United States Office of Research and EPA/600/AP-93/001 b
Environmental Protection Development July 1993
Agency Washington, DC 20460
&EPA Urban Soil Lead Review
Abatement Draft
Demonstration (Do Not
Project Cite or
Volume II: Part 1
Boston Report
Quote)
Notice
This document is a preliminary draft. It has not been formally
released by EPA and should not at this stage be construed to
represent Agency policy. It is being circulated for comment on its
technical accuracy and policy implications.
-------
(Do not Cite or Quote)
EPA/600/AP-93/001 b
July 1993
Urban Soil Lead Abatement
Demonstration Project
Volume II. Part 1
Boston Report
NOTICE
This document is a preliminary draft. It has not been formally
released by EPA and should not at this stage be construed to
represent Agency policy. It is being circulated for comment
on its technical accuracy and policy implications.
Environmental Criteria and Assessment Office
Office of Health and Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711 #
. Printed on Recycled Paper ~
-------
Table of Contents
Page
1 EXECUTIVE SUMMARY 1-1
1.1 LEAD POISONING AND LEAD CONTAMINATED SOIL
IN BOSTON 1-2
1.2 IDENTIFICATION AND ENROLLMENT OF
STUDY POPULATION 1-3
1.3 ENVIRONMENTAL MEASUREMENTS 1-4
1.3.1 Soil 1-4
1.3.2 Dust 1-4
1.3.3 Water 1-5
1.3.4 Paint 1-5
1.4 CHILD AND FAMILY MEASURES 1-5
1.4.1 Social and Behavioral Questionnaire 1-5
1.5 OUTCOME MEASURES 1-5
1.5.1 Blood Samples 1-5
1.5.2 Hand Lead Determinations 1-6
1.5.3 Environmental Interventions 1-6
1.6 ANALYSIS 1-7
1.7 RESULTS 1-8
1.7.1 Blood Lead Levels 1-9
1.7.1.1 Crude Analyses 1-9
1.7.2 Adjusted Analyses 1-10
1.7.3 Handwipe Lead Levels 1-11
1.8 CONCLUSION 1-12
1.9 IMPLICATIONS 1-14
2. BACKGROUND 2-1
2.1 LEAD POISONING IN BOSTON 2-2
2.2 CONTAMINATED SOIL IN BOSTON 2-3
2.3 IMPLEMENTATION 2-4
3. STUDY ADMINISTRATION 3-1
4. HUMAN STUDIES REVIEW 4-1
5. STUDY DESIGN 5-1
5.1 PURPOSE 5-1
5.2 IDENTIFICATION OF STUDY POPULATION 5-1
5.3 ELIGIBILITY CRITERIA 5-3
5.4 RATIONALE FOR ELIGIBILITY CRITERIA 5-4
5.5 INTERVENTION 5-7
ii
-------
Table of Contents (continued)
Page
5.6 STUDY SIZE 5-7
5.7 CHANGES IN STUDY DESIGN AND SAMPLE SIZE 5-10
5.8 ATTRITION AND FOLLOW-UP OF STUDY POPULATION .. 5-11
6. PARENT EDUCATION AND COMMUNITY RELATIONS
STRATEGIES 6-1
7. ENVIRONMENTAL MEASUREMENTS AND ANALYSIS 7-1
7.1 SOIL 7-1
7.2 PRELIMINARY SOIL SAMPLING TO DETERMINE
ELIGIBILITY 7-1
7.3 DETAILED SOIL SAMPLING 7-2
7.4 RECONTAMINATION ASSESSMENT SOIL SAMPLING 7-2
7.5 DUST 7-3
7.6 WATER 7-4
7.7 PAINT 7-4
7.8 QUALITY ASSURANCE FOR SOIL AND DUST
SAMPLING AND ANALYSIS 7-5
8. SOCIAL AND BEHAVIORAL QUESTIONNAIRE 8-1
9. BIOLOGICAL SAMPLING AND MEASURES 9-1
9.1 BLOOD SAMPLING 9-1
9.2 BLOOD SAMPLE ANALYTIC PROCEDURES 9-2
9.3 HAND LEAD DETERMINATIONS 9-2
9.4 QUALITY ASSURANCE AND CONTROL FOR
BLOOD LEAD MEASUREMENTS 9-3
9.5 REPORTING AND EVALUATION OF CLINICAL DATA .... 9-5
10. DETAILED DESCRIPTION OF THE INTERVENTIONS 10-1
10.1 LOOSE PAINT ABATEMENT 10-1
10.2 INTERIOR DUST ABATEMENT 10-1
10.3 SOIL ABATEMENT 10-2
10.3.1 Subsurface Fabric/Synthetic Barrier 10-4
10.3.2 Surface Covers 10-4
10.3.3 Soil Abatement Procedures 10-5
10.3.4 Soil Abatement Safety 10-8
10.3.5 Soil Disposal 10-9
10.3.6 Obstacles to Soil Abatement 10-11
iii
-------
Table of Contents (continued)
Page
11. INTERIOR AND EXTERIOR PAINT DELEADING 11-1
11.1 PRE-DELEADING PLANNING 11-1
11.2 DEVELOPMENT OF INSPECTION PROCEDURES 11-3
11.3 DELEADING ACTTVinES 11-4
11.3.1 Exterior Deleading 11-6
11.3.2 Interior Deleading 11-7
11.3.3 Temporary Housing 11-9
11.3.4 Damage Control 11-10
11.3.5 Clearance Sampling 11-11
12. SCHEDULE OF ACnVTITES . 12-1
13. DATA COLLECTION AND MANAGEMENT 13-1
14. DATA ANALYSIS 14-1
15. RESULTS 15-1
15.1 BLOOD LEAD LEVELS 15-1
15.1.1 Crude Analysis 15-1
15.2 CHARACTERISTICS OF FINAL STUDY POPULATION 15-5
15.2.1 Adjusted Analyses 15-10
15.3 HAND LEAD LEVELS 15-19
15.4 ENVIRONMENTAL LEAD LEVELS 15-21
15.4.1 Soil 15-22
15.4.2 Dust 15-32
15.4.3 Water 15-34
15.4.4 Paint 15-35
15.5 COST OF ABATEMENT ACTIVrnES 15-37
15.5.1 Soil Abatement 15-37
15.5.2 Interior Loose Paint and Dust Abatement 15-40
15.5.3 Interior Dust Abatement Costs 15-42
15.5.4 Deleading Costs 15-43
15.5.4.1 Total Deleading Costs 15-46
16. DISCUSSION 16-1
16.1 STUDY PROBLEMS AND THEIR RESOLUTION 16-2
16.1.1 Recruitment and Retention of Study Participants 16-2
16.1.2 Lead Contaminated Soil Disposal 16-3
16.1.3 Limited Funding 16-4
iv
-------
Table of Contents (continued)
Page
16.1.4 Concerns About Ethical, Legal, and Logistical
Constraints 16-4
16.1.5 Frozen Ground During Soil Abatement of the Study
Group 16-4
16.2 LIMITATIONS 16-5
16.2.1 Relatively Small Sample Size 16-5
16.2.2 Follow-up Limited To One Year 16-5
16.2.3 Mobility Of Families 16-6
16.2.4 Limitations Resulting From Study Design 16-6
16.2.5 Limitations To Generalizability 16-7
16.2.6 Misclassification 16-8
16.3 IMPLICATIONS OF FINDINGS 16-8
16.4 ONE YEAR EXTENSION 16-9
v
-------
Tables
Number Page
5-1 Percent of Children Judged Ineligible According to Reason for
Ineligibility 5-8
5-2 Follow-up Statistics by Participant Group 5-11
5-3 Interior and Exterior Paint Deleading Activities 5-13
12-1 Timetable of Activities 12-2
12 -2 Starting and Ending Dates for Interventions and Sampling 12-4
15-1 Crude Changes* in Blood Lead Levels Among All Participants ... 15-2
15-2 Blood Lead (/xg/dL) Distribution Over Time and According
to Group Excluding Children Who Became Lead Poisoned 15-4
* **
15-3 Crude Changes in Blood Lead Levels Excluding Children
Who Became Lead Poisoned 15-5
15-4 Adjusted* Differences in Blood Lead Levels Stratified
J **
by Pre-Abatement Blood Levels 15-8
15-5 Distribution of Children, Families, Units, and Premises
According to Group in the Final Study Population 15-8
15-6 Characteristics of Final Study Population 15-9
15-7 Distribution (%) of Calendar Months and Mean Interval Between
Blood Samples 15-11
15-8 Mean Blood Lead Level According to Calendar Month
of Sampling 15-12
15-9 Adjusted Analysis: Description of Variables Added to the
Base Model 15-13
15-10 Crude and Adjusted Changes in Blood Lead Levels 15-14
15-11 Crude and Adjusted POST2 Blood Lead Levels Among Children
in the Study Group According to Race 15-17
vi
-------
Tables (continued)
Number Page
15-12 Crude and Adjusted POST2 Blood Lead Levels Among Study
Group Participants According to Initial Soil Lead Level 15-17
15-13 Crude and Adjusted* POST2 Blood Lead Levels Among Study
Group Participants According to the Size of the Excavated
Yard Area 15-18
15-14 Handwipe Lead (/xg/pair of hands) Distributions Over Time
Adjusting for Maximum Field Blank Lead Level and
Excluding Children Who Became Lead Poisoned 15-20
15-15 Handwipe Lead (^g/pair of hands) Distributions Over Time
Adjusting for Median Field Blank Lead Level and
Excluding Children Who Became Lead Poisoned 15-21
15-16 Crude Changes* in Hand Lead Levels (/xg/pair oHiands)
Excluding Children Who Became Lead Poisoned 15-22
15-17 Crude Changes in Hand Levels (jig/pair of hands) Excluding
Children Who Became Lead Poisoned 15-23
15-18 Crude and Adjusted Changes in Hand Lead Levels (p.g/pair
of hands) Excluding Children Who Became Lead Poisoned .... 15-24
*
15-19 Distribution of Surface Soil Lead Concentrations Over Time
and According to Group 15-25
15-20 Distribution of Interior Floor Dust Lead Concentrations Over
Time and According to Group 15-26
* 2
15-21 Distribution of Interior Floor Dust Loading (mg/m ) Over Time
and According to Group 15-27
* 2
15-22 Distribution of Interior Floor Dust Lead Loading (jig/m ) Over
Time and According to Group 15-28
15-23 Distribution of Interior Window Well Dust Lead Concentrations
Over Time and According to Group 15-29
* 0
15-24 Distribution of Interior Window Well Loading (mg/m ) Over
Time and According to Group 15-30
vii
-------
Tables (continued)
Number Page
15-25 Distribution of Interior Window Well Lead Loading (/xg/m2)
Over Time and According to Group 15-31
15-26 QA/QC Results for Soil and Dust Analyses 15-32
*
15-27 Distribution of Water Lead Concentrations (/xg/L) According
to Group 15-35
15-28 Distribution (%) of Wall and Woodwork Paint Lead
Concentrations (mg/cm2) According to Group 15-36
15-29 Distribution (%) of Amount Interior Chipping Paint at Baseline
According to Group 15-36
15-30 1989 Soil Abatement Costs 15-38
15-31 1990 Soil Abatement Costs 15-39
16-1 Percentage of Children Expected to Have Blood Lead Levels
Exceeding 10, 15, arid 20 jtg/dL Assuming Various Mean
Blood Lead Levels 16-9
viii
-------
Number
Figures
Page
5-1 Study Designs 5-2
5-2 Eligibility Assessment and Recruitment Flow Chart 5-6
15-1 Relationship Between Pre And POST2 Blood Lead Levels 15-3
15-2 Crude Change in Blood Lead Levels Excluding Children Who
Became Lead Poisoned 15-6
15-3 Plots of PRE and POST2 Blood Lead Levels According to Group
Excluding Children Who Became Lead Poisoned 15-7
ix
-------
ACKNOWLEDGMENTS
This project was successfully accomplished only through the hard work and dedication of many
people working together towards a common goalto take an important step in reducing the
insidious impact of lead contamination on children. The project could not have been completed
without their focused drive. Listed below are the people from Region I, external to the Lead
Free Kids staff, who had major roles in bringing this about. Thanks to them and to the many
others, including the many unnamed contributors in Region I, and those from EPA
Headquarters, Research Triangle Park, EMSL, and CDC, who contributed to this difficult
endeavor.
EPA Region I
Michael Deland, Regional Administrator, 1987-1989
Julie Belata, Regional Administrator, 1990-1992
Paul Keough, Deputy Regional Administrator
Pat Meaney, Director, Planning and Management Division
Ed Conley, Director, Environmental Services Division
Tom Spittler, Ph.D., Chief, Technical Support Branch
David Mclntyne, Project Manager, 1987-1992
Mark Mahoney, Assistant Project Manager, 1988-1989
Beverly A. Fletcher, Assistant Project Manager, 1990-1992
Trustees of Health and Hospital of the City of Boston, Inc.
John Cristian, Vice President/General Manager
Stuart Goldstein, Program Development Manager
William Dunsford, Purchasing Manager
City of Boston
Rob Bauman, Office of the Mayor
Massachusetts Department of Environmental Protection
Iris Davis, Environmental Engineer, Division of Hazardous Waste
x
-------
Conservation Law Foundation
Stephanie Pollack, Esq., Lead Poisoning Project Director
Massachusetts Department of Public Health
Brad Prenney, Director, Childhood Lead Poisoning Prevention Program (CLPPP)
Roy Petre, Senior Planner, CLPPP
DISCLAIMER
Although the information in this document has been funded wholly or in part by the United
States Environmental Protection Agency under Assistance Agreement #X001822-01-7 to Trustees
of Health and Hospitals of the City of Boston, Inc., it may not necessarily reflect the views of
the Agency and no official endorsement should be inferred.
xi
-------
1. EXECUTIVE SUMMARY
Perceptions of the child lead poisoning problem have steadily changed as evidence has
accumulated demonstrating subtle but serious consequences of lead exposure levels previously
believed to be innocuous. Whereas concern 25 years ago was directed at symptomatic
children with blood lead levels of 60 /xg/dL and above, the Centers for Disease Control
recently redefined lead poisoning as a blood lead level greater than 10 fig/dL. It is estimated
that in 1984 17% of all children in the United States aged six months to five years had blood
lead levels of 15 /xg/dL or greater and that in many cities as many as 35-50% have blood
lead levels in excess of 10 /xg/dL. There is currently no lead level believed to be safe for
children.
Children have multiple potential sources of lead exposure. The most important
recognized sources include lead contaminated paint, dust, and water. Paint used on both
interior and exterior surfaces of houses through the 1950's and continuing, to some extent
through much of the 1970's, often contained high concentrations of lead. Dust is now
recognized as a major vector by which children are exposed to lead via normal hand-to-
mouth activities. Lead in house dust derives, in part, from deteriorating lead based paint
within the house, and in part from lead contaminated soil and dust from areas outside the
home. Children may also ingest lead from pottery, canned foods, and numerous other
sources, although these are generally viewed as minor sources of exposure for most children.
Concern has been raised recently that lead contaminated soil in older urban areas is
another important vector for children's exposure to lead. The sources of soil contamination
include deteriorated exterior paint, past deposition of airborne lead from gasoline, and point
sources such as smelters, incinerators, and other industrial activities. At present lead
contaminated soil is neither regularly removed as part of a comprehensive strategy to prevent
childhood lead poisoning, nor removed as part of environmental interventions on behalf of
children who have already suffered excessive exposure. In part, this is due to the lack of
data demonstrating the effectiveness of lead contaminated soil abatement.
There is general agreement that children's exposure to lead should be reduced as much
as possible and -that there is an urgent need to develop practical means for the prevention and
1-1
-------
treatment of low level lead exposure. In 1986 the reauthorization of the Superfund toxic
waste cleanup program (SARA) included a provision providing funds for projects to evaluate
the impact of residential lead contaminated soil abatement on children's blood lead levels.
Boston was chosen to implement one of these projects; Baltimore and Cincinnati are the sites
of the other two projects.
This report describes the randomized environmental intervention study conducted in
Boston to determine the effect of removing lead contaminated soil on children's blood lead
levels. The study was designed to test the following hypothesis:
A reduction of 1,000 PPM or more of lead in soil accessible to children will
result in a mean decrease of at least 3 /xg/dL in the blood lead levels of urban
preschool children living in areas with high soil lead levels, multiple potential
sources of lead exposure, and a high incidence of lead poisoning.
The report also describes the range of costs associated with lead contaminated soil,
dust, and paint abatement and practical issues that arose during these abatement activities.
The study was conducted by investigators from the Boston University Schools of
Medicine and Public Health, the Harvard School of Medicine, and the Boston Department of
Health and Hospitals with full approval of the Human Studies Committee of the Trustees of
Health and Hospitals of the City of Boston and in conjunction with the United States EPA
(Region I, and Research Triangle Park which is coordinating the Three-City Study).
1.1 LEAD POISONING AND LEAD CONTAMINATED SOIL IN
BOSTON
As in many cities in the United States, childhood lead poisoning is a common problem
in Boston. It has been estimated that approximately 24% of Boston children 6 months to
5 years of age have blood lead levels greater than 15 **g/dL and 69% have blood lead levels
greater than 10 /xg/dL. While occurring throughout most of the City, most of the lead
poisoning cases are concentrated within veiy limited geographic areas. Thirty percent of all
cases in the City between October 1979 and February 1985 occurred among the 4% of
preschool children who resided in 28 areas encompassing two-three city blocks. In these
areas more than one of every four children was poisoned during this period. Whereas the
1-2
-------
average surface soil lead concentration in Boston is approximately 600 PPM, the surface soil
lead level in these areas averaged more than 3,000 PPM in tests done before the start of our
study.
1.2 IDENTIFICATION AND ENROLLMENT OF STUDY
POPULATION
The study population was drawn from children living in and around the areas described
above who were under four years of age on August 1, 1989 and had finger stick blood lead
levels of 10-20 jig/dL determined as part of the screening efforts of the Boston Childhood
Lead Poisoning Prevention Program between January and June 1989. Additional children up
to four years of age who lived on the same premises as these children were also identified
for enrollment. Homes of potential participants were visited by study staff to determine if
they met the following additional eligibility criteria: the cumulative amount of chipping or
peeling paint did not exceed 30% of the total surface area on the exterior walls of the child's
home or exceed 40% on the walls of abutting premises (these percentages were determined
by visual inspection); premises had a yard of at least ten square feet composed of dirt and/or
grass that was accessible to the child; the mean or median surface soil lead level was
1,500 PPM or greater; the child resided in a dwelling with eight or fewer residential units,
was mobile, and had never been lead poisoned; and the family resided on premises for at
least three months and had no plans to move within the three months of enrollment.
All children meeting these criteria had venous blood lead determinations beginning in
August, 1989 and those with lead levels between 7 and 24 /xg/dL were enrolled. The
baseline venous lead levels were obtained prior to any environmental abatement activities.
Children with blood lead levels above 24 jig/dL were excluded because they met the former
definition of lead poisoning and were likely to undergo medical and environmental
interventions that could obscure any changes associated with the study interventions. All
these children were referred to the Boston Childhood Lead Poisoning Prevention Program
and followed according to Massachusetts state law and lead program case management
protocols.
1-3
-------
Informed consent for participation was obtained both from the parents and landlord.
One hundred and fifty two children were enrolled in the study. Participants were randomly
assigned to one of three groups: 54 in the Study Group, 51 in Control Group A, and 47 in
Control Group B. The Study Group received loose interior paint removal, interior dust
abatement, and soil abatement. Control Group A received loose interior paint removal and
interior dust abatement. Control Group B received only loose interior paint removal.
Several study groups were employed to enable separation of the effects of soil and interior
dust abatement. Children who moved during the study were traced and whenever possible,
interviews and blood, handwipe, and environmental samples were obtained at both the new
and original residence according to the study schedule.
1.3 ENVIRONMENTAL MEASUREMENTS
1.3.1 SoU
Soil sampling was conducted to determine eligibility of properties, characterize the
potential exposure of participants to lead from the soil, document lead levels after abatement
and monitor the rate of recontamination after abatement. A detailed protocol for soil
sampling and analysis was developed in conjunction with the EPA. After enrollment,
approximately eight composite surface and eight core samples at a depth of 15 cm were taken
at each property. Post abatement and recontamination assessment samples were taken at
every other previously sampled location. Soil samples were analyzed by x-ray fluorescence
by the EPA Region I Laboratory.
1.3.2 Dust
Household dust sampling was conducted to characterize the potential exposure of
children to lead from dust, to document the reduction in dust lead levels following
abatement, and to monitor rates of recontamination after abatement. Dust on upfacing
surfaces believed most accessible to the child was sampled. Six-seven samples in each
household were obtained from the following locations: entry floor, and the window wells
and floors from the kitchen, living room, and child's bedroom. Both the lead concentration
in the dust and the amount of dust per unit area (loading) were determined. A detailed
1-4
-------
protocol for dust sampling and quality assurance plan for the sampling and analysis of soil
and dust was developed by Region I of the EPA.
1.3.3 Water
Two water samples were taken during the course of the study. Each was a first flush
sample taken by the parent from the cold water faucet in the kitchen. Water samples were
analyzed by a private laboratory. Water lead sampling and analysis was conducted
according to the standard EPA protocol.
1.3.4 Paint
In the second year of the study (1990) portable x-ray fluorescence analyzers
(PGT XK-3) were used to identify lead in paint. Measurements were taken in the child's
bedroom, kitchen, and living room. One measurement was taken on the lower part of the
wall and one was taken on the window sill in each room according to a detailed protocol for
lead paint inspection.
1.4 CHILD AND FAMILY MEASURES
1.4.1 Social and Behavioral Questionnaire
Questionnaires were administered to parents to ascertain family demographic
characteristics and possible sources of lead exposure, to obtain information about
renovations, and to characterize children's exposure to lead in soil. Follow-up interviews
were conducted toward the end of the study to assess changes in child behavior, house
cleaning and new renovations.
1.5 OUTCOME MEASURES
1.5.1 Blood Samples
Venous samples were obtained to determine blood lead levels on three occasions: the
first was taken prior to any abatement activities, the second an average of six months after
1-5
-------
abatement activities, and the third an average of 11 months post-abatement. Serum ferritin
levels were obtained at baseline. Blood lead levels were determined using graphite furnace
atomic absorption and FEP levels were determined using a zinc protopoiphyrin
hematofluorometer. The detection limit was 1 /ig/dL for blood lead, and a total method
coefficient of variation was 13.8% at the 10 /ig/dL blood lead level. The laboratory
maintained a strict internal quality control system for the blood lead analyses. In addition,
the laboratory participated in the external quality control system developed and overseen by
the Centers for Disease Control.
1.5.2 Hand Lead Determinations
Handwipe samples were obtained each time blood samples were drawn. Parents were
asked not to wash the child's hands for the two hours immediately preceding sampling.
Wearing disposable gloves, a study staff member wiped all surfaces of each hand, front and
back up to the wrist, with three commercial wetwipes. To assess the extent of any
contamination during sampling, field blanks consisting of six additional wipes were handled
so as to simulate wiping the child's hands, and set aside to determine the background
wetwipe lead levels. Field blanks were taken for every tenth child. Each set of six
wetwipes was composited for chemical analysis and extraction of the lead utilized IN hot
HN03. The total quantity of lead was reported in fig per pair of hands.
1.5.3 Environmental Interventions
The purpose of the soil abatement was to remove lead contaminated soil accessible to
the children living on the premises. A six inch layer of topsoil was removed and replaced
with 8 inches of clean topsoil. A water permeable geotextile fabric barrier was laid directly
on top of the exposed subsurface immediately following removal of topsoil and prior to
placement of clean topsoil, so as to protect against recontamination by the subsurface soil.
The lead content of the surface soil was tested and then covered with sod, grass seed, bark,
or mulch. The abated lots ranged from 12 to 702 square meters, and 3-182 cubic yards of
soil were removed per lot. Soil disposal was accomplished in accordance with guidelines
developed in conjunction with the Massachusetts Department of Environmental Protection.
1-6
-------
Lead contaminated soil was removed to a location with limited access - a quarry abutting a
cemetery in Boston.
The purpose of loose paint abatement was to minimize lead based paint as a potential
source of children's exposure during the study period by removing loose chipping paint from
the inside of the home. Loose paint abatement consisted of vacuuming the loose paint areas
with HEPA (High Efficiency Particulate Aerosol Filter) vacuums, washing loose paint areas
with a trisodium phosphate and water solution, and painting the window wells with primer.
The purpose of the interior dust abatement was to significantly reduce the amount of
lead bearing dust in the treated homes. It consisted of HEPA vacuuming and wiping surfaces
with a wet cloth, or for furniture, with an oil treated rag. Floors, including carpeted areas,
woodwork, walls, and furniture surfaces were cleaned.
1.6 ANALYSIS
First, crude analyses were conducted of the change from baseline blood lead levels to
the first and second post-abatement blood lead levels. Analysis of variance was used to
compare mean blood lead changes among the intervention groups and paired t-tests were used
to determine whether the mean changes in blood lead levels within an intervention group
were significantly different from zero.
Analysis of covariance was used to compare the intervention groups with respect to
post-abatement blood lead levels adjusted for pre-abatement levels. The post-abatement
blood lead levels were reasonably normally distributed and did not require any
transformations. The base model that was used to obtain estimates of adjusted post-
abatement blood lead means in the intervention groups was:
Yi = bo + bjZn + + l^Xi + et
where for the ith child,
Yi = post-abatement blood lead level
Zjj = 1 if in Control Group A, otherwise 0
Zji = 1 if in Control Group B, otherwise 0
Xj = pre-abatement blood lead
ej = error term
1-7
-------
The coefficients, b0, b(, b2, and b3 were estimated using least squares methods, and
t-tests were used to test the null hypothesis that and b2 were equal to zero.
Potential confounders of the relationship between group assignment and post-abatement
blood lead were added to the base model one at a time to obtain adjusted estimates of the
group effect adjusted for baseline blood lead level and the potential confounder. More
complex models that controlled for several variables simultaneously were also developed.
Potential confounders included age, sex, race, socioeconomic status as measured by the
Hollingshead Index, mouthing behaviors, and environmental sources of lead (e.g., paint and
water). In most instances, the variables were categorized; cutoffs were based on the
frequency distribution of the particular variable or on external considerations.
1.7 RESULTS
Only three of the 152 (2%) children enrolled dropped out before completion of the
study. Another 22 (14.5%) moved from their original premises but were followed. Baseline
characteristics of children in the three groups were similar in most respects. The average
age of children was similar across groups, as was the proportion of subjects in the lowest
socioeconomic level according to the Hollingshead Index (Classes 4-5). However, the mean
pre-abatement blood lead level was higher among children assigned to the Study Group. The
proportion of Hispanics was higher in the Study Group than the Control Groups and the
proportion of Blacks was lower. There was also a larger proportion of males in the Study
Group. Median surface soil lead levels were, on average, about 800 PPM higher than those
taken at a depth of 15 centimeters.
Median interior floor dust lead levels were similar to the median surface soil levels and
median window well dust lead levels were five to seven times higher. The soil and floor
dust lead levels were similar across the intervention groups. Window well dust lead levels
were more variable across the groups but the differences were not statistically stable.
Median first flush tap water lead levels were all above 14 //g/dL and were similar across
groups. Lead-based paint was detected in almost all participants' homes and XRF readings
on the walls and woodwork were similar among the groups.
1-8
-------
1.7.1 Blood Lead Levels
1.7.1.1 Crude Analyses
Mean blood lead levels in all the three groups declined at the first post-abatement
sampling round (POST1) and rose at the second post-abatement sampling round (POST2)
although for no group did the mean return to the baseline. At POST1 the average blood lead
decline was 2.87 /xg/dL in the Study Group, 3.52 /xg/dL in Control Group A, and
2.04 /xg/dL in Control Group B. All declines were significantly different from zero.
Between POST1 and POST2 the average blood lead level increased 1.39 ng/dL in the Study
Group, 2.69 /xg/dL in Control Group A and 1.52 /xg/dL in Control Group B. The increases
in the two Control Groups were significantly different from zero but the increase in the Study
Group was not (p.08).
Two siblings in the Study Group became lead poisoned sometime between the POST1
and POST2 sampling rounds. Their blood lead levels were 19 /xg/dL and 12 /xg/dL at
baseline (PRE) (September 1989), 10 /xg/dL and 17 /xg/dL, respectively, at POST1 (March
1990) and 35 /xg/dL and 43 pg/dL, respectively, at POST2 (July 1990). No other children
experienced a blood lead rise of this magnitude during the study. In fact, these two
children's POST2 blood lead levels were more than three standard deviations higher than the
overall mean POST2 level. The increases were believed to be unrelated to the study
interventions since the elevated levels were detected many months after the abatement
activities and the siblings were exposed to leaded paint at another site that was being
renovated. Therefore, these two children were excluded from subsequent analyses. Without
these children, the mean blood lead level in the Study Group increased by only 0.46 /xg/dL
between POST1 and POST2.
Because the PRE and POST2 sampling rounds are most closely matched on season,
subsequent analyses focused on this comparison. The mean decline in blood lead was
2.44 /xg/dL in the Study Group (p=0.001), 0.91 /xg/dL in Control Group A (p=0.04) and
0.52 /xg/dL in Control Group B (p=0.31). The mean blood lead level of the Study Group
declined 1.53 /xg/dL more than that of Control Group A (95% Confidence Interval: - 2.87,
- 0.19) and 1.92 pg/dL more than that of Control Group B (95 % Confidence Interval:
- 3.28, - 0.56). The magnitude of the decline in blood lead associated with soil abatement
was independent of a child's baseline blood lead level.
1-9
-------
1.7.2 Adjusted Analyses
Potential confounding variables were added to the base model one at a time to obtain
adjusted estimates of the group effect. The POST2 blood lead levels adjusted for baseline
level were generally similar to crude levels. The adjusted mean difference between the Study
and Control Groups were slightly diminished but remained statistically significant. The
differences between the Study Group and Control Groups A and B were - 1.28 (p=.02) and
- 1.49 (p=.01), respectively. Group assignment was a significant predictor of POST2 blood
lead levels (p=0.02).
The results were also similar when the analysis included only children who lived on the
study premises for at least 300 days after the pre-abatement blood lead test thereby
eliminating children who moved during the follow-up period. Here, the differences between
the Study Group and Control Groups A and B were - 1.42 (p=.02) and - 1.49 (p=.02),
respectively.
The results were also quite similar when age, sex, socioeconomic status, ferritin levels,
mouthing and handwashing behaviors, spending time away from home, spending time outside
the study area, playing in the yard, eating food outdoors, sitting on the floor inside the
home, eating canned foods including those imported from foreign countries, lead related jobs
and hobbies and cigarette smoking among household residents, living in owner occupied
premises, the presence of chipping paint, the presence of pets that go outdoors, and tap water
lead levels were added to the base model one at a time. When the paint lead variables were
added, differences between the Study and Control Groups were somewhat diminished
(-1.19 and -1.34 /xg/dL for Control Groups A and B, respectively) and the group effect was
borderline significant (p=0.06). When race was added to the base model, differences were
also diminished (- 0.92 and -1.26 jtg/dL) and the group effect was not statistically
significant (p=0.09). However, no statistically significant differences in crude or adjusted
POST2 blood lead levels were seen among Study Group children of different races.
No "dose-response" relationship was observed between the mean change in blood lead
level and the starting soil lead level or the size of the excavated area. POST2 abatement
blood lead levels were quite similar for children in the lowest and highest pre-abatement soil
lead categories and the smallest and largest excavated yard areas. The lack of a trend should
be evaluated in light of the study eligibility criteria that restricted the soil and blood lead
1-10
-------
ranges. Only six children in the Study Group had median pie-abatement soil lead levels that
were less than 1,000 PPM, and pre-abatement blood lead levels were restricted to 7 through
24 fig/dL.
Exploratory multivariate analyses were also conducted to control simultaneously for
several potential confounding variables. Two variable selection methods were used. First,
a backward elimination procedure identified variables that were statistically significant
predictors of POST2 blood lead levels. When Pre-Pb, age, race, and lead jobs were
controlled simultaneously the adjusted POST2 blood lead levels were 10.36, 11.26, and
11.66 /ig/dL for Groups S, A, and B, respectively, and the adjusted differences between the
Study Group and Control Groups A and B were 0.90 and 1.31 /xg/dL, respectively. The
overall group effect was not statistically significant (p=.08).
Second, a potential confounding variable was selected for the multivariate model if its
inclusion in the base model altered the magnitude of difference between the Study Group and
either Control Group by more than 10%. The variables identified by this criterion were
race, socioeconomic status, and playing or sitting on the floor. In a model controlling these
variables and Pre-Pb, the adjusted differences between the Study Group and Control Groups
A and B were 0.80 and 1.21 /ig/dL, respectively. The overall group effect was not
statistically significant (p=.16).
1.7.3 Handwipe Lead Levels
Because the handwipe field blank lead levels varied considerably and were not
individually matched to the participants, background levels were taken into account by
subtracting the maximum or median field blank level for each sampling round. When the
maximum level was subtracted, the mean hand lead level in all groups declined from the pre-
abatement to the first post-abatement sampling round. The mean hand lead level in the Study
Group changed little at the second post-abatement sampling round while it increased in the
Control Groups. When the median level was subtracted, the mean hand lead level in the
Study Group declined at the first and second post-abatement sampling rounds. The mean
hand lead levels in the two Control Groups first declined and then rose to a level higher than
baseline.
1-11
-------
Because the PRE and P0ST2 sampling rounds are most closely matched on season, we
focused subsequent analyses on this comparison. When the maximum blank level was
subtracted, the mean hand lead level decreased by 3.61 fig in the Study Group (p=.02),
0.99 ng in Control Group A (p=.69), and 0.36 fig in Control Group B (p=.85). When the
median blank lead level was subtracted the mean hand lead levels declined by 2.75 pg in the
Study Group (p=.08), and 0.68 in Control Group A (p=.79) and increased by 0.76 in
Control Group B (p=.72).
When the POST2 hand lead levels were adjusted for baseline level the mean differences
between the Study Group and the two Control Groups were diminished; the magnitude of the
reduction was greater for the Control Group A comparison. Group assignment was not,
however, a significant predictor of POST2 hand lead levels (p values were .48 and .43,
respectively).
1.8 CONCLUSION
One of the most difficult aspects of the childhood lead problem is identifying the
sources of lead and determining their relative contribution to children's lead burden. Lead
based paint and household dust have received most of the attention to date. Far less attention
has been paid to urban outdoor sources of lead, especially soil, except in cases of stationary
sources such as smelters. Our findings suggest that lead contaminated soil does contribute to
the blood lead levels of urban children.
Numerous previous studies have shown that soil and dust lead levels are correlated
with children's blood lead levels. These studies have relied largely on cross-sectional data,
often from communities with point sources of lead such as smelters, where soil lead
concentrations are far greater than those typically found in urban settings. The current study
found that soil abatement alone (Study vs. Control Group A) was associated with a 0.8 to
1.4 fig/dL decline in blood lead levels and that soil and interior dust abatement combined
(Study Group vs. Control Group B) was associated with a 1.2 to 1.6 fig/dL decline. These
blood lead changes were observed approximately one year following soil abatement in which
surface soil lead levels were dropped an average of 1,856 PPM.
1-12
-------
Although designed and conducted to produce rigorous results, the study has several
limitations. Participants were chosen to be representative of the population of urban
preschool children who are at risk of lead exposure by using the Boston Childhood Lead
Poisoning Prevention Program to identify potential participants from neighborhoods with the
highest rates of lead poisoning and by using as wide a range of blood lead levels as was
practical. Since no study subjects had blood lead levels below 7 /tg/dL or in excess of
24 ng/dL at baseline, the study provides no information about the effect of lead contaminated
soil abatement for children with these lead levels. Similarly, a different effect might have
been found for children who had a greater blood lead contribution from soil, such as in
communities with smelters or other stationary sources where soil lead levels are substantially
higher than those seen in this study, or where differences in particle size result in differences
in bioavailability.
There are little data available about rates of change in children's blood lead levels
following a change in exposure to a potential source of lead. It is possible that the
intervention would have been associated with a greater reduction in children's blood lead
levels had they been followed for a longer period of time. In addition, all children in the
study were exposed to lead contaminated soil prior to enrollment and so we are unable to
investigate whether exposure to lead contaminated soil in the first year of life is associated
with higher blood lead levels. Lastly, the unit of abatement was the single premises rather
than clusters of premises. It is possible that the effect of lead contaminated soil abatement
on children's blood lead levels would have been greater had we also removed lead
contaminate soil from properties that surrounded Study Group children's premises.
In conclusion, this intervention study suggests that an average 1,856 PPM reduction in
soil lead levels results in a 0.8-1.6 ^ig/dL reduction in the blood lead levels of urban children
with multiple potential sources of exposure to lead.
This study provides information about soil abatement as a secondary prevention
strategy, that is the benefit to children already exposed to lead derived, in part, from
contaminated soil. It can not be used to estimate the primary prevention effect of soil
abatement. Since children's post-abatement blood lead levels reflect both recent exposure
and body burdens from past exposure, the benefit observed is probably less than the primary
1-13
-------
prevention benefit, that is the benefit of abating lead contaminated soil before children are
exposed to it so as to prevent increases in blood levels and body stores.
Soil lead tends to be concentrated at the surface and it is not rapidly removed by natural
processes. Once soil is contaminated with lead, it is likely to remain contaminated
indefinitely. In the future, soil is likely to become one of children's most intense sources of
lead as the current housing stock, 52% of which is estimated to have dangerous
concentrations of lead paint, ages and is replaced. Lead contaminated soil abatement may
well result in long-term reductions in environmental lead so that multiple future generations
of children benefit as they move onto abated properties. This thesis is currently untested,
however, and must be validated by monitoring abated properties for rates of reaccumulation
of lead.
1.9 IMPLICATIONS
Soil abatement in this study was associated with an approximately 0.8-1.6 ugldL
reduction in children's blood lead levels, slightly less than what was originally hypothesized.
The clinical and public health implications of a reduction of this magnitude are not readily
apparent. The magnitude of reduction in blood lead observed suggests that lead contaminated
soil abatement may not be a particularly useful clinical intervention for children with low
level lead exposure. It might be extremely useful, however, in specific situations, such as if
soil lead were extremely high or the particular child had pica for soil. It is also a relatively
inexpensive and low technology intervention. Although there are no data regarding the
relative safety of soil and lead based paint abatement, it seems unlikely that soil abatement is
as dangerous to children, families, and workers as lead based paint abatement can be.
Although the average benefit associated with abatement of lead-contaminated soil is
modest in this study, the societal impact may be substantial. Consider, for example, the
impact on the blood lead distribution of an average decline of 1 or 2 ^ig/dL in the mean
blood lead level of a population of children assuming a starting mean blood lead level of
12 /xg/dL, a standard deviation of 4, and a normal distribution. We also assume that the
amount of change (as opposed to the percentage of change) is constant for all starting values,
as we observed in our own sample in which the distribution of starting values was truncated.
1-14
-------
Specifically, this assumption may not apply to children with starting blood lead values greater
than 25 /xg/dL. A decline of 2 /xg/dL in the mean blood lead level results in 72% as many
children with levels exceeding 10 /xg/dL, 47% as many children with levels exceeding
15 Mg/dL, and 26% as many children with levels exceeding 20 /xg/dL (values of 10, 15, and
20 jxg/dL were chosen because they correspond to the new CDC definition of lead poisoning
and the new action levels for environmental and medical intervention, respectively). Even
a 1 /xg/dL decline in mean blood lead level results in 87%, 70%, and 52% as many children
with levels of 10, 15, and 20 jtg/dL, respectively. The percentage shifts may differ
somewhat in a more representative sample in which the distribution of starting values is
likely to be log normal.
Policy decisions regarding urban lead containinated soil abatement as a lead control
strategy will require numerous considerations. For example, are^other types of remediation
(e.g. planting grass cover and shrubs) equally effective but less expensive and intrusive?
How does the cost effectiveness of soil abatement compare to other lead exposure reduction
activities, such as paint abatement? Will it be practical to perform large scale abatements
without encountering problems regarding the disposal of lead contaminated soil? Will future
research help specify whether changes in children's blood lead levels of the magnitude seen
in this study are clinically relevant or prudent from a public health or societal perspective?
And will we develop and sustain the resolve and commit the resources needed to prevent
what remains the most important environmental health problem of children in the United
States?
1-15
-------
2. BACKGROUND
Perceptions of the child lead poisoning problem have steadily changed as evidence has
accrued demonstrating subtle but serious metabolic and developmental consequences of lead
exposure levels previously believed to be innocuous.1,2 Childhood lead poisoning was
initially perceived as a disease (often presenting as encephalopathy and sometimes resulting
in seizures, coma, or death) associated with the ingestion of peeling lead paint. Over the
past two decades, as scientific evidence has consistently revealed deleterious effects at lower
and lower lead levels, regulatory agencies have reduced the acknowledged level of children's
lead burden requiring environmental and medical intervention and clinical guidelines have
been revised accordingly. Whereas concern was initially directed at symptomatic children
with blood lead levels of 60 ^g/dL and above, lead poisoning is currently defined by the
Centers for Disease Control (CDC) as a blood lead level of 10 /ig/dL or greater.3 The
Agency for Toxic Substances and Disease Registry estimates that approximately 17% of all
children in the United States aged six months to five years have blood lead levels of
15 /xg/dL or greater.4 There is currently no lead level believed to be safe for children.
Children are exposed to lead from multiple sources.4"10 The most important sources
include lead contaminated paint, dust, soil, and water. Paint used on both the interior and
exterior of houses through the 1950's and continuing, to some extent through the 1970's,
often contained high concentrations of lead.11 It is estimated that 42 million homes in the
United States, or approximately 52% of all housing units, contain paint with more than
0.7 mg/cm sq. of lead.4 This enormous reservoir of lead, estimated to represent more than
three million tons, is easily accessible to young children.
More recently, concern has been raised that lead contaminated soil in older urban areas
is another important vector for children's exposure to lead.4,12,13 The sources of soil
contamination include lead paint chips from deteriorated exterior paint, past deposition of
airborne lead from gasoline, and point sources such as smelters and other industrial
activities.
House dust is, in part, composed of soil10,14 and can therefore be contaminated by
exterior lead sources. Other sources of house dust lead may be deteriorating lead based paint
2-1
-------
from furnishings or interior walls.6,8'15 Drinking water may contain high concentrations of
lead from old pipes or leaded solder. Children may also ingest lead from pottery and canned
foods although this is generally viewed as a minor contributor to exposure for most
children.4
There is a general consensus that children's exposure to lead should be reduced as
much as possible.4,16 With clear and growing evidence of long-term adverse cognitive and
1 4
behavioral deficits associated with levels of lead as low as 10 figldL, ' increasing numbers
of authorities have argued that there is an urgent need to develop practical and cost effective
approaches for the prevention and treatment of low level lead exposure.4,11,17 It was in
response to this mandate that the Boston Lead-In-Soil/Lead Free Kids Demonstration Project
was conducted. The study was designed to provide scientifically rigorous data about the
effectiveness of lead contaminated soil abatement in lowering children's blood lead levels, the
cost of removing lead contaminated soil, and a number of related questions relevant to
policymakers, public health officials, child advocates, and clinicians.
2.1 LEAD POISONING IN BOSTON
As in many U.S. cities, childhood lead poisoning is a widespread problem in Boston.
Children between the ages of nine months to six years are at greatest risk because they have
a high degree of hand-to-mouth activity, they absorb ingested lead more efficiently, and
because of the heightened vulnerability of their developing nervous systems to lead toxicity.
In recent years in Boston, the rate of identified lead poisoned children in this age group
ranged from 1.5% and 2.0%, on the basis of pre-1991 CDC guidelines (i.e., blood lead level
greater than 25 jtg/dL).
In order to identify the areas in Boston with the highest rates of childhood lead
poisoning, the Boston Department of Health and Hospitals' Office of Environmental Affairs
mapped all children in Boston identified as lead poisoned between October 1979 and
February 1985. These efforts demonstrated that lead poisoning in Boston, while occurring
throughout most of the City, was, to a surprising degree, concentrated within very limited
18
geographic areas. It showed that four high prevalence neighborhoods accounted for 87%
of the city's lead poisoned children but only 56% of the at-risk (nine months to six year old)
2-2
-------
population. It also showed that children living in 28 2-3 city block areas produced nearly
30% of Boston's child lead poisoning cases despite accounting for only 4% of the child
population aged nine months to six years. In each of these small areas, designated
Emergency Lead Poisoning Areas (ELPAs), an average of more than 30 children were lead
poisoned. This represents more than one of every four children.
2.2 CONTAMINATED SOIL IN BOSTON
The soil in Boston is contaminated by lead-based paint which has weathered or been
scraped off the exterior of buildings and by the deposition of lead in gasoline exhaust.
Scientific studies that correlate increases in blood lead levels with exposure levels have
not shown a significant contribution by exposure to soil with less than 500 parts per million
(PPM) lead. These studies have suggested that soil lead levels of 500-1,000 PPM can
significantly contribute to children's lead burdens, although other factors such as particle
size, distribution and lead species are important.4,10,13'19"21 At present, however, lead
contaminated soil is regularly not removed either as part of a comprehensive strategy to
prevent childhood lead poisoning, or an environmental intervention on behalf of children who
already have suffered excessive exposure. In part, this may be due to the lack of data
demonstrating the effectiveness of lead contaminated soil abatement. In the ELPA's
described above the surface soil lead level averaged more than 3,000 PPM, or 3-6 times the
A 4 A
"acceptable level" established by the CDC. ' Testing at numerous sites throughout Boston
has revealed much lower average lead levels of 600-700 PPM.
In October, 1986 the reauthorization of the Superfirad toxic waste cleanup program
(SARA) was signed into law. Included in the bill was a provision, Section 111 (a) (6),
providing funds for "a pilot program for removal, decontamination, or other action with
respect to lead-contaminated soil in one to three different metropolitan areas." Boston was
chosen to implement the first of the projects.
The EPA convened two workshops of lead experts to provide consultation on the design
of the study. The first workshop was held in April, 1987 in Raleigh, North Carolina.
It brought together individuals with expertise in the health effects of lead exposure,
epidemiology of lead toxicity, the biogeochemistry of lead, and the abatement of
2-3
-------
environmental sources of lead. A general design and evaluation for the study was drafted at
this workshop. A second workshop was held in Lexington, Massachusetts in June, 1987.
It was devoted to a continuing exploration of (1) possible study designs that could provide
scientifically rigorous data on the relationship between preschool children's exposure to lead
contaminated soil and blood lead levels, and the effectiveness of the removal of lead
contaminated soil in reducing low level lead exposure; (2) the ethical, legal, and logistical
constraints on the design of any such study conducted in Boston, Massachusetts; and (3) the
process by which broad-based scientific, pediatric, and public health acceptance of a
scientifically sound and implementable design could be achieved.
2.3 IMPLEMENTATION
The first phase of the study ran for ten months from May 28, 1987 to March 31, 1988.
It involved: (1) location and establishment of study facilities; (2) procurement of equipment
and supplies; (3) recruitment of some staff; (4) examination of scientific, legal and ethical
problems and issues; and (5) efforts directed at developing a study design in conjunction
with EPA staff. During this phase the significant implications of the Massachusetts Lead
Poisoning Prevention Law for the design and conduct of the study were explored.
In response to these issues, and in an effort to resolve related scientific and ethical
issues, a small group of medical, scientific, and public health experts assumed responsibility
for designing and implementing a lead contaminated soil abatement study in Boston,
Massachusetts early in 1988. They were: Michael Weitzman, M.D. (Principal Investigator),
Ann Aschengrau, Sc.D. (Coinvestigator), David Bellinger, Ph.D. (Coinvestigator) and
Mr. Ronald Jones, B.A. (Coinvestigator).
An initial draft of a proposed study design was submitted to the Environmental
Protection Agency on January 22, 1988. In May of 1988 the investigators hosted a meeting
attended by Evan Charney, M.D., representatives of the Massachusetts Department of Public
Health, the Conservation Law Foundation, the Centers for Disease Control, and the
Environmental Protection Agency. At this meeting the investigators presented their proposed
study design and two alternative designs, one by Dr. Renate Kimbrough of the
Environmental Protection Agency and another by Dr. Michael Rabinowitz, then of Harvard
2-4
-------
Medical School. The attendees unanimously endorsed and suggested ways to strengthen the
study design proposed by the investigators from the Boston Lead-in-Soil Demonstration
Project.
In August of 1988 a revised study design that incorporated suggestions from the May,
1988 meeting was submitted to the Environmental Protection Agency along with letters of
support from participants in the May meeting. The study was given the Environmental
Protection Agency's full support, contingent on the approval of the Human Studies
Committee of the Trustees of Health and Hospitals of the City of Boston. In the fall of 1988
the proposal was submitted to the Human Studies Committee of the Trustees of Health and
Hospitals of the City of Boston, and recruitment of staff began. In December of 1988 the
full approval of the Human Studies Committee was obtained. Enrollment of study
participants began in January 1989.
This document represents the final report to the Environmental Protection Agency, and
as such describes in detail the study's design, implementation, problems encountered, data
collection and analysis, and findings.
2-5
-------
3. STUDY ADMINISTRATION
The study was managed by a team consisting of a Principal Investigator
(Michael Weitzman, M.D.) and three Coinvestigators (Ann Aschengrau, Sc.D.,
David Bellinger, Ph.D., and Ronald Jones, B.A.). The day-to-day management was the
responsibility of a full time study Administrator (Natalie Zaremba) and Assistant
Epidemiologist (Julie Shea, MPH). An organizational chart can be found on page 23.
The study was designed to evaluate the impact of a laige scale environmental
intervention on the blood lead levels of a specific target population. It required a concerted
effort by the U.S. Environmental Protection Agency, the Trustees of Health and Hospitals of
the City of Boston, Inc., the City of Boston, and the study's investigators and staff.
Responsibilities for the various aspects of the study are listed below.
THE U.S. ENVIRONMENTAL PROTECTION AGENCY:
1. Provided funding to conduct the study.
2. Designated a Project Manager from EPA Region I.
3. Assisted in the development of Protocols.
4. Provided analyses of soil and dust environmental samples.
5. Provided representatives for community meetings and other activities conducted
as part of the Community Relations Plan.
6. Provided for or assisted with training and guidance in the collection of soil and
dust samples.
7. Assisted the study staff in the calibration of equipment.
THE TRUSTEES OF HEALTH AND HOSPITALS:
1. Provided bookkeeping, accounting, and other fiscal services.
2. Provided personnel management services for the study.
3. Provided internal fiscal audits for the study.
4. Provided for the long-term maintenance and storage of client records and data.
3-1
-------
THE CITY OF BOSTON:
1. Provided assurance of appropriate removal and disposal of lead contaminated
soil.
2. Ensured access through rights-of-way and easement necessary to the study.
3. Provided logistical support during the soil removal phase of the study.
4. Assisted in the resolution of the soil disposal controversy.
THE STUDY'S INVESTIGATORS AND STAFF:
1. Developed study Protocols and provided for their review.
2. Developed a Community Relations Plan and provided for its implementation.
3. Provided a Management Staff to supervise all study activities except those
specifically provided by the Trustees, EPA, or the City. Project Management
Staff coordinated all phases of field work.
4. Acquired and maintained suitable space for an operations center and for training
of study staff.
5. Acquired and maintained computerized data systems suitable for recording,
storing, and analyzing data generated in the course of the study.
6. Provided for recruitment of households in areas selected for the study.
7. Prepared and printed instructions, maps, questionnaires, consent forms, and other
materials to be used in the study.
8. Collected environmental samples for quality control purposes.
9. Furnished equipment and containers for soil and dust environmental samples and
provided appropriate sample preparation.
10. Contracted for the laboratory analysis of blood samples and provided appropriate
sample preparation.
11. Contracted for the laboratory analysis of water samples.
12. Provided personnel to conduct interviews, draw blood, collect handwipes and
environmental samples, and use the XRF machine for paint lead analysis.
13. Validated all environmental data and maintained the following:
a. Area maps used in the study;
b. Questionnaires;
c. Tabulation of blood lead and free erythrocyte protoporphyrin results by
name, age, and address; and
d. Forms recording lead readings by the XRF device.
14. Implemented quality assurance measures throughout the study, except for
laboratory work performed by the EPA.
15. Followed standard chain-of-custody measures for all samples.
16. Provided follow-up for all study children found to be lead poisoned according to
CDC guidelines.
17. Notified all parents of the results of blood and environmental tests and provided
an interpretation of the results.
18. Developed and implemented a data analysis plan in conjunction with EPA and
other appropriate organizations.
3-2
-------
19. Developed and implemented a plan to secure the cooperation of community
residents and property owners who were directly affected by the study but were
not the parents of children participating in the study.
20. Advertised for, negotiated, and managed contracts for interior dust and loose
paint abatements, soil abatement, moving, storage, and deleading.
21. Prepared draft and final study reports in conjunction with the EPA.
22. Provided necessary, authorized equipment.
23. Provided lead poisoning education, supportive services or referrals to parents of
children in the study.
3-3
-------
4. HUMAN STUDIES REVIEW
During the fall of 1988 the Boston Department of Health and Hospitals Human Studies
Review Committee reviewed the proposal submitted to Region I of the Environmental
Protection Agency in August, 1988 as well as all study protocols. The complex ethical,
legal, and scientific concerns raised about the study in 1987 and early 1988 by the
Massachusetts Department of Public Health, the legal counsels of the City of Boston and
Region I of the Environmental Protection Agency were also submitted to the Human Studies
Review Committee for the Committee's review. Full approval of the Human Studies Review
Committee was granted in December, 1988 and Annual Reports were submitted to the
Committee in December 1989 and 1990.
4-1
-------
5. STUDY DESIGN
5.1 PURPOSE
The purpose of the Boston Lead-in-Soil Demonstration Project was to determine the
effect of removing lead contaminated soil on children's blood lead levels. The hypothesis
tested was:
A significant reduction (equal to or greater than 1,000 PPM) of lead
in soil accessible to children will result in a mean decrease of at least
3 jtg/dL in the blood lead levels of children living in areas with
multiple possible sources of lead exposure and a high incidence of
lead poisoning.
