&EPA
United States
Environmental Protection
Agency
Hearth Effects Research
Laboratory
Research Triangle Park NC 27711
EPA-600/1-78-055
August 1978
Research and Development
Epidemiologic Study
of the Effects of
Automobile Traffic
on Blood Lead Levels
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL HEALTH EFFECTS RE-
SEARCH series. This series describes projects and studies relating to the toler-
ances of man for unhealthful substances or conditions. This work is generally
assessed from a medical viewpoint, including physiological or psychological
studies. In addition to toxicology and other medical specialities, study areas in-
clude biomedical instrumentation and health research techniques utilizing ani-
mals but always with intended application to human health measures.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/1-78-055
August 1978
EPIDEMIOLOGIC STUDY OF THE EFFECTS OF AUTOMOBILE
TRAFFIC ON BLOOD LEAD LEVELS
by
D. E. Johnson, R. J. Prevost, J. B. Tillery,
K. T. Kimball and J. M. Hosenfeld
Southwest Research Institute
3600 Yoakum Blvd.
Houston, Texas 77006
Contract No. 68-02-2227
Project Officer
Warren A. Galke
Population Studies Division
Health Effects Research Laboratory
Research Triangle Park, N.C. 27711
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
HEALTH EFFECTS RESEARCH LABORATORY
RESEARCH TRIANGLE PARK, N.C. 27711
-------
DISCLAIMER
This report has been reviewed by the Health Effects
Research Laboratory, U.S. Environmental Protection Agency,
and approved for publication. Approval does not signify
that the contents necessarily reflect the views and policies
of the U.S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute
endorsement or recommendation for use.
11
-------
FOREWORD
The many benefits of our modern, developing, industrial
society are accompanied by certain hazards. Careful assessment
of the relative risk of existing and new man-made environmental
hazards is necessary for the establishment of sound regulatory
policy. These regulations serve to enhance the quality of our
environment in order to promote the public health and welfare and
the productive capacity of our Nation's population.
The Health Effects Research Laboratory, Research Triangle
Park, conducts a coordinated environmental health research
program in toxicology, epidemiology, and clinical studies using
human volunteer subjects. These studies address problems in air
pollution, non-ionizing radiation, environmental carcinogenesis
and the toxicology of pesticides as well as other chemical
pollutants. The Laboratory participates in the development and
revision of air quality criteria documents on pollutants for
which national ambient air quality standards exist or are proposed,
provides the data for registration of new pesticides or proposed
suspension of those already in use, conducts research on hazardous
and toxic materials, and is preparing the health basis for non-
ionizing radiation standards. Direct support to the regulatory
function of the Agency is provided in the form of expert testimony
and preparation of affidavits as well as expert advice to the
Administrator to assure the adequacy of health care and surveillance
of persons having suffered imminent and substantial endangerment
of their health.
Lead, because of its variety of uses and its toxicity,
has been a pollutant of much concern. The present study is an
attempt to investigate the impact of automotive lead emissions
on the amount of lead in the body of persons living along
streets with typical urban traffic volume. The blood lead
levels of persons of different ages have been investigated in
relationship to traffic volumes of up to about 30,000 vehicles
per day.
F. G. Hueter, Ph. D.
Acting Director,
Health Effects Research Laboratory
111
-------
ABSTRACT
The objective of this research project was to character-
ize the. absorption of lead by people of different age-sex
groups exposed to automobile emissions of lead at traffic
densities from less than 1000 cars per day to 25,000 cars
per day. The relationships between traffic densities and
lead content of various environmental and household samples
were also examined. Participant selection was based upon
a strict set of criteria which eliminated anyone whose
blood-lead level was affected by exposure to sources of non-food
lead other than automobile emissions.
The degree of absorption by the participants was de-
termined by measuring blood-lead concentrations and blood-
FEP levels. Blood CO levels were also measured. A micro-
analytical technique was developed to accurately measure
blood-lead at concentrations less than 30 yg/100 ml in
100 yl capillary blood samples from children. Also, hand-
wipe samples from children were analyzed for lead.
The relationships between blood lead, handwipe lead
and blood CO levels of participants of different age-sex
groups and different traffic densities were examined. No
significant relationships with traffic densities were found
in the range of exposures studied.
Selected household samples were also measured for lead
content to eliminate them as possible sources of lead for the
participants and to examine the relationship of their lead
concentration to traffic densities. Household samples in-
vestigated included water from resident's kitchen tap, paint
from the interior and exterior surfaces of the residences,
28-day indoor dust samples, and window sill wipe samples.
IV
-------
Water was eliminated as a significant source of lead ingestion
and there were no relationships between any household samples'
lead content and traffic densities.
Outside environment a], samples were also analyzed for
lead. These included sell from participant residences,
outdoor dust from selected sites in the study area, and
air particulate matter from streets of different traffic
densities where participants lived. Traffic counts were
also made on these streets to determine traffic densities.
Physical and chemica] properties of the soil samples were
also determined.
