United States
Environmental Protection
Agency
Office of
Pollution Prevention and Toxics
Washinqton D.C., 20460
EPA747-R-95-006
July. 1995
REVIEW OF STUDIES ADDRESSING
LEAD ABATEMENT
EFFECTIVENESS
-------
July 24, 1995
EPA 747-R-95-006
FINAL REPORT
REVIEW OF STUDIES ADDRESSING
LEAD ABATEMENT EFFECTIVENESS
Prepared by
Battelle Memorial Institute
for
Technical Programs Branch
Chemical Management Division
Office of Pollution Prevention and Toxics
Office of Prevention, Pesticides, and Toxic Substances
U.S. Environmental Protection Agency
Washington, D. C. 20460
U.S. Environmental Protection Agency
Region 5, Library (PL-12J)
77 West Jackson Boulevard. 12th Floor
Chicago, !L 60604-3590
Recycled/Recyclable • Printed with Vegetable Based Inks on Recycled Paper (20% Postconsumer)
-------
DISCLAIMER
Mention of trade names, products, or services does not convey, and should not be interpreted
as conveying, official EPA approval, endorsement, or recommendation.
',,:• ,*',»•: •-'«,,,,;. - . , :
;' 'J •". ,''.],.'
-------
AUTHORS AND CONTRIBUTORS
This study was funded and managed by the U.S. Environmental Protection Agency. The
review was conducted by Battelle Memorial Institute under contract to the Environmental Protection
Agency. Each organization's responsibilities are listed below.
Battelle Memorial Institute (Battelle)
Battelle was responsible for conducting the literature search, obtaining and reviewing the
identified articles and reports, developing the conclusions and recommendations derived from the
review, and preparing this report. In addition, Battelle developed and utilized a biokinetic model of
the mobilization of bone-lead stores following an intervention.
U. S. Environmental Protection Agency (EPA)
The Environmental Protection Agency was responsible for managing the review, providing
guidance on the objectives for the review and report, contributing to the development of conclusions
and recommendations, and coordinating the EPA and peer reviews of the draft report. In addition,
EPA provided access to study results not yet available in the general literature. The EPA Work
Assignment Managers were Bradley D. Schultz and Samuel Brown. The EPA Project Officers were
Phil Robinson and Jill Hacker. Cindy Stroup, Barbara Leczynski, Dan Reinhart, Phil Robinson, and
John Schwemberger reviewed the draft report and provided comments.
-------
This page intentionally left blank.
-------
EXECUTIVE SUMMARY
INTRODUCTION
This report is a comprehensive review of the scientific literature regarding the effectiveness of
lead hazard intervention. One use of this review is to aid in assessing the potential benefits of Title X
rule-making activities. In this report, a lead hazard intervention is defined as any non-medical activity
that seeks to prevent a child from being exposed to the lead in his or her surrounding environment.
An intervention, therefore, may range from the in-home education of parents regarding the dangers of
a young child's hand-to-mouth activity to the abatement of lead-based paint. Interventions include
activities that attempt to remove or isolate a source of lead exposure, as well as activities that attempt
to reduce a child's lead exposure by modifying parental or child behavior patterns.
A number of studies have examined the effectiveness of abating the environment of lead
hazards associated with lead-based paint, elevated dust lead, and elevated soil lead. These studies have
emphasized hand-to-mouth activity as the primary pathway of childhood lead exposure and utilized
interventions that targeted this pathway. Generally, they have assessed whether a particular
intervention strategy effectively lowered an affected child's body-lead burden or the levels of lead in
his or her environment. Sixteen such studies are summarized in this report. In total, these studies
spanned 13 years, from 1981 to 1994. In all 16 cases, the interventions targeted primarily the child's
residential environment. Also, the studied interventions principally sought "secondary" rather than
"primary" prevention (e.g., assessing the effectiveness of lead hazard intervention on already exposed
rather than unexposed children). Ten of the 16 studies focused on the abatement of lead-based paint as
a primary form of intervention, five studies focused on dust or educational intervention, and one
study focused on soil abatement.
It is often infeasible to directly assess health benefits following an intervention because many
such benefits are subtle and, as such, are complicated and costly to measure directly. In this report,
therefore, the blood-lead concentrations of exposed children are utilized as the primary measure of
intervention efficacy. Blood-lead concentration can serve as a good surrogate health endpoint due to
the established association between elevated blood-lead levels and adverse health effects.
Page i
-------
MAJOR FINDINGS
The literature is very limited in its extent. However, it does indicate that blood-lead
concentrations declined after lead hazard intervention, at least for children with blood-lead levels
above 20 pg/dL.
The available literature only covers some of the intervention types and methods used in
practice. However, declines on the order of 18-34% were measured in exposed children's
blood-lead levels 6 to 12 months following a variety of intervention strategies. The evidence
for blood-lead concentration declines after intervention among children with pre-intervention
levels less than 20 jug/dL is mixed. With respect to changes in dust-lead levels, the declines
following intervention were larger than the blood-lead level declines. However, dust levels are
of limited relevance as a measure of actual exposure or health effects.
Four of the identified studies also simultaneously traced changes in blood-lead concentration
among a population of children not receiving the studied intervention strategy. The effect of
their interventions may then be estimated as the difference in the decline recorded for the
study population and that for the "control" population. The four studies examined distinct
intervention strategies: the abatement of damaged lead-based paint, the abatement of soil at
elevated lead levels, regular dust control measures, and in-home educational outreach efforts.
Using this measure, these four studies each would estimate the effect of their intervention to
be approximately 15%. That is, those receiving the intervention were better off than those
receiving partial or no interventions.
The evidence clearly indicates that short-term increases in exposed children's blood-lead
concentrations may result when abatements are performed improperly.
Declines in blood-lead concentrations followed several removal methods, as well as some
encapsulation and enclosure methods. In contrast, dry scraping and sanding with HEPA
vacuum attachments were both reported to produce considerable elevations in the blood-lead
levels of exposed children. Failure to clean-up post-abatement debris was also associated with
residential dust and blood lead elevations.
Page ii
-------
There is simply insufficient information available to identify a particular intervention strategy
as markedly more effective than others.
Evolution in the techniques associated with lead hazard control make comparison of the
effectiveness of different practices difficult. The literature cites comparable reductions in
blood-lead concentration resulting from the abatement of lead-based paint, dust at elevated
lead levels, and soil at elevated lead levels. Moreover, declines in blood-lead levels after in-
home educational efforts were observed in the same range as the other interventions, at least
up to one year following intervention. As for long-term effectiveness, there is virtually no
data on the effectiveness of any lead hazard intervention beyond one year following
intervention.
Information is especially lacking on the effectiveness of interventions for children with blood-
lead concentrations below 20 pg/dL. Also missing is data on effectiveness beyond one year after
intervention and on the efficacy achieved by trying to prevent elevated blood-lead concentrations
before they occur.
DISCUSSION
When considering the effectiveness of an intervention, it is important to recognize that
childhood lead exposure stems from a number of media (e.g., paint, soil, interior house dust, exterior
dust) across a range of environments (e.g., child's residence, school, playground, friend's residence).
Unless an intervention targets all the sources of a child's lead exposure, therefore, even an
intervention that fully abates the targeted source will not produce a 100% decline in the child's blood-
lead concentration. If other sources of lead remain unaffected by the intervention, lead exposure may
continue and the child's blood-lead concentration may remain elevated.
Another factor, bone-lead mobilization, can also cause blood-lead concentrations to remain
elevated following interventions that reduce the targeted lead exposure. An intervention which reduces
a child's lead exposure results in the mobilization of bone-lead stores into the blood. The available
scientific information on bone lead mobilization is minimal, but a simple model of this mobilization
was constructed in an effort to assess its impact. Bone lead mobilization modeling results in this
report suggest that observed declines of as little as 25% in a child's blood-lead concentration might be
possible for 6 months following an intervention which completely eliminates new lead exposure. The
results also suggest that 25% declines in blood-lead concentrations which are observed at least 12
Page iii
-------
months after an intervention indicate the intervention was less than 100% effective in reducing the
child's total lead exposure. However, mobilization of bone-lead stores is another reason why
prevention of lead poisoning before it ever occurs is important.
Finally, in planning future studies of lead hazard intervention effectiveness, the timing of
post-intervention measurements should be carefully considered. Environmental and blood lead
measurements taken one year after intervention are usually appropriate because both seasonal
variability and the effects of bone-lead mobilization are minimized. The timing of earlier measures
should be based on such factors as the importance of observing transient elevations in blood-lead
concentrations should they occur shortly after intervention, the importance of establishing a baseline
for assessing recontamination of environmental media, and a trade-off between the effects of seasonal
variability and bone-lead mobilization. Consideration should also be given to the population of
children examined by future studies. There is a particular lack of information on the effectiveness of
lead hazard intervention among children with blood-lead concentrations at or below 20 jig/dL Absent
too is information on effectiveness at time periods beyond one year. Perhaps most importantly,
information is lacking on the efficacy achieved by preventing elevated blood-lead concentrations
before they occur. Fortunately, some on-going intervention studies are examining these populations,
and should provide valuable information.
Page iv
-------
TABLE OF CONTENTS
Page
EXECUTIVE SUMMARY i
1.0 INTRODUCTION TO THE PROBLEM 1
1.1 ORGANIZATION OF THE REPORT 2
2.0 ASSESSING INTERVENTION EFFICACY 3
2.1 MEASURES OF INTERVENTION EFFICACY 3
2.2 IMPLICATIONS OF SOURCE APPORTIONMENT 5
2.3 INFLUENCE OF EXISTING BONE-LEAD STORES 11
2.3.1 Review of Evidence for Bone-Lead Mobilization 12
2.3.2 Modelling Bone-Lead Mobilization 13
2.3.2.1 Development of Parameter Values 13
2.3.2.2 A Simple Model for Bone-Lead Mobilization 16
2.3.3 Conclusions Regarding Influence of Bone-Lead Stores 18
3.0 REVIEW OF SCIENTIFIC EVIDENCE 23
3.1. LEAD HAZARD INTERVENTION STUDIES 25
3.1.1. Baltimore Dust Control Study 25
3.1.2 1982 St. Louis Retrospective Paint Abatement Study 27 .
3.1.3 Baltimore "Traditional"/"Modified" Paint Abatement Study 28
3.1.4 Boston Retrospective Paint Abatement Study 30
3.1.5 Baltimore Experimental Paint Abatement Studies 32
3.1.6 Central Massachusetts Retrospective Paint Abatement Study 34
3.1.7 Seattle Track-In Study 37
3.1.8 1990 St. Louis Retrospective Paint Abatement Study 38
3.1.9 Boston Three-City Soil Abatement Study 40
3.1.10 HUD Abatement Demonstration (HUD Demo) Study 44
3.1.11 Comprehensive Abatement Performance (CAP) Study 45
3.1.12 Milwaukee Retrospective Paint Abatement Study 47
3.1.13 New York Chelation Study 47
3.1.14 Milwaukee Retrospective Educational Intervention Study 49
3.1.15 Granite City Educational Intervention Study 50
3.1.16 Milwaukee Prospective Educational Intervention Study 53
3.2 SUMMARY OF SCIENTIFIC EVIDENCE 54
4.0 CONCLUSIONS 65
5.0 RECOMMENDATIONS FOR FUTURE INTERVENTION STUDIES 71
6.O REFERENCES 75
6.1 ADDITIONAL SOURCES OF INFORMATION 81
APPENDIX A: ABSTRACTS OF STUDIES ADDRESSING THE
EFFICACY OF LEAD HAZARD INTERVENTION A-l
Page v
-------
TABLE OF CONTENTS (continued)
Page
APPENDIX B: REVIEW OF ABATEMENT METHODS ASSOCIATED WITH
TEMPORARY INCREASES IN BLOOD-LEAD LEVELS B-l
LIST OF TABLES
Table 2-1. Baseline Lead Intake for a Two-Year-Old Child 9
Table 2-2. Lead Intake for a Two-Year-Old Child in an Urban Environment 9
Table 2-3. Lead Intake for a Two-Year-Old Child in a Non-Urban House with
Interior Lead-Based Paint 10
Table 2-4. Parameter Values Used in the Two Compartment Model of Bone-Lead
Mobilization 14
Table 2-5. Length of Time (in Months) Bone-Lead Stores Can Maintain PbB at
75% of Pre-Intervention Levels 19
Table 3-1. Summary Information Table for Identified Lead Intervention Studies 57
Table 3-2. Summary of Blood-Lead Concentration Results for Identified Lead
Hazard Intervention Studies 58
Table 3-3. Summary of Other Body-lead Burden Results for Identified Lead
Hazard Intervention Studies 60
Table 3-4. Summary of Environmental Media Results for Identified Lead Hazard
Intervention Studies 61
Table 4-1. Summary of Intervention Efficacy for Identified Lead Hazard
Intervention Studies 66
Table 5-1. Selection of Early Timepoint for Measuring Intervention Effectiveness 73
LIST OF FIGURES
Figure 2-1. Micro-environments and lead hazards to which a child may potentially
be exposed 6
Figure 2-2. Pathway diagram detailing sources of lead exposure and their
interactions within a particular micro-environment 7
Figure 2-3. Two compartment model of bone-lead mobilization 16
Figure 2-4. Blood-lead concentration versus time following a reduction in lead
uptake 17
Figure 2-5. Blood-lead concentration versus time following interventions which are
50%, 75%, and 100% effective 19
Figure 2-6. Percentage bias in blood-lead concentration due to masking versus
time following interventions which are 25%, 50%, and 75% effective 21
Figure 3-1. Timeline for the identified intervention studies 24
Figure 3-2. Arithmetic mean blood-lead concentration (jtg/dL) since abatement by
study group (Baltimore Dust Control Study) 26
Figure 3-3. Percentage of experimental homes with arithmetic mean dust-lead
loadings (/xg/ft2) in the defined range (Baltimore Dust Control Study) 26
Page vi
-------
TABLE OF CONTENTS (continued)
Page
Figure 3-4. Pre- and post-identification arithmetic mean blood-lead concentration
(jtg/dL) by status of residence abatement (1982 St. Louis
Retrospective Paint Abatement Study) 28
Figure 3-5. Arithmetic mean blood-lead concentration (jtg/dL) post-abatement by
population considered (Baltimore "Traditional"/"Modified" Paint
Abatement Study) 30
Figure 3-6. Arithmetic mean blood-lead concentration (/*g/dL) post-abatement by
study population considered (Boston Retrospective Paint Abatement
Study) 31
Figure 3-7. Geometric mean floor dust-lead loading (/tg/ft2) post-abatement by
abatement practice performed (Baltimore "Traditional"/"Modified"
and Experimental Practices Studies) 33
Figure 3-8. Distribution of the absolute change in blood-lead concentrations
(Central Massachusetts Retrospective Paint Abatement Study) 36
Figure 3-9. Pre-abatement and post-abatement blood-lead concentration by pre-
abatement blood-lead level (Central Massachusetts Retrospective Paint
Abatement Study) 36
Figure 3-10. Pre- and post-identification arithmetic mean blood-lead concentration
Oxg/dL) by status of residence abatement (1990 St. Louis
Retrospective Paint Abatement Study) 39
Figure 3-11. Arithmetic mean blood-lead concentration (/^g/dL) across sampling
rounds and experimental groups, Phase I (Boston 3-City Soil
Abatement Project) 41
Figure 3-12. Arithmetic mean blood-lead concentration (/xg/dL) across sampling
rounds and experimental groups, Phase I and II (Boston 3-City Soil
Abatement Project) 42
Figure 3-13. Arithmetic mean environmental lead level (for Study Group) across
sampling rounds (Boston 3-City Soil Abatement Project) 43
Figure 3-14 Percentage of components successfully abated to HUD Guidelines
Standards by abatement method and sampling location (HUD Demo
Study) 45
Figure 3-15. Estimated geometric mean dust-lead and soil-lead concentration (^tg/g)
in typical abated and control homes by sampling location (CAP
Study) 47
Figure 3-16. Arithmetic mean blood-lead concentration (/xg/dL) by population
considered (New York Chelation Study) 49
Figure 3-17. Initial and follow-up arithmetic mean blood-lead concentration (/xg/dL)
(Granite City Educational Intervention Study) 51
Figure 3-18. Absolute (on left) and percent change in blood-lead concentrations at
4-month and 12-month follow-up (Granite City Educational
Intervention Study) 52
Figure 3-19. Change in blood-lead concentration at 4-month and 12-month follow-
up plotted against initial blood-lead concentration (Granite City
Educational Intervention Study) 52
Page vii
-------
TABLE OF CONTENTS (continued)
Figure 4-1. Summary of blood-lead concentration results for identified lead hazard
intervention studies 67
Figure 4-2. Effect of lead hazard intervention as measured by declines in
children's blood-lead concentration 67
Page viii
-------
1.0 INTRODUCTION TO THE PROBLEM
This report is a comprehensive review of the scientific literature regarding the effectiveness of
lead hazard intervention. In addition, this review is intended to aid in assessing the potential benefits
of Title X rule-making activities. In this report, a lead hazard intervention is defined as any non-
medical activity that seeks to prevent a child from being exposed to the lead in his or her surrounding
environment. An intervention, therefore, may range from the abatement of lead-based paint to the
education of parents regarding the dangers of a young child's hand-to-mouth activity. Interventions
include activities that attempt to remove or isolate a source of lead exposure, as well as activities that
attempt to reduce a child's lead exposure by modifying parental or child behavior patterns.
In recent years substantial effort has focused on the development and demonstration of
methods for reducing childhood lead exposure and body-lead burden by applying interventions which
address environmental lead hazards. It is expected that these interventions will both prevent further
exposure and produce positive health outcomes. The extent to which the scientific literature supports
this expectation is characterized in this report. Currently available scientific information was
compiled on the effectiveness of lead hazard intervention in reducing childhood lead exposure. These
studies specifically address the efficacy of intervention strategies employed to reduce exposure to lead-
based paint, elevated soil-lead levels, and elevated dust-lead levels. Moreover, these studies all
sought to characterize interventions targeting already exposed children rather than unexposed children.
As such, the studies measured "secondary" rather than "primary" prevention efforts. The literature
currently reports no primary prevention studies, though some are being conducted.
Intervention has usually taken the form of abatements emphasizing the removal or enclosure
of the source of lead exposure. Three recent studies identified methods involving either
encapsulation, enclosure, or removal of lead-contaminated paint, dust, and soil. The Department of
Housing and Urban Development (HUD), for example, conducted the Lead-Based Paint Abatement
Demonstration Study to assess lead-based paint hazard abatement. The U.S. Environmental
Protection Agency (EPA) subsequently conducted the Comprehensive Abatement Performance (CAP)
Study to characterize the long-term efficacy of the paint abatement methods used in the HUD
Demonstration, and the Three City Urban Soil Lead Abatement Demonstration (3-City) Project to
investigate whether removal of leaded soil and dust from residential environments decreases the
blood-lead concentration of children living in those residences.
Page 1
-------
Even if proven efficacious, applying the source isolation or removal methods cited in these three
studies to our Nation's housing stock could prove to be prohibitively expensive. For this reason, a
number of studies have been conducted, or are now under way, to examine the efficacy of low-cost
abatement or in-place management intervention methods.
1.1 ORGANIZATION OF THE REPORT
Following in Section 2.0 is a discussion of the measures for assessing the efficacy of an
intervention strategy. Two issues which impact the effectiveness of an intervention, bone-lead
mobilization and source apportionment, are also considered. Section 3.0 is a review of the scientific
evidence. Specifically, a number of studies are discussed that have examined the extent to which lead
hazard intervention results in reduced lead exposure and lower blood-lead levels in children. While
not exhaustive, these studies were found to contain the most pertinent information. Sections 4.0 and
5.0 present the conclusions and recommendations derived from the review. References, an Appendix
A which contains abstracts of the studies, and an Appendix B containing a separate attachment
regarding hazardous abatement methods, are included at the end of the report.
Page 2
-------
2.0 ASSESSING INTERVENTION EFFICACY
Assessing the effectiveness of a lead hazard intervention strategy is not always a simple
undertaking. There are a wide range of parameters which may be measured to quantify the success
or efficacy of an intervention. Section 2.1 presents several measures of effectiveness and outlines the
parameters this report will rely upon in its review. There are also critical confounding factors which
make the comparison of one intervention study to another challenging. Two such potential factors,
• source apportionment of lead exposure, and
• existing bone-lead stores,
are discussed in Sections 2.2 and 2.3, respectively.
2.1 MEASURES OF INTERVENTION EFFICACY
In reviewing a series of studies assessing intervention effectiveness, there are a variety of
environmental, behavioral, and physiological parameters which may be measured to quantify efficacy.
The goal is to utilize a measure which adequately reflects the potential benefit or detriment resulting
from the intervention. Young children are the population most at risk from lead exposure and, as a
result, are the target group for most of the intervention procedures commonly employed. A suitable
measure of efficacy, therefore, should reflect the impact of the intervention on affected children.
Interventions are not performed merely to reduce or eliminate environmental lead levels, the aim is
always to positively impact the health of children or adults.
It would be ideal to precisely measure particular health outcomes, such as decreased learning
deficits or increased motor coordination, among children benefitting from intervention. In fact, one
study of moderately exposed children (detailed in section 3.1.13) did document increased cognitive
function six months following a set of interventions. Unfortunately, identifying health outcomes
following intervention is not always feasible. Such outcomes may not manifest themselves for a long
period of time. Many of the health benefits are subtle and, as such, are complicated and costly to
measure and verify. This assessment is made more difficult when considering interventions targeted
at children with low to moderate lead exposure. Recognizing lead-related health outcomes is
particularly difficult if the child was not exhibiting symptoms of lead poisoning before the
intervention was initiated. In such instances, intervention efficacy may have to be assessed using tests
of learning aptitude or intelligence quotient (IQ). The small differences recorded usually for these
Page 3
-------
measures require larger sample sizes to statistically verify the benefit following intervention. For
these reasons, it may be difficult and expensive to perform a sufficiently large study to demonstrate
an intervention's effectiveness in this manner. As will be seen in the reviews in Section 3.0, the
majority of identified studies did not measure specific health outcomes associated with their
interventional practice.
Given these limitations, measures of body burden such as blood-lead concentration may serve
as alternative biomarkers of lead exposure and intervention effectiveness. Such measures indicate the
extent to which the intervention impacts affected children and serve as a biomarker of lead exposure.
There is extensive evidence that body-lead burden is associated with lead levels in environmental
media (USEPA, 1986; CDC, 1991). Three of the measures of body-lead burden reported in the
literature are bone-lead content, blood-lead concentration, and erythrocyte protoporphyrin (EP) blood
concentration. Bone-lead levels are considered to be reflective of cumulative exposure to lead, but
their determination is currently either expensive or invasive. The accuracy and representativeness of
bone-lead concentrations measured externally by an x-ray fluorescence (XRF) instrument is questioned
by many researchers. Blood-lead and EP levels can be more readily measured, but often reflect a
varying mixture of long-term and more recent exposure. There is an extensive body of literature
relating blood-lead concentrations to specific health outcomes, though much of it examined children
with higher levels of exposure (usually indicative of lead poisoning) (USEPA, 1986). Evidence,
however, has been reported suggesting that even low levels of exposure, as measured via blood-lead
levels, are associated with learning deficits (CDC, 1991; Goyer, 1993; Schwartz, 1994). The Centers
for Disease Control (CDC) state that, "Data indicate significant adverse effects of lead exposure in
children at blood-lead levels previously believed to be safe. Some adverse health effects have been
documented at blood lead levels at least as low as 10 ^g/dL" (CDC, 1991).1 Reductions in blood-
lead concentration, therefore, can be used as an effective measure of the results of intervention.
It is important to note that the effect of an intervention on blood-lead concentration is the
change in concentration above and beyond that due to factors other than the strategy itself. The
blood-lead concentration of a child may decrease due simply to random variation (regression to the
1
Though the documented association between reduced blood-lead concentration and positive health outcomes
is not based on interventional studies (with the exception of Ruff et al, 1993), it is strongly suggestive. Blood-
lead concentration is associated with environmental lead exposure and linked to health outcomes. Moreover,
temporal and sampling variability suggest the child's blood-lead levels have a 50% chance of increasing further.
Intervention, therefore, eliminates the possibility that the exposure will be aggravated.
Page 4
-------
awareness of the health risk from lead. These decreases are usually characterized by examining a
comparable control population. As a number of the identified studies did not examine a control
population, we report the blood-lead concentration reductions of the studies population as the
effectiveness of the strategy employed. When the results for control populations are available, we
also report the estimated "effect" of the intervention.
When it is impractical or inappropriate to measure blood-lead concentrations, levels in
environmental media can provide valuable information. Such measures cannot demonstrate the
intervention's impact on affected children, nor has a quantitative relationship been established between
environmental lead exposure and health outcomes. Environmental measures can, however, be used to
evaluate the effectiveness of a particular procedure in reducing or eliminating a targeted lead hazard.
In addition, there is extensive evidence that elevated lead levels in environmental media are associated
with elevated blood-lead concentrations (USEPA, 1986; CDC, 1991). Environmental measures,
however, do confirm the effectiveness of a particular procedure in reducing or eliminating a targeted
lead hazard. Environmental measures may also be particularly appropriate for comparing different
abatement procedures implemented on the same lead hazards and for assessing how successfully a
particular source of the lead hazard is reduced. For example, dust-lead loading measurements on
surfaces following their abatement can be used to demonstrate the superiority of one practice over
another.
In the reviews and discussions that follow, any result which appears useful in assessing
intervention efficacy will be reported. The primary measure used in this report, however, will be the
blood-lead concentrations of exposed children. This measure of body-lead burden is commonly
employed in assessing lead exposure and was collected in a majority of the identified studies.
2.2 IMPLICATIONS OF SOURCE APPORTIONMENT
When considering the effectiveness of an intervention strategy for reducing a child's body-
lead burden, it is important to recognize the many different avenues by which a child may encounter
lead. Intervention will be most efficacious if it targets those sources and pathways of lead exposure
most responsible for the child's elevated lead burden. An intervention can reduce a child's lead
exposure no more than that consistent with the source of exposure targeted. For example, if lead-
based paint accounts for 50% of a child's lead uptake, even the complete abatement of that paint can
be no more than 50% effective.
Page 5
-------
A child's daily lead exposure may occur across a number of micro-environments and lead
hazards. Here, a micro-environment is defined as a location where a child spends a portion of their
time. A lead hazard is defined as a potential source of lead exposure. Figure 2-1 presents an
example of the micro-environments and lead hazards to which a child may be exposed. Note that the
potential lead hazards can vary across micro-environments. The studies discussed in this report each
involve the abatement or intervention of lead hazards at the primary residence of the child. The
actual micro-environments and lead hazards that constitute a child's exposure depend upon a myriad
of factors such as community, socio-economic status, age of child, and time of year. One child may
play in leaded dust at his residence and leaded soil at school. A second child may obtain his or her
exposure solely from lead-based paint contaminated dust at a friend's house. As a result, an
abatement can reduce a child's lead exposure to a degree no greater than the degree to which the
targeted source of exposure represents a hazard to the child. That is, if lead-based paint in the
primary residence were responsible for 50% of a child's lead exposure, even a 100% effective
abatement of the paint can only reduce the child's lead burden by 50%.
Playground
Soil
i
Exterior
Dust
' )
Soil
Figure 2-1. Micro-environments and lead hazards to which a child may potentially
be exposed.
Page 6
-------
The efficacy of an intervention within a particular micro-environment is affected by the
pathways of lead exposure targeted by the intervention. Each of the environmental lead hazards can
be categorized as either an original source of lead or an environmental medium which acts as a
reservoir for lead deposition. Major sources of lead in the environment include paint, industrial
emissions, gasoline and solder. Lead from these sources can then accumulate in environmental media
such as soil, dust, air, food and water. When an intervention strategy includes abatement of one of
these environmental media, it is important to determine whether or not the media will become
recontaminated from unabated sources. For example, abating elevated dust-lead levels within a
residence will potentially result in only transient declines in the blood-lead levels of resident children.
If the unabated source of lead (e.g., lead-based paint) recontaminates the dust, the child's blood-lead
concentration may rapidly return to its original level. In a similar way, an intervention may target an
existing reservoir of lead (e.g., lead-based paint), but not the intermediate media by which children
are exposed to that reservoir (e.g., lead contaminated dust). Depending upon the rate at which the
environmental media are recontaminated, the effectiveness of the intervention may be delayed.
Regulations on lead solder in cans and leaded gasoline emissions have greatly reduced the
concentrations of lead in food and air. As a result, lead in paint, household dust and soil have been
identified as the principal current sources of lead exposure for children (CDC, 1991). Potential
interactions among these three sources and their associated pathways vary within a particular micro-
environment. Figure 2-2 demonstrates some of the potential interactions and mechanisms by which
children may be exposed.
Airborne Pb
Exterior Paint Pb
Exterior Dust Pb
Interior Paint Pb
Soil Pb
Mouthing
Behavior
Pbin
Soft Tissue
HandPb
Blood Pb
Interior Dust Pb
Pb
in Bone
Figure 2-2. Pathway diagram detailing sources of lead exposure and their
interactions within a particular micro-environment.
Page 7
-------
Each of the intervention strategies that are discussed in this report can be viewed as an
attempt to reduce or eliminate one or more of the pathways that appear in the above diagram. Within
a micro-environment, intervention should be targeted at those exposure pathways that have the
greatest impact on the health of the child. The success of an intervention strategy is ultimately
determined by the magnitude of the reduction in the body-lead burden of a child. Potentially, an
intervention can be successful in reducing a particular environmental lead exposure and yet produce
no positive impact in a child only marginally exposed to the abated lead hazard.
Though the sources of lead exposure responsible for a child's elevated body-lead burden often
depend upon the individual, some attempts have been made to examine those issues for a "typical"
child. One of the most commonly cited examples is presented in the EPA Air Quality Criteria for
Lead (USEPA, 1986) which attributed typical human lead exposure to lead in food, water, dust, and
inhaled air. Furthermore, a minimal or baseline level of lead was recognized for all children. This
baseline body burden was, in turn, attributed to baseline levels of lead in numerous environmental
media. A lead hazard, in this sense, would represent the amount of lead in a particular medium
which greatly exceeded its baseline level. The EPA report (USEPA, 1986) estimated a typical child's
daily lead exposure from a range of media. As this report was written in 1986, its estimates are now
somewhat dated. To update them, a second EPA technical report (USEPA, 1989) was utilized.
Three tables estimating typical exposures of a two-year-old child were developed from these two
reports. Each table estimates the daily lead exposure from air, food, water, dust, and soil for a
particular type of residence.
Table 2-1 describes the average daily lead intake for a two-year-old child who is exposed to
only baseline lead levels. The occupational exposure represents the dust lead brought home from
work by the child's parents. A child whose lead intake resembles this profile would probably not
benefit from traditional interventions such as lead-based paint abatement. Note that the largest
contribution to exposure is from elevated dust-lead levels. This is consistent with the many scientific
literature citations of dust-lead as the primary pathway of exposure (CDC, 1991; Amitai et al., 1991;
Roberts et al., 1991; Chisolm et al., 1985).
Page 8
-------
Table 2-1. Baseline Lead Intake for a Two-Year-Old Child
Environmental
Media
Inhale Air
Food, Water, Beverages
Dust - Household
Soil
Dust - Occupational
Pb
Concentration
0.10/yg/m3
0.0033 fjg/g
300 fjg/g
90 fjg/Q
1 50 /yg/g
Daily Amount
Consumed
5 m3
1500g
0.05 g
0.04 g
0.01 g
Total
Daily Pb
Intake
0.5//g
5.0 fjg
15 fjg
4.5 fjg
~\.5 fig
26.5 fjg
% of Total
Intake
2
19
57
17
6
100
A child residing in an urban area will usually have greater lead intake (Table 2-2). Although
the daily consumption rate remains unchanged for each of the environmental media in this table, the
daily lead intake has increased due to higher concentrations of lead in household dust and soil in the
urban environment. Even though the concentration of lead in air has increased from 0.10 to 0.75
^g/m3, lead intake from inhaled air only accounts for about 3% of a child's total lead intake in the
urban environment. Urban children whose lead exposure resembles the profile in Table 2-2 may
benefit from intervention of exposure pathways associated with household dust and/or soil.
