Framework for Identifying and
Evaluating Lead-Based Paint Hazards
from Renovation, Repair, and Painting
Activities in Public and Commercial
Buildings
May 2014
EPA Office of Pollution Prevention and Toxics
1200 Pennsylvania Avenue, N.W.
Washington, DC 20460
-------
s
1. Introduction 2
2. Background 3
2.1. Previous Approach for Residences and COFs 3
2.1.1. Identifying Hazards 3
2.1.2. Evaluating Hazards 3
2.2. Considerations for an Approach for P&CBs 3
3. Discussion of a Potential P&CB Approach 4
3.1. Hazard Identification 5
3.2. Hazard Evaluation 6
3.3. Hazard Evaluation With Monte Carlo Analyses 7
3.4. Hazard Evaluation Using Child's IQ Decrements 7
3.5. Considerations for Using a Scenario-Specific Approach 9
3.5.1. Applicability/generalizability of Modeled Scenarios 9
3.5.2. Selection of Risk Reduction Measures 9
3.5.3. Exposure at Residences and COFs 9
4. References 10
Appendix A. Exposure and Risk Characterization A-l
References A-6
Appendix B. Variables Being Considered for Use in the Analysis and Designation of
Scenario or Monte Carlo Variables in the Model Framework A-7
Appendix C. Preliminary Analysis A-10
Figures
Figure 1. Evaluation of Hazard across Monte Carlo Percentiles and Scenario Means 8
Figure 2. Schematic of EPA's Exposure and Risk Assessment Process A-l
Figure 3. Modeling Framework for Exterior Renovation Activities A-4
Figure 4. Modeling Framework for Interior Renovation Activities A-5
Figure 5. Trends in Incremental IQ Change & Time Spent for Children Inside P&CBs
(Preliminary Analyses) A-ll
Figure 6. Trends in Incremental IQ Change & Room Size for Children Inside P&CBs
(Preliminary Analyses) A-ll
US EPA
May 2014
-------
Framework for Identifying and Evaluating Lead-Based Paint Hazards from Renovation, Repair, and Painting
Activities in Public and Commercial Buildings
Acronyms and Abbreviations
Acronym / Abbreviation
Stands For
AERMOD
American Meteorological Society (AMS) and the U.S. Environmental Protection
Agency (EPA) Regulatory Model Improvement Committee (AERMIC) Model
COFs
Child-Occupied Facilities
EPA
U.S. Environmental Protection Agency
IQ
Intelligence Quotient
OSHA
U.S. Occupational Safety and Health Administration
P&CBs
Public and Commercial Buildings
PbB
Blood Lead
RRP
Renovation, Repair, and Painting
SAB
Science Advisory Board
TSCA
Toxic Substances Control Act
US EPA
May 2014
-------
Framework for Identifying and Evaluating Lead-Based Paint Hazards from Renovation, Repair, and Painting
Activities in Public and Commercial Buildings
1. introduction
The U.S. Environmental Protection Agency (EPA) is currently in the process of determining whether or
not lead-based paint hazards are created by renovation, repair and painting (RRP) activities in public and
commercial buildings (P&CBs), as required under section 402(c)(3) of the Toxic Substances Control Act
(TSCA). For those renovation activities in P&CBs that create lead-based paint hazards, TSCA directs EPA
to address the hazards through regulation.
This document will explain, in general terms, how EPA could define what constitutes a "lead-based paint
hazard" for P&CBs undergoing renovations. EPA refers to this as "hazard identification", pursuant to
TSCA section 4031. This document also discusses how EPA could assess the expected impact of
renovation activities in P&CBs in order to determine whether or not RRP activities would be expected to
create a hazard. EPA refers to this as "hazard evaluation", pursuant to TSCA section 402(c)(3)2. If the
analysis indicates that hazards are created, EPA is directed by TSCA section 402(c)(3) to propose
appropriate regulatory controls to reduce or prevent exposure.
In two separate rulemakings, EPA has previously defined lead-based paint hazards in residences and
child-occupied facilities (COFs), and evaluated whether such hazards would result from RRP activities of
those buildings. For P&CBs, EPA is considering an approach that is different from our previous
residential and COF approach, both in the timing and the sequence of events, as well as in the manner in
which the Agency defines and evaluates hazards. EPA previously took a uniform approach to identifying
and evaluating hazards under TSCA for residences and COFs. This document describes a more tailored
approach that could be used for P&CBs. In addition, for P&CBs EPA would combine the identification of
hazards, the evaluation of whether hazards occur and any proposed regulatory requirements, if
appropriate, in one proposed rule. Because of these differences, EPA is providing an early opportunity,
prior to any proposal, for interested parties to review, consider, and provide feedback on the approach
described herein.
Because EPA is providing this information and opportunity for public comment prior to issuing any
proposed rule, the information contained in this document is limited and does not represent or
constitute an Agency decision or regulatory proposal. Additionally, this document does not provide
significant detail regarding modeling inputs and results, how EPA might apply the results of any analyses,
or a discussion regarding what magnitude of deleterious health effect would be considered to be
adverse. Further details and the results of such analyses would be provided for review and comment in
any future proposal. In addition, EPA plans to make public, and provide for peer review of any such
analyses.
115 U.S. Code § 2683
2 15 U.S.C. § 2682(c)(3).