We were also interested in a series of related questions relevant to explicating whether
lead contaminated soil is an important vector for lead exposure for children living in highly
lead contaminated environments and whether soil abatement is a feasible intervention: Is the
soil abatement more or less effective for certain subsets of children? Is interior dust
abatement effective in reducing children's blood lead levels and how quickly and to what
extent do soil and house dust become recontaminated after abatement? What is the cost of
soil and dust abatement? What problems can be anticipated if lead contaminated soil
abatement were widely adopted as a strategy for the primary and secondary prevention of
childhood lead poisoning?
The final study design, described in detail in the following sections, is illustrated on the
next page in Figure 5-1.
5.2 IDENTIFICATION OF STUDY POPULATION
The study used the ongoing city-wide screening efforts of the Boston Childhood Lead
Poisoning Prevention Program (BCLPPP) to identify potential participants. The BCLPPP
receives the results of capillary blood screening tests (blood lead levels and free erythrocyte
protoporphyrin levels) for many of the preschool children living in Boston. The source
5-1
-------
Pool of Eligible Participants
Enrolled Participants
(Random allocation)
i
Study Group Control Group A Control Group B
Year 1 Loose Paint,
Dust, and Soil
Abatement
Loose Paint
and Dust
Abatement
Loose Paint
Abatement
Year 2
Soil Abatement, *
Soil Abatement, *
* Interior and Exterior Paint Abatement if Indicated and Desired
Figure 5-1. Study Designs.
5-2
-------
neighborhoods of Boston who were under four years of age on August 1, 1989 who had
finger stick blood lead screening tests done as part of their routine health care between
January and June 1989 and whose screening levels were between 10 and 20 /xg/dL.
Additional children under four years of age who lived on the same premises as the BCLPPP
participant during the recruitment period were also identified for possible enrollment. The
map (Figure 5-2) on the following page shows the area of Boston involved in the Study.
5.3 ELIGIBILITY CRITERIA
Homes of potential study participants were visited by trained study field staff, and
families and landlords were contacted to determine if potential participants met the following
additional eligibility criteria:
1. The participant's paient(s) or caretakers) and, if applicable, landlord agreed to
participate;
2. Exterior walls of premises had little or no chipping paint. On a drive-by
inspection, study staff judged by visual inspection that (1) the cumulative amount
of chipping paint on the exterior walls (excluding trim, but including porches)
did not exceed 30% of the total surface area; and (b) the cumulative amount of
exterior chipping paint on the adjacent wall of an abutting premises (including
trim) did not exceed 40%.
3. Premises had a yard of at least ten square feet composed of dirt, grass or a
combination thereof and was accessible to the child.
4. The child resided in a dwelling with eight or fewer residential units.
5. Average or median surface soil lead level was 1,500 PPM or greater.
6. Child was mobile
7. Child had never been lead poisoned.
5-3
-------
/ v* ttr' *lป"A? ^"V ~*J\ nyo* -* .
> ซA a'
'V .-
\ R(AOV>Ll.C
>v*
MAtfAP
-r
x%16ir,'8ss. e"
^r^a# nrv -x \>
MILTON I .
[MStftorฎ
i? wear quihcy
;V\
VOL! ASTOnX. \J/ป
Mf MUTOH ^Vy
QUINCYyj
fc *4Vf * ฆ*.'' "
lปlinKtorJIl-SL--rtS' ^ ซ -- ;ฆ;ป / - *ป-ป<-*ป \j,. ^
^cx "/Tat - L. rs^?-
\ ^ > \ / J /wS \ ^t*^oVv a <*ซ. I / c** (
A ฆ
ซ ~~ ^*>2''* IN T R E E^
"'"^j qS^Sj" -""J
. I r**wmr^3
~ /
Figure 5-2. Map of study area.
-------
8. Family resided on premises for at least three months (as of the baseline venous
blood lead test).
9. Family had no definite plans to move within the next three months after
enrollment.
5.4 RATIONALE FOR ELIGIBILITY CRITERIA
The choice of eligibility criteria was motivated by both scientific and practical
considerations. The paramount concern from a scientific standpoint was to maximize the
ability of the study to detect a decrease in blood and hand lead levels following soil
abatement.
Children residing in Dorchester, Roxbury, Mattapan, and Jamaica Plain were targeted
because these areas were known to have a high incidence of childhood lead poisoning as well
as elevated soil lead levels. Only children up to four years of age at baseline were included
because these children still have a high degree of hand-to-mouth activity.
Participant children were required to be mobile, live in small to medium sized
residences, have accessible yards composed, at least in part, of contaminated soil because
these children would have the opportunity to come into both direct and indirect contact with
contaminated soil. The exact yard and premises size requirements were arbitrarily chosen.
The minimum soil contamination level was set at 1,500 PPM in order to make possible the
1,000 PPM decline in soil lead specified in the study hypothesis.
Based on the precision of the analytic method used to determine blood lead level, the
hypothesized drop in blood lead levels following abatement, and the sample size
requirements, we established 7 /*g/dL as the minimum baseline blood lead level required for
eligibility. Children whose baseline blood lead levels were above 24 /tg/dL were excluded.
They met the current definition of lead poisoning and so were likely to undergo medical and
environmental interventions during the follow-up period (i.e. chelation and paint deleading)
that could overwhelm the changes expected from the study interventions. Previously lead
poisoned children were excluded because of the possibility that their response to the study
interventions might differ from that of non-poisoned children due to their elevated body lead
burden.
5-5
-------
Homes and abutting properties were required to have little or no exterior chipping paint
in order to minimize the likelihood that any impact of the abatement would be attenuated by
the rapid recontamination of study premises' soil (please see Eligibility Protocol for a
description of how the extent of chipping was ascertained).
Three months residency at the current premises was required to ensure that a child's
baseline blood and hand lead levels reflect the lead levels in various environmental media
around the premises. Lastly, families were included only if they had no definite plans to
move within the three months following enrollment to minimize attrition during the course of
the study.
These eligibility criteria were applied both to children identified through the BCLPPP
and other children living on the premises. Children were not excluded if they attended a day
care center or day camp in the summer.
Eligibility criteria were generally evaluated in the following sequence: routine blood
screening, drive-by assessment to determine exterior condition of the home, sample soil,
interview family, search for other children residing on the premises, recruit landlord. All
children meeting the criteria received a venous blood lead determination beginning in August,
1989. If the blood lead level was between 7 and 24 /ug/dL and the child's parents and the
landlord of the premises agreed to participate, the child was enrolled in the study. This was
considered the child's baseline blood lead level for the purposes of this study. All baseline
venous blood lead levels were obtained prior to any environmental abatement activities.
At enrollment, informed consent for the subsequent environmental sampling, interview,
blood tests, hand lead determinations, etc. was obtained from the parent or caretaker.
Consent for the soil abatement was also obtained from the landlord.
The families of children with lead levels outside the eligible range were informed of the
results and of the reasons for excluding their children from the study. All children with
blood lead levels above 24 fig/dL were referred to the Boston Childhood Lead Poisoning
Prevention Program and followed according to Massachusetts state law and lead program
case management protocols.
Figure 5-3 Eligibility Assessment and Recruitment Flow Sheet depicts the eligibility
assessment and recruitment sequence, the numbers of children assessed and eligible at each
step of the process, and the final number of children enrolled. The numbers and percentage
5-6
-------
I
16,129 children
screened by BCLPPP January-June 1989
I
2,287 names received from BCLPPP
(children of appropriate age in target neighborhoods,
with screening blood lead levels between 10-20 iig.dl)
1,876 (82%) pursued by LFK
I
788 (42%) eligible after drive-by
I
645 (82%) eligible after preliminary soil sample
I
316 (49%) eligible after interview
153 other children located on same premises
Total of 469 eligible after interview
I
277 (85%) blood tests done
l
170 (72%) eligible after blood test
l
152 (89%) enrolled
Figure 5-3. Eligibility Assessment And Recruitment Flow Chart.
5-7
-------
of children judged ineligible by reason ineligible are given in Table 5-1. One hundred and
fifty two children were ultimately enrolled in the study.
5.5 INTERVENTION
As eligible participants were enrolled, they were randomly assigned to one of three
groups using a computer generated random number table: the Study Group, Control Group
A, or Control Group B (the randomization unit was the child's premises). Randomization
was used to enhance the probability that the three groups would be comparable with respect
to measured and unmeasured characteristics.
The study design is illustrated in Figure 5-1. The Study Group received loose interior
paint removal, interior dust abatement, and soil abatement in the first year. The unit of
abatement activity was the single premises on which the subject(s) lived. Control Group A
received loose paint removal and interior dust abatement in the first year of the study and
soil abatement in the second year. Control Group B received only loose interior paint
removal in the first year and soil abatement in the second year. The unit of abatement
activity for Control Groups A and B was also single premises.
The purpose of including several intervention groups was to enable separation of the
effects of soil and interior dust abatement. Removal of lead paint hazards was not included
as a study intervention but was suggested and facilitated, when indicated, after the second
follow-up study blood lead level was determined.
5.6 STUDY SIZE
We enrolled 152 children in the study: 54 children in the Study Group, 51 in Control
Group A, and 47 in Control Group B. Using data from other studies conducted with
comparable populations, we estimated that there would be, at most, a 15-20% attrition rate
during the course of the study and based our original statistical power calculations on a final
study size of 122 children, (See section on attrition and follow-up.) The statistical power of
the study to detect a 3 jug/dL difference in blood lead levels between the Study Group and
5-8
-------
TABLE 5-1. PERCENT OF CHILDREN JUDGED INELIGIBLE
ACCORDING TO REASON FOR INELIGIBILITY
Reason Ineligible Percent
More than eight residential units on premises 19.3
Yard less than ten square feet 1.6
Yard inaccessible to child 0.4
Cumulative exterior chipping paint on premises (excluding trim, S.2
including porches and walls) exceeded 30% of total surface area
Cumulative exterior chipping paint on adjacent side of an 2.3
abutting premises (including trim) exceeded 40%
At least 50% of the preliminary soil samples were not in excess of 6.2
1,500 PPM lead, or the average of the soil sample results did not exceed
1,500 PPM lead
Case manager unable to contact parent/guardian after five attempts 0.6
Paint deleading scheduled to be performed 0.9
Family intended to move in next three months 2.3
Parent/guardian was not interested in participating 4.1
Child's sibling had been lead poisoned while living at this residence 0.0*
and an environmental intervention was to be performed
Case manager unable to contact landlord after five attempts 0.8
Landlord not interested in participating 6.8
*
Data error unable to be resolved from screening form 0.1
Child did not have a permanent address 0.7
Case manager determined that landlord was not interested in 4.3
participating prior to landlord recruitment attempt
Child moved during eligibility phase 9.5
Child lived in public housing 2.5
Duplicate name/listing (e.g. child was screened twice during eligibility phase) 0.6
Premises located in unsafe area 0.7
Foster child 0.3
Child did not reside in Mattapan, Dorchester, Jamaica Plain, or Roxbury 0.0
Landlord and/or parent claimed that premises were scheduled to be deleaded 0.5
Residents claimed no children live on premises 0.3
Family was not located at the address. (This code was used when a case manager was not sure 1.0
if the family had moved, but could not find the family at the premises)
5-9
-------
TABLE 5-1 (cont'd). PERCENT OF CHILDREN JUDGED INELIGIBLE
ACCORDING TO REASON FOR INELIGIBILITY
*
Reason Ineligible Percent
Language barrier 0.5
Case manager unable to contact a resident on the premises (e.g., landlord) 0.7
for permission to sample soil
*
Child's venous baseline blood level was below 7/ig/dL 2.2
Child's venous baseline blood level was 25/xg/dL or greater 0.5
Child was not in the eligible age range 0.0
Phlebotomist unable to draw blood for baseline sample 0.0
Ineligibility codes issued in-house by the Assistant Epidemiologist.
Code became inapplicable during the course of the study.
Only one child in each of these categories.
BCLPPP = Boston Childhood Lead Poisoning Prevention Program.
**
i
***
+
in
ป*
Disposition of premises unable to be determined before enrollment 2.7
deadline (12/8/89)
Subject moved after receiving baseline blood draw but before abatement 0.1
was done
Parent claimed that child had been lead poisoned 0.2
Soil results not reported from EPA 0.1
Subject not interested in participating after baseline blood draw but 0.1
before abatement was done
Child lived in a dwelling in which there were insufficient number of 0.1
other children less than four years old
**
Interior chipping paint exceeded 16 square feet 0.2
No Date of Birth in BCLPPP+ data 3.6
No Pb Level in BCLPPP data 9.4
Pb in error in BCLPPP data 3.3
No address in BCLPPP data 1.6
Total 100.0
Note that ineligibility codes were issued for all children whose names were received from BCLPPP as well as
for "other" children who were recruited on the premises by case managers.
5-10
-------
each Control Group was excellent (.89-.91). The assumptions underlying these power
calculations were:
1. Mean blood lead level of 12.6.
2. Standard deviation of the blood lead level of 4.1.
3. Alpha level of 0.05 (two-tailed).
The estimates of mean and standard deviation were derived from the baseline study
data.
5.7 CHANGES IN STUDY DESIGN AND SAMPLE SIZE
Two major changes in the proposed study design were made after the study design was
submitted to the EPA in August 1988: elimination of one aim of the soil abatement group
and a reduction in sample size from 330 to 152. Both changes were made with the approval
of the Centers for Disease Control and the Environmental Protection Agency.
We originally intended to divide the Study Group into two subgroups. For one
subgroup, the unit of abatement activity would be the study subject's premises, and, for the
other, the unit would be an approximate six to seven house cluster of homes including and
adjacent to the study subject's home. The goal was to determine whether one abatement
strategy produced a greater decline in children's blood lead levels than the other. Difficulties
in recruiting landlords and budgetary constraints required us to abandon the cluster subgroup
and limit the Study Group to single residence abatement units. This reduction limited
somewhat the scope of the inferences that could be drawn about the impact of soil abatement
but in no way compromised the scientific integrity of the study.
In addition, we originally planned to include 330 children in the study: 130 in the
Study Group (65 each in the cluster and single premises subgroups) and 100 in each Control
Group. The power of this sample size to detect both 3 and 2 /zg/dL drops in blood lead
levels was excellent. Several reductions in the sample size were made during the course of
participant recruitment and 152 children were ultimately enrolled (approximately 50 in each
group). The final sample size met the minimum required by the Centers for Disease Control
5-11
-------
(i.e., 150 participants) and still had excellent power (.89 to .91) to detect a 3 ^g/dL decrease
in blood lead levels (even with 20% attrition).
The sample size was reduced because of difficulties recruiting landlords, budgetary
constraints, elimination of the cluster concept, and a larger than expected percentage of
baseline blood lead levels below the eligibility criteria of 7 fig/dL. The initial screening
blood lead determinations were based on finger stick samples, while the confirmatory
baseline levels were based on venous samples. We anticipated that almost all of the baseline
lead levels would fall within the eligible range (7 to 24 /xg/dL) but only about 70% did. We
suspect that the initial finger stick samples were prone to contamination during sampling.
5.8 ATTRITION AND FOLLOW-UP OF STUDY POPULATION
This study utilized randomization to achieve comparable Study and Control Groups.
Attrition because of population mobility and loss of interest can threaten comparability of
groups, terminate the assigned intervention, and/or result in a sample size so small as to
increase the likelihood of finding no effect when one does in fact exist. We recognized that
not all attrition could be avoided during the approximately 18 months that participant families
were involved in the study, and we anticipated 15-20% attrition. Table 5-2 lists by Study
Group the number of children initially enrolled, the number who moved during the study but
were followed so that blood and environmental samples could be obtained, the number
dropped from the study, and the number who remained at their original premises.
TABLE 5-2. FOLLOW-UP STATISTICS BY PARTICIPANT GROUP
Dropped Out Still at Original
Starting
Population of
Children
Moved But
Were Followed
Before 2nd
Follow-Up
Blood Test
Premises at 2nd
Follow-Up
Blood Test
Study Group
54
11 (20.4%)
0 (0.0%)
43 (79.6%)
Control Group A
51
8 (15.7%)
2 (3.9%)
41 (80.4%)
Control Group B
47
3 (6.4%)
1 (2.1%)
43 (91.5%)
Total
152
22 (14.5%)
3 (2.0%)
127 (83.6%)
5-12
-------
Valuable information on the effects of the intervention was gathered even from children
who moved during the course of the study. In effect these children received a partial
intervention. To obtain the necessary information on these children, the study staff made a
concerted effort to trace these families and obtain blood, handwipe, and environmental
samples according to the study schedule. Tracing methods included contacting friends,
neighbors, relatives, and landlords as well as the U.S. postal service. Information that would
facilitate tracing was collected at enrollment.
All families successfully traced were interviewed to obtain information on the date and
reason for moving. An attempt was made to obtain all environmental exposure data at both
the new and original residence as well as to obtain blood and handwipe samples. The
amount of time that a child resided at an abated residence was taken into account in certain
analyses.
In order to minimize the number of participants who dropped out due to lack of
sustained interest or commitment, parents and property owners were offered carefully
considered incentives to remain in the study until its completion. Incentives were also
available to families who moved away from their original address but remained involved in
the study. The incentives for participating families included a $25 per month gift certificate
at a local supermarket or general puipose store and at the end of the study a $150 gift
certificate at one of a variety of stores.
Property owners' cooperation also was vital to the study. In addition to having lead
contaminated soil removed, property improvement assistance in the form of assistance in
deleading proved to be a very effective incentive. The study offered to pay (1) the full cost
of interior and exterior paint deleading of owner-occupied homes; and (2) up to $2,000
towards the cost of interior and exterior paint deleading for non-owner occupied premises.
Table 5-3 lists the numbers of owner and non-owner occupied premises in the study,
households offered assistance with deleading, and the numbers agreeing to and receiving
assistance with interior lead paint abatement.
The money spent for participant incentives was a relatively inexpensive way to promote
good will and encourage continued participation in the study. Study participants performed
an invaluable service to children in these communities and ultimately throughout the
United States. Even though every effort was made to minimize the disruption to families,
5-13
-------
TABLE 5-3. INTERIOR AND EXTERIOR PAINT DELEADING ACTIVITIES *
Units in Units in
Owner-Occupied Non-Owner Occupied
Premises Premises
Total
Total
Offered Deleading
Assistance
Agreed to and Received
Both Interior and Exterior
Deleading Assistance
Received Only Exterior
Deleading
90
90
41
33
33
123
**
123
46
Deleading activities occurred from August to Februaiy, 1991 after all study samples were collected.
Six units already had deleading certificates (4 nonowner-occupied, 2 owner'occupied).
One owner did not receive a compliance letter.
***ซ . t
Several units were in the same premises.
Note that a single family home was counted as one unit; all single family homes were owner-occupied.
the study design required repeated visits to the homes of study participants for environmental
and biologic samplings and environmental interventions. These activities were intrusive and,
for many families, the puipose unclear. Thus, it was anticipated that some participants might
tire of the inconveniences related to the study and drop out. If this had happened in
sufficient numbers, the success of the study would have been seriously jeopardized and the
investments of time and money wasted. Participant incentives were critical in mitigating
these problems. Similarly, the study would not have been conducted without landlord
participation, necessitating our offer of assistance with lead paint abatement.
5-14
-------
6. PARENT EDUCATION AND COMMUNITY
RELATIONS STRATEGIES
Educational materials were developed and provided to all participating parents and the
public. They included information on the sources of lead in the environment, known effects
of lead on a child's well-being, and methods for reducing children's lead exposure (such as
wet mopping, washing hands). In addition, all parents were told the soil lead levels on their
premises and their children's blood lead levels.
The success of this study was dependent upon participant and community support and
cooperation. As a result, efforts to inform, educate, and involve potential participants and
community leaders were integral to conducting the study. The receptivity of parents and
property owners to enrolling and remaining in the study was related to a variety of factors.
For example, the degree to which participants and landlords perceived that participation was
in their best interest was a vital aspect of whether the individuals approached were willing to
enroll and remain involved despite substantial inconvenience. It was also reasoned that
whether participation resulted in positive recognition and improved standing in the
community or was a source of stigmatization would also impact upon enrollment and
retention. Thus, in addition to one-to-one educational activities and incentives for
participating families and landlords, a detailed community relations program was developed.
Community relations activities were designed to enhance the likelihood of positive responses
to the above mentioned concerns through activities that increased community acceptance of
the study.
An underlying objective of the community relations strategy was to ensure that the
study resulted in the least possible amount of intrusion for the community, and that, to the
extent possible, participants were recognized for their contribution to the success of the
study. A major priority was to ensure that communication with the community, especially
potential participants, was timely, forthright, and well presented. Thus, materials were
translated into the appropriate languages and study staff remained in close contact with
participating families and landlords and helped them plan and prepare for each of the study's
activities.
6-1
-------
It was also extremely important that respected and trusted members of the community
be apprised of the study's objectives, methods, progress, and difficulties. Thus, a
Community Advisory Committee was convened and met with the study's investigators during
the course of the study, both to keep them informed and to obtain their input into various
aspects of the study.
The effectiveness of the incentives employed and the community relations activities are
evidenced by the fact that the requisite number of families were enrolled and that only 3 of
the 152 (2%) children enrolled dropped out before completion of the study, despite numerous
intrusive and disruptive study related activities.
6-2
-------
7. ENVIRONMENTAL MEASUREMENTS
AND ANALYSIS
7.1 SOIL
Soil sampling was conducted to determine eligibility of properties, characterize the
potential exposure of participant children to lead from the soil, document the reduction in soil
lead levels, and monitor the rate of recontamination after abatement. Preliminary soil
sampling to determine eligibility was conducted from approximately April to November,
1989. The start date was chosen to allow for melting of snow and frost prior to sampling.
After eligible properties were identified and preabatement blood samples taken from the
children, detailed soil sampling of the properties was undertaken. Playgrounds frequented by
the children were also sampled. Follow-up sampling was conducted right after the soil
abatement to document the drop in soil lead levels and at approximately nine months after
abatement to assess the rate of recontamination. A detailed protocol for soil sampling and
analysis was developed in conjunction with the EPA and the other Lead-in-Soil
Demonstration Project teams from Baltimore and Cincinnati. Boston also participated in an
exercise to evaluate the merits of the wet digest method versus XRF and based on this, a
decision was reached to use XRF for soil analysis. The soil sampling process is summarized
below.
7.2 PRELIMINARY SOIL SAMPLING TO DETERMINE ELIGIBILITY
Following the drive-by survey of exterior paint conditions, a potential study participant
was approached by a study staff member who described the study and asked for permission
to sample the soil. If the potential participant was not available, other occupants of the
property were briefly told about the study and asked for permission to sample the soil.
In some cases preliminary soil sampling occurred before speaking with any residents, in
which case flyers about the study were left for all residents. In most cases, four composited
surface samples were taken within two meters of the house, one from each side of the house
where soil was present. Samples were analyzed by X-Ray Fluorescence (XRF) at the EPA
7-1
-------
Region I Laboratory. The property was eligible if the median or the mean soil lead level
was equal to or greater than 1,500 PPM.
7.3 DETAILED SOIL SAMPLING
After a child was enrolled, composited soil samples were taken throughout the property
both at the surface and at a depth of 15cm using one of three pattern sampling methods
described in the Appendix: line source, targeted, or small area patterns. The line source
soil sampling pattern used for most of the premises involved drawing lines parallel to the
premises about 0.5 meters away from the foundation and about 0.5 meters from the property
boundary. Depending on the size of the property, more parallel lines were added in between
the foundation and property boundary lines. Each parallel line was then divided into
segments seven meters in length. Composite soil samples were taken from a 2 by 2 foot
square at a random point along each line segment. The composite sample consisted of five
samples taken from the center and each comer of the square. The number of composited
surface and core samples taken varied according to the size of the yard. On average eight of
each were taken. Pattern selection was made according to the layout of the property.
Samplers made sketches showing property and sample locations. All soil samples were
transported to the EPA Region I laboratory and analyzed by XRF. Soil abatement was
documented immediately after landscaping by taking composite surface soil samples at every
other previously sampled location marked in copies of property sketches. On average, four
composite surface samples were taken at this stage.
7.4 ^CONTAMINATION ASSESSMENT SOIL SAMPLING
Approximately nine months after the initial soil abatement was conducted, composite
surface soil samples were taken to determine the extent of soil recontamination. These
samples were taken at every other previously sampled location marked on the property
sketches. On average about five composited samples were taken from each premises.
7-2
-------
The schedule for soil sampling was as follows:
Study Group Control Group A Control Group B
Summer-Fall 1989 Test
Abate*
Test**
Test
Test
Summer-Fall 1990 Test
Test
Abate*
Test**
Test
Abate*
Test**
* This refers only to soil abatement.
** These tests were conducted to document reductions in soil lead levels immediately
following the soil abatement.
Household dust sampling was conducted to characterize the potential exposure of
children to lead from dust, to document the reduction in dust lead levels following
abatement, and to monitor rates of recontamination after abatement. Our intent was to
sample dust on upfacing surfaces most accessible to the child (i.e., bare floors, window sills,
and wells). The Sirchee-Spittler modified dust buster was used to obtain the samples. This
instrument is a hand-held dust vacuum unit whose sampling head was modified to catch the
dust sample in a fine mesh (325) stainless steel screen. This modification enabled better
sample recovery than would be possible with the woven fiber cloth that is usually supplied
with commercially available dust busters. We generally took six samples in each household
from the following locations: the participant child's bedroom window well and floor, the
kitchen window well and floor, and the living room window well and floor. The three floor
dust samples were later composited into a single sample because the individual samples were
often lighter (less than 10 fig) than considered optimal for accurate XRF analysis. We
determined the lead concentration in the dust (PPM), the amount of dust per unit area
(mg/m2), and the lead loading (mg/m2).
A detailed protocol for the dust sampling was developed by Dr. Thomas Spittler of
Region I of the Environmental Protection Agency and is included in the Appendix.
7.5 DUST
7-3
-------
The schedule for dust sampling was as follows:
Study Group Control Group A Control Group B
Summer-Fall 1989 Test Test Test
Abate* Abate*
Test** Test**
Winter 1990 Test Test Test
Summer-Fall 1990 Test Test Test
* This refers only to interior dust abatement.
** These tests were conducted to document reductions in dust lead levels immediately after
the dust abatement with HEPA vacuuming.
7.6 WATER
Two water samples were taken during the course of the study. Each was a first flush
sample taken by the parent from the cold water faucet in the kitchen. Water samples were
analyzed by the Hall-Kimbrell Laboratory in Lawrence, Kansas. The water lead sampling
and analysis protocol can be found in the Appendix. Elevated water lead levels (i.e., above
50 fig/L) were reported to the participants. These participants were also informed of ways to
reduce the lead content of their drinking water.
The schedule for water sampling was as follows:
Study Group Control Group A Control Group B
Winter-Spring 1990 Test Test Test
Summer-Fall 1990 Test Test Test
7.7 PAINT
In the last year of the study portable x-ray fluorescence analyzers (PGT XK-3) were
used to identify lead in paint. Measurements were taken in the child's bedroom, kitchen, and
living room. One measurement was taken on the lower part of the wall and one was taken
7-4
-------
on the window sill in each room. A detailed protocol for lead paint inspection is included in
the Appendix.
The schedule for interior paint sampling was as follows:
Study Group Control Group A Control Group
Summer-Fall 1990 Test Test Test
7.8 QUALITY ASSURANCE FOR SOIL AND DUST SAMPLING AND
ANALYSIS
A quality assurance plan for the sampling and analysis of soil and dust was developed
by the EPA Region I and is included in the Appendix. It includes a description of the proper
procedures for soil sampling, sample custody, equipment calibration and analysis, internal
quality control checks and corrective actions.
7-5
-------
8. SOCIAL AND BEHAVIORAL QUESTIONNAIRE
After the baseline soil sampling, the case managers administered two questionnaires to
the child's parent or caretaker. The goals were to (1) gather information necessary to
characterize the study population and (2) assess factors that bear on a child's contact with
various sources of lead. Dr. Edmund Maes of the Centers for Disease Control initiated the
development of the questionnaire which was modified by Dr. Ann Asehengrau and
consultants from the Center for Survey Research at the University of Massachusetts, Boston.
Copies of the questionnaires are included in the Appendix.
One questionnaire, designated the "family questionnaire", was concerned with family
demographics (e.g., family size, parent's occupation and education, house cleaning), possible
sources of lead exposure (e.g., hobbies), and data on recent renovations and deleading
activities.
Another questionnaire, designated the "child questionnaire", collected information
intended to characterize each child's exposure to lead in soil. The respondent was asked to
identify the child's outside play areas (both in the immediate area of the dwelling and in the
neighborhood), to estimate the amount of time the child usually spends in each location, the
amount of time spent away from home (e.g., day care), handwashing, hand-to-mouth
activities, vitamin use, and nutritional data.
Because (1) the lead content of different foods may vary and (2) diet and nutritional
status may affect lead kinetics, specifically absorption, we also administered a food frequency
questionnaire.
After the interview, a study staff member measured the height and weight of the child
and obtained systolic and diastolic blood pressure readings. Parents' height and weight were
also obtained by interview.
Follow-up "family" and "child" interviews were done toward the end of the study to
assess changes in child behavior, house cleaning, new renovations, and increases in lead
related knowledge.
The study staff were trained in proper interviewing techniques by a staff member of the
Center for Survey Research of the University of Massachusetts. Each interviewer was
8-1
-------
required to tape her first few interviews and feedback was given. Interviews were translated
into Spanish, Portuguese, Creole, and Haitian Creole, and administered, as needed, in these
languages. These foreign languages were used in a total of 17.6% of the interviews.
8-2
-------
9. BIOLOGICAL SAMPLING ANฉ MEASURES
9.1 BLOOD SAMPLING
On three occasions during the study, blood samples of 2-3 ml each were drawn from
the antecubital vein for determination of blood lead, FEP (free erythrocyte protoporphyrin)
levels, and ferritin levels. The first sample was taken beginning in September 1989 prior to
any abatement activities, the second taken an average of six months (beginning in March
1990) after initial abatement activities, and the third taken an average of 11 months after
initial abatement activities (beginning July, 1990). Serum ferritin levels were obtained only
at baseline.
All laboratory analysis results were reviewed within one day of receipt from the
contract laboratory and health care providers were notified of the results. Any children with
blood lead levels of 25 ugldL or above were referred to the Boston Childhood Lead
Poisoning Prevention Program and followed according to Massachusetts state law and
accepted pediatric health practices.
An optional fourth blood sample was proposed in the original study design to be
obtained during 1991 if sufficient funds were available and if the effect of the soil abatement
was unclear.
The schedule for blood sampling was follows:
Study Group
Control Group A Control Group B
Fall 1989
Test
Test
Test
Winter 1990
Test
Test
Test
Fall 1990
Test
Test
Test
1991 Optional *
Test
Test
Test
* For children still residing on the original enrollment premises.
9-1
-------
9.2 BLOOD SAMPLE ANALYTIC PROCEDURES
We selected Environmental Sciences Associates (ESA) Laboratories in Bedford,
Massachusetts to perform the blood lead and free erythrocyte protoporphyrin (FEP) analyses
since they met all of our stringent performance criteria including experience in performing
biologic analyses for health studies, participation in proficiency testing programs, continuous
OSHA certification and a detection limit of 1 /xg/dL for blood lead. In quality assurance
testing administered by the CDC, ESA's coefficients of variation were 20.9, 13.8, and 7.1 %
at the 4, 10 and 46 /xg/dL blood lead levels. Bioran Laboratories in Cambridge,
Massachusetts performed ferritin analyses.
ESA Laboratories determined blood lead levels using graphite furnace atomic
absorption and EP levels using an ESA model zinc protoporphyrin hematofluorometer.
Protocols describing both methods are included in the Appendix.
9.3 HAND LEAD DETERMINATIONS
Handwipe samples were obtained each time blood samples were drawn. Parents were
asked not to wash the child's hands for the two hours immediately preceding sampling.
Wearing disposable gloves, a study staff member wiped all surfaces of each hand, front and
back up to the wrist, with three commercial wetwipes (Walgreen's Wetwipes). Sampling
took place inside the child's home. To assess the extent of any contamination during
sampling, field blanks consisting of six additional wipes were handled so as to simulate
wiping the child's hands, and set aside to determine the background wetwipe lead levels.
Field blanks were taken for every tenth child. Each set of six wetwipes was placed in a
sealed container, labelled and transported to Dennison Laboratories, Woburn, Massachusetts
where they were composited for chemical analysis. Extraction of the lead utilized IN hot
HN03. The total quantity of lead was reported in ng per pair of hands. Sampling and
Analysis protocols are in the Appendix.
9-2
-------
The schedule for hand lead determinations was as follows:
Study Group Control Group A Control Group B
Fall 1989 Test Test Test
Winter 1990 Test Test Test
Fall 1990 Test Test Test
1991 Optional * Test Test Test
* For children still residing on the original enrollment premises.
9.4 QUALITY ASSURANCE AND CONTROL FOR BLOOD LEAD
MEASUREMENTS
ESA Laboratories maintained strict internal quality control systems for their blood lead
analyses including composition of calibration curves with at least one reagent blank and three
standards, running standards both at the beginning and end of large runs, running known
control and QC material with every set of standards and every ten samples, and running
duplicates and spiked samples every ten samples.
In addition, ESA participated in the external quality control system developed and
overseen by Dr. Daniel Paschal of the Centers for Disease Control. CDC developed quality
assurance standards for specimen collection, preservation and shipping, analytic method
performance, bench and blind quality control materials, accuracy and blanks, and data
integrity that are described in detail in a protocol in the Appendix.
The protocol also includes the results of four whole bovine blood pool analyses
comparing ESA to the CDC, Cincinnati and Baltimore Lead-in-Soil Demonstration Project
laboratories. The conclusions drawn from these analyses were that: (1) ESA and the other
laboratory blood lead results were comparable; (2) each laboratory's blood lead data were
produced from analytical systems in statistical control (as defined by Shewhart); and
(3) no statistically significant time trends were observed.
9-3
-------
9.5 REPORTING AND EVALUATION OF CLINICAL DATA
The results of a child's blood lead, FEP, and ferritin analyses were provided to the
family and, with the family's permission, to the primary health care provider. Our staff's
relationship with primary health care providers, such as neighborhood health centers, was
vitally important to the success of the study. Study staff took every opportunity to encourage
study participants to maintain ongoing relationships with their primary health care provider.
9-4
-------
10. DETAILED DESCRIPTION OF THE
INTERVENTIONS
10.1 LOOSE PAINT ABATEMENT
The purpose of loose paint abatement was to remove safely any very loose chipping
paint from the inside of the home without generating dust or leaving behind small paint
chips. No children were allowed to be on-site during this process. Loose paint abatement
consisted of vacuuming the loose paint areas with HEPA (High Efficiency Particulate
Aerosol Filter) vacuums, washing loose paint areas with a trisodium phosphate and water
solution, and painting the window wells with primer.
Loose paint abatement was conducted to minimize lead based paint as a potential vector
for children's exposure during the study period. Loose paint abatement should be
distinguished from deleading, which was conducted after the collection of all environmental
and child-based samples was completed. Interior and exterior deleading is described in a
subsequent section of this report.
10.2 INTERIOR DUST ABATEMENT
The purpose of the interior dust abatement was to significantly reduce the amount of
lead bearing dust in the treated homes. It always followed the loose paint abatement.
Dust abatement was accomplished by HEPA vacuuming and wiping surfaces with a wet
cloth and furniture with an oil treated rag. Floors, woodwork on walls, window wells, and
furniture surfaces were cleaned. Only the living areas were abated. The common entries,
stairways, etc. were not cleaned. In the Study Group and Control Group A, interior dust
abatement took place at the beginning of the first year. Dust abatement was not done in
Control Group B.
Loose paint and dust abatements were performed from October 1989 through January
1990. It became readily apparent that the loose paint and dust abatements were time
consuming and logistically complicated to accomplish, largely because of the inconvenience
they caused to participating families. Cancellations and postponements by participating
10-1
-------
families were frequent. All scheduled work was confirmed twice and attempts were made to
have "backups" that could be used to fill openings in schedules made by cancellations.
Families were required to be out of their homes during these abatement activities.
Many families considered cancelling just prior to the actual abatements because they had
nowhere to go during the several hours it took to perform the abatement. Therefore it was
frequently necessary to arrange activities for them during the period in which the loose paint
and dust abatements were being conducted. Families were offered lunch at McDonald's
Restaurant and free access to the Children's Museum, the New England Aquarium, The
Tropical Rain Forest exhibit at the Franklin Park Zoo, or the Museum of Science. Study
staff provided transportation to these various sites.
Two case managers from the study staff supervised these abatement activities and used
documentation forms to record progress. ACP Cleaning Inc. of Maiden, Massachusetts,
performed all interior loose paint and dust abatement activities.
10.3 SOIL ABATEMENT
Soil lead was abated on 35 premises in the Study Group during the Fall of 1989. Soil
abatement was undertaken on 58 premises in the Control Groups during the Fall of 1990.
Eight premises were not abated in the Control Groups. The methods used differed somewhat
during the two phases and are described in detail later in this section.
The purpose of the soil abatement was to remove and provide a barrier between lead
contaminated soil and the children living on the study premises. The abatement was to be as
permanent as feasible given the practical limitations of the study. The strategy for soil
abatement involved removing a six inch layer of topsoil and placing a fabric or synthetic
barrier topped with 8" of clean topsoil.
The initial plan was to remove 6" of soil, test the soil at the 6" depth, and continue to
remove soil until a level was reached where lead was present at less than 500 PPM. This
approach called for on-site soil testing, to be carried out with all of the workers and
equipment standing by for the results of the analysis. After pilot testing this method during
the fall of 1987, it was decided that it was far too time consuming to be practical given the
10-2
-------
large number of'properties to be abated. A decision was made to remove soil to a set depth
of 6" and replace it with 8" of clean topsoil.
In those situations where the driveway consisted of soil and the ground was frozen the
area was capped with a minimum 3" layer of asphalt. This approach was employed in four
instances. Soil removal was not undertaken in areas where asphalt was used as a barrier.
After excavation to 6" and replacement with 8" of clean topsoil, surface soil or cover
was tested to insure that it was not contaminated with lead. Surface soil or cover was
retested on average seven months later to ascertain whether it had been tecontaminated by
subsurface or above surface sources.
Testing and removal protocols emphasized:
- Thorough sampling'of the yard.
- Adherence to removal safety procedures for Insuring thaffhe removal operation did
not spread contamination via dust or mishandled soil to other areas at the study
residence or neighboring premises.
- Insuring that replacement soil met requirements for low lead content.
The preparation of a site for soil abatement started well before the actual excavation.
The study's abatement coordinator attempted to meet with the property owner. Many yards
were found to have abandoned cars, trash, and other debris which had to be removed before
abatement. This work was done in large part by ACP Cleaning, Inc., the same contractor
who conducted the interior loose paint and dust abatements.
Several different methods were used to verify that the appropriate amount of soil had
been removed. One method involved running a string between two reference points on
objects such as the edge of a sidewalk or a fencepost. By measuring down from the string to
the soil surface before and after excavation a determination could be made as to how much
soil was removed. This method proved adequate for level yards, but it was not practical for
uneven terrain.
In most cases, permanent features of the property were used as reference points.
Before excavation, orange paint was sprayed onto fenceposts, building foundations, and tree
trunks at ground level, and notes were made on existing slopes and hills. In most yards this
worked well, but it was difficult to accurately measure 6" on very uneven yards.
Contractors were urged to err on the side of taking out too much, rather than too little soil.
10-3
-------
In situations where soil abatement involved frozen ground it was often impossible to remove
less than 12" of soil because the soil was excavated as large, thick, frozen slabs.
10.3.1 Subsurface Fabric/Synthetic Barrier
In most situations, the soil at the 6" level still contained significant amounts of lead.
A geotextile fabric barrier made of nonwoven polyethylene and polypropylene material which
is water permeable, very durable, and has the appearance of a thick grey felt was laid
directly on top of the exposed subsurface immediately following removal of topsoil and prior
to placement of clean topsoil. The placement of this barrier served two purposes:
1. It indicated the border between old subsurface and newly applied surface barrier to
determine how well the surface barrier would persist over time.
2. It protected against recontamination of the surface soil by the remaining
contaminated subsurface soil.
10.3.2 Surface Covers
One of the following surface covers were employed:
- 8" of clean topsoil topped with:
- sod
- hardy grass seeding
- bark or mulch where grass would not grow
- gravel, crushed stone or crushed bank in driveways and parking areas, walkways,
and areas susceptible to erosion
- 3" Asphalt (No soil removal or fabric barrier required).
Selection of surface cover for a particular area was based on:
- Appropriateness for site
- Least cost option that was acceptable to property owner
- Ease of maintenance.
Replacement soil was obtained by the contractor Franklin Environmental Services and
was tested by Alpha Analytical Laboratories for lead, 23 other metals, and a number of other
contaminants, such as volatile organic components. Laboratory confirmation was given to
the study staff indicating that the replacement soil lead level was undetectable.
10-4
-------
Where there was sufficient sunlight to support grass, the soil was covered with sod.
There were many yards where sod would not grow well. There were also many unpaved
driveways and paths where the soil had to be abated, but something other than replacement
soil was needed to cover the geotextile fabric. For parts of yards where grass would not
Stow, baric mulch was used. In these cases 6" of clean soil was put down, followed by 4" of
bark mulch. Gravel was used for driveways and heavily travelled walkways. In these cases
2" of clean soil was put down, covered by 6" of gravel.
10*3.3 Soil Abatement Procedures
Lot sizes varied from about 2,000 to 7,500 square feet (including the area occupied by
house and sidewalks, etc.). The lots that were abated in 1989 averaged 199 square meters
(2,141 sq. ft.) and ranged from 12 square meters to 702 squaresmeters. An average of
cubic yards of soil was removed in 1989 (range of 3-168 cubic yards). The lots that were
abated in 1990 averaged 178 square meters (1,918 sq. ft.) and ranged from 26 square meters
to 656 square meters. An average of 44 cubic yards of soil was removed in 1990 (range of
6-182 cubic yards).
Several different soil abatement methods were used on the 36 Study Group properties
abated in 1989. Initially, the soil was loosened with rototillers, then vacuumed into a truck
using an industrial vacuum similar to that used to pick up leaves. The second method was to
use a Bobcat (brand-name) tractor to dig up large areas and shovels for areas with narrow
access. a third method was adopted for digging up properties after the ground had frozen.
This called for jackhammers to loosen the soil and backhoes to remove it. Paving parts of
the property was another option used after the ground froze.
The first method, using the truck-mounted (Supersucker) vacuum was abandoned
*elatively quickly for a number of reasons. The soil had to be gathered into piles, then fed
mt0 the vacuum. At best, this meant handling each shovelful of soil twice. After heavy
ra"ls the soil was wet, requiring extra labor to feed the soil into the vacuum. The machine
Was so big that it could not move around the property, so all the soil from the backyard had
tฐ be taken to the front to be fed into the machine. Rental of the vacuum itself also was
extremely expensive. Six properties, including two double sized properties, were abated
using this method before it was abandoned.
10-5
-------
The majority of properties were abated using the Bobcat tractor combined with hand
labor. The Bobcat was able to lift the soil into a dump truck that had a ten cubic yard
capacity. In areas of the yard which were done by hand, soil was dug out with shovels, then
taken by wheelbarrow to a point where the Bobcat could scoop it up and place it in the truck.
This is an especially useful strategy because it can be adapted to almost any property.
Eighteen properties, including one double sized property, were excavated using Bobcat and
hand-labor. We were forced to abandon this approach when the ground froze in December
of 1989.
The last 12 properties in the Study Group were abated during an unusually severe cold
spell that began in late November and continued through December of 1989. The ground
quickly froze to a depth of over 14", making the use of Bobcats or hand tools impossible.
Jackhammer crews and backhoes were added to the work force.-The work was very slow,
and it became difficult to remove exactly six inches of soil. The backhoe would often
remove a slab of frozen soil 12" thick and ten square feet in area. The workday was
shortened due to the impact of windchill temperatures of -40 degrees Fahrenheit on workers
and equipment.
During this period we offered some property owners the option of having part or all of
their property paved with asphalt. One entire property and parts of three others were
ultimately paved. Since sod could not be planted during this period, grass seed was spread
on the new soil and repeated the following spring.
The general techniques used for soil abatement of properties in Control Groups A and
B in 1990 were similar to those used on the properties in the Study Group. The contractor
used a Bobcat tractor to excavate large areas. Smaller areas were excavated by hand, and
soil was wheelbaraowed to a place where the Bobcat could scoop it up and lift it into a truck.
On some occasions a ro to tiller was used to loosen the soil in preparation for hand digging.
There were, however, two important changes in the soil abatement procedures in 1990.
First, in place of gravel for driveways and walks, a material called "crushed bank" was used.
This mixture of ground stone (or stone dust) and gravel forms a packed surface which,
unlike gravel, is not subject to scattering. It creates an attractive and durable gray gravel-
like surface. A layer 8" deep was spread over the geotextile fabric and packed down with a
compacting machine. This was used extensively to resurface dirt paths and driveways.
10-6
-------
The second important change was that a single contractor did all of the work. .During
the previous year, two contractors were used. One did the excavation and the other did the
landscaping. In the second year, several types of contractors submitted bids, including
landscapes, hazardous waste firms, and deleading firms. The contract was awarded to
Franklin Environmental Inc., of Wrentham, Massachusetts. This company regularly
performs underground storage tank removals, hazardous waste removal and hauling, and
asbestos work. Using one contractor made it much easier to coordinate landscaping and
excavation activities.
The soil abatement schedule required that every property be prepared well in advance
of the commencement of excavation activities. It was clear, from our experience in 1989,
that additional staff would be needed to do advance work and to monitor abatement so three
"site monitors" were hired in August . 1990. They visited properties to be abated and met
with landlords to address the following issues in preparation for soil abatement:
Note presence of:
- Debris blocking access to yards
- Locked gates
- 6 foot access for bobcat bulldozer
- Dogs
- Abandoned cars
- Cars blocking access to yard
- Bad traffic or busy intersections
- Narrow streets
- Access to outdoor water spigot
Also:
- Ascertain owner preferences for sod, crashed bank, or baric mulch
- Plan for access to water if not available outside
- Drop off letter explaining process to owner
- Obtain signed cancellation form if owner did not want soil removal
During the fall of 1990 lead contaminated soil was removed from 58 properties on
which children in the Control Groups resided. The original schedule called for the soil
removal to be done soon after the second follow-up blood sample was obtained and after the
exterior deleading was complete in the Control Groups. This required that two properties be
10-7
-------
scheduled each week. Abatements did in fact proceed very close to original plans, despite
difficulties and unanticipated delays.
Soil removal for Control Groups A and B began in September 1990. Crew size ranged
from 6 to 12 people, with an average of 9. Crews typically included a foreman, who
coordinated movement of materials and people, a truck driver, a Bobcat operator, and five or
six laborers. The contractor worked on as many as four sites at once. Often four sites were
excavated in two to three days, and all landscaped simultaneously in the following two days.
Most sites were completed in one day. The largest sites took two to three days to complete.
The sites varied in size and difficulty. Some sites had to be abated entirely by hand
because there was no access for the Bobcat. This meant using a wheelbarrow to take all of
the contaminated soil out and bring all of the clean soil in. Most sites, however, consisted of
a combination of areas that could be excavated by Bobcat and smaller areas that had to be
excavated by hand.
There were only a few minor delays and the last property was abated on December 11,
1990.
10.3.4 Soil Abatement Safety
A soil abatement health and safety plan was developed to prevent the accidental
dispersal of lead-contaminated soil and to protect workers from lead exposure and accidents
while work was being done.
The same soil abatement health and safety plan was followed in both the 1989 and
1990 soil abatement phases.
Respirators were used by individuals conducting lead contaminated soil abatements only
for the pilot abatements conducted in 1989. During these abatements air monitoring was
conducted by Applied Occupational Health Systems (AOHS), an industrial hygiene consultant
firm from New Hampshire. Air monitors were put on the perimeter of the site and on the
dumpster into which the lead contaminated soil was placed. They were also clipped to the
shoulders of some of the workers. These monitors collected data on the amount of lead dust
escaping from the work area and the level of lead in air at the site. Area and personal
breathing zone air samples were collected by drawing air through 0.8 micron pore size,
37 millimeter diameter mixed cellulose ester filters mounted in closed face cassettes. Air
10-8
-------
was drawn through the filters using MSA or Gillian personal air sampling pumps with a flow
rate of at least 2.0 liters per minute as established by prior and post calibration using the
primary bubble tube method. At the end of the sampling period, the cassettes were capped,
sealed, labeled and hand-delivered to the AOHS, American Industrial Hygiene Association
(AIHA) accredited laboratory (#342), for analysis utilizing atomic absorption spectroscopy
(NIOSH Method 7082).
Monitoring results were compared to the OSHA Permissible Exposure limits (PEL)
and Action Level. The OSHA PEL for lead is 50 mg/m3. The highest time weighted
average exposure found during the pilot abatements was 0.82 mg/m3. Based on these
findings, respiratory precautions were abandoned.
Soil was prevented from becoming airborne by frequent watering using a garden hose
during excavation. This worked well as evidenced by air monitoring results during the first
abatements in 1989. When the ground was very dry, as it was during the first days of the
1990 abatements, the ground needed to be watered for several hours the day before
abatement was scheduled.
Safety measures for preventing soil from being tracked or spilled off site consisted of
establishing work areas, and surrounding the areas with plastic dropcloths. Weather
Permitting, decontamination areas were set up on the plastic, where workers would wash off,
^en remove their boots and tyvek suits. The decontamination area consisted of wading pools
filled with water, scrub-brushes for the boots, and trash bags for the disposal of tyvek suits
worn by the workers.
The waste water from the decontamination pools was poured back into the area that was
to* abated, before the geotextile fabric was in place. Equipment such as shovels, rototUlers,
^d Bobcat bulldozer blades and tires were hosed off in a place where the wash water would
diain back into the work area.
The Health and Safety Plan is included in the Appendix.
*0.3.5 Soil Disposal
Disposal of the lead contaminated soil in a safe and cost effective manner was critical
to the success of the study. This proved to be a difficult task throughout the study.
10-9
-------
Soil disposal was accomplished in accordance with guidelines developed in conjunction
with the Massachusetts Department of Environmental Protection. Lead contaminated soil
was removed to a location to which access was controlled, specifically a granite quarry in the
Hyde Park section of Boston (Barry's Quarry) that abuts and will become an extension of an
existing cemetery. This site was extremely useful because it was very close to the study
properties and was unlikely to result in exposure to the disposed soil.
We began using this site in October 1989. At that time, quarry owners had bulldozers
operating every day and arrangements were made to place soil from the study in a designated
area of the quarry. Each load of lead contaminated soil from the study was covered with
uncontaminated soil from other parts of the quarry to minimize the possibility that children
and other individuals would be exposed to the lead in the abated soil. This site also provided
minimal risk of contaminating water tables as underground wells-are not used and all
drinking water in the Boston area comes from a reservoir far west of the City.
On November 3, 1989, during the soil abatement phase of the Study Group, the City
Councilor from the Hyde Park section of Boston raised concerns about the safety of
disposing of lead contaminated soil in that section of the City. These included questions
about whether the lead contaminated soil would harm children or adults in the area,
contaminate the community water supply, and reduce property values of homes in this
neighborhood. Moreover, the justice and wisdom of taking soil that was believed dangerous
to children from one neighborhood and disposing of it in another neighborhood was also
questioned. This resulted in the project temporarily suspending use of the quany in that
neighborhood for the disposal of soil.
To stay on schedule, it was crucial that we not halt abatements and so it was essential
to use a temporary site for soil storage until the disposal site controversy was resolved.
Although soil was not considered hazardous waste by the Environmental Protection Agency
or by the Massachusetts Department of Environmental Protection, it was extremely difficult
to find an alternate permanent disposal site. Most landfills in the area were closed. The
landfills that were open were unwilling to accept lead contaminated soil because the operators
of these facilities feared that by accepting the soil they would incur penalties under future
regulations. A common concern voiced by the disposal industry is that material legally
10-10
-------
accepted today may be declared hazardous in the future and result in additional costs to
landfill owners.
We resorted to a temporary storage facility until access to the quarry was restored.
The temporary disposal site for soil from 14 properties was a parking lot at the Mattapan
Chronic Care Hospital. Permission to use the site was granted by Boston's Commissioner of
Health and Hospitals. A trench was cut in the pavement and a silt fence installed to prevent
run-off. Plastic drop-cloths were used to cover the soil and prevent dust from being blown
off the site. The parking lot was surrounded by woods on two sides and an abandoned
building on another. It was well within the grounds of the hospital and close to a rarely used
road. The combination of precautions taken to keep the soil in place and the remote location
made this a suitable temporary storage site for the lead contaminated soil.
The controversy was eventually resolved in a series of meetings involving city
councillors, representatives of the neighborhood in which the quarry is located, the study's
principal investigator, the EPA Region I Project Manager, and a representative of the
Mayor's Office. Input was also elicited from the Massachusetts Department of Public
Health, the Massachusetts Department of Environmental Protection, scientists and others
working with the study, and various lead experts not associated with the study. The
unanimous consensus of public officials and lead experts was that there was no danger to
local residents. The concerns of residents and the City councillor were allayed and
permission was granted to resume using the quarry as a disposal site on December 2, 1989.
Soil which had been temporarily stored at the Mattapan Hospital site was moved to the
quarry on December 7 and 8, 1989. Although successfully resolved, this episode occurred
despite substantial efforts at public awareness and community relations that preceded disposal
activities and jeopardized the study for a period of time. These issues may have significant
implications for future lead contaminated soil abatement and disposal efforts in other
communities.
10.3.6 Obstacles to Soil Abatement
Disposal of contaminated soil was clearly the greatest obstacle encountered in the study.
