) no drinking water variables
were included, and (4) the 99th percentile median air lead variable was included whereas the
indicator of the Top 20 highest counties was included in the other three models.

Tables 5-2 through 5-7 present the parameter estimates from each of the six multivariate national
models. The standard error and p-value associated with each predictor is also included.
Estimates are also presented for the four variance components that were included in the models -
ฐ*0 • ฐ"4.<5iian<* ฐ\  (related to the random intercept (8oO and slope (8u) terms), and 02Error
(related to the residual error term in the models that describe the geometric mean blood-lead
response)...

Following each table are two figures that provide information on the fit of the final models. The
first, a histogram of the residuals from the final model fit, helps determine whether or not it is
reasonable to assume that the random errors in a statistical process can be assumed to be drawn
from a normal distribution.  Figures 5-1,5-3 correspond to a typical  histogram of residuals from
the mixed model analysis of variance applied to the geometric mean. Figures 5-5,5-7,5-9, and
5-11 contain the residual histograms of the observed-predicted probabilities from each of the
four logistic regression models.  Please note that for these last four histograms - the model was
actually applied on the logit scale. However, since the logit is undefined for observed
proportions at zero and one, the histograms were applied to the original scale of measure.

The second set of figures plot the observed values versus the predicted values for each model.  If
the multivariate model fitted is appropriate, predicted values obtained from regressing the
observed values on the multivariate model's predicted values when plotted against observed
values, one would expect to all the points to be very close to the 45ฐ degree line.  Figures 5-2,5-
4,5-6,5-8,5-10, and 5-12 contain these comparison plots for each of the six national models,
                                          37
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed
to represent any Agency determination or policy.  .

-------
respectively. The plots for the last four models were conducted on both the observed probability
and logit probability scales, with observed data points at zero and one censored in the logit scale
plots.

In general, these plots suggest that the models are performing well. A weighted regression line
(blue line) fit to the observed vs. predicted plots shows a very high R2 value in most of the
models that mirrors the 45 degree line (shown in red) for the majority of the data. One trend
observed in these plots that is important to consider is that the Broad-Based National models
tend to underpredict for county/quarter combinations with higher geometric mean blood-lead
concentrations and higher proportions that exceed the 5,10,15 and 25 ug/dL threshold values.
Further exploration may be necessary to determine whether these higher values represent
county/quarter combinations with fairly sparse data (i.e. few observations) - which might explain
why they would have been less influential in Models 2 through 6 which are influenced by the
number of observations associated with each observed value. For the higher blood-lead
threshold categories, the model appears to over-predict the lower observed proportions -
suggesting the possibility of a regression to the mean effect.
                                             38
 This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
 quality guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed
 to represent any Agency determination or policy.

-------
Table 5-1. Summary of Variables Included in Final National Multivariate Model
Variable Type
Area and Time
Income
Race
Housing Cost
Occupancy
Family
Structure
Housing Age
Children's Age
Air Lead
Drinking Water
Lead
Funding
Model I
Region
Time*Region
Region ^Quarter

Percent Multiple
Race
Median Rent
Percent Rented
Percent Single
Parent
Median Year
Built

Median Air Lead,
95th percentile

lpbw*water
Current HUD
Funding * Time
Period
Model 2
Region
Time*Region
Region*Quarter
Percent of HHs
with No Wages
Percent Multiple
Race
Median Rent

Percent Single
Parent
Percent Built Pre-
1940

Median Air Lead,
95th percentile .

lpbw*water
Current HUD
Funding * Time
Period
Model 3
Region
Time*Region
Region*Quarter
Median Household
Income
Percent Asian


Percent Single Parent
Percent Occupied
Built Pre-1940

Median. Air Lead,
99* percentile
TRI Lead from Air
Stacks, 99lh percentile

Current HUD
Funding * Time
Period
Model 4
Region
Time*Region
Region*Quarter

Percent Asian

Percent Vacant
Percent Single
Parent
Percent Built Pre-
1950
Percent ฃ Six
Years Old
20 High Counties

lpbw*water
Current HUD
Funding * Time
Period
Model 5
Region
Time*Region
Region ^Quarter

Percent Black

Percent Vacant
Percent Single
Parent
Percent Built Pre-
1950

20 High Counties

lpbw*water
Current HUD
Funding * Time
Period
Model 6
Region .
Time*Region
Region*Quarter
Percent of HHs
with No Wages
.

Percent Vacant
Percent Single
Parent
Percent Occupied
Built Pre-1950
Percent < Six Years
Old
20 High Counties

Mean Water Lead,
95* percentile
Current HUD
Funding * Time
Period
                                                                   39
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality guidelines.  It has not been formally
disseminated by EPA. It does not represent and should not be construed to represent any Agency determination or policy.

-------
Table 5-2. Model 1 (Unweighted GM
Region 1
Effect
Region
Time*Region
Region*Quarter-l
Region*Quartei-2
Region*Quarter-3
Region*Quarter-4
Estimate
3.916
0.054
-0.085
-0.062
0.159
0.000
StdErr
0.125
0.018
0.039
0.039
0.039
•
P-Value
<.0001
0.0023
00282
0.1109
<0001

Region 4
Effect
Region
Time*Region
Region*Quarter-l
Region*Quaner-2
Region*Quarter-3
Region*Quaiter-4
Estimate
4.299
-0.166
-0.159
-0.015
0.081
0.000
StdErr
0.084
0.006
0.015
0.015
0.014

P-Value
<.000l
•e.OOOl
<0001
0.2934
<.0001

Region 7
Effect
Region
Time*Region
Region*Quarter-l
Region*Quaner-2
Region*Quarter-3
Region*Quarter-4
Estimate
4.853
-0.159
-0.293
-0.059
0.167
0.000
StdErr
0.074
0.008
0.018
0.018
0.018

P-Value
<.0001
<.0001
<0001
0.0012
<.0001

Region 10
Effect
Region
Time*Region
Region*Quarter-l
Region*Quaiter-2
Region*Quarter-3
Region*Quarter-4
Estimate
3.449
-0.127
-0.148
-0.116
0.076
0.000
StdErr
0.160
0.028
0.082
0.082
0.081

P-Value
<.0001
<.0001
0.0717
0.1591
i|
0.3504
ii
M
Parameter Estimates for Multivariate National Model
Region!
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quarter-3
Region*Quarter-4
Estimate
4.288
-0.097
-0.292
-0.125
0.122
0.000
StdErr
0.182
0.029
0.067
0.067
0.067
•
P-Valne
<0001
0.0009
<.0001
0.0627
0.0678

Region 3
Effect
Region
Time*Region
Region*Quarter-l
Region*Quaiter-2
Region*Quarter-3
Region*Quarter-4
Estimate
4.292
-0.209
-0.219
-0.130
0.075
0.000
StdErr
O.OS8
0.009
0.023
0.022
0.022

P-Value
<.0001
<.OOQ1
<.0001
<.0001
0.0006

Region S
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quarter-3
Region*Quarter-4
Estimate
3.881
-0.198
-0.267
-0.109
0.193
0.000
StdErr
0.076
0.006
0.015
0.015
0.015
•
P-Value
<.000l
<.0001
<.0001
<0001
<0001

Region 6
Effect
Region
Time*Region
Region*Quarter-l
Region*Quaiter-2
Region*Quarter-3
Region*Quaner-4
Estimate
4.116
-0.112
-0.155
-0.026
0.098
0.000
StdErr
0.082
0.008
0020,
0.020
0.020

P-Value
<.0001
<.0001
<.0001
0.2093
<.0001

Region 8
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quarter-3
Region*Quaner-4
Estimate
3.673
-0.106
-0.333
-0.187
0.029
0.000
StdErr
0.118
0.021
0.067
0.065
0.065

P-Value
<0001
<.0001
<.0001
0004
0.6546
•
Region 9
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quarter-3
Region*Quaner-4
Estimate
' 5.973
-0.251
-0.145
-0.179
-0.035
0.000
StdErr
0.162
0.027
0.077
0.077
0077

P-Value
<.OOOI

-------
      17.5

      15.0

      12.5
   ง  10.0
       7.5

       5.0

       2.5
           -5.4 -4.8 -4.2 -3.6  -3  -2.4 -1.8 -1.2 -0.6  0  0.6  1.2  1.8  2.4  3  3.6  4.2  4.8
                                    Residuals from Model-1 (GM)
Figure 5-1. Histograms of Residuals from Fitted National Multivariate Model 1
      12-


      10-
  (3

  ป    6-
  o

  ฃ    4 -
       2-


       0
         0
i I i  i i i t  i i i i I i  i i i i  i i i i I  i i i i  i i i i i  I i i i i  i

 4            6            8           10

       Observed (GM)
T~T
 12
                                R2 from Fitted Regression=0.814

Figure 5-2.  Plot of National Multivariate Model Predicted Values versus Observed with
             Fitted Regression Line and 45ฐ Reference Line for Unweighted Geometric
             Mean Response.
                                              41
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
Fable 5-3. Model 2 (Weighted GM)
Region 1
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quaiter-3
Region*Quarter-4
Estimate
2.646
0.028
-0.106
-0.067
0.150
0.000
StdErr
0.163
0.015
0.014
0.014
0014

P-Value
<.0001
0.0598
<.0001
<.0001
<0001

                                      Parameter Estimates for Multivariate National Model
Region 4
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quarter-3
Region*Quarter-4
Estimate
3.243
-0.168
-0.190
-0066
0.054
0.000
StdErr
0.125
0.006
0.013
0.013
0.012

P-Value
<.0001
<.0001
<.0001
<.0001
<.0001

Region 7
Effect
Region
Time*Region
Region*Quarter-l
Region*Quaiter-2
Region*Quaner-3
Region*Quarter-4
Estimate
3.792
-0.193
-0.310
-0.043
0.172
0.000
StdErr
0.129
0.008
0.019
0.019
0.019

P-Value
<.0001
<.0001
<0001
0.0231
<.0001

Region 10
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quarter-3
Region*Quarter-4
Estimate
2285
-0.113
-0.269
-0.178
0023
0000
StdErr
0.201
0.033
0.088
0.089
0.090

P-Value
<0001
0.0006
00021
0.046
0.7989

Region 2
Effect
Region
Time*Region
Region*Quarter-l
Region*Quaitet-2
Region*Quarter-3
Region*Quarter-4
Estimate
2.745
-0.024
-0.186
-0.046
0.143
0.000
StdErr
0.197
0.024
0.018
0.018
0.018

P-Value
<.0001
0.3093
<.0001
00114
<.0001

Region 3
Effect
Region
Time*Region
Region*Quaiter-l
Region*Quaner-2
Region*Quarter-3
Region*Quarter-4
Estimate
3.165
-0.229
-0.255
-0.120
0131
0.000
StdErr
0.137
0.009
0.017
0.017
0.016

P-Value
<.0001
<0001
<0001
<.0001
<0001

Region 5
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarier-2
Region*Quarter-3
Region*Quarter-4
Estimate
2.717
X).192
-0.411
-0.183
0.364
0.000
StdErr
0.128
0.006
0.010
0.010
0.009

P-Value
<.0001
<0001
•e.OOOl
<.0001
<.0001

Region 6
Effect
Region .
Time*Region
Region*Quaner-l
Region*Quarter-2
Region*Quarter-3
Region*Quarter-4
Estimate
3.075
-0.106
-0.093
0.001
0.098
0.000
StdErr
0.124
0.008
0.016
0.016
0.015

P-Value
<.0001
<.0001
<.0001
0.956
<.0001

Region 8
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quarter-3
Region*Quarter-4
Estimate
2.475
-0.069
-0.306
-0.161
0.059
0.000
StdErr
0.165
0.027
0.097
0.094
0.093

P-Value
<.0001
0.0114
0.0015
0.0848
0.5229

Region 9
Effect
Region
Hme*Region
Region*Quarter-l
Region*Quarter-2
Region*Quarter-3
Region*Quarter-4
Estimate
4.157
-0231
-0.252
-0.340
-0.169
0.000
StdErr
0.194
0.028
0.033
0.030
0029

P-Value
<.0001
<.0001
<.0001
<.0001
<.0001

Other Predictors
Effect
Percent No
Household Wage
Percent Multiple
Races
Median Rent ($):
Percent Single
Parent Households
Percent Built Pre-
1940
Median air lead, 95th
percenule
lpbw*water=l
lpbw*water=2
Current HUD
Funding * Quarter=l
Current HUD
Funding * Quarter=2
Current HUD
Funding * Quarter=3
Current HUD
Funding * Quarter=4
Estimate
0.562
-6.819
-0.070
0.896
2.044
0.271
0.048
0.004
-0.005
-0.029
0.031
0.024
StdErr
0.245
1.040
0015
0.176
0.142
0.053
0.007
0.019
0.002
0.002
0.002
0.002
P-Value
0.0219
<.0001
<.0001
<.0001
<0001
<0001
<.0001
0.8136
0.0321
<.0001
<0001
<.0001
Variance Components
Effect
<
a\*
<
^Error


Estimate
0.358
-0.042
0.014
32.576


StdErr






P-Value







                                                        42
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
      17.5

      15.0


      12.5

      10.0

       7.5

       5.0


       2.5

        0
r   1     i   ' T
                                                         r
                                                                  T~^~T
           -5.7 -4.9  -4.1  -3.3  -2.5 -1.7  -0.9  -0.1  0.7  1.5   2.3  3.1  3.9  4.7  5.5  6.3  7.1

                                Residuals from Model-2 (Weighted GM)


Figure 5-3. Histograms of Residuals from Fitted National Multivariate Model 2
  C5
      12-
      10
       8-
       6:
       2


       01

                !ฃ
                                   4
                                                                        10
                                                                                    12
                                     Observed (Weighted GM)

                                R2 from Fitted Regression=0.853

Figure 5-4.  Plot of National Multivariate Model Predicted Values versus Observed with
             Fitted Regression Line and 45ฐ Reference Line for Weighted Geometric
             Mean Response
                                              43
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
Fable 5-4. Model 3 (Proportion >5
Region 1
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quarter-3
Region*Quarter-4
Estimate
-2.397
-0.087
-0.173
-0.086
0.215
0.000
StdErr
0.114
0.012
0.008
0.008 >
0.007

P-Value
<.0001
<0001
<.0001
<0001
<.0001

ug/dl_) Parameter Estimates for Multivariate National Model

Region 2
Effect
Region
Time*Region
Region*Quaiter-l
Region*Quaiter-2
Region*Quarter-3
Region*Quarter-4
Estimate
-2.522
-0.116
-0.316
-0.181
0.133
0.000
StdErr
0.140
0.019
0.010
0.009
0.009

P-Value
<0001
<0001
<.0001
<.0001
<.0001








Region 3
Effect
Region
Time*Region
Region*Quaher-l
Region*Quarter-2
Region*Quarter-3
Region*Quarter-4
Estimate
-2.108
-0.181
-0.281
-0.128
0.141
0.000
StdErr
0.092
0.007
0.008
0.008
0.008

P-Value
<.0001
<.0001
•e.OOOl
<.0001
<.0001
•
Region 4
Effect
Region
Time*Region
Region*Quaner-l
Region*Quarter-2
Region*Quaner-3
Region*Quarter-4
Estimate
-1.835
-0.188
-0.205
-0.040
0.058
0.000
StdErr
0.081
0.004
0.006
0.006
0.006

P-Value
<.0001
<.0001
<.0001
<.0001
<.0001

Region 7
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quarter-3
Region*Quarter-4
Estimate
-2.346
-0.118
-0.302
-0.014
0.168
0000
StdErr
0.091
0.006
0.009
0.009
0.008

P-Value
<0001
<0001
<.0001
0.098
<.0001

Region 10
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quarter-3
Region*Quarter-4
Estimate
-2.751
-0.138
-0.382
-0.221
0.108
0.000
StdErr
0.134
0.024
0.050
0.050
0.048

P-Value
<.0001
•e.OOOl
<.0001
<.0001
0.0257

Region 5
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quaiter-3
Region*Quarier-4
Estimate
-2.510
-0.160
-0.352
-0.133
0.264
0.000
StdErr
0.094
0.005
0.004
0.004
0.004

P-Value
<.0001
<.0001
<.0001
<0001
<.0001









Region 6
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quarter-3
Region*Quarter-4
Estimate
-1.995
-0.115
-0.124
-0.023
0.089
0.000
StdErr
0.080
0.006
0.009
0.009
0.008

P-Value
•s.OOOl
<0001
<0001
0.0082
<0001

Region 8
Effect
Region
Time*Region
Region*Quarter-l
Region*Quaiter-2
Region*Quarter-3
Region*Quaner-4
Estimate
-2.535
'-0.099
-0.317
-0.266
0.017
0.000
StdErr
0.109
0.018
0.056
0.054
0.050

P-Value
<.0001
<0001
<.0001
<.0001
0.7429

Region 9
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quaiter-3
Region*Quarter-4
Estimate
-1.387
-0.151
-0.211
-0.397
-0.143
0.000
StdErr
0.120
0.019
0.021
0.020
0.019

P-Value
<.0001
<.0001
<.0001
<.0001
<.0001

Other Predictors
Effect
Median Household
Income
Percent Asian Alone
Percent Single
Parent Households
Percent Occupied
Built Pre-1950
Median air lead, 99th
percentile
TRI: 95"1 percentile
air stacks, lead only
Current HUD
Funding * Quarter=l
Current HUD
Funding * Quarter=2
Current HUD
Funding * Quarter=3
Current HUD
Funding * Quarter=4
Estimate
-0.005
-3.491
1.326
2.319
0.428
0.146
-0.009
-0.006
0.015
0.012
StdErr
0.001
0.525
0.150
0.096
0.092
0.091
0.001
0.001
0.001
0.001
P-Value
0.0008
<.0001
<.0001
<.0001
<.0001
0.1076
<.0001
<.0001
•e.OOOl
<.0001
Variance Components
Effect
<
ฐ*4 •
ฐ\



Estimate
0.210
-0.020
0.009



StdErr
0.007
0.001
0.000



P-Value







                                                            44
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines.. It-has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                   17.5

                   15.0

                   12.5

                g  10.0 '

                4  7.5

                   5.0

                   2.5

                    0
                       -0.7 -0.575 -0.45 -0.325 -0.2 -0.075 0.05  0.175  0.3  0.425  0.55  0.675  0.8
                                          Residuals from Model-3 (P5)

      Figure 5-5.  Histograms of Residuals from Fitted National Multivariate Model 3
                    1.0

                             R2 from Fitted Regression=0.859
                      0.0
                                 0.2
                                                                  0.8
                                                                             1.0
                                              Observed (PS)
Figure 5-6a.   Plot of National Multivariate Model Predicted Values versus Observed with
               Fitted Regression Line and 45ฐ Reference Line for Proportion of Children
               with BLL > 5 pg/dL
7-:
6 -
5 -

3-.
2 '
1 -
0 \
-1 -.
-2 -
-3 -
-4 -
-5-;
-6 j
                        R2 from Fitted Regressions 0.840
                          -5   -4    -3
                                       -2-101    23
                                          Observed (PS - Logit Scale)
Figure 5-6b.   Plot of National Multivariate Model Predicted Values versus Observed with
               Fitted Regression Line and 45ฐ Reference Line for Proportion of Children
               with BLL > 5 \ig/dL (Logit Scale)
                                              45
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed
to represent any Agency determination or policy.

-------
Fable 5-5. Model 4 (Proportion > 10 ug/dl_) Parameter Estimates for Multivariate National Model
Region 1
Effect
Region
Time*Region
Region*Quarter-l
Region*Quaner-2
Region*Quarter-3
Region*Quarter-4
Estimate
-4.601
-0.097
-0.248
-0.087
0.274
0.000
StdErr
0.155
0.011
0.015
0.015
0.013

P-Value
<0001
<.0001
<.0001
<.0001
<.0001

Region 2
Effect
Region
Time*Region
Region*Quaiter-l
Region*Quarter-2
Region*Quaiter-3
Region*Quarter-4
Estimate
-4.821
-0.175
-0.467
-0.205
0.239
0.000
StdErr
0.183
0.018
0.020
0.018
0.017

P-Value
<.0001
<.0001
<0001
<0001
<.0001

Region 3
Effect
Region
Time*Region
Region*Quaiter-l
Region*Quarter-2
Region*Quarter-3
Region*Quaiter-4
Estimate
-4.475
-0.202
-0.339
-0.106
0.233
0.000
StdErr
0.135
0.007
0.014
0.013
0.012

P-Value
<.0001
<0001
<.0001
<.0001
<.0001

Region 4
Effect
Region -
Time*Region
Region*Quaiter-l
Region*Quarter-2
Region*Quarter-3 .
Region*Quarter-4
Estimate
-4.451
. -0.196
-0.248
0.013
0.116
0.000
StdErr
0.134
0.005
0.015
0.015
0.014

P-Value
•c.OOOl
<.0001
<0001
0.3615
<.0001









Region 5
Effect
Region
Time*Region
Region*Quarter-l
Region*Quaiter-2
Region*Quarter-3
Region*Quaiter-4
Estimate
-4.806
-0.184
-0.483
-0.161
0.334
0.000
StdErr
0.140
0.005
0.007'
0.007
0.006
•
P-Value
<0001
<.0001
<.0001
<.0001
<.0001

Region 6
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quarter-3
Region*Quaiter-4
Estimate
-4.427
-0.146
-0.190
-0.071
0.053
0.000
StdErr
0.142
0.007
0.019
0.019
0.018

P-Value
<0001
•c.OOOl
•c.OOOl
0.0002
00039

Region 7
Effect
Region
Time*Region
Region*Quarter-l
Region*Quaiter-2
Region*Quarter-3
Region*Quaiter-4
Estimate
• -4.551
-0.158
-0.489
-0.033
0.187
0.000
StdErr
0.144
0.006
0.017
0.015
0.014

P-Value
<0001
•c.0001
<.0001
00276
<0001









Region 8
Effect
Region
Time*Region
Region*Quarter-l
Region*Quaiter-2
Region*Quaiter-3
Region*Quarter-4
Estimate
-4.930
-0.176
-0.600
-0.212
-0.163
0.000
StdErr
0.187
0.028
0.133
.0.113
0.107

P-Value

-------
                                -0.285 -0.21 -0.135 -0.06 0.015 0.09  0.165  0.24  0.315  0.39 0.465  0.54 0.615

                                                 Residuals trom Model-4(P10)


              Figure 5-7.  Histograms of Residuals from Fitted National Multivariate Model 4
                           0.0-
                                 R2 from Fitted Regression 0.878
                                                        0.4

                                                    Observed (P10)
                                                                                   0.8
Figure 5-8a.   Plot of National Multivariate Model Predicted Values versus Observed with Fitted
               Regression Line and 45ฐ Reference Line for Proportion of Children with BLL > 10ug/dL.
                               R2 from Fitted Regression 0.837
                           -8 •
                                                Observed (P10 - Loglt Scale)
Figure 5-8b.   Plot of National Multivariate Model Predicted Values versus Observed with Fitted
               Regression Line and 45ฐ Reference Line for Proportion of Children with BLL > 10 [jg/dL
               (Logit Scale)
                                                     47
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
Table 5-6.  Model 5 (Proportion > 15 ug/dL) Parameter Estimates for Multivariate National Model
Region 1
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quarter-3
Region*Quaiter-4
Estimate
-5.454
-0.087
-0.249
-0.047
0.344
0.000
StdErr
0.127
0.013
0.027
0.025
0.023

P-Value
<.0001
<.0001
<.0001
0.0588
<0001









Region 2
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quaiter-3
Region*Quarter-4
Estimate
-5.512
-0.207
-0.451
-0.151
0.300
0.000
StdErr
0.152
0.021
0.031
0.029
0.026
-
P-Value
<0001
<.0001
<.0001
<.0001
<0001

Region 3
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quarter-3
Region*Quarter-4
Estimate

-5.170
-0.225
-0.381
-0.081
0.266
0.000
StdErr
0.098
0.009
0.021
0.020
0.018

P-Value
<.0001
<.0001
<.0001
<.0001
<.0001

Region 4
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quaiter-3
Region*Quarter-4
Estimate
-5.369
-0.171
-0.284
0.078
0.146
0.000
StdErr
0.089
0.007
0.029
0.027
0.026

P-Value
<.0001
<.0001
<.0001
00033
<.0001









Region 5
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quarter-3
Region*Quarter-4
Estimate
-5.539
-0.178
-0.527
-0.130
0.400
0.000
StdErr
0.094
0.006
0.011
0.010
0.009

P-Value
<.0001
<.0001
<.0001
<0001
<.0001
•
Region 6
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quaiter-3
Region*Quatter-4
Estimate
-5.299
-0.118
-0.177
-0.109
-0.009
0.000
StdErr
0.095
0.010
0.033
0.033
0.031

P-Value
<.0001
<.OOOI
<.OOOI
00008
0.7778

Region 7
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quarter-3
Region*Quaiter-4
Estimate
-5.399
-0.151
-0.520
0.059
0.263
0.000
StdErr
0.100
0.008
0.027
0.024
0.023

P-Value
<.0001
<0001
<.0001
00127
<0001
•








Region 8
Effect
Region
Time*Region
Region*Quarter-l
Region*Quaner-2
Region*Quarter-3
Region*Quaner-4
Estimate
-5.621
-0.170
-0.850
-0.160
-0.193
0000
StdErr
0.193
0.041
0.234
0.179
0.171

P-Value
<.0001
<.0001
00003
0.3706
0.2582

Region 9
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quarter-3
Region "Quarter-4
Estimate
-3.675
-0.250
-0.388
-0.435
-0.129
0.000
StdErr
0.147
0.024
0.048
0.046
0.044

P-Value
•s.OOOl
<.0001
<0001
<.000l
0.0035

Region 10
Effect
Region
Time*Region
Region*Quatter-l
Region*Quarter-2
Region*Quarter-3
Region*Quarter-4
Estimate
-5.312
-0.169
-0.895
-0.424
-0.018
0.000
StdErr
0.213
0.041
0.172
0.154
0.143

P-Value
<.0001
<.0001
<.0001
0.0058
0.9023










Other Predictors
Effect
Percent Black AJone
Percent Vacant
Percent Single Parent
Households
Percent Built Before
1950
20 Counties Highest
Air Lead
Ipbw* waters 1
lpbw*water=2
Current HUD
Funding * Quarters 1
Current HUD
Funding * Quarter=2
Current HUD
Funding * Quaner=3
Current HUD
Funding * Quarter=4
Estimate
0.261
-0.496
1.203
2.815
0.603
0.030
0.045
0.005
-0.002
0.003
0.008
StdErr
0.177
0.189
0.339
0.148
0.161
0.011
0.028
0.002
0.003
0.003
0.002
P-Value
0.1406
00086
0.0004
<.0001
0.0002
00067
0.1108
0.0604
0.4442
0.2445
0.0001
Variance Components
Effect
<
a\A
<



Estimate
0.365
-0.018
0.009



StdErr
0.016
0.002
0.001



P-Value







                                                      48
This information is distributed solely forthe purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                  60


                  50


                  40


                  30


                  20


                  10
                   0 	1	1	1	T~"                   —I—
                     -0.138 -0.078 -0.018 0.042  0.102  0.162 0.222  0.282 0.342  0.402 0.462  0.522

                                        Residuals from Model-5 (P15)


      Figure 5-9. Histograms of Residuals from Fitted National Multivariate Model 5
                         from Fitted Regression 0.837
                                             Observed (P15)
Figure 5-1 Oa. Plot of National Multivariate Model Predicted Values versus Observed with
              Fitted Regression Line and 45ฐ Reference Line for Proportion of Children
              with BLL> 15 (jg/dL.
                &
                I   -si
                      R2 from Fitted Regression 0.777
                                        n—I—I—|—I—I—1—I—I—I—I—I—I—|—I—I—r

                                           -5          -3

                                         Observed (P15 - Logit Scale)
Figure 5-1 Ob. Plot of National Multivariate Model Predicted Values versus Observed with
               Fitted Regression Line and 45ฐ Reference Line for Proportion of Children
               with BLL > 15 pg/dL (Logit Scale)

                                              49
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
Fable 5-7. Model 6 (Proportion > 25 ug/dL) Parameter Estimates for Multivariate National Model
Region 1
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quarter-3
Region*Quarter-4
Estimate
-7.452
-0.075
-0.335
0.021
0.362
0.000
StdErr
0.244
0.016
0055
0.050
0.046

P-Value
<.0001
<0001
<.0001
0.6774
<.0001








Region 2
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quarter-3
Region*Quarter-4
Estimate
-7.271
-0.252
-0.385
-0.001
0.396
0.000
StdErr
0.270
0.026
0.061
0.055
0.052

P-Value
<0001
<.0001
<.0001
0.9921
<.0001

Region 3
Effect
Region
Time*Region
Region*Quarter-l
Region*Quaiter-2
Region*Quarter-3
Region*Quarter-4
Estimate
-7.022
-0.218
-0.481
-0.041
0.287
0.000
StdErr
0.236
0.013
0.045
0.041
0037

P-Value
<.0001
<0001
•c.OOOl
0.3149
<.0001

Region 4
Effect
Region
Time*Region
Region*Quarter-l
Region*Quaner-2
Region*Quaner-3
Region*Quarter-4
Estimate
-7.528
-0.142
-0.133
0.192
0.276
0.000
StdErr
0.245
0.010
0.062
0.057
0.056

P-Value
<.0001
<.0001
0.0303
0.0008
<0001

Region 5
Effect
•Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quarter-3
Region*Quarter-4
Estimate
-7.515
-0.158
-0.559
-0.034
0.494
0.000
StdErr
0.237
0.008
0.023
0.020
0.018

P-Value
<0001
<.0001
<.0001
0.0907
<.0001
•
Region 6
Effect
Region
Time*Region
Region*Quaiter-l
Region*Quarter-2
Region*Quaiter-3
Region*Quarter-4
Estimate
-7.465
-0.115
-0.155
-0.006
0.091
0000
StdErr
0.260
0.016
0.072
0.071
0.069

P-Value
<.0001
<.0001
0.0323
0.9335
0.1842

Region 7
Effect
Region
Time*Region
Region*Quarter-l
Region*Quarter-2
Region*Quarter-3
Region*Quaner-4
Estimate
-7.592
-0.121
-0.449
0.280
0.450
0.000
StdErr
0.249
0.012
0.059
0.049
0.048

P-Value
<.0001
<.0001
<0001
<.0001
<.0001

Region 8
Effect
Region
Time*Region
Region*Quaner-l
Region*Quarter-2
Region*Quarter-3
Region*Quarter-4
Estimate
-7.456
-0.088
-1.185
0.020
-0503
0.000
StdErr
0.376
0.071
0.515
0.329
0.355

P-Value
<0001
0.216
0.0215
0.9519
0.157

Region 9
Effect
Region
Time*Region
Region*Quaiter-l
Region*Quarter-2
Region*Quarter-3
Region*Quarter-4
Estimate
-5.956
-0.220
-0.340
-0.255
-0.153
0.000
StdErr
0.288
0.034
0.095
0.089
0.089

P-Value
<0001
<.0001
00003
0.0043
00861

Region 10
Effect
Region
Time*Region
Region*Quarter-l
Region*Quaner-2
Region*Quarter-3
Region*Quarter-4

Estimate
-7.354
-0.123
-0.225
0.015
0.009
0.000

StdErr
0.404
0.061
0.340
0.325
0.333

P-Value
<.0001
0.044
0.5089
0.9634
0.9781


Other Predictors
Effect
Percent No
Household Wage
Percent Vacant
Percent Single Parent
Households
Percent Occupied
Built Ple-1940
Percent Less than 6
Years of Age
20 Counties Highest
Air Lead
SDWIS: 95th
percentile water lead
Current HUD
Funding * Quarter=l
Current HUD
Funding * Quarter=2
Current HUD
Funding * Quarter=3
Current HUD
Funding * Quaner=4
Estimate
1.069
-1.303
1.426
3.059
4.484
0.312
-0.050
0.003
0.000
-0.001
0.013
StdErr
0.431
0.307
0.291
0238
1.924
0.203
0.025
0.005
0.005
1
0.005
0.004
P-Value
0.0133
<0001
<0001
<.OOOI
00198
0.1253
0.0513
0.5918
0.9437
0.8993
0.0003
Variance Components
Effect
<
a\*
ฐ\



Estimate
0.370
-0.018
0.010



StdErr
0.022
0.003
0.001



P-Value







                                                            50
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not. been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                   90

                   80

                   70

                   60

                   50

                   40

                   30

                   20

                   10
                   o ~~~T—                                          —T~	1	r
                     -0.063 -0.033 -0.003 0.027 0.057 0.087  0.117 0.147 0.177 0.207 0.237  0.267  0.297 0.327

                                         Residuals from Model-6 (P25)


      Figure 5-11. Histograms of Residuals from Fitted National Multivariate Model 6
                    0.00-
                          R2 from Fitted Regression 0.659
—i—
 0.07
                                            0.14

                                             Observed (P25)
                                                       i—
                                                      0.21
Figure 5-12a.  Plot of National Multivariate Model Predicted Values versus Observed with
               Fitted Regression Line and 45ฐ Reference Line for Proportion of Children
               with BLL > 25|jg/dL
                 I
                    -10-

                    -11 -
                      -11
                            R2 from Fitted Regression 0.658
                           -10
                                          -7    -6    -5    -4

                                          Observed (P25 - Logit Scale)
Figure 5-12b. Plot of National Multivariate Model Predicted Values versus Observed with
               Fitted Regression Line and 45ฐ Reference Line for Proportion of Children
               with BLL > 25 Mg/dL (Logit Scale)

                                              51
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
5.2    High Resolution Modeling Results

The Massachusetts final multivariate models were constructed similarly to the national models.
One basic difference in the Massachusetts models is that there was no EPA Region.  Thus, there
was no area variable included other than census tract.  As in the national model, time period and
quarter were significant predictors; however, in the Massachusetts model they are not interacted
with an area variable. Table 5-8 presents the full set of variables included in the final
multivariate models for Models 1  through 5. Model 6 was not fit for the Massachusetts data
because of the scarcity of data above 25 ug/dL.

Among the demographic variables, housing cost, occupancy, family structure, and housing age
were significant predictors in all five models. Median Rent was the selected housing cost
variable in all five models. For occupancy, Percent Rented was the selected variable in four of
the five models.  Percent of Single Parent Households is the family structure variable in all
models. Three housing age variables were included across the five models, but Percent Built
Pre-1950 was the included variable in Models 1 to 3.

Race and Income variables were included in four of the five final models. Median Household
Income and Percent Multiple Races were the two variables used in all four  models.  Children's
Age, Education, and Population each had a variable included in one of the final models.  Number
of Children less than or equal to six years old and Total Population were included in Model  3.
Percent Without 9th Grade Education was included in Model 4.

Unlike the national models, none  of the environmental variables were included in the final
multivariate models for Massachusetts. On the other hand, the housing inspection data from
Massachusetts were predictive and included in all of the final models.  The percentage of units
passing the Massachusetts standard of care (calculated using the MDPH method) was included in
all five models. Additionally, the percentage of units failing the Massachusetts standard of care
(calculated using the MDPH  method) was included in Models 4 and 5.

The selected programmatic funding variable was included in Models 1,2, and 5. Current State
Funding ($ per Child) was used in the GM models and Cumulative CDC Funding ($ per tract)
was used in Model 5.

As with the national models,  parameter estimates and associated standard errors and p-values are
presented for all models in Table  5-9. Figures 5-13 to 5-22 contain the histograms of residuals
and plots of observed versus  predicted values that allow assessment of the various model fits.

These plots suggest that models 1-3 are performing well, with Models 4 and 5 providing a
somewhat suboptimal fit (perhaps due to fewer children being observed above the 10 and 15
ug/dL threshold values in Massachusetts).  The weighted regression line fit to the observed vs.
predicted plots (shown  in blue) also demonstrates a systematic degradation in model
performance from Models 3 through 5, with the R2 value diminishing as the blood-lead threshold
value increases.  Similar to the National Models, the High Resolution Multivariate Models in
Massachusetts tend to underpredict for census-tract/quarter combinations with higher geometric

                                           52
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed
to represent any Agency determination or policy.

-------
mean blood-lead concentrations and higher exceedance proportions. Further exploration may be
necessary to determine whether these higher values represent county/quarter combinations with
fairly sparse data (i.e. few observations) - which might explain why they would have been less
influential in Models 2 through 6 which are influenced by the number of observations associated
with each observed value.
                                             53
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA.  It does not represent and should not be construed
to represent any Agency determination or policy.

-------
Table 5-8. Summar
Variable Type
Time
Income
Race
Housing Cost
Occupancy
Family Structure
Housing Age
Children's Age
Education
Population
Housing Inspection
Funding
/ of Variables Included in Final Massachusetts Multivariate Model
Model 1
Time Period, Quarter
Median Household
Income
Percent Multiple Race
Median Rent
Percent Rented
Percent Single Parent
Percent Built Pre-1950



P4 - % Passing
Standard of Care,
MDPH Method

Current State Funding
($ per Child)
Model 2
Time Period, Quarter
Median Household
Income
Percent Multiple Race
Median Rent
Percent Rented
Percent Single Parent
Percent Built Pre-1950



P4 - % Passing
Standard of Care,
MDPH Method

Current State Funding
($ per Child)
Model 3
Time Period, Quarter
Median Household
Income
Percent Multiple Race
Median Rent
Percent Rented
Percent Single Parent
Percent Built Pre-
1950
Number less than 6
years old

Total Population
P4 - % Passing
Standard of Care,
MDPH Method


Model 4
Time Period, Quarter
Median Household
Income
Percent Multiple Race
Median Rent
Percent Rented
Percent Single Parent
Percent Occupied
Built Pre-1940

Percent without 9*
Grade education

P4 - % Passing
Standard of Care,
MDPH Method
F4 - % failing
standard of care,
MDPH Method

Model 5
Time Period, Quarter


Median Rent
Percent Vacant
Percent Single Parent
Percent Occupied
Built Pre-1980



P4 - % Passing.
Standard of Care,
MDPH Method
F4 - % failing standard
of care, MDPH
Method
Cumulative CDC
Funding ($ per tract)
                                                                       54
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality guidelines.  It has not beenformatty
disseminated by EPA.  It does not represent and should not be construed to represent any Agency determination or policy.

-------
Table 5-9. Massachusetts Multivariate Model Estimates
Model
1
(Geometric
Mean)
2
(Weighted
Geometric
Mean)
Effect
Intercept
Time
Quarter (Season)
Median Household Income
Percent Multiple Races
Median Rent ($):
' Percent Rented Units
Percent Single Parent Households
Percent Units Built Before 1950
p4
Current State Funding
<
ฐ\*
ฐ\
&Em>r
Intercept
Time
Quarter (Season)
Median Household Income
Percent Multiple Races
Median Rent ($):
Percent Rented Units
Percent Single Parent Households
Percent Units Built Before 1950
p4
Current State Funding
ซ*
a\A
<
_2
ฐError
Levels
_
_
1
2
3
4

_

_
_
_
_
_
0.229
-0.026
0.004
0.191
_

1
2
3
4
_
_
_
1
_
_
_
_
0.218
-0.024
0.004
3.986
Estimate
2.290
-0.088
-0.187
-0.143
0.130
0.000
-0.008
3.909
-0.032
-0.568
0.702
0.849
-0.983
0.028
•
•
•
•
2.249
-0.087
-0.185
-0.137
0.127
0.000
-0.008
3.826
-0.032
-0.561
0.735
0.866
-0.933
0.031

•


Standard
Error
0.033
0.002
0.007
0.007
0.006
.
0.001
0.536
0.005
0.069
0.096
0.047
0.166
0.006




0.033
0.002
0.006
0.006
0.006
.
0.001
0.531
0.005
0.069
0.096
0.046
0.164
0.006




P-
Value
<.0001
<.0001
<.0001
<.0001
<.0001
.
<.0001
<.0001
<.0001
<.0001
<.0001
<.0001
<.0001
<.0001




<.0001
<.0001
<.0001
<.0001
<.0001
.
<.0001
<.0001
<.0001
<.0001
<.0001
<.0001
<.0001
<.0001




                                                55
This information is distributed solely for the purpose of pre-disseminat ion peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
Model


3
(Proportion
of Children
with Blood
Lead > 5
Hg/dL)
4
(Proportion
of Children
with Blood
Lead > 10
Hg/dL)
Effect
Intercept -
Time
Quarter (Season)
Median Household Income
Percent Multiple Races
Median Rent ($):
Percent Rented Units
Percent Single Parent Households
Percent Units Built Before 1950
Number Residents Less than Six Years of Age
Total Population
p4
<
ฐ
-------
Model
5
(Proportion
of Children
with Blood
Lead > 15
ug/dL)
Effect
Intercept
Time
Quarter (Season)
Median Rent ($):
Percent Vacant Units
Percent Single Parent Households
Percent Occupied Units Built Before 1980
f4
P4
Cumulative CDC Funding
<
a\A
ฐ\
Levels
_

1
2
3
4



_
_
_
_
0.249
-0.020
0.004
Estimate
-6.278
-0.093
-0.307
-0.093
0.332
0.000
-0.049
1.039
1.002
1.187
4.047
-1.476
0.000
0.035
0.008
0.002
Standard
Error
0.144
0.008
0.042
0.040
0.036
.
0.010
0.335
0.169
0.168
0.677
0.440
0.000



P-
Value
<.0001
<.0001
<.0001
0.019
<.0001
.
<.0001
0.0019
<.0001
<.0001
<.0001
0.0008
0.0127



                                                   57
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
20.0 "
17.5 '
15.0 -
12.5 "
10.0 '
7.5 "
5.0 '
2.5 "






/
^A.
i


i



-
/





(T






~N






\





\



L
A
PH_
        0    r
          -1.875  -1.275 -0.675  -0.075  0.525  1.125   1.725  2.325  2.925  3.525   4.125   4.725

                                   Residuals from Model-1 (GM)


Figure 5-13. Histograms of  Residuals from Fitted Massachusetts Multivariate Model 1
  o


  CL
8 -

7:

6:

5 :

4



2 -

1 •:

0-
          ;.;•: ->.    ;
          *i" ป < <  •
 -P^K^*:-;•'• •'
f&fEJui \' ':>ป. r.  . v
swTTv. ••' •- '
                                      pwra^^'
                                             4

                                       Observed (GM)
                                                                                  8
                              R2 from Fitted Regression = 0.697

Figure 5-14.  Plot of Massachusetts Multivariate Model Predicted Values versus
              Observed with Fitted Regression Line and 45ฐ Reference Line for
              Unweighted Geometric Mean Response
                                             58
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
      20.0

      17.5

      15.0

      12.5

      10.0

      7.5

      5.0

      2.5
M
        0 —r~     -V—T—V—V—Y1- ^T^     —r—T~ —r	1	1	1	r
           -1.8 -1.35 -0.9  -0.45  0   0.45   0.9 1.35   1.8  2.25  2.7  3.15  3.6  4.05  4.5   4.95

                                Residuals from Model-2 (Weighted GM)


Figure 5-15.  Histograms of Residuals from Fitted Massachusetts Multivariate Model 2
  I   ,
  01
  I   4
  •D
  %   3
                                   I I I I I I I I I I I I I ........ I I I I I ..... I ..... I I I I I ......... I

                                    345678
                                   Observed (Weighted GM)

                               R2 from Fitted Regression = 0.700

Figure 5-16.  Plot of Massachusetts Multivariate Model Predicted Values versus
              Observed with Fitted  Regression Line and 45ฐ Reference Line for Weighted
              Geometric Mean Response
                                             59
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
                       -0.34  -0.26 -0.18 -0.1  -0.02 0.06  0.14  0.22   0.3  0.38  0.46  0.54  0.62

                                         Residuals from Model-3 (P5)


  Figure 5-17. Histograms of Residuals from Fitted Massachusetts Multivariate Model 3
                   0.72-

                   0.24-


                   0.12-


                   0.00 -
                         R2 from Fitted Regression = 0.579
                                                                   0.72
                                                                          0.84
                                             Observed (P5)
Figure 5-18a. Plot of Massachusetts Multivariate Model Predicted Values versus
              Observed with Fitted Regression Line and 45ฐ Reference Line for
              Proportion of Children with BLL > 5 |jg/dL.
                ฃ  -21
                        R2 from Fitted Regression = 0.531
                                         Observed (P5 - Logit Scale)

Figure 5-18b. Plot of Massachusetts Multivariate Model Predicted Values versus
              Observed with Fitted Regression Line and 45ฐ Reference Line for
              Proportion of Children with BLL > ug/dL (Logit Scale)

                                             60
This information is distributed solely for the purpose ofpre- dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
35 "
30 "
25 '
20 '
15 "
10 "
5 '





/
A
_

/

0 — 1 	 T — ' — 1 	 1 	





"X
]]ffk___
( 1 l 1 1 1 l i j n^iyi i 1 1 1 i . 1 1 , ,y . [: . . | . . , j . , . 	 n 	 . . , 1
                       -0.12-0.09-0.06-0.03  0  0.03 0.06 0.09 0.12 0.15 0.18 0.21 0.24 0.27 0.3 0.33 0.36

                                         Residuals from Model-4 (P10)


  Figure 5-19.  Histograms of Residuals from Fitted Massachusetts Multivariate Model 4
                   0.4-
                I  0.2
                I
                   0.1 -
                        R  from Fitted Regression = 0.312
                      0.0
                                   0.1
                                                0.2
                                            Observed (P10)
                                                                           0.4
Figure 5-20a. Plot of Massachusetts Multivariate Model Predicted Values versus
              Observed with Fitted Regression Line and 45ฐ Reference Line for
              Proportion of Children with BLL > 10ug/dL
                I
                =  -2'
                !  -4
                I
                      R2 from Fitted Regression = 0.269
                     -6        -5        -4        -3        -2        -1         0
                                        Observed (P10 - Logit Scale)

Figure 5-20b. Plot of Massachusetts Multivariate Model Predicted Values versus
              Observed with Fitted Regression Line and 45ฐ Reference Line for
              Proportion of Children with BLL > 10jjg/dL (Logit Scale)
                                             61
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
                   60


                   50


                   40


                   30


                   20


                   10


                   0
                     -0.039 -0.015 0.009 0.033 0.057 0.081 0.105 0.129 0.153 0.177 0.201  0.225  0.249 0.273

                                        Residuals from Model-5 (P15)


  Figure 5-21. Histograms of Residuals from Fitted Massachusetts Multivariate Model 5
                ฃ
                ฃ  0.1
                        R2 from Fitted Regression = 0.119
                     0.0
                                            Observed (P15)
Figure 5-22a. Plot of Massachusetts Multivariate Model Predicted Values versus
              Observed with Fitted Regression Line and 45ฐ Reference Line for
              Proportion of Children with BLL > 15ug/dL.
                      R2 from Fitted Regression = 0.113
                     -5          -4          -3          -2          -1           0
                                        Observed (P15 - Logit Scale)

Figure 5-22b. Plot of Massachusetts Multivariate Model Predicted Values versus
              Observed with Fitted Regression Line and 45ฐ Reference Line for
              Proportion of Children with BLL > 15 \ig/dL (Logit Scale)

                                             62

This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
Appendix F presents predictions of areas of the country estimated to have the highest children's
blood lead levels.  These predictions were generated by averaging predicted values across the
four quarters of 2006.  Table F-l lists the 150 counties/townships in the U.S. with the highest
predicted GM blood lead levels (using Model 2) and proportion of children above 5,10,15, and
25 u,g/dL.  Table F-2 lists the 10 counties in each state with the highest levels of those same five
outcomes.  Table F-3 lists the 150 Massachusetts census tracts with the highest predicted GM
blood lead levels.  Figure F-l provides a map of these 150 Massachusetts census tracts.

These lists highlight the potential data issues with particular states.  For example, with Maine's
counties accounting for 19 of the highest 20 county-level GM blood lead levels, this raises the
possibility that Maine's surveillance data may hot have included low level blood tests.  Similarly,
many counties in Iowa have predicted blood lead levels between 4 and 5 ug/dL, putting them
among the highest in the country.  Because these levels are so widespread and similar throughout
the state, there is a possibility that there is a statewide data reporting problem.
                                             63
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. • It has not been formally disseminated by'EPA.  It does not represent and should not be construed
to represent any Agency determination or policy.

-------
6.0    GRAPHICAL PRESENTATION OF MODELING RESULTS

In addition to the discussion of the final multivariate modeling results in Section 5, it is
informative to be able to view the results visually. Two methods were utilized for graphical
presentation of the results - mapping and via use of an interactive software tool. Section 6.1
presents a subset of the maps generated, while Sections 6.2 and 6.3 discuss the interactive
software tool.

6.1     Maps of Observed and Predicted Blood Lead Outcomes

Mapping is an informative method to graphically present the results of the multivariate models.
Figure 6-1 contains maps displaying the observed levels of GM blood lead levels in 2000 and
2005 based on CDC's national surveillance data, and the comparable predicted GM blood lead
levels in 2000 and 2005. A key difference is that maps of observed levels contain many counties
with missing data either because they do not submit childhood lead surveillance data to CDC or
they have too few test records to be included in the analysis, while the maps of predicted levels
covers all counties in the country. Appendix G contains detailed maps from the national level
models of GM blood lead levels and proportion of children with DLLs > 10 ug/dL.

Because it is difficult to view many of the individual counties within the U.S.-level maps,
regional-level maps were also produced. Figures 6-2 and 6-3 contain examples of these for
Region V. Comparable maps for all regions are included in Appendix G.  With darker colors
representing areas of higher lead levels, it appears that lead levels are declining across Region V
from the 2000 to 2005 time period. Figure 6-4 contains maps of observed and predicted
proportion of children's blood lead levels in Massachusetts at the census tract level.  The Boston
area is enlarged to better show the tracts in that area.
                                          64
 This information is distributed solely for the purpose ofpre-dissemination-peer review under applicable information
 quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
 to represent any Agency determination or policy.

-------
                  Observed Blood-Lead Levels by County
                                 2000
             Observed Blood-Lead Levels by County
                            2005
     Geometric Mean PbB Level (ug/dL)
         0 - 1.999  I
         2-2.999  I
         3 - 3.999  I
         4 - 4.999
        LEGEND
Geometric Mean PbB Level (ug/dL)
    0 - 1.999  T •ฅ 5 - 5.999
    2 - 2.999  •• 6 - 7.999
    3 - 3.999  •• 8 +
    4 - 4.999      No data
                            United States
                       United States
                  Predicted Blood-Lead Levels by County
                                 2000
             Predicted Blood-Lead Levels by County
                            2005
     Geometric Mean PbB Level (ug/dL)
         0 - 1.999
         2 - 2.999
         3 - 3.999
         4 - 4.999
        LEGEND
Geometric Mean PbB Level (ug/dL)
    0 - 1.999  nra 5 - 5.999
    2 - 2.999  tm 6 - 7.999
    3-3.999  BBS +
  • 4 - 4.999  i   No data
Figure 6-1.  Observed and Predicted GM Blood Lead Levels in the United States by County, 2000 and 2005
                                                                            65
Thin information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality guidelines. It has not been formally
disseminated by EPA. It does not represent and should not be construed to represent any Agency determination or policy.

-------
          Observed Blood-Lead Levels by County
                        2000
                                                Observed Blood-Lead Levels by County
                                                              2005
           LEGEND
    Geometric Mean PbB Level (ug/dL)
       I - 1.999  BBS-5.999
       2 - 2.999
    BH 3 - 3.999
    BB 4 - 4.999
•i 6 - 7.999
••8 +
C3 No data
                               US EPA Region 5
       LEGEND
Geometric Mean PbB Level (ug/dL)
   .0-1.999  BBS-5.999
I   12 - 2.999  BB 5 - 7.999
i: _:. 3 - 3.999  BB 3 +
BB 4 - 4.999  CH No data
                                                                      US EPA Region 5
                                                               Predicted Blood-Lead Levels by County
                                                                             2005
           LEGEND
    Geometric Mean PbB Level (ug/dL)
        1-1.999  BBS-5.999
     _ 2 - 2.999
       3 - 3.999
    BB 4-4.999
  I 6 - 7.999
  18 +
  ] No data
                               US EPA Region 5
       LEGEND
Geometric Mean PbB Level (ug/dL)
   0-1.999  BBS-5.999
   2 - 2.999  BB 6 - 7.999
• 3 • 3.999  BB 8 +
BB 4 - 4.999  L™1 No data
                                                                     US EPA Region 5
Figure 6-2.  Observed and Predicted GM Blood Lead Levels in Region V by County, 2000
              and 2005
                                                    66
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed
to represent any Agency determination or policy.

-------
    Observed Proportion of Blood-Lead Levels >/= 10 ug/dL
                         2000
                          Observed Proportion of Blood-Lead Levels >/= 10 ug/dL
                                                2005
             LEGEND
    Proportion of PbB Levels >/'= 10 ug/dL
       0 - 0.0099    IB 0.075 - 0.0999
       0.01-0.0249  mf O.IO- 0.2499
    d. 0.025 - 0.0499 K 0.25 +
    L:~_ 0.05 - 0.0749  EH No data
US EPA Region 5
         LEGEND
 Proportion of PbB Levels >/= 10 ug/dL
   0 - 0.0099     BB 0.075 - 0.0999
   0.01-0.0249   •• O.IO -0.2499
d 0.025 - 0.0499  •• 0.25 +
•i 0.05 - 0.0749   CH No data
US EPA Region 5
    Predicted Proportion of Blood-Lead Levels >/= 10 ug/dL
                         2000
                           Predicted Proportion of Blood-Lead Levels >/= 10 ug/dL
                                                2005
             LEGEND
    Proportion of PbB Levels >/= 10 Ug/dL
      0 - 0.0099    Bl 0.075 - 0.0999
      0.01 - 0.0249  •• 0.10 - 0.2499
      0.025 • 0.0499 •ง 0.25 +
   •I 0.05 - 0.0749  di No data
US EPA Region 5
LEGEND
Proportion of PbB Levels >l= 10 ug/dL
0 - 0.0099 Hi 0.075 - 0.0999
I 0.0 1 - 0.0249 Ml 0. 1 0 - 0.2499
0.025 - 0.0499 •• 0.25 +
•I 0.05 - 0.0749 EH No data
US EPA Region 5
Fafcruwy :007
Figure 6-3.  Observed and  Predicted Proportion of Children with Blood Lead Levels > 10
               pg/dL in Region V by County, 2000 and 2005
                                                      67
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
               LEGEND
      Proportion of PbB Levels >/= I 0 ug/dL
         0 - 0.0099    •• 0.075 - 0.0999
         0.01 - 0.0249  •• 0.10 - 0.2499
         0.025 - 0.0499 •• 0.25 +
      ~1 0.05 - 0.0749  CZ1 No data
                           Massachusetts
           Predicted Proportion of Blood-Lead Levels >/= 10 ug/dL
                                2000
               LEGEND
      Proportion of PbB Levels >/= 10 Ug/dL
         0 - 0.0099    mm 0.075 - 0.0999
         0.01 - 0.0249  •• 0.10 - 0.2499
         0.025 - 0.0499 • 0.25 +
     i  ~J 0.05 - 0.0749  I   I No data
                           Massachusetts
         LEGEND
Proportion of PbB Levels >/= 10 ug/dL
   0 - 0.0099    •• 0.075 - 0.0999
   0.01 - 0.0249  M 0.10 - 0.2499
   0.025 - 0.0499 • 0.25 +
  1 0.05 - 0.0749  I   I No data
                     Massachusetts
         LEGEND
Proportion of PbB Levels >/= 10 ug/dL
   0 - 0.0099    •• 0.075 - 0.0999
   0.01 - 0.0249  Ml 0.10 - 0.2499
   0.02S - 0.0499 •• 0.25 +
 "I 0.05 - 0.0749  I   i No data
                     Massachusetts
Figure 6-4. Observed and Predicted Proportion of Children with Blood Lead Levels > 10 pg/dL in Massachusetts by Census
               Tract, 2000 and 2005
                                                                              68
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality guidelines. It has not been formally
disseminated by EPA.  It does not represent and should not be construed to represent any Agency determination or policy.

-------
6.2    Visualization Tool Development

In addition to generating maps, a software tool was developed to provide a flexible way for users
to quickly view data for particular areas and obtain information that led to the results being
viewed. To do this, Battelle utilized existing technology developed through internal research and
modified this technology to meet the needs of this study.  The software sews together a series of
static maps so that they can be viewed dynamically. This allows users view a movie of changes
in surfaces over time and space.

The software is written in C++. Users interact with the software via a Windows GUI that is
implemented using Microsoft Foundations Classes (MFC).  The 3-dimensional graphics within
the tool were implemented using an OpenGL graphics library.

The software visualizes the observed values and the predicted values  for the response variable of
each model (6 national models, 5 Massachusetts models). The software interpolates the
predicted values spatially within each state using a squared inverse distance algorithm; it
interpolates linearly in time. The predicted values are defined for each county. There are two
visualization modes: 1) a spatial surface moving in time; and 2) a time series. The tool was built
in a flexible way so that it can be easily adapted to accept updated data.

Figures 6-5 and 6-6 are screen shots from the visualization tool. Figure 6-5 provides an example
of a response surface generated by the tool to illustrate predicted blood lead levels across a
geographic area.  In this example the area is the state of Illinois.  Figure 6-6 provides an
example of a method the visualization provides to plot predicted blood lead levels in a given
geographic area over time.
                                            69
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
                                                              Unttticd - RipWi     O 3^S'5t::!
Figure 6-5.    Response Surface of Predicted Geometric Mean Blood-Lead Concentration
               Across the State of Illinois from the Visualization Tool
                                               70
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA, It does not represent and should not be construed
to represent any Agency determination or policy.

-------
Figure 6-6.   Time Series Plot of Observed and Predicted Geometric Mean Blood-Lead
              Concentration in Cook County Illinois from the Visualization Tool
                                              71
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
7.0    DISCUSSION AND FUTURE WORK

The goal of this study was to determine whether tools could be developed to differentiate
geographic areas (counties and Census tracts), based on their predicted risk of containing
children with elevated blood-lead levels. Statistical models .were developed that link CDC's
childhood blood-lead surveillance data to demographic predictor variables available in the 2000
U.S. Census. While earlier chapters of this report focus on the development and performance of
these statistical models, this chapter provides a discussion of the factors that should be
considered when using the models,  and some preliminary ideas for improvement.

7.1    Maior Findings

The results of this study suggest that longitudinal predictive models can be developed at the
county level across the nation bated on the use of quarterly summary information from CDC's
National Surveillance Database, and at the census-tract level within states that have a long
history of universal screening and reporting such as Massachusetts. These models can be used to
describe how risk of childhood lead poisoning changes over time within different regions of the
country,  as well as within small geographic areas within states (e.g. counties) and even smaller
geographic areas within counties (e.g. census tracts).  They can be used to predict the risk of
childhood lead poisoning in counties (or census tracts) with little or no surveillance data, and can
also be used to identify those counties (or census tracts) that are at highest risk at the end of the
period of observation (see Appendix F for a list of the ISO counties across the country at highest
risk predicted by each of the six models, as well as the top 10 counties within each state).

The statistical model chosen (a random effects model with separate intercepts and slopes
estimated within each county or census tract) also allows ranking of geographic areas based on
the rate of decline over time after accounting for the fixed effects variables of the model
(although only among those areas that provided adequate surveillance data). Within the context
of the Broad-Based National Model, these random effects would allow us to identify those
counties that are experiencing a more rapid reduction in risk of childhood lead poisoning over
time (to identify best practices) and those counties that are experiencing a significantly less rapid
decline over time (to identify areas in need of additional attention and resources for combating
lead poisoning), after already accounting for the demographic, programmatic, and environmental
factors included in the multivariate model.

Within the context of the series of Broad-Based National Models, the data suggest that there are
significant differences in the distribution of childhood blood-lead concentrations among the
different regions of the country, and that the manner in which these distributions change over
time and are impacted by seasonality is also regionally specific. The risk of childhood lead
poisoning had a statistically significant downward trend over time in all areas of the country with
the exception of EPA Re^on 1. The Arc-View Maps presented in Appendix G, as well as the
visualization tool can be used to assess whether there are any likely data quality issues within
one or more states within Region 1 that led to this unanticipated result. Our initial investigations
and thoughts on this issue are presented later in Section 7.3 in a section on data issues.
                                           72
 This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
 quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
 to represent any Agency determination or policy.

-------
After accounting for these regional differences, a number of demographic, environmental and
programmatic variables were found to be highly predictive of childhood blood-lead
concentrations among the different response variables modeled within this project. The specific
variables that were found to be predictive within the multivariate models varied based on the
response variable - however, there were certainly some variables that were found to be selected
in multiple models. In addition to various census demographic variables that were identified in
previous risk modeling, efforts (e.g., age of housing, percent single parent families, race/ethnicity,
etc.), we found that air modeling data was selected as a predictor variable in all six multivariate
models, variables constructed from EPA's Safe Drinking Water Information System were
selected in five of six multivariate models, and programmatic funding information from HUD
was selected in all six multivariate models. Variables indicating amount of lead emissions
within counties from EPA's Toxics Release Inventory were only selected in one of the six
models, likely due to the fact that these data are collinear with the air modeling data from EPA's
1999 National Air Toxics Assessment.

Within the context of the High Resolution Model developed using data from the Commonwealth
of Massachusetts, we also found a highly significant downward trend in the risk of childhood
lead  poisoning among the five models developed. Due to a very small number of children
observed at or above 25 ug/dL within Massachusetts over the 2000-2006 period of observation -
we were unable to fit this sixth model.  After accounting for the long term reduction over time
and seasonality using similar methods that were employed in the Broad-Based National Model,
we found that only the demographic and programmatic variables were predictive of the risk of
childhood lead poisoning at the census tract level. Of particular interest were the variables that
described the proportion of housing units within each census tract that were found to be in
compliance and out of compliance with the Massachusetts Standard of Care.  In all five of the
multivariate models, the risk of childhood lead poisoning was significantly reduced as the
proportion of housing units in compliance increased within a census tract.  In addition, for the L
last two models (which predicted proportion of children at or above 10 and 15 ug/dL), the risk of
childhood lead poisoning increased significantly as the proportion housing units out of
compliance increased within a census tract.

7.2     Comparison Between Results and NHANES

Due  to selection bias associated with surveillance data, it is expected that the CDC National
Surveillance dataset as well as the Massachusetts surveillance data may show higher proportions
of elevated blood-lead concentrations than found in the general population. For this reason, the
proportion of children with elevated blood-lead concentrations as well as the distribution of the
potential continuous summary measure derived from the surveillance data were compared with
those reported by the most recent six-years of available CDC National Health and Nutrition
Examination Survey (NHANES). Results of this comparison are presented graphically in Figure
7-1 - suggesting  that there is  a highly significant difference between the NHANES and National
Surveillance Data with respect to the geometric mean and proportion of children observed at or
above 5 ug/dL (with lesser differences observed for the proportion of children observed at or
above 10 and 15  ug/dL). In future work on this project, EPA might consider methods for
                                          73
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
calibrating the Surveillance data to better match the National Distribution of childhood
blood-lead concentrations using methods similar to those employed by Strauss, et. al.  2001.
2 5 •

IT 2.0-
^)
TJ
8 1.5-
ง
CD
g1 1.0-
ฃ
i
Z 0.5-
o.o-

Legend: NHANES
	 National Surveillance



"----.,_
"- --__

' """
0.5-

0.4-
u5"
II
A
•D 0.3 J
^
O
o
ffl 0.2-
3
o
0.
0.1-
o.o-

Legend: NHANES
National Surveillance

^
>
v
•^
s
V
>
->
•^^^
^•*^^
"* - -, ^
"^ -ป ^
      1994   1996   1998   2000   2002  2004   2006

                         time
1994   1996   1998  2000   2002   2004   2006

                  time
0.5 -

0.4-
o"
II
A
13 0.3 -1
O
0
m 0.2-
21
o
D.
0.1 -
o.o-

Legend: ' NHANES
----- National Surveillance




"""••>-.....,

0.5-

0.4
CT
n
A
T3 0.3'
(0
0
o
o
55 0.2-
r
o
a.
0.1 -
o.o-

Legend: NHANES
"' National Surveillance






       1994   1996   1998   2000  2002   2004   2006

                         time
1994   1996   1998   2000   2002   2004  2006

                   time
           Figure 7-1.  Comparison of National Surveillance Data to NHANES Data
                                                74
 This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
 quality guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed
 to represent any Agency determination or policy.

-------
7.3   • Data Issues

The models that were developed as part of this project are based on data sources that have both
strengths and limitations. In this section we consider five potentially limiting aspects of the data
- biases from the geocoding process, biases inherent in the surveillance data, predicting within-
area relationships with ecological models, and use of Census data from 2000.

7.3.1   Biases from Geocoding

The national surveillance data utilized in these analyses were summarized by county.  CDC
based this summarization on county FIPS codes reported by its grantees. This field is quite well
reported in CDC's CBLS database.  The Massachusetts surveillance data was summarized and
analyzed at the Census-tract level, with the geocoding of address data within the Massachusetts
data being 'conducted by MDPH staff.  While we have no reason to suspect lack of data quality
within the Massachusetts surveillance data, our experience is that the process of geocoding can
introduce some subtle biases into surveillance data.  Thus, we offer the following section as a
guide for EPA to consider for future modeling efforts in which state or local surveillance data is
geocoded to the census tract-level:

The geocoding process is highly dependent on the quality of address data recorded by the local
lead poisoning prevention programs with whom the blood-lead information originated. Several
factors could prevent an address from being successfully geocoded, such as:
    •   Erroneous, illegible, or purposefully misleading address information being provided to
       the childhood lead poisoning prevention program,
    •   Address data that contain either a P.O. Box or Rural Route as part of the street address,
       which typically cannot be successfully geocoded, or
    •   Errors in data entry.

While these problems with address data are likely to occur in all programs with a non-trivial
frequency, there may be a systematic bias that programs introduce (albeit unintentionally) when
correcting address data. It is likely that address data errors are identified and corrected with
higher frequency for children who have an elevated blood-lead level and require follow-up.

Given the potential bias introduced through the geocoding process, further research may be
worthwhile to determine whether there are reasonable approaches that could be used to adjust the
models for this bias.

7.3.2   Reporting Limits in Surveillance Data

Other naturally occurring biases in the surveillance data may influence the degree to which
models are representative of the true trends in childhood lead poisoning. For example, within
the context of the Broad-Based National Model, there may be differences between states and
localities in the manner in which childhood blood-lead testing results are reported to CDC. In
Sections 2 and 3, we discussed a screening algorithm that was applied to the surveillance data to
supress county/quarter data combinations for areas that were not conducting universal reporting

                                           75
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
of blood-lead testing results. While this algorithm appears to have successfully removed these
suspect data, there is another potential data anomaly mat we identified following the
development of the multivariate predictive models and visualizing the results through the Maps
provided in Section 6 and Appendix G, and the visualization tool described in Section 6.2.  The
result was discussed earlier - in which a significant decrease in the risk of childhood lead
poisoning in all regions of the country with the exception of EPA Region 1  (which contains the
New England States).  Further inspection of the observed data and the model predictions in
Region 1 reveals that the geometric mean blood-lead concentration observed within the State of
Maine was significantly higher than in most other areas of the country - suggesting that perhaps
the minimum reporting level of blood-lead concentrations for some laboratories within this State
may have been 5 ug/dL over at least part of the observation period. Prior to finalizing the results
of mis project, we suggest careful coordination with CDC to identify any such data anomalies
that might adversely impact the modeling results. If there were reporting limits in Maine (as
well as other areas) in which blood-lead results were not reported below a value of 5 ug/dL - we
could suppress the geometric mean and proportion > 5 ug/dL response variables in these areas
(leaving the proportion > 10,15, and 25 ug/dL response variables intact).

7.3.3   Selection Bias in Surveillance Data

Selection bias is perhaps the most serious bias that is yet left unaccounted for in the models that
have been developed, and may have severe impact on their predictive ability. Surveillance data
are observational by nature, and are not designed to be representative of the general population.
There are many competing forces that influence whether or not a child is screened at an
appropriate age, and recorded in the blood-lead surveillance database. Some have hypothesized
that surveillance data in the urban environment are representative of the affluent (who have
private health insurance) and the  poor (who receive Medicaid or other medical assistance), while
under-representing the working poor (who may have no health insurance, and no mechanism for
receiving appropriate preventive  medical testing). While this  may be true in general, many
outstanding lead poisoning prevention programs are currently extending outreach, education, and
screening services to areas with historically high incidence of childhood lead poisoning. These
programs generally provide assistance to all members  of the community, regardless of
entitlement status. While these services are typically offered in high-risk urban areas with the
infrastructure of a federally funded (CDC  and/or HUD) or state-funded lead poisoning
prevention program, they are typically less available in similar high-risk rural areas without
similar infrastructure. In addition to outreach, education and screening activities, many
childhood lead poisoning prevention programs (or partnering housing agencies) receive funding
from HUD's Office of Lead Hazard Control to conduct environmental investigations and reduce
lead hazards in the residential environment. Many of these activities generate targeted screening
of children living in' deteriorated, older housing - which is also a non-trivial source of selection
bias in the surveillance data.

An important question for EPA to address is how selection bias is  likely to  influence the relative
rankings of counties within a region or census tracts within a more localized area, as well as the
predictive ability of the models themselves.
                                           76
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
7.3.4  Limitations of Ecological Models for Predicting Within-Area Relationships

The models that were developed within this project are ecological models that describe quarterly
distributional summary statistics within geographic areas as a function of predictor variables
assessed within those same geographic areas. It also may be the case that some of these
predictor variables have significant variation within a county (or census tract) - and that this
within area variation is highly predictive of risk of childhood lead poisoning within these
geographic areas. Unfortunately, the data limitations within this study (for both the blood-lead
response variables as well as many of the predictor variables) prohibit us from ascertaining these
important.per^n-bwljrelationghips. This type of relationship can only be established by linking
individual blood-lead concentration data with individual-level environmental, demographic,
and/or programmatic information (which is usually not available).

Within the context of the High Resolution Massachusetts Model - it may be possible to link
individual blood-lead records with the longitudinal housing inspection information to assess the
loss of information associated with going from an individual level model to an area-based
ecological model. This type of assessment could be introduced in later stages of this project.

7.3.5  Use of 2000 Census Data and Other Time Invariant Data as Predictors

One potential criticism of the modeling effort is that we are linking blood-lead surveillance data
collected between 1995 and 2006 to Census data that were collected in 2000. Is the demographic
information collected in 2000 likely to remain unchanged over the course of time? The answer
probably depends on the variable under consideration. For example, age of housing in census
tracts or proportion of housing built prior to 1950 is not likely to change dramatically in census
tracts, unless there is a lot of demolition or new construction occurring.  On the flip side, average
income  is likely to change substantively over time.

Even though the demographic information contained in the 2000 Census is likely to change over
time, the more important question is what effect will that change have on our model predictions.
While the models would likely be improved with the use of more current census data for use as
predictors, we do not believe that the use of older (less current) information will result in poor or
inaccurate prediction. In fact, for the purpose of predicting current or future trends in childhood
lead poisoning, we are more concerned with the age of the surveillance data that is being used as
the response variable in this modeling exercise than with the age of the predictor variables.

Similar arguments can be made for the use of static air modeling data, and averaged information
from EPA's Toxic Release Inventory.

7.4    Model Validation issues

The risk index models developed as part of this project may require validation before being used
by childhood lead poisoning prevention programs throughout the country.  The following four
issues might be considered by EPA as being important to address as part of this validation
exercise:

                                           77
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
       1.  Within counties and/or census tracts that contribute blood-lead information to the
          models, how representative is the screened population of children (on which the
          models are based) of the general population of children?

       2.  Within counties and/or census tracts that do not contribute much information to the
          models (e.g. counties with low screening penetration), how well does the model
          perform at predicting relative risk and blood-lead distributions?

       3.  Can risk index models based on historical blood-lead data from 1995 through 2005
          accurately predict risk and blood-lead distributions in future years (e.g. can it be used
          to forecast towards the federal 2010 goal)?

       4.  Can the High Resolution Model developed in Massachusetts be generalized to predict
          risk and blood-lead distributions in other states across the Nation (or even within EPA
          Region 1)?

If EPA is to provide childhood lead poisoning prevention programs with a risk characterization
tool based on these models, a comprehensive validation should be pursued to address the above
four issues.

Validation of the Surveillance Data
The first issue  is related to the quality of the data supporting model development. For example,
if CDC's surveillance data are biased towards inclusion of high-risk children (as shown in the
comparisons to NHANES), the risk index models will also be biased and tend to over-predict
children at high risk. Note that if the bias is consistent among all counties and census tracts (i.e.,
it over-represents high risk children everywhere), the model predictions for the proportion of
children in each blood-lead category will likely be biased, while the ability for risk indices to
differentiate between high- and low- risk areas will be preserved.  If the biases occur differently
in different areas, non-trivial adjustments to the model would need to be pursued prior to use by
childhood lead poisoning prevention programs.

Since the unit of analysis in the development of the Broad-Based National Model is at the county
level, the goal  of a validation exercise would be to determine whether the distribution of
children's blood-lead concentrations that are included in the surveillance data for a sample of
census tracts are representative of the general population of children found within those census
tracts. One possible approach, would be to develop a field testing validation survey, in which a
stratified random sample of counties are selected for a short-term outreach campaign in which
eligible children are sampled in a representative manner. Stratification variables to be
considered would be Rural/Suburban/Urban, predicted level of risk from the model, and possibly
levels of socio-economic status. Obviously, development of such a survey would be costly,
difficult to implement, and likely beyond the scope of this project. Alternatively, CDC might be
able to reveal the specific counties that participated in various waves of NHANES - with
comparisons being made in those specific counties. Access to the identification of the specific
counties from  which NHANES study subjects were sampled (within the NHANES analysis
dataset) would provide this project with the best foundation to address the serious biases

                                           78
This information  is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
identified in Section 7.2 and calibrate the model to ensure that it is more reflective of the U.S.
population.

Validation of the Models in Areas with Low Screening Penetration
This second issue relates to the performance of the risk index models in predicting both relative
risk and the number of children in different blood-lead categories in the census tracts that
historically had low screening penetration.  Due to the fact that there is little to no data in these
geographic areas to determine the fit of the risk index models, we  would need to conduct some
field studies similar to the one described in the previous section to address this issue. The major
difference between the two field studies is that the census tracts chosen for this validation
exercise would be tracts in which the screening penetration is low.

A similar approach could be used to conduct this field validation exercise in which a stratified
sample of counties would be identified for the study, and a representative sample of children's
blood-lead levels would be obtained within those census tracts using an intense, brief outreach
effort.  The counties again  would be chosen using a stratified random sampling approach, to
obtain a sample of tracts that represents a combination of high-, medium- and low-risk areas in
the rural, suburban, and urban environments. This is an area for potential future collaboration
with CDC and perhaps some of their lead poisoning prevention  grantees.

Validation of the Models in Predicting Future Blood-Lead Concentrations
Validation of this third issue can be performed to a certain extent using data that are already
available as part of the modeling process.  For example, in the national model where data are
available from 1995 through 2005, data for a state or set of states can be removed for one or
more years and the missing data predicted by the model. If all 2005 data were removed, models
would be developed using  the data from 1996 through 2004, and then the "future" predictive
ability of those models can be assessed by applying them to the data from 2005.

Validation of the Models in Predicting Blood-Lead Concentrations in Other Geographic Areas
The last type of validation  involves the determination of synergies (or lack thereof) in prediction
between the Broad-Based National Model and the High Resolution Model. Conceptually, we
should be able to aggregate the modeling predictions from multiple census tracts within a county
from the low-resolution model and match the county-level predictions from the National Model.
Due to the fact that the National Model and Massachusetts Models were developed
independently,  using different data sources for the surveillance data (CDC and MDPH), and
utilizing different predictor variables - these synergies may not  exist.

Further work on integrating the Broad-Based Model with the High Resolution Model (or
multiple high resolution models if EPA is successful at expanding this project to include multiple
additional programs) can be done by fitting these two types of models jointly under the concept
of hierarchical linear modeling. This type of model, while more sophisticated and compute-
intense, can be developed using specialized software under a Monte-Carlo Markov Chain
Baysesian formulation.
                                           79
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
7.5    Other Recommendations for Immediate Future Work

The previous sections within Chapter 7 focus on various important issues related to the
development of models to predict risk of childhood lead poisoning at the geographic level,
including calibration to the nationally representative trends over time observed in NHANES,
assessment of the potential impact of a variety of important biases and other data quality issues,
and various model validation exercises that can be explored. EPA has also been including other
state and local lead poisoning prevention programs as part of the project conference calls in
anticipation of developing additional High Resolution Models as part of follow-up work to this
project. While these are all worthy tasks to pursue as part of future work, there are some
additional analyses that we recommend pursuing on the Broad-Based National Model as well as
the High Resolution Model within Massachusetts prior to approval of this report as a final report.
These activities include the following:

    •   Broad Based National Model
           o The aggregate blood-lead data that serves as the response variable in this report
             should be carefully scrutinized by collaborators at CDC to ensure that the
             information is appropriately handled in these analyses. CDC grantee relationship
             managers may have insight into data quality issues (such as  the previously
             discussed laboratory minimum reporting values,  and not following universal
             reporting guidelines) for specific geographic areas  and periods of time. The maps
             and visualization tool should help foster this review of the data.  In particular, we
             recommend careful scrutiny of the unanticipated upward trend of blood-lead
             concentrations in EPA Region 1. Additionally, the maps of Region 2 suggest that
             data from New York State were not included in the input data set (which was also
             not anticipated).  Once the CDC National Surveillance data  is reviewed and better
             understood by our collaborative investigative team - the Broad-Based National
             Model exploratory analyses and multivariate models can be efficiently re-fit to
             provide more informed analysis results.
           o We also recommend careful scrutiny and potential additional work on the
             environmental and programmatic predictor variables by  members of the
              collaborative investigative team.
           o  In addition to the above data quality issues - we  also recommend further
              investigation into using urban vs. rural status as a potential effect modifier in the
              analyses. Differentiating between urban and rural  areas  can be conducting in
              numerous ways, including:
                  •   Determining whether the county is  part of a Metropolitan Statistical Area
                     within the 2000 US Census,
                  •   Identification of the counties that contain the U.S. top 100 (or 200) cities
                     based on population size,
                  •   Use of a population density score (with a cut-off value).

              Use of this variable as a potential effect modifier might include  fitting separate
              intercepts and slopes for the effects of time and seasonality  within the different
              regions of the country, as well as the potential for using  different environmental,

                                           80
 This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
 quality guidelines.  It has not been formally disseminated by EPA. It does not  represent and should not be construed
 to represent any Agency determination or policy.

-------
              programmatic, and demographic predictor variables in these two area types in the
              multivariate predictive models.
          o  Once the proper way of handling the potential effect modifier for rural vs urban
              areas - the exploratory analyses that assess the predictive ability of each candidate
              environmental, programmatic and demographic variable should be refit in a
              manner consistent with the baseline effects that will be included in the model.
              Thus - rather than assessing the predictive ability of a candidate variable after
              adjusting for the downward trend of time, we should assess it after adjusting for
              region, region*time, region*seasonality, and potentially region*urban/raral.

High Resolution Model in Massachusetts
    •  Due to the fact that we know that Massachusetts followed universal screening and
       reporting guidelines during the entire period of observation (2000-2006), and the fact that
       these data have been used previously to support federally funded research projects - we
       are less concerned about some of the previously mentioned data quality issues. This does
       not mean that the Massachusetts data is not potentially biased or flawed, as there are still
       probable selection biases and potential geocoding biases that were introduced into the
       analysis dataset that supports the High Resolution Model. We invite our collaborators at
       the Massachusetts Department of Public Health to review and comment on this work, and
       add their insight and experience  in making recommendations on additional ways of
       handling the various data sources that were integrated into this model.

    •  We also recommend making comparisons between the observed and predicted data from
       the Broad-Based National Model for counties in Massachusetts (based on the input data
       received from CDC) with the observed and predicted data from the High Resolution
       Model (based on the input data received from MDPH) by aggregating the observed and
       predicted census tract data within Massachusetts to the county level.

    •  Finally, we recommend pursuit of some  additional analyses of the individual-level data
       from MDPH - by linking individual blood-lead testing results on children over time to
       the housing inspection results (as well as other census-tract level predictors that were
       used in the  current High Resolution Model). This will help identify the degree of
       information loss experienced by  pursuit of the ecological models of aggregate summary
       data.
                                           81
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
8.0    REFERENCES

24 CFR Part 35; 40 CFR Part 745, Lead;. Requirements for Disclosure of Known Lead-Based
       Paint and/or Lead-Based Paint Hazards in Housing; Final Rule (3/6/1996). Accessed at
       http://www.leadsafehomes.info/pdfs/all titleten fulltext englisn.pdf#search=%22HUD%
       201018%20Rule%22
40 CFR Part 745, Lead;. Identification of Dangerous Levels of Lead; Final Rule (1/5/2001).
       Accessed at http://www.epa.gov/fedrgstr/EPA!-TOX/2001/Januarv/Dav-05/t84.pdf
Battelle. Draft Quality Management Plan for the Targeting Elevated Blood-lead Levels in
       Children Pilot Study. February 2007.
CDC. 1997. Screening Young Children for Lead Poisoning: Guidance for State and Local Public
       Health Officials, edited by U.S. Department of Health and Human Sevices. Atlanta GA:
       Public Health Sevices, CDC.
HUD. 1995. The Relation of Lead Contaminated House Dust and Blood-Lead Levels Among
       Urban Children. Washington DC: U.S. Department of Housing and Urban Development.
Lanphear, BP, TD Matte, J Rogers, RP Clickner, B Dietz, RL Bornschein, P Succop, KR
       Mahaffey, S Dixon, W Galke, M Rabinowitz, M Farfel, C Rohde, J Schwartz, P Ashley,
       and DE Jacobs. 1998. The contribution of lead-contaminated house dust and residential
       soil to children's blood lead levels. A pooled analysis of 12 epidemiologic studies.
       Environ Res 79 (l):51-68.
Miranda, ML, DC Dolinoy, and MA Overstreet. 2002. Mapping for Prevention: GIS Models for
       Directing Childhood Lead Poisoning Prevention Programs. Environmental Health
       Perspectives 110 (9):947-53.
Miranda, ML, JM Silva, MA Overstreet Galeano, JP Brown, DS Campbell, E Coley, CS
       Cowan, D Harvell, J  Lassiter, JL Parks, and W Sandele. 2005. Building Geographic
       Information System Capacity in Local Health Departments: Lessons from a North
       Carolina Project. Am J Public Health 95 (12):2180-5.
Spivey, Angela. The Weight of Lead: Effects Add Up in Adults. Environmental Health
       Perspectives Volume 115. Number 1. January 2007.
Strauss, Warren, R Carroll, Steve Bortnick, John Menkedick, and B Schultz. 2001. Combining
       Datasets to Predict the Effects of Regulation of Environmental Lead Exposure in Housing
       Stock. Biometrics 57:203-210.
Strauss, Warren, Ramzi Nahhas, Leanna House, Amy Kurokawa, and Bradley Skarpness. 2001.
       Development of Models to Predict Risk of Childhood Lead Poisoning at the Census Tract
       Level. Columbus OH: Technical Report to the U.S. Centers for Disease Control and
       Prevention under Contract No. 200-98-0102.
Strauss, Warren, Tim Pivetz, P Ashley, John Menkedick, E Slone, and S Cameron. 2006.
       Evaluation of Lead Hazard Control Treatments in Four Massachusetts Communities
       through Analysis of Blood-lead Surveillance Data. Environmental Research 99
       (2):214-223.
U.S. Department of Housing and Urban Development. September 15,1999. Final Rule,
       Requirements for Notification, Evaluation and Reduction of Lead-Based Paint Hazards in
       Federally Owned Residential Property and Housing Receiving Federal Assistance.
       Washington DC: Federal Register, 50140-50231.