Table 2-2. Lead Intake for a Two-Year-Old Child in an Urban Environment
Environmental Media
Inhale Air
Food, Water, Beverages
Dust - Household
Soil
Dust - Occupational
Pb
Concentration
0.75fjg/m3
0.0033 //g/g
1000 fjg/g
150O fjg/g
1 50 /yg/g
Daily Amount
Consumed
5 m3
1500 g
0.05 g
0.04 g
0.01 g
Total
Daily Pb Intake
3.75 fjg
5.0 fjg
50 fjg
60 fjg
1.5/yg
120.75/yg
%of
Total
Intake
3
4
42
50
1
100
Table 2-3 shows the lead intake profile for children whose non-urban primary residence
contains lead-based paint. Deterioration of this paint results in an average increase in dust-lead levels
of 2200 /*g/g over baseline values (from 300 /xg/g to 2500 jig/g). In total, interior lead-based paint
Page 9
-------
accounts for an additional 110 /^g of lead in a child's daily lead intake through the paint-to-dust
pathway. Abating leaded dust alone within a residence may induce only transient declines in the lead
burden of resident children. If lead-based paint recontaminates the dust, then the lead burden of
resident children may rise again shortly after dust abatement. In a similar fashion, if an intervention
only targets the lead-based paint and ignores the resulting elevated dust-lead levels, there may be no
body-lead burden reduction for quite some time. However, a thorough abatement of both lead-based
paint and elevated dust-lead can theoretically reduce a child's lead intake by 80%, assuming that
household dust-lead concentrations decline to 300 /xg/g-
Table 2-3. Lead Intake for a Two-Year-Old Child in a Non-Urban
House with Interior Lead-Based Paint
Environmental Media
Inhale Air
Food, Water, Beverages
Dust - Household
Soil
Dust - Occupational
Pb
Concentration
0.10/vg/m3
0.0033 /yg/g
2500 fjg/g
90 fjg/g
1 50 fjg/g
Daily Amount
Consumed
5 m3
1500 g
0.05 g
0.04 g
0.01 g
Total
Daily Pb Intake
0.5/vg
5.0//g
125 jig
4.5/vg
1.5/yg
136.5/yg
%of
Total
Intake
0
4
92
3
1
100
These tables do not address a number of exposure scenarios potentially relevant to children
including residential exterior lead-based paint or elevated soil-lead levels. Such scenarios could be
developed but require additional assumptions and data. More importantly, Tables 2-1, 2-2, and 2-3
emphasize lead exposure exclusively from the child's primary micro-environment. Given the
widespread presence of lead, it is entirely plausible that a child experiences lead exposure outside his
or her home. Tables of total exposure could be developed for such scenarios, but they require
assumptions about the child's exposure away from home. For example, one might assume a child
spends 2/3 of his or her time at home and 1/3 away (e.g., at pre-school). Further, the pre-school
could be assumed to expose the child to some fraction of the total exposure at home.
Thus, Tables 2-1 through 2-3 provide an oversimplified picture of lead exposure for children
While consideration of such simplified models is useful in better understanding the lead exposure
problem, most researchers readily recognize the more complicated nature of the problem and that
Page 10
-------
non-residential micro-environments such as pre-schools and playgrounds also contribute to lead
exposure.
2.3 INFLUENCE OF EXISTING BONE-LEAD STORES
When considering measures of body-lead burden it is important to recognize that lead is not
stored in a single homogeneous pool within the body. Lead may be found in a number of organ
systems including the blood, bone, kidney, liver, and other soft tissues, and is stored within each
system at different concentrations. Therefore, a body-lead burden measure, such as blood-lead
concentration, is a combination of recent and long-term lead exposure. This integration of exposures
has implications for the assessment of intervention effectiveness. If the lead retained in the body from
long-term exposure causes the blood-lead concentration to remain elevated, an otherwise successful
intervention may appear to have had only marginal impact on the child's lead exposure.
Bone tissues exhibit a particular affinity for lead that results in the accumulation of lead
concentrations in bone that are many times greater than the lead concentrations in other body tissues.
Prolonged exposure to lead produces a considerable store of lead in the skeletal tissue (Barry and
Mossman, 1970; Barry, 1975; Barry, 1981). Approximately 70% of a child's total body-lead burden
is present in his or her skeleton, as compared to 95% in adults (Barry and Mossman, 1970; Barry,
1975; Schroeder and Tipton, 1968). For this reason, bone-lead content is often cited as an excellent
measure of cumulative lead exposure. Furthermore, at least a portion of this lead is available for
mobilization back into the blood (Barry, 1981; Rabinowitz et al., 1976).
The elimination of lead exposure sources in a child's environment should reduce the amount
of lead absorbed into the child's blood from environmental sources, which should in turn reduce the
child's blood-lead level. However, the resulting low blood-lead level is no longer in equilibrium with
the lead concentrations in the bone tissues, resulting in the transfer of lead from the bone to the blood
in an attempt to re-achieve an equilibrium. The lead which is mobilized from the bone to the blood
in this fashion will maintain higher blood-lead levels than would be expected based on the reduced
lead uptake from environmental sources. If blood-lead levels are maintained at significantly higher
levels for periods of six months or more, the effectiveness of the lead hazard intervention could be
seriously underestimated. This section examines the evidence for bone-lead mobilization within the
scientific literature and employs a simple model of bone-lead mobilization to facilitate that
examination.
Page 11
-------
2.3.1 Review of Evidence for Bone-Lead Mobilization
The kinetics of lead in bone have been predominantly examined in adults or in representative
animal models. These results have then been extended to children, though this extension is a source
of considerable debate. In many instances, bone tissue has been segmented into two categories: dense
cortical bone and the more spongy trabecular bone (Leggett et al., 1982). Cortical bone is stated to
have an average half-life for lead of 10-20 years, as compared to approximately 5 years for trabecular
bone (Nordberg et al., 1991). This may be contrasted with an estimated one month half-life for lead
in blood (Rabinowitz et al., 1976). Though bone is a significant storage site for lead, at least a
portion of this mass of lead is available for mobilization back into the bloodstream. The percentage
of the store available is unclear; some authors contend that bone may be further partitioned into pools
of available and non-available lead (Rabinowitz, 1991). This turnover of lead is enhanced in periods
of bone demineralization or reduced uptake of lead from environmental exposure (Nordberg et al.,
1991). The latter conclusion was developed from studies examining adult bone and blood-lead levels
following the elimination of occupational lead exposure (Hyrhorczuk et al., 1985). The body of
research on bone tissue in children does suggest that "skeletal turnover is highest among children
under 10 years of age [and] ... is strongly influenced by factors that include nutritional status, age and
pathological conditions such as osteoporosis" (Nordberg, 1991). The measured concentration of lead
in a child's blood, therefore, is a mixture of recent exposure and mobilized lead stores from bone and
other organ systems that function as lead stores, but do not accumulate lead to the degree of bone.
Only one study was identified for which bone-lead levels were measured in children before
and after an intervention; a study of the effectiveness of chelation therapy on low to moderately
exposed children (mean blood-lead concentration, 32 /xg/dL) measured bone-lead content by L-XRF
pre-intervention and six weeks following enrollment (see Section 3.1.13 or (Rosen et al., 1991;
Markowitz et al., 1993; Ruff et al., 1993)). Some researchers question the representativeness of L-
XRF measures of mobilizable bone lead content. The interventions included residential lead-based
paint abatement and CaNa2EDTA chelation therapy (administered within one week following
enrollment) for children with positive lead mobilization tests (a procedure administered to assess
whether chelation therapy may be appropriate). By six weeks post-enrollment, mean bone-lead
content declined significantly (23%) among 71 chelated children, but did not for 103 children
experiencing only lead-based paint abatement (3%). Results from a subset of these children report
mean bone-lead content had declined 41 % by six months following enrollment among an unreported
number (<29) of chelated children, but had risen by 3% among 30 non-chelated children. The lack
Page 12
-------
of decline among the 103 non-chelated children appears consistent with the marginal (9%) reduction
in their blood-lead levels six weeks post-enrollment as compared with a mean blood-lead
concentration reduction among the 71 chelated children of 19%.
This study suggests that bone-lead stores partially mobilize following reductions in blood-lead
concentration due to lead hazard intervention. Other studies of chelation therapy effectiveness
(Shannon et al., 1988; Graziano et al., 1988; Graziano et al., 1992) document a rebound in blood-
lead levels following completion of treatment, presumably due to the enhanced mobilization of bone-
lead stores. It remains unclear, however, the extent to which these existing stores may keep blood-
lead concentration elevated following the intervention.
2.3.2 Modelling Bone-Lead Mobilization
Two parameters are necessary for assessing the extent to which bone-lead mobilization may
mask the effectiveness of an intervention:
the bone-lead mass (MBONE) to blood-lead mass (MBLOOD) ratio, MASSRAT
(=MBONE/MBLOOD), and,
the ratio of the rate at which lead is eliminated from the body to the blood-lead mass,
KBLELIM, or normalized lead elimination rate.
In Section 2.3.2.1, data from the scientific literature is used to develop estimates of these parameters.
In Section 2.3.2.2, these estimates are used in a two-compartment model of bone-lead mobilization to
analyze the potential for masking effects.
2.3.2.1 Development of Parameter Values
In order to develop an estimate of the bone-lead mass to blood-lead mass ratio, autopsy data
reported by Barry (1981) was utilized. Barry reported average lead concentrations in the blood,
various bone tissues, kidney, liver, and other soft tissues of autopsied children in varying age groups.
An overall bone-lead concentration was calculated as an arithmetic average of the lead concentrations
reported for the rib, tibia and calvaria bone tissues.
A bone-lead concentration to blood-lead concentration ratio was then determined for each age
group. A regression equation was fitted to the concentration ratio data to produce a predictive
equation as a function of age. Values for the volume of blood and the mass of bone tissues in
Page 13
-------
children (Altman and Dittmer, 1962; Harley and Kneip, 1984) of varying ages were then applied to
produce a predictive equation for the bone-lead mass to blood-lead mass ratio (MASSRAT) as a
function of age. Values of this ratio are presented in the second column of Table 2-4 for children 1
through 7 years of age. For example, the bone-lead store of an average 3-year-old child is estimated
to be 20.66 times the mass of lead in the blood.
Table 2-4. Parameter Values Used in the Two Compartment Model of Bone-Lead
Mobilization
Child's Age
(years)
1
2
3
4
5
6
7
Bone-Lead Mass to Blood-Lead
Mass Ratio
[MASSRAT] (unitless)
14.72
17.33
20.66
24.05
27.42
30.75
34.04
Elimination Rate
Constant
IKBLELIMT (day ^
0.068
0.062
0.059
0.056
0.054
0.052
0.050
Developing an estimate for the normalized lead elimination rate, KBLELIM, is more difficult;
due in part to the complexity of the process of lead elimination from the body. KBLELIM can be
interpreted as the fraction (unitless) of the lead in the blood that is eliminated from the body per day.
Therefore, KBLELIM has units of days"1. Lead exits the body via urine, endogenous fecal excretion
of bile produced by the liver, and loss from other soft tissues such as hair, skin and nails. As a
result, KBLELIM may be written as a function of the urinary lead elimination rate (L1RATE),
endogenous fecal lead elimination rate (FRATE), and the rate of elimination via other soft tissues
(ORATE):
Page 14
-------
URATE + FRATE + ORATE
MBLOOD
URATE I FRATE ORATE
i + +
MBLOOD [ URATE URATE
Estimation of KBLELIM is based on the estimation of the following three ratios:
• ratio of urinary elimination rate to blood-lead mass (URATE/MBLOOD),
• ratio of endogenous fecal elimination rate to urinary elimination rate
(FRATE/URATE), and
• ratio of elimination rate via soft tissues to urinary elimination rate (ORATE/URATE).
URATE/MBLOOD is estimated to be 0.022 days"1 for adults based on eight adult studies
(Rabinowitz et al., 1976; Campbell et al., 1981; Carton et al., 1981; Folashade and Crockford, 1991;
He et al., 1988; Kawaii et al., 1983; Kehoe, 1961; Assenato et al., 1986). Each study contained data
on the mean and standard deviation of an observed sample of measurements on blood-lead
concentration and urinary-lead concentration (or urinary lead excretion rate). Whenever necessary,
values of 50 dL and 19 mL/kg/day were substituted for the volume of blood and the daily volume of
urine per kg body weight, respectively (Altaian and Dittmer, 1962; Spector, 1956). FRATE/URATE
is estimated to be 0.39 for adults based on two adult studies (Rabinowitz et al., 1976; Chamberlain et
al., 1978). ORATE/URATE is estimated to be 0.17 for adults based on one adult study (Rabinowitz
et al., 1976). Combining these three estimates using the above equation yields an KBLELIM estimate
of 0.035 days'1 for adults.
Since the estimate of KBLELIM was determined from adult data, it is necessary to extrapolate
appropriate values for children. This is accomplished by scaling the adult estimates inversely
proportional to body weight to the 1/3 power (Mordenti, 1986). That is,
Body Weight
KBLEUMcm -
Assuming an adult body weight of 70 kg, and using age-dependent mean body weights for children of
various ages (Spector, 1956), the elimination rate constants in the third column Table 2-1 were
derived.
Page 15
-------
2.3.2.2 A Simple Model for Bone-Lead Mobilization
In order to evaluate the potential for elevated blood-lead levels due to bone-lead mobilization,
it is necessary to adopt a model for the transfer of lead between the blood and bone tissues within the
body and for elimination of lead from the body. For this report, the simple model illustrated in
Figure 2-3 was adopted. In this model, lead is taken into the body (from the gastrointestinal tract and
lungs) via the blood, can transfer between the blood and bone tissue, and is eliminated from the body
via the blood. Transfer of lead between the blood and bone tissues and elimination of lead from the
blood are all assumed to follow a first-order kinetic relationship.
BONE
Uptake
BLOOD
Elimination
Figure 2-3. Two compartment model of bone-lead mobilization.
The adopted model is most certainly an oversimplification. However, the results produced
using this model will approximate those of more complicated models involving additional tissue
compartments for the following reasons:
• While lead does mobilize from non-bone tissues following a decrease in lead uptake,
the effects are believed to be limited to a period of days or weeks due to the lower
concentrations of lead amassed in these tissues, and
• While all lead elimination from the body does not occur via a direct pathway from the
blood, the derivation of KBLELIM in the previous section properly includes these
Page 16
-------
other pathways (endogenous fecal and via other soft tissues) as if they were direct
from the blood.
Using the model illustrated in Figure 2-3, blood-lead levels (PbB) after intervention would
follow the relationship illustrated in Figure 2-4. There would be an initial drop in blood-lead
concentration immediately after intervention to achieve a concentration that can be supported by the
amount of lead being transferred from the bone. After this initial drop, blood-lead concentrations
would follow an exponential decline toward the blood-lead concentration that can be supported by the
post-intervention exposure level with no additional lead from the bone.
Pre- Abatement
PbB Level
Immediate
Post-Abatement
PbB Level
Target
PbB Level
Ultimate
Post-Abatement
PbB Level
Time (days)
Figure 2-4. Blood-lead concentration versus time following a reduction in lead uptake
The emphasis in this analysis is the length of time a target blood-lead concentration can be
maintained by lead mobilized from the bone. The length of time is illustrated by the symbol "T" on
the horizontal axis in Figure 2-4. Specifically, the analysis considers a target blood-lead
concentration (PbBTarget) which is 75% of the pre-intervention blood-lead level (PbBpre). This is
because many of the studies to be discussed in Section 3.0 exhibit 25% reductions in blood-lead
concentrations by 6-12 months post-intervention. When the target blood-lead level is 75% of the pre-
intervention level, the maintenance time T is maximized by assuming that no initial drop in blood-lead
level occurs immediately after intervention. That is, the maintenance time T is maximized by
Page 17
-------
assuming that the lead mobilized from the bone is initially sufficient to maintain the pre-intervention
blood-lead concentration.
Using this assumption, Figure 2-5 illustrates the expected decline in blood-lead concentration
for a five-year-old child following three different intervention scenarios: 50% effective, 75%
effective, and 100% effective. The blood-lead level is plotted as a percentage of the pre-intervention
level. Note that the blood-lead concentration approaches 50% for the 50% effective intervention
scenario, approaches 25% for the 75% effective scenario, and approaches 0% for the 100% effective
scenario. The blood-lead concentration curves are governed by the following equation:
KBLELIM]
MASSRAT\
The time period for which the blood-lead level is maintained at or above the target level can,
therefore, be calculated from the equation:
PbB = PbBLmgTerm + (PbBpre - PbB^^J • exp
MaxTime = In
F PbBD_. - PbB
Pre * ""LongTerm
PbB Target ~ PbBLongTerm \
MASS RAT
KBLELIM
Table 2-5 contains values of this time period for PbBTarget = 0.75*PbBPre and PbBLongTerm
equal to 0.50*PbBpre, 0.25*PbBPre, and zero for interventions which are 50%, 75%, and 100%
effective, respectively. For these calculations, values of MASSRAT and KBLELIM were taken from
Table 2-4. Note that the values of 11.8, 6.9, and 4.9 months for a five-year-old correspond to the
time point at which the blood-lead concentration curves in Figure 2-5 cross the 75% line. For
example, according to these model calculations, if an intervention reduces a five year old child's lead
exposure by 75%, bone-lead mobilization can maintain blood-lead concentrations at or above 75% of
their pre-interventional level for at most 6.9 months.
2.3.3 Conclusions Regarding Influence of Bone-Lead Stores
As will be seen in Section 3.0, the literature suggests six to 12 month post-intervention
declines in blood-lead concentration of approximately 25%. Is it possible that the interventions were
significantly more than 25 % effective with bone-lead mobilization accounting for the elevated blood-
lead levels? First consider blood-lead concentration measurements taken six months post-intervention.
The last column of Table 2-5 would suggest that 100% effective interventions could maintain blood-
lead levels at 75% of pre-intervention levels for six months for children ages 6-7 years. Interventions
Page 18
-------
which are 75% effective could produce this result for children ages 4-7 years and interventions which
are 50% effective could produce this result for children of any age. Thus, depending on the age of
the child, interventions which are 50% effective or more could appear to be only 25% effective at six
months post-intervention due to bone-lead mobilization.
10O%
75%
50%
25%
O%
0 6 12 18 24 30 36 42
Time Since Intervention (months)
Effectiveness of Intervention: 100% 75% — 50%
48
Figure 2-5. Blood-Lead Concentration Versus Time Following Interventions Which Are 50%,
75%, and 100% Effective.
Table 2-5. Length of Time (in Months) Bone-Lead Stores Can Maintain PbB at 75% of Pre-
Intervention Levels
Child's Age
(years)
1
2
3
4
5
6
7
Effectiveness of the Intervention
50%
3.6
5.0
6.7
8.6
10.5
12.6
14.7
75%
2.1
2.9
3.9
5.0
6.2
7.3
8.6
100%
1.5
2.1
2.8
3.6
4.4
5.2
6.1
Page 19
-------
Now consider blood-lead concentration measurements taken one year post-intervention. The
last two columns of Table 2-2 would suggest that 75% effective and 100% effective interventions
cannot maintain blood-lead levels at 75 % of pre-intervention levels for one year for children of any
age. Interventions which are 50% effective could produce this result, but only for children ages 5-7
years. Thus, for children ages 1-4 years, interventions which are 50% or more effective should not
appear to be only 25% effective at one year post-intervention due to bone-lead mobilization. Further,
interventions which are 75 % or more effective should not appear to be only 25 % effective at one year
post-intervention for any age child.
After intervention of a particular source of lead exposure, the lead uptake experienced by an
affected child should decrease. This decrease will produce declines in the child's body-lead burden as
measured by the blood-lead concentration. However, as blood-lead levels decline, lead is
simultaneously mobilized from existing bone stores. This masking effect can be considerable. Figure
2-6 presents the bias due to masking as a percentage of the target blood-lead concentration versus the
time since the intervention. The modelling results generated for this report suggest that bone-lead
stores, consistent with autopsy data, may be sufficient to singlehandedly maintain highly elevated
blood-lead concentrations at six months post-intervention even when the interventions are highly
effective. However, by one year post-intervention, the masking effect of bone-lead stores is
diminished. Therefore, if blood-lead levels remain at 75% of pre-intervention levels for one year
post-intervention, the intervention may have been significantly less than 100% effective.
These conclusions depend upon the cited experimental data and analytical assumptions used in
developing the model of bone-lead mobilization. The estimated values of bone-lead to blood-lead
mass ratio (MASSRAT) and the elimination rate (KBLELIM) are critical to the results presented in
Table 2-5. The experimental data used to develop these estimates was culled from a thorough review
of the available literature, but there is still only limited investigation in this area. The limited data
implies that significant uncertainty is associated with the estimated parameter values and therefore, the
maximum maintenance times. Similarly, the scaling factor used to extend the adult parameters to
children has sound reasoning to support it, but there is little or no empirical data to validate its effect.
These cautionary notes are not documented to question the viability of the reported analysis; the
model's construction and the accompanying analytical procedures are valid. Rather, as additional
research in this area is reported, its implications to the results in this section should be considered.
Page 20
-------
300% H.
250%
o
K
200%
01
c
150%
o
0
8
m
m
100%
50%
\
\
0 6 12 18 24 30 36 42
Time Since Intervention (months)
Effectiveness of Intervention: 25% 50% - — 75%
48
Figure 2-6. Percentage bias in blood-lead concentration due to masking versus time
following interventions which are 25%, 50%, and 75% effective.
That interventions may be significantly less than 100% effective is quite plausible, since it is
extremely difficult to identify and successfully isolate all the potential sources of a child's exposure.
A study of 184 children (pre-intervention blood-lead concentrations exceeding 50 /ig/dL), who were
released following chelation therapy to "lead-free" residences, portrays the difficulty of reducing a
child's lead exposure to zero (Chisolm et al., 1985). Mean blood-lead concentrations increased
during the first three months following treatment and remained stable at these elevated levels
thereafter. The mean blood-lead concentrations were significantly different across types of housing,
with a greater mean blood-lead concentration observed for children living in or regularly visiting lead-
based paint abated residences than for children living in "lead-free" public housing or "totally gutted
and renovated old housing." The authors' suggested that blood-lead concentrations failed to decline
because the children were still exposed to lead hazards, as the abated residences could be categorized
into "adequately abated" and "inadequately abated." "It is quite possible that potentially hazardous
sources of lead in the old homes were not always identified. (Chisolm et al., 1985)"
Page 21
-------
3.0 REVIEW OF SCIENTIFIC EVIDENCE
A variety of approaches were employed in an attempt to identify studies addressing the
effectiveness of lead hazard intervention. These included the authors' knowledge of the available
literature, two focused literature searches, an examination of the referenced articles cited in identified
studies, and additional material provided by Bradley Schultz of the EPA. One literature search used
keywords to identify journal articles or reports of potential relevance. The primary keywords were
"lead" and "Pb." The secondary keywords were "intervention," "abatement," and "delead." An
article or report was selected if its title contained one of the primary keywords and one of the
secondary keywords. This search produced a list of 371 potential articles and reports. The second
search was focused on non-domestic studies of lead hazard intervention. An examination of articles
or reports written outside the United States and containing the keywords "lead" or "Pb" in the title
produced a list of 721 documents. The two lists were reviewed and relevant articles (in addition to
those already identified) were selected and acquired. These materials formed the literature base
examined for this review.
A number of studies have examined the effectiveness of abating the environment of lead
hazards associated with lead-based paint, elevated dust-lead, and elevated soil-lead. These studies
regard hand to mouth activity as the primary pathway of childhood lead exposure. Generally, they
assessed whether a particular abatement practice effectively lowered an affected child's blood-lead
level or the levels of lead in his or her environment. In all cases, the studied interventions sought
"secondary" rather than "primary" prevention (e.g., assessing the effectiveness of lead hazard
intervention on already exposed rather than unexposed children). The intervention strategy considered
also sometimes depended upon the assumed mechanism by which the child's environment becomes
contaminated. One practice attempted to halt the soil track-in mechanism, while another continually
reduced the elevated dust-lead levels within the home.
Sixteen such studies are summarized in this section. In total, these studies spanned almost 13
years, from 1981 to 1994. A timeline graph locating the period during which the interventions
examined in each study were conducted is presented in Figure 3-1.
Page 23
-------
Baltimore Dust Control -
1982 St. Louis Retrospective-
Baltimore Traditional/Modified -
Boston Retrospective -
Baltimore Experimental-
Massachusetts Retrospective-
Seattle Track-In
1990 St. Louis Retrospective
Boston 3-City-
HUD Demonstration / CAP Study-
Milwaukee Retrospective LBP-
New York Chelation -
Milwaukee Retro. Educational-
Granite City Educational -
Milwaukee Pro. Educational-
1981 1983 1985 1987 1989 1991 1993 1995
Year In Which Interventions Were Performed
Figure 3-1. Timeline for the identified intervention studies.
Ten of the 16 studies focused on the abatement of lead-based paint as a primary form of
intervention, five studies focused on dust abatement, and one study focused on soil abatement.
A detailed discussion of each of the 16 studies follows in chronological order. These studies
are presented chronologically to emphasize the evolutionary nature of some abatement practices. The
discussion includes the pertinent study objectives, the sampled population, the intervention approach
studied, the environmental and body burden measures collected, the study design and results, and the
conclusions relative to the efficacy of the intervention performed. Additional details are included in
the study abstracts in Appendix A.
Page 24
-------
3.1. LEAD HAZARD INTERVENTION STUDIES
3.1.1. Baltimore Dust Control Study
This 1981 study (Charney et al., 1983) sought to assess whether periodic dust-control
measures in addition to lead-based paint abatement would be more effective in reducing blood-lead
concentrations than lead-based paint abatement alone. Forty-nine children aged 15 to 72 months with
at least two confirmed blood-lead concentrations between 30 and 49 /ng/dL formed the study
population. Their residences had all undergone lead-based paint abatement entailing the removal of
all peeling, lead-containing interior and exterior paint from the residence, and the removal of all lead-
containing paint from chewable surfaces below 4 feet. No extensive clean-up procedures were
required following the abatement. In addition to the abatement, the homes of 14 children in the
experimental group received periodic dust-control measures involving two monthly visits by a dust-
control team which wet-mopped all rooms in the residence where the dust-lead loading was identified
in an initial survey as exceeding 100 /ttg/sq ft.
Venous blood samples were collected during regular visits to the clinic, approximately every 3
months during the course of the study. After 6 months, there was a significant reduction of 5.3
/ig/dL in mean blood-lead concentration among the 14 children in the experimental group (wet-
mopping and abatement) and a further decrease of 1.6 /*g/dL after 1 year (Figure 3-2). In contrast,
the mean value for the abatement only group did not change significantly over the 12 months. Resi-
dential dust-lead loadings were collected for the experimental group during recruitment. To assess the
success in cleaning, dust-lead loading measurements were also obtained from all areas within the
residence where the child spent a significant amount of time. These measurements were taken both
before and after the dust-control teams completed their work. The samples were collected with
alcohol-treated wipes within a 1 ft2 area of floor or from the entire window sill. Within experimental
residences, the bimonthly dust-control efforts reduced the dust-lead loading on measured surfaces
(Figure 3-3). Dust measures were not collected in the abatement only group in order to avoid
drawing attention to dust as a potential source of lead exposure.
Page 25
-------
40-1
-a
o
\
•D
O
20-
10-
0-
-8
-4
12
Months Following initiation of Dust Control
Study Group:
1 Wet Mopping & LBP Abatamant, n=U children
e-o-o LBP Abatement Only. n=35 children
Figure 3-2. Arithmetic mean blood-lead concentration fyig/dL) since abatement by study
group (Baltimore Dust Control Study).
Dust-Lead Loading (/^g/sq ft)
At Study Inception
Dust-Lead Loading (//,g/sq ft)
After 12 Months
400-800
800-1600
200-400
100-200
<100
>1600
100-200
200-400
<100
400-800
Figure 3-3. Percentage of experimental homes with arithmetic mean dust-lead loadings
(/tg/ft2) in the defined range (Baltimore Dust Control Study).
Page 26
-------
3.1.2 1982 St. Louis Retrospective Paint Abatement Study
This 1982 study (Copley, 1983) sought to demonstrate a significant difference between the
children living in abated environments after lead hazard intervention compared to children still
exposed to lead hazards. The comparison was made among children measured to have a blood-lead
concentration greater than 25 /*g/dL. These children were enrolled in the St. Louis Health Division
Childhood Lead Poisoning Prevention Program. The intervention entailed the abatement of the lead-
based paint hazard, identified using XRF, within the residence. Surfaces with peeling or broken lead-
based paint were enclosed, replaced, or had their lead-based paint removed. No extensive clean-up
procedures necessarily accompanied the abatement. Blood-lead concentration measurements were
collected during routine venipuncture screening.
A retrospective study compared those blood measurements which identified the child as lead
poisoned to follow-up samples collected 6 to 12 months following the initial identification. A total of
102 children had sufficient samples collected to allow this comparison. Follow-up blood-lead
concentrations in children whose lead hazards had been abated were found to be an average of 11.29
jttg/dL lower than their initial levels (Figure 3-4). Blood-lead levels decreased on average only 1.24
/xg/dL for children whose hazards had not yet been abated. The difference in these mean decreases
was statistically significant (p< 0.001).
The results indicated that abatement of lead-based paint hazards did significantly reduce the
lead burden being borne by children with elevated blood-lead levels. The magnitude of the mean
reported difference between initial and follow-up samples is impacted by the varying amount of time
that passed between the sample collections and their timing relative to the abatement.
Page 27
-------
3
TI
O
2
•o
o
j3
m
Figure 3-4.
50
40-
30-
20
10
0
Initial
Identification
6-12 months
Post-Identification
Sampling Round
Study Group: *-*-* Abated Homes. n=61 children
o-o-o Unabated Homes, n=41 children
Pre- and post-identification arithmetic mean blood-lead concentration (/tg/dL) by
status of residence abatement (1982 St. Louis Retrospective Paint Abatement
Study).
3.1.3 Baltimore "Traditional"/"Modified" Paint Abatement Study
The goal of this 1984-1985 study (Farfel and Chisolm, 1990) was to evaluate the health and
environmental impact of "traditional" and "modified" Baltimore practices for the abatement of lead-
based paint. The study examined children residing in 71 residences abated in urban Baltimore (53
traditional abatements, 18 modified abatements). Prior to abatement, all the residences had multiple
interior surfaces coated with lead-based paint and housed at least one child with a blood-lead
concentration greater than 30 /tg/dL. "Traditional" Baltimore abatement practices called for
addressing deteriorated paint on surfaces up to 4 ft from the floor, and all hazardous paint on
accessible surfaces which may be chewed on. Paint with a lead content greater than 0.7 mg/cm2 by
XRF or 0.5% by weight by wet chemical analysis was determined hazardous. Open-flame burning
and sanding techniques commonly were used to remove hazardous paint. Modified abatement
procedures included the use of heat guns for paint removal and the repainting of abated surfaces. In
Page 28
-------
clean-up including wet mopping with high-phosphate detergent, vacuuming with standard shop
vacuums, and disposal of debris off-site. Clean-up following traditional abatement procedures
typically entailed at most dry sweeping. Dust samples were obtained using a alcohol-treated wipe
within a defined area template (1 ft2). Blood samples were collected via venipuncture.
Serial measurements of lead in interior house dust (lead loading), and children's blood-lead
concentration were collected. Average increases of 1200 /xg/ft2 in floor dust-lead loadings were
measured immediately following traditional abatements (usually within 2 days) on or in close
proximity to abated surfaces (Figure 3-7), with 10-100 fold changes at individual sites. Dust-lead
levels measured after modified abatements were an average of 360 /tg/ft2 higher than pre-abatement
levels. Thus, modified abatement procedures resulted in elevated floor dust-lead loadings, but not to
the extent seen for traditional practices. At 6 months post-abatement, average dust-lead loadings were
65 /xg/ft2 higher than pre-abatement loadings for traditional abatements and 28 ng/ft2 higher than pre-
abatement loadings for modified abatements.