It was not, however, the only difficulty confronted. Listed below are some of the other
10-11
-------
problems that had to be addressed and that may have implications for future lead
contaminated soil abatement efforts:
Narrow streets that were difficult to negotiate with trucks
Narrow access for Bobcats (less than six feet wide)
Bulky trash items in yards
Non-functioning cars in yards
Fences that had to be taken down and replaced to gain access to yards
Availability of tested "clean soil"
10-12
-------
11. INTERIOR AND EXTERIOR PAINT DELEADING
Paint deleading was not included as a study intervention but we strongly recommended
it to homeowners and landlords after all study interventions, environmental sampling and the
second follow-up blood and handwipe sampling were completed. Deleading performed in
conjunction with the study met or exceeded minimum requirements of the Massachusetts
Lead Law. It included the removal of lead paint from chewable surfaces below five feet and
making intact all surfaces above five feet. Specifications written by the study's deleading
coordinator required that, whenever possible, dust generating methods be avoided. The
preferred deleading methods were off-site treatment of surfaces covered with leaded paint and
replacement with new materials. Exterior and common interior Seas of multi-unit housing
were deleaded, as well as the inside of participant's living units.
All contractors were monitored by the study's inspection staff to ensure that proper
safety and health considerations were addressed during the deleading and participant families
vacated the premises during deleading. Dust wipe samples were taken upon completion of
the deleading to confirm that the premises were safe for families to re-occupy.
11.1 PRE-DELEADING PLANNING
Planning the deleading activities began in January, 1990 when the loose paint and dust
abatement interventions were completed. At that time we anticipated deleading as many as
100 units. Because of the logistical complexities that this presented, advice was sought from
numerous sources.
All deleading contractors licensed to work in Massachusetts were invited to attend a
pre-request for bids "brainstorming" meeting on February 14, 1990. Marie Farfel and
Susan Guyeaux who have been involved in research and development of deleading
procedures in Baltimore, Maryland also were asked to attend and offer suggestions. The
scope of work was described, preliminary specifications explored, and feedback obtained
from local contractors. It quickly became apparent that several contractors would be
required to perform the work.
11-1
-------
The most prominent problem was scheduling deleading around soil abatement activities
in Groups A and B so that soil abatement could be completed before the cold weather months
when the ground freezes. It was also necessary to delead the exteriors of buildings before
soil removal in Groups A and B to prevent recontamination. (Group S received soil
abatement the previous year.) These problems led to the decision to develop separate
contracts for interior and exterior deleading activities so that these activities could be
performed independently, thereby preventing delays that could interfere with soil abatement.
It was decided that eight contracts would be needed (four for interior work and four for
exterior). This approach allowed for the soil abatement needs to be satisfactorily addressed
and allowed medium sized as well as larger deleading companies to bid on the activities.
Requests for bids were put out in three phases due to the time involved in preparing
lead paint inspection and bid documents. Each phase ended in competitive negotiations to
bring down the initial bids.
Exterior work included the removal of lead contaminated paint from chewable surfaces
below five feet on siding, porches, rails, stairs, windows and doorways of common areas as
well as the building's exterior surface. Loose paint above five feet was also made intact.
If these areas of the home were deteriorated, or if it was too difficult or hazardous to remove
the lead paint from a surface, items were removed and replaced with new materials of
similar workmanship as other items in the house or neighborhood. Columns with chewable
surfaces were covered or scraped to a height of five feet.
Interior deleading consisted of removing lead contaminated paint from chewable
surfaces below five feet and making intact all loose paint above five feet on walls within the
apartment or housing unit. To minimize deleading hazards, dust generating methods were
avoided whenever possible. Replacement and off site dipping were used whenever possible
although some use of dry scraping was unavoidable.
The following contractors performed the interior and exterior deleading. All were
licensed by the State of Massachusetts to engage in deleading activities:
Action Deleading Paint by Numbers
Point West Plaza P. O. Box 128
21 Torrey Street N. Easton, MA 02356
Brockton, MA 02401
11-2
-------
A. Escalada Painting Co.
633 Ferry Street
Marshfiled, MA 02050
Tri-State Restoration
16 Hazel Drive
Hampstead, NH 03841
Contractor hired by the one owner who chose to hire his own contractor:
Tolan and Sons Deleading
44 Coburn Street
Framingham, MA 01701
Subcontractor to Action Deleading:
Webster Environmental
161 Granite Avenue
Dorchester, MA 02124
11.2 DEVELOPMENT OF INSPECTION PROCEDURES
It was originally planned that an inspector from the Office of Environmental Affairs of
Boston's Department of Health and Hospitals would perform limited inspections of
participants' homes and provide information on the lead paint content in the premises. It was
essential for scientific considerations that we have a measure of children's exposure to lead in
paint and take this lead source into account when we determined the effectiveness of lead
contaminated soil abatement. Because this plan called for inspections to be performed by a
"code enforcement inspector," under Massachusetts law it would require that any unit found
to have lead paint be deleaded. Such an approach, however, was likely to discourage many
families and landlords from participating.
After meetings with representatives of the Massachusetts Department of Public Health,
it was decided that the study would hire private inspectors who were not bound by code
enforcement inspection requirements. The inspection reports generated by these individuals
would be filed with the State's Lead Paint Poisoning Prevention Program, but since the
participating children's blood lead levels were below 25 /zg/dL, the reports would not be
reviewed by state officials and so would not result in mandatory deleading. This allowed the
study to obtain the necessary scientific information without putting participating families or
owners in legal jeopardy.
11-3
-------
In April 1990 private inspectors were hired. Inspection procedures were developed that
were consistent with study needs and legal requirements. It was initially expected that these
inspectors would only perform a one-time inspection in each unit prior to any deleading
activities. It was planned that study case managers would be trained by an industrial hygiene
consulting firm to monitor deleading operations. Final deleading compliance letters were to
be issued by an inspector from the Office of Environmental Affairs after work was
completed. The Massachusetts Department of Labor and Industries, however, refused to
grant a waiver of their deleading regulations to allow on-site monitoring by study case
managers (even though there is no provision in the regulation for monitoring, i.e. the
regulation addresses inspections only). Thus, the only way to provide on-site safety
monitoring was to have the private inspectors do both on-site monitoring and inspections.
The final compliance letters were issued by the inspector who had performed the initial
inspection.
Initial inspections were performed between June and August 1990. Monitoring of
deleading activities took place between August and the end of December 1990.
11.3 DELEADING ACTIVITIES
The 152 children participating in the study lived in 123 housing units on 101 premises.
Deleading or assistance with deleading was offered to all 123 households. Thirty of these
households either moved or refused to have an inspection for lead based paint. Study staff
did not pursue deleading in these cases. In all cases, however, irrespective of whether an
extensive inspection was performed, at least six measurements of the lead content of interior
paint were obtained per unit. Using a PGT (Princeton Gamma Tech) x-ray fluorescence
instrument, the inspector obtained measurements of the lead in the paint on the woodwork
and on one wall in the kitchen, living room, and child's bedroom. If more than one child
lived in the unit, samples were obtained from each child's bedroom. Inspections were
refused generally over concern about the legal obligation to delead the unit if an inspection
revealed the presence of lead based paint. Study assistance with deleading was first offered
at the time families and landlords were recruited to participate in the study. Refusals to have
premises inspected or deleaded, however, occurred throughout the study.
11-4
-------
Ninety-five units were given full inspections for lead-based paint. Six of these units
had previously been deleaded and 32 families refused deleading after the inspection. In an
additional seven cases, deleading was refused after the deleading bids went out. Thus,
deleading was refused for 39 of the 95 inspected units. Reasons for refusal included
unwillingness to prepare for the move and disruptions to lifestyle that the move would entail
(e.g., children's transportation to school and adults' transportation to work).
A total of 92 deleading operations were conducted at 46 premises. Each deleading
operation refers to either the interior deleading of a unit or the exterior deleading activities
associated with that unit. Deleading related work started on August 20, 1990 and all work
that could be considered deleading was completed by December 31, 1990. There were tasks
that the contractor initially believed to be completed but were found to be incomplete when
post-deleading inspections were performed. These and other loose ends brought the
completion date for all deleading related activities to February 14, 1991.
Most deleading involved single units within buildings that had two or more living units.
Four of the addresses involved deleading two units within the same building and four
participating families lived in single family houses. Three non-owner occupied residences
participated in study assisted deleading. There were four units on these three properties.
The study offered to pay up to $ 2,000 towards deleading units in non-owner occupied
households. One owner who had two units at the same address chose to hire his own
deleading contractor to perform the work. The study paid $4,000 directly to the contractor
and monitored the deleading activities in the same manner as the other households in the
study. The remaining two non-owner occupied addresses were deleaded by contractors hired
by the study, as was the case for all owner occupied properties.
Forty-five deleading compliance letters were issued at the end of deleading activities.
Deleading compliance letters are official documents stating that a property is in compliance
with the Massachusetts Lead Law and that previously identified violations have been
rectified. One non-owner occupied address was not issued a deleading compliance letter
because only the building's exterior was deleaded under the study's guidance. Although the
interior of this unit had been previously deleaded, it was done under an older version of the
law. The law does not permit issuance of a certificate for exterior deleading alone and does
not contain a grandfather clause covering work performed under the old law. In three
11-5
-------
additional premises, only exterior deleading was done although we initially had planned to do
interior and exterior deleading at these sites. No deleading compliance letters were issued
for these premises.
In one dwelling, no compliance certificate was issued although both interior and
exterior deleading was completed. On initial inspection of this dwelling access to the
basement was prevented by a locked door. During the final inspection after deleading
activities were completed, this door was found open and was being used as an entrance to
office space created recently by the owners. The inspectors would not issue a compliance
letter because this allowed children access to an area that previously had not been inspected.
The owners must now have this area inspected and, if necessary, deleaded in order to obtain
a certificate of compliance. This is the responsibility of the owners since the study will not
be involved in any further inspection or deleading activities.
Weekly progress meetings were held to arrange and monitor deleading activities.
Contractors, inspectors, and the study's deleading coordinator attended all meetings and
whenever necessary, other members of the study staff attended to discuss issues or problems
that required their attention. This forum was used to provide updated information on all
changes in field activities, schedules, moving issues, etc., and keep inspectors, contractors,
and the project administrator informed of other changes that were required to accomplish
tasks in the necessary order.
11.3.1 Exterior Deleading
Exterior deleading woric: was performed on 46 properties covered under four separate
deleading contracts. This woric included all common interior areas; other buildings on the
properties such as garages; and exterior window sills (except for window sills in participating
families' units which were addressed as part of interior deleading). Deleading certificates of
compliance were issued for 42 of the 46 exterior deleading operations.
Exterior deleading required from 1 to 41 days per property. A total of 1,156 days
were required for the 46 exterior deleading operations. The average duration of exterior
deleading was 25 days per site. These figures include all of the work activities, including
non-hazardous finish work. All four of the exterior contracts were completed within the
scheduled time frame.
11-6
-------
Exterior deleading did not require occupants to relocate. All work was perfonned
without disruption to any of the building's occupants except for requiring alternate access
during the periods that work was being done on stairways. It was the responsibility of the
contractor to check the soil abatement schedule to make sure exterior deleading work did not
wterrupt the soil work. No exterior deleading took place when properties were undergoing
soil removal or lanHsrapifig
Lead painted exterior surfaces were freed of loose or peeling lead paint by chipping and
scraping and were then given a primer coat of paint. Application of finish coats of paint
were the owner's responsibility.
Common hallways in multi-unit buildings were addressed as part of the exterior
deleading contracts. The contractor was responsible for informing other building occupants
of the work activities occurring in these areas and to assure that alternate access rules were
observed. HEPA vacuum units were installed on the first floor at the entrance to buildings.
Containment barriers were set up to make sure that the work area was isolated and that no
contamination spread outside of the work area. A warning sign, as required by the
Massachusetts Lead Law, was affixed to the outside of the containment area entrance. Work
111 the common hallways/staircases began on the top floor and proceeded down to the first
floor level. Deleading was perfonned according to the methods developed and explained in
the study's specifications and were monitored by the study's inspection team to insure
compliance.
U.3.2 Interior Deleading
Interior deleading involved only the inside of the living units of families participating in
the study. Exterior window sills of these units were included in the interior deleading
contract activities. The unit's occupants and all of their belongings were relocated for the
duration of the interior deleading work. A moving contractor was hired by the study to
Groove the occupant's belongings and furniture prior to the deleading contractor's arrival.
AU belongings were fumigated to exterminate insects and placed into storage for the duration
ฐf the deleading. Damages caused by the movers were addressed prior to the final release of
a 15% retainer.
11-7
-------
The moving contractor billed additional charges for items that remained in storage for
over seven days at a rate of $6.00 per household per day. Additional storage per household
ranged from 1 to 43 days and averaged 9.5 days. There were a total of 391 extra storage
days, charged at a cost of $2,346. Forty-one households were moved under the
moving/storage contract; five households did not require moving assistance.
Case managers and contractors occasionally helped families with last minute packing.
In five instances extra charges were incurred when it was necessary to cancel moves since
occupants were unprepared. The cost of these cancellations was $1,800. There was only
one last minute cancellation that was never rescheduled because the occupants would not or
could not pack for moving. In the other four instances of cancellations, moves and deleading
were rescheduled. No compensation was made to the deleading contractor in cases where
sites were eliminated prior to the scheduled start date. The moving contractor, however, did
receive compensation for costs incurred when a move was cancelled or postponed. The total
cost of moving and storage (including additional storage costs and cancellation fees) was
$33,666.
The duration of interior deleading activities ranged from 3 to 58 days per unit,
averaging 15.6 days. A total of 716 days were needed to delead the 46 interior sites. This
included only the time during which occupants were required to be out of their homes.
Interior deleading work included 46 households that were addressed under four separate
contracts. Most work was started on schedule, but in several instances work was not
completed according to schedule. Some of these time overruns were unavoidable.
No penalties were assessed against contractors if the explanations for schedule delays were
reasonable. There were six cases, however, where these delays were avoidable and penalties
in the form of liquidated damages of $1,000 per day were assessed for a total cost of $6,000.
All of the penalties were assessed against the same contractor and only after other attempts
by study staff to rectify problems were exhausted. These penalties were used as a last resort
and only when it was absolutely necessary to maintain the study's best interests. Alterations
of time schedules were rarely permitted, and only with approval of the deleading coordinator.
Interior deleading involved the removal of lead paint from chewable surfaces below five
feet, and doors and windows and other chewable surfaces within the living unit. This was
accomplished by replacement, off-site dipping in paint removing chemical mixtures, and
11-8
-------
scraping. Owners had the option to have items that had ornamental detail taken off site to
have the lead paint removed by dipping. Loose paint above five feet was made intact. The
dipping process did cause some of the older deteriorated materials to separate or dissolve,
but this kind of damage was minimal.
Items that had little or no ornamental detail were replaced with #2 pine. When doors
were dipped, the door jamb was scraped free of lead paint to a height of five feet on site. In
general, this was the only instance where dry scraping was allowed during interior deleading.
Dry scraping or the use of chemical solvents was allowed only when there was an
architectural or structural reason for not removing the material from the site or when it was
required to satisfy the requirements of the Massachusetts Lead Law. When new doors were
installed the pre-hung/hollow-core type was used. This eliminated the need for scraping
jambs. Since the study investigators and staff believe that dry-scraping is an extremely
hazardous process, this type of deleading was kept to an absolute minimum.
All items deleaded off site and replaced, and all items that were deleaded on site were
given a coat of primer paint by the deleading contractor. This was done to make sure that
any fine dust film or residue left on the surface was sealed in. New materials were not
Painted by the contractor as this was the responsibility of the property owner. Similarly,
making the surfaces above five feet intact involved priming only the repaired surfaces. The
entire wall surface was left for painting by the owner after deleading was completed.
Prior to removal of the critical barriers, the contractor was required to HEPA vacuum
and wet wash all surfaces within the containment area. This cleaning process followed a set
sequence beginning with surfaces that were deleaded, then walls and vertical surfaces, then
horizontal surfaces, and finally floors. Critical barriers were then removed and the unit
HEPA vacuumed again. Wood floors were coated with polyurethane
11.3.3 Temporary Housing
The ability of participants to find short-term housing during interior deleading presented
a major obstacle. If the study were to provide temporary housing, these sites would have
had to be deleaded or the study would have had to delead them prior to housing families
there. Deleaded units were very difficult to locate and unavailable for short-term rental,
^he study staff investigated the possibility of deleading several units in exchange for
11-9
-------
temporary housing. This option was ruled out because of insurance, liability, and other legal
reasons. Participants were urged to find their own alternate housing with friends or family.
The study paid for lodging only after families had demonstrated a sincere effort and were
unable to find temporary housing. Study staff provided participants with a list of hotels/guest
houses that would provide lodging. Families made their own lodging arrangements and the
study staff set up purchase orders to handle payment.
Seven families were assisted in finding temporary lodging while deleading was being
performed. Their stays ranged from 1 to 19 nights, and the average length of temporary
lodging paid for by the study was 11 nights. A total of 78 nights of lodging were provided
by the study at a total cost of $11,612. An additional 39 families found alternative housing
without assistance from the study staff.
11.3.4 Damage Control
Pre-existing damage was recorded prior to the commencement of deleading. The
contractor was responsible for giving a written report on the pre-existing damage to the study
site monitor before beginning work. Pre-existing damage that was uncovered after the work
had started was brought to the attention of the study site monitor and recorded in the
monitor's daily log.
Study site monitors were on each site daily when deleading activities were being
conducted to assure that deleading activities were done safely with minimum damage. Due
to the nature of the work, however, some damage was inevitable. This was understood by
participating families and owners beforehand, and it was understood that certain corrections
would be the responsibility of the property owner.
In order to avoid damage and excess ripping of wallpaper the contractor was required
to cut a seam between door and window casings and wallpapered surfaces before removing
doors or windows. There was only one case where wallpaper was torn during the removal
of a window casing. There was no conflict since the owner was aware of possible damage.
Owners were requested to remove telephone and electrical cords that came into contact
with lead painted surfaces addressed as part of the deleading, before the scheduled start date.
If lines were left in place the contractor took appropriate action to work safely around these
areas. However, there were several cases where telephone lines left in place were cut or
11-10
-------
broken. The contractors were not held responsible for repairs except when there was
obvious lack of consideration when they were removed.
Any damage to walls was corrected by the contractor by filling with joint compound
and priming. All finish painting was the property owner's responsibility. Any damage that
occurred through neglect or carelessness of work crews was corrected by the responsible
party. In cases where there was a dispute as to where responsibility lay, the study site
monitors and deleading coordinator determined what course of action was appropriate.
All work sites were strictly monitored by on-site study representatives who interrupted
or redirected work for reasons of safety or requested corrections according to the
specifications.
11.3.5 Clearance Sampling
Massachusetts requires clearance sampling for dust lead levels after interior deleading
in situations where dust is visible. The State standard for acceptable dust lead levels after
deleading is 200 fig/square foot on the floor, 500 /ig/square foot on window sills, and
800 fig/square foot in window wells. When no dust is visible to the inspector, clearance
sampling is not required.
The study specifications required that visible dust be removed completely before the
work area was considered ready for clearance sampling. Clearance sampling was required
for every unit deleaded in conjunction with the study. The study did clearance sampling in
two rooms on each floor of each deleaded interior unit and in the common hallway areas.
Moreover, the study insisted that all clearance samples meet the lead levels indicated in the
Massachusetts Lead Law before the deleading operation was considered complete.
Samples were taken by lead inspectors from the floor, window sill, and window well
from each room sampled after contractors informed the inspectors that deleading was
complete. Inspectors did not inform contractors where the samples were going to be
obtained. The samples were obtained by wiping surfaces with commercially available "Wash
and Dries". Inspectors wore disposable gloves, which they changed between samples, and
they wiped one square foot on the floor and wiped measured window sills and wells.
Blanks were included with each set for quality control puiposes. In order to improve the
efficiency of sample preparation, study staff added hydrochloric acid to the samples to start
11-11
-------
the 15 hour digestion period. Analysis was performed by the Lead Lab at Boston City
Hospital using atomic absorption spectrophotometry (spectroscopy). By having the study
staff start the digestion process, the turn around time involved in preparation and analysis
was reduced by at least one day per unit.
Due to cost constraints, the original plan was to do clearance sampling only after
deleading was believed to be completed. However, pre-deleading clearance sampling was
performed in the interiors of 17 units to provide data for pre and post-deleading comparisons
Post-deleading clearance sampling of the 46 household interiors revealed that 32 (70%)
of the households had acceptable dust lead levels without additional clean-up. Fourteen
(30%) of the households were found to have unacceptable dust lead levels and required a
second cleaning and an additional set of samples. Two of these 14 required a third cleaning
before acceptable lead levels were obtained.
The failure rate of final wipe samples was very high despite the fact that the sites
appeared clean by visual inspection. All deleading contractors working for the study were
monitored during deleading activities and wipe samples were not taken unless areas appeared
clean of dust and dirt. This, along with limiting dry scraping to areas where it was
absolutely necessary, should have provided a work area that was as free of lead contaminate^
dust as possible. This highlights some of the dangers associated with interior deleading
activities and raises questions about the adequacy of visual inspections for dust post interior
deleading.
Clearance sampling was performed in conjunction with the final inspection activities
that occurred at the completion of interior deleading activities. It took approximately one
hour per site to obtain the samples and begin the digestion process.
11-12
-------
12. SCHEDULE OF ACTIVITIES
The schedule of activities is shown on Tables 12-1 and 12-2. Table 12-1 gives the
timetable of activities by month. The actual dates on which activities began and ended are
shown on Table 12-2.
12-1
-------
TABLE 12-1. TIMETABLE OF ACTIVITIES
1989 1990 1991
Jan Fab Mar Apr May Jun Jul AuoSaptOct Nซ Dec Jan Fab Mar Apr May Jun Jul Aug Sapt Oct Nov Dae Jar Fab Mar Apr May Jun Jul Aug Sapt
Identification of Study Pop
Eligibility Assessment
Enrollment
Pre-Abatement Detafled Soil Sampling
Pre-Abatement Dust Sampling
Pre-Abatement Blood and Hand Lead Tests*
Soil Abatement (Study Group)
Dust Abatement (Study Group, Control Group A)
loose Paint Abatement (AD Groups)
India! Interview
Water Sampling
Lead Paint Determination
Follow-Up:
Blood Lead Testa
Hand Lead Tests
Interview ....
BeoontamhmHon Assessment
Sofl
Dust ....
Soli Abatement
(Control Groups A and B)
Deleadlng
Data Clean Up and Analysis
Final Report Preparation
-------
TABLE 12-1. TIMETABLE OF ACTIVITIES
1989 1990 1991
Janfab Uatt* May Jun Jul AugSaptOd New Dae Jan Fab Mar Apr May Jun Jul.Aug 8apt Oct NovOaeJan Fab Mar Apr May
Identification of Study Pop
EHgibiHty Assessment
Enrollment
Pre-Abatement Detailed Sol Sampling
Pre-Abatement Dust 8ampฎng
Pre-Abatemert Blood and Hand Lead Tests*
Soil Abatement (Study Group)
Dust Abatement (Study Group, Control Group A)
Loose Paint Abatement (AR Groups)
InWal Interview
Water SampRng
Lead Paint Determination
FoBoaMJp:
Blood Lead Tests
Hand Lead Tests
Intordow
Reoontanriradon Assessment
So8
a
Dust
SoB Abatement
(Control Groups A and B)
Detoadhg
Data Clean Op and Analysts
Final Report Preparation
-------
TABLE 12-2. STARING AND ENDING DATES FOR
INTERVENTIONS AND SAMPLING
Started On
Ended On
Baseline Blood Sampling
9/14/89
1/8/90
POST1 Blood Sampling
3/19/90
6/13/90
POST2 Blood Sampling
7/19/90
11/17/90
Baseline Handwipe Sampling
9/13/90
12/20/89
POST! Handwipe Sampling
3/20/90
7/3/90
POST2 Handwipe Sampling
7/18/90
1/3/91
Preliminary Soil Sampling
4/4/89
11/18/89
Pre-Abatement Detailed Soil Sampling
8/4/89
6/20/90*
Soil Abatement (Study Group)
9/18/89
12/26/89
Soil Abatement (Control Groups A and B)
9/11/90
12/11/90
Soil Recontamination Sampling, Round #1
6/11/90
7/13/90
Pre-Abatement Interior Dust Sampling
8/17/89
1/16/90
Post-Abatement Interior Dust Sampling
10/6/89
3/9/90
Dust Recontamination Sampling, Round
3/27/90
6/14/90
Dust Recontamination Sampling, Round #2
7/2/90
12/15/90
Interior Dust Abatement
10/2/89
1/30/90
Interior Loose Paint Abatement
10/2/89
1/30/90
Water Sampling Round #1
2/8/90
7/23/90
Water Sampling Round #2
7/18/90
2/11/91
XRF Lead Paint Determinations
4/10/90
1/7/91
Initial Questionnaire Administration
2/2/90
.5/2/90
First Follow-up Questionnaire Administration
7/19/90
3/21/91
Paint Deleading
8/28/90
1/4/91
*
Approximately 80% of the detailed soil sampling was completed by 1/24/90. 23 soil samples were taken
between 1/24/90 and 6/20/90. Approximately 90% of the detailed soil sampling was completed by 5/25/90.
12 soil samples were taken between 5/25/90 and 6/20/90.
12-4
-------
13. DATA COLLECTION AND MANAGEMENT
The study generated and maintained data from the soil, dust, water, paint, blood lead,
and hand lead analyses. In addition, information collected from two sets of questionnaires
was collected and analyzed. Data collection instruments were developed to gather and code
all information.
Quality control and assurance measures were performed at each step of data collection.
These included the following measures:
1. All original paper forms were stored by premises identification number or in
chronological order in file cabinets enabling easy access and retrieval.
2. Original report sheets from ESA, EPA, Dennison, Bioran, and Hall- Kimbrell
laboratories were stored separately.
3. All completed forms and questionnaires were reviewed manually for accuracy and
completeness and any questions and problems were resolved in an ongoing manner.
4. The standard data entry validation tools (range checks, picture formats etc.) were
used for all data sets created through data entry.
5. The quality of identifiers was assured through table-lookup at data entry. The
principal identifiers were validated against a table of valid values and invalid values
were rejected. Valid values called up additional identifying information (name,
address, etc.) to verify a correct match.
6. Up-to-date source coding listings and coding manuals of all database files structures,
programs and documentation was maintained and available for easy access.
7. 100% verification of all data entered by the study's data entry cleric was conducted
visually.
8. There were daily and weekly backups of all important files as well as biweekly
archiving of all important files in the database format.
9. There was periodic inventory of all collected data.
13-1
-------
The study data base consisted of many data files which were either created by data
entry using the Paradox database management system, received from the EPA Region I
Laboratory as Lotus files, or received from the Boston Childhood Lead Prevention Program
(BCLPPP) as Dbase3 + files. All received files were imported into Paradox for data
correction. All Paradox files were converted into Statistical Analysis System (SAS) data sets
for data management and analysis.
The study had two distinct data collection phases, the "eligibility" phase and the
"study" phase, and each phase required its own set of data files. During the eligibility
phase, information pertaining to criteria for study enrollment was collected in order to
identify subjects who were willing and eligible to participate in the study. Computer files
generated during the subject recruitment period were used for study management and for
descriptive analyses of the non-participants. When the study phase began, data pertaining to
participants were transferred from the eligibility data files to the study files. The eligibility
data files were archived.
Data were collected about four separate units of observation: child, family, unit
(premises and apartment concatenated) and premises. Some data files contained repeat
measures type data, i.e. the same set of data items for the same unit of observation collected
at different times. For instance, blood lead test results were collected three times for each
child. The goal of the organization of the data base was to make it possible to (1) easily
match a child to the data that apply to bis premises or apartment, and to (2) easily match
information pertaining to a stage of a study (e.g., pre-abatement, post-abatement, etc) across
all files with repeated measures data.
The central file was the KID file which provided the means by which data from
different files could be combined to form composite case records. The unit of observation
for the KID file was a child and each observation contained all the identifiers for that child
(child id, family id, premises id, unit id). Any files that did not have identifiers in common
were merged through the KID file.
Repeated measure data contained a variable called PHASE that designated the phase of
the study in which the test was done. By selecting test results based on values of PHASE,
data from different stages of the study can be compared.
13-2
-------
A more detailed description of the data management plan including a database
configuration and file descriptions was also developed and is included in the Appendix.
13-3
-------
14. DATA ANALYSIS
The main puipose of the data analysis was to test the hypothesis that a reduction of at
least 1,000 PPM in the concentration of lead in soil with starting concentrations greater than
1,500 PPM results in at least a 3 /tg/dL reduction in children's blood lead levels over the
following year. First, we conducted crude analyses of the change from baseline blood lead
levels (i.e., before any abatement activities) to the first and second post-abatement blood lead
levels obtained an average of six and eleven months, respectively, after the abatement
activities. We used analysis of variance to compare mean blood lead changes among the
intervention groups and paired t-tests to determine whether mean changes in blood lead levels
within the intervention groups were significantly different from zero.
Following the crude analyses, we used analysis of covariance to compare the
intervention groups with respect to post-abatement blood lead levels adjusted for pre-
abatement blood levels. This was necessary because of slight differences in the baseline
blood lead levels of children in the three groups. The post-abatement blood lead levels were
reasonably normally distributed and did not require any transformations. The base model
that we used to obtain estimates of adjusted post-abatement blood lead means in the
Intervention groups was:
= + b,Zu + bjZa + + ej
where for the ith child,
Yj = post-abatement blood lead level
Zn = 1 if in Control Group A, otherwise 0
Zjj = 1 if in Control Group B, otherwise 0
Xj = pre-abatement blood lead
ej = error term
The coefficients, b0, blt t>2, and bj were estimated using least squares methods, and
t-tests were used to test the null hypothesis that bj and t>2C were equal to zero (i.e., Was the
mean adjusted post-abatement blood lead level in each Control Group different from that of
Study Group?).
14-1
-------
Potential confounders of the relationship between group assignment and post-abatement
blood lead were added to the base model one at a time to obtain estimates of the group effect
adjusted for baseline blood lead level and the potential confounder. More complex models
that controlled for several variables simultaneously were also developed. Potential
confounders included age, sex, race, socioeconomic status as measured by the Hollingshead
Index, mouthing and handwashing behaviors, and environmental sources of lead (e.g., paint
and water). In most instances, the variables were categorized; cutoffs were based on the
frequency distribution of the particular variable or on external considerations (e.g.,
regulatory standards for environmental sources of lead). In some instances, variables were
combined before being added to the base model. For example, the following set of variables
were developed to describe mouthing behaviors: pacifier use (yes/no), thumb sucking
(often/sometimes/rarely/never) and a count of the number of times any other mouthing
behaviors were reported at interview (zero/one/two/three-five).
There were several reasons why we decided to use analysis of covariance (ANCOVA)
with baseline Pb as a covariate to estimate the differences between groups instead of analysis
of variance (ANOVA) with the absolute change in Pb as the dependent variable. Consider
the following models:
ANCOVA Model Y = a, + /3X + E
ANOVA Model W = 7, + E
where Y = post treatment Pb,
X = baseline Pb,
and W = Y - X.
The relationship between the ANOVA and ANCOVA models depends on whether you
assume a fixed effect model or a mixed model for the ANOVA. The ANCOVA model may
be rewritten as a fixed effect ANOVA model if/3 = 1.
Y = ai + 0X + E
Y - X = ai + (0-1) X + E
W = ai + (0-1) X + E
14-2
-------
The ANCOVA model does not force the slope of the regression to be equal to one;
rather, it allows the slope to be estimated from the data. If the slope is, in fact, equal to
one, the two models are equivalent. Samuels22 shows that the ANOVA mixed model is a
special case of the ANCOVA model, with 0 = p, the correlation between Y and X.
In either case, the ANOVA model provides a less powerful test of treatment differences
than does the ANCOVA model. It is recommended only if there is a large imbalance in
baseline means, in which case the ANCOVA model may not be valid.22,23 In our study the
mean pre-abatement blood lead level was higher among children assigned to the Study
Group; however, this difference was not statistically significant.
An important assumption of the ANCOVA model is that the slopes of the regression
lines are equal in the treatment groups. We found this to be the case here, with the
interaction term not significantly different from zero (p>0.10).
The data analysis was conducted using SAS statistical software. The statistical methods
used (t-test, analysis of variance and analysis of covariance) are described in the SAS
manuals and standard statistics text books.
14-3
-------
15. RESULTS
15.1 BLOOD LEAD LEVELS
15.1,1 Crude Analysis
Table 15-1 describes the results of cnide analyses that examined the change in blood
lead levels among all participants following the abatement activities. Mean blood lead levels
in all the intervention groups declined at the first post-abatement sampling round (POSH)
and rose at the second post-abatement sampling round (POST2) although for no group did the
mean return to the baseline. At POST1 the average blood lead decline was 2.87 ugldL in
the Study Group, 3.52 /*g/dL in Control Group A, and 2.04 /*g/dL in Control Group B. All
declines were significantly different from zero. At POST2 the average blood lead level
increased 1.39 /tg/dL in the Study Group, 2.69 /xg/dL in Control Group A and 1.52 /ttg/dL in
Control Group B. The increases in the two Control Groups were significantly different from
zero but the increase in the Study Group was not (p=0.08).
Two siblings in the Study Group became lead poisoned sometime between the POST1
and POST2 sampling rounds. Their blood lead levels were 19 #ig/dL and 12 /*g/dL at
baseline (September 1989), 10 /xg/dL and 17 pg/dL, respectively, at POST1 (March 1990)
and 35 /xg/dL and 43 /tg/dL, respectively, at POST2 (July 1990). No other children in any
group experienced a blood lead rise of this magnitude during the course of the study. In
fact, these two children's POST2 blood lead levels were more than three standard deviations
higher than the overall mean POST2 level. Figure 15-1 depicts the relationship between
PRE and POST2 blood lead levels for all children and visually demonstrates that these two
children were outliers. Since the elevated levels were detected many months after the
abatement activities, we do not believe that the increases were related to the study
interventions. We hypothesize that the siblings were exposed to another source of lead,
probably leaded paint at another site, and have information from parent reports about their
exposure to renovations of an apartment containing lead contaminated paint to support this
hypothesis.
Therefore, with the approval of the EPA project officer and a consultant statistician, we
excluded these two children from subsequent analyses. Table 15-2 describes the blood lead
15-1
-------
TABLE 15-1. CRUDE CHANGES* IN BLOOD LEAD LEVELS
AMONG ALL PARTICIPANTS
STUDY PHASE
STUDY GROUP
CONTROL
GROUPA
CONTROL
GROUP B
Pm-AbctMTMnt
(Sept "89-Jan. *00)
13.18-1
(N-54)
Post-AbaMnwnt
POST1
(Mar. <90 Junป *90)
10.31-
(N-54)
POST2
(July '90 - Nov. 90)
12^37-,
(N-61J
-2.87
p-0,0001
8.86
(N-48)(
+1.39
p-o.oe
11-TO-J
(N-54)
12.02-!
(N-47)
352
p-0.0001
8.83 -
(n-46)
+2.69
p-0.0001
(N-49)
11JS-J
(n-46)
2.04
P-0.0001
+1.52
P-0.0001
distributions over time and Table 15-3 describes the average change in blood lead levels with
the two lead poisoned siblings excluded. Without these children, the mean blood lead level
in the Study Group increased by only 0.46 ng/dL at POST2. This increase was not
significantly different from zero (p=0.31). Figures 15-2 and 15-3 depict the blood lead
changes graphically.
Because the PRE and POST2 sampling rounds are most closely matched on season, we
focused subsequent analyses on this comparison. The mean decline in blood lead was
2.44 (ig/dL in the Study Group (p=0.001), 0.91 /xg/dL in Control Group A (p=0.04) and
0.52 ng/dL in Control Group B (p=0.31). The mean blood lead level of the Study Group
declined 1.53 pg/dL more than that of Control Group A (95% Confidence Interval: -2.87,
-0.19) and 1.92 (ig/dL more than that of Control Group B (95% Confidence Interval: -3.28,
-0.56). Over the course of the study, behavioral changes including more frequent
houseeleaning and handwashing were similar among the groups and so do not explain these
15-2
-------
Plot of BLD_PST2*BLD_PRE. Legend: A -1 obs, B ป 2 obs, etc.
BLD_PST2
45 --
40
35
30
25
20
15--
10
E
A
B
A
C A
A
A
A
A
B
A
C
A
C
B
B
A
B
B
A
A
A
A
A
C
B
A
A
A
B
A
C
B
A
A
A
A
A
A
B
B
B
B
B
A
A
A
A
C
A
B
C
E
A
A
C
B
A
A
A
B
C
A
B
A
B
C
B
A
A
C
B
B
B
A
A
A
A
A
A
C
A
A
I I I I I I I I I I I I I I I I I
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
BLD_PRE
Figure 15-1. Relationship Between Pre And Post2 Blood Lead Levels.
15-3
-------
TABLE 15-2. BLOOD LEAD 0*g/dL) DISTRIBUTION OVER TIME AND
ACCORDING TO GROUP EXCLUDING CHILDREN
WHO BECAME LEAD POISONED
S
A
B
Total
(N=52)
(N=51)
(N-47)
(N=150)
PRE
Minimum
7
7
7
7
25%
9.5
9
9
9
50%
13
12
12
12
15%
16
15
14
15
Maximum
22
23
22
23
Mean
13.10
12.37
12.02
12.51
Standard Deviation
4.36
4.26
3.71
4.13
POST1
(N=52)
(N=48)
3
II
t*
w
(N=146)
Minimum
2
3
4
2
25%
6
6
8
6
50%
10
8
9
9
75%
13
12
12
12
Maximum
22
17
18
22
Mean
10.19
8.85
9.83
9.64
Standard Deviation
4.63
3.79
3.49
4.04
POST2
(N=52)
(N=49)
(N=46)
(N=147)
Minimum
4
3
5
3
25%
7
9
8
8
50%
10
11
11.5
11
75%
14
14
14
14
Maximum
22
20
20
22
Mean
10.65
11.49
11.35
11.15
Standard Deviation
4.04
3.94
3.65
3.88
15-4
-------
TABLE 15-3. CRUDE CHANGES* IN BLOOD LEAD LEVELS EXCLUDING
CHILDREN** WHO BECAME LEAD POISONED
-------
P0ST1 P0ST2
-4
Figure 15-2. Crude Change in Blood Lead Levels Excluding Children Who became
Lead Poisoned.
Group. The age distribution varied across groups, however, the average age of the children
was quite similar. The proportions of subjects classified in the lowest socioeconomic level
according to the Hollingshead Index (Classes 4 and 5) were higher in the Study Group and
Control Group B than Control Group A. However, the proportions of owner occupied
premises and participant owned units were similar across groups.
Overall, median lead levels in preliminary surface soil samples were approximately
700 PPM higher than median levels in detailed surface samples taken throughout the yard.
We believe that this is because the preliminary samples were taken closer to the house.
Median surface soil lead levels were also about 800 PPM higher than those taken at a depth
of 15 centimeters. Median interior floor dust lead levels were generally similar to the
median surface soil levels while median window well dust lead levels were five to seven
15-6
-------
5
0
25
20
15
10
5
0
i
25
20
15
10
5
0
POST2 Pb (ng/dl)
Group S
10
POST2 Pb (ng/dl)
Group A
10
POST2 Pb (ng/dl)
Group B
Note: Diagonal indicates no change from PRE to POST2
ts of PRE and POST2 Blood Lead Levels According to Group
eluding Children Who Became Lead Poisoned.
15-7
-------
TABLE 15-4. ADJUSTED DIFFERENCES IN BLOOD LEAD LEVELS STRATIFIED
BY PRE-ABATEMENT BLOOD LEVELS**
Pre-Abatement Blood Lead Levels
S-A
S-B
<15 (N = 105)
-1.35
-1.69
S15 (N = 45)
-0.98
-0.72
<12 (N = 69)
-1.43
-1.47
2:12 (N = 81)
-1.17
-1.46
<10 (N = 44)
-0.99
-1.39
ฃ10 (N =106)
-1.35
-1.31
<10 (N = 44)
-1.00
-1.39
10-14 (N = 61)
-1.57
-1.70
<ฃ15 (N = 45)
-0.98
-0.72
Adjusted for pre-abatement blood lead level.
Excludes two children who became lead poisoned.
TABLE 15-5. DISTRIBUTION OF CHILDREN, FAMILIES, UNITS, AND PREMISES
ACCORDING TO GROUP IN THE FINAL STUDY POPULATION
GROUP
S
A
B
TOTAL
NUMBER OF:
Children
52
51
47
150
Families
43
43
39
125
Units*
42
42
38
122
Premises
34
36
30
100
*Units consists mainly of apartments in multi-unit dwellings.
times higher. Soil and floor dust lead levels were similar across the intervention groups.
Window well dust lead levels were more variable across the groups but the differences were
not statistically stable.
15-8
-------
TABLE 15-6. CHARACTERISTICS OF FINAL STUDY POPULATION
Study Group
Control Group A Control Group B
Soil, Dust, Loose
Dust, Loose
Loose Paint
Paint Abatement
Paint Abatement
Abatement
Total
(N=52)
(N=51)
(N=47)
(N 150)
^re-Abatement Blood
13.1
12.4
12.0
12.5
Lead Level (Mean)
Pre-Abatement Ferritin
25.7
24.6
26.6
25.6
Level (Mean)
Age in Months at Baseline
*9-24
34.6
21.6
25.5
27.3
* 25-36
26.9
45.1
29.8
34.0
*37-51
38.5
33.3
44.7
38.7
Age (Mean)
30.5
31.4
33.1
31.6
* Male
59.6
49.0
51.1
53.3
* Black
42.3
50.0
63.0
51.4
* White
7.7
8.3
4.4
6.9
* Hispanic
26.9
10.4
6.5
15.1
* Cape Verdean
19.2
16.7
15.2
17.1
* Other
3.9
14.6
10.9
9.6
X Class 1-3 SES*
30.0
54.2
26.1
36.8
* Class 4, SES
26.0
14.6
34.8
25.0
X Class 5, SES
44.0
31.3
39.1
38.2
X Owner Occupied Premises^
73.5
86.1
70.0
77.0
X Participant Owned Units***
25.6
29.3
29.0
27.9
Soil Lead Level (Median PPM,
2,722
3,163
3,111
2,904
Preliminaiy Sampling)
Soil Lead Level (Median PPM,
Detailed Sampling}**
At Surface
2,074
2,230
2,100
2,152
15 cm Depth
1,374
1,244
1,348
1,348
Dust Lead Levels (Median PPM)***
Floor
2,651
2,513
2,542
2,547
Window Wells
11,815
15,907
13,429
13,832
Water Lead Levels
(Median ngTL)***
14.8
14.7
22.0
17.0
Bunt Lead Levels (% Undetectable)***
Wall
30.6
47.4
. 20.0
33.7
Woodwork
5.1
0.0
11.1
5.2
^According to Hollingshead Index.
t Unit of analysis ia the premises.
Unit of analysis is the housing unit or apartment.
15-9
-------
Median first flush tap water lead levels were all above 14 ^tg/L and were similar across
the groups. Lead-based paint was detected in almost all homes and was more likely to be
detected on woodwork than walls. XRF readings were similar among the groups.
We also examined the final study population with respect to the calendar months and
time interval between children's PRE, POST1 and POST2 blood samples (Table 15-7). The
pre-abatement blood sampling round lasted 97 days from September through December
1989. A greater proportion of subjects in Control Group B had their pre-abatement blood
drawn towards the end of the round (November and December 1989) than did the Study
Group and Control Group A (42.5 vs. 23.1 and 25.5%). The POST1 abatement sampling
round lasted 86 days from March through June 1990. A larger proportion of Control Group
A was sampled in April compared to the other two groups. The POST2 sampling lasted
121 days and began and ended earlier in the year than the pre-abatement sampling (July
through November 1990). However, the mean number of days between PRE and POST1,
POST1 and POST2, and PRE and POST2 samples were similar across the groups.
Finally, we examined mean PRE, POST1, and POST2 blood lead levels according to
the calendar month of blood sampling (Table 15-8). Mean PRE blood lead levels varied
little by month of sampling (12.3 to 12.6 /xg/dL). Mean POST1 blood lead levels did vary
by sampling month (8.6 to 11.3 /xg/dL); the lowest mean level was seen in April
(8.6 /xg/dL). Mean POST2 blood lead levels increased slightly from August through October
(11.0 to 11.9 ^g/dL), the months when most of the sampling occurred.
15.2.1 Adjusted Analyses
la the analysis of covariance the intervention groups were compared with respect to
post-abatement blood lead levels adjusting for pre-abatement blood levels using a "base
model" that included only group variables. Potential confounding variables, described in
Table 15-9, were then added to the base model one at a time to obtain adjusted estimates of
the group effect.
The POST2 blood lead levels adjusted for baseline level were generally similar to crude
levels (Base Model, Table 15-10). The adjusted mean difference between the Study and
Control Groups were slightly diminished (columns S-A and S-B) but remained statistically
significant. 95% confidence intervals for S-A and S-B were -0.17 to -2.39 and -0.35 to
15-10
-------
TABLE 15-7. DISTRIBUTION (%) OF CALENDAR MONTHS AND MEAN
INTERVAL BETWEEN BLOOD SAMPLES
Groups
Pre Abatement*
September 1989
October 1989
November 1989
t>ecember 1989
42.6
30.8
13.5
9.6
47.1
27.5
5.9
19.6
B
36.2
21.3
25.5
17.0
Post-Abatement
Post 1
March 1990
April 1990
May 1990
June 1990
42.3
44.2
11.5
1.9
31.3
56.3
12.5
0.0
50.0
39.1
4.3
6.5
Post 2
July 1990
August 1990
September 1990
October 1990
November 1990
0.0
34.6
36.5
26.9
1.9
2.0
44.9
22.4
26.5
4.1
2.2
39.1
23.9
32.6
2.2
**.
Mean Number (SD ) of Days Between
PRE and POST1
POST1 and POST2
PRE and POST2
175 (30)
158 (24)
333 (28)
177 (38)
151 (33)
329 (24)
167 (33)
158 (29)
325 (21)
^All pro-abatement blood samples were taken before soil abatement occurred on &e premises.
SD = Standard Deviation.
15-11
-------
TABLE 15-8. MEAN BLOOD LEAD LEVEL ACCORDING TO
CALENDAR MONTH OF SAMPLING
Number in Mean Blood
Category Lead Level (/xg/dL)
Pre Abatement*
September 1989 65 12.6
October 1989 40 12.4
November 1989 22 12.5
December 1989 23 12.3
Post-Abatement
Postl
March 1990 60 10.4
April 1990 68 8.6
May 1990 14 10.9
June 1990 4 11.3
Post?
July 1990 2 10.1
August 1990 58 11.1
September 1990 41 11.0
October 1990 42 11.9
November 1990 4 6.5
-2.62, respectively. Group assignment was a significant predictor of POST2 blood lead
levels (p=0.02). In other words, which group a child was in (S, A, or B) was a significant
determinant of their POST2 blood lead level. The results were quire similar when the blood
lead levels were log transformed.
The results were similar when only one child randomly chosen from each family was
included in the analysis (18.4% of families had more than one child). The differences
between the Study Group and Control Groups A and B were -1.31 (p=.04) and -1.73
(p~ .01), respectively. When the analysis was limited to the first child initially identified on
15-12
-------
TABLE 15-9. ADJUSTED ANALYSIS: DESCRIPTION OF VARIABLES
ADDED TO THE BASE MODEL
Variable
Categories
Age (In Months at Baseline)
Gender
Socioeconomic Status
(According to the Hollingsbead Index)
Race
Mouthing Behaviors
Pacifier Use
Thumb Sucking
Number of All Other Mouthing
Behaviors Reported at the First
Follow-Up Interview
Spends Time Away From Home
Spends Time Outside of Study Area
Kays in Home Yard
Eats Food Outdoors
Hays or Sits on Floor Inside Home
(M Hours Per Day)
Handwashing
Before Meals and Snacks
After Playing Outdoors
Pets in Household that go Outdoors
Canned Food Intake
Number of Canned Food
Items Eaten in Last Six Months
Imported Canned Food Eaten
Ferritin Level
Lead Jobs Among Household Members in Last
Year
Cigarette Smoking Among Household Members
Lead Hobbies Among Household Members in Last
Year
Paint Lead Variables
Maximum XRF Reading on Wall
Maximum XRF Reading on Woodwork
Number of Places Lead Paint Detected
Amount of Interior Chipping Paint at Baseline
Water Lead Level
(Maximum Lead Ingested from Water,
Derived from Water Lead Concentration
and Daily Intake)
Owner Occupied Premises
< =30 Months / > 30 Months
Male/Female
Continuous Variable
Black/White/Hispanic/Cape Verdean/Other
Yes/No
Often/Sometimes/Rarely/Never
Zero/One/Two/Ihree-Five
Yes/No
Yes/No
Yes/No
Yes/No
< =One / Two-Four / > Four
Almost Always/Sometimes or Almost Never
Almost Always/Sometimes or Almost Never
Yes/No
Zero-Seven
Yes/No
< = 15 ng/ml / > 15 ng/ml
Yes/No
Not Detectable/0.5-9.9/10.0
Not Detectable/0.5-9.9/10.0
Zero-Six
0-50/51-200/ > 200 square inches
Not Detectable-6.0/6.1-24.9 /> =25.0
Yes/No
15-13
-------
TABLE 15-10. CRUDE AND ADJUSTED CHANGES IN BLOOD LEAD LEVELS*
Crude
**
Base Model
Base Model
Plus Age
Base Model
Plus Gender
Base Model
Plus Socioeconomic Status
Base Model Plus Race
Base Model
Plus Mouthing Variables
Base Model
Plus Paint Lead Variables
Base Model
Plus Chipping Paint Variable
Base Model
Plus Water Lead Level
Base Model
Plus Time Away From Home
Base Model
Plus Time Away From Study Area
Base Model
Plus Yard Play
Base Model
Plus Outdoor Eating
Base Model Plus Play
or Sit on Floor Inside
Base Model Plus Hand Washing
Before Meals
After Outdoors
Base Model
Plus Pets That go Outdoors
SAB
POST2 Blood Lead Levels
S-A1
S-B2
Overall3
Group
Effect
10.6S
11.49
11.35
-1.53
-1.92
10.26
11.54
11.74
-1.28
-1.49
.02
10.21
11.57
11.83
1.35
-p=,02
1.61
-p=.0l
.01
10.21
11.55
11.74
-1.34
p=.02
-1.53
p=.01
.02
10.30
11.42
11.80
-1.12
p=.06
-1.50
p=.01
.03
10.52
11.44
11.78
-0.92
-1.27
.09
10.80
12.04
12.31
-1.23
p=,04
-1.51
p=.02
.04
10.54
11.72
11.88
-1.19
-1.34
.05
10.23
11.55
11.77
-1.31
p=.02
-1.53
p=.01
.02
10.15
11.34
11.59
-1.20
p=.04
-1.44
p=.02
.03
10.23
11.47
11.70
-1.25
p=.03
-1.47
p= .01
.03
9.77
10.94
11.25
-1.17
p= .04
-1.48
P=.01
.03
10.33
11.60
11.85
-1.27
p=.03
-1.51
p=.01
.02
10.25
11.52
11.79
-1.27
p= .03
-1.54
p.01
.02
10,45
11.59
11.90
-1.14
p=s.06
-1.45
p=.02
.04
10.15
11.50
11.78
-1.35
p=.02
-1.63
pฎ.01
.02
9.96
11.35
11.49
-1.39
p= .02
-1.53
p=.01
.01
10.15
11.45
11.75
-1.30
p=.03
-1.60
p=,01
.02
15-14
-------
TABLE 15-10 (cont'd). CRUDE AND ADJUSTED CHANGES
IN BLOOD LEAD LEVELS*
SAB
POST2 Blood Lead Levels
S-A1
>
CO
Overall3
Group
Effect
Sported Canned Food
10.21
11.40
11.72
-1.19
p=.04
-1.51
p=.01
.03
W Model
Plus Ferritin Level
10.14
11.42
11.59
-1.28
p=.02
-1.45
p=.01
.02
W Model
Plus Lead Jobs
10.47
11.72
12.01
-1.25
p=.03
-1.54
p=.01
.02
Base Model
Plus Lead Hobbies
16.28
11.49
11.85
-1.21
p=.04
-1.57
p=.01
.02
^ase Model
Plus Smoking
10.26
11.52
11.82
-1.25
p=.03
-1.55
p=.01
.02
W Model
Plus Owner Occupied Premises
10.07
11.24
11.60
-1.17
p=.04
-1.53
p=.01
.02
Crude and adjusted blood lead difference between Group S and A. P values are given if the overall group
2effect is statistically significant.
Crude and adjusted blood lead difference between Group S and B. P values are given if the overall group
3effect is statistically significant.
P values associated with group assignment. Describes the statistical stability of the relationship between group
/ssignmait and POST2 blood lead levels.
>e chipping paint variable included an indicator term for missing.
^Excludes two children who became lead poisoned.
Base Model: POST2_Pb = Group + PreJPb.
* premises (N=85), the differences between the Study Group and Control Groups A and B
Weiฉ -0.92 (p=.17) and -1.46 (p=.04), respectively.
The results were also similar when the analysis included only children who lived on the
study premises for at least 300 days after the pre-abatement blood lead test, thereby
eliminating children who moved during the follow-up period. Here, the differences between
the Study Group and Control Groups A and B were -1.42 (p=.02) and -1.49 (p=.02),
respectively.
The results were also similar when we took into account the timing of the blood
samples using two different methods. First, we limited the analysis to subjects whose PRE
15-15
-------
and P0ST2 blood samples were most closely matched on calendar time (within sixty days of
one year or 305-420 days apart, N=134). Here, we found that the differences between the
Study and each Control Group were identical (1.41 fig/dL, p=.02). Second, we included all
150 study subjects and controlled for timing by including indicator terms for calendar month
of the PRE blood sample and the number of days between the PRE and POST2 samples.