                                          82
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
           Appendix A



Exploratory Analysis Summary Pages

-------
  
-------
                                    Median Household Income ($)
                                                           8-
                                                        TJ
                                                        W
                                                        o>
                                                         O
                                                        ^
                                                        m
                                                         CO
                                                         0)
                                                        'C 4 •
                                                        ซ

                                                        O
                                                        V
                                                        O
                                                           2-
                                                           
-------
    9

    8 1

    7
                                    Median Per Capita Income ($)
                                                            8
TJ
CO
3
                                                         o
                                                         CD
                                                         C
                                                         i
                                                         v
                                                            6
                                                         I  4
                                                         i
                                                         O
                                                         i
                                                         SI
                                                         Q_
Legend:
 - 1995-2000
-- 2002-2004
                                       2000-2002
                                       2004-2006
      7200  12000 16800 21600 26400 31200 36000 40800

                 Median Per Capita Income ($)
            10000    20000    30000   40000    50000

               Median Per Capita Income ($)
Figure A.3.    Median per Capita Income ($): Histogram and Linear Relationship with Geometric Mean Blood
               Lead Levels by Time
Table A.3a. Summary Information for Median per Capita Income ($) by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
18147.7
17768.0
17807.4
17840.8
17909.6
Standard
&ror
26.3
32.9
31.5
31.0
15.1
Minimum
7069
7069
7069
7069
7069
10th
Percentile
14102
13625
13682
13809
13838
25th
Percentile
15681
15308
15350
15408
15439
Median
17661.5
17218
17184
17218
17346
75th
Percentile
19800
19497
19479
19470
19600
90th
Percentile
22708
22370
22505
22505
22520
Maximum
41051
41051
44962
44962
44962
Table A.3b.    Model Information for the Relationship between Median per Capita Income ($) and Geometric
               Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Median_Per_CapitaJn
Intercept
time
Median_Per_CapitaJn
Intercept
time
Median_Per_Capita_ln
Intercept
time
Median_Per_Capita_ln
Intercept
time
Median_Per_Capita_Jn
Intercept
time
Median_Per_Capita_ln
Estimate
3.406
-0.155
0.000
3.395
-0.155
0.000
-1.500
-0.145
0.000
-3.313
-0.169
0.000
-4.458
-0.161
0.000
-5.935
-0.140
0.000
Standard
Error
0.018
0.003
0.000
0.018
0.003
0.000
0.012
0.002
0.000
0.015
0.003
0.000
0.018
0.003
0.000
0.023
0.005
0.000
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
0.143
<.001
<.001
0.630
-2 Log
Likelihood
153536


196140


250330


269484


300160


360987


Variance Components
Random Effects
an2 = 0.755
a212 = -0.070
0-22* = 0.020
0,^= 0.602
0212 = -0.054
a2/ = 0.016
a,,2 = 0.306
a212= -0.023
022Z= 0.010
ar2= 0.444
a?12= -0.009
0^= 0.008
a,,2 = 0.534
a?12 = -0.020
022^ = 0.010
a1t2= 0.557
a212 = -0.025
0-22* = 0.012
Error
CTe,or2= 0.474
ฐe™2= 38-338




                                                   Page A-3
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                            Percent Units with No Household Earnings
  c 4
  8
  0)
  Q.
                               \
                                                           8-
                                                        T3
                                                        8
                                                         O
                                                        _o
                                                        CD

                                                         8
                                                         o
                                                         0)
                                                        o
                                                        T3
                                                        0)
2-
                                                           o-
                                                               Legend:    	 1995-2000   	2000-2002
                                                                         	2002-2004   	2004-2006
      6.4 112.  16   20.8 25.6 30.4 35.2  40  44.8 49.6

                Percent No Household Earnings
         10
1 ' I ' '
 20
1' i' '
 30
1' i''
 40
50
60
                                                                       Percent No Household Earnings
Figure A.4.    Percent Units with No Household Earnings: Histogram and Linear Relationship with Geometric
              Mean Blood Lead Levels by Time

Table A.4a.    Summary Information for Percent Units with No Household Earnings by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
22.6
22.8
22.8
22.8
22.7
Standard
Error
0.0
0.1
0.0
0.0
0.0
Minimum
6.2
6.2
6.2
6.2
6.2
10th
Percentile
15.2
15.2
15.2
15.3
15.2
25th
Percentile
18.4
18.6
18.6
18.8
18.6
Median
22.4
22.6
22.7
22.8
22.6
75th
Percentile
26.4
26.7
26.4
26.4
26.4
90th
Percentile
30.2
30.5
30.3
30.2
30.3
Maximum
52.2
52.2
52.2
52.2
52.2
Table A.4b.    Model Information for the Relationship between Percent Units with No Household Earnings and
               Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Ffct_HH_No_Earnings
Intercept
time
Ftt_HH_No_Earnings
Intercept
time
Ftt_HH_No_Earnings



Intercept
time
Ffct_HH_No_Earnings
Intercept
time
Pct_HH_No_Earnings
Estimate
3.411
-0.155
2.582
3.398
-0.155
2.788
-1.496
-0.146
2.757



-4.458
-0.160
1.235
-5.929
-0.139
1.099
Standard
Error
0.018
0.003
0.243
0.018
0.003
0.240
0.012
0.002
0.176



0.018
0.003
0.285
0.023
0.005
0.351
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<001
<.001



<.001
<.001
<.001
<.001
<.001
0.002
-2 Log
Likelihood
1 53503


196104


250366





299926


360043


Variance Components
Random Effects
an2= 0.744
a?12 = -0.069
a222 = 0.020
a,,2= 0.595
a?12 = -0.054
a222= 0.016
a, ,z = 0.302
a?12= -0.022
0^= 0.010



a,,^ 0.528
a?12= -0.020
a2/= 0.010
0,^= 0.550
a?12= -0.025
cr22^= 0.012
Error
ฐem,2= 0.474
CTeTOf2= 38.342




                                                  Page A-4
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                              Percent Units with No Household Wage
    5 '
    4
  5.
  o.
    2
    1 •


                                                           8-
                                                        TJ
                                                        CD
o
CD
3
9
r
o
4>
O
1  2
'o
i
                                                           o-
                                                               Legend:    	 1995-2000   	2000-2002
                                                                         	2002-2004   	2004-2006
      7.6 12.4 17.2  22  26.8  31.6  36.4 41.2   46  50.8

                 Percent No Household Wage
            10
• • i''
 20
30
40
50
                                                 60
                                                                         Percent No Household Wage
Figure A.5.   Percent Units with No Household Wage: Histogram and Linear Relationship with Geometric Mean
              Blood Lead Levels by Time

Table A.5a.   Summary Information for Percent Units with No Household Wage by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
25.9
26.2
26.2
26.3
26.1
Standard
Error
0.0
0.1
0.1
0.0
0.0
Minimum
7.6
7.6
7.6
7.6
7.6
10th
Percentile
17.6
17.7
17.8
18.0
17.7
25th
Percentile
21.3
21.5
21.7
21.8
21.5
Median
25.6
26.3
26.3
26.4
26.2
75th
Percentile
30.3
30.6
30.5
30.5
30.5
90th
Percentile
33.6
34.2
34.1
34.0
34.0
Maximum
53.2
53.2
53.2
53.2
53.2
Table A.Sb.   Model Information for the Relationship between Percent Units with No Household Wage and
              Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Pct_HH_No_Wage
Intercept
time
Pct_HH_No_Wage
Intercept
time
Ftt_HH_No_Wage
Intercept
time
Pct_HH_No_Wage
Intercept
time
Fct_HH_No_Wage
Intercept
time
Pct_HH_No_Wage
Estimate
3.404
-0.155
2.767
3.396
-0.155
2.980
-1.499
-0.146
2.671
-3.313
-0.169
1.908
-4.457
-0.160
1.217
-5.927
-0.139
1.018
Standard
Error
0.018
0.003
0.224
0.018
0.003
0.222
0.012
0.002
0.164
0.015
0.003
0.228
0.018
0.003
0.267
0.023
0.005
0.330
p- value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
0.002
-2 Log
Likelihood
153466


196062


250286


269320


299766


359819


Variance Components
Random Effects
0,^= 0.741
cr?12 = -0.070
a222 = 0.020
a,,2 = 0.588
a?12 = -0.054
o222 = 0.016
on2= 0.303
a212= -0.023
a2/= 0.010
a,*= 0.437
a?12= -0.009
a222 = 0.008
a,,2= 0.528
a212= -0.020
a2/= 0.010
an2= 0.550
a212= -0.026
a2/ = 0.01 2
Error
CTe™2= 0.474
aerrorz= 38.340




                                                  Page A-5
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
    10
  c
  I
  <5
  a.
                              Percent Households on Public Assistance
                                                            8-
o
m

I
o
                                                        O

                                                        1  2
                                                        '•o
                                                         o>
                                                            o-
                                                                Legend:   	 1995-2000    	2000-2002
                                                                         	2002-2004    	2004-2006
       0.45  2.25  4.05  5.85 7.65 9.4511.2513.0514.85

             Percent Household on Public Assistance
1 I ' ' ' ' I '
 2     4
1 I ' '
 6
1 i > ' ' ' i '
 8    10
•""P"
 12
                                       14
T1"
 16
T
 18
                                                                    Percent Household on Public Assistance
Figure A.6.    Percent Households on Public Assistance: Histogram and Linear Relationship with Geometric
               Mean Blood Lead Levels by Time

Table A.6a.    Summary Information for Percent Households on Public Assistance by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
3.30
3.39
3.36
3.33
3.34
Standard
Error
0.01
0.01
0.01
0.01
0.01
Minimum
0.5
0.4
0.4
0.5
0.4
10th
Percentile
1.6
1.7
1.7
1.6
1.6
25th
Percentile
2.1
2.2
2.2
2.2
2.2
Median
2.9
3.0
3.0
3.0
3.0
75th
Percentile
4.1
4.1
4.1
4.0
4.1
90th
Percentile
5.5
5.6
5.5
5.4
5.5
Maximum
16.1
16.5
16.5
16.5
16.5
Table A.6b.    Model Information for the Relationship between Percent Households on Public Assistance and
               Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Ffct_HH_Public_Assist
Intercept
time
Pct_HH_Public_Assist
Intercept
time
Rat_HH_Public_Assist
Intercept
time
Rct_HH_Public_Assist
Intercept
time
Pct_HH_Public_Assist
Intercept
time
Pct_HH_Public_Assist
Estimate
3.413
-0.155
5.476
3.397
-0.156
5.443
-1.496
-0.146
5.716
-3.314
-0.169
4.367
-4.460
-0.161
2.554
-5.934
-0.140
2.750
Standard
Error
0.018
0.003
0.824
0.018
0.003
0.828
0.012
0.002
0.609
0.015
0.003
0.824
0.018
0.003
0.965
0.023
0.005
1.189
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
0.008
<.001
<.001
0.021
-2 Log
Likelihood
153566


196189


250628


269718


300259


360701


Variance Components
Random Effects
an2 = 0.768
cx,,2 = -0.070
a222 = 0.020
a,,^ 0.628
a212 = -0.056
a2/= 0.016
a,,2= 0.320
a212= -0.022
a^= 0.010
a,,2= 0.446
a212= -0.009
a2^= 0.008
an2= 0.532
ap12= -0.020
cr222= 0.010
a,,2= 0.552
a212= -0.025
a222= 0.012
Error
CTeror2= 0.474
aerror2= 38.340




                                                   Page A-6
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                              Percent Households below Poverty Line
  I
                                                         o
                                                         CD

                                                         ง
                                                         O
                                                         o
                                                         T3
                                                        I
                                                        Q.
                                                           o
                                                               Legend:        - 1995-2000
                                                                         	2002-2004
	2000-2002
	2004-2006
      2.8  7.6  12.4 17.2  22  26.8 31.6  36.4 41.2  46  50.8

                  Percent Below Poverty Line
1' r '
 10
1' i ''
 20
1' i' '
 30
40
50
TTT
 60
                                                                         Percent Below Poverty Line
Figure A.7.   Percent Households below Poverty Line: Histogram and Linear Relationship with Geometric Mean
              Blood Lead Levels by Time

Table A.7a.   Summary Information for Percent Households below Poverty Line by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
12.95
13.88
13.75
13.59
13.50
Standard
Error
0.04
0.06
0.05
0.05
0.02
Minimum
2.6
2.6
2.6
2.6
2.6
10th
Percentile
6.5
6.7
6.7
6.7
6.6
25th
Percentile
8.6
9.1
9.1
9.0
8.9
Median
11.6
12.5
12.5
12.4
12.2
75th
Percentile
16.2
17.5
17.3
17.0
16.9
90th
Percentile
21.2
22.9
22.6
22.3
22.0
Maximum
50.9
50.9
50.9
50.9
50.9
Table A.7b.    Model Information for the Relationship between Percent Households below Poverty Line and
               Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Pct_LT_Poverty
Intercept
time
Pct_LT_Poverty
Intercept
time
Pct_LT_Ft>verty
Intercept
time
Fbt_LT_Roverty



Intercept
time
Ftt_LT_ Poverty
Estimate
3.410
-0.155
1.421
3.395
-0.156
1.288
-1.501
-0.145
2.015
-3.315
-0.169
0.836



-5.935
-0.140
-0.592
Standard
Error
0.018
0.003
0.228
0.018
0.003
0.225
0.012
0.002
0.164
0.015
0.003
0.225



0.023
0.005
0.327
p- value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001



<.001
<.001
0.070
-2 Log
Likelihood
153574


196202


250550


269713





361060


Variance Components
Random Effects
(^,2 = 0.769
a?12 = -0.070
a222 = 0.020
a,,^ 0.628
a?12 = -0.056
a222 = 0.016
0^2= 0.314
0^2= -0.023
a222 = 0.010
a,* = 0.449
a?12 = -0.008
CT222 = 0.008



a1(2= 0.557
a?12= -0.025
a2/= 0.012
Error
Ge.or2= 0.474
aem)2= 38.339




                                                   Page A-7
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                     Percent Units with Family Income below Poverty Line
                                                           8-
                                                        CO
                                                        0)
                                                        o
                                                        o
                                                        CD
                                                        i
                                                        o
                                                        9
                                                        O
                                                        •
                                                        o
                                                           2-
                                                               Legend:   	 1995-2000   	2000-2002
                                                                        	2002-2004   	2004-2006
       1.6   6.4  11.2  16  20.8 25.6  30.4 35.2  40  44.8
            Percent Family Income Below Poverty Line
  10       20       30       40

Percent Family Income Below Poverty Line
50
Figure A.8.    Percent Units with Family Income below Poverty Line: Histogram and Linear Relationship with
              Percent Units with Geometric Mean Blood Lead Levels by Time

Table A.8a.    Summary Information for Percent Units with Family Income below Poverty Line by Time
Time
Period
1 995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
9.61
10.51
10.40
10.25
10.15
Standard
Error
0.04
0.05
0.04
0.04
0.02
Minimum
1.6
1.6
1.6
1.6
1.6
10th
Percentile
4.4
4.5
4.5
4.5
4.5
25th
Percentile
5.9
6.3
6.4
6.3
6.2
Median
8.2
9.1
9.0
9.0
8.8
75th
Percentile
12.0
13.3
13.2
13.0
12.9
90th
Percentile
16.6
18.3
18.0
17.6
17.6
Maximum
47.4
47.4
47.4
47.4
47.4
Table A.8b.   Model Information for the Relationship between Percent Units with Family Income below Poverty
              Line and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Pct_Family_lncome_LT
Intercept
time
Ftt_FamilyJncomeJ_T
Intercept
time
Ftt_FamilyJncome_LT
Intercept
time
Ftt_FamilyJncome_LT
Intercept
time
Ft t_Family_lnc ome_LT
Intercept
time
Ftt_FamilyJncome_LT
Estimate
3.409
-0.155
1.668
3.394
-0.156
1.526
-1.502
-0.145
2.342
-3.315
-0.169
0.943
-4.458
-0.161
-0.134
-5.935
-0.140
-0.670
Standard
Error
0.018
0.003
0.258
0.018
0.003
0.255
0.012
0.002
0.186
0.015
0.003
0.255
0.018
0.003
0.297
0.023
0.005
0.373
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
0.653
<.001
<.001
0.072
-2 Log
Likelihood
153571


196199


250537


269709


300336


361064


Variance Components
Random Effects
auz = 0.769
a212= -0.070
a222 = 0.020
a,,2 = 0.627
a212= -0.056
a2/= 0.016
a,,2 = 0.313
a212= -0.022
a2/ = 0.010
a,,2= 0.449
0-P12 = -0.008
a222 = 0.008
a,,2= 0.535
a?12 = -0.020
a222= 0.010
0,^= 0.558
a212= -0.025
a22a= 0.012
Error
ฐeror2= 0.474
aetror2= 38.338




                                                  Page A-8
This information is distributed solely for the purpose of p re-dissemination peer review under applicable information quality
guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                     Percent Units Spending Less than Five Years in Poverty
    10
  ~  6
  v
  Q-
                                  T-i-nri-
     0 ' T '   ' 'l' '    i i,i i i  ' V i

      0.075  1.125  2.175  3.225 4.275 5.325  6.375  7.425

               Percent Less than 5 Years in Poverty
                                                        1
                                                        0)
                                                        m
                                                           8-
                                                           6-
•—  4"

I
O

!*i
'1
Legend:
                       1995-2000
                       2002-2004
                                                                                              - 2000-2002
                                                                                                2004-2006
                                                                     Percent Less than 5 Years in Poverty
Figure A.9.    Percent Units Spending Less than Five Years in Poverty: Histogram and Linear Relationship with
              Geometric Mean Blood Lead Levels by Time

Table A.9a.    Summary Information for Percent Units Spending Less than Five Years in Poverty by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
1.49
1.61
1.59
1.56
1.56
Standard
Error
0.01
0.01
0.01
0.01
0.00
Minimum
0.1
0.1
0.1
0.1
0.1
10th
Percentile
0.7
0.7
0.7
0.7
0.7
25th
Percentile
0.9
1.0
1.0
0.9
0.9
Median
1.3
1.4
1.4
1.4
1.4
75th
Per centile
1.9
2.0
2.0
2.0
2.0
90th
Percentile
2.5
2.8
2.7
2.7
2.7
Maximum
7.9
7.9
7.9
7.9
7.9
Table A.9b.   Model Information for the Relationship between Percent Units Spending Less than Five Years in
              Poverty and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
R;t_LE_5Yrs_LT_ Rover
Intercept
time
Pct_LE_5Y rs_LT_Pover
Intercept
time
Ftt_LE_5Yrs_LT_R>ver



Intercept
time
Pct_LE_5Yrs_LT_Rover



Estimate
3.412
-0.155
9.870
3.396
-0.156
8.694
-1.500
-0.145
13.964



-4.459
-0.161
1.641



Standard
Error
0.018
0.003
1.664
0.018
0.003
1.648
0.012
0.002
1.206



0.018
0.003
1.904



p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001



<.001
<.001
0.389



-2 Log
Likelihood
153574


1 96203


250587





300297





Variance Components
Random Effects
a^z= 0.770
a212= -0.070
CT222 = 0.020
CT,,2 = 0.629
a2,2= -0.056
cr222 = 0.016
alt2 = 0.316
a?12 = -0.023
0^= 0.010



a,,2 = 0.534
a212= -0.020
cr222 = 0.010



Error
a 2 = 0.474
error
*.nor2 = 38.339




                                                  Page A-9
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                      Percent American Indian and Alaskan Native Alone
90 "


80

70


60


1 50 '
ง
O

30 '



20
10 '























0






















7.5 15 22.5 30 37.5 45 52.5 60 67.5 75
Percent Amer. Indian and Alaskan Native Alone
8-
"

-3 -
3 :
•o R _
o b .
o
m
c -
8 :
s
0 4:
1_ ^ .
O"
-
•a
o
t> 2:
'•a
o -



Legend: 1995-2000 2000-2002
	 2002-2004 	 2004-2006









	 	
	 	 	 	 	 	
X-"-X.'vฑr;-_.__
' --^-^^^-^^jw-.
— j-* =^v^ 	 — — - ^_
"*" *^*^^
~~" 	 r-~ป.


r""""i 	 i 	 i 	 i 	 I1111 	 i 	 "" i 	 '"i
0 10 20 30 40 50 60 70 8G
I • Al
Figure A.10.   Percent American Indian and Alaskan Native Alone: Histogram and Linear Relationship with
              Geometric Mean Blood Lead Levels by Time

Table A.lOa.   Summary Information for Percent American Indian and Alaskan Native Alone by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
1.36
0.97
0.95
0.91
1.07
Standard
Error
0.03
0.03
0.03
0.03
0.02
Minimum
0.0
0.0
0.0
0.0
0.0
10th
Percentile
0.1
0.1
0.1
0.1
0.1
25th
Percentile
0.2
0.2
0.2
0.1
0.2
Median
0.3
0.3
0.3
0.3
0.3
75th
Percentile
0.6
0.5
0.5
0.5
0.5
90th
Percentile
1.8
1.1
1.1
1.1
1.2
Maximum
79.8
79.8
79.8
79.8
79.8
Table A.lOb.  Model Information for the Relationship between Percent American Indian and Alaskan Native
              Alone and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
PcLAIANAP
Intercept
time
PcLAIANAP



Intercept
time
Pct_AIANAP
Intercept
time
Pct_AIANAP
Intercept
time
Pct_AIANAP
Estimate
3.419
-0.156
-1.000
3.402
-0.156
-1.037



-3.310
-0.170
-1.704
-4.459
-0.162
-1.979
-5.936
-0.141
-1.739
Standard
Error
0.018
0.003
0.341
0.018
0.003
0.371



0.015
0.003
0.368
0.018
0.003
0.470
0.023
0.005
0.650
p- value
<.001
<.001
0.003
<.001
<.001
0.005



<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
0.008
-2 Log
Likelihood
153603


1 96226





269827


300456


361009


Variance Components
Random Effects
CTll2 = 0.777
a?,2 = -0.070
CT2/ = 0.020
0-,,*= 0.635
a2,2 = -0.056
a222 = 0.016



a,,2= 0.448
a212= -0.008
a222 = 0.008
a,,2= 0.529
ap12 = -0.019
CT___,Z= 0.010
a,/= 0.553
a212= -0.025
a222= 0.012
Error
^error2 = ฐ.474
aerto,2= 38.338




                                                Page A-10
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines.  It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                           Percent Asian Alone
    80


    70


    60


    50
  ฃ40
  a.
30


20


10
_o
m
c

•C  4

i
O
                                                            2:

                                                            oH
                                                            Legend:        - 1995-2000
                                                                      	2002-2004
                                        2000-2002
                                        2004-2006
   0.4  5.2   10  14.8  19.6 24.4  29.2  34  38.8  43.6

                  Percent Asian Alone
              10       20       30

                    Percent Asian Alone
                                                                                                  40
50
Figure A.ll.   Percent Asian Alone: Histogram and Linear Relationship with Geometric Mean Blood Lead Levels
               by Time

Table A.I la.   Summary Information for Percent Asian Alone by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
0.91
0.87
0.89
0.84
0.88
Standard
Error
0.01
0.02
0.02
0.01
0.01
Minimum
0.0
0.0
0.0
0.0
0.0
10th
Percentile
0.1
0.1
0.1
0.1
0.1
25th
Percentile
0.2
0.2
0.2
0.2
0.2
Median
0.4
0.4
0.4
0.4
0.4
75th
Percentile
0.9
0.8
0.8
0.8
0.8
90th
Percentile
2.1
1.9
2.0
1.9
2.0
Maximum
45.4
45.4
45.4
45.4
45.4
Table A.lib.   Model Information for the Relationship between Percent Asian Alone and Geometric Mean Blood
               Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Ftt_Asian
Intercept
time
Ftet_Asian
Intercept
time
Ftt_Asian
Intercept
time
Ffct_Asian
Intercept
time
Ftt_Asian
Intercept
time
Ffct_Asian
Estimate
3.417
-0.155
-2.754
3.403
-0.156
-4.554
-1.490
-0.146
-4.312
-3.310
-0.169
-0.849
-4.460
-0.161
0.832
-5.936
-0.141
0.818
Standard
Error
0.018
0.003
0.710
0.018
0.003
0.667
0.012
0.002
0.519
0.015
0.003
0.675
0.018
0.003
0.733
0.023
0.005
0.820
p-value
<.001
<001
<.001
<.001
<.001
<001
<.001
<.001
<.001
<.001
<001
0.209
<.001
<.001
0.257
<.001
<001
0.319
-2 Log
Likelihood
153595


196186


250585


269722


300406


361015


Variance Components
Random Effects
CT11ii= 0.775
a212= -0.070
0^ = 0.020
a, ,2= 0.622
a212= -0.055
cr22^= 0.016
0^2 = 0.326
a212= -0.023
cr2/= 0.010
a,,2= 0.453
a212= -0.008
CT22^ = 0.008
a^2= 0.535
a212= -0.020
a2/= 0.010
(^,2= 0.556
a212= -0.025
cr22a= 0.012
Error
aerror-= 0.474
aerro,2= 38.342




                                                   Page A-11
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                          Percent Black Alone
  
'•5
u.
o:


Legend: 1 995-2000 2000-2002
	 2002-2004 	 2004-2006



	 ' 	 ' 	 	
	 	 „___,
	 	 -•



0 10 20 30 40 50 60 70 80 90
                      Percent Black Alone
                                                                            Percent Black Alone
Figure A.12.
Percent Black Alone: Histogram and Linear Relationship with Geometric Mean Blood Lead Levels
by Time
Table A.12a.   Summary Information for Percent Black Alone by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Nu m be r
Missing
0
0
0
0
0
Mean
9.08
9.94
9.97
9.97
9.70
Standard
Error
0.10
0.12
0.12
0.12
0.06
Minimum
0.0
0.0
0.0
0.0
0.0
10th
Percentile
0.1
0.1
0.1
0.1
0.1
25th
Percentile
0.3
0.4
0.5
0.5
0.4
Median
2.1
2.8
2.8
2.8
2.5
75th
Percentile
11.1
13.0
13.1
13.1
12.4
90th
Percentile
30.3
31.8
32.2
32.5
31.6
Maximum
84.2
86.1
86.1
86.1
86.1
Table A.12b.   Model Information for the Relationship between Percent Black Alone and Geometric Mean Blood
               Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Ftt_Black
Intercept
time
Ftt_Black
Intercept
time
Ffct_Black
Intercept
time
Ftt_Black
Intercept
time
FfcLBIack
Intercept
time
Ftt_Black
Estimate
3.419
-0.155
0.271
3.401
-0.156
0.209
-1.491
-0.146
0.777
-3.312
-0.170
0.353
-4.459
-0.161
-0.036
-5.933
-0.140
-0.339
Standard
Error
0.018
0.003
0.102
0.018
0.003
0.101
0.012
0.002
0.074
0.015
0.003
0.100
0.018
0.003
0.115
0.023
0.005
0.144
p- value
<.001
<.001
0.008
<.001
<.001
0.038
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
0.754
<.001
<.001
0.019
-2 Log
Likelihood
153607


196232


250677


269832


300307


360634


Variance Components
Random Effects
an2 = 0.776
a?12 = -0.070
a222 = 0.020
a, ,z= 0.634
a212 = -0.056
a222 = 0.016
0^2= 0.315
a212= -0.021
a22^= 0.010
a,,2= 0.450
apl2= -0.008
a^2 = 0.008
an2= 0.535
a212= -0.020
o222= 0.010
a,,2= 0.557
a212= -0.026
CT222= 0.012
Error
CTeTOr2 = ^.474
aerror2= 38.338




                                                  Page A-12
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                           Percent White Alone
    15.0
    12.5
                                                             8
                                                          E  6
                                                          o
                                                          ffl
                                                          c
                                                          8
                                                          2
                                                          'C  4
                                                          B
                                                          o
                                                          $
                                                          -a
                                                          a
        2.25 14.25 26.25 38.25 50.25 62.25 74.25 86.25 98.25
                                                             2-
                                                             OH
                                                                 Legend:    	 1995-2000
                                                                           	2002-2004
                                                                             	2000-2002
                                                                             	2004-2006
                                                                   10  20   30
                                                                 1 ' I ' ' ' ' I '
                                                                  40   50
' ' I ' '
 60
70
1 I ' ' ' ' I ' ' "T
80  90  100
                       Percent White Alone
                                                                              Percent White Alone
Figure A.13.
Percent White Alone: Histogram and Linear Relationship with Geometric Mean Blood Lead Levels
by Time
Table A.13a.   Summary Information for Percent White Alone by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
83.40
80.96
81.12
81.56
81.88
Standard
Error
0.12
0.16
0.15
0.15
0.07
Minimum
1.6
1.6
1.6
1.6
1.6
10th
Percentile
57.3
53.2
54.2
54.9
55.2
25th
Percentile
75.7
70.6
70.8
71.5
72.3
Median
90.5
88.7
88.6
89.1
89.3
75th
Percentile
96.3
95.8
95.8
96.0
96.1
90th
Percentile
97.8
97.6
97.6
97.7
97.7
Maximum
99.8
99.8
99.8
99.8
99.8
Table A.13b.   Model Information for the Relationship between Percent White Alone and Geometric Mean Blood
               Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
FtM/Vhite
Intercept
time
Ftt_White
Intercept
time
Ftt_White
Intercept
time
Ftt_White
Intercept
time
PcM/Vhite
Intercept
time
Ftt_White
Estimate
3.418
-0.155
-0.007
3.403
-0.156
0.186
-1.493
-0.146
-0.214
-3.311
-0.169
-0.033
-4.457
-0.161
0.146
-5.928
-0.139
0.418
Standard
Error
0.018
0.003
0.080
0.018
0.003
0.079
0.012
0.002
0.059
0.015
0.003
0.078
0.018
0.003
0.090
0.023
0.005
0.112
p-value
<.001
<.001
0.931
<.001
<.001
0.018
<.001
<001
<.001
<.001
<.001
0.673
<.001
<.001
0.106
<.001
<.001
<.001
-2 Log
Likelihood
153615


196231


250732


269777


300253


360434


Variance Components
Random Effects
a, ,2 = 0.779
a?12 = -0.070
a,/ = 0.020
a,,2 = 0.635

-------
                  Percent Native Hawaiian and Other Pacific Islander Alone
100




80



60
c
0)
o
l_
Q.

40



20

O-1-




n
















V
\
T "" ^ 1 	 1 	 1 	 1 	 1 	 1 	 1 	 1 	
D 1.2 2.4 3.6 4.8 6 7.2 8.4 9.6
Pct_NHOPI
8-
.
:
(0
a>
_i
o :
CD
C -
(0
0)
s

•s 4;
s •
E '
o ;
o :
•a ;
+1* O -
o
'•5
ซ
a.
0:

c

Legend: 1995-2000 2000-2002
	 2002-2004 	 2004-2006









•^^
^^\^^
^~^^^^
:^c^--- ^^^^
— ~L~~~~~;^~'---^^\^
— ~~-"~^~ ^~-~-T""~^>^^
~~ "^•^i^^Cr-
^5s^^Tr->-
^^\T

01 23456789 10 11
>ar/*ont Klatn/o Mauraiion onrl Othor Dar*!f!r* Iclonrlar Alnna
Figure A.14.   Percent Native Hawaiian and Other Pacific Islander Alone: Histogram and Linear Relationship
              with Geometric Mean Blood Lead Levels by Time

Table A.14a.   Summary Information for Percent Native Hawaiian and Other Pacific Islander Alone by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
0.04
0.05
0.05
0.03
0.04
Standard
Error
0.00
0.00
0.00
0.00
0.00
Minimum
0.0
0.0
0.0
0.0
0.0
10th
Percentile
0.0
0.0
0.0
0.0
0.0
25th
Percentile
0.0
0.0
0.0
0.0
0.0
Median
0.0
0.0
0.0
0.0
0.0
75th
Percentile
0.0
0.0
0.0
0.0
0.0
90th
Percentile
0.1
0.1
0.1
0.1
0.1
Maximum
10.3
10.3
10.3
10.3
10.3
Table A.14b.   Model Information for the Relationship between
              Islander Alone and Geometric Mean Blood Lead
Percent Native Hawaiian and Other Pacific
Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Ftt_NHOR
Intercept
time
FbtJXIHOR
Intercept
time
FctJMHOR
Intercept
time
Ftt_NHOR
Intercept
time
PCUNHOR
Intercept
time
Ftt_NHOR
Estimate
3.419
-0.155
-17.741
3.403
-0.156
-19.648
-1.490
-0.146
-10.777
-3.311
-0.169
-5.699
-4.458
-0.161
-2.003
-5.934
-0.140
-4.083
Standard
Error
0.018
0.003
4.018
0.018
0.003
3.890
0.012
0.002
3.045
0.015
0.003
3.959
0.018
0.003
4.350
0.023
0.005
5.131
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
0.150
<.001
<.001
0.645
<.001
<.001
0.426
-2 Log
Likelihood
153588


1 96204


250701


269751


300303


360767


Variance Components
Random Effects
a,,2= 0.773
a212= -0.070
a22* = 0.020
an2= 0.629
a?12= -0.056
a222 = 0.016
a,,2= 0.332
a212= -0.023
0^= 0.010
cr11z= 0.453
a?12= -0.008
cr222 = 0.008
a, ,2= 0.535
a212= -0.020
a222= 0.010
an2= 0.556
a?12= -0.025
CT222 = 0.012
Error
ฐ.™r2= ฐ-474
ป••- 38-339




                                                Page A-14
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                      Percent Other Race Alone
50 "
0




40





^ 30
8
0.

20




10


0.
















"

r
1 \




32




















t



0.26 0.5 0.74 0.98 1.22 1.46 1.7 1.94
Percent Other Race Alone
8-
•
;
CO
CD
— 1
TJ fi .
0 b .
0
m
c -
CO
CD
S.
0 4:
t :
E
o
o
C5
•a
S 2:
1 :
a :
o:



Legend: 1995-2000 ~ " 2000-2002
	 2002-2004 	 2004-2006










^•^^
^^^^^
^ป*-i-ซ- ^^^-N.

— - - — ^T"-"^^"*--ป ~~ ^^~"
~ ;"fc- x~ 	 C^ — — ^ ^
"^^x — — -_n^ — - 	 	



0123

Figure A.15.  Percent Other Race Alone: Histogram and Linear Relationship with Geometric Mean Blood Lead
              Levels by Time
Table A.15a. Summary Information for Percent Other Race Alone by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
0.07
0.06
0.07
0.07
0.07
Standard
Error
0.00
0.00
0.00
0.00
0.00
Minimum
0.0
0.0
0.0
0.0
0.0
10th
Percentile
0.0
0.0
0.0
0.0
0.0
25th
Percentile
0.0
0.0
0.0
0.0
0.0
Median
0.0
0.0
0.0
0.0
0.0
75th
Percentile
0.1
0.1
0.1
0.1
0.1
90th
Percentile
0.2
0.2
0.2
0.2
0.2
Maximum
1.2
1.2
2.1
2.1
2.1
Table A.lSb.  Model Information for the Relationship between Percent Other Race Alone and Geometric Mean
              Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Ftt_Other_Race
Intercept
time
Ftt_Other_Race



Intercept
time
Ftt_Other_Race
Intercept
time
Ffct_Other_Race
Intercept
time
Pct_Other_Race
Estimate
3.418
-0.155
-33.708
3.403
-0.156
-47.574



-3.31 1
-0.169
-11.116
-4.459
-0.161
-0.950
-5.935
-0.140
3.260
Standard
Error
0.018
0.003
13.077
0.018
0.003
12.623



0.015
0.003
12.849
0.018
0.003
14.400
0.023
0.005
16.661
p-value
<.001
<.001
0.010
<.001
<.001
<.001



<.001
<.001
0.387
<.001
<.001
0.947
<.001
<.001
0.845
-2 Log
Likelihood
153598


196212





269736


300303


360834


Variance Components
Random Effects
a,,2= 0.778
a?12= -0.070
a222= 0.020
at12= 0.632
a212= -0.056
a22^= 0.016



0,^= 0.453
a212= -0.008
a2/= 0.008
a,,2= 0.535
a?12= -0.020
a2/= 0.010
a,,2= 0.557
^,2= -0.025
cr22"= 0.012
&ror
CTe™2= 0.474
aeno2= 38.339




                                                  Page A-15
This information is distributed solely for the purpose ofpre- dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
    35
    30
    25
  o>
  o
  
                                                          TJ
                                                          a>
                                                                               - 1995-2000
                                                                           	2002-2004
2000-2002
2004-2006
                                                                              10
                                                                        20
         30
                     Percent Multiple Races
                                                                             Percent Multiple Races
Figure A.16.   Percent Multiple Races: Histogram and Linear Relationship with Geometric Mean Blood Lead
               Levels by Time

Table A.16a.   Summary Information for Percent Multiple Races by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
1.21
1.13
1.14
1.12
1.15
Standard
&ror
0.01
0.01
0.01
0.01
0.00
Minimum
0.0
0.0
0.0
0.0
0.0
10th
Percentile
0.4
0.4
0.4
0.4
0.4
25th
Percentile
0.6
0.6
0.6
0.6
0.6
Median
0.9
0.9
0.9
0.9
0.9
75th
Percentile
1.4
1.3
1.3
1.3
1.3
90th
Percentile
2.0
1.9
1.9
1.9
1.9
Maximum
24.0
24.0
24.0
24.0
24.0
Table A.16b.   Model Information for the Relationship between Percent Multiple Races and Geometric Mean
               Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Pct_Multi_Race
Intercept
time
Pct_Multi_Race
Intercept
time
Ftt_Multi_Race
Intercept
time
Rct_Multi_Race
Intercept
time
Rct_Multi_Race
Intercept
time
Rct_Multi_Race
Estimate
3.418
-0.155
-7.044
3.402
-0.156
-8.691
-1.491
-0.146
-5.276
-3.311
-0.169
-2.957
-4.458
-0.161
-1.331
-5.934
-0.140
-0.244
Standard
Error
0.018
0.003
1.248
0.018
0.003
1.228
0.012
0.002
0.938
0.015
0.003
1.244
0.018
0.003
1.418
0.023
0.005
1.701
p-value
<.001
<001
<.001
<.001
<.001
<001
<.001
<.001
<.001
<.001
<.001
0.017
<.001
<001
0.348
<.001
<001
0.886
-2 Log
Likelihood
153578


196182


250675


269740


300290


360789


Variance Components
Random Effects
0-,^= 0.771
a212= -0.070
o^ = 0.020
a,,2= 0.622
a212= -0.055
a22z = 0.016
o-,,2 = 0.329
a212= -0.022
0-^= 0.010
a,,2= 0.452
a?12= -0.008
o22a = 0.008
CT1,2= 0.535
a^2= -0.020
a22* = 0.010
0-,,* = 0.557
a212 = -0.025
a22^ = 0.012
Error
CTeTOr2= 0.474

-------
                                            Percent Hispanic
40

35
30

25
o> ,_
520-
Q_
15
10

5 ' '
O'^TT
0











lk\_
9 18 27 36 45 54 63 72 81 90
8-
•
•o
TO
_1

m
8 :
3 4:
4)
E •
o

O
5 2;
T3
ฃ
o.
o-


Legend: 1995-2000 2000-2002
	 2002-2004 	 2004-2006




	 	 	


~^ — 	 —


0 10 20 30 40 50 60 70 80 90 10
                       Percent His panic
                                                                              Percent Hispanic
Figure A.17.   Percent Hispanic: Histogram and Linear Relationship with Geometric Mean Blood Lead Levels by
               Time
Table A.17a.   Summary Information for Percent Hispanic by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
3.94
6.02
5.82
5.49
5.22
Standard
Error
0.06
0.10
0.09
0.09
0.04
Minimum
0.0
0.0
0.0
0.0
0.0
10th
Percentile
0.5
0.5
0.5
0.5
0.5
25th
Percentile
0.8
0.8
0.8
0.8
0.8
Median
1.6
1.7
1.7
1.6
1.6
75th
Percentile
3.7
4.7
4.6
4.2
4.2
90th
Percentile
7.8
14.1
14.1
12.7
11.5
Maximum
98.1
98.1
98.1
98.1
98.1
Table A.17b.   Model Information for the Relationship between Percent Hispanic and Geometric Mean Blood
               Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Pct_His panic
Intercept
time
Ffct_Hispanic



Intercept
time
Ftt_His panic
Intercept
time
Fct_His panic
Intercept
time
Ftt_His panic
Estimate
3.419
-0.155
-0.078
3.405
-0.156
-0.383



-3.309
-0.169
-0.183
-4.458
-0.161
-0.122
-5.932
-0.140
-0.418
Standard
Error
0.018
0.003
0.123
0.018
0.003
0.122



0.015
0.003
0.122
0.018
0.003
0.141
0.023
0.005
0.181
p-value
<.001
<.001
0.527
<.001
<.001
0.002



<.001
<.001
0.133
<.001
<.001
0.387
<.001
<.001
0.021
-2 Log
Likelihood
153614


196226





269762


300307


360745


Variance Components
Random Effects
a^z= 0.778
a212= -0.070
a2/ = 0.020
a^2 = 0.631
a212 = -0.055
CT222 = 0.016



a,,2= 0.453
a212 = -0.008
CT2/ = 0.008
a^2 = 0.535
a?12= -0.020
CT2/= 0.010
a^2 = 0.555
a212= -0.026
o22^= 0.012
Error
CTe.or2= 0.474
ฐ.nor2= 38-340




                                                  Page A-17
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                         Percent Rented Units
10 '



8





ซ- 6 •

I
<5
Q.
4-



2

n





"



/
/
1
\

I

1
1

/
I
a


-




V

\
-1 \
1 \
\
\
\
i-i \
Jl\
A
*

f






















8:
•
•D
CO
-1
T3 ft '.
O ฐ .
0
CD
C -
5 :
5
O ซ "
ฃ
Q)
E ;
o
o
o :
T> ;
1 * -
CL -
o-

Legend: 1995-2000 " ~ 2000-2002
	 2002-2004 	 2004-2006










	 _^ 	 	 	



= 	 	 	 	
— . 	 	 	 •- 	




10.8 18 25.2 32.4 39.6 46.8 54 61.2 68.4 75.6 10 20 30 40 50 60 70 80 90
Percent Rented Percent Rented
Figure A.18.  Percent Rented Units: Histogram and Linear Relationship with Geometric Mean Blood Lead
              Levels by Time

Table A.18a.  Summary Information for Percent Rented Units by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
25.95
25.79
25.80
25.64
25.80
Standard
Error
0.05
0.06
0.06
0.06
0.03
Minimum
10.5
10.5
10.5
10.5
10.5
10th
Percentile
18.5
18.1
18.0
17.9
18.1
25th
Percentile
21.0
20.9
20.8
20.6
20.8
Median
24.7
24.5
24.4
24.2
24.4
75th
Percentile
29.1
29.0
29.1
28.6
28.9
90th
Percentile
35.5
35.3
35.8
35.4
35.5
Maximum
69.4
69.4
69.4
80.5
80.5
Table A.18b.  Model Information for the Relationship between Percent Rented Units and Geometric Mean Blood
              Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Bct_Rented
Intercept
time
R3t_Rented
Intercept
time
R=t_Rented
Intercept
time
Ffct_Rented
Intercept
time
Pct_Rented
Intercept
time
Pct_Rented
Estimate
3.418
-0.155
0.275
3.402
-0.156
-0.095
-1.491
-0.146
0.073
-3.314
-0.169
0.967
-4.466
-0.162
1.298
-5.947
-0.142
1.088
Standard
Error
0.018
0.003
0.197
0.018
0.003
0.192
0.013
0.002
0.147
0.015
0.003
0.194
0.018
0.003
0.219
0.023
0.005
0.262
p- value
<.001
<.001
0.164
<.001
<.001
0.619
<001
<.001
0.618
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
153611


1 96235


250729


269994


301038


362209


Variance Components
Random Effects
CTl12= 0.777
a212= -0.070
a2/= 0.020
0^2= 0.636
cr212= -0.056
o2/= 0.016
0^2= 0.333
ap12= -0.022
a2/= 0.010
an2= 0.448
a212= -0.008
a^2 = 0.008
a112= 0.524
a?12= -0.019
P222= 0.010
a,,2= 0.548
a?12= -0.023
a222= 0.012
Error
ฐซma= ฐ-474
ฐen,r2= 38.339




                                                  Page A-18
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
  c 4
  d>
  o
                                              Median Rent ($)
                                                              8-
                                         cCD	CL_
       11	r I I I I I r	II l-^^-^-^l .1 .-^ I I	II      -.- —

      127.5  262.5 397.5  532.5  667.5  802.5 937.5 1072.5

                        Median Rent ($)
                                                           •o
                                                           CO
                                                           0)
                                                           o
                                                           CD

                                                           
                                                           t;  2
                                                              o-
                                                                  Legend:
           1995-2000
2000-2002
                                                                            	2002-2004    	2004-2006
200     400     600    800

          Median Rent ($)
1000
        1200
Figure A.19.   Median Rent ($): Histogram and Linear Relationship with Geometric Mean Blood Lead Levels by
               Time
Table A.19a. Summary Information for Median Rent ($) by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
361.62
355.61
357.20
356.13
357.93
Standard
Error
0.83
1.00
0.97
0.96
0.47
Minimum
125.0
125.0
125.0
125.0
125.0
10th
Percentile
235.0
230.0
230.0
231.0
231.0
25th
Percentile
278.0
271.0
271.0
272.0
273.0
Median
340.0
331.0
331.0
331.0
334.0
75th
Percentile
418.0
410.0
411.0
409.0
412.0
90th
Percentile
521.0
518.0
520.0
518.0
520.0
Maximum
1114.0
1114.0
1114.0
1114.0
1114.0
Table A.19b.  Model Information for the Relationship between Median Rent ($) and Geometric Mean Blood Lead
               Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Median_Rent
Intercept
time
Median_Rent
Intercept
time
Median_Rent
Intercept
time
Median_Rent
Intercept
time
Median_Rent
Intercept
time
Median_Rent
Estimate
3.406
-0.155
-0.002
3.402
-0.155
-0.002
-1.497
-0.145
-0.002
-3.310
-0.169
-0.001
-4.456
-0.160
0.000
-5.932
-0.140
0.000
Standard
&ror
0.018
0.003
0.000
0.017
0.003
0.000
0.012
0.002
0.000
0.015
0.003
0.000
0.018
0.003
0.000
0.023
0.005
0.000
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
0.002
<.001
<.001
0.414
-2 Log
Likelihood
153451


196013


250037


269220


299864
.

360483


Variance Components
Random Effects
a,,2= 0.735
a212 = -0.070
02Z* = 0.020
CT,,2 = 0.567
a212 = -0.052
a222 = 0.016
0^2 = 0.290
a?12= -0.024
a2/= 0.010
0^2 = 0.437
afz = -0.009
CT2/ = 0.008
CT,,2 = 0.532
a212= -0.020

-------
    25
    20
  ^ 15
  V
  a.
    10
                                            Housing Value ($)
                                                            8-
                                                         (0
                                                         0)
                                                         o
                                                         o
                                                         CD
                                                         (0
•=  4

o
O

f  2
'•5
k_
Qt
                                                               Legend:        - 1995-2000   	2000-2002
                                                                         	2002-2004   	2004-2006
      22500 172500 322500 472500 622500 772500 922500

                       Housing Value ($)
                 500000        1000000

                    Housing Value ($)
1500000
Figure A.20.   Housing Value ($): Histogram and Linear Relationship with Geometric Mean Blood Lead Levels by
               Time

Table A.20a.   Summary Information for Housing Value ($) by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
87234.72
84016.18
85413.63
85399.46
85659.42
Standard
Error
246.36
294.15
316.35
336.26
148.34
Minimum
22500.0
22500.0
22500.0
22500.0
22500.0
10th
Percentile
54300.0
49200.0
49400.0
50000.0
50900.0
25th
Percentile
64500.0
60000.0
60400.0
60900.0
61700.0
Median
81400.0
78000.0
78000.0
78000.0
79000.0
75th
Percentile
100200.0
97400.0
98100.0
97700.0
98600.0
90th
Percentile
126400.0
124300.0
127300.0
126400.0
126200.0
Maximum
469200.0
469200.0
577500.0
1000001.0
1000001.0
Table A.20b.   Model Information for the Relationship between Housing Value ($) and Geometric Mean Blood
               Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Housing_Value
Intercept
time
Housing_Value
Intercept
time
Housing_Value
Intercept
time
Housing_Value
Intercept
time
Housing_Value



Estimate
3.415
-0.155
0.000
3.404
-0.155
0.000
-1.491
-0.145
0.000
-3.308
-0.169
0.000
-4.456
-0.161
0.000



Standard
Error
0.018
0.003
0.000
0.018
0.003
0.000
0.012
0.002
0.000
0.015
0.003
0.000
0.018
0.003
0.000



p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001



-2 Log
Likelihood
153528


196103


250298


269449


300049





Variance Components
Random Effects
a^2= 0.750
a?12 = -0.070
CT22^ = 0.020
0,^= 0.588
a?12= -0.053
cr222 = 0.016
CT,,2 = 0.298
a?12 = -0.022
CT2/ = 0.010
a,,2= 0.440
cr212= -0.008
a222 = 0.008
0,^= 0.532
cr212= -0.020
a2/= 0.010



Error
CTeTOf2= ฐ-474
ฐe™2= 38.334




                                                   Page A-20
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                         Percent Vacant Units
    12
    10
  c
  
-------
                                  Percent Single Parent Households
       10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61

                    Percent Single Parent
8;


•o
CD
0)
O 6
O
m
c '•
s
~ 4 •
ฃ
o
0)
/R
c>
•O
4)
t> 2:
i_
a
o-

1

Legend: 1995-2000 2000-2002
	 2002-2004 	 2004-2006








_^ 	 	 ' 	 	 ~
^-~^* 	
_. 	 	 	 • -






0 20 30 40 50 60 70
                                                                           Percent Single Parent
Figure A.22.  Percent Single Parent Households: Histogram and Linear Relationship with Geometric Mean
              Blood Lead Levels by Time

Table A.22a.  Summary Information for Percent Single Parent Households by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
25.82
25.72
25.59
25.55
25.68
Standard
Error
0.05
0.06
0.06
0.05
0.03
Minimum
10.0
10.0
10.0
10.0
10.0
10th
Percentile
18.5
18.3
18.2
18.2
18.3
25th
Percentile
21.1
21.0
21.0
20.9
21.0
Median
24.7
24.6
24.5
24.4
24.6
75th
Percentile
29.0
28.8
28.8
28.8
28.9
90th
Percentile
34.3
34.4
34.3
34.2
34.3
Maximum
62.4
60.7
60.7
62.4
62.4
Table A.22b.  Model Information for the Relationship between Percent Single Parent Households and Geometric
              Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Ftt_Single_Rarent
Intercept
time
Ftt_Single_Fฃrent
Intercept
time
Ftt_Single_Rarent
Intercept
time
Ftt_Single_Rarent
Intercept
time
Ffct_Single_Fkrent



Estimate
3.419
-0.155
0.588
3.401
-0.156
0.548
-1.491
-0.146
1.601
-3.313
-0.169
1.315
-4.461
-0.161
0.889



Standard
Error
0.018
0.003
0.213
0.018
0.003
0.212
0.012
0.002
0.155
0.015
0.003
0.209
0.018
0.003
0.242



p- value
<.001
<.001
0.006
<.001
<.001
0.010
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001



-2 Log
Likelihood
1 53605


1 96228


250699


269936


300559





Variance Components
Random Effects
o-,,2= 0.773
a212 = -0.069
a222 = 0.020
a, ,2= 0.632
a212= -0.055
cr2,/ = 0.016
0^2 = 0.315
a2,2 = -0.022
CT2/= 0.010
a,* = 0.442
a2]2= -0.008
a222 = 0.008
a,,2= 0.528
a212= -0.019
a2/= 0.010



Error
CTeror2= 0.474
aerror2= 38.340




                                                 Page A-22
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                            Median Year Built
                                                             8
                                                          •a
                                                          s
                                                          CD
                                                          c
                                                          9
                                                          a)
                                                          •—  4
                                                          o
                                                          a>
                                                          O
                                                          •a
                                                          S  2
                                                          '•5
                                                          a>
                                                             0-
                                                                Legend:        - 1995-2000
                                                                          	2002-2004
                                                                                            2000-2002
                                                                                            2004-2006
1939 1945 1951 1957 1963 1969 1975 1981  1987 1993

                    Year Built
                                                             1930   1940   1950  1960  1970   1980   1990  2000

                                                                                 Year Built
Figure A.23.   Median Year Built: Histogram and Linear Relationship with Geometric Mean Blood Lead Levels
               by Time
Table A.23a. Summary Information for Median Year Built by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
1968.05
1969.07
1969.15
1969.01
1968.77
Standard
Error
0.07
0.08
0.08
0.08
0.04
Minimum
1939.0
1939.0
1939.0
1 939.0
1 939.0
10th
Percentile
1953.0
1954.0
1954.0
1954.0
1954.0
25th
Percentile
1961.0
1962.0
1 963.0
1962.0
1962.0
Median
1970.0
1971.0
1972.0
1971.0
1971.0
75th
Percentile
1975.0
1976.0
1976.0
1976.0
1976.0
90th
Percentile
1980.0
1980.0
1981.0
1981.0
1 980.0
Maximum
1993.0
1993.0
1993.0
1993.0
1993.0
Table A.23b.   Model Information for the Relationship between Median Year Built and Geometric Mean Blood
               Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Median_Yr_Built
Intercept
time
Median_Yr_Built



Intercept
time
Median_Yr_Built
Intercept
time
Median_Yr_Built
Intercept
time
Median_Yr_Built
Estimate
3.424
-0.156
-0.022
3.412
-0.156
-0.025



-3.303
-0.168
-0.028
-4.451
-0.159
-0.031
-5.935
-0.139
-0.034
Standard
Error
0.018
0.003
0.001
0.017
0.003
0.001



0.014
0.003
0.001
0.017
0.003
0.002
0.021
0.005
0.002
p-value
<.001
<.001
<.001
<.001
<.001
<.001



<.001
<001
<.001
<.001
<001
<.001
<.001
<.001
<001
-2 Log
Likelihood
153381


195953





269400


300399


362407


Variance Components
Random Effects
a,f = 0.708
a?12 = -0.068
0,2* = 0.020
a,,* = 0.552
a?12= -0.052
0,2* = 0.016




-------
                              Median Year Occupied Units were built
                                                           8-
                                                              Legend:
                  1995-2000
                     - 2000-2002
	2002-2004   	2004-2006
     1939 1945 1951 1957 1963 1969  1975 1981 1987 1993

                   Year Occupied Unit Built
1930  1940  1950  1960  1970  1980   1990   2000

              Year Occupied Unit Built
Figure A.24.  Median Year Occupied Units were built: Histogram and Linear Relationship with Geometric Mean
              Blood Lead Levels by Time

Table A.24a.  Summary Information for Median Year Occupied Units were built by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
1968.24
1969.31
1969.39
1969.26
1968.99
Standard
Error
0.07
0.08
0.08
0.08
0.04
Minimum
1939.0
1939.0
1939.0
1939.0
1939.0
10th
Percentile
1953.0
1955.0
1954.0
1954.0
1954.0
25th
Percentile
1961.0
1963.0
1963.0
1962.0
1962.0
Median
1971.0
1972.0
1972.0
1972.0
1971.0
75th
Percentile
1976.0
1976.0
1977.0
1977.0
1976.0
90th
Percentile
1980.0
1980.0
1980.0
1981.0
1980.0
Maximum
1993.0
1993.0
1993.0
1993.0
1993.0
Table A.24b.  Model Information for the Relationship between Median Year Occupied Units were Built and
              Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Median_Yr_Occ_Built
Intercept
time
Median_Yr_Occ_Built
Intercept
time
Median_Yr_Occ_Built
Intercept
time
Median_Yr_Occ_Built
Intercept
time
Median_Yr_Occ_Built
Intercept
time
Median_Yr_Occ_Built
Estimate
3.425
-0.156
-0.022
3.412
-0.156
-0.024
-1.485
-0.146
-0.017
-3.302
-0.168
-0.028
-4.451
-0.159
-0.031
-5.935
-0.139
-0.034
Standard
Error
0.018
0.003
0.001
0.017
0.003
0.001
0.012
0.002
0.001
0.014
0.003
0.001
0.017
0.003
0.002
0.021
0.005
0.002
p-value
<.001
<001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
153391


195964


250433


269427


300439


362487


Variance Components
Random Effects
0-^= 0.709
o-?12= -0.068
cr22^ = 0.020
0^2= 0.553
a?12 = -0.052
a22^ = 0.016
0-,^= 0.304
a?12 = -0.023
cv= 0.010
0-^2= 0.378
0-P12 = -0.012
0-^= 0.008
a, ,2= 0.430
CT212 = -0.021
o-22* = 0.010
0-^2 = 0.418
CT?12= -0.024
cr22^= 0.012
Error
aenor*= 0.474
a8rror2= 38.341




                                                  Page A-24
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines.  It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                    Percent Units Built Before 1940
6
1

-
5 • lr

4-


m
c
 r

1
2j
T-
f

1
r- f
_ /
Or i i i











^X

\
r-1 ^^ \
r \
_r ^ r

-. s

n,
<
1 T ' ' T 1 1 y 1 1^1 M ii 1 1 1:




















.
v























'^
0 4 8 12 16 20 24 28 32 36 40













































ttfl
Himi:ttซ=. 	 r
8
•
•o
(D
0)
o :
m

C -
S :
s
I *•
ซ
E
o
0)
O

•D
0)
E 2:
1 :
a !
0:

Legend: 1995-2000 	 2000-2002
	 2002-2004 	 2004-2006






^ — '
^^^^^
^
^^- 	 _^_

-r^*" ^___^_-j;ฑ,"l^J-'- ^^A — - — -**"
j, r *• *• 2,".^ -*~~"*~ 	 ^ — ~ "^
_ — — ~* ~^-*~ ' 	 ~*~ 	 __. r^~ ~^
^•f~~*~ _^_-r-^~^

— ^ ~*~'




t 	 ' 1 	 1 	 i i | i i i i r i i n | 	 T i i | i i i i > r i i i g i i i 	 |
44 48 52 56 60 Q 10 20 30 40 50 60 70
                   Percent Built Before 1940
                                                                           Percent Built Before 1940
Figure A.25.
Percent Units Built Before 1940: Histogram and Linear Relationship with Geometric Mean Blood
Lead Levels by Time
Table A.25a. Summary Information for Percent Units Built Before 1940 by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
19.54
17.66
17.59
17.93
18.27
Standard
Error
0.09
0.10
0.10
0.10
0.05
Minimum
0.3
0.4
0.4
0.4
0.3
10th
Percentile
5.0
4.7
4.8
4.9
4.8
25th
Percentile
8.6
7.7
7.6
7.7
7.8
Median
17.6
13.8
13.6
13.9
14.6
75th
Percentile
29.0
26.3
26.3
27.2
27.7
90th
Percentile
37.6
36.2
36.4
36.8
36.9
Maximum
59.8
59.8
60.0
60.0
60.0
Table A.25b.   Model Information for the Relationship between Percent Units Built Before 1940 and Geometric
               Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Ffct_Built_Pre_1 940
Intercept
time
Ftt_Built_Re_1940
Intercept
time
Pct_Built_Pre_1940
Intercept
time
Pct_Built_Pre_1940
Intercept
time
Ffct_Built_Pre_1940



Estimate
3.428
-0.156
1.762
3.418
-0.156
2.040
-1.482
-0.146
1.296
-3.297
-0.168
2.234
-4.444
-0.159
2.508



Standard
Error
0.018
0.003
0.116
0.017
0.003
0.115
0.012
0.002
0.087
0.014
0.003
0.111
0.017
0.003
0.126



p- value
<.001
<.001
<.001
<.001
<.001
<.001
<001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001



-2 Log
Likelihood
153396


195945


250449


269297


300081





Variance Components
Random Effects
a^2= 0.714
a212= -0.068
0,2*= 0.020
a, ,2= 0.554
a212= -0.052
a22^= 0.016
0,^= 0.308
a?12= -0.023
CT22^= 0.010
a,,2= 0.382
a?12= -0.012
a222= 0.008
a,,2= 0.440
a212= -0.023
0,2*= 0.010



Error
CTem>r2= 0.474
CTeTOr2= 38.344




                                                  Page A-25
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                    Percent Units Built Before 1950
  c 3
  4)
                                                            8
                                                         ฃ
                                                         m
                                                         c
                                                         n
                                                         o
                                                          ซ
                                                          o
                                                          v
                                                         03
                                                         •o
                                                          v
                                                         1
                                                            2-
                                                            o-
                                                                Legend:   	 1995-2000
                                                                          	2002-2004
                                                                                             	2000-2002
                                                                                           	2004-2006
       1.2  7.2
              13.2 19.2 25.2 31.2 37.2 43.2 49.2 55.2 61.2 67.2

                   Percent Built Before 1950
                                                              0
10    20     30     40    50

       Percent Built Before 1950
                                                                                                    60
70
Figure A.26.
               Percent Units Built Before 1950: Histogram and Linear Relationship with Geometric Mean Blood
               Lead Levels by Time
Table A.26a.   Summary Information for Percent Units Built Before 1950 by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
26.52
24.63
24.55
24.88
25.24
Standard
Error
0.10
0.11
0.11
0.11
0.05
Minimum
0.8
0.8
0.8
0.8
0.8
10th
Percentile
9.3
8.9
9.1
9.2
9.2
25th
Percentile
14.6
13.7
13.5
13.6
14.0
Median
25.3
21.6
21.3
21.7
22.2
75th
Percentile
37.3
35.0
34.9
35.2
35.7
90th
Percentile
46.0
44.3
44.9
45.5
45.5
Maximum
67.1
67.1
67.1
67.1
67.1
Table A.26b.   Model Information for the Relationship between Percent Units Built Before 1950 and Geometric
               Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Ftt_Built_Pre_1950
Intercept
time
Ftt_Built_Pre_1950



Intercept
time
Rct_Built_Pre_1950
Intercept
time
Ffct_Built_Pre_1 950



Estimate
3.426
-0.156
1.571
3.416
-0.156
1.802



-3.298
-0.168
2.047
-4.445
-0.159
2.292



Standard
Error
0.018
0.003
0.103
0.017
0.003
0.102



0.014
0.003
0.098
0.017
0.003
0.112



p- value
<.001
<.001
<.001
<.001
<.001
<.001



<.001
<.001
<.001
<.001
<.001
<.001



-2 Log
Likelihood
153395


1 95949





269269


300079





Variance Components
Random Effects
0^2= 0.713
a?12= -0.068
o2./ = 0.020
v^2= 0.552
a?12= -0.052
a22^= 0.016



^,2= 0.376
a212= -0.012
CT2/ = 0.008
a,,2= 0.430
a212= -0.022
CT22^= 0.010



Error
CTe.o,2= 0.474
CTenor2= 38.344




                                                   Page A-26
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality-
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                     Percent Units Built Before 1960
     3.5
     3.0
     2.5
     2.0
  0)

  Q.
     1.5
     1.0
     0.5
                                                              8
•a
ra
V
o

DO

c
o
0)
O

•a
Q)

t>  2
                                                                  Legend:    	 1995-2000

                                                                            	2002-2004
                                    	2000-2002

                                   	2004-2006
        3.6  10.8  18 25.232.439.646.8  54 61.268.475.6


                    Percent Built Before 1960
          '"I	I"	I'	I""	i	i	I"'""

          10    20    30     40    50    60    70    80


                  Percent Built Before 1960
Figure A.27.   Percent Units Built Before 1960: Histogram and Linear Relationship with Geometric Mean Blood

               Lead Levels by Time
Table A.27a. Summary Information for Percent Units Built Before 1960 by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
37.13
35.42
35.28
35.56
35.94
Standard
&ror
0.11
0.13
0.12
0.12
0.06
Minimum
3.2
3.2
3.2
3.2
3.2
10th
Percentile
17.1
16.5
16.7
16.6
16.8
25th
Percentile
25.0
23.8
23.6
23.6
24.0
Median
36.8
33.6
33.3
33.5
34.1
75th
Percentile
49.2
47.1
47.0
47.3
47.7
90th
Percentile
58.4
56.5
57.1
57.6
57.5
Maximum
79.1
79.1
79.1
79.1
79.1
Table A.27b.   Model Information for the Relationship between Percent Units Built Before 1960 and Geometric

               Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Rct_Built_Pre_1960
Intercept
time
R3t_Built_Pre_1 960



Intercept
time
Ftt_Built_Pre^1960
Intercept
time
Ftt_Built_Pre_1 960
Intercept
time
Ftt_Built_Pre_1960
Estimate
3.425
-0.156
1.340
3.412
-0.156
1.494



-3.301
-0.168
1.835
-4.449
-0.159
2.083
-5.934
-0.139
2.304
Standard
Error
0.018
0.003
0.094
0.017
0.003
0.093



0.014
0.003
0.089
0.017
0.003
0.101
0.021
0.005
0.124
p-value
<.001
<.001
<.001
<.001
<.001
<.001



<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
153419


195993





269404


300408


362573


Variance Components
Random Effects
CTn2= 0.718
CT212= -0.068
a222 = 0.020
a^2= 0.564
a212= -0.052
a2/= 0.016



an2= 0.376
a?12= -0.012
a2/ = 0.008
a,,2= 0.425
a?12= -0.021
a2/= 0.010
an2= 0.406
a212= -0.024
0,2*= 0.012
Error
ae|TOr2 = 0.474
aem,2= 38.342




                                                    Page A-27

This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality

guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency

determination or policy.

-------
                                   Percent Units Built Before 1970
4.0 "

3.5 '

3.0 '


2.5 '
c
0>
5 2.0 "
a.


1.5 '



1.0 •


0.5 "
n J
r
/ p
' f
f "
/

1
L
J
1

' f
1

J.

L
J
ni
/
,


x
A
-,n\
\
•INi
i _

-\ '
\
_ \ r

' "
\

\ -,

\
\
\
\
s


8;

•D
to
0)
o ^
o
m
i
Q)
O ,
•C 4
ซ)
E
o
a>
O
•D
0)
o 2 ;
1
Q_ ]

o-

Legend: 1995-2000 " 2000-2002
	 2002-2004 	 2004-2006





^^^
^
^^*-*~^'
^^-^^*~
_^*-*^^' ,---• — — — "•-
^ 	 ^ __ _ ,, -2. ^-'— -'-'~ "" " ~ 	
^^^**^^ *"•*'* ^/-^ — -~ *~ ^—r- — "" — "~
*-**~^, "^— -'"''" ~~ *~ —^ — ~^ ^
f - •• -*_^ ^--^ ^~~M'^~ ~-~- —
=-'— J ^^ 	 _^- — ""
	 	 — ' ~





| 1 I I | f 1 > 1 | 1 I I 1 | I I I 1 j 1 1 I I | I 1 I 1 | I I I I | 1 I !, | > I I I |
6.75 17.25 27.75 38.25 48.75 59.25 69.75 80.25 0 10 20 30 40 50 60 70 80 90
Percent Built Before 1970 n_. .- .
Figure A..2S.  Percent Units Built Before 1970: Histogram and Linear Relationship with Percent Units Geometric
              Mean Blood Lead Levels by Time

Table A.28a.  Summary Information for Percent Units Built Before 1970 by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
49.18
47.79
47.64
47.85
48.19
Standard
Error
0.11
0.13
0.12
0.12
0.06
Minimum
6.1
6.1
6.1
6.1
6.1
10th
Percentile
28.9
28.0
28.1
28.1
28.3
25th
Percentile
38.2
36.7
36.6
36.7
37.1
Median
49.1
47.4
47.0
47.2
47.8
75th
Percentile
61.1
59.5
59.3
59.9
60.1
90th
Percentile
69.8
68.8
69.0
69.3
69.2
Maximum
89.4
89.4
89.4
89.4
89.4
Table A.28b.  Model Information for the Relationship between Percent Units Built Before 1970 and Geometric
              Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Pct_Built_Pre_1970
Intercept
time
Ftt_Built_Pre_1 970
Intercept
time
Ffct_Built_Pre_1 970
Intercept
time
Ffct_Built_Pre_1970
Intercept
time
Pct_BuilLPre_1970
Intercept
time
Ffct_Built_Pre_J970
Estimate
3.424
-0.156
1.287
3.410
-0.156
1.396
-1.485
-0.146
1.089
-3.303
-0.168
1.784
-4.452
-0.160
2.015
-5.939
-0.140
2.236
Standard
Error
0.018
0.003
0.093
0.018
0.003
0.092
0.012
0.002
0.069
0.014
0.003
0.089
0.017
0.003
0.102
0.021
0.005
0.125
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<001
<.001
<.001
<.001
<.001
<.001
<001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
1 53432


196021


250462


269499


300594


363034


Variance Components
Random Effects
CT,,2^ 0.723
d212= -0.068
a222= 0.020
a,,2= 0.572
a212= -0.053
o^= 0.016
anz= 0.308
a?]2= -0.023
CT2/= 0.010
a1(2= 0.379
a?12= -0.011
CT22^ = 0.008
CT,,2= 0.429
a212= -0.021
a22^= 0.010
a,,2= 0.411
CT212= -0.023
a222= 0.012
Error
aeTOr2 = 0.474
^r2= 38.339




                                                 Page A-28
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                   Percent Units Built Before 1980
4.5 '

4.0 '


3.5 '


3.0 '


+*
g 2.5 '
o
c
ฐ- 2.0 -

1^ —
.5



1.0 '


0.5 '























'
y

0 i i i




i
S'
/|f

(

r
i
m
J
iff
,

/
/
Tn
.
Jr-
~\J ~
-






J"
N\n



i
v r
\-
\ " i
\

\

^
\ -.

>
\
\
k
8-
•
-Q '.

-------
                              Percent Occupied Units Built Before 1940
  V
                                V
                                                            8-
                                                         T5
                                                         s
o
m

I

o
•z


I
O
TJ
                                                         2
                                                         0.
                                                            4-
   2-
                                                            o-
       Legend:
                                                                               1995-2000
                                                                               2002-2004
2000-2002
2004-2006
      0.5  6.5  12.5 18.5 24.5 30.5 36.5 42.5 48.5 54.5

            Percent Occupied Units Built Before 1940
           11111
            10     20      30     40     50

           Percent Occupied Units Built Before 1940
         60
Figure A.30.   Percent Occupied Units Built Before 1940: Histogram and Linear Relationship with Geometric
               Mean Blood Lead Levels by Time

Table A.30a.   Summary Information for Percent Occupied Units Built Before 1940 by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
19.53
17.60
17.52
17.86
18.23
Standard
B-ror
0.09
0.10
0.10
0.10
0.05
Minimum
0.3
0.6
0.5
0.5
0.3
10th
Percentile
5.1
4.7
4.8
4.9
4.9
25th
Percentile
8.7
7.6
7.5
7.6
7.9
Median
17.5
13.5
13.2
13.6
14.6
75th
Percentile
29.3
26.6
26.6
27.3
27.7
90th
Percentile
37.7
36.4
36.5
36.8
36.9
Maximum
59.3
59.3
59.3
59.3
59.3
Table A.30b.   Model Information for the Relationship between Percent Occupied Units Built Before 1940 and
               Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Rct_Occ_Built_Pre_1 9
Intercept
time
Ftt_Occ_Built_Pre_19
Intercept
time
Pct_Occ_Built_Pre_1 9
Intercept
time
Ftt_Occ_Built_Pre_1 9
Intercept
time
Ffct_Occ_Built_Pre_19
Intercept
time
Ftt_Occ^Built_Pre_19
Estimate
3.429
-0.156
1.711
3.418
-0.156
1.989
-1.482
-0.146
1.269
-3.297
-0.168
2.203
-4.444
-0.159
2.484
-5.924
-0.137
2.688
Standard
Error
0.018
0.003
0.117
0.017
0.003
0.116
0.012
0.002
0.087
0.014
0.003
0.111
0.017
0.003
0.127
0.021
0.005
0.155
p-value
<.001
<.001
<.001
<.001
<.001
<001
<.001
<.001
<.001
<.001
<.001
<.001
<001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
153410


1 95961


250463


269317


300093


361379


Variance Components
Random Effects
a,,2= 0.717
a212= -0.068
o-222 = 0.020
a,,2 = 0.557
a2,2= -0.052
a2/ = 0.016
cr,,2^ 0.309
a?12= -0.023
CT2/ = 0.010
a,,2= 0.384
a212 = -0.012
o-222= 0.008
CTl12 = 0.442
a2,2= -0.023
cr2/ = 0.010
a1t2= 0.438
a212= -0.028
a2/= 0.012
Error
CTem>r2= 0-474
ฐenj= 38.345




                                                  Page A-30
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                             Percent Occupied Units Built Before 1950
  c
  I
  a>
  a
                                                         73
                                                         to
                                                         0)
                                                         o
                                                         m

                                                         ง
                                                            8-
                                                            6
•|  4
i
O
o
                                                            2
                                                                Legend:
                       1995-2000
2000-2002
                                                                         	2002-2004   	2004-2006
      1.2 7.2 13.2 19.2 25.2 31.2 37.2 43.2 49.2 55.2 61.2

            Percent Occupied Units Built Before 1950
          111111
           10
11 M i[ M 11 i! i rr p i

  20    30
n ri u j i 11 u IT 11 ] i i

   40     50
  -rrp-r

   60
                                                 70
                                                                    Percent Occupied Units Built Before 1950
Figure A.31.   Percent Occupied Units Built Before 1950: Histogram and Linear Relationship with Geometric
               Mean Blood Lead Levels by Time

Table A.31a.   Summary Information for Percent Occupied Units Built Before 1950 by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
26.38
24.44
24.34
24.68
25.06
Standard
Error
0.10
0.11
0.11
0.11
0.05
Minimum
1.0
1.0
1.0
1.0
1.0
10th
Percentile
9.5
8.8
8.9
9.1
9.1
25th
Percentile
14.7
13.6
13.4
13.5
13.8
Median
25.2
21.1
20.8
21.3
22.0
75th
Percentile
37.5
34.7
34.7
35.3
35.7
90th
Percentile
46.2
44.4
44.8
45.2
45.2
Maximum
65.6
65.9
65.9
65.9
65.9
Table A.31b.   Model Information for the Relationship between Percent Occupied Units Built Before 1950 and
               Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Pct_Occ_Built_Pre_19
Intercept
time
Ffct_Occ^Built_Pre_19
Intercept
time
Ffct_Occ_Built_Pre_1 9
Intercept
time
Ftt Occ Built Pre 19
Intercept
time
Pct_Occ_Built_Pre_19
Intercept
time
Rst.Occ .Built_Pre_19
Estimate
3.427
-0.156
1.540
3.416
-0.156
1.771
-1.483
-0.146
1.210
-3.298
-0.168
2.031
-4.445
-0.159
2.282
-5.925
-0.137
2.475
Standard
Error
0.018
0.003
0.104
0.017
0.003
0.103
0.012
0.002
0.077
0.014
0.003
0.099
0.017
0.003
0.112
0.021
0.005
0.137
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
153406


195961


250430


269294


300103


361607


Variance Components
Random Effects
CT,,2= 0.715
a212= -0.068
a222= 0.020
a,,2 = 0.555
a212 = -0.052
CT222 = 0.016
an2 = 0.306
a?,2= -0.023
a22^= 0.010
a,,2= 0.378
a,, 2= -0.012
a222 = 0.008
a,,2= 0.432
a212= -0.022
CT22^ = 0.010
a,,2= 0.422
a212= -0.026
CT22^= 0.012
Error
ฐeror2 = 0.474
ฐe^2= 38.344




                                                  Page A-31
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                             Percent Occupied Units Built Before 1960
4.0 "

3.5 '

3.0 -


2.5 '
c
I 2.0-
o_


1.5 "



1.0 '

0.5 -
0 J



.
iCK'
/
' r
t'
,
i
•
~
-i '

• i
i
f
t
^
y
A-
JJ1



"X
\

V
- > r
\
i .
\ r
\

\ -
H -ri
\
\
s
'ffL
lm>^
8;

•o
(0
o
•a B I
0 6 .
o
CD -
I :
g
•- 4 -
o
E j
o
fi
O
TJ
0>
.ซ 2"
o>
a ;
o:

Legend: 1995-2000 2000-2002
	 2002-2004 	 2004-2006





^^^^
^r***""****^
^^^~^
^ — _ - - ~ ~ ~ U"-- — 	
^^s~*^^~ ff *•"'•' i"^ — '•* *~ ~* ~~' 	 - - 	 	
--•—'"" _ _ - - - "2jZ- — ~r~^~ 	 , — "- ~^~ ~^
^ _, _, - - ;J"_^ f— — ' ' ' ^_j_ 	 ' -^-^ '"^
-r-^—'-^'""'^ 	 - 	 - — "*
__ — *• —
— • '




3.6 10.8 18 25.2 32.4 39.6 46.8 54 61.2 68.4 75.6 Q 10 20 30 40 50 60 70 8C
             Percent Occupied Units Built Before 1960
                                                                   Percent Occupied Units Built Before 1960
Figure A.32.  Percent Occupied Units Built Before 1960: Histogram and Linear Relationship with Geometric
              Mean Blood Lead Levels by Time

Table A.32a.  Summary Information for Percent Occupied Units Built Before 1960 by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
36.97
35.19
35.02
35.29
35.71
Standard
&ror
0.11
0.13
0.12
0.12
0.06
Minimum
3.0
3.0
3.0
3.0
3.0
10th
Percentile
17.4
16.5
16.6
16.5
16.8
25th
Percentile
24.8
23.5
23.2
23.2
23.7
Median
36.5
33.2
32.9
33.2
33.8
75th
Percentile
49.2
46.9
46.8
47.3
47.8
90th
Percentile
57.9
56.3
56.9
57.5
57.4
Maximum
79.6
79.6
79.6
79.6
79.6
Table A.32b.  Model Information for the Relationship between Percent Occupied Units Built Before 1960 and
              Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Pct_Occ_Built_Pre_19
Intercept
time
Ffct_Occ_Built_Pre_1 9



Intercept
time
Ffct_Occ_Built_Pre_19
Intercept
time
Ftt_Occ_Built_Pre_19
Intercept
time
Ffct_Occ_Built_Pre_19
Estimate
3.425
-0.156
1.327
3.413
-0.156
1.481



-3.301
-0.168
1.833
-4.449
-0.159
2.084
-5.934
-0.139
2.278
Standard
Error
0.018
0.003
0.094
0.017
0.003
0.093



0.014
0.003
0.089
0.017
0.003
0.102
0.021
0.005
0.124
p-value
<.001
<001
<.001
<.001
<.001
<001



<.001
<.001
<.001
<.001
<.001
<.001
<001
<.001
<.001
-2 Log
Likelihood
153425


195999





269428


300443


362627


Variance Components
Random Effects
a^= 0.719
CT?12= -0.068
o-2/ = 0.020
0^2= 0.565
o-pl2= -0.052
cr22a= 0.016



a,f= 0.377
a?12= -0.012
a22* = 0.008
0-,^= 0.426
a212= -0.021
a22*= 0.010
a,,2= 0.409
a212= -0.024
CT222= 0.012
Error
ฐe^= O-474
^e.or2= 38.342




                                                  Page A-32
    information w distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                             Percent Occupied Units Built Before 1970
4.0 -

3.5 '
3.0 *


2.5 -
O
5 2.0-
0.

1.5 "



1.0 '

0.5 '


















'
ntflfl
0 ' '
r.
X
I / .
,
^
/

/
•fl
/
B

'
r— i/

'
u/ '


i I I
•x
\

InflV
\

Aln
\
\-
\

\
\ T-,

\
\
\
s
\
run n-i

                                                           8-
                                                        •o
                                                        CO
                                                        0)
                                                        •ง 6i
                                                        m

                                                        I
                                                         i
                                                        o
                                                           2-
                                                   Legend:       - 1995-2000
                                                            	2002-2004
                                                                                                2000-2002
                                                                                                2004-2006
15   24   33   42  51   60   69   78   87

Percent Occupied Units Built Before 1970
                                                                  10   20   30   40   50   60   70   80   90

                                                                    Percent Occupied Units Built Before 1970
Figure A.33.   Percent Occupied Units Built Before 1970: Histogram and Linear Relationship with Geometric
               Mean Blood Lead Levels by Time

Table A.33a.   Summary Information for Percent Occupied Units Built Before 1970 by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
48.83
47.37
47.19
47.40
47.78
Standard
&ror
0.11
0.13
0.12
0.12
0.06
Minimum
5.9
5.9
5.9
5.9
5.9
10th
Percentile
29.2
27.9
27.9
27.8
28.1
25th
Percentile
37.6
36.5
36.1
36.1
36.7
Median
48.4
46.5
46.1
46.1
47.0
75th
Percentile
60.7
59.4
59.0
59.8
60.0
90th
Percentile
69.5
68.4
68.7
69.1
68.9
Maximum
89.1
89.1
89.1
89.1
89.1
Table A.33b.   Model Information for the Relationship between Percent Occupied Units Built Before 1970 and
               Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Ftt_Occ_Built_Pre_19
Intercept
time
Ftt_Occ_Built_Pre_19
Intercept
time
Ffct_Occ_Built_Pre_1 9
Intercept
time
Ftt_Occ_Built_Pre_19
Intercept
time
Ftt_Occ_Built_Pre_1 9
Intercept
time
Pct_Occ_Built_Pre^19
Estimate
3.424
-0.156
1.279
3.410
-0.156
1.388
-1.485
-0.146
1.086
-3.303
-0.168
1.785
-4.452
-0.160
2.019
-5.939
-0.140
2.213
Standard
Error
0.018
0.003
0.094
0.018
0.003
0.093
0.012
0.002
0.069
0.014
0.003
0.089
0.017
0.003
0.102
0.021
0.005
0.126
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
153436


196025


250472


269531


300653


3631 34


Variance Components
Random Effects
a,,z= 0.722
CT?12 = -0.068
Vy? = 0.020
CT,,2 = 0.572
a212= -0.053
CT222= 0.016
an2= 0.308
a?12= -0.023
cr22^= 0.010
a,,2= 0.379
a?12= -0.011
o^= 0.008
a,,2= 0.429
a?12= -0.020
cr222= 0.010
an2= 0.414
a212= -0.022
CT222= 0.012
Error
aarror2 = 0.474
ฐ8TOr2= 38.339




                                                  Page A-33
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                             Percent Occupied Units Built Before 1980
4.5 '

4.0 -
3.5 "
3.0 '

•*-
ง 2'5
9
ฐ- 2.0 -


1.5 -


1.0 '

0.5 '



"
/'*
n
r/ T
r
J '
rV
i •
1
•i

I •
1
i
J -r
/
r 71
rnJtejflf
0 i i i i i i

N n
•• \
r r
h
•\ r-.
\
\
L

\
\
\
S
\
\
Ik
8

1
0 61
0 ]
m -
c
CD
o>
S
•— 4 "
A
E :
o
i
o
o>
o 2"
1 -
a !
•
.
" ' ' "i1 	 i' " ' ' 'i" ' ' " u n

Legend: 1995-2000 '" 2000-2002
	 2002-2004 	 2004-2006




^^^
^^-^^
^^i^***^
^^-^"" ,,-— ^-'-:-~':";
^ — **" , - - ~ 3-^-' ^" ' — ""*" "" -J— ' *~
^f^'*' f. - ~ """ฃ•'-* *~ "^ ' ^-i— -1 	 ~J~r"
^Z^ -" ^ ' ^^~ —"^
J--— aJ
	 	 "






18 25.232.439.646.8 54 61.268.475.682.8 90 10 20 30 40 50 60 70 80 90 10
             Percent Occupied Units Built Before 1 980
                                                                   Percent Occupied Units Built Before 1980
Figure A.34.  Percent Occupied Units Built Before 1980: Histogram and Linear Relationship with Geometric
              Mean Blood Lead Levels by Time

Table A.34a.  Summary Information for Percent Occupied Units Built Before 1980 by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
67.66
66.46
66.29
66.31
66.75
Standard
Error
0.09
0.11
0.11
0.11
0.05
Minimum
18.3
18.3
18.3
18.3
18.3
10th
Percentile
50.5
49.1
49.0
48.8
49.2
25th
Percentile
58.9
57.4
57.3
57.1
57,7
Median
68.3
67.0
66.5
66.5
67.3
75th
Percentile
78.4
77.0
76.7
76.9
77.4
90th
Percentile
84.8
84.0
84.0
84.4
84.4
Maximum
94.3
95.0
95.0
95.0
95.0
Table A.34b.  Model Information for the Relationship between Percent Occupied Units Built Before 1980 and
              Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Pct_Occ_Built_Pre_19
Intercept
time
Pct_Occ_Built_Pre_1 9
Intercept
time
Ftt_Occ_Built_Pre_19
Intercept
time
Ffct_Occ_Built_Pre_19
Intercept
time
R:t_Occ_Built_Pre_19
Intercept
time
Ffct_Occ_Built_Pre_19
Estimate
3.422
-0.155
1.415
3.409
-0.156
1.523
-1.487
-0.146
1.181
-3.305
-0.168
1.918
-4.453
-0.159
2.144
-5.937
-0.139
2.443
Standard
Error
0.018
0.003
0.107
0.018
0.003
0.105
0.012
0.002
0.079
0.014
0.003
0.103
0.017
0.003
0.118
0.021
0.005
0.147
p- value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
153445


1 96038


250483


269497


300492


362692


Variance Components
Random Effects
CT,,2= 0.728
CT212= -0.068
a^= 0.020
a,,2= 0.577
CT?12= -0.053
a2/= 0.016
<7112= 0.312
a212= -0.024
a22^= 0.010
a,z= 0.388
ay,z= -0.011
CT22^= 0.008
a, ,2= 0.443
ap,2= -0.021
a22^= 0.010
a,,2= 0.426
a?12= -0.023
cr2/= 0.012
Error
ฐ^= OA74
aeror2= 38.339




                                                 Page A-34
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines.  It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                           Percent Residents Less than Six Years of Age
8


7

6



5

4


3

2



1





-
i
/

/

]
r|
I

1
f




1 1,1 II 1 1 1,1
3.8 5 6.2 7.4 8.6



•
_
\
IT

\

\
\
ji\
\
i\
\
i \
-
\
-A
)w_n
9.8 11 12.2 13.4 14.6
8-

•
to
3 :
o 6:
o ;
CD
C -
(0
i
••• 4 -
fl
E :
o :
o :
a)
TS 2:
'•5
0)
a I
o-


Legend: 1995-2000 2000-2002
	 2002-2004 	 2004-2006








"^^— — 	 -_^^^^^
=888*55- >,.
~~ ^- — ' — ^*"^ 3_— "*• — — ^ -,
— ' — ' ^— 	 ^-T-Z~^— —"-!!.""






3 4 5 6 7 8 9 10 11 12 13 14 15 16
               Percent Less than 6 Years of Age
                                                                      Percent Less than 6 Years of Age
Figure A.35.
Percent Residents Less than Six Years of Age: Histogram and Linear Relationship with Geometric
Mean Blood Lead Levels by Time
Table A.35a.  Summary Information for Percent Residents Less than Six Years of Age by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
8.97
9.09
9.07
9.02
9.03
Standard
Error
0.01
0.01
0.01
0.01
0.01
Minimum
3.7
4.8
4.8
3.9
3.7
10th
Percentile
7.5
7.5
7.5
7.5
7.5
25th
Percentile
8.2
8.2
8.2
8.2
8.2
Median
8.9
9.0
9.0
8.9
8.9
75th
Percentile
9.7
9.8
9.8
9.8
9.8
90th
Percentile
10.5
10.8
10.7
10.6
10.6
Maximum
15.5
15.5
15.5
15.1
15.5
Table A.35b.  Model Information for the Relationship between Percent Residents Less than Six Years of Age and
              Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Ftt_LE_Six
Intercept
time
R:t_LE_Six
Intercept
time
Ftt_LE_Six
Intercept
time
Pct_LE_Six
Intercept
time
Ftt_LE_Six
Intercept
time
R:t_LE_Six
Estimate
3.418
-0.155
-4.818
3.405
-0.156
-6.567
-1.489
-0.146
-4.949
-3.309
-0.169
-3.782
-4.456
-0.161
-3.616
-5.929
-0.139
-5.630
Standard
Error
0.018
0.003
1.095
0.018
0.003
1.105
0.012
0.002
0.825
0.015
0.003
1.122
0.018
0.003
1.308
0.023
0.005
1.631
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
0.006
<.001
<001
<.001
-2 Log
Likelihood
153590


196196


250652


269671


300145


360358


Variance Components
Random Effects
0,^= 0.773
o212= -0.070
<32* = 0.020
a,,2= 0.625
aff= -0.055
cr22^= 0.016
a, ,2= 0.328
a?12= -0.022
0^= 0.010
a,^= 0.451
a212= -0.009
0,2* = 0.008
CTn2= 0.533
a212= -0.020
a2/= 0.010
a,,2= 0.552
a?12= -0.026
a2/= 0.012
Error
ฐeTO,2= 0.474
CTeTO-2= 38.340




                                                 Page A-35
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                           Number Residents Less than Six Years of Age
    80


    70


    60


    50

  I
  fc 40
  0.

    30


    20


    10
                                                           8-
•o
s
E  e
_o
no
i
•I  4
5

O
9
O
   2-
                                                           o-
      Legend:   	 1995-2000   	2000-2002
               	2002-2004   	2004-2006
              24000000  48000000  72000000  96000000

                Number Less than 6 Years of Age
  O.OOE+00      5.00E+07       1.00E+08

             Number Less than 6 Years of Age
1.50E+08
Figure A.36.  Number Residents Less than Six Years of Age: Histogram and Linear Relationship with Geometric
              Mean Blood Lead Levels by Time

Table A.36a.  Summary Information for Number Residents Less than Six Years of Age by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
#######
#######
#######
#######
#######
Standard
&ror
30578.82
30423.42
27339.23
24313.95
14354.97
Minimum
13700.0
12000.0
12000.0
13200.0
12000.0
10th
Percentile
96500.0
82600.0
80100.0
78900.0
84000.0
25th
Percentile
166600.0
141500.0
139300.0
135700.0
145500.0
Median
#######
#######
#######
#######
#######
75th
Percentile
898900.0
767700.0
735700.0
697500.0
782100.0
90th
Percentile
2558800.0
2174600.0
2109100.0
1987500.0
2200900.0
Maximum
#########
#########
#########
#########
#########
Table A.36b.  Model Information for the Relationship between Number Residents Less than Six Years of Age and
              Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept.
time
Num_LE_Six
Intercept
time
Num_LE_Six



Intercept
time
Num_LE_Six
Intercept
time
Num_LE_Six
Intercept
time
Num_LE_Six
Estimate
3.419
-0.155
0.000
3.401
-0.156
0.000



-3.311
-0.169
0.000
-4.460
-0.161
0.000
-5.940
-0.141
0.000
Standard
Error
0.018
0.003
0.000
0.018
0.003
0.000



0.015
0.003
0.000
0.018
0.003
0.000
0.023
0.005
0.000
p-value
<.001
<.001
<.001
<.001
<.001
<.001



<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
153638


196258





269858


300609


361516


Variance Components
Random Effects
a,,2 = 0.777
a212 = -0.070
a222 = 0.020
0,^= 0.637
a2,2= -0.056
a2/= 0.016



a,,2 = 0.451
a212= -0.008
o^ = 0.008
a,,2= 0.528
a212= -0.018
a2/= 0.010
a,/= 0.548
a212= -0.023

-------
                    Percent Residents with Less than Ninth Grade Education
                                                           8-
                                                        "O  R
                                                        o  b
                                                        _o
                                                        CD
                                                        o
                                                        •a
                                                        
-------
                         Percent Residents without a High School Degree
  (5
  Q.
                                                        1
m

ง


I  4
o

O
1
ซ  2
'•o

-------
                            Percent Residents without College Education
    5
    4
  c
  
-------
                             Percent Residents without College Degree
7
0



6


5 '





c 4 '
o>
o
s.
<0
Q.
3 '


2 "



1 •








-|

J^







'









n /

35 39 43 47 51 55 59 63 67 71 75 79
Percent without College Degree







\

\

\
~
\

\

\

\
i
(
\
\
\
1\
- IJ 1 IIJ II Ij n-
83 87 91

8


•a
n
4)
TJ c
0 6
o
CO

c
(0
0)


•= 4^

0)
E
o
i
o
"S
t3 2;
'•5
0)
OL '.
o-
c


Legend: 1995-2000 " - 2000-2002
	 2002-2004 	 2004-2006












^
^^^r*^"*'^
^ff^~f**~r^
_^-— 	 "*~^ *-'~~~^-'-
^^^-^-^" ,,,-^-'"^-"::'~
— — ^^fi"^-- 	 ~" 	 ^ '
^ 	 	 *=--





1 	 1 	 1 	 1 	 1 	 1 	 1 	 " [
0 40 50 60 70 80 90 10
Pernpnt without Pnllpnp FVปnrFปp
Figure A.40.  Percent Residents without College Degree: Histogram and Linear Relationship with Geometric
              Mean Blood Lead Levels by Time

Table A.40a.  Summary Information for Percent Residents without College Degree by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
78.47
79.56
79.58
79.71
79.27
Standard
Error
0.05
0.06
0.06
0.06
0.03
Minimum
35.0
35.0
35.0
35.0
35.0
10th
Percentile
68.0
69.0
69.0
69.3
68.8
25th
Percentile
74.4
75.9
76.0
76.3
75.6
Median
80.0
81.1
81.1
81.2
80.9
75th
Percentile
83.9
85.1
85.1
85.1
84.9
90th
Percentile
86.9
87.8
87.9
87.8
87.6
Maximum
93.1
93.6
93.6
93.6
93.6
Table A.40b.  Model Information for the Relationship between Percent Residents without College Degree and
              Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Ftt_No_College_Degre
Intercept
time
Ftt_No_College_Degre
Intercept
time
Pct_No_College_Degre
Intercept
time
Ffct_No_College_Degre
Intercept
time
Ffct_No_College_Degre
Intercept
time
Fct_No_College_Degre
Estimate
3.409
-0.156
1 .630
3.397
-0.156
1.724
-1.499
-0.146
1.991
-3.312
-0.170
1.158
-4.458
-0.161
0.479
-5.934
-0.140
0.051
Standard
Error
0.018
0.003
0.191
0.018
0.003
0.184
0.012
0.002
0.137
0.015
0.003
0.188
0.018
0.003
0.214
0.023
0.005
0.254
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
0.025
<.001
<.001
0.840
-2 Log
Likelihood
153541


196150


250344


269423


300010


360729


Variance Components
Random Effects
an2= 0.756
a212= -0.069
a2/= 0.020
a,*= 0.606
a2]2= -0.054
a2/= 0.016
a,,^ 0.309
a212= -0.023
CT22^= 0.010
a,,2= 0.443
CT212= -0.008
a222 = 0.008
CT,,2= 0.532
a212= -0.020
o2^= 0.010
on2= 0.556
a?12= -0.025
a22z= 0.012
Error
CTerror2= 0-474
aem)r2= 38.338




                                                 Page A-40
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                           Total Housing Units
    90

    80

    70

    60

    50

    40

    30

    20

    10

     0
   6-
o
m
•S  4
I
O
3
O
s
•a
a>
   o-
      Legend
1995-2000
2002-2004
2000-2002
2004-2006
      30000   630000  1230000 1830000 2430000 3030000
                      Total Housing Units
             1000000    2000000    3000000

                   Total Housing Units
                       4000000
Figure A.41.   Total Housing Units: Histogram and Linear Relationship with Geometric Mean Blood Lead Levels
               by Time

Table A.41a.   Summary Information for Total Housing Units by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
53994.17
45568.66
44143.19
42969.61
47133.55
Standard
Error
1080.45
1082.30
977.22
904.39
513.88
Minimum
877.0
815.0
936.0
935.0
815.0
10th
Percentile
5219.0
4523.0
4346.0
4340.0
4634.0
25th
Percentile
8595.0
7509.0
7393.0
7337.0
7692.0
Median
16951.0
14953.0
14504.0
14098.0
15297.0
75th
Percentile
41394.0
34461 .0
33470.0
32726.0
35163.0
90th
Percentile
111762.0
93973.0
93070.0
90628.0
95800.0
Maximum
3270909.0
3270909.0
3270909.0
3270909.0
3270909.0
Table A.41b.   Model Information for the Relationship between Total Housing Units and Geometric Mean Blood
               Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Total_Housing_Units
Intercept
time
Total_Housing_Units



Intercept
time
Total_Housing_Units
Intercept
time
Total_Housing_Units
Intercept
time
TotaLHousing_Units
Estimate
3.418
-0.155
0.000
3.401
-0.156
0.000



-3.311
-0.169
0.000
-4.460
-0.161
0.000
-5.941
-0.142
0.000
Standard
Error
0.018
0.003
0.000
0.018
0.003
0.000



0.015
0.003
0.000
0.018
0.003
0.000
0.023
0.005
0.000
p-value
<.001
<.001
<.001
<.001
<.001
<.001



<.001
<.001
<001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
153641


196260





269875


300676


361719


Variance Components
Random Effects
a^ = 0.777
a212 = -0.070
CT222 = 0.020
a,,2= 0.637
a212 = -0.056
0^= 0.016



0^,2= 0.451
a212= -0.008
CT22^ = 0.008
a,,2= 0.527
a212= -0.018
o-2/= 0.010
au2 = 0.547
a212= -0.022
cr22^ = 0.012
Error
aerror2 = 0.474
ฐ.ro,2= 38-339




                                                  Page A-41
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                             Total Population
    80


    70


    60


    50
  +*

  S 40
  0.

    30


    20


    10
TJ
CO
"ง
m
CO

-------
                                             Housing Density
    100
     80
  „  60
  0)
  Q.
     40
     20
        300   5100  9900 14700 19500 24300 29100 33900

                        Housing Density
                                                             8-
                                                             6
                                                          o
                                                          m
—  4 -

i
ซ  2
1
                                                                Legend:
                    - 1995-2000    	2000-2002
                	2002-2004    	2004-2006
              10000      20000      30000      40000

                     Housing Density
Figure A.43.   Housing Density: Histogram and Linear Relationship with Geometric Mean Blood Lead Levels by
               Time
Table A.43a. Summary Information for Housing Density by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
113.87
102.19
103.32
132.32
113.30
Standard
Error
2.51
2.64
2.67
6.94
2.04
Minimum
0.5
0.3
0.3
0.3
0.3
10th
Percentile
8.3
7.5
7.2
7.6
7,7
25th
Percentile
13.9
13.0
12.9
13.1
13.3
Median
28.1
26.0
26.0
26.3
26.6
75th
Percentile
74.6
64.9
62.9
62.6
68.0
90th
Percentile
221.5
189.7
188.8
188.0
196.5
Maximum
7421.2
7421.2
7421.2
34756.7
34756.7
 Table A.43b.  Model Information for the Relationship between Housing Density and Geometric Mean Blood Lead
               Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
Housing_Density
Intercept
time
Housing_Density
Intercept
time
Housing_Density
Intercept
time
Housing^Density



Intercept
time
Housing_Density
Estimate
3.418
-0.155
0.000
3.402
-0.156
0.000
-1.491
-0.146
0.000
-3.311
-0.169
0.000



-5.936
-0.141
0.000
Standard
Error
0.018
0.003
0.000
0.018
0.003
0.000
0.012
0.002
0.000
0.015
0.003
0.000



0.023
0.005
0.000
p-value
<.001
<.001
0.635
<.001
<.001
0.279
<001
<.001
0.007
<.001
<.001
0.792



<.001
<.001
0.137
-2 Log
Likelihood
153631


196252


250723


269791





360997


Variance Components
Random Effects
e,,2= 0.779
a?12 = -0.070
CT222 = 0.020
a.,2= 0.636
a212 = -0.056
a222 = 0.016
a(12= 0.332
a212 = -0.022
o2/ = 0.010
an2= 0.453
cr212 = -0.008
a2/= 0.008



an2= 0.553
a212= -0.024
a222= 0.012
Error
ฐe^2= O-47^
aenor*= 38.338




                                                   Page A-43
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency-
determination or policy.

-------
                                            Average Air Lead
    90


    80


    70 '

    60


    50

    40

    30

    20

    10
m

S
•c
                                                            8-
   6-
4-
   2-
1
a
       Legend:
                   1995-2000
                   2002-2004
   - 2000-2002
	2004-2006
        0   0.028 0.056  0.084  0.112  0.14  0.168  0.196

                          Average Air
    0.00      0.05     0.10     0.15      0.20      0.25

                       Average Air
Figure A.44.   Average Air Lead: Histogram and Linear Relationship with Geometric Mean Blood Lead Levels by
               Time

Table A.44a.   Summary Information for Average Air Lead by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
0.00
0.00
0.00
0.00
0.00
Standard
Error
0.00
0.00
0.00
0.00
0.00
Minimum
0.0
0.0
0.0
0.0
0.0
10th
Percentile
0.0
0.0
0.0
0.0
0.0
25th
Percentile
0.0
0.0
0.0
0.0
0.0
Median
0.0
0.0
0.0
0.0
0.0
75th
Percentile
0.0
0.0
0.0
0.0
0.0
90th
Percentile
0.0
0.0
0.0
0.0
0.0
Maximum
0.2
0.2
0.2
0.2
0.2
 Table A.44b.  Model Information for the Relationship between Average Lead and Geometric Mean Blood Lead
               Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
air_avg
Intercept
time
air_avg
Intercept
time
air_avg
Intercept
time
air_avg
Intercept
time
air_avg
Intercept
time
air_avg
Estimate
3.418
-0.155
3.473
3.402
-0.156
2.222
-1.491
-0.146
1.780
-3.310
-0.169
7.581
-4.459
-0.161
8.770
-5.936
-0.141
6.328
Standard
Error
0.018
0.003
1.884
0.018
0.003
1.815
0.013
0.002
1.385
0.015
0.003
1.805
0.018
0.003
1.972
0.023
0.005
2.223
p-value
<.001
<.001
0.065
<.001
<.001
0.221
<.001
<.001
0.199
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
0.004
-2 Log
Likelihood
153605


196229


250722


269805


300468


361117


Variance Components
Random Effects
e^2= 0.776
a?12 = -0.070
a222 = 0.020
a,,2 = 0.635
a?12= -0.056
a222 = 0.016
a, ,2= 0.332
a2,2 = -0.022
CT22^= 0.010
a,,2 = 0.448
a?12= -0.008
cr222 = 0.008
a,,2 = 0.526
cr?12 = -0.019
a222= 0.010
a,,2 = 0.552
a212= -0.024
cr222= 0.012
&ror
CTemr2= 0.474
aeror2= 38.339




                                                   Page A-44
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency-
determination or policy.