Pre- and post-abatement blood-lead concentrations were available for 46 children who lived in
the abated residences and had not yet undergone any chelation therapy. The post-abatement samples
were collected within one month following the completion of the abatement activities. For traditional
abatements, average blood-lead levels in 27 children rose 6.84 /ng/dL (from 36.88 ng/dL to 43.72
jtg/dL) while a rise of only 1.03 /*g/dL (from 34.40 jig/dL to 35.43 /*g/dL) was observed for 19
children exposed to modified abatements (Figure 3-5). Moreover, a large number of children
required chelation therapy. Six months after abatement, a subset of 29 children (14 traditional, 15
modified) who had not undergone any chelation therapy exhibited blood-lead concentrations (mean,
30.67 jtg/dL) that were not significantly different from their pre-abatement levels (mean, 32.53
jug/dL). The 6-month results should be viewed with caution, since blood-lead levels were available
for relatively few non-chelated children six months after abatement, primarily due to the large number
of children requiring chelation therapy.
Despite the implementation of improved practices, modified abatements, like traditional
abatements, did'not result in any long-term reductions of levels of lead in house dust or the blood of
children with elevated pre-intervention blood-lead concentrations.
Page 29
-------
45
40
35
30
25
20
15
10
5
0
-2
Population Considered:
Months Since Abatement
TradlUonar Abatement, n=27 children
e-o-o "ModfflecT Abatement, n-19 children
see subset n=29 children (14 "McdHlecT)
Figure 3-5. Arithmetic mean blood-lead concentration (jtg/dL) post-abatement by population
considered (Baltimore "Traditional"/"Modified" Paint Abatement Study).
3.1.4 Boston Retrospective Paint Abatement Study
This 1984-1985 study (Amitai et al., 1991) sought to evaluate the extent to which the lead
poisoning of children is exacerbated during the abatement of lead-based paint within their residence.
The study population consisted of 114 children ranging in age from 11 to 72 months (median age of
24 months) with at least one blood-lead concentration (above 25 ^g/dL) obtained prior to deleading,
one blood-lead sample collected during deleading, and one blood-lead determination following the
completion of the deleading process. The deleading process consisted of the removal or permanent
coverage of any paint with a lead content greater than 1.2 mg/cm2 which was loose and peeling (at
any height), or present on chewable surfaces accessible to a child (below 4 ft). Clean-up using wet
washing with tri-sodium phosphate (TSP) was stressed, but not uniformly performed following the
abatement. Blood-lead concentration measurements were collected via venipuncture.
The mean blood-lead level in the 114 children rose 5.7 /ig/dL during deleading and then fell
8.6 fj-g/dL approximately 2 months following the completion of the deleading activities (Figure 3-6).
The statistically significant (p < 0.05) decrease in mean blood-lead concentration post-deleading is due
in part to 42 children who underwent chelation therapy between the mid- and post-deleading
measurements.
Page 30
-------
50
40
30
20
10
-2
Population Considered:
2 4
Months Since Abatement
Study Group, n=114 children
e-o-o Subset, n-59 non-chelated children
BBS subset, n=55 children (42 chelated)
Figure 3-6. Arithmetic mean blood-lead concentration Otg/dL) post-abatement by study
population considered (Boston Retrospective Paint Abatement Study).
In an effort to determine the effect of deleading activities alone, a subset of 59 children who
underwent no chelation therapy were examined. In this subset, an additional follow-up measure was
collected 236 to 264 days after completion of the deleading work. There was no evidence of a
change in mean blood-lead concentration during deleading. However, blood-lead levels fell an
average of 4.5 /xg/dL at the post-deleading collection (approximately 2 months) and fell an additional
5.5 ng/dL by the follow-up (approximately 8 months) deleading collection (Figure 3-6). Caution is
warranted when considering the results for the non-chelated children. By excluding children whose
blood-lead levels rose (the children were already bordering the point where chelation therapy was
appropriate), the declines in blood-lead concentration are not altogether surprising.
For 80 of the children, the specific method of deleading was available. Blood-lead levels in
affected children were considerably elevated by dry scraping and sanding methods (mean increase in
41 homes of 9.1 ^tg/dL) and torches (35.7 /*g/dL in 9 homes). By comparison, children exposed to
encapsulation, enclosure, or replacement abatement procedures (12 homes) experienced a mean
decrease of 2.25 jag/dL in their blood-lead burden during deleading. The study's results indicated
that deleading may produce a significant, transient elevation of blood-lead in many children. It was
most dangerous if accomplished with the use of torches, sanding, or dry scraping. The deleading
Page 31
-------
may have been efficacious long-term, however, in that blood-lead concentrations declined significantly
2 months after deleading.
The stability of blood-lead levels prior to deleading activities was characterized for a subset of
32 children who had two blood samples prior to deleading. The mean blood-lead concentration rose
from 35.4+1.3 /xg/dL to 36.0±1.1 jug/dL during the interval between these samples (73±23 days),
however, this change was not statistically significant (p>0.5).
3.1.5 Baltimore Experimental Paint Abatement Studies
These studies (Farfel and Chisolm, 1991; USEPA, 1987; Farfel, el al., 1994) sought to
demonstrate and evaluate experimental lead-based paint abatement practices developed in response to
the inadequacies uncovered for traditional Baltimore abatement procedures (see Section 3.1.3). The
literature examines two distinct sets of dwellings in urban Baltimore that were abated according to the
experimental method.
The first study (Farfel and Chisolm, 1991) evaluated the short-term efficacy (up to 9 months)
of the experimental abatement procedures in six older dwellings, built in the 1920s, that received
abatements in 1986-1987 as part of a pilot study examining the experimental procedures. Each
dwelling was a two-story six-room row home in poorly maintained condition with multiple lead-based
paint hazards. Four of the residences were vacant, two housed lead-poisoned children.
The second study (Farfel, el al., 1994) evaluated the longer-term efficacy (1.5 to 3.5 years) of
the experimental abatement procedures in 13 dwellings, which had received experimental abatements
by local pilot projects between 1988 and 1991. At least 6 pairs of pre- and immediately post-
abatement dust-lead loading measures, taken from the same locations, were available for each
dwelling. The dwellings were occupied and had not undergone major renovations since those
associated with the experimental abatement. Dust lead samples were collected in the 13 dwellings
during December 1991 and January 1992, in the same locations, where possible, that had been
sampled pre- and immediately post-abatement.
The experimental abatement procedure called for the floor to ceiling abatement of all interior
and exterior surfaces where lead content of the paint exceeded 0.7 mg/cm2 by XRF or 0.5% by
weight by wet chemical analysis. Lead-contaminated dust was contained and minimized during the
abatement, and extensive clean-up and disposal activities were utilized. Alcohol-treated wet wipes
were used to collect dust-lead loading samples from household surfaces within each residence. In
addition, surface soil samples were collected at the 13 dwellings in the second study.
Page 32
-------
In the 6 homes from the first study, serial measurements of lead in interior dust were made
immediately before abatement, during the abatement, after the final clean-up, and 1,3, and 6 to 9
months following the abatement. Floor dust-lead loadings immediately post-abatement were an
average of 390 /xg/ft2 lower than pre-abatement levels (Figure 3-7). By 6 to 9 months following the
abatements, average levels had decreased a further 74 /Mg/ft2. The dust-lead loading monitoring
before, during, and after the abatement activities also provided information on the effectiveness of
particular measures. All floor and window treatments were associated with significant (p<0.05)
decreases in dust-lead loading over time. Results also suggested that window replacement may have
been more effective in reducing dust-lead loading than stripping the lead-based paint. In addition,
vinyl floor coverings may have produced lower dust-lead loadings than sealing old wooden floors with
polyurethane.
10000
1000
100
10
-2
Abatement Practice:
Months Since LBP Abatement
* * * Traditional, n=53 homes
©-»« Modified, n=18 homes
Experimental, n=6 homes
Figure 3-7.
Geometric mean floor dust-lead loading (/tg/ft2) post-abatement by abatement
practice performed (Baltimore "Traditional"/"Modified" and Experimental
Practices Studies).
Page 33
-------
In the 13 homes considered in the second study, geometric mean PbD levels 1.5 to 3.5 years
post-abatement were significantly less than pre-abatement levels, despite some reaccumulation of lead
in dust. The geometric mean floor dust-lead loading was 40.9 /xg/ft2 1.5 to 3.5 years post-abatement,
compared with the pre- and immediately post-abatement levels of 254 /xg/ft2 and 13.9 /xg/ft2,
respectively. Similarly, the geometric mean window sill dust-lead loading was 103 fig/ft2 1.5 to 3.5
years post-abatement, compared with the pre- and immediately post-abatement levels of 1041 /^g/ft2
and 13.0 fj-g/ft2, respectively. Greater reaccumulation was observed for window wells, where the
geometric mean dust-lead loading was 600 /xg/ft2 1.5 to 3.5 years post-abatement, compared with the
pre- and immediately post-abatement levels of 14214 /xg/ft2 and 34.4 /xg/ft2, respectively. Seventy-
eight percent of all dust-lead loading measurements 1.5 to 3.5 years following intervention were
within Maryland's interim post-abatement clearance standards (200 /xg/ft2 for floors, 500 /tg/ft2 for
window sills, and 800 tig/ft2 for window wells); twenty-one of the 39 readings above the clearance
levels were from window wells. Soil-lead concentration was not found to be a significant factor in
explaining the change in dust-lead levels.
The results suggest that comprehensive lead paint abatement is associated with longer-term as
well as short-term control of residential dust-lead hazards. The experimental methods resulted in
substantial reductions in interior surface dust-lead levels immediately post-abatement which were
found to persist throughout a 6 to 9 month post-abatement period. Dust-lead levels were not
uniformly reduced to desired levels, particularly on window sill and window well surfaces that were
abated using paint removal methods by the 1.5 to 3.5 year post-abatement measures, however, 78%
of all readings remained below target levels. The magnitude of the decline in dust-lead loadings
following abatement may have been exaggerated in the first study since vacant units are likely to
contain more dust than occupied units.
3.1.6 Central Massachusetts Retrospective Paint Abatement Study
This 1987-1990 retrospective study (Swindell et al., 1994; Charney, 1995) examined the
effectiveness of residential lead-based paint abatements as conducted between 1987 and 1990 in
central Massachusetts. More stringent home deleading regulations were enacted in Massachusetts in
1988, during the period covered by the study. The sample population consisted of 132 children, 12
to 91 months of age, with a confirmed blood-lead concentration exceeding 25 itg/dL, and whose
homes were abated between 1987 and 1990. In addition, the children had at least one venous blood-
lead determination within 6 months prior to abatement and at least one venous blood-lead
Page 34
-------
determination 2 weeks to 6 months after the completion of abatement. Children who received
chelation therapy during that time, or who moved during the study period, were excluded from the
study. Although a venous blood-lead level above 25 jig/dL was a criterion for this retrospective
study, blood-lead concentrations immediately prior to abatement were less than 25 /ig/dL for some
children. In these cases, the authors suggest that the pre-abatement measure might have reflected
some early abatement or education effects.
Abatements prior to 1988 consisted of the removal or permanent coverage of any paint with a
lead content greater than 1.2 mg/cm2 which was loose or peeling, or present on chewable surfaces
accessible to the child (below 4 ft). No standard abatement methods, dust-control measures, or
cleanup procedures were mandated. The 1988 regulations required licensing of abatement contractors
(after completing a 3-day course and passing a certifying exam), specifically prohibited torching or
machine sanding of paint, permitted only hand-scraping and replacement as removal methods, and
required all occupants to vacate the dwelling during the entire abatement and cleaning process. There
were no dust samples required after abatement, but the mandated cleanup entailed vacuuming all
surfaces with a high-efficiency particle air (HEPA) filter vacuum, followed by wet-mopping and
sponging with a trisodium phosphate cleaning solution, and then a second HEPA vacuuming.
Children's blood-lead concentration measures at most 6 months prior to initiation of abatement
were compared to the last measurement collected within one year following abatement. The specific
timing of post-abatement measures ranged from 3 to 52 weeks after abatement. The mean blood-lead
level for the 132 children declined significantly (p< 0.001) following abatement, from 26.0 /xg/dL to
21.2 ^g/dL. Figure 3-8 summarizes the magnitude of the documented changes in blood-lead
concentrations. Although blood-lead levels declined in 103 children (78%) within one year of
abatement, the proportion who experienced a decline varied with initial blood-lead levels: 32 of 33
children (97%) with initial blood-lead levels exeeding 30 /ig/dL, 64 of 79 (81 %) with initial levels
between 20 /ig/dL and 29 /zg/dL, and only 7 of 20 (35%) with initial levels below 20 Mg/dL
experienced a decline following abatement. In fact, as shown in Figure 3-9, among children with
initial levels below 20 /*g/dL, mean blood-lead levels increased from 16.7 /tg/dL to 19.2 /xg/dL
following abatement (p=0.053). The calendar year of abatement did not appear to significantly
impact the measured declines. In 1987, before the regulations were enacted, 29 children experienced
a mean decline of 5.3 /xg/dL. This is comparable to mean declines of 5.7 /*g/dL in 1988 (n=48) and
4.7 ^g/dL in 1989 (n=42). The mean decline among 13 children whose residences were abated in
Page 35
-------
Number of Children
-* — M IO CM (* 4
3U1OUIOUIOU1C
1
OOOO NTsiV^ vvv\" vvxx^
^^^^
Absolute Change in Blood—Lead Concentration
I Decrease LXXXI Increase
Figure 3-8. Distribution of the absolute change in blood-lead concentrations
(Central Massachussets Retrospective Paint Abatement Study).
30
25
1 20
15
10
j
Pre~abatement
Pott—abatement
Pre-abatement
PbB Level:
Sampling Round
>30 ug/dL, n=33 children
e-o« 20-29 ug/dL. n=79 children
<20 /ig/dL. n=20 children
Figure 3-9. Pre-abatement and post-abatement blood-lead concentration by pre-abatement
blood-lead level (Central Massachusetts Retrospective Paint Abatement Study).
Page 36
-------
1990 was influenced by two children whose levels increased markedly (from 16.0 and 17.0 /xg/dL to
29.0 and 31.0 /xg/dL). A 2.0 /xg/dL median decline was measured for this group. Among 72
children with more than one pre-abatement measure, 40 children whose levels were declining (defined
as a decrease greater than 5 /xg/dL) prior to abatement exhibited only a modest further mean decline
of 1.9 /xg/dL. A more significant mean decline of 8.2 /xg/dL was observed for the 32 children whose
initial levels were relatively constant.
The study's results are consistent with other studies reporting declines in blood-lead
concentrations following lead-based paint abatement. The significant decline among 32 children with
stable levels prior to abatement suggests that regression to the mean cannot fully account for the
observed decline. Moreover, the anticipated efficacy appears to depend on the child's initial blood-
lead concentration. Children with high pre-abatement blood-lead concentrations were more likely to
experience declines (and of greater magnitude) than children with lower pre-abatement blood-lead
levels. In addition, it appears that stricter lead-based paint abatement regulations did not immediately
result in significantly greater effectiveness. It also appears that even carefully performed abatements
may result in post-intervention elevations in the blood-lead concentrations of children whose initial
levels are below 20 /xg/dL. Some caution in interpreting these results is warranted, however, because
of the highly variable timing of the post-abatement measures. The children's blood-lead
concentrations were measured anywhere between 3 and 52 weeks following completion of the
abatement. Seasonal and age variation in blood-lead concentrations could significantly impact the
observed decline, depending upon the period of time between the measures and the season in which
the measures were collected.
3.1.7 Seattle Track-In Study
This study (Roberts et al., 1991) sought to determine the extent to which low cost dust-control
measures successfully lowered household dust-lead loading. Forty-two homes in Seattle and Port
Townsend, Washington, built before 1950 formed the population studied from 1988-1990. The three
abatement procedures considered were strictly low-cost dust reduction procedures, namely, use of a
vacuum cleaner with an agitator bar in normal cleaning, removal of shoes at the entrance to the resi-
dence, and installation of walk-off mats. Dust samples were collected from rugs within the residence
using a Hoover Convertible vacuum cleaner. Soil samples were scraped from within 1 ft of the
residence's foundation.
Page 37
-------
The study employed step-wise regression analysis to assess which factors determine the dust-
lead loading within a residence. Significant associations were found between log transformed dust-
lead loading and removing shoes at the door and the presence of a walk-off mat (e.g., hall carpet in
an apartment building). Lower fine dust-lead levels (sieved before analysis) were found in homes
where the residents removed their shoes (29 jug/ft2) and/or used a walk-off mat (54 jig/ft2) compared
to those in homes whose residents did not (994 jig/ft2).
The occupants of three homes tested in the study began removing their shoes upon entry for at
least 5 months prior to the collection of a second dust-lead measurement from their carpets. In
addition, the occupants of one of these homes installed walk-off mats at both entrances and began
vacuuming twice weekly. The geometric mean dust-lead loading fell from 1588.6 /ig/ft2 to 23.2
^g/ft2 in these homes.
The data suggested that controlling external soil and dust track-in by removing shoes and/or
using a walk-off mat reduced the lead exposure from house dust. Lacking any blood measurements,
it was difficult to assess the impact these interventions may have had on childhood lead exposure.
3.1.8 1990 St. Louis Retrospective Paint Abatement Study
This 1989-1990 study (Staes et al., 1994) attempted to assess, via a retrospective cohort
study, the effectiveness of lead-based paint abatement in reducing children's blood-lead levels. The
sample population consisted of children under 6 years of age who were identified by the St. Louis
City Health Department as having a blood-lead concentration of at least 25 ^ig/dL, and residing in
dwellings with lead-based paint hazards. The intervention entailed the abatement of the lead-based
paint hazard, identified using XRF, within the residence. Surfaces with peeling or broken lead-based
paint were enclosed, replaced or had their lead-based paint removed. No extensive clean-up
procedures necessarily accompanied the abatement. The blood-lead concentrations were collected via
venipuncture.
The geometric mean blood-lead concentration among the 189 children selected was 33.6
jLtg/dL. Seventy-one of these children, 49 whose homes were abated and 22 whose homes had not
been abated, had blood-lead measures 10-14 months following the initial diagnosis. The geometric
mean blood-lead concentration of the 49 children from abated dwellings decreased by 23%, from 34.9
to 26.7 /ig/dL. This decline (Figure 3-10) was significantly greater (p=0.07) than the 12%
reduction, from 35.1 to 30.9 jug/dL, observed among the 22 children residing in unabated dwellings.
Page 38
-------
40
c
o
u
c
o
u
•o
a
T3
o
_0
CD
20-
10
0-
0 months 4—6 months 6—8 months 8—10 months
Sampling Round Following Identification
Study Group: *-*-* LBP Abated Homes. n<51 children
e-e-o Unabated Homes, ns63 children
10-14 months
Figure 3-10. Pre- and post-identification arithmetic mean blood-lead concentration (/tg/dL) by
status of residence abatement (1990 St. Louis Retrospective Paint Abatement
Study.)
A multiple linear regression model predicting the change in geometric mean blood-lead
concentration at 10-14 months following diagnosis was fitted. The dwelling's abatement status at the
time of the follow-up blood sample (e.g., abated or unabated) and whether the blood-lead level at
diagnosis exceeded 35 /*g/dL were statistically significant (p<0.10) factors in the analysis. The
geometric mean blood-lead concentration of children residing in abated dwellings was estimated to
decrease 13% (95% CI, -25% to 1%) more than that of children residing in unabated dwellings.
Moreover, the geometric mean blood-lead concentration of children with an initial blood-lead
concentration >35 /*g/dL was estimated to decline by 17% (95% CI, -27% to -5%) more than that of
children with lower blood-lead concentrations.
For lead-poisoned children in St. Louis, the decline in geometric mean blood-lead
concentration was greater for children whose dwellings underwent lead-paint hazard abatement than
for children whose dwellings did not. The magnitude of the efficacy appears to depend upon the
Page 39
-------
child's initial blood-lead concentration. The reported differences between initial and follow-up
samples were impacted by individual differences in the amount of time that passed between the
sample collections and their timing relative to the abatement. Many of the follow-up measures were
collected less than six months following the abatement.
3.1.9 Boston Three-City Soil Abatement Study
This 1989-1991 project (Weitzman et al., 1993; Aschengrau, et al., 1994) assessed whether a
significant reduction (> 1000 ppm) in the concentration of lead in residential soil will result in a
significant decrease (>3 /*g/dL) in the blood-lead concentration of children residing at the premises.
A total of 152 children were enrolled, each satisfying the following criteria: (1) less than or equal to 4
years of age, (2) blood-lead concentration between 10 and 20 /-ig/dL with no history of lead
poisoning, and (3) a minimum median residential soil-lead concentration of 1500 ppm. The project
employed four intervention procedures: (a) interior paint stabilization by removing peeling or chipping
paint, (b) interior dust abatement via wet mopping and HEPA vacuuming, (c) soil removal (to a depth
of 6 inches) and replacement, and (d) interior and exterior lead-based paint abatement. Dispersal of
soil during the abatement was retarded by wetting the soil, preventing track-in by workers, containing
the abatement site with plastic, and washing all equipment. Extensive environmental media and body
burden samples were collected: composite core soil samples; vacuum dust samples; first draw water
samples; interior and exterior paint assessment via portable XRF; venipuncture blood samples; and,
hand-wipe samples.
Each child enrolled was randomly assigned to one of three experimental groups: Study (54
children), Comparison A (51 children), or Comparison B (47 children). During Phase I, the Study
Group received interior paint stabilization, interior dust abatement, and soil abatement. Comparison
Group A received interior paint stabilization and interior dust abatement. Comparison group B
residences received only interior paint stabilization. During Phase II, which began approximately 12
months after the Phase I interventions, both comparison groups received soil abatement and all three
experimental groups were offered lead-based paint abatement. Environmental media and body burden
samples were collected at various times surrounding these intervention activities.
During Phase I, the average blood-lead concentrations in all three experimental groups
decreased at the first (6 months) post-abatement measurement (Figure 3-11). The statistically
significant decreases were: 2.9 /*g/dL for Study, 3.5 ^ig/dL for Comparison A, and 2.2 /xg/dL for
Comparison B. The following increases in average blood-lead concentration were recorded between
Page 40
-------
15
5
c
o
u
c
o
o
u
«
a
_o
m
10-
5
0
Pro—Abatement
Approx. 6 months
Post-abatement
Approx. 11 months
Post-abatement
Sampling Round
Experimental Group: *-*-* Study, n=52 children
e-o-o Comparison A, ns51 children
Comparison B, ns47 children
Figure 3-11. Arithmetic mean blood-lead concentration (/tg/dL) across sampling rounds and
experimental groups, Phase 1 (Boston 3-City Soil Abatement Project).
the first and second (11 months) post-abatement measurements: 0.5 /ng/dL for Study, 2.6 /ig/dL for
Comparison A, and 1.5 jig/dL for Comparison B. The increases for the two comparison groups were
significantly different from zero. The mean dust-lead levels from hand wipe samples for all groups
followed a similar pattern, though they exhibited considerably greater variability.
By the end of Phase II, 91 children were still participating and living at the same premises as
when they were enrolled. Of these children, 44 received both soil and lead-based paint abatement, 46
received only soil abatement, and 1 refused both interventions. Although some premises underwent
lead-based paint deleading during Phase II, no results on the additional efficacy of lead-paint
abatements were reported.
For children whose residence underwent soil abatement only, mean blood-lead concentrations
decreased by 2.44 jtg/dL for 52 children in the Study Group, 5.25 /*g/dL for 18 children in
Comparison Group A, and 1.39 jig/dL for 13 children in Comparison Group B, between pre- and
post-intervention measures (Figure 3-12). Blood-lead measures were taken an average of 10 months
Page 41
-------
post-abatement for the Study Group (during Phase I) and an average of 9 months post-abatement for
the comparison groups (during Phase II).
A repeated measures analysis was conducted using a restricted sample of 31 children from
Comparison Group A (N=18) and Comparison Group B (N = 13) who received only soil abatement
during Phase II and who had blood-lead measures at the beginning of Phase I, the end of Phase I, and
the end of Phase II. Study Group data were excluded for lack of a control period. Mean blood-lead
concentrations decreased by 0.64 /^g/dL during Phase I (before the soil abatements) and another
3.63 jug/dL during Phase II (a 33.9% decline overall). A trend in the magnitude of the decline in
blood-lead levels was apparent, with larger declines observed in children with larger initial blood-lead
levels.
D)
C
0)
u
C
o
o
T3
O
-a
o
o
15
10-
5
'—B ^"*^
Approx. 10 months
Pre-Abatement
Immediately
Pre-Abatement
Approx. 6 months
Post—Abatement
Approx. 10 months
F'ost-Abatement
Sampling Round
Experimental Group: * * * Study, n=52 children
e-o-o Comparison A, n=1B children
&ee Comparison B, n=13 children
Figure 3-12. Arithmetic mean blood-lead concentration (/tg/dL) across sampling rounds and
experimental groups, Phase I and II (Boston 3-City Soil Abatement Project).
The decline in median soil-lead concentration among Study group residences immediately
post-abatement averaged 1790 ppm (range: 160 ppm to 5360 ppm). Although many yards had
evidence of recontamination both at 6-10 months and 18-22 months post-abatement, follow-up median
soil-lead concentrations were generally less than 300 ppm (Figure 3-13). Similar results were
Page 42
-------
observed for the comparison groups following the soil abatements in Phase II. Dust-lead loadings
were less consistent. Composite floor dust-lead loadings declined significantly during the study.
Comparable declines were seen in all three groups during Phase I, despite Comparison Group B not
receiving any interior house dust abatement. Mean floor dust-lead loadings were relatively unchanged
for Comparison Groups A and B (P=0.95 and 0.15, respectively) during Phase II, despite the soil
abatement. By 18-22 months post-abatement, mean levels in the Study Group had risen, but remained
still significantly below initial levels (P=0.02). Mean window well dust-lead loadings declined in
Comparison Group A following the soil abatement, but rose in the Study Group and in Comparison
Group B.
100000
10000:
3
•a
a
s
£ 1000-
c
B
s
100:
10-
_____ _ B--
--•O
*
Pro—intervention
Approx. 10 month
Post-Intervention
Approx. 20 months
Post-Intervention
Media Levels for
Study Group:
Sampling Round
Soil-Lead Concentration
e-oo Floor Dust-Lead Lodaing (
BBQ Window Well Dust-Lead Loading (/^g/ft )
Figure 3-13. Arithmetic mean environmental lead level (for Study Group) across sampling
rounds (Boston 3-City Soil Abatement Project).
These results suggested that abatement of lead-contaminated soil around homes may result in a
modest decline in blood-lead levels. The reported declines, however, may be influenced by seasonal
variation in blood-lead levels. Seasonal variations in blood-lead concentrations of comparable
magnitude have been cited in other studies conducted in Boston and Milwaukee (Kinateder and
Menton, 1992; Schultz, 1993). In addition, relatively few children were available for the Phase II
Page 43
-------
analysis, which introduces the possibility of bias in the estimated declines due to low participation at
follow-up. Moreover, since no control populations were available for the Phase II results, it is
difficult to assess their larger declines.
3.1.10 HUD Abatement Demonstration (HUD Demo) Study
This study (HUD, 1991; HUD, 1990) was designed to determine and evaluate the overall
suitability and effectiveness of various methods of lead-based paint abatement. These methods were
tested in 1989-1990 in 172 FHA-foreclosed, single family housing units in seven urban areas:
Baltimore, Washington, D.C., Seattle, Tacoma, Indianapolis, Denver, and Birmingham. Six
abatement procedures were employed: (1) encapsulation by sealing the surfaces with durable coatings,
(2) abrasive removal of lead-based paint using mechanical removal equipment, (3) hand-scraping with
a heat gun to loosen and remove the lead-based paint, (4) chemical removal of lead-based paint using
a chemical stripper, (5) enclosure or covering the surface, and (6) removal of contaminated building
components and replacing with new or deleaded components. Because of the diversity of housing
components containing lead-based paint, it was generally true that no single abatement method could
be used uniformly throughout a given housing unit. Therefore an abatement strategy, consisting of
decision rules for choice of abatement method, was randomly assigned to each house. The method
used to characterize the unit abatement strategy was always the first-choice method and was used on
all components to the extent feasible. Second, third and fourth choice methods were also specified
for each strategy. XRF devices were used to identify components covered by paint with a lead
content greater than 1.0 mg/cm2. These components were abated in the houses selected for the study.
Following completion of the abatement, the units were extensively cleaned using HEP A vacuums and
wet washing with TSP. Surfaces were wet wiped to obtain dust-lead loading samples within a defined
area and core soil samples were collected. No blood-lead measures were collected, however, since
the units were vacant prior to abatement.
Pre-abatement dust-lead loadings generally were not collected. Once the lead-based paint had
been abated and the area cleaned, clearance wipe samples were collected to verify acceptable dust-lead
levels (Figure 3-14). The resulting dust-lead loading was compared to the appropriate standard in the
HUD Guidelines (HUD, 1990) - 200 ^g/ft2 for floors, 500 ^g/ft2 for window sills, and 800 ^g/ft2
for window wells. On average, 80% of floor wipe samples, 85% of window sill samples, and 65%
of window well samples passed the initial clearance test by measuring below the appropriate standard.
Additional cleaning, or other measures, were required for surfaces that did not pass. There were
Page 44
-------
significant differences in failure rates among the different abatement methods. The highest failure
rates were generally for components abated using chemical stripping (22.7%, 24.1%, and 45.7% for
floors, sills, and wells) and heat gun removal (28.8%, 24.4%, and 44.5% for floors, sills, and wells).
With the exception of abrasive sanding (the machines kept clogging), all the methods were
successfully implemented. To do so, however, required varying degrees of effort. Chemical
stripping and heat gun methods had lower success rates in meeting the HUD Guidelines than did
encapsulation, enclosure and replacement methods.
100 •]
90
80
70
60
SO
30
o
K 20
10
0
I
Floors
H I Window 1
Sills
I I Encapsulation f\VVl Enclosure
SSZ Scrape w/ Heat CBul Replacement
Window 1
Wells
Chemical Strip
Sample Type
Figure 3-14. Percentage of components successfully abated to HUD Guidelines Standards by
abatement method and sampling location (HUD Demo Study).
3.1.11 Comprehensive Abatement Performance (CAP) Study
The 1992 CAP study (Buxton et al., 1994) assessed the long-term effectiveness of two lead-
based paint abatement strategies: (1) encapsulation and enclosure methods and (2) removal methods.
Fifty-two FHA foreclosed, single family residences in Denver, Colorado, were examined. Thirty-five
of the residences were abated using the aforementioned methods as part of the HUD Demonstration
study. Each house was primarily classified according to the abatement category (i.e., encapsulation/
Page 45
-------
enclosure versus removal methods) accounting for the largest square footage of interior abatement.
The remaining 17 residences were unabated homes identified in the HUD Demonstration as containing
little or no lead-based paint. At the time of the environmental sampling approximately 1.5 to 2 years
following the abatements, the units were occupied. Vacuum dust-lead levels were measured at the
interior and exterior entry ways, floor perimeters, window sills, window wells, and air ducts of each
residence. Core soil samples were collected at the foundation, entryway, and boundary of the home.
No blood-lead measures were collected because the units were not reoccupied until several months
after their abatement.
The CAP Study found geometric mean lead concentrations in abated houses to be significantly
higher than those in unabated houses only at sampling locations where no abatement was performed
(Figure 3-15). Specifically, the differences were statistically significant for dust in the air ducts and
for soil at the foundation and boundary areas. Geometric mean dust-lead loadings on floors and
exterior entry ways were also significantly higher in abated houses than unabated houses, but these
differences were attributed to higher dust loadings. It should be noted that both floor and window sill
geometric mean dust-lead loadings in abated houses were found to be below their respective HUD
interim standards of 200 and 500 Mg/ft2- Geometric mean floor dust-lead loadings were also below
the EPA guidance (EPA, 1994) level of 100 jug/ft2. In contrast, geometric mean window well dust-
lead loadings in both abated and unabated houses were found to be well above the HUD value of
800 Mg/ft2.
Lead levels were somewhat higher, though not significantly higher, in houses abated by
encapsulation/enclosure methods than in houses abated by removal methods. When interpreting these
results it should be noted that encapsulation/enclosure houses typically had larger amounts of
abatement performed than removal houses. Therefore, the differences in lead levels noted above may
have been largely a result of the more severe initial conditions in encapsulation/enclosure houses.
Page 46
-------
1000-
5
c
,0
u
<§
•o
o
I
100-
10J
Air
Duct
Win.
Wall
Win.
Sill
Floor Entry Entry Entry Found. Bound.