Increased soil lead concentrations were observed with
increasing traffic density. Physical and chemical charac-
teristics cf the soil in the study area favor lead retention
in the soil matrix. Lead in air was found to be related to
increasing traffic density although the slope of the regression
line was not steep.
A positive relationship between smoking and blood lead
levels was found for both females and males in this study.
Females who smoke had significantly higher blood lead levels
than female exsmokers and nonsmokers. Male smokers and
exsmokers had higher blood lead levels than nonsmokers,
although this difference was not significant.
v
-------
TABLE OF CONTENTS
Page
I. Introduction 1
II. Methods 14
A. Site Selection 14
1. General Site Requirements 14
2. Study sites considered 15
3. Description of final study site 18
selected
4. Authorization of study by local 22
governments
B. Project Staff 25
1. Overall project 25
2. Field operations 28
C. Determination of the relationship 35
between air lead levels and
traffic flow characteristics
1. Design 35
2. Sample collection 48
3. Sample analysis and quality 55
control procedures
D. Determination of the relationship 65
between blood lead levels
and traffic density
1. Description of study 65
2. Data collection procedures 67
3. Sample analysis procedures 88
4. Statistical procedures 122
5. Participant recruitment 125
III. Results 138
A. Determination of the Relationship 138
between Air Lead Levels and
Traffic Flow Characteristics
VI
-------
Paqe
1. Results of traffic density 138
mini-study
2. Results of replicate hi-vol 141
mini-study
3. Results of particle size 144
mini-study
4. Results of distance from street 149
mini-study
5. Results of intersections 150
mini-study
6. Results of speed limit 158
mini-study
7. Results of indoor vs outdoor 160
air lead mini-study
8. Results of indoor vs outdoor 164
dustfall mini-study
9- Results of collection times less 166
than 24 hours mini-study
B. Determination of the Relationship 170
between Blood Lead Levels and
Traffic Density
1- Results of recruitment activities 170
2. Description of study participants 182
3. Environmental data 184
4. Biological data 201
5. Multivariate analysis 239
IV. Discussion 241
A. Air Monitoring Study 241
B. EpidemiologicStudy of Traffic Density 249
Relative to Levels of Lead in the
Environment and Blood of Residents
V. Conclusions 253
VI. Recommendations 256
References 258
Appendix A Justification for Change of Study Site 262
to Dallas
Appendix B Letters of Permission to Proceed from 270
Local Governments
Appendix C Justification for Household Health Survey 282
for Lead
Vll
-------
Page
Appendix D Air Lead Concentrations & Corresponding 292
Traffic Counts
Appendix E Variables Tap Water Lead, Soil Lead, 297
Indoor Dust Lead, Windowsill Wipe
Lead and Traffic for each Household
Appendix F Variables for Hand-wipe Lead, Blood 305
Lead, and Traffic Counts for each
Participant (Children only)
Appendix G Variables pertaining to Identification 309
of the Participant and Blood Analyses
Appendix H Fingerprick Samples 1 & 2 for each 325
Participant
Appendix I Paint Lead Concentration, Distance from 328
Street, and Composition of each
Household
Appendix J Report to Southwest Research Institute 335
by Geoderma Consultants - Dallas, Texas
Vlll
-------
LIST OF TABLES
Page
1. Analytical Parameters of Environmental 56
Samples
2. Analytical Parameters for Atomic Absorp- 61
tion Spectrophotometric Lead Analysis
3. Analytical Parameters of Environmental 93
Samples
4. Effect of Refrigeration on Blood Lead 102
Values of Whole Blood (CDC) Bovine Blood
5. Analytical Parameters of Biological 104
Samples
6. Determining Blood Volume in Capillary 112
Tube by Indirect Measurement
7. CDC Bovine Blood as Quality Control for 113
Capillary Blood Lead Analysis
8. Sample Opinion Survey re Air Pollution 134
Concern
9. Mean Air Lead Concentrations and Traffic 138
Counts at Each Location and Traffic Density
10. Lead Concentration in Five Particle Size 145
Ranges at Four Distances from the Street
11. Lead Concentration and Proportion in Five 147
Particle Size Ranges at Four Distances
from the Street
12. Concentration of Lead Suspended in Air at 148
Increasing Distances Expressed at as Percent
of Lead at Five Feet from the Street
13. Concentration of Lead Suspended in Air at 150
Four Distances from the Street for Two Days
at Three Traffic Densities
14. Intersection Study: Air Lead Concentration at 155
Intersections and Midblock Locations
15. Corner Home Study: Air Lead Concentration at 156
Intersections and Midblock Locations
IX
-------
LIST OF TABLES CONT'D
Page
16. Air Lead Concentrations at Two Speed Limits 158
17. Outdoor Dustfall Lead Concentrations from 164