-------
                                             Median Air Lead
    100
     80
     60
     40
     20
      0.00125 0.02625  0.05125 0.07625 0.10125 0.12625

                           Median Air
                                                             8-
                                                             4
                                                             2
                                                                Legend:   	 1995-2000
                                                                          	2002-2004
                              	2000-2002
                              	2004-2006
  r "               111| ii i         iii| ii i
0.00  0.02   0.04   0.06   0.08  0.10   0.12   0.14

                   Median Air
Figure A.45.   Median Air Lead: Histogram and Linear Relationship with Geometric Mean Blood Lead Levels by
               Time
Table A.45a.   Summary Information for Median Air Lead by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
0.00
0.00
0.00
0.00
0.00
Standard
Bror
0.00
0.00
0.00
0.00
0.00
Minimum
0.0
0.0
0.0
0.0
0.0
10th
Percentile
0.0
0.0
0.0
0.0
0.0
25th
Percentile
0.0
0.0
0.0
0.0
0.0
Median
0.0
0.0
0.0
0.0
0.0
75th
Percentile
0.0
0.0
0.0
0.0
0.0
90th
Percentile
0.0
0.0
0.0
0.0
0.0
Maximum
0.1
0.1
0.1
0.1
0.1
 Table A.45b.  Model Information for the Relationship between Median Air Lead and Geometric Mean Blood
               Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
air_med
Intercept
time
air_med
Intercept
time
air_med
Intercept
time
air_med
Intercept
time
air_med
Intercept
time
air_med
Estimate
3.419
-0.155
8.448
3.402
-0.156
6.028
-1.490
-0.146
3.804
-3.310
-0.169
14.489
-4.459
-0.161
15.660
-5.937
-0.141
1 1 .607
Standard
Error
0.018
0.003
2.772
0.018
0.003
2.663
0.012
0.002
2.032
0.015
0.003
2.611
0.018
0.003
2.865
0.023
0.005
3.227
p-value
<.001
<.001
0.002
<.001
<.001
0.024
<.001
<.001
0.061
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
153598


196225


250721


269809


300502


361193


Variance Components
Random Effects
a,,2= 0.773
a?12 = -0.070
CT2/= 0.020
o^2 = 0.633
a?12 = -0.055
a2/ = 0.016
a1,2 = 0.332
0212= -0.022
a22^= 0.010
at12= 0.445
cr?12 = -0.008
a222 = 0.008
a, ,2 = 0.522
a?12 = -0.018
0^= 0.010
a112= 0.549
a212 = -0.024
CT22^= 0.012
Error
ฐem,2= 0.474
^™r2= 38.339




                                                   Page A-45
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                         Air Lead 95th Percentile
    100
     80
     60 '
     40
     20
      Q   I ' '   ~~1	  	1	  	1	  	1	 	1	 	1	  	1	

       0.005  0.085  0.165  0.245  0.325 0.405 0.485  0.565
                                                             8-
                                                          -   4
                                                                 Legend:    	 1995-2000
                                                                           	2002-2004
                                                                              	2000-2002
                                                                              	2004-2006
                       95th Percentile Air
                                                0.0     0.1      0.2     0.3     0.4

                                                                95th Percentile Air
                                                                                                    0.5
0.6
Figure A.46.
Air Lead 95* Percentile: Histogram and Linear Relationship with Geometric Mean Blood Lead
Levels by Time
Table A.46a.   Summary Information for Air Lead 95th Percentile by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
0.01
0.01
0.01
0.01
0.01
Standard
Error
0.00
0.00
0.00
0.00
0.00
Minimum
0.0
0.0
0.0
0.0
0.0
10th
Percentile
0.0
0.0
0.0
0.0
0.0
25th
Percentile
0.0
0.0
0.0
0.0
0.0
Median
0.0
0.0
0.0
0.0
0.0
75th
Percentile
0.0
0.0
0.0
0.0
0.0
90th
Percentile
0.0
0.0
0.0
0.0
0.0
Maximum
0.6
0.6
0.6
0.6
0.6
 Table A.46b.  Model Information for the Relationship between Air Lead 95'  Percentile and Geometric Mean
               Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
air_p95
Intercept
time
air_p95






Intercept
time
air_p95



Estimate
3.418
-0.155
0.258
3.401
-0.156
-0.076






-4.459
-0.161
2.587



Standard
Error
0.018
0.003
0.729
0.018
0.003
0.703






0.018
0.003
0.768



p-value
<.001
<.001
0.724
<.001
<.001
0.914






<.001
<.001
<.001



-2 Log
Likelihood
153610


196232








300442





Variance Components
Random Effects
atl2= 0.778
ap,2 = -0.070
a222 = 0.020
0-,,*= 0.636
a?12 = -0.056
a2/ = 0.016






a112= 0.530
a?12 = -0.019
a 22*= 0.010



Error
aerror2 = 0.474
aerror2 = 38.339




                                                   Page A-46
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                       Current HUD Funding ($)
80


70


60;
(^KS
W
*— "
O)
.E 50-
T3
C
3
LL
Q 40-
13 ;
+*
c

>
i
!


i


c














c
1
c
c
c
c
[
e





=l

*
*











* i
* * *
* ฃ
* *
-
!
i 1
I I i
ill


<














<
ซ:


e
<



8

•
T>
(0
d>
o 6-
0 I
m-

c -
CO
*
&
y J
-— 4 ~
*-
0)
0
0)
C3
•o
1 2:
'•5
ซ
i_
Q.
o-


Legend: ^^^™" 1995-2000 	 2000-2002
	 2002-2004 	 2004-2006











^^ ___
^^k 	 	 	 — — '
^^^ ___. 	 — 	 	
^--^s^ — -=---rn""
"~^x- ^^v
~-x ^^.
""~V^NW
x^^
*v


0 10 20 30 40 50 60 70 80
        All Years
                 1995-2000   2000-2002  2002-2004  2004-2006
                                                                           Current HUD Funding ($)
Figure A.47.   Current HUD Funding: Histogram and Linear Relationship with Geometric Mean Blood Lead
               Levels by Time

Table A.47a.   Summary Information for Current HUD Funding by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
0.12
0.09
0.11
0.18
0.13
Standard
&ror
0.01
0.01
0.01
0.01
0.00
Minimum
0.0
0.0
0.0
0.0
0.0
10th
Percentile
0.0
0.0
0.0
0.0
0.0
25th
Percentile
0.0
0.0
0.0
0.0
0.0
Median
0.0
0.0
0.0
0.0
0.0
75th
Percentile
0.0
0.0
0.0
0.0
0.0
90th
Percentile
0.3
0.1
0.1
0.3
0.2
Maximum
27.7
30.9
35.5
75.9
75.9
 Table A.47b.  Model Information for the Relationship between Current HUD Funding and Geometric Mean
               Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
current
Intercept
time
current
Intercept
time
current
Intercept
time
current
Intercept
time
current
Intercept
time
current
Estimate
3.417
-0.155
-0.028
3.400
-0.156
-0.038
-1.493
-0.145
-0.048
-3.314
-0.169
-0.080
-4.462
-0.161
-0.106
-5.936
-0.140
-0.120
Standard
Error
0.018
0.003
0.003
0.018
0.003
0.001
0.013
0.002
0.001
0.015
0.003
0.001
0.018
0.003
0.003
0.023
0.005
0.006
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<001
<.001
<.001
<.001
<.001
<.001
<001
<.001
-2 Log
Likelihood
153550


195617


247645


269336


312172


397511


Variance Components
Random Effects
CT,,2= 0.779
a?12 = -0.070
cr22^ = 0.020
a,,2= 0.636
a212 = -0.056
CT222 = 0.016
an2 = 0.334
a212= -0.022
a^= 0.010

-------
                                     Cumulative HUD Funding ($)
400




g 300
0)
_^
^
c
LL
Q
3 200
,

Cumulati
-L
0
0




0





1
s|





















5
e
































|




i
















i




i i





*
>







%
X
B
i
*



.






<
c
















8-
"
•
•o
ta
^
0 I
m
c -
m
0)
5
•— 4 -
*-<
E :
o
4)
O
$ 2:
T3
g
0. I
o:




Legend: ^^^~ 1995-2000 ~~ 2000-2002
	 2002-2004 	 2004-2006







<
^t
\
•IT** JL
' *~*~^ 	 — . 	 	
^ "* — i_ """ "^ — • ~— — . 	 .
V


I 	 I 	 ' ' ' I 	 | i . I 1 1 I r . I |
0 100 200 300 40

        All Years   1995-2000  2000-2002   2002-2004  2004-2006
                                                                         Cumulative HUD Funding ($)
Figure A.48.   Cumulative HUD Funding: Histogram and Linear Relationship with Geometric Mean Blood Lead
               Levels by Time

Table A.48a.   Summary Information for Cumulative HUD Funding by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
2.06
2.79
3.53
4.87
3.26
Standard
Error
0.04
0.08
0.09
0.13
0.04
Minimum
0.0
0.0
0.0
0.0
0.0
10th
Percentile
0.0
0.0
0.0
0.0
0.0
25th
Percentile
0.0
0.0
0.0
0.0
0.0
Median
0.5
0.0
1.3
1.4
0.7
75th
Percentile
2.2
2.8
2.8
3.6
2.8
90th
Percentile
4.3
4.3
6.1
7.4
5.0
Maximum
110.3
147.1
181.6
310.0
310.0
 Table A.48b.  Model Information for the Relationship between Cumulative HUD Funding and Geometric Mean
               Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
cumulative
Intercept
time
cumulative
Intercept
time
cumulative
Intercept
time
cumulative



Intercept
time
cumulative
Estimate
3.419
-0.156
0.001
3.398
-0.155
-0.003
-1.495
-0.144
-0.004
-3.314
-0.168
-0.003



-5.933
-0.143
0.002
Standard
Error
0.018
0.003
0.001
0.018
0.003
0.001
0.013
0.002
0.000
0.015
0.003
0.000



0.023
0.005
0.001
p-value
<.001
<.001
0.111
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001



<.001
<.001
0.008
-2 Log
Likelihood
153621


196227


250649


269691





361196


Variance Components
Random Effects
a,,2= 0.779
a?12= -0.070
0-22* = 0.020
a, * = 0.635
a212= -0.056
a2/= 0.016
a^ = 0.334
ap12 = -0.022
CT22^= 0.010
crl12= 0.457
a?12 = -0.008
a 22* = 0.008



0^2= 0.551
a?12= -0.026
a222= 0.012
Error
CTe™r2= 0.474

-------
                                   Mean Water Lead Concentration
   ฃ

   i
   •o
   s
11
10
9
8:
7
6
5:
4

3

2
1
0
-1 •
-2

*






*
*





i







*






*
*
*

*
* - i *
1*1
& i i






                                                             8-
                                                         TJ c J
                                                          O '


                                                         m


                                                          (5
                                                          0)
                                                         I 4:

                                                          E
                                                          o
                                                          
-------
Model
Number
1
**f,H-<
2
•^<ง„*=

o,,2= 0.456
0012= -0.009
ฐ22Z =

.
.
.

0,,"= 0.562
oslซ= -0.026
ฐW =

Error
ฐ.m,rZ= 0.472
oeno,2= 38.214




/
                                                           Page A-50
   This information is distributed solely for the purpose ofpre-dissemiriation peer review under applicable information quality
   guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
   determination or policy.

-------
    100
     80
  ^  60
     40
     20
                                     TRI Compounds (Total Air)
                                                           8-
                                                        •o
                                                        a
                                                        T>  e
                                                        O  6
                                                        _O
                                                        m

                                                        (0
                                                        
-------
    100
     so
  ^  60
  0)
  o
  O
  Q.
     40
     20
                                   TRI Compounds (Fugitive Air)
                                                           8
m

ง

•-  4
I
O
0>
O

•**  o
O  ซ
'•5
                                                              Legend:
                    • 1995-2000
                   — 2002-2004
  - 2000-2002
	2004-2006
        600  9000  174002580034200426005100059400

                     Lead Total-Fugitive Air
                                        "' i1"
         10000 20000 30000 40000 50000 60000 70000
                                                                          Lead Total-Fugitive Air
Figure A.51.  TRI Compounds (Fugitive Air): Histogram and Linear Relationship with Geometric Mean Blood
              Lead Levels by Time

Table A.Sla.  Summary Information for TRI Compounds (Fugitive Air) by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
Mean
192.07
152.42
142.17
137.46
0 | 158.27
Standard
Error
18.57
19.44
17.37
17.20
9.15
Minimum
0.0
0.0
0.0
0.0
0.0
10th
Percentile
0.0
0.0
0.0
0.0
0.0
25th
Percentile
0.0
0.0
0.0
0.0
0.0
Median
0.0
0.0
0.0
0.0
0.0
75th
Percentile
1.2
0.7
0.6
0.5
0.8
90th
Percentile
53.0
39.9
35.5
33.0
39.9
Maximum
66985.5
66985.5
66985.5
66985.5
66985.5
Table A.Slb.  Model Information for the Relationship between TRI Compounds (Fugitive Air) and Geometric
              Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
air_fug
Intercept
time
air_fug
Intercept
time
air_fug
Intercept
time
air_fug
Intercept
time
air_fug
Intercept
time
air_fug
Estimate
3.418
-0.155
0.000
3.402
-0.156
0.000
-1.491
-0.146
0.000
-3.311
-0.169
0.000
-4.458
-0.161
0.000
-5.934
-0.140
0.000
Standard
Error
0.018
0.003
0.000
0.018
0.003
0.000
0.013
0.002
0.000
0.015
0.003
0.000
0.018
0.003
0.000
0.023
0.005
0.000
p-value
<.001
<.001
0.005
<.001
<.001
0.004
<.001
<.001
0.298
<.001
<.001
0.008
<.001
<.001
0.057
<.001
<.001
0.717
-2 Log
Likelihood
153626


196247


250744


269791


300354


360831


Variance Components
Random Effects
a^ = 0.776
apl2 = -0.070
a222 = 0.020
a,* = 0.633
a212 = -0.056
a222 = 0.016
0^,2 = 0.333
a?,2 = -0.022
cr222 = 0.010
a,,2= 0.451
0,,2= -0.008
CT222 = 0.008
a, ,* = 0.533
a:>12= -0.020
a2/= 0.010
^,2= 0.557
ap12= -0.025
CT2/= 0.012
Error
ฐ.norz = 0.474
aerror2= 38.339




                                                  Page A-52
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                              TRI Compounds-Air Lead from Stacks
    100
     80
  ^  60
     40
     20
                                                          8-
o

m



I
o

TJ
9

T5

'•5
o
Legend:
                                                                             1995-2000

                                                                             2002-2004
                                  — - 2000-2002

                                  	2004-2006
             12000  24000 36000 48000 60000 72000  84000



                       Lead Total-Stacks
           20000    40000    60000



                   Lead Total-Stacks
                              80000   100000
Figure A.52.  TRI Compounds (Air Lead from Stacks): Histogram and Linear Relationship with Geometric

              Mean Blood Lead Levels by Time
Table A.52a. Summary Information for TRI Compounds (Air Lead from Stacks) by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
393.51
314.32
303.75
298.00
331.68
Standard
Error
20.86
19.67
17.63
17.50
9.67
Minimum
0.0
0.0
0.0
0.0
0.0
10th
Percentile
0.0
0.0
0.0
0.0
0.0
25th
Percentile
0.0
0.0
0.0
0.0
0.0
Median
0.3
0.0
0.0
0.0
0.0
75th
Percentile
137.2
86.0
80.2
80.2
94.4
90th
Percentile
751.3
625.5
606.4
585.7
Maximum
85695.3
85695.3
85695.3
85695.3
646.2 | 85695.3
Table A.52b.  Model Information for the Relationship between TRI Compounds (Air Lead from Stacks) and

              Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
air_stk
Intercept
time
air_stk
Intercept
time
air_stk



Intercept
time
air_stk
Intercept
time
air_stk
Estimate
3.418
-0.155
0.000
3.401
-0.156
0.000
-1.491
-0.146
0.000



-4.459
-0.161
0.000
-5.935
-0.140
0.000
Standard
Error
0.018
0.003
0.000
0.018
0.003
0.000
0.013
0.002
0.000



0.018
0.003
0.000
0.023
0.005
0.000
p- value
<.001
<.001
0.871
<.001
<.001
0.750
<.001
<.001
0.678



<.001
<.001
0.060
<.001
<.001
0.440
-2 Log
Likelihood
153633


196255


250745





300390


360883


Variance Components
Random Effects
a^= 0.779
a212 = -0.070
a222 = 0.020
a,,2 = 0.636
a212= -0.056
CT22" = 0.016
0-,^= 0.333
a?12 = -0.022
a2/= 0.010



an2 = 0.533
a?12= -0.019
cr2/ = 0.010
a,* = 0.557
a?,2= -0.025
a2/= 0.012
Error
aerror2 = 0.474
ฐBnor2= 38-339




                                                 Page A-53


This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality

guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency

determination or policy.

-------
    100
     80
  *.  60

  0)
  o
     40
     20
                                   TRI Compounds-Water Surface


                                                           8-
T3
(0
<0


•a
o
_o

m


a
0)
                                                        ?  4-
1
o
V
O

I
TS  2
                                                           o-
                                                              Legend:
                    ' 1995-2000

                   -- 2002-2004
                            - 2000-2002

                         	2004-2006
        600  9000  17400 25800 34200 42600 51000 59400


                    Lead Total-Water Surface
                                        " T " '


                     30000 40000 50000  60000 70000
10000 20000


       Lead Total-Water Surface
Figure A.53.  TRI Compounds (Water Surface): Histogram and Linear Relationship with Geometric Mean Blood

              Lead Levels by Time



Table A.53a.  Summary Information for TRI Compounds (Water Surface) by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
110.89
79.73
72.99
68.17
84.69
Standard
Error
14.85
13.35
11.87
11.72
6.67
Minimum
0.0
0.0
0.0
0.0
0.0
10th
Percentile
0.0
0.0
0.0
0.0
0.0
25th
Percentile
0.0
0.0
0.0
0.0
0.0
Median
0.0
0.0
0.0
0.0
0.0
75th
Percentile
0.6
0.2
0.1
0.1
0.3
90th
Percentile
49.0
33.5
32.1
29.7
35.1
Maximum
66686.8
66686.8
66686.8
66686.8
66686.8
Table A.53b.  Model Information for the Relationship between TRI Compounds (Water Surface) and Geometric

              Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
water_surf
Intercept
time
water_surf
Intercept
time
water_surf
Intercept
time
water^surf
Intercept
time
water_surf
Intercept
time
water_surf
Estimate
3.418
-0.155
0.000
3.401
-0.156
0.000
-1.491
-0.146
0.000
-3.311
-0.169
0.000
-4.459
-0.161
0.000
-5.934
-0.140
0.000
Standard
Error
0.018
0.003
0.000
0.018
0.003
0.000
0.013
0.002
0.000
0.015
0.003
0.000
0.018
0.003
0.000
0.023
0.005
0.000
p-value
<.001
<.001
0.292
<.001
<001
0.500
<.001
<.001
0.981
<.001
<.001
0.743
<.001
<.001
0.714
<.001
<.001
0.501
-2 Log
Likelihood
153631


1 96254


250744


269789


300341


360851


Variance Components
Random Effects
0-^2 = 0.779
a212 = -0.070
0222 = 0.020
an2 = 0.636
a212= -0.056
a 22*= 0.016
a, ,2= 0.333
a2,2= -0.022
cr2/= 0.010
a, ,2= 0.454
CT212= -0.008
cr222= 0.008
a,,2= 0.535
a212= -0.020
CT222= 0.010
a,,2 = 0.557
a?12 = -0.025
CT2/= 0.012
Error
CTeTO,2= ฐ'474
*eror2= 38.339




                                                  Page A-54

This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality

guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency

determination or policy.

-------
                               TRI Compounds - Water by Injection
    100
     80
  0)
  u
     60
     40
     20
                                                        1
                                                        o
                                                        o
                                                        m
                                                        c
•-  4 -
s
ฃ
o
o
o
T3
I  2:
                                                        ฃ
                                                        Q.
                                                           o-
                                                              Legend:
                    • 1995-2000
                	2002-2004
  - 2000-2002
	2004-2006
              40000  80000  120000 160000 200000  240000
                 Lead Total-Underwater Injection
                100000        200000

              Lead Total-Underwater Injection
          300000
Figure A.54.   TRI Compounds (Water by Injection): Histogram and Linear Relationship with Geometric Mean
               Blood Lead Levels by Time

Table A.54a.   Summary Information for TRI Compounds (Water by Injection) by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
35.80
93.16
128.25
127.84
92.75
Standard
Error
17.21
38.08
42.81
42.53
17.52
Minimum
0.0
0.0
0.0
0.0
0.0
10th
Percentile
0.0
0.0
0.0
0.0
0.0
25th
Percentile
0.0
0.0
0.0
0.0
Median
0.0
0.0
0.0
0.0
0.0 | 0.0
75th
Percentile
0.0
0.0
0.0
0.0
0.0
90th
Percentile
0.0
0.0
0.0
0.0
0.0
Maximum
245603.4
245603.4
245603.4
245603.4
245603.4
Table A.54b.   Model Information for the Relationship between TRI Compounds (Water by Injection) and
               Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
lime
underjnj
Intercept
time
underjnj
Intercept
time
underjnj
Intercept
time
underjnj
Intercept
time
underjnj
Intercept
time
underjnj
Estimate
3.418
-0.155
0.000
3.401
-0.156
0.000
-1.491
-0.146
0.000
-3.311
-0.169
0.000
-4.459
-0.161
0.000
-5.934
-0.140
0.000
Standard
Error
0.018
0.003
0.000
0.018
0.003
0.000
0.013
0.002
0.000
0.015
0.003
0.000
0.018
0.003
0.000
0.023
0.005
0.000
p-value
<.001
<.001
0.411
<.001
<.001
0.363
<.001
<.001
0.775
<.001
<.001
0.778
<.001
<.001
0.661
<.001
<.001
0.383
-2 Log
Likelihood
153634


196256


250746


269790


300341


360841


Variance Components
Random Effects
a1t2 = 0.779
a212= -0.070
a ^ = 0.020
a,,2 = 0.636
a?12 = -0.056
a222 = 0.016
a,,2= 0.333
a21z = -0.022
o^= 0.010
an2= 0.453
a?12= -0.008
a2/ = 0.008
an2= 0.535
a212= -0.020
a222 = 0.010
a,,2= 0.556
a?12= -0.025
CT2/ = 0.01 2
Error
a 2 = 0.474
error
aenor2= 38.339




                                                 Page A-55
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
    100
     80
  „  60

  ง
  0)
  a.
     40
     20
TRI Lead Only (Total Air)

                      8"f Legend:
                                                          to
                                                          v
                                                          o
                                                          _o
                                                          m

                   |
                   o
                   
-------
                                    TRI Lead Only (Fugitive Air)
100



80




^ 60 '
S
0.
40



20
n
















\
\
8;
•
•D -
(U
-1 '.
o 6 ~
o
CD -
c -
CO
V
~ 4 -
E
0
o
0
•D -
% 2:
'•a
ฃ
a .
0-

Legend: ^^^ 1995-2000 	 2000-2002
	 2002-2004 	 2004-2006






^^^^^
^ -**
-*-— ^•*-"
	 	 	
	 	 	 ~~ 	 	
— — ' —
,





















0 2000 4000 6000 8000 100001200014000 0 5000 10000 15000 20000
Lead Total-Fugitive Air Legd TotahRjgititfe Kr
Figure A.56.  TRI Lead Only (Fugitive Air): Histogram and Linear Relationship with Geometric Mean Blood
              Lead Levels by Time

Table A.56a.  Summary Information for TRI Lead Only (Fugitive Air) by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
54.66
39.06
38.66
39.45
43.76
Standard
Error
3.71
3.38
3.11
3.03
1.70
Minimum
0.0
0.0
0.0
0.0
0.0
10th
Percentile
0.0
0.0
0.0
0.0
0.0
25th
Percentile
0.0
0.0
0.0
0.0
0.0
Median
0.0
0.0
0.0
0.0
0.0
75th
Percentile
1.1
0.5
0.4
0.4
0.7
90th
Percentile
34.8
22.0
19.0
18.9
25.0
Maximum
15513.2
15513.2
15513.2
15513.2
15513.2
Table A.56b.  Model Information for the Relationship between TRI Lead Only (Fugitive Air) and Geometric
              Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
underjnj
Intercept
time
underjnj
Intercept
time
underjnj
Intercept
time
underjnj
Intercept
time
underjnj
Intercept
time
underjnj
Estimate
3.418
-0.155
0.000
3.401
-0.156
0.000
-1.491
-0.146
0.000
-3.311
-0.169
0.000
-4.459
-0.161
0.000
-5.936
-0.140
0.000
Standard
Error
0.018
0.003
0.000
0.018
0.003
0.000
0.013
0.002
0.000
0.015
0.003
0.000
0.018
0.003
0.000
0.023
0.005
0.000
p-value
<.001
<.001
0.689
<.001
<.001
0.652
<.001
<.001
0.954
<.001
<.001
0.932
<.001
<.001
0.661
<.001
<.001
0.555
-2 Log
Likelihood
1 53632


1 96254


250743


269787


300332


360904


Variance Components
Flandom Effects
a,,2= 0.779
a?12 = -0.070
0,22 = 0.020
an2 = 0.636
a212 = -0.056
cr222 = 0.016
a, ,2= 0.333
cr?12= -0.022
a22^ = 0.010
0^2= 0.454
a212= -0.008
a222 = 0.008
an2 = 0.535
a212= -0.020
a2/= 0.010
a,f = 0.555
a?12= -0.025
0,2* = 0.012
Error
CTe^2= 0.474
aem)r2= 38.339




                                                  Page A-57
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines.  It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                               TRI Lead Only-Air Lead from Stacks
    100
     80
  „  60

  ง
  0)
  a.
     40
     20
                                                           8-
                                                        "8
                                                        0)
jo
CD
(0
8
•S  4
I
o
4)
O
                                                        ฃ
                                                        a.
                                                              Legend:
                    • 1995-2000
                	2002-2004
   - 2000-2002
	2004-2006
            1050 2100 3150 4200 5250  6300 7350 8400
                       Lead Total-Stacks
~
0
            2000    4000     6000

                    Lead Total-Stacks
   8000
10000
Figure A.57.  TRI Lead Only (Air Lead from Stacks): Histogram and Linear Relationship with Geometric Mean
              Blood Lead Levels by Time

Table A.57a.  Summary Information for TRI Lead Only (Air Lead from Stacks) by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
103.02
84.52
83.07
88.02
90.60
Standard
Error
3.66
3.84
3.55
3.79
1.86
Minimum
0.0
0.0
0.0
0.0
0.0
10th
Percentile
0.0
0.0
0.0
0.0
0.0
25th
Percentile
0.0
0.0
0.0
0.0
0.0
Median
0.0
0.0
0.0
0.0
0.0
75th
Percentile
6.5
3.1
2.7
2.7
3.9
90th
Percentile
120.0
92.0
92.0
93.6
101.7
Maximum
8940.3
9009.6
8940.3
8940.3
9009.6
Table A.57b.  Model Information for the Relationship between TRI Lead Only (Air Lead from Stacks) and
              Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
air_tot
Intercept
time
air_tot
Intercept
time
air_tot
Intercept
time
air_tot
Intercept
time
air_tot



Estimate
3.418
-0.155
0.000
3.401
-0.156
0.000
-1.490
-0.146
0.000
-3.310
-0.169
0.000
-4.458
-0.161
0.000



Standard
Error
0.018
0.003
0.000
0.018
0.003
0.000
0.012
0.002
0.000
0.015
0.003
0.000
0.018
0.003
0.000



p-value
<.001
<.001
0.538
<.001
<.001
0.681
<.001
<.001
0.040
<.001
<.001
<.001
<.001
<.001
<.001



-2 Log
Likelihood
153631


1 96253


250741


26981 1


300451





Variance Components
Random Effects
a,,2 = 0.778
a212 = -0.070
a222 = 0.020
a,,2= 0.636
a212= -0.056
a22a = 0.016
CT,,2= 0.332
a212= -0.022
a22^= 0.010
a^= 0.449
a212= -0.008
a222 = 0.008
a,,2= 0.526
a2,2= -0.019
a2/= 0.010



Error
CTenor2= ฐ'474
ฐ.nor2= 38-339




                                                 Page A-58
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                    TRI Lead Only-Water Surface
    100
     80
  „  60
     40
     20
                                                            8-
                                                          to
                                                            2
'•a
o
                                                            o-
                                                                Legend:
                     • 1995-2000
                    — 2002-2004
   - 2000-2002
	2004-2006
             720 1440 2160 2880 3600 4320 5040 5760 6480

                    Lead Total-Water Surface
          1000  2000  3000  4000   5000   6000  7000

                 Lead Total-Water Surface
Figure A.58.   TRI Lead Only (Water Surface): Histogram and Linear Relationship with Geometric Mean Blood
               Lead Levels by Time
Table A.58a. Summary Information for TRI Lead Only (Water Surface) by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
14.56
11.82
11.10
10.99
12.27
Standard
Error
1.47
1.45
1.29
1.29
0.70
Minimum
0.0
0.0
0.0
0.0
0.0
10th
Percentile
0.0
0.0
0.0
0.0
0.0
25th
Percentile
0.0
0.0
0.0
0.0
0.0
Median
0.0
0.0
0.0
0.0
0.0
75th
Percentile
0.0
0.0
0.0
0.0
0.0
90th
Percentile
2.6
1.6
1.5
1.5
1.8
Maximum
6508.0
6508.0
6508.0
6508.0
6508.0
Table A.58b.   Model Information for the Relationship between TRI Lead Only (Water Surface) and Geometric
               Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
water_surf
Intercept
time
water_surf
Intercept
time
water_surf



Intercept
time
water_surf
Intercept
time
water_surf
Estimate
3.418
-0.155
0.000
3.401
-0.156
0.000
-1.491
-0.146
0.000



-4.459
-0.161
0.000
-5.935
-0.140
0.000
Standard
Error
0.018
0.003
0.000
0.018
0.003
0.000
0.012
0.002
0.000



0.018
0.003
0.000
0.023
0.005
0.000
p- value
<.001
<.001
0.774
<.001
<.001
0.876
<.001
<.001
0.212



<.001
<.001
0.017
<.001
<001
0.020
-2 Log
Likelihood
153628


196250


250739





300363


360919


Variance Components
Random Effects
a, ,2= 0.779
a?12= -0.070
CT2/ = 0.020
an2= 0.636
a212= -0.056
CT2/= 0.016
at12= 0.332
a212= -0.022
a %,* = 0.010



au2= 0.533
a212 = -0.020
cr2/= 0.010
a1t2= 0.554
a2,2= -0.025
o2./= 0.012
Error
ae™2= 0.474
aemr== 38.339




                                                  Page A-59
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                 TRI Lead Only - Water by Injection
    100
     80
  ^  60

  ง
  0)
  CL
     40 J
     20
                                                            8-
                                                                               1995-2000
                                                                          	2002-2004
                            2000-2002
                        	2004-2006
             6400  12800 19200 25600 32000 38400 44800

                 Lead Total-Underwater Injection
10000    20000    30000    40000

   Lead Total-Underwater Injection
50000
Figure A.59.   TRI Lead Only (Water by Injection): Histogram and Linear Relationship with Geometric Mean
               Blood Lead Levels by Time
Table A.59a. Summary Information for TRI Lead Only (Water by Injection) by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
2.27
26.11
23.17
22.89
17.43
Standard
Error
2.26
8.98
7.98
7.93
3.40
Minimum
0.0
0.0
0.0
0.0
0.0
10th
Percentile
0.0
0.0
0.0
0.0
0.0
25th
Percentile
0.0
0.0
0.0
0.0
0.0
Median
0.0
0.0
0.0
0.0
0.0
75th
Percentile
0.0
0.0
0.0
0.0
0.0
90th
Percentile
0.0
0.0
0.0
0.0
0.0
Maximum
45822.5
45822.5
45822.5
45822.5
45822.5
Table A.59b.   Model Information for the Relationship between TRI Lead Only (Water by Injection) and
               Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
underjnj
Intercept
time
underjnj
Intercept
time
underjnj
Intercept
time
underjnj
Intercept
time
underjnj
intercept
time
underjnj
Estimate
3.418
-0.155
0.000
3.401
-0.156
0.000
-1.491
-0.146
0.000
-3.311
-0.169
0.000
-4.459
-0.161
0.000
-5.936
-0.140
0.000
Standard
Error
0.018
0.003
0.000
0.018
0.003
0.000
0.013
0.002
0.000
0.015
0.003
0.000
0.018
0.003
0.000
0.023
0.005
0.000
p-value
<.001
<.001
0.689
<.001
<.001
0.652
<.001
<.001
0.954
<.001
<.001
0.932
<.001
<.001
0.661
<.001
<.001
0.555
-2 Log
Likelihood
153632


1 96254


250743


269787


300332


360904


Variance Components
Random Effects
al12= 0.779
a?12 = -0.070
CJ222 = 0.020
0^2= 0.636
a?12 = -0.056

-------
                                      TRI Total Lead (Total Air)
    100
     80
  ^  60

  ง
  4)
  D.
     40
     20
   81


TO
_i

I  ^
CD
                                                i
                                                c
                                                1
                                                s
                                                            o-
                                                               Legend:
                                                                    • 1995-2000
                                                                  -— 2002-2004
	2000-2002
	2004-2006
0    18000  36000  54000  72000  90000  108000

              Lead Total-Total Air
                                                                   20000  40000  60000  80000  100000 120000

                                                                            Lead Total-Total Air
Figure A.60.   TRI Total Lead (Total Air): Histogram and Linear Relationship with Geometric Mean Blood Lead
               Levels by Time

Table A.60a.   Summary Information for TRI Total Lead (Total Air) by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
Number
Missing
0
0
0
0
67295 | 0
Mean
727.58
579.13
557.13
552.35
612.07
Standard
Error
35.56
34.13
30.57
30.32
16.65
Minimum
0.0
0.0
0.0
0.0
0.0
10th
Percentile
0.0
0.0
0.0
0.0
0.0
25th
Percentile
0.0
0.0
0.0
0.0
0.0
Median
11.8
4.9
4.3
4.0
6.2
75th
Percentile
314.3
244.6
233.0
232.8
264.1
90th
Percentile
1241.9
1057.2
1024.8
1024.8
1100.3
Maximum
118778.1
118778.1
118778.1
118778.1
118778.1
Table A.60b.   Model Information for the Relationship between TRI Total Lead (Total Air) and Geometric Mean
               Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
air_tot
Intercept
time
air_tot
Intercept
time
air_tot
Intercept
time
airjot
Intercept
time
air_tot



Estimate
3.418
-0.155
0.000
3.401
-0.156
0.000
-1.490
-0.146
0.000
-3.310
-0.169
0.000
-4.458
-0.161
0.000



Standard
Error
0.018
0.003
0.000
0.018
0.003
0.000
0.012
0.002
0.000
0.015
0.003
0.000
0.018
0.003
0.000



p- value
<.001
<.001
0.538
<.001
<.001
0.681
<001
<.001
0.040
<.001
<.001
<.001
<.001
<.001
<.001



-2 Log
Likelihood
153631


196253


250741


26981 1


300451





Variance Components
Random Effects
a,,2 = 0.778
a212= -0.070
1 o2^= 0.020
a,,2 = 0.636
a?12 = -0.056
CT22^= 0.016
a^ = 0.332
CT?12= -0.022
CT2!/= 0.010
an2= 0.449
a?12= -0.008
a222 = 0.008
a,f = 0.526
a?12= -0.019

-------
                                    TRI Total Lead (Fugitive Air)
100 •




80





60 '
c
0)
o
0)
Q.

40 '



20 '

A -^
•




















I
\
\
8-
-
•O
(B
0)
— '
TJ R J
0 b .
0
m
c -
n
0)
5
1 4^

0)
E
o
0)
O
€ 2:
'•6
ฃ
Q.
o-

Legend: ^^^~ 1995-2000 — " 2000-2002
	 2002-2004 	 2004-2006




^^^^f
^^^^^^^
^^^^^^
^^^^^
^^^^^^
^******^^ .--~-~~'^
.^^^^^^^ " ~ ^-"•"^ ^"
^^^ .,-^'-^'"''
f _ - ^*^ -^ 	 „ . 	 	 	 "
_. 	 ^."-T — "~ 	 . 	 •
^ _ *• ^."J^ -*1 •** 	 	 . 	 	 	 	
~ฃ^ -~ ~* 	 	 	 "

"J~^~




























600 9000 174002580034200426005100059400 0 10000 20000 30000 40000 50000 60000 70000
Lead Total-Fugitive Air ,_. Tota. r^ra.... Air
Figure A.61.  TRI Total Lead (Fugitive Air): Histogram and Linear Relationship with Geometric Mean Blood
              Lead Levels by Time

Table A.61a.  Summary Information for TRI Total Lead (Fugitive Air) by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
Sample
Size
20286
14434
16235
16340
All Years | 67295
Number
Missing
0
0
0
0
0
Mean
237.11
185.02
175.89
172.56
195.49
Standard
Error
18.76
19.59
17.53
17.37
9.24
Minimum
0.0
0.0
0.0
0.0
0.0
10th
Percentile
0.0
0.0
0.0
0.0
0.0
25th
Percentile
0.0
0.0
0.0
0.0
0.0
Median
0.0
0.0
0.0
0.0
0.0
75th
Percentile
8.6
5.5
5.0
4.7
5.8
90th
Percentile
200.6
156.0
150.0
145.8
166.2
Maximum
66985.5
66985.5
66985.5
66985.5
66985.5
Table A.61b.  Model Information for the Relationship between TRI Total Lead (Fugitive Air) and Geometric
              Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
air_fug
Intercept
time
air_fug



Intercept
time
air_fug



Intercept
time
air_fug
Estimate
3.418
-0.155
0.000
3.402
-0.156
0.000



-3.311
-0.169
0.000



-5.935
-0.140
0.000
Standard
Error
0.018
0.003
0.000
0.018
0.003
0.000



0.015
0.003
0.000



0.023
0.005
0.000
p-value
<.001
<.001
0.506
<.001
<.001
0.469



<.001
<.001
0.067



<.001
<.001
0.076
-2 Log
Likelihood
153630


196251





269791





360929


Variance Components
Random Effects
a,,2= 0.778
a212= -0.070
a22" = 0.020
a,/= 0.636
a?,2= -0.056
CT2/= 0.016



a,,2= 0.453
a212= -0.008
a2/= 0.008



0,,*= 0.554
a212= -0.025
a222= 0.012
Error
ฐemr2= 0.474
aerror-= 38.339




                                                 Page A-62
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                               TRI Total Lead-Air Lead from Stacks
    100
     80
  _  60 '
  4>
  CL
     40
     20
                                                           8-
                                                        •D
                                                        CO
                                                        0>
OQ

(8
(I)
I
O
0)
O
T5
0)
                                                           2-
                                                        ซ
                                                        k~
                                                        Q.
                                                              Legend:
               ^^ 1995-2000
               	2002-2004
	2000-2002
	2004-2006
            12000 24000 36000 48000 60000 72000 84000

                       Lead Total-Stacks
            20000    40000    60000

                    Lead Total-Stacks
    80000    100000
Figure A.62.  TRI Total Lead (Air Lead from Stacks): Histogram and Linear Relationship with Geometric Mean
              Blood Lead Levels by Time

Table A.62a.  Summary Information for TRI Total Lead (Air Lead from Stacks) by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
490.47
394.11
381.24
379.79
416.58
Standard
&ror
22.58
21.25
19.05
18.97
10.46
Minimum
0.0
0.0
0.0
0.0
0.0
10th
Percentile
0.0
0.0
0.0
0.0
0.0
25th
Percentile
0.0
0.0
0.0
0.0
0.0
Median
5.2
1.6
1.3
1.3
2.3
75th
Percentile
223.5
155.8
149.1
149.3
180.5
90th
Percentile
947.9
821.7
821.7
816.2
891.4
Maximum
92189.9
92189.9
92189.9
92189.9
92189.9
Table A.62b.  Model Information for the Relationship between TRI Total Lead (Air Lead from Stacks) and
              Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
air_stk
Intercept
time
air_stk
Intercept
time
air_stk
Intercept
time
air_stk
Intercept
time
air_stk
Intercept
time
air_stk
Estimate
3.418
-0.155
0.000
3.401
-0.156
0.000
-1.491
-0.146
0.000
-3.310
-0.169
0.000
-4.458
-0.161
0.000
-5.935
-0.141
0.000
Standard
Error
0.018
0.003
0.000
0.018
0.003
0.000
0.012
0.002
0.000
0.015
0.003
0.000
0.018
0.003
0.000
0.023
0.005
0.000
p-value
<.001
<.001
0.755
<.001
<.001
0.984
<.001
<.001
0.050
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
0.001
-2 Log
Likelihood
153630


1 96252


250741


269808


300449


361128


Variance Components
Random Effects
a,S = 0.779
a212 = -0.070
a222 = 0.020
a,,2= 0.636
a?12 = -0.056
a22a= 0.016
0,^ = 0.332
a?12= -0.022
a2/ = 0.010
a,,2= 0.448
0^2 = -0.008
a222 = 0.008
a,,2 = 0.526
a212= -0.019
cr2/= 0.010
a,,2= 0.547
ap12= -0.024
a22^ = 0.012
&ror
aerTOr^ = 0.474
CTelTOr2 = 38.339




                                                  Page A-63
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
    100
     80
  _  60
  ง
  0)
  a.
     40
     20
                                    TRI Total Lead-Water Surface
                                                            8-
                                                         TJ
                                                         n
                                                         
                                                            o-
                                                               Legend:
                    • 1995-2000
                   — 2002-2004
2000-2002
2004-2006
        600  9000 17400 25800 34200 42600 51000 59400
                    Lead Total-Water Surface
      r "  '"i1"  ' "i'"               '"i	i'"
     0    10000  20000  30000 40000 50000 60000  70000

                 Lead Total-Water Surface
Figure A.63.   TRI Total Lead (Water Surface): Histogram and Linear Relationship with Geometric Mean Blood
               Lead Levels by Time
Table A.63a. Summary Information for TRI Total Lead (Water Surface) by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
123.96
90.22
83.10
78.01
95.71
Standard
Error
14.91
13.43
11.94
11.79
6.71
Minimum
0.0
0.0
0.0
0.0
0.0
10th
Percentile
0.0
0.0
0.0
0.0
0.0
25th
Percentile
0.0
0.0
0.0
0.0
0.0
Median
0.0
0.0
0.0
0.0
0.0
75th
Percentile
2.5
1.7
1.6
1.5
1.8
90th
Percentile
71.4
49.7
51.3
47.8
55.7
Maximum
66686.8
66686.8
66686.8
66686.8
66686.8
Table A.63b.   Model Information for the Relationship between TRI Total Lead (Water Surface) and Geometric
               Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
water surf
Intercept
time
water_surf
Intercept
time
water surf



Intercept
time
water surf
Intercept
time
water_surf
Estimate
3.418
-0.155
0.000
3.401
-0.156
0.000
-1.491
-0.146
0.000



-4.459
-0.161
0.000
-5.935
-0.140
0.000
Standard
Error
0.018
0.003
0.000
0.018
0.003
0.000
0.012
0.002
0.000



0.018
0.003
0.000
0.023
0.005
0.000
p- value
<.001
<.001
0.774
<.001
<.001
0.876
<.001
<.001
0.212



<.001
<.001
0.017
<.001
<.001
0.020
-2 Log
Likelihood
1 53628


196250


250739





300363


360919


Variance Components
Random Effects
an2= 0.779
a?12 = -0.070
cr2/ = 0.020
o,,2= 0.636
c?12= -0.056
a22^= 0.016
o^2 = 0.332
a?]2 = -0.022
a 22* = 0.010



a,,2= 0.533
a?12= -0.020
cr222= 0.010
a,,2= 0.554
3,,* = -0.025
<722'!= 0.012
Error
CTe.o,2= O-474
ฐenor2= 38.339




                                                   Page A-64
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
    100
     80
  „  60
  d>
  Q.
     40
     20
                                 TRI Total Lead - Water by Injection
                                                            8
                                                         s
                                                         CO
                                                         c
I  ^


I
1
tป  2

1
ex
                                                               Legend:    — 1995-2000
                                                                         	2002-2004
                                     - 2000-2002
                                   	2004-2006
              40000  80000  120000  160000  200000  240000

                 Lead Total-Underwater Injection
                 100000        200000

              Lead Total-Underwater Injection
300000
Figure A.64.   TRI Total Lead (Water by Injection): Histogram and Linear Relationship with Geometric Mean
               Blood Lead Levels by Time

Table A.64a.   Summary Information for TRI Total Lead (Water by Injection) by Time
Time
Period
1995-2000
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
20286
14434
16235
16340
67295
Number
Missing
0
0
0
0
0
Mean
38.07
119.26
151.41
150.72
110.18
Standard
Error
17.36
39.12
43.54
43.26
17.85
Minimum
0.0
0.0
0.0
0.0
0.0
10th
Percentile
0.0
0.0
0.0
0.0
0.0
25th
Percentile
0.0
0.0
0.0
0.0
0.0
Median
0.0
0.0
0.0
0.0
0.0
75th
Percentile
0.0
0.0
0.0
0.0
0.0
90th
Percentile
0.0
0.0
0.0
0.0
0.0
Maximum
245603.4
245603.4
245603.4
245603.4
245603.4
Table A.64b.   Model Information for the Relationship between TRI Total Lead (Water by Injection) and
               Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
underjnj
Intercept
time
underjnj
Intercept
time
underjnj
Intercept
time
underjnj
Intercept
time
underjnj
Intercept
time
underjnj
Estimate
3.418
-0.155
0.000
3.401
-0.156
0.000
-1.491
-0.146
0.000
-3.311
-0.169
0.000
-4.459
-0.161
0.000
-5.936
-0.140
0.000
Standard
Error
0.018
0.003
0.000
0.018
0.003
0.000
0.013
0.002
0.000
0.015
0.003
0.000
0.018
0.003
0.000
0.023
0.005
0.000
p-value
<.001
<001
0.689
<.001
<.001
0.652
<.001
<.001
0.954
<.001
<.001
0.932
<.001
<.001
0.661
<.001
<001
0.555
-2 Log
Likelihood
153632


1 96254


250743


269787


300332


360904


Variance Components
Random Effects
CT,,2 = 0.779
a?12 = -0.070
a2/ = 0.020
a,,2 = 0.636
a212 = -0.056
a222 = 0.016
a,,2= 0.333
a212= -0.022
CT2/ = 0.010
a112= 0.454
a?12= -0.008
CT2/= 0.008
a112= 0.535
a212= -0.020
CT2/= 0.010
an2= 0.555
a212= -0.025
a222= 0.012
Error
CTe™2= 0-474
aerror2= 38.339




                                                   Page A-65
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
        30-
        20-
     LLI
     o
     EC
     LJJ
     0.
        10-
EPA Region
             8-
                                                         T3
                                                          n
                                                            6:
                                                         m
                                                          c
                                                          ra
          I  4

          o
                                                         (3
                                                         I
                                                         •
                                                         0)
                                                            o-l
                                                                 Legend:
                                   1
                                   4
                                   7
                                  10
                                                             1994
                                                                     1996
                                                                            1998
                                                10
                                                                                   2000

                                                                                   time
                                                                                           2002
                                                                                                  2004
                                                                                                         2006
Figure A.65.   EPA Region: Histogram and Linear Relationship with Geometric Mean Blood Lead Levels by Time

Table A.65a.   Model Information for the Relationship between EPA Region and Geometric Mean Blood Lead
               Levels
Model
Number
1
2
Factor
Intercept
time
regionl
region2
regions
region4
regions
region6
region?
regions
region9
regionl 0
Intercept
time
regionl
region2
regions
region4
regions
regione
region/
regions
regions
regionl 0
Estimate
2.421
-0.157
1.380
1.069
0.820
0.982
0.568
1.000
1.814
0.298
1.862
0.000
2.383
-0.158
1.389
1.085
0.854
0.988
0.649
1.016
1.874
0.276
1.174
0.000
Standard
Error
0.097
0.003
0.121
0.150
0.103
0.100
0.100
0.102
0.102
0.125
0.128

0.112
0.003
0.131
0.153
0.117
0.114
0.114
0.116
0.116
0.149
0.142

p- value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
0.017
<.001

<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
0.064
<.001

-2 Log
Likelihood
152618











195377











Variance Components
Random Effects
0-,^= 0.623
a212= -0.075
a222= 0.020









a,f= 0.494
Q212= -0.057
Q222= 0.016









Error
CTerror2 = ฐ'474
ฐe™?= 38.297
                                                   Page A-66
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
Table A.65a.  Model Information for the Relationship between EPA Region and Geometric Mean Blood
               Levels (Continued)
Lead
Model
Number
3
4
5
6
Factor
Intercept
time
regionl
region2
regions
region4
regions
regions
region?
regions
regions
regionIO
Intercept
time
regionl
region2
regions
region*
regions
regions
region?
regions
region9
regionl 0
Intercept
time
regionl
region2
regions
region4
regions
regions
region?
regions
regions
regionl 0
Intercept
time
regionl ,
region2
regions
region4
regions
regions
region?
regions
regions
regionIO
Estimate
-2.257
-0.146
0.969
0.430
0.762
0.809
0.572
0.895
0.967
0.293
0.860
0.000
-3.910
-0.169
0.964
0.383
0.674
0.441
0.533
0.544
0.341
-0.034
1.392
0.000
-4.836
-0.162
0.744
0.340
0.545
0.141
0.383
0.220
0.694
-0.253
1.303
0.000
-5.993
-0.142
0.413
0.085
0.319
-0.229
0.128
-0.253
0.330
-0.475
0.882
0.000
Standard
Error
0.093
0.002
0.112
0.134
O.O98
0.095
0.096
0.097
0.097
0.118
0.118
.
0.136
0.003
0.157
0.183
0.142 ,
0.139
0.139
0.141
0.141
0.174
0.166

0.167
0.003
0.186
0.211
0.173
0.170
0.170
0.172
0.172
0.219
0.198
.
0.223
0.005
0.240
0.261
0.230
0.227
0.226
0.230
0.229
0.304
0.257
•
p- value
<.001
<.001
<.001
0.001
<.001
<.001
<.001
<.001
<.001
0.013
<.001

<.001
<.001
<.001
0.037
<.001
0.002
<.001
<.001
<.001
0.844
<.001

<.001
<.001
<.001
0.107
0.002
0.405
0.024
0.200
<.001
0.249
<.001
.
<.001
<.001
0.085
0.745
0.166
0.313
0.572
0.271
0.150
0.118
<.001
•
-2 Log
Likelihood
250539
.
.
.
.
.
.
.
.
.
.
.
269706
.
.
.
.
.
.


.
.

300842
.


.

.
.

.
.
.
362973
.
.
.
.


.
.

.
.
Variance Components
Random Effects
a,,2 = 0.311
o,,2 = -0.024
a^= 0.010









0,,*= 0.416
o212= -0.012
a^= 0.008









0,,"= 0.476
o,,2= -0.021
0^= 0.010









o,,2= 0.486
os,2= -0.025
oas2= 0.012









Error

i


                                                    Page A-67
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
.guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                   Appendix B



Massachusetts Data: Exploratory Analysis Summary Pages

-------
    10
Median Family Income ($)
                    o -
                       Legend:
                                                     1
                                                      o
                                                     _o
                                                     CD

                                                      RJ
                                                      I
                                                     O
                                                     I
                                                     S 2
                                                        o^
                                                                        - 2000-2002
                                                                    	2005-2006
2003-2004
      14000  42000  70000  98000 126000  154000 182000

                  Median Family Income ($)
                              50000     100000    150000

                                 Median Family Income ($)
      200000
Figure B.I.   Median Family Income ($): Histogram and Linear Relationship with Geometric Mean
              Blood Lead Levels by Time

Table B.la.   Summary Information for Median Family Income ($) by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
62118.2
62126.9
62160.6
62132.8
Standard
Error
204.5
249.9
250.7
133.8
Minim um
12354
12354
12354
12354
10th
Percentile
32215
32255
32215
32255
25th
Percentile
45000
45000
45000
45000
Median
59717
59717
59821
59717
75th
Percentile
74091
74135
74358
74135
90th
Percentile
93772
93774.5
93777
93777
Maximum
191062
191062
191062
191062
Table B.lb.   Model Information for the Relationship between Median Family Income ($) and Geometric
              Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
Median_FamilyJncome
Intercept
time
Median_Family_lncome
Intercept
time
Median_Family_lncome
Intercept
time
Median_Family_lncome
Intercept
time
Median_Family_lncome
Estimate
2.647
-0.084
0.000
2.634
-0.082
0.000
-1.739
-0.139
0.000
-3.641
-0.129
0.000
-4.939
-0.098
0.000
Standard
Error
0.016
0.002
0.000
0.016
0.002
0.000
0.015
0.002
0.000
0.022
0.004
0.000
0.030
0.007
0.000
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
51727


48155


86502


139627


178071


Variance Components
Random Effects
an2= 0.306
a212 = -0.028
a222= 0.004
0,,* = 0.299
a212= -0.027
o222= 0.004
on2= 0.246
a212= -0.009
a222= 0.004
a,,2= 0.362
a212= -0.018
a222= 0.004
a,,2= 0.344
a212 = -0.023
a222= 0.003
Error
"error2 = 0.207
"error2 = 4.429



                                                PageB-1
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
I
9
ฐ-4l
                                  Median Household Income ($)

                                                        8-
                                                     •o
                                                     s
                                                     E  6
                                                     o
                                                     m

                                                     a
                                                     0)
                                                     •S  4
                                                     I

                                                     O
                                                     
                                                     C3
                                                     •a
                                                     •
                                                     i
                                                        0-
                                                         Legend:
                                                                         2000-2002
                                                                         2005-2006
2003-2004
     10500   34500  58500  82500 106500 130500  154500

                Median Household Income ($)
                                                                50000     100000     150000    200000

                                                                  Median Household Income ($)
Figure B.2.   Median Household Income ($): Histogram and Linear Relationship with Geometric Mean
              Blood Lead Levels by Time

Table B.2a.   Summary Information for Median Household Income ($) by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
52858.9
52859.9
52873.3
52863.3
Standard
Error
175.5
214.1
214.8
114.7
Minimum
9895
9895
9895
9895
10th
Percentile
27520
27540
27540
27520
25th
Percentile
37719
37722
37722
37720.5
Median
50505
50485
50505
50505
75th
Percentile
63488
63488
63488
63488
90th
Percentile
80818
80818
80818
80818
Maximum
157189
157189
157189
157189
Table B.2b.   Model Information for the Relationship between Median Household Income ($) and
              Geometric Mean Blood Lead Levels
Model
Num ber
1
2
3
4
5
Factor
Intercept
time
Medlan_HH_lncome
Intercept
time
Median_HH_lncome
Intercept
time
Median_HH_lncome
Intercept
time
Median_HH_lncome
Intercept
time
Median_HHJncome
Estimate
2.645
-0.084
0.000
2.633
-0.082
0.000
-1.739
-0.139
0.000
-3.638
-0.129
0.000
-4.936
-0.098
0.000
Standard
Error
0.016
0.002
0.000
0.015
0.002
0.000
0.015
0.002
0.000
0.022
0.004
0.000
0.030
0.007
0.000
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
51689


48115


86375


139433


177920


Variance Components
Random Effects
a,,2= 0.296
a212= -0.027
a222= 0.004
a,,2 = 0.289
a212 = -0.026
a222 = 0.004
a,*= 0.235
0212= -0.010
a222 = 0.004
an2= 0.346
a212= -0.018
a222= 0.004
a,,2= 0.333
a212= -0.023
a222= 0.003
Error
"error2 = ฐ.207
^error2 = 4.429



                                                Page B-2
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
    14
                                  Median Per Capita Income ($)
                                                         8-
                                                      a
                                                      
-------
    12
    10
  o>
  g  •
  0.
                           Percent Units with No Household Earnings
                                                        8-
                                                     •o
                                                     s
                                                      O
                                                      _o
                                                      m
1  4
i
o
i
C3
                                                     13
                                                      0)

                                                        2-
                                                        0-
                                                            Legend:
                   - 2000-2002
                	2005-2006
2003-2004
      0.75  9.75  18.75 27.75 36.75 45.75 54.75 63.75 72.75

                Percent No Household Earnings
          10    20    30   40   50    60    70

               Percent No Household Earnings
        80
Figure B.4.   Percent Units with No Household Earnings: Histogram and Linear Relationship with
              Geometric Mean Blood Lead Levels by Time

Table B.4a.   Summary Information for Percent Units with No Household Earnings by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sam pie
Size
15617
10450
10417
36484
Num ber
Missing
0
0
0
0
Mean
20.1
20.1
20.1
20.1
Standard
Error
0.1
0.1
0.1
0.0
Minim um
0.0
0.0
0.0
0.0
10th
Percentile
11.0
11.0
11.0
11.0
25th
Percentile
14.8
14.8
14.8
14.8
Median
18.8
18.8
18.8
18.8
75th
Percentile
24.2
24.2
24.2
24.2
90th
Percentile
30.5
30.4
30.5
30.5
Maxim um
64.4
74.5
74.5
74.5
Table B.4b.   Model Information for the Relationship between Percent Units with No Household
              Earnings and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
Pct__HH_No_Earnings
Intercept
time
Pct_HH_No_Earnings
Intercept
time
Pct_HH_No_Earnings
Intercept
time
Pct_HH_No_Earnings
Intercept
time
Pct_HH_No_Earnings
Estimate
2.647
-0.084
2.421
2.635
-0.082
2.505
-1.738
-0.139
3.018
-3.645
-0.126
3.022
-4.940
-0.095
2.557
Standard
Error
0.017
0.002
0.142
0.017
0.002
0.144
0.017
0.002
0.189
0.024
0.004
0.230
0.031
0.007
0.251
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
52020


48446


86853


139532


177347


Variance Components
Random Effects
on2= 0.349
C212= -0.028
o222 = 0.004
o,,2= 0.340
a212= -0.027
0222 = 0.004
a,,2= 0.322
a212= -0.008
a222 = 0.004
au2= 0.471
a212= -0.022
a222= 0.004
0^2= 0.429
a212= -0.027
a222= 0.003
Error
ฐe,ror2= 0.207
"error2 = *.429



                                                Page B-4
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                             Percent Units with No Household Wage
    10
  ~  6
  i
  o
  ha
  tD
  OL
                                  n i n.i
                                          n.n
                                                         8-
                                                      T3
                                                      CO
T5  B
O  6
_O
m

CO
0)
'w  ^ "
C)

o
CD
O
T3
0)
t5  2-
                                                         o-
                                                            Legend:       - 2000-2002
                                                                      	2005-2006
                                      2003-2004
      0.75  9.75 18.75 27.75 36.75 45.75 54.75 63.75 72.75

                 Percent No Household Wage
          10    20   30   40    50    60

                Percent No Household Wage
70
80
Figure B.5.   Percent Units with No Household Wage: Histogram and Linear Relationship with
              Geometric Mean Blood Lead Levels by Time

Table B.5a.   Summary Information for Percent Units with No Household Wage by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
22.5
22.5
22.5
22.5
Standard
Error
0.1
0.1
0.1
0.0
Minim um
0.0
0.0
0.0
0.0
10th
Percentile
13.4
13.4
13.4
13.4
25th
Percentile
17.4
17.4
17.4
17.4
Median
21.4
21.4
21.4
21.4
75th
Percentile
26.4
26.4
26.4
26.4
90th
Percentile
33.0
33.0
33.0
33.0
Maximum
65.0
74.5
74.5
74.5
Table B.5b.   Model Information for the Relationship between Percent Units with No Household Wage
              and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
Pct_HH_No_Wage
Intercept
time
Pct_HH_No_Wage
Intercept
time
Pct_HH_NoJA/age
Intercept
time
Pct_HH_No_Wage
Intercept
time
Pct_HH_No_Wage
Estimate
2.645
-0.084
2.390
2.633
-0.082
2.469
-1.738
-0.139
2.957
-3.645
-0.126
2.969
-4.939
-0.095
2.461
Standard
Error
0.017
0.002
0.146
0.017
0.002
0.149
0.017
0.002
0.195
0.024
0.004
0.238
0.031
0.007
0.261
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
52040


48467


86858


139460


177185


Variance Components
Random Effects
0^2 = 0.346
a212 = -0.027
a222= 0.004
a,,2= 0.338
a212= -0.026
a222 = 0.004
a,,2 = 0.326
0212= -0.008
o222= 0.004
0^2 = 0.476
a?12= -0.022
a222 = 0.004
o,,2= 0.437
a212= -0.027
o222 = 0.003
Error
ฐerrar2= ฐ-207
aerror2= 4.428



                                                Page B-5
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                             Percent Households on Public Assistance
                                                      •o
                                                       CD
                                                       (1)
                                                       o
                                                       m
                                                       o
                                                       0)
                                                       T3
                                                       0)
                                                         8-
                                                         6
                                                         2
                                                          o-
                                                             Legend:   	2000-2002
                                                                      	2005-2006
                                 2003-2004
       0.4  3.6  6.8  10 13.2 16.4 19.6 22.8  26 29.2 32.4

            Percent Household on Public Assistance
0
    1 i •        ' i '
    10        20         30

Percent Household on Public Assistance
                                          40
Figure B.6.   Percent Households on Public Assistance: Histogram and Linear Relationship with
              Geometric Mean Blood Lead Levels by Time
Table B.6a. Summary Information for Percent Households on Public Assistance by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Num ber
Missing
0
0
0
0
Mean
3.27
3.26
3.27
3.27
Standard
Error
0.03
0.04
0.04
0.02
Minimum
0.0
0.0
0.0
0.0
10th
Percentile
0.4
0.4
0.4
0.4
25th
Percentile
1.0
1.0
1.0
1.0
Median
1.9
1.9
1.9
1.9
75th
Percentile
3.9
3.9
3.9
3.9
90th
Percentile
8.4
8.3
8.4
8.4
Maximum
34.1
34.1
34.1
34.1
Table B.6b.   Model Information for the Relationship between Percent Households on Public Assistance
              and Geometric Mean Blood Lead Levels
Model
Num ber
1
2
3
4
5
Factor
Intercept
time
Pet HH Public Assist
Intercept
time
Pet HH Public Assist
Intercept
time
Pet HH Public Assist
Intercept
time
Pet HH Public Assist
Intercept
time
Pct_HH_Public_Assist
Estimate
2.651
-0.084
5.762
2.638
-0.083
5.876
-1.736
-0.139
7.156
-3.641
-0.129
7.326
-4.945
-0.097
6.285
Standard
Error
0.017
0.002
0.303
0.017
0.002
0.302
0.017
0.002
0.395
0.023
0.004
0.463
0.031
0.007
0.479
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
51964


48386


86854


139837


178218


Variance Components
Random Effects
a,,2= 0.349
a212= -0.029
a222= 0.004
on2 = 0.340
a212 = -0.028
C222= 0.004
a,,2= 0.303
a212 = -0.007
a222= 0.004
a,,2= 0.427
a212= -0.019
a222= 0.004
a,,2= 0.385
a2,2= -0.024
a222 = 0.003
Error
aerror2= 0.207
ฐe™2= 4-429



                                                 Page B-6
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                            Percent Households below Poverty Line
20.0 "


17.5 "
15.0 "


12.5 '
t .
c
0>
5 10.0 -
ex


7.5 '



5.0 '


2.5 '
0J
















-



,
/


r
0














_



f
r




6













































-



1

















~v


-



2
























i


















\

























X

~


1



















\
\
\
JJTT_V

8 24 30 36 42 48 54 60
8 1

;
CO -
o 6:
o "
m
CO I
0)
s

~ 4 -
CD
ฃ
o
CD
O
•a
CD -
o 2 ~
1 ;
Q.

o:


Legend: 2000-2002 ~ 2003-2004
	 2005-2006









^ — • 	 ^
^ 	 "~~_^^ ---""
__f*—~r~ — ' ' --"'""'" — 	
_l^-^— — ^~ T* '-~~>~l^~-'~~"f~~~
_-r^-^~r^~^*^~- ~f"~~^.-r'-''~"~
-~ 	 ~rJ~^^\-'-r ~~~'_r -'*""""''"
_--••*"' ~ j- 	 ~*"^^
r*-^ *"'




0 10 20 30 40 50 60 70
                  Percent Below Poverty Line
                                                                    Percent Below Poverty Line
Figure B.7.   Percent Households below Poverty Line: Histogram and Linear Relationship with
              Geometric Mean Blood Lead Levels by Time

Table B.7a.   Summary Information for Percent Households below Poverty Line by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
10.40
10.40
10.41
10.40
Standard
Error
0.08
0.10
0.10
0.05
Minimum
0.0
0.0
0.0
0.0
10th
Percentile
2.4
2.4
2.4
2.4
25th
Percentile
3.7
3.7
3.7
3.7
Median
6.5
6.5
6.5
6.5
75th
Percentile
13.7
13.7
13.7
13.7
90th
Percentile
25.4
25.4
25.4
25.4
Maximum
63.3
63.3
63.3
63.3
Table B.7b.   Model Information for the Relationship between Percent Households below Poverty Line
              and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
Pct_LT_Poverty
Intercept
time
Pct_LT_Poverty
Intercept
time
Pct_LT_Poverty
Intercept
time
Pct_LT_Poverty
Intercept
time
Pct_LT_Poverty
Estimate
2.647
-0.084
2.169
2.635
-0.082
2.248
-1.736
-0.139
2.777
-3.639
-0.128
2.990
-4.938
-0.097
2.643
Standard
Error
0.017
0.002
0.116
0.017
0.002
0.116
0.016
0.002
0.151
0.023
0.004
0.178
0.030
0.007
0.188
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
51974


48390


86778


139558


177839


Variance Components
Random Effects
an2= 0.343
a212= -0.028
a222= 0.004
0,^= 0.335
a212= -0.027
a222= 0.004
a,,2= 0.298
a212= -0.007
a222= 0.004
0,^= 0.415
a212= -0.019
o222= 0.004
o,,2= 0.369
o212= -0.023
0222= 0.003
Error
0error2= 0.207
ฐ,,J= 4.430



                                               Page B-7
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                     Percent Units with Family Income below Poverty Line
    20.0


    17.5


    15.0


    12.5
  a>
  g 10.0
  Q_
     7.5


     5.0


     2.5
                                T" ' '-r^'^r "-• i-
             6  12  18  24  30  36 42   48  54
            Percent Family Income Below Poverty Line
                                                        8-
      •a
      o
      o
      CO
      I 4
      i
      i
      o
      1
      T3 2
      '•5
      o>
                                                         o-
             Legend:   	2000-2002
                      	2005-2006
2003-2004
60
                 10     20     30     40     50     60

                 Percent Family Income Below Poverty Line
        70
Figure B.8.   Percent Units with Family Income below Poverty Line: Histogram and Linear Relationship
              with Percent Units with Geometric Mean Blood Lead Levels by Time
Table B.8a. Summary Information for Percent Units with Family Income below Poverty Line by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Num ber
Missing
0
0
0
0
Mean
8.16
8.16
8.16
8.16
Standard
Error
0.08
0.09
0.09
0.05
Minimum
0.0
0.0
0.0
0.0
10th
Percentile
1.2
1.2
1.2
1.2
25th
Percentile
2.1
2.2
2.1
2.1
Median
4.4
4.4
4.4
4.4
75th
Percentile
10.2
10.2
10.2
10.2
90th
Percentile
21.6
21.5
21.6
21.6
Maxim um
61.6
61.6
61.6
61.6
Table B.8b.   Model Information for the Relationship between Percent Units with Family Income below
              Poverty Line and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
Pct_Family_lncome_LT
Intercept
time
Pct_Family_lncome_LT
Intercept
time
Pct_Family_lncome_LT
Intercept
time
Pct_FamilyJncome_LT
Intercept
time
Pct_FamilyJncome_LT
Estimate
2.649
-0.084
2.247
2.637
-0.083
2.309
-1.736
-0.139
2.781
-3.641
-0.128
2.955
-4.942
-0.097
2.588
Standard
Error
0.017
0.002
0.124
0.017
0.002
0.124
0.017
0.002
0.162
0.023
0.004
0.190
0.031
0.007
0.200
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
51991


48412


86847


139686


177952


Variance Components
Random Effects
a^= 0.351
a212 = -0.028
o2/ = 0.004
0,^= 0.342
0212 = -0.028
a222= 0.004
a,,2= 0.307
a212 = -0.007
a222= 0.004
0n2 = 0.430
a212 = -0.019
0222= 0.004
o,,2= 0.383
a212= -0.024
a222 = 0.003
Error
ฐeTO,2 = 0-207
aerror2 = 4.429



                                                Page B-8
This information is distributed solely for the purpose of'pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                    Percent Units Spending Less than Five Years in Poverty
30




25


20 '

c
CD
a



10 '




5 "

0

















,





,



i-













'






































































\
\
\
\
\
\
TfliK
lmrtw_
8:

•
•D
RJ
a>
o 6 -
o
CD
1 :
s

•5 .
E
o
0)
O
IS :
o 2;
'•5
0)

a -
o-

Legend: 2000-2002 	 2003-2004
	 2005-2006







,
_^'~ " 	 -
-~^rr^--~"~
^,^-^tr^-""^"^ -"""""'^
^ — ^^~ " *" *^ "
^ — T'-*""*'" ^- —
^^-^^T^lT^ " " "
j.J.-''^^-p'"^
~"~





           1.6  3.2 4.8  6.4  8  9.6 11.212.814.4  16

              Percent Less than 5 Years in Poverty
  4    6    8    10   12   14   16

Percent Less than 5 Years in Poverty
18
Figure B.9.   Percent Units Spending Less than Five Years in Poverty: Histogram and Linear
              Relationship with Geometric Mean Blood Lead Levels by Time

Table B.9a.   Summary Information for Percent Units Spending Less than Five Years in Poverty by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
1.10
1.10
1.10
1.10
Standard
Error
0.01
0.02
0.02
0.01
Minim um
0.0
0.0
0.0
0.0
10th
Percentile
0.0
0.0
0.0
0.0
25th
Percentile
0.2
0.2
0.2
0.2
Median
0.5
0.5
0.5
0.5
75th
Percentile
1.3
1.3
1.3
1.3
90th
Percentile
3.2
3.2
3.1
3.2
Maxim um
12.7
17.3
17.3
17.3
Table B.9b.   Model Information for the Relationship between Percent Units Spending Less than Five
              Years in Poverty and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
Pct_LE_5Yrs_LT_Pover
Intercept
time
Pct_LE_5Y rs_LT_Pover
Intercept
time
Pct_LE_5Yrs_LT_Pover



Intercept
time
Pct_LE_5Yrs_LT_Pover
Estimate
2.650
-0.084
12.221
2.637
-0.083
12.722
-1.737
-0.139
14.741



-4.945
-0.097
13.343
Standard
Error
0.017
0.002
0.726
0.017
0.002
0.733
0.017
0.002
0.962



0.031
0.007
1.169
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001



<.001
<.001
<.001
-2 Log
Likelihood
52025


48446


86925





178034


Variance Components
Random Effects
0,^= 0.361
0212= -0.029
o222= 0.004
a,,^ 0.353
02]2= -0.028
a222= 0.004
o,,2= 0.323
a212= -0.008
a222= 0.004



a,,2= 0.404
a212= -0.025
a222= 0.003
Error
ฐerror2= 0.207
<>erra,2= 4.429



                                               Page B-9
This information is distributed solely for the purpose of pre-dissetninat ion peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                     Percent American Indian and Alaskan Native Alone
60 '



50


40


4— i
I
lao
0.



20




10




















^



0



















-v
























\
\
IK*W_ _
0.6 1.2 1.8 2.4 3 3.6 4.2 4.8
8:

;
(0
4)
•O c I
0 b
o
m
c
n
0)
5
'C 4 "
'3
E
o
o> :
TO •
0)
o 2:
'•5
0)
t-
CL
o-


Legend: 2000-2002 " ™ 2003-2004
	 2005-2006










	 —
	 ~
^ 	 	 --*"
^ — • " 	 	 	 — ^_~ 	 — —
•~~^^--'^-~"^---~' 	
IT--- •*• -^ ~ ~ ~ ~~ '




\ 	 i 	 i 	 ' ' i ' ' ' ' ' ' ' ' ' i • ' • 	 'i 	 '"i
0123456
         Percent Amer. Indian and Alaskan Native Alone
                                                             Percent Amer. Indian and Alaskan Native Alone
Figure B.10. Percent American Indian and Alaskan Native Alone: Histogram and Linear Relationship
             with Geometric Mean Blood Lead Levels by Time

Table B.lOa. Summary Information for Percent American Indian and Alaskan Native Alone by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sam pie
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
0.19
0.19
0.19
0.19
Standard
Error
0.00
0.00
0.00
0.00
Minimum
0.0
0.0
0.0
0.0
10th
Percentile
0.0
0.0
0.0
0.0
25th
Percentile
0.0
0.0
0.0
0.0
Median
0.0
0.0
0.0
0.0
75th
Percentile
0.3
0.3
0.3
0.3
90th
Percentile
0.5
0.5
0.5
0.5
Maximum
5.0
5.0
5.0
5.0
Table B.lOb.  Model Information for the Relationship between Percent American Indian and Alaskan
              Native Alone and Geometric Mean Blood Lead Levels
Model
Num ber
1
2
3
4
5
Factor
Intercept
time
PC t_ A IAN A
Intercept
time
Pct_AIANA
Intercept
time
Pct_AIANA
Intercept
time
Pct_AIANA
Intercept
time
Pct_AIANA
Estimate
2.648
-0,084
15.931
2.635
-0.082
16.154
-1.735
-0.140
22.164
-3.646
-0.126
21.553
-4.944
-0.094
15.435
Standard
Error
0.018
0.002
3.824
0.017
0.002
3.881
0.018
0.002
5.027
0.025
0.004
5.978
0.032
0.007
6.367
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
0.015
-2 Log
Likelihood
52259


48692


87121


139740


177392


Variance Components
Random Effects
a,,2= 0.381
a212= -0.027
a222= 0.004
a,,2= 0.373
a212= -0.026
a22J = 0.004
o,,2= 0.367
a212= -0.006
a222 = 0.004
a,,2= 0.532
a212= -0.022
a222= 0.004
a.,2= 0.482
0^2 = -0.028
a222= 0.003
Error
"error2 = 0.207
ฐerror2 = 4.428



                                               Page B-10
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
    50
    40
  ^ 30

    20
    10
                                         Percent Asian Alone
                                                         8
                                                       _o
                                                       CD
i
o
"S
'•5
0)
                                                             Legend:       - 2000-2002
                                                                      	2005-2006
                                      2003-2004
       0.75  9.75  18.75 27.75 36.75  45.75 54.75 63.75

                     Percent Asian Alone
          10    20   30    40    50

                   Percent Asian Alone
60
     70
           80
Figure B.ll.  Percent Asian Alone: Histogram and Linear Relationship with Geometric Mean Blood
              Lead Levels by Time

Table B.I la.  Summary Information for Percent Asian Alone by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sam pie
Size
15617
10450
10417
36484
Num ber
Missing
0
0
0
0
Mean
3.76
3.76
3.77
3.76
Standard
Error
0.05
0.05
0.06
0.03
Minim um
0.0
0.0
0.0
0.0
10th
Percentile
0.1
0.1
0.1
0.1
25th
Percentile
0.6
0.6
0.6
0.6
Median
1.7
1.7
1.7
1.7
75th
Percentile
4.7
4.7
4.8
4.7
90th
Percentile
10.0
10.1
10.1
10.0
Maximum
71.6
71.6
71.6
71.6
Table B.llb. Model Information for the Relationship between Percent Asian Alone and Geometric Mean
              Blood Lead Levels
Model
Num ber
1
2
3
4
5
Factor
Intercept
time
Pct_Asian
Intercept
time
Pct_Asian
Intercept
time
Pct_Asian



Intercept
time
Pct_Asian
Estimate
2.649
-0.084
-0.979
2.636
-0.082
-0.943
-1.736
-0.140
-0.820



-4.945
-0.094
0.030
Standard
Error
0.018
0.002
0.230
0.018
0.002
0.233
0.018
0.002
0.304



0.032
0.007
0.393
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
0.007



<.001
<.001
0.939
-2 Log
Likelihood
52264


48699


87140





177385


Variance Components
Random Effects
0,^ = 0.388
a212= -0.028
o222 = 0.004
0,,?= 0.380
a212 = -0.027
a222= 0.004
a,,2= 0.376
a212= -0.006
a222= 0.004



a,,2 = 0.488
0212= -0.029
o222= 0.003
Error
v*,J= 0.207
^error2 = 4.428



                                                Page B-11
This information is distributed solely for the purpose ofpre- dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                        Percent Black Alone
fin "
\j\j


50 '



40


*j
ง
5 30
0.



20

10
0

-,


















.. •"
1




















fr^_
7 13 19 25 31 37 43 49 55 61 67 73 79 85 91
8:


•o
ซ ':
•a R '.
0 6 .
o
m
c
(0
5 -
o .-_
"C
B

E
o
V
O
1 =
ฃ I
o-


Legend: 2000-2002 2003-2004
	 2005-2006











^ 	
	 . — - — ' 	
^ 	 ___ — '~~~ 	
^ — • — ' 	 	 --~ 	 _- ~- • "
rrr^n^---1


0 1 0 20 30 40 50 60 70 80 90 10
                    Percent Black Alone
                                                                        Percent Black Alone
Figure B.12.  Percent Black Alone: Histogram and Linear Relationship with Geometric Mean Blood
              Lead Levels by Time

Table B.12a.  Summary Information for Percent Black Alone by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sam pie
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
5.59
5.58
5.59
5.59
Standard
Error
0.10
0.12
0.12
0.06
Minim um
0.0
0.0
0.0
0.0
10th
Percentile
0.1
0.1
0.1
0.1
25th
Percentile
0.5
0.5
0.5
0.5
Median
1.5
1.5
1.5
1.5
75th
Percentile
4.5
4.5
4.5
4.5
90th
Percentile
13.1
13.1
13.1
13.1
Maximum
91.0
91.0
91.0
91.0
Table B.12b.  Model Information for the Relationship between Percent Black Alone and Geometric Mean
              Blood Lead Levels
Model
Num ber
1
2
3
4
5
Factor
Intercept
time
Pct_Black
intercept
time
Pct_Black
Intercept
time
Pct_Black
Intercept
time
Pct_Black
Intercept
time
Pct_Black
Estimate
2.649
-0.084
1.119
2.636
-0.082
1.141
-1.735
-0.140
1.364
-3.642
-0.128
1.365
-4.943
-0.096
1.189
Standard
Error
0.017
0.002
0.103
0.017
0.002
0.103
0.017
0.002
0.132
0.024
0.004
0.152
0.031
0.007
0.154
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
52170


48599


87051


139781


177691


Variance Components
Random Effects
0^2= 0.362
0212= -0.027
0222= 0.004
o,,2= 0.353
02,2= -0.026
o222= 0.004
0,^= 0.342
0212= -0.005
a222= 0.004
0,^= 0.489
a?12= -0.019
a222= 0.004
0-^2= 0.440
a212= -0.026
o222= 0.003
Error
ae,,0,2= 0.207
ฐer,or2 = 4'429



                                                Page B-12
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
    17.5
    15.0
    12.5
  0)
  a
  0)
  0.
    10.0
                                         Percent White Alone
                                                          8-
     7.5
TJ
CO

-------
                  Percent Native Hawaiian and Other Pacific Islander Alone
    100
     80
  _  60
  ง
  4)
  Q.
     40
     20
                                                        8 I
TJ
en
4)

E  6
o
m
CD
ซ
o
o>
O
+*  o -
o  *~
0)
a
                                                        0
                                                            Legend:
                     2000-2002
                     2005-2006
2003-2004
            0.15  0.3  0.45  0.6 0.75  0.9  1.05  1.2  1.35

                         Pet NHOPI
    0.0    0.2    0.4    0.6    0.8    1.0    1.2    1.4

    Percent Native Hawaiian and Other Pacific Islander Alone
Figure B.14.  Percent Native Hawaiian and Other Pacific Islander Alone: Histogram and Linear
              Relationship with Geometric Mean Blood Lead Levels by Time

Table B.14a.  Summary Information for Percent Native Hawaiian and Other Pacific Islander Alone by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
0.02
0.02
0.02
0.02
Standard
Error
0.00
0.00
0.00
0.00
Minim um
0.0
0.0
0.0
0.0
10th
Percentile
0.0
0.0
0.0
0.0
25th
Percentile
0.0
0.0
0.0
0.0
Median
0.0
0.0
0.0
0.0
75th
Percentile
0.0
0.0
0.0
0.0
90th
Percentile
0.0
0.0
0.0
0.0
Maximum
1.4
1.4
1.4
1.4
Table B.14b.  Model Information for the Relationship between Percent Native Hawaiian and Other
              Pacific Islander Alone and Geometric Mean Blood Lead Levels
Model
Num ber
1
2
3
4
5
Factor
Intercept
time
Pct_NHOPI
Intercept
time
Pct_NHOPI
Intercept
time
Pct_NHOPI
Intercept
time
Pct_NHOPI
Intercept
time
Pct_NHOPI
Estimate
2.649
-0.084
7.425
2.636
-0.082
8.036
-1.735
-0.140
-0.786
-3.647
-0.126
3.538
-4.946
-0.094
-6.263
Standard
Error
0.018
0.002
14.698
0.018
0.002
14.811
0.018
0.002
19.257
0.025
0.004
22.631
0.032
0.007
24.829
p-value
<.001
<.001
0.613
<.001
<.001
0.587
<.001
<.001
0.967
<.001
<.001
0.876
<.001
<.001
0.801
-2 Log
Likelihood
52273


48707


87136


139740


177379


Variance Components
Random Effects
a,,2= 0.387
0212= -0.027
a222 = 0.004
0^2 = 0.379
a212 = -0.026
a222 = 0.004
an2= 0.374
0212= -0.006
o222= 0.004
0^2 = 0.541
a212= -0.022
a222= 0.004
a,,2= 0.488
a212 = -0.029
a222= 0.003
Error
aerror2= 0.207
"error2 = ".428



                                                Page B-14
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
    90

    80


    70

    60

    50 1


    40

    30

    20

    10
                                     Percent Other Race Alone
                                                         8-
o
m
8
o
•-  4
E
o
a
C3
T3
O
   2
       Legend:       - 2000-2002
                	2005-2006
2003-2004
       0.4  3.6  6.8  10  13.216.419.622.8  26 29.232.4

                  Percent Other Race Alone
               10
                                              40
                                                                       Percent Other Race Alone
Figure B.15.  Percent Other Race Alone: Histogram and Linear Relationship with Geometric Mean
              Blood Lead Levels by Time

Table B.lSa.  Summary Information for Percent Other Race Alone by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
0.753
0.750
0.744
0.749
Standard
Error
0.019
0.023
0.023
0.012
Minim um
0.00
0.00
0.00
0.00
10th
Percentile
0.00
0.00
0.00
0.00
25th
Percentile
0.00
0.00
0.00
0.00
Median
0.14
0.14
0.14
0.14
75th
Percentile
0.59
0.59
0.59
0.59
90th
Percentile
1.45
1.45
1.41
1.45
Maximum
35.01
35.01
35.01
35.01
Table B.lSb.  Model Information for the Relationship between Percent Other Race Alone and Geometric
              Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
Pct_Other_Race
Intercept
time
Pct_Other_Race
Intercept
time
Pct_Other_Race
Intercept
time
Pct_Other_Race
Intercept
time
Pct_Other_Race
Estimate
2.648
-0.084
4.852
2.636
-0.082
4.982
-1.734
-0.140
6.528
-3.641
-0.128
7.799
-4.936
-0.097
6.985
Standard
Error
0.017
0.002
0.538
0.017
0.002
0.543
0.017
0.002
0.690
0.024
0.004
0.769
0.031
0.007
0.747
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
52203


48633


87055


139688


177606


Variance Components
Random Effects
an2 = 0.362
a212= -0.026
a222 = 0.004
a,,2 = 0.353
a212= -0.025
a222 = 0.004
a,,2= 0.345
a212= -0.005
a222= 0.004
a,,2= 0.479
a212= -0.019
o222= 0.004
0-,,* = 0.410
a212= -0.023
a222= 0.003
Error
ฐerror2= 0.207
ฐ error2 = 4.429



                                                Page B-15
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA, It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                        Percent Multiple Races
    14
    12
    10
  c  8

  I
  0)
  0.
o
m
•8  4
s

i
o
T3

-------
    35
    30
                                           Percent Hispanic
                                                          8-
TJ
(tj
0)
                                                       oo

                                                       ง
                                                       o
                                                       0)
                                                       C3
                                                       
-------
                                       Percent Rented Units
  0)
  a









I
0




-


_
/


/
/'







-
/



/-^N
-i







i-.










6 12 18 24 30 36 4248





V
\
"



\



s
V
\
_. S




^k, j-
•s









-v









-









ffi
54 60 66 72 78 84 90 96
8;
•o
1
E 6
o
ffl
I
a
1 4
E
o
9)
O
% 2
TJ
Q.
0

Legend: 2000-2002 	 2003-2004
	 2005-2006





__-^r___H^
r--"^'''J'Jl~"


0 10 20 30 40 50 60 70 80 90 10
                      Percent Rented
                                                                         Percent Rented
Figure B.18.  Percent Rented Units: Histogram and Linear Relationship with Geometric Mean Blood
              Lead Levels by Time

Table B.18a.  Summary Information for Percent Rented Units by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
39.62
39.66
39.65
39.64
Standard
Error
0.20
0.25
0.25
0.13
Minimum
0.0
0.0
0.0
0.0
10th
Percentile
10.0
10.0
10.0
10.0
25th
Percentile
17.9
18.0
18.0
18.0
Median
34.8
34.9
34.8
34.8
75th
Percentile
59.9
59.9
59.9
59.9
90th
Percentile
75.7
75.7
75.7
75.7
Maximum
100.0
100.0
100.0
100.0
Table B.18b.  Model Information for the Relationship between Percent Rented Units and Geometric
              Mean Blood Lead Levels
Model
Num ber
1
2
3
4
5
Factor
Intercept
time
Pct_Rented
Intercept
time
Pct_Rented
Intercept
time
Pet Rented
Intercept
time
PcLRented
Intercept
time
Pct_Rented
Estimate
2.647
-0.084
0.827
2.634
-0.082
0.859
-1.734
-0.140
1.222
-3.634
-0.130
1.402
-4.934
-0.098
1.283
Standard
Error
0.016
0.002
0.047
0.016
0.002
0.047
0.016
0.002
0.059
0.022
0.004
0.070
0.030
0.007
0.078
p-value
<.001
<.001
<.001
<.001
<.001
<.001
•e.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
52007


48421


86682


139427


177819


Variance Components
Random Effects
a^= 0.327
0212= -0.026
0222= 0.004
a,,2= 0.319
a212= -0.026
a222= 0.004
0^= 0.271
a212= -0.006
a222= 0.004
a,,2= 0.370
0^= -0.017
0222= 0.004
a,,2= 0.337
a212= -0.022
o222= 0.003
Error
aerror2= 0.207
aerror2= 4.430



                                               Page B-18
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                           Median Rent ($)
                                                          8-
                                                       T3
                                                       to
                                                       o
                                                       CD
                                                          6-
                                                       r
                                                       o
                                                       V
                                                       o
                                                       l.H
                                                       '•5
                                                       g
                                                          o-
                                                              Legend:       - 2000-2002
                                                                       	2005-2006
                     2003-2004
            240  480  720  960  1200  1440  1680  1920

                       Median Rent ($)
1000          2000

   Median Rent (S)
3000
Figure B.19.  Median Rent ($): Histogram and Linear Relationship with Geometric Mean Blood Lead
              Levels by Time
Table B.19a. Summary Information for Median Rent ($) by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sam pie
Size
15617
10450
10417
36484
Num ber
Missing
0
0
0
0
Mean
653.7
654.6
654.4
654.1
Standard
Error
1.8
2.3
2.3
1.2
Minim um
0
0
0
0
10th
Percentile
414
415
415
415
25th
Percentile
504
505
505
504
Median
619
619
619
619
75th
Percentile
755
755
755
755
90th
Percentile
931
935.5
933
933
Maxim um
2001
2001
2001
2001
Table B.19b. Model Information for the Relationship between Median Rent ($) and Geometric Mean
              Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
Median_Rent
Intercept
time
Median_Rent
Intercept
time
Median_Rent
Intercept
time
Median_Rent
Intercept
time
Median_Rent
Estimate
2.650
-0.084
-0.001
2.637
-0.083
-0.001
-1.739
-0.139
-0.001
-3.652
-0.126
-0.001
-4.952
-0.095
-0.001
Standard
Error
0.017
0.002
0.000
0.017
0.002
0.000
0.017
0.002
0.000
0.024
0.004
0.000
0.031
0.007
0.000
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
52004


48440


86943


139972


177952


Variance Components
Random Effects
0^2 = 0.361
0212= -0.029
a222 = 0.004
0^2= 0.352
0^2 = -0.028
a222 = 0.004
0^2= 0.324
a212= -0.010
a222 = 0.004
a,,2= 0.476
a212= -0.023
a222= 0.004
a,,2= 0.445
a212= -0.029
a222= 0.003
Error
aerror2= 0.207
ฐerror2 = ^.428



                                                 PageB-19
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
    14
                                          Housing Value ($)
                                                         8-

                                                      _o
                                                      m
                                                      i
                                                      i
                                                      o
                                                      0)
                                                      o
                                                      TJ
                                                      0)
                                                         6-
                                                         2-
                                                            Legend:     — 2000-2002
                                                                     	2005-2006
                     2003-2004
            160000  320000 480000 640000 800000 960000

                      Housing Value ($)
500000        1000000

   Housing Value ($)
1500000
Figure B.20.  Housing Value ($): Histogram and Linear Relationship with Geometric Mean Blood Lead
              Levels by Time

Table B.20a.  Summary Information for Housing Value ($) by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
203914.1
204453.3
204436.7
204217.7
Standard
Error
1043.4
1279.9
1281.5
683.9
Minimum
0
0
0
0
10th
Percentile
98200
98200
98500
98200
25th
Percentile
130000
130100
130200
130100
Median
172000
172100
172100
172100
75th
Percentile
233800
235100
235100
234600
90th
Percentile
341300
342600
342200
342200
Maximum
1000001
1000001
1000001
1000001
Table B.20b.  Model Information for the Relationship between Housing Value ($) and Geometric Mean
              Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
Housing_Value
Intercept
time
Housing. .Value
Intercept
time
Housing_Value
Intercept
time
Housing_Value
Intercept
time
Housing_Value
Estimate
2.652
-0.084
0.000
2.638
-0.083
0.000
-1.738
-0.139
0.000
-3.652
-0.126
0.000
-4.953
-0.095
0.000
Standard
Error
0.017
0.002
0.000
0.017
0.002
0.000
0.017
0.002
0.000
0.024
0.004
0.000
0.031
0.007
0.000
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
52095


48521


87033


140054


178202


Variance Components
Random Effects
a,,2= 0.374
a212= -0.030
a222 = 0.004
a,,2= 0.363
a212= -0.029
0222= 0.004
a1l2= 0.337
a212= -0.009
a222 = 0.004
a,,2 = 0.489
a2]2= -0.024
a222= 0.004
a,,2= 0.443
a212= -0.028
a222= 0.003
Error
ฐ,,,ar2= 0-207
aerror2= 4.429



                                                Page B-20
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
  40

  35

  30

  25

|2ปJ
o_

  15 '

  10
                                        Percent Vacant Units
                                                         8-
                                                      T5
                                                      CO
                                                      V
                                                      T3
                                                      O
                                                      CD
                                                      a
                                                      o>
                                                      •S  4
                                                      i

                                                      O
                                                      O
                                                      O
                                                         o-
                                                             Legend:       - 2000-2002
                                                                      	2005-2006
                             2003-2004
       0.75 9.75  18.75 27.75 36.75 45.75 54.75 63.75 72.75

                       Percent Vacant
10   20   30   40  50   60   70   80   90

           Percent Vacant
Figure B.21.  Percent Vacant Units: Histogram and Linear Relationship with Geometric Mean Blood
              Lead Levels by Time

Table B.21a.  Summary Information for Percent Vacant Units by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Num ber
Missing
0
0
0
0
Mean
5.33
5.36
5.34
5.34
Standard
Error
0.05
0.07
0.07
0.04
Minimum
0.0
0.0
0.0
0.0
10th
Percentile
1.5
1.5
1.5
1.5
25th
Percentile
2.1
2.1
2.1
2.1
Median
3.2
3.2
3.2
3.2
75th
Percentile
5.6
5.7
5.7
5.7
90th
Percentile
10.1
10.1
10.1
10.1
Maximum
80.5
78.0
66.3
80.5
Table B.21b.  Model Information for the Relationship between Percent Vacant Units and Geometric
              Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
PcM/acant
Intercept
time
Pct_Vacant
Intercept
time
Pct_Vacant
Intercept
time
Pct_Vacant
Intercept
time
Pct_Vacant
Estimate
2.640
-0.084
1.540
2.629
-0.082
1.686
-1.740
-0.140
1.877
-3.646
-0.126
2.027
-4.940
-0.094
1.812
Standard
Error
0.017
0.002
0.153
0.017
0.002
0.166
0.018
0.002
0.212
0.024
0.004
0.272
0.032
0.007
0.323
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
•c.001
-2 Log
Likelihood
52187


48618


87004


139452


177021


Variance Components
Random Effects
Qn2 = 0.357
a212= -0.026
o222= 0.004
o,,2= 0.351
a212= -0.025
o222 = 0.004
a,,2= 0.352
a212 = -0.006
a222 = 0.004
an2= 0.512
a212 = -0.022
a222 = 0.004
a,,2= 0.468
a212= -0.029
a222= 0.003
Error
ฐerror2 = 0.207
ฐerror2 = 4.429



                                                PageB-21
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                Percent Single Parent Households
                                                        8-
                                                     •o
                                                     (0
                                                     o

                                                     T3
                                                     O
                                                     CD
                                                     a
                                                     5
                                                     •S 4
                                                      E
                                                      o
                                                      a>
                                                     O
                                                     TJ
                                                     1
                                                     a
                                                            Legend:   	2000-2002
                                                                     	2005-2006
                             2003-2004
       1   7  13 19 25  31  37 43 49 55 61 67 73 79  85

                   Percent Single Parent
1 I ' ' ' ' I ' ' ' ' | • ' ' ' I '           i i | i i i i | i i
10   20   30   40   50   60   70  80   90
                                                                       Percent Single Parent
Figure B.22.  Percent Single Parent Households: Histogram and Linear Relationship with Geometric
              Mean Blood Lead Levels by Time

Table B.22a.  Summary Information for Percent Single Parent Households by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sam pie
Size
15617
10450
10417
36484
Num ber
Missing
0
0
0
0
Mean
27.59
27.61
27.60
27.60
Standard
Error
0.14
0.17
0.17
0.09
Minimum
0.0
0.0
0.0
0.0
10th
Percentile
9.5
9.5
9.5
9.5
25th
Percentile
14.1
14.1
14.1
14.1
Median
22.7
22.8
22.8
22.7
75th
Percentile
38.0
38.1
38.0
38.0
90th
Percentile
53.6
53.6
53.6
53.6
Maximum
87.9
87.9
87.9
87.9
Table B.22b.  Model Information for the Relationship between Percent Single Parent Households and
              Geometric Mean Blood Lead Levels
Model
Num ber
1
2
3
4
5
Factor
Intercept
time
Pct_Single_Parent
Intercept
time
Pct_Single_Parent
Intercept
time
Pct_Single_Parent
Intercept
time
Pct_Single_Parent
Intercept
time
Pct_Single_Parent
Estimate
2.650
-0.084
1.595
2.636
-0.083
1.635
-1.736
-0.139
2.100
-3.638
-0.129
2.208
-4.940
-0.099
1.906
Standard
Error
0.016
0.002
0.062
0.016
0.002
0.061
0.015
0.002
0.079
0.022
0.004
0.096
0.030
0.007
0.105
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
51748


48156


86543


139654


178339


Variance Components
Random Effects
0^2 = 0.304
a212 = -0.027
a222 = 0.004
0^2 = 0.294
CJ212= -0.026
a222 = 0.004
a,,2 = 0.245
a212= -0.008
a222 = 0.004
0^2 = 0.346
a212= -0.017
a222= 0.004
a,,2= 0.314
a212= -0.021
0222= 0.003
Error
aerrar2= 0.207
aerror2= 4.430



                                               Page B-22
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
    35
    30
    25
    20
  (D
  CL
    15
    10
                                         Median Year Built
                                                         8-
TJ
s
_o
CO
CD
9
•—  4 •

1
o
o
O
                                                      S  2-
                                                      B
                                                            Legend:       - 2000-2002
                                                                     	2005-2006
                                      2003-2004
      1939.2 1947.6  1956  1964.4 1972.8 1981.2 1989.6  1998

                         Year Built
    1930  1940  1950  1960  1970  1980   1990   2000

                      Year Built
Figure B.23.  Median Year Built: Histogram and Linear Relationship with Geometric Mean Blood Lead
              Levels by Time

Table B.23a.  Summary Information for Median Year Built by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
1954.4
1954.4
1954.4
1954.4
Standard
Error
0.1
0.1
0.1
0.1
Minimum
1939
1939
1939
1939
10th
Percentile
1939
1939
1939
1939
25th
Percentile
1939
1939
1939
1939
Median
1954
1954
1954
1954
75th
Percentile
1966
1966
1966
1966
90th
Percentile
1974
1974
1974
1974
Maximum
1999
1999
1999
1999
Table B.23b.  Model Information for the Relationship between Median Year Built and Geometric Mean
              Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
Median_Yr_Built
Intercept
time
Median_Yr_Built
Intercept
time
Median_Yr_Built
Intercept
time
Median_Yr_Built
Intercept
time
Median_Yr_Built
Estimate
2.651
-0.084
-0.017
2.637
-0.082
-0.017
-1.732
-0.140
-0.026
-3.634
-0.130
-0.030
-4.937
-0.099
-0.026
Standard
Error
0.016
0.002
0.001
0.015
0.002
0.001
0.015
0.002
0.001
0.021
0.004
0.001
0.030
0.007
0.001
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
51950


48361


86621


139739


178275


Variance Components
Random Effects
0,,= = 0.301
o212= -0.024
0222 = 0.004
o,,2= 0.288
o212= -0.023
o222= 0.004
o,,2= 0.231
0212= -0.004
o222= 0.004
a,,2= 0.318
a212= -0.015
a222= 0.004
0,,2= 0.312
a212= -0.022
a222= 0.003
Error
"error2 = 0.207
ฐซJ = 4.431



                                                Page B-23
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
    30
    25
    20
  tซ
  0.
    10
                            Median Year Occupied Units were built
                                                        8-
CO
c
1
o
0)
C3
                                                     B  2

                                                     1
                                                     ฃ
                                                        OH
                                                           Legend:     	 2000-2002
                                                                    	2005-2006
                                     2003-2004
     1939.21947.6  1956  1964.41972.81981.21989.6  1998
                  Year Occupied Unit Built
   1930  1940  1950   1960   1970  1980  1990  2000

                 Year Occupied Unit Built
Figure B.24.  Median Year Occupied Units were built: Histogram and Linear Relationship with
              Geometric Mean Blood Lead Levels by Time

Table B.24a.  Summary Information for Median Year Occupied Units were built by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
1954.5
1954.6
1954.6
1954.6
Standard
Error
0.1
0.1
0.1
0.1
Minim um
1939
1939
1939
1939
10th
Percentile
1939
1939
1939
1939
25th
Percentile
1939
1939
1939
1939
Median
1954
1954
1954
1954
75th
Percentile
1966
1966
1966
1966
90th
Percentile
1974
1975
1975
1975
Maximum
1999
1999
1999
1999
Table B.24b.  Model Information for the Relationship between Median Year Occupied Units were Built
              and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
Median_Yr_Occ_Built
Intercept
time
Median_Yr_Occ_Built
Intercept
time
Median_Yr_Occ_Built
Intercept
time
Median_Yr_Occ_Built
Intercept
time
Median_Yr_Occ_Built
Estimate
2.651
-0.084
-0.016
2.637
-0.082
-0.017
-1.732
-0.140
-0.026
-3.635
-0.130
-0.029
-4.937
-0.099
-0.026
Standard
Error
0.016
0.002
0.001
0.016
0.002
0.001
0.015
0.002
0.001
0.021
0.004
0.001
0.030
0.007
0.001
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
51967


48378


86641


139749


178259


Variance Components
Random Effects
a,,2= 0.303
a212= -0.024
cr222 = 0.004
a,,2 = 0.291
a2]2= -0.023
a222= 0.004
a,,^ 0.235
a212= -0.004
a222= 0.004
a,,2= 0.325
a212= -0.016
a222= 0.004
a,,2= 0.317
a212= -0.022
a222= 0.003
Error
ฐem?= 0.207
"error2 = 4.431



                                               Page B-24
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                  Percent Units Built Before 1940
  5
  o
  0)
  a.
                                                         8-
                                                      13
                                                      o
                                                      o
                                                      o
                                                      m
                                                      to
                                                      E
                                                      o
                                                      Q)
                                                      O
                                                      TJ
                                                      4)
                                                      2
                                                      Q.
                                                            Legend:
           - 2000-2002
       	2005-2006
	2003-2004
       0  6  12 18 24 30 36 42 48 54 60 66 72  78 84 90

                  Percent Built Before 1940
            1 • i • •     i i | 11
10  20  30  40   50   60   70  80  90  100

        Percent Built Before 1940
Figure B.25.  Percent Units Built Before 1940: Histogram and Linear Relationship with Geometric Mean
              Blood Lead Levels by Time

Table B.25a.  Summary Information for Percent Units Built Before 1940 by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Num ber
Missing
0
0
0
0
Mean
36.40
36.38
36.37
36.39
Standard
Error
0.17
0.21
0.21
0.11
Minim um
0.0
0.0
0.0
0.0
10th
Percentile
10.2
10.2
10.2
10.2
25th
Percentile
17.9
17.9
17.8
17.9
Median
34.9
34.9
34.8
34.9
75th
Percentile
53.3
53.3
53.3
53.3
90th
Percentile
66.2
66.1
66.1
66.1
Maximum
90.9
90.9
90.9
90.9
Table B.25b.  Model Information for the Relationship between Percent Units Built Before 1940 and
              Geometric Mean Blood Lead Levels
Model
Num ber
1
2
3
4
5
Factor
Intercept
time
Pct_Built_Pre_1940
Intercept
time
Pct_Built__Pre_1940
Intercept
time
Pct_Built_Pre_1940
Intercept
time
Pct_Built_Pre_1940
Intercept
time
Pct_Built_Pre_1940
Estimate
2.649
-0.084
1.140
2.635
-0.082
1.177
-1.731
-0.141
1.801
-3.631
-0.131
2.076
-4.934
-0.099
1.771
Standard
Error
0.015
0.002
0.054
0.015
0.002
0.054
0.014
0.002
0.064
0.021
0.004
0.077
0.029
0.007
0.091
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
51924


48335


86505


139477


178111


Variance Components
Random Effects
a,,2= 0.283
a212= -0.023
a222 = 0.004
a,,2 = 0.273
cr212 = -0.022
a 222 = 0.004
a,,2= 0.214
a212= -0.003
cr222= 0.004
a,,2= 0.293
a212= -0.015
a222= 0.004
an2= 0.301
a212= -0.022
a222 = 0.003
Error
ฐe,J= 0-207
ฐer,or2= 4.431



                                                Page B-25
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency-
determination or policy.