Int. Ext. Soil Soil Soil
Sample Type
Control, n=17 homes oooa IBP Abated, n=35 homes
Figure 3-15. Estimated geometric mean dust-lead and soil-lead concentration (jig/g) in typical
abated and unabated homes by sampling location (CAP Study).
3.1.12 Milwaukee Retrospective Paint Abatement Study
This on-going study is examining the effectiveness of the lead-based paint abatement strategies
implemented in the Milwaukee area in 1989-1992 (Schultz, 1993). Damaged, painted surfaces with
lead loadings exceeding 1.0 mg/cm2 were abated. Abatement method and clean-up procedures varied
depending upon the practices of the particular abatement contractor. Only preliminary results from
this study were available, but are worth noting. Blood-lead concentrations were collected from 104
children before and (mostly) 3 to 12 months after the lead-based paint abatement. The arithmetic
mean blood-lead concentration reduced from 34 /xg/dL pre-abatement to 26 /xg/dL post-abatement, a
24% decline.
3.1.13 New York Chelation Study
This 1989-1990 study (Rosen et al., 1991; Markowitz et al., 1993; Ruff et al., 1993) was an
effort to ascertain the efficacy of a particular chelation therapy procedure on moderately lead-poisoned
Page 47
-------
children. Two hundred and one children with blood-lead levels between 25 and 55 /xg/dL were
administered a lead mobilization test (LMT) to determine whether chelation therapy might prove
effective. Children with a positive LMT underwent chelation therapy. For all children enrolled,
visual and XRF inspections of the paint in their residences were performed. Residences of 89% of
the children had sufficient lead-based paint to warrant an abatement. In addition to taking blood-lead
measurements, the authors measured cognitive ability and bone-lead content, using the net corrected
photon count (CNET) by L-XRF for the latter.
The reported results for this study emphasized overlapping subsets of the enrolled population.
The first set of analyses examined a subset of 174 children (71 chelated, 103 control). Six to seven
weeks following enrollment, average blood-lead levels among the 103 non-chelated children had fallen
2.5 jug/dL (mean at enrollment, 29.0 /xg/dL) and average bone-lead levels had fallen 3.3 CNET
(mean at enrollment, 125.3 CNET). The second set of analyses considered a subset of 154 children
(61 chelated, 93 control). Cognitive index rose 3.6 points (from 79.0 to 82.6), on average, among a
subset of 126 children (both chelated and non-chelated) six months following enrollment. The authors
concluded that cognitive index increased approximately one point for every 3 ^g/dL decrease in
blood-lead level. The third subset was of 59 children, 30 of whom were non-chelated. Mean blood-
lead levels among the 30 non-chelated children had fallen 6 /xg/dL by 6 weeks post-enrollment (from
29 /xg/dL to 23 /*g/dL) and fell an additional 2 /xg/dL (to 21 /xg/dL) by 24 weeks post-enrollment
(Figure 3-16). This represents a 28% decline as compared to an average decline of 37% among the
chelated children (39.5 /xg/dL to 25 jttg/dL). Mean bone-lead levels did not change among the non-
chelated children during this time period.
Though sifting through the various subsets is difficult, there was evidence that lead-based
paint abatement lowered blood-lead levels. Furthermore, the authors concluded that the results
suggest an association between declines in blood-lead levels and positive health outcomes (in addition
to the lowered blood-lead concentration).
Page 48
-------
40-1
u
c
o
u
TJ
D
O
_o
m
20-
10-
0-
-o
n
Population Considered:
2 4
Months Following Enrollment
*-*-* Subset |1. n=103 non-chelated children
e-o« Subset #2, n=126 children (chelated & non-chelated)
B-e-B Subset #3. n=30 non-chelated children
Figure 3-16. Arithmetic mean blood-lead concentration (jig/dL) by population considered (New
York Chelation Study).
3.1.14 Milwaukee Retrospective Educational Intervention Study
This study assessed the effectiveness of in-home education efforts in Milwaukee in 1991-1994.
The sample population consists of 431 children, 6 years of age and under, who were identified by the
Milwaukee Health Department as having blood-lead concentrations between 20-24 /xg/dL, had a
follow-up blood-lead measure, and had not moved before the follow-up measure. Of these, 195
children received an in-home educational visit and had a follow-up blood-lead measure after the in-
home visit. The control group of 236 children did not receive a health department in-home visit,
either because they were identified before the educational outreach program was in place, or because
the family could not be contacted after three attempts. The in-home educational visits were conducted
by para-professionals. Visits lasted approximately one hour and included education on nutrition and
behavior change, as well as housekeeping recommendations to reduce childhood lead exposure. Both
venous and capillary blood-lead measurements were used. Follow-up measurements were collected
Page 49
-------
2-15 months after initial blood-lead measures. Blood-lead concentrations were adjusted for the effects
of age and seasonal variations.
Blood-lead levels decreased for 154 of the 195 (79%) children who received an in-home
educational visit, compared to 124 of the 236 (53%) children in the control group. The arithmetic
mean blood-lead concentration for the Educational Outreach group declined by 18%, from 22 )ng/dL
to 18 jitg/dL, compared to the Control group decline of 5%, from 22 /xg/dL to 21 /xg/dL. The
difference between these declines is highly statistically significant (p= 0.001). These results imply
that educational intervention is effective in reducing children's blood-lead levels, although blood-lead
concentrations usually remained above 10
3.1.15 Granite City Educational Intervention Study
This 1991 study (Kimbrough, 1992, 1994; IDPH, 1995) included an effort to evaluate the
efficacy of educational interventions in reducing blood-lead concentrations in exposed individuals.
Children, under six years of age and recruited in Granite City, Illinois, constituted the sampled
population. Most homes in the community were built prior to 1920 and contained lead-based paint.
In addition, a secondary lead smelter had been in operation until 1983. Extensive educational efforts
were aimed at the children and families exposed to elevated levels of lead in the surrounding
environment. Instruction included identifying where lead-based paint was commonly found,
explaining available abatement procedures, detailing how to perform house cleaning procedures, and
reviewing hygienic procedures for young children. Venous blood samples, soil samples, dust samples
from within the residence, tap water samples, and an assessment of the lead content in interior paint
were collected. When possible, follow-up blood-lead measures were collected from children with
elevated blood-lead levels four months and twelve months after the initial sample.
Blood-lead levels were initially measured during the months of August and September 1991.
Of the 490 children under age 6, 78 (16%) had blood-lead levels greater than 9 ^g/dL. Of these, 5
had levels greater than 25 /ig/dL. Between the initial and four month samples, the families of
children with elevated blood-lead levels received extensive counseling in the prevention of lead
exposure. Mean blood-lead concentrations decreased significantly from an initial level of 14.6 /xg/dL
to 7.8 Mg/dL at the four month post-abatement measurement, but rose again to 9.6 /xg/dL by the
twelve month measure, as shown in Figure 3-17. Despite the rise in blood-lead levels, the 12 month
averages remained significantly below initial levels.
Page 50
-------
In addition, four month follow-up blood-lead concentrations were significantly lower than
initial levels in a small number of older children with elevated blood-lead levels. For 7 children aged
6 to 14 years, the arithmetic mean decrease was 5.9 /ig/dL, and for 3 children age 15 years or older,
blood-lead levels decreased 7.0
40
TJ
O
3
30
20
10
0-
4 8
Sampling Period Following Enrollment (Months)
12
Figure 3-17. Initial and follow-up arithmetic mean blood-lead concentration (/tg/dL) (Granite
City Educational Intervention Study).
The absolute decrease and percent decrease at the four month and twelve month follow-up
measures are shown in Figure 3-18 for a subgroup of 24 children under age 6 with initially elevated
blood-lead concentrations, for whom both follow-up measures were available. The magnitude of the
decrease in blood-lead concentration appears to be directly related to the magnitude of the initial
blood-lead concentration, as shown in Figure 3-19.
The striking declines in blood-lead levels provide evidence of the effectiveness of educational
efforts. The full implications of these declines in blood-lead concentration, however, could not be
ascertained since no measurements were collected for a control group of children. The one year
follow-up results for the subset of children suggest that seasonal variation alone does not account for
the observed decline in blood-lead concentrations, but may very well explain why the levels at 4
months were lower than those at 12 months.
Page 51
-------
10
5-
0
c?
-s^
J -5.
m
Q_
C
s, ~10
E
O
u
3
|
< -20-
-25
-30
B-
1
i|
o-
o
o
o
a
o
e
0
o
i,
8
1
0 1
a
a
o
B
II g
In s
In B
ST °
0-L
Q
D
50
40
30
20
10
?
0 |
-10 o
o
-20 J
-30 =
-40 ^
-50 ^
-60
-70
-80
-90
4 Months 12 Months
Sampling Parfod Follawtng Enrollmant
Figure 3-18. Absolute (on left) and percent change in blood-lead concentrations at 4-month
and 12-month follow-up (Granite City Educational Intervention Study).
10H
0
-10
-20
-30
o +
3
V , +O
» °
O
} 10 20 30
Blood—Lead Concentration at Enrollment
Timing Following Enrollment: 0004 Months + + +12 Months
40
Figure 3-19. Change in blood-lead concentration at 4-month and 12-month follow-up plotted
against initial blood-lead concentration (Granite City Educational Intervention
Study).
Page 52
-------
3.1.16 Milwaukee Prospective Educational Intervention Study
This study sought to assess the effectiveness of in-home educational efforts by an educational
outreach program established in Milwaukee in 1991 (Schultz, 1995). Children between 9 and 72
months of age were identified by the Milwaukee Health Department in June 1993. The children
selected for this prospective study had a blood-lead concentration greater than 20 jig/dL but had not
received a in-home educational visit in the previous year. Nor had their dwelling been abated in the
previous year. Two groups of children were prospectively examined.
(a) Standard In-home Educational Outreach.
The Educational Outreach group consisted of 54 children whose initial blood-lead
concentration was between 20-24 /*g/dL. Follow-up blood-lead measures were collected an average
of 2 months after the in-home visit for these children. For comparison, a control group of 122
children was selected from those previously identified in the Milwaukee Retrospective Study (see
Section 3.1.14), whose follow-up blood-lead measurement was taken within approximately 3 months
of the initial measurement. Children received in-home educational visits conducted by para-
professionals. Visits lasted approximately one hour and educated the families on nutrition, behavior
change, and housekeeping recommendations to reduce childhood lead exposure. Both venous and
capillary blood-lead measurements were used. Blood-lead levels were adjusted for the effects of
seasonal variation.
Only preliminary results from this study were available, but are worth noting. Both the
Educational Outreach and Control groups had initial mean blood-lead levels of approximately 22
/Lig/dL. On average, a significantly greater decline between initial and follow-up blood-lead
concentration was observed for the Educational Outreach group than for the Control group. The
decline was about 3 jtg/dL more in the Educational Outreach group.
(b) Pre-abatement Educational Outreach.
The Pre-abatement Educational Outreach group consisted of 28 children whose initial blood-
lead concentration was between 25 /ig/dL and 40 jug/dL. Lead-based paint abatements were required
for these children, but had not been implemented at the time of the in-home visit and follow-up blood-
lead measurement. Follow-up blood-lead measures were collected 2-6 months after the initial
measurement for the Pre-abatement Educational Outreach group. Children in the Pre-abatement
Educational Outreach group received the same visit as the study (a), plus an additional visit from a
Page 53
-------
Public Health Nurse, who conducted a general health assessment of the child and family and also
answered any questions about lead. In addition, a paint inspection was performed in all pre-
abatement homes. Both venous and capillary blood-lead measurements were used, and blood-lead
levels were adjusted for the effects of seasonal variation.
Again, only preliminary results from this study were available. Mean blood-lead
concentrations for the Pre-abatement Educational Outreach group started at 29 /xg/dL and declined by
19%. No control group was available for the pre-abatement educational outreach group, as extensive
efforts were made to contact all families where children had blood-lead levels at or above 25 /ig/dL.
Approximately the same percentage decline in blood-lead levels was observed in both the Educational
Outreach and Pre-abatement Educational Outreach groups. On average, larger absolute declines were
observed in the Pre-abatement group, suggesting that greater declines on an absolute scale may be
associated with higher initial blood-lead concentrations. This is consistent with results reported for
other lead-hazard intervention methods.
The results of these studies imply that educational intervention is effective in reducing
children's blood-lead levels, although blood-lead concentrations usually remained above 10 /ttg/dL.
3.2 SUMMARY OF SCIENTIFIC EVIDENCE
The available literature on lead hazard intervention efficacy focused on impeding the hand-to-
mouth pathway of childhood exposure to environmental lead sources. The emphasis on this exposure
pathway seems appropriate since it is recognized in the literature as the predominant pathway in
young children (USEPA, 1986; CDC, 1991; ATSDR, 1988). The pathway may be disrupted by a
variety of means including the abatement of lead-based paint, dust-lead level reduction procedures,
and elevated soil-lead abatement.
The literature is limited in its scope. It only covers some of the many abatement types and
methods used in practice. However, the studies suggest that both "in-place management" and "source
isolation or removal" methods were at least partially effective in reducing blood-lead concentrations.
There was no definitive evidence in the literature that one of these categories of methods was more
efficacious than the other. Source isolation or removal methods often had an accompanying risk of at
least short-term elevation of residents' blood-lead levels that must be factored into any summary of
intervention efficacy. In-place management methods, in turn, usually required sustained effort to
maintain their effectiveness.
A summary table for the identified lead intervention studies is presented in Table 3-1.
Page 54
-------
Summary tables for blood-lead concentration results and other body-lead burden results are presented
in Tables 3-2 and 3-3, respectively. A summary of environmental media efficacy is displayed as
Table 3-4. The 10 paint abatement studies examined all employed source removal or isolation
methods to abate the lead-based paint hazard. The literature suggests that the efficacy of these
methods depends in part on the safeguards employed to protect the occupants and their residential
environment during abatement. In the 1984-1985 Boston Retrospective (Amitai et al., 1991) and
1984-1985 Baltimore Traditional/Modified (Farfel and Chisolm, 1990) Paint Abatement studies,
average blood-lead levels were observed to increase 16% to 19%, on average, during abatement and
remain elevated following the intervention. The levels in Baltimore were elevated one month
following intervention, but in Boston they had decreased by two months post-abatement. In the case
of the Baltimore study, the authors suggested that the increase stemmed from incomplete abatement or
insufficient clean-up following the abatement. Dust-lead levels within the dwelling were exacerbated,
which led the authors to the conclusion that environmental exposure had merely been shifted from one
medium to another. In both the Boston and Baltimore studies, elevated blood-lead levels were
associated particularly with the dry-scraping and heat-gun methods of source removal which were
performed in 1984-1985.
In the Boston Retrospective study, lead-based paint abatement methods such as encapsulation,
enclosure, and replacement were associated with an average reduction of 2 to 3 /ng/dL in blood-lead
concentrations. The results of the HUD Demo study (HUD, 1991; HUD, 1990) also suggest that
clearance standards may be easier to meet via encapsulation and enclosure methods than via removal
methods. The CAP study (Buxton et al., 1994) indicated that long-term interior dust-lead levels were
somewhat higher, though not statistically higher, in encapsulation/enclosure homes than in removal
homes. However, as was noted earlier, this may have been largely a result of the more severe initial
conditions in encapsulation/enclosure houses. Still, in samples collected from floors and window
sills, both types of abatement method resulted in 18 to 24 month follow-up dust-lead levels below
HUD Guidelines (HUD, 1990) standards. Since the HUD Demo and CAP studies followed units that
were vacant before abatement, no changes in residents' blood-lead levels were available.
Lead-based paint removal methods were shown to lower the blood-lead levels of inhabitants in
the Boston Retrospective (Amitai et al., 1991), Central Massachusetts Retrospective (Swindell et al.,
1994), 1982 St. Louis Retrospective (Copley, 1983), 1990 St. Louis Retrospective (Staes et al.,
1994), New York Chelation (Rosen et al., 1991; Markowitz et al., 1993; Ruff et al., 1993), and
Milwaukee (Schultz, 1993) studies. These studies reported 18 to 29% declines in the blood-lead
Page 55
-------
concentration of affected residents. Comparable or larger declines post-intervention were identified
for other body-lead burden measures in the New York Chelation (Rosen et al., 1991; Markowitz et
al., 1993; Ruff et al., 1993) and 1982 St. Louis Retrospective (Copley, 1983) studies. The declines
were manifest as soon as 6 weeks after abatement. The magnitudes of these reductions are less than
the 80% potential discussed in Section 2.2. The remaining lead in the blood (20-29% declines leave
about 3/4 still present) may be due to any number of reasons including the mobilization of bone-lead
stores, the incomplete abatement of the lead-based paint and elevated dust-lead, and the potential for
exposure from other micro-environments. Since the analysis discussed in Section 2.3 suggests bone-
lead stores could not by themselves keep blood-lead levels elevated for even six months post-
abatement, the latter reasons seem plausible as contributors to elevated blood-lead concentrations.
There is evidence that lead-based paint abatement, by itself, may not be fully effective
because of the potential recontamination from unabated sources. In the CAP Study (Buxton et al.,
1994) geometric mean lead concentrations in unabated air ducts and soils were found to be
significantly higher in abated houses as compared to control houses. Moreover, geometric mean dust-
lead loadings in window wells were above HUD Guideline levels (800 /tig/ft2) for both abated and
control houses.
The two non-educational dust abatement studies primarily employed in-place management
methods. It seems unlikely that these methods aggravate childhood lead exposure if performed
improperly. Once such techniques are discontinued, however, the dust-lead hazard may return. The
Baltimore Dust Control Study (Charney et al., 1983) focused on managing the dust-lead hazard after
removing or isolating the lead-based paint hazard identified within the residence. The Baltimore study
noted that, "in most homes the initially high [dust-lead] levels were again present within 2 weeks after
the first visit" (Charney et al., 1983), although eventually dust-lead levels remained low between
visits. Similarly, the one-time dust abatement and paint stabilization performed in the Boston 3-City
Soil Abatement study (Weitzman et al., 1993) reduced window well dust-lead loadings for only a
short period of time.
Regular, extensive dust-lead hazard management efforts by trained personnel produced an
18% decline in mean blood-lead concentration and a 29% decline in FEP concentration for affected
residents; a control population exhibited only a 2% decline in mean blood-lead concentration (see
Baltimore Dust Control study (Charney et al., 1983)). The Seattle Track-In study (Roberts et al.,
1991) reported significantly lowered dust-lead levels after residents removed their shoes and used a
walk-off mat (no blood-lead measures were collected).
Page 56
-------
I
"S
5**
o
f
2
£
V
^
|
•o «
si
ll
V
S
I
o
(/)
I'-
ll
\
I
•
•s
(0 >
1 1
a> (g
3 0
w
!
1 §
££.
4
1
"o
£
n™
*;
Q
, ,
O
(A
^P
ll
if
*• lo'
: c
- (0
S
*M >
« c
fl-
^>
E |
O *
-|£
£&
E
3
1
|| ADatement
Studied
•
(U
82 St. Louis Retrospect
05
*~
•
•a
J>
90 St. Louis Retrospect
O5
"~
•
9
T3
3
•*-*
UD
O
E
03
0
Q
I
•
9
-D
D
0.
0
•
Q.
m
i
waukee Retrospective
—
•*
•
1
•
New York Chelation
*
•
^
Baltimore Dust Control
_c
r
o
E
0)
C
ID
0]
ID
Milwaukee Retrospectiv
Educational
U)
Q
Granite City Educationa
Milwaukee Prospective
Educational
A
•
•
0
0
C*)
o
0
00
'5
CO
n
TO
O>
'c
'(5
o
y
•o
ID
_OJ
H—
o
»
I
(D
JD
(D
_
Q.
O
O
T3
(D
I
(U
(U
(C
D.
73
0>
in
to
TJ
-------
i
o
£
e
_o
c
ea
ost-intervi
Up
a.
c
o
co
A
,
>
2.
S.
e
o
+i
01
*T
+*
in
0
a.
.c
i
up
T-
"5
1
•
c
§>
E C
10 (0
a £
SO
^sP
0
C
Months
nterventio
•4*
£ .M
(0 W
Q]
Ol
10 (0
ID f.
S o
^?
»
Si
0. *••
l>
i *
s -
a>
a 0)
ES
co
•al MassachL
Retrospectiv
c
CO
O
xp
CO
CM
6
CT>
0
C
'
,
ro
•
| Abated |
V)
3
C/3
O
en
en
vP
CM
6
CM
CM
O
C
'
,
ro
•
| Unabated
CO
Retrospectiv
vP
O)
o
CM
in
S?
CM
CM
CO
CM
in
CO
CM
in
T3
^
o
en
S?
00
CM
CO
00
CM
in
| Comparison A |
>s
O
CO
c
o
*-•
to
0
03
CD
O
CD
00
CO
CO
CM
•*
| Comparison B
CO
V)
a
0_
*•?
1^
O)
00
u
c
•
,
CO
CO
| Comparison A
>»
+-•
b
CO
o
•*-•
en
o
CQ
^
CO
en
CO
0
c
•
,
CO
1 Comparison B
CO
in
to
0.
^o
CM
CM
CO
O
U
C
•
,
CO
o
>•
i
Q.
CD
i
Milwaukee
itrospective 1
0)
DC
U
•
,
en
in
CO
o
^""
.
CO
o
1 No Chelation
•
s
1
^p
CO
CM
CO
O
CO
CM
in
o
CO
en
0
CO
No Chelation
1 (Subset) 1
New YorK
Chelation
Page 58
-------
O
1
S
I
2 a
T
1
S
a>
A
O
ss
t-lntervent
Up
«
£
j:
i
^
T5
S
c!
S5
*
g
1 i
s |
1*
c c
II
ths Post-
irvention
c w
1*
a>
ll
a U)
V)
3
Sf
I
"a «
1 'w
J
6
^^
*O
3
5?
0
c
,
•
#
00
in
CM
in
o>
CM
CM
in
CO
1 -=
.1 8
+•» m
ID £
3-3
•0 O
LJJ
o
c
,
•
in
in
CM
CO
CO
CM
CM
CM
CO
CO
CM
"o
w
*••
c
o
0
C
o
CO
o
3
CD S
.*
3 CD
CO >
5 '•*-*
— u
;= CD
S a.
CO
o
CD
a:
CM
CO
CM
O)
CM
vp
51-
in
,«.
in
in
*"~
a>
in
TJ
3
ti
0
CM
CM
ss
t
^.
CM
in
•"
CM
'> *
< •»-•
c - 2
"D o
1 <->
o
£ (0
(\ C
.2
.i± «!
cz cj
CO 3
>- TJ
O ai
o
c
,
•
S?
00
CM
in
CM
CM
in
I •*=
.1 So
*-* m
ca £:
o *;
^0
LU
c
,
•
in
CM
CM
CM
T—
CM
CM
CM
CM
~O
£
O
o
u
c
,
•
#
O)
CO
CM
00
CM
O)
CM
00
CM
CD "CO _
E g-g
2 -B S
co co S.
J3 o t;
CO 3 3
«S°
Q.
C
O
'+3
CO
o
CD 3
CD TJ
-* UJ
3
CO CD
? ^
_ -M
= O
2 CD
a
«
o
I
+-*
c
CD
CD
CO
O
'•M
co
o
•
o>
c
w
CD
TJ
•S.
U>
S
•£ s!
CD <-• O
- 5 —
TJ
ii 3 CD CD
S -5 «
I 5 =
g io g
4! -S2 a)
2 c
-
CD
.2 S £
CU c
:I
cu .*_. 3 •*-
E c w °
I | If
w £ c E
CD 3 CU 'Tl
CU CO CO ~
2 £ » »
O CO —
c c c
Page 59
-------
_o
o
u.
c
1
th Post-lnterve
Up
O
(D
A
i
J
S
£
ost-lnterver
Up
Q.
•B
o
5
(O
*-
T5
C
0)
O)
a
e .N
<§ «
(0
s>
c c
| 0
Months
Post*
tervention
c
f.8
g «
m
is
l!
0)
a. o
« 25
W
Q.
3
O
O
£
in
a
0)
TD
3
in
0)
CM
IN
s
^
CM
CM
CO
it
CO
O
CM
^
«~
3
Q
~° 2
"2 u
*-*
-Q
o
CM
in
CO
**P
CO
CO
CO
CO
CO
CM
in
CO
73
(U
(C
.a
<
~
5
a.
LU
u.
"o
+j
:;
2^
00
CO
o
CO
^
CM
in
O
CO
0
00
O
CO
c
o
_ro (u
U M
0
Z
•o
3
a.
CO
CO
o
CO
#
CO
in
o
CO
£
O
CO
c
o
•*-* '+-<~
ro a>
<-> 5
o
Z
"0 ~° 2
S * c
-2 0 3
i <" 0
™ 0 °
CQ u a>
•^ C
o .2
>• "*-*
£ -c
z <->
Page 60
-------
£
I
§
1
a.
I
«
£
g
5
1
|
a.
1
o
1
ill
fa
0
II
w
**t
to
01 -
•S 3 §
|£*
is
« W
jj
*DI
w
io
g> CO
O
f S
II
g
a
M
8
O *r 2
Z ° g
X
»
'c
IB
I
a
2
o
•o
OT
CO
O
*"
CO
CO
CO
CM
w
X
00
o
-
o
1
CM
O
in
CN
o
00
CM
CO
CM
a.
s
i n
LL_ Q
CO
CO
CO
X
00
CM
«-
in
en
in
CO
CM
O
oo
CO
CO
en
CN
CO
in
CM
a.
5 S
0 3
co
en
*~
00
to
CM
8
0)
X
00
*
•-
co
in
co
en
CO
^
CO
en
in
0
in
CO
in
CM
.a
a.
g I
0 Q
T5 _
li
"CD £
._ ^
is*
i^ °n5 ^
8 Q. <
CO
CO
•~
CO
CO
in
w
X
3
CM
i—
O
in
co
O
in
oo
T—
00
00
CM
CM
00
00
CN
f.
a.
O u
- ^
u. O
CO
CN
1—
CO
CO
CO
to
X
00
00
r—
•-
o
co
co
co
CN
o
00
in
en
CO
CM
1
5 a
O 3
co
CO
*~
en
00
CM
CM
CO
(O
X
00
CO
CO
«-
CO
oo
0
CO
CN
CO
• _
.E "5
> £
m ^
=5^8
o c ro
2? 'ro -Q
» CL <
^
*ro *-•
d) C "O £
o '& £ E
E =5 =6 ^ °J >
•;= ro o c ro 12
fO I— 2 CD -° ^
O3 s 5 0_ < C/J
CO
£
*~
co
in
CO
CD
_
00
•-
o
co
o
^
0
CN
in
o
CO
CN
_o
D.
s
O w
O D
a: Q
CD
CM
co
*~
en
CO
_
CO
CO
<-
in
CM
CO
in
CO
CO
CM
00
o
co
^>
CO
co
CM
_o
Q.
S
g a
O 3
1 =
5 w
to
CM
T-
CO
o
o
T-
s
_
CO
CO
oo
CO
CT)
CO
CO
CM
CM
cn
CN
00
CM
CD
CM
a.
S
«
o Q
-o _
11
c o
— CT1
S.3
uj Q.
* ^
"5
c *-
.,0) C
CO c QJ
o ~ E
E S *- 2 >
• ™ u. ^ ;Z ~
m «" a. < t/j
Page 61
-------
i
o
S
i
I
$
1
tt
(A
S
3
s
I
i
jj
<
$
0.
S
u_
1
f ! i
p <
§ °
a V>
O
II
4,4
2 ° E
to
|££
i=a-<
.
§ Q
5 to
O
£ c
o S
w
t*
a
V)
EQ
S V)
o
II
4
z E
d
r~~
^
;
0)
CO
IV
'
in
CM
CM
CO
CM
a.
o S
E Q
CM
CO
*~
;
00
0
CD
tn
_
:
o
CO
^
LO
1
I
CO
m
CO
CM
a.
§
5*-«
(O
O 3
In
CM
CO
^
J
§
CO
*
_
!
^.
CO
•*
1
CM
•>*
co
CO
CM
CL.
5
0
•a _ *j
.£ "5 ^
5 5 Q
"ro £
r~ ro
~ Q)
E ^
a0?
5 *—
1 ~
(D ^ ^~
i 1 „ | >-
To uj 'to •& +j
CQ = Q. < (/]
O
CM
tn
O
CO
T—
CO
o
CJ)
CM
10
CO
CO
CM
T—
*~
CO
O
CM
CM
in
CO
CO
a>
a.
S
'5
o
CM
CM
CM
rv
o
CM
"-
o
in
CO
o>
^.
T—
in
en
co
IV
CM
CM
CM
1
11
C Q
O
CM
5
O
CO
"-
O
in
CO
IV
IV
co
CM
in
T—
CO
CM
in
o
i—
en
CM
23
a.
S
5
o
•a — t-i
111
T3
cy5
en
CM
IV
E
CM
CO
!
1
1
CO
0
CM
•
oo
in
CO
CM
CO
CO
O)
a.
§
'o
C/5
en
CO
CM
CO
in
«-
,
!
1
1
en
CM
in
en
CM
CM
CM
CM
CM
CM
a.
b w
O 3
tL a
en
CM
en
IV
5
in
tn
!
'
1
00
tn
CO
IV
CO
00
CM
CM
CM
CM
CM
CM
.£>
0-
g
0
~o ^
II 3
> $ Q
<
C
0
.w
(O
Q.
E
o
0
en
|V
CM
O
CO
(D
•CM
!
'
1
en
CM
*~
en
en
CM
CM
CO
CM
CO
CM
D)
CL
'6
in
en
[v
CM
in
en
CM
CM
,
:
•
in
Tfr
CO
CM
CM
CM
CM
CM
CM
CL
O 0)
0 3
ti- Q
en
CM
O
in
O
CM
IV
co
CM
,
!
'
1
CO
CO
IV
00
CM
CM
CM
CM
£
5
o
"D _ <-
£-& %
§5:0
m
c
o
to
a
o
O
-U
•Q C
— We
I 2-1 1
"^ cj 2 o*
CD CO < it
T3
(0
?
T3
C
(0
E
o
0> 0)
Q. O)
1°
O T3
£ S
I!
(A
OJ T>
s-s
o to
1- '>
(D 03
o «»
O T3
<» E
i_ 03
is
is S
El
1!
w 5
-------
The three educational intervention studies also employed in-place management methods. In-
home educational visits emphasized proper housecleaning methods to reduce dust-lead levels,
improved hygiene habits to reduce hand-to-mouth lead exposure, and educated families on proper
nutrition to reduce the health effects of elevated body-lead levels. No abatements were performed in
the study homes. The Granite City Educational Intervention Study (Kimbrough et al., 1992, 1994)
found a 32% drop in mean blood-lead level from extensive educational outreach (a drop from 15
ftg/dL, on average). The implication of this decline was difficult to ascertain, however, since no
measurements were collected for a control group of children. Both the Milwaukee Retrospective
Educational Intervention Study (Schultz, 1994) and Milwaukee Prospective Educational Intervention
Study (Schultz, 1994) reported 18% declines in blood-lead concentrations following in-home
educational visits. The declines following educational intervention for these studies were significantly
greater than declines observed in control children.
The one study of soil abatement employed both source isolation or removal methods and in-
place management methods. The Boston 3-City Soil Abatement Study (Weitzman et al., 1993)
removed and replaced soil exhibiting elevated lead levels, but also stabilized the peeling paint and wet
mopped the interior dust. Soil-lead and floor dust-lead levels in the abated residences remained low
post-abatement. Blood-lead concentrations among affected inhabitants oscillated after abatement, but
did not return to pre-abatement levels. In fact, a modest decline of 1 to 2 ^g/dL in average blood-
lead concentration (19% of pre-abatement levels, on average) was reported approximately 1 year
following the abatements in Phase I. Similar temporal variation in the average blood-lead levels of
residents of unabated dwellings used as controls in the study was observed, with declines after 1 year
of 7.1% and 5.6% for Comparison Groups A and B, respectively. In Phase I, the control residences
underwent the same one-time paint stabilization procedure as the study residences. A subset of the
comparison populations underwent soil abatement in Phase II, and exhibited 41% (Comparison Group
A) and 13% (Comparison Group B) declines in mean blood-lead concentration nine months post-
abatement. It was unclear exactly why the unabated residents experienced temporal variation in Phase
I, though seasonal variation of a comparable magnitude has been identified previously in children's
blood-lead levels (Kinateder and Menton, 1992; Schultz, 1993). This was a potential complicating
factor in several of the efficacy studies. Also, the reductions reported for the control populations may
have reflected the impact of age and behavioral factors stemming from an increased environmental
awareness of the health hazard from lead.