US EPA
2
May 2014
-------
Framework for Identifying and Evaluating Lead-Based Paint Hazards from Renovation, Repair, and Painting
Activities in Public and Commercial Buildings
Aground
2.1. Previoi proach for Residences and COFs
2.1.1. Identifying Hazards
For residences and COFs, EPA promulgated a uniform standard under TSCA 403 defining a level of lead in
various media (i.e., dust, soil, paint, etc.) that EPA would consider to be dangerous, as a persistent
condition, based on available scientific literature about the health effects of lead to adults and children
and expected exposure in residences and COFs. In 2001, EPA promulgated a regulation that identified,
for residential dwellings and COFs, three different types of lead-based paint hazards: paint-lead hazards,
dust-lead hazards, and soil-lead hazards (EPA 2001). EPA relied on the Centers for Disease Control (CDC)
"level of concern" at the time3, as well as other information as described in the preamble to that (403)
rule, to inform our decision about what health endpoint and magnitude would be considered an adverse
health effect. At the time, the CDC level of concern was defined as a blood lead level for children greater
than or equal to 10 ng/dL (CDC, 1991). The EPA residential and COF hazard standards were developed to
provide a relatively low chance that an individual child would experience a blood lead level at or above
10 ng/dL.
2.1.2. Evaluating Hazards
TSCA section 402(c)(3) requires EPA to evaluate whether or not renovation activities create lead-based
paint hazards. For residences and COFs, the hazard finding was made by first determining the dust-lead
levels that would be generated by renovation activities and then comparing these levels to the hazard
standard. This approach was used for the 2008 RRP Rule (US EPA, 2008), wherein EPA compared the
observed dust-lead levels from the renovations tested in the Dust Study (US EPA, 2007) to the dust-lead
hazard standard promulgated in 2001 (i.e., 40 ng/ft2 on floors or 250 ng/ft2 on interior window sills). Any
renovation activities that created dust-lead levels that exceeded the dust-lead hazard standard for floors
were considered to create hazards. This approach formed the basis for EPA's determination that all
renovation activities that disturb paint in target housing and COFs create lead-based paint hazards.
2.2. Considerations for an Approach for P&CBs
The 2001 hazard standards apply only to target housing and COFs. EPA has explored setting hazard
standard(s) for P&CBs. Attempting to identify and apply a "one size fits all" numerical dust-lead level or
3 Since EPA promulgated the hazard standards in 2001, CDC has updated its recommendations on children's blood lead levels.
Experts now use a reference level of 5 micrograms per deciliter to identify children with blood lead levels that are much
higher than most children's levels. This new level is based on the U.S. population of children ages 1-5 years who are in the
highest 2.5% of children when tested for lead in their blood.
US EPA
3
May 2014
-------
Framework for Identifying and Evaluating Lead-Based Paint Hazards from Renovation, Repair, and Painting
Activities in Public and Commercial Buildings
other media metric, however, may not be appropriate for P&CBs. Exposure times in P&CBs are much
more varied than those in homes and COFs. Children and adults may spend only a few minutes in certain
P&CBs, while other exposure may last much longer, such as in a school or similar facility for children 6
years and older, and workplaces for adults. P&CBs where children under 6 years spend significant
amounts of time are already regulated under the 2008 RRP rule. The tailored approach described in this
Framework avoids the creation of a uniform hazard standard that may be more protective than
necessary in some cases and not protective enough in others.
In addition, P&CBs vary greatly in terms of their size, shape, and room configurations, compared to
residences and COFs. EPA hopes to take this broad variability into account and is therefore considering
an approach to more appropriately account for risk. The approach described in this document, referred
to as the "scenario-specific approach", is designed to account for the variable amounts of time spent in
P&CBs, the broad heterogeneity in building sizes and configurations, and the short-term nature of the
exposure resulting from renovation activities in order to assess risk.
If selected, the scenario-specific approach described in this document would be used to make a hazard
finding inside P&CBs resulting from renovations of the same, or nearby P&CBs. This approach would be
used to support a hazard finding for these situations inside P&CBs where existing hazard standards do
not exist. As described here, the analytical approach and analysis plan would not address hazards to
passers-by or individuals that could be exposed to lead outside of P&CBs through air, soil, or other
media, or for exposure at homes and COFs where hazard standards are already established.
3. Discussion of a Potential P&CB Approach
Using the tailored approach, EPA would model exposure to lead from renovations in a wide distribution
of building types, resulting from various renovation activities and taking into consideration the different
age groups of occupants and the varied exposure times that might occur. This distribution of activities
would allow EPA to characterize risk across all the exposure conditions and situations that may occur.
Once this probabilistic distribution is developed EPA would review the modeling results including
estimates of blood lead changes and appropriate health endpoints, such as IQ decrements for young
children. To inform the hazard finding, EPA would consider the magnitude of any predicted deleterious
health effects as well as how frequently that exposure situation is likely to occur. This approach allows
EPA the ability to take into account the varied exposure times for individuals, and the variability in
buildings and renovations when characterizing risks. This tailored approach would provide the ability to
apply any regulatory requirements to only those situations that most closely resemble those modeled
scenarios in which the probabilistic modeling predicted that an adverse health effect would occur. This
allows EPA the ability to limit the scope of any regulatory intervention in a way not possible when
relying on a uniform hazard standard.
The adaptability and tailored nature of the scenario-specific approach is evident when reviewing the
results of the preliminary analysis shown in Appendix C. However, preliminary analysis results are not
US EPA
4
May 2014
-------
Framework for Identifying and Evaluating Lead-Based Paint Hazards from Renovation, Repair, and Painting
Activities in Public and Commercial Buildings
representative of all scenarios that could be analyzed, and the modeling inputs and parameters have not
been finalized. Therefore, the preliminary findings reported in Appendix C should not be construed as
final and may change pending subsequent and more comprehensive analyses. The preliminary analysis
trends displayed in Figure 5 of Appendix C, indicate that, where other factors such as room size are held
equal, renovations in buildings where children spend longer amounts of time result in considerably
larger deleterious health effects (Incremental IQ change) than do renovations in buildings where
children spend shorter amounts of time. As another example, reflected in Figure 6 of Appendix C, the
preliminary analysis indicates that, where other factors such as time of exposure are held equal,
renovations that take place when children are in the renovation workspace and renovations in smaller
rooms result in larger adverse health effects (Incremental IQ change). The advantage of using this
approach is that it is possible for EPA, by modeling exposures specific to different scenarios, to avoid
reliance on a uniform hazard standard that may be more protective than necessary in some cases and
not protective enough in others. As a result, EPA expects that the scenario specific approach would
enable the Agency to evaluate risk and target the application of any potential regulatory requirements
to situations in a more efficient and appropriate manner. In addition, EPA would be able to more
accurately ascertain if there are renovation situations in which regulatory intervention is necessary
because the risk to building occupants or visitors is particularly significant.