-------
                                 Percent Units Built Before 1950
  0)
  u
  0)
  Q.





















/
IJ
1
1










TL,








/
/


/
~i f>
' t
f
j
/




i i i
[-,



^r*~~-~~~


/
/
/
-( ~~
/









f












~
















" r
_










-






















s


-












^












-
\
\












\
\









-

\









-


\







-




\
's








-


\
\




-


\

lit III 1 1 1 1 1
7 13 19 25 31 37 43 49 55 61 67 73 79 85





















v
•fV"
1 1
8^


•a
(0
0)
o 6
0
m
c
1
S
•c 4 •
o
i
O
^J
a>
t) 2;
'•a
ฃ
a
0:

Legend: 2000-2002 2003-2004
	 2005-2006










^^^__^
_^-r*-r~f~~r~^^^ - -'
^ r^--- -~"'"'~-i—"
^ r~~~f~^^j _.--— ~~^ 	 ^ — -^ -^ ' ^~
^ r*^ - — ~~L--^^" ~"~'~r~^
r^^^Hi-"-^--1"*""""
~*



91 97 0 10 20 30 40 50 60 70 80 90 10
                  Percent Built Before 1950
                                                                      Percent Built Before 1950
Figure B.26.  Percent Units Built Before 1950: Histogram and Linear Relationship with Geometric Mean
              Blood Lead Levels by Time

Table B.26a.  Summary Information for Percent Units Built Before 1950 by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sam pie
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
44.98
44.96
44.94
44.96
Standard
Error
0.18
0.22
0.22
0.12
Minimum
0.6
0.6
0.6
0.6
10th
Percentile
14.7
14.6
14.7
14.7
25th
Percentile
25.2
25.2
25.2
25.2
Median
44.4
44.5
44.4
44.4
75th
Percentile
64.1
64.1
64.1
64.1
90th
Percentile
76.7
76.7
76.7
76.7
Maximum
96.5
96.5
96.5
96.5
Table B.26b.  Model Information for the Relationship between Percent Units Built Before 1950 and
              Geometric Mean Blood Lead Levels
Model
Num ber
1
2
3
4
5
Factor
Intercept
time
Pct_Built_Pre_1950
Intercept
time
Pct_Built_Pre_1950
Intercept
time
Pct_Built_Pre_1950
Intercept
time
Pct_Built_Pre_1950
Intercept
time
Pct_Built_Pre_1950
Estimate
2.649
-0.084
1.076
2.636
-0.082
1.110
-1.731
-0.141
1.699
-3.632
-0.131
1.933
-4.935
-0.099
1.641
Standard
Error
0.015
0.002
0.049
0.015
0.002
0.049
0.014
0.002
0.058
0.021
0.004
0.071
0.029
0.007
0.084
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
51898


48308


86483


139547


178189


Variance Components
Random Effects
0,^= 0.284
0212= -0.023
a222 = 0.004
a,,2= 0.272
a2,2 = -0.022
a222= 0.004
a,,2= 0.209
0212= -0.004
a222= 0.004
an2= 0.290
c?12= -0.015
a222= 0.004
a,,2= 0.299
a212= -0.022
a222= 0.003
Error
O6rrorz= 0.207
"error2 = 4.431



                                               Page B-26
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                  Percent Units Built Before 1960
    4.0




    3.5




    3.0




    2.5
I
Q.
    2.0
    1.5




    1.0




    0.5




      0
ffl
                                                         8-
                                            TJ
                                            to
                                            V
                                            o
                                            o

                                            m


                                            CO
                                            0)
                                            •S 4
                                            €

                                            i
                                            (3
                                              2
                                            1
                                            CL
                                              01
                                                  Legend:       - 2000-2002

                                                           	2005-2006
                                                                                           2003-2004
        3  9 15 21 27 33 39 45 51 57 63 69 75 81  87 93 99


                   Percent Built Before 1960
                                                   ^ i • • • • i • •     • • i' •     • • i •'f • i • • • • i • • • • i ''

                                                0   10  20  30   40   50  60  70  80   90  100


                                                            Percent Built Before 1960
Figure B.27.  Percent Units Built Before 1960: Histogram and Linear Relationship with Geometric Mean

              Blood Lead Levels by Time



Table B.27a.  Summary Information for Percent Units Built Before 1960 by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Num ber
Missing
0
0
0
0
Mean
57.91
57.88
57.88
57.89
Standard
Error
0.18
0.21
0.21
0.11
Minim um
2.9
2.9
2.9
2.9
10th
Percentile
27.1
27.0
27.0
27.0
25th
Percentile
40.8
40.7
40.6
40.7
Median
59.7
59.7
59.7
59.7
75th
Percentile
76.6
76.6
76.6
76.6
90th
Percentile
85.2
85.2
85.2
85.2
Maximum
99.4
99.4
99.4
99.4
Table B.27b. Model Information for the Relationship between Percent Units Built Before 1960 and

              Geometric Mean Blood Lead Levels
Model
Num ber
1
2
3
4
5
Factor
Intercept
time
PcLBuilt_Pre_1960
Intercept
time
Pct_Built_Pre_1960
Intercept
time
Pct_Built_Pre_1960
Intercept
time
Pct_Built_Pre^1 960
Intercept
time
PcLBuilt_Pre_1 960
Estimate
2.651
-0.084
1.025
2.637
-0.082
1.060
-1.732
-0.140
1.602
-3.636
-0.130
1.822
-4.938
-0.098
1.567
Standard
Error
0.016
0.002
0.053
0.016
0.002
0.052
0.015
0.002
0.064
0.021
0.004
0.079
0.030
0.007
0.092
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
51960


48374


86653


139698


178062


Variance Components
Random Effects
a,,2 = 0.306
a212 = -0.025
a222 = 0.004
0^2 = 0.294
a212 = -0.024
a222= 0.004
0^2 = 0.241
a212= -0.005
a222 = 0.004
a,,2 = 0.341
a212= -0.017
a222 = 0.004
0,,* = 0.335
a212 = -0.023
a222= 0.003
Error

-------
                                 Percent Units Built Before 1970
  0>
  Q.






















™Wtf
7 13 19 25 :





















	
'
M






















/






















J

17



















^
/


/


















"
/



13
















i
/
r




/















/
r






i<













/
/








>












/
/









5










/
/





~





5









/
f



_








(








/
r



~









i1







/



—



















/•















6










-












7





p

~ — s




-




~





73




-,



\ -
V
s












79











\











{












\










)J







-T





V
\








i ฃ






~









\
^





1












-





y




c



















\



)'




















\

]
r
8-

•
•o
5 :
\ e:
o
CD -
C -
flj
fl)
S
~ 4 -
t) -
E -
o
0)
0
"D
^ 2:
TJ
ซ
CL
o -


Legend: 2000-2002 2003-2004
	 2005-2006













^^^_ — -
— ^— "•""" — TTT--—
,— ^^^^ — ^TT^--~~'"V— — " ^" —
^^^cr^^7-7!:^^- --"•" " ~ ^
— '— '-"-~ ~"~" — " ~


i > ' ' ' i 	 1 1 1 1 1 1 1 1 1 1 1 1 1 1 , 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
0 10 20 30 40 50 60 70 80 90 10
                  Percent Built Before 1970
                                                                     Percent Built Before 1970
Figure B.28.  Percent Units Built Before 1970: Histogram and Linear Relationship with Percent Units
              Geometric Mean Blood Lead Levels by Time

Table B.28a.  Summary Information for Percent Units Built Before 1970 by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
69.68
69.66
69.66
69.67
Standard
Error
0.15
0.19
0.19
0.10
Minim um
6.3
6.3
6.3
6.3
10th
Percentile
41.6
41.5
41.6
41.6
25th
Percentile
57.2
57.2
57.2
57.2
Median
74.0
74.0
74.0
74.0
75th
Percentile
85.0
85.0
85.0
85.0
90th
Percentile
91.7
91.7
91.7
91.7
Maximum
99.4
99.4
99.4
99.4
Table B.28b.  Model Information for the Relationship between Percent Units Built Before 1970 and
              Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
Pct_Built_Pre_1970
Intercept
time
Pct_Built_Pre_1970
Intercept
time
Pct_Built_Pre_1970
Intercept
time
Pct_Built_Pre_1970
Intercept
time
Pct_Built_Pre_1 970
Estimate
2.651
-0.084
1.004
2.637
-0.082
1.045
-1.732
-0.140
1.579
-3.637
-0.129
1.807
-4.937
-0.098
1.608
Standard
Error
0.016
0.002
0.062
0.016
0.002
0.062
0.016
0.002
0.078
0.022
0.004
0.095
0.030
0.007
0.110
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
52052


48469


86806


139776


177977


Variance Components
Random Effects
a,,2= 0.329
a212= -0.026
a222= 0.004
a,,2= 0.317
a212= -0.025
o222= 0.004
a1JL2= 0.274
a212= -0.005
a222= 0.004
an2= 0.383
a212= -0.017
0222= 0.004
o,,2= 0.358
a212= -0.023
a222= 0.003
Error
ฐerror2= ฐ-207
aerror2= 4.430



                                               Page B-28
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                   Percent Units Built Before 1980
                                                          8-
                                                       TJ
                                                       8
                                                       H  eH
                                                       _o
                                                       m
                                                       1
                                                       0)
                                                       o
                                                       
-------
                            Percent Occupied Units Built Before 1940
  0)
  Q-
                                                        8-
                                                      o
                                                      m
                                                      n
                                                      0)
                                                      a

                                                      I
                                                      0
                                                     T3
                                                      V
                                                        o-
                                                            Legend:   	 2000-2002
                                                                     	2005-2006
                                 2003-2004
       0  6  12  18 24 30 36 42 48 54 60 66  72 78 84 90
           Percent Occupied Units Built Before 1940
0   10   20   30   40  50  60  70   80   90  100

      Percent Occupied Units Built Before 1940
Figure B.30.  Percent Occupied Units Built Before 1940: Histogram and Linear Relationship with
              Geometric Mean Blood Lead Levels by Time

Table B.30a.  Summary Information for Percent Occupied Units Built Before 1940 by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
36.13
36.12
36.11
36.12
Standard
Error
0.17
0.21
0.21
0.11
Minim um
0.0
0.0
0.0
0.0
10th
Percentile
10.2
10.2
10.2
10.2
25th
Percentile
17.5
17.5
17.5
17.5
Median
34.4
34.4
34.4
34.4
75th
Percentile
52.9
52.8
52.9
52.9
90th
Percentile
65.9
65.8
65.8
65.8
Maximum
90.7
90.7
90.7
90.7
Table B.30b.  Model Information for the Relationship between Percent Occupied Units Built Before 1940
              and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
Pct_Occ_Built_Pre_1 9
Intercept
time
Pct_Occ__Built_Pre_19
Intercept
time
Pct_Occ_Built_Pre_1 9
Intercept
time
Pct_Occ_Built_Pre_19
Intercept
time
Pct_Occ_Built_Pre_19
Estimate
2.649
-0.084
1.134
2.635
-0.082
1.171
-1.731
-0.141
1.802
-3.631
-0.131
2.070
-4.934
-0.099
1.762
Standard
Error
0.015
0.002
0.054
0.015
0.002
0.054
0.014
0.002
0.064
0.021
0.004
0.077
0.029
0.007
0.091
p-value
<.001
<.001
•s.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
51931


48342


86512


139476


178078


Variance Components
Random Effects
a,,2 = 0.283
0212= -0.023
a222 = 0.004
0^2= 0.273
a212= -0.022
a222 = 0.004
a,,2= 0.214
a212= -0.003
a222= 0.004
a,,2= 0.295
a212= -0.015
a222= 0.004
o,,2= 0.303
a212= -0.022
a222= 0.003
Error
ฐerror2 = 0.207
aerror2= 4.431



                                               Page B-30
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                            Percent Occupied Units Built Before 1950
  4)
  CL
4.0



3.5



3.0



2.5



2.0



1.5



1.0



0.5
                                                      TJ
                                                      (0
                                                      T3
                                                      O
                                                      O

                                                      GO

                                                      re
                                                      
O
•o
0>
ซ  2

0)
                                                         o-
                                                             Legend:   	2000-2002
                                                                      	2005-2006
                                      2003-2004
        0  61218 24 30 36 42 48 54 60 66 72 78 84 90 96

            Percent Occupied Units Built Before 1950
         10   20   30   40  50  60   70   80  90  100


          Percent Occupied Units Built Before 1950
Figure B.31.  Percent Occupied Units Built Before 1950: Histogram and Linear Relationship with
              Geometric Mean Blood Lead Levels by Time


Table B.31a.  Summary Information for Percent Occupied Units Built Before 1950 by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
44.65
44.63
44.61
44.63
Standard
Error
0.18
0.22
0.22
0.12
Minim um
0.5
0.5
0.5
0.5
10th
Percentile
14.4
14.4
14.4
14.4
25th
Percentile
24.9
24.9
24.9
24.9
Median
44.2
44.2
44.2
44.2
75th
Percentile
63.7
63.7
63.7
63.7
90th
Percentile
76.4
76.4
76.4
76.4
Maxim um
96.5
96.5
96.5
96.5
Table B.31b.  Model Information for the Relationship between Percent Occupied Units Built Before 1950
              and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
Pct_Occ_Built_Pre_19
Intercept
time
Pct_Occ_Built_Pre_19
Intercept
time
Pct_Occ_Built_Pre_19
Intercept
time
Pct_Occ_Built_Pre_19
Intercept
time
Pct_Occ_Built_Pre_1 9
Estimate
2.649
-0.084
1.061
2.636
-0.082
1.095
-1.731
-0.141
1.687
-3.632
-0.131
1.918
-4.935
-0.099
1.627
Standard
Error
0.015
0.002
0.050
0.015
0.002
0.049
0.014
0.002
0.058
0.021
0.004
0.071
0.029
0.007
0.084
p-value
<.001
<.001
<,001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
51911


48321


86498


139549


178149


Variance Components
Random Effects
0,,* = 0.285
C212= -0.023
o222 = 0.004
a,,2= 0.274
a212= -0.022
o222= 0.004
a,,2 = 0.210
a212= -0.003
a222= 0.004
a, ,2= 0.293
a212= -0.015
a222= 0.004
a,,2= 0.303
0212= -0.022
a222= 0.003
Error
ฐem,2= 0-207
"error2 = 4.431



                                                PageB-31

This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                            Percent Occupied Units Built Before 1960
    4.0



    3.5



    3.0



    2.5
    2.0
    1.5



    1.0



    0.5



     0
1
                                                         8-
                                           10
                                           0)
                                           "O  fi-l
                                           o  "1
                                           o
                                           m

                                           i
                                           ฃ  4H
                                           i

                                           o
                                           i
                                           O
                                           •o
                                           9
                                           .5  :
                                           0)
                                           a
                                                 Legend:   	2000-2002
                                                          	2005-2006
2003-2004
        3  9  15 21 27 33 39 45 51 57 63 69 75 81 87 93 99

            Percent Occupied Units Built Before 1960
                                                   10  20   30   40  50  60  70  80   90  100


                                                     Percent Occupied Units Built Before 1960
Figure B.32.  Percent Occupied Units Built Before 1960: Histogram and Linear Relationship with

              Geometric Mean Blood Lead Levels by Time


Table B.32a.  Summary Information for Percent Occupied Units Built Before 1960 by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sam pie
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
57.57
57.54
57.54
57.55
Standard
Error
0.18
0.22
0.22
0.12
Minim um
2.1
2.1
2.1
2.1
10th
Percentile
26.5
26.4
26.4
26.4
25th
Percentile
40.2
40.1
40.1
40.1
Median
60.1
60.1
60.1
60.1
75th
Percentile
76.3
76.3
76.3
76.3
90th
Percentile
85.2
85.2
85.1
85.2
Maximum
99.4
99.4
99.4
99.4
Table B.32b.  Model Information for the Relationship between Percent Occupied Units Built Before 1960
              and Geometric Mean Blood Lead Levels
Model
Num ber
1
2
3
4
5
Factor
Intercept
time
Pct_Occ_Built_Pre_19
Intercept
time
Pct_Occ_Built_Pre_19
Intercept
time
Pct_Occ_Built_Pre_19
Intercept
time
Pct_Occ_Built_Pre_19
Intercept
time
Pct_Occ_Built_Pre_1 9
Estimate
2.651
-0.084
0.996
2.637
-0.082
1.032
-1.732
-0.140
1.571
-3.637
-0.130
1.790
-4.938
-0.098
1.545
Standard
Error
0.016
0.002
0.053
0.016
0.002
0.052
0.015
0.002
0.064
0.022
0.004
0.079
0.030
0.007
0.092
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
51976


48390


86675


139713


178044


Variance Components
Random Effects
a,,2= 0.309
a212= -0.025
a222= 0.004
a,,2= 0.297
a2]2= -0.024
a222= 0.004
a,,2= 0.245
a212= -0.005
a222 = 0.004
a,,2= 0.345
a212= -0.017
o222= 0.004
a,,2= 0.338
a212= -0.023
a222 = 0.003
Error
ฐ error' = ฐ.207
ฐ=rrof2 = 4.430



                                                Page B-32
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                            Percent Occupied Units Built Before 1970
  0)
  Q.
                                                        8-
                                                        6-
                                                      _o
                                                      m
                                                      o
                                                      ซ
                                                      O
                                                      •a
                                                      4)
                                                      B 2
                                                            Legend:       - 2000-2002
                                                                     	2005-2006
                             2003-2004
       5 11  17 23 29 35 41 47 53 59 65 71 77 83 89 95
            Percent Occupied Units Built Before 1970
10  20  30  40  50   60   70   80  90  100

  Percent Occupied Units Built Before 1970
Figure B.33.  Percent Occupied Units Built Before 1970: Histogram and Linear Relationship with
              Geometric Mean Blood Lead Levels by Time

Table B.33a.  Summary Information for Percent Occupied Units Built Before 1970 by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
69.33
69.31
69.32
69.32
Standard
Error
0.16
0.19
0.19
0.10
Minim um
4.9
4.9
4.9
4.9
10th
Percentile
40.6
40.6
40.6
40.6
25th
Percentile
56.5
56.5
56.5
56.5
Median
73.6
73.6
73.7
73.6
75th
Percentile
85.1
85.1
85.1
85.1
90th
Percentile
91.5
91.5
91.5
91.5
Maximum
99.4
99.4
99.4
99.4
Table B.33b.  Model Information for the Relationship between Percent Occupied Units Built Before 1970
              and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
Pct_Occ_Built_Pre_1 9
Intercept
time
Pct_Occ_Built_Pre_19
Intercept
time
Pct_Occ_Built_Pre_19
Intercept
time
Pct_Occ_Built_Pre_1 9
Intercept
time
Pct_Occ_Built_Pre_19
Estimate
2.651
-0.084
0.952
2.637
-0.082
0.998
-1.732
-0.140
1.523
-3.637
-0.129
1.753
-4.938
-0.098
1.568
Standard
Error
0.016
0.002
0.062
0.016
0.002
0.062
0.016
0.002
0.078
0.022
0.004
0.095
0.030
0.007
0.109
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
52070


48486


86830


139799


177979


Variance Components
Random Effects
a,,2= 0.333
a212= -0.026
a222= 0.004
0,^ = 0.321
a212= -0.025
a222= 0.004
an2 = 0.278
02,2= -0.005
a222= 0.004
a,,2 = 0.388
a212= -0.017
a222 = 0.004
a,,2= 0.361
a212= -0.023
a222= 0.003
Error
aerror2= 0.207
ฐerror2= 4.430



                                               Page B-33
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                            Percent Occupied Units Built Before 1980
                                                         8-
                                                      T3
                                                      as
                                                      a>
                                                      o
                                                      o
                                                      m

                                                      c
                                                      n
                                                      v
                                                      v

                                                      o
                                                      C3

                                                      I
                                                      tS  2


                                                      '1
                                                         o-
  Legend:   	 2000-2002

           	2005-2006
                                                                                            2003-2004
     21.75 30.75 39.75 48.75 57.75 66.75 75.75 84.75 93.75


            Percent Occupied Units Built Before 1980
20   30   40    50    60    70   80   90


      Percent Occupied Units Built Before 1980
100
Figure B.34.  Percent Occupied Units Built Before 1980: Histogram and Linear Relationship with

              Geometric Mean Blood Lead Levels by Time



Table B.34a.  Summary Information for Percent Occupied Units Built Before 1980 by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
81.60
81.58
81.58
81.59
Standard
Error
0.11
0.14
0.14
0.07
Minimum
21.7
21.7
21.7
21.7
10th
Percentile
61.1
61.0
61.0
61.0
25th
Percentile
74.2
74.1
74.1
74.1
Median
85.7
85.7
85.7
85.7
75th
Percentile
92.5
92.5
92.5
92.5
90th
Percentile
96.1
96.2
96.1
96.1
Maximum
100.0
100.0
100.0
100.0
Table B.34b.  Model Information for the Relationship between Percent Occupied Units Built Before 1980

              and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
Pct_Occ_Built_Pre_19
Intercept
time
Pct_Occ_Built_Pre_19
Intercept
time
Pct_Occ_Built_Pre_19
Intercept
time
Pct_Occ_Built_Pre_19
Intercept
time
Pct_Occ_Built_Pre_19
Estim ate
2.651
-0.084
1.256
2.637
-0.082
1.314
-1.732
-0.140
1.967
-3.638
-0.129
2.263
-4.937
-0.098
2.075
Standard
Error
0.017
0.002
0.085
0.016
0.002
0.085
0.016
0.002
0.108
0.023
0.004
0.133
0.030
0.007
0.153
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
•e.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
52084


48501


86860


139777


177867


Variance Components
Random Effects
a,,2 = 0.338
a212 = -0.026
a222 = 0.004
a,,2^ 0.326
a212= -0.025
a222= 0.004
a,,2= 0.289
a212= -0.005
a222= 0.004
a,,2 = 0.405
a212 = -0.018
0222= 0.004
o1,2= 0.366
a212= -0.023
a222= 0.003
Error
terror2 = 0.207

-------
                         Percent Residents Less than Six Years of Age
10



8



6
c
o
3
Q.

4 "



2
0J



















ffrflf
0.75 3.25


















1
|T

















/
/

















y
/


5








i
/
/
/
L






T
















y
/












/













-



/
















-

' N

















\
>


















\J












\










T


\
\
_t







\
A






\

-


















\




















\
T



















\



















w_
8;
•
•O
TO
-I
•a RI,
0 6 .
o
CD -
ง
5
0 4:
'= H.
0)
E
o
5
O
ง 2-
1 ;
a.
o:

Legend: 2000-2002 ~ 2003-2004
	 2005-2006














~^^ 	 ' 	 ' 	 — - — -^— _
	



75 8.25 10.75 13.25 15.75 18.25 20.75 0 10 20 30
               Percent Less than 6 Years of Age
                                                                  Percent Less than 6 Years of Age
Figure B.35.  Percent Residents Less than Six Years of Age: Histogram and Linear Relationship with
              Geometric Mean Blood Lead Levels by Time

Table B.35a.  Summary Information for Percent Residents Less than Six Years of Age by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
8.89
8.89
8.88
8.89
Standard
Error
0.02
0.03
0.03
0.01
Minimum
0.8
0.8
0.8
0.8
10th
Percentile
5.8
5.8
5.8
5.8
25th
Percentile
7.3
7.3
7.3
7.3
Median
8.9
8.9
8.9
8.9
75th
Percentile
10.4
10.4
10.4
10.4
90th
Percentile
11.9
11.9
11.9
11.9
Maximum
21.8
21.8
21.8
21.8
Table B.35b.  Model Information for the Relationship between Percent Residents Less than Six Years of
              Age and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
Pct_LE_Six
Intercept
time
Pct_LE_Six
Intercept
time
Pct_LE_Six



Intercept
time
Pct_LE_Six
Estimate
2.649
-0.084
0.161
2.636
-0.082
0.096
-1.735
-0.140
-0.851



-4.949
-0.094
1.287
Standard
Error
0.018
0.002
0.494
0.018
0.002
0.502
0.018
0.002
0.651



0.032
0.007
0.865
p-value
<.001
•e.001
0.745
<.001
<.001
0.848
<.001
<.001
0.191



<.001
<.001
0.137
-2 Log
Likelihood
52280


48714


87127





177615


Variance Components
Random Effects
a,,2= 0.388
a212= -0.028
a222= 0.004
a,,2= 0.379
a212= -0.027
o222= 0.004
an2= 0.373
a212= -0.006
o222= 0.004



an2= 0.489
a212= -0.029
0222= 0.003
Error
aerror2= 0.207
"error2 = 4.428



                                              Page B-35
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                         Number Residents Less than Six Years of Age
10


8 '




.
ซ- 6
1
9
a.

4 '



2

0

















y
/
J










r


/

1
1










-

1
I
j













/







































1\
\

























-

















I

\
\
\




















\
-\




















\





















1' n 1 	 1 	 =-
8


1
E 6
0
m
c
re
V
5
'C 4
w
E
8
o
1
^rf O
o ^

a
0:

Legend: 2000-2002 2003-2004
	 2005-2006











~^~~1- — ^^^^^^
^ ^ ^^r77^^-^ ^_
" " " "" ^^"~tr=^=^=^.
ia=i=i;;=i|


      2000  30000  58000  86000114000142000170000

               Number Less than 6 Years of Age
50000     100000    150000

Number Less than 6 Years of Age
200000
Figure B.36.  Number Residents Less than Six Years of Age: Histogram and Linear Relationship with
              Geometric Mean Blood Lead Levels by Time

Table B.36a.  Summary Information for Number Residents Less than Six Years of Age by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Nu m b e r
Missing
0
0
0
0
Mean
42635
42582
42621
42616
Standard
Error
168
205
206
110
Minimum
1600
1400
1400
1400
10th
Percentile
18800
18800
18800
18800
25th
Percentile
27600
27500
27600
27500
Median
40000
39900
39900
39900
75th
Percentile
54400
54200
54400
54400
90th
Percentile
69600
69600
69600
69600
Maximum
190800
190800
190800
190800
Table B.36b.  Model Information for the Relationship between Number Residents Less than Six Years of
              Age and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
Num_LE_Six
Intercept
time
Num_LE_Six
Intercept
time
Num_LE_Six
Intercept
time
Num_LE_Six



Estimate
2.645
-0.084
0.000
2.633
-0.082
0.000
-1.732
-0.141
-0.001
-3.636
-0.127
-0.001



Standard
Error
0.017
0.002
0.000
0.017
0.002
0.000
0.017
0.002
0.000
0.024
0.004
0.000



p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001



-2 Log
Likelihood
52243


48678


86913


138996





Variance Components
Random Effects
a^2= 0.360
0212= -0.026
a222 = 0.004
a,,2 = 0.353
a212 = -0.025
a222 = 0.004
0,^= 0.344
a212 = -0.005
a222 = 0.004
a,,2= 0.497
a212= -0.020
a222= 0.004



Error
ฐem,2= 0.207
aerro,2= 4.428



                                              Page B-36
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                  Percent Residents with Less than Ninth Grade Education
14


12 '


10



S 8 -
0)
o
5
Q.
6



4 "



2
0"


"
















j
/
f


.1J-1.,,1




-.









p








-\
^S






048


















\
\



1
.1




















\
\
\
hX
mnnLTfttrfi^iL^^
8;
•
•Q J
CO
CO
_J
•D fi I
0 b .
0
CD
C -
CO
0)
5

._ 4 ~
fe -
ซ
E
o
ซ
•D -

*^ o —
O
•5 -
ฃ
Q.
0-

Legend: 2000-2002 	 2003-2004
	 2005-2006












^ 	 *""*'
^— — "" 	 *^^3- -"""''*
^ 	 — — """~~*'_ป^'^ "
-****^***^m^"^*--*-""*
-*-****T^-^*^- *-'" ^ "" " "
_,^"~"^_ 	 — '-^'
_!• *— "^




12 16 20 24 28 32 36 40 0 10 20 30 40 5C
Percent Less than 9th Grade _ .„ซ..,ซ,-
Figure B.37. Percent Residents with Less than Ninth Grade Education: Histogram and Linear
             Relationship with Geometric Mean Blood Lead Levels by Time

Table B.37a. Summary Information for Percent Residents with Less than Ninth Grade Education by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
6.22
6.20
6.21
6.21
Standard
Error
0.05
0.06
0.07
0.03
Minim um
0.0
0.0
0.0
0.0
10th
Percentile
1.0
0.9
0.9
1.0
25th
Percentile
2.0
1.9
1.9
1.9
Median
3.8
3.8
3.8
3.8
75th
Percentile
7.7
7.7
7.7
7.7
90th
Percentile
15.2
15.1
15.2
15.1
Maximum
41.2
41.2
41.2
41.2
Table B.37b. Model Information for the Relationship between Percent Residents with Less than Ninth
             Grade Education and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
Pct_LT_9th_Grade
Intercept
time
Pct_LT_9th_Grade
Intercept
time
Pct_LT_9th_Grade
Intercept
time
Pct_LT_9th_Grade
Intercept
time
Pct_LT_9th_Grade
Estimate
2.651
-0.084
3.135
2.637
-0.083
3.173
-1.736
-0.139
4.259
-3.643
-0.128
4.305
-4.944
-0.097
3.405
Standard
Error
0.017
0.002
0.179
0.017
0.002
0.179
0.016
0.002
0.228
0.023
0.004
0.270
0.031
0.007
0.291
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
52007


48436


86829


139727


177859


Variance Components
Random Effects
a,,2 = 0.347
a212= -0.028
a222 = 0.004
0^ = 0.336
a212= -0.027
a222= 0.004
0,^= 0.301
a212= -0.008
a222= 0.004
0^2= 0.437
a212= -0.020
a222= 0.004
a,,2 = 0.405
a212= -0.024
a222= 0.003
Error
aerror2= 0.207
"error2 = 4.428



                                              Page B-37
This information is distributed solely for the purpose ofpre- dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                        Percent Residents without a High School Degree
                                                        8
                                                      T3

                                                      s
                                                      _o

                                                      CD


                                                      re
                                                      a>
                                                     I 4
                                                      o>


                                                      o
                                                      o>
                                                     O



                                                     1,

                                                     TJ
                                                      0)
                                                            Legend:   	2000-2002

                                                                     	2005-2006
                2003-2004
      0.75  8.25 15.75 23.25 30.75 38.25 45.75 53.25 60.75


              Percent without High School Degree
30
40
50
10    20



  Percent without High School Degree
                  60
                  70
Figure B.38.  Percent Residents without a High School Degree: Histogram and Linear Relationship with

              Geometric Mean Blood Lead Levels by Time



Table B.38a.  Summary Information for Percent Residents without a High School Degree by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
17.23
17.19
17.21
17.21
Standard
Error
0.10
0.12
0.12
0.07
Minimum
0.4
0.4
0.4
0.4
10th
Percentile
5.2
5.1
5.1
5.2
25th
Percentile
8.5
8.3
8.3
8.3
Median
13.3
13.3
13.3
13.3
75th
Percentile
22.4
22.3
22.3
22.3
90th
Percentile
35.8
35.7
35.8
35.8
Maximum
64.3
64.3
64.3
64.3
Table B.38b.  Model Information for the Relationship between Percent Residents without a High School

              Degree and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
Pct_No_HS_Degree
Intercept
time
Pct_No_HS_Degree
Intercept
time
PctJMo_HS_Degree
Intercept
time
Pct_No_HS_Degree
Intercept
time
Pct_No_HS_Degree
Estimate
2.651
-0.084
2.038
2.638
-0.083
2.061
-1.737
-0.139
2.713
-3.643
-0.128
2.731
-4.947
-0.098
2.249
Standard
Error
0.016
0.002
0.090
0.016
0.002
0.090
0.016
0.002
0.115
0.022
0.004
0.139
0.030
0.007
0.153
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
51853


48280


86710


139858


178347


Variance Components
Random Effects
a,,2 = 0.331
a212= -0.028
a222= 0.004
a.,2= 0.320
a212 = -0.027
a222= 0.004
0,,* = 0.271
a212= -0.009
a222= 0.004
0,^= 0.394
a,,2= -0.019
a222 = 0.004
at12= 0.366
a^= -0.023
o222= 0.003
Error
aerror2= 0.207
a.TOr*= 4.428



                                               Page B-38

This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality

guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency

determination or policy.

-------
                          Percent Residents without College Education
  o
  53
  0.
                                \
                                                        8-
                                                     •o
                                                     re
                                                     u>


                                                     •O
                                                     O
                                                     o

                                                     m
                                              •| H






                                              I


                                              1
                                              t5 2

                                              '•6
                                               V
                                                        01
                                                     Legend:
                                                                          2000-2002

                                                                          2005-2006
2003-2004
5 11 17 23 29 35 41 47 53 59 65 71  77 83 89


          Percent without any College
                                                          0   10   20   30   40   50   60   70    80   90


                                                                     Percent without any College
Figure B.39.  Percent Residents without College Education: Histogram and Linear Relationship with

              Geometric Mean Blood Lead Levels by Time



Table B.39a.  Summary Information for Percent Residents without College Education by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
44.81
44.74
44.75
44.77
Standard
Error
0.14
0.17
0.17
0.09
Minimum
4.2
4.2
4.2
4.2
10th
Percentile
20.5
20.5
20.5
20.5
25th
Percentile
32.1
32.0
32.0
32.1
Median
45.1
45.1
45.0
45.1
75th
Percentile
57.6
57.6
57.6
57.6
90th
Percentile
68.5
68.4
68.5
68.5
Maximum
88.8
88.8
88.8
88.8
Table B.39b.  Model Information for the Relationship between Percent Residents without College

              Education and Geometric Mean Blood Lead Levels
Model
Num ber
1
2
3
4
5
Factor
Intercept
time
Pct_No_College
Intercept
time
Pct_No_College
Intercept
time
Pct_No_College
Intercept
time
Pct_No_College
Intercept
time
Pct_No_College
Estimate
2.652
-0.084
1.510
2.639
-0.083
1.524
-1.739
-0.139
2.011
-3.647
-0.128
2.020
-4.951
-0.098
1.692
Standard
Error
0.016
0.002
0.061
0.016
0.002
0.062
0.016
0.002
0.079
0.022
0.004
0.100
0.030
0.007
0.115
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
51787


48218


86729


140104


178689


Variance Components
Random Effects
a,,2= 0.326
a212 = -0.029
cr222= 0.004
CT112= 0.315
0212= -0.028
a222= 0.004
CT112= 0.261
a212= -0.009
a222 = 0.004
0,^= 0.389
a?12 = -0.019
CJ222 = 0.004
a,,2= 0.363
a212= -0.023
a222= 0.003
Error
"error' = ^.207
"error2 = 4.428



                                               Page B-39

This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality

guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency

determination or policy.

-------
                            Percent Residents without College Degree
  o
                                                         8:
                                                      ฃ
                                                      CO
                                                      c
                                                      as
o>

o
0)
O
TJ

%
                                                         2
                                                         o-
                                                            Legend:
                     2000-2002
                     2005-2006
2003-2004
      10 16 22 28 34  40 46 52 58 64 70 76  82 88 94

               Percent without College Degree
         10   20   30   40  50  60  70   80   90  100

              Percent without College Degree
Figure B.40.  Percent Residents without College Degree: Histogram and Linear Relationship with
              Geometric Mean Blood Lead Levels by Time

Table B.40a.  Summary Information for Percent Residents without College Degree by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
63.51
63.46
63.45
63.48
Standard
Error
0.15
0.18
0.18
0.10
Minim um
9.9
9.9
9.9
9.9
10th
Percentile
35.9
36.0
36.0
36.0
25th
Percentile
52.3
52.1
52.2
52.2
Median
66.0
65.9
65.9
66.0
75th
Percentile
77.5
77.4
77.4
77.4
90th
Percentile
85.5
85.4
85.4
85.5
Maximum
97.2
97.2
97.2
97.2
Table B.40b.  Model Information for the Relationship between Percent Residents without College Degree
              and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
Pct_No__College_Degre
Intercept
time
Pct_No_College_Degre
Intercept
time
Pct_No_College_Degre
Intercept
time
Pct_No_College_Degre
Intercept
time
Pct_No_College_Degre
Estimate
2.651
-0.084
1.433
2.638
-0.083
1.450
-1.739
-0.139
1.910
-3.648
-0.128
1.926
-4.951
-0.097
1.622
Standard
Error
0.016
0.002
0.061
0.016
0.002
0.061
0.016
0.002
0.079
0.023
0.004
0.100
0.031
0.007
0.116
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
51822


48251


86760


140085


178626


Variance Components
Random Effects
a,,2 = 0.331
a212= -0.029
a222 = 0.004
0^ = 0.320
0212= -0.028
a222= 0.004
a,,2= 0.268
a212= -0.009
a222 = 0.004
an2 = 0.402
a212= -0.020
a222= 0.004
a112= 0.370
C212= -0.024
o222 = 0.003
Error
ฐe,J= 0.207
ฐซm,2= 4.428



                                                Page B-40
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
  eu
  o
                                        Total Housing Units
                                                        8-
                                                      T5
                                                      CO
                                                      V
                                                      o
                                                      m
                                                      CD
                                                      •=  4
                                                      E
                                                      o
                                                      o

                                                      1,
                                                      T3
                                                      O
                                                         0-
                                                            Legend:       - 2000-2002
                                                                     	2005-2006
                            2003-2004
      75   825  1575 2325 3075  3825  4575 5325 6075

                    Total Housing Units
1000  2000   3000  4000  5000  6000  7000

         Total Housing Units
Figure B.41.  Total Housing Units: Histogram and Linear Relationship with Geometric Mean Blood
              Lead Levels by Time

Table B.41a.  Summary Information for Total Housing Units by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sam pie
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
1954
1955
1956
1955
Standard
Error
6
7
7
4
Minimum
262.0
85.0
85.0
85.0
10th
Percentile
1084.0
1084.0
1084.0
1084.0
25th
Percentile
1375.0
1375.0
1376.0
1375.0
Median
1869.0
1868.0
1869.0
1869.0
75th
Percentile
2414.0
2414.0
2415.0
2414.0
90th
Percentile
2971.0
2971.0
2971.0
2971.0
Maxim um
6549.0
6549.0
6549.0
6549.0
Table B.41b.  Model Information for the Relationship between Total Housing Units and Geometric Mean
              Blood Lead Levels
Model
Num ber
1
2
3
4
5
Factor
Intercept
time
TotaLHousing_Units
Intercept
time
Total_Housing_Units
Intercept
time
TotaLHousing_Units
Intercept
time
Total_Housing_Units
Intercept
time
Total_Housing_Units
Estimate
2.648
-0.084
0.000
2.636
-0.082
0.000
-1.733
-0.140
0.000
-3.642
-0.126
0.000
-4.933
-0.094
0.000
Standard
Error
0.018
0.002
0.000
0.017
0.002
0.000
0.018
0.002
0.000
0.024
0.004
0.000
0.031
0.007
0.000
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
52287


48721


87091


139462


176701


Variance Components
Random Effects
an2= 0.382
a212 = -0.027
a222 = 0.004
au*= 0.374
0212 = -0.026
a222= 0.004
an2 = 0.364
a212= -0.005
a222 = 0.004
0^= 0.519
a?12= -0.022
a222= 0.004
a,,2 = 0.456
a212= -0.027
a222= 0.003
Error
aerror2 = 0.207
^error2 = 4.428



                                                Page B-41
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                            Total Population
  c 4
  s
  0)
  a
                                                           8
                                                        •a
                                                        a
ffl
I

1  4
i
o
i
                                                       •a
                                                        0)
                                                       •
                                                        0)
                                                        a
      Legend:
                                                                             2000-2002
                                                                             2005-2006
2003-2004
      500  2000  3500  5000  6500  8000  9500  11000

                       Total Population
     0   2000  4000  6000  8000  10000 12000 14000

                    Total Population
Figure B.42. Total Population: Histogram and Linear Relationship with Geometric Mean Blood Lead
              Levels by Time
Table B.42a. Summary Information for Total Population by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
4773
4770
4773
4772
Standard
Error
14
17
17
9
Minim um
405.0
388.0
388.0
388.0
10th
Percentile
2622.0
2630.0
2635.0
2630.0
25th
Percentile
3451.0
3448.0
3451.0
3448.0
Median
4611.0
4610.0
4616.0
4611.0
75th
Percentile
5945.0
5942.0
5945.0
5945.0
90th
Percentile
7277.0
7277.0
7277.0
7277.0
Maximum
12051.0
12051.0
12051.0
12051.0
Table B.42b. Model Information for the Relationship between Total Population and Geometric Mean
              Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
TotaLPop
Intercept
time
TotaLPop
Intercept
time
TotaLPop
Intercept
time
TotaLPop



Estimate
2.645
-0.084
0.000
2.632
-0.082
0.000
-1.733
-0.141
0.000
-3.636
-0.126
0.000



Standard
Error
0.017
0.002
0.000
0.017
0.002
0.000
0.017
0.002
0.000
0.024
0.004
0.000



p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001



-2 Log
Likelihood
52223


48661


86910


138948





Variance Components
Random Effects
an2 = 0.359
a2,2= -0.026
0222= 0.004
a,,2= 0.353
a212= -0.025
a222 = 0.004
on2= 0.341
a212= -0.005
a222= 0.004
a,,2= 0.487
a212= -0.021
a222= 0.004



Error
aerror2= 0.207
aerror2= 4.428



                                                 Page B-42
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency-
determination or policy.

-------
                                            Housing Density
    35
    30
    25
  c 20
  I
  9)
  a.
    15
    10
                                                          8-
                                                          6-
                                                        s
o
m
to
S
1  4
3
o
a>
O
1
o
'•5
v
                                                          o-
                                                             Legend:      	 2000-2002
                                                                       	2005-2006
                                      2003-2004
            8400  16800 25200  33600  42000 50400 58800

                       Housing Density
         10000 20000 30000 40000  50000  60000  70000

                    Housing Density
Figure B.43.  Housing Density: Histogram and Linear Relationship with Geometric Mean Blood Lead
              Levels by Time
Table B.43a. Summary Information for Housing Density by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
2935.7
2933.1
2921.4
2930.9
Standard
Error
35.7
43.6
43.0
23.2
Minimum
10.3
10.3
10.3
10.3
10th
Percentile
144.8
144.8
145.3
144.8
25th
Percentile
370.6
362.2
370.6
364.7
Median
1293.4
1293.4
1293.4
1293.4
75th
Percentile
3766.4
3766.4
3766.4
3766.4
90th
Percentile
7387.5
7387.5
7387.5
7387.5
Maximum
60388.6
60388.6
60388.6
60388.6
 Table B.43b. Model Information for the Relationship between Housing Density and Geometric Mean
              Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
Housing_Density
Intercept
time
Housing_Density
Intercept
time
Housing_Density
Intercept
time
Housing_Density
Intercept
time
Housing_Density
Estimate
2.648
-0.084
0.000
2.635
-0.082
0.000
-1.734
-0.140
0.000
-3.643
-0.127
0.000
-4.939
-0.095
0.000
Standard
Error
0.017
0.002
0.000
0.017
0.002
0.000
0.018
0.002
0.000
0.024
0.004
0.000
0.031
0.007
0.000
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
52273


48697


87068


139580


177260


Variance Components
Random Effects
Q,,2= 0.374
a212= -0.027
o222= 0.004
o,,2= 0.365
a212 = -0.026
ฐ222 = 0.004
a,,2 = 0.353
o212= -0.005
a222 = 0.004
on2= 0.500
ay,2= -0.021
a222= 0.004
a,/ 0.450
a212= -0.027
a222= 0.003
Error
"error2 = 0.207
aerror2= 4.429



                                                 Page B-43
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                  Air Dispersion (ASPEN) Model
60




50



40


^
C
lac
0.



20



10

0






















1


1





















>T
\
\

1 1 1 1
8-


"D
(0
ซi
_l
0 6-
o
m
c
(0
8
^
•^ 4;
(I)
E '
0
V
o :
•o
V
•*-' O
o 2
'•5
ฃ !
Q.
0_


Legend: 2000-2002 " 2003-2004
	 2005-2006
















	 . 	

	 	




      0.0004  0.006  0.0116 0.0172  0.0228 0.0284  0.034

                 Air Dispersion (ASPEN) Model
0.000 0.005 0.010 0.015 0.020 0.025 0.030  0.035 0.040

           Air Dispersion (ASPEN) Model
Figure B.44.  Air Dispersion (ASPEN) Model: Histogram and Linear Relationship with Geometric Mean
              Blood Lead Levels by Time

Table B.44a.  Summary Information for Air Dispersion (ASPEN) Model by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Combined
Sam pie
Size
15617
10450
10417
36484
Num ber
Missing
0
0
0
0
Mean
0.0015
0.0015
0.0015
0.0015
Standard
Error
0.0000
0.0000
0.0000
0.0000
Minim um
0.0001
0.0001
0.0001
0.0001
10th
Percentile
0.0006
0.0006
0.0006
0.0006
25th
Percentile
0.0009
0.0009
0.0009
0.0009
Median
0.0012
0.0012
0.0012
0.0012
75th
Percentile
0.0016
0.0016
0.0016
0.0016
90th
Percentile
0.0022
0.0021
0.0022
0.0022
Maxim um
0.0356
0.0356
0.0356
0.0356
 Table B.44b. Model Information for the Relationship between Air Dispersion (ASPEN) Model and
              Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
aspen
Intercept
time
aspen
Intercept
time
aspen
Intercept
time
aspen
Intercept
time
aspen
Estimate
2.649
-0.084
-3.315
2.636
-0.082
-3.473
-1.735
-0.140
-1.779
-3.647
-0.126
-2.885
-4.946
-0.094
-2.154
Standard
Error
0.018
0.002
6.448
0.018
0.002
6.588
0.018
0.002
8.535
0.025
0.004
10.296
0.032
0.007
10.895
p-value
<.001
<.001
0.607
<.001
<.001
0.598
<.001
<.001
0.835
<.001
<.001
0.779
<.001
<.001
0.843
-2 Log
Likelihood
52275


48708


87137


139740


177377


Variance Components
Random Effects
a^2= 0.387
0212 = -0.027
a222= 0.004
0,^= 0.379
a2J2 = -0.026
a222 = 0.004
a,,2= 0.374
0212= -0.006
a222 = 0.004
o,,2= 0.541
a?12= -0.022
a222 = 0.004
a,,2= 0.489
a212= -0.029
a222= 0.003
Error
aerror2= 0.207
aerror2= 4.428



                                                Page B-44
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
    50
    40
  .. 30
  ง
  s.
    20
    10
                                 Air Exposure (HAPEM5) Model
                                                        8-
         r
                    4-
                 m
                 c
                 (0
                 9
u
S
o
a
O
                                                     S 2
                                                     o ซ
                                                      0
                                                      1_
                                                      Q.
                                                        o-
                                                           Legend:
                                    - 2000-2002
                                   — 2005-2006
                                     2003-2004
     0.00015  0.00285  0.00555 0.00825  0.01095  0.01365

                  Exposure (HAPEM5) Model
                    0.000 0.002 0.004 0.006 0.008 0.010 0.012 0.014 0.016

                                Exposure (HAPEM5) Model
Figure B.45.  Air Exposure (HAPEM5) Model: Histogram and Linear Relationship with Geometric
              Mean Blood Lead Levels by Time

Table B.45a.  Summary Information for Air Exposure (HAPEM5) Model by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Combined
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
0.0007
0.0007
0.0007
0.0007
Standard
Error
0.0000
0.0000
0.0000
0.0000
Minimum
0.0000
0.0000
0.0000
0.0000
10th
Percentile
0.0003
0.0003
0.0003
0.0003
25th
Percentile
0.0004
0.0004
0.0004
0.0004
Median
0.0005
0.0005
0.0005
0.0005
75th
Percentile
0.0007
0.0007
0.0007
0.0007
90th
Percentile
0.0010
0.0010
0.0010
0.0010
Maximum
0.0144
0.0135
0.0144
0.0144
 Table B.45b. Model Information for
              Geometric Mean Blood
the Relationship between Air Exposure (HAPEM5) Model
Lead Levels
                                        and
Model
Number
1
2
3
4
5
Factor
Intercept
time
hapem
Intercept
time
hapem
Intercept
time
hapem
Intercept
time
hapem
Intercept
time
hapem
Estimate
2.649
-0.084
-9.643
2.636
-0.082
-10.194
-1.735
-0.140
-5.991
-3.647
-0.126
-8.459
-4.946
-0.094
-5.788
Standard
Error
0.018
0.002
16.424
0.018
0.002
16.931
0.018
0.002
21.925
0.025
0.004
26.690
0.032
0.007
28.371
p-value
<.001
<.001
0.557
<.001
<.001
0.547
<.001
<.001
0.785
<.001
<.001
0.751
<.001
<.001
0.838
-2 Log
Likelihood
52273


48706


87135


139738


177375


Variance Components
Random Effects
a,,2 = 0.387
a212= -0.027
a222 = 0.004
a,,2= 0.379
a212 = -0.026
a222 = 0.004
a,,2= 0.374
a212= -0.006
0222 = 0.004
a,,2= 0.541
a212 = -0.022
o222= 0.004
a,,2 = 0.489
a212= -0.029
o222= 0.003
Error
"error2 = 0.207
aerror2= 4.428



                                               Page B-45
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                     Air Hazard Quotient (HQ)
    50
    40
  „ 30


  ง
  o>
  Q.
    20
    10
                                                      •o
                                                      TO
                                                      
-------
                               Current HUD Funding ($ per Child)
9-
"

8;

.
2~ 7:
'j-
o
1 6:
Q
O) c '.
C 5 .
'•5
C
i A -
LJ_ t|
D
0 -
4— < O
C
o>
^
3 0 "
0 2.
1 ^
:
o-


•








1









•
1
j



•








1









•

.



•







I








•



I



•



N

•

i








•



•







i

• •







i





i










•







•




•

J

8 -


•o
01
o
o 6
o
CD
01
0)
a
ฃ 4
o>
E
o

-------
                            Cumulative HUD Funding ($ per Child)
400-


^^
^
~
0 300-
1
f/>
5*
O)
C
ง 200-
0
3
X
V

'I
1 100-
0


o-





i









1



I
!




I









1



I
F
Mn



•~


i
I




•







1

I
1



•







1

•
1


1









1




•











1



•


u



8:
:
eo
0)
— i
-a R l
o " .
o
CD
c
CO
0)
2
i 4-
E
%
O

0)
ง 2
"f
S:
Q.
0;

Legend: 2000-2002 2003-2004
	 2005-2006




















0 100 200 300 40

AII Years 2000-2002 2003-2004 2005-2006 Cumulative HUD Rinding ($ per Child)
Figure B.48. Cumulative HUD Funding: Histogram and Linear Relationship with Geometric Mean
             Blood Lead Levels by Time

Table B.48a. Summary Information for Cumulative HUD Funding by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
89.45
119.05
141.25
112.72
Standard
Error
0.53
0.84
0.96
0.44
Minim um
30.90
36.61
47.76
30.90
10th
Percentile
32.10
39.31
51.29
33.70
25th
Percentile
34.09
43.34
54.26
44.68
Median
64.37
109.39
128.54
84.54
75th
Percentile
119.49
153.89
184.80
153.89
90th
Percentile
206.34
290.03
341.59
226.35
Maximum
275.77
329.21
378.88
378.88
 Table B.48b. Model Information for the Relationship between Cumulative HUD Funding and Geometric
             Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
hf_cum
Intercept
time
hf_cum
Intercept
time
hf_cum
Intercept
time
hf_cum
Intercept
time
hf_cum
Estimate
2.634
-0.080
0.000
2.623
-0.079
0.000
-1.765
-0.131
-0.001
-3.673
-0.119
-0.001
-4.960
-0.090
0.000
Standard
Error
0.018
0.003
0.000
0.018
0.003
0.000
0.019
0.003
0.000
0.026
0.005
0.000
0.034
0.008
0.000
p-value
<.001
<.001
0.002
<.001
<.001
0.008
<.001
<.001
<.001
<.001
<.001
0.007
<.001
<.001
0.203
-2 Log
Likelihood
52287


48723


87163


139805


177444


Variance Components
Random Effects
a,,2 = 0.388
a212= -0.027
a222 = 0.004
a,,2= 0.379
a212= -0.026
a222= 0.004
a,,2= 0.375
a212= -0.006
a222= 0.004
a,,2= 0.545
a2]2= -0.023
a222= 0.004
a,,2= 0.492
a^j2= -0.029
a222= 0.003
Error
aerror2= 0.207
ฐerror2= 4.428



                                              Page B-48
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                              Current State Funding ($ per Child)
6-
-
"
-
5-
"D
~
ฃ.
O
| 4;

^
O)
c
"O - .
c J
LL
Si -
1 -
3
O
1 -
.
.
-
o-



•










i



•


i



•
•





















•



•










i



•


i



•
• •










•



•


f
• •










1 •



I I


r *
I










•



•


F
1


• XX
8:

•o
CD
O
TJ c
0 b
O
m
CD
9
S
o ,,
•- 4
s -
E -
0
0)
o :
-n
1 2-
'•5
0)
Q. I

.
o-

Legend: 2000-2002 ~ 2003-2004
	 2005-2006















	 , 	 	 — 777777777
-• — 	 ~"




1 	 | 	 	 | 1 I 	 1 | 1 III 	 | 1 ! 	 > I | 	 |
0123456

All Years 2000-2002 2003-2004 2005-2006 Current State Funding ($ per Child)
Figure B.49.  Current State Funding: Histogram and Linear Relationship with Geometric Mean Blood
              Lead Levels by Time

Table B.49a.  Summary Information for Current State Funding by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
2.18
2.18
2.18
2.18
Standard
Error
0.01
0.01
0.01
0.01
Minimum
0.00
0.00
0.00
0.00
10th
Percentile
0.92
0.92
0.92
0.92
25th
Percentile
0.92
0.92
0.92
0.92
Median
1.59
1.59
1.59
1.59
75th
Percentile
3.02
3.02
3.02
3.02
90th
Percentile
5.14
5.14
5.14
5.14
Maximum
5.14
5.14
5.14
5.14
 Table B.49b. Model Information for the Relationship between Current State Funding and Geometric
             Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
sf_cur
Intercept
time
sf_cur
Intercept
time
sf_cur
Intercept
time
sf_cur
Intercept
time
sf_cur
Estimate
2.649
-0.084
0.098
2.636
-0.082
0.104
-1.736
-0.139
0.117
-3.646
-0.126
0.127
-4.944
-0.095
0.099
Standard
Error
0.017
0.002
0.009
0.017
0.002
0.009
0.018
0.002
0.011
0.024
0.004
0.013
0.032
0.007
0.015
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
52162


48583


87014


139707


177504


Variance Components
Random Effects
a,,2= 0.363
Q212= -0.027
CT222 = 0.004
a,,2= 0.354
a212= -0.027
a222 = 0.004
CT,,2= 0.352
CT212= -0.008
cr222= 0.004
a,,2= 0.503
cr212= -0.022
cr222= 0.004
a112= 0.455
a212= -0.027
CT222= 0.003
Error
aem)r2= 0.207
^eTO,2= 4.430



                                              Page B-49
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                              Cumulative State Funding ($ per Child)
         300-
      O
        200-
      cn

      'o

      LL

      €
      Si
      %
      3
      E

      o
100-
          o-


                                               o
                                               O

                                               m

                                               CO
                                               (U
                                               •c  4
                                               o>

                                               I
                                               O
                                               ?
                                                       TS  2
                                                       '•5
                                                       01
                                                          o
                                                              Legend:
                                                                    2000-2002

                                                                    2005-2006
2003-2004
                                                                         100
                                                                               1 I '  ' '

                                                                               200
                                                                                                     300
           All Years
                      2000-2002
                                 2003-2004
                                            2005-2006
                                                                   Cumulative State Funding ($ per Child)
Figure B.50.  Cumulative State Funding: Histogram and Linear Relationship with Geometric Mean

              Blood Lead Levels by Time
Table B.SOa. Summary Information for Cumulative State Funding by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
75.33
97.10
114.55
92.76
Standard
Error
0.41
0.64
0.75
0.34
Minimum
0.00
0.00
0.00
0.00
10th
Percentile
29.38
39.48
46.82
33.06
25th
Percentile
35.81
44.08
51.42
44.99
Median
55.68
68.66
81.13
68.40
75th
Percentile
106.12
135.85
160.00
130.31
90th
Percentile
159.41
215.98
257.12
195.42
Maximum
205.79
246.90
288.05
288.05
 Table B.SOb. Model Information for the Relationship between Cumulative State Funding and Geometric

              Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
sf_cum
Intercept
time
sf_cum
Intercept
time
sf_cum
Intercept
time
sf_cum
Intercept
time
sf_cum
Estimate
2.695
-0.098
0.002
2.686
-0.097
0.002
-1.658
-0.163
0.003
-3.567
-0.150
0.002
-4.881
-0.113
0.002
Standard
Error
0.018
0.003
0.000
0.018
0.003
0.000
0.019
0.003
0.000
0.026
0.005
0.000
0.033
0.008
0.000
p-value
<.001
<001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
52201


48618


87044


139740


1 77460


Variance Components
Random Effects
CT,,2 = 0.371
CT212 = -0.028
aZ22 = 0.004
a ^2= 0.363
a212= -0.028

-------
                               Current CDC Funding ($ per Child)
5-
.
.
.
-
•— • 4-
2
!E
O
fl>
a

Q 3 -
O)
'•5
c
~
fc^
8 2-
O
'c
V
^
3
1
-
"
-
o-
*


















41









^


















It





• H
IH


8 -
•
\
•a
S :
_J
O
o
m -
c -
a 1
a>
S
•s 4;
ซ5 •
ฃ -
0
V
o :
a>
t> 2:
'•a
ฃ
a! '
-
o-




Legend: 2000-2002 ~ — 2003-2004
	 2005-2006














_
	 	 	 "
_^"







012345

           All Years
                     2000-2002
                                2003-2004
                                           2005-2006
                                                                    Current CDC Funding ($ per Child)
Figure B.51.  Current CDC Funding: Histogram and Linear Relationship with Geometric Mean Blood
              Lead Levels by Time
Table B.Sla. Summary Information for Current CDC Funding by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
0.65
0.58
0.56
0.61
Standard
&ror
0.01
0.00
0.00
0.00
Minimum
0.38
0.56
0.52
0.38
10th
Percentile
0.38
0.56
0.53
0.38
25th
Percentile
0.38
0.56
0.53
0.39
Median
0.39
0.60
0.59
0.53
75th
Percentile
0.44
0.60
0.59
0.59
90th
Percentile
0.44
0.60
0.59
0.60
Maximum
4.96
0.61
0.60
4.96
 Table B.Slb. Model Information for the Relationship between Current CDC Funding and Geometric
              Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
cf_cur
Intercept
time
cf_cur
Intercept
time
cf_cur
Intercept
time
cf_cur
Intercept
time
cf_cur
Estimate
2.649
-0.084
-0.006
2.637
-0.083
-0.007
-1.736
-0.140
0.004
-3.649
-0.125
0.016
-4.953
-0.092
0.048
Standard
Error
0.018
0.002
0.005
0.018
0.002
0.005
0.018
0.003
0.006
0.025
0.004
0.011
0.032
0.007
0.019
p-value
<.001
<.001
0.251
<.001
<.001
0.132
<.001
<.001
0.462
<.001
<.001
0.160
<.001
<.001
0.013
-2 Log
Likelihood
52288


48721


87152


139760


177456


Variance Components
Random Effects
at,2= 0.388
a212= -0.028

-------
                            Cumulative CDC Funding ($ per Child)
70-



•5- 60;
!E
o
o

^.
O)
c
'o
ง 40-
O
Q
O
o
> 30-
I
3
E
5 20 ;
.
•
10-


i
1
I
i
*
*
*
1


^



—
i





i
i
Ii
^
ปt
x
*
$


*

*

9 B
a



8-

i
•o
S ]
"o 6 -
o :
CQ
c •
TO
g
^
'C .
E :
8 :
o :
•o
ง 2i
1 !
ol '
o-

Legend: 2000-2002 2003-2004
	 2005-2006












	 , 	
	 	



10 20 30 40 50 60 70

AII Years 2ooo-2002 2003-2004 2005-2006 Cumulative CDC Funding ($ per Child)
Figure B.52. Cumulative CDC Funding: Histogram and Linear Relationship with Geometric Mean
Blood Lead Levels by Time
Table B.52a. Summary Information for Cumulative CDC Funding by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
24.72
30.51
35.09
29.34
Standard
Error
0.07
0.11
0.11
0.06
Minimum
19.23
24.31
28.97
19.23
10th
Percentile
19.63
24.33
28.99
20.41
25th
Percentile
20.42
25.53
30.17
22.39
Median
21.57
26.70
31.29
26.70
75th
Percentile
23.28
27.83
32.34
31.28
90th
Percentile
42.68
57.31
61.96
48.42
Maximum
56.16
60.82
65.27
65.27
 Table B.52b. Model Information for the Relationship between Cumulative CDC Funding and Geometric
             Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
cf_cum
Intercept
time
cf_cum
Intercept
time
cf_cum
Intercept
time
cf_cum
Intercept
time
cf_cum
Estimate
2.653
-0.085
0.001
2.644
-0.085
0.001
-1.701
-0.150
0.004
-3.605
-0.138
0.005
-4.908
-0.105
0.004
Standard
Error
0.019
0.003
0.001
0.019
0.003
0.001
0.021
0.004
0.001
0.029
0.006
0.002
0.036
0.009
0.002
p-value
<.001
<.001
0.590
<.001
<.001
0.284
<.001
<001
0.002
<.001
<.001
0.005
<.001
<.001
0.037
-2 Log
Likelihood
52293


48725


87136


139706


177331


Variance Components
Random Effects
a,f = 0.386
a212= -0.027
a222 = 0.004
an2= 0.377
a2,2= -0.026
a222= 0.004
a,* = 0.370
a2]2= -0.006
a222 = 0.004
a,f= 0.533
a212= -0.022
a222 = 0.004
a,,2= 0.481
a2,2= -0.028
a222= 0.003
Error
ฐenJ= 0.207

-------
                              Current Total Funding ($ per Child)
16

15
14
— 13
TJ
Z 12
O
i"
ซฃ 10
ฃ> 9
'•5 _
c 8
*• 7
$
5 6
c 5i
ฃ -
i- 4 -
3 :
3 =
2i
1 :

0:


*
*
*
•






i

•

1
I






1

I






X
•







1


1

1
•
•






1


I


•
*






1






8-


TJ
8 :
_i
? 6:
0
CD
(0 I
0>
s
I 4^

E
2 :
O
t> 2-
1 ;
Q.


Legend: 2000-2002 2003-2004
	 2005-2006











	



I'" 'I""!""!""!""!""!""!""!" "l""l""l""l""l""l ""l""l
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

All Years 2000-2002 2003-2004 2005-2006 Current Total Funding ($ per Child)
Figure B.53.  Current Total Funding: Histogram and Linear Relationship with Geometric Mean Blood
              Lead Levels by Time

Table B.53a.  Summary Information for Current Total Funding by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
5.99
5.36
5.40
5.64
Standard
Error
0.03
0.03
0.03
0.02
Minimum
0.78
1.90
1.13
0.78
10th
Percentile
2.22
2.92
2.56
2.37
25th
Percentile
2.70
3.33
3.47
3.04
Median
6.08
3.86
5.08
5.06
75th
Percentile
7.54
7.30
6.26
6.92
90th
Percentile
13.62
11.91
11.88
11.91
Maximum
15.11
12.90
11.95
15.11
 Table B.53b. Model Information for the Relationship between Current Total Funding and Geometric
             Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
tot_cur
Intercept
time
tot_cur






Intercept
time
tot_cur
Estimate
2.646
-0.083
0.007
2.633
-0.081
0.007






-4.950
-0.092
0.015
Standard
Error
0.018
0.002
0.002
0.017
0.002
0.002






0.032
0.007
0.006
p-value
<.001
<.001
0.004
<001
<.001
<.001






<.001
<.001
0.011
-2 Log
Likelihood
52283


48712








177377


Variance Components
Random Effects
0^2= 0.386
a2]2= -0.028
o222 = 0.004
a,,2 = 0.377
a212= -0.027
cr222 = 0.004






a(12= 0.482
a212= -0.028
a222= 0.003
Error
aerror2 = 0.207
ฐeTO,2= 4.428



                                              Page B-53
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                            Cumulative Total Funding ($ per Child)
800-
"

700;

2
0 600-
1 :
4ft
"ro 50ฐ-
c
'•D
= 400;
1- 300 ;
o
1
= 200-
O
100-
,
o-


#







•


i


•
}







•


i


•




*
1

?
1
i
> '.
5
> \
>
1 i jt
*i

^M
I I I


All Years 2000-2002 2003-2004 2005-2



















8;


CO
o

E 6
0
m
c
CO
0)
S
1 4
i
o
•a
ซ 0
o •*-
1
V
n
o:

Legend: 2000-2002 2003-2004
	 2005-2006









__________




i 	 r ' 	 i " 	 i • " 	 i 	 ' ' • 'i 	 ' • • i 	 i 	 ' • ' f i
0 100 200 300 400 500 600 700 80

006 Cumulative Total Funding ($ per Child)
Figure B.54  Cumulative Total Funding: Histogram and Linear Relationship with Geometric Mean
             Blood Lead Levels by Time

Table B.54a. Summary Information for Cumulative Total Funding by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
1 89.50
246.65
290.89
234.82
Standard
Error
0.93
1.45
1.67
0.78
Minimum
50.16
60.96
76.76
50.16
10th
Percentile
98.67
129.72
158.20
107.16
25th
Percentile
1 1 1 .69
143.08
170.95
136.41
Median
149.00
211.00
250.20
178.67
75th
Percentile
205.16
277.83
324.24
273.48
90th
Percentile
408.44
563.32
660.66
449.63
Maximum
537.61
636.83
732.14
732.14
 Table B.54b. Model Information for the Relationship between Cumulative Total Funding and Geometric
             Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
tot_cum
Intercept
time
tot_cum
Intercept
time
tot_cum






Estimate
2.662
-0.088
0.000
2.653
-0.087
0.000
-1.715
-0.146
0.000






Standard
Error
0.018
0.003
0.000
0.018
0.003
0.000
0.020
0.004
0.000






p-value
<.001
<.001
0.014
<.001
<.001
0.002
<.001
<.001
0.018






-2 Log
Likelihood
52292


48722


87146








Variance Components
Random Effects
a1(2= 0.384
a212 = -0.028
cr222 = 0.004
a,,2 = 0.376
a2]2= -0.027
cr222 = 0.004
a,,2= 0.372
a212= -0.006
a222= 0.004






Error
aenor2= 0.207
<^r2= 4.429



                                              Page B-54
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                          Current HUD Funding ($ per Census Tract)
9ooo:
:
8000-
tf
2 :
H 7000 -
in
c
 2-
'•6
ฃ
a '.
o-

Legend: 2000-2002 2003-2004
	 2005-2006











— 	 — 	 ^— ^^_
~^^^- 	 ^^^ 	 -^ 	 	
"~ — 	
~~~


0 1000 2000 3000 4000 5000 6000 7000 8000 900
All Years 2000-2002 2003-2004 2005-2006 Current HUD Rjnding ($ per Census Tract)
Figure B.55. Current HUD Funding: Histogram and Linear Relationship with Geometric Mean Blood
             Lead Levels by Time

Table B.55a. Summary Information for Current HUD Funding by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
1291
1043
1070
1157
Standard
Error
10
8
9
6
Minimum
6
19
9
6
10th
Percentile
112
329
205
159
25th
Percentile
190
484
379
339
Median
942
766
783
817
75th
Percentile
1990
1329
1434
1636
90th
Percentile
3110
2133
2369
2639
Maximum
8932
7758
6737
8932
 Table B.55b. Model Information for the Relationship between Current HUD Funding and Geometric
             Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
hf_cur
Intercept
time
hf_cur
Intercept
time
hf_cur
Intercept
time
hf_cur
Intercept
time
hf_cur
Estimate
2.649
-0.084
0.000
2.636
-0.082
0.000
-1.736
-0.139
0.000
-3.645
-0.127
0.000
-4.937
-0.095
0.000
Standard
Error
0.018
0.002
0.000
0.018
0.002
0.000
0.018
0.002
0.000
0.025
0.004
0.000
0.032
0.007
0.000
p-value
<.001
<.001
0.450
<.001
<001
0.666
<.001
<.001
0.133
<.001
•e.001
0.116
<.001
<.001
0.007
-2 Log
Likelihood
52303


48737


87168


1 39723


177172


Variance Components
Random Effects
a^2= 0.386
Q2]2= -0.027
o222= 0.004
a,*= 0.379
a212= -0.027
cr222 = 0.004
a,,2= 0.377
a212= -0.006
cr222 = 0.004
an2= 0.535
a212= -0.022
a222 = 0.004
au2= 0.478
CT212= -0.028
CT222= 0.003
Error
^TOr2= 0.207
ฐeTOr2= 4.428



                                              Page B-55
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                       Cumulative HUD Funding ($ per Census Tract)
500000 -

1 :
u 400000-
3

J"
•
ffl
ฐ- 300000-
^
O)
c
TJ
C

"- 200000-
Q
I
S
-fi
•5 100000-
1 '-_
o-



















I


I
















|


i





















1
!










i

•
\ i












1
1


1 !















1



1
8-
-
•a
(0
4)
T3 e J
0 b .
O
m
C •
8 :
s
." 4:
L_ .
ฃi
0)
E
o
0)
o
"S
•^-* O
O
'•5
ฃ

o:

Legend: 2000-2002 2003-2004
	 2005-2006














— ^_^_^_^_^^
~~~~^ 	 	 	 — - 	 .
— — - 	 •' 	 	



























0 100000 200000 300000 400000 500000
Ail Years 2000-2002 2003-2004 2005-2006 Cumulative HUD Finding ($ per Cens us Tract)
Figure B.56.  Cumulative HUD Funding: Histogram and Linear Relationship with Geometric Mean
             Blood Lead Levels by Time

Table B.56a.  Summary Information for Cumulative HUD Funding by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10^50
10417
36484
Number
Missing
0
0
0
0
Mean
36034
47907
56925
45399
Standard
Error
249
395
461
209
Minimum
5oe
64c
767
506
10th
Percentile
9068
11430
14467
10723
25th
Percentile
14238
18231
22734
17129
Median
25860
35641
42697
32913
75th
Percentile
48422
66264
78390
61557
90th
Percentile
75560
97864
1 1 5487
95537
Maximum
299162
357143
411041
411041
 Table B.56b. Model Information for the Relationship between Cumulative HUD Funding and Geometric
             Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
hf_cum
Inte-cept
time
hf_cum
Intercept
time
hf_cum



Intercept
time
hf cum
Estimate
2.642
-0.082
0.000
2.629
-0.080
0.000
-1.766
-0.131
0.000



-4.962
-0.088
0.000
Standard
Error
0.018
0.002
0.000
0.018
0.002
0.000
0.019
0.003
0.000



0.033
0.008
0.000
p-value
<.001
<.001
<.001
<001
<.001
<.001
<.001
<001
<.001



<.001
<.001
<.001
-2 Log
Likelihood
52307


48740


87098





177128


Variance Components
Random Effects
a,,2= 0.386
0^= -0.028
a222= 0.004
a,,2= 0.378
a212= -0.026
a222 = 0.004
a,,2= 0.369
a212= -0.006
a222= 0.004



a,,2= 0.488
a2l2 = -0.029
CT222= 0.003
Error
ฐem,2= 0.207
CTeTOr2= 4'428



                                              Page B-56
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                           Current State Funding ($ per Census Tract)
       6000-
       5000-
    H
    
    s
    I
       4000-
       3000-
     c
     3
       2000-
    ฃ
    o
       1000-
          o-
             *
             *

 x
 t
                                   *
                                   *
*
*
                                              X
                                              *
*
*
9
           All Years
                     2000-2002
                                2003-2004
                                           2005-2006
                                                                1000   2000   3000   4000   5000

                                                               Current State Funding ($ per Census Tract)
                                                                                                  6000
Figure
       B.57.  Current State Funding: Histogram and Linear Relationship with Geometric Mean Blood
              Lead Levels by Time
Table B.57a.  Summary Information for Current State Funding by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
887
886
887
887
Standard
&ror
6
7
7
4
Minimum
0
0
0
0
10th
Percentile
244
244
244
244
25th
Percentile
406
404
405
405
Median
565
663
665
664
75th
Percentile
1162
1162
1162
1162
90th
Percentile
1863
1863
1863
1863
Maximum
5579
5579
5579
5579
 Table B.57b. Model Information for the Relationship between Current State Funding and Geometric
              Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
sf_cur
Intercept
time
sf_cur
Intercept
time
sf_cur
Intercept
time
sf_cur
Intercept
time
sf_cur
Estimate
2.652
-0.084
0.000
2.639
-0.083
0.000
-1.737
-0.139
0.000
-3.651
-0.126
0.000
-4.953
-0.094
0.000
Standard
Error
0.018
0.002
0.000
0.018
0.002
0.000
0.018
0.002
0.000
0.025
0.004
0.000
0.032
0.007
0.000
p-value
<001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
52232


48659


87178


140076


178005


Variance Components
Random Effects
a,,2= 0.390
a212= -0.029
a222= 0.004
a,,2= 0.381
a212= -0.028
a222= 0.004
a1(2= 0.373
a212= -0.008
a222= 0.004
a,,2= 0.536
a2,2= -0.023
a222 = 0.004
a,,2= 0.487
a212= -0.029
cr222 = 0.003
Error

-------
                        Cumulative State Funding ($ per Census Tract)
400000 -

^
s
n
3
w 300000-
O
,_
D.
^.
0)
~ 200000-
c
3
LL
tt)
75
55
0>
.2 100000-
J5
3
1 '
o-




















i
i
















i
!




















i
!










t
i
\ i











1
I
\ \















I

1
8-
|
•o
n
^ :
O ^
o !
CQ -
(0 -
0>
• — 4 -
i_ ^ .
•5
E '
o
o> :

•a
0)
t> 2:
'•a
fll
ฃ
a.
0:

Legend: 2000-2002 2003-2004
	 2005-2006












	 	 — 	 — ^. 	 _^ — ^- "~ ~
_^— ~- — "~ — T7^--^~ !-"—--''' —
___---""^^^ — -^-" ^-
"'"*"' _i- •ซ— ~* ~ — * ~*~
^j"

























i 	 i ' ' i ' ' i ' 'i
0 100000 200000 300000 400000
AII Years 2000-2002 2003-2004 2005-2006 Cumulative State Finding (S per Census Tract)
Figure B.58. Cumulative State Funding: Histogram and Linear Relationship with Geometric Mean
Blood Lead Levels by Time
Table B.58a. Summary Information for Cumulative State Funding by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
30603
39407
46543
37676
Standard
Error
200
312
369
166
Minimum
0
0
0
0
10th
Percentile
8328
10885
12792
9779
25th
Percentile
13888
18024
21278
16575
Median
23041
29732
35126
28002
75th
Percentile
39888
51403
60844
48882
90th
Percentile
63538
82026
97123
78225
Maximum
223170
267804
312438
312438
Table B.58b. Model Information for the Relationship between Cumulative State Funding and Geometric
              Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
sf cum
Intercept
time
sf_cum
Intercept
time
sf_cum
Intercept
time
sf_cum
Intercept
time
sf_curn
Estimate
2.689
-0.096
0.000
2.679
-0.095
0.000
-1.684
-0.155
0.000
-3.606
-0.140
0.000
-4.919
-0.105
0.000
Standard
Error
0.018
0.002
0.000
0.018
0.002
0.000
0.019
0.003
0.000
0.026
0.005
0.000
0.033
0.008
0.000
p-value
<.001
<.001
<.001
<.001
<.001
•e.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<001
-2 Log
Likelihood
52211


48638


87223


140117


178019


Variance Components
Random Effects
0^2= 0.387
CT912= -0.028
CJ222 = 0.004
0^2= 0.378
a212= -0.027
CT222 = 0.004
ฃ7^2= 0.371
CT?12= -0.007
cr222 = 0.004
a^2= 0.533
CT212= -0.023
tJ222 = 0.004
a1t2= 0.485
a212= -0.029
a222 = 0.003
Error
<*eTOf2= 0.207
ฐ^?= 4-429



                                                Page B-58
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                           Current CDC Funding ($ per Census Tract)
     in
    8
     I
    O)
    '•5
    c
    LL
    o
    o
    o
       6000
       5000-
       4000-
       30001
       2000 H
       1000-
*
*

*
*
                                    *
                                    *
                                    *
                                    *
                                               *
                                               *

                                                          8-

o
CO
CO
o>


I4
o
0)
O
                                                          oH
                                                             Legend:
                                                                            2000-2002
                                                                            2005-2006
                                      2003-2004
           All Years
                      2000-2002
                                 2003-2004
                                            2005-2006
                                                                 1000   2000   3000   4000   5000

                                                                 Current CDC Funding ($ per Census Tract)
                                                                                                    6000
Figure
       B.59.  Current CDC Funding: Histogram and Linear Relationship with Geometric Mean Blood
              Lead Levels by Time
Table B.59a. Summary Information for Current CDC Funding by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
1 0450
10417
36484
Number
Missing
0
0
0
0
Mean
887
886
887
887
Standard
Error
6
7
7
4
Minimum
0
0
0
0
10th
Percentile
244
244
244
244
25th
Percentile
406
404
405
405
Median
665
663
665
664
75th
Percentile
1162
1162
1162
1162
90th
Percentile
1863
1863
1863
1863
Maximum
5579
5579
5579
5579
 Table B.59b. Model Information for the Relationship between Current CDC Funding and Geometric
              Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
cf_cur
Intercept
time
cf_cur
Intercept
time
cf_cur



Intercept
time
cf_ctr
Estimate
2.649
-0.084
0.000
2.637
-0.083
0.000
-1.735
-0.140
0.000



-4.948
-0.093
0.000
Standard
Error
0.018
0.002
0.000
0.018
0.002
0.000
0.018
0.002
0.000



0.032
0.007
0.000
p-value
<.001
<.001
0.043
<.001
<.001
0.015
<.001
<001
0.764



<.001
<.001
0.390
-2 Log
Likelihood
52297


48730


87163





177463


Variance Components
Random Effects
a,,2= 0.388
a212= -0.028
a222 = 0.004
a,,2= 0.380
a212= -0.027
<7222 = 0.004
a,,2= 0.374
0212= -0.006
cr222= 0.004



a,,2= 0.487
a2,2= -0.029
CT222= 0.003
Error
ฐam,2= 0.207
aerror2 = 4.427



                                                Page B-59
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                        Cumulative CDC Funding ($ per Census Tract)
400000 -
'
?•

"
H
W
w 300000 -
<ง
1
&•
O)
.E 200000 -
"D

LL
fj
ซ
0
(1)
•I 100000-
I
3
1 :
o-



















i


i













1


1





















•
!









l

•
\ i









|

1


i (
















!
8:
-
•o
as
9>
—1
"O c
0 ฐ "
O
m
c -
S :
•- 4:
^;
0)
O
B
O
•o
0)
o 2
1
C
Q.

o:

Legend: 2000-2002 2003-2004
	 2005-2006












_^— — — T^^:^^ '' '" ~~ '
___---'"'" 1/1- — -^ *~
"* """* —T*—~^^^
' 	 "'




























0 100000 200000 300000 400000
All Years 2000-2002 2003-2004 2005-2006 Cumulative CDC Finding ($ per Census Tract)
Figure B.60. Cumulative CDC Funding: Histogram and Linear Relationship with Geometric Mean
             Blood Lead Levels by Time
Table B.60a. Summary Information for Cumulative CDC Funding by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
30603
39407
46543
37676
Standard
Error
200
312
369
166
Minimum
0
0
0
0
10th
Percentile
8328
10885
12792
9779
25th
Percentile
13888
18024
21278
16575
Median
23041
29732
35126
28002
75th
Percentile
39888
51403
60844
48882
90th
Percentile
63538
82026
97123
78225
Maximum
223170
267804
312438
312438
 Table B.60b. Model Information for the Relationship between Cumulative CDC Funding and Geometric
              Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
cf cum
Intercept
time
cf_cum
Intercept
time
cf cum
Intercept
time
cf_cum
Intercept
time
cf_cum
Estimate
2.641
-0.082
0.000
2.628
-0.080
0.000
-1.752
-0.135
0.000
-3.667
-0.119
0.000
-4.961
-0.087
0.000
Standard
Error
0.018
0.003
0.000
0.018
0.003
0.000
0.020
0.003
0.000
0.026
0.005
0.000
0.033
0.008
0.000
p-value
<.001
<.001
0.142
<.001
<.001
0.153
<.001
<.001
0.024
<.001
<.001
0.022
<.001
<.001
0.046
-2 Log
Likelihood
52304


48737


87122


139571


176972


Variance Components
Random Effects
0^2 = 0.385
Q212= -0.028
0222 = 0.004
0^2 = 0.377
cr212= -0.026
o2/= 0.004
a,,2= 0.371
a212= -0.006
CT222= 0.004
a,,2= 0.536
a212= -0.022
a222 = 0.004
0^2= 0.482
a212= -0.029
a222= 0.004
Error
^r2= 0-207
CT8TO,2= 4-428



                                               Page B-60
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                          Current Total Funding ($ per Census Tract)
30000 -
-

?•

1-
(A
3
C
,5 20000-
1

0)
'a
C
u.
5 10000-
^
_
t.
o-









*
*
*








•
!








1
1









=+ *
* *
* *






1

!






•
I
! 1
*






i i
\ \






i
\
8:

•a
CO
9
•a e
o b
0
m
i
&
~ 4
E
I
O
•a
S 2
'•5
ฃ
a.
0:

Legend: 2000-2002 • •" 2003-2004
	 2005-2006












	 	 	 	 —
~- 	 '"* " 	 — 	 =s 	
- — ' 	

























0 10000 20000 30000
           All Years
                     2000-2002
                                2003-2004
                                          2005-2006
                                                              Current Total Funding ($ per Census Tract)
Figure B.61.  Current Total Funding: Histogram and Linear Relationship with Geometric Mean Blood
              Lead Levels by Time

Table B.61a.  Summary Information for Current Total Funding by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
3065
2814
2843
2930
Standard
Error
18
21
20
11
Minimum
13
59
49
13
10th
Percentile
884
900
925
902
25th
Percentile
1467
1390
1484
1446
Median
2491
2183
2389
2359
75th
Percentile
4071
3510
3594
3769
90th
Percentile
6055
5632
5237
5751
Maximum
20091
18917
17896
20091
 Table B.61b. Model Information for the Relationship between Current Total Funding and Geometric
              Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
toLcur
Intercept
time
tot_cur
Intercept
time
tot_cur
Intercept
time
tot_cur
intercept
time
tot_cur
Estimate
2.648
-0.084
0.000
2.635
-0.082
0.000
-1.738
-0.139
0.000
-3.648
-0.126
0.000
-4.945
-0.094
0.000
Standard
Error
0.018
0.002
0.000
0.018
0.002
0.000
0.018
0.003
0.000
0.025
0.004
0.000
0.032
0.007
0.000
p-value
<.001
<.001
0.495
<.001
<.001
0.199
<.001
<.001
0.007
<.001
<.001
0.659
<.001
<.001
0.810
-2 Log
Likelihood
52303


48736


87174


139791


177371


Variance Components
Random Effects
a1(2 = 0.388
a212= -0.028
o222= 0.004
a,,z = 0.380
a212 = -0.027
a222= 0.004
a,,2 = 0.379
a?12= -0.006
a222 = 0.004
o^= 0.542
Q212= -0.022
a222= 0.004
0^2= 0.488
a2,2= -0.029
CT222= 0.003
Error
aerror2= 0.207
CTenor2= 4'428



                                               Page B-61
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality-
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                        Cumulative Total Funding ($ per Census Tract)
1100000;
.
"
•^ 1000000 :
o :
E
•- 900000 -
w
in -
c 800000 :
O :
ซ 700000 :
o> 600000 -
c
'•5 :
c 500000 ;

u- :
 200000 :
E --
3
0 looooo:
o:



















i


!

















i


i







y












i

!










•
•

\ \











•
i


















i


•

8-



•o
(0
_i
I ^
CO
c •
S :
•— 4 '
t>
E
o
4)
O
~o
*~* o
O
'•5
C
0.