Page 63
-------
4.0 CONCLUSIONS
This report assesses whether lead hazard intervention effectively improved the health of
exposed children. Ideally, efficacy would be demonstrated in specific health outcomes among
children benefitting from the intervention. However, only one such study was identified in the
literature (Rosen et al., 1991; Markowitz et al., 1993; Ruff et al., 1993), and its interventions
included chelation therapy. The lack of such studies is not surprising given the cost and difficulty
associated with measuring health outcomes, especially among asymptomatic children with low to
moderate lead exposure. Blood-lead concentration can serve as a surrogate health endpoint due to the
recognized association between elevated blood-lead levels and adverse health effects. Such measures
may not demonstrate the complete benefit of the intervention received by the child, but do illustrate
its successful impact. In addition, their quantitative character allows for comparisons of different
intervention strategies.
There is evidence within the literature that intervention did reduce exposed children's blood-
lead concentrations (Table 4-1). Declines on the order of 18-34% were measured in exposed
children's blood-lead levels six to 12 months following a variety of intervention strategies (Figure 4-
1). Declines were reported for extensive, carefully managed projects which abated or isolated sources
of lead, as well as routine cleaning procedures or educational instructions employed to alleviate the
lead exposure of children. It should be noted, however, that short-term elevations in exposed
children's blood-lead concentrations can result when abatements are performed improperly.
Moreover, the evidence for post-intervention blood-lead concentration declines among children with
pre-intervention levels less than 20 ^tg/dL is mixed.
Given the documented declines in blood-lead concentration, it is reasonable to investigate the
degree to which they represent the actual effect of the intervention (e.g., that decline beyond changes
due to unrelated factors). Four of the 16 identified studies also simultaneously traced changes in
blood-lead level among a population of children not receiving the studied intervention strategy. The
effect of their intervention may then be estimated as the difference in the decline recorded for the
study population and that for the control population (Figure 4.2). The Milwaukee Retrospective
Educational Study (Schultz, 1995) results indicate a 13.6% decline 2 to 15 months following
intervention as the effect of their in-home educational outreach efforts. Dust control measures,
conducted in the Baltimore Dust Control Study (Charney et al., 1983), were associated with a 16.1%
effect 12 months following initiation. Soil abatements, performed in the Boston 3-City Soil
Abatement Study (Weitzman et al., 1993; Aschengrau et al., 1994), exhibited an 11.5% effect by 11
Page 65
-------
months post-intervention. Finally, the 1990 St. Louis Paint Abatement Study (Staes et al., 1993) also
reported an 11.5% effect on the blood-lead levels of resident children 10 to 14 months following the
abatement of damaged lead-based paint (recall that a multiple linear regression model predicted a 13%
effect). Though the data are limited, these results suggest that these intervention strategies are
comparable in their effect on blood-lead concentrations.
Table 4-1. Summary of Intervention Efficacy for Identified Lead Hazard Intervention Studies
Study
Baltimore Dust Control
1982 St. Louis Retrospective
Baltimore 'Traditional"/"Modified"
Boston Retrospective
Baltimore "Experimental"
Central Massachusetts Retrospective
Seattle Track-In
1990 St. Louis Retrospective
Boston 3-City
HUD Abatement Demonstration
Comprehensive Abatement Performance
Milwaukee Retrospective LBP
New York Chelation
Milwaukee Retro. Educational
Granite City Educational
Milwaukee Pro. Educational
Effective in Reducing
Targeted Pathway of
Exposure?
Yes
nc
No
nc
Yes
nc
Yes
nc
Yes
Yes
Yes
nc
nc
na2
na2
na2
Effective in Reducing
Exposed Child's
Body-Lead Burden?
Yes
Yes
No*
Yes
na1
Yes
nc
Yes
Yes
na1
na1
Yes
Yes
Yes
Yes
Yes
nc - Measurements necessary to make an assessment of the intervention's effectiveness were
not collected.
* Modified abatement practices were associated with fewer increases in children's blood-lead
concentrations.
na1 Not applicable — Interventions performed on vacant housing.
na2 Not applicable — Educational interventions were utilized in this study; no environmental
interventions were performed.
Page 66
-------
Bolt. DustCntrl
Bolt. Tradltnl.
£ Bolt. Modified-
c
s: Man. Retro.
| Mast. <20 ug/dL
TE
g Boston 3-Clty-
a
Mllw. Retro. Ed. -
Granite City Ed.
Mllw. Pro, Ed.
12 months J _1fl% •
6- 12 months • —2355 *
1 month * +18% *
1 month . 3^
VOTb
7-9 months * —29% *
< 1 2 months T « ov*
— 1 on
<\ 2 month. ^^
10-14 months * —23% *
1 0 month* *')~xjy
5 mOmni OA?
2- 15 months -18%
12 months ^-TA5S*
2—6 months ^_^ otf
10 15 20 25 30 35 40
Blood-Lead Concentration (/ig/dL)
45
50
Figure 4-1. Summary of blood-lead concentration results for identified lead
hazard intervention studies.
Figure 4-2.
Bait. DustCntrH
in
c
£ 1 990 St. Louis -{
Boston 3-Clty
Mllw. Retro. Ed.-I
16.1X
12 months
10—14 months
11 months
2—15 months
11.5*
11.5X
13.6X
0 5 10 15 20 25 30 35 40
Blood—Lead Concentration 0-tg/dL)
Study Population: Experimental Control
Effect of lead hazard intervention as measured by declines in children's blood-
lead concentration.
Page 67
-------
The magnitude of the cited declines in blood-lead concentration are a function of a number of
factors. Issues of source apportionment may limit observed blood lead reductions. Unless the micro-
environments and lead hazards targeted by the intervention represent the full range of exposures, the
intervention can only be partially successful. The timing of the measurements (relative to the
intervention itself) and the kinetics of lead within the body, in turn, can impact the measured decline
in body-lead burden. Mobilization of bone-lead stores following an intervention is capable of
partially masking the effectiveness of the intervention. The length of time the masking can be
maintained depends upon the degree of masking and the true effectiveness of the intervention. The
analysis reported in Section 2.3 suggests that the observed 25% reductions in blood-lead level one
year following the intervention are not the result of bone-lead mobilizations masking a fully effective
intervention. In fact, it is unlikely that the interventions reduced the children's lead exposure by
more than 50%.
Evolution in the techniques and guidelines associated with lead hazard abatement make
comparison of the efficacy of different practices difficult. The literature does cite, however, beneficial
results from the abatement of lead-based paint, elevated dust-lead levels, and elevated soil-lead levels.
Moreover, declines in blood-lead levels after in-home educational efforts were observed in the same
range as the other interventions, at least up to one year following intervention. The successful lead-
based paint abatements described included removal methods, as well as encapsulation and enclosure
methods. In contrast, dry scraping without HEPA vacuum attachments and open-flame burning of
lead-based paint were both reported to produce considerable elevations in the lead burden of exposed
children. It is still unclear whether more costly, large-scale abatement strategies are more successful
than less expensive (though sometimes more labor intensive), in-place maintenance practices.
This dilemma emphasizes the limitations in the data currently available regarding the
effectiveness of lead hazard intervention. Some studies are currently under way to further examine
the efficacy of lead hazard interventions, including in-place management methods. The EPA is
conducting the Lead-Based Paint Abatement and Repair and Maintenance Study in Baltimore to
compare comprehensive and low-cost methods for lead-based paint abatement in terms of their
efficacy for reducing the levels of lead in residential house dust and children's blood. The EPA is
completing a study in Jersey City of strategies requiring lower up-front abatement costs. The 10 first-
year recipients of HUD Abatement Grants will also soon provide information on currently
implemented abatement practices. In a joint effort, the Centers for Disease Control and the EPA are
sponsoring low-cost lead-based paint abatement evaluations in Baltimore, Cleveland, and Boston. In
Page 68
-------
addition, results of the EPA 3-City Soil Abatement Demonstration Projects in Cincinnati and
Baltimore should be released soon. So too should the results of an educational intervention in
Leadville. Also, the National Institute of Environmental Health Sciences (NIEHS) is currently
sponsoring a clinical drug trial to determine whether treatment with the drug succimer provides any
additional benefit beyond environmental intervention. The trial is being conducted in four cities and
examines children with moderately elevated (20-44 /*g/dL) blood-lead concentrations. The results of
these studies will shed additional light on the effectiveness of lead hazard intervention and the trade-
offs between different in-place management and source isolation or removal strategies.
Page 69
-------
Page 70
-------
5.0 RECOMMENDATIONS FOR FUTURE INTERVENTION STUDIES
A critical question in planning studies to assess intervention effectiveness is the timing of the
measurements following the interventions. Pre-intervention measures should be collected to provide a
basis for comparison, but when should post-intervention measures be scheduled to best assess the
effectiveness of an intervention? Frequent measurement collection is difficult and costly, especially
for measures of health outcomes or body-lead burden. This issue is particularly pertinent when
measuring a child's blood-lead concentration.
One timepoint is straightforward, one year post-intervention. The one-year timepoint
minimizes the effect of seasonal variation (when compared to pre-intervention measurements), but
does allow time for the effects of the intervention to become more fully manifest. The literature on
bone-lead mobilization and the analysis detailed in Section 2.3 suggest that bone-lead stores mobilize
over many months rather than days. Furthermore, one-year post-intervention measures should
identify the effects of any recontamination of environmental media, and assess more fully the efficacy
of intervention strategies which do not fully interdict a pathway of lead exposure. Bone-lead stores
do have the potential for maintaining blood-lead concentrations above levels consistent with post-
intervention exposure for more than one year. However, fully effective interventions should manifest
much of their efficacy by the one year milestone.
It is also valuable to assess the progress of an intervention strategy by measuring its
effectiveness before one year. The most suitable timing of such a measure, however, depends upon
the age of the children targeted by the intervention, and the expected effectiveness of the intervention.
Both seasonal variation and bone-lead mobilization may potentially impact a measured blood-lead
concentration. Seasonal variation also impacts environmental lead measures. Table 5-1 presents
some of the advantages and disadvantages to measuring intervention efficacy at 1, 3, and 6 months
following the intervention. The bone-lead mobilization results presented in Table 2-5 are used to
determine which children may be assessed post-intervention without detrimental confounding of the
results due to mobilization of bone-lead stores.
Just as the timing of post-intervention measures is important, so too is the population of
children to be examined. As is apparent in Section 3.0, the majority of identified studies examined
children with considerably elevated blood-lead levels. These levels appear particularly elevated given
the current 10 ^g/dL level of concern cited by the CDC and the EPA. These studies provide valuable
information on the potential benefit of lead hazard intervention, but do not address a significant
portion of lead exposed children today. There is particular lack of information on the effectiveness of
Page 71
-------
lead hazard intervention among moderately exposed children, specifically children with blood-lead
concentrations at or below 20 pg/dL. Studies of such populations have occurred (Weitzman et al.,
1993; Markowitz et al, 1993; Ruff et al., 1993; Kimbrough et al., 1994), but their results are mixed
or not entirely relevant when considering non-medical interventions. The additional on-going
research described in Section 4.0 should help in this regard, but other work may be appropriate.
Absent too is information on effectiveness at time periods beyond one year, and more importantly, on
the efficacy achieved by preventing elevated blood-lead concentrations before they occur (primary
prevention). Measuring abatement effectiveness for moderately exposed children is particularly
difficult, but the results are necessary in order to determine the role intervention should play in
reducing childhood lead poisoning for this large population of children.
Page 72
-------
V)
^
12
Disadvantages
W
«
o>
«
+•<
<0
3
<
c
•2
*; 5
.£ «
0 >
II
1= *i»
ft
o
a.
• allows masking due to bone-lead
mobilization for all children,
regardless of the effectiveness of
the intervention
? °
0) C
5 ~
O ro
0 C
T3 J2 'E
<0 C (/> CO
— c *i
"S g-s §
•? -2 g S
TJ (0 >
8 £$ o
J3 "53 c ^
•£ C "a C
t3 * E
c •— c co
O « t in
S -2 § o S
•g § - s s
" -2 o > fe .S
1 5 I 8 8 1
<" "T CO Q. :=
S 2 2 o S 5
w c g <-• co c
(0 ~ j^ QJ "^ ~;
S E o ^ (o E
.H C „ T3 «S 2
1 > I SJ | >
E O> "S ^ Q. Q)
^
c
0
^~
•0 w £ 0 T3
m C ro
or^.o) (D*-* .c ooo
?O**-Q. SS"Om£^**~ -•- *t KJ Q-
•— **~ *"' c Q. +•* _n Q 03 >
•*-• "F
||
If
— §
d) 3
3 0
T3 >
go
!i2 «
1 > E
« CM .2
CD *i
N c c:
IIS
C = (D
E-5.E
(0
o
E
00
• allows masking due to bone-lead
mobilization for children 5 years
or older, and for younger children
experiencing partially effective
interventions
• allows considerable seasonal
variation in blood-lead and
environmental-lead levels
• same recontamination and
transient elevation disadvantages
as noted for 3 months
O)
c
3
i- . O
o >• >•
x- C
c ro o
.2 g o
5 g a
S •£ "5 g
g g. S .2
c H- co *;
•o S > |
1 It 1
1 II 5
° 6 0 0
4) CD "> >2
T3 §"_ "S
~ M ^ "§>
E * g\i
1 1 1 1
|c := S Q.
E o .£
-------
6.0 REFERENCES
Adurns, M.A. (1991) "Total Diet Study: Dietary Intakes of Lead and Other Chemicals." Chemical
Spedation and Bioavailability. 1991 (3): 37-41.
Agency for Toxic Substances and Disease Registry. (1988) "The Nature and Extent of Lead
Poisoning in Children in the United States: A Report to Congress." Atlanta: U.S. Department of
Health and Human Services, Public Health Service. DHHS Document No. 99-2966.
Altman, P.L., Dittmer, D.S. (1962) "Growth Including Reproduction and Morphological
Development." Biological Handbook. Federation of American Societies for Experimental Biology.
Washington, D.C.
Amitai, Y., Graef, J.W., Brown, M.J., Gerstle, R.S., Kahn, N., and Cochrane, P.E. (1987)
"Hazards of 'Deleading' Homes of Children with Lead Poisoning." American Journal on Diseases of
Children. 141:758-760.
Amitai, Y., Brown, M.J., Graef, J.W., and Cosgrove, E. (1991) "Residential Deleading: Effects on
the Blood Lead Levels of Lead-Poisoned Children." Pediatrics. 88(5).-893-897.
Aschengrau, A., Beiser, A., Bellinger, D., Copenhafer, D., Weitzman, M. (1994) "The Impact of
Soil Lead Abatement on Urban Children's Blood Lead Levels: Phase II results from the Boston Lead-
in-Soil Demonstration Project." Environmental Research. 67:125-148.
Assenato, G., Paci, C., Baser, M., Molinini, R., Candela, R.G., Altamura, B.M., and Giorgino, R.
(1986) "Sperm Count Suppression Without Endocrine Dysfunction In Lead-Exposed Men." Archives
of Environmental Health. 41 (6).-387-390.
Barry, P.S.I, and Mossman, D.B. (1970) "Lead Concentration in Human Tissues." British Journal of
Industrial Medicine. 27:339-351.
Barry, P.S.I. (1975) "A Comparison of Concentrations of Lead in Human Tissues. British Journal of
Industrial Medicine. 32:119-139.
Barry, P.S.I. (1981) "Concentrations of Lead in the Tissues of Children," British Journal of
Industrial Medicine. 38:61-71.
Battelle Memorial Institute and Midwest Research Institute, "Comprehensive Abatement Performance
Study." Draft Final Report to U.S. Environmental Protection Agency. January, 1994.
Battelle Memorial Institute (1994a) "Reported Effectiveness Derived from Methods for Lead Hazard
Intervention." Memorandum to U.S. EPA, Office of Pollution Prevention and Toxics, prepared under
Contract No. 68-D2-0139. September 1994.
Bellinger, D., Leviton, A., Waternaux, C., Needleman, H., and Rabinowitz, M. (1987)
"Longitudinal Analysis of Prenatal and Postnatal Lead Exposure and Early Cognitive Development."
New England Journal of Medicine. 316:1037-1043.
Page 75
-------
Burgoon, D.A., Rust, S.W., and Schultz, B.D. (1993) "A Summary of Studies Addressing the
Efficacy of Lead Abatement." Draft proceedings for ASTM Lead in Paint, Soil, and Dust
Conference, 1993.
Buxton, B.E., Rust, S.W., Kinateder, J.G., Schwemberger, J.E., Lim, B., Constant, P., and Dewalt,
G. (1994) "Post-Abatement Performance of Encapsulation and Removal Methods for Lead-Based
Paint Abatement," Lead in Paint, Soil, and Dust. Health Risks, Exposure Studies, Control Measures,
Measurement Methods, and Quality Assurance, ASTM STP 1226, Michael E. Beard and S. D. Allen
Iske, Eds., American Society for Testing and Materials, Philadelphia.
Campbell, B.C., Elliott, H.L., and Meredith, P.A. (1981) "Lead Exposure and Renal Failure: Does
Renal Insufficiency Influence Lead Kinetics?" Toxicology Letters. 9:121-124.
Carton, A., Maradona, A., and Arribas, M. (1987) "Acute-Subacute Lead Poisoning: Clinical
Findings and Comparative Study of Diagnostic Tests." Archives of Internal Medicine. 147:697-703.
Centers for Disease Control. (1991) "Preventing Lead Poisoning in Young Children - A Statement by
the Centers for Disease Control". Atlanta: U.S. Department of Health and Human Services, October
1991.
Chamberlain, A.C., Heard, M.J., Little, P., Newton, D., Wells, A.C., and Wiffen, R.D. (1978)
"Investigations Into Lead From Motor Vehicles." Report No AERE-R 9198; AERE Harwel,
Oxfordshire.
Charney, E., Kessler, B., Farfel, M., and Jackson, D. (1983) "Childhood Lead Poisoning: a
Controlled Trial of the Effect of Dust-Control Measures on Blood Lead Levels." New England
Journal of Medicine. 309:1089-1093.
Chisolm, J.J. Jr., Mellits, E. D., and Quaskey, S. A. (1985) "The Relationship between the Level of
Lead Absorption in Children and the Age, Type, and Condition of Housing." Environmental
Research. 38:31-45.
Copley, C.G. (1983) "The Effect of Lead Hazard Source Abatement and Clinic Appointment
Compliance on the Mean Decrease of Blood Lead and Zinc Protoporphyrin Levels." Mimeo. City of
St. Louis, Department of Health and Hospitals, Division of Health, Office of the Health
Commissioner, St. Louis, MO.
Farfel, M.R. and Chisolm, J.J. Jr. (1990) "Health and Environmental Outcomes of Traditional and
Modified Practices for Abatement of Residential Lead-Based Paint." American Journal of Public
Health. 80(10):1240-1245.
Farfel, M.R. and Chisolm, J.J. Jr. (1991) "An Evaluation of Experimental Practices for Abatement of
Residential Lead-Based Paint: Report on a Pilot Project." Environmental Research. 55:199-212.
Farfel, M.R., Chisolm, J.J. Jr., and Rohde, C.A. (1994) "The Longer-Terrn Effectiveness of
Residential Lead Paint - Abatement." Environmental Research. 66:199-212.
Page 76
-------
Feldman, R.G. (1978) "Urban Lead Mining: Lead Intoxication Among Deleaders." New England
Journal of Medicine. 298:1143-1145.
Fischbein, A., Anderson, K.E., Sassa, S., Lilis, R., Kon, S., Sarkozi, L., and Kappas, A. (1981)
"Lead Poisoning from 'Do-It-Yourself Heat Guns for Removing Lead-Based Paint: Report of Two
Cases." Environmental Research. 24:425-431.
Folashade, O.O. and Crockford, G.W. (1991) "Sweat Lead Levels in Persons with High Blood Lead
Levels: Experimental Evaluation of Blood Lead by Ingestion of Lead Chloride." The Science of the
Total Environment. 108:235-242.
Goyer, R.A. (1993) "Lead Toxicity: Current Concerns." Environmental Health Perspectives.
100:177-187.
Graziano, J.H., Lolacono, N.J., and Meyer, P. (1988) "Dose-Response Study of Oral 2,3-
Dimercaptosuccinic Acid in Children with Elevated Blood Lead Concentrations." Journal of
Pediatrics. 113:751-757.
Graziano, J.H., Lolacono, N.J., Moulton, T., Mitchell, M.E., Slavkovich, V., and Zarate, C. (1992)
"Controlled Study of Meso-2,3-dimercaptosuccinic acid for the management of childhood lead
intoxication." Journal of Pediatrics. 120:133-139.
Harley, N.H. and Kneip, T.H. (1984) "An Integrated Metabolic Model for Lead in Humans of All
Ages." Report to USEPA from New York University, Department of Environmental Medicine,
Contract No. B44899.
He, F., Zhang, S., Li, G., Zhang, S., Huang, J., and Wu, Y. (1988) "An Electroneurographic
Assessment of Subclinical Lead Neurotoxicity." International Archives of Occupational and
Environmental Health. 61:141-146.
Hyrhorczuk, D., Rabinowitz, M., Hessel, S., Hoffman, D., Hogan, M., Mallin, K., Finch, H.,
Orris, P., and Berman, E. (1985) "Elimination Kinetics of Blood Lead in Workers with Chronic Lead
Intoxication." American Journal of Industrial Medicine. 8:33-42.
Kawaii, M., Toriumi, H., Katagiri, Y., and Maruyama, Y. (1983) "Home Lead-Work as a Potential
Source of Lead Exposure for Children." International Archives of Occupational and Environmental
Health. 53:37-46.
Kehoe, R.A. (1961) "The Metabolism of Lead in Man in Health and Disease: The Normal
Metabolism of Lead." Journal of the Royal Institute of Public Health. 24:81-98.
Kimbrough, R.D. (1992) Statement to the Subcommittee on Investigations and Oversight, Committee
on Public Works and Transportation, U.S. House of Representatives, June 9, 1992.
Kimbrough, R.D., LeVois, M., and Webb, D.R. (1994) "Management of Children with Slightly
Elevated Blood-Lead Levels." Pediatrics. 93(2): 188-191.
Page 77
-------
Kinateder, J. and Menton, R. (1995) "Seasonal Rhythms of Blood-Lead Levels." Final Report from
Battelle to the U.S. Environmental Protection Agency. March 1995.
Leggett, R.W., Eckerman, K.F., and Williams, L.R. (1982) "Strontium-90 in Bone: A Case Study in
Age-Dependent Dosimetric Modeling." Health Physics. 43(3):307-322.
Markowitz, M.E., Bijur, P.E., Ruff, H.A., and Rosen, J.F. (1993) "Effects of Calcium Disodium
Versenate (CaNa2EDTA) Chelation in Moderate Childhood Lead Poisoning." Pediatrics. 92(2):265-
271.
McMichael, A.J., Baghurst, P.A., Wigg, N.R., Vimpani, G.V., Robertson, E.F., and Roberts, R.J.
(1988) "Port Pirie Cohort Study: Environmental Exposure to Lead and Children's Abilities at Age
Four Years." New England Journal of Medicine. 319:468-475.
Mordenti, J. (1986) "Man Versus Beast: Pharmacokinetic Scaling in Mammals." Journal of
Pharmacological Science. 75:11:1028-1040.
Nordberg, G.R., Mahaffey, K.R., and Fowler, B.A. (1991) "Introduction and Summary.
International Workshop on Lead in Bone: Implications for Dosimetry and Toxicology."
Environmental Health Perspectives. 91:3-7.
Office of Air Quality Planning and Standards. (1989) "Review of the National Ambient Air Quality
Standards for Lead: Exposure Analysis Methodology and Validation," U.S. Environmental Protection
Agency, Research Triangle Park, NC 27711. EPA-450/2-89-11. June 1989.
Rabin, R., Brooks, D.R., and Davis, L.K. (1994) "Elevated Blood Lead Levels among Construction
Workers in the Massachusetts Occupational Lead Registry." American Journal of Public Health.
84:1483-1485.
Rabinowitz, M.B. (1991) "Toxicokinetics of Bone Lead." Environmental Health Perspectives. 91:33-
37.
Rabinowitz, M.B., Wetherill, G.W., and Kopple, J.D. (1976) "Kinetic Analysis of Lead Metabolism
in Healthy Humans." Journal of Clinical Investigation. 58:260-270.
Rey-Alvarez, S., and Menke-Hargrave, T. (1987) "Deleading Dilemma: Pitfall in Management of
Childhood Lead Poisoning." Pediatrics. 79:214-217.
Roberts, J.W., Camann, D.E., and Spittler, T.M. (1991) "Reducing Lead Exposure from Remodeling
and Soil Track-In in Older Homes." Presented at the 84th Annual Meeting of the Air and Waste
Management Association. June 16-21, 1991.
Rosen, J., Markowitz, M., Bijur, P., Jenks, S., Wielopolski, L., Kalef-Ezra, P., and Slatkin, D.
(1991) "Sequential Measurements of Bone Lead Content by L-X-Ray Fluorescence in CaN^EDTA-
Treated Lead-Toxic Children." Environmental Health Perspectives. 91:57-62.
Page 78
-------
Rosen, J.F. (1985) "Metabolic and Cellular Effects of Lead: A Guide to Low Level Lead Toxicity in
Children." In: Dietary and Environmental Lead: Human Health Effects (K.R. Mahaffey, Ed.)
Elsevier, New York and Amsterdam, pp 157-185.
Ruff, H.A., Bijur, P.E., Markowitz, M.E. Ma, Y., Rosen, J.F. (1993) "Declining Blood-Lead
Levels and Cognitive Changes in Moderately Lead-Poisoned Children." Journal of the American
Medical Association. 269(IB)-.1641-1646.
Schroeder, H.A., Tipton, I.H. (1968) "The Human Body Burden of Lead." Archives of
Environmental Health. 17:965-977.
Schwartz, J. (1994) "Low-level lead exposure and children's IQ: A meta-analysis and search for a
threshold." Environmental Research. 64:42-55.
Schwartz, J., Otto, P. (1987) "Blood Lead, Hearing Thresholds, and Neurobehavioural Development
in Children and Youth." Archives of Environmental Health. 42:153-160.
Schwartz, J., Angle, C., Pitcher, H. (1986) "Relationship Between Childhood Blood Lead Levels and
Stature." Pediatrics. 77:281-288.
Shannon, M., Graef, J., Lovejoy, F.H. (1988) "Efficacy and Toxicity of D-Penicillamine in Low-
Level Lead Poisoning." Journal of Pediatrics. 112:799-804.
Spector, W. (1956) "Handbook of Biological Data."
Staes, C., Matte, T., Copley, G., Flanders, D., and Binder, S. (1994) "Retrospective Study of the
Impact of Lead-Based Paint Hazard Remediation on Children's Blood Lead Levels in St. Louis,
Missouri." American Journal of Epidemiology. 139(10): 1016-1026.
Swindell, S.L., Charney, E., Brown, M.J., Delaney, J. (1994) "Home Abatement and Blood Lead
Changes in Children With Class III Lead Poisoning." Clinical Pediatrics. September: 536-541.
U.S. Department of Housing and Urban Development. (1991) "The HUD Lead-Based Paint
Abatement Demonstration (FHA)." Washington, D.C. August 1991.
U.S. Department of Housing and Urban Development. (1990) "Comprehensive and Workable Plan
for the Abatement of Lead-Based Paint in Privately Owned Housing: Report to Congress."
Washington, D.C. December 1990.
U.S. Department of Housing and Urban Development. (1990) "Lead-Based Paint: Interim Guidelines
for the Hazard Identification and Abatement in Public and Indian Housing." Office of Public and
Indian Housing. September 1990.
U.S. Department of Housing and Urban Development. (1994) "Guidelines for the Evaluation and
Control of Lead-Based Paint Hazards in Housing." Washington, D.C. May 1994.
U.S. Department of Labor (1993) "Lead Exposure in Construction; Interim Final Rule." Federal
Register. 88(84):26590-26649.
Page 79
-------
U.S. Environmental Protection Agency. (1986) "Air Quality Criteria for Lead." Environmental
Criteria and Assessment Office, USEPA, Research Triangle Park, NC. Volumes I-IV.
U.S. Environmental Protection Agency. (1987) "Baltimore Integrated Environmental Management
Project - Phase II Report: Reducing the Hazards from Abatement of Lead Paint." Final Report.
U.S. Environmental Protection Agency (1994) "Guidance on Residential Lead-Based Paint, Lead-
Contaminated Dust, and Lead-Contaminated Soil." Office of Prevention, Pesticides, and Toxic
Substances Memorandum, July 14, 1995.
Weitzman, M., Aschengrau, A., Bellinger, D., Jones, R., Hamlin, J. S., Beiser, A. (1993) "Lead-
Contaminated Soil Abatement and Urban Children's Blood Lead Levels." Journal of the American
Medical Association. 269(13)-.1647-1654.
Yankel, A.J., von Lindern, I.H., and Walter, S.D. (1977) "The Silver Valley Lead Study: The
Relationship Between Childhood Blood Lead Levels and Environmental Exposure." Journal of the Air
Pollution Control Association. 27(8):763-767.
Page 80
-------
6.1 ADDITIONAL SOURCES OF INFORMATION
Charney, E. (1995), Personal Communication, February 1995.
Illinois Department of Public Health (IDPH) (1995). Additional information on the Granite City
Educational Intervention Study was provided by John R. Lumpkin, M.D., Director, and the Staff of
the IDPH.
Schultz, B. D. (1993) "Variation in Blood Lead Levels by Season and Age." Draft Memorandum on
data from the Milwaukee blood screening program. June 1993.
Schultz, B.D. (1995), Personal Communication, April 1995.
Page 81
-------
50272-101
REPORT DOCUMENTATION
PAGE
1. REPORT NO.
EPA 747-R-95-006
3. Recipient's Accession No.
4. Title and Subtitle
Review of Studies Addressing Lead Abatement Effectiveness
5 Report Date: July 1995
7 Author^) David A. Burgoon, Steven W. Rust, Nancy A. Niemuth, Brandon J. Wood
8. Performing Organization Rept. No.
9. Performing Organization Name and Address
Battelle Memorial Institute
505 King Avenue
Columbus, Ohio 43201-2693
10. Project/Task/Work Unit No.
G001017-03
11. Contract(C) or Grant(G) No.
(C) 68-D2-0139
(G)
12. Sponsoring Organization Name and Address
U.S. Environmental Protection Agency
Office of Pollution Prevention and Toxics
401 M Street, S.W.
Washington, D.C. 20460
13. Type of Report & Period Covered
Final Report
14.
15. Supplementary Notes
16. Abstract (Limit 200 words)
This report is a comprehensive review of the scientific literature regarding the effectiveness of lead hazard
intervention. One use of this review is to aid in assessing the potential benefits of Title X rule-making activities.
The literature is limited in its extent, and it only covers some of the intervention methods used in practice. However,
the scientific literature does indicate that blood-lead concentrations declined after lead hazard intervention for children with
blood-lead levels > 20 /zg/dL. Declines of 18-34% were measured in exposed children's blood-lead levels 6-12 months
following a variety of intervention strategies. Also, evidence suggests that short-term elevations in exposed children's blood-
lead concentrations may result when abatements are performed improperly.
Evolution in the techniques associated with abatement make comparison of the effectiveness of different practices
difficult. There is insufficient information available to identify a particular intervention strategy as markedly more effective
than another. The literature cites comparable reductions in blood-lead concentration resulting from the abatement of lead-
based paint, dust at elevated lead levels, and soil at elevated lead levels. Moreover, declines in blood-lead levels after in-
home educational efforts were observed in the same range as the other interventions.