3.1. Hazard Identification
The scenario-specific approach would define lead-based paint hazards inside P&CBs in terms of
exposures created by P&CB renovation conditions that result in adverse health effects.
TSCA Section 401(10) defines a "lead-based paint hazard" as:
any condition that causes exposure to lead from lead-contaminated dust, lead-
contaminated soil, lead-contaminated paint that is deteriorated or present in accessible
surfaces, friction surfaces, or impact surfaces that would result in adverse human health
effects established by the Administrator under this title.4,5
Under this approach, EPA would consider the health implications of lead exposure and define lead-
based paint hazards - solely for the purposes of evaluating exposures from renovations in P&CBs - as
any condition that causes exposure to lead-based paint dust that would result in health effects that the
Agency finds to be adverse. The determination of whether or not a health effect is adverse would be
4 15 U.S.C. § 2681(10) (stating that EPA must classify as a lead-based paint hazard, "any condition that causes exposure to lead
from lead contaminated dust, lead-contaminated soil, lead contaminated paint that is deteriorated or present in accessible
surfaces, friction surfaces, or impact surfaces that would result in adverse human health effects" as established by EPA).
5 The definition in section 401 encompasses both exposure and effects, which by definition are components of a risk assessment.
US EPA
5
May 2014
-------
Framework for Identifying and Evaluating Lead-Based Paint Hazards from Renovation, Repair, and Painting
Activities in Public and Commercial Buildings
informed by a variety of considerations including available scientific literature, measurability, frequency
of occurrence, biological significance, etc.
EPA is currently evaluating the literature and the conclusions included in both the Integrated Science
Assessment for Lead (US EPA, 2013) and the National Toxicology Program Monograph on Health Effects
of Low-Level Lead (NTP, 2012) to develop the concentration-response functions for the health effects,
such as renal and cardiovascular, associated with lead exposure for both older children and adults. As
appropriate, those concentration-response functions and the magnitude of associated health effects
resulting from incremental changes in blood lead level from the modeled renovation scenarios would
form the basis for EPA's identification of hazard. For children, intelligence quotient (IQ) change and
possibly other endpoints would be evaluated as a quantifiable health effect. However, EPA notes that
most P&CBs where children under 6 years would spend longer amounts of time are already regulated
under the 2008 RRP rule as COFs6.
For example, EPA might find that a particular IQ decrement in young children is an adverse health effect.
A hazard, then, would be identified as any renovation condition that causes exposure to lead-based
paint dust that would result in that particular IQ decrement for children. A similar identification could be
made for adults based on other health endpoints as well.
3.2. Hazard Evaluation
Under this approach, EPA would evaluate whether or not exposure to leaded dust from specific
renovation scenarios in P&CBs would result in the health effects identified as adverse. For those that do,
EPA would determine that lead-based paint hazards are created. To perform this analysis, EPA would
assess elevations in lead exposure resulting from a broad range of scenarios, considering variations in
types of renovation activities, building types, sizes and configurations, use and occupancy patterns,
cleaning frequencies, etc., which are designed to be reflective of actual P&CB settings. EPA could
evaluate those exposures to both children and adults and consider the implications on blood lead levels
and health effects. For those scenarios where our modeling predicts that the magnitude of health effect
change due to the renovation activity is significant enough to be considered adverse, EPA would make
the finding that lead-based paint hazards are created. EPA would consider how to generalize the results
of the analysis and craft any regulatory requirements to apply only to those renovation activities and
P&CB categories where exposure is reflective of the modeled scenarios in which a hazard was found.
6 40 C.F.R. 745.80 Child-occupied facility means a building, or portion of a building, constructed prior to 1978, visited regularly
by the same child, under 6 years of age, on at least two different days within any week (Sunday through Saturday period),
provided that each day's visit lasts at least 3 hours and the combined weekly visits last at least 6 hours, and the combined annual
visits last at least 60 hours.
US EPA
6
May 2014
-------
Framework for Identifying and Evaluating Lead-Based Paint Hazards from Renovation, Repair, and Painting
Activities in Public and Commercial Buildings
3.3. Hazard Evaluation With Monte Carlo Analyses
EPA would utilize a Monte Carlo method to evaluate hazards inside P&CBs if using a scenario-specific
approach. The Monte Carlo analysis is described in Appendix A.2 and the variables under consideration
are described in Appendix B. The Monte Carlo analysis would repeatedly sample from distributions of
these variables to create a probabilistic distribution of exposure for each renovation scenario. EPA
would then consider these single scenario exposures within the overall context of the general
population. To do so, the variability within each exposure scenario and the probability of occurrence for
each exposure scenario would be estimated.
For children, EPA would estimate the resultant IQ level changes due to lead exposure from both RRP
activities and background sources (from non-renovation activities), and then subtract out the IQ change
related to background alone. The magnitude of the incremental IQ change may be affected by
background levels, since the background level affects the shape of the concentration-response curve.