Again, we found similar differences between the Study Group and Control Groups A and B
(1.22, p=.03 and 1.30, p=.03, respectively). The latter results were also similar when the
model included an interaction term between calendar months and number of days. The
interaction was not significant.
The results were also quite similar when age, sex, socioeconomic status, ferritin levels,
mouthing and handwashing behaviors, spending time away from home, spending time outside
of the study area, playing in the yard, eating food outdoors, sitting on the floor inside the
home, eating canned foods including those imported from foreign countries, lead related job
and hobbies and cigarette smoking among household residents, living in owner occupied
premises, the presence of chipping paint, the presence of pets that go outdoors, and tap water
lead levels were added to the base model one at a time (Table 15-10). The results were
virtually identical when the continuous variables were treated as such. However, when the
paint lead variables were added, differences between the Study and Control Groups were
somewhat diminished (-1.19 and -1.34 pg/dL for Control Groups A and B, respectively) and
the group effect was borderline significant (p=0.05). When race was added to the base
model, differences were also diminished (-0.92 and -1.27 /xg/dL) and the group effect was
also not statistically significant (p=0.09). However, no statistically significant differences in
crude or adjusted POST2 blood lead levels were seen among Study Group children of
different races (Table 15-11).
No "dose-response" relationship was observed between the mean change in blood lead
level and the starting soil lead level or the size of the excavated yard area (Tables 15-12 and
15-13). POST2 abatement blood lead levels were similar for children with the lowest and
highest pre-abatement soil lead levels and the smallest and largest excavated yaixi areas. The
lack of a trend should be evaluated in light of the study eligibility criteria that restricted the
soil and blood lead ranges. Only six children in the Study Group had median pre-abatement
15-16
-------
TABLE 15-11. CRUDE AND ADJUSTED* POST2 BLOOD LEAD LEVELS AMONG
CHILDREN IN THE STUDY GROUP ACCORDING TO RACE
Race
Crude
POST2
Blood Lead Level **
Adjusted
POST2
Blood Lead Level **
Black (N=22)
11.51
11.27
White (N=4)
11.00
11.54
Hispanic (N=14)
10.05
9.78
Cape Vendean (N=10)
10.76
11.32
Other (N=2)
11.64
11.99
%
^Adjusted for pre-abatement blood lead level.
Differences between racial groups were not statistically significant.
TABLE 15-12. CRUDE AND ADJUSTED* POST2 BLOOD LEAD LEVELS AMONG
STUDY GROUP PARTICIPANTS ACCORDING TO INITIAL SOIL LEAD LEVEL
Median
Pre-Abatement
Soa
Lead Level (PPM)
***
Number
In
Categoiy
Crude
POST2
Blood Lead
Level **
Adjusted
POST2
Blood Lead
Level **
< 1,000
1,001 - 2,000
2,001 - 3,000
> 3,000
6
18
16
12
9.00
12.00
9.81
10.58
%ฃdjustad for pre-abatement blood lead level.
^Differences were not statistically significant.
Adjusted osmg die weighing factors derived from the inteicalibiwtiaa. study.
9.17
11.47
10.75
10.04
soil lead levels that were less than 1,000 PPM, and pie-abatement blood lead levels were
restricted to 7 through 24 pg/dL.
We also conducted exploratory multivariate analyses to control simultaneously for
several potential confounding variables. Two variable selection methods were used. First, a
backward elimination procedure identified variables from the list in Table 15-9 that were
statistically significant predictors of POST2 blood lead levels. When Pre-Pb, age, race, and
15-17
-------
TABLE 15-13. CRUDE AND ADJUSTED* POST2 BLOOD LEAD LEVELS
AMONG STUDY GROUP PARTICIPANTS ACCORDING TO THE SIZE OF THE
EXCAVATED YARD AREA
Excavated
Yard Area
(Sq. Feet)
Number In
Category
Crude POST2
Blood Lead
**
Level
Adjusted POST2
Blood Lead
**
Level
<; 1,000
11
9.82
10.80
1,001 - 2,000
18
10.67
11.66
> 2,000
23
11.04
10.58
'Adjusted for pre-abatement blood lead level.
Differences were not statistically significant.
lead jobs were controlled simultaneously the adjusted POST2 blood lead levels were 10.36,
11.26, and 11.66 /xg/dL for Groups S, A, and B, respectively, and the adjusted differences
between the Study Group and Control Groups A and B were 0.90 (95 % CI +0.23 to -2.04)
and 1.31 /xg/dL (95% CI -0.14 to -2.47), respectively. The overall group effect was not
statistically significant (p=.08). These results were quite similar when the blood lead levels
were log transformed.
Second, a potential confounding variable was selected for the multivariate model if its
inclusion in the base model altered the magnitude of difference between the Study Group and
either Control Group by more than 10%. The variables identified by this criterion were
race, socioeconomic status, and playing or sitting on the floor. In a model controlling these
variables and Pie-Pb, the adjusted POST2 blood lead levels were 10.71, 11.51, and
11.92 /xg/dL for Groups S, A, and B, respectively, and the adjusted differences between the
Study Group and Control Groups A and B were 0.80 (95% CI +0.45 to -2.05) and
1.21 /xg/dL (95% d +0.06 to -2.48), respectively. The overall group effect was not
statistically significant (p=.16). The results were quite similar when the blood lead levels
were log transformed.
We also examined the data for the presence of effect modification, that is, differences
in soil abatement effectiveness according to the child's characteristics. No statistically
significant interactions were seen for age, sex, socioeconomic status, length of residence
(since birth or not), residence in an owner occupied home, and behavioral characteristics
15-18
-------
such as amount of time spent playing in the yard, eating food outside, handwashing after
outside play, thumb sucking, and other mouthing behaviors. However, the data suggested
that the effect of the soil abatement was enhanced among children who played in their yards
more than 15 hours per week (S-A -2.17 and S-B -3.56) and among children who lived in
non-owner occupied housing (S-A -3.08 and S-B -2.44).
15.3 HAND LEAD LEVELS
Tables 15-14 and 15-15 describe the handwipe lead distributions over time and
Tables 15-16 and 15-17 describe the results of erode analyses that examined the average
change in hand lead levels among the participants following abatement activities. The
handwipe Held blank lead levels varied considerably within and across phases. The means
and standard deviations were 6.0 and 1.9 for PRE, 8.4 and 3.5 for POST1, and 12.3 and 5.4
for POST2. Because the blanks were so variable and were not individually matched to the
participants, background levels were taken into account by subtracting the maximum and
median field blank level for each sampling round. (Any negative value was treated as zero.)
When the maximum level was subtracted, the mean hand lead level in all groups declined
from the pre abatement to the first post-abatement sampling round. The mean hand lead
level in the Study Group changed little at the second post-abatement sampling round while it
increased in the Control Groups (Table 15-16). When the median level was subtracted, the
mean hand lead level in the Study Group declined at the first and second post-abatement
sampling rounds. The mean hand lead levels in the two Control Groups first declined and
then rose to a level higher than baseline (Table 15-17).
Because the PRE and POST2 sampling rounds are most closely matched on season, we
focused subsequent analyses on this comparison. When the maximum blank level was
subtracted, the mean hand lead level decreased by 3.61 pg in the Study Group (p=.02),
0.99 ftg in Control Group A (p=.69), and 0.36 fig in Control Group B (p=.85). When the
median blank lead level was subtracted the mean hand lead levels declined by 2.75 fig in the
Study Group (p=.08), and 0.68 in Control Group A (p=.79) and increased by 0.76 in
Control Group B (p=.72).
15-19
-------
TABLE 15-14. HAND WIPE LEAD (#tg/pair of hands) DISTRIBUTIONS OVER TIME
ADJUSTING FOR MAXIMUM FIELD BLANK LEAD LEVEL* AND
EXCLUDING CHILDREN WHO BECAME LEAD POISONED
S
A
B
Total
as
(N=52)
P
II
(N=47)
fN=148^
Minimum
0
0
0
0
25%
2.5
2.5
1.5
2.5
50%
4.5
3.5
3.5
4.5
75%
8.0
6.5
8.5
7.5
Maximum
52.5
42.5
35.5
52.5
Mean
6.67
5.67
6.60
6.31
Standard Deviation
8.21
7.08
7.41
7.56
POST1
fN=51)
-------
TABLE 15-15. HANDWIPE LEAD (fig/pair of hands) DISTRIBUTIONS OVER TIME
ADJUSTING FOR MEDIAN FIELD BLANK LEAD LEVEL* AND EXCLUDING
CHILDREN WHO BECAME LEAD POISONED
S
A
B
Total
SEE
(N=52)
(N=491
(N=47)
P
1!
ฃ
oo
Minimum
0
0
0
0
25%
4.75
4.75
3.75
4.75
50%
6.75
5.75
5.75
6.75
75%
10.25
8.75
10.75
9.75
Maximum
54.7
44.75
37.75
54.75
Mean
8.79
7.79
8.75
8.45
Standard Deviation
8.33
7.20
7.51
7.67
PQSTl
fN=5n
fN=48^
(N=46)
(N=145)
Minimum
0
0
0
0
25%
1.5
2.5
1.3
1.5
50%
3.5
5.5
3.5
4.5
75%
8.5
8.5
9.5
8.5
Maximum
30.5
22.5
64.5
64.5
Mean
6.30
5.96
7.89
6.69
Standard Deviation
6.80
4.86
12.68
8.64
POST2
(N=52)
fN=49^
(N=46)
(N=147)
Minimum
0
0
0
0
25%
0
0
0.5
0
50%
3.5
2.5
4.5
3.5
75%
8.0
6.5
12.5
9.5
Maximum
58.5
96.5
63.5
96.5
Mean
6.04
6.58
9.42
7.28
Standard Deviation
9.34
16.82
14.31
13.74
Negative levels were assigned the value mo.
15.4 ENVIRONMENTAL LEAD LEVELS
Tables 15-19 through 15-25 describe the distributions of soil, dust, water, and paint
lead levels among the Study and Control Groups. Table 15-26 describes the QA/QC results
for the soil and dust analyses. Median soil and dust lead levels, maximum first flush water
lead levels and maximum wall and woodwork paint lead levels were used to characterize
15-21
-------
TABLE 15-16. CRUDE CHANGES* IN HAND LEAD LEVELS (jig/pair of hands)*
EXCLUDING CHILDREN WHO BECAME LEAD POISONED**
STUDY PHASE
STUDY GROUP
CONTROL
GROUPA
CONTROL
GROUP B
Pre-Abatsment
(Sopt 89-Doc. *89)
6.67 -i
(N-52)
-3.81
p-0.002
6.67-,
(N-49)
-2.96
p-0.01
6.60
(N-47)
-1.95
P-0.35
Post-Abatement
POST1
(Mar. *90 - July *90)
2.00 -
(N-52)
+0.22
p-0.86
2.10 -
(N-48]
+2.13
p-0.30
4.67 -
(f>-46)
+1.69
P-0.4S
POSTS
(July *80-Jan. >91)
3.06 -
(N-62)
4.14-
(N-49)
6.15 -
(n-46)
each child's unit or premises. Detection limits for soil and dust were each 100 PPM, for
water it was 1 ^tg/L, and for paint it was 0.5 mg/cm . The soil and dust concentrations were
adjusted using weighing factors designed to make the Boston project's soil and dust lead
levels comparable to those of the Cincinnati and Baltimore Lead-in-Soil Demonstration
projects. These weighing factors (1.0370 for soil and 1.1527 for dust) were derived from
the results of the intercalibration study conducted under the supervision of Dr. Robert Elias.
Dr. Elias is with the U.S. EPA Environmental Criteria Assessment Office and has the
responsibility to facilitate the successful completion of the Lead-in-Soil Demonstration
Projects.
15.4.1 SoU
At baseline the median surface soil lead levels were slightly higher in Control Group A
(2,230 PPM) than the Study Group and Control Group B (2,074 and 2,100 PPM). Sampling
15-22
-------
TABLE 15-17. CRUDE CHANGES IN HAND LEVELS fog/pair of hands)*
EXCLUDING CHILDREN WHO BECAME LEAD POISONED**
CONTROL
GROUPA
CONTROL
GROUPB
6.74-,
(Nซ40j
6.98 -
(N-48;
6.98-J
(N-49)
8.14
(N-47)
-1.29
p-0.31
7.89 -|
(n-ซ)
40.89
p-0.76
9.80 -
(n-46)
-0.77
P-0.73
+1.63
P-0.52
STUDY PHASE
STUDY GROUP
Pw-Abatanwrt
(Sept *89-Doc. '89)
Post-Abatamant
POST1
(Mar. '90 July *90)
8.79
(N-52)
6.30
(N-52)
POSTS
(July "90-Jan. "91)
6.04 -
(N-52)
2.63
p-O.OS
056
p-0.86
conducted within a few weeks of the soil abatement documented the reduction in lead levels
in the Study Group. The drop in median soil lead levels ranged from 166 to 5,558 PPM, the
average drop was 1,856 PPM. However, many samples still had detectable lead levels at
post-abatement sampling (Median Post Abatement Level: 109 PPM).
About nine months after soil abatement, median surface soil lead levels in the Study
Group had not increased but several properties had evidence of recontamination. Eight
properties (2396) had median soil lead levels ranging from 156 to 1,867 PPM. The
concentration of lead in soil for these eight properties in PPM, were 156, 171, 202, 228,
249, 259, 389, and 1,867. The surface soil lead levels in Control Groups A and B did not
change substantially over this period.
15-23
-------
TABLE 15-18. CRUDE AND ADJUSTED CHANGES IN HAND LEAD LEVELS
Otg/pair of hands) EXCLUDING CHILDREN WHO BECAME LEAD POISONED*
SAB
POST2 Blood Lead Levels S-A
S-B
Overall'
Group
Effect
Crude
Adjusting for Pre-
Abatement Hand
Lead Level
Crude
Adjusting for
Pre-Abatement Hand
***
Lead Level
3.06
4.14 6.15 -2.62
5.86
-3.25
2.90 4.47 6.04 -1.56 -3.14 p=.48
6.04 6.58 9.42 -2.07 -3.51
6.95 9.29 -1.08 -3.43 p=.43
*Two children in the Study Group became poisoned between POST1 and POST2 sampling rounds.
Adjusts for maximum field handwipe blank lead level.
Adjusts for median field handwipe blank lead level.
15.4.2 Dust
Tables 15-20 through 15-22 describe the distribution of interior floor dust lead
2 1
concentration (PPM), dust loading (mg/m ), and dust lead loading (jig/m ) over time in the
Study and Control Groups. At baseline median dust lead concentrations were similar across
the three groups (2,513-2,651 PPM). Median floor dust lead concentrations in the Study
Group and Control Group A were reduced by 53% and 49%, respectively, an average of 4-5
weeks after the interior dust abatement (Post Abatement). Floor dust lead levels remained
substantially below baseline levels an average of 33 weeks after interior dust abatement for
both the Study Group (67%) and Control Group A (54%) (Recontamination 2). During this
period Control Group B experienced a comparable decline (42%) in floor dust lead levels.
(Control Group B received loose paint abatement but not interior dust abatement.)
At baseline the median floor dust loading was higher in Control Group B than in the
other two groups (40 vs. 24 and 25 mg/m ) In the Study Group median floor dust loading
increased by 50% an average of 4-5 weeks after the interior dust abatement. (Mean floor
dust loading was essentially unchanged.) Median dust loading in the Study Group then
15-24
-------
TABLE 15-19. DISTRIBUTION OF SURFACE SOIL LEAD CONCENTRATIONS*
OVER TIME AND ACCORDING TO GROUP
SAB Total
Pre Abatement (N=34) (N=36) (N=30) (N100)
(Aug.'89 - June '90)
Minimum
415
747
985
415
25%
1,556
1,374
1,452
1,452
50%
2,074
2,230
2,100
2,152
75%
2,644
3,215
3,422
3,163
Maximum
5,704
6,948
4,563
6,948
Mean
2,255
2,524
2,401
2,395
Standard Deviation
1,165
1,381
1,195
1,248
Post (Oct.'89 - Dec.'89)
Abatement (N=25) N/A N/A (N=25)
Minimum 52 52
25% 83 83
50% 109 109
75% 166 166
Maximum 249 249
Mean 123 123
Standard Deviation 56 56
Recontamination (N=34)
(June '90 - July '90)
Assessment
Minimum 52
25% 52
50% 52
75% 109
Maximum 1,867
Mean 145
Standard Deviation 315
(N=34) (N=30) (N=98)
622
954
52
1,556
1,556
52
2,23
1,970
1,556
3,007
3,059
2,385
5,755
5,081
5,755
2,437
2,315
1,605
1,226
1,144
1,443
Adjusted using the weighing factor derived from the intercalibrstton study. The median soil concentration was
used to characterize the premises. Detection limit was 100 PPM. Undetectable levels were assigned the value
50 PPM. N/A = Not Applicable.
15-25
-------
TABLE 15-20. DISTRIBUTION OF INTERIOR FLOOR DUST LEAD
CONCENTRATIONS* OVER TIME AND ACCORDING TO GROUP
S
A
B
Total
Pre Abatement
(N=41)
(N=40)
(N=36)
3
II
w
(Aug.'89 - Jan.'90)
Minimum
150
646
496
150
25%
1,164
1,406
1,354
1,303
50%
2,651
2,513
2,542
2,547
75%
4,288
4,369
5,314
4,380
Maximum
107,201
22,823
46,108
107,201
Mean
6,623
4,202
5,178
5,350
Standard Deviation
16,786
5,075
8,272
11,291
Post (Oct. 89 - March '90)
Abatement
(N=31)
(N=34)
N/A
(N=65)
Minimum
450
519
450
25%
807
865
865
50%
1,233
1,274
1,268
75%
2,190
2,121
2,132
Maximum
12,680
5,141
12,680
Mean
2,420
1,822
2,107
Standard Deviation
3,058
1,352
2,327
Recontamination 1 (March '90 - June '90)
(N=40)
(N=35)
(N=37)
(N=112)
Minimum
334
426
184
184
25%
657
807
807
795
50%
939
1,279
1,095
1,095
75%
1,712
1,568
1,499
1,562
Maximum
49,566
6,052
4,253
49,566
Mean
3,108
1,458
1,493
2,059
Standard Deviation
8,257
1,108
1,084
5,033
Recontamination 2 (July '90 -
Dec.'90)
(N=33)
(N=35)
(N=34)
(N=102)
Minimum
346
311
288
288
25%
692
749
968
784
50%
876
1,153
1,475
1,193
75%
1,349
1,568
2,017
1,694
Maximum
4,841
3,804
10,374
10,374
Mean
1,294
1,300
1,886
1,494
Standard Deviation
1,094
774
1,777
1,300
Adjusted using the weighing factor derived from the intercalibration study. A single composited floor dust
sample was used to characterize a child's living unit. N/A Not Applicable.
15-26
-------
TABLE 15-21. DISTRIBUTION OF INTERIOR FLOOR DUST LOADING* (mg/m2)
OVER TIME AND ACCORDING TO GROUP
S
A
B
Total
Pre-Abatement (Aug.'89 - Jan. '90)
2
II
W
(N=40)
(N=35)
(N=116)
Minimum
4
7
5
4
25%
13
11
15
12
50%
24
25
40
29
75%
69
51
71
67
Maximum
363
246
141
363
Mean
51
41
47
46
Standard Deviation
67
46
36
52
Post (Oct;89 - March '90)
(N=31)
(N=34)
N/A
(N*=65)
Abatement
Minimum
9
3
3
25%
15
9
31
12
50%
36
19
31
29
75%
59
37
31
45
Maximum
254
117
31
254
Mean
52
30
31
41
Standard Deviation
58
31
31
47
decontamination 1 (March '90 - June '90)
(N=40)
(N=35)
(N=37)
(N=112)
Minimum
2
3
2
2
25%
14
15
12
13
50%
24
28
32
27
75%
62
48
56
55
Maximum
366
195
278
366
Mean
57
41
47
49
Standard Deviation
74
43
53
59
Recontamination 2 (July '90 - Dec.'90)
(N=32)
(N=33)
(N=33)
(N=98)
Minimum
2
2
2
2
25%
9
10
11
9
50%
16
17
19
17
75%
32
41
32
35
Maximum
136
153
115
153
Mean
27
31
29
29
Standard Deviation
31
36
29
32
*A single composited floor dust sample was used to characterize a child's living unit. N/A = Not Applicable.
15-27
-------
TABLE 15-22. DISTRIBUTION OF INTERIOR FLOOR DUST LEAD
LOADING* Oig/m2) OVER TIME AND ACCORDING TO GROUP
S
A
B
Total
Pre-Abatement (Aug.*89 - Jan.'90)
(N=41)
?
II
5
(N=35)
1)
5
Minimum
9
9
3
3
25%
35
36
38
36
50%
61
68
87
75
75%
126
203
208
153
Maximum
7,976
437
3,354
7,976
Mean
344
117
291
250
Standard Deviation
1,280
113
623
836
Post (Oct.89 - March '90)
(N=31)
(N=34)
N/A
(N =65)
Abatement
Minimum
10
3
3
25%
22
12
17
50%
47
27
41
75%
74
71
71
Maximum
2,547
191
2,547
Mean
145
49
94
Standard Deviation
452
50
315
Recontamination 1 (March '90 - June '90)
ง
II
s
(N=35)
(N=37)
(N=112)
Minimum
1
6
1
1
25%
13
14
11
13
50%
28
32
31
31
75%
80
67
82
77
Maximum
3,087
259
929
3,087
Mean
236
53
83
128
Standard Deviation
6S8
58
156
410
Recontamination 2 (July '90 - Dec. *90)
(N=32)
(N=33)
. (N=33)
(N98)
Minimum
2
1
2
1
25%
7
9
14
9
50%
18
21
25
21
75%
28
46
62
46
Maximum
224
226
527
527
Mean
38
39
65
47
Standard Deviation
60
50
107
77
*
Adjusted using the weighing factor derived from the intercalibration study. A single composited floor dust
sample was used to characterize a child's living unit. N/A = Not Applicable.
15-28
-------
TABLE 15-23. DISTRIBUTION OF INTERIOR WINDOW WELL DUST
LEAD CONCENTRATIONS* OVER TIME AND ACCORDING TO GROUP
S
A
B
Total
fte-Abatement (Aug.'89 - Jan.'90)l
(N=40)
(N=41)
(N=34)
(N=115)
Minimum
231
1,153
58
58
25%
3,458
3,170
2,594
3,170
50%
11,815
15,907
13,429
13,832
75%
28,818
38,039
28,818
33,428
Maximum
121,034
74,926
147,546
147,546
Mean
19,481
22,429
27,285
22,839
Standard Deviation
23,039
20,722
37,314
27,299
Post (Oct. 89 - March '90) (N=32) (N=32) N/A (N=64)
Abatement
Minimum 58 58 58
25% 2,795 778 11,166 1,037
50% 7,550 2,190 11,166 3,400
75% 16,138 3,689 11,166 10,778
Maximum 46,108 63,399 11,166 63,399
Mean 10,789 4,766 11,166 7,777
Standard Deviation 10,492 11,166 11,166 11,168
Recontamination 1 (March'90 - June *90) (N=41) (N=35) (N=36) (N=112)
Minimum
1,153
1,383
692
692
25%
4,323
5,418
3,977
4,409
50%
12,103
10,086
11,527
11,181
75%
18,443
25,359
32,276
23,486
Maximum
44,955
55,330
98,844
98,844
Mean
12,493
15,671
21,464
16,370
Standard Deviation
10,559
14,156
22,998
16,806
Recontamination 2 (July '90 - Dec.'90) (N=34) (N=32) (N=30) (Nซ96)
Minimum
807
865
548
548
25%
3,573
5,418
2,900
3,631
50%
8,213
13,832
9,942
10,807
75%
27,665
29,207
42,650
34,581
Maximum
109,507
96,827
103,743
109,507
Mean
19,815
20,674
24,761
21,647
Standard Deviation
23,830
23,123
27,604
24,676
Adjusted using the weighing factor derived from the intercalibration study. The median window well dust
concentration was used to characterize a living unit. N/A = Not Applicable.
15-29
-------
TABLE 15-24. DISTRIBUTION OF INTERIOR WINDOW WELL LOADING*
(mg/m2) OVER TIME AND ACCORDING TO GROUP
S
A
B
Total
Pre-Abatement (Aug.'89 - Jan.'90)
(N=40)
(N=41)
(N=34)
3
11
w
Minimum
20
23
0
0
25%
52
68
89
63
50%
155
280
216
219
75%
525
578
401
547
Maximum
2,122
6,542
5,701
6,542
Mean
339
564
443
450
Standard Deviation
472
1,100
968
884
Post (Oct.89 - March *90)
(N=32)
(N=32)
N/A
(N=64)
Abatement
Minimum
0
0
0
25%
27
16
244
18
50%
79
29
244
39
15%
162
122
244
160
Maximum
2,018
1,293
244
2,018
Mean
185
115
244
150
Standard Deviation
372
244
244
314
Recontamination 1 (March '90 - June '90)
(N=41)
(N=35)
(N=36)
N=(112)
Minimum
32
22
9
9
25%
123
176
85
118
5096
387
284
213
278
15%
646
542
558
615
Maximum
3,431
3,207
8,905
8,905
Mean
513
489
634
545
Standard Deviation
615
601
1,509
984
Recontamination 2 (July '90 - Dec.'90)
(N=34)
(N=32)
2"
II
s
(N=96)
Minimum
30
38
15
15
25%
125
216
158
153
50%
295
438
418
341
75%
546
760
744
700
Maximum
3,310
2,867
3,457
3,457
Mean
559
562
641
585
Standard Deviation
726
559
837
707
*
The median window well loading was used to characterize a child's living unit. N/A = Not Applicable.
15-30
-------
TABLE 15-25. DISTRIBUTION OF INTERIOR WINDOW WELL LEAD LOADING*
(jig/m2) OVER TIME AND ACCORDING TO GROUP
S
A
B
Total
Pre-Abatement (Aug. '89 - Jan. '90)
(N=40)
(N=41)
(N=34)
3
II
~-ป
Minimum
5
32
0
0
25%
213
351
279
282
50%
2,262
3,825
2,236
3,145
75%
9,298
17,122
11,465
12,829
Maximum
56,614
451,193
657,170
657,170
Mean
7,861
25,545
26,976
19,817
Standard Deviation
12,606
73,487
111,910
75,178
Post (Oct. 89 - March '90)
Abatement
(N=32)
(N=32)
N/A
(N=64)
Minimum
0
0
0
25%
13
27
50%
114
68
249
75%
687
381
1,398
Maximum
3,280
11,561
46,529
Mean
46,529
674
1,942
Standard Deviation
3,211
2,073
6,114
8,267
Recontamination 1 (March '90 - June *90)
(N=41)
(N=35)
(N=36)
(N=112)
Minimum
98
83
9
9
25%
1,066
644
672
803
50%
3,547
3,810
3,503
3,531
75%
7,817
13,293
8,683
8,711
Maximum
19,869
38,116
173,592
173,592
Mean
5,007
8,544
11,955
8,346
Standard Deviation
4,961
10,416
29,861
18,210
Recontamination 2 (July '90 - Dec.'90)
(N=34)
(N=32)
(N=30)
(N=96)
Minimum
41
44
6
6
25%
1,754
1,540
1,340
1,549
50%
5,102
5,104
5,318
5,186
75%
11,234
12,797
18,690
13,434
Maximum
169,584
80,981
52,078
169,584
Mean
12,443
10,750
11,406
11,555
Standard Deviation
28,946
16,129
14,160
20,920
*
Adjusted using the weighing factor derived from the intercalibration study. The median window well dust
concentration, and loading were used to characterize a child's living unit. N/A = Not Applicable.
15-31
-------
TABLE 15-26. QA/QC RESULTS FOR SOIL AND DUST ANALYSES
EMSL RESULTS* BOSTON RESULTS*
95 % Biweight % of Results
Consensus Biweight Distribution Within EMSL
Mean Mean Bounds Mean Range 95% Bounds
**
SOIL
CINL
290
315
204-426
399
207-570
61.3
BALH
934
1,017
847-1,187
1,044
747-1,244
73.3
BOSM
5,759
6,219
4,742-7,696
6,786
6,015-7,549
100.0
CINH
12,376
12,729
11,361-14,096
14,074
11,407-16,592
50.0
DUST**
CIN02
242
233
93-372
331
115-461
64.7
BAL03
1,334
1,438
1,091-1,786
1,232
980-1,441
92.0
CINOi
2,933
2,617
1,422-3,812
2,671
2,075-3,228
100.0
***
***
***
***
BOSOl
11,783
10,374-15,561
Adjusted using the weights derived from the intercalibration study.
^Acronyms stand for the source of the sample.
BOSOl was not included in the intercalibration study because of lack of material.
decreased to a level 33 % below baseline by the end of the recontamination assessment
period. In Control Group A median floor dust loading decreased by 24% after dust
abatement and remained substantially below baseline levels during subsequent sampling.
Median levels in Control Group B decreased by 53% from baseline levels over this time
period.
At baseline floor dust lead loading was higher in Control Group B than the other groups
(87 vs. 61 and 68 uglm ). An average of 4-5weeks after the interior dust abatement, lead
loading had decreased by 23 % in the Study Group and by 60% in Control Group A. By the
end of the recontamination assessment period, floor lead loading had declined by 70% in the
Study Group and 69% in Control Group A. Control Group B declined by 71 % over this
time period.
15-32
-------
Tables 15-23 through 15-25 describe the distribution of interior window well dust lead
concentration, dust loading, and dust lead loading over time in the Study and Control
Groups. At baseline median dust lead concentrations were higher in Control Group A than
the Study Group and Control Group B (15,907 vs. 11,815 and 13,429 PPM). Median
window well dust lead concentrations in the Study Group and Control Group A were reduced
by 36% and 8696, respectively, an average of 4-5 weeks after the interior dust abatement
(Post Abatement). Window well dust lead levels increased in these groups over the
recontamination assessment period but were still below baseline levels (13-30%) an average
of 33 weeks after interior dust abatement (Recontaminafion 2). During this period the
window well dust lead concentrations in Control Group B declined by 26%.
At baseline the median window well dust loading (mg/m2) was higher in Control Group
A than in the other groups (280 vs. 155 and 116 mg/m2). An average of 4-5 weeks after the
interior dust abatement, the median window well dust loading levels decreased by 49% in the
Study Group and 90% in Control Group A. Over die recontamination assessment period,
median window well dust loading substantially rose in both groups and were 56-90% above
baseline by the end of this period. Median levels in Control Group B also rose by 94% by
the end of this period.
At baseline the median window well dust lead loading was higher in Control Group A
than the other groups (3,825 vs. 2,262 and 2,236 /tg/m2). An average of 4-5 weeks after the
interior dust abatement, lead loading had decreased by 70% in the Study Group and by 98%
in Control Group A. Over the recontamination assessment period window well lead loading
rose (from baseline levels) by 126% in the Study Group and 33% in Control Group A.
Control Group B increased by 138% over this time period.
Lastly, no dose-response relationship was seen when we modeled the second
recontamination assessment floor and window well dust measures as a function of the change
in soil lead concentration.
The results of the external audit sample analyses conducted by EMSL, the external
QA/QC contract laboratory in Las Vegas, indicate that our soil and dust lead data were
generally of good quality (Table 15-26). While our mean soil concentrations were
consistently higher than the consensus and biweight means, the majority of our results fell
within the 95 % biweight distribution bounds provided by EMSL. The best agreement was
15-33
-------
seen for lead levels that encompassed most of our soil samples (BALH and BOSM). Our
dust lead results were also in good agreement with those of the EMSL laboratory; the
majority of our results also fell within the 95 % biweight bounds and the best agreement was
seen for lead levels that encompassed most of our floor dust samples (BAL03 and CIN01).
Because of the lack of material, no EMSL data are available for the highest dust lead
category (BOSOl) where a large portion of our window well dust samples fell.
15.4.3 Water
Table 15-27 describes the distribution of water lead levels in the Study and Control
Groups. The maximum levels in two first flush tap water samples was used to characterize
each living unit. Water lead levels ranged from undetectable to 560 fig/L. The median
concentration in Control Group B was higher than the other two groups (36 vs. 20 and
18 /tg/L).
TABLE 15-27. DISTRIBUTION OF WATER LEAD
CONCENTRATIONS* (jig/h) ACCORDING TO GROUP
S
A
B
Total
(N=34)
(N34)
(N=29)
(N=97)
Minimum
1
UD
2
UD
25%
11
8
11
9
50%
20
18
36
20
75%
43
58
61
57
Maximum
350
387
560
560
Mean
43
54
76
56
Standard Deviation
68
84
115
90
The maximum of two first flush samples was used to characterize the living unit of the child. UD means
undetectable. The detection limit was 1.0 figfL. To calculate the mean and standard deviation, 0.S fig/L was
used to characterize undetectable levels.
15-34
-------
15.4.4 Paint
Table 15-28 describes the distribution of wall and woodwork paint lead concentrations
among the Study and Control Groups. The maximum XRF readings for the wall and
woodwork were used to characterize each living unit. A smaller proportion of units in
Control Group B had detectable lead paint on walls (20.0%) than the Study Group and
Control Group A (30.6% and 47.4%, respectively). Almost all units had detectable lead
paint on woodwork.
Table 15-29 describes the amount of chipping paint that was present at baseline inside
subjects' living units Case managers conducted the chipping paint assessments by visual
inspection in all but 20 units. A larger proportion of units in the Study Group had significant
amounts (>200 square inches) of chipping and peeling paint compared to Control Groups A
and B. All groups received loose paint abatement as part of the intervention.
TABLE 15-28. DISTRIBUTION (%) OF WALL AND WOODWORK PAINT LEAD
CONCENTRATIONS (rag/cm2) ACCORDING TO GROUP
S
A
B
Total
(N=39)
(N=40)
(N=36)
(N=115)
Wall
Undetectable
30.6
47.4
20.0
33.7
0.5-1.0
25.0
18.4
33.3
25.0
1.1-9.9
25.0
10,5
16.7
17.3
10.0
19.4
23.7
30.0
24.0
Woodwork
Undetectable
5.1
0.0
11.1
5.2
0.5-1.0
10.3
15.0
27.8
17.4
1.1-9.9
25.6
30.0
22.2
26.1
10.0
59.0
55.0
38.9
51.3
The maximum XRF level for the wall and woodwork were used to characterize the living unit of the child.
Detection limit was 0.5 mg/cm2-
15-35
-------
TABLE 15-29. DISTRIBUTION (%) OF AMOUNT INTERIOR CHIPPING PAINT*
AT BASELINE ACCORDING TO GROUP
SAB Total
(N34) (N=36) (N=32) (N=102)**
Amount of Chipping Paint at
Baseline (Square Inches)
0 - 50 55.9 58.3 46.9 53.9
51-200 14.7 22.2 34.4 23.5
> 200 29A 19A lO 22.5
The presence of interior chipping paint at baseline was assessed by visual inspection of all rooms in the living
unit.
**
The chipping paint assessment was not performed on 20 units (8 in S, and 6 each in A and B). Percentages
exclude units with missing data.
15.5 COST OF ABATEMENT ACTIVITIES
15.5.1 Soil Abatement
In this section the actual cost of soil abatement is presented, along with the cost
breakdown for the various components of the abatement including soil sampling, excavation,
disposal and replacement. Costs are described separately for abatements conducted in 1989
and 1990 because of differences in contractors and abatement requirements. Average cost
per property, per square meter, and per cubic yard of soil excavated and replaced are
presented. These figures must be interpreted with caution given the many unique conditions
under which the abatements were conducted for this study. Alternative cost estimates are
also provided that perhaps better reflect future costs of lead contaminated soil abatement.
Lead contaminated soil was abated from thirty-six properties in 1989. The abated areas
of these properties averaged 2,141 square feet, or 199 square meters and ranged from 12 to
702 square meters. It is estimated that an average of 41 cubic yards of soil were excavated
and replaced at each site. This estimate is cnide since 12 of the 36 abatements were
conducted after the ground had frozen and consequently large slabs of earth were often
removed that were difficult to measure. In some cases, measurements were not possible.
Since the yard sizes were somewhat smaller, on average, for the premises abated in 1990
15-36
-------
(178 versus 199 square meters), and since the measurement of cubic feet excavated in 1990
was easier and more accurate and resulted in an estimate of 44 cubic yards per property, we
have assumed that the cubic feet abated per property were similar in 1989 and 1990, Thus,
the estimated cost per cubic meter in 1989 is probably greater than the actual cost incurred.
The costs of soil abatements are shown in Tables 15-30 and 15-31.
Many of the soil abatement related costs incurred in both 1989 and 1990 may not be
applicable to future soil abatement activities in this and other communities. In 1989, for
example, $69,668 was spent for contract development and supervision by Applied
Occupational Health Systems. This cost was incurred because of lack of experience with
lead contaminated soil abatement. Probably only the $19,401 spent for abatement
supervision is applicable for other settings. Other portions of the Contract
Development/Supervision expenses might be applicable as one time costs for future soil
abatement activities. Similarly, of the $52,307 sprat for miscellaneous extra costs
($l,453/property), only the $2,425 ($67/property) spent for pre-abatement yard cleaning may
be applicable for future soil abatement activities, although it might be argued that the cost for
hoses, sprinklers, and taips may also be needed for future soil abatements. If these costs are
included miscellaneous costs totalled $7,627, or $220/property.
While $26,190 was spent on soil disposal in 1989 ($725/property), soil disposal is
likely to be the most variable and unpredictable expense associated with future abatements
and it may be more useful to estimate abatement costs without including the cost of soil
disposal.
For the 36 properties abated in 1989, the total cost for soil sampling and analysis
($10,933), pre-abatement yard cleaning ($2,425), soil excavation and replacement
($186,420), and supervision of abatement activities was $219,179 or approximately $6,100
per property. The cost per square meter of soil abated and replaced was $31, and the cost
per cubic yard of soil replaced was $140. In 1990, the total cost of soil sampling and
analysis ($17,535), pre- abatement yard cleaning ($1,550), soil excavation and replacement
($307,995), and supervision of abatements ($39,247) was $366,327, or $6,315 per property.
The cost per square meter of soil abated and replaced was $35, and the cost per cubic yard
of soil replaced was $143. Thus, the average costs were quite similar in 1989 and 1990.
15-37
-------
TABLE 15-30. 1989 SOIL ABATEMENT COSTS
1. Soil Sampling and Analysis $3,780
Labor, Sampling 36 Properties @ $105 Each 36
2 Core Sampling Tubes @ $17.95 Each 74
2 Core Sampling Tubes @ $36.80 Each 25
1,000 Plastic Bags @ $25 Per 1,000 200
Miscellaneous Supplies 6,818
Analyses of Approximately 20 Samples/Site $ 10,933
@ $9.47 Each x 36
Total for Soil Sampling and Analysis
2. Contract Development/Supervision by Applied
Occupational Health Systems (AOHS)
Development $ 20,503
Pilot Abatement 9,450
Abatement Supervision 19,401
Dosimeters 365
Final Report 2,449
LFK Field Operations Coordinator 6 Months 17,500
Total Contract Development/Supervision $ 69,668
3. Abatement Contract 157,740
33 Properties @ $4,780 Each 28,680
3 Properties @2x $4,780 Each $ 186,420
Total Abatement Contract
4. Soil Disposal
Use of Barry's Quarry
Mattapan Costs, Prep, Clean-up
Total Soil Disposal
5. Miscellaneous Extra Costs $ 2,425
Yard Cleaning Pie-abatement 1,402
Hoses and Sprinklers 3,800
Extra Poly Taips, 19 Sites 30,795
Cold Weather Abatement 13,885
Asphalt, 4 Properties
Total Miscellaneous Extra Costs
Total Cost for 36 Soil Abatements in 1989 $345,518
Average Cost Per Property, including all factors listed above $ 9,598
Cost Per Square Meter of Soil Abated $ 48
Cost Per Cubic Yard of Soil Replaced $ 218
15-38
-------
TABLE 15-31. 1990 SOIL ABATEMENT COSTS
1. Soil Sampling and Analysis
Labor, 58 Sites Sampled @ 105 Per Site $ 6,090
2 Core Sampling Tubes @ 17.95 Each 36
2 Core Sampling Tubes ฎ 36.80 Each 74
2,000 Plastic Bags @ $25 Per 1,000 50
Miscellaneous Supplies 30
Analyses of approximately Samples/Site 0
@ $9.47 Each x 58 10,985
Total for Soil Sampling and Analysis $ 17,535
2. Supervision of Abatements
3 Site Monitors 3 Months Each $ 25,550
Travel for Site Monitors 1,697
LFK Field Operations Coordinator 4 Months 12,000
Total for Supervision $ 39,247
3. Abatement Contract
51 Properties @ $4,738.38 Each $241,657
7 Properties @ 2 x $4,738.38 Each 66,337
Total Abatement Contract $307,995
4. Soil Disposal
Use of Barry's Quarry $ 2,500
Materials for Temporary Storage 1,724
Taking Stored Soil to Quarry 1,400
Site Monitor at Quany 1,528
Gravel for Quany Driveway 485
Bulldozer Rental for Covering Soil - 3 Months 9,350
Total Soil Disposal $ 16,987
5. Miscellaneous Extra Costs $ 1,550
Yard Cleaning/Change Orders 655
Hoses and Sprinklers 795
Transit Level Rental $ 3,000
Total Miscellaneous Extra Costs
Total Cost of 58 Soil Abatements in 1990 $384,764
Average Cost Per Property, Including all Factors Listed Above $ 6,634
Cost Per Square Meter of Soil Abated $ 37
Cost Per Cubic Yard of Soil Replaced $ 150
15-39
-------
15.5.2 Interior Loose Paint and Dust Abatement
Actual cost for units that received interior paint abatement alone, and for units that
received both interior paint and dust abatement are presented. Actual cost of dust abatement
alone cannot be provided as no units in the study received dust abatement in the absence of
interior loose paint abatement. Possible cost for dust abatement alone is estimated using two
different approaches. Costs are provided for the abatement activities, and for the costs
associated with pre-abatement preparation, abatement monitoring, and the costs associated
with cancellations. The contractor who performed the interior loose paint and dust
abatements charged different unit rates depending on unit size. Actual costs for loose paint
abatement and loose paint and dust abatement, and estimates of costs for dust abatement
alone are therefore provided by the cost category charged by the contractor. Average costs
are also provided.
A total of 129 units had interior abatements: 40 units had only loose paint abatement,
and 89 units had loose paint and dust abatement.
While participants were asked to prepare their units for abatement activities it quickly
became apparent that most could not accomplish this. Thus, the contractor hired to do the
interior abatements was paid to prepare the units at a rate of $20/hour. This included
moving all furniture items to the middle of the room. A total of 407.25 hours at $20/hour
($8,145) was spent on preparation activities for the 129 units that received interior
abatements. In calculating cost estimates we have used the average unit preparation cost as
$63/unit ($8,145/129 units). This assumes that these costs did not vary by size of unit.
All interior abatement activities were monitored by study staff. It is estimated that the
total cost of monitoring was $15,832. Although the monitoring cost may have varied by unit
size and abatement category (loose paint abatement alone took approximately 1/2 day per unit
whereas loose paint and dust abatement took approximately one day per unit to complete) the
average cost of monitoring was used ($15,832/129 or $123/unit).
15-40
-------
Interior Loose Paint Abatement Costs
Abatement Work
Preparation
Monitoring
Cancellations
32 units @ 499/unit $15,968
8 units @ $988/unit 7.904
Total for 40 units $23,872
40 units @ $63/unit $ 2,520
40 units @ $123/unit $ 4,920
4 units @ $499/unit $ 1,996
Thirty-two units were abated at a cost of $499 per unit. Eight units were considered
oversized and were abated at a cost of $988 per unit. Thus, for 75% the cost of loose paint
abatement was $499 and for 25% the cost was $988. The average cost for all 40 units was
$597/unit. These figures do not include the costs of preparation work, monitoring, or the
cancellations.
If the costs of preparation work ($63/unit) and monitoring ($123/unit) are included, the
cost for interior loose paint abatement was $685/unit for 75% of the units, $l,174/unit for
25%, and the average cost was $783/unit.
If the costs of cancellations are added, then each figure would be increased by $50
($1,996/40).
Interior Loose Paint and Dust Abatement Costs
Abatement Work - 84 units @ $873/unit
4 units @ $l,748/unit
1 unit @ $1,310
Total for 89 units
Cancellations - 9 units @ 873/unit
$73,332
6,992
1.31Q
$81,634
$ 7,857
For 84 units the cost of loose paint and dust abatement combined was $873, for four
units the cost was $l,748/unit, and for one unit the cost was $1,310. The average cost for
all 89 units was $917/unit. These figures do not include the costs of preparation work,
monitoring, or cancellations.
If the costs of preparation woric and monitoring are included the respective costs are
$1,059, $1,934, and $1,496 with an average cost of $1,103. If the costs of cancellations are
added, and cancellations were common, occurring in 10% of cases, then each figure above
should be increased by $50.
15-41
-------
15.5.3 Interior Dust Abatement Costs
No units in the study underwent dust abatements without associated loose paint
abatement. The figures presented therefore represent hypothetical cost estimates of interior
dust abatements. Estimates are provided of costs estimated in two different ways.
First, the cost per unit for loose paint and dust abatement was divided in half as it took
approximately 1/2 day to do loose paint abatements alone and approximately one day to do
loose paint and dust abatement. Among the 89 units that had both loose paint and dust
abatements, the estimated cost of dust abatements alone were: $437 each for 84 units, $674
each for four units, and $650 for one unit. The estimated average cost was $458 overall not
including the costs of preparation and monitoring. If preparation and monitoring costs are
included the respective costs were $525, $867, and $748 with an average cost of $552
overall.
Secondly, the average cost for dust abatement was estimated by subtracting the average
cost of loose paint abatement alone, with and without the costs of preparation and
monitoring ($783 and $597, respectively), from the average cost of loose paint and dust
abatement combined, again with and without the cost of preparation and monitoring included
($1,103 and $917, respectively). By this method, the average cost of dust abatement was
$134 without including preparation and monitoring costs and $320 if they were included.
All estimates provided for the cost of interior loose paint, loose paint and dust, and dust
abatement do not include costs associated with identifying units in need of abatement,
recruiting landlords, making arrangements for families to be off the premises during
abatement activities, and pre or post-environmental sampling.
15.5.4 Deleading Costs
Interior and exterior deleading activities are described in detail in another section of the
report. The study offered to pay in full the cost of exterior and interior deleading for owner
occupied units, and $2,000 towards the cost of these activities for non-owner occupied units.
Non-owner occupied properties were viewed as businesses and therefore were believed to be
responsible for bringing their properties into compliance with the Massachusetts Lead Law
and so were not offered full coverage of the deleading costs. A total of 46 exterior and
46 interior deleading operations were facilitated and paid for in total or in part. These
15-42
-------
92 operations were performed by four licensed deleading contractors under eight separate
contracts. One contractor subcontracted some of the woric to a fifth licensed deleader.
The study incurred a variety of costs that may or may not be relevant to or included in
cost estimates of deleading activities, such as those associated with lead based paint
inspections, including the purchase of a portable XRF machine, respirator use by inspectors
while monitoring deleading (monitoring is not currently required in Massachusetts), clearance
sampling to assure that the units were free of lead contaminated dust (this is also not required
in Massachusetts but it is conducted if dust is discernible on visual inspection), moving and
furniture storage charges, and alternate housing for families during deleading. Many of these
costs may not be applicable to deleading activities undertaken as part of the environmental
management of children with elevated lead levels, and they may in large part reflect
idiosyncrasies associated with this study. Many of these costsjnay, however, be quite
relevant to future endeavors where deleading is undertaken on a large scale or as part of a
comprehensive approach to the primary prevention of low level lead toxicity among children.
All actual cost estimates are presented.
Costs Associated with Lead Paint Inspections
Development of the inspection process
$l,800/wk x 10 weeks
Conducting Inspections of units
$2,000/wk x 12 weeks
Monitoring Deleading Activities
$2,000/wk X 24 weeks
Cost of One Portable XRF Machine
Total Cost of Inspecting and Monitoring
$18,000
24,000
48,000
4.147
$94,147
As described earlier, for the purposes of this study it was necessary to hire private lead
paint inspectors. In other situations, code enforcement inspectors may woric for public
regulatory agencies, such as health departments, or families who do not have children with
elevated lead levels but want inspections may hire licensed private inspectors. Three lead
paint inspectors and one assistant accomplished the inspection related activities. They had
the additional assistance of an intern inspector who woiked on the project.
15-43
-------
The total st of inspector related activities for this study was $94,147, or
approximately $1,000 per deleading operation. This figure is probably substantially higher
than usual because of factors unique to this study. For example, the $18,000 spent to
develop the inspection process is most likely best viewed as a cost idiosyncratic to this study,
or as a one time cost that might be incurred by other cities or projects that were starting up
large scale deleading activities. Similarly, private lead inspectors do not typically monitor
deleading activities, but rather perform lead inspections to determine the need for deleading
and to issue certificates of compliance after deleading has been completed. The cost of the
portable XRF could also be viewed either as a one time cost, or not included as a cost since
any lead related regulatory agency must have this type of equipment. Therefore, it is
probably most realistic to include only the costs of actually conducting inspections of units in
arriving at the total cost of deleading.
Ninety-two deleading operations were performed, and the cost of lead paint inspections
associated with these operations was $24,000, or approximately $260 per operation. Since
certificates of compliance are issued only after both interior and exterior deleadings in
Massachusetts, a more accurate estimate of this inspection/compliance cost might be $520 per
unit deleaded.
Costs Associated With Actual Deleadings
46 Exterior Deleading Operations $262,278
46 Interior Deleading Operations 343 242
Total Cost of Exterior and Interior Deleading $605,520
The average cost of an exterior deleading operation was $5,702 and the average cost of
an interior deleading operation was $7,462. Thus the average total cost per unit of both
interior and exterior deleading operations was $13,164.
Cost of Respirators
4 PAPT Respirators @ 546/each $ 2,186
15-44
-------
Respirators were purchased for use by inspectors during deleading monitoring. While
it would be accurate to factor their cost into the study related deleading costs, this is a one-
time cost and is not regularly required for lead paint inspectors. Deleaders are required to
wear respirators while performing deleading operations.
Cost Associated With Moving and Storage
41 Moves @ 720/each $29,520
5 Cancellations @ 360/each 1,800
Extra Storage 2.346
Total Cost $33,666
$33,666 was spent on moving and storing the property of 41 families. Five other
families were able to find alternative housing without any financial assistance from the
project. Thus, to accomplish 46 interior and 46 exterior deleading operations, 41 required
financial assistance with temporary moves and the average cost for those requiring moving
assistance was $820 per family.
Cost Associated With Alternative Housing
Seven families could not locate suitable alternative housing and the study identified and
paid for temporary housing for those individuals. A total of 78 nights of alternative housing
was provided for these seven families at a total cost of $11,612, or approximately
$l,650/family.
Cost Associated With Clearance Samples
Materials (centrifuge tubes and wash and drys) $ 1,476
Acid dispenser and centrifuge shaker 997
Analyses of 619 samples at $14.50/each 8.975
Total Cost $11,448
A total of $11,448 was spent obtaining and analysing clearance samples. No estimate
is available of the cost associated with the inspectors' time involved in obtaining clearance
samples. The average cost of performing clearance sampling for the 46 interior units is
therefore approximately $250/unit deleaded. Since it is not currently standard public health
15-45
-------
practice to conduct clearance sampling with all interior deleading operations, it may be best
to exclude the cost of clearance sampling in estimating total costs of deleading operations.
15.5.4.1 Total Deleading Costs
$741,712 was spent for the 46 interior and 46 exterior deleading operations. This
includes the cost associated with inspecting and monitoring deleading activities, the actual
deleading activities, financial assistance with moving and storage, alternative housing for
selected families and clearance samples for all interior deleading operations. Thus, the
average cost per combined interior and exterior deleading operation was $741,712/46, or
approximately $16,124 per combined operation. If the cost of moving, storage, and
alternative housing are removed, the total cost was $696,434/46, or approximately $15,000
per combined operation. If the cost of moving, storage, alternative housing, and inspections
and clearance samples are removed, the total cost was $594,072/46, or approximately
$13,000 per combined operation. This cost may be likened to the cost of deleading a single
family home, or one interior operation and an exterior deleading operation in a multiunit
home. Deleading of subsequent units in multiunit homes would have costs closer to that of
interior unit deleading alone.
Separate costs associated with exterior deleading operations may be about
$5,700/exterior operation, although there is no data to suggest how comparable costs would
be in other cities with different sizes or types of homes. The average cost per interior
deleading activity, not including the costs associated with the unique situations of this study
(e.g., moving, storage, alternative housing, inspections, monitoring, and clearance samples)
was $7,500/interior deleading operation. If one assumes that many of the unique costs are
applicable to future large scale deleading activities in other communities, the actual cost may
be closer to $10,000-$10,500 per unit.
15-46
-------
16. DISCUSSION
One of the most difficult aspects of the childhood lead problem is identifying the
sources of lead and determining their relative contribution to children's lead burden. Lead
based paint and household dust have received most of the attention to date. Far less attention
has been paid to urban outdoor sources of lead, especially soil, except in cases of stationary
sources such as smelters. Our findings suggest that lead contaminated soil does contribute to
the blood lead levels of urban children. We found that soil abatement alone (Study vs.
Control Group A) was associated with a 0.8 to 1.4 /zg/dL decline in blood lead levels and
that soil and interior dust abatement combined (Study Group vs. Control Group B) was
associated with a 1.2 to 1.6 /xg/dL decline. (These numbers are the range of adjusted point
estimates.) These blood lead changes were observed approximately one year following soil
abatement in which surface soil lead levels were dropped an average of 1,856 PPM.