17. Document Analysis
a. Descriptors
Lead Poisoning, Abatement, Effectiveness, Reviews
b. Identifiers/Open-Ended Terms
Lead Hazard Intervention, Abatement Effectiveness, Literature Survey, Blood-Lead Concentration
c. COSATI Field/Group
18. Availability Statement
Release Unlimited
19. Security Class (This Report)
Unclassified
20. Security Class (This Page)
Unclassified
21. No. of Pages
144
22. Price
(SeeANSI-239.18)
See Instructions on Reverse
OPTIONAL FORM 272 (4-77)
(Formerly NTIS-35)
Department of Commerce
-------
APPENDIX A
ABSTRACTS OF STUDIES ADDRESSING
LEAD ABATEMENT EFFECTIVENESS
-------
This page intentionally left blank.
-------
APPENDIX A
ABSTRACTS OF STUDIES ADDRESSING
LEAD ABATEMENT EFFECTIVENESS
A.I Baltimore Dust Control Study
Reference. Charney, E., Kessler, B., Farfel, M., and Jackson, D. (1983) "Childhood Lead
Poisoning: A Controlled Trial of the Effect of Dust-Control Measures on Blood Lead Levels." New
England Journal of Medicine. 309:1089-1093.
Pertinent Study Objectives. This study sought to assess whether periodic dust-control measures in
addition to lead-based paint abatement would be more effective in reducing blood-lead concentrations
than lead-based paint abatement alone.
Sampled Population. Forty-nine children aged 15 to 72 months with at least two confirmed blood-
lead concentrations between 30 and 49 /ig/dL formed the study population. These children were
divided into an experimental group of 14 children and a control group of 35 children. All the
children were patients at the Lead Poisoning Clinic of the John F. Kennedy Institute in Baltimore,
Maryland.
Intervention Strategy. Both the experimental and control group underwent lead-based paint
abatement which entailed removing all peeling lead-containing interior and exterior paint from the
residence. In addition, all child accessible surfaces (below 1.2 m) which may be chewed on were
covered or rendered lead-free. For the experimental group only, periodic dust-control involved twice
monthly visits by a dust-control team who wet-mopped all rooms in the residence where the dust-lead
loading was greater than 100 /^g/sq.ft.
Measurements Taken. Dust-lead loading measurements were collected from all areas within the
residence where the child spent time. The samples were collected with alcohol-treated wipes within a
1 ft2 area of floor or from the entire window sill. Blood-lead (PbB), free erythrocyte protoporphyrin
(FEP), and hematocrit levels were measured during regular visits to the clinic.
Study Design and Results. Measurements of PbB, FEP and hematocrit were taken approximately
every three months during the course of the study. For the experimental group, there was a
significant reduction in mean PbB and FEP after six months, and a further decrease after one year
(Tables A. 1-1, A. 1-2). In contrast, the mean value for the control group did not change significantly
over the twelve months.
For many of the children, PbB levels six months prior to the study were available. Although
the PbB levels for most children were stable prior to the study, a mean increase of 1.0 /xg/dL was
reported in each group, but these increases were not statistically significant. At the start of the study,
PbB levels for 8 of 11 children in the experimental group and 15 of 24 children in the control group
remained within 3 /ug/dL of their respective levels six months before.
Residential dust-lead loadings were collected for the experimental group during recruitment.
No dust-lead measurements were collected in the control residences so as to avoid drawing attention
to dust as a source of lead exposure. To assess the success in cleaning, dust-lead loading
measurements were also obtained before and after the dust-control teams completed their work.
Page A-l
-------
Though the cleaning was effective, elevated levels returned in most homes within two weeks
following cleaning. It took several months before all the study residences had persistent reductions in
dust-lead levels. After 12 months, however, the cleaning was effective. Within experimental
residences, the bimonthly dust-control efforts reduced the dust-lead loading on measured surfaces
(Table A. 1-3).
Table A. 1-1. Blood-Lead Concentration (/tg/dL) by Study Group and Time
Group
Experiment
Control
Time
Start
6 months
1 2 months
Start
6 months
1 2 months
N
14
14
14
35
33
35
Mean
38.6
33.3
31.7
38.5
38.7
37.8
Std. Error
5.2
3.6
2.6
5.2
2.6
7.9
Table A.l-2. Free Erythrocyte Protoporphyrin Concentration Oig/dL) by Study
Group and Time
Group
Experiment
Control
Time
Start
6 months
1 2 months
Start
6 months
1 2 months
N
14
14
14
35
33
35
Mean
203
158
144
231
216
208
Std. Error
99
76
82
103
125
130
Table A.l-3. Number of Experimental Residences by Maximum Dust-Lead Loading
and Time
Max. PbD (//g/ft2)
<100
100-200
200-400
400-800
800-1600
>1600
Study Inception
1
1
1
3
4
4
After 12 months
5
4
1
4
0
0
Page A-2
-------
Conclusions (including caveats). The lead loading of house dust may be reduced by regular and
focused dust-control efforts within the residence, and the blood-lead levels in children residing in
those homes can be significantly lowered. The children examined in this study were already lead-
poisoned, so it is unclear how efficacious such procedures would be with children exhibiting lower
blood-lead concentrations. The dust-lead loadings return rapidly to elevated levels if the cleaning
procedures are discontinued.
Page A-3
-------
A.2 1982 St. Louis Retrospective Paint Abatement Study
Reference. Copley, C. G. (1983) "The Effect of Lead Hazard Source Abatement and Clinic
Appointment Compliance on the Mean Decrease of Blood Lead and Zinc Protoporphyrin Levels."
Mimeo. City of St. Louis, Department of Health and Hospitals, Division of Health, Office of the
Health Commissioner, St. Louis, MO.
Pertinent Study Objectives. This study sought to demonstrate a significant difference between the
children living in abated environments after lead hazard intervention compared to children still
exposed to lead hazards.
Sampled Population. The comparison was made among children enrolled in the St. Louis Health
Division's Childhood Lead Poisoning Prevention Program and measured to have a blood-lead
concentration greater than 25 /tg/dL.
Intervention Strategy. The lead hazard intervention entailed the enclosure or removal of paint from
surfaces with peeling or broken leaded paint. Chewable surfaces with evidence of damage were
completely stripped. Extensive cleanup procedures were not required and were likely implemented
infrequently at best. Replacement of building components in question (e.g., window sills,
baseboards) was recommended, but used infrequently.
Measurements Taken. The blood-lead concentration measurements were collected during routine
venipuncture screening. Lead containing paint was identified using XRF.
Study Design and Results. A retrospective study compared those blood measurements which
identified the child as lead poisoned to follow-up samples collected six to twelve months following the
initial identification. A total of 102 children had sufficient samples collected to allow this
comparison. Follow-up blood-lead concentrations in children whose lead hazards had been abated
were found to be significantly lower than their initial levels (Table A.2-1). This was not the case for
children whose hazards had not yet been abated. Blood-lead concentrations fell significantly in
children regardless of the extent to which their guardians successfully met their scheduled
appointments. Similar results were seen for zinc protoporphyrin concentrations (Table A.2-2).
Table A.2-1. Blood-Lead Concentration (jig/dL) by Study Group
Study Group
All Children
Abated Homes
Unabated Homes
Compliance >50%
Compliance <50%
Sample Size
102
61
41
63
39
Initial Sample Arith.
Mean
46.13
48.31
42.88
47.60
43.74
Follow-up Sample
Arith. Mean
38.87
37.02
41.63
38.82
38.95
Page A-4
-------
Table A.2-2. Zinc Protoporphyrin Concentration (jig/dL) by Study Group
Study Group
All Children
Abated Homes
Unabated Homes
Compliance ^50%
Compliance <50%
Sample Size
102
61
41
63
39
Initial Sample
Arith, Mean
111.04
119.43
98.56
126.10
86.72
Follow-up Sample
Arith. Mean
87.19
81.56
95.56
96.06
72.85
Conclusions (including caveats). The results indicate that abatement of lead-based paint hazards
does significantly reduce the lead burden being borne by lead-poisoned children. The magnitude of
the reduction, however, is confounded with the timing of the sampling (seasonal variation may play a
role) and the age of the children (PbB levels usually peak at 2 years of age).
Page A-5
-------
A.3 Baltimore Traditional/Modified Paint Abatement Study
Reference. Farfel, M. R., and Chisolm, J. J. Jr. (1990) "Health and Environmental Outcomes of
Traditional and Modified Practices or Abatement of Residential Lead-Based Paint." American
Journal of Public Health. 80(10): 1240-1245.
Pertinent Study Objectives. The goal of this study was to evaluate the health and environmental
impact of traditional and modified practices for the abatement of lead-based paint.
Sampled Population. The study examined children residing in 71 residences abated in urban
Baltimore (53 traditional abatements, 18 modified abatements). Prior to abatement all the residences
had multiple interior surfaces coated with lead-based paint and housed at least one child with a blood-
lead concentration greater than 30
Intervention Strategy. Traditional abatement practices called for addressing deteriorated paint on
surfaces up to four feet from the floor, and all hazardous paint on accessible surfaces which may be
chewed on. Paint with a lead content greater than 0.7 mg/cm2 by XRF or 0.5% by weight by wet
chemical analysis was denoted hazardous. For traditional abatements, blow torches and/or dry
sanding were commonly used, the abated surfaces were not repainted, and clean-up typically entailed,
at most, dry sweeping. Modified practices excluded the use of open-flame burning and sanding
techniques and included the repainting of abated surfaces. In addition, it called for more extensive
clean-up efforts entailing wet-mopping with a high phosphate detergent, vacuuming with a standard
shop vacuum, and disposal of debris off-site. In addition, worker training, protection, and
supervision were provided.
Measurements Taken. Dust samples were obtained using a alcohol-treated wipe within a defined
area template (1 ft2). Blood samples were collected via venipuncture.
Study Design and Results. Serial measurements of lead in interior house dust-lead loading (PbD),
and children's blood-lead concentration (PbB) were collected. Increased dust-lead loadings were
measured immediately following traditional abatements (usually within two days) on or in close
proximity to abated surfaces (Tables A. 3-1, A. 3-2, and A. 3-3). Dust-lead levels measured after
modified abatements were also higher than pre-abatement levels, but not to the extent seen for
traditional practices. At six months post-abatement, PbD levels were comparable to, or greater than,
their respective pre-abatement loadings in both study groups. It should be noted that neither
traditional nor modified practices entailed the abatement of window wells within the residence.
Page A-6
-------
Table A.3-1. Floor Dust-Lead Loading (/tg/ft2) by Group and Time*
Time Period
Pre-Abatement
Post-Abatement
6 Months Post-
Abatement
Study Group
Traditional
Modified
Traditional
Modified
Traditional
Modified
Sample Size
280
82
271
50
234
57
Geometric Mean
251
288
1440
650
316
316
95% Conf.
Interval
(223, 288)
(204, 390)
(1198, 1719)
(455, 920)
(269, 362)
(242, 418)
Table A.3-2. Window Sill Dust-Lead Loading (pig/ft2) by Group and Time*
Time Period
Pre-Abatement
Post-Abatement
6 Months Post-
Abatement
Study Group
Traditional
Modified
Traditional
Modified
Traditional
Modified
Sample Size
249
45
246
64
199
66
Geometric Mean
1338
1802
3595
604
1542
1635
95% Conf.
Interval
(1096, 1616)
(1356, 2406)
(2889, 4459)
(446, 818)
(1226, 1942)
(1152, 2323)
Table A.3-3. Window Well Dust-Lead Loading (/tg/ft2) by Group and Time*
Time Period
Pre-Abatement
Post-Abatement
6 Months Post-
Abatement
Study Group
Traditional
Modified
Traditional
Modified
Traditional
Modified
Sample Size
150
37
139
24
100
32
Geometric Mean
15496
18274
14354
8083
12468
24879
95% Conf.
Interval
(11585, 20745)
(11316, 29515)
(11223, 18348)
(4887, 13369)
(9012, 17243)
(15301, 40450)
Dust-lead loading results were converted from mg Pb per m2 to /xg Pb per ft2
Page A-7
-------
In order to assess blood-lead concentration, an additional 25 modified practices abated
residences were considered. No dust samples were collected in these residences. All the residences
under consideration housed a total of 151 children eligible for PbB analysis. Seventy-eight of these
children had sufficient follow-up venous samples to allow their consideration in at least one
component of the analysis. Forty-six children who did not undergo any chelation therapy had pre-
and post-abatement samples. The post-abatement samples were collected within one month following
the completion of the abatement activities. In residences abated using either practice, PbB levels in
resident children rose significantly (Table A.3-4). At six months following abatement, 29 children
with no history of chelation therapy (14 residing in traditionally abated dwellings, 15 in modified)
continued to suffer from elevated PbB levels (arithmetic mean: 32.53 fj.g,ldL) which were not
significantly different from their pre-abatement measures (arithmetic mean, 30.67 fig/dL).
Table A.3-4. Blood-Lead Concentration (/tg/dL) by Group and Time*
Study Group
Traditional
Modified
Sample Size
27
19
Pre-Abatement
Arith. Mean
36.88
34.40
Std. Error
1.45
2.07
One Month
Post-Abatement
Arith. Mean
43.72
35.43
Std, Error
2.69
2.49
Blood lead concentrations were converted from /itnol/L to /ig/dL by multiplying by 20.72.
Conclusions (including caveats). Despite the implementation of improved practices, modified
abatements, like traditional abatements, did not result in any long-term reductions of levels of lead in
house dust or the blood of children with elevated pre-abatement PbB levels. In addition, the activities
further elevated blood-lead concentrations.
Page A-8
-------
A.4 Boston Retrospective Paint Abatement Study
Reference. Amitai, Y., Brown, M. J., Graef, J. W., Cosgrove, E. (1991) "Residential Deleading:
Effects on the Blood Lead Levels of Lead-Poisoned Children." Pediatrics. 88(5):893-897.
Pertinent Study Objectives. This study sought to evaluate the extent to which the lead poisoning of
children is exacerbated during the abatement of lead-based paint within their residence.
Sampled Population. The study population consisted of 114 children ranging in age from 11 to 72
months (median: 24 months) with at least one blood-lead concentration above 25 /xg/dL obtained prior
to deleading, one blood-lead sample collected during deleading, and one blood-lead determination
following the completion of the deleading process. All the children were enrolled in the
Massachusetts Department of Public Health's Childhood Lead Poisoning Prevention Program.
Intervention Strategy. The deleading process consisted of the removal or permanent coverage of
any paint with a lead content greater than 1.2 mg/cm2 which was loose and peeling, or present on
chewable surfaces accessible to the child (below 4 ft). Abatement was accomplished using an
unspecified combination of methods including dry scraping and sanding, blow torch burning, and
replacement or permanent enclosure of building components. Detailed cleanup practices (i.e., HEPA
vacuuming and TSP washing) and relocation of the occupants during the abatements were
recommended, but not uniformly followed.
Measurements Taken. The blood-lead concentration measurements were collected via venipuncture.
Study Design and Results. The geometric mean PbB in the 114 children rose during deleading and
fell following the completion of the deleading activities (Table A.4-1). Post-deleading measures were
determined an average of 49 ±8 days (mean ± standard error) after the deleading activities were
completed. The mid-deleading measures were obtained 63+4 days following the pre-deleading
samples. The decrease in geometric mean PbB post-deleading is due in part to 42 children who
underwent chelation therapy between the mid- and post-deleading measurements.
Table A.4-1. Blood-Lead Concentration Otg/dL) by Time
Time Period
pre-deleading
mid-deleading
post-deleading
Geometric Mean
36.4
42.1
33.5
Standard Error
0.6
1.5
1.0
In an effort to determine the effect of deleading activities alone, a subset of 59 children who
underwent no chelation therapy were examined. In this subset, an additional follow-up measure was
collected 250+14 days after completion of the deleading work. No evidence of a change in
geometric mean PbB was found, but blood-lead levels did fall at the post-deleading collection and fell
even further by the follow-up deleading collection (Table A.4-2).
Page A-9
-------
Table A.4-2. Blood-Lead Concentration Oig/dL) by Time
Time
pre-deleading
mid-deleading
post-deleading
follow-up
Geometric Mean
35.7
35.5
31.0
25.5
Standard Error
0.9
0.8
1.0
0.9
For 80 of the children, the specific method of deleading was available. Dry scraping and
torches considerably elevated the blood-lead levels of the affected children (Table A.4-3). By
comparison, children exposed to encapsulation, enclosure, or replacement abatement procedures
experienced a mean decrease in their blood-lead burden. The stability of PbB prior to deleading
activities was characterized for a subset of 32 children who had two blood samples prior to deleading.
The mean PbB rose from 35.4±1.3 jig/dL to 36.0+1.1 /*g/dL during the interval between these
samples (73+23 days).
Table A.4-3. Change in Mid-Deleading PbB (/tg/dL) by Method of Abatement
Method
dry scraping and sanding
encap, enclose, or replace
torches employed
# of Homes
41
12
9
Arith. Mean
+ 9.1
-2.25
+ 35.7
Std. Error
2.4
2.4
10.8
Conclusions (including caveats). Deleading may often produce a significant, transient elevation of
PbB in many children. It is most dangerous if accomplished with the use of torches, sanding, or dry
scraping. The results for non-chelated children should be viewed with caution, since their long-term
reductions may be due to reasons other than paint abatement. Perhaps the children were not chelated
because their levels were falling naturally.
Page A-10
-------
A.5 Baltimore Experimental Paint Abatement Study
Reference. Farfel, M. R. and Chisolm, J. J. Jr. (1991) "An Evaluation of Experimental Practices
for Abatement of Residential Lead-Based Paint: Report on a Pilot Project." Environmental Research.
55:199-212.
Farfel, M. R., Chisolm, J. J. Jr., and Rohde, C. A. (1994) "The Longer-Term Effectiveness
of Residential Lead Paint - Abatement." Environmental Research. 66:217-221.
Pertinent Study Objectives. The study sought to demonstrate and evaluate experimental lead-based
paint abatement practices developed in response to the inadequacies uncovered for traditional
abatement procedures (see A.3).
Sampled Population. The literature on this study examined two distinct subsets of dwellings in
urban Baltimore. The first set is composed of six older dwellings in Baltimore City which were built
in the 1920s. Each dwelling was a two-story six-room row home in poorly maintained condition with
multiple lead-based paint hazards. Four of the residences were vacant, two housed lead-poisoned
children. The second set consisted of 13 dwellings which had previously been abated between 1988
and 1991 according to Maryland regulations. Each dwelling had 1) at least six pairs of dust-lead
loading (PbD) measures taken from the same locations pre- and immediately post-abatement; 2) no
major renovations performed since abatement; and 3) occupancy by a family providing written
informed consent.
Intervention Strategy. The experimental practices called for the floor to ceiling abatement of all
interior and exterior surfaces where lead content of the paint exceeded 0.7 mg/cm2 by XRF or 0.5%
by weight by wet chemical analysis. Several methods were tested, including encapsulation, off-site
and on-site stripping and replacement. The abatements took place either in unoccupied dwellings or
the occupants were relocated during the abatement process. Lead-contaminated dust was contained
and minimized during the abatement, and extensive clean-up activities included HEPA vacuuming and
off-site waste disposal. In addition, extensive worker training and protection were provided.
Measurements Taken. Alcohol-treated wet wipes were used to collect dust-lead loading samples
from household surfaces within each residence. Soil samples were taken with a 15.24 cm stainless
steel probe.
Study Design and Results. This abstract considers the reported results for the two nonoverlapping
subsets of this study. The first set of analyses (Farfel et al., 1991) examined serial measurements of
lead in interior dust samples in six homes. Dust samples were collected immediately before initiating
abatement (pre-abatement), during the abatement, after the final clean-up (post-abatement), and one,
three, and six to nine months following the abatement. Dust-lead loadings immediately post-
abatement were significantly lower than pre-abatement levels (Tables A.5-1 through A.5-3). By six
to nine months following the abatements, these levels either improved further or remained unchanged.
PbD monitoring before, during, and after the abatement activities also provided information
on the effectiveness of particular measures. All floor and window treatments were associated with
significant decreases in PbD over time (Tables A.5-4a,b). Window replacement was reported to be
more effective in reducing dust lead loading than stripping the lead-based paint. In addition, vinyl
floor coverings produced lower dust-lead loadings than sealing old wooden floors with polyurethane.
Page A-11
-------
Table A.5-1. Floor Dust-Lead Loading (/tg/ft2) in Six Homes by Time Period*
Time Period
Pre-abatement
Post-abatement
6 Months Post
Sample Size
70
70
63
Geometric Mean
520
130
56
95% Conf. Int.
(390, 697)
(102, 176)
(46, 74)
Table A.5-2. Window Sill Dust-Lead Loading (/tg/ft2) in Six Homes by Time Period*
Time Period
Pre-abatement
Post-abatement
6 Months Post
Sample Size
34
35
31
Geometric Mean
4608
325
409
95% Conf. Int.
(3019, 7024)
(195, 557)
(242, 669)
Table A.5-3. Window Well Dust-Lead Loading (/tg/ft2) in Six Homes by Time Period*
Time Period
Pre-abatement
Post-abatement
6 Months Post
Sample Size
28
31
24
Geometric Mean
29422
938
1003
95% Conf. Int.
(18060, 47938)
(567, 1561)
(548, 1849)
Dust-lead loading results were converted from mg Pb per m2 to /*g Pb per ft2.
The second set of analyses (Farfel et al., 1994) examined wipe-dust samples (n = 179)
collected from thirteen study dwellings between December 1991 and January 1992. These measures
were made prior to abatement, immediately post-abatement, and 1.5 to 3.5 years post-abatement.
Dust-lead loadings 1.5 to 3.5 years post-abatement were significatly lower from pre- and immediately
post-abatement levels (Table A.5-5). Soil-lead concentration which ranged from 209 to 1962 /*g/g
(GM = 688 Mg/g, Log SD = 0.69) was not found to be a significant factor in explaining the change in
dust-lead levels. 1.5 to 3.5 years following intervention 78% of all dust lead loading measurements
were within Maryland's interim post-abatement clearance standards (200 jug/ft2 for floors, 500 /tg/ft2
for window sills, and 800 /xg/ft2 for window wells). Twenty-one of the 39 readings above the
clearance levels were from window wells. Ratios of dust-lead loadings at 1.5 to 3.5 years post-
abatement to those pre- and immediately post-abatement were calculated (Table A.5-6). Despite some
reaccumulation of lead in dust, geometric mean PbD levels 1.5 to 3.5 years post-abatement were
significantly less than pre-abatement levels.
Page A-12
-------
Table A.5-4a. Geometric Mean PbD Loadings (/tg/ft2) in Six Homes
by Type of Surface and Treatment*
Surface and
Treatment
Pre-Abatement
N
G. Mean
Post-
Abatement
N
G. Mean
Post-
Treatment
N
G. Mean
Post Clean-up
N
G. Mean
Floors
Urethane
Vinyl tile
58
12
492
688
58
12
2313
3391
56
12
28
622
58
12
139
102
Window Sills
On-site caustic strip
Off-site caustic strip
26
8
4069
6912
27
-
10043
24
7
1133
1115
27
8
316
372
Window Wells
New Window
On-Site caustic strip
11
17
15580
44361
13
16
2230
29376
13
12
1208
5639
14
17
502
1589
Table A.5-4b. Geometric Mean PbD Loadings (/ig/ft2) in Six Homes
by Type of Surface and Treatment*
Surface and
Treatment
1 Month Post
N
G. Mean
3 Months Post
IM
G. Mean
6-9 Months Post
N
G. Mean
Floors
Urethane
Vinyl tile
43
10
102
46
54
12
93
37
51
12
74
19
Window Sills
On-site caustic strip
Off-site caustic strip
24
6
418
1143
25
8
539
483
23
8
353
604
Window Wells
New Window
On-Site caustic strip
10
15
251
5314
13
15
474
7107
14
10
465
2945
Dust-lead loading results were converted from mg Pb per m2 to ^g Pb per ft2.
Page A-13
-------
Table A.5-5. Geometric Mean PbD Loadings (fig/ft2) in Thirteen Homes by Surface Type*
Surface Type
Floor
Window Sill
Window Well
Pre-abatement PbD
N
42
53
31
G. Mean
(95% CD
254
(143,452)
1041
(542,2007)
14214
(7339,27778)
Post-abatement PbD
N
47
54
41
G. Mean
(95% CD
13.9
(7.4,25.1)
13.0
(7.4,22.3)
34.4
(22.3,53.0)
1.5 to 3.5 Years
Post-abatement
N
71
59
49
G. Mean
(95% CD
40.9
(25.1,68.8)
103
(66,161)
600
(345,1041)
Table A.5-6. Ratios of 1.5 to 3.5 Year Dust-Lead Loadings (PbD) to Those
Pre- and Post-abatement for Thirteen Homes.
Surface
Type
Floor
Window Sill
Window Well
1.5 to 3.5 Years
Pre-abatement
(95% CD
0.16
(0.09,0.31)
0.10
(0.05, 0.20)
0.04
(0.02, 0.08)
1.5 to 3.5 Years
Immediate Post-abatement
(95% CD
2.9
(1.5, 6.3)
7.9
(4.4, 15)
17
(9.1,31)
Dust-lead loading results were converted from mg Pb per m2 to /tg Pb per ft2.
Conclusions (including caveats). The experimental methods resulted in substantial reductions in
interior surface dust-lead levels immediately post-abatement which were found to persist throughout a
6- to 9-month post-abatement period. By the 1.5 to 3.5 year post-abatement measures, 78% of the
readings remained below target levels (< 140 Mg/ft2)- Dust-lead concentrations at this time were
reduced to 16, 10, and 4% of pre-abatement levels for floors, window sills, and window wells,
respectively. This suggests that comprehensive lead-paint abatement is associated with short-term as
well as the longer-term control of residential dust-lead hazards. Reaccumulation of dust was greatest
for window wells. The magnitude of the decline in dust-lead loadings following abatement may have
been exaggerated for the first subset since vacant units are likely to contain more dust than occupied
units.
Page A-14
-------
A.6 Central Massachusetts Retrospective Paint Abatement Study
Reference. Swindell, S. L., Charney, E., Brown, M. J., Delaney, J. (1994) "Home Abatement and
Blood Lead Changes in Children With Class III Lead Poisoning." Clinical Pediatrics.
September:536-541.
Charney, E., Personal Communication, February, 1995.
Pertinent Study Objectives. This retrospective study was designed to assess the effect of residential
lead-based paint abatements practiced between 1987 and 1990 in central Massachusetts. More
stringent home deleading regulations were enacted in Massachusetts in 1988 during the conduct of the
study.
Sampled Population. The sample population consisted of 132 children ranging in age from 12 to 91
months (mean: 35 months) who were identified by the Massachusetts Department of Public Health as
having a blood-lead concentration (PbB) >25 Mg/dL between 1987 and 1990, and whose homes were
abated during this period. Moreover, the child must have had at least one venous PbB determination
within 6 months prior to abatement; at least one venous PbB detemination 2 weeks to 6 months
following abatement; must not have received chelation therapy during that time period; and must have
resided in the same dwelling throughout the study period.
Intervention Strategy. Interventions prior to 1988 consisted of the removal or permanent coverage
of any paint with a lead content greater than 1.2 mg/cm2 which was loose or peeling, or present on
chewable surfaces accessible to the child (below 4 ft). No standard abatement methods, dust-control
measures or cleanup procedures were mandated. After 1988, only hand-scraping and replacement of
parts were acceptable removal methods and all occupants were removed from the dwelling during the
entire deleading and cleaning process. Cleanup involved vacuuming all surfaces with a high-
efficiency particle air (HEPA) filter vacuum, followed by wet-mopping and sponging with a trisodium
phosphate cleaning solution and then a second HEP A vacuuming. Abatement contractors were
licensed, which required completion of a 3-day course and passing a certifying exam.
Measurements Taken. Blood-lead concentration measurements were collected via venipuncture.
Study Design and Results. Childrens' blood-lead concentration measures at most 6 months prior to
initiation of abatement were compared to the last measurement collected within one year following
abatement. The actual range of post-abatement measures was 3 to 52 weeks following abatement.
Although a venous PbB level of >25 /^g/dL was chosen as a criterion for this retrospective study,
blood-lead concentration immediately prior to the abatement were less than 25 /ig/dL for some
children. In these cases, the authors suggest that the pre-abatement PbB measure might have reflected
some early abatement or education effects. Of the total 132 children, 103 (78%) showed a reduction
in PbB within one year following intervention. Table A.6-1 presents the number of children whose
blood-lead levels increased or decreased between pre- and post-abatement measures. However, this
reduction varied with pre-abatement level (Table A.6-2). In fact, mean blood-lead levels for subjects
with initial PbB >20 /-ig/dL decreased, while mean blood-lead concentrations for subjects with pre-
abatement PbB <20 /xg/dL increased from 16.7 to 19.2 jtg/dL.
Page A-15
-------
Table A.6-1. Number of Children Whose PbB Changed Between Pre- and
Post-abatement Measurements by Amount of Change
Change in PbB Between
Pre- and Post-abatement levels (//g/dL)
Decrease
No Change
Increase
>12
9-12
5-8
1-4
0
1-4
5-8
9-12
>12
Number of Children
12
20
31
40
4
9
5
5
6
Table A.6-2. Blood-Lead Concentration (/tg/dL) by Pre-abatement Level
Pre-abatement
PbB Level
Complete
Sample
> 30
20- 29
< 20
Sample
Size
132
33
79
20
# (%) in Sample
with Decreased
Post-abatement
PbB
103
(78%)
32
(97%)
64
(81%)
7
(35%)
Mean PbB Level
Pre-abatement
26.0
34.2
24.9
16.7
Post-abatement
21.2
23.2
20.8
19.2
Mean
Change in
PbB Level
(%)
-4.8
(-18%)
-11.0
(-32%)
-4.1
(-16%)
+ 2.5
( + 15%)
There were no significant differences in the reduction of blood-lead concentration between
males and females, nor were there indications of differences among age groups. However, all age
groups showed significant decreases post-abatement. Despite the more stringent regulations beginning
in 1988, reduction in PbB levels by calendar year of abatement were not found to be meaningful
(Table A.6-3). The increase of 0.1 /^g/dL in 1990 was significantly different from the decreases
observed in the three previous years. This increase was accounted for by a small sample size (n=13)
and the presence of two children whose levels increased markedly (from 16.0 and 17.0 /xg/dL to 29.0
and 31.0 /xg/dL). The median change from pre- to post-abatement measures was a decrease of 2.0
/xg/dL in that year, while the mean PbB remained relatively unchanged when the two children were
excluded.
Among 72 children with more than one pre-abatement measure, 40 children whose levels
were declining (defined as a decrease greater than 5 jig/dL) prior to abatement exhibited only a
modest
Page A-16
-------
Table A.6-3. Blood-Lead Concentration (/tg/dL) Change by Year
Year
1987
1988
1989
1990
Sample
Size
29
48
42
13
Mean PbB Level
Pre-abatement
27.5
26.1
24.3
23.9
Post-abatement
22.2
20.4
19.6
24.0
Mean Change in PbB
Absolute
-5.3
-5.7
-4.7
+ 0.1
Percent
-19%
-21%
-19%
+ 0.4%
Table A.6-4. Blood-Lead Concentration (/ig/dL) by Timing of Post-Abatement Measure
# of Days
Post-abatement
14-90
91-180
181-270
271-365
Sample Size
29
47
31
25
Mean PbB Level
Pre-abatement
24.2
25.3
27.5
25.8
Post-abatement
21.2
20.0
23.4
19.4
Mean Change
in PbB
-3.0
-5.3
-4.1
-6.4
further mean decline of 1.9 /ig/dL. A more significant mean decline of 8.2 )ng/dL was observed for
the 32 children whose initial levels were relatively constant.
Seasonal variations in blood-lead concentrations may be a factor in confounding the declines,
especially since post-abatement measures were taken from 3 to 52 weeks after the abatement process
(Table A.6-4). But all mean blood-lead concentrations decreased by their post-abatement measure
indifferent of timing.
Conclusions (including caveats). These results demonstrate that abatement of lead-contaminated
paint in residential homes is associated with a modest decline in blood-lead levels. The significant
decline among 32 children with stable levels prior to abatement suggests that regression to the mean
cannot fully account for the observed decline. Moreover, the magnitude of the decline appears to
depend upon the child's initial PbB. For children with blood-lead levels >25 ^g/dL, and particularly
above 30 jug/dL, lead-based paint abatement as practiced between 1987 and 1990 was associated with
an approximate 18% mean decline in blood-lead concentrations. The decline was not as significant in
children whose pre-abatement PbB levels were less than 25 ^g/dL, and particularly below 20 //.g/dL.