0

Legend: 2000-2002 2003-2004
	 2005-2006













	 , 	 	 ; — -.- — -^-S----- 	
••"" J^-j-— 	 	
•



























1 	 1 1 1— 1 1 	 1 1 1 I 1 1 1 1 1 1 1
0 500000 1000000 1500000
Alivears 2000-2002 2003-2004 2005-2006 Cumulative Total Funding ($ per Census Tract)
Figure B.62. Cumulative Total Funding: Histogram and Linear Relationship with Geometric Mean
Blood Lead Levels by Time
Table B.62a. Summary Information for Cumulative Total Funding by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15617
10450
10417
36484
Number
Missing
0
0
0
0
Mean
97240
126721
150011
120752
Standard
Error
301
944
1104
502
Minimum
1115
2563
2967
1115
10th
Percentile
31087
40888
49211
36712
25th
Percentile
48955
64246
76790
58823
Median
77293
101723
121431
94927
75th
Percentile
119183
155541
1 84701
1 50549
90th
Percentile
190075
247844
291289
236432
Maximum
745502
892751
1035917
1035917
 Table B.62b. Model Information for the Relationship between Cumulative Total Funding and Geometric
             Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
tot_cum
Intercept
time
tot_cum
Intercept
time
tot_eum
Intercept
time
tot_cum
Intercept
time
tot_cum
Estimate
2.664
-0.089
0.000
2.653
-0.087
0.000
-1.731
-0.141
0.000
-3.646
-0.126
0.000
-4.944
-0.095
O.OOC
Standard
Error
0.018
0.003
0.000
0.018
0.003
0.000
0.019
0.003
0.000
0.026
0.005
0.000
0.033
0.008
0.000
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
52300


48732


87180


139780


177448


Variance Components
Random Effects
CT,,2 = 0.389
cr212= -0.028
CT222 = 0.004
0^2= 0.380
cr21^= -0.027
a222 = 0.004
a,,2 = 0.374
a212= -0.006
CT222 = 0.004
cr]12= 0.541
a212= -0.022
cr222 = 0.004
a,,2= 0.488
a212= -0.029
cr222 = 0.003
Error
aerlor2= 0.207
o.mr2= 4-429



                                              Page B-62
This information is distributed solely for the purpose ofpre -dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
    100
     80
  s  60
  i
  to
  a.
     40
     20
                                   TRI Compounds (Total Air)
                                                        8-
                                                     T3
                                                     O
                                                     _o
                                                     m
                                                     c
                                                     3
                                                     o>
•S
I
o
o>
O
•o
0)
                                                        4
                                                        0^
                                                           Legend:
                   - 2000-2002
                 	2004-2006
2002-2004
            180  360  540   720  900 1080 1260 1440

                  Lead Compounds-Total Air
         200   400  600   800  1000  1200  1400  1600

                Lead Compounds-Total Air
Figure B.63.  TRI Compounds (Total Air): Histogram and Linear Relationship with Geometric Mean
              Blood Lead Levels by Time

Table B.63a.  Summary Information for TRI Compounds (Total Air) by Time
Time
Period
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
10417
10432
11749
32598
Number
Missing
0
0
0
0
Mean
3.68
3.68
3.67
3.68
Standard
Error
0.53
0.53
0.50
0.30
Minimum
0.00
0.00
0.00
0.00
10th
Percentile
0.00
0.00
0.00
0.00
25th
Percentile
0.00
0.00
0.00
0.00
Median
0.00
0.00
0.00
0.00
75th
Percentile
0.00
0.00
0.00
0.00
90th
Percentile
0.00
0.00
0.00
0.00
Maximum
1500.15
1500.15
1500.15
1500.15
Table B.63b.  Model Information for the Relationship between TRI Compounds (Total Air) and
              Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
air_tot
Intercept
time
air_tot
Intercept
time
air_tot
Intercept
time
air_tot
Intercept
time
air tot
Estimate
2.649
-0.084
0.000
2.636
-0.082
0.000
-1.735
-0.140
-0.001
-3.647
-0.126
-0.001
-4.945
-0.094
-0.001
Standard
Error
0.018
0.002
0.000
0.018
0.002
0.000
0.018
0.002
0.000
0.025
0.004
0.000
0.032
0.007
0.000
p-value
<.001
<.001
0.644
<.001
<.001
0.655
<.001
<.001
0.069
<.001
<.001
0.022
<.001
<.001
0.151
-2 Log
Likelihood
52295


48729


87154


139761


1 77396


Variance Components
Random Effects
a,,*= 0.387
a?12= -0.027
CT222 = 0.004
a,,2= 0.379
a212= -0.026
cr222 = 0.004
a,,2= 0.373
a212= -0.006
CT222 = 0.004
a,,2= 0.538
a2]2= -0.022
a222 = 0.004
a,,2= 0.487
a212= -0.029
a222 = 0.003
Error
ฐeTOr2= 0.207
*.™or2= 4'428



                                               Page B-63
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                 TRI Compounds (Fugitive Air)
    100
     80
  _  60
  d>
  g
  O
  a.
     40
     20
                                                       8-
                                                    T3
                                                     CO
                                                     0)
CO
ง
1  4
I
o
O
•a
                                                       o^
                                                           Legend:
                   - 2000-2002
                  -- 2004-2006
2002-2004
        0  61218 24 30 36 42 48 54 60 66 72 78 84 90 96

                 Lead Compounds-Fugitive Air
        10  20  30   40   50   60   70   80   90   100

               Lead Compounds-Fugitive Air
Figure B.64. TRI Compounds (Fugitive Air): Histogram and Linear Relationship with Geometric Mean
             Blood Lead Levels by Time

Table B.64a. Summary Information for TRI Compounds (Fugitive Air) by Time
Time
Period
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
10417
10432
11749
32598
Number
Missing
0
0
0
0
Mean
0.33
0.34
0.34
0.34
Standard
Error
0.04
0.04
0.04
0.03
Minimum
0.00
0.00
0.00
0.00
10th
Percentile
0.00
0.00
0.00
0.00
25th
Percentile
0.00
0.00
0.00
0.00
Median
0.00
0.00
0.00
0.00
75th
Percentile
0.00
0.00
0.00
0.00
90th
Percentile
0.00
0.00
0.00
0.00
Maximum
98.50
98.50
98.50
98.50
Table B.64b.  Model Information for the Relationship between TRI Compounds (Fugitive Air) and
              Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
air_fug
Intercept
time
airjug
Intercept
time
airjug
Intercept
time
airjug
Intercept
time
airjug
Estimate
2.649
-0.084
-0.005
2.636
-0.082
-0.005
-1.735
-0.140
-0.009
-3.648
-0.126
-0.012
-4.946
-0.094
-0.009
Standard
Error
0.018
0.002
0.003
0.017
0.002
0.003
0.018
0.002
0.004
0.025
0.004
0.005
0.032
0.007
0.005
p-value
<.001
<.001
0.079
<.001
<001
0.069
<.001
<.001
0.024
<.001
<.001
0.014
<.001
<.001
0.110
-2 Log
Likelihood
52288


48721


87146


1 39750


1 77400


Variance Components
Random Effects
a1(2 = 0.386
a212 = -0.027
a222= 0.004
a,* = 0.377
a212= -0.026
a222= 0.004
a,,2= 0.373
a212= -0.006
a222= 0.004
an2= 0.541
cr212= -0.023
cr222 = 0.004
a,,2= 0.488
a212= -0.029
CT222 = 0.003
Error
ฐ81TOr2= 0.207
ฐerror2 = 4.428



                                               Page B-64
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                            TRI Compounds-Air Lead from Stacks
    100
     80
  I
  o
  0)
  Q.
     40
     20
                                                       8-
                                                       6-
                                                    O
                                                    CD

                                                    a
                                                    to
E
o
i
o
T3
0)
                                                       2
                                                       o-
      Legend:
                                                                        2000-2002
                                                                        2004-2006
        2002-2004
            180  360  540  720  900  1080 1260 1440

                  Lead Compounds-Stacks
     0   200   400   600   800
1000  1200  1400  1600
                                                                    Lead Compounds-Stacks
Figure B.65. TRI Compounds (Air Lead from Stacks): Histogram and Linear Relationship with
             Geometric Mean Blood Lead Levels by Time

Table B.65a. Summary Information for TRI Compounds (Air Lead from Stacks) by Time
Time
Period
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
10417
10432
11749
32598
Number
Missing
0
0
0
0
Mean
3.34
3.34
3.33
3.34
Standard
Error
0.53
0.53
0.50
0.30
Minimum
0.00
0.00
0.00
0.00
10th
Percentile
0.00
0.00
0.00
0.00
25th
Percentile
0.00
0.00
0.00
0.00
Median
0.00
0.00
0.00
0.00
75th
Percentile
0.00
0.00
0.00
0.00
90th
Percentile
0.00
0.00
0.00
0.00
Maximum
1500.15
1500.15
1500.15
1500.15
Table B.65b. Model Information for the Relationship between TRI Compounds (Air Lead from Stacks)
             and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
air_stk
Intercept
time
air_stk
Intercept
time
air_stk
Intercept
time
air_stk
Intercept
time
air_stk
Estimate
2.649
-0.084
0.000
2.636
-0.082
0.000
-1.735
-0.140
-0.001
-3.647
-0.126
-0.001
-4.945
-0.094
-0.001
Standard
Error
0.018
0.002
0.000
0.018
0.002
0.000
0.018
0.002
0.000
0.025
0.004
0.000
0.032
0.007
0.000
p-value
<.001
<.001
0.750
<.001
<001
0.765
<.001
<001
0.099
<.001
<.001
0.035
<.001
<.001
0.186
-2 Log
Likelihood
52296


48729


87154


139762


177396


Variance Components
Random Effects
a, ,2 = 0.387
a212= -0.027
cr222 = 0.004
0-^2 = 0.379
a212= -0.026
CT222 = 0.004
a^2= 0.374
ap12 = -0.006
cr222= 0.004
0.538
-0.022
0.004
a, ,2= 0.488
a2]2= -0.029
a222= 0.003
Error
ฐerror2 = 0.207
ฐeTOr2= 4.428



                                              Page B-65
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                 TRI Compounds-Water Surface
    100
     80
  _  60
  0)
  g
  0)
  Q_

     40
     20
                                                        8-
                                                     CO
                                                     0)
                                                     "O  ft
                                                     1
                                                     CO
•S  4
I
o
1
TJ
0)
                                                           Legend:
                     2000-2002
                     2004-2006
2002-2004
        0.1 0.9  1.7 2.5 3.3  4.1  4.9 5.7  6.5 7.3 8.1  8.9

                Lead Compounds-Water Surface
             2345678

              Lead Compounds-Water Surface
        10
Figure B.66.  TRI Compounds (Water Surface): Histogram and Linear Relationship with Geometric
              Mean Blood Lead Levels by Time

Table B.66a.  Summary Information for TRI Compounds (Water Surface) by Time
Time
Period
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
10417
10432
11749
32598
Number
Missing
0
0
0
0
Mean
0.02
0.02
0.02
0.02
Standard
Error
0.00
0.00
0.00
0.00
Minimum
0.00
0.00
0.00
0.00
10th
Percentile
0.00
0.00
0.00
0.00
25th
Percentile
0.00
0.00
0.00
0.00
Median
0.00
0.00
0.00
0.00
75th
Percentile
0.00
0.00
0.00
0.00
90th
Percentile
0.00
0.00
0.00
0.00
Maximum
9.50
9.50
9.50
9.50
Table B.66b.  Model Information for the Relationship between TRI Compounds (Water Surface) and
              Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
water_surf
Intercept
time
water_surf
Intercept
time
water_surf
Intercept
time
water_surf
Intercept
time
water_surf
Estimate
2.649
-0.084
0.007
2.636
-0.082
0.011
-1.735
-0.140
-0.030
-3.647
-0.126
0.008
-4.946
-0.094
0.034
Standard
Error
0.018
0.002
0.039
0.018
0.002
0.038
0.018
0.002
0.050
0.025
0.004
0.058
0.032
0.007
0.059
p-value
<.001
<.001
0.848
<.001
<.001
0.781
<.001
<.001
0.557
•c.001
<.001
0.896
<.001
•=.001
0.560
-2 Log
Likelihood
52286


48719


87147


1 39753


1 77392


Variance Components
Random Effects
CT,,2 = 0.387
ay,2= -0.027
cr222 = 0.004
an2= 0.379
a212= -0.026
a222= 0.004
a,,2= 0.374
a212= -0.006
a222 = 0.004
a,,2= 0.541
a212 = -0.022
CT222 = 0.004
al12= 0.489
a212= -0.029
a222= 0.003
Error
ฐe™,2= 0.207
ฐeror2= 4.428



                                               Page B-66
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
    100
     80
  s  60
  <ป
  Q.
     40
     20
TRI Lead Only (Total Air)
                    o -
                        Legend:
                                                      •o
                                                      CD
                  O 6
                  o
                  m

                  |


                  | 4
                  o
                  i
                  O
                  •D
                  2 2
                  '•5
                  0)
                                                         o-
                                                                         - 2000-2002
                                                                     	2004-2006
2002-2004
         0  32  64  96  128 160 192  224 256 288 320

                      Lead Only-Total Air
                                100        200       300

                                    Lead Only-Total Air
        400
Figure B.67.  TRI Lead Only (Total Air): Histogram and Linear Relationship with Geometric Mean
              Blood Lead Levels by Time

Table B.67a.  Summary Information for TRI Lead Only (Total Air) by Time
Time
Period
2000-2002
2002-2004
2004-2006
All Years
Combined
Sample
Size
10417
10432
11749
32598
Number
Missing
0
0
0
0
Mean
1.27
1.27
1.27
1.27
Standard
Error
0.15
0.15
0.14
0.08
Minimum
0.00
0.00
0.00
0.00
10th
Percentile
0.00
0.00
0.00
0.00
25th
Percentile
0.00
0.00
0.00
0.00
Median
0.00
0.00
0.00
0.00
75th
Percentile
0.00
0.00
0.00
0.00
90th
Percentile
0.00
0.00
0.00
0.00
Maximum
336.50
336.50
336.50
336.50
Table B.67b.  Model Information for the Relationship between TRI Lead Only (Total Air) and Geometric
              Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
airjot
Intercept
time
air_tot
Intercept
time
air_tot
Intercept
time
air_tot
Intercept
time
air_tot
Estimate
2.649
-0.084
0.001
2.636
-0.082
0.001
-1.735
-0.140
0.002
-3.647
-0.126
0.002
-4.946
-0.094
0.001
Standard
Error
0.018
0.002
0.001
0.018
0.002
0.001
0.018
0.002
0.001
0.025
0.004
0.001
0.032
0.007
0.001
p-value
<.001
<.001
0.149
<.001
<.001
0.127
<.001
<.001
0.090
<.001
<.001
0.166
<.001
<.001
0.426
-2 Log
Likelihood
52291


48724


87153


139763


177405


Variance Components
Random Effects
onz= 0.387
a212 = -0.028
a222= 0.004
0^2 = 0.379
a212= -0.026
a222= 0.004
a,,2= 0.374
a212= -0.006
a222 = 0.004
0-^2 = 0,540
a212= -0.022
a222= 0.004
0^2 = 0.488
a212= -0.029
CT222= 0.003
Error
aerror2= 0.207
aerror2 = 4.428



                                                Page B-67
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                   TRI Lead Only (Fugitive Air)
    100
     80
  ^  60
  0>

  0.
     40
     20
         r1	i"	1	1	1	1	1	1	1	1	1—


         0  32  64  96  128  160 192 224 256 288 320



                     Lead Only-Fugitive Air
                                                       T3
                                                       o
                                                       _o

                                                       CO


                                                       ง
o
0)

O
•
0)
                                                          2
                                                          o-
                                                              Legend:
                    - 2000-2002

                  	2004-2006
                                                                                              2002-2004
               100        200        300



                   Lead Only-Fugitive Air
                                               400
Figure B.68.  TRI Lead Only (Fugitive Air): Histogram and Linear Relationship with Geometric Mean

              Blood Lead Levels by Time
Table B.68a. Summary Information for TRI Lead Only (Fuj
Time
Period
2000-2002
2002-2004
2004-2006
All Years
Combined
Sample
Size
10417
10432
11749
32598
Number
Missing
0
0
0
0
Mean
0.68
0.68
0.68
0.68
Standard
Error
0.12
0.12
0.11
0.07
Minimum
0.00
0.00
0.00
0.00
10th
Percentile
0.00
0.00
0.00
0.00
*itive Air) by Time
25th
Percentile
0.00
0.00
0.00
0.00
Median
0.00
0.00
0.00
0.00
75th
Percentile
0.00
0.00
0.00
0.00
90th
Percentile
0.00
0.00
0.00
0.00
Maximum
336.00
336.00
336.00
336.00
Table B.68b. Model Information for the Relationship between TRI Lead Only (Fugitive Air) and

              Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
air fug
Intercept
time
air fug
Intercept
time
air fug
Intercept
time
air fug
Intercept
time
air fug
Estimate
2.649
-0.084
0.002
2.636
-0.082
0.002
-1.735
-0.140
0.003
-3.647
-0.126
0.003
-4.946
-0.094
0.002
Standard
Error
0.018
0.002
0.001
0.017
0.002
0.001
0.018
0.002
0.001
0.025
0.004
0.002
0.032
0.007
0.002
p-value
<.001
<.001
0.081
<.001
<.001
0.066
<001
<.001
0.056
<.001
<.001
0.048
<.001
<.001
0.199
-2 Log
Likelihood
52290


48723


87151


139759


177404


Variance Components
Random Effects
a,,2= 0.386
a212= -0.027
a222= 0.004
a,,2= 0.378
a212 = -0.026
CT222 = 0.004
an2= 0.373
a212= -0.006
a222 = 0.004
a,,2= 0.539
a212 = -0.022
cr222 = 0.004
a,,2= 0.488
a2,2= -0.029
a222= 0.003
Error
ฐeTO,2= 0.207
ฐซ^= 4-428



                                                 Page B-68

This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality

guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency

determination or policy.

-------
                              TRI Lead Only-Air Lead from Stacks
    100
     80
     6ฐ
  O
  a.
     40
     20
                                                        8-
T3
as

-------
                                  TRI Lead Only-Water Surface
100




80





^ 60 '
I
5
a

40 '





20 '
O-*-1





















^
\
\
\
0 0.3 0.6 0.9 1.2 1.5 1.8 2.1 2.4
8:


i
0)
—1
•a fi
o *>
0
ffl
c
8
•= 4
A
0)
E
o
4)
a
•a
V
T5 2
'•a
ฃ
a.
o-


Legend: 2000-2002 '" 2002-2004
	 2004-2006









^^~~-
^___ 	
^ —
^— - — ' — — " " - - ~ ~^- ^-~"~
_-— 	 	 ,_---~,Z^-~"'J "" — "
.^^^^^^^^ .- — •* ** *" Z-"'- — "^ " "
—**^^ - _-""l^ 1T-^^"" '
_ - - ^_~Ji __ 	 • ~" "
— •^ *



l ' ' ' ' ' ' ' ' ' i ' ' ' 	 [—i—i—i—i 	 |
0123
                   Lead Only-Water Surface
                                                                      Lead Only-Water Surface
Figure B.70.  TRI Lead Only (Water Surface): Histogram and Linear Relationship with Geometric
              Mean Blood Lead Levels by Time

Table B.70a.  Summary Information for TRI Lead Only (Water Surface) by Time
Time
Period
2000-2002
2002-2004
2004-2006
All Years
Combined
Sample
Size
10417
10432
11749
32598
Number
Missing
0
0
0
0
Mean
0.00
0.00
0.00
0.00
Standard
Error
0.00
0.00
0.00
0.00
Minimum
0.00
0.00
0.00
0.00
10th
Percentile
0.00
0.00
0.00
0.00
25th
Percentile
0.00
0.00
0.00
0.00
Median
0.00
0.00
0.00
0.00
75th
Percentile
0.00
0.00
0.00
0.00
90th
Percentile
0.00
0.00
0.00
0.00
Maximum
2.60
2.60
2.60
2.60
Table B.70b.  Model Information for
              Geometric Mean Blood
the Relationship between TRI Lead Only (Water Surface) and
Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
water_surf
Intercept
time
water_surf
Intercept
time
water_surf
Intercept
time
water_surf
Intercept
time
water_surf
Estimate
2.649
-0.084
0.441
2.636
-0.082
0.421
-1.735
-0.140
0.449
-3.647
-0.126
0.497
-4.946
-0.094
0.314
Standard
Error
0.018
0.002
0.151
0.017
0.002
0.153
0.018
0.002
0.195
0.025
0.004
0.219
0.032
0.007
0.234
p-value
<.001
<.001
0.004
<.001
<.001
0.006
<.001
<.001
0.021
<.001
<.001
0.023
<.001
<.001
0.179
-2 Log
Likelihood
52274


48708


87138


139746


1 77386


Variance Components
Random Effects
a,,2= 0.385
CT212 = -0.027
a222 = 0.004
a,,z= 0.377
a212= -0.026
a222= 0.004
0^2= 0.373
a?12= -0.006
CT222= 0.004
a,,2= 0.540
a2]2= -0.023
0222 = 0.004
CT,,2= 0.489
a212= -0.029
d222= 0.003
Error
ฐem,2= 0.207
ฐ.nor2= 4-428



                                               Page B-70
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                                     TRI Total Lead (Total Air)
    100
     80
„  60 '

I
Q.
   40
     20
                                                      T3
                                                      8
                                                      _
                                                      CD
                                                      c
                                                      CO
                                                      0)
                                                         6-
                                                      •= 4 -


                                                      I
                                                      O
                                                      S
                                                         2-
                                                     •
                                                     o
                                                         0
                                                           Legend:       - 2000-2002
                                                                    	2004-2006
                                                                                            2002-2004
             180  360  540  720  900 1080 1260 1440

                      Lead Total-Total Air
                                                              200   400  600

                                                                       Lead Total-Total Air
     "T"  '"I"'  "'i'"
800  1000  1200  1400 1600
Figure B.71.  TRI Total Lead (Total Air): Histogram and Linear Relationship with Geometric Mean
              Blood Lead Levels by Time

Table B.71a.  Summary Information for TRI Total Lead (Total Air) by Time
Time
Period
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
10417
10432
11749
32598
Number
Missing
0
0
0
0
Mean
4.80
4.80
4.80
4.80
Standard
Error
0.55
0.55
0.52
0.31
Minimum
0.00
0.00
0.00
0.00
10th
Percentile
0.00
0.00
0.00
0.00
25th
Percentile
0.00
0.00
0.00
0.00
Median
0.00
0.00
0.00
0.00
75th
Percentile
0.00
0.00
0.00
0.00
90th
Percentile
0.00
0.00
0.00
0.00
Maximum
1500.15
1500.15
1500.15
1500.15
Table B.71b.  Model Information for the Relationship between TRI Total Lead (Total Air) and
              Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
air_tot
Intercept
time
air_tot
Intercept
time
air_tot
Intercept
time
air_tot
Intercept
time
air_tot
Estimate
2.649
-0.084
0.000
2.636
-0.082
0.000
-1.735
-0.140
0.000
-3.647
-0.126
-0.001
-4.945
-0.094
0.000
Standard
Error
0.018
0.002
0.000
0.018
0.002
0.000
0.018
0.002
0.000
0.025
0.004
0.000
0.032
0.007
0.000
p-value
<.001
<.001
0.956
<.001
<.001
0.986
<.001
<.001
0.194
<.001
<.001
0.071
<.001
<.001
0.254
-2 Log
Likelihood
52296


48729


87155


139760


177392


Variance Components
Random Effects
a,,2 = 0.387
a?12 = -0.027
a222 = 0.004
au2= 0.379
a212= -0.026
a222 = 0.004
a,,2 = 0.374
a212= -0.006
o222 = 0.004
a,,2 = 0.539
a212= -0.022
a222 = 0.004
a,,2= 0.488
cr2,2 = -0.029
CT222 = 0.003
Error
ฐ8TOr2= 0.207
ฐeno,2= 4-428



                                                PageB-71
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
    100
     80
„  60
m
o
o>
a
   40
     20
                                   TRI Total Lead (Fugitive Air)
                                                         8-
                                                      •a
                                                      ซo
                                                      o
                                                      CD

                                                      (0
                                                      •s  4;
                                                      O
                                                      a)
                                                    TJ
                                                    a>
                                                         o^
                                                            Legend:
                                                                        - 2000-2002
                                                                      	2004-2006
                        2002-2004
            32  64  96  128 160 192 224 256 288 320

                    Lead Total-Fugitive Air
                                                         M
                                                         0
100       200        300

    Lead Total-Fugitive Air
400
Figure B.72.  TRI Total Lead (Fugitive Air): Histogram and Linear Relationship with Geometric Mean
              Blood Lead Levels by Time

Table B.72a.  Summary Information for TRI Total Lead (Fugitive Air) by Time
Time
Period
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
10417
10432
11749
32598
Number
Missing
0
0
0
0
Mean
0.97
0.98
0.97
0.97
Standard
Error
0.12
0.12
0.12
0.07
Minimum
0.00
0.00
0.00
0.00
10th
Percentile
0.00
0.00
0.00
0.00
25th
Percentile
0.00
0.00
0.00
0.00
Median
0.00
0.00
0.00
0.00
75th
Percentile
0.00
0.00
0.00
0.00
90th
Percentile
0.00
0.00
0.00
0.00
Maximum
336.00
336.00
336.00
336.00
Table B.72b.  Model Information for the Relationship between TRI Total Lead (Fugitive Air) and
              Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
alrjfug
Intercept
time
air_fug
Intercept
time
air_fug
Intercept
time
air_fug
Intercept
time
air_fug
Estimate
2.649
-0.084
0.001
2.636
-0.082
0.001
-1.735
-0.140
0.002
-3.647
-0.126
0.002
-4.945
-0.094
0.001
Standard
Error
0.018
0.002
0.001
0.017
0.002
0.001
0.018
0.002
0.001
0.025
0.004
0.002
0.032
0.007
0.002
p-value
<.001
<.001
0.265
<.001
<001
0.237
<.001
<.001
0.268
<.001
<.001
0.240
<.001
<.001
0.411
-2 Log
Likelihood
52292


48725


87154


139762


177402


Variance Components
Random Effects
an2= 0.387
0^2 = -0.027
o222 = 0.004
a1(2 = 0.378
a212= -0.026
a222= 0.004
an2 = 0.374
ap12 = -0.006
a222= 0.004
an2 = 0.540
0^2 = -0.022
crZ22 = 0.004
a,,2= 0.488
a212= -0.029
CT222 = 0.003
Error
ฐ.ro,z= ฐ-207
ฐeTOr2= 4.428



                                                Page B-72
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality-
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
    100
     80
  _  60
  O
  0.
     40
     20
                              TRI Total Lead-Air Lead from Stacks


                                                        8-
                                                      s
m


CO
o>
ฃ  4

%


O
o>
O
                                                        2-
                                                        o-
      Legend:
                                                                          2000-2002
             2002-2004
                                                                    ---- 2004-2006
            180  360  540  720   900  1080  1260 1440


                      Lead Total-Stacks
         200  400  600


                   Lead Total-Stacks
          '"i1"

800  1000  1200  1400  1600
Figure B.73.  TRI Total Lead (Air Lead from Stacks): Histogram and Linear Relationship with

              Geometric Mean Blood Lead Levels by Time



Table B.73a.  Summary Information for TRI Total Lead (Air Lead from Stacks) by Time
Time
Period
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
10417
10432
11749
32598
Number
Missing
0
0
0
0
Mean
3.83
3.83
3.82
3.83
Standard
Error
0.53
0.53
0.50
0.30
Minimum
0.00
0.00
0.00
0.00
10th
Percentile
0.00
0.00
0.00
0.00
25th
Percentile
0.00
0.00
0.00
0.00
Median
0.00
0.00
0.00
0.00
75th
Percentile
0.00
0.00
0.00
0.00
90th
Percentile
0.00
0.00
0.00
0.00
Maximum
1500.15
1500.15
1500.15
1500.15
Table B.73b.  Model Information for the Relationship between TRI Total Lead (Air Lead from Stacks)

              and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
air_stk
Intercept
time
air_stk
Intercept
time
air_stk
Intercept
time
air_stk
Intercept
time
air_stk
Estimate
2.649
-0.084
0.000
2.636
-0.082
0.000
-1.735
-0.140
-0.001
-3.647
-0.126
-0.001
-4.945
-0.094
-0.001
Standard
Error
0.018
0.002
0.000
0.018
0.002
0.000
0.018
0.002
0.000
0.025
0.004
0.000
0.032
0.007
0.000
p-value
<.001
<.001
0.753
<.001
<001
0.773
<.001
<.001
0.108
<.001
<.001
0.029
<.001
<.001
0.167
-2 Log
Likelihood
52296


48729


87154


139761


177393


Variance Components
Random Effects
CT112= 0.387
a?12 = -0.027

-------
                                  TRI Total Lead-Water Surface
    100
     80
  ^  60

  I
  0.
     40
     20
                                                     •o
                                                     s
                                                     _o
                                                     m
•g  4
3
E
o
|

I
a.
                                                        o-
       Legend:    	 2000-2002
                	2004-2006
                                                                                           2002-2004
        0.1 0.9  1.7 2.5 3.3 4.1 4.9  5.7 6.5 7.3  8.1 8.9

                   Lead Total-Water Surface
                 34567

                 Lead Total-Water Surface
                                              10
Figure B.74.  TRI Total Lead (Water Surface): Histogram and Linear Relationship with Geometric
              Mean Blood Lead Levels by Time

Table B.74a.  Summary Information for TRI Total Lead (Water Surface) by Time
Time
Period
2000-2002
2002-2004
2004-2006
All Years
Sample
Size
10417
10432
11749
32598
Number
Missing
0
0
0
0
Mean
0.02
0.02
0.02
0.02
Standard
Error
0.00
0.00
0.00
0.00
Minimum
0.00
0.00
0.00
0.00
10th
Percentile
0.00
0.00
0.00
0.00
25th
Percentile
0.00
0.00
0.00
0.00
Median
0.00
0.00
0.00
0.00
75th
Percentile
0.00
0.00
0.00
0.00
90th
Percentile
0.00
0.00
0.00
0.00
Maximum
9.50
9.50
9.50
9.50
Table B.74b.  Model Information for the Relationship between TRI Total Lead (Water Surface) and
              Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
water_surf
Intercept
time
water_surf
Intercept
time
water surf
Intercept
time
w ater surf
Intercept
time
water surf
Estimate
2.649
-0.084
0.019
2.636
-0.082
0.021
-1.735
-0.140
-0.014
-3.647
-0.126
0.019
-4.946
-0.094
0.046
Standard
Error
0.018
0.002
0.038
0.018
0.002
0.038
0.018
0.002
0.050
0.025
0.004
0.058
0.032
0.007
0.058
p-value
<.001
<.001
0.626
<.001
<001
0.582
<.001
<.001
0.777
<.001
<.001
0.737
<.001
<.001
0.427
-2 Log
Likelihood
52285


48719


87147


139753


177393


Variance Components
Random Effects
a,,2= 0.387
a212= -0.027
a222= 0.004
a,,2 = 0.379
a2,2= -0.026
a222 = 0.004
a1(2= 0.374
a212= -0.006
a222= 0.004
a,*= 0.541
a212 = -0.022
CT222 = 0.004
a1(2= 0.489
a?12= -0.029
CT222= 0.003
Error
^2= 0.207
ฐe™2 = 4.428



                                                Page B-74
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
     PI: Proportion of Housing Units Passing MA Standard of Care:  Naive Method 1
*- 0.7-
TJ
O
ฃ
0)
S 0.6
o>
i
g 0.5;

Q.
0 0.1 -
f
O
a.
o
a 0.0-

j




i

!








i


E

ป




•
i
!








i


!

i





s
!
I











!

k





f
!
1








, i


! !

1= =1




.
I
i
•
\








t 1


i i

N




L
1
6
-
!








i


!
                                                    E
                                                    o
                                                    m
                                                    ro
                                                    0

                                                     o
                                                       8-
6-
                                                    t3 2:
                                                    TJ
                                                    0)
                                                           Legend:
                  2000-2002
                  2005-2006
2003-2004
           All Years
                     2000-2002
                               2003-2004
                                          2005-2006
                                                        0.0   0.1    0.2    0.3    0.4   0.5   0.6    0.7

                                                        Proportion Passing MA Standard of Care: NaTve Method 1
Figure B.75. PI: Proportion of Housing Units Passing MA Standard of Care: Naive Method 1:
             Histogram and Linear Relationship with Geometric Mean Blood Lead Levels by Time

Table B.75a. Summary Information for PI: Proportion of Housing Units Passing MA Standard of
             Care; Naive Method 1 by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15614
10446
10413
36473
Number
Missing
3
4
4
11
Mean
0.04
0.05
0.05
0.05
Standard
Error
0.00
0.00
0.00
0.00
Minimum
0.00
0.00
0.00
0.00
10th
Percentile
0.01
0.01
0.01
0.01
25th
Percentile
0.01
0.01
0.01
0.01
Median
0.02
0.03
0.03
0.02
75th
Percentile
0.06
0.06
0.06
0.06
90th
Percentile
0.11
0.12
0.12
0.11
Maximum
0.66
0.66
0.66
0.66
Table B.75b. Model Information for the Relationship between PI:  Proportion of Housing Units Passing
             MA Standard of Care: Naive Method 1 and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
pi
Intercept
time
Pj
Intercept
time
pi
Intercept
time
P1
Intercept
time
Pj
Estimate
2.655
-0.086
1.626
2.643
-0.084
1.667
-1.724
-0.143
2.425
-3.630
-0.131
2.795
-4.930
-0.100
2.931
Standard
Error
0.017
0.002
0.192
0.017
0.002
0.190
0.017
0.003
0.260
0.024
0.004
0.311
0.031
0.007
0.325
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
52214


48643


87070


1 39775


177816


Variance Components
Random Effects
0-,^= 0.371
a212= -0.028
a222 = 0.004
an2= 0.363
a212= -0.027
a222= 0.004
0-^2 = 0.344
CT2,2 = -0.006
cr222= 0.004
a,,2= 0.492
a212= -0.022
a222= 0.004
0^2= 0.434
CT212= -0.027
a222= 0.003
Error
ฐerror2 = ฐ-207
CT9TO,2= 4-430



                                              Page B-75
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
     Fl: Proportion of Housing Units Failing MA Standard of Care:  Nai've Method 1
     o
     I
    
-------
                 Nl: Proportion of Housing Units Assessed: Naive Method 1
,- 0.30 :
"O •
0 0.28 :
.!_
4-/
g 0.26 :
B
> 0.24 i
1 i
0.22 :
ID
Fj 0.20:
O
"S 0.18:
I ฐ-16l
c n 14 ^
S ' ':
rf 0.12:
I
1) ฐ'10^
= 0.08:
'co :
LL
C :
•| 0.04:
O :
g- 0.02:
1- I
a 0.00 :
>

ซ
I










1




5
1

ซ
1










1




!
1

1
>









1





5
•

!
S :
1|

3






I

I



! 5



ซ
ft
N

:








' •



! 5















1



3
8-


•o

-------
     P2:  Proportion of Housing Units Passing MA Standard of Care: Naive Method 2
     •o
     o
ง
        0.8-
        0.7-
        0.6-
     1  0.4;
     <0
        0.3-
     O
     & 0.2-1
     O
        0.1 -
       *
       *
        o.o -L
          All Years
                  >t
                  4
                                  *
                                             *
                                             *
*
*
i         I        I
                                                       8-
                                                     (0
                                                     d>
                                                    m

                                                     TO
                                                     0)
                                               2  4
                                               I

                                               O
                                               0)
                                               C3
                                               •o
                                               ซ>  .
                                                     •D
                                                     V
                                                       o-
                                                           Legend:
                                                                    2000-2002
                                                                    2005-2006
                                             2003-2004
                                                          | I I I I I

                                                         0.0
II I | I I I I I M I I [ I I

 0.1   0.2
                     2000-2002
                                2003-2004
                                          2005-2006
        0.3   0.4   0.5   0.6

Proportion Assessed: Naive Method 1
                                                                                       rrprr

                                                                                       0.7
                                                     0.8
Figure B.78. P2: Proportion of Housing Units Passing MA Standard of Care: Nai've Method 2:
             Histogram and Linear Relationship with Geometric Mean Blood Lead Levels by Time

Table B.78a. Summary Information for P2:  Proportion of Housing Units Passing MA Standard of
             Care: Naive Method 2 by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15614
10446
10413
36473
Number
Missing
3
4
4
11
Mean
0.04
0.04
0.05
0.04
Standard
Error
0.00
0.00
0.00
0.00
Minimum
0.00
0.00
0.00
0.00
10th
Percentile
0.01
0.01
0.01
0.01
25th
Percentile
0.01
0.01
0.01
0.01
Median
0.02
0.02
0.03
0.02
75th
Percentile
0.05
0.06
0.06
0.05
90th
Percentile
0.09
0.11
0.12
0.10
Maximum
0.66
0.66
0.66
0.66
Table B.78b.  Model Information for the Relationship between P2: Proportion of Housing Units Passing
              MA Standard of Care: Naive Method 2 and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
P2
Intercept
time
P2
Intercept
time
P2
Intercept
time
P2
Intercept
time
P2
Estimate
2.659
-0.087
1.146
2.647
-0.086
1.167
-1.717
-0.145
1.834
-3.618
-0.135
2.534
-4.912
-0.106
2.995
Standard
Error
0.017
0.002
0.172
0.017
0.002
0.170
0.018
0.003
0.242
0.024
0.004
0.309
0.031
0.007
0.335
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
52241


48671


87101


1 39768


177801


Variance Components
Random Effects
a,,2 = 0.375
cr212 = -0.028
CT222 = 0.004
a,,2= 0.367
a212= -0.027
a222= 0.004
a,,2= 0.350
a?12= -0.007
a222 = 0.004
a1(2= 0.498
CT212= -0.023
a222= 0.005
at12= 0.434
a212= -0.028
a222 = 0.004
Error
"error2 = 0.207
ฐeno,2= 4.431



                                               Page B-78
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
     F2: Proportion of Housing Units Failing MA Standard of Care:  Naive Method 2
CM 0.30
H
o 0.28
5 0.26
.> 0.24
1
0.22
a>
a ฐ-2ฐ
•5 ฐ-18
1 ฐ-16
1 ฐ-14
; 0.12
0) ฐ'1ฐ
c
•- 0.08 :
'5 :
"• 0.06 i
c
•| 0.04:
g- 0.02:
l_
ฐ- 0.00 :


*
*

*











!











!


>t
* *
it
*F
*








i


{






i

i


1 !






•
i




I i





i




i
8 j
"
:
To :
0)
— J
o 6^
o
CD •
CO
0> j
S
1 4"
1 !
s :
o :
1 2;
•~
0)
u_
D.

0

Legend: 2000-2002 2003-2004
	 2005-2006





^^-~-"~"
^^^"
^^^-^ -
^^^-^-^^ ~~~"
~^~~~~





\ 	 i 	 i 	 i 	 1 1 1 1 1 	 1 1 	 |
00 0.05 0.10 0.15 0.20 0.25 0.3
All Years 2000-2002 2003-2004 2005-2006 Proportion Failing MA Standard of Care: Naive Method 2
re B.79. F2: Proportion of Housing Units Failing MA Standard of Care: Naive Method 2:
             Histogram and Linear Relationship with Geometric Mean Blood Lead Levels by Time

Table B.79a. Summary Information for F2:  Proportion of Housing Units Failing MA Standard of Care:
             Naive Method 2 by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15614
10446
10413
36473
Number
Missing
3
4
4
11
Mean
0.03
0.03
0.04
0.03
Standard
Error
0.00
0.00
0.00
0.00
Minimum
0.00
0.00
0.00
0.00
10th
Percentile
0.01
0.01
0.01
0.01
25th
Percentile
0.01
0.01
0.01
0.01
Median
0.02
0.03
0.03
0.02
75th
Percentile
0.04
0.05
0.05
0.04
90th
Percentile
0.06
0.07
0.08
0.07
Maximum
0.28
0.26
0.20
0.28
Table B.79b. Model Information for the Relationship between F2: Proportion of Housing Units Failing
             MA Standard of Care: Naive Method 2 and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
f2
Intercept
time
f2
Intercept
time
f2
Intercept
time
(2
Intercept
time
f2
Estimate
2.667
-0.089
3.155
2.652
-0.087
2.835
-1.702
-0.149
5.190
-3.585
-0.145
8.138
-4.884
-0.116
8.450
Standard
Error
0.017
0.002
0.329
0.017
0.002
0.319
0.017
0.003
0.426
0.023
0.004
0.537
0.030
0.007
0.588
p- value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
52200


48645


87046


1 39861


1 78375


Variance Components
Random Effects
a,,2= 0.358
a212= -0.028
a222 = 0.004
0-,^= 0.352
a212= -0.027
CJ222 = 0.004
a,,2= 0.316
c212= -0.006
a222= 0.004
a^2= 0.425
a212= -0.021
a222 = 0.005
o,,2= 0.364
a212= -0.026
CT222= 0.005
Error
peTOf2= 0.208
ฐerro,2= ^.434



                                             Page B-79
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                 N2: Proportion of Housing Units Assessed:  Nai've Method 2
0.8;
0.7-
•0
O
I ฐ-6;
^z
z
TJ
$ 0.4;

-------
     P3: Proportion of Housing Units Passing MA Standard of Care: Naive Method 3
ซ 0.7-
TJ
O
s 0.6-
0)
i
i> 0.5-
.5
10.4
re
1
(/> 0.3-
O)
(0
CO
Q.
o o.1 -
0
Q.
2
a 0.0 -
j




1



1
1
k



"




I
I
>


$
i
:
'
i



i
(
* *


r
I

(
H



i
•
i e
i< >

.
ง
3




i
i
; E
1=

L,
!
i
j
i



i

1
                                                        8
                                                     T3
                                                     n
                                                     0)
                                                     o
                                                     m
                                                     CD
                                                     B
                                                     s

                                                     14
                                                     o
                                                     O
                                                     T3
                                                     S 2
                                                           Legend:
                     2000-2002
                     2005-2006
2003-2004
           All Years
                     2000-2002
                                2003-2004
                                          2005-2006
                                                         0.0   0.1    0.2    0.3    0.4    0.5    0.6    0.7

                                                        Proportion Passing MA Standard of Care: Naive Method 3
Figure B.81.  P3: Proportion of Housing Units Passing MA Standard of Care: Naive Method 3:
              Histogram and Linear Relationship with Geometric Mean Blood Lead Levels by Time

Table B.Sla.  Summary Information for P3:  Proportion of Housing Units Passing MA Standard of
              Care; Na'i've Method 3 by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15614
10446
10413
36473
Number
Missing
3
4
4
11
Mean
0.04
0.04
0.05
0.04
Standard
&ror
0.00
0.00
0.00
0.00
Minimum
0.00
0.00
0.00
0.00
10th
Percentile
0.01
0.01
0.01
0.01
25th
Percentile
0.01
0.01
0.01
0.01
Median
0.02
0.02
0.03
0.02
75th
Percentile
0.05
0.06
0.06
0.05
90th
Percentile
0.09
0.11
0.12
0.10
Maximum
0.66
0.66
0.66
0.66
Table B.Slb.  Model Information for the Relationship
              MA Standard of Care: Naive Method 3
between P3:  Proportion of Housing Units Passing
and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
p3
Intercept
time
p3
Intercept
time
p3
Intercept
time
p3
Intercept
time
P3
Estimate
2.659
-0.087
1.146
2.647
-0.086
1.167
-1.717
-0.145
1.834
-3.618
-0.135
2.534
-4.912
-0.106
2.995
Standard
Error
0.017
0.002
0.172
0.017
0.002
0.170
0.018
0.003
0.242
0.024
0.004
0.309
0.031
0.007
0.335
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
52241


48671


87101


1 39768


177801


Variance Components
Random Effects
CT,,2= 0.375
a212= -0.028
cr222= 0.004
a^2= 0.367
a212= -0.027
a222= 0.004
a,,2= 0.350
ap12= -0.007
CT222= 0.004
a, ,2= 0.498
a212= -0.023
a222 = 0.005
an2= 0.434
a212= -0.028
cr222 = 0.004
Error

-------
     F3:  Proportion of Housing Units Failing MA Standard of Care: Naive Method 3
co 0.30 :
"O •
0 0.28:
1 :
5 0.26:
o
.> 0.24:
"RJ
Z 0.22 :
IP
in 0.20:
0
O ~
I 0.16 ;
~o -
8 ' '-
< 0.12 z
S 0.10:
O>
c :
= 0.08:
'5 :
c ฐ-06i
•| 0.04:
g- 0.02:
i_ I
a 0.00 :
A

;
ป
a
•










i







5
II Y

E
',
t




















I
ears

>
;

i
I

1
7
!






I





•

5
2000

|
!
i j
1 i
t H

3







I
1




•

! 5
-2002 2003



c
(
=

t















i s
-2004 2005
















I






2
-20
8:


•a
re
5 :
__
o 6^
o
CD -

CO -
a>
•
•ง I
E -
s :
0
S 2-
,*ฃ
~a
a>
^
a

o-
0
36

Legend: 2000-2002 	 	 2003-2004
	 2005-2006









^__^_- — - — "
~~^~~~~ "T-*"""**""*
^— — '""^•-^ '- - "-• •" "
^^^^^^^^
'w^1-^-'~'






\ 	 i 	 ' ' ' i 	 i 	 ' ' i 	 i 	 i
00 0.05 0.10 0.15 0.20 0.25 0.3
Proportion Failing MA Standard of Care: NaTve Method 3
Figure B.82.  F3: Proportion of Housing Units Failing MA Standard of Care: Naive Method 3:
             Histogram and Linear Relationship with Geometric Mean Blood Lead Levels by Time

Table B.82a.  Summary Information for F3: Proportion of Housing Units Failing MA Standard of Care:
             Naive Method 3 by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15614
10446
10413
36473
Number
Missing
3
4
4
11
Mean
0.04
0.04
0.04
0.04
Standard
Error
0.00
0.00
0.00
0.00
Minimum
0.00
0.00
0.00
0.00
10th
Percentile
0.01
0.01
0.01
0.01
25th
Percentile
0.02
0.01
0.01
0.01
Median
0.03
0.03
0.03
0.03
75th
Percentile
0.06
0.05
0.05
0.05
90th
Percentile
0.09
0.08
0.08
0.09
Maximum
0.30
0.27
0.21
0.30
Table B.82b. Model Information for the Relationship between F3:  Proportion of Housing Units Failing
             MA Standard of Care: Naive Method 3 and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
f3
Intercept
time
f3
Intercept
time
f3
Intercept
time
(3
Intercept
time
f3
Estimate
2.634
-0.079
4.312
2.621
-0.078
4.113
-1.756
-0.133
5.572
-3.678
-0.118
8.092
-4.985
-0.087
7.679
Standard
Error
0.017
0.002
0.313
0.016
0.002
0.302
0.016
0.003
0.346
0.023
0.004
0.451
0.030
0.007
0.488
p- value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
•=.001
-2 Log
Likelihood
52109


48548


86917


1 39803


178521


Variance Components
Random Effects
a,,2= 0.338
a212= -0.026
a222= 0.004
a,,2 = 0.328
a212= -0.025
cr222 = 0.004
a,,2= 0.290
a?12 = -0.004
a222= 0.004
a,,2= 0.391
a212= -0.016
a222= 0.004
an2= 0.339
a212= -0.021
a222 = 0.003
Error
crelror2 = 0.207
CTe,or2= 4'432



                                             Page B-82
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                 N3:  Proportion of Housing Units Assessed: Naive Method 3
     eo
      o
         0.8-
         0.7-
        0.6
     •o
      8 o-4 d
      in
      < 0.3-
      o
      r
      ง. 0.2 H
      o
        0.1 -
*
HI
        o.o-.
           All Years
+
*
+
                        i
*
I
ฃ
*
                     y        e
                                                        8 1
                                        •O  fi J
                                        16H
                                        o
                                        C
                                        5
                                                      O
                                        B  2:
                                        '•5
                                        V
                                                        o-
                                                            Legend:
                                                           - 2000-2002
                                                         -— 2005-2006
                                                        2003-2004
                                                         0.0
                     2000-2002
                                2003-2004
                                           2005-2006
                            0.1   0.2   0.3   0.4   0.5    0.6   0.7

                              Proportion Assessed: Naive Method 3
                                                                                                   0.8
Figure B.83.  N3: Proportion of Housing Units Assessed: Naive Method 3: Histogram and Linear
              Relationship with Geometric Mean Blood Lead Levels by Time

Table B.83a.  Summary Information for N3:  Proportion of Housing Units Assessed: Naive Method 3 by
              Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15614
10446
10413
36473
Number
Missing
3
4
4
11
Mean
0.0801
0.0841
0.0869
0.0832
Standard
Error
0.0006
0.0008
0.0009
0.0004
Minimum
0.0007
0.0007
0.0007
0.0007
10th
Percentile
0.0148
0.0161
0.0167
0.0159
25th
Percentile
0.0280
0.0297
0.0311
0.0292
Median
0.0549
0.0572
0.0583
0.0565
75th
Percentile
0.1037
0.1086
0.1114
0.1074
90th
Percentile
0.1792
0.1856
0.1926
0.1838
Maximum
0.7064
0.7113
0.7122
0.7122
Table B.83b.  Model Information for the Relationship between N3: Proportion of Housing Units
              Assessed: Naive Method 3 and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
n3
Intercept
time
n3
Intercept
time
n3
Intercept
time
n3
Intercept
time
n3
Estimate
2.658
-0.087
1.581
2.645
-0.085
1.632
-1.719
-0.145
2.648
-3.620
-0.135
3.071
-4.924
-0.104
2.943
Standard
Error
0.017
0.002
0.137
0.017
0.002
0.137
0.017
0.003
0.183
0.023
0.004
0.215
0.030
0.007
0.224
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
52161


48586


86996


1 39865


1 78257


Variance Components
Random Effects
a^2= 0.352
0^2= -0.027
CT222= 0.004
a,,2= 0.343
a212= -0.026
a222 = 0.004
an2= 0.305
a212= -0.004
a222 = 0.004
a,,2= 0.426
CT212= -0.018
a222= 0.004
^,2= 0.368
a212= -0.022
a222= 0.003
Error
aerror2 = 0.207
ฐ*™= 4-431



                                               Page B-83
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
     P4:  Proportion of Housing Units Passing MA Standard of Care: MDPH Method
1
I 0.6-
Q.
Q
5
0> 0.5
CD
O
•5 :
•o 0.4-
CB
T5
C
S
w 0.3-
S
O)
c
•55 0.2-
n
Q.
C
Proportio
o o
b '-*
*




i
i
0
I
™









•
\








1
1
* * *







i
5? )
*








i
1







i
•
i E
i
$

i








•
i
3 \







1

!
                                                       8-
                                                    o
                                                    o
                                                    m
                                                    CO
                                                    d>
                                                    o
                                                    o
                                                    CD
                                                    o
                                                    "S
                                                    I  2^
                                                    T3
                                                    0)
                                                          Legend:
2000-2002
2005-2006
                                                                                        2003-2004
          All Years
                     2000-2002
                               2003-2004
                                          2005-2006
                                                        0.0    0.1    0.2    0.3    0.4    0.5    0.6   0.7

                                                         Proportion Passing MA Standard of Care: MDPH Method
Figure B.84. P4:  Proportion of Housing Units Passing MA Standard of Care: MDPH Method:
             Histogram and Linear Relationship with Geometric Mean Blood Lead Levels by Time

Table B.84a. Summary Information for P4: Proportion of Housing Units Passing MA Standard of
             Care: MDPH Method by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15614
10446
10413
36473
Number
Missing
3
4
4
11
Mean
0.0381
0.0451
0.0497
0.0434
Standard
&ror
0.0004
0.0006
0.0006
0.0003
Minimum
0.0000
0.0000
0.0000
0.0000
10th
Percentile
0.0053
0.0065
0.0074
0.0061
25th
Percentile
0.0101
0.0122
0.0134
0.0116
Median
0.0195
0.0239
0.0267
0.0228
75th
Percentile
0.0483
0.0562
0.0613
0.0541
90th
Percentile
0.0906
0.1098
0.1229
0.1052
Maximum
0.6582
0.6598
0.6598
0.6598
Table B.84b. Model Information for the Relationship between P4: Proportion of Housing Units Passing
             MA Standard of Care: MDPH Method and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
p4
Intercept
time
p4
Intercept
time
p4
Intercept
time
p4
Intercept
time
p4
Estimate
2.659
-0.087
1.153
2.647
-0.086
1.174
-1.717
-0.145
1.834
-3.618
-0.135
2.570
-4.912
-0.106
3.019
Standard
Error
0.017
0.002
0.173
0.017
0.002
0.170
0.018
0.003
0.242
0.024
0.004
0.309
0.031
0.007
0.334
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
52240


48671


87099


1 39769


1 77809


Variance Components
Random Effects
an2 = 0.375
a?12= -0.028
a222= 0.004
a,,2 = 0.367
a212= -0.027
a222 = 0.004
an2 = 0.349
aP1z= -0.007
a222 = 0.004
a,,2= 0.496
a212= -0.023
cr222 = 0.005
a,,2= 0.433
a212= -0.028
a222 = 0.004
Error
aanJ= 0.207
ฐe.or2= 4.431



                                              Page B-84
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent any Agency
determination or policy.

-------
     F4: Proportion of Housing Units Failing MA Standard of Care: MDPH Method
•o ฐ-30:
o •
ฃ 0.28 :
o :
S 0.26:
I :
0 054:

0.22 z
dj
fe 0.20:
"o 0.18:
TJ :
n 0.16:
•o :
ง 0.14:
t/5
ซ-• 0.1 2 z
2
a 0.10 1
= 0.08:
ii?
c 0.06 :
o '•
'f 0.04 :
O :
0 0.02 z
ฐ~ 0.00 :
A
3

•
i
>











i




5
MY
1
i
ซ
I
i











|




!
ears
i
I
>
5











i





s
2000
j

!
f i
1 >
si

31








g

I



3 5
-2002 2003



<
e
<

c










•



! 5
2004 2005

















i



!
20C
8-

.
•a
0)
d)
	 I

o ฎ ~
o :j
CQ -
(0
0)
S
•— 4 ~

o
O
V
O
•a
S 2!
1 :
D.
o:
0
6

Legend: 2000-2002 2003-2004
	 2005-2006








^^^^
^ — ~~~~
^ 	 ' 	 ^ •
_^— -~"tI^--""'~""
^^~~^ZZ^~~~-~~~''~
•*-*~~^~\- •* *~ ~ ~^ •*--* — '^
^"-:''J"'



00 0.05 0.10 0.15 0.20 0.25 0.3
Proportion Failing MA Standard of Care: MDPH Method
Figure B.85.  F4: Proportion of Housing Units Failing MA Standard of Care: MDPH Method:
             Histogram and Linear Relationship with Geometric Mean Blood Lead Levels by Time

Table B.85a.  Summary Information for F4: Proportion of Housing Units Failing MA Standard of Care:
             MDPH Method by Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15614
10446
10413
36473
Number
Missing
3
4
4
11
Mean
0.0422
0.0391
0.0372
0.0399
Standard
&ror
0.0003
0.0003
0.0003
0.0002
Minimum
0.0000
0.0000
0.0000
0.0000
10th
Percentile
0.0078
0.0075
0.0073
0.0076
25th
Percentile
0.0151
0.0144
0.0137
0.0145
Median
0.0315
0.0300
0.0287
0.0302
75th
Percentile
0.0553
0.0512
0.0487
0.0521
90th
Percentile
0.0906
0.0827
0.0786
C.0854
Maximum
0.2960
0.2698
0.2066
0.2960
Table B.85b. Model Information for the Relationship between F4: Proportion of Housing Units Failing
             MA Standard of Care: MDPH Method and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
f4
Intercept
time
f4
Intercept
time
f4
Intercept
time
f4
Intercept
time
f4
Estimate
2.635
-0.080
4.351
2.622
-0.078
4.155
-1.756
-0.133
5.597
-3.677
-0.118
8.141
-4.984
-0.087
7.698
Standard
Error
0.017
0.002
0.314
0.016
0.002
0.303
0.016
0.003
0.348
0.023
0.004
0.454
0.030
0.007
0.491
p-value
<.001
<.001
<.001
<.001
<.001
<,001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
52107


48546


86920


1 39809


178518


Variance Components
Random Effects
an2= 0.338
a212= -0.026
o222= 0.004
0^,2= 0.329
a212= -0.025
a222= 0.004
a,,2= 0.290
a212= -0.004
a222= 0.004
a,,2= 0.391
a212= -0.016
cr222= 0.004
a,,2= 0.340
a212= -0.021
cr222= 0.003
Error
aerror2= 0.207
ฐeTO,2= 4.432



                                             Page B-85
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent any Agency
determination or policy.

-------
                N4: Proportion of Housing Units Assessed:  MDPH Method
0.8-
•

0.7-

TJ
o
IJ 0.6-
5 :
z
Q_
0 0.5
E-

1 0.4 ;
in

o
"D
* ซ "
o 2:
l_
0.
-
c

Legend: 2000-2002 2003-2004
	 2005-2006








^_^-—
^^^^^-
..^^-^^^^ -.-•'"""'^ -^"-"""— ""
_-r*J~f*~*^ __-*"~~* __--• — "" '
_-*•*******"**-• --"""""— — ^ "*""""'
-r^j^T''- ~ " " ~^ — " ~ ~^ """ "
-•'*"" __— f*~~^
~*~




.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 O.t
All Years 2000-2002 2003-2004 2005-2006 Proportion Assessed: MDPH Method
Figure B.86.  N4: Proportion of Housing Units Assessed: MDPH Method: Histogram and Linear
             Relationship with Geometric Mean Blood Lead Levels by Time

Table B.86a.  Summary Information for N4: Proportion of Housing Units Assessed:  MDPH Method by
             Time
Time
Period
2000-2002
2003-2004
2005-2006
All Years
Sample
Size
15614
10446
10413
36473
Number
Missing
3
4
4
11
Mean
0.0802
0.0842
0.0869
0.0833
Standard
Error
0.0006
0.0008
0,0009
0.0004
Minimum
0.0007
0.0007
0.0007
0.0007
10th
Percentile
0.0148
0.0161
0.0167
0.0159
25th
Percentile
0.0280
0.0297
0.0311
0.0292
Median
0.0550
0.0572
0.0583
0.0565
75th
Percentile
0.1037
0.1086
0.1114
0.1075
90th
Percentile
0.1805
0.1860
0.1926
0.1848
Maximum
0.7064
0.7113
0.7122
0.7122
Table B.86b. Model Information for the Relationship between N4:  Proportion of Housing Units
             Assessed: MDPH Method and Geometric Mean Blood Lead Levels
Model
Number
1
2
3
4
5
Factor
Intercept
time
n4
Intercept
time
n4
Intercept
time
n4



Intercept
time
n4
Estimate
2.658
-0.087
1.584
2.645
-0.085
1.636
-1.719
-0.145
2.656



-4.924
-0.104
2.943
Standard
Error
0.017
0.002
0.137
0.017
0.002
0.137
0.017
0.003
0.183



0.030
0.007
0.223
p- value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001



<.001
<.001
<.001
-2 Log
Likelihood
52160


48585


86995





1 78260


Variance Components
Random Effects
a,,2 = 0.352
a212= -0.027

-------
                          Appendix C

Detailed Exploratory Analyses of 95th and 99th Percentile Variables in
                        National Models

-------
95th percentile or above for TRI Lead Only (Stacks)
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tn_as1
Intercept
time
tn_as1



Intercept
time
tn_as1
Intercept
time
tn_as1
Intercept
time
tri_as1
Estimate
3.419
-0.155
-0.027
3.404
-0.156
-0.048



-3.326
-0169
0.325
-4.478
-0.161
0.383
-5.953
-0.141
0.332
Standard
Error
0.019
0.003
0.074
0.019
0.003
0.070



0.016
0.003
0.069
0.019
0.003
0.076
0.023
0.005
0.087
p-value
<.001
<.001
0.710
<.001
<.001
0.490
.


<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
153615


196236





269815
.

300488


361216
.

Variance Components
Random Effects
0,,2 = 0.779
0212 = -0.070
O.,.,* = 0.020
o,,2 = 0.637
0212= -0.056
022"== 0.016


.
o,,2= 0.446
0212= -0.008
Oyf= 0.008
o,,2= 0.522
0212 = -0.018
022Z = 0.010
o,,2= 0.543
0312= -0.024
0^* = 0.012
Error
ฐenorZ= 0.474
oenor2= 38.339




95th percentile or above for TRI Lead Compounds (Stacks)
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tn_as2
Intercept
time
tn_as2
Intercept
time
tri_as2
Intercept
time
tri_as2
Intercept
time
tri_as2
Intercept
time
tri_as2
Estimate
3.417
-0.155
0.025
3.401
-0.156
0.001
-1.495
-0.146
0.083
-3.325
-0.169
0283
-4.477
-0.161
0.351
-5954
-0.141
0.341
Standard
Error
0.019
0.003
0.073
0.019
0.003
0069
0.013
0.002
0.053
0.016
0.003
0.069
0.019
0.003
0.075
0.023
0.005
0.085
p-value
<.001
<.001
0.732
<.001
<.001
0.991
<.001
<.001
0.120
<001
<.001
<.001 •
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
153615


196237


250730


269824


300504

.
361290
.

Variance Components
Random Effects
0,,2= 0.778
Oa,2 = -0.070
0K* = 0.020
o,,2= 0.636
0212 = -0.056
02/= 0.016
0,,2= 0.332
0212 = -0.022
0^= 0.010
0,,2= 0.448
0212 = -0.008
aK*= 0.008
0,,2= 0.524
0212= -0.018
0^= 0.010
0,,2= 0.544
0212= -0.024
022* = 0.012
Error
Oerorz= 0.474
Oeror2= 38.339




                                                   C-l
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent
any Agency determination or policy.

-------
95th percentile or above for TRI Total Lead (Stacks)
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tri_as3
Intercept
time
tri_as3
Intercept
time
tri_as3
Intercept
time
tn_as3






'Estimate
3.414
-0.155
0.079
3.399
-0.156
0.051
-1.497
-0.146
0.147
-3.329
-0.169
0.389
.
.
.
.


Standard
Error
0.019
0.003
0.074
0.019
0.003
0.070
0.013
0.002
0.054
0.016
0.003
0.069






p-valiie
<.001
<.001
0.283
<.001
<.001
0.464
<.001
<.001
0.006
<.001
<.001
<.001




.
•
-2 Log
Likelihood
153614


196236


250728


269837








Variance Components
Random Effects
0,,2 = 0.777
02,2= -0.070
O^ 0.020
0,,2 = 0.635
0212= -0.056
0^= 0.016
o,,2= 0.331
0212= -0.022
022' = 0.010
0,,2= 0.443
0212 = -0.007
0^= 0.008






• Error
Oeno2= 0.474
ฐ.TOr2 = 38.339




95th percentile or above for TRI Lead Only (Fugitive Air)
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tn_af1
Intercept
time
tn_af1
Intercept
time
tri_af1
Intercept
time
tri_af1
Intercept
time
tri_af1



Estimate
3.410
-0.155
0.190
3.395
-0.156
0.135
-1.496
-0.146
0.124
-3.327
-0.169
0.335
-4.478
-0161
0394



Standard
Error
0.019
0.003
0.076
0.019
0.003
0.071
0.013
0002
0055
0016
0.003
0.070
0.019
0.003
0.077



p-value
<.001
<.001
0.012
<001
<.001
0.056
<.001
<.001
0.024
<.001
<.001
<.001
<.001
<.001
<.001



-2 Log
Likelihood
153609


196233
.

250731


269847


300567





Variance Components
Random Effects
o,,2 = 0.775
0212= -0.070
022' = 0.020
o,,2= 0.634
0212= -0.056
0^= 0.016
o,,2= 0.332
0212= -0.022
022* = 0.010
0,,2= 0.448
0212= -0.008
022'= 0.008
On2= 0.526
0212= -0.019
022*= 0.010



Error ;
0STO2= 0.474
oem>2= 38.339




                                                   C-2
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed 10 represent
any Agency determination or policy.

-------
95th percentile or above for TRI Lead Compounds (Fugitive Air)
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tri_af2
Intercept
time
tri_af2
Intercept
time
tri_af2
Intercept
time
tri_af2
Intercept
time
tri_af2
Intercept
time
tri_af2
Estimate
3.421
-0.155
-0.068
3.406
-0.156
-0.101
-1.490
-0.146
-0.018
-3.316
-0.169
0.097
-4.468
-0.161
0.184
-5945
-0.140
0.181
Standard
Error
0.019
0.003
0073
0.019
0003
0.069
0.013
0.002
0053
0.016
0003
0.070
0.019
0.003
0.077
0.023
0.005
0.088
p-value
<.001
<001
0353
<.001
<001
0.144
<.001
<.001
0.741
<.001
<.001
0.164
<.001
<.001
0.017
<.001
<.001
0.039
-2 Log
Likelihood
153614


196235


250725


269785


300403


361063


Variance Components
Random Effects
onz= 0.780
02I2 = -0.070
022Z= 0.020
0,,2 = 0.637
a,*= -0.056
0^= 0.016
onz= 0.333
0212 = -0.022
a?f= 0.010
0,,* = 0.453
o?12= -0.008
0^= 0.008
Onz= 0.532
0212= -0.019
022* = 0.010
0,,2= 0.554
0212 = -0.025
022Z = 0.012
Brror
ฐ-.2 = ฐ-474
oero,2= 38.339




95th percentile or above for TRI Total Lead (Fugitive Air)
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tn_af3
Intercept
time
tri_af3
Intercept
time
tri_af3
Intercept
time
tri_af3
Intercept
time
tri_af3
Intercept
time
tri_af3
Estimate
3.418
-0.155
-0.001
3.402
-0.156
-0.014
-1.492
-0.146
0.030
-3.319
-0.169
0.177
-4.472
-0.161
0.259
-5.949
-0.140
0.248
Standard
Error
0019
0.003
0.074
0.019
0.003
0.071
0.013
0.002
0.054
0.016
0.003
0.070
0.019
0.003
0.077
0.023
0.005
0.087
p-value
<.001
<.001
0.986
<.001
<.001
0.844
<.001
<.001
0.585
<.001
<.001
0.012
<.001
<.001
<.001
<.001
<.001
0.004
-2 Log
Likelihood
153615


196237


250727


269806


300462


361218


Variance Components
Random Effects
0,,2= 0.779
0212= -0.070
0^'= 0.020
Onz= 0.636
02,2= -0.056
022-== 0.016
o,,2= 0.333
0212 = -0.022
022'= 0.010
o,,2= 0.451
0212= -0.008
o22-= = 0.008
on2= 0.530
0212= -0.019
022'= 0.010
0U2= 0.552
0212= -0.025
0^= 0.012
Error
ฐeno,2= 0.474
08TOZ= 38.339




                                                   C-3
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not beenfonnallv disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
95th percentile or above for TRI Lead Only (Total
Air)
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tri_at1
Intercept
time
tri_at1
Intercept
time
tn_at1
Intercept
time
tn_atl
Intercept
time
tn_at1
Intercept
time
tn_at1
Estimate'
3.416
-0.155
0.048
3.402
-0.156
-0.005
-1.495
-0.146
0.101
-3.327
-0.170
0.344
-4.478
-0.161
0.407
-5.954
-0.141
0.360
.Standard
Error
0.019
0.003
0.074
0.019
0.003
0.070
0.013
0.002
0.054
0.016
0.003
0.070
0.019
0.003
0.077
0.023
0.005
0.087
p-value
<.001
<.001
0.517
<.001
<.001
0.949
<.001
<.001
0.062
<001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
153615
.
.
196237


250729


269818


300516


361318


• Variance Components .
Random Effects
o,,2= 0.778
0212 = -0.070
022* = 0.020
o,,2 = 0.636
0212 = -0.056
022' = 0.016
o,,2= 0.331
0212= -0.022
022'= 0.010
0,,2= 0.445
0212= -0.008
022Z= 0.008
o,,z= 0.521
0212= -0.018
ฐ2ie = 0.010
o,,2= 0.542
02,2= -0.024
022Z= 0.012
Error
ฐeno,2 = 0.474
08TO2= 38.339




95th percentile or above for TRI Lead Compounds (Total Air)
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tri_at2
Intercept
time
tn_at2
Intercept
time
tn_at2
Intercept
time
tn_at2
Intercept
time
tn_at2



Estimate
3.416
-0.155
0.032
3.401
-0.156
0.009
-1.494
-0.146
0.067
-3.323
-0.169
0.259
-4.477
-0.161
0.360



Standard
Error
0.019
0.003
0.073
0.019
0.003
0.069
0.013
0.002
0.053
0.016
0.003
0.069
0.019
0.003
0.075



p-value
<.001
<.001
0.662
<.001
<.001
0.902
<.001
<.001
0.210
<.001
<.001
<.001
<.001
<.001
<.001



-2 Log
Likelihood
153615


196237


250729


269823


300517





Variance Components
Random Effects
0,,* = 0.778
0212= -0.070
0^= 0.020
0,,2= 0.636
0212= -0.056
ฐ22* = 0.016
0^= 0.332
0212= -0.022
ฐW = 0.010
0,,2= 0.448
0212= -0.008
022'= 0.008
o,,2= 0.523
0212= -0.018
022*= 0.010



Error
ฐeno,2 = 0.474
OenD2= 38.339




                                                   C-4
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
95th percentile or above for TRI Total Lead (Total Air)
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tri_at3
Intercept
time
tri_at3
Intercept
time
tri_at3
Intercept
time
tri_at3
Intercept
time
tri_at3
Intercept
time
tri_at3
Estimate
3.413
-0.155
0.103
3.399
-0156
0.045
-1.496
-0.146
0.119
-3.328
-0.169
0.352
-4.481
-0.161
0.443
-5.957
-0.141
0.390
Standard
Error
0.019
0.003
0.074
0019
0.003
0.069
0.013
0.002
0.054
0.016
0.003
0.069
0.018
0.003
0.075
0.023
0.005
0.085
p-value
<.001
<.001
0.161
<.001
<.001
0.519
<.001
<.001
0.026
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
153613


196237


250728


269831


300567


361430


Variance Components
Random Effects
0,,2= 0.776
0212 = -0.070
0^ = 0.020
On2= 0.635
0212= -0.056
o22*= 0.016
0,,2= 0.331
02,2= -0.022
azz"= 0.010
0^= 0.445
o!M2= -0.008
022Z= 0.008
0,,2= 0.519
0212= -0.018
022-== 0.010
0,,2= 0.540
0212= -0.023
022* = 0.012
Error
ฐe™,2= 0-474
Oeiro2= 38.339




95th percentile or above for TRI Lead Only (Water Surface)
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tn_ws1
Intercept
time
tn_ws1
Intercept
time
tn_ws1
Intercept
time
tri_ws1
Intercept
time
tri_ws1



Estimate
3.416
-0.155
0.033
3.402
-0.156
-0.01 1
-1.494
-0.146
0.076
-3324
-0.169
0.260
-4.472
-0.161
0.262



Standard
Error
0019
0003
0.073
0.019
0.003
0.069
0.013
0.002
0.053
0.016
0.003
0.069
0.019
0.003
0.075



p-value
<.001
<.001
0653
<.001
<.001
0.878
<.001
<.001
0.153
<.001
<.001
<.001
<.001
<.001
<.001



-2 Log
Likelihood
153615
.

196237


250731


269830


300483





Variance Components
Random Effects
0,,2= 0.778
0212= -0.070
022"== 0.020
0,,2= 0.636
0212= -0.056
0..2* = 0016
On2= 0332
0212 = -0.022
022"= 0.010
o,,2= 0.449
0212 = -0.008
022"= = 0.008
0,,2= 0.528
0212= -0019
022'= 0.010
.


Error
ฐ8m,,2 = 0.474
ฐซ2- 38.339




                                                   C-5
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
95th percentile or above for TRI Lead Compounds (Water Surface)
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tn_w s2
Intercept
time
tri_w s2
Intercept
time
tri_w s2
Intercept
time
tn_w s2



Intercept
time
tri_w s2
Estimate
3.413
-0.155
0.111
3.394
-0.156
0.144
-1.496
-0.146
0.108
-3.321
-0.169
0.196


.
-5.947
-0.140
0.212
Standard
Error
0.019
0.003
0.071
0.019
0.003
0.069
0.013
0.002
0.052
0.016
0.003
0.068



0.023
0.005
0088
p-value
<.001
<.001
0.119
<.001
<.001
0.037
<.001
<.001
0.040
<.001
<.001
0.004



<.001
<.001
0.016
•2 Log
Likelihood
153613


196233


250728


269800





361066


Variance Components
Random Effects
0,,2 = 0.779
0212= -0.070
aK<= 0.020
o,,2= 0.635
0212 = -0.056
022"== 0.016
0,,"= 0.332
0212 = -0.022
022*= 0.010
0^= 0.451
0212= -0.008
a?? = 0.008



0,,2= 0.553
0212= -0.025
0^= 0.012
Error
ฐ.TO,2 = 0.474
ฐeno,2 = 38.340




95th percentile or above for TRI Total Lead (Water Surface
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tn_w s3
Intercept
time
tri_w s3
Intercept
time
tn_w s3
Intercept
time
tri_ws3
Intercept
time
tn_w s3
Intercept
time
tri_w s3
Estimate
3.413
-0.155
0.101
3.395
-0.156
0.118
-1.497
-0.146
0.122
-3.324
-0.169
0.251
-4.473
-0.161
0.261
-5.949
-0.140
0245
Standard
Error
0.019
0.003
0.072
0.019
0.003
0.068
0.013
0.002
0.052
0.016
0.003
0.068
0.019
0.003
0.075
0.023
0005
0086
p-value
<.001
<.001
0.159
<.001
<.001
0.084
<.001
<.001
0.020
<.001
<.001
<001
<.001
<.001
<.001
<.001
<.001
0.005

-2 Log
Likelihood
153613

.
196234


250728


26981 1


300458


361127


Variance Components
Random Effects
0,^= 0.779
0212 = -0.070
022" = 0.020
0,,2= 0.636
0212 = -0.056
022* = 0.016
0,,2 = 0.332
0212= -0.022
a^= 0.010
0,,2= 0.450
0212= -0.008
022Z = 0.008
0,,2= 0.530
0212= -0.019
O^s 0.010
0,,2= 0.551
0212= -0.025
022* = 0.012
Error
ฐ,™= 0.474
oem>2= 38.339




                                                   C-6
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent
any Agencv determination or policv.

-------
95th percentile or above for tri_ui1
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tri_ui1
Intercept
time
tn_ui1
Intercept
time
tn_ui1
Intercept
time
tri_ui1
Intercept
time
tri_ui1



Estimate
3.418
-0.155
0.000
3.401
-0.156
0.000
-1.491
-0.146
0000
-3.311
-0.169
0000
-4459
-0161
0.000



Standard
Error
0.018
0.003

0.018
0003

0013
0002

0015
0.003

0.018
0.003




p-value
<.001
<.001

<001
<.001

•s.001
<001

<001
<.001

<.001
<.001




-2 Log
Likelihood
153612
.

196233


250721


269765


300313





Variance Components
Random Effects
0,,'= 0.779
O2,z= -0070
022' = 0.020
o,/= 0.636
021Z= -0.056
0^'= 0016
o,,'= 0.333
02,"= -0.022
022'= 0.010
0,,'= 0.453
021Z= -0008
022'= 0.008
0,,'= 0.535
021Z= -0020
022'= 0010



Error

-------
95th percentile or above for TRI Total Lead (Underground Injection)
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tn_ui3
Intercept
time
tri_ui3
Intercept
time
tri_ui3
Intercept
time
tri_ui3
Intercept
time
tn_ui3



Estimate
3.418
-0.155
0.000
3.401
-0.156
0.000
-1.491
-0.146
0.000
-3.31 1
-0.169
0.000
-4.459
-0.161
0.000



Standard
&ror
0.018
0.003

0.018
0.003

0.013
0.002

0.015
0.003

0.018
0.003




p-value
<.001
<.001

<.001
<.001

<.001
<.001

<.001
<.001

<.001
<001




-2 Log
Likelihood
153612


196233


250721


269765


300313





Variance Components
Random Effects
0,^= 0.779
0212= -0.070
0^= 0.020
o,,2= 0.636
0212 = -0.056
Oyf= 0.016
o,,2= 0.333
02,2= -0.022
022' = 0.010
0A12= 0.453
0212= -0.008
o22* = 0.008
o,,2= 0.535
0212= -0.020
022* = 0.010



&ror
ฐaTO,2 = 0.474
Oeror2= 38.339




95th percentile or above for Average Air
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
air_avg
Intercept
time
air_avg
Intercept
time
air_avg
Intercept
time
air_avg
Intercept
time
air_avg
Intercept
time
air_avg
Estimate
3.406
-0.155
0.279
3.394
-0156
0.169
-1.498
-0.146
0.158
-3.336
-0.170
0.542
-4.488
-0.162
0.585
-5.963
-0.142
0.499
Standard
Error
0.019
0.003
0.074
0.019
0.003
0.070
0.013
0.002
0.054
0016
0.003
0.069
0.018
0.003
0.075
0.023
0.005
0.084
p-value
<.001
<.001
<.001
<.001
<.001
0.016
<.001
<.001
0.003
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
153601


196231


250726


269826


300587
.

361524


Variance Components
Random Effects
0,,2= 0.767
0212= -0.069
aK"= 0.020
o,,2= 0.630
0212= -0.055
022* = 0.016
On2= 0.331
0212= -0.022
022"== 0.010
0,^= 0.434
0212 = -0.007
022Z = 0.008
0,,2= 0.508
0212= -0.017
022*= 0.010
0,,2= 0.530
0!)12= -0.022
0^= 0.012
Error
ฐซ,2 = 0.474
oeTO2 = 38.339




                                                   C-8
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
95th percentile or above for Median Air
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
air_med
Intercept
time
air_med
Intercept
time
air_med
Intercept
time
air_med
Intercept
time
air_med



Estimate
3.402
-0.155
0.359
3.390
-0.156
0.231
-1.500
-0.146
0.185
-3340
-0.170
0.600
-4.493
-0.162
0.674



Standard
Error
0.019
0.003
0.072
0.019
0.003
0.069
0.013
0.002
0.053
0.016
0.003
0.067
0018
0.003
0.073



p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001



-2 Log
Likelihood
153591


196226


250725


269820


300619





Variance Components
Random Effects
0,,2= 0.764
0212= -0.069
022* = 0020
o,,2 = 0.629
0212= -0.055
v&= 0.016
Olt2 = 0.331
0212= -0.022
0^= 0.010
0,,2= 0.432
0212 = -0.007
a2*'= 0.008
o,,2= 0.500
Oj,,2 = -0.017
022Z= 0.010



Error
Oanorz= 0.474
Oeno2= 38.338




95th percentile or above for 95th Percentile Air
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
air_p95
Intercept
time
air_p95
Intercept
time
air_p95



Intercept
time
air_p95
Intercept
time
air_p95
Estimate
3.416
-0.155
0.041
3.402
-0.156
-0.022
-1.494
-0.146
0.070


•
-4.477
-0.161
0.362
-5.951
-0.141
0.288
Standard
Error
0.019
0.003
0.074
0.019
0.003
0.070
0.013
0.002
0.054



0.019
0.003
0.077
0.023
0005
0.088
p-value
<.001
<.001
0.580
<.001
<.001
0.755
<.001
<.001
0.195



<.001
<.001
<.001
<.001
<.001
0.001
-2 Log
Likelihood
153615


196237


250728





300478


361164


Variance Components
Random Effects
0,,2 = 0.778
0212= -0.070
0^= 0.020
0,,2= 0.637
0212= -0.056
022* = 0.016
0,,2= 0.332
0212 = -0.022
022-== 0.010

.

0,,2= 0.525
0212= -0.019
022'= 0.010
On2= 0.549
0212= -0.024
022"== 0.012
Error
ฐ—2= ฐ-474
Oero2= 38.339




                                                    C-9
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination orpolicv.

-------
95th percentile or above for Drinking Water Lead
Model
Number
1
2
3
4
5
6
: Factor
Intercept
time
sdwis
Intercept
time
sdwis
Intercept
time
sdwis
Intercept
time
sdwis



Intercept
time
sdwis
Estimate
3.416
-0.156
0.060
3.400
-0.156
0.035
-1.491
-0.146
0.018
-3.31 1
-0.169
-0.008
.
.
-
-5932
-0140
-0043
Standard
Error
0.018
0.003
0.018
0.018
0.003
0.010
0.013
0.002
0.005
0.015
0.003
0.008

.
.
0.023
0.005
0.025
s p-value
<.001
<.001
0.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
0.313



<.001
<.001
0.089
-2 Log
Likelihood
153607


196228


250697


269792





360772


Variance Components
Random Effects
0^2= 0.778
0212 = -0.070
OJ2* = 0.020
0,^ = 0.636
0212 = -0.056
022-== 0.016
0,,"= 0.332
0212 = -0.022
032* = 0.010
a,* = 0.453
0212 = -0.008
O^ 0.008



0,,*= 0.558
0212 = -0.025
022* = 0.012
Error
ฐaro,2= 0.474
ฐeTOr2 = 38.332




99th percentile or above for TRI Lead Only (Stacks)
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tn_as1
Intercept
time
tri_as1
Intercept
time
tri_as1
Intercept
time
tri_as1



Intercept
time
tn_as1
Estimate
3.416
-0.155
0.172
3.401
-0.156
0.067
-1.493
-0.146
0.208
-3.315
-0.169
0.454



-5.937
-0.140
0.305
Standard
- Error
0.018
0.003
0.157
0.018
0.003
0.153
0.013
0.002
0.116
0.015
0.003
0.151



0.023
0.005
0.195
p-value
<.001
<001
0.274
<.001
<.001
0.663
<.001
<.001
0.072
<.001
<.001
0.003



<.001
<.001
0.117
-2 Log
Likelihood
153612


196235


250721


269774



.
-
360886
.

Variance Components
Random Effects
0,,*= 0.778
02,2= -0.070
022* = 0.020
0,,2= 0.636
0212= -0.056
0^= 0.016
0^ = 0.332
0212= -0.022
022Z = 0.010
0,,2= 0.451
0212= -0.008
022*= 0008



o,,2= 0.554
0212 = -0.025
022*= 0.012
Error
ฐซ.2- 0.474
oeno,2= 38.339




                                                   C-10
77ป'5 information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
99th percentile or above for TRI Lead Compounds (Stacks)
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tri_as2
Intercept
time
tn_as2
Intercept
time
tn_as2



Intercept
time
tn_as2
Intercept
time
tri_as2
Estimate
3.417
-0.155
0.049
3.400
-0.156
0.139
-1.492
-0.146
0.162



-4.463
-0.161
0.444
-5.938
-0.140
0.316
Standard
Error
0.018
0.003
0.168
0.018
0.003
0.156
0.013
0.002
0.121



0.018
0.003
0.170
0.023
0.005
0.190
p-value
<.001
<.001
0.769
<.001
<.001
0.372
<.001
<.001
0.182



<.001
<.001
0.009
<.001
<.001
0.097
-2 Log
Likelihood
153613


196235


250726





300367


360902


Variance Components
Random Bfects
0,,2= 0.778
0212= -0.070
Vy? = 0.020
o,,2= 0.635
0,,* = -0.056
0^= 0.016
0,,2= 0.332
0212= -0.022
022Z = 0.010



0,,2= 0.532
0212= -0.020
0^= 0.010
0,,2= 0.555
0212= -0.025
0^= 0.012
Error
OenD2 = 0.474
oeno,2= 38.339




99th percentile or above for TRI Total Lead (Stacks)
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tn_as3
Intercept
time
tn_as3



Intercept
time
tri_as3
Intercept
time
tri_as3



Estimate
3.419
-0.155
-0139
3.401
-0.156
-0.008



-3.315
-0.169
0.426
-4.464
-0.161
0.534



Standard
Error
0.018
0.003
0.165
0.018
0.003
0.156



0.015
0.003
0.156
0.018
0.003
0.169



p-value
<.001
<.001
0.401
<.001
<.001
0.957



<.001
<.001
0.006
<.001
<.001
0.002"



-2 Log
Likelihood
153613


196235





269785


300386





Variance Components
Random Bfects
o,,2= 0.779
0!>12= -0.070
Qyf= 0.020
0,,2= 0.636
0212= -0.056
022Z= 0.016



0,,2= 0.451
0212= -0.008
O^ = 0.008
On2= 0.530
0?12= -0.019
0^= 0010



Error
ฐ.TO,2 = 0.474
oero2= 38.339




                                                  C-ll
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
99th percentile or above for TRI Lead Only (Fugitive Air)
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tri_af1
Intercept
time
tn_af1
Intercept
time
tri_af1
Intercept
time
tri_afl






Estimate
3.416
-0.155
0.290
3.399
-0.156
0.323
-1.493
-0.146
0.220
-3.315
-0.169
0.488
.




•
Standard
Bror
0.018
0.003
0.171
0.018
0.003
0.161
0.013
0.002
0.125
0.015
0.003
0.160





•
p-value
<.001
<.001
0.090
<.001
<.001
0.046
<.001
<.001
0.078
<.001
<.001
0.002





•
-2 Log
Likelihood
153610
.

196231


250723
.
.
269776








Variance Components
Random Effects
0,,2= 0.777
0212 = -0.070
022* = 0.020
0,,2 = 0.635
0212 = -0.056
022* = 0.016
0,^= 0.332
a,* = -0.022
022* = 0.010
On2= 0.451
0212= -0.008
o^ = 0.008
.





Error
ฐero,2= 0.474
ฐeTOr2 = 38.339




99th percentile or above for TRI Lead Compounds (Fugitive Air)
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tri_af2
Intercept
time
tn_af2
Intercept
time
tri_af2
Intercept
time
tri_af2
Intercept
time
tri_af2
Intercept
time
tn_af2
Estimate
3.419
-0.155
-0.086
3.402
-0.156
-0.052
-1.491
-0.146
0.044
-3.314
-0.169
0.332
-4.463
-0.161
0.408
-5.938
-0.140
0.342
Standard
Error
0.018
0.003
0.164
0.018
0.003
0.154
0.013
0.002
0.119
0.015
0.003
0.152
0.018
0.003
0.167
0.023
0.005
0.186
p-value
<.001
<.001
0.599
<.001
<.001
0.734
<.001
<.001
0.711
<.001
<.001
0.029
<.001
<.001
0.014
<.001
<.001
0.066
-2 Log
Likelihood
153613


196235
.
.
250724


269781


300365


360916


Variance Components
Random Effects
a,,2= 0.779
0212 = -0.070
0^ = 0.020
0,,2 = 0.637
0212= -0.056
022Z = 0.016
0,,2= 0.333
a 2= -0.022
022* = 0.010
o,,2= 0.452
a212 = -0.008
Oyf = 0.008
0,,2= 0.532
a212= -0.019
0^= 0.010
0,,2= 0.554
02,2= -0.025
022Z= 0.012
Error
ฐ-e™,2 = 0.474
oenD2= 38.339




                                                  C-12
777/5 information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not beenfonnally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
99th percentile or above for TRI Total Lead (Fugitive Air)
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tri_a!3
Intercept
time
tri_af3



Intercept
time
tn_af3
Intercept
time
tri_af3



Estimate
3.417
-0.155
0.106
3.400
-0.156
0.134



-3.315
-0.169
0.500
-4.464
-0.161
0.569



Standard
Error
0.018
0.003
0.164
0.018
0.003
0.156



0.015
0.003
0.155
0.018
0.003
0.170



p-value
<.001
<.001
0.518
<.001
<.001
0.392
.
.

<.001
<.001
0.001
<.001
<.001
<.001



•2 Log
Likelihood
153613


196235

.

.

269773

.
300363





Variance Components
Random Effects
o,,2 = 0.778
o212= -0.070
022" = 0.020
on2= 0.635
o,,2= -0.056
022'= 0016



On2 = 0.450
0212= -0.008
022Z = 0.008
o,,2 = 0.529
0212 = -0.019
0^= 0.010



Error
ฐoro,2= 0.474
ฐenor2= 38.339




99th percentile or above for TRI Lead Only (Total Air)
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tn_at1
Intercept
time
tri_at1
Intercept
time
tn_at1
Intercept
time
tn_at1



Intercept
time
tn_atl
Estimate
3.417
-0.155
0.048
3.401
-0.156
0.090
-1.492
-0.146
0.182
-3.314
-0.169
0382



-5.940
-0.140
0.588
Standard
Error
0.018
0.003
0.160
0.018
0.003
0.157
0.013
0.002
0.120
0.015
0.003
0.158



0.023
0.005
0.195
p-value
<.001
<.001
0.765
<.001
<.001
0.565
<.001
<.001
0.127
<.001
<.001
0.016



<.001
<.001
0.003
-2 Log
Likelihood
153613


196235


250722


269766





360975


Variance Components
Random Effects
o,,2= 0.779
021Z = -0.070
022* = 0.020
0,,2= 0.636
0212 = -0.056
022*= 0.016
0,,2= 0.332
0212= -0.022
022Z = 0.010
0,,2= 0.451
0212 = -0.008
O^ = 0.008



On2= 0.550
0212= -0025
Oyf= 0.012
Error
0BTO2= 0.474
Oenor2= 38.339




                                                   C-13
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent
any Agency determination or policy.

-------
99th percentile or above for TRI Lead Compounds (Total Air)
Model
Number
1
2
3
4
5
6

Factor
Intercept
time
tn_at2
Intercept
time
tri_at2
Intercept
time
tri_at2
Intercept
time
tri_at2
Intercept
time
tn_at2
Intercept
time
tri_at2
Estimate
3.418
-0.155
-0.007
3.401
-0.156
0.040
-1.492
-0.146
0.096
-3.314
-0.169
0.331
-4.463
-0.161
0.442
-5.938
-0.140
0.292
Standard
Error
0.018
0.003
0.168
0.018
0.003
0.156
0.013
0.002
0.121
0.015
0.003
0.155
0.018
0.003
0.168
0.023
0.005
0.189
p-value
<.001
<.001
0.967
<.001
<.001
0.795
<.001
<.001
0.429
<.001
<.001
0.033
<.001
<.001
0.009
<.001
<.001
0.121
-2 Log
Likelihood
153613


196235


250726


269784


300376


360892


Variance Components
Random Effects
0,^= 0.779
0212 = -0.070
022Z = 0.020
otl2= 0.636
0212 = -0.056
022* = 0.016
O,,2 = 0.333
0212= -0.022
o22z= 0.010
On2= 0.452
0212 = -0.008
022" = 0.008
0,,*= 0.532
o,*= -0.019
022Z = 0.010
0^= 0.555
0212= -0.025
022Z= 0.012
Error
ฐeTO,2 = 0.474
Oeror2= 38.339




Intercept
99th percentile or above for TRI Total Lead (Total Air)
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tri_at3
Intercept
time
tn_at3
Intercept
time
tri_at3
Intercept
time
tn_at3
Intercept
time
tri_at3
Intercept
time
tri_at3
Estimate
3.418
-0.155
-0.074
3.401
-0.156
-0.015
-1.492
-0.146
0.119
-3.315
-0169
0.418
-4.464
-0.161
0.519
-5.939
-0.140
0.432
Standard
Error
0.018
0.003
0.163
0.018
0.003
0.153
0.013
0.002
0119
0.015
0.003
0.153
0.018
0.003
0.166
0.023
0005
0184
p- value
<.001
<.001
0.647
<.001
<.001
0.920
<.001
<.001
0.316
<.001
<.001
0.006
<.001
<.001
0.002
<.001
<.001
0.019
-2 Log
Likelihood
153613


196235


250725


269784


300380
.
.
360945


Variance Components
Random Effects
0,,2= 0.779
0212 = -0.070
022Z= 0.020
o,,2= 0.636
0212= -0.056
02ZZ= 0.016
0,^= 0.333
0212 = -0.022
022* = 0.010
0,^= 0.451
0212= -0.008
Oyf = 0.008
0,^= 0.530
0212= -0.019
as2e = 0.010
0112= 0.552
0212= -0.025
022* = 0.012
Error
ฐenor2= 0-474
Oarror2= 38.339




                                                  C-14
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
99th percentile or above for TRI Lead Only (Water Surface)
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tri_ws1
Intercept
time
tri_wst
Intercept
time
tn_ws1
Intercept
time
tri_ws1
Intercept
time
tn_ws1
Intercept
time
tn_ws1
Estimate
3.417
-0.155
0.094
3.401
-0.156
0.049
-1.493
-0.146
0.204
-3.316
-0.169
0508
-4.465
-0.161
0.529
-5.939
-0.140
0.331
Standard
Error
0.018
0.003
0155
0.018
0.003
0.146
0.013
0.002
0.113
0.015
0.003
0.144
0.018
0.003
0.157
0.023
0.005
0.179
p-value
<.001
<.001
0.544
<.001
<.001
0.738
<.001
<.001
0.070
<.001
<001
<.001
<.001
<.001
<.001
<.001
<.001
0.064
-2 Log
Likelihood
153613


196235


250725


269783


300379


360901


Variance Components
Random Effects
0,,2= 0.778
0212= -0.070
022Z= 0.020
0,,2 = 0.636
a,* = -0.056
022< = 0.016
0,,2 = 0.332
0?12= -0.022
022* = 0.010
0,^= 0.450
0212 = -0.008
ฐ22Z = 0.008
0^= 0.531
0212= -0.019
ฐK~ 0.010
o,,2= 0.555
0212= -0.025
022* = 0.012
Error
ฐซซ2- 0.474
Oeror2= 38.339




99th percentile or above for TRI Lead Compounds (Water Surface)
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tri_w s2
Intercept
time
tn_w s2



Intercept
time
tri_w s2
Intercept
time
tri_w s2
Intercept
time
tn_w s2
Estimate
3.416
-0.155
0.137
3.400
-0.156
0.135



-3.313
-0169
0240
-4.461
-0.161
0.214
-5.936
-0.140
0.160
Standard
Error
0.018
0.003
0.158
0018
0.003
0150



0.015
0.003
0.148
0.018
0.003
0164
0.023
0.005
0.188
p-value
<.001
<.001
0.386
<.001
<.001
0.369



<.001
<.001
0.106
<.001
<.001
0.192
<.001
<.001
0.394
-2 Log
Likelihood
153613


196235

.



269777


300342


360857


Variance Components
Random Effects
0,,* = 0.779
0,*= -0.070
022* = 0.020
o,,2= 0.636
0212= -0.056
ฐK= 0.016



o,,2= 0.453
0212= -0.008
OJB* = 0.008
0^= 0534
02I2= -0.020
02/= 0.010
0,,2= 0.556
0212= -0.025
022Z = 0.012
Error
ฐero,2 = 0.474
0BTO2= 38.339




                                                  C-15
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent
any Agency determination or policy.

-------
99th percentile or above for TRI Total Lead (Water Surface
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tri_w s3
Intercept
time
tri_w s3
Intercept
time
tri_w s3
Intercept
time
tri_w s3
Intercept
time
tn_w s3
Intercept
time
tri_w s3
Estimate
3.417
-0.155
0.117
3.400
-0.156
0.102
-1.492
-0.146
0.101
-3.314
-0.169
0.316
-4.461
-0.161
0.268
-5.937
-0.140
0.197
Standard.
Error
0.018
0.003
0.161
0.018
0.003
0.151
0.013
0.002
0.116
0.015
0.003
0.149
0.018
0.003
0.164
0.023
0.005
0.183
p-value
<.001
<.001
0.469
<.001
<.001
0.500
<.001
<.001
0.383
<.001
<.001
0.034
<.001
<.001
0.101
<.001
<.001
0.283

-2 Log
Likelihood
153613
.
.
196235


250726


269783
.
.
300353
.

360879


• Variance Components
Random Effects
0,,2= 0.779
0212= -0.070
Oa/ = 0.020
o,,2 = 0.636
0212 = -0.056
022'= 0.016
0,,2 = 0.333
0212= -0.022
ฐ22Z= 0.010
0,,2= 0.452
0212= -0.008
0^= 0.008
On2= 0.533
0212= -0.019
0^= 0.010
On2= 0.555
0312= -0.025
0^= 0.012
Error
OerlD2= 0.474
oeror2= 38.339




99th percentile or above for TRI Lead Only (Underground Injection)
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tn_ui1
Intercept
time
tn_ui1
Intercept
time
tri_ui1
Intercept
time
tn_ui1
Intercept
time
tn_ui1



Estimate
3.418
-0.155
0.000
3.401
-0.156
0.000
-1.491
-0.146
0000
-3.311
-0.169
0.000
-4.459
-0.161
0.000



Standard
Error
0.018
0.003

0.018
0.003

0.013
0.002

0.015
0.003

0018
0.003




p-value
<.001
<.001

<.001
<.001

<.001
<.001

<.001
<.001

<.001
<.001




-2 Log
Likelihood
153612


196233


250721


269765


300313





Variance Components
Random Effects
0,,* = 0.779
0212= -0.070
o22z = 0.020
On2= 0.636
0212 = -0.056
022"= 0.016
o,,2= 0.333
02,2= -0.022
ฐ22Z= 0.010
o,,2= 0.453
0212= -0.008
022*= 0.008
o,,2= 0.535
0212 = -0.020
022'= 0.010



Error
ฐ~2 = 0.474
0Bro2= 38.339




                                                   C-16
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
99th percentiie or above for TRI Lead Compounds (Underground Injection)
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
tri_ui2
Intercept
time
tn_ui2
Intercept
time
tn_ui2
Intercept
time
tn_ui2
Intercept
time
tri_ui2



Estimate
3.418
-0.155
0.000
3.401
-0.156
0.000
-1.491
-0.146
0.000
-3.311
-0.169
0.000
-4.459
-0.161
0.000



Standard
Error
0.018
0.003

0.018
0.003

0.013
0.002

0.015
0.003

0.018
0.003




p-value
<.001
<.001

<.001
<.001

<.001
<.001

<.001
<.001

<.001
<.001




-2 Log
Likelihood
153612


196233


250721


269765


300313
.