However, in evaluating distributions of results for the scenario-specific approach, only the renovation-
related (incremental) exposures and effects would be considered in making a hazard finding. For IQ
change in children, or for any other health effect chosen for the analysis, the concentration-response
function can be used to create a distribution of incremental health effect changes.
The task of identifying adverse adult health effects for exposure scenarios anticipated to occur during
renovation of P&CBs presents some unique difficulties. The median blood lead level in the general US
population, based on the latest NHANES data, is now approximately 1 |-ig/dL, and the 95th percentile is
3.5 ng/dL (CDC, 2013). However, the literature on health effects at this exposure level is sparse,
especially for adult health endpoints. It is expected that renovation exposure scenarios in P&CBs would
result in a short-term increase in exposures while the renovation is underway, and then the exposures
will quickly decline after the work is completed. The contribution of short-term exposures to total blood
or bone lead concentration can be modeled, but the overall impact of the changes to total blood and
bone lead from P&CB renovation scenarios is expected to be quite small.
3.4. Hazard Evaluation Using Child's IQ Decrements
EPA would use Monte Carlo predictions of incremental IQ decrements for children to evaluate hazards
to children and make the hazard finding for exposure inside P&CBs from interior and exterior
renovations. Many scenarios could return small mean IQ changes related to the renovation. Relatively
larger IQ changes, however, may occur in two different situations. In the first case, consider a scenario
resulting in a small mean IQ change (Figure 1A). The Monte Carlo simulation provides an estimate of the
distribution of IQ changes for that scenario, and the mean might be the point shown with the "X" in the
figure. When examined based on percentiles, however, the higher percentiles could represent a lead
hazard (the red portion of curve within the "Hazard" shaded area). Thus, the mean exposures can be
low, but a certain fraction of the population in that exposure scenario is expected to have IQ changes
that result in a hazard. In the second case, consider the mean IQ changes across all the different
US EPA
7
May 2014
-------
Framework for Identifying and Evaluating Lead-Based Paint Hazards from Renovation, Repair, and Painting
Activities in Public and Commercial Buildings
scenarios (Figure IB). Most of the mean IQ changes (the blue bars) may not result in a hazard. The IQ
changes from some fraction of those scenarios, however, could result in a lead hazard (the red bars that
reach the "Hazard" shaded area).
To analyze potential hazards, EPA would examine the various distributions across percentiles in a single
scenario (Figure 1A) and across different scenarios (Figure IB). The collection of these distributions helps
account for the total variability in exposure owing to environmental, lifestyle, and biokinetic differences
across the population. EPA would place the less frequent, high incremental IQ changes within the
context of expected (mean) incremental IQ changes. Because each scenario is not equally likely, EPA
would provide a discussion of the relative frequency of these high incremental IQ changes. The hazard
finding would be made based on an overall judgment of the frequency and magnitude of incremental
health effect changes resulting from P&CB renovations.
Figure 1. Evaluation of Hazard across Monte Carlo Percentiles and Scenario Means
A. One Scenario, 20,000
Monte Carlo Iterations
B. 15,000 Scenarios, Mean
Across Monte Carlo
Iterations
Hazard
Hazard
Scenario Scenario Scenario
1 10,000 15,000
Scenarios
50th 95th
Percentiles
US EPA
8
May 2014
-------
Framework for Identifying and Evaluating Lead-Based Paint Hazards from Renovation, Repair, and Painting
Activities in Public and Commercial Buildings
3.5. Considerations for Using a Scenario-Specific Approach
3.5.1. Applicability/generalizability of Modeled Scenarios
The scenario-specific approach seeks to more accurately address hazards and minimize regulatory
burden. However, if appropriate, EPA must also be able to use the approach to propose and implement
regulatory requirements that can be understood and adopted by the regulated community. It is possible
that the analysis could indicate hazards in scenarios that are unlikely to occur, or occur in only very
limited instances affecting a small number of people. EPA would consider how to generalize the hazard
findings from the scenario-specific analysis to assess the effects of a potential regulation on the
population (and on sensitive sub-populations) in considering regulatory options.
3.5.2. Selecti Risk Reduction Measures
The exposure modeling construct of the scenario-specific approach allows a more targeted way for EPA
to evaluate the effect of a variety of mitigation measures in reducing exposures that occur inside P&CBs.
For example, this approach can examine the effectiveness of potential regulatory requirements, such as
preventing occupants of P&CBs from being present in the workspace during the renovation, the use of
plastic barriers, prohibiting certain work practices and requiring the use of HEPA dust capture systems
on equipment, and cleaning requirements to contain or otherwise limit exposure to leaded dust. This is
accomplished by including separate model simulations implementing each work practice option and
estimating the resulting difference in incremental blood lead levels or health effect changes with and
without the various regulatory options.
3.5.3. Exposure at Residences and COFs
The lead dust created during P&CB renovations can move downwind to surrounding COFs, residences
and P&CBs. This dust can penetrate these buildings via air infiltration. In addition, dust can deposit on
the ground or hard surfaces, leading to elevated lead levels and subsequent track-in of dust which would
contribute to interior dust levels. These elevated media concentrations may impact the children and/or
adults within these nearby buildings. EPA is considering how to evaluate hazards created in nearby
residences and COFs as a result of exterior renovations of P&CBs. Because, as discussed earlier, there is
an existing hazard standard that applies to residences and COFs, EPA could evaluate whether hazards
are created inside nearby residences and COFs by comparing estimated dust lead levels to the existing
residential and COF hazard standard of 40 ng/ft2 for floors, and 250 ng/ft2for interior window sills.
US EPA
9
May 2014
-------
Framework for Identifying and Evaluating Lead-Based Paint Hazards from Renovation, Repair, and Painting
Activities in Public and Commercial Buildings
4. References
Centers for Disease Control and Prevention, (1991). Preventing Lead Poisoning in Young Children.