The decline may be confounded by the quality of the abatement process itself. Although more
stringent regulations were enacted in 1988, the prescribed methods may not have been used
immediately. Also, seasonal and age variation in blood-lead concentrations could significantly impact
the observed decline, depending upon the period of time between the measures and the season in
which the measures were collected.
Page A-17
-------
A.7 Seattle Track-In Study
Reference. Roberts, J. W., Camann, D. E., Spittler, T. M. (1991) "Reducing Lead Exposure from
Remodeling and Soil Track-In in Older Homes." Presented at the 84th Annual Meeting of the Air
and Waste Management Association. June 16-21, 1991.
Pertinent Study Objectives. The study sought to determine the extent to which low cost dust-control
measures successfully lower household dust-lead loading (PbD).
Sampled Population. Forty-two homes in Seattle and Port Townsend, Washington built before 1950
formed the sample populations.
Intervention Strategy. The abatement procedures considered between 1988 and 1990 were strictly
low-cost dust reduction procedures: use of a vacuum cleaner with an agitator bar, removing shoes at
the entrance to the residence, and installation of walk-off mats.
Measurements Taken. Dust samples were collected from rugs within the residence using a Hoover
Convertible vacuum cleaner. Soil samples were scraped from within one foot of the residence's
foundation. Total dust-lead loadings and fine dust-lead loadings (seived before analysis) were
measured.
Study Design and Results. The study employed piece-wise regression analysis to assess which
factors determine the dust-lead loading within a residence. Significant pairwise correlations were
found between In(PbD) and removing shoes at the door (r=-0.62) and the presence of a walk-off mat
at the home's entrance (r = -0.48). Lower dust-lead levels were found in homes where the residents
removed their shoes and/or utilized a walk-off mat (Table A.7-1).
Table A.7-1. Dust Levels within Residences by Abatement Procedure
Measure
Number of Homes
Total Dust Loading (mg/ft2)*
Fine Dust Loading (mg/ft2)*
Fine Dust-Lead Loading (//g/ft2)*
Fine Dust-Lead Concentration (ppm)
Shoes Off
5
325.2
74.3
28.8
320
Shoes On
32
2415.5
929
994.1
780
Walk-off Mat
6
622.5
157.9
53.9
430
Loadings converted from amount Pb per m2 to per ft2.
The occupants of three homes tested in the study began removing their shoes upon entry for at
least five months prior to the collection of a second PbD measurement from their carpets. In
addition, the occupants of one of these homes installed walk-off mats at both entrances and began
vacuuming twice weekly. The geometric mean dust-lead loading fell from 1588.6 jig/ft2 to 23.2
fig/ft2 in these homes.
Conclusions (including caveats). The data presented here suggest that the control of external soil
and dust track-in by removal of shoes and/or the use of a walk-off mat will reduce the lead exposure
from house dust. Lacking any blood measurements, the impact these interventions may have had on
childhood lead exposure are somewhat difficult to ascertain.
Page A-18
-------
A.8 1990 St. Louis Retrospective Paint Abatement Study
Reference. Staes, C., Matte, T., Copley, G., Flanders, D., and Binder, S. (1994) "Retrospective
Study of the Impact of Lead-Based Paint Hazard Remediation on Children's Blood Lead Levels in St.
Louis, Missouri." American Journal of Epidemiology . 139(10): 1016-1026.
Pertinent Study Objectives. The study attempted to assess, via a retrospective cohort study, the
effectiveness of lead-based paint abatement in reducing children's blood-lead concentration.
Sampled Population. The sample population consisted of children under six years of age who were
identified by the St. Louis City Health Department's Childhood Lead Poisoning Prevention Program
as having a blood-lead concentration >25 /xg/dL between January 1, 1989 and December 31, 1990.
Moreover, the child had to reside, for six months prior and six months following diagnosis, in a
dwelling with an identified lead-based paint hazard (at least one chipping or peeling paint surface with
a lead content >0.7mg/cm2). Children who experienced chelation therapy were excluded. One
hundred and eighty-five children met the criteria for the study, of which 71 had follow-up blood-lead
measures 10-14 months following diagnosis.
Intervention Strategy. Intervention entailed the abatement of peeling or chipping lead-based paint
(as identified by XRF to have a lead content >0.7 mg/cm2) within the dwelling via enclosure or
removal and replacement. No extensive clean-up procedures, other than removal of obvious
remediation debris, accompanied abatement. Most likely, the families were not relocated from the
dwelling during the intervention. In addition, educational interventions were initiated following the
child's diagnosis of elevated blood-lead levels.
Measurements Taken. The blood-lead concentrations (PbB) were collected via venipuncture.
Study Design and Results. The geometric mean PbB among the 189 children selected was 33.6
/xg/dL (range, 25-53 ^g/dL). Seventy-one of these children had their blood-lead concentration
measured 10-14 months following the initial diagnosis. 49 of these 71 children lived in dwellings
which had been abated prior to follow-up measures. Blood-lead levels for children living in abated
dwellings decreased by 23% (Table A. 8-1). This decline was significantly (p=0.07) greater than the
12% reduction observed in geometric mean PbB among the 22 children residing in unabated
dwellings.
Table A.8-1. Blood-Lead Concentration (jig/dl) by Abatement Group
Group
Abated
Unabated
N
49
22
Range
25-51
28-45
Geo. Mean
34.9
35.1
Follow-up
Geo. Mean
26.7
30.9
Percent
Decline
23%
12%
A multiple linear regression model predicting the change in geometric mean PbB at 10-14
months following diagnosis was fitted. The dwelling's abatement status at the time of the follow-up
blood sample (e.g., abated or unabated) and whether the blood-lead level at diagnosis exceeded 35
were statistically significant (p<0.10). The geometric mean PbB of children residing in abated
Page A-19
-------
dwellings was estimated to decrease 13% (95% CI, -25% to 1%) more than that of children residing
in unabated dwellings. Moreover, the geometric mean PbB of children with an initial PbB >35
jtg/dL was estimated to decline by 17% (95% CI, -27% to -5%) more than that of children with
lower PbB.
The decline in PbB 10-14 months following diagnosis was found to increase as the length of
time since the lead-based paint abatement had occurred increased. The abatements usually occurred
sometime after diagnosis. Finally, the reported 10-14 month post-diagnosis declines may be
underestimates, since children with such extended follow-up measures were found to experience
smaller declines in PbB at 2-4 months post-diagnosis than children without such extended measures.
Conclusions (including caveats). For lead-poisoned children in St. Louis, the decline in geometric
mean PbB is greater for children whose dwellings undergo lead-paint hazard abatement than for
children whose dwellings do not. The magnitude of the efficacy appears, however, to depend upon
the child's initial blood-lead concentration. The follow-up measures were reported with respect to the
timing of the diagnosis, rather than the abatement, so the potential masking effect of bone-lead
mobilization cannot be assessed. The reference suggests many of the follow-up measures were
collected less than six months following the abatement.
Page A-20
-------
A.9 Boston 3-Citv Soil Abatement Project
Reference(s). Weitzman, M., Aschengrau, A., Bellinger, D., Jones, R., Hamlin, J. S., Beiser, A.
(1993) "Lead-Contaminated Soil Abatement and Urban Children's Blood Lead Levels." Journal of the
American Medical Association. 269(13): 1647-1654.
Aschengrau, A., Beiser, A., Bellinger, D., Copenhafer, D., Weitzman, M. (1994) "The
Impact of Soil Lead Abatement on Urban Children's Blood Lead Levels: Phase II results from the
Boston Lead-in-Soil Demonstration Project." Environmental Research. 67:125-148.
Pertinent Study Objectives. This project endeavored to assess whether a significant reduction
(> 1000 ppm) in the concentration of lead in residential soil results in a significant decrease (> 3
/tg/dL) in the blood-lead concentration (PbB) of children residing at the premises.
Sampled Population. Volunteers were sought among children residing in areas in Boston already
known to have high incidence of childhood lead poisoning and elevated soil-lead concentrations. A
total of 152 children were enrolled, each satisfying the following criteria:
• less than or equal to four years of age,
• blood-lead concentration between 10 and 20 ^tg/dL with no history of lead poisoning, and
• a minimum median residential soil-lead concentration of 1500 ^g/g (ppm).
Intervention Strategy. This project employed four lead hazard interventional procedures: 1) an
initial one-time interior paint stabilization by removing exposed paint chips; 2) one-time interior dust
abatement via wet mopping and HEPA vacuuming; 3) extensive soil abatement; and finally 4) interior
and exterior lead-based paint abatement. Interior paint stabilization consisted of vacuuming loose
paint areas with a HEPA vacuum, washing loose paint areas with a TSP solution, and painting the
window wells with primer. Deleading consisted of removing leaded paint from mouthable surfaces
below 5 feet, and making intact all paint above 5 feet inside the home, and exterior areas. Soil
abatement consisted of removing surface soil to a depth of 6 in. for homes, and replacing with top
soil containing minimum lead levels. Dispersal of soil during the abatement was retarded by wetting
the soil, preventing track-in by workers, containing the abatement site with plastic, and washing all
equipment.
Measurements Taken. Extensive environmental media and body burden samples were collected:
• composite core soil samples,
• vacuum dust samples,
• first draw water samples,
• interior and exterior paint assessment via portable XRF,
• venipuncture blood samples to assess blood lead concentration, and free erythrocyte
protoporphyrin (FEP) and ferritin levels; and,
• hand-wipe samples.
Study Design and Results. Each child enrolled was randomly assigned to one of three experimental
groups: Study (54 children), Comparison A (51 children), or Comparison B (47 children). During
Phase I, the Study group received interior paint stabilization, interior dust abatement, and soil
abatement. Comparison Group A received interior paint stabilization and interior dust abatement.
Only interior paint stabilization was performed for Comparison Group B. During Phase II, which
took place an average of 12 months after Phase I interventions, both comparison groups received soil
abatement and all three experimental groups were offered lead-based paint abatement (Table A.9-1).
Page A-21
-------
Table A.9-1. Schedule of Activities.
Phase
Phase 1
Phase II
Activity
Baseline Blood Sample
(9/89-1/90)
Intervention
I
(9/89-1/90)
Post-Abatement Blood
Sample I
(7/90-11/90)
Intervention
II
(9/90-1/91)
Post-Abatement Blood
Sample
II
(7/91-8/91)
Study Group
N = 54/54
Soil and Interior
Dust Abatement,
Loose Paint
Stabilization
N = 54/54
N = 54/54
Paint Deleading
N = 23/54
N = 33/54
Group A
N = 51/51
Interior Dust
Abatement,
Loose Paint
Stabilization
N = 51/51
N = 49/51
Soil
Abatement
N = 47/49
Paint Deleading
N - 1 8/49
N = 32/49
Group B
N = 47/47
Loose paint
Stabilization
N = 47/47
N = 46/47
Soil
Abatement
N = 42/46
Paint
Deleading
N = 1 6/46
N = 26/46
Environmental media and body burden samples were collected at various times surrounding the
interventional activities. Soil samples were also collected immediately following soil abatement to
confirm its effectiveness.
During Phase I, the average blood-lead concentrations in all three experimental groups
decreased at the first (6 months) post-abatement measurement. The statistically significant decreases
were: 2.9 ng/dL for Study, 3.5 /ug/dL for Comparison A, and 2.2 pg/dL for Comparison B. The
following increases in average blood-lead concentration were recorded between the first and second
(11 months) post-abatement measurements: 0.5 /*g/dL for Study, 2.6 fig/dL for Comparison A, and
1.5 /ng/dL for Comparison B. The increases for the two comparison groups were significantly
different from zero. The mean dust-lead levels from hand-wipe samples for all groups followed a
similar pattern, though they exhibited considerably greater variability.
By the end of Phase II, 91 children were still participating and living at the same premises as
when they were enrolled. Of these children, 44 received both soil and lead-based paint abatement, 46
received only soil abatement, and 1 refused both interventions. Mean blood-lead concentrations in all
three experimental groups were taken an average of 10 months post-abatement for Phase I and an
average of 9 months post-abatement for Phase II. Although some premises underwent lead-based
paint deleading during Phase II, no results on the additional efficacy of lead-paint abatements were
reported.
For children whose premise underwent soil abatement only, mean blood-lead concentrations
decreased between pre- and post-intervention measures (Table A.9-2). Study Group results are an
Page A-22
-------
average of 10 months post-abatement, and both comparison group results are an average of 9 months
post-abatement. Two children in the Study Group and one child in Comparison Group B were
excluded as outliers from this analysis.
Table A.9-2. Blood-Lead Concentration (/tg/dL) by Experimental
Group and Sample Period*
Group
Study
Comparison A
Comparison B
Study,
Comparison A and
B combined
Sample Size
52
18
13
83
Pre-abatement
13.10
12.94
10.54
12.66
Post-abatement
10.65
7.69
9.15
9.77
Mean Decline
2.44
5.25
1.39
2.89
Study Group results are from Phase I and both Comparison
Group results are from Phase II.
A repeated measures analysis was conducted for the restricted sample (N = 31) of children
from Comparison Group A (N = 18) and Comparison Group B (N=13) who had PbB data at all three
times. Study Group data was excluded for lack of a control period. Mean blood-lead concentrations
decreased by 0.64 jitg/dL during Phase I and another 3.63 Mg/dL during Phase II (a 33.9% decline
overall). For the 31 children of the restricted sample, variation was seen in the decline of blood-lead
levels depending upon initial PbB (Table A.9-3). A trend in the magnitude of the decline in blood-
lead levels was apparent, with larger declines observed in children with larger initial blood-lead
levels.
Table A.9-3. Blood-Lead Concentration (/tg/dL) by Experimental Group and Sample Period
Initial PbB
7-9
10-14
15-22
Change in PbB for
Phase I
+ 0.30
+ 0.18
-2.50
Phase II
-1.45
-3.82
-5.60
Overall Percentage
Change
-18.1%
-31.8%
-30.3%
Although many yards had evidence of recontamination both at 6-10 and 18-22 months post-
abatement, follow-up median soil-lead concentrations were generally less than 300 ppm (Table A.9-
4). Similar results were observed for the comparison groups during Phase II. Dust-lead loadings
were less consistent. Floor dust samples from within the residence were composited to produce a
single dust measure for each residence. Dust-lead loadings declined significantly during the study
Page A-23
-------
Table A.9-4. Surface Soil-Lead Concentration (ppm) by
Experimental Group and Sample Period
Group
Study
Comparison A
Comparison B
Period
Pre-Abate.
6-12 months Post-Abate.
1 8-22 months Post-Abate.
Pre-Abate.
6-12 months Post-Abate.
1 8-22 months Post-Abate.
Pre-Abate.
6-12 months Post-Abate.
1 8-22 months Post-Abate.
Sample Size
35
35
34
31
32
N/A
26
26
N/A
Arith. Mean
2206
141
160
2358
171
N/A
2299
180
N/A
Std. Oev.
1123
299
115
1203
172
N/A
1129
127
N/A
(Table A.9-5). Comparable declines were seen in all three groups during Phase I, despite
Comparison Group B not receiving any interior house dust abatement. Mean floor dust-lead loadings
at 6-12 months post-abatement fell significantly for the Study Group (P<0.001), during Phase I, but
remained relatively unchanged for Comparison Groups A and B (P=0.95 and 0.15, respectively)
during Phase II, despite the soil abatement. At 18-22 months post-abatement, mean levels in the
Study Group rose, but were still significantly below baseline (P=0.02). No significant declines were
seen in the lead loading, lead concentration, or dust loading measures for window well samples (Table
A.9-6).
Conclusions (including caveats). These results demonstrate that a reduction of 2060 ppm in lead-
contaminated soil around homes is associated with a modest decline in blood-lead levels. The
magnitude of reduction in blood-lead level observed, however, suggests that lead-contaminated soil
abatement is not likely to be a useful clinical intervention for the majority of urban children in the
United States with low-level lead exposure. Furthermore, the decline is confounded with the efficacy
of lead-based paint stabilization and seasonal variation in blood-lead levels.
Page A-24
-------
Table A.9-5. Interior Floor Dust-Lead Loading 0*g/ft2) by
Experimental Group and Sample Period*
Group
Study
Comparison A
Comparison B
Period
Pre- Abate.
6-12 months Post-Abate.
1 8-22 months Post-Abate.
Pre-Abate.
6-12 months Post-Abate.
1 8-22 months Post-Abate.
Pre-Abate.
6-12 months Post-Abate.
1 8-22 months Post-Abate.
Sample Size
21
14
11
22
15
N/A
22
12
N/A
Geo. Mean
769
96
207
295
315
N/A
261
295
N/A
Std. Dev.
54
35
22
29
28
N/A
45
27
N/A
Table A.9-6. Interior Window Well Dust-Lead Loading Gig/ft2) by
Experimental Group and Sample Period*
Group
Study
Comparison A
Comparison B
Period
Pre-Abate.
6-12 months Post-Abate.
1 8-22 months Post-Abate.
Pre-Abate.
6-12 months Post- Abate.
1 8-22 months Post-Abate.
Pre-Abate.
6-12 months Post-Abate.
1 8-22 months Post-Abate.
Sample Size
19
15
11
22
15
N/A
21
12
N/A
Arith. Mean
11524
28773
31420
28373
15417
N/A
21487
37205
N/A
Std. Dev.
143
75
41
58
92
N/A
63
102
N/A
Lead loadings converted from /ig Pb per m2 to /xg Pb per ft2.
Page A-25
-------
A. 10 HUD Abatement Demonstration (HUD Demo) Study
Reference(s). U.S. Department of Housing and Urban Development. "The HUD Lead-Based Paint
Abatement Demonstration (FHA)." Washington, D. C. August 1991.
U.S. Department of Housing and Urban Development. "Comprehensive and Workable Plan
for the Abatement of Lead-Based Paint in Privately Owned Housing: Report to Congress."
Washington, D. C. December 1990.
Pertinent Study Objectives. The study was designed to determine and evaluate the overall usability
and effectiveness of various methods of lead-based paint abatement.
Sampled Population. These methods were tested between 1989 and 1990 in 172 FHA-foreclosed,
single family housing units in seven urban areas: Baltimore, MD; Washington, D. C.; Seattle, WA;
Tacoma, WA; Indianapolis, IN; Denver, CO; and Birmingham, AL. Three of these houses had only
pilot abatements performed, while the other 169 were completely abated.
Intervention Strategy. Six abatement procedures were employed:
1) encapsulation - coating and sealing of surfaces with durable coatings,
2) abrasive removal - removal of lead-based paint using mechanical removal equipment,
3) hand-scraping with a heat gun - removal of lead-based paint using a heat gun to loosen
the paint,
4) chemical removal - removal of lead-based paint using a chemical stripper,
5) enclosure - resurfacing or covering of the surface, and
6) removal and replacement - removing contaminated substrates and replacing with new or
deleaded components.
Following the abatements, units were cleaned using HEPA vacuums and a high phosphate
wash until HUD clearance standards were met. Debris was disposed of off-site. In practice, abrasive
removal was not feasible for most surfaces and was not used.
Measurements Taken. X-Ray fluorescence (XRF) determination of lead content in paint, wet wipe
sampling of surfaces within a defined area, and core soil samples were collected.
Study Design and Results. The specific units to be abated were selected by first identifying older
housing likely to contain lead-based paint and then testing painted surfaces for lead using portable
XRF. Units included in the study were those found to have a large number of structural components
covered by paint with a high concentration (> 1.0 /xg/cm2) of lead. An abatement strategy,
consisting of decision rules for choice of abatement method, was randomly assigned to each house.
The method used to characterize the unit abatement strategy was always the first-choice method and
was used on all components to the extent feasible. Second, third, and fourth choice methods were
specified for each strategy. Because of the diversity of housing components containing lead-based
paint, it was generally true that no single abatement method could be used uniformly throughout a
given housing unit.
Once the lead-based paint had been abated from a component and the area cleaned, clearance
wipe samples were collected to verify the abatement. The resulting dust-lead loadings were compared
to the appropriate standard in the HUD Guidelines: 200 ^g/ft2 for floors, 500 /^g/ft2 for window sills,
and 800 /ig/ft2 for window wells. Eighty percent of floor wipe clearance samples passed by
measuring below the 200 /xg/ft2 standard (Table A. 10-1). Units predominantly abated using
Page A-26
-------
replacement methods were most often measured to have floor dust-lead loading below the standard,
87.5%. The differences seen in the rates at which particular methods failed to meet the standard are
statistically significant (p< 0.001).
Table A.10-1. Distribution of Wipe Samples Gig/ft2) on Floors by Clearance Standard
on Initial Wipe Test by Predominant Unit Abatement Strategy
Wipe Value
< 200
/vg/ft2
> 200
//g/ft2
ALL
Encaps.
188
(86.2%)
30
(13.8%)
218
(100%)
Enclose
96
(80.0%)
24
(20.0%)
120
(100%)
Chemical
276
(77.3%)
81
(22.7%)
357
(100%)
Scrape w
Heat Gun
163
(71.2%)
66
(28.8%)
229
(100%)
Replacement
203
(87.5%)
29
(12.5%)
232
(100%)
Total
926
(80.1%)
230
(19.9%)
1156
(100%)
The highest failure rates among window sill wipe clearance samples were for chemical
stripping and heat gun removal units (Table A. 10-2). Overall, the window sill samples passed 84.7%
of the time. There were significant differences among the different abatement methods.
Table A.10-2. Distribution of Wipe Samples (/*g/ft2) on Window Sills by Clearance Standard
on Initial Wipe Test by Predominant Unit Abatement Strategy
Wipe Value
< 500
//g/ft2
> 500
//g/ft2
ALL
Encaps.
157
(95.2%)
8
(4.8%)
165
(100%)
Enclose
78
(91.8%)
7
(8.2%)
85
(100%)
Chemical
173
(75.9%)
55
(24.1%)
228
(100%)
Scrape w
Heat Gun
124
(75.6%)
40
(24.4%)
164
(100%)
Replacement
137
(92.6%)
11
(7.4%)
148
(100%)
Total
669
(84.7%)
121
(15.3%)
790
(100%)
Window well clearance wipe samples were more problematic than the other sample types;
only 65% were measured below 800 fig/ft2 (Table A. 10-3). Units predominantly abated using
chemical stripping and heat gun removal methods had approximately 45 % of their clearance wipes
above the standard. This is significantly different than the 21% failure rate encountered for units
predominantly abated using replacement methods.
Page A-27
-------
Table A.10-3. Distribution of Wipe Samples 0*g/ft2) on Window Wells by Clearance Standard
on Initial Wipe Test by Predominant Unit Abatement Strategy
Wipe Value
< 800
//g/ft2
> 800
//g/ft2
ALL
Encaps.
75
(74.3%)
26
(25.7%)
101
(100%)
Enclose
45
(76.3%)
14
(23.7%)
59
(100%)
Chemical
95
(54.3%)
80
(45.7%)
175
(100%)
Scrape w
Heat Gun
61
(55.5%)
49
(44.5%)
110
(100%)
Replacement
79
(79.0%)
21
(21.0%)
100
(100%)
Total
355
(65.1%)
190
(34.9%)
545
(100%)
Core soil samples were collected before and after the abatement procedures were employed.
An examination of a subset of 130 homes suggests that the abatement procedures may have elevated
lead levels in the surrounding soil (Table A. 10-4).
Table A. 10-4. Comparison of Pre- and Post-Abatement Soil
Lead Concentration (ppm) by Urban Area
Urban Area
Baltimore / Washington
Birmingham
Denver
Indianapolis
Seattle / Tacoma
Number
of Homes
17
23
38
27
25
Arithmetic
Mean Change
in PbS
179.67
61.76
54.55
122.59
227.39
Number with
Increased
PbS Levels
11
12
28
24
22
Percent with
Increased PbS
Levels
64.7
52.2
73.7
88.9
88.0
Conclusions (including caveats). Five of the six methods successfully abated the lead-based paint
hazard, but required varying degrees of effort. Abrasive sanding was not successfully implemented
because the machines kept clogging. If encapsulation and enclosure methods are found to exhibit
long-term efficacy, their low-cost and minimal waste make them ideal for most abatement tasks.
There is evidence that soil-lead levels surrounding the residences increased due to the abatement
procedures.
Page A-28
-------
A.ll Comprehensive Abatement Performance (CAP) Study
Reference. Battelle Memorial Institute and Midwest Research Institute, "Comprehensive Abatement
Performance Study." Draft Final Report to U. S. Environmental Protection Agency. January, 1994.
Buxton, B.E., Rust, S.W., Kinateder, J.G., Schwemberger, J.E., Lim, B., Constant, P., and
Dewalt, G. "Post-Abatement Performance of Encapsulation and Removal Methods for Lead-Based
Paint Abatement", Lead in Paint, Soil, and Dust. Health Risks, Exposure Studies, Control Measures,
Measurement Methods, and Quality Assurance, ASTM STD 1226, Michael E. Beard and S.D. Allen
Iske, Eds., American Society for Testing Materials, Philadelphia. 1994.
Pertinent Study Objectives. The CAP Study sought to assess the long-term effectiveness of two
lead-based paint abatement strategies: 1) encapsulation and enclosure methods, and 2) removal
methods.
Sampled Population. Fifty-two FHA foreclosed, single family residences in Denver, Colorado were
examined. Thirty-five of the residences were abated using the aforementioned methods as part of the
HUD Abatement Demonstration (HUD Demo) Study. The remaining 17 residences were control
(unabated) homes identified in the HUD Demo Study to contain little or no lead-based paint.
Intervention Strategy. Surfaces identified via XRF to contain lead-based paint were abated using
either encapsulation, enclosure, or one of four removal methods (chemical stripping, abrasive
stripping, heat-gun stripping, and complete removal or replacement of painted components).
Measurements Taken. Vacuum dust samples within a specified area (to permit both lead loading and
lead concentration calculations) and core soil samples were collected.
Study Design and Results. Because of the diversity of housing components containing lead-based
paint, it was generally true that no single abatement method could be used uniformly throughout a
given housing unit. For the CAP Study, each house was primarily classified according to the
abatement category (i.e., encapsulation/enclosure versus removal methods) accounting for the largest
square footage of interior abatement. However, at many HUD Demonstration houses, a great deal of
exterior abatement was also performed. Therefore, the data interpretation also considered which
specific methods were used on both the interior and exterior of the house.
Dust-lead levels were measured in two abated and one unabated room in each abated
residence, and two rooms in each control residence. In each room, air duct, floor, window sill, and
window well samples were obtained. Interior and exterior dust samples were also collected at the
entryway to the residence. Soil samples were taken at the foundation, entryway, and boundary of the
home.
The study results may be summarized as follows:
• Lead levels were often found to be higher in abated houses than in control houses, primarily
in sampling locations where no abatements were performed. The most significant differences
in dust-lead loadings were found for air duct vacuum samples and exterior entryway vacuum
samples. The most significant differences in lead concentrations were found for air duct
vacuum samples (Table A. 11-1).
• Soil-lead concentrations were significantly higher at abated houses than at control houses
(Table A. 11-1).
Page A-29
-------
< W <1>
_ O W
.£ £ 3
| o|
^ <8 c
"two
0 3 O
.2 o
cc
o o
O §
^•^ ^B
.£ c «
<0 "O W
c
5w .5
<0 *!
0 1° LU
1 I
e 1
I*
Of
M (0 *t
Q -J c
•a d _o>
J J 1
"8 "8%
-•-. si
1*1:
c
•a d o>
cn c? ***-
3° 5
t3 -0*^
o ^
-J -J §;
ll|
*X3 O O)
(D C "S.
* o S5
-1 o :s-
-D -Ot
ra ra 2^
°? °. O)
-J -J ^.
c
0)
0
Q.
E
o
o
O
00
<-
*
q
CN
*
*
cn
CO
*
*
CO
*
*
in
*
*
0
^
CM
0
CM
CM
CO
CO
CD
to
Q
O
Q
k_
pNn
CO
0
CO
in
°
00
00
d
00
cn
6
CO
CO
d
m
00
LO
00
0
CD
"55
o
c
CM
^~
"-
r-
LO
CM
cn
q
0
00
CM
CD
00
CO
CO
0
-o
1
:
CO
cn
0
.
|
:
CO
"3)
Q.
O
O
LL
LO
LO
«-
O
CO
*
*
CM
O
CM
*
LO
CO
r~
CO
q
*co
^
00
CO
CO
(Vacuum
V—
o
o
LL
^-
CN
«-
LO
cn
d
LO
"
cn
in
00
d
LO
q
in
0
CO
00
cn
ra
>-
•*->
LU
L_
g
0
4-»
ps»
q
*-
q
cn
q
"
*
in
cn
T—
cn
r—
CM
CM
CM
in
00
0
CN
CM
(0
L_
C
LU
g
*L_
to
X
LU
;
CO
CM
.
*
00
't
:
|
CO
CN
'6
0 <£
(0
§
^
LU
O
r-
d
.
*
00
00
i
in
00
\
w
c.
o
V-i
CO
Tl
C
o
LI-
CM
*00
CD
T—
:
CO
00
'o
CO
•a
c
D
O
QQ
I
I
2
1
•o
u
d
•s
1
•a
o
(A
O
&
H ~
- 4J -3
,> d>
8 " «
f§*
S t-H m
O +-> -4—>
tt rt ™
-------
Only for air duct samples was there a significant difference in lead concentrations between
houses abated by different methods. The most significant differences in dust-lead loadings
were found for air duct vacuum samples and floor vacuum samples. However, it should be
noted that for almost every sample type where lead concentrations were higher in abated
houses than in control houses, lead concentrations were also typically higher in houses abated
by encapsulation/enclosure methods than in houses abated by removal methods. The same
was true for lead loadings. (Table A. 11-1).
Lead levels were often lower, and sometimes significantly lower, in control rooms of abated
houses (i.e., rooms that did not require abatement) than in abated rooms of these same
houses, although the differences observed were only of marginal significance (10% level).
There were no statistically significant differences in lead concentrations, and lead loadings
were (marginally) lower in window well samples and floor vacuum samples. (Table A. 11-2)
Floor dust-lead loadings in abated houses were below the HUD interim standard of
200 Mg/ft2, as well as the EPA guidance level (July 14, 1994) of 100 /xg/ft2. Window sill
dust-lead loadings also were below the HUD interim standard of 500 /*g/ft2. However,
window well dust-load loadings in both abated and unabated houses were typically greater
than the HUD interim standard of 800 /xg/ft2.
Table A.ll-2. Ratio of Levels of Control Rooms to Those in Abated Rooms
Component
Air Duct
Window Channel
Window Stool
Floor (Vacuum)
Interior Entryway
Lead Loading
0.73
0.39*
0.67
0.56*
1.63
Lead Concentration
0.79
0.61
0.69
0.87
1.28
Dust Loading
0.91
0.65
0.96
0.65*
1.31
Significant at the 10 percent level.
Conclusions (including caveats). While, lead levels typically remained higher for abated houses than
for control homes the abatement methods appear efficacious in the long-term. The study results also
indicate that the lead levels after encapsulation/ enclosure abatement were typically, though not
significantly, higher than those with removal abatement. When interpreting these results it should be
noted that encapsulation/enclosure houses typically had larger amounts of abatement performed than
removal houses. Therefore, the differences in lead levels noted above may be largely a result of the
more severe initial conditions in encapsulation/enclosure houses, that is, the greater amount of
abatement required in encapsulation/enclosure houses. In addition, abated houses were, on average,
17 years older than control houses. Thus, differences in soil-lead levels between abated and control
houses could be due to the differences in age, the current or past presence of lead paint, or both.
Page A-31
-------
A.12 Milwaukee Retrospective Paint Abatement Study
References. Schultz, B.D. (1993). "Variation in Blood Lead Levels by Season and Age." Draft
Memorandum on data from the Milwaukee Blood Screening Program. June, 1993.
Pertinent Study Objectives. Examine the effectiveness of the lead-based paint abatement strategies
implemented in the Milwaukee area in 1989-1992.
Sampled Population. Children residing in houses whose lead-based paint hazard has been at least
partially abated.
Intervention Strategy. Damaged, painted surfaces with lead loadings exceeding 1.0 mg/cm2 were
abated. The exact method of abatement was not available. Clean-up procedures varied depending
upon the practices of the particular abatement contractor.
Measurements Taken. Pre- and post-abatement blood samples were collected. Most of the post
abatement samples were collected 3 to 12 months following abatement.