Numerous previous studies have shown that soil and dust lead levels are correlated with
children's blood lead levels.4,6"10'12"14'19"25 These studies have relied largely on cross-
sectional data, often from communities with point sources of lead such as smelters, where
soil lead concentrations were far greater than those typically found in urban settings. Many
of the smelter area studies were conducted in response to crises and were not designed as
research studies so that important design features such as study size and timing of
intervention could not be planned. These studies have produced widely differing results with
slope estimates of the soil lead - blood lead relationship that vary over nearly an order of
magnitude.25
Removal of lead contaminated soil in this study was associated with a 0.8-1.6 ptg/dL
reduction in children's blood lead levels, suggesting that uiban soil lead is biologically
available and contributes to low level lead absoiption in children. The clinical and public
health implications of a reduction of this magnitude are not readily apparent. The magnitude
of reduction in blood lead observed suggests that lead contaminated soil abatement may not
be a particularly useful clinical intervention for children with low level lead exposure.
It might be extremely useful, however, in specific situations, such as if soil lead were
extremely high-or the particular child bad pica for soil. It is also a relatively inexpensive
16-1
-------
and low technology intervention. Although there are no data regarding the relative safety of
soil and lead based paint abatement, it seems unlikely that soil abatement is as dangerous to
children, families, and workers as lead based paint abatement can be.
16.1 STUDY PROBLEMS AND THEIR RESOLUTION
16.1.1 Recruitment and Retention of Study Participants
A potential problem was the recruitment and retention of study participants. The
success of the study depended, in part, on recruiting and retaining sufficient numbers of
participants who were representative of the general population of urban preschool children
who are at risk for low level lead exposure. These concerns were important so as to provide
a large enough sample so that we had sufficient power to test the study hypothesis and be
generalizable to other children. The issue of generalizability was addressed by using the
Boston Childhood Lead Poisoning Prevention Program (BCLPPP) for identification of
children. This program has data on the majority of children screened in the neighborhoods
of Boston of interest (i.e., those neighborhoods with the highest rates of lead poisoning). We
also attempted to improve generalizability by choosing as wide a range of blood lead levels
as was practical. That is, we could not choose children with blood lead levels greater than
24 jtg/dL because of concern that they would receive medical and possibly environmental
interventions that might confound study results. We chose as a lower limit blood lead values
of 7 /xg/dL because of concern that it would be difficult to ascertain the effect that soil
abatement would have on lower blood lead levels. Our range for blood lead levels at entry
was therefore 7-24 /*g/dL. A related concern was that the BCLPPP screening data were
derived from fingerstick lead tests. We addressed this by confirming all potential subjects'
blood lead levels with venous blood samples before final enrollment.
Recruitment of participants was further supported by six approaches:
1. An active and visible community relations program and subject education effort was
mounted. This ensured that residents of the target communities were aware of the
lead poisoning problem in their communities, the risks that lead poisoning posed for
their children, and of the program at the time that study staff attempted to recruit
them. In addition, study staff were educated in the epidemiology, long-term effects,
prevention, and treatment of childhood lead poisoning so that they could discuss
these issues with potential subjects and convince them of the importance of the study
and its potential benefits to their children and community.
16-2
-------
2. Each month families were given $25 gift certificates for local supermarkets and
general purpose stores as long as they participated in the study. Even if they moved
they were eligible for these incentives as long as they stayed in touch with the study
staff and agreed to provide access for environmental and biologic sampling and study
interviews. All families completing the study were given a $150 gift certificate.
3. Study staff were enthusiastic, well trained, well supported by the study's
management, often experienced in home visitation, and frequently came from the
target communities. This led to a close and effective rapport with participating
families.
4. Families were not enrolled if during intake they stated that they had plans to move
during the next three months.
5. Study activities were very intrusive and disruptive to families and we made every
effort to minimize family disruption by scheduling study activities at their
convenience, taking children to museums and restaurants during interior abatement
work, and offering alternative housing if necessary during interior lead paint
abatement.
6. There was great concern about landlord consent to participate and the early stages of
recruitment supported this concern. That is, landlords were initially reluctant to
participate because of concern that if lead paint were found on the interior surfaces
of their houses, they would be forced to pay for deleading. This problem was
addressed by offering landlords of non-owner occupied premises $2,000 towards the
cost of interior lead paint deleading and landlords of owner occupied premises the
full cost of interior lead paint deleading. Moreover, we pointed out that (1) these
properties were not in compliance with Massachusetts law and that at some time in
the future they would have to be deleaded; (2) the study would facilitate and pay for
part or all of the cost; and (3) if the landlords did not delead at the end of the study
and a child became lead poisoned, they might be found liable if the family chose to
sue them. We also pointed out that the soil contained high levels of lead and that we
would remove this soil at no cost to them.
16.1.2 Lead Contaminated Soil Disposal
A great deal of energy went into identifying a location to dispose of lead contaminated
soil. After exploring multiple options, some of which were not used because of distance
from the excavation site or political concerns, a quarry was identified that abutted a cemetery
in a Boston neighborhood not involved in the study. This worked well until the City
councillor from that neighborhood raised concerns about the potential hazard of this soil to
this neighborhood. This problem was resolved by temporarily storing the abated soil on a
city-owned property while the EPA Project Manager, the principal investigator and other
16-3
-------
members of the Lead Free Kids Staff, and representatives from the Mayor's Office met with
the City councillor and concerned citizens and convinced them that dumping the soil in the
quarry and covering it with unleaded soil posed no risk to residents of this community.
16.1.3 Limited Funding
Early in the study the Boston Lead-in-Soil Project was subjected to a cut in funds
available from the EPA due to the other projects having budgetary needs that had to be
addressed. This was dealt with by calculating the minimum number of families who needed
to be recruited and retained to have sufficient power to test the study hypothesis. The CDC
was helpful in supporting these estimates. This budget cut, plus the need to offer landlords
substantial incentives to participate, led us to abandon the cluster arm of the Study Group and
focus only on the effects of abating individual properties.
16.1.4 Concerns About Ethical, Legal, and Logistical Constraints
A series of very complicated ethical, legal and logistical constraints, documented in
detail in documents produced by Region I of the EPA, submissions to and correspondence
with the Institutional Review Board of the Trustees of Health and Hospitals, and
correspondence with the Massachusetts Department of Public Health, led to great concern
about the feasibility of carrying out a scientifically rigorous study in Boston. This problem,
or series of problems, was dealt with by assembling a credible and capable leadership team
with an established and respected record of scientific and public health accomplishment. Hie
team assumed leadership for all aspects of the study, met regularly, provided daily oversight
of all study activities, and worked closely and effectively with local and national EPA and
public health officials, lead advocates, and nationally renowned leaders in the field of
childhood lead poisoning research and treatment. This is a very truncated discussion of a
substantial number of very complicated legal and ethical issues.
16.1.5 Frozen Ground During Soil Abatement of the Study Group
The ground froze during lead contaminated soil abatement of the Study Group in the
Winter of 1989-1990. The study had a very narrow window of time in 1989 in which to
accomplish soil abatement for the Study Group. Before these abatements were completed the
i
16-4
-------
soil froze, as Boston experienced the coldest December in recorded history. This problem
was addressed by using jack hammers to loosen soil so that abatements could proceed.
16.2 LIMITATIONS
Although designed and conducted to produce rigorous results, the study has a number
of limitations that deserve mention:
16.2.1 Relatively Small Sample Size
Despite a sample size with adequate power to detect the hypothesized overall effect of
the intervention, the relatively small sample size did result in a number of limitations:
1. Randomization was undertaken to maximize the probability that all three groups were
comparable as regards measured and unmeasured characteristics. Because of the
small sample size, randomization did not result in groups that were entirely
comparable at baseline. In the analyses, we adjusted for the measured variables that
were potential confounders.
2. Outliers had a greater influence on the study results. Two siblings in the Study
Group had significant increases in blood lead levels between the first and second
post-abatement sampling rounds. We hypothesized that they became poisoned
because they were spending time at the father's home that had lead-based paint and
was being renovated. The crude analysis was conducted both with and without these
children and all results are reported. If the sample size were larger, however, the
influence of the outliers would have been attenuated.
3. The relatively small sample size limited the stability of our stratified analyses on
children with particular characteristics. For example, the small sample size resulted
in a limited number of children with blood lead levels of IS /xg/dL and greater at
baseline. Thus, our estimates are unstable regarding the effectiveness of soil
abatement among children with lower versus higher starting blood lead levels.
16.2.2 Follow-up Limited To One Year
There are virtually no data available on the rate of change in children's blood lead
levels following a change in lead exposure. It is possible that the intervention would have
been associated with a greater reduction in children's blood lead levels had we followed them
for a longer period of time. We have consequently applied to the EPA for a no cost
lfc-5
-------
extension to obtain blood lead levels on participants who remained on the study premises
during the summer of 1991. This will allow us to compare the blood lead levels of children
who had soil abatement two years ago to those who had soil abatement during the fall of
1990.
16.2.3 Mobility Of Families
Twenty two families (14.5%) moved by the second post-abatement blood sampling
round, and more families from the Study Group moved (20,4%) than families from Control
Group A (15.7%) or Control Group B (6.4%). We followed all of these families for the
duration of the study and obtained blood and environmental samples whenever possible. In
addition it did not appear that the movement of the families reduced the magnitude of the
treatment effect. Children who lived on the study premises for at least 300 days had a
similar reduction in blood lead levels as the entire group.
16.2.4 Limitations Resulting From Study Design
Several aspects of the design of the study may have limited the observed effectiveness
of the intervention.
1. All children in the study, irrespective of group assignment, were exposed to lead
contaminated soil prior to enrollment. An alternative study design which would have
been logistically more difficult to execute would have involved conducting lead
contaminated soil abatement prior to biith. Such a design would have enabled us to
investigate whether exposure to lead contaminated soil abatement in the first year of
life is associated with lower blood lead levels.
This study provides information about soil abatement as a secondary prevention
strategy, that is the benefit to children already exposed to lead derived, in part, from
contaminated soil. It can not be used to estimate the primary prevention effect of
soil abatement. Since children's post-abatement blood lead levels reflect both recent
exposure and body burdens from past exposure, the benefit observed is probably less
than the primary prevention benefit, that is the benefit of abating lead contaminated
soil before children are exposed to it so as to prevent increases in blood levels and
body stores.
16-6
-------
2. Lack of Cluster Groups - Due to budgetary constraints and difficulties enrolling
landlords, we abandoned our original plan to study clusters as the unit of abatement.
We therefore evaluated only the effect of single premises abatements. It is possible
that the effect of lead contaminated soil abatement on children's blood lead levels
would be even greater had we abated lead contaminated soil from properties that
surrounded Study Group children's premises.
3. Study staff regularly visited all participating families and provided education about
lead poisoning, and while educational efforts were identical among the groups, this
may have resulted in decreased group differences.
4. Children were already 31 months old, on average, at the outset of the study, well
above the age at which mean blood leads are highest.32'33
These limitations all would tend to drive the results toward the null (Type n error),
rather than produce false positive results (Type I error), making it likely that the study
underestimates the full impact of urban soil abatement.
16.2.5 Limitations To Generalizability
There are limitations to the generalizability of the results stemming from the
characteristics of the study population. For example, the abatement might have had a
different effect among children with more or less exposure to soil lead. It might also have
been different among children of higher socioeconomic status because of better diets,
foundation shrubbery, more grass cover, or other reasons. The results therefore can be
generalized to inner city children 1 to 4 years of age who have soil lead levels greater than
1,500 PPM, blood lead levels of 7 to 24 pg/dL, whose families' place of residence are
reasonably stable (only 15% moved during the course of this study), and whose exterior lead
paint is in fairly good condition. The study provides no information about the effect of lead
contaminated soil abatement for children with lead levels outside of the eligible range (7 to
24 pg/dL). Similarly, the results may not be generalizable to children who live in
communities with smelters or other stationary sources where soil lead levels are substantially
higher than those seen in this study, or where differences in particle size result in differences
in bioavailability.5
16-7
-------
16.2.6 Misclassification
Errors in the sampling and measurement of lead in the environmental media and the
blood and handwipes may have resulted in exposure and outcome misclassification. Because
these errors were just as likely to occur in the Study Group as the Control Groups, they are
more likely to result in bias toward the null than toward falsely positive results. We suspect
that handwipe data was subject to non-differential misclassification because of sampling
problems (e.g. the parent may have washed the child's hands shortly before sampling), and
the highly variable background lead levels in the wipes (ranging from 2 to 18 /xg).
Deficiencies in parental memory and report may have led to inaccuracies in the
interview data. Most of the variables collected at interview were considered potential
confounders of the relationship between soil abatement and blood and hand lead levels.
Therefore any misclassification would have reduced our ability to control for confounding.
16.3 IMPLICATIONS OF FINDINGS
Although the average benefit associated with abatement of lead-contaminated soil is
modest, the societal impact may be substantial. Consider, for example, the impact on the
blood lead distribution of an average decline of 1 or 2 /xg/dL in the mean blood lead level of
a population of children assuming a starting mean blood lead level of 12 pg/dL, a standard
deviation of 4, and a normal distribution (Table 16-1). We also assume that the amount of
change (as opposed to the percentage of change) is constant for all starting values, as we
observed in our own sample in which the distribution of starting values was truncated.
A decline of 2 /xg/dL in the mean blood lead level results in 72% as many children with
levels exceeding 10 jtg/dL, 47% as many children with levels exceeding 15 /xg/dL, and 26%
as many children with levels exceeding 20 /xg/dL (values of 10, 15, and 20 pg/dL were
chosen because they correspond to the new CDC definition of lead poisoning, the new
screening guideline, and the new action level for medical intervention). Even a 1 /xg/dL
decline in mean blood lead level results in 87%, 70%, and 52% as many children with levels
of 10, 15, and 20 figldh, respectively. The percentage shifts may differ somewhat in a more
representative sample in which the distribution of starting values is likely to be log normal.
16-8
-------
TABLE 16-1. PERCENTAGE OF CHILDREN EXPECTED TO HAVE
BLOOD LEAD LEVELS EXCEEDING 10, 15, AND 20 jtg/dL
ASSUMING VARIOUS MEAN BLOOD LEAD LEVELS *
Mean
% > 10 /xg/dL
% > 15 /xg/dL
% > 20 /xg/dL
12
69.1
22.7
2.3
11
59.9
15.9
1.2
10
50.0
10.6
0.6
*
A constant standard deviation of 4 pg/dL is assumed for all mean blood lead levels.
Hie results of this study suggest that lead contaminated soil contributes to the lead
burden of urban children and that abatement of lead contaminated soil around their homes
results in a modest decline in blood lead levels. Thus it may be prudent to include soil
inspection and abatement as part of primary prevention strategies in communities with high
rates of childhood lead poisoning and as part of the environmental intervention on behalf of
selected lead poisoned children.
Policy decisions regarding urban lead contaminated soil abatement as a lead control
strategy will require numerous considerations. For example, are other types of remediation
(e.g., planting grass cover and shrubs) equally effective but less expensive and intrusive?
How does the cost effectiveness of soil abatement compare to other lead exposure reduction
activities, such as paint abatement? Will it be practical to perform large scale abatements
without encountering problems regarding the disposal of lead contaminated soil? Will future
research help specify whether changes in children's blood lead levels of the magnitude seen
in this study are clinically relevant or prudent from a public health or societal perspective?
And will we develop and sustain the resolve and commit the resources needed to prevent
what remains the most important environmental health problem of children in the United
States?
16.4 ONE YEAR EXTENSION
In December, 1990 the investigators requested that unexpended funds that resulted from
paint deleading refusals be used to support a no cost extension for one year. In May 1991
16-9
-------
the extension was granted and signed. The extension will be used to accomplish a number of
related objectives.
The first objective is to conduct detailed analyses of data already collected that are vital
to our understanding of how lead contaminated soil abatement affects children's blood lead
levels. A great deal of environmental and child based data were collected as part of this
study. As anticipated from the outset of the project, only a selected number of analyses
directed at answering the primaiy hypothesis posed by the project, that is, in the aggregate,
did lead contaminated soil abatement result in significant reductions in children's blood lead
levels, could be conducted by the end of May, 1991. The extension will enable the
investigators to complete more detailed analyses regarding the impact of behavioral and
environmental variables on the change in blood lead levels.
Second, analyses conducted to date suggest that soil abatement was associated with a
0.8-1.6 /xg/dL reduction in children's blood lead levels, somewhat less than what was
originally hypothesized. The clinical and public health implications of a reduction of this
magnitude are not readily apparent. It is possible that larger differences in mean blood lead
levels between the experimental and Control Groups may be found at two years post-
abatement. If this is the case, then it might imply that this environmental intervention is
prudent public policy. The one year extension enabled the investigators to obtain an
additional blood and hand lead level measurement among children who still live at their
original premises. This will also allow the investigators to examine the impact of paint
deleading on blood lead levels. Additional soil and dust samples will also be obtained and
analyzed for lead content so that recontamination can be further studied. The possibility of
obtaining a fourth blood lead level if the results were inconclusive or if the financial
resources were available was discussed in the original grant application submitted in August,
1988.
A one year extension also enables this project to be completed simultaneously with the
other projects in Baltimore and Cincinnati and will facilitate our input in understanding how
the data from each of the three projects complement each other. It also ensures that the
investigators will be available to work with EPA officials in writing the final report to
Congress combining the results of all three projects.
16-10
-------
The Trustees of Health and Hospitals of the City of Boston has agreed to continue
housing the grant. Michael Weitzman, M.D., although having moved to the University of
Rochester School of Medicine and Dentistry, will continue to be the principal investigator,
and in that capacity will continue to be responsible for the overall operation of the project
during the extension period. Ann Aschengrau, Sc.D. at Boston University will continue to
oversee the day-to-day data collection and analyses and she, along with Michael Weitzman,
M.D., David Bellinger, Ph.D. of Harvard Medical School, and Alexa Beiser, Ph.D. at
Boston University, will collaborate on the production of all reports.
16-11
-------
Boston Lead Free Kids Study
Protocols and Other Documents
A. Protocols
1. Blood Sampling and Processing
2. Blood Lead Analysis
3. Erythrocyte Protoporphyrin Analysis
4. Handwipe Sampling
5. Handwipe Analysis
6. Child Height Measurement
7. Child Weight Measurement
8. Soil Sampling
9. Soil Lead Analysis
10. Interior Dust Sampling
11. Interior Dust Lead Analysis
12. Water Sampling
13. Water Lead Analysis
I
14. Lead Paint and Site Inspection
15. Interior Dust Abatement
16. Interior Loose Paint Abatement
17. Quality Assurance Plan: Soil and Dust Analyses
18. CDC External Quality Assurance Plan: Blood Lead Analyses
19. US EPA/EMSL External Quality Assurance Plan: Soil, Dust, and
Handwipes
-------
BOSTON LEAD FREE KIDS STUDY
PROTOCOL FOR SAMPLING BLOOD FOR LEAD ANALYSIS
Venipuncture Method
1. Make sure the consent form is signed.
2. Educate the patient according to their level of comprehension, with pgrent present.
3. Assure the patient of minimal discomfort.
4. Inspect the patient's arm and hands for best venipuncture site.
5. Determine the best method of venipuncture for the patient (butterfly 18 ga. or
conventional needle 21 ga. assembly).
6. Clean venipuncture site using Becton Dickinson (B-D) alcohol prep until alcohol prep
shows clean. Let air dry or dry with clean gauze.
7. Be sure the patient is properly restrained.
8. Apply tourniquet.
9. Don gloves.
10. Palpate for vein.
11. Clean site again and dry after palpating.
12. Insert needle assembly.
13. Draw 2 B-D pediatric 3 ml vacutainer evacuated tubes with EDTA preservative. Mix
well; invert 3-5 times.
14. Loosen tourniquet before last tube is full or before withdrawing needle.
15. Withdraw needle.
16. Apply pressure to venipuncture site until bleeding is stopped, then apply band-aid.
17. Write in patient's name, other coded information, and sign labels. Attach labels to
tubes.
18. Put tubes in cooler.
Processing Equipment:
1. Consent form
2. Butterfly 18 ga. or 21 ga. by 3/4", 12" tubing infusion set, vacutainer multiple sample
Luer-Adapter, Becton Dickinson vacutainer with EDTA preservative, arid vacutainer
holder.
3. Becton Dickinson (B-D) alcohol swab.
4. Tourniquet
5. Cooler with "Blue Ice" packs to keep sample cool.
6. Trained and qualified person to obtain blood samples (i.e., medical technician, nurse,
etc.).
All waste materials should be red-tagged. Upon failure to achieve venipuncture, the
alternative tingerstick procedure as described by CDC may be used.
-------
Revised 8/89
Boston Lead-in-Soil Demonstration (Lead Free Kids) Project
Whole Blood Collection and Processing Protocol
For Blood Lead and FEP determinations
Blood
One 7 ml Lavender Top Vacutainer
i
Invert Several Times To Ensure Proper Mixing
I 1
Blood Lead Blood FEP
4
Initiate Chain Of Custody Document
i
Refrigerate at 4ฐC Within 30 Minutes Of Collection
And Store At Same Temperature Until Shipment
1
Pack Blood in Styrofoam Shipping Containers Provided By Laboratory
i
Transport Blood Via Lead Free Kids Messenger
to Environmental Science Associates (E.S.A.) Laboratories
of Bedford, Massachusetts
-------
LEAD FREE KIDS STUDY
BLOOD LEAD ANALYSIS
ANALYTE:
Lead
MATRIX:
Blood
LABORATORY: ESA Laboratories, Bedford, MA
PROCEDURE: Graphite Furnace Atomic Absorption (GFAAS)
METHOD: A 1:12 dilution with a Matrix Modifier, FGAAS
RANGE:
0-50 /ig/dL
DATE:
1. PRINCIPLE OF THE METHOD
1.1 A 100 fxL aliquot of well mixed blood is diluted to 1.2 mL with a matrix
1.2 The sample is then run by GFAAS using polarized Zeeman background
2.1 The sensitivity of this method is 1 fig/dL. The upper range is 50 ^ig/dL,
samples above this range should be diluted and re-run.
3. INTERFERENCES
3.1 Normal constituents of blood and urine do not interfere. Zeeman background
correction will adequately correct for all background interference at this
dilution.
modifier.
correction.
2. RANGE AND SENSITIVITY
4.
REAGENTS
4.1 Triton X-100
4.2 Dibasic ammonium phosphate (NHA) 2HP04
-------
4.3 16 M nitric acid
4.4 Matrix Modifier - To approximately 1000 mL of DIW stir in 20 mL of Triton-X
and 4 grams of dibasic ammonium phosphate. Then add 4mL of NH03 and
dilute to 2000 mL with DIW.
5. STANDARDS
5.1 Stock Standard - 1000 /xg/mL Pb Fisher certified or equivalent.
5.2 Working Standard - 10 /Ag/mL Pb - Dilute 1 mL of the Stock Standard to 100
mL with DIW.
5.3 Curve Standard - 5,10, 20 and 50/xg/dL - Dilute 0.5, 1.0, 2.0 and 5.0 mL of the
Working Standard to 100 mL with the Matrix Modifier.
6. SAMPLE PREP
6.1 Using a micromedic pumping system, draw up 100 uL of well mixed blood, and
dispense along with 500 fxL of Matrix Modifier into a sample cup.
6.2 Draw up 100 /xL of Matrix Modifier and dispense along with 500 ixL of Matrix
Modifier into the same sample cup from 6.1.
6.3 Controls - Follow Steps 6.1 - 6.2 using a known blood control.
6.4 Calibration Curve - Using a micromedic pumping system, draw up 100 /j,L of
1 a <5 /tg/dL blood sample, and dispense along with 500 /xL of Matrix Modifier
into a sample cup.
6.5 Draw up 100 ptL of the lowest Curve Standard and dispense along with 500 jxL
of Matrix Modifier into the same sample cup from 6.4. Continue steps 6.4 -6.5
with remaining standards. Run the Matrix Modifier at the beginning of the run
to ensure it is lead free.
7. INSTRUMENT PARAMETERS
HITACHI SIMULTANEOUS MULTIELEMENT ATOMIC ABSORPTION
SPECTROPHOTOMETER
Element: Pb
Sample: Blood
Analyst: Lee
-------
Date: 89.10.26
A/S PROGRAM
Signal Mode:
BKG Corrected
Measurement Mode:
Absorbance
Sample Blank:
No
STD Replicates:
1
Sample Replicates:
2
Statistics:
Mean, SD, RSD
Auto Sampling:
Yes
Sample Volume:
20 /xL
Dilution:
Off (Sample)
Cone. Times:
1
Modif. Add.:
No
Stop Position:
16
Reslope -Standard:
No
-Interval:
1
Result on Record:
Yes (Cone. +Abs)
Chart Speed:
1
GROUP 1
Calculation Pb:
Peak Height
Slicing Height
(Peak width only) Pb:
10%
Carrier Gas Int.:
Yes
Opt. Temp. Contr.:
On
GkOUP 2
Calculation
Slicing Height
(Peak width only)
Carrier Gas Int.:
Yes
Opt. Temp. Contr.:
On
STANDARD SAMPLE
SI S2 S3 S4 S5 S6
INSTRUMENTAL CONDITION
Unit: ppb
Time Constant: 0.2 sec
Cuvette: Pyro
-------
Carrier Gas:
Interrupted Gas:
200 Ml/min
30 Ml/min
TEMPERATURE PROGRAM
Stage
No. Temperature (C)
Start End
Time (Sec)
Dry
Ash
Ash
Atom
Clean
1
2
3
4
5
50 120
120 720
720 720
2400 2400
3000 3000
60 Group l:Pb
30
30
5
5
W.L.
Pb 283.3 nm
Lamp Dimension
7.5 mA Linear
8. QUALITY CONTROL
8.1 A calibration curve composed of a minimum of reagent blank and three
standards is prepared.
8.2 If a large number of samples are to be run, run a set of standards at the
beginning and the end of the run. Average standards.
8.3 Run at least one known control, QC material, NBS or Quebec, etc. with every
set of standards and every 10 samples.
8.4 At least one duplicate sample should be run every 10 samples.
8.5 At least one spiked sample should be run every 10 samples.
9. REFERENCE RANGES
BEI Lead in Blood - 50 /ig/100 mL
10. NOTES
10.1 All plasticware used in this procedure must be soaked in 5% HN03 and
rinsed several times with DIW.
10.2 A well seasoned pyro-coated cuvette seems to work better for this method.
It reduces the carbon buildup and minimizes the splattering of the sample.
-------
REFERENCE
Miller DT, Paschal DC, Gunter EW, Stroud PE, and D'Angelo J: Analyst 1982, 112,
1701.
-------
ACCURACY AND PRECISION OF LABORATORIES
OC Sample
TV
"Low Bench"
1.6
"High Bench"
45.6
"Low Blind"
4.3
"High Blind"
10.3
Boston
ESA
0.2 (0.5)
47.1 (1.6)
4.0 (0.6)
10.6 (0.7)
-------
LEAD FREE KIDS STUDY
ERYTHROCYTE PROTOPORPHYRIN PROCEDURE
(Hematofluorometer)
A. Principle
Erythrocyte Protoporphyrin (EP) is beneficial in identifying a biological response to
lead (Pb) absorption (causing increased blood lead levels), as well as iron-deficiency
anemia. A simple dedicated portable fluorometer, which is called a
hematofluorometer, measures the zinc protoporphyrin level by using front surface
illumination of a drop of blood on a glass microscope cover slip.
The instrument measures the ratio between zinc protoporphyrin (ZnP) and hemoglobin
(oxyhemoglobin) in blood. Consequently, the raw data from the instrument is
presented in ZnP/Hgb. The instrument has been calibrated to take the raw reading
and convert it electronically to units or equivalent up EPP/lOOmL of whole blood at
a fixed hematocrit (_42_) to conform to the CDC definition of the Modified Piomelli
Technique (see manufacturer's manual for more detail).
B. Apparatus
1. ZnP Manual 4000 Hematofluorometer.
2. 25 - /xL micropipet, Eppendorf.
3. Oxygenation apparatus.
4. Glass cover slips.
5. Wood applicator sticks.
C. Basic Operation
1. Introduction
a. Gloves, safety glasses, and lab coats are to be worn in the blood lead
laboratory at all times.
b. Power to the unit is on at all times.
c. Handle the calibration slides by the edge only.
2. Calibration Procedure
a. Checker slides and internal calibration
-------
1. Pull the slide assembly forward to the stop position. Place a low-level
(green) calibration slide into the assembly. Push the slide assembly until
it engages the first detent position. The "wait" indicator light will
immediately light for a period of approximately one second and then go
out. The "advance slide" indicator light will light. Push the slide assembly
into the second position and wait until the "advance slide" indicator light
again lights. Now push the slide assembly into the third position; take the
digital reading when the "read" light illuminates.
2. Record three digital readings for the green checker slide in the
Hematofluorometer Daily QC notebook. The average of these three
readings should be within the range given by the manufacturer. If the
average value is out of range, refer to Section C.2.c. of this procedure to
calibrate the instrument.
3. Record the values for the medium level (yellow) and high level (red)
checker slides following the above procedure.
4. Record the values for internal calibration: with no slide in the assembly,
advance to the detent position. The digital display is the value for the
internal low calibration. Advance to the second position and record the
internal high calibration value. Advance to the third position and record
the dark current. If the internal calibration values differ greatly from the
previously recorded values and the checker slides are out of range, refer
to Section C.2.b. of this procedure.
b. Cleaning the internal slides.
Periodically it may be necessary to remove dust that has accumulated on the
internal slides. If you feel this process must be completed, notify the
laboratory supervisor at once. If she/he sees fit, PROCEED WITH GREAT
CARE. The internal slides are extremely fragile. They cannot be repaired or
replaced so don't break them!
Remove the black cover on the right side panel of the hematofluorometer.
Loosen the small screw within and pull out the slide assembly. You will notice
that the top side of the assembly is very dusty. That is not our concern. On
the underside, you will see portion of the internal slides that is exposed to the
radiation. If there is dust on this portion, remove it in the following manner.
Moisten a cotton swab with methanol and VERY CAREFULLY attempt to
remove the dust particles. Allow the slides to dry completely.
Replace the slide assembly and tighten the screw to the desired tension.
With no slide in the assembly, advance to the first, second, and third positions
-------
and make note of the digital readings at each position. If the problem seems
to have been rectified, analyze the checker slides and continue. Record in the
QC notebook that this section was taken.
If the problem persists, notify the lab supervisor and if necessary, notify the
ESA, Inc., Service Department.
c. Recalibration
If the. internal calibration values are acceptable but one or more of the
checker slides is out of range, complete the following. Record in the QC
notebook that recalibration was done.
Zero Offset Calibration
If the low level check slide is out of range, place the slide in the assembly and
advance to the third position. Depress the "Push to Cal" button on the front
panel and hold while adjusting the "Zero Control" located on the rear panel
until the desired reading is obtained. Check the low value calibration two or
three times.
Slope Correction (High Value Calibration)
Place the high level (red) calibration slide into the assembly. Advance the
slide to its third position. If calibration is necessary, depress the "Push to Cal"
button and hold while adjusting the "Cal" control on the front panel.
Recheck the zero and high calibration points. Repeat calibration procedure
if necessary. When the green and red slides are within range, the yellow slide
must be within range since the calibration curve is linear. If it is not, notify
the lab supervisor.
3. Sampling Procedure
a. Rock venous samples on an aliquot mixer for one hour to insure they are
completely mixed and have reached room temperature.
NOTE: Protoporphyrin is light sensitive. Protect samples from excessive light
exposure.
b. Turn on oxygen by turning the small black knob on the tank gauge. Turn until
the water tap bubbles gently.
c. Place 25 /*L of whole blood on a 25 X 25 mm coverslip. Spread blood to
completely cover a 3/8 inch diameter spot on the center of the slide using the
pipet tip or a wood applicator stick.
-------
NOTE: Use a new pipet tip for each specimen.
d. Ensure that no clots, air bubbles, or other debris are located on the cover slip.
e. Place coverslip in oxygenation chamber for at least one minute.
f. Place the coverslip on the slide assembly and advance to the third position.
The EP value expressed in /xg EP/lOOmL will appear on the digital readout.
Continue to take readings until the value has stabilized. (The blood may
continue to oxygenate causing the value to drop. The value will stabilize when
the blood is completely oxygenated.)
g. Record the last three readings taken on the green worksheet.
The digital readout represents EP or FEP. To convert EP to ZPP, multiply
the EP value by 1.1.
h. NOTE:
1. Lysed blood will give questionable data.
2. Elevated bilirubin concentration will give increased EP.
3. Previously frozen samples will give questionable data due to lysed red
blood cells. The same may happen with blood over two weeks old.
Quality Control
a. Analyze a high, medium, and low control daily and record the values in the
Hematofluorometer Daily QC notebook.
b. These controls are taken from client samples, selected by the lab supervisor.
c. If the QC values differ by more than 15 ju.g/100mL from the previous value,
see the lab supervisor.
Maintenance
The hematofluorometer must be calibrated by the manufacturer (ESA, Inc.) once
a year. At that time, values will be assigned to the checker slides.
-------
Updated 2/90
LEAD FREE KIDS STUDY
PROTOCOL FOR HAND LEAD DETERMINATIONS (HAND WIPES)
Testing of hand lead will be conducted each time a blood sample is taken for lead
analysis. There wili be a total of three hand lead determinations: the first baseline test will
be done before any abatement activities occur, and the second and third follow-up tests will
occur 4-6 months and 9-12 after abatement activities are completed.
Since studies indicate that hand dust reaches equilibrium within two hours after
washing, case managers will make every effort to conduct the hand lead testing more than
two hours after the last hand washing reported by the parent or guardian.
Case managers will wear disposable gloves when obtaining a hand wipe. Lead in dust
on children's hands will be sampled by wiping each hand of the child with three separate
commercial wet-wipes. The Walgreen's brand wet-wipes will be used for the LFK study.
All surfaces of the hands, front and back up to the wrists will be wiped thoroughly with each
of the three wet-wipes. All six wet wipes will be placed inside the container provided by the
analytic laboratory. The container will be labeled with the child's name and LFK number.
Each case manager team will also prepare hand wipe blanks at regular intervals
during the sampling period (i.e. every tenth child). The hand wipe blank will be prepared
by removing six wipes from the wet-wipe container and handling them in such a manner as
to simulate wiping the child's hands. These wipes will be placed into a container labeled
"BLANK", dated and submitted to the laboratory along with the regular samples. Blind
external quality control samples prepared by the EPA with dummy (seemingly correct)
identifiers will also be submitted to the lab.
Chain of custody forms will be initiated when hand wipe samples are taken. All
samples will be transported to Dennison Laboratories of Woburn, Massachusetts by the
Lead Free Kids Study driver.
The six wipes will be composited for chemical analysis. The method of extraction of
the lead from the wet wipes is currently being determined. The total quantity of lead found
will be reported as ug/pair of hands.
-------
LABORATORY ANALYSIS OF HANDWIPES
Report of Analysis
Method
1 M Nitric Acid Extraction
1. Place each sample in a labeled, acid-washed 800 ml beaker.
2. To each sample, add 100 ml of 1 M nitric acid prepared with deionized water.
3. Swirl each sample for 10 seconds.
4. Cover each sample with a vvatchglass and allow it to extract at room temperature for
2 hours.
5. Decant the acid solution from the handwipes into a labeled, acid-washed 250 ml
beaker.
6. Add 50 ml of 1 M nitric acid to the handwipes in the 800 ml beaker.
7. Swirl the sample for 10 seconds.
8. Decant the acid solution into the same 250 ml beaker to composite the acid rinse.
9. Repeat steps 6, 7, and 8 a second time for a total acid solution of about 200 ml.
10. Cover the samples with a watchglass which is elevated above the beaker rim with glass
hooks. (The watchglass must be elevated to prevent "bumping" of the sample during
evaporation).
11. Place the samples on the hotplate at about 250ฐC.
12. Evaporate the samples to dryness.
13. Add about 3-5 ml of 1 M nitric acid to each sample, rinsing the watchglass and the
sides of the beaker.
14. Heat the samples gently on a hotplate at 120-150ฐC to redissolve lead.
15. a. Filter the samples using Whatman, rinsing the beaker/filter paper/funnel with IM
HN03.
b. Evaporate to about 5 raL on a hotplate.
16. Transfer to an acid-washed 10 mL graduated volumetric flask, rinsing and diluting with
IM HN03.
17. Shake sample well and transfer to borosilicate test tube and cover.
18. Measure lead concentrations using a Varian 1475 Atomic Absorption Spectrometer.
Report results in total /xg/sample.
-------
LEAD FREE KIDS STUDY
PROTOCOL FOR MEASURING THE LFK CHILD'S HEIGHT
The LFK child should be in his/her stocking feet when you measure his/her height.
First, find an open wall in a room that has no carpeting or thin pile carpeting (the kitchen
may be best). Place the yardstick up against the wall with the centimeter side facing you.
Make sure that the bottom of the yardstick is resting squarely on the floor and that numbers
on the yardstick are increasing as you look up from the floor. Have the child stand straight
(no slouching!) up against the yardstick and measure his/her height to the nearest one-half
centimeter (i.e. 40.0, 40.5, 41.0, 41.5). This means that you should round up to the nearest
one-half centimeter if the height is in between half-centimeters. Thus, 40.25 should be
rounded to 40.5 and 40.75 should be rounded to 41.0. The height measurement should be
taken right at the top of his/her head (big hairdos should not be counted!!). A small ruler
held across the top of the child's head may help you read the correct number on the
yardstick. If the child is taller than the yardstick, mark his/her height on the wall with a light
pencil and measure the distance from the floor to the mark with the yardstick. Record the
height IN CENTIMETERS on the form provided.
-------
LEAD FREE KIDS STUDY
PROTOCOL FOR MEASURING LFK CHILD'S WEIGHT
USING THE SECA INTEGRA SCALE
SETTING UP THE SCALE
The scale is already assembled for use with the digital indicator head fitted at the back of the platform
the connecting cable stored in the compartment underneath the head.
INSTALLING TI1E BATTERY
The seca integra operates with a standard 9-V alkaline battery. Remove the digital display head from the
bracket or base in order to open the battery compartment underneath. Connect the battery terminals,
then insert the battery and close the cover. Replace the head on the bracket or base.
HOW TO WEIGHT CORRECTLY
Selected lbs. or kg measurement using the switch on the underside of the digital display.
Switch on the scale by pressing the ON button.
Weight yourself when display switches to 0.0
Your weight is indicated after a short time (approximately 4 seconds).
The scale switches off automatically after 30 seconds.
How to use the seca integra when you wish to weight a small child for instance who cannot stand alofl6
on the scale:
First, weigh yourself as described above.
Remain standing on the scale and press the ON button once again. 000.0 lb (or tArE kg) appears
the digital display.
Wait until the display switches to 0.0 and then take the child in your arms.
9 After a few seconds, the child's weight appears in the display.
StOP lbs (or SUP kg) signals that the scale has been overloaded.
WHAT TO DO IF....
...No weight display appears under load?
Remove load from scale
...appears?
Press the ON button
...Err appears?
Remove load from scale, press ON and wait for 0.0
...bAt appears?
Change battery.
-------
LEAD FREE KIDS STUDY
SUPERFUND SOIL LEAD ABATEMENT DEMONSTRATION PROJECT
PROTOCOL FOR SOIL SAMPLING AND ANALYSIS
1. SOIL SAMPLING
1.1. SITE DESCRIPTION
1.1.1. General Site Description
For each location, a detailed drawing should be made that shows the boundary
of the lot, the position of the main building and any other buildings such as
storage sheds or garages, the position of the sidewalks, driveways, and other
paved areas, the position of the play areas if obvious, and the position of the
areas with exposed soil (grassy or bare) (See attachment A). Show down
spouts and general drainage patterns. Identify each soil subarea by letter or
number. If a large soil area needs to be divided into smaller patches for
sampling convenience, show how this division was made.
In addition to the diagram, briefly describe the location, including the
following information:
Type of building construction
Condition of main building
Condition of lot (debris, standing water, vegetation cover)
Nature of adjacent property
Presence and type of fence
Animals on property
Apparent use of yard (toys, sandbox, children present)
Underground utilities
1.1.2. Subarea Description
For each soil subarea identified on the general diagram, draw a full page
diagram showing the approximate dimensions and position relative to the
building foundation (see Attachment A). Indicate vegetation and bare soil
areas, as well as obvious traffic patterns. Identify the category of landuse, such
as roadside, property boundary, adjacent to foundation, play area. Select an
appropriate sampling scheme and mark the sample locations on the diagram.
1.1.3. Sampling Schemes. The-sample scheme selected for each subarea must
adequately characterize the potential exposure of children to lead in the dust
-------
from this soil. It must identify the areas of high lead concentrations, and the
general distribution pattern of lead concentrations at the soil surface. For
abatement puiposes, the depth to which lead has penetrated the soil profile
must be determined. Consequently, selected the most appropriate sampling
scheme is the ciitical element in the site description. Several options are
offered for the best judgement of the investigator.
Line Source Pattern. This pattern can be used whenever the source of the lead
is thought to be linear, such as along a building foundation, a fencerow, a
street, or beside a garage. Draw a line parallel to the source, such as the
foundation of the main building, approximately 0.5 meters (20 inches) from the
foundation. Repeat at the property boundary if the subplot is more than three
meters wide (10 ft), and add a third parallel line between the first two if the
subarea exceeds five meters (16 ft) in width. Divide each line into segments
that do not exceed 7 meters (20ft) in length. Take one composite of 5-10 cores
along each line segment. A subarea, for example, that is at the side of the
main building and measures 12 X 7 meters would have three lines of two
segments each. The lines would be parallel and approximately three meters
apart. They would be 12 meters long and consist of two 6 meter segments
each, making a total of six samples, each being a composite of at least five
cores divided into a top 2 cm sample and a bottom 2 cm sample.
Targeted Pattern. This method is intended to be used in conjunction with the
line source or grid patterns as a means of sampling obvious areas that would
be missed by the regular patterns. In using the targeted pattern, the
investigator should select those locations within the subarea that are likely to
reflect potential exposure to lead in soil dust. These may be play areas, paths,
drainage collection areas, or areas that are likely to contribute dust to other
surfaces that children use. Determine the number of samples to be taken by
identifying distinctive landuse characteristics (path, swingset, sandbox), and
take a composite of 5-10 cores for each sample.
Small Area Pattern. When the subarea is less than two meters in each
dimension, or when the accessible area of a larger plot is less than four square
meters, a single composited sample may be taken if it appears that such a
sample would adequately represent the subarea.
Grid Pattern. Establish a rectangular grid of intersecting lines 2-10 meters
apart, and sample each rectangular area. For larger areas, randomly select the
rectangles to be sampled. In each rectangular area, mark three lines parallel
to the longest axis, and composite 5-10 cores along each line. Since the
rectangle should not exceed four meters, there is no need to divide the line into
segments. Therefore, each rectangle should have six samples of 5-10
composites each. Use this pattern when the subarea is generally uniform and
there is no reason to suspect large variations in lead concentrations.
-------
When the sample sites have been located on the subarea diagram and the
sample collection is ready to proceed, locate each sample with a flag and
visually confirm an even and representative distr ibution of sample locations.
SAMPLE COLLECTION
The flags or other markers represent the center of the sample location for the targeted and
small area patterns. For the line source and grid patterns, the flags indicate the sampling
lines. Take at least five but not more than ten cores randomly selected from within the
sampling area of the targeted and small area sampling patterns. For the line source
sampling pattern, select a random location on each line and take subsamples within a 2'
by 2' square area. Take these subsamples from the four corners and the middle of the
square with the middle point being on the line. When the line exceeds 7 meters and is
broken into segments, take a composited sample in the above manner on each segment.
The cores make a composite identified as a single sample. A sample record sheet is used
to record information about the composite.
The corer should be clean and free of lead contamination. Vegetation and debris can be
removed at the point of insertion, but do not remove any soil or decayed litter. The corer
should be driven into the ground to a depth of at least 10 cm, 15 cm if possible. If the
10 cm depth cannot reached, the corer should be extracted and cleaned, and another
attempt made nearby. If the second attempt does not permit a 10 cm core, the sample
should be taken as deep as possible, and the maximum depth of penetration noted on the
sample record sheet. Every effort should be made to take all cores of a composited
sample at the same depth.
The cores of each plot should be examined for debris, artifacts, and any other evidence
of recent soil disturbance. These should be noted on the subarea description sheet, as
should a brief description of the soil color and soil type.
For each sample location, the top 2 cm segment of each of the cores are composited into
one sample, and the bottom 2 cm segment combined into a second. For the surface
segment, debris and leafy vegetation should not be included with the sample. However,
no soil or decomposed litter should be removed, as this is the most critical part of the soil
sample and is likely to be the highest in lead concentration.
The soil core segments should be composited in sealable polyethylene containers suitable
for prevention of contamination and loss of the sample. The sample identification number
should be placed on the container and the sample record sheet. After each sample
composite, the corer should be cleaned by reinsertion in the next sampling area. Store
the composited soil sample at ambient temperature until returned to the lab.
A field blank should be taken for each sample crew day. This is normally done by taking
a sample container with clean quartz sand into the field, opening it to expose the
container for a period of time representing normal sample procedures, then returning the
container to the lab in the same manner as other soil samples. The puipose of the field
-------
blank is to detect accidental or incidental contamination during the sampling process.
1.3 SAMPLING HANDLING AND STORAGE
The sample containers should be sealed to prevent loss or contamination of the sample.
Shipping containers should also be airtight. Storage should be in a cool, dry location.
1.4 RECORD-KEEPING AND SAMPLE CUSTODY
Soil sample records for each location consist of a location diagram and description, a plot
diagram for each distinct soil plot, and sample record sheet for each sample in a plot.
The sample record sheets should also contain space for chain-of-custody documentation
(See Attachment B).
Samples should be sequentially numbered within each subarea. Each location diagram,
subarea description, and sample record sheet should bear all sample numbers and the
signature of the person responsible for verifying the quality of the infomiation collected.
This signature certifies that there has been no misuse of the sample protocol, no mistake
in recording the information, and that the information is sufficient to clearly identify these
samples for comparison with other types of samples taken at the same location, such as
street dust, house dust, house paint, blood, and hand dust. These documents also establish
the chain of custody required for the Quality Assurance Plan.
When the sample is delivered to the laboratory, custody is relinquished by the field
investigator and received by the lab supervisor by signatures on the sample record form.
2. SAMPLE ANALYSIS
2.1 METHOD OF ANALYSIS
Three methods of analysis have been considered. They are Atomic Absoiption
Spectroscopy (AAS), Inductively Coupled Plasma Emission Spectroscopy (ICP), and X-
Ray Fluorescence (XRF). The XRF method is the approved method for routine analyses,
whereas the AAS method should be used for standardization.
2.1.1. Sample Definition. The representative urban soil sample is defined as the soil
from 0-2 cm depth that passes a 250 pm stainless steel sieve. This fraction is
comprised of small particles, and the concentration of lead believed to be
closely related to that of particles on the hands of children. The fraction is
also homogenous enough to allow reliable analysis by X-Ray fluorescence.
2.1.2 Sample Preparation. Sample preparation requires that the sample be air dried
and separated by particle size before being digested by wet chemistry. Drying
is done at room temperature overnight, or until the sample can be easily
-------
disaggregated by hand or with a rolling pin. The full sample should be
brought to complete disaggregation by passing through a 2 mm sieve, using the
fingers or a stainless steel tool to crush the larger soil particles. Material larger
than 2 mm should be discarded. Soil should not be milled to a fine powder
with a mortar and pestle or any other grinding device.
The fraction that passes the 2 mm sieve is now called the total soil fraction.
A portion of this sample is retained for possible reference analysis, but the
larger fraction is passed through a #60 mesh sieve (250 pm), giving a fine soil
fraction identified as the "Urban Soil Sample." The portion that does not pass
the #60 mesh sieve should be discarded, as only the total soil fraction (<2 mm)
and the fine soil fraction will be analyzed.
About 5-10% of the retained total soil samples should be analyzed. An aliquot
is ground so that it all passes a #60 mesh (250 pm) sieve, mixed well and
analyzed. Grinding is necessary to provide low/appropriate variance in XRF
analysis.
During the processing of the sample, it should be remembered that small soil
particles may individually be as high as 50,000 pg Pb/g, and paint fragments
as high as 300,000 pg/g. Care should be taken to clean equipment between
samples. The sieves may be cleaned by tapping on a hard surface to remove
residual particles, or any other dry method. Wet washing is not recommended
as this will interfere with the size calibration.
Care should also be taken to thoroughly homogenize the separated sample
before removing the aliquot for analysis. Shaking will cause separation.
Tumbling or stirring is recommended.
2.1.3. Atomic Absorption Spectroscopy (To be used for primary standards)
2.1.3.1. Wet Digestion. The extraction procedure used for solubilizing soil
lead is critical to the interpretation of the results of the Superfund
Soil Lead Abatement Demonstration Projects. Even in the absence
of analytical errors, the data may not represent the same lead
concentrations from sample to sample unless the correct extraction
procedure is used. The method selected here does not represent the
total extraction of lead, but the breakdown of the organic material
and the leaching of lead from the inorganic soil fraction. The
methods measure total non-matrix soil lead, because no other
extractable fraction has been experimentally shown to measure
bioavailable, or non-HF extractable, soil lead. Hot HN03 has been
repeatedly shown to extract total non-matrix soil lead, or at least
>95% of soil lead, compared to a total soil dissolution method
(HF). The 1.0 .N HN03 cold shake method has been shown to
extract as much lead as the hot HN03 extract, except for unpolluted
-------
soils where a higher fraction of the total soil lead is within the
matrix of soil particles.
The sample should be oven dried at 105ฐC for 24 hours or until a
constant weight is achieved. The aliquot should be placed in a 150
ml beaker and covered with a watch glass. Class A borosilicate
glassware and stainless steel tools should be used throughout the
sample processing. Low density conventional polyethylene
containers may be used to store the solution prior to analysis.
An aliquot of 1 g soil is normally considered representative of the
whole sample if the soil is well mixed. Prior to removing the
aliquot, the sample should be stirred with a spatula or rod.
Shaking the container can cause the sample to separate by particle
size.
2.1.3.1.1 Hot HNQ, Extraction. Add 50 ml 7N HNO cover and digest
gently at 95ฐC for 2 hours, stirring occasionally. If excessive
foaming occurs, remove from the heat periodically until foaming
subsides. Maintain at least 25 ml in the beaker by adding 7N
HNOj as necessary.
Cool and dilute with 10 ml IN HN03. Filter through Whatman
No. 42 filter paper into a volumetric flask. Rinse filter and
labware with IN HN03, and dilute to volume.
2.1.3.1.2. Cold HNCh Extraction. Weight the 1 g aliquot into a 4 oz.
urinalysis cup. Add 50 mL 1.0 N HN03 to each cup. Screw the
lid on tightly and place on a reciprocal shaker. Adjust the speed
of the shaker to maintain a suspension of the soil particles. Shake
for one hour, then filter through a Whatman 111-V filter. Rinse
with 1.0 N HN03. Dilute to standard volume.
2.1.3.2. Analysis. Analysis by flame AAS should be at 283.3 nm, with
background correction. Working standards should be prepared
fresh daily, in the range of 2-50 pg/g, in a 1.0 N HNO, matrix.
2.1.4. XRF Analysis. Approximately 2 g of loose soil sample are poured into sample
cups (Somar Labs, Inc., Cat No. 340), fitted with windows of 1/4 mil thick X-
ray polypropylene film (Chemplex Industries, Inc., Cat No. 425). The sample
cup should be at least half full. The sample cup is sealed with a sheet of
microporous film (Spex Industries, Inc., Cat No. 352A) held in place by the
snap-on sample cup cap. The exact weight of the sample is not important, but
should be in the range of 2-6 g.
The instrument configuration for the Kevex Delta Analyst Energy Dispersive
-------
X-ray Spectrometer is:
1. Kevex Analyst 770 Excitation/Detection Subsystem:
a. X-ray tube: Kevex high output rhodium anode
b. Power supply: Kevex 60 kV, 3.3 mA,
c. Detector/cryostat: Kevex Quantum - UTW lithium, drifted silicon. 165
eV FWHM resolution at 5.9 KeV.
2. Kevex Delta Analyzer:
a. Computer mainframe: Digital Equipment Corporation, PDP 11/73
b. Computer software: Kevex XRF Toolbox II, Version 4.14
c. Disk drives: Iomega Bernoullik box, dual drives, 10 MB
d. Pulse processor: Kevex 4460
e. Energy to digital converter: Kevex 5230
3. Operating Conditions:
a. Excitation mode: Mo secondary target with 4 mil thick Mo filter.
b. Excitation conditions: 30 kV, 1.60 mA
c. Acquisition time: 300 livetime seconds
d. Shaping time constant: 7.5 microseconds
e. Sample chamber atmosphere: air
f. Detector collimator: Ta
4. Analytical Conditions:
a. Escape peaks, but not background be removed from all spectra.
b. The intensity ratio, defined as the integral of counts in the Pb (LA)
window divided by the integral of the counts in the Mo (KA) Comptom
scatter window, should be determined for each spectrum
c. The intensity ratios for the standards should be used to determine a
linear least squares calibration curve.
The acquisition time (3c) may be reduced at the discretion of the lab
supervisor.
2.1.5. QA/QC. By blind insertion into the sample stream (where possible), the
QA/QC officer will provide the following blanks at the indicated frequency.
At the discretion of the project director, the field team will collect one blank
per day by carrying a sample of clean quartz sand into the field in a normal
sample container. The sample container will be opened and exposed during the
collection of one sample, then closed and returned to the lab. The field blank
can be split into two aliquots. One aliquot, the field blank, can be analyzed
directly with no further treatment. The second aliquot (the sample blank) can
-------
be analyzed after it has passed through the sample stream (except seiving).
The field blank represented contamination added in the field, the sample blank
represents contamination added in the field and during storage and sample
preparation.
A project standard soil sample will be prepared and distributed at the beginning
of the study. This will be used as a lab control. For XRF analysis, there is no
need for a reagent blank.
Field blank 1/field sampling day
Sample blank 1/field sampling day
Lab control 1/20 samples
Reagent blank 3/reagent batch
Additionally, split sample (duplicate) analyses and spiked samples will be
deteimined as follows:
Split soil 1/20 samples
Spiked soil 1/20 samples
The spiked soil samples will be prepared by mixing dried and sieved soil of
known concentration with the sample. Spiked soil samples may be used at the
discretion of the project director. Additional split soil samples will be sent to
a designated QA/QC laboratory for analysis using the hot HN03 method, one
for each 40 samples.