Atlanta, GA: CDC.
Centers for Disease Control and Prevention, (2013). Fourth Report on Human Exposure to
Environmental Chemicals, Updated Tables, (September, 2013). Atlanta, GA: U.S. Department of
Health and Human Services, Centers for Disease Control and Prevention.
http://www.cdc.gov/exposurereport/
Lanphear B; Hornung R; KhouryJ; Yolton K; Baghurst P; Bellinger D; Canfield R; Dietrich K; Bornschein R;
Greene T; Rothenberg S; Needleman H; Schnaas L; Wasserman G; Graziano J; Roberts R. (2005).
Low-level environmental lead exposure and children's intellectual function: an international
pooled analysis. Environmental Health Perspectives 113:894-899.
National Toxicology Program. (2012). National Toxicology Program Monograph on Health Effects of Low-
Level Lead. Available at http://ntp.niehs.nih.gov/?objectid=4F04B8EA-B187-9EF2-
9F9413C68E76458E.
OSHA. (2004) Safety and Health Regulations for Construction, Occupational Health and Environmental
Controls: Lead. 29 CFR - 1926.62.
https://www.osha. gov/pls/oshaweb/owadisp.show_document?p_table=STANDARDS&p_id=106
41.
US EPA. (1998). Risk Analysis to Support Standards for Lead in Paint, Dust, and Soil. Appendix G: Health
effects associated with exposure to lead and internal lead doses in humans. EPA 747-R-97-006.
Washington, DC: US Environmental Protection Agency, Office of Pollution Prevention and Toxics.
Available online at: http://www.epa.gov/lead/pubs/toc.pdf.
US EPA. (2001). Lead; Identification of Dangerous Levels of Lead; Final Rule. Federal Register Volume 66,
Number 4 (Friday, January 5, 2001) [Pages 1206-1240], http://www.epa.gov/fedrgstr/EPA-
TOX/2001/January/Day-05/t84.pdf
US EPA. (2007). Draft Final Report on Characterization of Dust Lead Levels After Renovation, Repair, and
Painting Activities. Prepared for EPA's Office of Pollution Prevention and Toxics (OPPT). Available
online at: http://www.epa.gov/lead/pubs/duststudy01-23-07.pdf
US EPA. (2008) Lead; Renovation, Repair, and Painting Program; Final Rule. Federal Register Volume 73,
Number 78 (Tuesday, April 22, 2008)[Pages 21692-21769] Docket number: EPA-HQ-OPPT-2005-
0049; FRL-8355-7, http://www.gpo.gov/fdsys/pkg/FR-2008-04-22/html/E8-8141.htm.
US EPA. (2010a). Advanced Notice of Proposed Rulemaking for Public and Commercial Buildings. Docket
number: EPA-HQ-OPPT-2010-0173. http://www.gpo.gov/fdsys/pkg/FR-2010-05-06/html/2010-
10097.htm.
US EPA. (2013). Integrated Science Assessment for Lead. U.S. Environmental Protection Agency,
Washington, DC, EPA/600/R-10/075F.
US EPA
10
May 2014
-------
Appendix A. Exposure and Risk Characterization
A.l. Exposure Assessment Process
The left-hand portion of Figure 2 shows the exposure assessment process, as described in EPA's
Guidelines for Exposure Assessment (US EPA, 1992). Each numbered box represents a separate step in
the assessment, from planning, to finding models and input values, to estimating media concentrations
and dose, to characterizing uncertainty. The right-hand portion of the figure incorporates the analysis of
health effects and characterization of risk as described in EPA's Framework for Human Health Risk
Assessment to Inform Decision Making (US EPA, 2014). The Framework takes into account
recommendations on risk assessment processes described in the National Research Council's (NRC) 2009
report, Science and Decisions: Advancing Risk Assessment (NRC, 2009) to use a risk-based approach to
establish whether a hazard exists from renovations of P&CBs with lead-based paint. The figure indicates
how health effects and concentration-response relationships are developed so that the exposure
assessment metrics (intake or dose) can be linked to specific health effects.
Figure 2. Schematic of EPA's Exposure and Risk Assessment Process
Exposure Assessment Toxicity Assessment
1. Plan the Exposure Assessment
Formulate the problem to determine the purpose, scope, and
level of detail
2. Gather and Develop Data for the Exposure
Assessment
Perform literature searches to select models and develop
input parameters
A. Review Health Effects Literature
Characterize the strengths and weaknesses of each study
B. Evidence Synthesis
Indicate the health effects that present a hazard based on
conclusions from the literature
3. Use Data to Estimate Exposure and Dose
a. Estimate relevant media concentrations (air, dust,
soil, etc.)
b. Estimate activity patterns, exposure factors, intake or
internal dose
C. Dose-response Analysis
Using statistical methods, determine the dose-response
relationship for each critical health effect
4. Health Effects
Estimate the incidence or degree of the critical
effect in the exposed population
5. Assess Uncertainty
Assess uncertainty in the scenario, parameters,
and model. Distinguish between population
variability and uncertainty.
Sources:
EPA Guidelines for Exposure Assessment; Federal Register 57{104):22888-22938
EPA Draft Framework for Risk Assessment to Inform Decision Making; Docket ID: EPA-HQ-ORD-2G12-QS79
-------
Framework for Identifying and Evaluating Lead-Based Paint Hazards from Renovation, Repair, and Painting
Activities in Public and Commercial Buildings
A.2. Risk Characterization Inside P&CBs
This section describes how EPA would assess the hazard, exposure, and risk to the U.S. population inside
P&CBs from the renovation, repair, and painting of P&CBs using the scenario-specific approach7. Lead
dust can persist in the room being renovated or migrate to nearby rooms, even after the renovation is
complete. The level of exposure to children and adults will depend on, among other things, the extent to
which they spend time in the renovation room or an adjacent room, either during or after the
renovation (or both), and the amount of time they spend in the building.