Study Design and Results. Only preliminary results are available at this time. Blood-lead
concentrations were collected from 104 children. The arithmetic mean blood-lead concentration fell
from 34 /ng/dL pre-abatement to 26 fig/dL post-abatement which represents a 24% decline.
Conclusions (including caveats). The above results seem to imply that lead-based paint abatement
does reduce blood-lead levels. However, as mentioned previously the results are preliminary at this
time.
Page A-32
-------
A.13 New York Chelation Study
References. Markowitz, M.E., Bijur, P.E., Ruff, H.A., Rosen, J.F. (1993). "Effects of Calcium
Disodium Versenate (CaNa2EDTA) Chelation in Moderate Childhood Lead Poisoning." Pediatrics.
92(2):265-271.
Ruff, H.A., Bijur, P.E., Markowitz, M.E., Ma, Y., Rosen, J.F. (1993). "Declining Blood-
Lead Levels and Cognitive Changes in Moderately Lead-Poisoned Children." Journal of the American
Medical Association. 269(13): 1641-1646.
Rosen, J.F., Markowitz, M.E., Bijur, P.E., Jenks, S.T., Wielopolski, L., Karlef-Ezra, J.A.,
and Slatkin, D.N. (1991). "Sequential Measurements of Bone Lead Content by L-X-Ray Fluorescence
in CaNa2EDTA-Treated Lead-Toxic Children." Environmental Health Perspectives. 91:57-62
Pertinent Study Objectives. The study was an effort to ascertain the efficacy of a particular
chelation therapy procedure on moderately lead-poisoned children. Effectiveness was assessed both
short-term (6-7 weeks post-treatment) and long-term (6 months post-treatment).
Sampled Population. The study examined a subset of the children, 1 to 7 years of age, referred by
their physicians to the Montefiore Medical Center Lead Clinic. A child was eligible for the study if
their blood-lead levels were between 25 /xg/dL and 55 j^g/dL, their erythrocyte protoporphyrin (EP)
levels were greater than 35 jug/dL and the child had not required chelation therapy before.
Intervention Strategy. Paint hazard abatements were initiated at the child's residence following
identification that the child was moderately lead-poisoned. The abatements were either completed
before the child was released from the hospital following chelation therapy or lead-free alternative
housing was found until the abatement had been completed. The specific details of the abatement
procedure were not cited in the reference.
Measurements Taken. Measurements of the lead-based paint loading on interior residential surfaces
were collected via XRF both at enrollment and six weeks following enrollment (lead-based paint
abatement was performed and completed within that time). LXRF tibial bone, EP, blood, and urinary
measurements were collected from each child at enrollment, 6-7 weeks post-enrollment, and 6 months
post-enrollment. In addition, an index of cognitive functioning was administered at about the same
time as the biochemical measurements.
Study Design and Results. A total of 201 children were enrolled in this study. Each enrolled child
was administered an 8-hour edetate calcium disodium (CaNa2EDTA) lead mobilization test (LMT)
and tested for iron deficiency or depletion. A positive LMT suggested chelation therapy might prove
effective in reducing their blood-lead levels, and iron deficiency was treated with iron supplements.
The New York Department of Health used visual inspection and XRF measurements to assess whether
the child's residence required lead-based paint abatement. It was determined that the residences of
89% of the children in this study required abatement.
Page A-33
-------
Table A. 13-1. Pre-Treatment Measures by Treatment Group
Measure
Blood-Lead
Level (fjg/dl)
EP Level
(X/g/dL)
Bone-Lead CNET1
HES2
Chelated Group (n = 71)
Mean
37.3
143.3
191.4
164.4
Standard
Deviation
8.1
88.0
105.4
193.8
Control Group (n= 103)
Mean
29.0
78.0
125.3
110.6
Standard
Deviation
5.6
43.2
87.1
153.3
1 Corrected net counts.
2 Home environment scale (XRF reading multiplied by 0-3 score for paint's condition, 0 = intact, 3=peeling,
summed across all surfaces assessed within the residence).
This abstract considers the reported results for three overlapping subsets of this study. These
subsets were cited in the three journal articles outlining the results of this study. The first subset is
addressed in Markowitz et al (1993), the second in Ruff et al (1993), and the third in Rosen et al
(1991).
The first set of analyses examined a subset of 174 children. Seventy-one of the children had a
positive LMT and underwent chelation therapy. The remaining 103 children underwent no chelation
therapy and were used as a control population. Twenty-two of the chelated children and 50 of the
non-chelated children were administered iron supplements. For all the children enrolled, body burden
and environmental measures were collected (Table A. 13-1).
Six to seven weeks following enrollment, mean blood-lead levels among the 103 non-chelated
children had fallen 2.5 fig/dL and mean bone-lead levels had fallen 3.3 CNET (Table A. 13-2). An
analysis of the measured changes found significant differences for blood-lead concentration, bone-lead
counts, and EP concentration between the chelated and control groups (p<0.05). The differences in
HES were not statistically significant between the two groups. Average HES did fall from 129 to 82
among the abated residences, but 79% of the residences were still identified as having a problem with
peeling paint following abatement. Recall that the control and chelated groups had significantly
different initial mean levels for blood-lead, bone-lead, and EP levels (Table A. 13-1). To assess this
disparity, the declines six weeks post-enrollment were re-examined after developing chelated and
Table A.13-2. Mean Change in Body Burden Measures at 6 Weeks
Post-Treatment, by Treatment Group
Measure
Blood-Lead (//g/dL)
Bone-Lead (CNET)
EP (yug/dL)
Chelated Group (n = 71)
-7.2
-44.9
-37.4
Control Group (n= 103)
-2.5
-3.3
-12,6
Page A-34
-------
control groups matched by initial levels. The reanalysis found the declines were not significantly
different across the two groups for any of the measured parameters.
The second subset consisted of 154 of the 201 children enrolled, 126 of which had complete
data. Sixty of the 154 children were iron deficient, 93 underwent no chelation therapy, 35 were
treated once, 19 were treated twice, and 7 were chelated three times. In addition to results at
enrollment and 6-7 weeks post-enrollment, 6 month post-enrollment measurements were cited for this
subset (Table A. 13-3). Despite finding no association between blood-lead levels and cognitive index
(CI) at enrollment, the authors did note that mean CI increased to a small but significant degree over
the six months following enrollment. More importantly, though changes in CI were not related to
changes in blood-lead level short-term, CI was stated to increase as blood-lead concentration
decreased long-term (Table A. 13-4). Furthermore, the authors suggested that CI increased
approximately 1 point for every 3 /xg/dL decrease in blood-lead level.
Table A. 13-3. Mean and Standard Deviation Measures,
by Time Point
Measure1
Blood-Lead Level (/vg/dL)
Cognitive Index
Bayley Score
(n = 56)
Stanford-Binet Score
(n = 53)
Enrollment
31.2
(6.5)
79.0
(13.0)
76.9
(14.0)
83.5
(10.2)
6 Weeks Post-
Enrollment
26.9
(6.3)
83.1
(13.6)
78.4
(14.1)
87.0
(10.3)
6 Months Post-
Enrollment
23.9
(6.5)
82.6
(13.3)
76.6
(13.3)
88.1
(11.2)
Subset of 126 children seen at all three time points.
Table A. 13-4. Mean and Standard Deviation Cognitive Index,
by Treatment Group and Time Point
Treatment Group
Non-Chelated
(n = 80)
Chelated
(n = 49)
Enrollment
78.8
(11.4)
79.2
(15.1)
6 Month Post-Enrollment
81.5
(12.3)
83.8
(14.8)
The third subset was of 59 children, 30 of which were non-chelated. In addition to
enrollment and 6-7 week post-enrollment results, 6 month post-enrollment results for blood-lead,
bone-lead, and EP levels were reported (Table A. 13-5). Some of the 29 chelated children in the
subset underwent two rounds of chelation therapy. Mean blood-lead levels among the 30 non-
Page A-35
-------
Table A.13-5. Mean Blood-Lead, Bone-Lead and EP Levels,
by Treatment Group and Sampling Period
Blood-Lead Level
(fjgldl)
At Enrollment
After 6 Weeks
After 6 Months
Control (n = 30)
29
23
21
Chelated Once1
37
26
24
Chelated
Twice1
42
32
26
Bone-Lead
(CNET)
At Enrollment
After 6 Weeks
After 6 Months
Control (n = 30)
117
120
121
Chelated Once1
211
132
125
Chelated
Twice1
217
161
115
EP
ifjgldl)
At Enrollment
After 6 Weeks
After 6 Months
Control (n = 30)
80
82
42
Chelated Once1
110
100
45
Chelated
Twice1
138
88
43
1 References did not specify the sample size of these two groups, though together they equal the 29
chelated children.
chelated children had fallen 6 /xg/dL by 6 weeks post-enrollment and an additional 2 /ig/dL by 6
months post-enrollment. Mean bone-lead levels did not change significantly among the non-chelated
children neither by 6-weeks nor 6-months post-enrollment.
Conclusions (including caveats). Since the lead-based paint abatements were not the focus of this
study, hypotheses surrounding the abatement's efficacy were not tested. The reported results,
however, did suggest the abatements were effective. There is evidence that blood-lead, bone-lead and
EP levels declined following the partial abatement of residential lead-based paint. More importantly,
the blood-lead declines were associated with increases in cognitive index among moderately lead-
poisoned children. In addition, the study's authors determined that initial bone-lead levels impacted
the extent to which bone-lead declined following intervention. It should be noted, however, that
seasonal variation may have played a role in the body burden declines cited 6 months post-enrollment.
Page A-36
-------
A.14 Milwaukee Retrospective Educational Intervention Study
Reference. Schultz, B.D. (1995) Personal Communication, April, 1995.
Pertinent Study Objectives. This study sought to assess the effectiveness of in-home education
efforts in Milwaukee in 1991-1994.
Sampled Population. The sample population consists of 431 children under 6 years of age, who
were identified by the Milwaukee Health Department as having initial blood-lead concentrations
between 20-24 /xg/dL, who had a follow-up blood-lead measure and had not moved before the follow-
up measure. Of these, 195 children received and in-home educational visit and had a follow-up
blood-lead measure after the in-home visit. The control group of 236 children did not receive a
health department in-home visit, either because they were identified before the educational outreach
program was in place, or because the family could not be contacted after several attempts.
Intervention Strategy. The in-home educational visits were conducted by para-professionals. The
visits lasted approximately one hour and educated the families on nutrition, behavior change, and
housekeeping recommendations to reduce childhood lead body burden.
Measurements Taken. Venous and capillary blood-lead measurements were taken.
Study Design and Results. Follow-up blood-lead concentrations (PbB) were collected 2-15 months
after initial PbB measures. The arithmetic mean blood-lead concentration for the Educational
Outreach group declined by 18 percent, as compared to the Control group decline of 5 percent (Table
A. 14-1). The difference between these decline is highly statistically significant. Also, blood-lead
levels decreased for 154 of the 195 (79%) children in the Educational Outreach group, compared to
124 of the 236 (53%) children in the Control group. Blood-lead concentrations were adjusted for the
effects of age and seasonal variations.
Table A. 14-1. Blood-Lead Concentrations
by Study Group
Study Group
Educational Outreach
Control
N
195
236
Mean Initial
PbB (//g/dL)
22
22
Mean Decline
in PbB
4
1
Mean %
Decline in PbB
18
5
Conclusions (including caveats). The results seem to imply that educational intervention does appear
to reduce blood-lead levels. Although children receiving visits receive significant health benefits on
average, their blood-lead concentrations were still usually above 10 /xg/dL and even above 15
Page A-37
-------
A.15 Granite City Educational Intervention Study
Reference. Kimbrough, R.D., LeVois, M., and Webb, D.R. (1994) "Management of Children with
Slightly Elevated Blood-Lead Levels." Pediatrics. 93(2): 188-191.
Statement of Dr. Renate D. Kimbrough, Subcommittee on Investigations and Oversight,
Committee on Public Works and Transportation, U. S. House of Representatives, June 9, 1992.
Kimbrough, R.D., Personal Communication, January, 1995.
Illinois Department of Public Health (IDPH) (1995). Additional information on the Granite City
Educational Intervention Study was provided by John R. Lumpkin, M.D., Director, and the Staff of
the IDPH.
Pertinent Study Objectives. The study's objectives included an effort to evaluate the efficacy of
educational interventions to reduce blood-lead concentrations (PbB) in exposed individuals.
Sampled Population. 827 volunteers, including 490 children under six years of age, were recruited
from 388 households in Granite City, Illinois. Most homes in the community were built prior to
1920 and contained lead-based paint. In addition, a secondary lead smelter was closed in 1983 and
had been declared a superfund site.
Intervention Strategy. During home visits where the entire family was present, extensive and
intensive educational efforts were aimed at the children and families exposed to elevated levels of lead
in the surrounding environment. Instruction included identifying where lead-based paint was
commonly found, how to perform house cleaning procedures, and hygienic procedures for young
children. Suggestions were also made to carefully remove peeling paint or make it inaccessible by
installing barriers.
Measurements Taken. Venous blood samples, soil samples, dust samples from within the residence,
tap water samples, and an assessment of the lead content in interior paint were collected.
Study Design and Results. Of the 490 children under age 6, 78 (16%) had PbB levels greater than 9
jug/dL. Of these 5 had PbB levels greater than 25 /*g/dL. When possible, follow-up measures were
collected at four months and twelve months after the initial sample. In the interim between the initial
and four month samples, the families of these children received extensive counseling in the prevention
of lead exposure.
These follow-up measures were available for a subset of 59 children. Two groups were
considered for analysis. The first group consisted of all 59 children, and the second group consisted
of a total of 24 who had data for all three measures. Mean blood-lead concentrations in both groups
decreased significantly (P=0.001) at the four month post-abatement measurement, and rose again by
the twelve month measure (Table A. 15-1). Despite the rise in blood-lead levels, the 12 month
averages remained significantly (P=0.001) below initial levels. For all 59 children, variation was
seen in the decline of blood-lead concentrations depending upon initial PbB (Table A. 15-2). A trend
in the magnitude of the decline in blood-lead levels was apparent, with larger declines observed in
children with larger initial blood-lead levels.
In addition, four month follow-up blood-lead concentrations were significantly lower than
initial levels for a small number of older children with elevated blood-lead levels. For 7 children aged
6 to 14 years, the arithmetic mean decrease was 5.9 /*g/dL, and for 3 children 15 years or older,
blood-lead levels decreased 7.0
Page A-38
-------
Conclusions (including caveats). There is evidence to suggest that the educational efforts lowered
blood-lead levels. These differences, however, cannot be separated from possible seasonal or age
variations. In addition, the full implications of the observed declines in blood-lead concentration are
somewhat difficult to ascertain without a control group.
Table A. 15-1. Blood-Lead Concentration
by Time Period
Timing Period
Pre-intervention
4 Months
Post-intervention
1 2 Months
Post-intervention
Measurement
N
Geo. Mean
(LSD)
Arith. Mean
(SD)
N
Geo. Mean
(LSD)
Arith. Mean
(SD)
Arith. Decline
Arith. % Decline
N
Geo. Mean
(LSD)
Arith. Mean
(SD)
Arith. Decline
Arith. % Decline
Sampling Group
Al! Children
59
13.83
(0.31701)
14.6
(5.37)
54
7.36
(0.33440)
7.76
(2.42)
6.83
44.8%
29
9.08
(0.33822)
9.58
(3.08)
5.59
32.1%
Complete
24
14.17
(0.37112)
15.25
(6.81)
24
7.72
(0.29324)
8.03
(2.20)
7.23
47.4%
24
8.73
(0.34014)
9.20
(2.94)
6.05
39.7%
Page A-39
-------
Table A.15-2. Blood-Lead Concentration (/ig/dL)
by Initial PbB
Initial PbB
<10*
10-15
15-20
20-25
>25
4 Months
Sample
Size
(N)
10
26
9
7
2
Arith.
Mean
Decline
4.11
5.10
7.62
11.69
22.45
Percent
Decline
(%)
42.6
41.7
43.5
55.0
65.9
1 2 Months
Sample
Size
(N)
4
14
6
3
2
Arith.
Mean
Decline
1.68
2.87
8.07
8.67
20.40
Percent
Decline
(%)
17.2
24.1
46.3
42.5
59.9
All children were between 9 and 10 /tg/dL.
Page A-40
-------
A.16 Milwaukee Prospective Educational Intervention Study
Reference. Schultz, B.D. (1995) Personal Communication, April, 1995.
Pertinent Study Objectives. This study sought to assess the effectiveness of in-home education
efforts in Milwaukee in 1993.
Sampled Population. Children selected were between 9 and 72 months old, had not received a in-
home visit in the previous year, and their dwelling had not undergone an abatement within the
previous year. Three separate groups of children were examined. The Educational Outreach Group
contained 54 children whose initial PbB was between 20-24 /ig/dL. For comparison, a Control Group
of 122 children was selected from those previously identified in the Milwaukee Retrospective Study,
whose follow-up blood-lead measurement was taken within 3 months of the initial measurement. The
Pre-abatement Educational Outreach Group consisted of 28 children who had an initial PbB between
25-40 jig/dL. Abatements were required for these children, but had not been implemented at the time
of the in-home visit and follow-up PbB measurement. Follow-up blood-lead measures were collected
an average of 2 months after the in-home visit for the Educational Outreach and Control Groups and
an average of 2 to 6 months for the Pre-abatement Educational Outreach Group.
Intervention Strategy. The in-home educational visit was conducted by para-professionals. The
visits lasted approximately one hour and educated the families on nutrition, behavior change, and
housekeeping recommendations to reduce childhood lead body burden. No paint abatements were
performed in these homes. Children in the 25-40 jig/dL range received an additional visit from a
Public Health Nurse, who conducted a general health assessment of the child/family and also
answered any questions about lead.
Measurements Taken. Venous and capillary blood-lead measurements were taken.
Study Design and Results. Both the Educational Outreach and Control Groups had initial mean PbB
levels of approximately 22 /xg/dL. The Educational Outreach Group mean decline was about 3 /xg/dL
more than the Control Group at the 2 month follow-up. This difference was statistically significant.
Mean blood-lead concentrations for the Pre-abatement Educational Outreach Group declined by
19.3% (from 28.9 to 22.6 /xg/dL). Both the Educational Outreach and Pre-abatement Educational
Outreach Groups had approximately the same percentage drop. However the Pre-abatement Group
had a greater absolute decline, suggesting greater declines with greater initial blood-lead
concentrations. Blood-lead levels were adjusted for the effects of age and seasonal variations.
Conclusions (including Caveats). These results seem to imply that educational intervention does
appear to reduce blood-lead levels, although blood-lead concentrations usually remained above 10
Page A-41
-------
Page A-42
-------
APPENDIX B
REVIEW OF ABATEMENT METHODS ASSOCIATED WITH
TEMPORARY INCREASES IN BLOOD-LEAD LEVELS
-------
This page intentionally left blank.
-------
APPENDIX B
REVIEW OF ABATEMENT METHODS ASSOCIATED WITH
TEMPORARY INCREASES IN BLOOD-LEAD LEVELS
In some cases residential lead-based paint abatement has resulted in a temporary increase in
blood-lead (PbB) levels in children and abatement workers. Abatement methods most associated with
this phenomenon include sanding, dry scraping, and the use of heat guns or torches to soften paint so
that it may be scraped more easily. These procedures can create large amounts of lead contaminated
dust or airborne lead, which is not easily removed from the residence.
The following paragraphs provide short descriptions of the use of sanding, dry scraping, heat
guns, and torches in lead-based paint abatements, followed by a summary of the scientific evidence on
each abatement method. In addition to the prospective and retrospective studies cited, there are many
case reports in the literature that document elevated PbB in children or abatement workers following
the removal of leaded paint using these methods (e.g., Rey-Alvarez and Menke-Hargrave, 1987;
Fischbein, et al, 1981; Feldman, 1978; Amitai, et al, 1987). Other sources document such elevated
PbB levels, but do not associate the elevated levels with specific abatement practices (e.g., Rabin, et
al, 1994; Swindell, et al, 1995).
In some cases, sample personal air exposure (PAE) data were available. These data may be
compared to the Occupational Safety and Health Administration (OSHA) permissible exposure limit
(PEL) of 50 /xg/rn3 or the OSHA action limit of 30 /xg/m3. The OSHA PEL is the current standard to
protect construction workers against chronic exposure to airborne lead. Medical monitoring of PbB
levels is required for workers who are exposed to lead levels above the action limit. Both the PEL
and action limit are based on time weighted average exposure over an eight hour work day.
B.I SANDING
B.I.I Description of Sanding
Leaded paint can be removed by manual sanding, or machine sanding with ordinary circular,
reciprocating, belt, or palm sanders. Related methods include the use of needle guns, grinders,
brushes, or abrasive blasting tools to abrasively or percussively remove leaded paint. Power tools
used for lead-based paint abatement are sometimes fitted with a high-efficiency particulate
accumulator (HEPA) dust collection system. In some instances, the surface is misted to contain dust
and avoid aggravating the lead hazard. Use of a respirator is recommended when applying any of
Page B-l
-------
these abrasive removal methods. A more powerful respirator may be required when using power
tools without HEPA dust collection systems (Labor, 1993).
B.1.2 Scientific Evidence on Sanding
Use of this method was limited in the U.S. Department of Housing and Urban Development
(HUD) Demonstration, because many surfaces were not flat, or could not endure abrasive action (e.g.
drywall). When applied, abrasive removal by machine sanding was found to generate a large amount
of potentially hazardous dust and HEPA attachments to collect this dust were found to be ineffective
in most instances (HUD, 1991).
Personal air lead exposure data, for housing abatement projects where power tools with
HEPA dust collection systems were in use, were reported for 28 samples, averaging 185 jig/m3 with
individual sample exposures ranging from .2 /xg/m3 to 1596 ^g/m3. PAEs for vacuum blasting were
reported for 4 samples, averaging 169 ^g/m3 with individual sample exposures ranging from 2 ^g/m3
to 665 /xg/m3 (Labor, 1993).
Sanding is frequently used in combination with other abatement methods. Consequently,
studies reporting results from such abatements do not always distinguish among the methods applied.
Additional results on abatements where sanding was employed, including documented increases in
PbB levels of resident children following abatement, are presented in the sections on dry scraping and
torch methods.
B.2 DRY SCRAPING
B.2.1 Description of Dry Scraping
Dry scraping is the traditional method of surface preparation for home renovation and
remodeling, whereby paint is removed by hand-scraping with a putty knife or similar tool. In terms
of lead-based paint abatement, this method is time-consuming and generates a large amount of lead
containing dust. Wet scraping, where the surface is misted to reduce dust levels, is preferred for
work on lead-based paint surfaces (HUD, 1994). However, dry scraping usually has been deemed
acceptable for small surfaces, e.g. near electrical outlets.
B.2.2 Scientific Evidence on Dry Scraping
In a retrospective study of preschool children in Boston, dry scraping and sanding were
associated with an increase of 9.1 /^g/dL in the PbB levels of 41 children during abatement, compared
Page B-2
-------
with their pre-abatement levels. Abatements were performed in accordance with state regulations at
the time, which required that lead-based paint be removed from or permanently covered on all
chewable surfaces below 4 feet and that loose or peeling paint be made intact on all other surfaces.
For these abatements, dry scraping was used to remove paint and sanding was used to feather the
edges to prevent additional deterioration and prepare the surface for repainting. As such, the effects
of dry scraping and sanding cannot be isolated. The PbB level during abatement was determined
about 2 months after the pre-abatement PbB measurement. Abatements were usually completed in 3
to 4 months (Amitai, et al, 1991).
PAEs for dry scraping were reported for 6 samples, averaging 45 ^ig/m3 with individual
sample exposures ranging from 6 /*g/m3 to 167 /xg/m3 (Labor, 1993).
B.3 HAND SCRAPING WITH HEAT GUN
B.3.1 Description of Hand Scraping with Heat Gun
This procedure entails using a heat gun to soften the paint, which may then be removed by
hand-scraping. Heating the paint can generate high levels of volatilized lead, which creates a risk
hazard to the abatement worker and makes the final cleanup process more difficult. Commercial heat
guns typically produce air temperatures of approximately 1000°F at the gun nozzle. A respirator is
strongly recommended during this procedure due to the potential release of volatilized lead and
organic compounds (HUD, 1991; NIOSH, 1992; Labor, 1993).
B.3.2 Scientific Evidence on Handscraping with Heat Gun
Despite softening the paint, this procedure had mixed success in abating lead-based paint from
floors and windows in the HUD Demonstration. Residences abated by hand-scraping with heat gun
failed the initial clearance test 28.8, 24.4, and 44.5 percent of the time for floors, window sills, and
window wells, respectively. In comparison, residences abated by replacement of building components
coated with lead-based paint failed the initial clearance test 12.5, 7.4, and 21.0 percent of the time for
floors, window sills, and window wells (HUD, 1991).
Although a 700°F temperature restriction was placed on heat guns used in the HUD
Demonstration, PAEs exceeded the OSHA action limit in 17.5 percent of 360 samples collected by
HUD contractors when a heat gun was in use. In addition, 6 of 10 PAE samples collected by
National Institute for Occupational Safety and Health (NIOSH) investigators during interior heat gun
work exceeded the OSHA PEL. The PAE samples collected by HUD contractors and NIOSH
Page B-3
-------
investigators tended to be over short time periods, however it was assumed that the sampling periods
were representative of full shift exposure (NIOSH, 1992; HUD, 1991).
A slight (though not statistically significant) increase in mean PbB was observed within one
month of abatement for 19 children living in 18 Baltimore homes abated by this method, even with
thorough (wet cleaning with high-phosphate detergent together with dry vacuuming with standard shop
vacuums) cleanup following the abatements. Post-abatement dust-lead levels in these homes showed
decreases in window sill and window well dust-lead levels, but increased levels on floors. By six
months post-abatement, dust-lead levels were similar to, or greater than, pre-abatement levels (Farfel
and Chisolm, 1990).
PAEs during heat gun use were reported for 380 samples, averaging 26 /xg/m3 per hour over
8 hours with individual sample exposures ranging from 0.4 /-tg/m3 to 916 jitg/m3 (Labor, 1993).
B.4 TORCH METHODS
B.4.1 Description of Torch Methods
Similar to the heat gun method, a propane torch can be used to soften lead containing paint,
before removing it by hand-scraping. Alternatively, the torch can be used to burn the paint, with
residue removed by sanding. As with heat guns, high levels of lead and organic compounds can be
volatilized. Indeed, since there is less control over temperature, volatilized lead levels are likely to be
greater.
B.4.2 Scientific Evidence on Torch Methods
In the Boston Retrospective study, an average increase of 35.7 jtg/dL in PbB during
abatement, compared with pre-abatement PbB, was reported for 4 children whose homes were abated
using torches to soften lead containing paint prior to scraping (Amitai, et al, 1991).
In 53 Baltimore homes, traditional methods of abatement (usually entailing open-flame
burning and sanding) resulted in 3 to 6 fold increases in lead-contaminated house dust over pre-
abatement levels, with 10 to 100 fold increases at abated sites. By 6 months post-abatement dust-lead
levels were similar to, or greater than, pre-abatement levels. The mean PbB of 27 children living in
these homes increased by 6.8 /xg/dL within one month of abatement (Farfel and Chisolm, 1990).
In the same study, a subset of 19 dwellings were monitored more frequently. Downward
trends in lead dust on floors and window sills were observed following abatement. These trends were
greatest during the first month. By 3 months post-abatement, dust-lead levels were similar to the
Page B-4
-------
levels that would be observed in these homes at 6 months post-abatement. At 6 months post-
abatement, dust-lead levels in these homes were similar to the levels observed in all study homes
(Farfel and Chisolm, 1990).
A Baltimore study of 184 children who received inpatient chelation therapy found that PbB
increased during the first 3 months after discharge and remained stable for the remainder of the 12
month study period. Discharge was keyed to abatement of the dwelling, or relocation of the family to
lead-free public housing. Study homes were usually abated by using a gas torch to soften paint so
that it could be scraped off, followed by sanding down to bare wood. Children discharged to abated
homes, or who routinely visited unabated lead containing homes or abated homes, had significantly
higher PbB than children residing exclusively in lead-free public housing (38.5 vs. 28.8 /ig/dL at 3
months). In addition, PbB increased by about 10 /xg/dL within about 3 months in a small number of
children who moved from lead-free public housing to abated homes, and decreased similarly in those
who moved from abated homes to lead-free public housing (Chisolm, et al, 1985).
B.5 DISCUSSION
The studied abatements often combined methods which can produce large amounts of lead
contaminated dust or volatilized lead with poor cleanup. Though improved cleanup may mitigate the
effects of increased dust-lead levels, it is not a complete solution. Children and workers may be
exposed during the deleading process, prior to cleanup. Also, improved cleanup was not helpful in
Baltimore (Farfel and Chisolm, 1990) or Massachusetts, where more stringent abatement regulations
were enacted in 1988 (Swindell, et al, 1994). Indeed, it may be impossible to completely remove
leaded dust from carpets, as demonstrated in Cincinnati (EPA, 1993).
The abatement methods under consideration have been reviewed by HUD and the U.S.
Department of Labor, Occupational Safety and Health Administration (OSHA). Although their focus
was on worker protection rather than reducing childhood lead exposure, the recommendations of these
agencies regarding the abatement methods under consideration may be of interest. Table 1
summarizes the recommendations in "Guidelines for the Evaluation and Control of Lead-Based Paint
Hazards in Housing" (HUD, 1994) and "Lead Exposure in Construction; Interim Final Rule" (Labor,
1993) regarding sanding, dry scraping, hand scraping with heatgun, and torch methods. Also, the
results of the U.S. Environmental Protection Agency Renovation and Remodelling Study will be
available soon. The focus of this study is on worker protection as well, however, the additional
information on personal air exposure to airborne lead during paint removal by sanding and dry
scraping which will be provided may be of interest.
Page B-5
-------
Although a lengthy discussion of alternative methods is beyond the scope of this document, it
should be noted that a variety of intervention strategies have been studied. In addition to sanding and
hand scraping with a heat gun, the HUD Demonstration tested abatement strategies of chemical
removal of lead-based paint, encapsulation, enclosure, and replacement of building components coated
with lead-based paint. Most of these alternative methods produced less hazardous waste and were
less likely to produce high levels of airborne lead. In addition, abated residences were more likely to
pass initial clearance tests (HUD, 1991). In Boston, significant declines in PbB levels during
abatement, compared with pre-abatement PbB, were observed in 12 children whose homes were
abated by replacing or permanently covering painted surfaces (Amitai, et al. 1991). Alternative
intervention strategies including in-place management, dust control, and educational interventions have
been used effectively, as well as these alternative abatement strategies. Many of these methods have
been summarized and their effectiveness reviewed (Battelle, 1994a; Battelle, 1994b; Burgoon, et al,
1993).
Table B-l. Summary of HUD and OSHA Recommendations Regarding Sanding, Dry Scraping,
Hand Scraping with Heat Gun, and Torch Abatement Methods.
Abatement Method
Sanding
-Manual
-Powerb, with HEPA Dust Collection
-Power6, no HEPA Dust Collection
Dry Scraping
Hand Scraping with Heat Gun
- Below 1100°F
-Above 1100°F
Torch Methods
HUD Guidelines
Not Mentioned
Recommended
Banned
Not Recommendedd
Recommended
Banned
Banned
OSHA Rule8
Halfmask APR
Halfmask APR
Powered0 APR
Halfmask APR
Hnlfmask APR
Halfmask APR
Not Mentioned
Respirators are required unless air sampling at the specific job site indicates lower lead levels than usual.
Halfmask air purifying respirators (APR) should have a protection factor of 10. Powered APRs should
have a protection factor of at least 25, with higher protection factors required for some tasks.
This includes power grinders, brushes, needle guns, sanders, and abrasive blasting devices.
Higher protection factor required for abrasive blasting devices than for other power tools.
Dry scraping is not recommended by HUD, nor is it banned. Wet scraping is preferred over dry scraping,
however, when wet scraping is not safe or practical, e.g. near electrical outlets, dry scraping is permissible.
Page B-6
-------
U.S. Environmental Protection Agency
J^'O" 5» Library (Pt-l2JJ
77 West Jackson Botrtevard. 12tti Floor
Chicago, II 60604-3590
-------
------- |