An interlaboratory comparison, similar to the soil pilot study, will be conducted
during each six month period, with 10-20 samples from each laboratory,
including the QA/QC lab. These samples will be dried, but not sieved.
-------
LEAD FREE KIDS STUDY
Additional Soil Sampling Methods
Protocol for Preliminary Soil Sampling
The goal of the preliminary sampling is to determine whether the soil surrounding the
premise of a potential participant contains high levels of lead. For a premise to be eligible, two
or more samples must contain at least 1500 parts per million, or the mean of all the samples must
be 1500 parts per million or greater.
A total of up to five samples will be taken, which in most cases represents the four sides
of the house and a separate play area if one exists. To start, draw a rough sketch of the house
and surrounding property. Indicate areas that are paved and those with soil or grass. Label the
sides of the house F, L, R, and B, for front, left, right, and back, respectively. Right and left are
always from the perspective of standing on the sidewalk looking at the front of the house. Take
one sample from each side of the house where there is soil. If there is an area of soil that is not
directly adjacent to the house, but appears to be a potential play area, a sample should be taken
there as well. Areas of soil that are on the same side of a house but are separated by a porch or
stairs may be sampled separately, or combined as one sample.
Sampling Instructions:
Materials needed:
Trowel
paper towels
plastic bucket
zip-lock sandwich bags
marker
labels for bag
Chain-of-custody forms
To get a representative sample, you will use a technique called "composite sampling".
This involves taking several sub-samples in an area and then mixing them together to make one
composite sample.
For areas adjacent to the house, take five sub-samples along a line parallel to the house,
at a distance of one meter from the foundation. The subsamples should be fairly evenly spaced
along the foundation for the length of that side, or as much of that side as is not paved. Each
subsample should consist of scoop of soil 5 cm in diameter and 2 cm deep. Mix the five
sub-samples together in the bucket to make the composite sample from that side of the house.
Put the composite sample in a zip-lock bag and place the identifying label on the bag. Fill out
the label, giving premise address, premise ID, and sample letter. The sample identification
number will consist of the premise number, followed by the letter corresponding to the side of
the house. If more than one sample is taken on a side, then follow the letter with a number, for
-------
example, Fl, F2, etc. Indicate all sample locations on sketch.
For play areas that are not adjacent to the house, follow the composite sampling
guidelines, treating the area as a rectangle or a square. Take sub-samples from the four corners
and the middle of this area. Label samples from such play areas "P".
For each sample, initiate a chain-of-custody form. Between samples, wipe the trowel and
bucket with paper towels to remove any residual soil.
-------
LEAD FREE KIDS STUDY
Additional Sampling Methods
Protocol for Soil Recontamination Sampling
For each property that was in the project after the baseline blood sample, soil samples
need to be taken to check for changes in soil lead. Some of the properties have been abated,
and others have not. You will get a sampling pattern for each property. Recontamination
samples will be taken at every other location where a detailed sample was taken before. In
other words, you will take half the number of samples. The locations which need to be
sampled will be highlighted on the map. Locations which are not highlighted can be ignored.
At each location on the map, there will be a number and a little box like this ~
Sometimes the box will be on a line, like this . The box is where the sample should be
taken. It represents an area of about two square feet. At each location, get as close as you can
to where the box appears on the map, and take five surface scoops of soil in an area of about
two square feet. Mix these samples in the plastic container and put about a 1/2 cup of the
mixed soil in the sample bag.
The sample bags can be written on using an indelible marker. Put the address, premid,
and sample number on each bag. You can do these in advance to save time in the field.
Please use a separate paper bag and chain-of-custody form for each property. If you are
unable to take a sample or if there is some other problem with a property, please write a note
separate from the chain-of-custody form and put it in the paper bag with the samples.
Samples should be numbered using the number on the sample plan, with the addition
of the letters "RE". For example, RE2, RE4, RE6, etc.
-------
Revised 2/90
LEAD FREE KIDS STUDY
HOUSEHOLD DUST SAMPLING PROTOCOL*
For this study, the household dust samples are defined as the samples that are most
likely to come into contact with a child's hands during indoor activity. This would include
dust on upfacing surfaces accessible to the child such as bare floors, carpets, window sills and
wells, furniture, as well as dust on toys and other objects likely to be handled by children.
Dust sampling has two components that are important to interpreting lead exposure:
the concentration of lead in the dust and the amount of dust or loading on the surface. The
concentration of lead in dust appears to be closely related to the amount of lead on
children's hands whereas the amount of dust on surfaces is an indicator of the importance
of this route of human exposure.
Dust Collection and Sample Handling
There is no standard procedure for collecting dust samples. The following protocol
was decided upon after reviewing other available methods (such as the personal air pump)
and finding them inadequate. The dust sampling method chosen was the Sirchee-Spittler
modified dust buster. We believe that it is the best method for collecting numerous
household dust samples within a reasonable amount of sampling time. Other necessary
equipment to conduct the sampling are a ruler to measure the sampling area, a 25" by 25"
template for designating the floor sampling area, paper envelopes to which the dust samples
will be transferred, tape to seal the envelopes, and a cylinder of compressed air for cleaning
the sample collection screen.
Before collection, make certain that the Sirchee-Spittler modified dust buster is fully
charged. You can tell this by running the dust buster for a few seconds and listening for a
high pitched sound from the motor. Another way to monitor the charge in the dust buster
is to keep track of the number of samples taken on a change. A maximum of 18 samples
(roughly three households) should be taken on one charge. Also, when starting a sampling
round in a household make sure that the sample collection screen is clean. Use the
compressed air cylinder to blow the screen clean.
Seven dust samples should be taken in each LFK household from each of the following
locations: entry floor (i.e. right inside the front door of the house or apartment), LFK child's
bedroom window well and floor, kitchen window well and floor, and living room window well
and floor. You may choose which window to sample in a room. The floor samples should
be taken roughly from the center of the room. Sometimes it will not be possible to get all
six samples in a household because of windows that are nailed shut, obstructed by air
conditioners, etc. In these instances, obtain as many samples as possible from the designated
locations.
-------
8/89
HOUSEHOLD DUST SAMPLING PROTOCOL
Pg.2
Once the individual sampling locations are decided upon, the size of the sampling area
must be measured. For the window wells, measure the sampling area with a ruler. For the
floors, set down the 25" x 25" template. If the floor is very clean, it may be necessary to
vacuum a surface area larger than 25" x 25". In these cases, vacuum an area whose size is
double or triple the template area. Be sure to obtain an amount of dust that is adequate
for analysis (at least 5 mg).
The sampling sequence should be as follows: Collect the bedroom, kitchen and living
room floor samples first. Then, collect the floor sample from the entry way. Finally, collect
the window well samples.
To collect a dust sample, switch on the dust buster and vacuum the designated area
with back and forth strokes about 1-2 inches in width. The vacuum is most efficient if the
head is held parallel to the ground and titled about 5 degrees in the direction of the motion.
When the surface has been vacuumed, keep your finger on the switch while raising the
vacuum to an upright position. The constant air flow will prevent loss of dust from the filter
before it is in an upright position. Switch off the power and carefully remove the vacuum
head without tilting it significantly. Reach in and remove the filter screen with a gentle
clockwise motion.
Transfer the dust sample to the paper envelope in the following way. Empty the
contents of the filter screen into the paper envelope. Tap the envelope to cause the sample
to collect in one end. Next, tap the filter ring several times into the open envelope on a
hard surface.
Tap the dust to the bottom of the envelope and then seal the envelope and fold over
1/2 inch of the top of the envelope and crease carefully. Tape the folded part of the
envelope down with at least a 10 inch long piece of Scotch tape. Each envelope should be
labelled with the following information: LFK child's name, LFK number, sample location
(i.e. bedroom window well) and size of sample area. It would be best if these envelopes and
labels were prepared beforehand. Remember to handle the dust containing envelopes
carefully; keep them upright in an envelope box. We want to avoid any loss of dust from
the envelopes.
Replace the filter screen with a counterclockwise motion, attach the vacuum head and
collect the other samples in the household using the same method. When you are finished
sampling a household, clean out the filter screen and the vacuum head with a blast of
compressed air.
* Parts of this protocol were adapted from -Dr. Tom Spittler's 12/88 protocol "Instructions
for Operation and Maintenance of Sirchee-Spittler Hand-Held Dust Vacuum Units".
-------
Updated 1/90
LEAD FREE KIDS STUDY
WATER SAMPLING PROTOCOL
We wish to obtain a tap water sample that will be predictive of the child's blood lead
level. Since a standing water sample (i.e. water that has been standing in the pipes for at
least 8 hours) is thought to be most predictive, it will be necessary for the parent or guardian
to take the water sample. The case managers should give the following instructions to this
individual:
The tap water sample should be taken from the cold water faucet of the kitchen. It
should be a first flush sample of water that has been standing in the pipes from 8 to
18 hours. We foresee two main options for the time a sample is taken: (1) it can be
taken first thing in the morning, or (2) if all of the residents of the household have
been out of the house for the entire day it can be taken at the end of the day (i.e.
dinner time).
We will provide a labelled plastic bottle for the sample. The bottle should be
completely filled with the water. The bottle contains a small amount of acid
preservative and so you should store it unopened in a safe place until you take the
sample. We will return to pick up the sample at a convenient time.
Before dropping off a water collection bottle case managers will fill out and affix the
label provided by the laboratory. The chain of custody form will be initiated when case
managers pick up the water sample. The water samples will be shipped to the Hall-Kimbrell
laboratoiy in Lawrence Kansas by U.S. Postal Service,
-------
LEAD FREE KIDS STUDY
WATER ANALYSIS PROTOCOL
A. Reference: Method 239.2 (Atomic Absorption, furnace technique) EPA - 600/4-79-
020
Optimum Concentration Range: 5-100 /xg/L
Detection Limit: 1 jtg/L
B. Application:
Tests for lead are carried out using the graphite furnace atomic absorption technique
as described herein. Samples, blanks, quality control, replicate, and spike test
solutions are prepared as described and placed in trays for automatic sampling. This
instrument setup and analysis steps are performed using the parameters defined.
C. Preparation of Standard Solution:
1. Stock lead solution: Commercially available containing 1000 mg/L (1000 ppm)
of lead.
2. Matrix modifier - ammonium monobasic phosphate + magnesium nitrate
solution: Transfer 4 grams of NH4H2P04 monobasic Ultrex reagent and 0.2
grams of Mg ($03)2, Suparapure, to a 100-mL volumetric flask and makeup to
mark with deiottized distilled water (DW) containing 0.5% (v.v) HN03.
3. Working lead solution: Dilute the stock solution to the ratios needed as
calibration standards at the time of analysis. The calibration standards and
reagent blank must be prepared with the same acid, i.e., 0.5% (v/v) HNO3. The
reagent blank used in all subsequent dilutions is prepared by diluting 5 mL conc.
HNO3 to 1 L with DW. A 1-ppm solution is prepared by dilution of the 1000-
ppm stock solution with reagent blank. This 1-ppm solution is used to obtain
calibration standards of 0, 5, 10, 25, 50, and 100 ppb lead. To obtain the
calibration standards, withdraw appropriate aliquots of the 1 ppm solution and
dilute to 100 mL with reagent blank.
D. Sample Preparation
All samples solutions for analysis are acidified in the field and contain 0.5% (v/v)
conc. HNOv
-------
E. Instrument Parameters for Lead Analysis
1.
Drying Time and Temp:
40 sec. - 120 degrees C
2.
Charring Time and Temp:
40 sec. - 1000 degrees C
3.
Atomizing Time and Temp:
5 sec. - 1800 degrees C
4.
Cleaning Time and Temp:
5 sec. - 2600 degrees C
5.
Cooling Time and Temp:
20 sec. - 25 degrees C
6.
Purge Gas Atmosphere:
Argon
7.
Wavelength:
283.3 nm
8.
Slit:
0.7 nm
9.
Tub/site:
Pyro coated tube with L'vov platform
10.
Matrix Modifier Setting:
5 fxL
11.
Sample and Standard
Quality Setting:
20 fiL
12.
Max power:
30
13.
Background correction mode:
On
14.
Lamp:
Electrodeless discharge lamp (EDL)
Note: Parameters 1, 2, 4, and 5 use 1 second ramp time. Parameter 3 uses 0 second
ramp time and gas stop flow.
F. INSTRUMENT USED
Perkin-Elmer Zeeman model 5100 atomic absorption spectrophotometer equipped
with a model AS-60 autosampler and an HGA model 600 graphite analyzer.
G. LABORATORY USED
Hall-Kimbrell Laboratory, Kansas City, Kansas
-------
Drafted June 1990
Revised August 1990
LEAD FREE KIDS STUDY
LEAD PAINT AND SITE INSPECTION PROTOCOL
LFK participants' homes will be inspected to provide information on the extent of
leaded paint to deleading contractors and the project epidemiologist. The contractors will
be given this information so that they can make informed estimates on the cost of interior
and exterior deleading. The project epidemiologist will use the measurements for scientific
purposes to estimate the contribution of leaded paint to participant children's blood and
hand lead levels.
The first part of this document describes how lead paint inspections will be conducted
to gather information for the deleading contractors. The second part describes how this and
additional information will be used for scientific purposes.
Lead Paint Inspection
Lead paint inspections will be performed according to current Massachusetts
Department of Public Health requirements by registered inspectors. The following forms
will be used to record the needed information on all properties:
1. Adapted Massachusetts lead paint inspection forms
2. LFK interior deleading information form
3. LFK exterior deleading information form
Instructions for filling out these forms are as follows:
Make sure the address of each property is recorded on each page, of each form and
that the participant child's room is designated on the appropriate form. Also record which
machine (PGT or Microlead) was used to measure the amount of leaded paint. The sides
of the house will be labelled as follows: A - front, B - left, C - rear, and D - right. Window
and doors in each room will be numbered from left to right. Window measurements should
be taken from the header to the sill and from casing to casing. A list of definitions and
abbreviations that may be used on these forms is attached.
Lead Paint Measurements
Lead in paint will be measured using x-ray fluorescence (XRF). Two different brands
of XRF machines will be used to measure lead in paint for the deleading contractors:
Princeton Gamma-Tech (PGT) XK-3 and Microlead. The two different brands will be used
because they are the only machines that are available to the study and both are needed to
conduct the inspections in a timely fashion. Only PGT XK-3 measurements will be used for
the scientific study data since the two machines are not sufficiently comparable for research
purposes.
-------
BOSTON LEAD FREE KIDS STUDY
LEAD PAINT AND SITE INSPECTION PROTOCOL
Page 2
Differences between the machines are as follows: The possible measurements on the
PGT range from 0 to 10 mg/cm2 and those on the Microlead range from 0 to approximately
50 mg/cm2. In general, the Microlead XRF reads leaded paint many more inches below the
surface than the PGT does. When we tested the comparability of the two machines, we
observed that repeated Microlead readings of the LFK conference room windowsill were 2.5,
2.2, 2.2 and 2.9 and repeated PGT readings of the same spot were 0.2, 0.7, 1.4, and 0.6.
(Note: the first two readings were taken on one day and the second two readings were taken
two days later).
XRF Machine Calibration
Both machines will be calibrated twice a day: once in the morning and again in the
early afternoon. An XRF calibration form will be filled out each time a machine is
calibrated (see attached). Calibration will involve making two sets of ten readings. The first
set of ten readings will be done using a zero standard and the other set will be done using
known lead standards of various levels (i.e. 1.45, 3.5 mg/cm2).
XRK Machine Use in the Field
XRF readings of lead paint concentrations are read directly from the digital read-out
on the machine. If the reading is 2.0 mg/cm2 or less, three readings will be taken and the
average will be recorded on the lead paint inspection form. If the inspector believes that
there is lead present on a surface despite a negative or very low XRF reading, sodium
sulfide will be used to test for leaded paint. The results of both the XRF measurement and
the sodium sulfide test will be recorded on the inspection form.
XRF measurements will be taken on painted and on (non-vinyl) wallpapered surfaces.
The determination of what constitutes an appropriate surface will be made by the inspector.
Measurements will be taken on the interior and exterior of the participant's dwelling. The
interior is defined as the apartment or living quarters of the LFK participant. The exterior
is defined as the common hallways, stairs, entrances, porches, accessible basements as well
as the exterior walls of the building. The exterior may also include any other buildings (i.e.
garages) and fences on the property. Interior measurements will be taken on walls and
woodwork including baseboards, windowsill, etc. in each room of the participant's unit.
Ceiling measurements will be taken only if the paint on the ceiling is peeling.
-------
BOSTON LEAD FREE KIDS STUDY
LEAD PAINT AND SITE INSPECTION PROTOCOL
Page 3
Additional Deleading Information and Instructions
Besides taking the lead paint readings, the inspectors will record other pertinent
information/instructions for the deleading contractors. Examples of such instructions are:
1. In general, baseboards will be made intact and capped with quarter round moldings.
When lead painted decorative moldings are present, record the width that will be
needed for replacement.
2. When porch rails or other items require replacement, specify materials and
workmanship common to the area. Also note that this will require further
negotiation with the landlord.
3. Indicate whether the door and window trim are decorative or flat. Flat boards will
be replaced with #2 pine. Decorative moldings will be dipped off-site.
4. Ceilings will be tested for lead only if they are peeling. If peeling ceilings are not
accessible, note that they should be made intact on the comment sheet.
5. Lead painted basement windows wherever possible will be covered with plexiglass.
6. Measure rails and count ballisters on exterior porches.
7. Exterior window sills and wells will be covered with aluminum and caulked.
Lead Paint Measurements for Scientific Purposes
Since the Microlead and PGT XRF machines are not sufficiently comparable, only
the PGT measurements taken by the lead paint inspectors will be used for the project's
scientific data. Thus, only about 50% of the properties initially inspected will have
measurements useful to test the study hypothesis. Once the lead paint inspectors finish
gathering all the data needed for deleading, they will return to the properties where the
Microlead was used to take the measurements and will re-take six measurements using the
PGT XK-3.
The six measurements will re-taken in each of the following rooms since it is likely
that the participant child spends most of his/her time there: the child's bedroom, the kitchen,
and the living room. One measurement will be taken on the lower part of the wall and one
on the window sill (i.e. woodwork) in each of these rooms. The calibration and
-------
BOSTON LEAD FREE KIDS STUDY
LEAD PAINT AND SITE INSPECTION PROTOCOL
Page 4
measurement procedures described previously will also be followed during this round of
measurements. Special study data collection forms will be developed for recording these
data. These same data will be abstracted from the inspection forms for the properties that
were originally tested using the PGT.
Abbreviations and Definitions for Lead Paint Inspections
n/a = not accessible
cov = covered
rep = replace
y = yes
n = no
dip = off-site removal of lead from surface by an approved method
R & R = remove and replace (unless otherwise noted, the replacement material will be #2
pine)
neg = negative
pos = positive
upper walls = walls above five feet
lower walls = walls below five feet
mit = make intact
porch = the area extending from the house, the wall the porch is attached to is the exterior
of the house.
scrape = delead on-site
interior = the apartment or living quarters of only the LFK participant, excludes common
areas within the building.
exterior = the common hallways, stairs, entrances and porches as well as the exterior walls
of the building, and all other buildings and fences located on the property.
All other abbreviations are described on the individual forms.
-------
LEAD FREE KIDS STUDY
INTERIOR DUST ABATEMENT
Description of abatement process
The purpose of the dust abatement is to significantly reduce the amount of
lead-bearing dust in the treated homes. It follows the loose paint abatement in the
Study Group and the Control Group A. The methods used are similar to those used
during the loose paint abatement. The most important distinction is that no loose
paint is removed during the dust abatement. This clean-up focuses on cleaning dust
off surfaces where it accumulates.
The two primary activities involved in this process are vacuuming with a
HEPA vacuum and wiping surfaces with either a wet cloth or an oil-treated rag(for
furniture). The surfaces treated in this manner are floors, woodwork, walls, and
furniture. For the dust abatement, the vacuuming on floors is timed. Carpets are
vacuumed for 3 minutes per sqare yard. Wood and tiled floors are vacuumed for 2
minutes per square yard and washed with a TSP solution. Area rugs are vacuumed
on each side, then rolled up so that the floor beneath can be vacuumed and washed.
Because the loose paint abatement and the dust abatement are so similar,
checklists are used in both cases to document that all necessary steps are taken. In
the dust abatement no tyvek suits and plastic dropcloths are required, but care is
taken to do the rooms in such an order that no dust is tracked from an uncleaned
area to a cleaned area.
Summary of dust abatement
1. Furniture is moved as needed to expose floor.
2. Top (horizontal) surfaces of woodwork(doorframes, windowsills, etc.) are
HEPA vacuumed.
3. Walls and other vertical surfaces are HEPA vacuumed.
4. Vertical surfaces are wiped wiped with wet cloths(TSP solution). Cloths are
used once and thrown away.
5. Horizontal surfaces are wiped with wet cloths.
6. Furniture is wiped with oil-treated cloths.
7. Area rugs are vacuumed on both sides and rolled up to expose floor.
-------
8. Floors:
Wood, tile, or linoleum floors are vacuumed for 2 minutes per square yard,
then washed with wet cloths.
Carpeted floors are vacuumed for 3 minutes per square yard.
9. Furniture is moved back to original positions. Tyvek foot coverings are used
by anyone who needs to enter a cleaned area.
-------
LEAD FREE KIDS STUDY
INTERIOR PAINT ABATEMENT
Description of abatement process
The purpose of the interior paint abatement is to safely remove any very
loose, chipping paint from the inside of the home without generating dust, or
leaving behind any small chips of paint. The techniques used for the interior paint
abatement are simple in principle. Loose paint will be vacuumed using a HEPA
vacuum. HEPA stands for "high efficiency particle accumulator". This vacuum is
equipped with a special filter which catches dust that would pass through an
ordinary vacuum. The peeling surfaces are then washed off using disposable cloths
and a solution of Trisodium Phosphate (TSP) and water. TSP is a detergent which
is good for picking up lead.
Obviously only very loose paint will come off during this process. This
abatement is not "deleading". We are only trying to remove the paint most readily
available to the children living in the house, and most likely to fall off and
contribute to lead in the house dust.
Summary of loose paint abatement process
NOTE: This work is monitored by the case managers to ensure that no steps are
skipped. All children are absent from the house during the abatement. All workers
put on tyvek suits and overshoes before entering a work area and remove the
tyvek overshoes before leaving the work area.
1. A case manager walks through premise with the contractor to identify areas
to be washed. All window wells and trim are washed even if loose paint is
not evident. In some cases, the window wells and trim are the only areas
needing abatement. In others, walls and trim also need work. In general, we
do not remove any ceiling paint, because ceiling paint rarely contains lead.
A work plan for the premise is agreed upon, including the order in which
the rooms will be worked on, what furniture will be moved, etc.. The entire
floor of the apartment is HEPA vacuumed, and all toys are put in plastic
trash bags.
2. Work proceeds room by room, starting from the far end of the unit and
working back towards the entrance. In each room, furniture is moved away
from windows, baseboards, or any other area of chipping paint. The floor
around these areas is vacuumed to pick up paint that may already have
fallen. Furniture and floors are then covered over with plastic sheets as
needed to provide a work area within which paint can be contained. The
plastic should be attached to walls or baseboards below any areas needing
abatement, and should extend out into the room from the point of
-------
attachment. In some cases the plastic may be put up under a window, run
out to the middle of the room, and up over furniture or up a wall to form a
basin.
3. Workers vacuum with a HEPA vacuum all areas of chipping paint. They do
not use the vacuum to chip or scrape paint, simply to pick up whatever
paint readily comes off. They then wash down chipping surfaces with
disposable cloths soaked with a solution of water and sodium sulfide. The
cloth may be folded over carefully for a second pass over one area, but the
last time a surface is cleaned a new cloth should be used. Cloths will be
thrown away after each use to prevent spreading dust and chips.
4. When all the surfaces to be cleaned in a room are finished, all plastic
dropcloths are wiped off. The cleaned surfaces and the plastic are then
cleaned with a HEPA vacuum. Window wells are painted with primer paint
to "Jock down" any remaining paint and dust.
5. All equipment is decontaminated by washing. The plastic is taken up. The
workers start at the edges, and carefully roll the plastic inward towards
them from all sides, until they stand in the center of a ring of rolled-up
plastic. They step off the plastic, leaving behind the tyvek overshoes. The
plastic can then be placed in a trash bag. The floor will be HEPA
vacuumed to catch any residual dust. Any furniture that the contractors had
to move will be moved back into place.
6. All rooms are treated in this manner. For Control Group B this concludes
the abatement process for the fall of 1989. The Study Group and Control
Group A have dust abatement immediately following the loose paint
abatement.
-------
LEAD FREE KIDS STUDY
Interior Paint and Dust Abatement
Site Documentation Form
Address
Premise ID
Date
Monitor(s)
Please check appropriate circle to indicate that each step has been completed.
Part A: Completed
1. Walkthrough with contractor
to assess work to be done [ ]
2. HEPA vacuum entire floor of residence [ ]
3. All toys sealed in plastic bags [ ]
Part B-
Note: In all activities of washing or wiping surfaces with cloths, the cloth should never be put
into the wash water after use. All cloths are to be disposed of in a plastic bag immediately
after use. A bag should be placed close to the work area to avoid unnecessarily tracking dust
from immediate area.
Comments and Notes on Interior Paint abatement. Please note any unusual circumstances
or conditions in this house:
On the next page, fill in one of the following room codes at the top of each column.
K = Kitchen P = Pantry LR = Living Room DR = Dining Room BR = Bedroom (BR2
= 2nd Bedroom, etc.) BT = Bathroom
O = Other (specify below)
O
02
03
-------
Check steps when completed
Room Codes
1. Shades\curtains removed
2. Furniture moved away from loose paint
paint abatement areas
3. Polyethylene in place for loose paint
abatement
4. Workers wearing Tyvek suits, respirators
and foot covering
5. Procedure observed for leaving or entering
work area
6. Initial vacuuming of areas of loose paint
7. Washing of loose paint areas with TSP
and water solution
8. Final HEPA vacuuming of loose paint areas
9. Wipedown of polyethylene sheets to collect
any paint and dust
10. Workers removed foot coverings, and head
covering, etc. when leaving work area
11. Floor coverings rolled up and removed in such
a way that no paint could fall outside plastic
12. Floor area under poly vacuumed
13. Window wells painted with primer paint
14. Return furniture and other articles to
original position
OR
15. Proceed to Dust Abatement(Part C.)
-------
Part C. Interior Dust Abatement Checklist
Note: If it is not possible to move furniture to the next room before the dust abatement, the room may
be cleaned one half at a time. The furniture can be moved to one half of the room while the other half
is cleaned. If a piece of furniture can't be moved at all, every attempt should be made to vacuum and
or wash as far under it as possible.
Check steps when completed
Room Codes
i. Move furniture as needed to clean floor
[]
G
[]
d
ii
[]
2. HEPA vacuum top surface of woodwork and other
horizontal surfaces
[]
[)
ii
[]
i]
[]
3. HEPA vacuum walls and other vertical surfaces
D
I]
D
[]
[]
[]
4. Damp wipe horizontal surfaces (except floors)
[]
[]
n
[]
n
[]
5. Damp wipe vertical surfaces (trim, etc.)
[]
[]
[]
n
[]
[]
6. Wipe furniture with oil-treated rags
I]
[]
[]
n
n
[]
7. HEPA vacuum area rugs on both sides
(top and bottom)
a
u
0
ii
[]
[]
8. Floors (see part D. also)
A. Wall-to-wall carpets:
HEPA vacuum (with beater bar)
3 minutes per square yard
D
n
[]
n
D
I]
B. Wood, tile, etc.
1. Wash with clean rags using
TSP solution
ii
ii
[]
ii
I]
[]
2. HEPA vacuum at a rate of
2 minutes per square yard
(see schedule for each room
part D)
n
[]
ii
[]
[]
[]
C Area Rugs
HEPA vacuum on both sides, roll up rug,
clean floor
i)
ii
[]
[]
[]
[]
9. Move articles back to original positions
wearing foot covering
ii
ii
ii
ii
[1
a
-------
INTERIOR ABATEMENT DOCUMENTATION
PART D.
Room Code
Primary floor type
Wood, carpet,
linoleum, etc.
Area Rugs
Check if
Yes
HEPA vacuum time
required
Rugs sq. yds. X 3
Others sq. yd X 2
HEPA vacuum
timed and
completed
-------
I flilMkSx . 1 I LANDMARK ONE
IvVvJj/J k I ONE VAN DE GRAAFF DRIVE
UNq^UVJ ซ BURLINGTON. MA 01803
0ESK3NERSAMNSULTAMTS (617) 229-2050 FAX: (617) 229-0046
ESAT PROJECT
May 29, 1991
K-l-05-11
Mr. Scott Clifford
ESAT Deputy Project Officer
Environmental Services Division
U.S. EPA Region I
60 Westview Street
Lexington, Massachusetts 0217 3
Re: TID No. 01-9104-50
Standard Operating Procedure
Columbia X-MET 820 XRF
Dear Mr. Clifford:
Environmental Services Assistance Team (ESAT) member
Paul Killian has completed the Standard Operating Procedure
(SOP) for the Columbia X-MET 820 x-ray fluorescence
instrument. The task was requested by Beverly Fletcher and
Scott Clifford, EPA Task Monitors, and authorized under
Technical Instruction Document (TID) Number 01-9104-50.
Enclosed is the SOP. The ESAT demonstration on how to
operate the instrument has been scheduled for June 7, 1991 at
10:00 am. Please contact Paul Killian at 617/229-2050 should
you have any questions or comments.
Very truly yours,
ROY F. WESTON, INC
Paul F. Killian
Associate Project Scientist
Team Manager
ESAT Region I
/pfk
Enclosures
cc: Beverly Fletcher
-------
STANDARD PRACTICES MANUAL
ESAT Division
Operating Practice
Effective Initiated Reviewed Approved
Date: By: By: By: SP No.
May 29, 1991 ESAT-010083
Environmental Services Assistance Team
EPA Region I
STANDARD OPERATIONS
for the
COLUMBIA X-MET 820 X-RAY FLUORESCENCE INSTRUMENT
TABLE OF CONTENTS
1.0 PURPOSE
2.0 SCOPE
3.0 DEFINITIONS AND ACRONYMS
4.0 PRINCIPLES OF OPERATION
5.0 APPARATUS AND MATERIALS
5.1 SAMPLE PREPARATION
5.2 SAMPLE ANALYSIS
6.0 PROCEDURES
6.1 SAMPLE PREPARATION
6.1.1
Sample Container Preparation
6.1.2
Sample Preparation
6.2
SAMPLE ANALYSIS
6.2.1
Definition of Fields
6.2.2
Instrument Operation
6.2.3
LFK Order of Analysis
6.3
SAMPLE QUANTITATION
6.3.1
Determination of Peak Height
6.3.2
Determination of Sample Concentration
7.0 ATTACHMENTS
Revision 0
Page l of 11
-------
STANDARD PRACTICES MANUAL
ESAT Division
Operating Practice
Effective
Date:
Initiated
By:
Reviewed
By:
Approved
By:
SP No
May 29, 1991
ESAT-01-0083
1.0 PURPOSE
To enable the operator to analyze soils samples for lead
content using the Columbia X-MET 820 X-Ray Fluorescence Instrument.
2.0 SCOPE
This SOP will allow the operator to determine the
concentration of lead in soil samples. This SOP covers the
preparation of soil samples, operation of the Columbia X-MET 820
XRF instrument, and calculation of the results from the printed
spectra. Modifications to this SOP can be made to determine lead
content in other matrices as well as other elements in various
matrices.
3.0 DEFINITIONS AND ACRONYMS
EPA Environmental Protection Agency
ESAT Environmental Services Assistance Team
LCS Laboratory Control Sample
ppm Parts Per Million
RTN Hard Return on the Terminal
SOP standard Operating Procedures
SPL Sample
X-MET Columbia X-MET 820 X-Ray Fluorescence Instrument
XRF Energy Dispersive X-ray Fluorescence
Revision 0
Page 2 of 11
-------
STANDARD PRACTICES MANUAL
ESAT Division
Operating Practice
Effective
Date:
Initiated
By:
Reviewed
By:
Approved
By:
SP No
May 29, 1991
ESAT01-0083
4.0 PRINCIPLES OF OPERATION
When an atom is bombarded with x-rays from a radioisotope x-
ray source, it looses an electron from its inner shell. As a
result, one of the atom's outer electrons is repositioned to the
inner shell and emits energy. The energy emitted is at a specific
Kiloelectron Volt (KeV) depending on which element and which outer
electron was repositioned. The concentration of each element can
be determined by examining the height of the peak at the specified
KeV. This process is know as Energy Dispersive X-ray Fluorescence.
The advantage of the Energy Dispersive X-ray Fluorescence
(XRF) technique is that the sample is not destroyed in the
analysis. The sample remaining in a stable state, enables the
analyst to reanalyze the sample at a later date, or digest and
analyze the sample using other techniques such as inductively
coupled plasma (ICP) or atomic absorption (AA) spectroscopy, other
advantages include the quick turnaround of sample results, and the
ease of operating the instrument.
5.0 APPARATUS AND MATERIALS
5.1 SAMPLE PREPARATION
Sample cups, plastic, spectro-cup, Cat. No. 340, Somar Lab.
Inc., New York or equivalent.
Mylar film, 6 micron
60 Mesh sieve
Sample weigh boats
Powder funnel
Revision 0
Page 3 of ll
-------
STANDARD PRACTICES MANUAL
ESAT Division
Operating Practice
Effective
Date:
Initiated
By:
Reviewed
By:
Approved
By:
SP No
May 29, 1991
ESAT-01-0083
5.2 SAMPLE ANALYSIS
Columbia X-MET 820 X-Ray Fluorescence Instrument, with:
Cm-244 source
Texas Instrument 703 Data Terminal.
Printer paper - Thermal Fax Paper
6.0 PROCEDURES
6.1 SAMPLE PREPARATION
6.1.1 Sample Container Preparation
6.1.1.1 Invert cup and place a piece of 6 micron mylar film
over the bottom aperture.
6.1.1.2 Snap a retaining o-ring over the film onto the base
of the cup (o-ring teeth down).
6.1.1.3 Place cup upright and add enough soil to uniformly
cover the mylar film bottom of the cup.
6.1.1.4 Snap cap into place on top of the cup.
6.1.1.5 Label the sample cup with the sequential laboratory
Identification number and record that in the
instrument logbook.
6.1.2 Sample Preparation
6.1.2.1 An aliquot of the soil, 2 to 3 table spoons (10 to
15 grams), is removed with a spoon or spatula and
placed in a sample weigh boat.
Revision 0
Page 4 of 11
-------
STANDARD PRACTICES MANUAL
ESAT Division
Operating Practice
Effective Initiated Reviewed Approved
Date: By: By: By: SP No.
May 29, 1991 ESAT-010083
6.1.2.2 The weigh boat is marked with the laboratory
identification number and allowed to air dry under
a hood overnight at ambient laboratory temperature.
6.1.2.3 The dried soil sample is manually shaken in a 60
mesh sieve until approximately 1 gram of fines have
been collected. (Typically 10 to 15 seconds is
adequate.)
6.1.2.4 The fines are then transferred to the analysis
sample container using a glass powder funnel placed
over the sample container. The cap is then placed
on the sample container.
6.1.2.5 All excess soils from sample preparation are
discarded in the waste barrel in the preparation
hood.
6.1.2.6 The powder funnel, sieve, and spoon (or spatula)
will be cleaned between samples to remove soil
particles. The funnel and sieve will be blown free
of dust with compressed air. The spoon will be
wiped with disposal tissues.
6.2 SAMPLE ANALYSIS
6.2.1 Definition of Fields
6.2.1.1 >
This field details the type of sample analyzed. If the
instrument is being recalibrated, STD (Standard) would be
entered. Since the curves are prepared separately, no STD
should be entered. The analyst should enter SPL, for sample.
There is no default value for this field.
Revision 0
Page 5 of 11
-------
STANDARD PRACTICES MANUAL
ESAT Division
Operating Practice
Effective
Date:
Initiated
By:
Reviewed
By:
Approved
By:
SP No
May 29, 1991
ESAT-01-0083
6.2.1.2 LATEST?
This field details which spectra to print. Since the
spectra are not named, only the latest measurement may be
recalled. Therefore, the field should be left blank and the
hard return key pressed. The default for this field is the
last sample.
6.2.1.3 FIRST CHANNEL: 0 ?
This field details which point the spectra is to start.
The total spectra goes from Channel 0 to Channel 255. Since
the Lead peak is located around Channel 166, the spectra
should be viewed from Channel 140 to Channel 190. Therefore
the proper input for this field is 140. The default for this
field is 0.
6.2.1.4 LAST CHANNEL: 255 ?
This field details which point the spectra is to stop. The
total spectra goes from Channel 0 to Channel 255. Since the
Lead peak is located around Channel 166, the spectra should be
viewed from Channel 140 to Channel 190. Therefore the proper
input for this field is 190. The default for this field is
255.
NOTE: If you enter a Last Channel that is lower than the
First Channel, the instrument will print out an error message
and ask you to reenter the Last Channel value.
6.2.1.5 WINDOW: 1 ?
This field details how frequent the channels will be
printed in the spectra. The choices are from 1 to 4. If 1 is
chosen then every channel will be printed. If 4 is chosen
then every fourth channel will be printed. The proper input
for this field is 2. The default for this field is 1.
Revision 0 Page 6 of 11
-------
STANDARD PRACTICES MANUAL
ESAT Division
Operating Practice
Effective Initiated Reviewed Approved
Date: By: By: By: SP No.
May 29, 1991 ESAT-01-0083
6.2.1.6 RANGE,lower: 0 ?
This field details the lower scale of the curve. Since the
baseline of the peak is drawn from the two low points on
either side of the peak, the graph does not need to extend
below these two points. Therefore, the lower range is set at
250. The default for this field is 0.
NOTE: On occasions, one or both of the low points may fall
below the 250 mark. When this occurs, the graph should be
reprinted with the RANGE, lower set at 200. If the points
still fall below 200, the sample is to be reported as non-
detected.
6.2.1.7 RANGE,upper: ### ?
This field details the upper scale of the curve. Since the
Lead peak is the highest point from Channel 140 to Channel
190, the curve does not have to extend any farther than just
above the value of the peak. The value ### is the highest
point on the graph from Channel 140 to Channel 190.
Therefore, the upper range is set at the next higher multiple
of 25. (i.e. if ### equals 459, then enter 475.) The
default for this field is #/#.
6.2.1.8 40 CHARACTER PER LINE ?
This field details the size of the print. When a larger
Character Per Line is entered, the graph expands over more of
the paper. Therefore, the Character Per Line is set at 80.
The default for this field is 40.
6.2.2 Instrument Operation
6.2.2.1 Turn on the instrument by pressing the switch on
the back left face of the instrument.
Revision 0
Page 7 of 11
-------
STANDARD PRACTICES MANUAL
ESAT Division
Operating Practice
Effective Initiated Reviewed Approved
Date: By: By: By: SP No.
May 29, 1991 ESAT-01-0083
6.2.2.2 Turn on the printer by pressing the switch on the
back right face of.the printer.
6.2.2.3 Allow the instrument to warm up for 30 minutes.
6.2.2.4 Place the sample into the instrument by:
A. sliding the holder towards you;
B. opening the holder by lifting the top;
C. placing the sample into the open holder and
closing the top;
D. sliding the holder back into place.
6.2.2.5 Type in sample identification (i.e. 300 STD)
6.2.2.6 Press the START 1 key on the instrument.
Instrument will respond:
DATE: dd,mm,yy TIME: hh-mm-as
MEASURING:
MODEL 10 PROBE 1 50 SECONDS
After 50 seconds the analysis is complete, the instrument
will signal by beeping. The instrument prints:
ASSAYS: PB ##.##
>
6.2.2.7 Type the following:
Instrument response Analyst RgBPPngg
a. > SPL
b. LATEST?
c. FIRST CHANNEL: 0 ? 140
d. LAST CHANNEL: 255 ? 190
e. WINDOW: 1 ? 2
Revision 0
Page 8 of 11
-------
STANDARD PRACTICES MANUAL
ESAT Division
Operating Practice
Effective Initiated Reviewed Approved
Date: By: By: By: SP No.
May 29, 1991 ESAT-01-0083
f. RANGE,lover: 0 ? 250
g. RANGE,upper: ### ?. next higher multiple
Of 25
(i.e. if ### ซ= 408 then enter 425)
h. 40 CHARACTERS PER LINE ? 80
The spectra for the sample is printed, and the
instrument responds:
LATEST?
If all the points on the spectra fall above the
baseline (250), proceed to step 6.2.2.8. Otherwise,
reprint the spectra with the baseline (RANGE, lower) set
at 200. This is done by repeating steps 6.2.2.7.b -
6.2.2.7.h, and entering 200 at step 6.2.2.7.f instead of
250. Regardless of whether or not the points still fall
below the baseline (200), proceed to step 6.2.2.8.
6.2.2.8 Press the ESCAPE key twice.
6.2.2.9 Follow steps 6.2.2.4 through 6.2.2.7 for the
remaining samples to be analyzed.
6.2.3 LFK Order of analysis
6.2.3.1 The following standards are run from low to high:
(Empty sample cup)
(Laboratory # 5103)
(Laboratory # 5113)
(Labeled as 1600 STD)
(Laboratory # 4873)
(Laboratory t 4903)
Laboratory Control Sample (LCS) (Labeled as 880 STD)
a.
blank
standard
b.
300
PPm
standard
c.
900
PPm
standard
d.
1600
ppm
standard
e.
6000
PPm
standard
f.
13000
ppm
standard
Revision 0
Page 9 of 11
-------
STANDARD PRACTICES MANUAL
ESAT Division
Operating Practice
Effective
Date:
Initiated
By:
Reviewed
By:
Approved
By:
SP No.
Hay 29, 1991
ESAT-010083
6.2.3.2 Ten laboratory samples are analyzed. (Both
duplicates and replicates are considered laboratory
samples.)
NOTE: Duplicates are prepared during sample preparation
at a rate of one per twenty. Replicates are
analyzed at a rate of one per twenty.
6.2.3.3 One of the standards (b - e) is analyzed.
6.2.3.4 Steps 6.3.2.2 and 6.3.2.3 are repeated until the
analysis batch is complete, rotating the standards
(b - e).
6.2.3.5 Once the analysis batch is complete, all standards
are analyzed, including the LCS, as in step 2.2.1.
6.3 SAMPLE QUANTITATION
6.3.1 Determination of Peak Height
6.3.1.1 A straight line is drawn connecting the two low
points on either side of the peak.
6.3.1.2 The peak height is then measured, in millimeters,
from the straight line to the highest point on the
peak.
6.3.1.3 The corresponding number of counts is then
determined by:
(RANGE, upper - RANGE, lower)
132.5 mm (Length of full scale)
X peak height (mm)
Counts
Revision 0
Page 10 of 11
-------
STANDARD PRACTICES MANUAL
ESAT Division
Operating Practice
Effective
Date:
Initiated
By:
Reviewed
By:
Approved
By:
SP No
May 29, 1991
ESAT-01-0083
6.3.2 Determination of Sample Concentration
6.3.2.1 The analysis results (counts and concentration) of
all standards, except the LCS results, are
tabulated.
6.3.2.2 Two standard curves are then created using linear
regression. A low concentration curve consisting
of the blank, 300, 900, 1600, and 6000 standards
are used for all sample results less than 6000 ppm.
The high concentration curve consisting of blank,
1600, 6000, and 13000 standards are used for all
sample results greater than 6000 ppm. Both
standard curves are plotted through the point zero,
zero.
6.3.2.3 The slope of the appropriate curve is then
multiplied by the sample's counts to determine the
sample concentration.
6.3.2.4 The LCS results are determined as in 6.3.2.3 (using
the low standard curve) . The results must fall
within 20% of the true value (880 ppm).
7.0 ATTACHMENTS
Enclosed are two copies of the spectra of a Laboratory Control
Sample analysis. The first is the copy is unmarked. The second
copy details steps from the SOP for printing the spectra and
calculating the peak height in counts.
Revision 0 Page 11 of 11
/
-------
WHAT?
> LCS
DATE: 14.05.91 TIME* 13-21-47
MEASURING:
MODEL 10 PROBE 1 50 SECONDS
ASSAYStPB 85.97
> SPL
LATEST?
FIRST CHANNEL: 0 ? 140
LAST CHANNELS ฃ55 ? 190
WINDOW! 1 ? 2
RANGE,lower: 0 ? 250
RANGE,uppers 725 ? 750
40 character per line ?80
~~~~ LATEST SPECTRUM MEAS.TIME* 50 DATE: 14.05.91 13-23-25
CHANNEL COUNTS 250 500 750
I 1 1
140
692
142
647
144
531
146
502
148
453
150
421
152
416
154
478
156
491
158
565
160
586
162
629
164
725
166
703
168
690
170
650
172
615
174
546
176
527
178
478
180
449
182
423
184
431
186
478
188
490
190
485
LATEST?
-------
WHAT?
'SUp
> LCS
DATE* 14.05.91 TIME: 13-21-47
MEASURING*
MODEL 10 PROBE 1 50 SECONDS
ASSAYStPB 65.97
> SPL
LATEST?
FIRST CHANNEL8 0 ? 140
LAST CHANNEL: 255 ? 190
UIMDOUs l ? 2
RANEE.lower* 0 ? 250
RANGE,upper* 725 ? 750
40 character per line ?30
LATEST SPECTRUM HEAS
5*ซf
.a.a/7
CHANNEL
COUNTS
140
692
142
647
144
531
146
502
148
453
150
421
152
416
154
478
156
491
158
565
160
586
162
629
164
725
166
703
168
690
170
650
172
615
174
546
176
527
178
478
180
449
182
423
184
431
186
478
188
490
190
485
250
I-
TIME: 50 DATE: 14.05.91
500
1
13-23-25
750
j
SVtf (" 3 ' *iL
Pf ev\C- V\"ซ IvA-
(7SQ c.H - asvc^) % gl ^ . 3^.7c
mr>
LATEST?
(ff, 3.\ซI
i
-------
MANAGERS
DESGNER&OONSUIDMIS
ESAT PROJECT
LANDMARK ONE
ONE VAN DE ORAAFF DRIVE
BURLINGTON, MA 01603
(617) 229-2050 FAX: (617) 22ป-0046
February 12, 1991
K-l-02-05
Mr. Scott Clifford
ESAT Deputy Project Officer
Environmental Services Division
U.S. EPA Region I
60 Westview Street
Lexington, Massachusetts 02173
Re: TID No. 01-9102-17
Lead Free Kids Project
Review of the LFK Protocols Report
Dear Mr. Clifford:
Environmental Services Assistance Team (ESAT) member
Paul Killian has completed the review of the Lead Free Kids
(LFK) Project Protocols. The request was made by Beverly
Fletcher, EPA Task Monitor, and authorized under Technical
Instruction Document (TID) number 01-9102-17. The requested
start date was February 11, 1991. The estimated completion
date was February 12, 1991.
The task was initiated on February 11, 1991, and
completed on February 12, 1991. The task required reviewing
the LFK Protocols Report, Section II Soil Analysis Protocol,
pages A-4 through A-7, and Section IV Dust Analysis pages A-
11 through A-13. The methods were compared to the methods
submitted by ESAT on February 8, 1991 (Correspondence Number
01-9102-17). The following discrepancies were noted:
Soil Analyses
The procedure in which ESAT received soil samples is
slightly different than the procedure presented on Page
A-4 of the LFK Protocols Report. The procedure ESAT
followed is: The samples are received from LFK staff.
The Chain-of-Custodies (COCs) are checked to verify that
all samples are present. The COCs are signed and dated,
noting the time of sample receipt. The COCs are then
copied, returning the'original COC to the LFK staff.
Laboratory Identification numbers are then placed on the
sample bags and on the laboratory copy of the COCs.
-------
Scott Clifford
Two
February 12, 1991
K-l-02-05
The procedure in which ESAT prepared soil samples is not
the same as the procedures presented in the LFK
Protocols Report, steps 3-14, pages A-4 through A-6.
The procedures that ESAT followed are outlined in
Appendix B-4, sections 6.3 and 6.4. (Attachment I)
ESAT did not use the Kevex X-Ray Fluorescence (XRF)
instrumentation to analyze the soil samples. Soil
samples were analyzed using the Oxford LAB-X 1000 until
May 1990; then analyses were performed using the
Columbia X-MET 800 XRF. The procedures followed by ESAT
for the analysis of soil samples on the Oxford LAB-X
1000 XRF are presented in section 7.0 of Appendix B-4.
The procedures followed by ESAT for the analysis of soil
samples on the Columbia X-MET 800 XRF are presented in
Attachment II.
Dust Analyses
The procedure in which ESAT received dust samples is
slightly different than the procedure presented on Page
A-ll of the LFK Protocols Report. The procedure ESAT
followed is: The samples are received from LFK staff.
The Chain-of-Custodies (COCs) are checked to verify that
all samples are present. The COCs are signed and dated,
noting the time of sample receipt. The COCs are then
copied, returning the copied C0C to the LFK staff.
Laboratory Identification numbers are then placed on the
sample bags and on the laboratory copy of the COCs. The
original COCs are returned when analysis has been
completed.
The procedure in which ESAT prepared dust samples is
similar to the procedures presented in the LFK Protocols
Report, steps 2-6, pages A-ll through A-12. However,
step 5a, page A-12, states that "The minimum acceptable
sample is 20 mg.M In actuality there was no minimum
acceptable amount of sample. Several of the samples had
only 1 mg of sample.
The procedure in which ESAT analyzed dust sample is more
detailed than the procedures presented in the LFK
Protocols Report, pages A-12 through A-13. ESAT
followed section 7 of -Appendix B-3. (Attachment III)
-------
Mr. Scott Clifford February 12, 1991
Page Three K-l-02-05
The LFK Protocols Report, step 3c, page A-13, states an
acquisition time of 100 seconds; however, ESAT used an
acquisition tine of 30 seconds.
The LFK Protocols Report, step 4c, page A-13, states
that standards should be used to determine a linear
least squares calibration curve; however, ESAT
determined sample concentrations by directly comparing
the sample peak height to appropriate standard peak
height. ESAT followed the procedure detailed in section
7.6 of Appendix B-3.
ESAT reviewed the remaining sections of the LFK Protocol
Report; however, no comments were made because the sections
pertained to areas of the project with which ESAT was not
involved. Please contact Paul Killian at 617/229-2050 should
you require any additional information.
Very truly yours,
ROY F. WESTON, INC.
Paul F. Killian
Associate Project Scientist
John J. Hagopian, P.G.
Team Manager
ESAT Region I
/pfk
Enclosures
cc: Beverly Fletcher
-------
ATTACHMENT I
-------
APPENDIX B-3
STANDARD OPERATING PROCEDURE:
LABORATORY SCREENING METHOD FOR LEAD IN HOUSE DUST
USING ENERGY DISPERSIVE X-RAY FLUORESCENCE
(KEVEX 0700)
-------
Pflqg 1 9f 1?
CATEGORY: TITLE:
Field
Technical
Lead Free Kids
Demonstration Project
No.
ฆDate; ?/9Q
Revision: 0
1.0 SCOPE AND APPLICATION
1.1 Lead in household dust nay be determined by energy
dispersive X-ray fluorescence (XRF) spectrometry. This
method is simple, rapid, and applicable to Lead in
various matrices with little or no sample preparation
(i.e., digestion is not required prior to analysis).
1.2 Detection limits, sensitivity, and optimum ranges of
the metals will vary with regard to sample matrix as
well as the model of XRF instrument utilized.
1.3 This method is applicable for use by Region I ESD and
ESAT staff for performing XRF screening analyses in lead
in house dust samples as part of the LFK Demonstration
Project.
2.0 SUMMARY OF METHOD
This method may be used for the semi-quantitative
screening analysis of house dust samples for lead. The dust
sample is thoroughly sieved, and placed in a plastic sample
cup for XRF analysis. The intensity of the sample response
at the L-alpha energy region of lead is compared to known
lead reference standards for quantitation.
3.0 INTERFERENCES
Certain elements, such as , if
present in the soil at concentrations times that of
lead, could present difficulties in the identification and
quantitation of lead.
4.0 APPARATUS AND MATERIALS
4.1 Energy Dispersive X-Rav Fluorescence Spectrometer
A Kevex Model 7000 XES equipped with:
(a)
source;
(b)
detector
(c) sixteen (16) place rotating sample holder? and
-------
Page 2 of 13
CATEGORY: TITLE:
Field Lead Free Kids Ho.
Technical PeWPnstmiPn Prgiect Date: 3/gn
Revision: o
(d) computerized data system for analyzing, comparing
and storing sample spectra.
4.2 8 inch Floppy Data diskettes, IBM, or equivalent.
4.3 Sample cups, plastic, consisting of cup, o-ring, and
cap, Spectra-Cup, Cat. No. 340, Somar Labs. Inc., New
York, or equivalent.
4.4 Mylar film, 6 micron.
5.0 REAGENTS
5.1 U.S. Department Commerce. National Bureau of
Standards. Standard Reference Materials
Unit Certified Lead
SEH Type &iฃฃ cpn?sntmipn
1579 Powdered Lead 35g 11.87%
Base Paint
1633a Coal Fly Ash 75g 72.4 ug/g
1645 River Sediment 70g 714 ug/g
1646 Estuarine Sediment 75g 28.2 ug/g
1648 Urban Particulate 2g 0.655%
5.2 US EPA. Environmental Monitoring and Surveillance
Laboratory (EMSL1. Quality Control Reference Standards
5.3 Instrument Calibration Standards
Dust M-10 2500 ppm 10 ng.
Dust M-50 2500 ppm 50 mg.
Dust H-10 25,000 ppm 10 mg.
Dust H-50 25,000 ppm 50 ng.