Variance Components
Random Effects
0,,2 = 0.779
0212= -0.070
0^= 0.020
0,,2= 0.636
0212 = -0.056
022* = 0.016
0,,2 = 0.333
0212 = -0.022
022Z= 0.010
o,,2= 0.453
0212= -0.008
Oyf = 0.008
o,,2= 0.535
0212= -0.020
0^= 0.010



Error

-------
99th percentile or above for Average Air
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
air_avg
Intercept
time
air_avg



Intercept
time
air_avg



Intercept
time
air_avg
Estimate
3.414
-0.155
0.450
3.397
-0.156
0.465

.
.
-3.316
-0.169
0.581
.
.

-5.938
-0.140
0.378
Standard
Error
0.018
0.003
0157
0.018
0.003
0.155



0.015
0.003
0.152



0.023
0.005
0.196
p-value
<.001
<.001
0.004
<.001
<.001
0.003



<.001
<.001
<.001



<.001
<.001
0.054
-2 Log
Likelihood
153605


196226





269764





360904


Variance Components
Random Effects
0,,2= 0.775
0212= -0.070
0^= 0.020
0^= 0.632
0212= -0.055

-------
99th percentile or above for 95th Percentile Air
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
air_p95
Intercept
time
air_p95
Intercept
time
air_p95
Intercept
time
air_p95
Intercept
time
air_p95
Intercept
time
air_p95
Estimate
3.416
-0.155
0.220
3.399
-0.156
0.261
-1.493
-0.146
0.208
-3.314
-0.169
0.376
-4.463
-0161
0.426
-5.938
-0.140
0.390
Standard
Error
0.018
0.003
0.161
0.018
0.003
0.155
0.013
0.002
0.118
0.015
0.003
0.155
0.018
0.003
0.172
0.023
0.005
0.196
p-value
<.001
<.001
0.171
<.001
<.001
0.094
<.001
<.001
0.077
<.001
<.001
0.015
<.001
<.001
0.013
<.001
<.001
0.046
-2 Log
Likelihood
153612


196233


250723


269771


300348


360908


Variance Components
Random Effects
a,*= 0.778
0212 = -0.070
022'= 0.020
0,,"= 0.635
0212 = -0056
022'= 0.016
0,2= 0.332
o212 = -0.022
022"== 0.010
0,,2= 0.452
0212 = -0.008
032* = 0.008
0^= 0.532
0212 = -0.020
022"== 0.010
0,,2= 0.554
0212= -0.025
022Z= 0.012
Error
ฐe™,2= 0.474
OeiI02= 38.339




99th percentile or above for Drinking Water Lead
Model
Number
1
2
3
4
5
6
Factor
Intercept
time
sdwis
Intercept
time
sdwis
Intercept
time
sdwis
Intercept
time
sdwis
Intercept
time
sdwis
Intercept
time
sdwis
Estimate
3.417
-0.156
0.240
3.402
-0.157
0.213
-1 491
-0.146
0.139
-3.312
-0.170
0.267
-4.459
-0.162
0.222
-5.935
-0141
0.232
Standard
Error
0018
0.003
0.036
0.018
0.003
0.023
0.012
0.002
0.011
0.015
0.003
0.021
0.018
0003
0.034
0.023
0.005
0.066
p-value
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<.001
<001
<001
<.001
<.001
<.001
<.001
<.001
<.001
-2 Log
Likelihood
153572


196154


250548


269358


300320


360871


Variance Components
Random Effects
0,^= 0.778
0,,2= -0.070
022* = 0.020
0,,2= 0.635
0212 = -0.056
as2C = 0.016
0,,2= 0.332
02,z= -0.022
022* = 0.010
0,,2= 0.453
0212= -0.008
022" = 0.008
o,,2= 0.534
0212= -0.020
022*= 0010
o,,2= 0556
0212 = -0.025
022'= 0.012
Error
ฐ—f- ฐ-473
oero,2 = 38.285




                                                   C-19
This information is distributed solely for the purpose of pre-disseminalion peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
County identified by EPA as high air lead concentration county
Model
Number
1
' 2
3
4
5
6
Factor
Intercept
time
high_airpb
Intercept
time
high_airpb
Intercept
time
high_airpb
Intercept
time
high_airpb
Intercept
time
high_airpb
Intercept
time
high_airpb
Estimate
3.418
-0.155
0.471
3.402
-0.156
0.489
-1.491
-0.146
0.296
-3.311
-0.169
• 0.502
-4.458
-0.161
• 0.477 '
.- -5.934
-0.140
0.186
Standard
Error
.0.018
0.003
0.169
0.018
0.003
0.171
0.012
0.002
0.125
0.015
0.003
,. 0.166
0.018
0.003
0.188
0.023
0.005
0.229
p-value
<.001
<.001
• 0.005
. <.001
- <.001
0.004
<.001
• <.001
0.018
. <.001
.. <.001 ,
0.002
<.001
. <.001
; 0.01 1
" ,<.001
.<.001
0.417
-2 Log
Likelihood
153606


196227


250719


269759
.

300328


360820 .


Variance Components
Random Effects
0,,2= 0.775
0212= -0.070
02Z'= 0.020
o,,2= 0.633
0212= -0.056
0^= 0.016
o,,2= 0.332
0212= -0.022
Oyf= 0.010
0,," = 0.451
0212= -0.008
0^= 0.008
o,,2= 0.532
o212 = -0.020.
o22' = 0.010
aA*= 0.556
o2,2= -0.025
0^= 0.012
Error •'
09TO2= 0.474
Oenor2= 38.339




                                                   C-20
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy..

-------
                   Appendix D



Detailed Discussion of National Exploratory Analyses

-------
D.     Relationship between National Blood Lead Data and Explanatory Variables

D.I    Analyses of National Blood Lead Data by Demographic Variables

Demographic information from the 2000 U.S. Census is being utilized in both the high and low
resolution models, with data being acquired at the county level for the entire nation and at the Census
tract level for Massachusetts.  Initially, 50 demographic variables within 10 general categories are being
explored, most of which had been previously used by Battelle in a CDC-sponsored study to predict risk
of elevated blood-lead concentrations at the census tract level (Strauss, 2001). In many cases, these
variables are constructed from counts or summary statistics published in the detailed U.S. Census
Tables. For example, within each geographic area, the census provided the number of houses that were
built before 1950 and the median income of all households.  In order for this study to draw comparisons
from tract to tract and/or county to county, however, the census variables need to be manipulated in a
fashion that depended upon the format  of the variable.

D. 1.1   Income Variables

Initial results from exploring and modeling the income-related variables are presented in Tables Al to
A9.  Highlights from the analysis of each exploratory variable are listed below.

Median Family Income - Similar relationships between GM blood  lead and Median Family Income are
seen across the four time periods, with  GMs decreasing as family income increases. Blood lead levels
decline steadily across the four periods. Mean and Median Family Income show just a slight decrease
from the initial time periods to the later three periods. Median Family Income is significant in models 1
to 5, but not 6.
Median Household Income - Similar relationships between GM blood lead and Median Household
Income are seen across the four time periods, with GMs decreasing as household income increases.
Blood  lead levels  decline steadily across the four periods.  Mean and Median Household Income show
just a slight decrease from the initial time periods to the later three periods. Median Household Income
is significant in models 1 to 4, but  not in 6. Model 5 did not converge.
Median Per Capita Income - Consistent relationships between GM blood lead and Median Per Capita
Income are seen across the four time periods, with GMs decreasing as median per capita income
decreases.  Mean and median Median Per Capita Income show a slight decrease from the initial time
periods to the later three periods. Median  Per Capita Income is significant in models 1 to 4, but not in
Models 5 and 6.
Percent Units with No Household Earnings - Consistent relationships between GM blood lead and
Percent of Units with No Household Earnings are seen across the four time periods, with GMs
increasing as percent of units with no household earnings increases.  Mean and Percent of Units with No
Household Earnings are stable across the four time periods.  Percent of Units with  No Household
Earnings is significant in models 1 to 3 and 5 and 6, but did not converge in Model 4.
Percent Units with No Household Wage - Consistent relationships between GM blood lead and Percent
of Units with No Household Wages are seen across the four time periods, with GMs increasing as
Percent of Units with No Household Wages. Mean and median Percent of Units with No Household

                                              D-l
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
Wages show a slight increase across the four time periods. Percent of Units with No Household Wages
is significant in all models.
Percent Households on Public Assistance - Consistent relationships between GM blood lead and
Percent of Households on Public Assistance are seen across the four time periods, with GMs increasing
as Percent of Households on Public Assistance increases. Mean Percent of Households on Public
Assistance shows a slight increase from the initial time period to the later three periods. Percent of
Households on Public Assistance is significant in all models.
Percent Households Below Poverty Line - Consistent relationships between GM blood lead and
Percent of Households Below the Poverty Line are seen across the four time periods, with GMs
increasing as Percent of Households Below the Poverty Line increases. Mean and median Percent of
Households Below the Poverty Line increase from the initial time period to the later three periods.
Percent of Households Below the Poverty Line is significant in Models 1 to 4, but not in Model 6 and it
failed to converge in Model 5.
Percent Units with Family Income Below Poverty Line - Consistent relationships between GM blood
lead and Percent of Units with  Family Income Below the Poverty Line are seen across the four time
periods, with GMs increasing gradually as  percent of units increases. Mean and median Percent of Units
with Family Income  Below the Poverty Line increase from the initial time period to the later three
periods. Percent of Units with Family Income Below the Poverty Line is significant in Models 1 to 4,
but not in  Models 5 and 6.
Percent Units Spending Less than Five Years in Poverty - Consistent relationships between GM blood
lead and Percent of Units Spending Less than Five Years in Poverty are seen across the four time
periods, with GMs increasing gradually as  percent of units increases. Mean and median Percent of Units
Spending  Less than Five Years in Poverty increase from the initial time period to the later three periods.
Percent of Units Spending Less than Five Years in Poverty is significant in Models 1 to 3, but not in
Model 5, and failed to converge in Models 4 and 6.

A few trends stand out across these nine related variables. One, average levels of some poverty statistics
increase slightly from the 1995-1999 time period to the later three periods.  This may indicate a slightly
different mix of counties included in the 1995-1999 time period. Also, each of the nine variables is
significant in predicting children's blood lead levels in at least three of the six models, and usually more.
As seen in Table 4-1, the variable Percent No Household Wage provides the best fit in five of the six
models, with Median Household Income providing the best fit for Model 3.

D.I.2   Race Variables

Results from the eight race variables explored  are presented in Figures/Tables A10 to A17, with
highlights listed below.

Percent American Indian and Alaskan Native Alone - Most counties have very low percentages of
American Indian and Alaskan Natives, with the 90th percentile averaging 1.2 percent.  The 1995-1999
and 2004-2006 time  periods show a similar relationship between blood lead levels and percent American
Indian and Alaskan Native. The 2000-2001 and 2002-2003 periods display a stronger downward slope
with blood lead levels declining as percent American Indian  increases. The mean percent of American

                                              D-2
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
Indian and Alaskan Native decreases slightly from the 1995-1999 time period to the later three periods.
Percent of American Indian and Alaskan Native is a significant predictor of the outcome measure in five
of the six models.  It failed to converge for Model 3.
Percent Asian Alone - Consistent relationships between GM blood lead and  percent Asian are seen
across the four time periods, with GMs decreasing as percent Asian increases. Mean percent Asian
declines slightly from the initial time period to the later three periods. Percent Asian is significant in
Models 1 to 3, but not in Models 4 to 6.
Percent Black Alone - Consistent relationships between GM blood lead and  percent Black are seen
across the four time periods, with GMs remaining flat or increasing slightly as percent Black increases.
Mean and median percent Black increases slightly from the initial time period to the later three periods,
perhaps indicating that counties with higher percentage of Black populations  are included in the later
time periods. Percent Black is significant in Models 1 to 4 and 6, but not in Model 5.
Percent White Alone - The relationship between GM blood lead and percent White is relatively flat
across the four time periods, with GMs declining steadily with each time period. Mean and median
percent White decreases slightly from the initial time period to the later three periods. Percent White is
significant in Models 2, 3, and 6 but not the other three.
Percent Native Hawaiian and Other Pacific Islander Alone - GM blood lead levels decline
consistently as percent Native Hawaiian increases across all four time periods; however, the decline
appears steeper in the 1995-2000 time period.  Mean and median percent Native Hawaiian remain fairly
stable across time periods. Percent Native Hawaiian is significant in Models  1 to 3, but not in Models 4
to 6.
Percent Other Race Alone - GM blood lead levels decline as percent Other Race increases across all
four time periods; however, the decline appears steeper in the 1995-2000 and 2000-2002 time periods.
Mean and median percent Other Race remain fairly stable across time periods, although the average
percentages are very low. Percent Other Race  is only significant in Models 1 and 2, but did not
converge in Model 3 and was not significant in Models 4 to 6.
Percent Multiple Races - GM blood lead levels consistently decline as percent Multiple Races increases
across all four time periods, although there is some slight differences in the slopes across the time
periods. Mean percent Multiple Races declines slightly from the initial time  period to the later three
periods. Percent Multiple Races is significant  in  Models 1 to 4, but was not significant in Models 5 and
6.
Percent Hispanic - The relationship between GM blood lead and percent Hispanic is relatively flat
across the four time periods, although there is a slight negative relationship in the 2002-2004 time
period. Mean percent Hispanic increases significantly from the initial time period (3.94%) to the later
three periods (5.49 to 6.02%). Percent Hispanic is significant only in Models 2 and 6, but failed to
converge for Model 3 and was not significant in Models 1, 4, and 5.

The eight race variables are not as consistently predictive of blood lead levels as the income variables.
Additionally, some display differing relationships across the time periods, which may be related to
changes in the distributions of the variables over the time periods. Three different variables provided
best-fitting models as seen in Table 4-1.  Percent Multiple races provided the best fit for Models 1 and 2;
Percent Asian provided the best fit for Models  3 and 4; and Percent White provided the best fit for
Models 5 and 6.
                                               D-3
This information is distributed solely for the purpose of pre-disseminalion peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
D. 1.3    Housing Cost Variables

Figure/Table A19 and A20 contain the exploratory results of the two variables related to housing cost.

Median Rent - GM blood lead levels consistently decline as Median Rent increases across all four time
periods, although the slope of the decline appears to be decreasing with each successive time period.
Mean and median Median Rent declines slightly from the initial time period to the later three periods.
Median Rent is significant in all models except Model 6.
Housing Value - Predicted GM blood lead levels consistently decline as Housing Value increases
across all four time periods, although similar to Median Rent the slope of the decline appears to be
decreasing with each successive time period as the intercept decrease.  Mean and median Housing Value
declines from the initial time period to the later three periods. Housing Value is significant in Models 1
to 5, but Model 6 failed to converge.

The log-likelihood statistics in Table 4-1 indicate that Median Rent provided a better fit in each model
than Housing Value.

D. 1.4    Occupancy Variables

Figure/Table A18 contains results of exploring Percent of Rental Units and Figure/Table A21 contains
results of exploring Percent of Vacant Units.

Percent Rented Units - Consistent relationships between GM blood lead and percent Rented Units are
seen across the four time periods, with GMs slightly increasing as percent Rented Units increases. Mean
and median percent Rented Units declines very slightly from the initial time period to the later three
periods.  Percent  Rented Units is significant in Models 4 to 6, but not in Models 1 to 3.
Percent Vacant Units - The relationship between predicted GM blood lead and percent Vacant Units is
relatively flat across the four time periods, although there appears to be a slight negative relationship in
the 1995-2000 time period  and very slight positive relationship in the other time periods. Mean and
median percent Vacant Units increases from the  initial time period to the later three periods.  Percent
Vacant Units is significant  in Models 3 to 6, but  not in Models 1 and 2.

The log-likelihood statistics in Table 4-1 indicate that Percent Vacant Units provided a better model fit
for Models 2 through 6. Model  1 was the exception in which Percent Rented Units provided a better fit,
although the  log-likelihood statistics from Model 1 are nearly identical.

D. 1.5    Single Parent Status Variable

Exploratory analysis results are contained in Figure/Table A22.

Percent Single Parent Households - The relationship  between predicted GM blood lead and percent
Single Parent Households is slightly increasing across the four time periods, although there appears to be
a slightly more positive relationship in the 1995-2000 time  period that is flattening  out in the 2004-2006
                                               D-4
This information is distributed solely for the purpose of pre-dissemmation peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
period.  Mean and median percent Single Parent Households decrease very slightly across the four time
periods.  Percent Single Parent Households is significant in Models 1 to 5, but Model 6 failed to
converge.

D. 1.6    Housing Age Variables

A number of variables related to housing age by county were investigated to identify those that best
predict children's blood lead levels. The results of the 12 variables explored are contained in
Figure/Tables A23 to A34.

Median Year Built - Consistent relationships between GM blood lead and Median Year Built are seen
across the four time periods with blood lead levels declining as age of housing increases, although it
appears that the 1995-2000 period has a slightly steeper slope. Mean and median Median Year Built
increase by about a year from the 1995-2000 time period to the later three periods.  Median Year Built is
significant in all models except Model 5, which failed to converge.
Median Year Occupied Units were Built - The results for Median Year Occupied Units were Built is
very similar to those observed for Median Year Built.  Consistent relationships between GM blood lead
and Median Year Occupied Units were Built are seen across the four time periods with blood lead levels
declining as age of housing increases, although it appears that the 1995-2000 period has a slightly
steeper slope. Mean and median Median Year Occupied Units were Built increase by about a year from
the 1995-2000 time period to the later three periods. Median Year Occupied Units were Built is
significant in all models.
Percent Units Built Before 1940 - Consistent relationships between GM blood lead and Percent Units
Built Before 1940 are seen across the four time periods with blood lead levels increasing as the
percentage increases, although it appears that the 1995-2000 period increases at a slightly faster rate.
Mean and median Percent Units Built Before 1940 decrease from the 1995-2000 time period to the later
three periods with the mean dropping by nearly two percent.  Percent Units Built Before 1940 is
significant in Models 1 to  5,  but failed to converge in Model 6.
Percent Units Built Before 1950 - Consistent relationships between GM blood lead and Percent Units
Built Before 1950 are seen across the four time periods with blood lead levels increasing as the
percentage increases, although it appears that the 1995-2000 period increases at a slightly faster rate.
Mean and median Percent Units Built Before 1950 decrease from the 1995-2000 time period to the later
three periods with the median dropping by nearly four percent. Percent Units Built Before 1950 is
significant in Models 1, 2,4, and 5, but failed to converge in Models 3 and 6.
Percent Units Built Before 1960 - Similar results are seen with this variable as with the previous two
(percent of units built before 1940 and 1950). Percent Units Built Before 1960 is significant in all
Models except Model 3 that failed to converge.
Percent Units Built Before 1970 - Again, these results are very similar to the other Percent Units Built
Before variables.  Each time period has a consistently positive relationship between blood lead levels
and percent  built before 1970 and mean and median percentages decline from the 1995-2000 time period
to the others. Percent Units Built Before 1970 is significant in all models.
Percent Units Built Before 1980 - Results are  very similar to those for Percent Units Built Before 1970.
Percent Units Built Before 1980 is  significant in all models.

                                               D-5
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA  It does not represent and should not be construed to represent
any Agency determination or policv.

-------
Percent Occupied Units Built Before 1940 - The results of the exploratory analyses on this variable are
very similar to those seen from the Percent Units Built Before 1940 variable, and most of the other
Percent Units Built Before variables. Percent Occupied Units Built Before 1940 is significant in all
models.
Percent Occupied Units Built Before 1950 - Similar results. Percent Occupied Units Built Before 1950
is significant in all models.
Percent Occupied Units Built Before 1960 - Similar results. Percent Occupied Units Built Before 1960
is significant in all Models except Model 3 that failed to converge, which also occurred with the Percent
Units Built Before 1960 variable.
Percent Occupied Units Built Before 1970 - Similar results. Percent Occupied Units Built Before 1970
is significant in all models.
Percent Occupied Units Built Before 1980 - Similar results. Percent Occupied Units Built Before 1980
is significant in all models.

The exploratory results of all 10 of the percentage variables appeared to be quite similar. Whether
percent of all housing units or percent of occupied units was used did not seem to make a difference in
the results. Similarly, the cutoff year used also did not seem to impact results.  Four different percentage
variables were among the best-fitting models, in addition to the Year Built variable.  As seen in Table 4-
1, Year Built provided the best fit for Model 1; Percent Built Before 1940 provided the best fit for
Model 2; Percent Occupied Units  built Before 1950 provided the best fit for Model 3; Percent Built
Before 1950 provided the best fit for Models 4 and 5; and Percent Occupied Units built Before 1940
provided the best fit for Model 6.  Thus, it appears that the cutoff years 1940 and 1950 for housing age
generally provide better model fits.

D.I.7    Children's Age Variables

Figure/Table A35 and A36 contain the exploratory results for percentage and number of residents less
than six years old, respectively.

Percent Less Than  6 Years of Age - Consistent relationships between GM blood lead and Percent Less
Than 6 Years of Age are seen across the four time periods with blood lead levels declining slightly
percentage less than six increases. The mean and median percents are fairly stable across the time
periods.  Percent Less Than 6 Years of Age is significant in all models.
Number Less Than 6 Years of Age - The relationship between GM blood lead and Number Less Than 6
Years of Age is not consistent across the four time periods. In the 1995-2000 and 2000-2002 data,
predicted blood lead levels  increase  as number of children increases. In the 2002-2004 data, a slight
negative relationship is seen between predicted blood lead levels and number of children, while the
relationship is flat in the most current data. Mean and median Number Less Than 6 Years of Age
decrease steadily across the four time periods. Number Less Than 6 Years of Age is significant in all
models except Model 3 that failed to converge.
                                               D-6
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
Not surprisingly, as reported in Table 4-1, the Percent Less Than 6 Years of Age variable provided the
best fit to all six models between these two variables related to amount of children less than six in each
county.

D.I.8    Education Level Variables

Exploratory analyses were conducted upon the four percentage variables. Results are detailed in
Figure/Table A37 to A40.

Percent Residents with Less Than P"' Grade Education - Consistent relationships between GM blood
lead and Percent Residents with Less Than 9th Grade Education are seen across the four time periods
with blood lead levels increasing slightly as the percentage increases, although it appears that the 1995-
2000 period increases at a slightly faster rate. Mean and median Percent Residents with Less Than 9lh
Grade Education increase from the 1995-2000 time period to the later three periods with the mean
increasing by over one percentage point.  Percent Residents with Less Than 9th Grade Education is only
significant in Models  1 and 2, failed to converge in Model 3, and is not significant in Models 4 to 6.
Percent Residents Without a High School Degree - The exploratory results on this variable are similar
to those for the less than 9th Grade variable. Consistent relationships between GM blood lead and
Percent Residents Without a High School Degree are seen across the four time periods with blood lead
levels increasing slightly as the percentage increases, although less of a relationship is seen in the 2004-
2006 data.  Mean and median Percent Residents Without a High School Degree increase from the 1995-
2000 time period to the later three periods.  Percent Residents Without a High School Degree is
significant in Models  1 to 4, failed to converge in Model 6, and is not significant in Model 5.
Percent Residents Without College Education - Consistent relationships between GM blood lead and
Percent Residents Without College Education are seen across the four time periods with blood lead
levels increasing slightly as the percentage increases, although the 2004-2006 data has less steep of a
slope.  Mean and median Percent Residents Without College Education increase from the 1995-2000
time period to the later three periods.  Percent Residents Without College Education is significant in
Models 1, 2,4, and 5, failed to converge in Model 3, and is not significant in Model 6.
Percent Residents Without College Degree - Similar results were obtained for this variable.  The
relationship between predicted blood lead levels and Percent Without College Degree is consistent
across time, with higher lead levels associated with higher percentages and a flattening of that
relationship in the current data. Mean and median  Percent Residents Without College Degree increase
slightly from the 1995-2000 time period to the later three periods. Percent Residents Without College
Degree is significant in all models expect Model 6.

Similar exploratory results were seen across these four variables focusing on parents' education level.
In general, higher lead levels are associated with less education.  Also, the percentage of residents
included in the analyses without various education levels increases slightly across the four time periods.
Table 4-1 reports that Percent Residents Without Any College and Percent Residents Without College
Degree provide the best model fits across the six models.  Percent Residents Without Any College
provided the best fit for Models 2, 5, and 6, while Percent Residents Without College Degree provided
the best fit for models 1, 3, and 4.

                                               D-7
This information is distributed solely for the purpose of pre-disseminalion peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
D.I.9    Population Variables

The detailed results of the exploratory analyses on the three variables in this category are included in
Figure/Table A41 to A43.

Total Housing Units - The relationship between GM blood lead and Total Housing Units is not
consistent across the four time periods. In the 1995-2000 and 2000-2002 data, predicted blood lead
levels increase as number of housing units increases. In the 2002-2004 data, a slight negative
relationship is seen between predicted blood lead levels and number of housing units, while the
relationship is flat in the most current data.  Mean and median Total Housing Units in each county
decrease steadily across the four time periods. Total Housing Units is significant in all models except
Model 3 that failed to converge.
Total Population - The results for Total Population are very similar to Total Housing Units.  The
relationship between GM blood lead and Total Population is not consistent across the four time periods.
In the 1995-2000 and  2000-2002 data, predicted blood lead levels increase as population increases.  In
the 2002-2004 data, a slight negative relationship is seen between predicted blood lead levels and
population, while the relationship is flat in the most current data. Mean and median Total Population in
each county decrease steadily across the four time periods.  Total Population is significant in  all models.
Housing Density - The Housing Density results are skewed by a very large maximum density in the
2004-2006 data that is much higher than in the other time periods (34,757 compared to 7,421).  In
general, mean and median Housing Density declines from the initial time period to  the later three time
periods, although the mean of the 2004-2006 period is pulled up by the outlier value. Housing Density
is only significant in Model 3.  Model 5 failed to converge, but Housing Density was not significant in
the other four models. These results may change if the outlier county is removed.

Generally, the Total Population and  Total Housing Units variables appeared similar. Surprisingly, from
among the three population-focused variables, Housing Density provided the best fit to the data for four
of the six models - Models 1,2,4, and 6. As reported  in Table 4-1, Total Population provided the best
for Models 3 and 5.

D.2   Analyses of National Blood Lead Data by Environmental Variables

Environmental data analyzed for the national models included the ASPEN air modeling data, TRI data,
and drinking water data.  Presented below are exploratory analysis results from investigating  these three
variable types.

D.2.1    Air Lead Variables

The model estimates toxic air pollutant concentrations  for every census tract in the  continental United
States, however - these data are only available for 1999.  Three variables representing modeled
estimates oftoxic  air pollutant concentrations were obtained from the ASPEN model for each county and
investigated - average air lead level, median air lead level,  and 95th percentile of air lead level.
Figures/Tables A44 to A46 contain the detailed exploratory analysis results.

                                              D-8
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
Average Air Lead - The Average air lead levels are impacted by a small number of very large values.
For example, in the 1995-2000 data, the 90lh percentile is .0044 and the maximum is nearly 50 times
higher at.2132. Consistent relationships between predicted GM blood lead levels and average air lead
levels are seen across the four time periods, generally with higher average air levels associated with
higher blood lead levels. That relationship, however, appears to be flattening in the more current data.
Mean and median Average Air Lead decrease slightly across the time periods.  Average Air Lead was
significant in Models 4 through 6, but was not significant in the other three models.
Median Air Lead - Median Air Lead also appears to be impacted by some relatively high values.
Consistent relationships between predicted GM blood lead levels and Median Air Lead are seen across
the four time periods, with higher Median Air Lead associated with  higher blood lead levels. That
relationship also appears to be flattening in the more current data. Mean and median Median Air Lead
decrease slightly from the initial time period to the later time periods. Average Air Lead was significant
in all models except Model 3, in which it was borderline significant (p-value=.061).
Air Lead 95"' Percentile - The relationship between Air Lead 95th Percentile and predicted GM blood
lead levels is not as evident. While the first two time periods (1995-2000 and 2000-2002) show a slight
positive relationship, that relationship appears flat in the 2002-2004 data and with  a slight negative
association in the most current data. Mean and median Air Lead 95lh Percentile also decrease slightly
from the initial time period to the later time periods.  Air Lead 95th Percentile was  significant in Model
5, not significant in Models 1 and 2, and the other three models failed to converge.

Table 4-1 reports that Median Air Lead provided the best model fits of the three air lead-related
variables for Models 1  to 3.  Average Air Lead provided the best model fits for Models 4 and 6.  95th
Percentile Air Lead provided the best model fit for Model 5.

D.2.2    Toxics Release Inventory Variables

Three types of TRI variables were utilized - total compounds, lead only, and total  lead. Within each
types, five pollution variables were explored - total lead in the air, lead in fugitive air, lead from
smokestacks, lead in surface water, and lead in water by injection. The results for the  15 TRI variables
are discussed below and presented in Figures/Tables A50 to A64.

TRI Compounds (Total Air) - Higher predicted blood lead levels are associated with higher levels of
lead compounds in total air in each of the four time periods; however, the slopes become less positive
with each successive period. The distributions of lead compounds in total air decline consistently across
the four time periods as the mean declines from 585.6 in the 1999-1999 period to 435.5 in the 2004-
2006 period. TRI Compounds (Total Air) is significant in Models 4 and 5 and not significant in the
other four models.
TRI Compounds (Fugitive Air) - The results from the analysis of TRI Compounds in fugitive air are
similar to those for TRI Compounds (Total Air).  Higher predicted blood lead levels are associated with
higher levels of lead compounds in total air in each of the four time periods, with an apparent slight
decline in the slopes over time.  The distributions of TRI Compounds (Fugitive Air) decline consistently
across the four time periods as the mean declines from 192.1 in the  1999-1999 period to 137.5 in the
2004-2006 period. TRI Compounds (Fugitive Air) is significant in Models 1, 2, and 4, but not
significant in the other three models.
                                               D-9
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy

-------
TRI Compounds (Air Lead from Stacks) - Generally, higher predicted blood lead levels are associated
with higher TRI Compounds (Air Lead from Stacks) levels; however, the relationship flattens over time.
Mean levels of TRI Compounds (Air Lead from Stacks) consistently decline with each successive time
period. The TRI Compounds (Air Lead from Stacks) variable is not significant in Models 1 to 3,5, and
6, and Model 4 failed to converge.
TRI Compounds (Water Surface) - The relationship between predicted GM blood lead levels and TRI
Compounds (Water Surface) is negative across all four time periods, although it flattens slightly in the
most recent period. Mean and 90  percentile levels of TRI Compounds (Water Surface) decline with
each successive time period. TRI Compounds (Water Surface) is not significant in any of the six
models.
TRI Compounds (Water by Injection) - Slightly lower predicted GM blood lead levels are associated
with higher levels of TRI Compounds (Water by Injection) across each time period. Although over 90
percent of quarterly county records have values of 0 in each time period, mean levels of TRI
Compounds (Water by Injection) increase over time, with the 2002-2003 and 2004-2006 periods being
nearly equal. TRI Compounds (Water by Injection) is not significant in any of the six models.
TRI Lead Only (Total Air) - Although there is a positive relationship between predicted GM blood lead
levels and TRI Lead Only (Total Air) in the 1995-1999 period, the relationship appears to be flat in the
ensuing three periods. Mean and 90lh percentile levels of TRI Lead Only (Total Air) decline over the
first three time periods, and then increase slightly in the 2004-2006 data. TRI Lead Only (Total Air) is
significant in Models 3  to 5, not significant in Models 1 and 2, while Model 6 failed to converge.
TRI Lead Only (Fugitive Air) - While a positive relationship between predicted GM blood lead levels
and TRI Lead Only (Fugitive Air) is evident in the 1995-1999 period, this relationship flattens out in the
ensuing three periods where it is only slightly positive. Mean and 90th percentile levels of TRI Lead
Only (Fugitive Air) are higher in the 1995-1999 period than in the subsequent three periods.  TRI Lead
Only (Fugitive Air) is not significant in any of the six models.
TRI Lead Only (Air Lead from Stacks) - While a positive relationship is evident between predicted GM
blood lead levels and TRI Lead Only (Air Lead from Stacks)  in the 1995-1999 period, this relationship
is flat in the ensuing three periods. Mean and 75lh and 90lh percentile levels of TRI Lead Only (Air Lead
from Stacks) are higher in the 1995-1999 period than in the subsequent three periods,  although there is a
slight increase in the most current period. TRI Lead Only (Air Lead from Stacks) is significant in
Models 3 to 5 and not significant in any Models 1 and 2, while Model 6 failed to converge.
TRI Lead Only (Water Surface) - The relationship between predicted GM  blood lead levels and TRI
Lead Only (Water Surface)  is positive across the first three time periods, but in the 2004-2006 period
lower predicted blood lead levels are associated with higher water lead  levels. Over 75 percent of
county-level values are 0 across all four time periods. Mean and 90lh percentile levels of TRI Lead Only
(Water Surface) are higher in the 1995-1999 period than in the subsequent three periods. TRI Lead
Only (Water Surface) is significant for Models 5 and 6, but not significant for Models 1 to 3, while
Model 4  failed to converge.
TRI Lead Only (Water by Injection) - Generally, slightly lower predicted GM blood  lead levels are
associated with higher levels of TRI Compounds (Water by Injection), although the relationship appears
to be flattening over time. Over 90 percent of quarterly county records have values of 0 in each time
period. Mean levels of TRI Compounds (Water by Injection) are significantly lower in the 1995-1999

                                             D-10
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
period than in the later three periods.  TRI Compounds (Water by Injection) is not significant in any of
the six models.
TRI Total Lead (Total Air) - Higher predicted blood lead levels are associated with higher levels of
Total  Lead in total air, although the slopes are becoming less positive over time.  Mean and median
levels of TRI Total Lead (Total Air) decline with each successive time period. TRI Total Lead (Total
Air) is significant in Models 3 to 5, not significant in Models 1 and 2, while Model 6 failed to converge.
TRI Total Lead (Fugitive Air) - Higher predicted GM blood lead  levels are associated with higher
levels of TRI Total Lead (Fugitive Air) in each time period. Mean, 75th percentile, and 90lh percentile
values of TRI Total Lead (Fugitive Air) decline with each successive time period. TRI Total Lead
(Fugitive Air) is not significant in Models 1, 2, 4, and 6, while Models  3 and 5 failed to converge.
TRI Total Lead (Air Lead from Stacks) - Although higher predicted GM blood lead levels are
associated with higher levels of TRI Total Lead (Air Lead from Stacks) from  1995-2001, the
relationship flattens in the 2002-2003 period and even becomes slightly negative in the most current
period.  The distribution of TRI Total Lead (Air Lead from Stacks) is much higher in the 1995-1999
period than in the later periods and continues to decline slightly over time after that. TRI Total Lead (Air
Lead  from Stacks) is significant in Models 4 to 6, is borderline significant in Model 3 (p-value=0.05),
and is not significant in Models 1 and 2.
TRI Total Lead (Water Surface) - The relationship between predicted GM blood lead levels and TRI
Total  Lead (Water Surface) is negative across the four time periods as predicted blood lead level
decrease as lead in water increases; however, the relationship appears to be flattening over time. The
distribution of TRI Total Lead  (Water Surface) is declining across time, as  mean levels fall from 124 in
the 1995-1999 period to 78 in the 2004-2006 period. TRI Total Lead (Water  Surface) is significant in
Models  5 and 6, not significant in Models 1 to 3, while Model 4 failed  to converge.
TRI Total Lead (Water by Injection) - In each time period, lower predicted GM blood lead levels are
associated with higher levels of TRI Compounds (Water by Injection).  Over  90 percent of quarterly
county records have values of 0 in each time period. Mean levels of TRI Compounds  (Water by
Injection) are significantly lower in the 1995-1999 period than in the later three periods. TRI
Compounds (Water by Injection) is not significant in any of the six models.

The TRI variables were mixed in their ability to predict children's blood lead levels. The three Water by
Injection variables were not significant in any models, while the Water Surface variables were only
significant for Models 5 and 6  for the Lead Only and Total Lead variables.  The air lead variables were
significant in only 32 percent of their 54 variable/model combinations.  Although the log-likelihood
ratios from each model are similar across the 15 variables for each model, TRI Compounds (Fugitive
Air) provided the best fit for Models 1, 2, and 6. The best fit for Model 3 was shared by two lead in
surface  water variables - TRI Lead Only (Water Surface) and TRI Total Lead (Water Surface). The best
fit for Models 4 and 5 was shared by TRI Lead Only (Fugitive Air), TRI Lead Only (Water by
Injection), and TRI Total Lead (Water by Injection).

D.2.3   Lead in Drinking Water

Figure/Table A49 contain the exploratory results from analyzing the Mean  Water Lead Concentration
variable from the Drinking Water Information System.

                                             D-ll
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should  not be construed to represent
any Agency determination or policy.

-------
Mean Water Lead Concentration -The relationship between predicted GM blood lead levels and Mean
Water Lead Concentration is inconsistent across the four time periods.  Over the 1995-1999 time period,
there is a highly negative relationship between blood lead levels and water lead levels, with higher water
lead levels associated with lower blood lead levels. In the 2000-2001 and 2002-2003 periods, there is a
positive relationship between the variables, although the 2002-2003 period has a more positive slope.
The relationship appears to flat in the 2004-2006 time period.  Over 90 percent of mean values are
reported as -2.3. The distribution of Mean Water Lead Concentration is nearly identical across the four
time periods, except for the very upper ends, with the maxima ranging from 1.1 to 10.6. Mean Water
Lead Concentration is significant in Models 1 to 4.

D.3   Analyses of National Blood Lead Data by Programmatic Variables

The programmatic data explored  for the national model include (1) HUD grant funding related to lead
hazard control and lead poisoning and (2) EPA Region.

D.3.1    Programmatic Funding

The funding data analyzed at the  national level include grant funding histories for HUD only at this
point. Two variables were generated from these data and analyzed - current and cumulative per-capita
dollars allocated to each  county to combat childhood lead poisoning. The detailed exploratory analysis
results for these two variables are presented in Figures/Tables A47 and A48.  Note that these analyses
may be impacted by the relatively low percentage of counties that have received HUD funding.

Current HUD Funding - The relationship between predicted GM blood lead levels and Current HUD
Funding is not consistent across the four time periods. In the earlier time periods (1995-2000 and 2000-
2002), higher funding amounts are associated with lower blood lead levels. That association appears flat
in the 2002-2004 data and appears slightly positive in the current data; however, the 2004-2006 time
period includes a maximum funding amount of $76M that may be impacting that relationship. Mean
Current HUD Funding changes across the time periods, while the median Current HUD Funding levels
are 0.0 for all periods. Current HUD Funding is found to be a significant predictor of GM blood lead
levels across all six models.
Cumulative HUD Funding - Similar inconsistencies are evident in the exploratory analyses on
Cumulative HUD Funding levels. In 1995-2000, there appears to be a strong relationship between
higher funding levels and lower GM lead levels. The relationship is less strong but still in the same
direction in the 2000-2002 data, but flattens in the 2002-2004  and 2004-2006 data. Mean and median
Cumulative HUD Funding levels increase over the four time periods, which makes sense because of the
nature of this variable. Cumulative HUD Funding is significant in Models 2 to 4 and 6, but Model 5
failed to converge and it was not  significant in Model 1.

Table 4-1 reports that Current HUD Funding provided lower log-likelihood statistics and, thus, better
model fits than Cumulative HUD Funding for all models except Model 6.
                                             D-12
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
D.3.2    EPA Region

EPA Region was investigated as a potential predictor of children's blood lead levels to determine if that
high-level geographic indicator should be included in multivariate models. Figure/Table A65 contains
the exploratory results for this variable.  The line plot in Figure A65 indicates that in nearly all regions
predicted CM blood lead levels decline over time expect in Region 1, where the predicted blood lead
levels increase over the period from 1995-2006.  The model output in Table A65 indicates that blood
lead levels in each individual EPA Region from Region 1 to Region 9 are significantly greater than
levels in Region 10, which serves as the baseline. In Models 2 and 4, only Region 8 is not significantly
different than Region 10. In Model 5, Regions 2,4, 6, and 8 are not significantly different than Region
10. In Model 6, only Region 9 is found to be significantly different than Region 10.
                                                D-13
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not beenfonnally disseminated by EPA.  It does not represent and should not be construed to represent
any Agency determination or policy.

-------
                      Appendix E



Detailed Discussion of Massachusetts Exploratory Analyses

-------
E.     Relationship between Local Blood Lead Data and Explanatory Variables

E.I    Analyses of Local Blood Lead Data by Demographic Variables

As noted above, the identical set of 43 Census variables across nine general categories were explored to
determine their association with the Massachusetts surveillance data.

E. 1.1  Income Variables

Initial results from exploring and modeling the income-related variables are presented in Figures/Tables
Bl to B9. Highlights from the analysis of each exploratory variable are listed below.

Median Family Income - Similar relationships between GM blood lead and Median Family Income are
seen across the three  time periods, with GMs decreasing as Median Family Income increases. Blood
lead  levels decline slightly between each consecutive time periods. The distribution of Median Family
Income is stable across the three time periods.  Median Family Income is significant in Models 1 to 5.
The findings for Massachusetts were very similar to those seen  in the national data.
Median Household Income - Similar relationships between GM blood lead and Median Household
Income are seen across the three time periods, with GMs decreasing as Median Household Income
increases. Blood lead levels decline steadily across the three periods.  Median Household Income is
stable across the three time periods. Median Household Income is significant in Models 1 to 5.
Median Per Capita Income - Consistent relationships between GM blood lead and Median Per Capita
Income are seen across the four time periods, with GMs decreasing as Median Per Capita Income
decreases. The distribution of Median Per Capita Income shows a very slight increase across the three
periods, about 0.1 percent from the 2000-2002 period to the 2005-2006 period. Median Per Capita
Income is significant in Models 1 to 5.
Percent Units with No Household Earnings - Consistent relationships between GM blood lead and
Percent of Units with No Household Earnings are seen across the three time periods, with GMs
increasing as percent of units with no household earnings increases. Mean and median Percent of Units
with No Household Earnings do not change across the three time periods.  Percent of Units with No
Household Earnings  is significant in Models 1 to 5.
Percent Units with No Household Wage - This variable performs similarly to the previous one.
Consistent relationships between  GM blood lead and Percent of Units with No Household Wages are
seen across the three  time periods, with GMs increasing as the percentage increases. Mean and median
Percent of Units with No Household Wages is stable across the three time periods.  Percent of Units
with No Household Wages is significant in Models 1 to 5.
Percent Households  on Public Assistance - Consistent relationships between GM blood lead and
Percent of Households on Public  Assistance are seen across the three time periods, with GMs increasing
as Percent of Households on Public Assistance increases. Mean and median Percent of Households on
Public Assistance are stable across the three periods. Percent of Households on Public Assistance is
significant in Models 1 to 5.
Percent Households  Below Poverty Line - Consistent relationships between GM blood lead and
Percent of Households Below the Poverty Line are seen across the three time periods, with GMs

                                              E-l
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
increasing as Percent of Households Below the Poverty Line increases. Mean and median Percent of
Households Below the Poverty Line are stable across the three periods. Percent of Households Below
the Poverty Line is significant in Models 1 to 5.
Percent Units with FamUy Income Below Poverty Line - Consistent relationships between GM blood
lead and Percent of Units with Family Income Below the Poverty Line are seen across the three time
periods, with GMs increasing gradually as percentage of units increases. Mean and median Percent of
Units with Family Income Below the Poverty Line are stable across the three periods. Percent of Units
with Family Income Below the Poverty Line is significant in Models 1 to 5.
Percent Units Spending Less than Five Years in Poverty - Consistent relationships between GM blood
lead and Percent of Units Spending Less than Five Years in Poverty are seen across the three time
periods, with GMs increasing as percent of units increases. Mean and median Percent of Units Spending
Less than Five Years in Poverty do not change across the three periods, although the latter two  periods
have a significantly higher maximum value (17.3 vs. 12.7 percent).  Percent of Units Spending  Less than
Five Years in Poverty is significant in Models 1 to 3 and 5, but failed to converge in Model 4.

Across these nine variables, very few differences were evident from the exploratory analyses. In all
cases, each of the three time periods had similar distributions and similar relationships with predicted
GM blood lead levels. Of the nine variables in the Income category, 8 were significant for Models 1
through 5, with only Model 6 (proportion of children with blood lead levels over 25 ng/dL) failing to
converge. As seen in Table 4-2, the variable Median Household Income provided the best fit in Models
1 to 4, while Percent Units with No Household Wage provided the best fit for Model 5.

E.I.2    Race Variables

Results from the eight race variables explored are presented in Figures/Tables BIO to B17, with
highlights listed below.

Percent American Indian and Alaskan Native Alone - As expected, most Census tracts in
Massachusetts have very low percentages of American Indian and Alaskan Natives, with the  90lh
percentile being 0.5 percent even lower than the 1.2 percent nationally. The three time periods display
similar relationships between blood lead levels and percent American  Indian and Alaskan Native, as
blood lead levels increase as Percent American Indian increases. The mean percent of American Indian
and Alaskan Native is steady across the three time periods at 0.19 percent. Percent of American Indian
and Alaskan Native is a significant predictor of the outcome measure in Models 1 to 5.
Percent Asian Alone - The relationship between GM blood lead and percent Asian appears to be
changing slightly across the three time periods. While the relationship between GM blood levels and
Percent Asian was nearly flat with just a slight negative slope, in each the two successive time periods
the slopes are more negative.  This indicates that  in more recent years  lower blood lead levels are
associated with Census tracts with higher percentages of Asian people. The distribution of Percent
Asian is stable across the three time periods.  Percent Asian is significant in Models 1 to 3, but  not in
Model 5, while Model 4 failed to converge.
Percent Black Alone - Percent Black Alone in each Census tract ranges from 0 to 91 percent.
Consistent relationships between GM blood lead  and Percent Black are seen across the three  time

                                               E-2
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent
any Agency determination or policy.

-------
periods, with GMs increasing as Percent Black increases. Mean and median Percent Black are stable
across the three periods.  Percent Black is a significant predictor of the outcome variable in Models 1 to
5.
Percent White Alone - The distribution of Percent White is also wide-ranging with Census tracts
containing 0.1 to 100 Percent White. The relationship between GM blood lead and Percent White is
consistent across the three time periods, with GMs declining gradually as Percent White increases. The
distribution of Percent  White is stable across the three time periods, as mean and median Percent White
stay about the same. Percent White is a significant predictor of the outcome variable in Models 1 to 5.
Percent Native Hawaiian and Other Pacific Islander Alone - Very few Native Hawaiians live in
Massachusetts as evidenced by a 90lh percentile of Percent Native Hawaiian of 0.0 percent. The
distribution of Percent  Native Hawaiian is stable across the three time periods with a mean of 0.02
percent for each period. The relationship between GM blood lead levels and Percent Native Hawaiian is
consistently flat  across the three time periods. Percent Native Hawaiian is not significant in Models 1 to
5.
Percent Other Race Alone - The relationship between Percent Other Race Alone and GM blood lead
levels in Massachusetts is opposite of the national trend. GM blood lead levels increase steadily as
percent Other Race increases across all three time periods; however, the increase appears slightly steeper
in the 2000-2002 time  period. Mean and median percent Other Race  remain fairly stable across time
periods, with an average mean percentage of 0.75  across the time periods. Percent Other Race is
significant in Models 1 to 5.
Percent Multiple Races - Predicted GM blood lead levels consistently increase as Percent Multiple
Races increases  across the three time periods, although the 2000-2002 period again has a slightly more
positive slope. Mean and median Percent Multiple Races are stable across the three periods.  Percent
Multiple Races is significant in Models 1 to 5.
Percent Hispanic - Predicted GM blood lead levels generally increase gradually as Percent Hispanic
increases, however the 2000-2002 time period had a slightly less positive slope. Mean and median
Percent Hispanic are stable across the three time periods. Percent Hispanic  is significant in Models 1  to
5.

Similar to the national  data, in the Massachusetts data the eight race variables were not as consistently
predictive of blood lead levels as the income variables; however six of the eight variables  were
significant predictors in the five models that consistently converged - Models 1 to 5. The only two
variables that yielded different results were Percent Asian and Percent Native Hawaiian. The race
composition of the Census tracts included in each  of the three time periods was consistent across the
periods.

Two different variables provided best-fitting models as seen in Table 4-2. Percent Multiple Races
provided the best fit for Models 1, 2, 3, and 4; and Percent Native Hawaiian provided the best fit for
Model 5 (proportion of children with blood lead levels above 15 ug/dL), although it was not a
significant predictor in that model. Percent Asian provided the second best  fit in Model 5, but it also
was not significant.
                                               E-3
This information is distributed solely for the purpose of pre-dissemmation peer review under applicable information quality
guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
E. 1.3    Housing Cost Variables

Two variables related to housing costs in Massachusetts, Median Rent and Housing Value, were
analyzed to explore their relationships with children's blood lead levels.  Figure/Table B19 and B20
contain the exploratory results from these two variables.

Median Rent - Predicted GM blood lead levels decline sharply as Median Rent increases across all
three time periods, with similar slopes. The distribution of Median Rent is quite stable across the three
time periods, as the means are within $1 of each other and medians are all $619. Median Rent is
significant in Models 1 to 5.
Housing Value - Predicted GM blood lead levels consistently decline as Housing Value increases
across all three time periods, although the slopes are not as steep as with Median Rent.  Mean Housing
Value in each time period is around $204,000 while median Housing Value is about $172,000 in each
time period. Housing Value is significant in Models  1 to 5.

As in the national model, Median Rent provided a better fit than Housing Value in each model, as
evidenced by the log-likelihood statistics in Table 4-2.

E. 1.4    Occupancy Variables

Figure/Table B18 contains results of exploring Percent of Rental Units and Figure/Table B21 contains
results of exploring Percent of Vacant Units.

Percent Rented Units - Consistent relationships between GM blood lead and Percent Rented Units are
seen across the three time periods, with GMs increasing gradually as Percent Rented Units increases.
The distribution of Percent Rented Units is stable across time periods, with means of about 39.6 percent
and medians about 34.8 percent. Percent Rented Units is  significant in Models 1 to 5.
Percent Vacant Units - The relationship between predicted GM blood lead and percent Vacant Units is
similar across the three time periods, with blood lead levels increasing as percentage increases.  The
2000-2002 time period has a more positive slope than the other two periods.  The distribution of Percent
Vacant Units is stable across the time periods, with means of about 5.35 percent and medians at 3.2
percent.  Identical to the modeling results for Percent Rented Units, Percent Vacant Units is significant
in Models 1 to 5.

The log-likelihood statistics in Table 4-2 indicate that Percent Rented Units provided a better model fit
for Models 1 through 4.  Model 5 was the exception in which Percent Vacant Units provided a better fit.
This is counter to the national models, in which Percent Vacant Units achieved the better fit in all but
one model.

E. 1.5    Single Parent Status Variable

Exploratory analysis results are contained in Figure/Table B22.
                                               E-4
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
Percent Single Parent Households - Predicted GM blood lead levels increase as Percent Single Parent
Households increases across the three time periods.  The distribution of Percent Single Parent
Households is nearly identical across the three time periods.  Percent Single Parent Households is
significant in Models  J to 5. For comparison, the fit of Models 1 to 3 evaluating Percent Single Parent
Households is better than any models from the previous two variable categories.

E. 1.6    Housing Age Variables

The results of the 12 variables explored are contained in Figure/Tables B23 to B34.

Median Year Built - Note that the minimum Year Built is 1939. According to the histogram in Figure
B23, this category accounts for the average Median Year Built in over 30 percent of all Census tracts.
Consistent relationships between GM blood lead and Median Year Built are seen across the three time
periods with blood lead levels declining as age of housing increases, although it appears that the 2000-
2002 period has a slightly steeper slope. The distribution of Median Year Built is identical across the
three periods. Median Year Built is significant in all models.
Median Year Occupied Units were Built - The results for Median Year Occupied Units were Built is
very similar to those observed for Median Year Built.  Consistent relationships between predicted GM
blood lead and Median Year Occupied  Units were Built are seen across the three time periods with
blood lead levels  declining as age of housing increases, although it appears that the 2000-2002 period
has a slightly steeper slope. The distribution of Median Year Occupied Units were Built across the three
time periods are nearly identical.  Median Year Occupied Units were Built is significant in Models J to
5.
Percent Units Built Before 1940 - Consistent relationships between GM blood lead and Percent Units
Built Before 1940 are seen across the three time periods with blood lead levels increasing  as the
percentage increases, although it appears that the 2000-2002 period increases at a slightly  faster rate.
The distribution of Percent Units Built  Before  1940 across the three time periods are nearly identical.
Percent Units Built Before 1940 is significant in Models 1  to 5.
Percent Units Built Before 1950 - Similar results are seen with this variable as with Percent Units Built
Before 1940.  Consistent relationships between GM blood  lead and Percent Units Built Before 1950 are
seen across the three time periods with  blood lead levels increasing as the percentage increases, although
it appears that the 2000-2002 period increases at a slightly faster rate. The distributions of Percent Units
Built Before 1950 across the three time periods are nearly identical.  Percent Units Built Before 1950 is
significant in Models  1 to 5.
Percent Units Built Before 1960 - Similar results are seen with this variable as with the previous two
(percent of units built before 1940 and  1950).  Percent Units Built Before 1960 also is significant in
Models 1  to 5.
Percent Units Built Before 1970 - Again, these results are very similar to the other Percent Units Built
Before variables.  Each time period has a consistently positive relationship between blood lead levels
and percent built  before 1970, although the intercepts are slightly lower. The distributions of Percent
Units Built Before 1970 are stable across the three time periods.  Percent Units Built Before 1970 is
significant in Models  1 to 5.


                                               E-5
77ii5 information is distributed solely for the purpose of pre-disseminalion peer review under applicable information quality
guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
Percent Units Built Before 1980 - Results are nearly identical to those for Percent Units Built Before
1970. Percent Units Built Before 1970 is significant in Models 1 to 5.
Percent Occupied Units Built Before 1940 - The results of the exploratory analyses on this variable are
very similar to those seen from the Percent Units Built Before 1940 variable, and most of the other
Percent Units Built Before variables. Percent Occupied Units Built Before 1940 is significant in all
models.
Percent Occupied Units Built Before 1950 - Similar results. Percent Occupied Units Built Before 1950
is significant in all models.
Percent Occupied Units Built Before 1960 - Similar results. Percent Occupied Units Built Before 1960
is significant in all models.
Percent Occupied Units Built Before 1970 - Similar results. Percent Occupied Units Built Before 1970
is significant in all models.
Percent Occupied Units Built Before 1980 - Similar results. Percent Occupied Units Built Before 1980
is significant in all models.

As with the national county-level data set, the exploratory results of all 10 of the percentage variables in
the Census-tract level Massachusetts data appeared to be quite similar. Whether percent of all housing
units or percent of occupied units was used did not seem to make a difference in the results. Similarly,
the cutoff year used also did not seem to impact results. Three different percentage variables were
among the best-fitting models. As seen in Table 4-2, Percent Units Built Before 1940 provided  the best
fit for Models 1, 2, and 3.  Percent Occupied Units  Built Before  1940 provided the best fit for Model 4,
although Percent Units Built Before 1940 had a nearly identical log-likelihood statistic.  Percent
Occupied Units Built  Before 1980 provided the best fit for Model 5, whose outcome measure is
proportion of children with blood lead levels of 15  u,g/dL or greater.

E. 1.7    Children's Age Variables

Figures/Tables B35 and B36 contain the exploratory results for percentage and number of residents less
than six years old, respectively.

Percent Less Than 6  Years of Age - During the 2000-2002 time period, it appears that slightly  lower
GM  blood lead levels are associated with higher percentages of children less than six years old.  For the
other two time periods, the relationship between predicted GM blood lead level and Percent Less Than 6
is flat. The distributions of Percent Less Than 6 are stable across the three time periods. Percent Less
Than 6 Years of Age  is not significant in Models 1  to 3 and 5, while Model 4 failed to converge.
Number Less Than 6 Years of Age - Consistent relationships between GM blood lead and Number Less
Than 6 Years of Age  are seen across the three time periods with blood lead levels decreasing as the
percentage increases,  although it appears that the 2000-2002 period decreases at a slightly faster rate.
Mean and median Number Less Than 6 Years of Age are fairly stable across the three time periods.
Number Less Than 6  Years of Age is significant in Models 1 to 4, while Model 5 failed to converge.
                                               E-6
This information is distributed solely for the purpose ofpre-disseminalion peer review under applicable information quality
guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
As reported in Table 4-2, the Number Less Than 6 Years of Age variable provided the better fit for
Models 1 to 4, while Percent Less Than 6 Years of Age provided the better fit for Model 5, although this
is because the model with Number Less Than 6 Years of Age did not converge.

E. 1.8    Education Level Variables

Four variables were constructed and analyzed related to education level. Results are detailed in
Figures/Tables B37 to B40.

Percent Residents with Less Than &h Grade Education - Consistent relationships between predicted
GM blood lead and Percent Residents with Less Than 9lh Grade Education are seen across the three time
periods with blood lead levels increasing as the percentage increases. The distribution of Percent
Residents with Less Than 9th Grade Education is stable across the three periods. Percent Residents with
Less Than 9th Grade Education  is significant in all models.
Percent Residents Without a High School Degree - The exploratory results on this variable are similar
to those for the less than 9th Grade variable.  Consistent relationships between GM blood lead and
Percent Residents Without a High School Degree are seen  across the four time periods with blood lead
levels increasing as the percentage increases. The distribution of Percent Residents Without a High
School Degree is stable across the three periods.  Percent Residents Without a High School Degree is
significant in Models 1 to 5.
Percent Residents Without College Education - Consistent relationships between GM blood lead and
Percent Residents Without College Education are seen across the three time periods with blood lead
levels increasing as the percentage increases. Mean and median Percent Residents Without College
Education change only slightly  across the three periods.  Percent Residents Without College Education
is significant in Models 1 to 5.
Percent Residents Without College Degree - Similar results were obtained  for this variable. The
relationship between predicted blood lead levels and Percent Without College Degree is consistent
across time, with higher lead levels associated with higher percentages. The distribution of Percent
Residents Without a High School Degree is stable across the three periods.  Percent Residents Without
College Degree  is significant in all models.

Very similar exploratory results were seen across these four variables focusing on parents' education
level. Tn general, higher lead levels are associated with less education.  Also, the percentage of residents
included in the analyses without various education levels increases slightly across the four time periods.
Table 4-2 reports that Percent Residents Without College Education provides the best model fits for
Models 1 and 2, which model GM blood lead levels.  Percent Residents without a High School Degree
provided the best fit for Model  3. Percent Residents with Less Than 9lh Grade Education provided the
best fit for Models 4 and  5.

E. 1.9    Population Variables

The detailed results of the exploratory analyses on the three variables related to population are included
in Figures/Tables B41 to B43.

                                               E-7
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agencv determination or policy.

-------
Total Housing Units - The relationship between predicted GM blood lead and Total Housing Units is
mostly consistent across the three time periods with predicted blood lead levels decreasing slightly as
number of housing units per Census tract increases. The relationship appears to be flattening slightly
with each successive time period. The distribution of Total Housing Units within each Census tract is
similar for each time period. Total Housing Units is significant in all models.
Total Population - The results for Total Population are very similar to Total Housing Units. The
relationship between predicted GM blood lead and Total Population decreases slightly as number of
housing units per Census tract increases, with the relationship flattening slightly with each successive
time period. The distribution of Total Population within each Census tract is similar for each time
period.  Total Population is significant in Models 1 to 4, but Model 5 did not converge.
Housing Density - In each time period, predicted  GM blood lead levels increase as housing density
increases; however, it appears that the relationship flattens slightly with  each successive time period.
The distribution of Housing Density within each Census tract is similar for each time period. Housing
Density is significant in all models.

Among the population variables, Total Population provided the best model fit for Models 1 to 4. For
Model 5, that did not converge for Total Population, Total Housing Units provided the better fit.

E.2    Analyses  of Local Blood Lead Data by Environmental Variables

Environmental data acquired for this project will include air and groundwater monitoring data
aggregated at the county level for the low resolution model and at higher resolutions when possible for
the MA analyses. In cases where the data are available for a limited number of air-monitoring stations
or drinking water samples available for the region(s) being investigated, geo-spatial modeling
techniques might be used as appropriate to develop predictions across the entire region. Existence of
industrial sources of lead within each county as indicated by the toxics release inventory will also be
included as an environmental data source.  Currently, impacts of three air lead-related variables have
been analyzed for their impacts on children's blood lead levels in Massachusetts. Additionally, twelve
variables were generated from Toxics Release Inventory (TR1) data and analyzed in conjunction with
Massachusetts childhood lead data.

E.2.1    Air Lead Variables

The variable Air Dispersion (ASPEN) Model captures the output from the ASPEN model for each
Massachusetts Census tract investigated.  Exploratory results are presented in Figure/Table B44.  The
Air Exposure (HAPEM5) Model variable captures the predicted exposure data from the HAPEM5
model.  Results are presented in Figure/Table B45. The third air lead variable considered, Air Hazard
Quotient (HQ), is derived from the 1999 National  Scale Air Toxics Assessment data. This variable
represents lifetime exposure for children at the centroids of each Census tract.  Lifetime exposure is
calculated based on considering annual exposures and yearly activity patterns.  Results from the Air
Hazard Quotient  variable are  in Figure/Table B46.

Air Dispersion (ASPEN) Model - Consistent relationships between predicted GM blood lead levels and
Air Dispersion Model levels are seen across the three time periods, generally with higher air lead levels
                                               E-8
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policv.

-------
associated with slightly lower blood lead levels. The distribution of the Air Dispersion Model across the
three time periods is identical.  Air Dispersion Model was not significant in any models.
Air Exposure (HAPEM5) Model - The Air Exposure Model variable performs similarly to the Air
Dispersion Model variable. Generally, higher exposure levels are associated with slightly lower
predicted GM blood lead levels in each time period. The distribution of the Air Exposure Model across
the three time periods is nearly identical. Air Exposure Model was not significant in any models.
Air Hazard Quotient - The Air Hazard Quotient also performs similarly to the other two air lead
variables. Generally, higher Air Hazard Quotient is associated with slightly lower predicted GM blood
lead levels in each time period, although the relationship is nearly flat. The distribution of the Air
Exposure Model across the three time periods is nearly identical. Air Hazard Quotient was not
significant in any models.

Table 4-2 reports that the log-likelihood ratios from the models containing the three air lead variables
are very similar.  Although the log-likelihoods are nearly identical across the three variables, Air Hazard
Quotient had the lowest ratio in each model.

E.2.2    Toxics Release Inventory Variables

EPA's Toxic Release Inventory (TRI) catalogs various sources of lead, based on information provided
by industrial facilities. This data source was used to generate Census tract-level estimates within
Massachusetts of the total amount of lead and/or lead-containing compounds that are released by
industrial facilities into the environment via air, surface water, or underwater injection, three types of
TRI variables were  utilized - total compounds, lead only, and total lead.  Within each  types, four
pollution variables were explored - total lead in the air, lead in fugitive air, lead from smokestacks, and
lead in surface water.  Thus, 12 total TRI data variables were evaluated. The results from investigating
the 12 TRI data variables are presented in Figures/Tables B63 to B74.

TRI Compounds (Total Air) - Generally, lower predicted blood lead levels are associated with higher
levels of lead compounds in  total air.  The 2000-2002 and 2003-2004 time periods have slight negative
slopes, while the 2005-2006 relationship is nearly flat. The distributions of lead compounds in total air
are very similar across the three time periods, with  the 90th percentiles of the distributions equal to 0.00.
TRI Compounds (Total Air) is significant in Model 4 and borderline significant in Model 3 (p-
value=.064), but is not significant in the other three models.
TRI Compounds (Fugitive Air) - The relationship between TRI Compounds in fugitive air and
predicted GM blood lead levels is more consistent across the three time periods than the prior variable.
Again, there is a negative relationship as lower blood lead levels are associated with higher levels of
fugitive air lead.  The distributions of the three time periods are nearly identical, as again the 90th
percentiles all equal 0.00.  TRI Compounds (Fugitive Air) is  significant in Models 3 and 4, but not
significant in the other models, although p-values from those three models range from 0.069 to 0.11.
TRI Compounds (Air Lead from Stacks) - Predicted blood lead levels decline slightly as TRI
Compounds (Air Lead from  Stacks) increases.  The 2000-2002 and 2003-2004 periods have slight
negative slopes between the  two variables while the 2005-2006 period appear to be  flat. The
distributions of the three time periods are nearly identical, as  again the 90th percentiles all equal 0.00.

                                               E-9
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agencv determination or policy.

-------
TRI Compounds (Air Lead from Stacks) is only significant for Model 4, but not in the other four
models.
TRI Compounds (Water Surface) - The relationship between predicted GM blood lead levels and TRI
Compounds (Water Surface) appears to be flat in all three time periods.  The distribution of TRI
Compounds (Water Surface) is identical across the three time periods, as again the 90th percentiles all
equal 0.00. TRI Compounds (Water Surface) is not significant in any of the five models.
TRI Lead Only (Total Air) - The relationship between predicted GM blood lead levels and TRI Lead
Only (Total Air) is positive as blood levels increase with  air lead levels. The distribution of TRI Lead
Only (Total Air) is identical across the three time periods, as again the 90th percentiles all equal 0.00.
TRI Lead Only (Total Air) is not significant in any of the five models.
TRI Lead Only (Fugitive Air) - The relationship between predicted GM blood lead levels and TRI Lead
Only (Fugitive Air) is positive as blood levels increase with air lead levels in all three time periods.  The
distribution of TRI Lead Only (Fugitive Air) is  nearly identical across the three time periods, as again
the 90lh percentiles all equal 0.00.  TRI Lead Only (Fugitive Air) is only significant at the .05 level for
Model 4, while the p-values from Models 1 to 3 range from .056 to .081.
TRI Lead Only (Air Lead from Stacks) - The relationship  between  predicted GM blood lead levels and
TRI Lead Only (Air Lead from Stacks) is negative over the 2000-2004 data as blood levels decrease as
air lead levels increase, while the relationship is flat in  the 2005-2006 data. The distribution of TRI
Lead Only (Air Lead from Stacks) is identical across the  three time periods, as again the 90lh percentiles
all equal 0.00. TRI Lead Only (Air Lead from Stacks) is not significant in any of the  five models.
TRI Lead Only (Water Surface) - The relationship between predicted GM blood lead levels and TRI
Lead Only (Water Surface) is positive across the three  time periods with the 2000-2002 period have a
slightly more positive slope than the other two period.  The distribution of TRI Lead Only (Water
Surface) is identical across the three time periods, as again  the 90th percentiles all equal 0.00.  TRI Lead
Only (Water Surface) is significant for Models  1 to 4, but not significant for Model 5.
TRI Total Lead (Total Air) - Generally, lower predicted blood lead levels are associated with higher
levels of Total Lead in total air. The 2000-2002 and 2003-2004 time periods have slight negative
slopes, while the 2005-2006 relationship is nearly flat.  The distributions of TRI Total Lead (Total Air)
are very similar across the three time periods, with the  90lh  percentiles of the distributions equal to 0.00.
TRI Total Lead (Total Air) is not significant in  any of the five models.
TRI Total Lead (Fugitive Air) - The relationship between  predicted GM blood lead levels  and TRI
Total Lead (Fugitive Air) is positive as blood levels increase with air lead levels in all three time
periods. The distribution of TRI Total Lead (Fugitive Air)  is nearly identical across the three time
periods, as again the 90lh percentiles all equal 0.00.  TRI Total Lead (Fugitive Air) is  not significant in
any of the five models.
TRI Total Lead (Air Lead from Stacks) - The  relationship between predicted GM blood lead levels and
TRI Total Lead (Air Lead from Stacks) is slightly negative over the 2000-2004 data as blood levels
decrease as air lead levels  increase, while the relationship appears to be flat in the 2005-2006 data. The
distribution of TRI Total Lead  (Air Lead from Stacks)  is identical across the three time periods, as again
the 90lh percentiles  all equal 0.00. TRI Total Lead (Air Lead from Stacks) is significant in Model 4, but
is not significant in any of the other four models.
                                              E-10
TTi is information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
TRI Total Lead (Water Surface) - The relationship between predicted GM blood lead levels and TRI
Total Lead (Water Surface) is slightly positive across the three time periods as predicted blood lead level
increases very slightly as lead in water increases. The distribution of TRI Total Lead (Water Surface) is
identical across the three time periods, as again the 90lh percentiles all equal 0.00. TRI Total Lead
(Water Surface) is not significant in any of the five models.

Across the 12 TRI variables, the vast majority of measurements for Census tracts in Massachusetts are
0.00, over 90 percent for each variable. Of the 60 variable/model combinations, the TRI variables were
only significant predictors of blood lead levels for 10 of those models.  Although the log-likelihood
ratios from each model are similar across the 12 variables for each model, TRI Lead Only (Water
Surface) provided the best model fit for all five models.

E.3    Analyses of Local Blood Lead Data by Programmatic Variables

Most of the explanatory variables being explored in this project are considered risk factors for childhood
lead poisoning. It is anticipated that the level and characteristics of programmatic support from either
federal, state, or local sponsors will contribute towards meaningful reductions in the prevalence of
childhood lead poisoning. In the high resolution models, the various characteristics of the programs
(information from housing inspections and case  management services) are being explored within the
statistical models.

E.3.1    Programmatic Funding Variables

For the  high-resolution model in the Commonwealth of Massachusetts, information on within-state
funding levels was obtained and analyzed.  Within-state funding data were available down to the
township level. That funding was then allocated to Census tracts in two ways. First, in the Not
Standardized version, total dollars were allocated to a Census tract proportionally based on population
and analyzed. These models are identified as $ per Census Tract in each variable name.  Second, in the
Standardized version, the dollars allocated  to each tract were divided by the number of children less than
six years old and the funds per child were analyzed.  These models are identified as $ per Child in each
variable name.

The HUD funding and  CDC funding  information was also explored with the Massachusetts data. Four
variables were generated from these data and analyzed - current and cumulative funding allocated to
each Census tract in Massachusetts to combat childhood lead poisoning, both Standardized by number
of children per tract and Not Standardized. The state, HUD, and CDC funding data also were combined
to create Total Funding variables, including both current and cumulative levels and both standardized
and not standardized versions. The detailed exploratory analysis results for the 16 state, HUD, and total
funding variables are presented  in Figures/Tables B47 to B62. Results for each variable are discussed
below.

Current HUD Funding ($per Child) - The relationship between predicted GM blood lead  levels and
Current HUD Funding  is not consistent across the three time periods.  In the earlier time periods (2000-
2002 and 2003-2004), the relationship appears to be nearly flat, with a slight negative slope in the first

                                              E-ll
7/7/5 information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
time period and a slight positive slope in the second time period. Meanwhile, the most current time
period has a more negative slope, with lower lead levels associated with higher funding amounts. The
distributions of Current HUD Funding also change across the time periods. The 2000-2002 period has
higher mean and median Current HUD Funding levels than the other two periods.  Current HUD
Funding (Standardized) is found to be a significant predictor of GM blood lead levels only in Model 3,
while it is not significant in the other four models.
Cumulative HUD Funding ($per Child) - The relationship between Cumulative HUD Funding levels
and predicted GM blood lead levels is flat in each of the three time periods. The distribution of
Cumulative HUD Funding levels increases over the three time periods as expected given the nature of
this variable. Cumulative HUD Funding (Standardized) is significant in Models 1 to 4, it was not
significant in Model 5.
Current State Funding ($per Child) - The relationship between Current State Funding levels and
predicted GM blood lead levels is consistent and positive across  the three time periods, with increases in
state funding associated with slightly higher blood lead levels. The distribution of Current State
Funding is identical across the three time periods, as all Census tracts are included in each time period.
Current State Funding (Standardized) is a significant predictor of children's blood lead levels in all
models.
Cumulative State Funding ($per Child) - The relationship between Cumulative State Funding levels
and predicted GM blood lead levels is consistent across the three time periods, with increases in state
funding associated with slightly higher blood lead levels, although there appears to be a slight flattening
in that relationship with each successive time period.  The distribution of Cumulative State Funding
increases over the three time periods as expected given the nature of this variable.  Cumulative State
Funding (Standardized) is a significant predictor of children's blood lead levels in all models.
Current CDC Funding ($ per Child) - The distribution of Current CDC Funding differs between the
2000-2002 time period and the other two periods. While the range of Current CDC Funding ($ per
Child) across Census tracts in Massachusetts is about $4.50  per child in the 2000-2002 period, the range
is only $0.05 and $0.08 in the other two periods.  From Figure B51, it appears the relationship between
Current CDC Funding levels and predicted GM blood lead levels is slightly positive for the 2000-2002
period; however, the relationship is unable to be determined for the other two periods. Current CDC
Funding ($ per Child) is only significant in Model 5, but is not significant in the other four models.
Cumulative CDC Funding ($per Child) - The relationship between Cumulative CDC Funding ($ per
Child) and predicted GM blood lead levels is slightly positive in the 2000-2002 time period, but appears
to be nearly flat in the subsequent two periods. The distribution  of Cumulative CDC Funding ($ per
Child) increases over the three  time periods as expected. Cumulative CDC Funding ($ per Child) is
significant in Models 3 to 5, but not in Models 1  and 2.
Current Total Funding ($per  Child) - The relationship between Current Total Funding levels and
predicted GM blood lead levels is  generally flat across the three  time periods, although there is a slight
positive slope in the middle time period. Mean and median  levels of Current Total Funding ($ per
Child) across Massachusetts Census tracts decrease from the 2000-2002 period to the later two periods.
Current Total Funding ($ per Child) is significant in Models 1, 2, and 5, but Models 3 and 4 failed to
converge.
Cumulative Total Funding ($per Child) - The relationship between Cumulative Total Funding ($ per
Child) levels and predicted GM blood lead levels is consistent across the three time periods, with
                                             E-12
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
increases in state funding associated with slightly higher blood lead levels, although there appears to be
a slight flattening in that relationship in the later two time periods. The distribution of Cumulative Total
Funding ($ per Child) increases over the three time periods as expected given the nature of this variable.
Cumulative Total Funding ($ per Child) is significant in Models 1,2, and 3, but Models 4 and 5 failed to
converge.
Current HUD Funding ($per Census Tract) - Lower predicted GM blood lead levels are associated
with higher levels of Current HUD Funding ($ per Census Tract) in the 2000-2002 and 2005-2006 time
periods, while the middle period displays a mostly flat relationship. The distributions of Current HUD
Funding ($ per Census Tract) also change across the time periods. The 2000-2002 time period has
higher mean and median Current HUD Funding levels than the other two periods.  Current HUD
Funding ($ per Census Tract) is significant in Model 5, but not significant in the other four models.
Cumulative HUD Funding ($per Census Tract) - Generally, slightly lower blood lead levels are
associated with higher levels of Cumulative HUD Funding ($ per Census Tract), however, the
relationship appear to be flattening with each successive time period.  As with all the cumulative
variables, the distribution of Cumulative HUD Funding ($ per Census Tract) levels increases over the
three time periods. Cumulative HUD Funding ($ per Census Tract) is significant in all models, except
Model 4 that failed to converge.
Current State Funding ($per Census Tract) - The relationship between Current State Funding ($ per
Census Tract) levels and predicted GM blood lead levels is mainly consistent across the three time
periods, with increases in state funding associated with slightly higher blood lead levels, although the
slope of this relationship increases with each time period (because blood lead levels are decreasing over
time while the Current State Funding distribution remains constant. The distribution of Current State
Funding ($ per Census Tract) is identical across the three time periods, as all Census tracts are included
in each time period. Current State Funding ($ per Census Tract) is significant in all models.
Cumulative State Funding ($per Census Tract) - The relationship between Cumulative State Funding
($ per Census Tract) levels and predicted GM blood lead levels is consistent across the three time
periods, with increases in state funding associated with slightly higher blood lead levels, although the
slope appears to be more positive with each successive time  period. The distribution of Cumulative
State Funding ($ per Census Tract) increases over the three time periods as expected given the nature of
this variable.  Cumulative State Funding ($ per Census Tract) is significant in all models.
Current CDC Funding ($per Census Tract)-The relationship between Current CDC Funding ($ per
Census Tract) levels and predicted GM blood lead levels is mainly consistent across the three time
periods, with increases in state funding associated with slightly higher blood lead levels, although the
slope of this relationship increases with each time period.  The distribution of Current CDC Funding ($
per Census Tract) is nearly identical across the three time periods. Current CDC Funding ($ per Census
Tract)  is significant in Models  1 and 2, not significant in Models 3 and 5, while Model 4 failed to
converge.
Cumulative CDC Funding ($per Census Tract) - The relationship between Cumulative CDC Funding
($ per Census Tract) levels and predicted GM blood lead levels is consistent across the three time
periods, with increases in state funding associated with slightly higher blood lead levels.  The
distribution of Cumulative CDC Funding ($ per Census Tract) increases over the three time periods as
expected given the nature of this variable. Cumulative CDC Funding ($ per Census Tract) is significant
in all Models 3 to 5, but is not significant in Models 1 and 2.

                                              E-13
7/1/5 information is distributed solely for the purpose ofpre-disseinination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent
any Agency determination or policy.

-------
Current Total Funding ($per Census Tract) - The relationship between Current Total Funding ($ per
Census Tract) levels and predicted GM blood lead levels is inconsistent across the three time periods.
During the 2000-2002 time period, predicted GM blood lead levels decline slightly as Current Total
Funding ($ per Census Tract) increases, while the other two time periods have slight positive slopes.
The distribution of Current Total Funding ($ per Census Tract) across Massachusetts Census tracts
generally decrease from the 2000-2002 period to the later two periods, although the median decreases
from $2,491 to $2,183 between the first time periods and returns to $2,389 in the most current period.
Current Total Funding ($ per Census Tract) is significant only in Model 3, and is not significant in the
other four models.
Cumulative Total Funding ($per Census Tract) - The relationship between Cumulative Total Funding
($ per Census Tract) levels and predicted GM blood lead levels is slightly different across the three time
periods, with a flat relationship in 2000-2002 but becoming slightly more positive in the successive time
periods. The distribution of Cumulative Total Funding ($ per Census Tract) increases over the three
time periods as expected given the nature of this variable. Cumulative Total Funding ($ per Census
Tract) is significant in all models.

Analyses of these programmatic funding variables provided surprising results for Massachusetts,
although potential explanations can be suggested.  For the State funding variables, higher amounts of
funding do not appear to be associated with lower blood lead levels.  A few of the HUD funding
variables did indicate some association between increased funds and lower blood lead levels, but for
some time periods there does not appear to be a relationship between them. The total funding analyses
show mainly flat to positive relationships, perhaps caused by the state funding data. When increased
funding is associated with higher lead levels, a potential explanation is that funds are being successfully
targeted at areas that have worse lead poisoning problems.

Table 4-2 reports that Current State Funding ($ per Child) provided the lowest log-likelihood statistics
for Models 1 to 3. Note that as levels of this variable increase, predicted GM blood lead levels increase
consistently across each time period (see Figure B49). Cumulative CDC Funding ($ per Census tract)
provided the best model fits for Models 4 and 5.

E.3.2   Housing Inspection Variables

As discussed in Section 3.4.3, Massachusetts supplied a data set of housing inspections conducted
within the state. The detailed exploratory analysis results for the 12 housing inspection variables are
presented in Figures/Tables B75 to B86. Results for each variable are discussed below.

PI: Proportion of Housing Units Passing Massachusetts Standard of Care, Naive Method 1 - The
relationship between PI variable levels and predicted GM blood lead levels is consistently positive
across the three time periods, as higher blood lead levels are associated with a higher percentage of units
passing the standard of care.  The distribution of the PI variable is similar across the three time periods,
although the mean and median are one percentage point less in the 2000-2002 time period.  PI is
significant in all models.
Fl: Proportion of Housing Units Failing Massachusetts Standard of Care, Naive Method 1 - Across
all three time periods, predicted GM blood lead levels increase as levels of the Fl  variable increase. The
                                              E-J4
This information is distributed solely for the purpose ofpre-dissernination peer review under applicable information quality
guidelines.  It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent
any Agency determination or policy.

-------
distribution of the Fl variable is similar across the three time periods, although the maximum proportion
of failing units decreases with each successive time period.  Fl is significant in all models.
Nl: Proportion of Housing Units Assessed, Naive Method 1 - The relationship between Nl variable
levels and predicted GM blood  lead levels is consistently positive across the three time periods, as
higher blood lead levels are associated with a higher percentage of units assessed. The distribution of
the Nl variable is nearly identical across the three time periods. Nl is significant in all models.
P2: Proportion of Housing Units Passing Massachusetts Standard of Care, Naive Method 2 - The
relationship between P2 variable levels and predicted GM blood lead levels is positive across the three
time periods, as higher blood lead levels are consistently associated with a higher percentage of units
passing the standard of care.  The distribution of the P2 variable is similar across the three time periods.
P2 is significant in all models.
F2: Proportion of Housing Units Failing Massachusetts Standard of Care, Naive Method 2 -  Across
all three time periods, predicted GM blood lead levels increase as levels of the F2 variable increase, with
the 2000-2002 appearing to have a slightly more positive slope than the other two periods.  The
distribution of the F2 variable is similar across the three time periods, although the maximum proportion
of failing units decreases with each successive time period.  F2 is significant in all models.
N2: Proportion of Housing Units Assessed, Naive Method 2 - The relationship between N2 variable
levels and predicted GM blood  lead levels is consistently positive across the three time periods, as
higher blood lead levels are associated with a higher percentage of  units assessed. The slopes  appear to
be a bit less positive with each successive time period. Mean and median levels of the N2 variable
increase across the three time periods as more units are assessed each year.  N2 is significant in all
models.
P3: Proportion of Housing Units Passing Massachusetts Standard of Care, Naive Method 3 - The
relationship between P3 variable levels and predicted GM blood lead levels is positive across the three
time periods, as higher blood lead levels are consistently associated with a higher percentage of units
passing the standard of care, although the slopes appear to be declining over time. The distribution of
the P3 variable is similar across the three time periods, although the mean and median increase by a
percentage point  in the 2005-2006 period. P3 is significant in all models.
F3: Proportion of Housing Units Failing Massachusetts Standard of Care, Naive Method 3 - Across
all three time periods, predicted GM blood lead levels increase as levels of the F3 variable increase, with
slopes appearing  to be consistent  across time.  The distribution of the F3 variable is similar across the
three time periods, although the maximum proportion of failing units decreases with each successive
time period. As with all the previous  housing inspection variables, F3 is significant in all models.
N3: Proportion of Housing Units Assessed, Naive Method 3 - The relationship  between N3 variable
levels and predicted GM blood  lead levels is positive across the three time periods,  as higher blood lead
levels are consistently associated  with a higher percentage of units  passing the standard of care, although
the slopes appear to be declining over time.  The distribution of the N3 variable increases across the
three time periods as more units are assessed each year. N3 is significant in all models.
P4: Proportion of Housing Units Passing Massachusetts Standard of Care,  MDPH Method - The
relationship between P4 variable levels and predicted GM blood lead levels is positive across the three
time periods, as higher blood lead levels are consistently associated with a higher percentage of units
passing the standard of care, although the slopes appear to be declining over time. Mean and median

                                              E-15
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy.

-------
levels of the P4 variable increase across the three time periods as higher percentages of units pass the
MA standard of care. P4 is significant in all models.
F4: Proportion of Housing Units Failing Massachusetts Standard of Care, MDPH Method - Across
all three time periods, predicted GM blood lead levels increase as levels of the F4 variable (% failing)
increase, with slopes appearing to be consistent across time. Mean, median, and maximum failing rates
decrease with each successive time period. F3 is also significant in all models.
N4: Proportion of Housing Units Assessed, MDPH Method - The relationship between N4 variable
levels and predicted GM blood lead levels is positive across the three time periods, as higher blood lead
levels are consistently associated with a higher percentage of units passing the standard of care, although
the slopes appear to be declining over time. The distribution of the N4 variable increases slightly across
the three time periods as the mean, median, and maximum go up. N4 is significant in all models, except
Model 4 that failed to converge.

In summary, positive relationships are evident between all the housing inspection variables and
predicted GM blood lead levels - passing rates, failure rates, and assessment rates. All variables were
significant for predicting blood lead levels across all models,  except the N4 variable for Model 4. As
reported in Table 4-2, F4 provides the best fit for Models 1 and 2, although the Fl and F3 variables were
very close behind. This reversed for Model 3 with Fl and F3 providing the best fit and F4 close behind.
The P2 and P3 variables provided the best fit for Models 4 and 5. Note that P2 and P3 yielded identical
fit statistics for all models, as did Fl and F3.
                                               E-16
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by EPA.  It does not represent and should not be construed to represent
any Agency determination or policy.

-------
                      Appendix F

U.S. Counties and Massachusetts Census Tracts with Highest
                    Predicted BLLs

-------
Table F-l. List of U.S. Counties with Highest Predicted Blood Lead Levels across Five Outcomes
CM (Model 2)
State
AZ
ME
ME
MI-
ME
ME
ME
ME
ME
ME
ME
ME
ME
ME
MI-
ME
NH
ME
IA
IA
IA
IA
NH
IA
IA
NH
CA
IA
IA
IA
IA
IA
IA
NH
IA
County
Maricopa
Somerset
Kennebcc
Cumberland
Pcnobscol
Sagaclahoc
Washington
York
Hancock
Oxford
Knox
Arooslook
Waldo
Franklin
Andioscoggm
Lincoln
Coos
Piscaiaquis
Dubuque
Scon
Black Hawk
Delaware
Cheshire
Linn
Buchanan
Belknap
San Joaqum
Benton
Ida
Chickasaw
Brenier
O'Bnen
Crawford
Graft on
Wright
GM
6789
6111
5984
5.877
5.851
5.848
5.733
5.716
5.695
5 643
5642
5605
5601
5.558
5518
5.504
5.432
5.369
5273
5 150
5032
5022
5016
5012
4986
4.930
4.891
4.884
4.826
4805
4.785
4773
4735
4699
4600
PS
State
CA
AZ
IN
IL
LA
LA
AZ
CA
CA
MO
IL
CA
NH
LA
LA
LA
LA
LA
LA
OH
LA
IL
LA
MO
LA
LA
LA
LA
AL
LA
LA
LA
IL
LA
PA
County
San Joaqum
Maricopa
Vanderburgh
Rock Island
Morchouse Parish
De Soto Parish
Cochise
Stanislaus
Solano
Worth
Mercer
Sonoma
Coos
Rapidcs Parish
West Carroll Parish
Concordia Parish
East Carroll Parish
Webster Parish
East l-elicianu Parish
Lucas
Tensas Parish
Henry
Richland Parish
Carroll
Vemon Parish
Franklin Parish
La Salle Parish
East Baton Rouge Parish
Dallas
Ibena Parish
Caklwell Parish
Iberville Parish
Peon a
Red River Parish
Philadelphia
PS
0627
0.607
0.576
0421
0.411
0.404
0403
0380
0.332
0331
0328
0322
0318
0.315
0313
0310
0305
0302
0.301
0.298
0.294
0.294
0294
0289
0282
0281
0277
0275
0275
0274
0.271
0271
0266
0.261
0261
PJO
State
CA
CA
CA
CA
CA
CA
CA
AL
CA
CA
CA
CA
LA
AL
IA
IA
VT
CA
IL
CA
CO
MO
NH
IA
NV
IA
CA
LA
OH
VT
IL
PA
IA
IL
IA
County
San Joaqum
Stanislaus
Solano
Madera
Sonoma
Humboldt
Monterey
Dallas
Sania Clara
Sacramento
Imperial
Napa
Orleans Parish
Pike
Wayne
Ida
Rutland
Comra Costa
Cook
Colusa
Denver
St Louis city
Coos
Union
White Pine
O'Bncn
Glenn
West Carroll Parish
Cuyahoga
Washington
Greene
Northumberland
Decatur
Kankakee
Cherokee
P10
0440
0.247
0234
0171
0167
0 151
0 142
0 140
0123
0 112
0104
0079
0073
0.072
0072
0072
0072
0071
0.071
0.070
0.069
0.067
0066
0066
0066
0063
0062
0062
0.062
0061
0.061
0061
0060
0059
0058
PIS
State
CA
CA
CA
AL
CA
CA
CA
CA
IA
AL
LA
CA
VT
IL
VT
PA
CA
CO
MO
OH
VT
IA
PA
IA
CA
CA
IL
IL
PA
PA
KS
PA
IA
MO
KS
County
Stanislaus
Santa Clara
San Joaquin
Dallas
Madera
Sonoma
Monterey
Solano
Union
Pike
Orleans Pansh
Humboldt
Rutland
Vermilion
Franklin
Lancaster
Contra Costa
Denver
St Louis city
Cuyahoga
Washington
O'Brien
Northumberland
Woodbury
San Mateo
Sacramento
Greene
Cook
Berks
York
Neosho
Huntingdon
Lucas
Clark
Edwards
PIS
0063
0062
0049
0045
0044
0.043
0042
0036
0033
0.028
0028
0028
0027
0.026
0025
0.025
0.025
0.024
0.024
0.024
0.024
0.024
0023
0023
0.023
0.023
0.023
0.023
0023
0022
0022
0022
0021
0021
0021
P25
State
CA
CA
IA
IL
VT
CA
CO
CA
PA
PA
CA
VT
IA
VT
VT
VT
IL
CA
PA
PA
MO
IA
PA
IL
IA
KS
PA
CA
CA
PA
IA
CT
PA
IA
VT
County
San Mateo
Santa Clara
O'Bnen
Vermilion
Franklin
Sonoma
Denver
Monterey
Delaware
Chester
Stanislaus
Washington
Woodbury
Rutland
Orleans
Chittenden
Cook
Humboldl
Indiana
Northumberland
St. Louis city
Marshall
Dauphin
Union
Taylor
Alchison
Lycoming
Tehama
Contra Costa
Lancaster
Webster
New Haven
Berks
Des Moines
Orange
P2S
0.01 1
0.010
0010
0009
0009
0.008
0.008
0007
0007
0.007
0.007
0.006
0006
0.006
0.006
0.006
0.006
0.006
0.006
0.006
0.006
0.006
0.006
0.006
0.006
0.005
0.005
0.005
0.005
0.005
0.005
0.005
0.005
0.005
0.005
                                                                          F-l
This information is distributed solely for the purpose of pre-dissemmation peer review under applicable information quality guidelines. It has not been formally disseminated by
EPA. It does not represent and should not be construed to represent any Agency determination or policy.

-------
GM (Model 2)
State
NH
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
LA
IA
NH
IA
IA
IA
NH
IA
IA
IA
KS
IA
WY
IA
IA
IA
IA
IA
IA
County
Mem muck
Franklin
Marshall
Decalui
Cherokee
Jones
Grundy
Hardin
Fayeite
Butler
Carroll
Jackson
Palo Alio
Clayton
Hamilton
Page
Cerro Gordo
Taylor
Morehousc Parish
Clay
Strafford
Floyd
Tama
Lyon
Carroll
Hancock
Mitchell
Greene
Cowley
Monona
Albany
Buena Vista
Emmet
Lucas
Audubon
Worth
Sac
GM
4595
4.577
4.562
4.561
4.520
4.508
4.487
4.484
4483
4453
4437
4367
4361
4349
4 345
4 343
4.317
4268
4.254
4.233
4.233
4.230
4.228
4.220
4.197
4.188
4.181
4.174
4.081
4.081
4066
4059
4056
4.037
4033
3996
3993
PS
State
LA
Ml
CA
LA
KS
LA
KS
VT
LA
MO
MO
CA
LA
CO
LA
PA
KY
NH
LA
IA
VT
OH
IA
VT
NE
LA
AZ
NM
LA
TX
LA
IA
OK
KS
LA
CA
NM
County
Winn Parish
Lenawce
Imperial
Orleans Parish
Chaulauqua
St. Mary Parish
Edwards
Rutland
Natchitoches Parish
Atchison
St Louis city
Madcra
Union Parish
San Juan
Ouachita Parish
Northumberland
Fulion
Sullivan
Beauregard Pansh
Scott
Windham
Seneca
Ida
Caledonia
Fillmore
Livingston Pansh
Santa Cruz
Harding
Grant Pansh
Col lings worth
St. Helena Parish
Page
Jefferson
Harper
Bienville Pansh
Monterey
San Juan
PS
0.259
0.258
0.256
0254
0248
0247
0.247
0.247
0244
0.241
0239
0239
0238
0.236
0.236
0.234
0234
0234
0232
0231
0230
0229
0229
0228
0228
0.227
0.226
0.225
0.224
0.224
0224
0223
0.223
0.222
0.221
0221
0.220
P10
State
AL
IL
CA
IL
PA
IA
MO
PA
CA
PA
IA
IA
LA
LA
VT
AL
IL
IA
PA
IL
CA
PA
AZ
IL
NE
IA
PA
IL
CT
VT
CA
VA
IA
MO
VT
IL
KS
County
Macon
Peona
Tehama
Edgar
Blair
Monona
Worth
Lancaster
Sierra
York
Woodbury
Des Moines
Tensas Parish
De Soto Parish
Windham
Coffee
Fulton
Taylor
Berks
Alexander
Tulare
Schuylkill
Cochise
Stephenson
Fill more
Rmggold
Lycoming
Vermilion
New Haven
Franklin
Yolo
Winchester city
Lucas
Caldwell
Bennington
Rock Island
Atchison
P10
0.058
0058
0058
0.057
0.057
0.056
0.056
0.056
0.055
0.055
0.054
0.054
0053
0.053
0053
0.053
0052
0.052
0052
0052
0051
0051
0050
0050
0050
0050
0049
0049
0049
0049
0.048
0048
0048
0.048
0.048
0.048
0048
PIS
State
IL
VT
IL
IA
IA
PA
IA
PA
TX
VT
PA
PA
CT
IL
PA
MO
NE
KY
VT
VT
IL
PA
TX
PA
CA
LA
IL
PA
IN
VA
IL
NE
IL
NH
KS
CA
CA
County
Carroll
Orange
Peoria
Taylor
Cass
Indiana
Des Moines
Dauphin
Young
Bennington
Lycoming
Schuylkill
New Haven
Alexander
Delaware
Buchanan
Jefferson
Harrison
Windham
Caledonia
Rock Island
Venango
Bosque
Chester
Kings
De Soto Pansh
Knox
Blair
Rush
Petersburg city
Slephenson
Otoe
Wabash
Slrafford
Alchison
Tehama
Imperial
PIS
0.020
0.020
0.020
0020
0020
0020
0.019
0.019
0019
0.019
0.019
0.019
0.018
0.018
0.018
0.018
0.017
0.017
0.017
0.017
0.017
0.017
0.017
0.017
0.017
0.017
0.017
0.017
0.017
0.017
0.017
0.016
0.016
0.016
0.016
0.016
0.016
P25
State
PA
IL
MO
CA
PA
OH
MO
IL
PA
NH
MO
IA
KS
ME
NE
ME
IL
IL
CA
VT
NE
NH
IA
PA
LA
IA
IN
IL
CT
CT
KS
IA
IN
IA
IL
CT
IA
County
York
Cass
Caldwell
Madera
Schuylkill
Cuyahoga
Buchanan
Henry
Perry
Belknap
Audrain
Benton
Coffey
Somerset
Fillmore
Sagadahoc
Peona
Knox
Solano
Bennington
Hall
Slrafford
Van Buren
Lehigh
Orleans Parish
Wayne
Rush
Adams
Hartford
Litchfield
Edwards
Buena Vista
Elkhurt
Hamilton
Morgan
New London
Cherokee
P2S
0.005
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0004
0004
0.004
0.004
0.004
0004
0.004
0.004
0.004
0.004
0004
0004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0003
0.003
0.003
0.003
0003
                                                                               F-2
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality guidelines.  It has not been formally disseminated by
EPA.  It does not represent and should not be construed to represent any Agency determination or policy.