Potential P&CB exposures can be characterized by estimating the media concentrations (soil and dust)
for each renovation situation and combining these concentrations with age-dependent and building
specific activity patterns to estimate the resulting blood lead levels. Finally, these levels may be linked to
appropriate, measureable, and quantifiable health effects via concentration-response functions, derived
from empirical studies (if available and appropriate for the analysis). Incremental blood lead level
changes and health effect changes resulting from the renovation will be developed by subtracting the
estimated blood lead level due to background exposure from the full renovation model simulation.
Each individual has some exposure to lead via sources other than renovation activities, and this
background exposure will vary across the population. The specific health effects, such as incremental IQ
change in children, can be estimated by comparing the expected background effect to the expected
post-renovation effect via appropriate concentration-response functions. The change in IQ for each unit
change in blood lead (PbB) level is proportionately larger for children with lower background exposures.
Thus, background exposure is an important piece in estimating any lead-related health effect.
Variability would be accounted for by performing a probabilistic Monte Carlo analysis as part of the
overall exposure analysis framework (see Appendix A). First, EPA would determine a set of scenario
variables and representative ranges of values for these variables, to which the exposure estimates are
highly sensitive. EPA would also determine a range of values for each scenario variable (see Appendix B).
EPA performed some preliminary analyses to determine the impact of different variables on incremental
IQ and PbB level changes for children. This analysis helped identify variables that require
characterization using distributions (higher impact variables) versus variables that can be parameterized
using a single point estimate. Additional information is provided in Appendix C.
To perform the Monte Carlo analysis, each scenario would be run 20,000 times (where each run is
referred to as an "iteration"). Preliminary testing indicates that 20,000 iterations would be appropriate
in order to optimize the combination of accuracy and run-time efficiency. Each different scenario would
be defined by a specific combination of values for the scenario variables listed in Appendix B. For each of
the 20,000 iterations within each scenario, a value would be selected randomly for every Monte Carlo
variable from its associated distribution. After all iterations are completed, the exposure estimates
would be averaged to estimate a mean value, and other summary statistics, for that exposure scenario.
7 For information on how EPA has previously assessed hazards using the hazard standard approach please refer to the preamble
to the 2008 final Renovation, Repair, and Painting Rule (US EPA, 2008).
US EPA A-2 May 2014
-------
Framework for Identifying and Evaluating Lead-Based Paint Hazards from Renovation, Repair, and Painting
Activities in Public and Commercial Buildings
Note that each model iteration is essentially a hypothetical but demographically representative
individual's exposure within that scenario. The distribution of characteristics of these 20,000
hypothetical individuals per scenario will, by design, mimic the expected distributions of those
characteristics in the US population. After all 20,000 hypothetical individuals have been generated and
evaluated (for expected background blood lead levels and for expected post-renovation blood lead
levels), the mean and other summary statistics for these blood lead levels (or for their health effect
endpoints, via concentration-response functions) will be calculated. At this point these summary
statistics are describing the distribution of the population-level response to the particular exposure
scenario.
To estimate the total population exposure, EPA would determine the probability of occurrence of each
exposure scenario based on both population and building survey data. The final population exposure
estimate represents the population-wide estimate of exposure to lead due to P&CB renovations. EPA
would use the distributions from the Monte Carlo analysis among the various scenarios to evaluate
whether a hazard exists inside P&CBs.
EPA would conduct sensitivity analyses to determine the influence of each variable on the model
outputs. Uncertainties associated with many variables would also be documented and described in the
analysis.
^ < V* Is Being ? msidered for Use in the Full Analysis
EPA has three specific goals for the models expected to be used in the analysis:
1. To use peer-reviewed models wherever possible,
2. To avoid models that require more inputs than can be realistically determined using existing data,
and
3. To remain consistent with the 2008 RRP analysis while updating any models that have been
improved since then.
For the exterior analysis (Figure 3), EPA will likely use the AERMOD (US EPA, 2009) modeling system to
model the dispersion of the lead-containing dust downwind of the renovated building. AERMOD is an
EPA-recommended (http://www.epa.gov/scram001/dispersion prefrec.htm) Gaussian dispersion model
that incorporates meteorological and land-use characteristics, and adjustments for obstructions.
To estimate the indoor dust lead loadings resulting from the airborne renovation-derived lead at nearby
receptors (Figure 4), a mass-balance indoor dust model that the Science Advisory Board (SAB) originally
reviewed in 2010 (US EPA, 2010b) is likely to be used. The model incorporates air exchange between the
indoor and outdoor environments, indoor deposition, indoor re-suspension, cleaning frequency and
efficiency, and track-in of dust from outdoor soil.
A Monte Carlo model can be developed using the Python programming language. The model would
consist of several modules that perform lookups, sampling, and calculations based on input files
specifying the distributions of parameter values, such as the AERMOD air concentration files, indoor
dust lead and outdoor soil lead concentration files, and the Leggett response surface tables. Each lookup
US EPA
A-3
May 2014
-------
Framework for Identifying and Evaluating Lead-Based Paint Hazards from Renovation, Repair, and Painting
Activities in Public and Commercial Buildings
and calculation step would be designed as a separate function in the code. These modules would
provide the overall model with data about how environmental concentrations vary over time, and how
hypothetical individuals of different ages come into contact with these environmental concentrations
based on age-driven activity patterns and exposure factors. After verifying the calculations being
performed, test cases would be run in the model, and trends across model scenarios would be
examined, for the purposes of further quality assurance, to ensure the trends and magnitudes of results
match expectations. The code would use a random number seed to ensure reproducibility of the model
results.