-------
Page 3 of n
CATEGORY: TITLE:
Field Lead Free Kids No.
Technical Demonstration Project Pate; 3/90
Revision: o
6.0 Sample Collection and Transfer of Custody to the U.S.
EPA
Samples are collected in the field by Lead-Free-Kids
staff, placed in labeled individual envelopes, and submitted
with chain-of-custody (COC) documentation to the U.S. EPA New
England Regional Laboratory (NERL) for XRF analysis. . U.S.
EPA personnel or their contractors will acknowledge receipt
of custody by signing and dating the COC document in the
presence of the LFK dust sample courier. The COC document
is retained until sample analysis has been completed and
results have been entered onto it. Then the original COC is
returned to LFK with a cover letter.
6.1 Sample Preparation
6.1.1 Samples are assigned unique laboratory
identification numbers, a sequential five-digit
number, which is subsequently recorded on the
sample envelope, chain-of-custody document, XRF
Dust preparation worksheet, XRF analytical result
summary sheet, and on the cover of the sample
analysis container.
6.1.2 Under the ventilation hood, the sample envelope
is carefully opened at one end (with scissors)
and the dust is placed into a 60 mesh sieve.
6.1.3 The sieve is manually shaken for approximately
15 to 20 seconds.
6.1.4 All the fines are then transferred to the pre
weighed sample analysis container using a glass
powder funnel centered over and touching the
center of the mylar window of the sample
container.
6.1.5 Information from the chain-of-custody, including
weight of sample, and laboratory ID number is
recorded on the analytical results summary form.
6.1.6 All of the excess (non-filtered) soil/dust from
the sample preparation is discarded in a special
barrel in the laboratory, in some cases filtered
-------
Paq
-------
Paoe 5 of n
CATEGORY: TITLE:
Field Lead Free Kids No.
Technical Demonstration Project Date: 3/90
Revision: 0
6.2.8 Label the sample container cap with the correct
sequential laboratory sample ID number.
6.3 Standards Preparation
Study Control standards are prepared from previously
analyzed and concentration verified house dust samples.
Standard concentrations should be prepared at concentration
levels and weigh ranges as presented below.
Sample
Calibration
Height
Stat Ranoe
C/tSff
M-10 or 0.0 - ฆฃ)ra4g '
H-10 e,oz5""q
M-10 or Q-ซ Bftg or
greater
K10
Dust
M-10
2500
ppm
10
mg
Dust
M-50
2500
ppm
50
mg
Dust
H-10
25,000
ppm
10
mg
Dust
H-50
25,000
ppm
50
mg
6.4 Sample Preservation and Handling
No preservation is required. Handling of the sample,
once it is placed in the analysis cup, must be done in a
gentle manner to keep the sample centered in the middle of
the mylar. This is especially important for samples
requiring replicate analysis.
7.0 ANALYSIS PROCEDURE
7.1 The use of the Kevex 7000 XRF is relatively
straightforward. The Kevex is normally left in the
standby mode (target .8, 30 kVf and 0.5 mA) between
analyses to prevent x-ray tube damage. House dust
samples for lead are analyzed under the following
instrumental conditions: target .4, 30 kV, .5 mA.
(Detailed instructions can be found in the user's Manual
for Kevex XRF Software.)
-------
Page 6 of n
CATEGORY: TITLE:
Fir1.d Lea* Free Kids No.
Technical PftfflPngtratlPn Prpjggt Data; 3/09
Revision: 0
7.2 instrument Set-Vp
7.2.1 Turn the video monitor end plotter power on.
7.2.2 Insert the Master floppy disk into disk drive
No. 0 (DYO)
7.2.3 Insert formatted floppy disk into disk drive No.
1 (DY1).
7.2.4 Boot the operating system by pressing the "Shift"
and "Reset" keys simultaneously. Next, press the
"Q Vantx" and then the "Enter" key.
7.2.5 When prompted on the screen, enter the current
date.
7.2.6 After the current date has been entered, the
spectral region of interest for lead must be
established. This is accomplished by pressing
the blue double-headed arrow (< >) key.
The region of interest that should be obtained
is from 7.04 Kilo-electron Volts (KeV) to 17.28
KeV, where the lead L-alpha (L-a) peak is 10.2S
KeV and the lead L-beta (L-b) is 10. KeV.
After the spectral region has been established
for lead analysis, wait for the asterisk (*)
prompt and type in ATO, PBS0IL4. Type in sample
ID Numbers as 5 digit numbers followed by -0- for
each number at the end.
ex: Lab ID I 143 entered as 00143-D-
7.2.7 The first carousel run on the Kevex for the day
must contain all four calibration standards.
Each additional carousel run must include one of
the four study control standard on a rotating
basis. Calibration standards are run manually
and not on the ATO program.
7.3 landing ths Kevsx Sampler (Carousel)
7.3.1 Push the "Reset"'key (red) to shut-off the x-ray
beam. (As a safety precaution, the lid will not
open when the x-ray beam is functioning).
-------
Page ?
-------
Page 8 of
CATEGORY: TITLE:
Field Lead Free Kids No.
Technical Demonstration Project Date; 3/90
Revision: 0
7.4.7 When you have stopped the peak at its desired
height (2.5) type SMO to smooth the curve. If
the peak now falls below 2.5 it may be necessary
to continue acquiring the peak for a couple more
seconds and again hit stop to halt peak.
Alternating between acquire, stop and smooth may
be done an unlimited number of times until the
peak appears in the right position as long as the
time count is below 30 seconds. Time of analysis
may not run over 30 seconds.
Note: Only calibration standards will be run on manual
not dust samples.
7.4.8 Await (*); type "RฃA"d, press "Enter".
7.4.9 Await (*)? type "SAV"e, press "Enter".
7.4.10 Prompt: General Comments.
7.4.11 Response: Section is ignored, press "Enter".
7.4.12 Prompt: Enter Unit: (1) or (2).
7.4.13 Response: Type "1", press "Enter".
* 7.4.14 Prompt: Enter Sample ID":
7.4.15 Response: type in Sample ID as assigned in XRF
dust preparation worksheet.
* Manual analysis does not automatically add a 4
onto the end of the identification label and
therefore the 4 is not needed for recall
purposes.
7.5 Automatic Analysis Procedure
7.5.1 At asteric on screen type ato.pbsotla Enter.
7.5.2 Enter the last sample position but do not
include standards that will run manually.
7.5.3 Enter lab ID numbers for each corresponding
position.
-------
Page 9 Pf 11.
CATEGORY: TITLE:
Field Lead Free Kids No.
Technical PgfflPnstmign Project Date: 3/90
Revision: 0
7.5.4 As the program runs you roust be present to
observe each lead sample peak as it acquires
for 30 seconds.
7.5.5 With screen parameters of <7.04 and 17.28> the
compton scatter peak will be the last peak
visible on the right hand side. The lead peak
will appear directly above the blue arrow at
the bottom of the screen.
7.5.6 If the lead peak rises faster then the compton
peak it will be calibrated using the high
standard. If the lead sample peak does not
rise above the compton peak, the medium
standard will be used.
7.5.7 To determine if the 10 standard or the 50
standard is to be used, identify the weight of
the sample. The sample is:
O.OOg - 0.024g use 10 standard
0.025g - O.lOOg or above use 50 standard.
7.6 Manual Quantitation and Comparison of Dust Samples
7.6.1 Await (*); type "RCL" (recall), press "Enter".
NOTE: The RCL (recall) command is used to recall a
previously analyzed spectra that has been stored
on the floppy diskette (DY1). In this case, a
previously analyzed lead in dust calibration or
reference standard for comparison to the various
dust samples analyzed and stored on the same
diskette.
7.6.2 Prompt: Enter Unit: 1 or 2.
7.6.3 Response: Type "1", press "Enter".
7.6.4 Prompt: Enter ID:
7.6.5 Response: Type the standard/label ID, press
"Enter".
7.6.6 Prompt: Smooth Recalled Spectrum (Y/N)?
-------
Page 10 of
CATEGORY: TITLE:
Field Lead Free Kids Ho.
Technical Demonstration Prelect Date; 3/90
Revision: 0
7.6.7 Response: Press "Enter".
7.6.8 Await (*); type "OVR" (overlay), press "Enter".
The overlay command is used to compare and
normalize spectra from the disk. The
normalization feature (OVR) allows the operator
to mark regions within the displayed spectrum
as a basis for normalization. This feature
aids in the visual interpretation of data and
reduces channel~to-channel statistical
fluctuations.
7.6.9 Prompt: Enter ID: add -D-4 to the end of each
ID.
7.6.10 Response: Enter the sample ID, press "Enter".
7.6.11 Prompt: Smooth Recalled Spectrum (Y/N)?
7.6.12 Response: Press "Enter".
7.6.13 Prompt: Mark Peak(s) or Region(s) Hit Enter When
Ready a cursor will appear on the screen.
7.6.14 Response: Mark the regions to be used for
normalization by moving the cursor with the left
and right green arrow function keys. The peak
to be painted is the compton scatter peak. The
screen parameters should be 9.60 - 19.84 use the
green" equal (ฆ) key to paint the desired area.
Note: the paint cursor will move in the
direction it was last set. Press the "Enter" key
when finished.
7.6.15 The screen display will now include the standard
spectrum overlaid by the sample spectrum
normalized to the same energy region of the
spectrum. Direct comparison of the lead (L-a)
peaks can be made and a concentration (in ppm)
can be determined.
Note: The red peak is the standard peak which should
read 2.5 (use the Blue up and down arrows to set
this). The white peak is the sample peak. Use
the blue up and down arrows to best compare the
sample peak value ppm. Although the height of
-------
Page 11 of n
CATEGORY: TITLE:
Field Lead Free Kids No.
Technical Demonstration Project Date: 3/90
Revision: 0
the red and white peaks will change the ppm value
of the red (standard) will always remain the same
2500 ppm or 25,000 ppm depending on the standard
used.
7.6.16 The OVR sequence can be repeated for each sample
on the disk (DY1).
7.7 A Modified Quantitation Procedure - This is basically
the same procedure as described above.
Dr. T. Spittler, USEPA Region I, Technical Services
Branch Chief, Lexington, Massachusetts initiated the use of
a quick and easy method for the semi-quantitative analysis
of lead in soil samples.
Dr. Spittler has determined that, when acquiring data
for the 2000 ppm lead in soil standard at an attenuation of
512 and the energy level for the compton's back scattering
energy peak at 15 KeV is at 50 percent intensity, each
horizontal screen division is equivalent to the response of
ca. 800 ppm lead. To utilize this technique for dust, follow
the XRF instrument set-up guidelines as previously described
in Sections 7.2, 7.3, and 7.4 (7.4.1 to 7.4.5). To acquire,
quantify, and store data, utilize the following procedure:
.7.1 Check sampler position at n0n.
.7.2 Await (*); press the yellow "ACQ" key.
.7.3 Wait for energy level at 15.- KeV to reach 50
percent scale at a range of 512.
.7.4 Press the yellow "Stop** key.
.7.5 Await (*); type "SMO", press "Enter".
.7.6 Await (*); type "REA", press "Enter".
.7.7 Await (*)t type "SAV", press "Enter".
.7.8 Prompt: General Comments.
.7.9 Response: Section is ignored, press "Enter".
-------
Page 12 of 13
CATEGORY: TITLE:
Field
Technical
Lead Free Kids
Demonstration Project
No.
Date; 3/90
Revision: 0
7.7.10 Prompt: Enter Unit: (1) or (2).
7.7.11 Response: Type "1", press "Enter".
7.7.12 Prompt: Enter Sample ID:
7.7.13 Response: type in sample ID as assigned in the
XRF logbook.
7.7.14 Quantify the L(a) lead peak using the following
scale:
7.7.15 Await (*)? type "CI*" (clear), press "Enter".
7.7.16 Advance the sample tray one space and repeat the
analysis procedure.
8.0 QUALITY CONTROL
8.1 All quality control data should be maintained and
available for easy reference or inspection.
8.2 At the beginning of each operating shift all 4 study
control standards are analyzed on the first carousel.
On following carousel runs analyze one standard (one per
sixteen) This is done to assess method accuracy and to
correct for normal standard drift and results should
agree within ฑ20 percent of the true value.
8.3 At least one laboratory replicate should be analyzed for
every 20 samples to verify precision of the method.
Replicate samples may be run at the end of an analytical
day in their own carousel.
Attenuation
Concentration Range of Lead
fvertical scale division concentration)
64
128
256
512
1024
0 to 700 (100 ppm)
0 to 1400 (200 ppm)
0 to 2800 (400 ppm)
0 to 5600 (800 ppm)
0 to 11,200 (1600 ppm)
8.4
At least one laboratory replicate should be analyzed at
a frequency of 1 per 20 samples to verify precision of
-------
Pace 13 of 13
CATEGORY: TITLE:
Field Lead Free Kids No.
Technical Pempnstratipn Project Date: 3/90
Revision: o
the method. Replicate samples maybe run at the end of
an operation shift.
NOTE: True replicates of soil and dust samples are
usually not possible since chemicals such as lead
are typically not uniformly distributed in these
materials. Additional handling of the sample may
cause the dust to migrate away from the center
of the mylar. Care must be taken when handling
samples. Care must be taken in the
interpretation of soil and dust replicate anal-
ytical results.
9.0 METHOD REFERENCE
9.1 Precision and accuracy data are not available at this
time.
9.2 The performance characteristics for a dust sample free
from interferences are:
Optimum Concentration Range: N/A ug/g
Detection Limit: NA ug/g
N/A: not available at this time.
-------
ATTACHMENT XX
-------
F-MBT 800 m
Standard Operating Procedures
1.0 SAMPLE PREPARATION
All procedures detailed in section 6.0 of Appendix B-4 were
followed for sample preparation.
2.0 SAMPLE ANALY8X8
2.1 Instrument Operation
2.1.1 Turn on the instrument by pressing the switch in back left
face of the instrument.
2.1.2 Turn on the printer by pressing the switch in back right
face of the printer.
2.1.3 Allow the instrument to warm up for 30 minutes.
2.1.4 Place the sample into the instrument by:
A. sliding the holder towards you;
B. opening the holder by lifting the top;
C. placing the sample into the open holder and closing the
top;
D. sliding the holder back into place.
2.1.5 Turn the printer off line by alternating the ON LINE switch
away from the
2.1.6 Type in sample identification (i.e. 300 STD)
2.1.7 Turn the printer on line by aternating the ON LINE switch
towards the
2.1.8 Press the START 1 key on the instrument.
Instrument will respond:
DATE: dd,mm,yy TIME: hh-mm-ss
MEASURING:
MODEL 10 PROBE 1 50 SECONDS
After 50 seconds the analysis is complete, the instrument will
signal by beeping. The instrument prints:
ASSAYS: PB #*.##
>
-------
2.1.9
Type the following:
Instrument resDonse
Analvst ResDonse
a.
>
SPL
b.
LATEST?
c.
FIRST CHANNEL: 0 ?
140
d.
LAST CHANNEL: 255 ?
190
e.
WINDOW: 1 ?
2
f.
RANGE, lower: 0 ?
250
g-
RANGE, upper: ill ?
next
higher multiple of
25
(i.e. if III * 108 then
enter 125)
h.
40 CHARACTERS PER LINE ?
80
The spectra for the sample is printed, and the instrument responds:
LATEST?
If all the points on the spectra fall above the baseline (250),
proceed to step 2.1.10. Otherwise, reprint the spectra with the
baseline (RANGE, lower) set at 200. This is done by repeating
steps 2.1.9.b - 2.1.9.h, and entering 200 at step 2.1.7.f instead
of 250. Regardless of whether or not the points still fall below
the baseline (200), proceed to step 2.1.10.
2.1.10 Press the ESCAPE key twice.
2.1.11 Follow steps 2.1.4 through 2.1.8 for the remaining samples
to be analyzed.
2.2 LFK Order of analysis
2.2.1 The following standards are run from low to high:
(Empty sample cup)
(Laboratory # 5103)
(Laboratory # 5113)
(Labeled as 1600 STD)
(Laboratory # 4873)
(Laboratory # 4903)
g. Laboratory Control Sample (LCS) (Labeled as 880 STD)
2.2.2 Ten laboratory samples are analyzed. (Both duplicates and
replicates are considered laboratory samples.)
NOTE: Duplicates are prepared during sample preparation at a
rate of one per twenty. Replicates are analyzed at a
rate of one per twenty.
2.2.3 One of the standards (b - e) is analyzed.
2.2.4 Steps 2.2.2 and 2.2.3 are repeated until the analysis batch
is complete, rotating the standards (b - e).
2.2.5 Once the analysis batch is complete, all standards are
analyzed, including the LCS, as in step 2.2.1.
a.
blank
standard
b.
300
ppm
standard
c.
900
ppm
standard
d.
1600
ppm
standard
e.
6000
ppm
standard
f.
13000
ppm
standard
-------
3 . 0 SAMPLE QUANTITATION
3.1 Determining Peak Height
3.1.1 A straight line is drawn connecting the two low points of
the curve.
3.1.2 The peak height is then measured, in millimeters, from
the straight line to the highest point on the peak.
3.1.3 The corresponding number of counts is then determined by:
(RANGE, upper - RANGE, lower)
x peak height (mm) * Counts
132.5 mm (Length of full scale)
3.2 Determining Sample Concentration
3.2.1 The analysis results (counts and concentration) of all
standards, except the LCS results, are tabulated.
3.2.2 Two standard curves are then created using linear
regression. A lower curve consisting of the blank, 300,
900, 1600, and 6000 standards are used for all sample
results less than 6000 ppm. The high curve consisting of
blank, 1600, 6000, and 13000 standards are used for all
sample results greater than 6000 ppm. Both standard
curves are plotted through the point zero, zero.
3.2.3 The slope of the appropriate curve is then multiplied by
the sample's counts to determine the sample
concentration.
3.2.4 The LCS results are determined as in 3.2.3 (using the low
standard curve). The results must fall within 20% of the
true value (880 ppm).
-------
ATTACHMENT ZIZ
-------
APPENDIX B-4
STANDARD OPERATING PROCEDURE:
LABORATORY SCREENING METHOD FOR
LEAD IN SOIL USING ENERGY
DISPERSIVE X-RAY FLUORESCENCE
OXFORD LXIOOO
-------
Paqe2 pt 12
CATEGORY:
Field
Technical
TITLE:
Lead Free Kids
Demonstration Project
No.
Date: 3/90
1.0 SCOPE AND APPLICATION
1.1 Metals in a solution nay be readily determined by energy
dispersive x-ray fluorescence (XRF) spectrometry. The method
is simple, rapid, and applicable to a large number of metals
in various matrices with little or no sample preparation
(i.e., digestion is not required prior to analysis).
1.2 Detection limits, sensitivity, and optimum ranges of the
metals will vary with the sample matrices and the models of
XRF spectrometers utilized.
1.3 This method is applicable to Region I ESD and ESAT staff
performing laboratory screening analyses for lead in soil
samples collected as part of the LFK Demonstration project.
2.0 SUMMARY OF METHOD
This method is used for the semi-quantitative screening of
lead in soil. The soil sample is homogenized, an aliquot is
removed and placed in a sampling container. The sample is then
analyzed using XRF.
3.0 INTERFERENCES
Certain elements, such as f present in the
soil sample could interfere with the analysis, if present in
concentrations greater than times that of lead.
4.0 APPARATUS AND MATERIALS
4.1 Energy Dispersive X-Rav Fluorescence
An Oxford Analytical Instrument LAB-X 1000 equipped with:
excitation source: Cadmium 109
typical activity: 3 milli Curies (3MCi)
half life: 1.3 years
principal energy level: silver, K, 22 KeV
atomic no range:(K) spectra, 24-42; (L) spectra, 72-92
detector: xenon filled proportional counter
six (6) position motorized turntable
microprocessor control consisting of:
-------
Pflqg3 Ot 12
CATEGORY: TITLE:
Field Lead Free Kids Ho.
Teghnical Demonstration Project Date: 3/90
display: 40 column, 2 line liquid crystal display
printer: 40 column, 2 color dot matrix with
graphics, uses 70mm wide plain paper
keypad: 20 Key alphanumeric membrane pad.
4.2 Printer paper, 70mm wide.
4.3 Printer ribbon.
4.4 Sample cups, plastic, spectro-cup, Cat. No. 340, Somar Lab.
Inc,, New York or equivalent.
4.5 Mylar film, 6 micron
4.6 A stable power supply, whose requirements of 100-120 volt AC,
45-165 Hz, 50 VA maximum consumption are critical to
instrument performance. Extreme temperature ranges also
effect instrument performance.
5.0 REAGENTS
5.1 U.S. Department of Commerce. National Bureau of Standards.
standard Reference Material*
Unit Certified Lead
ฃฃH Type Sizฃ Concentration
1579 Powdered Lead Base 35? 11.87%
Paint
1633a Coal Fly Ash 75g 72.4 ug/g
1645 River Sediment 70g 714 ug/g
1646 Estuarine Sediment 75g 28.2 ug/g
1648 Urban Particulate 2g 0.655%
5.2 us EPA. Environments 1 Monitoring and Surveillance Laboratory
IEMSL1, Quality control Reference Standards
5.3 Tnatruihent Calibration Standards
Not available at this tine.
-------
Paoc4 of 12
CATEGORY: TITLE:
Field Lead Free Kids No.
Technical Pgnomrtratipn Project pate; 3/qo
6.0 SAMPLE COLLECTION, PRESERVATION, AND HANDLING
6.1 Samples are collected in the field, placed in labelled,
individual, zip-lock plastic bags, and submitted to the ESD
laboratory for analysis. Samples are logged into the
laboratory logbook and assigned a laboratory identification
number.
6.2 Soil samples are thoroughly mixed (homogenized) in the zip-
lock bag. An aliquot of the soil, 2 to 3 table spoons (10 to
15 grams), is removed with a spoon or spatula and placed in
a wang dish or appropriate drying vessel. The dish is marked
with the laboratory identification number and allowed to air
dry overnight at ambient laboratory temperature.
6.2.1 Excess sample in the zip lock bag will be stored until
the analytical report has been finalized then
discarded. However, selected soil samples maybe kept
longer for additional testing.
6.3 Sample Preparation
Dried soil samples will be passed through a 60 mesh sieve
until approximately 1 gram of fines have been passed. The sieve
will be manually shaken, typically 10 to 15 seconds is adequate.
The fines are then transferred to the analysis sample container
using a glass powder funnel which is placed over the sample
container.
6.3.1 All excess soils from sample preparation will be
discarded in a special barrel in the laboratory.
6.3.2 The powder funnel, sieve, drying vessel, and spoon (or
spatula) will be cleaned between samples to remove
soil particles. The funnel and sieve will be blown
free of dust with compressed air. The spoon will be
wiped with disposal tissues and drying vessel washed
vigorously with hot water.
6.4 Sample Container Preparation
6.4.1 Invert cup and place a piece of 6 micron mylar film
over the bottom aperture.
6.4.2 Snap a retaining o-ring over the film onto the base
of the cup (o-ring. teeth down).
6.4.3 Place cup upright and add enough soil to uniformly
cover the mylar film bottom of the cup.
-------
Page5 of 13
CATEGORY: TITLE:
Field Lead Free Kids No.
Technical Demonstration Project Date; 3/so
.4.4 Snap cap into place on top of the cup.
6.4.5 Label the sample cup with the sequential laboratory
I.D. No. and record that in the XRF instrument
logbook.
NOTE: Information to be recorded in the XRF logbook would
include:
field identification numbers;
laboratory identification numbers;
date samples prepared;
date samples and analyzed;
analysis parameters; and
analyst's initials affiliation and date.
6.5 Standards Preparation
Calibration standards are prepared from previously analyzed and
concentration verified soil samples or known reference standards.
Standard concentrations should be prepared at concentration levels
of lead at approximately:
50 - 100 ppm (ug/g)
100 - 500 ppm
500 - 1000 ppm
1000 - 2000 ppm
2000 - 5000 ppm
6.6 No special preservation or handling procedures are
required.
7.0 ANALYSIS PROCEDURE
The use of the Oxford Analytical Instrument Model LAB-X 1000
XRF is relatively simple. (Detailed instructions for its use can
be found in the LAB-X 1000 Instruction Manual.1
7.1 Instrument Set-Up
7.1.1 Turn power on.
7.1.2 Wait for menu to appear in video display.
7.1.3 Press key n3M to select Utilities routine.
-------
Paq
-------
Paoe7 of 12
CATEGORY: TITLE:
Field Lead Free Kids No.
Technical PgTOPngtratiPn Project Data; T/c>ft
7.2.9 Prompt: Is "Sample Label" Inserted?
7.2.10 Response: press "Yes" key.
7.2.11 The measurement cycle now begins. The turntable will
rotate 60 degrees, carry out an Energy Lock for Ca.
10 seconds prior to further rotation which transports
the sample to the required sampling head. The
operator may terminate a measurement by pressing the
"Esc"ape key before the programmed time has elapsed.
7.2.12 After completion of the measurement cycle, select
option 2, Print Scan. Press key "2".
7.2.13 After the scan has been printed, determine if the lead
L-alpha peak is on scale and measurable. If not,
select one of the three (3) scaling options: 5, 10,
or 20. Press the appropriate key.
7.2.14 After the scale scan has been printed, select option
4, Turn Page, to return to the Analyses Menu. Press
key "4".
7.2.15 Place another sample into the sample holder and repeat
the analysis process.
7.3 Quantification
7.3.1 A series of calibration standards are analyzed at each
scaling factor; 0, 5, 10, and 20. An average response
factor (RF) is determined using a minimum of three (3)
concentrations and one (1) reagent blank analyzed at
least three times.
7.3.2 The peak height of the lead L-alpha (at ) is
measured for each sample. This peak height is
multiplied by the RF to determine the concentration
of lead (ppm) in the sample.
8.0 QUALITY CONTROL
8.1 All quality control data should be maintained and available
for easy reference or inspection.
8.2 A set of calibration standards at each scaling factor should
be analyzed in the laboratory prior to initiating field
studies. These calibration standards should consist of a
minimum of three (3) standards and one (1) reagent blank
("clean soil").
-------
Paae8 of 12
CATEGORY: TITLE:
Field Lead Free Kids No.
Technical Demonstration Project Date: 3/90
8.3 A minimum of one (1) reagent blank and one (1) standard at or
near the mid-range of the calibration curve should be analyzed
daily to verify instrument reproducibility. These values
should agree within ฑ20 percent of the initial calibration.
8.4 If forty-five (45) or more samples per day are analyzed or if
samples from more than one site are to be analyzed in one day,
then the working standard curve must be verified by analyzing
a mid-range standard for every thirty (30) samples or for each
siter whichever is more frequent. These check standard
results must be within ฑ20 percent of the true value.
8.5 At least one (1) field laboratory duplicate sample should be
analyzed with every twenty (20) samples to verify the
precision of the method.
NOTE: True replicates of soil samples are usually not
possible since chemicals such as lead are typically
not uniformly distributed in these materials. Care
must be taken in the interpretation of soil replicate
analytical results.
8.6 At least one (1) lead-in-soil standard reference sample should
be analyzed daily or per site, which ever is more frequent.
The result should agree within ฑ 20 percent of the true value.
9.0 METHOD PERFORMANCE
9.1 Precision and accuracy data are not available at this time.
9.2 The performance characteristics for a soil sample free from
interferences are:
Optimum Concentration Range: N/A ug/g
Detection Limit: N/A ug/g
NA: not available at this time.
-------
ROY F. WESTON. INC.
LANDMARK ONE
ONE VAN DE GRAAFF DRIVE
BURLINGTON. MA 01803
(617)229-2050
March 1, 1990
D-0-2-30
Mr. Scott Clifford
ESAT Deputy Project Officer
Environmental Services Division
US EPA - Region I
60 Westview Street
Lexington, Massachusetts 02173
Re: TID No. 01-9001-25
LFK Demonstration Project
Quality Assurance Project Plan
Revisions to Interim DRAFT
Dear Mr. Clifford:
Environmental Service Assistance Team (ESAT) member Jay
Markarian prepared an interim revised draft of the Boston
Lead Free Kids Demonstration Project Quality Assurance
Project Plan (QAPjP). The task, requested by David Mclntyre,
EPA task monitor, was authorized under Technical Instruction
Document (TID) No. 01-9001-25. The requested start date was
January 29, 1990. An interim revised draft QAPjP was
requested for submission on February 27, 1990 with the final
version due on April 27, 1990.
The first revision of the April 28, 1989 QAPjP was
initiated on January 29, 1990. J. Markarian met with D.
Mclntyre on February 8, 1990 to discuss the scope of
revisions required. After this meeting D. Mclntyre edited
sections 1.0 through 6.0 of the QAPjP, addressing program
changes, which included addition of the dust sampling and
analysis elements of the project. J. Markarian has reviewed
the edits made by D. Mclntyre and edited the remaining
sections of the QAPjP, which included the following:
All tables and figures;
Sections 7.0 through 11.0;
the Detailed sampling SOP;
Deletion of former Appendices B-2, B-3, B-5 and B-
6?
-------
su
Mr. Scott Clifford
Page Two
March 1, 1990
D-0-2-30
Development of a dust sample preparation and
analysis SOP and;
Incorporation of a dust sampling SOP developed by
LFK staff, into the sampling guidelines found in
Appendix A of the QAPjP.
The enclosed diskette contains a merger of edits
provided by D. Mclntyre and J. Markarian under the file name
QAPP3Bl.Jay. "Redline" and "Strikeout" modes, available on
WordPerfect software version 5.0, have been used to assist
D. Mclntyre with the QAPjP's review.
Current status of this TID is summarized below:
The interim draft QAPjP has been submitted by ESAT
to USEPA for comment;
76 of the 120 assigned labor hours have been
expended (approximately 64 percent);
The task appears to be on schedule and on budget,
barring extensive comments by the US EPA.
Please contact Jay Markarian at 617/229-2050 should you
require any additional information.
Very truly yours
ROY F. WESTON, INC.
P'ay Markarian, P.G., CHMM
Senior Investigation Coordinator
Joseph D. Mastone
Team Manager
ESAT Region I
JSM/dam
cc: D. Mclntyre, US EPA
-------
Section No. 1
Revision No. 1
Date: 02/28/90
Page 1 of 55
1.0 QUALITY ASSURANCE PROJECT PLAN
Project Title: Lead Free Kids (LFK) Demonstration Project
EPA Project Manager: David Mclntyre
LFK Principal Investigator: Michael Weitzman, M.D;
Performing Organization: Trustees of Health and Hospitals
of the City of Boston, Inc. (Trustees)
Lead Free Kids Project, Department of Health and Hospitals city of
Boston
Duration: 2 years (See Project Design)
Type of Project: Superfund Epidemiological/Soil Abatement
Demonstration Project
Supporting Organization: EPA Region I
Approvals:
EPA, Region I
Name: David Mclntyre Title: Project Manager
Signature Date
Name: Carol Wood Title: Quality Assurance Coordinator
Signature Date
Name: Dr. Thomas Spittler Title: Technical Project Director
Signature Date ;
LFK Project
Name: Michael Weitzman, M.D. Title: Principal Investigator
Signature Date
Name: John L. Christian Title: VP Trustees of Health & Hospitals
Signature: Date
-------
Section No. 2
Revision No. 1
Date: 02/28/90
Page 2 of 55
2.0 TABLE OF CONTENTS
Section No. Page
1.0 QUALITY ASSURANCE PROJECT PLAN . 1
2.0 TABLE OF CONTENTS 2
3.0 PROJECT DESCRIPTION 9
3.1 Introduction 9
3.2 Project Summary 9
3.2.1 Project Background 9
3.2.2 Demonstration Project 10
3.3 Project Objectives 12
3.3.1 Major Task Summary 12
/4.0 PROJECT ORGANIZATION AND RESPONSIBILITIES 14
-4.1 Organization 14
4.2 Responsibility for Quality Assurance 14
'5.0 QUALITY ASSURANCE OBJECTIVES 18
5.1 General 18
5.2 Representativeness 18
5.3 Precision and Accuracy 18
5.4 Completeness 19
5.5 Comparability 19
5.6 Quality Assurance Objectives 19
6.0 SAMPLING PROCEDURES 21
6.1 general 21
6.2 Eguipment List 22
6.3 Sample Collection 22
6.4 Sample Handling and Storage 23
6.5 Record keeping 23
6.6 Preliminary Soil Sampling 23
6.6.1 Site Description 23
6.6.2 Sampling Schemes 24
6.7 periled s
-------
Section No. 2
Revision No. l
Date: 02/28/90
Page 3 of 55
2.0 TABLE OF CONTENTS (continued)
8*3 Calibration Frequency Procedures for X-Rav
Fluorescence 31
8.3.1 Calibration Procedures ฃsฃ Soils
Analysis using the Oxford XRF 31
8.3.2 Calibration Procedures for Dust Analysis
Using KEVEX 0700 XRF 32
9.0 ANALYTICAL PROCEDURES 34
9.1 general 34
9.2 Method of Analysis 34
9.3 sample Preparation (icp/xrf) 34
9.4 Atomic Emission Spectroscopy - TCP 35
9.4.1 Wet Digestion 35
9.4.2 Analysis 35
9.4.3 Quality Assurance/Oualitv Control
(ICP) 35
9.5 X-Rav Fluorescence 36
9.5.1 Sample Preparation for Dust and Soils. 36
9.5.2 Analysis 36
9.5.3 Quality Assurance/Oualitv Control ... 37
9.5.3 Quality Assurance/Oualitv Control ... 38
-------
Section No. 2
Revision No. 1
Date: 02/28/90
Page 4 of 55
2.0 TABLE OF CONTENTS (concluded)
t/^4.0 DATA ASSESSMENT 4 9
14.1 General 49
14.1.1 Phase 1 49
14.1.2 Phase 2 49
14.1.3 Phase 3 49
14.2 Precision and Accuracy 50
14.3 Completeness 51
15.0 CORRECTIVE ACTION 52
15.1 Immediate Corrective Action 52
15.2 Long-Term Corrective Action 53
16.0 REPORTS TO MANAGEMENT 55
-------
Section No. 2
Revision No. 1
Date: 02/28/90
Page 5 of 55
Quality Assurance Project Plan
Distribution List
A copy of this Quality Assurance Project Plan has been
provided to the following organizations and their project
representatives:
Natalie Zaremba, LFK
Ann Aschengrau, LFK
Michael Wietzman, LFK
John Christian, -LFK
Tom Spittler, U.S. EPA
Carol Wood, U.S. EPA
David Mclntyre, U.S. EPA
Jay Markarian, Roy F. Weston, Inc.
-------
Section No. 2
Revision No. l
Date: 02/28/90
Page 6 of 55
LIST OF APPENDICES
A - Sampling Guidelines
A-l Preliminary Sampling - Soil
A-2 Detailed Sampling - Soil
A-3 Post-Abatement Sampling - Soil
A-4 Dust Sampling
B - Analytical Procedures
B-l Method 3050: Acid Digestion of Sediments, Sludges, and
Soil
B-2 Method 6010: Inductively Coupled Plasma Atomic Emission
Spectroscopy
B-3 Standard Operating Procedure: Laboratory Screening
Method for Lead in House Dust Using Energy Dispersive X-
Ray Fluorescence (Kevex 0700)
B-4 Standard Operating Procedure: Laboratory Screening
Method for Lead in Soil Using Energy Dispersive X-Ray
Fluorescence (Oxford LX1000)
-------
Section No. 2
Revision No. l
Date: 02/28/90
Page 7 of 55
LIST OF FIGURES
3-1 Demonstration Project Flow Chart
4-1 Project Organization
5-1 Data Quality Objectives for Lead in Soils and Dust Analysis
6-1 Five-point Composite Sample Scheme - Soil
6-2 Detailed Site Diagram - Soil
7-1 Chain of Custody for Soil
7-2 Chain of Custody for Dust
7-3 Sample Container with sample label for Soil
7-4 Sample Container with Sample label for Dust
7-5 Sample and Data Progression through Laboratory - Dust
12-1 Quality Assurance Audit Report
15-1 Corrective Action Documentation Form
-------
Section No. 3
Revision No. 1
Date: 02/28/90
Page 8 of 55
3.0 PROJECT DESCRIPTION
3.1 Introduction
The purpose of the Quality Assurance Project Plan (QAPjP)
for the Boston Lead-in-Soil/LFK Demonstration Project is to
indicate prime responsibilities and prescribe requirements for
assuring that the project is planned and executed in a manner
consistent with defined quality assurance objectives. This QAPjP
provides guidance and specifications to assure that:
1. All field sampling, methodologies and documentation,
sample preparation, handling, and transportation are
conducted consistently according to established
procedures;
2. All laboratory determinations and analytical results
are valid through preventative maintenance, instru-
ment calibration, and analytical protocols;
3. Samples are identified and controlled through sample
tracking systems and chain-of-custody protocols;
4. Records are retained as documentary evidence of the
sample integrity, applied processes, equipment used,
and analytical results;
5. Generated data is validated and its use in calcula-
tions documented; and
6. Calculations and evaluations are accurate, approp-
riate, and consistent throughout the project. l
The requirements of this QAPjP apply to the EPA Region I (and
its subcontractor activities) and to the Trustees (and' its sub-
contractor activities) as appropriate for the Demonstration
Project.
3.2 Project gyflromrY
The following information summarizes the specific tasks
required for this Demonstration Project as well as other pertinent
information.
3.2.1 Project Background
In 1985, the Centers for Disease Control (CDC), Atlanta,
Georgia, published a report entitled, Preventing Lead Poisoning in
Young Children, which stated that "lead in soil and dust appears
-------
Section No. 3
Revision No. l
Date: 02/28/90
Page 9 of 55
to be responsible for blood lead levels in children increasing
above background levels when the concentration in the soil or dust
exceeds 500-1000 parts per million (ppm)."
Data from the City of Boston Childhood Lead Poisoning Preven-
tion Program, coupled with the CDC report, led to the following
conclusions:
a. Children playing in the area of exposed, lead-con-
taminated soil may ingest lead in the course of their
normal hand-to-mouth activities.
b. Congestion of lead"thvmk direct contact with lead-
cofitaminatS?3~soii may result? m^n increased body
burden of the contaminant.
c. Exposure of humans to lead through ingestion or
inhalation can result in toxic effects in the brain,
central and peripheral nervous system, kidney, and
hematopoietic system. Anemia is an early manifesta-
tion of lead poisoning. Peripheral neuropathy also
results from lead poisoning. Young children under
the age of six are especially prone to the most
profound and deleterious effects of lead exposure.
Chronic exposure to low levels of lead can cause per-
manent learning disabilities in children.
3.2.2 Demonstration Project
The Boston Lead Free Kids Demonstration Project shall involve
sampling approximately 150 selected children for blood lead levels
to determine base line data, sampling their yards to establish soil
lead levels, and their residences to determine dust lead levels,
removing contaminated soil and dust, and resampling the children
during the following year to observe the effects of the soil/dust
removal. This QAPjP addresses the soil and dust sampling and
analysis.
Preliminary soil sampling will be conducted during 1989, as
necessary, at the selected children's properties. Should the
preliminary sampling indicate lead concentrations equal to or
greater than 1500 ppm in two or more samples, residents of those
properties shall be contacted for enrollment in the project.
Detailed soil sampling will be conducted at those sites enrolled
in the project in order to define the nature and extend-of the lead
contamination. The surface of those properties which are included
in the project will be removed to a depth of 15cm (six inches).
Post abatement sampling and analysis will be conducted to evaluate
the effectiveness of abatement activities, and to monitor the
levels of lead in soil at the'project properties it later dates.
This is illustrated in the flow diagram presented as Figure 3-1.
-------
Section No. 3
Revision No. l
Date: 02/28/90
Page 10 of 55
Once the properties are signed on to the project, interior
dust sampling and interior dust abatement (extensive vacuuming and
cleaning) will commence. Dust samples will be analyzed at the EPA
lab and the data forwarded to LFK. Dust abatement activity details
are available in the project design document prepared by LFK.
Preliminary Soil Sampling to Determine Eligibility
Three to four composited surface samples will be collected
from properties with children chosen to be potential study partici-
pants. One composited surface sample will also be taken from any
obvious play areas. Sketches of the properties indicating key ,
landmarks and sample locations will be made. Samples will be
analyzed by XRF at the EPA Region I laboratory. The property will L(C
be eligible if two or more sample results are equal to or greater
than 1500 ppm.
Detailed Soil Sampling
After properties are selected and signed up to be in the
project, they will undergo detailed soil sampling. Soil samples
will be collected at the surface and from corings 15 centimeters
below surface by the project staff according to attached protocols,
using one or more of the defined patterns: line source, targeted
area, small area, or grid patterns. Pattern selection will be
based upon the layout of the subject property at the discretion of
the sample crew chief. Sketches indicating property details and
sample locations will be made by the samplers. The samples will
be transported to the EPA Region 1 laboratory as described in the
protocols and analyzed on XRF. Results of the detailed sampling
and analysis will be forwarded to LFK for interpretation.
Post Abatement Soil Sampling
Properties will be sampled after abatement activities. The
purpose of this is two-fold. Firstly, to document the effective-
ness of abatement activities, and establish a control point (i.e.,
the abated property). Secondly, to document lead concentrations
in the soil at later dates in order to determine if lead concentra-
tions in the soil have changed since abatement activities. Soil
sampling conducted immediately after abatement will be confined to
the abated areas, and be conducted in the same pattern as was
previously used. Periodic, post abatement sampling will be
conducted in areas which previously had the highest concentrations
of lead, play areas, or any locations otherwise specified by the
principal investigator. Soil sampling will follow the protocol
utilized during the preliminary sampling phase of the project.
Analysis will be by XRF at the EPA Region I laboratory.
-------
Section No. 3
Revision No. 1
Date: 02/28/90
Page 11 of 55
Dust Sampling
Dust sampling will be conducted in interior areas on hard
surfaces, according to the LFK protocol. Modified "Dust Busters"
will be used to collect dust from areas identified by a template
(24x24 or 25x25 inches), the sample will be placed in a paper,
legal sized envelope, sealed, and delivered to the EPA lab for XRF
analysis.
3.3 project Qblectiygg
The main objective of the Boston Lead-in-Soil Demonstration
Project is: to determine the effects of lead contaminated soil
removal from inner city residential areas on blood lead levels of
children living on the contaminated properties. A secondary
objective is to measure the effects, if any, on the blood lead
levels of children after dust abatement on those same properties.
3.3.1 Maior Task Summary
The Boston Lead-in-Soil Demonstration Project will include0^ ^
field and laboratory nrHvltlfts hv.JBPA Region I and Trustees .'""a ^i"4*
summary of tasks (covered by this QAPj>) is presented below. j. +>>-r
of
preparation of a Health and Safety Plan that will
identify potential hazards associated with the planned
field activities, establish the level of protection,
and provide information and procedures needed to
mitigate these hazards for on-site workers;
preparation of Sampling and Analysis procedures to be
utilized by personnel in obtaining soil and dust
samples for analysis and for personnel conducting soil
and dust analyses in the laboratory. For soil, this
will include procedures for conducting site surveys
for the purpose of preparing site grids, and obtaining
preliminary, detailed and post-abatement soil samples,
and procedures for laboratory analysis. For dust,
this will include procedures for collecting and
analyzing;
conduct preliminary soil sampling at selected
properties according to established protocols;
conduct detailed soil sampling at selected properties
with full site documentation as specified in the
protocols;
conduct dust sampling at selected properties as specified
in the protocols;
-------
Figure 3-1
Demonstration Project Flow Chart
Preliminory surface soil sampling
I
Analyze soil with XRF
+
Data to LFK
Property not eligible
for project.
I
[Pb]> 1500ppm
in two or more samples and
children reside at residence
YES
Property maybe enrolled in
project. After enrollment,
analyze children's blood
for lead level, perform detailed
soil sampling, sample
interior dust.
Excavate soil to a depth
of 6 inches 15 cm (6 inches)
J
Analyze soil ond dust with XRF
and remove soil. Fill excavated
area with clean fill; sample
abated areas utilizing preliminary
surface soil sampling phase
protocol. Abate interior dust.
I
Data to LFK
-------
FIGURE 4-1
PROJECT ORGANIZATION
U.S. EPA
Region I
Project Manager
David Mclntyre
T rustees
Principal Investigator
Wiezman
Quality Assurance Coordinator yAQ>
*Carol Wood *
Lead Free Kids
Project (DHH
Dr. Thomas Spittler
Technical Project Director
Jay Markarian
Task Leader
Laboratory
Analysis
Field Operation
(Soil)
Bart Hoskins
Dust Sampling
Dr. Ann
Aschengrau
-------
Section No. 3
Revision No. 1
Date: 02/28/90
Page 12 of 55
analysis of all soil and dust samples using the tech-
niques of energy dispersive X-ray fluorescence, and
of a fraction of the samples using Inductively Coupled
Plasma Emission Spectrophotometry; and
post abatement soil sampling of properties according
to protocols.
-------
Section No. 3
Revision No. 1
Date: 02/28/90
Page 13 of 55
4.0 PROJECT ORGANIZATION AND RESPONSIBILITIES
4.1 Organization
This project is being financed through the EPA under SARA as
part of a three project effort to define the lead-in-soil/childhood
lead poisoning relationship. Each of the three projects are
responsible for their own operation and findings. EPA Region I and
the Trustees of Health and Hospitals of the City of Boston, Inc.
have entered into a cooperative agreement to conduct this Demon-
stration Project. The Trustees are responsible for the operation
and findings of the Boston project, including all scientific and
logistical facets. The EPA's responsibility, as a partner in the
cooperative agreement with the Trustees, is to ensure that (1) the
money allocated to the Boston project is spent appropriately
according to federal regulations, (2) that involved federal
agencies are coordinated, and (3) that EPA's input as defined in
the Special Conditions is provided. Zt is Region I's position that
it will closely monitor the activities of the Trustees and work
with them on the project, but that the Trustees are running the
project. The responsibilities and project organization are
discussed below.
4.2 Responsibility for Quality Assurance
EPA support to the Project
The Region I Environmental Services Division (ESD)
Laboratory and their contractors will provide
personnel and facilities for energy dispersive x-ray
fluorescence spectrometry (XRF), and inductively
coupled plasma spectroscopy (ICP) analyses for lead
in soil.
The ESD and their contractors will also provide
technical guidance and services related to the
collection and chemical analysis of soil samples
obtained during field activities and will serve as
sample custodian during sample analysis. In addition,
they will provide support in developing the 'QAPjP,
perform analytical data review and report generation.
Trustees Support to the Project
The Trustees will provide equipment, laboratory sup-
plies and personnel necessary for on-site sampling
soil removal activities, blood sampling, dust
sampling, and all pther project activity not
specifically provided by EPA.
-------
Section No. 4
Revision No. 1
Date: 02/28/90
Page 14 of 55
Figure 4-1 shows the project organization and its principal
lines of communications. The responsibilities of the EPA and
Trustees' project staff positions and support organizations are
summarized below:
For the EPA:
The Project Manager is responsible for:
1. Maintaining coordination between the EPA and
the Trustees.
Uซ ' Kbs. oป_ cpuA^ouL ซs- ~**ป*.
2. Monitoring all projoct aotivitics. v
3. Providing overall direction for preparation of
work plans, sampling plans, and analytical
procedures relative to soil and dust.
4. Administering all contracts with EPA contrac-
tors.
The Project Manager is David Mclntvre.
The Technical Project Director is responsible for:
1. Approving, maintaining, and implementing this QAPjP for
EPA-conducted activities, i.e., sample analysis.
2. Indicating the types of QA records to be maintained for
the analytical portion of the project.
3. Approving analytical procedures and operating
systems.
The Technical Project Director is Dr. Thomas
SPltUer or designee.
The Quality Assurance Coordinator (OAC1 will be
responsible for:
1. Evaluating and approving this QAPjP.
2. Scheduling and conducting systems and performance
audits on-site and in the Laboratory.
3. Providing QA reports to the Project Manager on the
results of audits * and the need for preventative or
corrective actions.
-------
Section No. 4
Revision No. l
Date: 02/28/90
Page 15 of 55
4. Developing and initiating preventative and
corrective actions as needed in conjunction
with the Project Manager and the Technical
Project Director.
The Quality Assurance Coordinator (QAC) is Carol Wood
or designee.
For the Trustees:
The Principal Investigator is responsible for staffing
and conducting the project, except for activities
having to do with sample analysis which will be
provided by EPA. As part of his QA responsibilities
he will:
1. Approve, maintain, and implement this QAPjP as
it relates to LFK activities, i.e., sample
collection.
2. Indicate the types of QA records to be retained for the
LFK aspects of the project, and retain such records.
3. Provide for QA audits by EPA.
4. Approve task plans and operating systems.
The Principal Investigator is Dr. Michael Weitzman.
LFK Task Leaders are responsible for specific
engineering and scientific operations. As part of
this responsibility they will:
1. Initiate, develop and check subtask plans
including initiating, monitoring, and accepting
support services and products.
2. Identify safety hazards and ensure that the as-
sociated risks are at acceptable levels.
3. Supervise and participate in operations,
analyses, data collection, and data reduction.
4. Maintain samples and their identification.
-------
Section No. 4
Revision No. l
Date: 02/28/90
Page 16 of 55
5. Develop site sketches, identify sample loca-
tions, buildings, and appropriate structures,
identify notable site specific conditions and
observations to include photographs, identify
clean soil zones and those to be abated.
6. Generate required QA records.
7. Implement corrective actions, when required.
LFK Task Leaders will be named by the Principal
Investigator.
-------
TABLE 5-1
Data Quality Objectives for Lead in Soil Analyses
Method
Detection . . Complete- Method
Parameter Method Instrument Reference Precision Accuracy* ness Comparability Limit0
Lead (Pb)
-Sol I XRF1
Lead (Pb)
-Oust XRF
Oxford LX1000
Kevex 0700
ICP Parkin Elmer
6010
120
~20
~10
~20
~10
~10
90
90
90
mg/kg dry wt
mo/kg dry wt
ซ0/kg dry ut
200 (mg/kg)
100 (mg/kg)
42 (ug/L)
a. Energy dispersive x-ray fluorescence spectrometry
b. Copies of methods are attached as appendices
c. SV 846, 3rd Ed.
d. Percent relative standard deviation from mean or true value
e. Relative percent difference
f. Percent
g. Nominal detection limits, soil HDL estimated from semple size end concentretion factors (units) mg/kg
h. Air dry overnight in laboratory hood at anfcient temperatures
I. Inductively Coupled Plasma Emission Spectrophotometry
TB0 ฆ To be determined
-------
Section No. 5
Revision No. 1
Date: 02/28/90
Page 17 of 55
5.0 QUALITY ASSURANCE OBJECTIVES
5.1 general
The quality of measurements made during this study will be
determined by the following characteristics: accuracy; precision;
representativeness; completeness; and comparability. Specific
objectives for each characteristic are established to develop
sampling protocols, to identify applicable documentation, and to
perform sample handling and measurement procedures. These objec-
tives are established based on site conditions, objectives of the
project, and knowledge of available measurement systems. The
subsequent use of these measurements in calculations and evalua-
tions is also subjected to aspects of this QAPjP as described in
the following sections. The Quality Assurance Objectives for
chemical analyses conducted in conjunction with the Boston Lead
Free Kids project are presented in Table 5-1.
The Trustees (Lead Free Kids) will collect all samples and
provide site-specific field documentation and transport samples to
the ESD Laboratory maintaining chain of custody from collection to
delivery at the ESD Laboratory. Sample collection and field
handling will be in accordance with the sampling protocols estab-
lished in this QAPjP. Soil and dust samples will be analyzed at
the EPA Region I ESD Laboratory in Lexington, Massachusetts.
Analytical laboratory QA/QC discussion is presented in Section 9.0.
5.2 Representativeness
Sampling procedures (Section 6.0) will be used to assure that
samples collected are representative of the media. Sample handling
protocols (e.g., storage and transportation) protect the repre-
sentativeness of the collected sample. Proper documentation will
ensure that protocols; have been followed and that sample iden-
tification and integrity are assured.
Sample preparation procedures (Section 9.3.1 - Soil and Dust)
will be used to assure that the samples analyzed are representative
of the fraction which poses the greatest risk to the public.
5.3 Precision and Accuracy
Precision, the ability to replicate a value, and accuracy, the
ability to reproduce a true value, are addressed for all data
generated by EPA. Data quality objectives for precision and
accuracy are established for each major parameter to be measured
at the site. These objectives are based on prior knowledge of the
capabilities of the measurement system to be employedtj and are in
turn selected in accordance with the requirements of the project.
The precision and accuracy requirements vary, dependent on their
-------
Section No. 5
Revision No. 1
Date: 02/28/90
Page 18 of 55
intended use. For example, a screening tool to identify the
general extent of chemical distribution will not require the same
precision and accuracy required to define the exact nature and
amount of chemicals present at specific locations.
5.4 CPfflPlgtenggg
The characteristic of completeness is a measure of the amount
of valid data obtained compared to the amount that was specified
to be obtained under normal conditions. The amount of valid data
specified is established based on the measurements required to
accomplish project objectives. The extent of completeness must be
reviewed on a relative basis for sample collection activities.
Completeness of data handling systems is described in Sections
10.0, 11.0, 12.0, and 14.0. Examples of completeness objectives
for specific measurement systems are also provided in Table 5-1.
5*5 Comparability
The characteristic of comparability reflects both internal
consistency of measurements made at the site and expression of
results in units and methodologies consistent with other organ-
izations reporting similar data. Each value reported for a given
measurement should be similar to other values within the same data
set and within other related data sets. Comparability of data and
measuring procedures must also be addressed. This characteristic
implies operating within the calibrated range of an instrument and
utilizing analytical methodologies which produce comparable
results, (e.g., data obtained for lead (ICP) is not directly
comparable to data obtained for lead (XRF). However, it is a
Quality Assurance Objective to define the limits of comparability
by submitting samples analyzed by XRF for ICP analysis and
comparing their results).