-------
CM (Model 2)
State
MO
WY
IA
NY
KS
KS
MO
IA
MO
NH
IA
KS
IA
IA
IA
KS
IA
KS
IA
IA
KS
NE
KS
KS
KS
LA
LA
KS
LA
KS
NE
LA
IA
CA
KS
NE
IA
County
Wonh
Natrona
Alkimakee
Chauiauquu
Crawford
Prall
Livingston
Montgomery
Iron
Sullivan
Humboklt
Lincoln
Union
Webster
Guihne
Barber
Clarke
Jewell
Calhoun
Wayne
Maishull
Pawnee
Elk
Comanche
Neosho
Richland Parish
De Soto Parish
Clay
Orleans Parish
Edwards
Nemaha
Tangipahoa Parish
Kmggold
Sonoma
Wood son
Dakota
Adams
CM
3.964
3.946
3.943
3938
3927
3.917
3916
3913
3.905
3.892
3886
3 885
3878
3.877
3.851
3.847
3.842
3.811
3.784
3779
3766
3765
3.764
3761
3759
3.755
3.749
3.745
3744
3741
3741
3738
3738
3738
3725
3724
3.721
P5
State
LA
CA
PA
NE
KS
CA
OK
KS
IL
A2
MO
MO
PA
IL
TX
KS
MO
LA
IA
MO
OK
IA
NE
OK
PA
AL
OK
LA
KS
PA
NE
MO
LA
IL
LA
LA
MO
County
Caddo Parish
Colusa
Schuylkill
Otoe
Coffey
Humboklt
Harmon
Barber
Alexander
Apache
DeKalb
Gentry
Blair
Edgar
Shackelfbrd
Clay
Shelby
Washington Parish
Delaware
Iron
Choclaw
Union
Boone
Tillman
Delaware
Macon
Alfalfa
Jackson Parish
Greenwood
Venango
Franklin
Vernon
Evangelme Parish
Fulton
Catahoula Parish
Bossier Pansh
Scotland
PS
0220
0220
0219
0217
0217
0216
0216
0216
0215
0.215
0.215
0215
0215
0.215
0.215
0.214
0214
0214
0213
0.212
0.211
0.211
0.208
0207
0207
0.206
0.205
0.205
0.204
0.204
0.203
0.203
0.203
0.203
0203
0203
0202
P10
State
LA
PA
VT
CA
NH
VT
PA
IL
PA
KY
NH
PA
KS
KS
KY
KS
LA
MO
IA
PA
MO
PA
NV
MO
KS
IA
VT
NH
VA
NH
CA
NV
IA
IL
AZ
IN
IA
County
Morehousc Pansh
Philadelphia
Caledonia
Kings
Straff ord
Orange
Venango
Adams
Indiana
Fulton
Sullivan
Huntingdon
Crawford
Neosho
Hamson
Edwards
Richland Parish
Iron
Page
Delaware
Clark
Dauphin
Mineral
Audram
Brown
Cass
Addison
Mem mack
Clifton Forge city
Bel knap
Inyo
Lincoln
Marshall
Wabash
Greenlee
Rush
Clarke
P10
0047
0.047
0.047
0.047
0047
0047
0.047
0046
0046
0046
0.046
0046
0045
0.045
0.045
0.045
0.045
0.045
0.045
0.045
0.045
0.044
0.044
0044
0044
0044
0.044
0.044
0044
0044
0044
0043
0043
0.043
0043
0.043
0.043
PIS
State
IA
IA
HI
IA
KS
PA
PA
KS
NH
KS
IA
KS
CA
CA
KS
MO
IL
VT
NV
NE
CA
IA
IA
PA
MO
1A
WI
ME
KS
VT
MO
IL
NE
KS
NE
IA
MO
County
Crawford
Cherokee
Kalawuo
Franklin
Ottawa
Lehigh
Bradford
Labelte
Belknap
Montgomery
Hardm
Kiowa
Yolo
Colusa
Coffey
Audram
Mercer
Orleans
White Pine
York
Glenn
Marshall
Lee
Fayelte
Wonh
Page
Milwaukee
Somerset
Crawford
Addison
Caldwell
Lee
Bun
Comanche
Richardson
Iowa
Clinton
PJS
0016
0.015
0.015
0.015
0.015
0015
0.015
0015
0015
0.015
0.015
0015
0014
0014
0.014
0014
0014
0014
0014
0014
0.014
0.014
0.014
0014
0014
0.014
0.014
0.014
0.014
0.014
0.014
0.014
0.014
0.014
0.014
0.013
0.013
P25
State
1A
IL
IA
IA
IA
IA
PA
ME
PA
MO
IA
NE
KY
KS
VA
MO
KS
IA
KS
NE
KS
IL
PA
IA
PA
NE
IA
IL
CA
OK
WI
IN
CO
IA
OK
NE
MN
County
Lucas
Rock Island
Louisa
Henry
Brenier
Tama
Blair
York
Washington
Penis
Hardm
Jefferson
Campbell
Clay
Clifton Forge city
Clark
Cowley
Black Hawk
Ottawa
Pawnee
Labette
Slephenson
Montgomery
Boone
Lackawanna
Richardson
Lee
Warren
Sacramento
Adair
Sheboygan
Fountain
Lake
Mills
Hughes
Franklin
Cotton wood
P25
0.003
0.003
0.003
0.003
0.003
. 0.003
0.003
0.003
0.003
0.003
0.003
0.003
0003
0003
0003
0.003
0003
0003
0.003
0003
0003
0003
0.003
0003
0003
0.003
0.003
0.003
0003
0003
0003
0003
0.003
0.003
0003
0003
0.003
                                                                               F-3
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality guidelines.  It has not been formally disseminated by
EPA.  It does not represent and should not be construed to represent any Agency determination or policy.

-------
GM (Model 2)
State
KS
MO
LA
IA
IA
KS
NE
IA
LA
MO
NY
CA
NE
MO
KS
KS
KS
NE
IA
KS
IA
KS
KS
IA
PA
IA
LA
KS
CA
KS
LA
IA
KS
KS
KS
IL
IA
County
Mitchell
Vemon
Washington Parish
Climon
Dallas
Smith
Richardson
Howard
West Carroll Parish
Carroll
Montgomery
Stanislaus
Franklin
Si Clair
Coffey
Greenwood
Dickinson
Arthur
Shelby
Harper
Powcshick
Republic
Chautauqua
Pocahontas
Philadelphia
Winnebago
Rapidcs Parish
Morns
Madera
Cheyenne
Ea.sl Carroll Parish
Madison
Osborne
Stafford
Alch^on
Peona
Plymouth
GM
3721
3720
3718
3718
3713
3.712
3709
3701
3700
3696
3 695
3 695
3692
3691
3690
3686
3682
3.676
3670
3.668
3.666
3.654
3.648
3.642
3638
3636
3.633
3632
3627
3624
3.610
3.609
3.603
3.588
3 588
3.587
3.573
P5
State
LA
VA
KS
TX
CT
KS
IA
OH
OK
VT
MO
NH
NM
VA
MO
NH
MO
NV
NE
OH
OK
LA
NM
TX
VT
PA
VT
KS
KS
•IA
CA
KS
VT
IL
IA
CA
LA
County
Sabinc Parish
Northampton
Rooks
Falls
New Haven
Woodson
Montgomery
Noble
Hughes
Benmnglon
Reynolds
Belknap
Mora
Middlesex
Harrison
Grafton
Bates
White Pine
Richardson
Cuyahoga
Kiowa
West Feliciana Parish
Eddy
Terrell
Addison
Dauphin
Orange
Ottawa
Rice
Fremont
Sacramento
Kiowa
Franklin
Greene
Decalur
Napa
Lafayette Parish
PS
0.202
0.202
0.201
0.201
0.201
0.200
0.200
0.200
0.199
0.199
0.199
0.198
0.198
0.198
0.198
0.197
0.197
0.197
0.197
0196
0 195
0194
0.194
0193
0193
0193
0192
0192
0192
0191
0191
0 191
0191
0189
0189
0189
0188
RIO
Stale
PA
CA
IL
IA
PA
KY
MO
IA
IL
LA
IL
MO
Wl
NE
MO
KY
CA
KS
VA
IA
KS
IA
CA
VA
AL
VT
AL
CA
ME
CA
KS
IL
DC
IA
KS
IA
CT
County
Carbon
Alpine
Lee
Hardm
Bradford
Campbell
Buchanan
Crawford
Hancock
Franklin Parish
Cass
Si Francois
Milwaukee
Jefferson
Carroll
Warren
Modoc
Coffey
Danville city
Lee
Ottawa
Jackson
San Mateo
Petersburg city
Tallapoosa
Orleans
Limestone
Lassen
Somerset
Shasta
Comanche
Menard
District of Columbia
Franklin
Wilson
Delaware
Windham
TLO
0043
0.043
0042
0042
0042
0042
0.042
0042
0042
0042
0042
0041
0041
0041
0041
0041
0041
0041
0041
0041
0040
0040
0040
0.040
0.040
0.040
0.039
0.039
0.039
0.039
0039
0039
0039
0039
0038
0038
0038
P15
State'
NM
IA
CT
IA
IA
IA
NH
KY
KS
LA
IL
PA
IL
VA
LA
KS
IA
MS
NH
LA
DC
PA
ME
NE
MO
VA
CA
IA
ME
VT
IA
OK
KS
KS
NH
IL
IA
County
Harding
Mitchell
Litchfield
Clinton
Muscatine
Monona
Coos
Kenton
Barton
Franklin Parish
Adams
Lackawanna
Morgan
Clifton Forge city
Caldwell Parish
Wilson
Sac
Lauderdale
Merrimack
Morehouse Parish
District of Columbia
Carbon
Sagadahoc
Dixon
Lewis
Danville city
Sierra
Washington
Franklin
Chiltenden
Benton
Alfalfa
Cowley
Jackson
Sullivan
Edgar
Wayne
PIS
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0012
0012
0012
0.012
0.012
0012
0.012
0.012
P2S
State
IA
KS
NE
AL
MO
VT
NE
PA
TN
TN
IL
KS
NE
IL
Wl
CT
MN
MO
KS
VA
IN
NE
CA
VT
PA
CO
TX
NE
NE
OH
MO
KS
MS
Wl
IN
IA
SD
County
Union
Bourbon
Greeley
Pike
Linn
Windsor
Kearney
Luzeme
Monroe
Coffee
Fulton
Jewell
York
Ogle
Milwaukee
Windham
Nobles
Carroll
Elk
Covington city
Vigo
Webster
Siskiyou
Addison
Fayette
San Juan
Nacogdoches
Furnas
Burl
Hamilton
Atchison
Lyon
Lauderdale
Ozaukee
Allen
Ida
Aurora
P2S
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0003
0.003
0.003
0.003
0003
0003
0.003
0.003
0.003
0003
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0003
0003
0003
0003
0003
0.003
0003
0003
                                                                               F-4
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality guidelines.  It has not been formally disseminated by
EPA.  It does not represent and should not be construed to represent any Agency determination or policy.

-------
GM (Model 2V
State
IA
MO
MO
KS
County
Dickinson
Gentry
Scotland
Anderson
GM
3.568
3.560
3.559
3.559
PS
State
NE
OH
LA
LA
County
Thursion
Morgan
Si Tammany Parish
Lincoln Parish
PS
0188
0188
0188
0187
PIO
State
IL
TX
MO
VA
County
McDonough
Bosque
Scotland
Northampton
PLO
0038
0038
0038
0.038
P15
State
OH
TX
MS
CT
County
Lucas
Falls
Smith
Windham
IMS
0012
0012
0012
0012
P25
State
NE
SD
VT
IA
County
Otoe
Hyde
Caledonia
Clinton
P2S
0.003
0.003
0.003
0.003
                                                                                F-5
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality guidelines.  It has not been formally disseminated by
EPA.  It does not represent and should not be construed to represent any Agency determination or policy.

-------
Table F-2. List of Top 10 Counties in each State with Highest Predicted Blood Lead Levels across Five Outcomes

State
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AK
AK
AK
AK
AK
AK
AK
AK
AK
AK
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AR
AR
AR
AR
GM
County*
Dallas
Macon
Lumar
Butler
Tallupoosa
Lowndes
Coffee
Sumtcr
Chambers
El more
Hames Bor
Wade Humpion CA
Yukon-Koyukuk CA
Skagway-Hoonah-
Angoon CA
Aleutians East Bor
Bnsiol Buy Bor.
Lake and Peninsula Bor.
Kenai Peninsula Bor.
Southeast Fairbanks CA
Northwest Arctic Bor
Mancopa
Cochise
Santa Cruz
Apache
Greenlce
Graham
Navajo
Lit Paz
Yavapai
Coconino
Lee
Phillips
Jefferson
Bradley

GM
3.52
3.28
2.97
2.97
2.87
286
2.76
273
2.66
2.66
1.42
1.38
1.38
1.35
1.29
.28
24
23
19
18
6.79
353
291
2.89
2.80
2.62
2.48
2.42
2.40
2.37
3.09
3.08
301
2.98

State
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AK
AK
AK
AK
AK
AK
AK
AK
AK
AK
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AR
AR
AR
AR
1>5
County
Dallas
Macon
Laniar
Pike
Butler
Chambers
Coffee
Lnudcrdale
Gcnevii
Clarke
Skagway-Hoonah-Angoon CA
Nome Census Area
Wrangell-Petersburg CA
Yukon-Koyukuk CA
Yakutat City and Bor
Lake and Peninsula Bor.
Ketchikan Gateway Bor
Wade Hampton ca
Northwest Arctic Bor
Bnsiol Bay Bor.
Mancopa
Cochise
Santa Cruz
Apache
Greenlce
Puna
Nava|O
Coconino
Graham
Yuma
Lee
Phillips
Bradley
Jefferson

P5
0.275
0.206
0.184
0.177
0.165
0.160
0.151
0 149
0149
0 141
0.044
0039
0039
0038
0.038
0.036
0.034
0.034
0033
0032
0607
0403
0.226
0215
0119
0100
0091
0089
0085
0084
0 142
0 140
0131
0127

State
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AK
AK
AK
AK
AK
AK
AK
AK
AK
AK
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AR
AR
AR
AR
P10
County
Dallas
Pike
Macon
Coffee
Tallapoosa
Limestone
Lowndes
Butler
Hale
St. Glair
Northwest Arctic Bor.
Nome Census Area
Skagway-Hoonah-Angoon
CA
Ketchikan Gateway Bor
Wrangell-Petersburg CA
Wade Hampton CA
Dillmgham CA
Yukon-Koyukuk CA
Yakutat City and Bor
Lake and Peninsula Bor.
Cochise
Greenlee
Mancopa
Apache
Graham
Navajo
La Paz
Yuma
Mohave
Yavapai
Phillips
Lee
Arkansas
Bradley

P10
0.140
0.072
0.058
0053
0.040
0.039
0.034
0.028
0.028
0.026
0.006
0.006
0.006
0.006
0006
0.005
0.005
0.005
0.005
0.005
0050
0043
0035
0.030
0028
0.025
0.020
0.019
0.017
0017
0.024
0021
0018
0018

-State
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AK
AK
AK
AK
AK
AK
AK
AK
AK
AK
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AR
AR
AR
AR
P15
County :,:f ,-t.
Dallas
Pike
Coffee
Macon
Etowah
Limestone
Tallapoosa
B arbour
St. Clair
Crenshaw
Ketchikan Gateway Bor.
Wrangell-Peiersburg CA
Skagway-Hoonah-
Angoon CA
Juneau City and Bor.
Sitka City and Bor.
Nome C A
Haines Bor
Northwest Arctic Bor.
Aleutians East Bor.
North Slope Bor.
Mancopa
Greenlee
Graham
Cochise
Yuma
Navajo
Mohave
Apache
Gila
La Paz
Phillips
Lee
Bradley
Arkansas

,P1S
0.045
0.028
0.011
0.010
0.009
0.008
0.008
0.008
0.007
0.007
0.003
0.002
0.002
0.002
0.002
0002
0002
0.002
0.002
0.002
0010
0009
0.007
0.007
0.006
0.005
0005
0005
0005
0005
0007
0007
0.006
0006

State
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AK
AK
AK
AK
AK
AK
AK
AK
AK
AK
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AR
AR
AR
AR
P2S
County
Pike
Coffee
Dallas
Si Clair
Lowndes
Shelby
Escambia
Jefferson
Macon
Crenshaw
Wrangell-Peiersburg CA
Ketchikan Gateway Bor
Juneau City and Bor.
Nome CA
Wade Hampton CA
North Slope Bor
Northwest Arctic Bor.
Sitka City and Bor.
Bethel CA
Anchorage Municipality
Santa Cruz
Mohave
Graham
Mancopa
Cochise
Greenlee
Yavapai
Apache
Yuma
Gila
Phillips
Lee
St. Francis
Desha

P2S
0.003
0.002
0.002
0.002
0.002
0.002
0.002
0.002
0.002
0001
0001
0001
0001
0.001
0001
0001
0001
0.001
0.001
0001
0.002
0.002
0.002
0002
0002
0002
0002
0001
0001
0001
0002
0001
0001
0.001
This information
EPA. It does not
                                                          F-6
is distributed solely for the purpose of pre-dissemination peer review under applicable information quality guidelines. It has not been formally disseminated by
represent and should not be construed to represent any Agency determination or policy.

-------
CM
Stale
AR
AR
AR
AR
AR
AR
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CT
CT
CT
CT
CT
CT
CT
CT
DE
DE
DE
Counly*
Monroe
Chicol
Woodruff
Scare v
Praine
Arkansas
San Joaqum
Sonoma
Stanislaus
Maclcra
Monterey
Solano
Sierra
Huinboldi
Imperial
Colusa
Sun Juan
Scdgwick
Las Am mas
Crowlev
Phillips
Denver
Otero
Washington
Lake
Beni
Wmdham
New Haven
Han ford
New London
Tolland
Fairfield
Litchfield
Middlesex
Keni
Sussex
New Castle
CM
2.95
2.92
291
2.91
2.90
2.89
4.89
3.74
369
363
332
308
3.02
297
295
2.79
3.25
2.85
2.84
2.76
276
273
269
2.66
2.61
258
3.04
2.90
2.70
261
249
229
228
222
307
304
2.49
P5
Slate
AR
AR
AR
AR
AR
AR
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CT
CT
CT
CT
CT
CT
CT
CT
DE
DE
DE
Counly
Monroe
Arkansas
Lafayette
Chicol
Union
Ouachita
San Joaqum
Stanislaus
Solano
Sonoma
Imperial
Muderu
Monterey
Colusa
Humboldl
Sacramento
Sun Juan
Denver
Sedgwick
Las Animas
Lake
Crowley
Huerfano
Bent
Otero
Washington
New Haven
Wmdham
Hartford
New London
Tolland
Litchfield
Fairfield
Middlesex
Sussex
New Castle
Kent
PS
0 124
0123
0121
0 121
0.119
0119
0627
0.380
0332
0322
0256
0.239
0.221
0.220
0216
0.191
0.236
0.157
0129
0.129
0127
0 127
0121
0.120
0.117
0.104
0201
0.183
0.154
0 140
0 130
0113
0108
0094
0084
0074
0072
P10
State
AR
AR
AR
AR
AR
AR
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CT
CT
cr
CT
CT
CT
CT
CT
DE
DE
DE
County
St. Francis
Monroe
Union
Ouachiia
Woodruff
Chicot
Sun Jouquin
Stanislaus
Solano
Madeni
Sonoma
Humboldl
Monterey
Santa Clara
Sacramento
Imperial
Denver
Lake
Las Animus
San Juun
Pueblo
Sedgwick
Bent
Olero
Crowley
Phillips
New Huven
Wmdham
New London
Hunford
Litchfield
Fairfield
Tolland
Middlesex
Sussex
New Castle
Kent
PJO
0018
0017
0017
0017
0.017
0017
0440
0.247
0234
0171
0167
0151
0.142
0.123
0.112
0.104
0.069
0.020
0.020
0.020
0.016
0014
0013
0013
0012
0012
0049
0038
0031
0030
0026
0022
0.022
0017
0014
0012
0011
PIS
Stale
AR
AR
AR
AR
AR
AR
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CT
CT
CT
CT
CT
CT
CT
CT
DE
DE
DE
County
Union
Ouachiia
Si. Francis
Chicot
Monroe
Desha
Stanislaus
Sunlu Clara
Sun Jouquin
Madera
Sonoma
Monterey
Solano
Humboldt
Contra Costa
San Maleo
Denver
Lake
Las Animus
San Juan
Sedgwick
Crowley
Pueblo
Bent
Otero
Phillips
New Haven
Litchfield
Windhum
Hartford
Tolland
New London
Fairfield
Middlesex
Sussex
New Castle
Kent
PJS
0.006
0.006
0006
0006
0.006
0.006
0063
0.062
0049
0044
0043
0.042
0.036
0028
0025
0.023
0.024
0008
0006
0.006
0005
0005
0005
0.005
0.004
0.004
0.018
0.013
0012
0011
0011
0010
0.008
0006
0003
0.003
0.003
P2S
State
AR
AR
AR
AR
AR
AR
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CT
CT
CT
CT
CT
CT
CT
CT
DE
DE
DE
County
Chicol
Bradley
Woodruff
Monroe
Cnttenden
Mississippi
San Maleo
Santa Clara
Sonoma
Monterey
Stanislaus
Humboldl
Tehama
Contra Costu
Madera
Solano
Denver
Luke
Sun Juan
Otero
Las Animas
Sedgwick
Crowley
Phillips
Beni
Washington
New Haven
Hunford
Lnchfield
New London
Windhum
Fairfield
Tolland
Middlesex
Sussex
Kent
New Castle
P2S
0001
0001
0.001
0.001
0.001
0001
0.011
0.010
0.008
0007
0007
0006
0.005
0.005
0.004
0.004
0.008
0.003
0.003
0.002
0.002
0.002
0.002
0002
0.002
0.002
0.005
0.004
0.004
0003
0.003
0.002
0.002
0.002
0.001
0.001
0.001
                                                                                F-7
77715 information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality guidelines.  It has not been formally disseminated by
EPA. It does not represent and should not be construed to represent any Agency determination or policy.

-------
CM
Stale
DC
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
HI
HI
HI
HI
HI
ID
ID
ID
ID
ID
ID
ID
ID
ID
ID
IL
County*
Dismcl of Columbia
DeSolo
Harclec
Gadsdcn
Sumlcr
Osceola
Piisco
Duval
Jackson
Highlands
Hcmando
Calhoun
Wonh
Slewail
Terrell
Gordon
Taluiferro
Eiirly
Clay
Warren
Ben Hill
Kalawao
Honolulu
Kauai
Hawaii
Maui
Bear Uike
Shoshonc
Oncida
Lewis
Benewah
Camas
Franklin
Lincoln
Clark
Butie
Peoria
CM
2.55
3.04
296
294
292
287
287
285
281
2.79
2.75
3.04
303
3.02
3.00
2.95
2.94
2.94
2.93
2.93
2.93
2.54
1.17
1.03
0.98
0.92
2.14
2.09
2.02
1.99
1.90
I 87
1.85
1.85
1.83
1.77
3.59
PS
Stale
DC
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
HI
HI
HI
HI
HI
ID
ID
ID
ID
ID
ID
ID
ID
ID
ID
IL
County
District of Columbia
Sumtcr
Glades
Baker
Highlands
Lafaycltc
Pasco
Holmes
Li ben y
Calhoun
Jackson
Wonh
Terrell
Evans
Stewart
Ben Hill
Calhoun
Early
Wheeler
Taylor
Oglethorpe
Kalawao
Hawaii
Kauai
Maui
Honolulu
Shoshone
Bear Lake
Lewis
Oneida
Franklin
Washington
Lincoln
Goodmg
Twin Falls
Camas
Rock Island
PS
0.173
0124
0 120
0107
0 100
0097
0085
0081
0080
0074
0074
0185
0180
0158
0157
0.145
0.140
0137
0.137
0136
0.135
0.095
0.051
0.026
0.022
0.018
0.077
0.063
0.061
0.058
0.056
0.052
0.051
0.050
0.050
0.049
0421
P10
State
DC
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
HI
HI
HI
HI
HI
ID
ID
ID
ID
ID
ID
ID
ID
ID
ID
IL
County
District of Columbia
Highlands
Baker
Suniter
Pasco
Jackson
Franklin
Volusia
Hardce
Hamilton
DeSoto
Warren
Early
Clay
Wonh
Richmond
Columbia
Hancock
Bryan
Crisp
Burke
Hawaii .
Kalawao
Kauai
Honolulu
Maui
Shoshone
Franklin
Oneida
Bear Lake
Lewis
Lincoln
Clark
Washington
Twin Falls
Goodmg
Cook
PJO
0039
0026
0023
0016
0014
0014
0.013
0012
0.011
0011
0010
0033
0.032
0025
0024
0.024
0023
0022
0022
0021
0.021
0.036
0.024
0.008
0.008
0.007
0.013
0.010
0009
0.009
0.009
0008
0008
0008
0.008
0.008
0.071
PIS
Slate
DC
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
HI
HI
HI
HI
HI
ID
ID
ID
ID
ID
ID
ID
ID
ID
ID
IL
County
Distnct of Columbia
Highlands
Franklin
Sumter
Hardce
Jackson
Hemando
Dixie
Volusia
Pasco
Gadsden
Worth
Turner
Hancock
Warren
Taliaferro
Charlton
Pulaski
Richmond
Treutlen
Stewan
Kalawao
Hawaii
Kauai
Maui
Honolulu
Shoshone
Oneida
Franklin
Lewis
Bear Lake
Lincoln
Washington
Twin Falls
Gooding
Nez Perce
Vermilion
PIS
0013
0.008
0.007
0.005
0.005
0.004
0.004
0.004
0004
0.004
0.004
0.008
0.008
0007
0.007
0006
0006
0.006
0.006
0.006
0006
0.015
0011
0.006
0.003
0.002
0.006
0004
0004
0004
0004
0.004
0004
0003
0.003
0.003
0026
P2S
Slate
DC
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
HI
HI
HI
HI
HI
ID
ID
ID
ID
ID
ID
ID
ID
ID
ID
IL
County
District of Columbia
Pasco
Highlands
Volusia
Sumter
Jackson
Columbia
Duval
Osceola
Hardee
Jefferson
Hall
Turner
Hancock
Cherokee
Bibb
Emanuel
Baker
Stewan
Taliaferro
Charlton
Kauai
Hawaii
Kalawao
Maui
Honolulu
Shoshonc
Franklin
Lewis
Oneida
Bear Lake
Washington
Lincoln
Twin Falls
Gooding
Nez Perce
Vermilion
P2S
0.003
0002
0001
0001
0.001
0.001
0.001
0.001
0.001
0001
0001
0002
0002
0002
0.002
0.002
0.002
0.002
0.002
0.002
0.002
0.001
0.001
0.001
0.001
0.001
0.002
0.002
0.002
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.009
                                                                               F-8
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality guidelines. It has not been formally disseminated by
EPA.  It does not represent and should not be construed to represent any Agency determination or policy.

-------
CM
Stale
IL
IL
IL
IL
IL
IL
IL
IL
IL
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
KS
KS
KS
KS
KS
KS
KS
KS
Counly*
Rock Island
Shelbv
Mercer
Tazewell
Henry
Vermilion
Jo Daviess
Menard
Fulton
Vanderburgh
Royd
Posey
Gibson
Rush
Wurnck
Harrison
Vermillion
Warren
Bcnion
Dubuquc
Scoit
Black Hawk
Delaware
Linn
Buchanan
Bcnton
Ida
Chickasaw
Bremer
Cowley
Crawford
Prali
Lincoln
Barber
Jewell
Marshall
Elk
CM
3.55
3.47
3 19
3 15
306
306
300
296
2.93
3.33
2.52
2.51
243
233
2.27
227
226
220
2.15
5.27
5 15
503
5.02
501
499
4.88
4.83
4.81
4.79
408
393
392
389
3.85
3.81
3.77
3.76
PS
Slate
IL
IL
IL
IL
IL
IL
IL
IL
IL
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
KS
KS
KS
KS
KS
KS
KS
KS
County
Mercer
Henry
Peona
Alexander
Edgar
Fulion
Greene
Perry
Slephenson
Vanderburgh
Posey
Rush
Floyd
Wamck
Fountain
Blackford
Elkhart
Si. Joseph
Vigo
Scot!
Ida
Page
Delaware
Union
Montgomery
Fremont
Decalur
O'Bnen
Monona
Chautauqua
Edwards
Harper
Coffey
Barber
Clay
Greenwood
Rooks
PS
0.328
0.294
0.266
0215
0215
0203
0.189
0.178
0178
0576
0.176
0.160
0.140
0129
0 124
0118
0 113
0.112
0 109
0.231
0.229
0223
0213
0.211
0200
0191
0 189
0.185
0.168
0.248
0.247
0.222
0217
0216
0214
0.204
0.201
P10
State
IL
IL
IL
IL
IL
IL
IL
IL
IL
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IA
IA
IA
IA
IA
IA
IA
IA
1A
IA
KS
KS
KS
KS
KS
KS
KS
KS
County
Greene
Kankakee
Peoria
Edgar
Fulion
Alexander
Slephenson
Vermilion
Rock Island
Rush
Fountain
Si. Joseph
Decalur
Vigo
Allen
Wayne
Elkhan
Blackford
Fayelle
Wayne
Ida
Union
O'Brien
Decalur
Cherokee
Monona
Woodbury
Des Moines
Taylor
Atchison
Crawford
Neosho
Edwards
Brown
Coffey
Ottawa
Comanche
PJO
0.061
0.059
0058
0057
0.052
0052
0050
0049
0.048
0.043
0.037
0032
0028
0026
0.026
0025
0025
0025
0.021
0.072
0.072
0.066
0063
0060
0.058
0.056
0.054
0.054
0.052
0.048
0.045
0045
0.045
0.044
0041
0040
0039
PIS
Slate
IL
IL
IL
IL
IL
IL
IL
IL
IL
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
KS
KS
KS
KS
KS
KS
KS
KS
County
Greene
Cook
Carroll
Peona
Alexander
Rock Island
Knox
Stephenson
Wabash
Rush
Allen
Vigo
Royd
Si. Joseph
Elkhart
Fountain
Gibson
Clinton
Wayne
Union
O'Brien
Woodbury
Lucas
Taylor
Cass
Des Moines
Crawford
Cherokee
Franklin
Neosho
Edwards
Atchison
Ottawa
tabetic
Montgomery
Kiowa
Coffey
IMS
0.023
0023
0.020
0.020
0018
0017
0.017
0.017
0.016
0.017
0011
0.011
0.010
0.010
0.009
0009
0.008
0008
0008
0.033
0.024
0023
0.021
0.020
0.020
0019
0016
0.015
0.015
0022
0021
0.016
0.015
0.015
0.015
0.015
0.014
P2S
Slate
IL
IL
IL
IL
IL
IL
IL
IL
IL
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
KS
KS
KS
KS
KS
KS
KS
KS
County
Cook
Union
Cass
Henry
Peoria
Knox
Adams
Morgan
Rock Island
Rush
Elkhan
Fountain
Vigo
Allen
Lake
St Joseph
Vermillion
Carroll
Benton
O'Brien
Woodbury
Marshall
Taylor
Webster
Des Moines
Benton
Van Buren
Wayne
Buena Vista
Atchison
Coffey
Edwards
Clay
Cowley
Ottawa
Labetle
Bourbon
P25
0.006
0.006
0.004
0.004
0.004
0.004
0.004
0.003
0.003
0.004
0.003
0003
0003
0003
0002
0.002
0.002
0002
0.002
0.010
0.006
0.006
0.006
0005
0005
0004
0.004
0004
0004
0.005
0.004
0.004
0.003
0.003
0.003
0.003
0.003
                                                                               F-9
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality guidelines. It has not been formally disseminated by
EPA.  It does not represent and should not be construed to represent any Agency determination or policy.

-------
GM
Stale
KS
KS
KY
KY
KY
KY
KY
KY
KY
KY
KY
KY
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
ME
ME
ME
ME
ME
ME
ME
ME
ME
ME
MD
MD
MD
MD
MD
County*
Coniiinche
Neosho
Fulton
Carlisle
Boyle
Magoffin
Harrison
Graves
Hickman
Criuenden
Robertson
Nicholas
Morehousc Parish
Richland Parish
De Soto Parish
Orleans Parish
Tangipahoa Pansh
Washington Pansh
West Carroll Pansh
Rapides Parish
East Carroll Parish
St Helena Pansh
Somerset
Kennebec
Cumberland
Penobscot
Sagadahoc
Washington
York
Hancock
Oxford
Knox
Somerset
Worcester
Kent
Wicomico
Washington
GM
376
376
346
3 14
3 12
302
2.97
294
294
2.92
2.89
282
4.25
3.76
3.75
3.74
3.74
3.72
3.70
363
3.61
3.55
611
5.98
5.88
5.85
585
573
5.72
5.70
5.64
5.64
2.73
2.54
2.49
2.43
2.42
PS
State
KS
KS
KY
KY
KY
KY
KY
KY
KY
KY
KY
KY
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
ME
ME
ME
ME
ME
ME
ME
ME
ME
ME
MD
MD
MD
MD
MD
County
Woodson
Ottawa
Fulton
Carlisle
Hamson
Graves
Campbell
Barren
Magoffin
Warren
Union
Gal latin
Morehouse Parish
De Soto Parish
Rapides Pansh
West Carroll Parish
Concordia Parish
East Carroll Parish
Webster Parish
Easi Fehciana Pansh
Tensas Pansh
Richland Pansh
Somerset
Franklin
Piscataquis
Sagadahoc
Oxford
Kennebec
Hancock
Knox
York
Washington
Dorchester
Worcester
Baltimore city
Allegany
Somerset
P5
0.200
0.192
0.234
0 175
0164
0 139
0138
0.130
0127
0126
0 125
0125
0411
0404
0315
0313
0310
0305
0302
0301
0294
0294
0148
0134
0123
0120
0.120
0.112
0 110
0106
0 103
0103
0 109
0102
0099
0.090
0.089
P10
State
KS
KS
KY
KY
KY
KY
KY
KY
KY
KY
KY
KY
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
ME
ME
ME
ME
ME
ME
ME
ME
ME
ME
MD
MD
MD
MD
MD
County
Wilson
Chautauqua
Fulton
Hamson
Campbell
Warren
Carlisle
Henderson
Kenton
Henry
Mason
Magoffin
Orleans Pansh
West Carroll Pansh
Tensas Pansh
De Soto Pansh
Morehouse Pansh
Richland Pansh
Franklin Pansh
Claiborne Parish
St Helena Parish
Webster Parish
Somerset
Franklin
Knox
Sagadahoc
York
Oxford
Androscoggm
Piscataquis
Kennebec
Hancock
Dorchester
Baltimore
Worcester
Somerset
Wicomico
P10
0038
0037
0046
0045
0.042
0041
0.031
0030
0028
0024
0.023
0.022
0.073
0.062
0.053
0.053
0.047
0045
0042
0035
0034
0034
0.039
0032
0031
0028
0025
0022
0022
0022
0018
0018
0.020
0.018
0016
0014
0013
PIS
State
KS
KS
KY
KY
KY
KY
KY
KY
KY
KY
KY
KY
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
ME
ME
ME
ME
ME
ME
ME
ME
ME
ME
MD
MD
MD
MD
MD
County
Crawford
Comanche
Hamson
Kenton
Warren
Henderson
Mason
Campbell
Carlisle
Fulton
Bracken
Criuenden
Orleans Pansh
De Soto Parish
Franklin Parish
Caldwell Parish
Morehouse Parish
Ouachita Parish
West Carroll Pansh
St. Mary Pansh
Claiborne Pansh
East Carroll Pansh
Somerset
Sagadahoc
Franklin
York
Piscataquis
Knox
Oxford
Androscoggm
Kennebec
Lincoln
Dorchester
Kent
Somerset
Baltimore
Worcester
PIS
0014
0014
0017
0.013
0012
0.0 II
0011
0010
0010
0010
0010
0009
0028
0.017
0.013
0.013
0.013
0.012
0.011
0.011
0.011
0.011
0.014
0.013
0.013
0010
0.010
0.009
0008
0.007
0007
0.006
0009
0005
0005
0005
0004
P2S
State
KS
KS
KY
KY
KY
KY
KY
KY
KY
KY
KY
KY
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
ME
ME
ME
ME
ME
ME
ME
ME
ME
ME
MD
MD
MD
MD
MD
County
Jewell
Elk
Campbell
Fulton
Hopkins
Harrison
Nelson
Grant
Kenton
Morgan
Magoffin
Wayne
Orleans Pansh
Ouachita Pansh
Franklin Pansh
Morehouse Pansh
St. James Pansh
Iberville Pansh
Jackson Parish
Ibena Parish
Caldwell Parish
Grant Parish
Somerset
Sagadahoc
York
Franklin
Piscalaquis
Penobscot
Knox
Washington
Androscoggm
Waldo
Dorchester
Kent
Somerset
Wicomico
Washington
P25
0.003
0003
0003
0.003
0.002
0.002
0.002
0.002
0.002
0.002
0.002
0.002
0.004
0003
0003
0002
0002
0002
0002
0.002
0.002
0.002
0.004
0.004
0.003
0.003
0.002
0.002
0.002
0.002
0.002
0001
0002
0001
0001
0001
0001
                                                                               F-10
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality guidelines. It has not been formally disseminated by
EPA.  It does not represent and should not be construed to represent any Agency determination or policy.

-------
CM
State
MD
MD
MD
MD
MD
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
Ml
Ml
Ml
Ml
Ml
Ml
Ml
Ml
Ml
Ml
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MS
MS
County*
Dorchester
Queen Anne's
Caroline
Allegany
Baltimore
Naniucket
Dukes
Hampden
Franklin
Worcester
Suffolk
Hampshire
Berkshire
Bnstol
Essex
Lenuwee
Jackson
Muskegon
Kcwccnaw
Sagmaw
Bav
Hillsdiile
Monroe
Huron
Kent
Kandiyohi
Lincoln
Renville
Brown
Ramsey
Redwood
Swift
Pine
Rock
Itasca
Jefferson Davis
Covington
GM
2.26
1.99
1.97
195
1 84
3 II
289
287
276
254
240
230
2.29
2.29
217
350
2.83
2.62
' 2.46
237
2.33
228
213
2.13
204
290
2.60
2.52
2.48
2.46
245
2.45
2.38
2.37
236
332
329
PS
Stale
MD
MD
MD
MD
MD
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
Ml
Ml
Ml
Ml
Ml
Ml
Ml
Ml
Ml
Ml
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MS
MS
County
Wicomico
Baltimore
Caroline
Kent
Washington
Hampden
Nantucket
Berkshire
Dukes
Franklin
Worcester
Bnstol
Essex
Suffolk
Hampshire
Lenawee
Jackson
Wayne
Hillsdalc
Keweenaw
Sagmaw
Kent
Bay
Branch
Mason
Lincoln
Ramsey
Big Stone
Kittson
Kandiyohi
Cottonwood
Cook
Norman
Renville
Yellow Medicine
Washington
Lauderdale
PS
0087
0076
0.061
0060
0.057
0.153
0.147
0.122
0 114
0113
0.097
0.097
0.083
0.080
0.065
0258
0156
0 115
0110
0103
0.103
0100
0097
0.095
0.093
0.138
0.107
0.100
0.100
0.098
0093
0093
0091
0089
0087
0 177
0150
P10
State
MD
MD
MD
MD
MD
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
Ml
Ml
Ml
Ml
Ml
Ml
Ml
Ml
Ml
Ml
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MS
MS
County
Baltimore city
Kent
Allegany
Caroline
Garreti
Berkshire
Franklin
Nantucket
Dukes
Hampden
Suffolk
Worcester
Bristol
Essex
Plymouth
Jackson
Wayne
Kent
Emmet
Branch
Presque Isle
Keweenaw
Lenawee
Muskegon
Bern en
Koochichmg
Ramsey
Norman
Martin
Fillmore
Kittson
Houston
Cottonwood
Nobles
Freebom
Pike
Jones
P10
0013
0012
0011
0009
0008
0.024
0.021
0021
0020
0020
0.015
0.015
0.015
0.015
0012
0.027
0.026
0.025
0.023
0020
0.020
0020
0017
0017
0015
0027
0025
0019
0.018
0018
0018
0018
0017
0.017
0.017
0030
0029
PIS
Stale
MD
MD
MD
MD
MD
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MI
MI
MI
MI
Ml
MI
Ml
Ml
Ml
MI
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MS
MS
County
Wicomico
Harford
Allegany
Talbot
Baltimore city
Berkshire
Dukes
Franklin
Hampden
Essex
Bristol
Nantucket
Worcester
Suffolk
Barnstabte
Wayne
Kent
Jackson
Muskegon
Gogebic
Keweenaw
Hillsdale
Huron
Branch
Kalamazoo
Watonwan
Cottonwood
Norman
Nobles
Ramsey
Kittson
Freebom
Fillmore
Big Stone
Jackson
Lauderdale
Smith
PIS
0.004
0.003
0.003
0.003
0.002
0.008
0.007
0.007
0.006
0.005
0.005
0.004
0.004
0.004
0.004
0009
0008
0008
0.007
0006
0006
0006
0005
0.005
0005
0010
0.010
0009
0008
0008
0008
0008
0008
0007
0.007
0013
0.012
P2S
State
MD
MD
MD
MD
MD
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
Ml
Ml
Ml
Ml
Ml
MI
Ml
Ml
Ml
Ml
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MS
MS
County
Baltimore
Talbot
St. Mary's
Allegany
Baltimore city
Berkshire
Barnsiable
Norfolk
Essex
Plymouth
Franklin
Bristol
Hampden
Worcester
Hampshire
Kent
Branch
Hillsdale
Wayne
Keweenaw
Jackson
Bern en
Presque Isle
Sanilac
Gogebic
Cot ton wood
Nobles
Ramsey
Mower
Norman
Big Stone
Yellow Medicine
Pipestone
Sibley
Faribault
Lauderdale
Pike
P25
0001
0001
0001
0001
0001
0002
0002
0002
0002
0.001
0001
0001
0001
0001
0.001
0002
0002
0.002
0002
0002
0.002
0.001
0.001
0.001
0.001
0.003
0.003
0.002
0002
0.002
0002
0.002
0002
0002
0.002
0003
0.002
                                                                               F-ll
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality guidelines. It has not been formally disseminated by
EPA.  It does not represent and should not be construed to represent any Agency determination or policy.

-------
CM
Stale
MS
MS
MS
MS
MS
MS
MS
MS
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MT
MT
MT
MT
MT
MT
MT
MT
MT
MT
NE
NE
NE
NE
NE
NE
NE
NE
NE
County*
Forrest
Perry
Marion
Humphreys
Sionc
Lincoln
Luniiir
Sharkey
Worth
Livingston
Iron
Vernon
Carroll
St Clair
Gentry
Scotland
Atchi.son
Washington
Prairie
Golden Valley
Petroleum
Wibaux
Wheatland
Daniels
Deer Lodge
Musselshell
Garfleld
Judith Basin
Pawnee
Ncmaha
Dakota
Richardson
Franklin
Arthur
Keya Paha
Thayer
Fillmore
GSM
3 18
3 IS
292
291
284
282
281
280
396
392
391
372
370
369
356
356
353
353
293
284
283
282
278
278
277
269
269
263
377
374
372
371
3.69
368
355
353
352
PS
State
MS
MS
MS
MS
MS
MS
MS
MS
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MT
MT
MT
MT
MT
MT
MT
MT
MT
MT
NE
NE
NE
NE
NE
NE
NE
NE
NE
County
Lincoln
Covington
Humphreys
Sharkey
Coahoma
Quitman
Wilkinson
Choctaw
Worth
Carroll
Alchison
Si. Louis city
DcKalb
Gentry
Shelby
Iron
Vernon
Scotland
Wheatland
Deer Lodge
Prairie
Dawson
Golden Valley
Powell
Petroleum
Daniels
Wibaux
Treasure
Fillmore
Otoe
Boone
Franklin
Richardson
Thurslon
Nemaha
Polk
Pawnee
PS
0 148
0 140
0 130
0.128
0123
0 123
0122
0 122
0331
0289
0241
0239
0.215
0.215
0.214
0.212
0203
0202
0.135
0 131
0.115
0110
O.I 10
0.106
0103
0102
0 101
0099
0228
0217
0208
0.203
0197
0188
0.184
0.176
0.170
P10
State
MS
MS
MS
MS
MS
MS
MS
MS
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MT
MT
MT
MT
MT
MT
MT
MT
MT
MT
NE
NE
NE
NE
NE
NE
NE
NE
NE
County
Washington
Yazoo
Laudcrdale
Jefferson Davis
Amite
Smith
Sharkey
Leake
St. Louis city
Wonh
Caldwell
Iron
Clark
Audrain
Buchanan
St. Francois
Carroll
Scotland
Wheatland
Prairie
Deer Lodge
Petroleum
Golden Valley
Daniels
Treasure
Wibaux
Chouteau
Powell
Fillmore
Jefferson
Otoe
Thayer
Pawnee
York
Richardson
Boone
Franklin
PJO
0.026
0.025
0022
0.021
0021
0020
0019
0019
0067
0.056
0048
0.045
0.045
0.044
0042
0041
0041
0038
0.016
0012
0.012
0.012
0.012
0011
0010
0.010
0.010
0.010
0.050
0.041
0.037
0.031
0030
0.030
0.029
0.029
0.029
PIS
State
MS
MS
MS
MS
MS
MS
MS
MS
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MT
MT
MT
MT
MT
MT
MT
MT
MT
MT
NE
NE
NE
NE
NE
NE
NE
NE
NE
County
Pike
Pontotoc
Yazoo
Tallahatchie
Coahoma
Washington
Webster
Tunica
St. Louis city
Clark
Buchanan
Audrain
Worth
Caldwell
Clinton
Lewis
St. Francois
Carroll
Deer Lodge
Prairie
Wheailand
Golden Valley
Daniels
Petroleum
Chouteau
Custer
Treasure
Wibaux
Jefferson
Otoe
York
Burt
Richardson
Dixon
Fillmore
Franklin
Pawnee
PJ5
0.01 1
0009
0008
0007
0007
0007
0.007
0006
0024
0021
0018
0.014
0.014
0.014
0.013
0.013
0.012
0.012
0.005
0.005
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.017
0.016
0.014
0014
0014
0.013
0.01 1
0.011
0.010
P2S
State
MS
MS
MS
MS
MS
MS
MS
MS
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MT
MT
MT
MT
MT
MT
MT
MT
MT
MT
NE
NE
NE
NE
NE
NE
NE
NE
NE
County
Tallahatchie
Humphreys
Jones
Neshoba
Harrison
Pontoioc
Montgomery
Sunflower
St Louis city
Caldwell
Buchanan
Audrain
Pettis
Clark
Linn
Carroll
Atchison
Chanton
Wheatland
Deer Lodge
Golden Valley
Fergus
Prairie
Custer
Powell
Roosevelt
Chouteau
Blaine
Fillmore
Hall
Jefferson
Pawnee
Richardson
Franklin
Greeley
Kearney
York
P25
0.002
0.002
0.002
0.002
0001
0.001
0.001
0.001
0.006
0.004
0.004
0.004
0003
0.003
0.003
0.003
0.003
0003
0002
0002
0.002
0.002
0.001
0.001
0.001
0.001
0.001
0.001
0.004
0.004
0.003
0003
0003
0.003
0.003
0.003
0.003
                                                                               F-12
7/1/5 information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality guidelines. It has not been formally disseminated by
EPA.  It does not represent and should not be construed to represent any Agency determination or policy.

-------
CM
Stale
NE
NV
NV
NV
NV
NV
NV
NV
NV
NV
NV
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NM
NM
NM
NM
NM
NM
County*
Grant
While Pine
Lincoln
Mineral
Storey
Pershing
Carson Cily
Esmenildu
Nye
Humboklt
Lander
Coos
Cheshire
Belknap
Graflon
Mem mack
Stratford
Carroll
Sullivan
Rockingham
Hillsborough
Union
Cumberland
Salem
Somerset
Essex
Mercer
Moms
Sussex
Hunierdon
Atlantic
Harding
Mora
Guudalupe
Union
De Baca
Col fax
GM
3.49
2.96
293
274
265
254
246
243
238
2.35
2.35
543
502
4.93
470
460
423
420
3.89
297
2.71
3.51
3.36
332
331
330
3 16
3 16
3.01
2.99
2.88
3.39
3.18
3.09
306
302
294
PS
Slate
NE
NV
NV
NV
NV
NV
NV
NV
NV
NV
NV
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NM
NM
NM
NM
NM
NM
County
Sherman
White Pine
Lincoln
Mineral
Pershing
Esmeralda
Storey
Humboldt
Carson City
Churchill
Elko
Coos
Sullivan
Belknap
Graflon
Strafford
Mem mack
Cheshire
Carroll
Hillsborough
Rockingham
Cumberland
Essex
Salem
Mercer
Union
Atlantic
Passaic
Hunierdon
Camden
Hudson
Harding
San Juan
Mora
Eddy
Union
Guadalupe
PS
0168
0197
0144
0 141
0.122
0.117
0.112
0.096
0.095
0.094
0092
0.318
0.234
0198
0197
0 180
0179
0.172
0 146
0.127
0.069
0.170
0156
0 127
0 113
0.109
0102
0097
0077
0077
0072
0.225
0.220
0198
0194
0186
0173
P10
State
NE
NV
NV
NV
NV
NV
NV
NV
NV
NV
NV
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NM
NM
NM
NM
NM
NM
County
Deuel
While Pine
Mineral
Lincoln
Pershing
Storey
Humboldt
Carson City
Elko
Esmeralda
Churchill
Coos
Strafford
Sullivan
Mem mack
Belknap
Cheshire
Hillsborough
Grafton
Carroll
Rockingham
Cumberland
Essex
Salem
Union
Mercer
Atlantic
Passaic
Ocean
Somerset
Hunierdon
Harding
Guadalupe
Mora
Union
Dona Ana
De Baca
P10
0.028
0.066
0044
0043
0036
0032
0031
0030
0.029
0.029
0.028
0.066
0.047
0.046
0.044
0044
0037
0035
0032
0028
0.015
0.029
0.026
0.021
0019
0016
0014
0013
0012
0011
0010
0035
0029
0.024
0.023
0.022
0019
PIS
Slate
NE
NV
NV
NV
NV
NV
NV
NV
NV
NV
NV
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NM
NM
NM
NM
NM
NM
County
Deuel
White Pine
Lincoln
Mineral
Storey
Pershing
Carson City
Esmeralda
Churchill
Washoe
Humboldl
Slrafford
Belknap
Coos
Merrimack
Sullivan
Hillsborough
Carroll
Cheshire
Grafton
Rockingham
Cumberland
Union
Essex
Ocean
Salem
Passaic
Hunierdon
Mercer
Somerset
Middlesex
Harding
Guadalupe
Union
Grant
Mora
De Baca
IMS
0.010
0.014
0.009
0.009
0008
0008
0.007
0007
0007
0.006
0006
0016
0015
0013
0013
0012
0.011
0.010
0.010
0.009
0.006
0009
0008
0007
0005
0005
0005
0.004
0.004
0003
0003
0.013
0.009
0008
0.007
0.007
0.007
P2S
State
NE
NV
NV
NV
NV
NV
NV
NV
NV
NV
NV
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NM
NM
NM
NM
NM
NM
County
Websier
White Pine
Lincoln
Storey
Carson City
Churchill
Pershing
Mineral
Washoe
Lyon
Clark
Belknap
Strafford
Hillsborough
Grafton
Coos
Carroll
Mem mack
Sullivan
Rockingham
Cheshire
Cumberland
Union
Essex
Passaic
Ocean
Middlesex
Gloucester
Hudson
Burlington
Salem
Harding
Mora
Union
Socorro
San Miguel
Grant
P25
0.003
0.002
0.002
0.002
0.002
0.002
0.002
0.002
0.001
0.001
0001
0004
0.004
0003
0002
0002
0002
0002
0002
0.001
0.001
0002
0.002
0.002
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.002
0.001
0.001
0.001
0.001
0.001
                                                                               F-13
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality guidelines. It has not been formally disseminated by
EPA.  It does not represent and should not be construed to represent any Agency determination or policy.

-------
CM
State
NM
NM
NM
NM
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
OH
OH
OH
County*
Sun Miguel
Hidalao
Quay
Sierra
Chaulauqua
Montgomery
Cayuaa
Albany
Hcrkimer
Fullon
Wyoming
Otsego
Orleans
Lewis
Tyrrell
Halifax
Scotland
Surry
Warren
Bertie
Gates
Northampton
Caswell
Washington
Slope
Burke
Sheridan
Nelson
Divide
Cnggs
Eddy
LaMoure
Einmons
Golden Valley
Manon
Belmont
Lucas
GM
286
277
276
276
394
370
355
351
348
346
3.46
3.46
3.46
345
2.89
2.73
2.67
2.67
264
2.58
2.58
256
253
253
3 19
296
293
292
2.90
2.85
284
282
2.80
2.77
345
3.30
3.15
PS
Slate
NM
NM
NM
NM
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
OH
OH
OH
County
Col fax
De Baca
San Miguel
Hidalgo
Chautauqua
Montaomery
Fulton
Herkimer
Orleans
Cayuga
Otsego
Cattaraugus
Chemung
Wyoming
Tyrrell
Halifax
Hyde
Gates
Warren
Surry
Northampton
Lenoir
Bertie
Caswell
Slope
Benson
Barnes
Sheridan
Ransom
Golden Valley
Burke
Ramsey
Nelson
Emmons
Lucas
Seneca
Noble
PS
0162
0142
0 140
0.138
0.156
0153
0 133
0 129
0.127
0.123
0.123
0.123
0.120
0.120
0153
0143
0 141
0.130
0.126
0123
0 121
0119
0117
0 115
0132
0.124
0.117
O.I II
0.109
0 108
0108
0.107
0.106
0104
0298
0.229
0.200
P10
State
NM
NM
NM
NM
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
OH
OH
OH
County
Taos
Colfax
Quay
Hidalgo
Montgomery
Chautauqua
Orleans
Schcnectady
Cayuga
Fulton
Chemung
Herkimer
Cortland
Wyoming
Caswell
Surry
Duplin
Halifax
Edgecombe
Lenoir
Nash
Hyde
Warren
Yadkin
Benson
Ransom
Slope
Golden Valley
Burke
Nelson
Grand Forks
Eddy
Sheridan
Emmons
Cuyahoga
Lucas
Ashlabula
PJO
0.018
0.017
0016
0016
0026
0025
0020
0.020
0020
0020
0019
0019
0018
0018
0.019
0.016
0.015
0.014
0.013
0.012
0.012
0011
0011
0.011
0.017
0017
0016
0012
0012
0011
0011
0011
0010
0010
0062
0034
0031
PIS
State
NM
NM
NM
NM
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
OH
OH
OH
County
San Juan
Colfax
Curry
Taos
Montgomery
Chaulauqua
Schenectady
Cayuga
Orleans
Chemung
Fulton
Wyoming
Kings
Herkimer
Caswell
Duplin
Nash
Gates
Lenoir
Jones
Onslow
Surry
Stokes
Halifax
Slope
Benson
Burke
Golden Valley
Nelson
Eddy
Sheridan
Ransom
Mclntosh
LaMoure
Cuyahoga
Lucas
Mahomng
. PIS
0.006
0.006
0.006
0.005
0.007
0.007
0.006
0006
0.006
0.006
0.005
0.005
0.005
0.005
0.006
0006
0005
0004
0004
0.004
0.004
0.004
0.004
0.004
0.006
0.005
0.004
0.004
0.004
0.004
0.004
0004
0.004
0.004
0024
0.012
0.011
P2S
State
NM
NM
NM
NM
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
OH
OH
OH
County
Guadalupc
Hidalgo
Colfax
Dona Ana
Montgomery
Chautauqua
Kings
Orleans
Cayuga
Herkimer
Wyoming
Chemung
Cortland
Otsego
Duplin
Surry
Nash
Hertford
Granvillc
Caswell
Jones
Greene
Onslow
Lenoir
Ransom
Benson
Slope
Barnes
Emmons
Traill
Sheridan
Nelson
Eddy
Steele
Cuyahoga
Hamilton
Clark
P25
0001
0001
0001
0.001
0.002
0.002
0.002
0.002
0.001
0.001
0.001
0.001
0.001
0.001
0002
0002
0001
0001
0001
0.001
0.001
0.001
0.001
0.001
0.002
0.002
0002
0002
0.002
0.002
0.002
0.001
0.001
0.001
0004
0.003
0002
                                                                              F-14
This information is distributed solely for the purpose ofpre-disseminalion peer review under applicable information quality guidelines. It has not been formally disseminated by
EPA.  It does not represent and should not be construed to represent any Agency determination or policy.

-------
CM
Stale
OH
OH
OH
OH
OH
OH
OH
OK
OK
OK
OK
OK
OK
OK
OK
OK
OK
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
County"
Scioto
Richlund
Seneca
Morrow
Williams
Cuyahoga
Ashiabula
Bcckham
Alfalfa
Harmon
Gram
Kiowa
Jefferson
Hughes
Ellis
Cimarron
Greer
Sherman
Gilliam
Wheeler
Wai Iowa
Baker
Union
Uike
Malhcur
Multnomah
Hamey
Philadelphia
Delaware
Blair
Northumberland
Schuylkill
Indiana
Montgomery
Union
Venango
Carbon
GM
3.12
3.03
290
284
283
277
2.76
336
3.35
331
323
322
3.16
3 10
309
304
3.01
226
222
2.17
2 10
203
1.96
1.94
1 92
I.8S
1 83
364
334
3.12
3.06
3.06
287
2.86
2.80
2.78
2.78
PS
State
OH
OH
OH
OH
OH
OH
OH
OK
OK
OK
OK
OK
OK
OK
OK
OK
OK
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
County
Cuyahoga
Morgan
Coshoclon
Ashtabula
Muskingum
Wood
Muhoning
Jefferson
Harmon
Choclaw
Tillman
Alfalfa
Hughes
Kiowa
Beckhum
Cotton
Noble
Multnomah
Malheur
Clatsop
Gilliam
Sherman
Wheeler
Buker
Wai Iowa
Union
Tillamook
Philadelphia
Northumberland
Schuylkill
Blair
Delaware
Venango
Dauphin
Indiana
Berks
York
PS
0196
0188
0 146
0.144
0.134
0134
0124
0223
0.216
0.211
0.207
0.205
0.199
0.195
0 187
0185
0.181
0.098
0098
0072
0066
0065
0064
0063
0.061
0.058
0053
0261
0234
0219
0215
0207
0204
0.193
0187
0.182
0.170
P10
State
OH
OH
OH
OH
OH
OH
OH
OK
OK
OK
OK
OK
OK
OK
OK
OK
OK
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
County
Mahoning
Morgan
Clark
Noble
Coshoclon
Monroe
Hamilton
Jefferson
Alfalfa
Kiowa
Hughes
Grant
Ellis
Aloku
Okmulgee
Harmon
Cimarron
Multnomah
Wasco
Gilliam
Sherman
Yamhill
Wheeler
Baker
Umalilla
Washington
Wallowa
Northumberland
Blair
Lancaster
York
Berks
Schuylkill
Lycoming
Philadelphia
Venango
Indiana
P10
0.031
0029
0027
0026
0025
0.024
0024
0.037
0033
0.032
0.030
0.028
0.026
0.025
0023
0022
0022
0020
0015
0012
0.011
0010
0010
0.010
0.010
0.009
0.009
0.061
0057
0056
0055
0052
0051
0049
0047
0047
0046
PIS
State
OH
OH
OH
OH
OH
OH
OH
OK
OK
OK
OK
OK
OK
OK
OK
OK
OK
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
County
Coshoclon
Hamilton
Ashtabula
Clark
Holmes
Knox
Ene
Alfalfa
Jefferson
Hughes
Logan
Grant
Tillman
Ellis
Harmon
Okmulgee
Beckham
Multnomah
Umatilla
Gilliam
Yamhill
Sherman
Wasco
Wheeler
Wallowa
Baker
Union
Lancaster
Northumberland
Berks
York
Huntingdon
Indiana
Dauphin
Lycoming
Schuylkill
Delaware
P15
0.010
0.010
0.009
0.009
0.009
0008
0008
0012
0.011
0011
0010
0.010
0.010
0.009
0008
0.008
0.008
0.008
0.007
0.006
0.005
0.005
0.005
0.005
0.004
0004
0004
0025
0023
0.023
0.022
0.022
0.020
0019
0019
0019
0.018
P2S
State
OH
OH
OH
OH
OH
OH
OH
OK
OK
OK
OK
OK
OK
OK
OK
OK
OK
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
County
Lucas
Coshoclon
Hardin
Miami
Mahoning
Muskingum
Jefferson
Adair
Hughes
Carter
Washington
Payne
Jefferson
Harmon
Tillman
Alfalfa
Okmulgee
Multnomah
Sherman
Yamhill
Union
Baker
Umalilla
Wheeler
Gilliam
Columbia
Tillamook
Delaware
Chester
Indiana
Northumberland
Dauphin
Lycoming
Lancaster
Berks
York
Schuylkill
P25
0.002
0.002
0.002
0.002
0002
0002
0.002
0003
0.003
0002
0.002
0002
0002
0002
0002
0.002
0002
0002
0002
0002
0.002
0002
0002
0001
0001
0.001
0.001
0.007
0.007
0.006
0.006
0.006
0.005
0.005
0.005
0.005
0.004
                                                                               F-15
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality guidelines. It has not been formally disseminated by
EPA.  It does not represent and should not be construed to represent any Agency determination or policy.

-------
CM
Stale
Kl
Rl
Kl
Rl
Kl
SC
SC
SC
SC
SC
SC
SC
SC
SC
SC
SD
SD
SD
SD
SD
SD
SD
SD
SD
SD
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TX
TX
County*
Providence
Newport
Washington
Bristol
Kent
McCormick
Allenclale
Bambcrs
Chester
Marlboro
Dillon
Union
Barn well
Calhoun
Lee
Miner
Sanborn
Aurora
Jerauld
Clark
Hyde
Bon Homme
Douglas
Turner
Spink
Hamblen
Scott
Campbell
Anderson
Mudison
Lake
Haywood
Grainger
Claiborne
Hardm
Mason
Baylor
CM
208
176
1.75
1.73
1.67
2.73
2.72
268
267
264
259
259
2.58
2.57
255
3.22
3.21
3.18
3.13
3 12
3.10
3.10
3.07
3.05
3.00
3.12
3.05
2.95
2.93
2.91
2.88
2.87
2.86
2.85
2.85
3.32
3.28
PS
Slate
Rl
Rl
Rl
Rl
Rl
SC
SC
SC
SC
SC
SC
SC
SC
SC
SC
SD
SD
SD
SD
SD
SD
SD
SD
SD
SD
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TX
TX
County
Providence
Newport
Washington
Bristol
Kent
Allendale
Union
Chester
Newberry
McCormick
Marion
Marlboro
Bamberg
Fairfield
Dillon
Miner
Sanborn
Aurora
Clark
Roberts
Moody
Turner
Hyde
Day
Jerauld
Hardm
Marshall
Unicoi
Bedford
Fayelte
Carter
Lewis
Johnson
Meigs
Madison
Col lings worth
Shackelford
P5
0 137
0085
0076
0073
0061
0 112
0098
0.095
0092
0091
0090
0090
0089
0087
0087
0139
0 138
0135
0132
0125
0 124
0123
0 122
0122
0121
0164
0146
0122
0118
0115
0115
0 114
0 110
0109
0 109
0224
0215
P10
State
Rl
Rl
Rl
Rl
Rl
SC
SC
SC
SC
SC
SC
SC
SC
SC
SC
SD
SD
SD
SD
SD
SD
SD
SD
SD
SD
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TX
TX
County
Providence
Newport
Washington
Bristol
Kent
Allendale
Chester
Union
Dillon
Marlboro
Bamberg
Greenwood
Manon
Newberry
Pickens
Miner
Sanborn
Clark
Aurora
Moody
Hyde
Turner
Hanson
Douglas
Charles Mix
Bedford
Hawkins
Coffee
Marshall
Putnam
Washington
Hardeman
Unicoi
Monroe
Lincoln
Bosque
Jones
P10
0.032
0.021
0.020
0.020
0.012
0.016
0.015
0014
0013
0013
0013
0012
0012
0.012
0012
0017
0.017
0.016
0.016
0016
0015
0015
0.014
0014
0.014
0.036
0.032
0.030
0.026
0.024
0.023
0.020
0.020
0.019
0.018
0.038
0.037
PIS
Stale
Rl
Rl
Rl
Rl
Rl
SC
SC
SC
SC
SC
SC
SC
SC
SC
SC
SD
SD
SD
SD
SD
SD
SD
SD
SD
SD
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TX
TX
County
Providence
Newport
Bristol
Washington
Kent
Allendale
Chester
Union
Marlboro
Bamberg
McCormick
Dillon
Greenwood
Orangeburg
Calhoun
Miner
Aurora
Sanborn
Clark
Moody
Turner
Douglas
Bon Homme
Hanson
Jerauld
Coffee
Bedford
Marshall
Putnam
Madison
Morgan
Loudon
Lake
Hardeman
Unicoi
Young
Bosque
PIS
0.011
0.007
0.006
0.006
0.003
0.005
0.005
0005
0004
0004
0004
0004
0004
0004
0004
0007
0006
0006
0006
0.006
0.006
0.005
0.005
0.005
0.005
0010
0009
0.008
0.008
0.008
0.007
0.007
0.006
0.006
0.006
0.019
0017
P25
State
Rl
Rl
Rl
Rl
Rl
SC
SC
SC
SC
SC
SC
SC
SC
SC
SC
SD
SD
SD
SD
SD
SD
SD
SD
SD
SD
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TX
TX
County
Providence
Bristol
Washington
Newport
Kent
Allendale
Chester
Union
Marlboro
Lee
Bamberg
Manon
Newberry
Dillon
Bamwell
Aurora
Hyde
Miner
Sanborn
Turner
Clark
Moody
Charles Mix
Buffalo
Hanson
Monroe
Coffee
Putnam
Crockett
Dickson
Lawrence
White
Unicoi
Roane
Loudon
Nacogdoches
Falls
P25
0.002
0.002
0.002
0.002
0.001
0001
0001
0001
0.001
0.001
0001
0001
0.001
0.001
0.001
0.003
0.003
0.003
0.003
0.002
0.002
0002
0002
0002
0002
0003
0003
0.002
0.002
0.002
0002
0.001
0001
0.001
0001
0.003
0002
                                                                              F-16
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality guidelines.  It has not been formally disseminated by
EPA.  It does not represent and should not be construed to represent any Agency determination or policy.

-------
CM
Slate
TX
TX
TX
TX
TX
TX
TX
TX
UT
UT
UT
UT
UT
UT
UT
UT
UT
UT
VT
VT
VT
VT
VT
VT
VT
VT
VT
VI-
VA
VA
VA
VA
VA
VA
VA
VA
VA
County*
Jones
Conic
Colemun
Collingsworth
Throckmorton
Mcnurd
Shackelforcl
Donley
Piute
Juab
Sanpele
Millard
Wayne
Sevier
Duchcsne
Beaver
GarTield
Umiah
Essex
Grand Isle
Rutland
Caledonia
Windham
Addison
Franklin
Windsor
Benninglon
Washington
Clifton Forge city
Northampton
Lancaster
Highland
Middlesex
Accomack
Caroline
Nottoway
Mecklenburg
CM
3.27
3.26
3.24
3.23
3.21
321
321
3 19
267
260
237
233
229
2.29
2.27
2.25
2.19
2.17
237
2.03
2.02
1.99
1.99
185
183
1.81
1.80
1.79
3.07
2.91
274
260
255
2.53
249
247
2.46
PS
State
TX
TX
TX
TX
TX
TX
TX
TX
UT
UT
UT
UT
UT
UT
UT
UT
UT
UT
VT
VT
VT
VT
VT
VT
VT
VT
VT
VT
VA
VA
VA
VA
VA
VA
VA
VA
VA
County
Falls
Terrell
Wheeler
Red River
Mitchell
East land
Morris
Dickens
Juab
Piute
Umtah
Duchesne
Beaver
Sevier
Wayne
Garficld
Cache
Millard
Rutland
Windham
Caledonia
Benninglon
Addison
Orange
Franklin
Washington
Essex
Windsor
Northampton
Middlesex
Clifton Forge cily
Lancaster
WincheMer city
Caroline
Essex
Mecklenburg
Richmond
PS
0.201
0193
0186
0.184
0.183
0.182
0.180
0.177
0.146
0.089
0.088
0076
0075
0073
0071
0070
0070
0.070
0.247
0.230
0.228
0199
0193
0 192
0.191
0187
0.178
0.173
0202
0.198
0.185
0.181
0169
0146
0 138
0.138
0 130
P10
State
TX
TX
TX
TX
TX
TX
TX
TX
UT
UT
UT
UT
UT
UT
UT
UT
UT
UT
VT
VT
VT
VT
VT
VT
VT
VT
VT
VT
VA
VA
VA
VA
VA
VA
VA
VA
VA
County
Baylor
Collings worth
Young
Red River
DeWitt
Refugio
Hill
Lavaca
Juab
Piute
Box Elder
Sanpete
Carbon
Millard
Duchesne
Beaver
Sevier
Wayne
Rutland
Washington
Windham
Franklin
Bennington
Caledonia
Orange
Addison
Orleans
Essex
Winchester city
Clifton Forge city
Danville city
Petersburg city
Northampton
Richmond city
Orange
Middlesex
Essex
P10
0.035
0035
0032
0032
0031
0031
0031
0.030
0.027
0.009
0.009
0.008
0008
0007
0007
0007
0.006
0.006
0.072
0.061
0.053
0.049
0.048
0047
0047
0044
0.040
0035
0.048
0.044
0.041
0.040
0038
0.028
0023
0023
0.023
PIS
State
TX
TX
TX
TX
TX
TX
TX
TX
UT
UT
UT
UT
UT
UT
UT
UT
UT
UT
VT
VT
VT
VT
VT
VT
VT
VT
VT
VT
VA
VA
VA
VA
VA
VA
VA
VA
VA
County
Falls
Lavaca
Kimble
Collingsworth
Hill
Wilbarger
Refugio
Schleicher
Juab
Piute
Box Elder
Sanpctc
Carbon
Millard
Sevier
Beaver
Weber
Wayne
Rutland
Franklin
Washington
Orange
Bennington
Windham
Caledonia
Orleans
Addison
Chittenden
Petersburg city
Clifton Forge cily
Danville city
Northampton
Richmond city
Accomack
Lancaster
Orange
Southampton
pas
0.012
0012
0.012
0.011
0.0 II
0.010
0010
0.010
0003
0003
0003
0003
0003
0002
0002
0002
0.002
0.002
0027
0.025
0.024
0.020
0.019
0017
0017
0.014
0014
0012
0.017
0.013
0.013
0011
0.009
0.009
0007
0007
0.007
P2S
State
TX
TX
TX
TX
TX
TX
TX
TX
UT
UT
UT
UT
UT
UT
UT
UT
UT
UT
VT
VT
VT
VT
VT
VT
VT
VT
VT
VT
VA
VA
VA
VA
VA
VA
VA
VA
VA
County
Schleicher
Lampasas
Grayson
Lamar
Medina
Ellis
Baylor
Anderson
Juab
Box Elder
Piute
Sanpete
Sevier
Beaver
Millard
Uintah
Weber
Carbon
Franklin
Washington
Rutland
Orleans
Chittenden
Orange
Bennington
Windsor
Addison
Caledonia
Clifton Forge city
Covmgton city
Petersburg city
Roanoke city
Halifax
Danville cily
Suffolk city
Bristol city
Richmond city
P2S
0.002
0.002
0.002
0.002
0.002
0.002
0002
0.002
0.002
0001
0001
0001
0001
0001
0.001
0.001
0.001
0.001
0009
0006
0.006
0.006
0.006
0.005
0.004
0.003
0.003
0.003
0003
0003
0.002
0002
0002
0002
0.002
0002
0002
                                                                               F-17
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality guidelines. It has not been formally disseminated by
EPA.  It does not represent and should not be construed to represent any Agency determination or policy.

-------
CM
State
VA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WV
WV
WV
WV
WV
WV
WV
WV
WV
WV
Wl
W[
Wl
Wl
Wl
Wl
Wl
Wl
Wl
Wl
WY
WY
WY
WY
WY
WY
County"
Northumberland
Garfield
Columbia
Okunogan
Asolm
Yakima
Lincoln
Grays Harbor
Walla Walla
Wahkiakum
Whuicom
Boone
Kanawha
Taylor
Doddndgc
Ritchie
Lewis
Roane
Monroe
Greenbricr
Tucker
Milwaukee
Wmnebago
Clark
Brown
Kenosha
Waupaca
Sauk
Rusk
Douglas
Ozaukee
Albany
Natrona
Frcmoni
Carbon
La ramie
Niobrara
GM
2.46
233
221
216
2 14
2.05
1.98
1 89
189
185
185
308
283
270
2.58
2.54
250
249
249
248
246
323
252
232
2.18
216
215
204
203
202
202
407
395
3 17
2.99
2.93
2.89
PS
State
VA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WV
WV
WV
WV
WV
WV
WV
WV
WV
WV
Wl
Wl
Wl
Wl
Wl
Wl
Wl
Wl
Wl
Wl
WY
WY
WY
WY
WY
WY
County
Danville city
Okanogan
Garfield
Whatcom
Columbia
Grays Harbor
Lincoln
Wahkiakum
Douglas
Asolin
Ferry
Doddridge
Taylor
Ritchie
Pocahontas
Tucker
Ohio
Lewis
McDowell
Preston
Hampshire
Milwaukee
Wmnebago
Waupaca
Richland
Sauk
Vernon
Green
Racine
Shcboygan
Kenosha
Goshen
Niobrara
Fremont
Big Horn
Natrona
Sheridan
PS
0.130
0.091
0.086
0.082
0068
0.066
0.065
0.060
0.058
0.057
0.055
0.174
0.168
0.160
0.149
0.121
0.119
0.118
0.117
0.109
0.108
0.117
0.115
O.I II
0.096
0.095
0.094
0.092
0.091
0.090
0.090
0.150
0.132
0.105
0.102
0.093
0.091
P10
State
VA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WV
WV
WV
WV
WV
WV
WV
WV
WV
WV
Wl
Wl
Wl
Wl
Wl
Wl
Wl
Wl
Wl
Wl
WY
WY
WY
WY
WY
WY
County
Roanoke city
Garfield
Columbia
Wahkiakum
Asolin
Grays Harbor
Clark
Chclan
Lincoln
Walla Walla
Cowlitz
Doddndge
Taylor
Ritchie
Preston
Lewis
Tyler
Pocahontas
McDowell
Manon
Monongaha
Milwaukee
Sheboygan
Richland
Sauk
Lafayelle
Kenosha
Manilowoc
Green
Jefferson
Racine
Niobrara
Big Horn
Converse
Johnson
La ramie
Natrona
P10
0022
0.013
0.013
0.010
0.010
0010
0.010
0.009
0.009
0.009
0.008
0.029
0.027
0.026
0026
0021
0020
0020
0.017
0017
0.017
0.041
0.028
0.025
0022
0022
0.021
0020
0020
0020
0018
0.012
0.009
0.008
0.007
0007
0007
PIS
Slate
VA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WV
WV
WV
WV
WV
WV
WV
WV
WV
WV
Wl
Wl
Wl
Wl
Wl
Wl
Wl
Wl
Wl
Wl
WY
WY
WY
WY
WY
WY
County
Suffolk city
Columbia
Garfield
Wahkiakum
Yakima
Asotin
Grays Harbor
Lincoln
Chelan
Walla Walla
Whitman
Doddndge
Taylor
Preston
Monongalia
Lewis
McDowell
Brooke
Tyler
Berkeley
Ohio
Milwaukee
Sheboygan
Richland
Ozaukee
Iron
Sauk
Waupaca
Jefferson
Mamtowoc
Columbia
Niobrara
Big Horn
Johnson
Laramie
Platte
Sheridan
PIS
0.007
0.006
0.006
0005
0004
0004
0004
0.004
0.004
0.004
0004
0012
0.008
0.008
0.007
0.007
0.007
0.006
0.006
0.006
0.005
0.014
0.011
0.009
0009
0008
0.008
0.007
0.006
0006
0006
0004
0.003
0.003
0.003
0002
0002
P25
State
VA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WV
WV
WV
WV
WV
WV
WV
WV
WV
WV
Wl
Wl
Wl
Wl
Wl
Wl
Wl
Wl
Wl
Wl
WY
WY
WY
WY
WY
WY
County
Washington
Columbia
Garfield
Grays Harbor
Yakima
Lincoln
Wahkiakum
Asotin
Grant
Pierce
Lewis
Taylor
Monongalia
Doddridge
Ohio
Preston
McDowell
Lewis
Berkeley
Gilmer
Manon
Sheboygan
Milwaukee
Ozaukee
Richland
Lafayette
Waupaca
Pepin
Kenosha
Jefferson
Sauk
Niobrara
Big Horn
Sheridan
Natrona
Goshen
Johnson
P25
0.002
0.002
0002
0002
0002
0.002
0.001
0001
0001
0001
0001
0.002
0.002
0.002
0.002
0.002
0.002
0001
0001
0001
0001
0003
0.003
0.003
0.002
0.002
0.002
0.002
0.002
0002
0002
0002
0.001
0.001
0.001
0001
0001
                                                                               F-18
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality guidelines. It has not been formally disseminated by
EPA.  It does not represent and should not be construed to represent any Agency determination or policy.

-------
CM
Stale
WY
WY
WY
WY
County*
Sheridan
Converse
Goshcn
Bis Horn
GM
2.83
2.68
2.50
2.40
PS
Slate
WY
WY
WY
WY
County
Albany
Carbon
Johnson
La ramie
I>5
0.090
0.083
0.080
0074
P10
Stale
WY
WY
WY
WY
County
Carbon
Sheridan
Plane
Lincoln
IMO
0.007
0.007
0.006
0006
PIS
State
WY
WY
WY
WY
County
Carbon
Washakie
Goshen
Nairona
P15
0.002
0.002
0002
0.002
P2S
State
WY
WY
WY
WY
County
Platte
Converse
Washakie
Hot Springs
P25
0.001
0.001
0.001
0.001
* CA = Census Area, Bor. = Borough
                                                                              F-19
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality guidelines.  It has not been formally disseminated by
EPA.  It does not represent and should not be construed to represent any Agency determination or policy.

-------
Table F-3. Top 150 Census Tracts in Massachusetts with Highest Blood Lead Levels
GM
Census Tract
25013801700
25013801800
25005651400
25013801300
25013802100
25013802300
250 1380 1401
25005651300
25013802601
25013812300
25013801900
25023510800
25015820102
25013802200
25013810403
25005651600
25027707200
25005651100
250235 11400
25027732600
25005652100
25027710700
25023511500
25013812701
25009221400
25009221700
25005652700
25013812702
25023510900
25005631600
25013800500
25013800100
25009221500
25013800600
25013802000
County
Humpdcn
Hampdcn
Bristol
Hampden
Hampdcn
Hampden
Hampden
Bristol
Hampdcn
Hampdcn
Hampdcn
Plymouth
Hampshire
Hampden
Hampden
Bristol
Worcester
Bristol
Plymouth
Worcester
Bnstol
Worcester
Plymouth
Hampden
Essex
Essex
Bnstol
Hampdcn
Plymouth
Bnstol
Hampdcn
Hampden
Essex
Hampden
Hampden
GM
3.791
3.745
3702
3.627
3597
3558
3511
3.500
3.419
3.396
3.384
3.350
3.346
3 334
3.306
3283
3261
3 235
3.224
3224
3221
3218
3217
3200
3 191
3 179
3.176
3.152
3 135
3 134
3.127
3.110
3.098
3093
3084
PS
Census Tract
25005651400
25005651300
25013801700
25013801300
25027710700
25013802100
25013801800
25023510800
25013802601
25023511400
25013802300
25023511500
25013810403
25003900100
25005651600
25027711000
25005651100
25005652700
250 138 12701
25003900600
25027754300
25009268300
25009221500
25027710800
25003921200
25013812300
25013801401
25003900200
25027732600
25005652500
25015820102
25009221400
25013802200
25005652100
25009260100
County
Bristol
Bristol
Hampden
Hampden
Worcester
Hampden
Hampden
Plymouth
Hampden
Plymouth
Hampden
Plymouth
Hampden
Berkshire
Bnstol
Worcester
Bristol
Bnstol
Hampden
Berkshire
Worcester
Essex
Essex
Worcester
Berkshire
Hampden
Hampden
Bcikshire
Worcester
Bnstol
Hampshire
Essex
Hampden
Bnstol
Essex
PS .
0280
0276
0.254
0245
0.239
0235
0.235
0234
0233
0232
0232
0.230
0.224
0224
0.224
0224
0.221
0.217
0212
0.212
0.212
0.211
0.208
0.208
0.208
0.208
0207
0203
0.201
0.201
0201
0200
0 199
0.199
0198
P10
Census Tract
25005651400
25005651300
25023510800
25005652100
25013801800
25005652700
25005651600
25013802601
25003922100
25013801300
25027710700
25005631400
25023510400
25005651900
25013802100
25009206500
25005631600
25023511400
25013801700
25005651100
25025091200
25015820102
25005650600
25005652600
25005651700
25027707200
25003900200
25009260200
250 138 II 800
25003900100
25005652500
25015822000
25023511600
25027710800
25009221500
County
Bristol
Bristol
Plymouth
Bnstol
Hampden
Bnstol
Bnstol
Hampden
Berkshire
Hampden
Worcester
Bnstol
Plymouth
Bristol
Hampden
Essex
Bnstol
Plymouth
Hampden
Bnstol
Suffolk
Hampshire
Bristol
Bnstol
Bristol
Worcester
Berkshire
Essex
Hampden
Berkshire
Bnstol
Hampshire
Plymouth
Worcester
Essex
P10
0.076
0.070
0.056
0.053
0.050
0.049
0047
0047
0046
0.046
0045
0044
0044
0044
0043
0043
0043
0043
0.042
0.042
0042
0042
0042
0040
0040
0040
0040
0040
0040
0040
0.039
0.038
0.038
0.038
0.038
PIS
Census Tract
25005651300
25023510800
25005651400
25003900100
25013801800
25013801300
25015820600
25005651900
25027710700
25005651800
25013801700
25027710800
25015820102
25013802100
25027732600
25003900600
25025082000
25005631600
25023510900
25025091700
25009260200
25005651100
25013802300
25005651200
25023510400
25023511500
25013812300
25023511400
25003900200
25005652700
25027707200
25005652100
25013812701
25015822000
25023511000
County
Bristol
Plymouth
Bnstol
Berkshire
Hampdcn
Hampden
Hampshire
Bnstol
Worcester
Bnstol
Hampden
Worcester
Hampshire
Hampden
'Worcester
Berkshire
Suffolk
Bristol
Plymouth
Suffolk
Essex
Bristol
Hampden
Bristol
Plymouth
Plymouth
Hampden
Plymouth
Berkshire
Bnstol
Worcester
Bnstol
Hampden
Hampshire
Plymouth
P15
0.017
0.016
0015
0.015
0013
0013
0.013
0.013
0.013
0.012
0.012
0.012
0.012
0.012
0.012
0.012
0.01 1
0.011
0011
0011
0011
0010
0010
0010
0.010
0.010
0.010
0.010
0.010
0.010
0.010
0010
0010
0010
0.010
                                                                         F-20
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality guidelines.  It has not been formally disseminated by
EPA. It does not represent and should not be construed to represent any Agency determination or policy.

-------
CM
Census Tract
250 138 13301
25005651700
25023510400
250 1380 1501
25013802400
25013800400
25013810902
25027731400
25027757300
25005652500
25003900100
25003922100
25005641800
25005652600
25013802500
25025091800
25003900600
25027707300
25005631400
25013800900
25003900200
25003921400
25003922200
25013801402
25027754300
2501 104 1300
25005651800
25027732400
25005651900
25013801200
25013813401
25005650600
25009268300
2501 1040701
25027731201
25005641900
25013813403
County
Humpdcn
Bristol
Plymouth
Hampden
Hampdcn
Hampden
Hampden
Worcester
Worcester
Bristol
Berkshire
Berkshire
Bristol
Bristol
Humpdcn
Suffolk
Berkshire
Worcester
Bristol
Hampdcn
Berkshire
Berkshire
Berkshire
Hampdcn
Worcester
Franklin
Bristol
Worcester
Bristol
Hampden
Hampden
Bristol
Essex
Franklin
Worcester
Bristol
Hampden
GM
3070
3065
3.063
3.048
3047
3047
3.041
3.039
3.035
3030
3.025
3003
3002
3001
3.000
2.999
2996
2996
2.989
2984
2.983
2979
2974
2967
2.965
2.963
2962
2.961
2.953
2.952
2.951
2.950
2.936
2934
2932
2932
2914
P5
Census Tract
25005631600
25027730402
25003921300
25005650600
25023510900
25005631400
25027707200
25027710100
25027726200
25003922100
25009221700
25003922200
25005641900
25005651800
25003921400
25027757300
25005641800
25023510400
25009260200
25023511600
25005651900
25013801900
2501 1040701
25025091200
25009260600
25025091900
25005642300
25005652200
25027707300
2501 1041300
25027710600
25005640900
25025091800
25009204400
25027732400
250 138 13301
25027744300
County
Bristol
Worcester
Berkshire
Bristol
Plymouth
Bnstol
Worcester
Worcester
Worcester
Berkshire
Essex
Berkshire
Bristol
Bristol
Berkshire
Worcester
Bnstol
Plymouth
Essex
Plymouth
Bnstol
Hampden
Franklin
Suffolk
Essex
Suffolk
Bnstol
Bristol
Worcester
Franklin
Worcester
Bristol
Suffolk
Essex
Worcester
Hampden
Worcester
PS
0198
0197
0196
0 194
0193
0192
0 191
0 190
0.189
0189
0.186
0.186
0.185
0185
0 185
0185
0183
0183
0.182
0182
0 181
0181
0 179
0.178
0.176
0175
0 174
0174
0 174
0173
0 172
0172
0.171
0170
0 169
0169
0.169
P10
Census Tract
25023511500
25009204400
25003925100
25023510900
25005641800
25003922200
25009268300
25009204500
25009221700
25003900600
25027732400
25003921200
250 138 12701
25013801900
25027710600
2501 1040500
25027732600
25027731201
25005651800
25013802200
25025091800
25027744300
25025092300
25027731600
25023511000
25009260800
25009221400
25025060300
25027710100
2501 104 1300
25025091500
25027711000
25027732500
25025091700
25025091100
25027730802
25003921300
County
Plymouth
Essex
Berkshire
Plymouth
Bnstol
Berkshire
Essex
Essex
Essex
Berkshire
Worcester
Berkshire
Hampdcn
Hampden
Worcester
Franklin
Worcester
Worcester
Bnstol
Hampden
Suffolk
Worcester
Suffolk
Worcester
Plymouth
Essex
Essex
Suffolk
Worcester
Franklin
Suffolk
Worcester
Worcester
Suffolk
Suffolk
Worcester
Berkshire
P10
0038
0038
0037
0037
0.037
0036
0036
0036
0.036
0.035
0.034
0.034
0034
0034
0034
0.034
0033
0033
0.033
0.033
0.033
0.033
0.032
0.032
0.032
0032
0032
0032
0.031
0031
0031
0.031
0031
0031
0.030
0.030
0.030
P15
Census Tract
25025080100
25025090700
25003922100
25025081500
25027744300
25005613800
25005652600
25023511301
25005651700
25005631400
25013802200
25025010102
25025091500
25023511600
25025090400
25001010100
25025092300
25003925100
25025120300
25009206000
25005650700
25013801900
250 138 II 800
25009251200
25027757300
25009204500
25017310800
25011041400
2501 1041300
25027754300
25027731900
25013800600
25015820400
25013802000
25027732400
25027732500
25005651500
County
Suffolk
Suffolk
Berkshire
Suffolk
Worcester
Bnstol
Bristol
Plymouth
Bristol
Bristol
Hampden
Suffolk
Suffolk
Plymouth
Suffolk
Bamstable
Suffolk
Berkshire
Suffolk
Essex
Bnstol
Hampden
Hampden
Essex
Worcester
Essex
Middlesex
Franklin
Franklin
Worcester
Worcester
Hampden
Hampshire
Hampden
Worcester
Worcester
Bnstol
P15
0.010
0.010
0.010
0009
0009
0009
0009
0.009
0009
0.009
0009
0.009
0.009
0.009
0.009
0.009
0.009
0.009
0.009
0.009
0.008
0008
0008
0008
0008
0008
0008
0008
0008
0008
0008
0008
0008
0008
0.008
0.008
0008
                                                                               F-21
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality guidelines. It has not been formally disseminated by
EPA.  It does not represent and should not be construed to represent any Agency determination or policy.

-------
CM
Census Tract
25025091700
25027726200
25005650700
25027730402
25003921200
25013800201
25013801605
25027711000
25009204300
25025092300
2501 1040100
25009260100
250 138 II 800
25025091200
25009206500
25025091900
25023511000
25027731900
25027733000
25005651200
25027732700
25013810300
25003921300
25013810200
25027732500
25005650500
25005652200
25005650800
25013801503
25027757200
25023511600
25009204400
25013801502
25013810901
25011041400
25005640200
25025092000
County
Suffolk
Worcester
Bnslol
Worcester
Berkshire
Humpclen
Hampden
WorceMer
Essex
Suffolk
Franklin
Essex
Hampdcn
Suffolk
Essex
Suffolk
Plymouth
Worcester
Worcester
Bristol
Worcester
Hampden
Berkshire
Hampden
Worcester
Bristol
Bristol
Bnslol
Hampden
Worcester
Plymouth
Essex
Hampdcn
Hampden
Franklin
Bnslol
Suffolk
GM
2.911
2.904
2.903
2.893
2.890
2.888
2.880
2.879
2876
2871
2.870
2.868
2.866
2.863
2.857
2.856
2854
2.854
2.852
2.849
2.848
2.848
2.844
2.843
2.837
2.831
2831
2.827
2.824
2.810
2.808
2.808
2805
2804
2.803
2.802
2799
P5
Census Tract
25013812702
25005650500
25009260800
25005650700
25013800100
25005651700
25025091700
25027732500
25005651200
25027731201
25003925100
25005650800
25005640500
25009204300
250 1380 1501
25005652600
25009266300
25027731600
25025100300
25009250900
25013800500
25027725100
25027731400
250 138 13401
25023510501
25027731300
25025092300
25013802500
25005652000
25003931100
25005640600
25027731900
25027757200
25013802400
25027723100
25027703200
25025091300
County
Hampden
Bristol
Essex
Bristol
Hampden
Bnslol
Suffolk
Worcester
Bristol
Worcester
Berkshire
Bnsiol
Bnslol
Essex
Hampden
Bnslol
Essex
Worcester
Suffolk
Essex
Hampden
Worcester
Worcester
Humpden
Plymouth
Worcester
Suffolk
Hampden
Bristol
Berkshire
Bnslol
Worcester
Worcester
Hampden
Worcester
Worcester
Suffolk
PS
0 169
0168
0168
0167
0167
0167
0166
0.166
0.166
0.165
0.163
0163
0.162
0.161
0.160
0.159
0.158
0158
0158
0156
0154
0 154
0 154
0 154
0154
0153
0 152
0.151
0.151
0.151
0.151
0150
0.150
0150
0150
0 149
0 149
P10
Ceasus Tract
25005613800
25025080100
25013812300
25025090600
25025082000
25009268400
25003900900
25015820600
25025091300
25027707300
250IJS02300
25005652000
25005655300
25025091400
25025100300
25009250600
25013812702
25027754300
25005631500
25005650400
25027731500
2501 1040701
25005640500
25009260600
25027731900
25025091900
25027731400
25025081500
25025090700
25005640900
25009204200
25023510700
25023510501
25013810902
25023511301
25027703300
2501 104 1400
County
Bristol
Suffolk
Hampden
Suffolk
Suffolk
Essex
Berkshire
Hampshire
Suffolk
Worcester
Hampden
Bristol
Bnstol
Suffolk
Suffolk
Essex
Hampdcn
Worcester
Bristol
Bnstol
Worcester
Franklin
Bnstol
Essex
Worcester
Suffolk
Worcester
Suffolk
Suffolk
Bristol
Essex
Plymouth
Plymouth
Hampden
Plymouth
Worcester
Franklin
P10
0.030
0.030
0.030
0.030
0030
0030
0030
0030
0030
0029
0029
0029
0028
0028
0.028
0.028
0.028
0.028
0.028
0.028
0.028
0.028
0.028
0.028
0.028
0.028
0.028
0027
0027
0027
0027
0027
0027
0027
0.027
0026
0.026
PIS
Census Tract .
25005651600
2501 1040500
25027731600
25025100300
25027711000
25001012400
2500101 1700
25005652000
25005640600
25005641800
25013800800
25009205500
25009221400
25027731300
25025090200
25025091400
25025081700
25025081800
25025090600
25009260600
25013802601
25025091600
25009204400
25005631500
25017311100
25005650600
25009204300
25001011600
25009206800
25003921500
250 1 38 10901
25013811600
25005650800
25003900300
25027733000
25025091300
25023510501
Count)
Bnstol
Franklin
Worcester
Suffolk
Worcester
Bamstable
Bamstable
Bnslol
Bnslol
Bristol
Hampden
Essex
Essex
Worcester
Suffolk
Suffolk
Suffolk
Suffolk
Suffolk
Essex
Hampden
Suffolk
Essex
Bnstol
Middlesex
Bristol
Essex
Bamstable
Essex
Berkshire
Hampden
Hampden
Bristol
Berkshire
Worcester
Suffolk
Plymouth
P15
0.008
0.008
0008
0.008
0008
0008
0008
0.008
0.008
0.008
0008
0.008
0.008
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0007
0007
0007
0007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
                                                                               F-22
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality guidelines. It has hot been formally disseminated by
EPA.  It does not represent and should not be construed to represent any Agency determination or policy.