EPA will likely use the Leggett model (Leggett, 1993) as the PbB model for estimating both childhood
and adult changes in PbB level. The Leggett model, in contrast with the EPA Integrated Exposure Uptake
Biokinetic (IEUBK) model, allows incorporation of short-term (less than one month) changes in exposure
levels and modeling of both children and adults. EPA is considering updating the tissue volumes and key
age-dependent pharmacokinetic parameters used in the Leggett model. If these revisions are made,
model performance will likely be validated using key bone and PbB data sets, including the National
Health and Nutrition Examination Survey (NHANES).
Figure 3. Modeling Framework for Exterior Renovation Activities
20,000 Realizations
Compile AH
Health Effect
Summary
Statistics
Total soil >
concentration time
series y
f Total air
concentration time
v. series
Fraction of Paint
Emitted
Renovated Building
Characteristics
Leggett Blood Lead
Model
Blood Lead
Geometric
Standard Deviation
Fate and Transport
Modeling Oatasets
(AERMOD)
Obstruction
Adjustments
Residence/P&CB
Receptor
Characteristics
Background Media
Concentrations
Total in door dust ^
concentration time
Total indoor dust
loading time series
For Each Monte Carlo Realization:
Changes in Health
Effects
US EPA
A-4
May 2014
-------
Framework for Identifying and Evaluating Lead-Based Paint Hazards from Renovation, Repair, and Painting
Activities in Public and Commercial Buildings
Figure 4. Modeling Framework for Interior Renovation Activities
20,000 Realizations
For Each Monte Carlo Realization:
Renovati on-related
air concentration
Blood Lead Averaged
Over Critical Window
v Child and Adult y
Population
Distribution of B
. Lead
Total soil
Compile All
lealth Effect
Summary
.Statistics y
Changes in Health
Effects
Total indoor dust
concentration tinre
f Total afr
concentration time
V series ,
Renovation-related
dust loading time
V series
Renovated Building
Characteristics
Leggett Blood Lead
Model
Control options
Blood Lead
Geometric
Standard Deviation
Time spent In
public/commercial
building
Background Media
Concentrations
Person allowed in
renovation room
Cleaning
Efficiency and
Frequency
Job
Characteristics
Renovation-related
indoor dust loading
in multiple rooms
Indoor Dust Loading -
Concentration
Conversion
Parameters
US EPA
A-5
May 2014
-------
Framework for Identifying and Evaluating Lead-Based Paint Hazards from Renovation, Repair, and Painting
Activities in Public and Commercial Buildings
References
Leggett RW. (1993). An age-specific kinetic model of lead metabolism in humans. Environmental Health
Perspectives 101:598-616.
NRC 2009. Science and Decisions: Advancing Risk Assessment (The "Silver Book"). National Academy of
Sciences. Washington, DC: National Academy Press.
http://www.nap.edu/catalog.php?record_id=12209
US EPA. (1992). Guidelines for Exposure Assessment. U.S. Environmental Protection Agency, Risk
Assessment Forum, Washington, DC, EPA/600/Z-92/001.
US EPA. (2009). AERMOD modeling system "version 09292" [Software], US Environmental Protection
Agency.
US EPA. (2010b). Proposed Approach for Developing Lead Dust Hazard Standards for Residences. SAB
Consultation Draft. Washington, DC: US Environmental Protection Agency, Office of Pollution
Prevention and Toxics (OPPT). Prepared for July 6-7, 2010 consultation.
US EPA. (2014). Framework for Human Health Risk Assessment to Inform Decision Making. U.S.
Environmental Protection Agency, Risk Assessment Forum, Washington, DC, EPA/100/R-14/001.
US EPA
A-6
May 2014
-------
Framework for Identifying and Evaluating Lead-Based Paint Hazards from Renovation, Repair, and Painting
Activities in Public and Commercial Buildings
Appendix B. Variables Being Considered for Use in the Analysis and Designation
of Scenario or Monte Carlo Variables in the Model Framework
Variable
Scenario or Monte Carlo (MC)
Exterior
Renovation Characteristics:
Interior
Renovation Activity
Scenario
Scenario
Containment options
Scenario
Scenario
Containment efficiency
MC
MC
Size of crew performing renovation
MC
Fraction emitted as aerosol by renovation activity
MC
Fraction emitted as bulk by renovation activity
MC
N/A8
Work room loading per square foot disturbed, post
work
Work room loading per square foot disturbed, post
cleaning
N/A9
MC
MC
Renovated building characteristics:
Use type of renovated building
Scenario
Job size within renovated building
Scenario ,==
Location of person within building (in workspace or
adjacent)
N/A
Scenario ,=
Time spent in building
B
MC 4—1
Location of job in building
B
MC ^—1
Size of renovated building
Scenario
N/A
Fraction of renovated building that is glass
B
MC ^—1
N/A
Containment Area
B
MC 4—1
Variable not Required for Interior Analysis
9 Variable not Required for Exterior Analysis
US EPA
A-7
May 2014
-------
Framework for Identifying and Evaluating Lead-Based Paint Hazards from Renovation, Repair, and Painting
Activities in Public and Commercial Buildings
Variable
Scenario or Monte Carlo (MC)
Exterior
Interior
Vintage of renovated building
Scenario
Scenario
Lead content in paint
b
MC ^—1
B
MC ^—1
Layers of paint on wall
b
MC ^—1
N/A
Climate region
b
MC ^—1
Rain frequency
b
MC ^—1
N/A
Obstruction Adjustment
b
MC ^—1
Receptor building characteristics:
Distance of receptor from renovated building
Scenario
N/A
Receptor use type
Scenario ,=
Same as renovated
building use type
Time spent in building
b
MC ^—1
Same as renovated
building time spent
Ceiling height
b
MC 4—1
B
MC 4—1
Area of building
b
MC 4—1
N/A
Receptor location (urban or rural): NEW
Scenario ,==