5.6 Quality Assurance Pfrlegtiveg
The quality assurance objectives for the Demonstration Project
are:
to produce documented, traceable, and consistent data;
to collect and analyze sufficient trip blank and field
duplicate samples to allow an assessment of sample
collection protocol precision;
-------
Section No. 5
Revision No. 1
Date: 02/28/90
Page 19 of 55
to provide sufficient laboratory duplicates, blanks,
replicates, and reference standards to allow an
assessment of analytical precision and accuracy.
Sufficiency of analytical QC procedures is specified
by the referenced methods (see Section 9.0);
to produce documented, consistent, and technically
defensible data reports; and
to conduct site sampling and site-specific documenta-
tion according to established protocols.
to define comparability of XRF and ICP obtained analytical
data so as to allow XRF data to be compared to other
organizations conducting similar studies.
-------
Section No. 6
Revision No. 1
Date: 02/28/90
Page 20 of 55
6.0 SAMPLING PROCEDURES
6.1 general
ฃฃjJL
Three soil sampling events will be conducted during the
Demonstration Project. They are
Preliminary Sampling;
Detailed Sampling; and
Post Abatement Sampling.
Preliminary sampling is the initial phase of sampling, and will
qualify a property for further participation in the project.
Detailed sampling will indicate the nature and extent of lead
contamination. Post abatement sampling will indicate effectiveness
of abatement, and provide additional data to be utilized in future
lead abatement efforts.
All sampling locations will consist of a five point composite.
The center point of the composite will fall upon the pre-determined
sample station. Each sample location will consist of soil col-
lected from the center point and four corners of the square. This
is illustrated in Figure 6-1.
Mo preservatives will be required for soil samples. Sample
containers will consist of plastic "Zip-Lock" bags. Sample
collection schemes and field documentation will differ, based upon
which phase of sampling (preliminary, detailed, or post abatement)
is being conducted.
Dust
Dust sampling events will be conducted in conjunction with
soil sampling and other scheduled activities on-going throughout
the Demonstration Project.
For this study, the household dust samples are defined as the
samples that are most likely to come into contact with a child's
hands during indoor activity. This would include dust on upfacing
surfaces accessible to the child such as bare floors, carpets,
window sills, and wells, furniture, as well as dust on toys and
other objects likely to be handled by children.
Dust sampling has two components that are important to
interpreting lead exposure: the concentration of lead in the dust
and the amount of dust or loading on the surface. The concentra-
tion of lead in dust appears to be closely related to the amount
of lead on children's hands whereas the amount of dust on surface
is an indicator of the importance of this route of human exposure.
-------
Section No. 6
Revision No. 1
Date: 02/28/90
Page 21 of 55
All sampling locations within a home will consist of a
measured suarea from which dust will be collected using a
modified ha(nd--hfld vacuum (Oust Buster).
No
transfer:
envelope
6.2 Equipment List
Several items are required to collect soil and dust samples
and document sampling events, site description, etc. Because many
properties shall be investigated and several samples collected from
each property, an adequate supply of equipment should be available
at all times.
A list of equipment is provided in each of the four sampling
protocols found in Appendix A. Additional equipment not on the
list may be required on a site specific, as-needed basis.
preservation will be required for dust samples. Sample
red from this dust: blaster will be contained in a paper
6.3 SflBBlS-EallSSfciSD
Soil
Each sampling point will consist of an approximate two-foot
square area. The center of the square will fall upon the decided
sampling point, and a composite sample will be collected, composed
of soil taken from the four corners and center of the square.
Preliminary Phase and Post Abatement Phase samples will be col-
lected from 0 to 2 cm below the surface. Detailed Phase samples
will be collected from 0 to 2 cm below the surface, and an addi-
tional core sample will be collected from 13 to 15 cm below the
surface, from the same point. Figure 6-1 illustrates A - Prelimi-
nary and Post Abatement, and B - Detailed Soil Sample Collection.
In depth discussion of sampling techniques, to be utilized during
all phases of sampling, are provided in Appendices A-l, A-2, A-3
and A-4.
Each sampling point will be collected from/a known fequare area
using a modified Dust Buster. Floor and hallway areas will be
sampled first using a 25" x 25" template. Tnese^ samples will be
collected from the center of the floor. sWmpAea/will then be
collected from window wells and other areas with^xhe square area
sampled documented. A minimum-of 5 milligrams of sample is
-------
FIGURE 6-1
5 Point Composite Sample Scheme
-1.5'
AO O O
'1.5 Oฎ O
tooo
Preliminary and Post
Abatement Phase
Surface Sample
Composite Location Schematic
Legend
Sampling Station
O Sample Composite Location
(3 Field Duplicate Sample
Composite Locations
(At selected sampling stations duplicates
wB be colected at mid-points between
sample composite locations.)
B
Detailed Phase
Surface and Subsurface Sample
Composite Location Schematic
surface
$
2 cm
Sample collected from
surface to 2cm
below surface.
surface
o
2 cm
13 cm
0
15 cm
First sample collected from surface
to 2 cm below surface.
Second sample collected from 13 cm
to 15 cm below surface in same boring location.
-------
FIGURE 6-2
Detailed Site Diagram
Address: Date:
Cement Sidewalk
Street
-------
Section No. 6
Revision No. l
Date: 02/28/90
Page 22 of 55
required for analysis. This may necessitate sampling and recording
a larger floor or window well surface area. In-depth discussion
of sampling procedures to be utilized during dust sampling are
provided in Appendices A-4.
6.4 Sample Handling and Storage
The soil sample containers (polyethylene "Zip-Lock'1 bags, or
equivalent) should be sealed to prevent loss or cross-contamination
of the sample. Dust sample containers (paper envelopes) should be
folder over 1/2 inch from the top and then taped down to prevent
loss or cross contamination of the sample. No special considera-
tions will be given to sample packaging or shipping papers as the
samples will be delivered to Environmental Services Division (ESD)
Laboratory by a member of the LFK staff. Pending delivery to EPA, ^ n
samples should be stored in a cool, dry, and secure location with '
limited, controlled access.
6.5 Record keeping
Sampling records, maintained by LFK, for each property will
consist of a site sketch for soils and floor plan for dust with
location descriptions and chain of custody record for each sample
collected. Samples shall be assigned LFK field identification
numbers to include premises identification number with a sequential
alpha numeric designation, and located on the site sketch and
summarized in a Blood Field Sampling log book. Chain of custody
shall be established as described in Section 7.0.
6.6 Preliminary Soil Sampling
6.6.1 Site Description
During this phase of sampling, LFK will generate a detailed
drawing that indicates the boundary of the lot, the position of the
main building and any other buildings such as storage sheds or
garages, the position of the side walks, driveways, and other paved
areas, the position of the play areas, if obvious, and the areas
with exposed soils, as illustrated in Figure 6-2. The property
should be divided into separate sub-areas if necessary for clarity
or detail and be identified with alpha designation. Sub-areas may
include isolated areas of the site such as front, rea*r, and side
yards. Sample locations will be identified on this drawing,
indicating approximate distances from buildings and other
landmarks. The resulting sketch will be placed in the property
file at the LFK office.
-------
Section No. 6
Revision No. l
Date: 02/28/90
Page 23 of 55
In addition to the sketch, the following information will be
included on the page:
Address0j
Date and name of artist;.
Apparent use of yard if any (toys, sandbox,
children present);
Debris, standing water, vegetation, cover and bare
spots, animals on property; and
Any notable unusual feature)(
.6.2 Sampling Schemes
The sampling scheme utilized for each property during the
Preliminary Phase will be the same for each one, and will involve
sampling in the immediate area of the house and in obvious play
areas. This protocol is described in Attachment A-l.
6.7 Detailed Soil Sampling
6.7.1 Site Description
During the Detailed Soil Sampling Phase for each location, the
project log should briefly describe the sampling locations and
sampling schemes used, and include the following information if not
provided by preliminary investigation:
address ;Q
date and name of Artist ;ฃ)
type of building construction;
condition of main building;
condition of lot (debris, standing water, vegetation
cover);
nature of adjacent property;
presence and type of fence;
animals on property;
-------
Section No. 6
Revision No. 1
Date: 02/28/90
Page 24 of 55
apparent use of yard (toys, sandbox, children
present); and
underground utilities.
6.7.2 Sampling Schemes
The sample scheme selected must adequately characterize the
potential exposure of children to lead in the soil. The scheme
utilized should reflect the size and proportions of the area to be
sampled (see the protocols for details). Several options are
offered for the best judgement of the investigator, and include:
Line Source (LS) Pattern;
Targeted Method (TM);
Small Area (SA) Pattern; and
Grid (6) Pattern.
Sampling Schemes are detailed in Appendix A-2.
6.8 Post-Abatement Sampling
Post-abatement sampling will be conducted in order to deter-
mine the effectiveness of abatement activities, and to monitor lead
levels in the soil. Sampling shall be conducted in the same
sampling locations as were used for the Preliminary Soil Phase
Sampling (Section 6.6). The principal investigator will designate
number of samples and specific location of these points. Detailed
description of the sampling procedure is found in Appendix A-3.
-------
Section No. 7
Revision No. 1
Date: 02/28/90
Page 25 of 55
7.0 CHAIN-OF-CUSTODY
7.1 General
EPA has established a program of sample chain-of-custody that
is followed during sample handling activities in both field and
laboratory operations.
Chain-of-custody procedures document the sample history and
constitute a crucial part of sampling and analysis programs.
Chain-of-custody documentation verifies the identification and
history of a sample from collection through the time of analysis.
The objective of sample custody identification and control is
to ensure that:
all samples scheduled for collection, as appropriate
for the data required, are uniquely identified;
the correct samples are analyzed and are traceable to
specific analysis records;
important sample characteristics are preserved;
samples are protected from loss or damage;
any alteration of samples (e.g., filtration, preservation)
is documented;
a record of sample integrity is established for legal
and technical purposes; and
The chain-of-custody record is used to:
document sample handling procedures, including sample
location, and sample number; and
describe the chain-of-custody process.
The chain-of-custody description section requires:
the sample number;
the name(s) of the sampler(s) and the person shipping
the samples;
the date and time that the samples were delivered
for shipping; and
the names of those responsible for receiving the samples
at the laboratory.
-------
Section No. 7
Revision No. l
Date: 02/28/90
Page 26 of 55
Samples of a chain-of-custody record for soils and dust samples
are shown in Figure 7-1 and Figure 7-2 respectively.
As samples are collected, entries are made on the chain-of-
custody forms. Data to be noted include:
Date/Time;
Samplers;
Sample phase description ie. for soils: preliminary,
Special instructions/notes and
Sample Identification information to includ D and
Soil and dust sample containers, will be labelled by LFK
samplers with an indelible marker with station number/sample number
and other appropriate information necessary to match the sample
container to the chain-of-custody Record, (see Figure 7-3 and 7-
4).
The station number/sample number shall be such as to allow
tracking of the sample from its source of collection through
analyses and be consistent with other site sample location iden-
tification systems.
When samples are received at the laboratory, the Laboratory
Task Leader or Analyst will verify each and every sample against
the chain-of-custody forms, note any discrepancies or losses of
samples, and then sign for receipt of the samples. The laboratory
task leader may also contact field personnel to resolve
deficiencies, irregularities, discrepancies, etc., prior to
accepting the samples. Samples will remain under the control of
the laboratory task leader until samples are ultimately disposed
of .
A sample is considered to be in custody if it:
is in the physical possession of the responsible party;
is in view of the responsible party;
is secured by the responsible party to prevent tamper-
is secured by the responsible party in a restricted area.
Field ID numbers
ing; or
-------
Figure 71
- Soil
Lead-in-Soil Demonstrtation Project
Chain ofCustody Record
Prem. ID
Samp ID
Address
Dist.
Date
Collected
Collected
By
Relinquished By
Received By
Date
Tine
Comments/Instructions
-------
Figure 7-2
Dust -
Lead-in-Soil Demonstrtation Project
Chain-of-Custody Record
DUST SAMPLES DATA FORM, STUDY PHASE
PREMID: _ Address: _________________________________ Apt No.
Date sample taken: Taken by:
LFK children in this apartment: NAME, LFK NUMBER, and BEDROOM CODE (if needed)
ROOM
CODE
PLACE
CODE
NOTE
AREA
OR
CIRCLE SAMPLE SAMPLE
TEMPLATE TYPE WEIGHT PPM
LABID
1. _
2. __
3. _
4.
5. __
6. ___
7. ___
8. __
9. __
10. __
ROOM CODES:
X - Kitchen
L - Living Room
D - Dining Room
Bl - Bedroom 1
B2 - Bedroom 2 (etc.)
01 - other
x
x
X
X
X
X
X
X
X
X
25X25 24X24
25x25 24X24
25X25 24X24
25x25 24X24
25X25 24X24
25x25 24X24
25X25 24X24
25X25 24X24
25x25 24X24
25X25 24X24
PLACE CODES:
F - Floor
W - Window
O - Other
Rtlinquiihcd by: r$>pnซnป*/
0ซtt
'Timt
Rwtivซd by: (ittmmnl
Rtlinqunnrt by: (Sipfivnl
Dcti
Tim#
Rmivrt by:
fUfineutllMd by: ISttntrunt
Oft* /Tim*
R*e*ivซd by: fx*wnav/
Rdinouilhtd by:
0ปtป
Tim*
Rtaimtf by: thr-mnl
fuiinqutthvd by: r**wn>r><
Dim / Tim*
R*o*iซ*d lor Uborttery by:
Dim /Tim*
Rwnarki
-------
FIGURE 7-3
Soil
Sample Container with
Sample Label
Plastic Ziplock Bag or Equivalent - Soil
-------
FIGURE 7-4
Dust
Sample Container with
Sample Label
Paper Envelope - Dust
To Scale
-------
Section No. 7
Revision No. 1
Date: 02/28/90
Page 27 of 55
7.1 Sample Receipt
All soil samples will be delivered to the Environmental
Services Division Laboratory in Lexington, MA. by a member of the
LFK field sampling team. Upon receipt chain-of-custody and sample
integrity are to be checked and any problems recorded. Samples
will then be logged in by EPA personnel or their contractors who
will accept and sign the chain-of-custody record. The EPA project
Manager will be informed of any deficiencies and will advise the
laboratory on the desired disposition of the samples. Chain-of-
custody forms and deficiency notices are maintained in the Labora-
tory's Project file.
Each sample that is received by the laboratory is assigned a
unique sequential Laboratory Identification number which will
identify the sample in the laboratory's internal tracking system.
The flow of samples and analytical data through the laboratory is
shown in Figure 7-5 for soil and dust.
7.2 Sample Storage
All soil samples will be stored in a secure sample storage
bank at the Region I ESD Laboratory facility.
Original sample containers (plastic baggies for soil and
envelopes for dust), and laboratory analysis containers will be
stored until each data report is finalized. Soil containers
(plastic baggies) will be disposed of and envelopes returned to LFK
staff. Analysis containers will be kept for duration of the
project, a minimum of 3 years.
-------
FIGURE 7-5
Sample and Data Progression Through Laboratory
Somple Receipt
I
Samples logged into loborotory by
assigning a lab ID number to sample container label
I
Temporary sample storage
I
Sample Lab ID and Field ID entry into
instrument analysis work sheets
I
Samples prepared for XRF onalysis
J
Samples loaded into XRF (KEVEX or Oxford) with sample
ID entered into XRF computer software
I
XRF Analysis Run-
Spectra stored on diskette
(or hardcopy) with
results transcribed to into
Instrument Analysis
work sheets
Samples to storage
cups, bags ond envelopes
Selected soil samples to OC
Analysis via ICP
*
Preliminary Report
T.
Data Review^ป-One set of all row
J* dolo lo project file
Final Report
1-copy to U.S. EPA
1-copy to LFK
1-copy lo project file
Plastic bags disposed of
after final report.Envelopes
returned to LFK
-------
Section No. 8
Revision No. l
Date: 02/28/90
Page 28 of 55
8.0 CALIBRATION PROCEDURES AND FREQUENCY
8.1 overview
Before any instrument can be used as a measurement device, the
instrumental response to known reference materials must be deter-
mined. The manner in which the various instruments are calibrated
depends upon the particular instrumentation and the intended use
of the instrument. All sample measurements will be made within the
calibrated range of the instrument. Preparation of all reference
materials used for calibration will be documented in a standards
preparation notebook. Good laboratory practices require general
calibration procedures that should include:
Preparation of. standards that represent the range of
concentrations of concern in the samples;
For soil establishment of a concentration/response
factor with a minimum of three points, using the stan-
dards prepared above and for dust, establishment of
concentration/response spectra with a minimum of two
points in both high and low sample weight ranges also
using the standards prepared above;
A set of secondary standards that can ultimately be
traced to National Bureau of Standards (NBS) primary
standards.
Inductively coupled Plasma Spectrophotometry and X-ray Fluores-
cence are the two methods of analysis for the Demonstration
Project. A separate discussion of calibration procedures and
frequency for each of these measurement systems including each of
the two X-Ray units used presented below.
8.2 Calibration and Frequency Procedures for Inductively Coupled
Plasma Spectrophotometer
8.2.1 Calibration Procadn-res
The methods employed will be adapted from established EPA
Methods as outlined in "Test Methods for Evaluating Solid Waste",
SW 846, 3rd Ed., U.S. EPA Office of Solid Waste and Emergency
Response, Nov. 1986. The quality assurance protocols are based
upon the guidelines in "Handbook for Analytical Quality Control in
Hater and Wastewater Laboratories", EPA 600/4-79-0019, March 1979;
"Methods for the Chemical Analysis of water and Wastewater", EPA
600/4-79-020, March 1983; and "Test Methods for Evaluating Solid
Waste", EPA SW846, Nov. 1986.
-------
Section No. 8
Revision No. 1
Date: 02/28/90
Page 29 of 55
Inductively Coupled Plasma spectrophotometer (ZCP) will be
calibrated each day before samples are analyzed. Calibration
standards will be prepared from certified reference materials.
Working calibration standards should exceed 6 month past date of
preparation or exceed expiration date of the reference material.
The working standards will include a blank and a minimum of three
(3) concentrations to cover the anticipated range of measurement.
Duplicate injections will be made for each concentration. At
least one of the calibration standards will be at the desired
instrument detection limit. The correlation coefficient of the
plot of known versus found concentrations will be at least 0.990
in order to consider the responses linear over the range to be
tested. If a correlation coefficient of 0.990 cannot be achieved,
the instrument will.be recalibrated prior to analysis of samples.
If a secondary wavelength is being used to detect lead a
calibration must also conform with calibration procedures below.
Calibration data, to include the correlation coefficient, will
be entered into laboratory notebooks to maintain a permanent record
of instrument calibrations.
The following standards should be utilised for ICP analyses:
initial Calibration - ICP. New standards are prepared
for each calibration sequence. Initial Calibration is
performed using a blank sample and at least three
standards. A regression analysis is used to construct
the calibration curve. Any regression with a coeffi-
cient of correlation below 0.990 is considered unaccep-
table and a re-calibration is required. Instrument
calibrations are from microprocessor outputs, with
chart-recorder graphs as supplemental documentation.
Continuing calibration. one of the calibration
standards preferably at mid-range is analyzed every 10
samples to verify instrument stability. Results for the
continuing calibration analysis must fall within the
control limit of +10 percent of the established mean
value or re-calibration is required. Verify calibration
every 10 samples and at the end of the analytical run,
using a calibration and a single point check standard.
Reference standard. Independent reference* standards
traceable to NBS standards are analyzed to verify instrument
performance. Any reference standard value outside the 95
percent confidence interval is considered suspect and re-
calibration is requiredj. This standard must be from a
separate source than that of the certified reference
material.
-------
Section No. 8
Revision No. l
Date: 02/28/90
Page 30 of 55
Interference Check Solution. This solution should be
analyzed at the beginning of each operating shift to verify
the ability to detect lead at the specified wavelength
without spectral interferences. The lead result should be
within + 20 percent of true value.
8.3 calibration Frequency Procedures for X-Rav Fluorescence
8.3.1 Calibration Procedures for Soils Analysis using the
QXfPrti XRF
Tlie methods employed are based on accepted analytical proce-
dures utilized in XRF analyses. The quality assurance protocols
are based upon the guidelines outlined in "Test Methods for
Evaluating Solid Wastes", EPA SW846, Nov. 1986.
The Oxford energy dispersive x-ray fluorescence spectrometer
(XRF) will be calibrated prior to, during, and at the end of each
day of use. A series of study control standards will be prepared
from appropriate reference materials. Study control standards
should include a blank and a minimum of four (4) concentrations
spanning the anticipated range of measurement. Replicate analyses
of study control standards will be performed for each concentration
through the analysis day. At least one of the working study
control standards will be at or below 1/2 the lowest action level.
Calibration data to include calculation of the daily response
factor will be entered into the laboratory analysis work sheets and
placed in project files to maintain a permanent record of
instrument calibration.
The following calibration standards should be utilized for
Oxford XRF analyses:
* Study Control Standards fSCS). Daily calibration is
performed using a blank sample and six SCS's. These
will include the following concentrations of SCS's if
available:
0 ppm
250 ppm
400 ppm
950 ppm
1200 ppm
2400 ppm
4400 ppm
One of the SCS's is analyzed every 20 samples on a rotating
basis to verify instrument stability. Results for this check
sample must fall within the control limit of +/-20 percent of the
day's established mean value or re-calibration is required.
-------
Section No. 8
Revision Mo. 1
Date: 02/28/90
Page 31 of 55
Reference standard. Independent reference standards
traceable to NBS standard^ are analyzed to verify
instrument performance, fiaj available. Any reference
outside of the 90 percent confidence interval is
considered suspect and re-calibration is required.
8.3.2 - Calibration Procedures for Dust Analysis Using KEVEX
P7QP XRF.
The methods employed are based on analytical procedures
developed by Dr. Thomas Spittler USEPA tor the KEVEX 0700 XRF
analysis of dust. The quality assurance protocols are based upon
the guidelines outlined in "test method for evaluating solid
wastes", EPA SW846, Nov. 1986.
The KEVEX energy dispersive X-Ray Fluorescence Spectrometer
(XRF) will be calibrated daily using a blank and a series of the
four study control standards to generate calibration spectra for
comparison to unknowns.
The first analysis run of the day will contain all four study
control standards. In each of the following runs one of the 16
available analysis positions (in KEVEX Carousel) will contain one
of these standards (* All standards are run manually).
Replicate analysis will be performed for each SCS through the
analysis day. calibration data to include results of the SCS
analysis will be entered onto the laboratory analysis worksheets
and the instrument software and placed in project files to maintain
permanent record of instrument calibration.
Following calibration standards should be utilized for XRF
analysis:
Study Control Standards (SCS): Calibration is performed
using a blank sample and the four SCS standards. These
will include the following concentrations of standards and
weight ranges.
Samples to be
Concentration Weight Name: used on
2500 ppm 10 mg DustM-10-sequential # 0 - . 024g
2500 ppm 50 mg DustM-50-sequential # > 0.25g
25,000 ppm 10 mg DustH-10-sequential # 0 - .024g
25,000 ppm 50 mg DustH-50-sequential # > 0.25g
-------
Section No. 8
Revision No. 1
Date: 02/28/90
Page 32 of 55
Reference standard: Independent reference standards
traceable to NBS standards are analyzed to verify instrument
performance if available. Any reference outside of the 90
percent confidence interval is considered suspect and re-
calibration is required.
SCS's have been prepared according to sample preparation
procedures however, analysis is done manually as opposed to
automatic analysis which is used for unknown samples. Daily
calibration for each operating shift will contain all four
standards and a blank. The set of SCS's will be run on the first
carousel load of the day and will be used to generate comparison
spectra used to calculate results of unknowns. The remaining 11
positions available in the carousel will be loaded will samples for
analysis. In each of the following carousel runs, one of the 16
positions will contain one of these standards on a rotating basis.
Results for this check control samples must fall within the control
limit of + 20 percent of the days established mean value or
recalibration is required.
-------
Section No. 9
Revision No. 1
Date: 02/28/90
Page 33 of 55
9.0 ANALYTICAL PROCEDURES
9.1 general
Analytical methods are routinely conducted as outlined in
published sources (EPA, Standard Methods, ASTM, AOAC, etc.).
Modifications to these methods may be necessary in order to provide
accurate analyses of particularly complex matrices. When modifica-
tions to standard analytical methods are performed, the specific
alternatives as well as the reason for the change will be reported
with the results of analyses.
Laboratory reagents will be of a quality to minimize or
eliminate background concentrations of the analyte to be measured.
Reagents must also not contain other contaminants that will
interfere with the analyte of concern.
9.2 Method Qt Analysis
The methods of analysis to be used in this project are
Inductively Coupled Plasma (ICP) spectrometry and X-Ray
Fluorescence (XRF). The XRF method is the suggested method for
routine analyses. The ICP method should be used to verify lead
concentrations in soil and dust and for other Quality
Control/Quality Assurance determinations. All soil and dust
samples for the LFK Demonstration Project will be prepared and
analyzed according to the following procedures. A detailed
description of sample analysis is found in Appendix B.
9.3 sample Preparation (icp/xrf)
A representative "urban soil sample" or "urban household dust
sample" is defined as the sample fraction which passes through a
250 micron (#60 mesh) sieve.
Sample preparation requires that the samples be allowed to air
dry overnight. Particle separation involves passing soil or dust
through a 250 micron sieve. Light grinding of soils may be
required to bring soils to disaggregation prior to the 250 micron
sieving. This is necessary to provide low/appropriate variance in
XRF analysis. For soil, aliquots of fines are then collected for
both XRF and/or ICP analysis. Dust sample preparation will not
require grinding prior to 250 micron sieving. However, the sample
must be completely sieved and weighed. A minimum of 5 milligrams
of dust is required for XRF analysis and no aliquots for ICP
analysis will be prepared.
-------
Section No. 9
Revision No. 1
Date: 02/28/90
Page 34 of 55
During the processing of the sample, it should be remembered
that small soil and dust particles may individually be as high as
50,000 ug Pb/g, and paint fragments as high as 300,000 ug/g. Care
should be taken to clean equipment (spatula, sieves, powder
tunnels) between samples. The sieves and powder tunnels may be
cleaned by tapping on a hard surface and blown free using com-
pressed air to remove residual particles. Wet washing is not
recommended as this will interfere with the sieving process.
Detailed procedures for soil and dust simple preparation are found
in Appendix B.
9.4 Atomic Emission Spectroscopy - TCP
9.4.1 wet Piqestipn
The procedure used for digesting (solubilizing) the lead in
soil is critical to the interpretation of the results of the Lead
Free Kids Project - soil sample and dust sample analysis . EPA
Method 3050, SW846 is a heated mineral acid digestion capable of
leaching lead from the soil matrix and into aqueous matrix.
9.4.2 Analysis
Analysis by ZCP, EPA Method 6010, SW846 3rd Edition, should be
at 220.353 nm. This is the suggested protocol.
9.4.3 Quality Assurance/Quality Control (ICP)
The Laboratory Task Leader will provide the following QC
samples:
Laboratory Duplicates. One sample out of 20 is prepared
and analyzed in duplicate. A control limit of +/-20 percent
relative percent difference is used.
Method BlanKff' Procedural blanks are prepared and analyzed
at a 5 percent frequency or one per batch digested if less
than 20 samples.
Matrix gpito/Mfltrix SplKe Duplicate (ms/msd) . Duplicate
samples are matrix spiked at ten times the detection
limit prior to digestion at a frequency of 20 percent.
Samples producing either spike recovery outside 75 to
125 percent control limits are re-analyzed by the
"Method of Standard Additions". The matrix spike
duplicate must fall within ฑ20 percent of true value.
-------
Section No. 9
Revision No. l
Date: 02/28/90
Page 35 of 55
Laboratory Control Samples (LCS). Laboratory Control
samples are prepared and analyzed according to each of
the procedures applied to the samples. One LCS is
prepared and analyzed once per operating shift. Percent
recovery control limits of 80 to 120 percent must be
attained. These maybe obtained through US EPA EMSL Las
Vegas, Nevada and should be in a soil matrix.
Field Blank. The field team will collect one blank per
day by carrying a sample of clean quartz sand into the
field in a normal sample container. The sample
container will be opened and exposed during the
collection of one sample, then closed and returned to
the laboratory. The field blank represents contamina-
tion added in the field during storage and sample
preparation.
Study Control Standards (SCS). Project soil and dust
study samples standards will be prepared and distribut-
ed at the beginning of the study. These will be
analyzed in conjunction with IฃS's. SCS samples will
be used as calibration standards for XRF analysis.
Their analysis via ICP will assist in assessing data
quality of XRF data.
These QA/QC analysis will be performed at the frequency
detailed below:
Field blank
1
per
20 samples
Laboratory control sample
1
per
operating shift
Laboratory duplicate soil
1
per
20 samples
MS/MSD
1
per
20 samples
Method blank
1
per
20 samples or one per batch
SCS
1
per
20 samples
9.5 x-Rqy Fluorescence
9.5.1 sample Preparation for Dust and Soils.
Refer to Section 9.3 for overview and B Appendices for detailed
procedures.
9.5.2 Analysis
The oxford LX 1000 XHRay Fluorescence Spectrophotometer is used
for the identification and quantitation of lead in soil samples.
The Kevex 0700 X-Ray Fluorescence Spectrophotometer is used for the
-------
Section No. 9
Revision No. 1
Date: 02/28/90
Page 36 of 55
identification and quantitation of lead in dust samples. These
protocols are presented in the appendices 3 and 4 respectively.
Quantification
The samples and standards are analyzed under identical
conditions and the resulting peak heights of the standard and
samples are compared for the Oxford, and after normalizing on the
Compton scattering peak for the Mo target for the Kevex. The
concentration of the lead in the sample is calculated by direct
proportions of peak heights to standard concentrations.
9.5.3 Quality Assurance/Quality Control
Soil
The Laboratory Task Leader will provide the following QC
samples:
Study Control Standard (SCS). Project soil samples will
be prepared and distributed at the beginning of this
study. The SCS are prepared and analyzed according to
the same procedure as applied to the samples. One set
of SCS's are analyzed twice per operating shift once at
the beginning and once at the end. During an operating
shift individual SCS's, will be analyzed on a rotating
basis,at the frequency of 1 per 20 samples. Percent
recovery control limits of 80 to 120 percent should be
attained.
Method Blank. Procedural blanks consisting an empty
analysis cup are analyzed at frequency of twice per
operating shift.
Laboratory Duplicates. One sample out of 20 is prepared
and analyzed in duplicate. A control limit of 20
percent relative percent difference is suggested.
Confirmatory TCP Analysis. Selected samples from XRF
analysis will be submitted for ICP confirmatory analysis at
the frequency of 1 per 20 samples analyzed. Selected
samples will be include SCS's, Duplicate and Replicate
samples, and field blanks. The frequency may be decreased
to 1 per 40 or more if data suggests a good correlation
between ICP and XRF results.
-------
Section No. 9
Revision No. l
Date: 02/28/90
Page 37 of 55
* Field Blank. The field team will collect one blank per
day by carrying a sample of clean quartz sand into the
field in a normal sample container. The sample
container will be opened and exposed during the
collection of one sample, then closed and returned to
the laboratory. The field blank represents contamina-
tion added in the field during storage and sample
preparation.
Laboratory Replicate. One sample per 20 is analyzed in
replicate. A control limit of 20 percent relative percent
difference is suggested.
These QC analyses will be performed at the following frequency:
Laboratory control samples At minimum one set if SCS
analyzed twice per operating
shift, at the beginning and end.
Then, one SCS per 20 samples on
a rotating basis during the
shift,
Field blank 1 per field sampling day
Method blank 2 per operating shift
Laboratory duplicate 1 per 20 samples
Laboratory Replicate 1 per 20 samples
Confirmatory ZCP Sample 1 per 20 samples (subject to
change)
DUSt
The Laboratory Task Leader will provide the following QC
samples:
Study Control Standards (SCS). Project dust samples
will be prepared and distributed at the beginning of
this study. The SCSs are prepared and analyzed
according to the same procedure as applied to samples
submitted for anlaysis. One set (consisting of 4
standards) of SCS's are analyzed at the beginning of
an operating shift . During an operating shift
individual SCS's, will be analyzed on a rotating basis,
at the frequency of 1 per 16 samples. Percent recovery
control limits of 80 to 120 percent should be attained.
* Method Blank. Procedural blanks consisting an empty
analysis cup are analyzed at frequency of twice per
operating shift.
-------
Section No. 9
Revision No. l
Date: 02/28/90
Page 38 of 55
Laboratory Duplicates. Due to the small quantity of
sample typically submitted for analysis laboratory
duplicates analysis will not be performed.
Confirmatory TCP Analysis. Selected samples may be
submitted for ICP confirmatory analysis at the frequency of
1 per 20 samples analyzed if practical. Selected samples
will include SCS's, and samples which contain sufficient
quantity for analysis, a minimum of 1 gram requried for
analysis. The frequency may be decreased if data suggests
a good correlation between ICP and XRF results.
Field Blank. No field blanks will be prepared for this
Laboratory Replicate. One sample per 20 is analyzed in
replicate. A control limit of 20 percent relative percent
difference is suggested.
These QC analyses will be performed at the following frequency:
study
Laboratory control samples
At minimum one set of SCSs are
analyzed at the beginning of a
shift. Then, one of SCS's per
15 analysis on a rotating basis
during the shift.
Method blank
2 per operating shift
Laboratory Replicate
Confirmatory ICP Sample
1 per 20 samples
1 per 20 samples (if applicable)
-------
Section No. 10
Revision No. 1
Date: 02/28/90
Page 39 of 54
10.0 DATA ANALYSIS, VALIDATION, AND REPORTING
10.1 Data Analysis and Reduction
In addition to the data collected in the field and recorded
on the chain-of-custody forms, data describing the processing and
analysis of samples will be accumulated on the laboratory and
recorded on laboratory Analytical work sheets. Laboratory
analytical work sheet will contain:
Date of processing or analysis;
Laboratory sample identification numbers;
Field identification sample number;
Analyses or operation performed;
Calibration data;
Quality control samples data;
Concentrations/dilutions required;
Instrument readings;
Special observations (operational); and
Analyst's, reviewer's, and person making calculations
signature.
Data reduction is performed by the individual analysts which
consists of calculating concentrations in samples from the raw data
obtained from the measuring instruments. The complexity of the
data reduction will be dependent on the specific analytical method
and the number of discrete operations (extractions, dilutions,
weighing (dust) and concentrations) involved in obtaining a sample
that can be measured.
IฃE
For XCP the analytical method which utilizes a calibration
curve, sample responses will be applied to the linear regression
line to obtain an initial raw result which is then factored into
equations to obtain the estimate of the concentration in the
original sample. Rounding will not be performed until after the
final result is obtained to minimize rounding errors. Results will
not normally be expressed in more than two significant figures.
Soil
XRF analysis of soils using the Oxford utilizes a response
factor methodology. A response factor is generated on a daily
basis by analyzing the SCS. which fall in each of the four
concentration ranges of the Oxford instrument. The response factor
is calculated by measuring the peak height of the L-Alpha line (in
millimeters) for the SCS and then dividing it into the SCS's
assigned value. All other unknowns, for a specific concentration
-------
Section No. 10
Revision No. 1
Date: 02/28/90
Page 40 of 54
range are then determined by multiplying their peak heights by the
response factor. Rounding is performed for the response factor to
the nearest 10. Rounding is performed on final results so they may
be expressed in no more than two significant figures.
PUSt
XRF analysis of dust using the Kevex utilizes a peak comparison
methodology. In this procedure SCS Spectra are generated for
comparison with unknowns. L-Alpha peaks of the SCS and unknown are
overlayed on a video screen after normalizing for the compton
scatter and concentrations are read directly by comparison using
the screen's grid marks. Measurement of unknowns is dependent on
sample weight and concentration. An unknown spectra is generated
first, then based on weight and concentration, the appropriate SCS
is selected for overlay and the concentration is determined.
Rounding is done at the time concentration determination with
results expressed in two significant figures.
Copies of all raw data and the calculations used to generate
the final results will be retained on file to allow reconstruction
of the data reduction process at a later date. ICP laboratory data
and Kevex XRF diskettes containing spectra for dust analysis will
be maintained by US EPA in Lexington, MA. All other data will be
maintained in contractor files until project termination.
10.2 Data Validation and Review
Validation of measurements is a systematic process of review-
ing a body of data to provide assurance that the data are adequate
for their intended use. The process includes the following
activities:
editing,
screening,
checking,
auditing,
verification,
certification, and
review.
Data validation activities will be documented and records kept
of any necessary corrective or remedial action.
System reviews are performed at all levels. The individual
analyst constantly reviews the .quality of data through calibration
checks, quality control sample results, and performance evaluation
-------
Section No. 10
Revision No. l
Date: 02/28/90
Page 41 of 54
samples (ICP Analysis only). These reviews are performed prior to
submission to the Data Reviewer or the Task Leader.
The Data Reviewer and/or the Task Leader review data for
consistency and reasonableness with other generated data and
determine if program requirements have been satisfied. Selected
hard copy output of data (chromatograms, spectra, laboratory work
sheets, etc.) will be reviewed to ensure that results are inter-
preted correctly. Unusual or unexpected results will be reviewed,
and a resolution will be made as to whether the analysis should be
repeated. in addition, the Task Leader or Data Reviewer will
recalculate selected results to verify the calculation procedure.
The Region I ESD Quality Assurance Officer independently
conducts a complete review of selected projects to determine if
laboratory and quality assurance/quality control requirements have
been met. Discrepancies will be reported to the Project Manager
and/Technical Project Director for resolution.
!(r*-
10.3 Data RepprUnq
Laboratory reports of data will be edited by comparing with
original calculations. Subsequent data tabulations will be edited
by comparing with the laboratory analytical work sheets.
Reports will contain final results (uncorrected for blanks and
recoveries), methods of analysis, levels of detection (ICP Analysis
only), and method blanks data. In addition, special analytical
problems, and/or any modifications of referenced methods will be
noted. The number of significant figures reported will be
consistent with the limits of uncertainty inherent in the
analytical method. Consequently, most analytical results will be
reported to no more than two significant figures.- Data are
normally reported in units commonly used for the analyses
performed. Concentrations in liquids are expressed in terms of
weight per unit volume (e.g., milligrams per liter). Concentra-
tions in solid or semi-solid matrices are expressed in terms of
weight per unit weight of sample (e.g., milligrams/kilogram, ppm).
Reported detection limits will be the concentration in the
original matrix corresponding to the low level instrument calibra-
tion standard after concentration, dilution, and/or extraction
factors are accounted for.
Prior to issuance of a report for soil and dust analysis,
results reported for each sample are verified to assure proper
identification by comparing the chain-of-custody forms, laboratory
analytical work sheets, and raw. data. Based on the results of this
-------
Section No. 10
Revision No. 1
Date: 02/28/90
Page 42 of 54
validation procedure, the laboratory certifies that the results are
in compliance with the quality assurance objectives for accuracy
and precision. Upon certification by the Task Leader, the report
is reviewed by the QAC (if deemed necessary), then provided to the
Project K.anager for distribution.
ICP results reported for dust and soil analysis are verified
and reported according to US EPA, Lexington, MA Laboratory
procedures.
-------
Section No. 11
Revision No. 1
Date: 02/28/90
Page 43 of 55
11.0 INTERNAL QUALITY CONTROL CHECKS
n.i general
For each major measurement parameter, appropriate quality
control checks shall be established. Field blanks should be taken
to check for contamination introduced during sample collection and
shipping. Study control samples should be analyzed to determine
the accuracy of the analytical technique. Laboratory replicates
and duplicates should be analyzed to determine the precision of the
analysis. Spiked (fortified) samples should be used to determine
the accuracy of the analysis (1CP only).
I CP Conifirmatory Analysis of selected soil and dust will assist
in assessing the comparability between XRF and ICP results.
11.2 Field Blanks
No field blanks for dust are required at this time. Field
blank samples for soil/sediment matrices are not readily available,
however quartz sand will be used. Field blanks will be prepared
at the rate of one per sampling event day. Enough sample will be
prepared for both XRF and ICP analysis. A Field Blank will be
submitted for XRF analysis for each batch of samples associated
with a sampling event day. Samples selected for ICP analysis will
be accompanied by selected Field Blanks not necessarily
representing their sampling event day(s).
11.3 Laboratory Duplicate
For soils and ICP analysis duplicate soil samples should be
prepared in the laboratory at a rate of one per every 20 samples
analyzed. For dust laboratory duplicate will not be prepared due
to the small amount of sample typically collected.
11.4 Laboratory Control Samples (ICP only)
Laboratory control samples (LCS) consisting of secondary
standards ultimately traceable to National Bureau of Standards
(NBS) or EPA Environmental Monitoring and Surveillance Laboratory
(EMSL) primary standards will be prepared for ICP, The LCS will
be analyzed at least once per operating shift and should fall
within the established recovery control limits of 80 to 120
percent.
-------
Section No. 11
Revision No. 1
Date: 02/28/90
Page 44 of 55
11.5 Completeness
Completeness of scheduled sample collection will be controlled
in the field by comparing predetermined sample locations on the
gridded lot plan for each residential property with samples
collected in the field at that site each day. Daily checking of
site log books and chain-of-custody logs will provide further
control on documentation and completeness. The determination of
the completeness objective will be the responsibility of the QAC.
11.6 Study Control Standards (SCS)
XRF analysis Study Control standards for soil and dust have
been made available by Dr. Thomas Spittler U.S. EPA. For soil
analysis the SCS's are analyzed as a set twice per operating shift
once at the beginning and once at the end with individual SCS's
analyzed during the shift at a frequency of 1 per 20 samples
analyzed on a rotating basis. The daily SCS results should fall
within control limit of 80 to 120 percent of their respective
concentrations. For dust SCS's are analyzed once at the beginning
of the shift and results should also fall within 80 to 120 percent
or their respective concentrations.
11.7 Laboratory Replicates
Replicate soil and dust samples analysis will be conducted at
a rate of one per 20 samples analyzed.
11-8 Comparability - ICP Confirmatory Analysis
Selected soil and dust XRF samples will be submitted for
confirmatory analysis using Inductively Coupled plasma Emission
Spectrophotometry at a frequency of one per 20 samples analyzed
(this frequency may be decreased at a later date). ICP and XRF
analysis will be compared to define the limits of comparability
between these two methods.
-------
Section No. 12
Revision No. 1
Date: 02/28/90
Page 45 of 55
12.0 AUDITS
Quality assurance audits are performed to assure and document
that quality control measures are being utilized to provide data
of acceptable quality and that subsequent calculations, interpreta-
tion, and other project outputs are checked and validated.
The Quality Assurance coordinator (QAC) will conduct system
and performance audits and reviews of interpretations and reports
based on the measurement system outputs. If any of the procedures
to assess precision and accuracy described in Section 14.2 indicate
potential data problems, an audit will be initiated, if approp-
riate.
12.1 system? Audit
A systems audit will be conducted on all components of
measurement systems to determine proper selection and utilization.
The systems audit includes evaluation of both field and laboratory
procedures.
Organization and personnel. The project organization is
reviewed for compliance with the proposed organization and for
clarity of assigned responsibility. Personnel assigned to the
project will be reviewed to determine that assigned responsibility,
skill, and training of the personnel are properly matched. The
Technical Director maintains firsthand knowledge of his team's
capabilities and will discuss the organization's efficacy with the
QAC. Assigned personnel may be interviewed by the QAC during an
audit.
Facilities and Equipment. The audit will address whether field
tools and analytical instruments are selected and used to meet
requirements specified by the project objectives stated in the
QAPjP. Equipment and facilities provided for personnel health and
safety will also be evaluated. Documentation procedures used in
the field will receive special attention.
Analytical Methodology. Routine external performance evalua-
tions as well as blind internal performance evaluations are
generally conducted. A review of analytical methodology in regard
to the data requirements for the project will also be performed.
An in-laboratory observation of analyst technique, data reduction,
and record keeping may be performed if determined necessary.
Periodic review of precision and accuracy data is essential.
sampling and Sample Handling Procedure. An audit of scheduled
samples vs. samples collected vs. samples received for analysis may
be performed. Field documentation will be reviewed. If deemed
necessary, a site visit will be made to assure that designated
control procedures are practiced during sampling activities.
-------
Section No. 12
Revision No. 1
Date: 02/28/90
Page 46 of 55
Data Handling. During a systems audit, the QAC will review
data handling procedures with the TD and Task. Leaders. Accuracy,
consistency, documentation, and appropriate selection of methodolo-
gies will be discussed.
12.2 Performance Audits
These audits are intended primarily for analytical and data
generation systems. The EPA Region I ESD laboratory regularly
participates in, and successfully completes U.S. EPA Performance
Evaluations (WS and WP Series). Ongoing performance evaluations
include duplicates, matrix spikes, QC check samples, etc., with
regard to ICP analysis.
12.3 Project Audits
Project audits encompass the aspects of both the systems audit
and the performance audits. The project audit typically occurs at
least twice for a short-term project and more often during long-
term projects. Timing is keyed to the systems involved and the
project objectives.
12.4 ovmiitv Assurance (OA) Audit Report
A written report (Figure 12-1) of the QA audit may be prepared
to include:
an assessment of project team status in each of the
major project areas;
clear statements of areas requiring improvement or
problems to be corrected. Recommendations and assis-
tance will be provided regarding proposed corrective
actions or system improvements. If no action is
required, the report will state that the QA audit was
satisfactorily completed;
a timetable for any corrective action required; and
a follow-up to assure that recommendations have been
implemented.
-------
Figure 12-1
QUALITY ASSURANCE AUDIT REPORT
Project: _______________________________
Project Ho.s
Quality Assurance Coordinator:
Project Aspects Audited:
Laboratory/Technical Director:
Audit Conducted By:
for the period to
Date of Audit:
Personnel Interviewed:
1.0 Purpose and Objectives of the Project Aspects Audited
2.0 Brief Description of the Sampling and Analytical Requirements
3.0 Organization and Personnel
4.o Facilities Utilized
s.o Analytical Methodologies
6.o sampling and Sample Handling
7.0 Quality control Measures Utilized
8.o Pata Handling
9.0 Quality Assurance Deficiencies
10.0 Recommended Correction Actions and Schedule
Section No. 12
Revision No. 1
Date: 02/28/90
Page 47 of 55
SignedDate
Title
Distribution:
Reviewed by Date
Title
-------
Section No. 13
Revision No. 1
Date: 02/28/90
Page 48 of 55
13.0 PREVENTATIVE MAINTENANCE
Preventative maintenance of field equipment proceeds routinely
before each sampling event. More extensive maintenance would be
performed based on the number of hours in use. Preventative
maintenance of EPA Region I ESD laboratory analytical equipment is
the responsibility of that laboratory.
-------
Section No. 14
Revision No. l
Date: 02/28/90
Page 49 of 55
14.0 DATA ASSESSMENT
14.1 general
The purpose of data quality assessment is to assure that data
generated under the program are accurate and consistent with
project objectives. The quality of data will be assessed based on
the precision, accuracy, consistency, and completeness of the data
that are measured or generated.
Data quality assessment will be conducted in three phases:
14.1.1 PhaB8 1
Prior to data collection, sampling, and analysis, procedures
are evaluated in regard to their ability to generate the approp-
riate, technically acceptable information required to achieve
project objectives. This QA/QC Plan meets this requirement by
establishing project objectives defined in terms of required
sampling analysis protocols.
14.1.2 Phase 2
During data collection, results will be assessed to assure
that the selected procedures are efficient and effective and that
the data generated provide sufficient information to achieve
project objectives. Precision and accuracy of measurement systems
will also be evaluated. In general, evaluation of data will be
based on performance audits, results of duplicate and reference
sample analyses, and review of completeness objectives.
Documentation may include:
number of duplicate and reference samples analyzed;
identification of statistical techniques, if used, to
measure central tendency, dispersion, or testing for
outliers;
use of historical data and its reference; and
identification of analytical method.
14.1.3 Phflgg 3.
Throughout the data collection activities, an assessment of
the adequacy of the data base generated in regard to completing
project objectives will be undertaken throughout the project.
Recommendations for improved quality control will be developed, if
appropriate. In the event that data gaps are identified, the QAC
may recommend the collection of additional raw data to fully
support the project's findings and recommendations.
-------
Section No. 14
Revision No. 1
Date: 02/28/90
Page 50 of 55
14.2 Precision and Accuracy
Assessment of precision and accuracy of analytical data is
accomplished via review of duplicate analyses (precision) and
reference standard (accuracy) in soil. Precision is generally
expressed as the relative percent difference (RPD). Accuracy is
expressed as percent recovery. Precision must be assessed for each
matrix, since distribution of contaminants may be non-homogeneous,
especially in soil. Precision in samples must be reviewed with
knowledge of the matrix and level of analyte present. Corrective
action or documentation of substandard precision is the labora-
tory^ responsibility. Accuracy is also impacted by matrix
interferences. Each method which provides quality control require-
ments and acceptance criteria also specifies the method of generat-
ing the data to be reviewed. It is the laboratory*s responsibility
to attempt to identify the source of substandard recoveries and
either take corrective action or document the cause.
Calculations are presented below:
Accuracy;
Kit* observed value ffm
theoretical value
where: %R ป percent recovery
Precision control requirements and acceptance criteria also
specifies the method of generating the data to be reviewed. It is
the laboratory's responsibility to attempt to identify the source
of substandard recoveries and either take corrective action or
document the cause.
-------
Section No. 14
Revision No. 1
Date: 02/28/90
Page 51 of 55
Precision (as determined by Relative percent difference):
(A1 - A,)
RPD - X 100
-------
Section No. 15
Revision No. 1
Date: 02/28/90
Page 52 of 55
15.0 CORRECTIVE ACTION
Corrective or preventative action is required when potential
or existing conditions are identified that may have an adverse
impact on data quantity or quality. Corrective action could be
immediate or long-term. In general, any member of the program
staff who identifies a condition adversely affecting quality can
initiate corrective action by notifying in writing his or her
supervisor and the QAC. The written communication will identify
the condition and explain how it may affect data quality or
quantity.
Ap analysis or analytical system is considered to be out-of-
control when it does not conform to the conditions specified by the
method or standard operating procedures which apply. To confirm
that an analysis or analytical system is in control, the laboratory
routinely performs instrument calibration checks, analysis of
method blanks and method blank spikes. These results are compared
to the results of quality control samples to laboratory control
charts or analytical protocol criteria (e.g., U.S. EPA-CLP).
A Corrective Action Documentation Form, Figure 15-1, is to be
completed for each out-of-control situation. The analyst, working
with his or her supervisor or Task Leader, will attempt to deter-
mine the cause of the problem and take appropriate corrective
action. Analysis may not resume until the problem has been
corrected and it is determined that the analysis is back in
control. Demonstration of the restoration of analytical control
will normally be accomplished by generating satisfactory calibra-
tion and/or quality control sample data. This documentation will
be attached to the corrective action documentation form to be
placed in the project files.
is.i immediate Corrective Action
Immediate corrective action is applied to spontaneous, non-
recurring problems, such as an instrument malfunction. The
individual who detects or suspects nonconformance to specify the
previously established criteria or protocol in equipment, instru-
ments, data, methods, etc., will immediately notify his/her
supervisor. The supervisor and the appropriate Task Leader will
then investigate the extent of the problem and take the necessary
corrective steps. If a large quantity of data is affected, the
Task Leader must prepare a memorandum to the Project Manager, the
Technical Project Director and the QAC. These individuals will
collectively decide how to proceed. If the problem is limited in
scope, the Task Leader will decide on the corrective action
measure, document the solution in the appropriate workbook, and
notify the Project Coordinator,- the Technical Project Director, and
the QAC.
-------
Section No. 15
Revision No. 1
Date: 02/28/90
Page 53 of 55
15.2 Long-Term Corrective Action
Long-term corrective action procedures are devised and
implemented to prevent the recurrence of a potentially serious
problem. The QAC will be notified of the problem and will conduct
an Investigation to determine the severity and extent of the
problem. The QAC will then file a corrective action request with
the Project Manager.
Corrective actions may also be initiated as a result of other
activities, including:
Performance Audits;
System Audits;
Laboratory field/comparison studies; and
QA Program Audits.
The QAC will be responsible for documenting all notifications,
recommendations, and final decisions. The Project Coordinator and
the QAC will be jointly responsible for notifying program staff and
implementing the agreed upon course of action. The QAC will be
responsible for verifying the efficacy of the implemented actions.
The development and implementation of preventative and corrective
actions will be timed, to the extent possible, so as to not
adversely impact either project schedules ojr subsequent data
generation/processing activities. However, scheduling delays will
not override the decision to correct any data collection deficien-
cies or inaccuracies before proceeding with additional data collec-
tion. The QAC will also be responsible for developing and im-
plementing routine program controls to minimize the need for
corrective action.
-------
Section No. 15
Revision No. 1
Date: 02/28/90
Page 54 of 55
FIGURE 15-1
CORRECTIVE ACTION DOCUMENTATION FORM
DESCRIPTION OF PROBLEM and when identified:
State cause of problem if known or suspected:
SEQUENCE OF CORRECTIVE ACTION: (If no responsible person is
identified, bring this form directly to the QA Coordinator)
State date, person, and action planned: _______________________
CA Initially Approved By:
Date:
Follow-up dates:
Description of follow-up:
Final CA Approved By:
Date:
-------
Section No. 16
Revision No. 1
Date: 02/28/90
Page 55 of 55
16.0 REPORTS TO MANAGEMENT
Periodic summary reports will be prepared to inform management
of project status. The reports vill include:
periodic assessment of measurement data accuracy,
precision, and completeness;
results of performance audits and/or systems audits;
significant QA problems and recommended solutions; and
status of solutions to any problems previously
identified.
Periodic Analytical Summary Progress Reports which includes
results for soil and dust sample analysis and an on-going
status of samples received, analyzed and reported.
Additionally, any incidents requiring corrective action will
be fully documented. Procedurally, the QAC will prepare the
reports to management. These reports will be addressed to the
Project Manager, with copies to the Technical Project Director and
the QAC. The summary of findings shall be factual, concise, and
complete. Any required supporting information will be appended to
the report.
ฆfou.s. GOVERNMENT PRINTING OFFICE: 1993 - 750-068/60015
------- |