-------
CM
Census Tract
250 1380 II 02
25013812402
25013813207
25025010102
25005640900
25009206700
25025091300
25025100300
25009266300
25009260600
25005652000
25013810800
25025 10 1102
25013800300
25023510300
25009204200
250 1380 1601
25009260200
25013800202
25027710100
25009206600
25013813404
25013800800
25017351400
25013802602
25025091100
25013810700
25027710800
25025091400
2501381 1 101
25025090400
25015820600
25027744300
25013800700
25013812200
25027731600
25003931100
County
Hampden
Hampden
Hampden
Suffolk
Bristol
Essex
Suffolk
Suffolk
Essex
Essex
Bristol
Hampden
Suffolk
Hampden
Plymouth
Essex
Hampden
Essex
Hampden
Worcester
Essex
Hampden
Hampden
Middlesex
Hampden
Suffolk
Hampden
Worcester
Suffolk
Hampden
Suffolk
Hampshire
Worcester
Hampden
Hampden
Worcester
Berkshire
GM
2.794
2790
2788
2786
2780
2778
2775
2.769
2.768
2.766
2762
2762
2.760
2.758
2.750
2.748
2.748
2743
2.741
2738
2735
2734
2732
2732
2730
2728
2.727
2.726
2722
2.722
2.721
2.718
2713
2712
2.698
2697
2691
Pa
Census Tract
25009250700
25025090700
25005613800
25025091100
25015821902
25025091500
25027754400
250 138 II 800
25027733000
25009204200
25009206600
25003922300
25003900900
25009250300
25005640200
25025 101 102
25023511000
25025091400
25027710500
25005631300
25005613600
25005655300
25013813403
25023511301
25027703300
25013800400
25011041500
25025090600
25013810902
25009204500
25027732700
25005613700
25009206500
25027731500
25025080100
25011040500
25013810300
County
Essex
Suffolk
Bristol
Suffolk
Hampshire
Suffolk
Worcester
Hampden
Worcester
Essex
Essex
Berkshire
Berkshire
Essex
Bristol
Suffolk
Plymouth
Suffolk
Worcester
Bristol
Bnstol
Bristol
Hampden
Plymouth
Worcester
Hampden
Franklin
Suffolk
Hampden
Essex
Worcester
Bnstol
Essex
Worcester
Suffolk
Franklin
Hampden
P5
0148
0147
0147
0 146
0 146
0.146
0.145
0.144
0.144
0143
0.143
0.143
0.142
0.141
0.141
0.141
0 141
0140
0 140
0140
0 140
0.140
0.140
0.139
0.138
0138
0138
0138
0.138
0137
0.137
0136
0.136
0136
0.136
0.135
0.135
P10
Census Tract
25005640600
250 1380 1401
25009205900
25027726200
25009206800
25005651500
25025100601
25005630100
25027732203
25005652300
25005650800
25009260500
25015822700
25005650700
25009266300
25009206400
25025160500
25017353300
25013800400
2501 1040100
25009206900
25025092000
25025 10 II 02
25005652200
25005641900
25003921400
25009206700
25009206300
25001010100
25009210800
25013800500
25027757200
25009206600
25027757300
25027730402
25009204300
25009260100
County
Bnsiol
Hampden
Essex
Worcester
Essex
Bnsiol
Suffolk
Bristol
Worcester
Bristol
Bristol
Essex
Hampshire
Bnstol
Essex
Essex
Suffolk
Middlesex
Hampden
Franklin
Essex
Suffolk
Suffolk
Bnstol
Bnstol
Berkshire
Essex
Essex
Bamstable
Essex
Hampden
Worcester
Essex
Worcester
Worcester
Essex
Essex
P10
0026
0.026
0.026
0.026
0.026
0.026
0026
0026
0025
0025
0025
0025
0025
0025
0025
0025
0025
0.025
0.024
0.024
0.024
0024
0024
0024
0024
0024
0.024
0024
0.023
0.023
0.023
0.023
0.023
0.023
0.023
0023
0.023
PIS
Census Tract
25013810902
25005614000
25025091200
25009207100
25009206500
25009221500
25009250600
25027730300
25005640900
25009251700
25025091 100
25027709702
25025091000
25005613700
2501381 1700
25003921300
25025050300
2501 1040701
25009207200
25001012500
25005652500
25027757200
25017310700
25009250200
25017337300
25003900900
25027731201
25009206100
25025100601
25027737200
25005640500
25023510700
25013810403
25009206200
25027707300
25009266300
25025140201
County
Hampden
Bnsiol
Suffolk
Essex
Essex
Essex
Essex
Worcester
Bnstol
Essex
Suffolk
Worcester
Suffolk
Bristol
Hampden
Berkshire
Suffolk
Franklin
Essex
Bamstable
Bristol
Worcester
Middlesex
Essex
Middlesex
Berkshire
Worcester
Essex
Suffolk
Worcester
Bristol
Plymouth
Hampden
Essex
Worcester
Essex
Suffolk
P15
0007
0007
0007
0007
0007
0007
0.007
0007
0007
0007
0007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0007
0007
0.007
0007
0007
0007
0007
0.007
0007
0007
0007
0007
0007
0.007
0.007
0.007
0007
0.007
                                                                              F-23
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality guidelines. It has not been formally disseminated by
EPA.  It does not represent and should not be construed to represent any Agency determination or policy.

-------
CM
Census Tract
25009207200 ,-
25011041500
25009205900'
25025091500
County
Essex
Franklin
Essex
Suffolk
GM
2.684
2.684 '
2.682
2.680
P5
Census Tract
. 25009260500
25025010102
25005631500
25009250200
County
Essex
Suffolk
Bristol
Essex
PS
- 0.135--
0.134
0.134
0.134
P10
Census Tract
25025060800
2501 1041200
• 25027754400
25005650500
County
Suffolk
Franklin
Worcester
Bristol
P10
0.023
0.023
0.023
0.022
PIS
Census Tract
25027731500
25023511200
25009206700
25025092200
County
Worcester
Plymouth
Essex
Suffolk
PIS
0.007
0.006
0006
0006
                                                                              F-24
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality guidelines. It has not been formally disseminated by
EPA.  It does not represent and should not be construed to represent any Agency determination or policy.

-------
            Census Tracts with Highest Predicted Blood-Lead Levels
                            in 2006 (Top 150)
            Census Tracts with Highest Predicted Proportion
           of Blood-Lead Levels >/= 5 ug/dL in 2006 (Top 150)
                                                                                    LEGEND
                                                                                    of PbB Levels >/= 5 ug/dL
    Geometric Mean PbB Level (ug/dL)
            2.680 - 2.795
            2.796-2.915
            2.9I6-3.I25
         ••3.126-3.420
         •13.421 +
        0.1338-0.1460
        0.1461 -0.1630
        0.1631 -0.1865
        0.1866-0.2170
        0.2171 +
                            Massachusetts
                         Massachusetts
           LEGEND
       rtion of PbB Levels >/= I 0 ug/dL
           0.0225 - 0.0265
           0.0266 - 0.0320
           0.0321 -0.0395
        •1 0.0396 - 0.0565
        •1 0.0566 +
        LEGEND
Proportion of PbB Levels >/= 15 Ug/dL
        0.0065 - 0.007
        0.0071 -0.008
        0.0081 - 0.009
        0.0091 -0.010
        0.0101 +
                            Massachusetts
                         Massachusetts
Figure F-l. 150 Census Tracts in Massachusetts with Highest Predicted 2006 Blood Lead Levels

                                                                                   F-25
This information ix distributed solely for the purpose ofpre-dissemination peer review under applicable information quality guidelines.  It has not been formally disseminated by
EPA. It does not represent and should not be construed to represent any Agency determination or policy.

-------
                Appendix G



Detailed Maps of National and State Model Output

-------
                     Region 1 - Geometric Mean Blood Lead Levels
        Observed Blood-Lead Levels by County
                       2000
                                                         Observed Blood-Lead Levels by County
                                                                        2005
          LEGEND
  Geometric Mean PbB Level (ug/dL)
    I I - 1.999  MS -5.999
2 - 2.999
3 - 3.999
4 - 4.999
                  6 - 7.999
                  8 +
                  No data
                              US EPA Region 1
        LEGEND
Geometric Mean PbB Level (ug/dL)
  H I - 1.999   ^B 5 -5.999
I   j 2 - 2.999   •• 6 - 7.999
I   i 3 - 3.999   • 8 +
                                                                                US EPA Region 1
        Predicted Blood-Lead Levels by County
                       2000
                                                          Predicted Blood-Lead Levels by County
                                                                        2005
          LEGEND
  Geometric Mean PbB Level (ug/dL)
    _' I - 1.999  •• 5-5.999
      2 - 2.999
      3 - 3.999
     I 4 - 4.999
            I 6 - 7.999
            |8 +
            I No data
                               US EPA Region 1
                                                           LEGEND
                                                   Geometric Mean PbB Level (ug/dL)
                                                      Zl-l.999  Ml 5-5.999
    2 - 2.999
    3 - 3.999
    4 - 4.999
I 6 - 7.999
|8 +
] No data
                                                                                US EPA Region 1
                                                   G-l
This information is distributed solely for the purpose ofpre -dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA.  It does not represent and should not be construed
to represent any Agency determination or policy.

-------
                          Region 1 - Proportion of BLLs >  10
  Observed Proportion of Blood-Lead Levels >/= 10 ug/dL
                         2000
                                        Observed Proportion of Blood-Lead Levels >/= 10 ug/dL
                                                               2005
            LEGEND
   Proportion of PbB Levels >/= I 0 ug/dL
      0 - 0.0099     BB 0.075 - 0.0999
      0.01 -0.0249
      0.025 - 0.0499
      0.05 - 0.0749
O.IO-0.2499
0.25 +
No data
                                 US EPA Region 1
           LEGEND
  Proportion of PbB Levels >/= ID Ug/dL
[   I 0 - 0.0099    •  0.075 - 0.0999
LO.OI -0.0249  Blo.io-o.2499   US EPA Region 1
    0.025 - 0.0499 •  0.25 +
• 0.05-0.0749  •  No data
  Predicted Proportion of Blood-Lead Levels >/= 10 ug/dL
                         2000
                                        Predicted Proportion of Blood-Lead Levels >/= 10 ug/cIL
                                                               2005
            LEGEND
   Proportion of PbB Levels >/= 10 ug/dL
 I    I 0 - 0.0099     •• 0.075 - 0.0999
 I    I 0.01 - 0.0249   •• 0.10 - 0.2499
 I    : 0.025 - 0.0499 •ง 0.25 +
 •I 0.05 - 0.0749   !"-• No data
            US EPA Region 1
          LEGEND
 Proportion of PbB Levels >/= 10 ug/dL
rn 0 - 0.0099    • 0.075 - 0.0999
n 0.01 - 0.0249  • 0.10 - 0.2499
L~3 0.025 - 0.0499 • 0.25 +
• 0.05 - 0.0749  ~l No data
US EPA Region 1
                                                      G-2
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA.  It does not represent and should not be construed
to represent any Agency determination or policy.

-------
                    Region 2 - Geometric Mean Blood Lead Levels
        Observed Blood-Lead Levels by County
                      2000
                                              Observed Blood-Lead Levels by County
                                                             2005
         LEGEND
 Geometric Mean PbB Level (ug/dL)
  L I  - 1.999   Hi S - 5.999
     2 - 2.999
     3 - 3.999
   • 4. 4.999
I 6 - 7.999
18 +
1 No data
                              US EPA Region 2
        LEGEND
Geometric Mean PbB Level (ug/dL)
  ~ I - 1.999  • 5 -5.999
I   i 2 - 2.999  •• 6 • 7.999
I   1 3 - 3.999  •18 +
•I 4 - 4.999  • No data
                                                                     US EPA Region 2
        Predicted Blood-Lead Levels by County
                       2000
         LEGEND
  Geometric Mean PbB Level (ug/dL)
      I - 1.999  •• 5 - 5.999
      2 - 2.999  ปi 6 - 7.999
  I    3 - 3.999  • 8 +
  • 4 - 4.999  • No data
              US EPA Region 2
                                               Predicted Blood-Lead Levels by County
                                                             2005
                                                LEGEND
                                         Geometric Mean PbB Level (ug/dL)
                                         I    I - 1.999  H 5 -5.999
                                            2 - 2.999
                                            3 - 3.999
                                            I 4 - 4.999
               I 6 - 7.999
               18 +
               ] No data
                             US EPA Region 2
                                                   G-3
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA.  It does not represent and should not be construed
to represent any Agency determination or policy.

-------
                          Region 2 - Proportion of BLLs >  10 ng/dL
  Observed Proportion of Blood-Lead Levels >/= 10 ug/dL
                         2000
                            Observed Proportion of Blood-Lead Levels >/= 10 ug/dL
                                                   2005
            LEGEND
  Proportion of PbB Levels >/= I 0 ug/dL
     0 - 0.0099
     0.0 1  - 0.0249
 I   I 0.025 - 0.0499
 tm 0.05 - 0.0749
I 0.075 - 0.0999
I o.io-0.2499   US EPA Region 2
I 0.25 +
] No data                    nun.™ 2007
                                      LEGEND
                            Proportion of PbB Levels >/= 10 Ug/dL
                             Z 0 - 0.0099    Bi 0.075 - 0.0999
                               0.01  -0.0249
                             • 0.025 - 0.0499
                             • 0.05 - 0.0749
I O.IO-0.2499
I 0.25 +
i No data
                                                          US EPA Region 2
  Predicted Proportion of Blood-Lead Levels >/= 10 ug/dL
                         2000
                            Predicted Proportion of Blood-Lead Levels >/= 10 ug/dL
                                                   2005
            LEGEND
  Proportion of PbB Levels >/= I 0 ug/dL
   Hi 0 - 0.0099    • 0.075 - 0.0999
   Zl 0.01 - 0.0249  •• O.IO- 0.2499
     0.025  - 0.0499 •ง 0.25 +
 •• 0.05 - 0.0749  C3 No data
US EPA Region 2
                                      LEGEND
                            Proportion of PbB Levels >/= 10 ug/dL
                            I  I 0 - 0.0099    • 0.075 - 0.0999
                             __ 0.01  -0.0249
                            EJS3 0.025 - 0.0499
                            • 0.05 - 0.0749
I O.IO-0.2499
I 0.25 +
I No data
             US EPA Region 2
                                                      G-4
This information ix distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
                     Region 3 - Geometric Mean Blood Lead Levels
         Observed Blood-Lead Levels by County
                       2000
  Geometric Mean PbB Level (ug/dL)
      I - 1.999  •15-5.999
       - 2.999  M6-7.999
   L  ,3- 3.999
   • 4-4.999  M No data
                                                    Geometric Mean PbB Level (ug/dL)
                                                    i_^ 0 - 1.999  •• 5 - 5.999
                                                    Cj 2-2.999  MM 6 -7.999
                                                    CZ3 3 - 3.999  • 8 +
                                                    • 4-4.999  • No data
        Predicted Blood-Lead Levels by County
                       2000
                                                           Predicted Blood-Lead Levels by County
                                                                         2005
          LEGEND
  Geometric Mean PbB Level (ug/dL)
  I   : I - 1.999  MS -5.999
 -2.999
3 - 3.999
4-4.999
                  6- 7.999
                  8 +
                  No data
                               US EPA Region 3
                                                                 LEGEND
Geometric Mean PbB Level (ug/dL)
CH 0-1.999  MS-5.999
CH 2 - 2.999  ma 6 - 7.999
     - 3.999  •• 8 '
   14-4.999  B No data
                                                                                 US EPA Region 3
                                                   G-5
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA.  It does not represent and should not be construed
to represent any Agency determination or policy.

-------
                          Region 3 - Proportion of BLLs > 10 fig/dL
  Observed Proportion of Blood-Lead Levels >/= 10 ug/dL
                         2000
                            Observed Proportion of Blood-Lead Levels >/= 10 ug/dL
                                                   2005
            LEGEND
   Proportion of PbB Levels >/= I 0 ugldL
  I   ' 0 - 0.0099     • 0.075  - 0.0999
   Z! 0.01 - 0.0249   •• 0.10 - 0.2499
  I   10.025 - 0.0499  • 0.25 +
  •i 0.05 - 0.0749   HZ No data
US EPA Region 3
          LEGEND
 Proportion of PbB Levels >/= I 0 Ug/dL
!   0 - 0.0099     • 0.075  - 0.0999
ZZJ 0.01 - 0.0249   Hi 0.10 - 0.2499
E31 0.025 - 0.0499  • 0.25 +
IH 0.05 - 0.0749   EH] No data
US EPA Region 3
  Predicted Proportion of Blood-Lead Levels >/= 10 ug/cIL
                         2000
                            Predicted Proportion of Blood-Lead Levels >/= 10 ug/dL
                                                   2005
            LEGEND
   Proportion of PbB Levels >/= 10 ug/dL
   Z 0 - 0.0099    1H 0.075 - 0.0999
  ZZ 0.01  - 0.0249  •• 0.10 - 0.2499
  ETJ 0.025 - 0.0499 ^0.25 +
  •I 0.05 - 0.0749  Lฑ3 No data
US EPA Region 3
          LEGEND
 Proportion of PbB Levels >/= 10 ug/dL
 H 0 - 0.0099     • 0.075 - 0.0999
iZT! 0.01 - 0.0249   • 0.10 - 0.2499
1   0.025 - 0.0499  •• 0.25 +
^ 0.05 - 0.0749   HH No data
US EPA Region 3
                                                      G-6
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA.  It does not represent and should not be construed
to represent any Agency determination or policy.

-------
                     Region 4 - Geometric Mean Blood Lead Levels
        Observed Blood-Lead Levels by County
                       2000
                                 Observed Blood-Lead Levels by County
                                                2005
         LEGEND
  Geometric Mean PbB Level (ug/dL)
      I - 1.999  M 5 - 5.999
  CH 2 -2.999  M6-7.999
      3-3.999  M8 +
  •U-4.999  Ml No data
US EPA Region 4
        LEGEND
Geometric Mean PbB Level (ug/dL)
CZ I-1.999  •15-5.999
CH 2-2.999  Ml 6-7.999
iH] 3 - 3.999  BH 8 +
• 4 - 4.999  HI No data
US EPA Region 4
                                                                                                  February 2007
         Predicted Blood-Lead Levels by County
                       2000
                                  Predicted Blood-Lead Levels by County
                                                2005
         LEGEND
  Geometric Mean PbB Level (ug/dL)
   m I - 1.999  • 5 - 5.999
       -2.999  M6-7.999
      3-3.999  H8 +
      4-4.999  M No data
US EPA Region 4
        LEGEND
Geometric Mean PbB Level (ug/dL)
CHI -1.999  •15-5.999
EH 2 - 2.999  mt 6 - 7.999
     -3.999  ••••:
   14-4.999  ^ No data
US EPA Region 4
                                                   G-7
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
                          Region 4 - Proportion of BLLs > 10
  Observed Proportion of Blood-Lead Levels >/= 10 ug/dL
                         2000
                             Observed Proportion of Blood-Lead Levels >/= 10 ug/dL
                                                    2005
            LEGEND
  Proportion of PbB Levels >/= I 0 Ug/dL
 l~"l 0 - 0.0099    Hi 0.075 - 0.0999
 LZZ 0.01 - 0.0249  MOJO- 0.2499
 C~\ 0.025  - 0.0499 M 0.25 +
 •I 0.05 - 0.0749  F~1 No data
US EPA Region 4
          LEGEND
 Proportion of PbB Levels >/= I 0 Ug/dL
 __ _ 0 - 0.0099     Bl 0.075 - 0.0999
IZZ! 0.01 - 0.0249   BHf 0.10 - 0.2499
 S 0.025 - 0.0499  Bl 0.25 +
 ,, j 0.05 - 0.0749      I No data
US EPA Region 4
  Predicted Proportion of Blood-Lead Levels >/= 10 ug/dL
                         2000
                             Predicted Proportion of Blood-Lead Levels >/= 10 ug/dL
                                                   2005
            LEGEND
  Proportion of PbB Levels >/= I 0 ug/dL
 I   , 0 - 0.0099    Bl 0.075 - 0.0999
 ZZ 0.01 - 0.0249  •• 0.10 - 0.2499
 LZH 0.025 - 0.0499 •• 0.25 +
 •i 0.05 - 0.0749  E3 No data
US EPA Region 4
          LEGEND
 Proportion of PbB Levels >/= 10 ug/dL
I   ! 0 - 0.0099     Bl 0.075 - 0.0999
I   0.01 - 0.0249   •ง 0.10 - 0.2499
i Z 0.025 - 0.0499  IB 0.25 +
•I 0.05 - 0.0749   SHU No data
US EPA Region 4
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA.  It does not represent and should not be construed
to represent any Agency determination or policy.

-------
                     Region 5 - Geometric Mean Blood Lead Levels
        Observed Blood-Lead Levels by County
                       2000
                                 Observed Blood-Lead Levels by County
                                               2005
         LEGEND
  Geometric Mean PbB Level (ug/dL)
   HI- 1.999   MS-5.999
       -2.999   IB 6-7.999
     3 - 3.999   •• 8 +
     4 . 4.999   $B No data
US EPA Region 5
       LEGEND
Geometric Mean PbB Level (ug/dL)
 ZJO- 1.999  MS-5.999
I   12 - 2.999  •ง 6 - 7.999
   3 - 3.999  •18 +
• 4 - 4.999  IB No data
US EPA Region 5
        Predicted Blood-Lead Levels by County
                       2000
                                 Predicted Blood-Lead Levels by County
                                               2005
         LEGEND
Geometric Mean PbB Level (ug/dL)
Z 1 - 1.999 Ml 5 -5.999
I : 2 - 2.999 •• 6 - 7.999
3 - 3.999 IB 8 +
•1 4 - 4.999 Bl No data
US EPA Region 5

Fobruary 2007
                                                                LEGEND

                                                        Geometric Mean PbB Level (ug/dL)
                                                            0- 1.999  MS-5.999
                                                            2 - 2.999
                                                             - 3.999
                                                            4 - 4.999
                                         I 6 - 7.999
                                         18 +
                                         H No data
                                                                                     US EPA Region 5
                                                  G-9
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
                          Region 5 - Proportion of BLLs > 10 ug/dL
  Observed Proportion of Blood-Lead Levels >/= 10 ug/dL
                         2000
                                          Observed Proportion of Blood-Lead Levels >/= 10 ug/dL
                                                                 2005
            LEGEND
   Proportion of PbB Levels >/= I 0 ug/dL
     0 - 0.0099    mm 0.075 - 0.0999
   HI 0.01 - 0.0249  Ki 0.10 - 0.2499
  E3 0.025 - 0.0499 • 0.25 +
  IH 0.05 - 0.0749  E3 No data
              US EPA Region 5
          LEGEND
 Proportion of PbB Levels >/= 10 ug/dL
I   10 - 0.0099     •• 0.075 - 0.0999
 H 0.01 - 0.0249   • 0.10 - 0.2499
   "0.025-0.0499  • 0.25+
fM 0.05 - 0.0749   UTI No data
US EPA Region 5
  Predicted Proportion of Blood-Lead Levels >/= 10 ug/dL
                        2000
                                          Predicted Proportion of Blood-Lead Levels >/= 10 ug/dL
                                                                2005
            LEGEND
  Proportion of PbB Levels >/= I 0 ug/dL
     0 - 0.0099    • 0.075 - 0.0999
     0.01 - 0.0249
    ; 0.025 - 0.0499
    I 0.05 - 0.0749
I O.IO-0.2499
I 0.25 +
! No data
                                US EPA Region 5
          LEGEND
 Proportion of PbB Levels >/= 10 ug/dL
I   ! 0 - 0.0099    • 0.075 - 0.0999
I   : 0.01 - 0.0249  Hi O.IO- 0.2499
 ~ 0.025 - 0.0499 • 0.25 +
IH 0.05 - 0.0749  IB No data
                                                                        US EPA Region 5
                                                     G-10
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
                     Region 6 - Geometric Mean Blood Lead Levels
         Observed Blood-Lead Levels by County
                       2000
                                 Observed Blood-Lead Levels by County
                                                2005
          LEGEND
  Geometric Mean PbB Level (ug/dL)
    ZO- 1.999  MS-5.999
  I   12 - 2.999  M 6 - 7.999
    _ 3 - 3.999  • 8 +
  H 4 - 4.999  B No data
US EPA Region 6
                                  LEGEND
                           Geometric Mean PbB Level (ug/dL)
                             Z I - 1.999  ••5-5.999
                           [	12 - 2.999
                           ZZ 3 - 3.999
                           •I 4 - 4.999
              I 6 - 7.999
              ] No data
                            US EPA Region 6
        Predicted Blood-Lead Levels by County
                       2000
                                 Predicted Blood-Lead Levels by County
                                                2005
          LEGEND
  Geometric Mean PbB Level (ug/dL)
    ^10- 1.999  MS-5.999
   ZZ 2 - 2.999  •• 6 - 7.999
     :3-3.999  Hi 8 +
   •• 4 - 4.999  B No data
US EPA Region 6
       LEGEND
Geometric Mean PbB Level (ug'dL)
  Zl- 1.999  ซ5-5.999
 CH 2 - 2.999  •ง 6 - 7.999
 CZ 3 - 3.999  • 8 +
 • 4-4.999  M No data
US EPA Region 6
                                                  G-ll
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
                          Region 6 - Proportion of BLLs > 10 jig/dL
   Observed Proportion of Blood-Lead Levels >/= 10 ug/dL
                          2000
             LEGEND
   Proportion of PbB Levels >/= 10 ug/dL
   HI 0 - 0.0099    •! 0.075 - 0.0999
      0.01 - 0.0249  Hi 0.10 - 0.2499
  d! 0.025 - 0.0499 Hi 0.25 +
  •10.05 - 0.0749  M No data
US EPA Region 6
                            Observed Proportion of Blood-Lead Levels >/= 10 ug/dL
                                                   2005
          LEGEND
 Proportion of PbB Levels >/= 10 ug/dL
 Zl 0 - 0.0099     •ง 0.075 - 0.0999
EH 0.01 - 0.0249   Ml 0.10 - 0.2499
EZ3 0.025 - 0.0499 Hi 0.25 +
•I 0.05 - 0.0749   QZ] No data
US EPA Region 6
   Predicted Proportion of Blood-Lead Levels >/= 10 ug/dL
                         2000
             LEGEND
   Proportion of PbB Levels >/= I 0 ug/dL
  r~ 0 - 0.0099    •ง 0.075 - 0.0999
  I   10.01 - 0.0249  •• 0.10- 0.2499
      0.025 - 0.0499 •ง 0.25 +
  •1 0.05 - 0.0749  F53 No data
US EPA Region 6
                            Predicted Proportion of Blood-Lead Levels >/= 10 ug/dL
                                                  2005
          LEGEND
 Proportion of PbB Levels >/= I 0 ug/dL
:   0 - 0.0099     HI 0.075 - 0.0999
I   10.01 - 0.0249   HI 0.10 - 0.2499
d 0.025 - 0.0499  •ง 0.25 +
•1 0.05 - 0.0749   IH No data
US EPA Region 6
                                                                                                        February 2007
                                                      G-12
TTz/.y information is distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA.  It does not represent and should not be construed
to represent any Agency determination or policy.

-------
                     Region 7 - Geometric Mean Blood Lead Levels
        Observed Blood-Lead Levels by County
                       2000
                                 Observed Blood-Lead Levels by County
                                                2005
         LEGEND
  Geometric Mean PbB Level (ug/dL)
   Z I - 1.999  •5-5.999
  I   : 2 - 2.999  M 6 - 7.999
  dj 3 - 3.999  M 8 +
  • 4 - 4.999  H No data
US EPA Region 7
        LEGEND
Geometric Mean PbB Level (ug/dL)
  Zl- 1.999  MS-5.999
 ZZ 2 - 2.999  mi 6 - 7.999
    3 - 3.999  mm 8 +
 • 4-4.999  •No data
US EPA Region 7
        Predicted Blood-Lead Levels by County
                       2000
                                 Predicted Blood-Lead Levels by County
                                                2005
         LEGEND
  Geometric Mean PbB Level (ug/dL)
    Zl- 1.999  MS-5.999
  (ZZ2-2.999  M6-7.999
  I   ! 3 - 3.999  M 8 +
  • 4 - 4.999  B No data
US EPA Region 7
       LEGEND
Geometric Mean PbB Level (ug/dL)
 CHI -1.999  MS-5.999
 CH2-2.999  ซ6-7.999
     -3.999  B18 +
   H-4.999  • No data
US EPA Region 7
                                                  G-13
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
                          Region 7 - Proportion of BLLs > 10 ug/dL
  Observed Proportion of Blood-Lead Levels >/= 10 ug/dL
                        2000
                            Observed Proportion of Blood-Lead Levels >/= 10 ug/dL
                                                  2005
            LEGEND
  Proportion of PbB Levels >/= I 0 ug/dL
  I] 0 - 0.0099    •• 0.075 - 0.0999
 CH 0.01 - 0.0249  M 0.10 - 0.2499
 CTJ 0.025 - 0.0499 •• 0.25 +
 tm 0.05 - 0.0749  I   I No data
US EPA Region 7
          LEGEND
 Proportion of PbB Levels >/= I 0 ug/dL
   0 - 0.0099     • 0.075 - 0.0999
I   : 0.01 - 0.0249   •• 0.10 - 0.2499
   0.025 - 0.0499  •• 0.25 +
•1 0.05 - 0.0749   B No data
US EPA Region 7
  Predicted Proportion of Blood-Lead Levels >/= 10 ug/dL
                        2000
                            Predicted Proportion of Blood-Lead Levels >/= 10 ug/dL
                                                   2005
            LEGEND
  Proportion of PbB Levels >l- I 0 ug/dL
     o - 0.0099    mm 0.075 - 0.0999
 rj o.o i - 0.0249  mm o. i o - 0.2499
 I   I 0.025 - 0.0499 • 0.25 +
 WM 0.05 - 0.0749  SB No data
US EPA Region 7
          LEGEND
 Proportion of PbB Levels >/= I 0 ug'dL
    0 - 0.0099     WM 0.075 - 0.0999
    0.01 - 0.0249   • 0.10 - 0.2499
El 0.025 - 0.0499 M 0.25 +
213 0.05 - 0.0749   I   I No data
US EPA Region 7
                                                     G-14
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
                     Region 8 - Geometric Mean Blood Lead Levels
        Observed Blood-Lead Levels by County
                       2000
                                  Observed Blood-Lead Levels by County
                                                2005
          LEGEND
  Geometric Mean PbB Level (ug/dL)
    Z I - 1.999  H5 -5.999
   CZl 2 - 2.999  •• 6 - 7.999
   LZJ 3 - 3.999 . ง• 8 +
   •I 4 - 4.999  SB No data
US EPA Region 8
        LEGEND
Geometric Mean PbB Level (ug/dL)
  Zl- 1.999  MS-5.999
 I  12 - 2.999  Wm 6 - 7.999
    3 - 3.999  •• 8 +
 •i 4 - 4.999  B No data
US EPA Region 8
         Predicted Blood-Lead Levels by County
                       2000
                                  Predicted Blood-Lead Levels by County
                                                2005
          LEGEND
  Geometric Mean PbB Level (ug/dL)
    Z I - 1.999  Hi 5 -5.999
  IZZ2-2.999  B6-7.999
    I 3-3.999  B8 +
  • 4 - 4.999  Bl No data
US EPA Region 8
        LEGEND
Geometric Mean PbB Level (ug/dL)
    I - 1.999  BBS -5.999
 CH2-2.999  B6-7.999
 CZj 3 - 3.999  •• 8 +
 •14-4.999  ^ No data
US EPA Region 8
                                                  G-15
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
                          Region 8 - Proportion of BLLs > 10 ng/dL
  Observed Proportion of Blood-Lead Levels >/= 10 ug/dL
                         2000
                             Observed Proportion of Blood-Lead Levels >/= 10 ug/dL
                                                   2005
            LEGEND
  Proportion of PbB Levels >/= 10 ug/dL
   ZJ 0 - 0.0099    Bl 0.075 - 0.0999
     0.01 - 0.0249  HI 0.10 - 0.2499
  I  I 0.025 - 0.0499 •ง 0.25 +
  •I 0.05 - 0.0749  E3 No data
US EPA Region 8
          LEGEND
 Proportion of PbB Levels >/= 10 ug/dL
dJ 0 - 0.0099     Bl 0.075 - 0.0999
   o.oi-0.0249   MO.IO-0.2499    US EPA Region 8
dj 0.025 - 0.0499  • 0.25 +
•I 0.05 - 0.0749   • No data
  Predicted Proportion of Blood-Lead Levels >/= 10 ug/dL
                         2000
                             Predicted Proportion of Blood-Lead Levels >/= 10 ug/dL
                                                   2005
            LEGEND
  Proportion of PbB Levels >/= 10 ug/dL
     0 - 0.0099    •• 0.075 - 0.0999
     0.01 - 0.0249  • 0.10 - 0.2499
 C   0.025 - 0.0499 •ง 0.25 +
 •1 0.05 - 0.0749  EH No data
US EPA Region 8
          LEGEND
 Proportion of PbB Levels >/= 10 ug/dL
[ -~ 0 - 0.0099     •! 0.075 - 0.0999
[ ~I 0.01 - 0.0249   M 0.10 - 0.2499
! Zl 0.025 - 0.0499  •! 0.25 +
   ] 0.05 - 0.0749   I   I No data
US EPA Region 8
                                                     G-16
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
                     Region 9 - Geometric Mean Blood Lead Levels
         Observed Blood-Lead Levels by County
                       2000
          LEGEND
  Geometric Mean PbB Level (ug/dL)
    Z> I - 1.999  ซB5 -5.999
       - 2.999
      3 - 3.999
      4 - 4.999
I 6 - 7.999
18 +
! No data
                               US EPA Region 9
Geometric Mean PbB Level (ug/dL)
I   iQ- 1.999  HIS - 5.999
HI]2-2.999  M6-7.999
Lr: 3-3.999  •H8
•I 4 - 4.999  LZD No data
         Predicted Blood-Lead Levels by County
                       2000
                                                Predicted Blood-Lead Levels by County
                                                               2005
         LEGEND
  Geometric Mean PbB Level (ug/dL)
    Z I - 1.999  MS -5.999
  I   12 - 2.999  •• 6 - 7.999
  CZ 3 - 3.999  • 8 +
  • 4 - 4.999  H No data
              US EPA Region 9
        LEGEND
 Geometric Mean PbB Level (ug/dL)
 HZ]0-1.999  ซ5-5.999
 LZJ 2 - 2.999  im 6 - 7.999
 IZJ 3 - 3.999  • 8 +
 ^ 4 - 4.999  HI No data
US EPA Region 9
                                                   G-17
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
                          Region 9 - Proportion of BLLs > 10 ug/dL
  Observed Proportion of Blood-Lead Levels >/= 10 ug/dL
                         2000
                             Observed Proportion of Blood-Lead Levels >/= 10 ug/dL
                                                   2005
            LEGEND
   Proportion of PbB Levels >/= 10 ug/dL
  I   10 - 0.0099    M 0.075 - 0.0999
  I   10.01  - 0.0249  M 0.10 - 0.2499
  ED 0.025 - 0.0499 Hi 0.25 +
  • 0.05  - 0.0749  13 No data
US EPA Region 9
          LEGEND
 Proportion of PbB Levels >/- 10 ug/dL
d " 0 - 0.0099     Bi 0.075 - 0.0999
CT 0.01 - 0.0249  IB 0.10 - 0.2499
33 0.025 - 0.0499 • 0.25 +
ง• 0.05 - 0.0749  • No data
US EPA Region 9
  Predicted Proportion of Blood-Lead Levels >/= 10 ug/dL
                         2000
                             Predicted Proportion of Blood-Lead Levels >/= 10 ug/dL
                                                   2005
            LEGEND
  Proportion of PbB Levels >/= I 0 ug/dL
  LH. 0 - 0.0099    H 0.075 - 0.0999
  di 0.01 • 0.0249  • 0.10 - 0.2499
  I  0.025 - 0.0499 M 0.25 +
  •i 0.05 - 0.0749  Hi No data
US EPA Region 9
          LEGEND
 Proportion of PbB Levels >/= ID ug/dL
LZ 0 - 0.0099     ง• 0.075 - 0.0999
LZi 0.01 - 0.0249   • 0.10 - 0.2499
[ ~. 0.025 - 0.0499  Hi 0.25 +
•• 0.05 - 0.0749
US EPA Region 9
                                                     G-18
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA. It does not represent and should not be construed
to represent any Agency determination or policy.

-------
                     Region 10 - Geometric Mean Blood Lead Levels
         Observed Blood-Lead Levels by County
                        2000
          LEGEND
  Geometric Mean PbB Level (ug/dL)
    ~ I - 1.999   MS-5.999
    II 2 - 2.999   •• 6 - 7.999
   [U 3 - 3.999   • 8 +
   mm4 - 4.999   EH No data
US EPA Region 10
                                  Observed Blood-Lead Levels by County
                                                 2005
        LEGEND
Geometric Mean PbB Level (ug/dL)
 L JO- I.999  M5-S.999
 LZj 2 - 2.999  mm 6 - 7.999
 d3 3 - 3.999  mm 8 +
 mm 4 - 4.999  B No data
US EPA Region 10
         Predicted Blood-Lead Levels by County
                       2000
                        190 360   720
         LEGEND
  Geometric Mean PbB Level (ug/dL)
    ~\-\.999  MS-5.999
   Z 2 - 2.999  mm 6 - 7.999
  ED 3 - 3.999  mm 8 +
  mm 4 - 4.999  mm No data
US EPA Region 10
                                  Predicted Blood-Lead Levels by County
                                                2005
                                                                               BS 370   740
       LEGEND
Geometric Mean PbB Level (ug/dL)
C~.IO- I.999  MS-5.999
L_J 2 - 2.999  mm 6 - 7.999
n 3 - 3.999  mm 3 +
mm 4 - 4.999  mm No data
US EPA Region 10
                                                  G-19
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA.  It does not represent and should not be construed
to represent any Agency determination or policy.

-------
                         Region  10 - Proportion of BLLs  > 10
  Observed Proportion of Blood-Lead Levels >/= 10 ug/dL
                         2000
                          B5 370    740
            LEGEND
  Proportion of PbB Levels >/= I 0 ug/dL
  P7T 0 - 0.0099    • 0.075 - 0.0999
     0.01 -0.0249  •• O.I0-0.2499
  CIJ 0.025 - 0.0499 •• 0.25 +
  •I 0.05 - 0.0749  ED No data
US EPA Region 10
                             Observed Proportion of Blood-Lead Levels >/= 10 ug/dL
                                                    2005
                                                                                     85 370   740
          LEGEND
 Proportion of PbB Levels >/= 10 ug/dL
d 0 - 0.0099     H 0.075 - 0.0999
CHi 0.01 - 0.0249   K 0.10 - 0.2499
CH 0.025 - 0.0499  !• 0.25 +
•I 0.05 - 0.0749   US No data
US EPA Region 10
  Predicted Proportion of Blood-Lead Levels >/= 10 ug/dL
                         2000
                       0 185 370    740
            LEGEND
  Proportion of PbB Levels >/= 10 ug/dL
     0 - 0.0099    •• 0.075 - 0.0999
 I   10.01 - 0.0249  •• 0.10 - 0.2499
     0.025 - 0.0499 • 0.25 +
 •I 0.05 - 0.0749  EC3 No data
US EPA Region 10
                             Predicted Proportion of Blood-Lead Levels >/= 10 ug/dL
                                                    2005
                                       LEGEND
                             Proportion of PbB Levels >/= I 0 ug/dL
                            ^0-0.0099    •0.075-0.0999
                            CJO.OI -0.0249
                            LIJ 0.025 - 0.0499
                            • 0.05 - 0.0749
                 I O.I0-0.2499
                 I 0.25 +
                 1 No data
                               US EPA Region 10
                                                      G-20
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information
quality guidelines.  It has not been formally disseminated by EPA.  It does not represent and should not be construed
to represent any Agency determination or policy.

-------
                                      Massachusetts - Geometric Mean Blood Lead Levels
               Observed Blood-Lead Levels by Census Tract
                                2000
            LEGEND
    Geometric Mean PbB Level (ug/dL)
         I - 1.999  MIS- 5.999
         2 - 2.999  mm 6 - 7.999
         3 - 3.999  mm 8 +
         4 - 4.999  I  I No data
                           Massachusetts
               Predicted Blood-Lead Levels by Census Tract
                                 2000
            LEGEND
    Geometric Mean PbB Level (ug/dL)
         I - 1.999  •15-5.999
         2 - 2.999  mm 6 - 7.999
         3 - 3.999  mm 8 +
     •N-4.999  EU No data
                           Massachusetts
        LEGEND
Geometric Mean PbB Level (ug/dLj
    I - 1.999   M15-5.999
    2 - 2.999   mm 6 - 7.999
    3 - 3.999
 (HI 4 - 4.999   EH No data
                                                                                                    Massachusetts
        LEGEND
Geometric Mean PbB Level (ug/dL)
    I - 1.999   mm 5- S.999
    2 - 2.999   mm 6 - 7.999
    3 - 3.999   mm a +
   J 4 - 4.999   i"~l No data
                                                                                                    Massachusetts
                                                                       G-21
This information is distributed solely for the purpose ofpre-dissemination peer review under applicable information quality guidelines.  It has not been formally
disseminated by EPA. It does not represent and should not be construed to represent any Agency determination or policy.

-------
                                            Massachusetts - Proportion of BLLs > 10
                LEGEND
       Proportion of PbB Levels >/= 10
         0 - 0.0099    BB 0.075 - 0.0999
         0.01 -0.0249  Mi O.I 0-0.2499
         0.025 - 0.0499 •• 0.25 +
       Z3 0.05 - 0.0749  rn No data
                             Massachusetts
                LEGEND
       Proportion of PbB Levels >t~ 10 ug/dL
         0 - 0.0099    HI 0.075 - 0.0999
         0.01 - 0.0249  •• 0.10 - 0.2499
         0.025 - 0.0499 •• 0.25 +
         O.OS - 0.0749   HI No data
                             Massachusetts
                                                                                         Observed Proportion of Blood-Lead Levels >/= 10 ug/dL
                                                                                                               2005
-
 Proportion of PbB Levels >/= 10 ug/dL
   0 - 0.0099    •• 0.075 - 0.0999
   0.01 -0.0249  Wm O.I 0-0.2499
   0.025 - 0.0499 •• 0.25 +
   0.05 - 0.0749  i~	I No data
                       Massachusetts
 Proportion of PbB Levels >J= 10 ug/dL
   0 - 0.0099    •• 0.075 - 0.0999
   0.01 - 0.0249  •• 0.10 - 0.2499
   0.025 - 0.0499 •• 0.25 +
   I 0.05 - 0.0749  C_j No data
                       Massachusetts
                                                                            G-22
Thin information /.y distributed solely for the purpose of pre-dissemination peer review under applicable information quality guidelines. It has not been formally
disseminated by EPA. It does not represent and should not be construed to represent any Agency determination or policy.

-------
                                       United States - Geometric Mean Blood Lead Levels
                   Observed Blood-Lead Levels by County
                                  2000
                                  Observed Blood-Lead Levels by County
                                                 2005
             LEGEND
     Geometric Mean PbB Level (ug/dL)
         0- 1.999   ma 5- 5.999
         2 - 2.999   BB 6 - 7.999
         3-3.999   mm 8 +
      Om 4 - 4.999   [1J No data
105 210  43)   630
                             United States
        LEGEND
Geometric Mean PbB Level (ug/dL)
    0- 1.999   Om 5 -5.999
    2 - 2.999   mm 6 - 7.999
    3 - 3.999   mm & +
    4 - 4.999   I  I No data
                   Predicted Blood-Lead Levels by County
                                  2000
                                  Predicted Blood-Lead Levels by County
                                                 2005
             LEGEND
     Geometric Mean PbB Level (ug/dL)
         0- 1.999   mm S -5.999
         2 - 2.999   mm 6 - 7.999
         3 - 3.999   mm a +
      •14-4.999   L_ No data
                             United States
                             LEGEND
                     Geometric Mean PbB Level (ug/dL)
                         0- 1.999  MIS- 5.999
                         2 - 2.999  mm 6 - 7.999
                         3 - 3.999  mm 8 +
                      ••4-4.999  LU No data
                                            United States
                                                                        G-23
This information is distributed solely for the purpose of pre-dissemination peer review under applicable information quality guidelines. It has not been formally
disseminated by EPA. It does not represent and should not be construed to represent any Agency determination or policy.

-------
                                             United States - Proportion of BLLs > 10 ng/dL
             Observed Proportion of Blood-Lead Levels >/= 10 ug/dL
                                    2000
          Observed Proportion of Blood-Lead Levels >/= 10 ug/dL
                                 2005
             LEGEND
    Proportion of PbB Levels >/= 10 ug/dL
      0 - 0.0099    •ง 0.075 - 0.0999
      0.01 - 0.0249  •• 0.10 - 0.2499
      0.025 - 0.0499 • 0.25 +
    13 0.05 - 0.0749   	I No data
                              United States
             LEGEND
    Proportion of PbB Levels >/= 10 ug/dL
      0 - 0.0099    HI 0.075 - 0.0999
      0.01  - 0.0249  mi 0.10- 0.2499
      0.025 - 0.0499 mt 0.25 +
    • 0.05 - 0.0749  i   I No data
                              United States
          LEGEND        ^^ป
 Proportion of PbB Levels >/= 10 ug/dL
    0 - 0.0099     mt 0.075 - 0.0999
    0.01 - 0.0249  •• 0.10 - 0.2499
    0.025 - 0.0499 •• 0.25
EZ3 0.05 - 0.0749  L	i No data
                            United States
                                                                                          Predicted Proportion of Blood-Lead Levels >/= 10 ug/dL
                                                                                                                 2005
          LEGEND
 Proportion of PbB Levels >/= 10 ug/dL
   0 - 0.0099     BB 0.075 - 0.0999
   0.01 - 0.0249   •• 0.10 - 0.2499
   0.025 - 0.0499  BH 0.25 +
Cll 0.05 - 0.0749   L	I No data
                           United States
                                                                            G-24
This information is distributed solely for the purpose ofpre -dissemination peer review under applicable information quality guidelines.  It has not been formally
disseminated by EPA. It does not represent and should not be construed to represent any Agency determination or policy.

-------
                      Appendix H



Data Dictionaries for National and Massachusetts Databases

-------
                               PILOT STUDY OF TARGETING ELEVATED BLOOD-LEAD LEVELS IN CHILDREN
                                                 National Blood Lead Data Aggregated to the County Level
                                                                        Data Dictionary
                                                                        March 2, 2007
   # Field Name
Data Type
Size  Description
Primary KeySecondary
Table: National_Analysis_Dataset:  Dataset used for analysis purposes that contains data merged from all data sources.
   1 County_FIPS
   2 Water
   3 State
   4 time
   5 county
   6 ntot
   7 P5
   8 p10
   9 p15
  10 p25
  11 gm
  12 am
  13 flag
  14 Ipbw
  15 TotaLPop
  16 Pct_White
  17 PcLBIack
  18 Pct_AIANAP
  19 Pct_Asian
  20 Pct_NHOPI
  21 Pct_Other_Race
  22 Pct_Multi_Race
  23 Pct_Hispanic
  24 Pct_LE_Six
  25 Pct_Smgle_Parent
  26 Median_HH_lncome
  27 Pct_HH_No_Earnings
  28 Pct_HH_No_Wage
  29 Pct_HH_Pubhc_Assist
  30 Median_Family_lncome
  31 Median_Per_Capita_lncome
  32 Pct_LT_Poverty
  33 Pct_LE_5Yrs_LT_Poverty
  34 Pct_Family_lncome_LT_Poverty
  35 Pct_LT_9th_Grade
Text
Number, Long Integer
Text
Number, Double
Text
Text
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Integer
Number, Double
Number, Long Integer
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Long Integer
Number, Long Integer
Number, Double
Number, Double
Number, Double
Number, Double
255
4
255
8
255
255
8
8
8
8
8
8
2
8
4
8
8
8
8
8
8
8
8
8
8
8
8
8
8
4
4
8
8
8
8
                         Concatenation of State & County FIPS numbers to create unique identifier for a
                         SDWIS. Indicator variable for County Level Monitoring
                         US Census State FIPS number
                         Time category for adjusting HUD Funding trends
                         US Census County FIPS number
                         CDC: Total Number of Children Screened for PbB
                         CDC: Percent of Children Observed with PbB > = 5 ug/dL
                         CDC: Percent of Children Observed with PbB > = 10 ^g/dL
                         CDC: Percent of Children Observed with PbB > = 15 ug/dL
                         CDC: Percent of Children Observed with PbB > = 25 ug/dL
                         Geometric Mean
                         Arithmetic Mean
                         0/1 Indicator regarding Blood-lead data quality
                         EPA SDWIS. Log Transformed Average 90th percentile values
                         Total US Population
                         Percent US Population where race is Caucasian
                         Percent US Population where race is African-American
                         Percent US Population where race is American Indian/Alaska Native alone
                         Percent US Population where race is Asian
                         Percent US Population where race is Native Hawaiian and Other Pacific Islander
                         Percent US Population where race is Some Other Race alone
                         Percent US Population where race is Two or More Races
                         Percent US Population where ethnicity is Hispanic
                         Percent US Population who are less than or equal to six years of age
                         Percent US Households with Single Parents
                         Median Household income
                         Percent US Households with no earnings
                         Percent US Households with no wages
                         Percent US Households on Public Assistance
                         Median Family Income
                         Median Per Capita Income
                         Percent US Population living below poverty level
                         Percent US Population less than or equal to five years of age living below poverty
                         Percent Households with family income less than poverty level
                         Percent of US Population with less than a 9th grade education
                                                                              Yes (Duplicates OK)
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     NO
                                                                                     No
                                                                            Page H-l

-------
                                PILOT STUDY OF TARGETING ELEVATED BLOOD-LEAD LEVELS IN CHILDREN
                                                  National Blood Lead Data Aggregated to the County Level
                                                                          Data Dictionary
                                                                           March 2, 2007
   # Field Name
Data Type
Size  Description
Primary KeySecondary
Table: National_Analysis_Dataset:  Dataset used for analysis purposes that contains data merged from all data sources.
  36 Pct_No_HS_Degree
  37 Pct_No_College
  38 Pct_No_College_Degree
  39 Total_Housing_Units
  40 Pct_Vacant
  41 Pct_Rented
  42 Pct_Built_Pre_1940
  43 Pct_Built_Pre_1950
  44 Pct_Built_Pre_1960
  45 Pct_Built_Pre_1970
  46 Pct_Built_Pre_1980
  47 Median_Yr_Built
  48 Pct_Occ_Built_Pre_1940
  49 Pct_Occ_BuiltJ>re_1950
  50 Pct_Occ_Built_Pre_1960
  51 Pct_Occ_Built_Pre_1970
  52 Pct_Occ_Built_Pre_1980
  53 Median_Yr_Occ_Built
  54 Median_Rent
  55 Housmg_Value
  56 tri_at1
  57 tn_af1
  58 tn_as1
  59 tn_ws1
  60 tri_ui1
  61 tn_at2
  62 tn_af2
  63 tn_as2
  64 tri_ws2
  65 tn_ui2
  66 tri_at3
  67 tn_af3
  68 tri_as3
  69 tn_ws3
  70 tn_ui3
  71 air_avg
  72 air med
Number, Double
Number, Double
Number, Double
Number, Long Integer
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Long Integer
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Long Integer
Number, Double
Number, Long Integer
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double
8
8
8
4
8
8
8
8
8
8
8
4
8
8
8
8
8
4
8
4
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
                          Percent of US Population with no high school degree
                          Percent of US Population with no college experience
                          Percent of US Population with no college degree
                          Total number US housing units
                          Percent US Housing units vacant
                          Percent US Housing units rented
                          Percent US Housing units built before 1940
                          Percent US Housing units built before 1950
                          Percent US Housing units built before 1960
                          Percent US Housing units built before 1970
                          Percent US Housing units built before 1980
                          Median year units built
                          Percent Occupied US Housing units built before 1940
                          Percent Occupied US Housing units built before 1950
                          Percent Occupied US Housing units built before 1960
                          Percent Occupied US Housing units built before 1970
                          Percent Occupied US Housing units built before 1980
                          Median Year Occupied US Housing units built
                          Median rent
                          Housing value
                          TRI: Total Air Emissions of Lead Only (Ibs)
                          TRI: Fugitive Air Emissions of Lead Only (Ibs)
                          TRI. Air Stack Emissions of Lead Only (Ibs)
                          TRI: Surface Water Emissions of Lead Only (Ibs)
                          TRI: Underground Injection of Lead Only (Ibs)
                          TRI- Total Air Emissions of Lead Containing Compounds (Ibs)
                          TRI: Fugitive Air Emissions of Lead Containing Compounds (Ibs)
                          TRI. Air Stack Emissions of Lead Containing Compounds (Ibs)
                          TRI. Surface Water Emissions of Lead Containing Compounds (Ibs)
                          TRI: Underground Injection of Lead Containing Compounds (Ibs)
                          TRI- Total Air Emissions of Total Lead (Ibs)
                          TRI: Fugitive Air Emissions of Total Lead (Ibs)
                          TRI- Air Stack Emissions of Total Lead (Ibs)
                          TRI: Surface Water Emissions of Total Lead  (Ibs)
                          TRI. Underground Injection of Total Lead (Ibs)
                          1999 EPA NATA. Average Air Pb Concentration
                          1999 EPA NATA: Median Air Pb Concentration
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        NO
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                                        No
                                                                              Page H-2

-------
                                PILOT STUDY OF TARGETING ELEVATED BLOOD-LEAD LEVELS IN CHILDREN
                                                 National Blood Lead Data Aggregated to the County Level
                                                                        Data Dictionary
                                                                         March 2, 2007
   # Field Name
Data Type
Size  Description
Primary KeySecondary
Table: National_Analysis_Dataset: Dataset used for analysis purposes that contains data merged from all data sources.
  73 air_p95
  74 numkids
  75. hud_cum
  76 cdc_cum
  77 hud_cur
  78 cdc_cur
  79 quarter
  80 t_cat
  81 tot_cur
  82 tot_cum
  83 region
  84 h2o_p95
  85 h2o_p99
  86 air_med_p95
  87 air_med_p99
  88 tn_af1_p95
  89 tri_af1_p99
  90 tn_ui3_p95
  91 tri_ui3_p99
  92 tri_as1_p95
  93 tn_as1_p99
  94 tn_ui1_p95
  95 tn_ui1_p99
  96 high_airpb
Number, Double          8  1999 EPA NATA. 95th Percentile Air Pb Concentration
Number, Long Integer      4  CDC Total Number of Children in the county
Number, Double          8  Cumulative Funding from HUD Allocated to Counties
Number, Double          8  Cumulative Funding from CDC Allocated to Counties
Number, Double          8  Current Funding from HUD Allocated to Counties
Number, Double          8  Current Funding from CDC Allocated to Counties
Number, Integer          2  Quarter term used for seasonality adjustments
Number, Integer          2  Time Category for Adjusting HUD Funding Trends
Number, Double          8  Total Current Funding (HUD+CDC) Allocated to Counties
Number, Double          8  Total Cumulative Funding (HUD+CDC) Allocated to Counties
Text                  255  Categorical Variable representing 10 EPA Regions
Number, Double          8  EPA SDWIS: Indicator >= 95th percentile of Water Pb Concentration
Number, Double          8  EPA SDWIS: Indicator >= 99th percentile of Water Pb Concentration
Number, Double          8  1999 EPA NATA. Indicator >= 95th percentile of  Median Air Pb Concentration
Number, Double          8  1999 EPA NATA1 Indicator >= 99th percentile of  Median Air Pb Concentration
Number, Double          8  TRI: Indicator >= 95th percentile of Lead  Only Fugitive Air Emissions
Number, Double          8  TRI- Indicator >= 99th percentile of Lead  Only Fugitive Air Emissions
Number, Double          8  TRI. Indicator >= 95th percentile of Total  Lead Underground Injection
Number, Double          8  TRI- Indicator >= 99th percentile of Total  Lead Underground Injection
Number, Double          8  TRI- Indicator >= 95th percentile of Lead  Only Air Stack Emissions
Number, Double          8  TRI: Indicator >= 99th percentile of Lead  Only Air Stack Emissions
Number, Double          8  TRI- Indicator >= 95th percentile of Lead  Only Underground Injection
Number, Double          8  TRI: Indicator >= 99th percentile of Lead  Only Underground Injection
Number, Double          8  1999 EPA NATA: Indicator for 20 Counties with High Pb Concentration
                                                                                      No
                                                                               Yes (Duplicates OK)
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                             Page H-3

-------
                                PILOT STUDY OF TARGETING ELEVATED BLOOD-LEAD LEVELS IN CHILDREN
                                                 National Blood Lead Data Aggregated to the County Level
                                                                         Data Dictionary
                                                                         March 2, 2007
   # Field Name
Data Type
             Size  Description
                                                                    Primary KeySecondary
Table: tblCensus2000County: US Census Summary File 3 data.
   1 COUNTY_FIPS
   2 Total_Pop
   3 Pct_White
   4 Pct_Black
   5 Pct_AIANA
   6 Pct_Asian
   7 Pct_NHOPI
   8 Pct_Other_Race
   9 Pct_Multi_Race
  10 Pct_Hispanic
  11 Pct_LE_Six
  12 Num_LE_Six
  13 Pct_Smgle_Parent
  14 Median_HH_lncome
  15 Pct_HH_No_Earnings
  16 Pct_HH_No_Wage
  17 Pct_HH_Pubhc_Assist
  18 Median_Family_lncome
  19 Median_Per_Capita_lncome
  20 Pct_LT_Poverty
  21 Pct_LE_5Yrs_LT_Poverty
  22 Pct_Family_lncome_LT_Poverty
  23 Pct_LT_9th_Grade
  24 Pct_No_HS_Degree
  25 Pct_No_College
  26 Pct_No_College_Degree
  27 Total_Housing_Units
  28 Pct_Vacant
  29 Pct_Rented
  30 Pct_Built_Pre_1940
  31 Pct_Built_Pre_1950
  32 Pct_Built_Pre_1960
  33 Pct_Built_Pre_1970
  34 Pct_Built_Pre_1980
  35 Median_Yr_Built
Text
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Double
Double
Double
Double
Double
Double
Double
Double
Double
Double
Double
Double
Double
Double
Double
Double
Double
Double
Double
Double
Double
Double
Double
Double
Double
Double
Double
Double
Double
Double
Double
Double
Double
Double
5  Concatenation of State & County FIPS numbers to create unique identifier
8  Total Population
8  Percent Population where race is Caucasian
8  Percent Population where race is African-American
8  Percent Population where race is American Indian/Alaska Native alone
8  Percent Population where race is Asian
8  Percent Population where race is Native Hawaiian and Other Pacific Islander
8  Percent Population where race is Some Other Race alone
8  Percent Population where race is Two or More Races
8  Percent Population where ethnicity is Hispanic
8  Percent Population who are less than or equal to six years of age
8  Number Population who are less than or equal to six years of age
8  Percent Households with Single Parents
8  Median Household income
8  Percent Households with no earnings
8  Percent Households with no wages
8  Percent Households on Public Assistance
8  Median Family Income
8  Median Per Capita Income
8  Percent Population living below poverty level
8  Percent Population less than or equal to five years of age living below poverty
8  Percent Households with family income less than poverty level
8  Percent Population with less than a 9th grade education
8  Percent Population with no high school degree
8  Percent Population with no college experience
8  Percent Population with no college degree
8  Total number housing units
8  Percent Housing units vacant
8  Percent Housing units rented
8  Percent Housing units built before 1940
8  Percent Housing units built before 1950
8  Percent Housing units built before 1960
8  Percent Housing units built before 1970
8  Percent Housing units built before 1980
8  Median year units built
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
                                                                             Page H-4

-------
                              PILOT STUDY OF TARGETING ELEVATED BLOOD-LEAD LEVELS IN CHILDREN
                                                                                                                         ^
                                               National Blood Lead Data Aggregated to the County Level
                                                                     Data Dictionary
                                                                      March 2,2007
   # Field Name
Data Type
Size Description
Primary KeySecondary
Table: tblCensus2000Couhty: US Census Summary File 3-data.
  36 Pct_Occ_Built_Pre_1940
  37 Pct_Occ_Built_Pre_1950
  38 Pct_Occ_Built_Pre_1960
  39 Pct_Occ_BuiltiPre_1970
  40 Pct_Occ_Built_Pre_1980
  41 Median_YrJDcc_Built
  42 Median_Rent .
  43 Housing_Value
  44 Housing_Density

  45 statej_numje_six
  46 pct_le_six_by_state'
  47 REGION
  48 STATE
  49. COUNTY
  50 NAME
  51 INTPTllAT
  .52 INTPJLON
  53.SQMILE
Number, Double
Number, Double
Number, Double
Number, Double-
Number, Double
Number, Double
Number, Double
Number, Double
Number, Double

Number, Double
Number, Double
Text-
Text
Text
Text
Text
Text
Number, Double
   8  Percent Occupied Housing units built before 1940
   8  Percent Occupied Housing units built before 1950 .
   8  Percent Occupied Housing units built before'1960
   '8 . Percent Occupied Housing units built'before 1970
   8  Percent Occupied Housing units built before 1980
   8  Median Year Occupied Housing units built
   8  Median rent
   8  Housing value
   8  Housing Density (calculated variable. Housing_Density = Total_H6using_Units /
     SQMILE)
   8  Number Population less than or equal to six years of age for applicable state
   8  Percent Population less than or equal to six years of age for applicable state
   1  US Census Region
   2  State FIPS
   3  County FIPS '
  90  Name of County,
   9  Latitude of Interior Point
  10  Longitude of Interior Point
   8  Square miles of County
               No
               No
               No
               No
               No
               No
               No
               No
               No'

               -No
               No
               No
               No
               No
               No
               No
               No
               No
                                                                         Page H-5

-------
                              PILOT STUDY OF TARGETING ELEVATED'BLOOD-LEAD LEVELS IN CHILDREN
                                              National Blood Lead Data Aggregated to the County Level
                                                                    Data Dictionary
                                                                     March 2,2007
   # Field Name
 Data Type
Size Description
Primary KeySecbndary
Table: tbIDrinkingWaterData:  EPA National Review of LCR Implementation and Drinking Water Lead Reduction Plan Data
   1  COUNTY_FIPS
   2  STATE
   3  COUNTY
   4  PWSID
  '5  PWSName
   6  EPA_Region
   7  Geographyjype
   8  State_Abbrev
   9  CountyJMame
  10  PWSType
  11  Gw_Sw
  12  PSource
  13  SizeCat3_pt
  14  SizeCat11_pt
  15  Pop-
  16  CYMonPerBegin
  17'MonPerBegin
  18  MonPerEnd
  19, Measure
  20  AboveAL
 Text
 Text
 Text
 Text
 Text
 Text
 Text
 Text
 Text
 Text
 Text
 Text
 Text
 Text
 Number, Long Integer
 Number, Integer
 Date/Time
 Date/Time.
. Number, Double
 Number.'Byte  -
   5  Concatenation of State & County FIPS numbers to create unique identifier for a
   2  -US Census State FIPS number
   3  US Census County FIPS number
  15  Water Source ID
  50  Water Source Name
   2  EPA Region
   5  Geography Type
   2  State Abbreviation
  50  County Name
  10  Primary Water Source Type
   2  Primary Source of Water Indicator
   3  Primary Source of Water
  50  Size category, based on average daily retail population served:
  50  Size category, based on average daily retail population served:
   4  Population
   2  Calendar Year Monitoring Period Began
   8  Monitoring. Period Began
   8  Monitoring Period Ended
   8  Measure of Lead in Water
   1.  Above Action Level
        Yes (Duplicates OK)
               No
               No
               No
               No
               No
               No
               No
               No
               No
               No
               No
               No
               No
               No
               No
               No
               No
               No
               No
Table: tblNATA1999AmbientPb_State_County: 1999 EPA NATA Data Ambient Lead data by county
   1 County_Fips
   2 air_avg
   3 air_med
   4 air_p95
 Text
 Number, Double
 Number, Double
 Number, Double
   5 Concatenation of State & County FIPS numbers to create unique identifier
   8 Average Air Pb Concentration
   8 Median Air Pb Concentration
   8 95th Percentile Air Pb Concentration
               No
               No
               No
               No
                                                                         Page H-6

-------
                              PILOT STUDY OF TARGETING ELEVATED BLOOD-LEAD LEVELS IN CHILDREN
                                               National Blood Lead Data Aggregated to the County Level
                                                                     Data Dictionary
                                                                      March 2,2007
   # Field Name
Data Type
            Size  Description
Primary KeySecondary
Table: tbINationalDataByCountyFIPS:  National Blood Lead Level data from. CDC.
   1 COUNTY_FIPS
   2 Year
   3 Quarter
   4 StateAbbrev
   5 State
   6 County
   7 Total
   8 PercentGS
   9 PercentGIO
  10 PercentG15
  11 PercentG25
  12 GeometricMean
  13 ArithmeticMean
Text
Number,
Number,
Text
Text
Text
Number,
Number,
•Number,
Number,
'Number,
Number,
Number,
               5 ' Concatenation of State & County PIPS numbers to create unique identifier for a
Long Integer     4 Sampling year
Long Integer     4 Sampling quarter (1 = Jan - Mar; 2=Apr-Jun; 3=Jul-Sep; 4=Oct-Dec)
              50 US Postal Service state abbreviation
               2 US Census State PIPS number
              50 US Census County PIPS number
Long Integer     4 Total Number of Children Screened for PbB
Double          8 Percent of Children Observed with PbB > = 5 ug/dL
Double          8 Percent of Children Observed with PbB > = 10 ug/dL
Double          8 Percent of Children Observed with PbB > = 15 ug/dL
Double          8 Percent of Children Observed with PbB > = 25 ug/dL
Double          8 GeometricMean
Double          8 Arithmetic Mean
        Yes (Duplicates OK)
               No
               No
               No
               No
               No
               No
               No
               No
               No
               No
               No
               No
Table: tblNationalFunding_CDC: Funding Data Provided by CDC and Allocated Across Counties
   1  COUNTY_FIPS
   2  Funding_Year
   3  Funding
   4  Funding_Source
Text
Number, Integer
Number, Long Integer
Number, Integer
               5 Concatenation of State & County FIPS numbers to create unique identifier for a
               2 Year'of Funding from CDC
               4 Amount of Funding from CDC
               2 Source of Funding from CDC (0=no funding; 1 =county-level funding; 2= state-level
                  funding)
        Yes (Duplicates OK)
               No
               No
               No
Table: tblNationalFunding_HUD: Funding Data Provided by HUD and Allocated Across Counties
   1  COUNTY.FIPS
   2  Fundmg_Year
   3  Funding
   4  Fiinding_Source
Text
Number, Integer
Number, Long Integer
Number, Integer
               5 Concatenation of State & County FIPS numbers to create unique identifier for a
               2 Year of Funding from HUD
               4 Amount of Funding from HUD
               2 Source of Funding from HUD (0=no funding; 1=county-level funding; 2= state-level
                  funding)
        Yes (Duplicates OK)
               No
               No
               No-
                                                                         Page H-7

-------
                            PILOT STUDY OF TARGETING ELEVATED BLOOD^LEAD LEVELS IN CHILDREN
                                           National Blood Lead Data Aggregated to the County Level
                                                                Data Dictionary
                                                                March 2,2007
  # Field Name
Data Type
Size Description
Primary KeySecondary
Table: tblTRI_Data: Toxics Release Inventory Data for Lead and Lead Components only (1998- 2004).
   1 TRIFID                             Text
   2 COUNTY FIPS                       Text
   3 REPORTING_YEAR                    Number, Integer
   4 CHEMICALJMAME                     Text
   5 CAS NUMBER                       Text
   6 TOTAL_AIR_EMISSIONS                Number, Double
   7 TCTAL_FUG_AIR EMISS                Number, Double
   8 TOTAL NUMBER_RECEIVING_STREAMS    Number, Byte'
   9 TOTAL_STACK_AIR EMISS              Number, Double
  10 TOTAL_SURFACE WATER_DISCHARGE    Number,'Double
  11 TOTAL UNDERGROUNDJNJECTION       Number, Double
  12 UNIT_OF_MEASURE                   Text
                   20  Toxics Release Inventory Facility ID
                    5  Concatenation of State & County FIPS numbers to create unique identifier
                    2  Reporting Year
                   100  Name of Chemical ("Lead", 'Lead Compounds')
                   15  CAS Number
                    8  Total Air Emissions
                    8  Total Fugitive Air Emissions,
                    1  Total Number Receiving Streams
                    8  Total Stack Air Emissions
                    8  Total Surface Water Discharge
                    8  Total Underground Injection
                    .6  Unit of Measure (pounds)
                                                                            No
                                                                      Yes (Duplicates OK)
                                                                            No
                                                                            No
                                                                            No
                                                                            No
                                                                            No
                                                                            No
                                                                            No
                                                                            No
                                                                            No
                                                                            No
                                                                    PageH-8

-------
                                  PILOT STUDY OF TARGETING ELEVATED BLOOD-LEAD LEVELS IN CHILDREN
                                                      National Blood Lead Data Aggregated to the County Level
                                                                                   Data Diagram
                                                                                  March 2, 2007
tWCensuslOOOCounty
COUNTY_FIPS
Total.Pop
Pct_white
Pct_Black
PctJUANA
Pct_Asan
Pct_NHOPI
PcM3ther_Race
Pct_Multi_Race
Pct_Hisparuc
PctJ.E_Six
Num_LE_Six
Pct_Smgle_Parent
Median_HH_Income
Pct_HH_No_Earnmgs
Pct_HH_No_Wage
Pct_HH_Pubhc_Assist
Median_Family_Incoine
Median_Per_Capita_Income
Pct_LT_Poverty
Pct_LE_5Yrs_LT_Poverty
Pct_Family_Income_LT_Poverty
Pet J.T_9th_Grade ~
Pct_No_HS_Degree
Pct_No_College
Pct_No_College_Degree
Total_Housng_Uruts
Pct.Vacant
Pct_Rented
Pct_Built_Pre_19-W
Pct_Built_Pre_1950
Pct_5uilt_Pre_1960
Pct_Buili_Pre_1970
Pct_Built_Pfe_1980
Median_Yr_Built
Pct_Occ_Built_Pre_1940
Pct_Oซ_Built_Pre_1950
Pct_Occ_Built_Pre_1960
Pct_Occ_Built_Pre_1970
Pct_Occ_Built_Pre_1980
Median_Yr_Occ_Built
Median.Rent
Housmg_Value
Housing_Density
state_num_le_six
pct_le_six_by_state
REGION
STATE
                              J
                                                                                                tblDrinkingWaterOata
                                                                                                                                            National_Analysis_Dataset
                                                 COUNTY.FIPS
                                                 REPORTING.YEAR
                                                 CHEMICAL.NAME
                                                 CAS.NUMBER
                                                 TOTAL_AIR_EMISSIONS
                                                 TOTAL_FUG_AIR_EMISS
                                                 TOTAL .NUMBER _RECBVING_STREAT"1S
                                                 TOTAL_STACK_AIR_EMISS
                                                 TOTAL_SURFACE_\VATER_OISCHARGE
                                                 TOTAL_UNDERGROUND_INJECTION
                                                 UNrT_OF_MEASURE
COUNTY.FIPS
STATE
COUNTY
PWSID
PWSMame
EPA_Region
Geography_type
State _Abbrev
County_Name
PWSType
Gw_Sw
PSource
SizeCat3j>t
SizeCatll_pt
Pop
CYMonPerBegin
MonPerBegm
MonPerEnd
Measure
AboveAL
                                                                                               tbIMationalOataByCountyFIPS
                                                                                               OOUNTY_FIPS
                                                                                               Year
                                                                                               Quarter
                                                                                               StateAbbrev
                                                                                               State
                                                                                               County
                                                                                               Total
                                                                                               PercentGS
                                                                                               PercentGlO
                                                                                               PercentGIS
                                                                                               PercentG25
                                                                                               GeometncMean
                                                                                               AnthmetacMean
County_FIPS
Water
State
tame
county
ntot
PS
plD
PIS
p25
gm
am
flag
Ipbw
Total.Pop
Pct.White
Pct.Black
Pct_AIANAP
Pct_Asan
Pet_NHOPI
Pct_Other_Race
Pct_Multi_Race
Pct.Hispanic
Pct_LE_Six
Pct_Smgle_Parent
Median_HH_Income
Pct_HH_No_Earnlngs
Pct_HH_No_Wage
Pct_HH_Public_Assist
Median_Family_Income
Median_Per_Capita_Income
Pct_LT_Poverly
Pct_LE_5Yrs_LT_Poverty
Pct_Family_Income_LT_Poverty
Pct_LT_9th_Grade
Pct_No_HS_Degree
Pet_No_CoDege
Pct_No_College_Degree
Total_Houang_Units
Pct_Vacant
Pct_Rented
Pct_Built_Pre_19'W
Pct_puat_Pre_1950
Pct_8uilt_Pre_1960
Pct_Built_Pre_1970
Pct_Built_Pre_1930
Median_Yr_Built
Pct_Occ_Built_Pre_1910
Pct_Occ_Built_Pre_1950
                                                                                      Page H-9

-------
                                PILOT STUDY OF TARGETING ELEVATED BLOOD-LEAD LEVELS IN CHILDREN
                                           Massachusetts Blood Lead Data Aggregated to the Census Tract Level
                                                                        Data Dictionary
                                                                         March 2, 2007
   # Field Name
Data Type
Size  Description
Primary KeySecondary
Table: MA_Analysis_Dataset:  Dataset used in SAS that contains data from all data sources.
   1 FullTract

   2 time
   3 Total_Pop
   4 PCL.AIANA
   5 Pct_Multi_Race
   6 Pct_LE_Six
   7 Num_LE_Six
   8 Pct_Single_Parent
   9 Median_HH_lncome
  10 Pct_HH_No_Wage
  11 Pct_LT_9th_Grade
  12 Pct_No_HS_Degree
  13 Pct_No_College
  14 Total_Housing_Units
  15 Pct_Vacant
  16 Pct_Rented
  17 Pct_Built_Pre_1950
  18 Pct_Occ_Built_Pre_1940
  19 Pct_Occ_Built_Pre_1980
  20 Median_Rent
  21 numhouse
  22 p4
  23 f4
  24 cf_cum
  25 sf_cur
  26 ntot
  27 p5
  28 p10
  29 p15
  30 p25
  31 gm
  32 hq
  33 water_surf
  34 quarter
Text                   11  Concatenation of State & County FIPS & Census Tract numbers to create unique
                          identifier
Number, Double          8  Time (Measured iin Years - Centered at Yr 2000)
Number, Long Integer      4  Total Population
Number, Double          8  Percent Population where race is American Indian/Alaska Native alone
Number, Double          8  Percent Population where race is Two or More Races
Number, Double          8  Percent Population who are less than or equal to six years of age
Number, Long Integer      4  Number Population who are less than or equal to six years of age
Number, Double          8  Percent Households with Single Parents
Number, Double          8  Median Household income
Number, Double          8  Percent Households with no wages
Number, Double          8  Percent Population with less than a 9th grade education
Number, Double          8  Percent Population with no high school degree
Number, Double          8  Percent Population with no college experience
Number, Long Integer      4  Total number housing units
Number, Double          8  Percent Housing units vacant
Number, Double          8  Percent Housing units rented
Number, Double          8  Percent Housing units built before 1950
Number, Double          8  Percent Occupied Housing units built before 1940
Number, Double          8  Percent Occupied Housing units built before 1980
Number, Double          8  Median rent
Number, Long Integer      4  Number of Households
Number, Double          8  MDPH: Housing Interventions - Proportion in compliance
Number, Double          8  MDPH: Housing Interventions - Proportion out of compliance
Number, Double          8  CDC- Cumulative Funding Allocated to Census Tract
Number, Double          8  MDPH: Current Funding Allocation from Massachusetts Loans
Number, Double          8  MDPH: Total Number of Children Screened
Number, Double          8  MDPH: Proportion of Children Screened >= 5 ug/dL
Number, Double          8  MDPH: Proportion of Children Screend >= 10 ug/dL
Number, Double          8  MDPH: Proportion of Children Screend >= 15 M9/dL
Number, Double          8  MDPH. Proportion of Children Screend >= 25 ug/dL
Number, Double          8  MDPH: Geometric Mean Blood-Lead Concentration
Number, Double          8  NATA Air Concentration, Hazard Quotient Model
Number, Double          8  TRI- Pounds of Lead Emitted into Surface Water
Number, Integer          2  Seasonality Adjustment Term (Quarter of the Year)
                                                                               Yes (Duplicates OK)

                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                               Yes (Duplicates OK)
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      NO
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                               Yes (Duplicates OK)
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      No
                                                                                      NO
                                                                                      No
                                                                                      No
                                                                            Page H-10

-------
                                PILOT STUDY OF TARGETING ELEVATED BLOOD-LEAD LEVELS IN CHILDREN
                                           Massachusetts Blood Lead Data Aggregated to the Census Tract Level
                                                                        Data Dictionary
                                                                         March 2, 2007
   # Field Name
Data Type
Size Description
Primary KeySecondary
Table: tblCensus2000Tract:  US Census Summary File 3 data.
   1 FullTract

   2 REGION
   3 STATE
   4 COUNTY
   5 TRACT
   6 INTPTLAT
   7 INTPTLON
   8 SQMILE
   9 Total_Pop
  10 PcLWhite
  11 Pct_Black
  12 Pct_AIANA
  13 Pct_Asian
  14 Pct_NHOPI
  15 Pct_Other_Race
  16 Pct_Mul1i_Race
  17 Pct_Hispanic
  18 Pct_LE_Six
  19 Num_LE_Six
  20 Pct_Smgle_Parent
  21 Median_HH_lncome
  22 Pct_HH_No_Eamings
  23 Pct_HH_No_Wage
  24 Pct_HH_Pubhc_Assist
  25 Median_Family_lncome
  26 Median_Per_Capita_lncome
  27 Pct_LT_Poverty
  28 Pct_LE_5Yrs_LT_Poverty
  29 Pct_Family_lncome_LT_Poverty
  30 Pct_LT_9th_Grade
  31 Pct_No_HS_Degree
  32 Pct_No_College
  33 Pct_No_College_Degree
  34 Total_Housing_Units
  35 Pct_Vacant
Text                   11  Concatenation of State & County FIPS & Census Tract numbers to create unique
                          identifier
Text                    1  US Census Region
Text                    2  State FIPS
Text                    3  County FIPS
Text                    6  Census Tract number
Text                    9  Latitude of Interior Point
Text                   10  Longitude of Interior Point
Number, Double          8  Square miles of Census Tract
Number, Double          8  Total Population
Number, Double          8  Percent Population where race is Caucasian
Number, Double          8  Percent Population where race is African-American
Number, Double          8  Percent Population where race is American Indian/Alaska Native alone
Number, Double          8  Percent Population where race is Asian
Number, Double          8  Percent Population where race is Native Hawaiian and Other Pacific Islander
Number, Double          8  Percent Population where race is Some Other Race alone
Number, Double          8  Percent Population where race is Two or More Races
Number, Double          8  Percent Population where ethnicity is Hispanic
Number, Double          8  Percent Population who are less than or equal to six years of age
Number, Double          8  Number Population who are less than or equal to six years of age
Number, Double          8  Percent Households with Single Parents
Number, Double          8  Median Household income
Number, Double          8  Percent Households with no earnings
Number, Double          8  Percent Households with no wages
Number, Double          8  Percent Households on Public Assistance
Number, Double          8  Median Family Income
Number, Double          8  Median Per Capita Income
Number, Double          8  Percent Population living below poverty level
Number, Double          8  Percent Population less than or equal to five years of age living below poverty
Number, Double          8  Percent Households with family income less than poverty level
Number, Double          8  Percent Population with less than a 9th grade education
Number, Double          8  Percent Population with no high school degree
Number, Double          8  Percent Population with no college experience
Number, Double          8  Percent Population with no college degree
Number, Double          8  Total number housing units
Number, Double          8  Percent Housing units vacant
                                                                                     No

                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     NO
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     NO
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                                     No
                                                                            Page H-11

-------
                               PILOT STUDY OF TARGETING ELEVATED BLOOD-LEAD LEVELS IN CHILDREN
                                          Massachusetts Blood Lead Data Aggregated to the Census Tract Level
                                                                       Data Dictionary
                                                                        March. 2,2007"
   # Field Name
 Data Type
             Size Description
                                                                   Primary KeySecondary
Table: tblCensus2000Tract: US Census Summary File 3 data.
  36 PcURented
  37 Pct_Built_Pre_1940
  38 Pct_Built_Pre_1950
  39 Pct_Built_Pre 1960
  40 Pct_Built_Prei1970
  41 Pct_Built_Pre_1980
  42 Median_Yr_Built
  43 Pct_Occ_Built Rre_1940
  44 Pct_Occ_Built_Pre_1950
  45 "Pct_Occ_Built Pre_1960
  46 Pct_Occ_Built_Pre_1970
  47 Pct_Occ_Buiit_Pre_1980
  48 Median_Yr_Occ_Built
  49 Median_Rent
  50 Housing_Value
  51 Housing_Density
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Number,,
Number,
Number,
Number,
Number,
Number,
Number,
Number,
Double
Double
Double
Double
Double
Double
Double.
Double
Double
Double
Double
Double
Double
Double
Double
Double
8  Percent Housing units rented
8  Percent Housing units built before 1940
8  Percent Housing units built before 1950
8  Percent Housing units built before 1960
8. Percent Housing units built before 1970
8  Percent Housing units built before 1980
8  Median year units built
8  Percent Occupied Housing units built before 1940
8  Percent Occupied Housing units built before 1950
8  Percent Occupied Housing units built before 1960
8  Percent Occupied Housing units built before 1970
8  Percent Occupied Housing units built before 1980
8  Median Year Occupied Housing units built
8  Median rent
8  Housing value
8  Housing Density (calculated variable. Housing_Density = Total_Housing_Units /
   SQMILE)
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
Table: tbIMADataByCensusTract: Aggregated Massachusetts blood lead level data by census tract.
   1 FullTract

   2 Year
   3 Quarter
   4 StateAbbrev
   5 Total
   6 PercentGS
   7 PercentGIO
   8 PercentG15
   9 PercehtG25
  10 GeometricMean
  11 ArithmeticMean
Text                  11  Concatenation of State & County FIPS & Census Tract numbers to create unique
                          identifier.
Number, Long Integer     4  Sampling year
Number, Long Integer     4  Sampling quarter (1= Jan - Mar; 2=Apr-Jun; 3=Jul-Sep; 4=Oct-Dec)
Text                   2  US Postal Service state abbreviation
Number, Long Integer     4  Total Number of Children Screened for PbB
Number, Double          8  Percent of Children Observed with PbB > = 5 ug/dL
'Number, Double          8  Percent of Children Observed with PbB > = 10 pg/dL
Number, Double          8  Percent'of ChHdren Observed with PbB > = 15 ug/dL
Number; Double          8  Percent of Children Observed with PbB > = 25 |jg/dL
Number, Double          8  Geometric Mean
Number, Double          8  Arithmetic Mean
                                                                                           Yes (Duplicates OK)

                                                                                                  No
                                                                                                  No
                                                                                                  No
                                                                                                  No
                                                                                                  No
                                                                                                  No
                                                                                                  No
                                                                                                  No
                                                                                                  No
                                                                                                  No
                                                                           Page H-12

-------
                              PILOT STUDY OF TARGETING ELEVATED BLOOD-LEAD LEVELS IN CHILDREN
                                         Massachusetts Blood Lead Data Aggregated to the Census Tract Level
                                                                     Data Dictionary
                                                                      March 2, 2007
   # Field Name
Data Type
Size Description
Primary KeySecondary
Table: tblMAFunding_CDC: National CDC Grantee Funding Allocated to the Census Tract Level for the Applicable Massachusetts Grantees.
   1 FullTract

   2 Fundmg_Year
   3 Funding
   4 Funding_Source
Text

Number, Integer
Number, Long Integer
Number, Integer
  11  Concatenation of State & County FIPS & Census Tract numbers to create unique
     identifier
   2  Year of Funding from CDC
   4  Amount of Funding from CDC
   2  Source of Funding from CDC (0=no funding; 1 =county-level funding, 2= state-level
     funding)
        Yes (Duplicates OK)

               No
               No
               No
Table: tbIMAHouseholdData:  Massachusetts environmental data.
   1  Address! D
   2 Activity Date
   3 Event_Code
   4 EventDesc
   5 Event_outcome
   6 County
   7 Census_Tract
   8 Event Result
   9 In Final Compliance
  10 Event_Outcome_Code
  11  Month
  12 Quarter
  13 Year
  14 FullTract
Text                  14  Address ID
Date/Time              8  Activity Date
Number, Integer          2  Event Code
Text                  40  Event Description
Text                  20  Event Outcome
Text                 255  County
Text                 255  Census Tract
Text                   1  Event Result
Text                   1  In Final Compliance
Number, Integer          2  Event Outcome Code
Number, Integer          2  Month of sample
Number, Integer          2  Quarter of Sample
Number, Integer          2  Year of Sample
Text                  11  Concatenation of State & County FIPS & Census Tract numbers to create unique
                         identifier
                                                                                  No
                                                                                  No
                                                                                  No
                                                                                  No
                                                                                  No
                                                                                  No
                                                                                  No
                                                                                  No
                                                                                  No
                                                                                  No
                                                                                  No
                                                                                  No
                                                                                  No
                                                                           Yes (Duplicates OK)
Table: tbIMAStateFunding:  Funding from the MDPH allocated across census tracts for the state of Massachusetts.
   1  FullTract

   2  Funding_Year
   3  Funding
   4  Fundmg_Source
Text                  11

Number, Integer          2
Number, Long Integer     4
Number, Integer          2
     Concatenation of State & County FIPS & Census Tract numbers to create unique
     identifier
     Year of Funding from MDPH
     Amount of Funding from MDPH
     Source of Funding from MDPH (0=no funding; 3 = funding from MDPH)
        Yes (Duplicates OK)

               No
               No
               No
                                                                         Page H-13

-------
                              PILOT STUDY OF TARGETING ELEVATED BLOOD-LEAD LEVELS IN CHILDREN
                                        Massachusetts Blood Lead Data Aggregated to the Census Tract Level
                                                                    Data Dictionary
                                                                     March 2, 2007
   # Field Name
Data Type
Size Description
                                                                Primary KeySecondary
Table: tbINATA1999CAANUM182_Tract
   1  fulltract                               Text

   2  ASPENTotal                           Number, Double
   3  HAPEMTotal                           Number, Double
   4  HQTotal                              Number, Double
                     11  Concatenation of State & County FIPS & Census Tract numbers to create unique
                        identifier
                      8  Air Concentration, Assessment System for Population Exposure Nationwide Model
                      8  Air Concentration, Hazardous Air Pollutant Exposure Model
                      8  Air Concentration, Hazard Quotient Model
                                                                          Yes (No Duplicates)

                                                                                 No
                                                                                 No
                                                                                 No
Table: tblNationalFunding_MA_Tracts: 1999 EPA NATA Data Ambient Lead data by census tract.
   1  FullTract

   2  Funding_Year
   3  Funding
   4  Funding_Source
Text

Number, Integer
Number, Long Integer
Number, Integer
11  Concatenation of State & County FIPS & Census Tract numbers to create unique
   identifier
 2  Year of Funding from HUD
 4  Amount of Funding from HUD
 2  Source of Funding from HUD (0=no funding; 1 =county-level funding; 2= state-level
    funding)
                                                                          Yes (Duplicates OK)

                                                                                 No
                                                                                 No
                                                                                 No
Table: tblTRI_Data_MA: Toxics Release Inventory Data at the Census Tract Level for the state of Massachusetts (Lead and Lead Compounds only, 1998 - 2004 data).
   1  TRIFID                               Text
   2  FullTract                              Text

   3  REPORTING_YEAR                     Number, Integer
   4  CAS NUMBER                         Text
   5  CHEMICALJMAME                      Text
   6  TOTAL_AIR_EMISSIONS                 Number, Double
   7  TOTAL_FUG_AIR_EMISS                 Number, Double
   8  TOTAL_NUMBER_RECEIVING_STREAMS   Number, Byte
   9  TOTAL_STACK_AIR_EMISS              Number, Double
  10  TOTAL_SURFACE_WATER_DISCHARGE    Number, Double
  11  TOTAL_UNDERGROUND_INJECTION       Number, Double
  12  UNIT_OF_MEASURE                    Text
                     20 Toxics Release Inventory Facility ID
                     11 Concatenation of State & County FIPS & Census Tract numbers to create unique
                        identifier
                      2 Reporting Year
                     15 CAS Number
                    100 Name of Chemical ("Lead", "Lead Compounds")
                      8 Total Air Emissions
                      8 Total Fugitive Air Emissions
                      1 Total Number Receiving Streams
                      8 Total Stack Air Emissions
                      8 Total Surface Water Discharge
                      8 Total Underground Injection
                      6 Unit of Measure (pounds)
                                                                                 No
                                                                          Yes (Duplicates OK)

                                                                                 No
                                                                                 No
                                                                                 No
                                                                                 No
                                                                                 No
                                                                                 No
                                                                                 No
                                                                                 No
                                                                                 No
                                                                                 No
                                                                        Page H-14

-------
                                  PILOT STUDY OF TARGETING ELEVATED BLOOD-LEAD LEVELS IN CHILDREN
                                              Massachusetts Blood Lead Data Aggregated to the Census Tract Level
                                                                                 Data Diagram
                                                                                 March 2,  2007
tblOnsus2000Tract
REGION
STATE
COUNTY
TRACT
INTPTIAT
INTPTLON
SQMILE
Total_Pop
Pct_White
Pct_Black
Pct_AIANA
Pct_Asian
Pct_NHOPI
Pct_Qther_Race
Pct_Multi_Race
Pct_Hispanic
Pct_LE_Six
Num_LE_Six
Pct_Single_Parent
Median_HH_Income
Pct_HH_No_Earnings
Pct_HH_No_Wage
Pct_HH_Public_Assist
Median_Family_!ncome
Median_Per_Capita_Income
Pct_LT_Poverty
Pct_lE_SYrs_LT_Poverty
Pct_Family_Income_LT_Poverty
Pct_LT_9th_Grade
Pct_No_HS_Degree
Pct_No_College
Pct_No_College_Degree
TotalJ-iousingJJnits
Pct_Vacant
Pct_Rented
Pet _Built_Pre_ 1930
Pct_Built_Pre_1950
Pct_Built_Pre_1960
Pct_Built_Pre_1970
Pct_Buili_Pre_19SO
Median_Yr_6uilt
Pct_Occ_Built_Pre_19'W
Pct_Occ_Built_Pre_ 19 50
Pct_Occ_Built_Pre_1960
Pct_Occ_Built_Pre_1970
Pct_Occ_Built_Pre_19SO
Median_Yr_Occ_Built
Median_Rent
Housing_Value
  tblMAFundmeCDC
  FulTract
  FundhgLYear
  Funding
  Funding_Source
jFuirract
 Fin*ig_Year
I Funding
|JFunding_Source
  tbiMADataBvC?n<;u5Tra(:t
  FJTract
  Year
  Quarter
  StateAbbrev
  Total
  PercentGS
  PercentG 10
  PercentG 15
  PercentG25
  GeornetricMean
  ArithmeticMean
                                                                                          MA. .AnAlv$ts_.D3taset
tbINattonalFunding_MA__Tr
acts
FufTract
Fvndmg_Year
Funding
Funding_Source
FufTract
tine
Total_Pop
Pct_AIAMA
Pct_Multi_Race
Pct_LE_Six
Num_LE_Six
Pct_Single_Parent
Median_HH_Income
Pct_HH_No_Wage
Pct_LT_9th_Grade
Pct_No_HS_Degree
Pct_No_College
Tot3l_Housing_L)nits
Pct_Vacant
Pct_Rented
Pct_Built_Pre_19SO
Pct_Oa_Built_Pre_1943
Pct_Occ_Built_Pre_1930
Median_Rent
numhouse
P4
f4
cf_cum
sf_cur
ntot
p5
plO
pis
p25
gm
hq
water_surf
quarter
                                                                                                                                tbiMAHousehDldPdta
 AddressID
 Activity Date
 Event_Code
 EventDesc
 Event_outcome
 County
 CensusJTract
 Event Re suit
 In Final Compliance
 Event_Outcome_Code
 Month
 Quarter
 Year
 Fuim-act
                                                                               fijltract
                                                                               ASPE^^•otal
                                                                               HAPEMTotal
                                                                               HQTotal
                                                                               tbiTRi Dซta  Mi
TRIFIO
FullTract
REPORTTNG_YEAR
CAS_NUMBER
CHEMICAL_NAME
TOTAi_AIR_EMISSIONS
TOTAL_FUG_AIR_EMISS
TOTAL_NUMBER_RECaVING_STREAMS
TOTAL_STACK_AIR_EMISS
TOTAL_SURFACE_WATER_DISCHARGE
TOTAJ._UNDERGROUNO_INJECTION
UNIT OF MEASURE

-------