Location of receptor relative to renovated building
b
MC ^—1
Vintage of receptor building
b
MC 4—1
N/A
Height of receptor building
B
MC ^—1
Ground type (hard surface or soil): NEW
B
MC 4—1
Background Media Concentrations:
Background air concentrations
b
MC 4—1
o
MC 4—1
Background dust lead loadings by vintage
B
MC ^—1
B
MC ^—1
Background soil concentration by vintage
B
MC 4—1
B
MC 4—1
US EPA
A-8
May 2014
-------
Framework for Identifying and Evaluating Lead-Based Paint Hazards from Renovation, Repair, and Painting
Activities in Public and Commercial Buildings
Scenario or Monte Carlo (MC)
Variable
Exterior Interior
Dust Model Variables:
Cleaning frequency
MC
MC
Percentage of receptor that is carpeted
MC
Scenario
Floor dust lead loading to concentration conversion
MC
MC
Floor cleaning efficiency (carpets and hard surfaces)
MC
MC
Air Exchange Rate
MC
N/A
Particulate Track-in Rate
MC
Fraction of Particulate on Doormat
MC
MC
Fraction Bulk
MC
MC
Track-in Porosity
MC
MC
Track-in Depth
MC
MC
Track-in Rain Efficiency
MC
MC
Dust Model Penetration
Dust Model Resuspension
Dust Model Deposition
Volume of building
MC
MC
MC
MC
N/A
Blood Lead Model Variables:
Blood lead model inputs
MC
MC
Concentration Response Curves:
IQ equation
MC
MC
Other health endpoint equations
MC
MC
US EPA
A-9
May 2014
-------
Framework for Identifying and Evaluating Lead-Based Paint Hazards from Renovation, Repair, and Painting
Activities in Public and Commercial Buildings
Appendix C. Preliminary Analysis
EPA performed a limited set of preliminary analyses to determine the impact of different variables on
incremental IQ and blood lead (PbB) change estimates for children. EPA is considering similar analyses
for adults. The preliminary analysis was deterministic and does not fully account for variation in
scenarios evaluated. Therefore, preliminary analysis results are not representative of all scenarios that
could be analyzed. This analysis provided an initial assessment of the variables that may require
characterization using distributions (higher impact variables) versus those variables that may be
parameterized using a single point estimate. See Appendix B for the proposed characterization of
variables as scenario - requiring only a point estimate at each proposed value for the scenario - or
Monte Carlo (MC) - requiring an entire distribution to describe the important range of the variable. EPA
will determine whether these preliminary results are reproducible once more robust analyses are
performed. Therefore, the preliminary findings reported herein should not be construed as final and
may change pending the subsequent and more comprehensive analyses.
Figures 5 and 6 show the range of results from the interior analysis, where trends based on the time
spent in the building and the square footage of the room are shown. The data points are separated by
color according to whether or not the exposed child was allowed in the workspace during the
renovation. These metrics are represented as "incremental" changes, indicating they represent the
change related to the renovation. The metrics are incremental IQ change, averaged over ages 1 through
7, for a child who experiences a renovation at age 1. In other words, the hypothetical child experiences
the renovation at age 1, which results in a short-term spike in blood lead. The child's blood lead then
returns to the pre-renovation level, typically within 1 year. For this preliminary analysis, the child's
AVERAGE blood lead, over ages 1 through 7, is then used to approximate the actual long-term impact on
the child's IQ.
Figure 5 shows that the incremental IQ changes and PbB levels are highly sensitive to the amount of
time the exposed person spends in the building. The amount of time that children spend in P&CBs will
vary. Most P&CBs where young children would spend longer amounts of time are COFs, and are already
regulated. The ranges of time spent values in Figure 5 represent a hypothetical range from low to high,
and would be refined for the final analysis.
Figure 6 shows that exposure concentrations that lead to incremental IQ and PbB changes tend to
increase with decreasing room size, although this is not strictly observed across all the different room
sizes. In addition to the room sizes varying, the ceiling heights were also varied in an effort to represent
actual room types. Examples include hospital rooms (smaller area, higher ceiling), offices (smaller area,
lower ceiling), and larger retail spaces (larger area, higher ceiling). Thus, the trend is due to both the
floor area and the ceiling height.
US EPA
A-10
May 2014
-------
Framework for Identifying and Evaluating Lead-Based Paint Hazards from Renovation, Repair, and Painting
Activities in Public and Commercial Buildings
Figure 5. Trends in Incremental IQ Change & Time Spent for Children Inside P&CBs (Preliminary Analyses)
• Allowed in Workspace
~ Not Allowed In Workspace
Time Spent Key:
1: 1 hour per month
2:1 hour per week
3:1 hour per day, 2 days per week for four
weeks
4: 2 hours per day, 3 days per week, for 8
weeks
5: 59 hours straight
6: 2.75 hours per day, 7 days a week
7: 24 hours a day, 1 day per week
1
2
i l
3 4 5
Time Spent Category
Figure 6. Trends in Incremental IQ Change & Room Size for Children Inside P&CBs (Preliminary Analyses)
1.8
1.6
2* 1.4
C
0
112
QO 1-0
£
fO
u 0.8
a
™ 0.6
E
01
I °-4
c
- 0.2
0.0
• Allowed in Workspace
~ Not Allowed in Workspace
~t m "w
\ : :
1 » !
• i _i
2 3 4
Room Category
t
i
6
Room Size Key:
1: 15 m2
2: 30 m2
3: 35 m2
4: 135 m2
5: 825 m2
6: 1600 m2
US EPA
A-ll
May 2014
------- |