Ten Locations with Corresponding Indoor
Dustfall Lead Concentrations and Traffic
Counts
18. Collection Times Less Than 24 Hours: Air Lead 167
Concentrations and Traffic Counts for Five
Collection Times at Three Traffic Densities
19. Recruitment Results I"72
20. Number of Participants by Age, Sex, and 176
Traffic Level
21. Number of Participating Households at 187
Intersections
22. Participant Demographic Characteristics- 183
23. Test for Differences Among Sites in Tap 191
Water Using Kruskal-Wallis Test
24. Paint Lead Concentration: Means, Standard 208
Errors and Sample Sizes for Indoor and
Outdoor Paint Lead at Each Traffic Density
25. Results of Test for Extreme Values (Dixon's 212
B-^ Test) on Blood Samples for Participants
Potentially Exposed to other Lead Sources
26. Two-Way ANOVA of the Effects of Sites and 213
Samples on Log (Blood Lead) for Each
Age-Sex Group
27. Blood Lead Concentrations: Means, Confidence 214
Limits, and Sample Sizes for Each Age-Sex
Group at Each Site
28. Two-Way ANOVA of the Effects of Sex and Age 217
on Log (Blood Lead) at Sites 1,2,3 and 4
and All-Sites
29. FEP for Each Age-Sex Group at Each Site 223
30. HCT for each Hematocrits for Each Age-Sex 224
Group at Each Site
x
-------
LIST OF TABLES CONT'D
Page
31. Variables and Regression Coefficients 230
Used in Least Squares Regression Analysis
32. Simple Correlation Coefficients among All 231
Variables for Children (above) and Adults
(below)
33. Correlations with the 25 Independent Variables 235
of the Six Principal Component Factors that
Contribute Most to R2 (Data from Children Only)
34. Correlations with the 25 Independent Variables 238
of the Six Principal Component Factors that
Contribute Most to R2 (Data from Adults Only)
XI
-------
LIST OF FIGURES
Page
1. Traffic Artery Map of Dallas Metro Area 20
2. Principal Study Areas within the Dallas 23
Metro Area
3. Qualifications for On-Site Coordinator 29
and Sample Task Descriptions
4. Dallas Traffic Lead Study Instructions 31
for Performing Household Surveys
5. General Criteria for Placement of High 36
Volume Particulate Sampler
6. Placement of Samplers for Distance from 39
Road Mini-study
7. Placement of Samplers for Intersection 41
Mini-study
8. City of Dallas, Western Section 44
9. City of Dallas, North Central Section 45
10. City of Arlington 46
11. Placement of Samplers for Collection Times 47
Less Than 24 Hours Mini-study
12. Air Sampling and Traffic Counting 53
13. Apparatus for the Acid Digestion of 60
Air, Windowsill Wipe, and Hand-Wipe
Samples
14. Analytical Curve for Lead in Air Particulate 63
15. Analytical Curve for Lead in Outdoor Dust 64
16. Traffic Lead Household Questionnaire 68-69
17. Traffic Lead Individual Questionnaire 70-71
18. Neighborhood Lead Study - Participants and 72
Household Checklist
-------
LIST OF FIGURES CONT'D
Page
19. Household Validation Form 83
20. Traffic Lead Validation 34
21. Analytical Curve for Lead in Soil 92
22. Analytical Curve for Lead in Water 94
23. Analytical Curve for Lead in Indoor Dust 95
24. Analytical Curve for Lead in Windowsill 97
Wipes
25. X-ray Fluorescence Analyzer - Rhenium 98
Filter-Calibration Curve
26. X-ray Fluorescence Analyzer - Lead Filter- 99
Calibration Curve
27. Analytical Curve for Lead in Hand-wipes 101
28. Analytical Curve for Lead in Venous Blood 107
29. Analytical Curve for Lead in Capillary Blood 116
30. General Information for Participants of 127
Traffic Lead Public Health Survey
31. Letter of Introduction 133
32. Volunteer's Informed Consent 136-137
33. Air Lead Levels by Traffic Count 139
34. Frequency Distribution of the Variable Air 140
Lead
35. Replicate Air Samples 142
36. Proportion of Total Lead Found in each 146
Particle Size Fraction vs Distance
from Street
37. Air Lead vs Distance from Street 148
38. Air Lead Concentrations as a Function of Distance 151
from Road and Traffic Density
Kill
-------
LIST OF FIGURES CONT'D
39. Intersection Study: Air Lead Levels at 153
Intersection and Midblock Locations
40. Corner Home Study: Air Lead Levels at 157
Midblock and Intersection Locations
41. Air Lead Concentrations vs Two Speed Limits 160
42. Indoor vs Outdoor Air Lead Concentrations 162
at Two Traffic Locations
43. Lead in Dust (Indoor and Outdoor) vs 165
Traffic Density
44. Indoor vs Outdoor Dust Lead Concentrations 166
at Nine Matched. Locations
45. Air Lead Levels vs Traffic Density for Four 169
Collection Times at Three Traffic Levels
46. Frequency Distribution of Soil Lead 185
47. Frequency Distribution of Soil Lead at 185
Each Traffic Density
48. Soil Lead vs Traffic Density 186
49- Frequency Distribution of Tap Water Lead 190
50. Frequency Distribution of Tap Water Lead 191
at Each Traffic Density
51. Tap Water Lead vs Traffic Density 192
52. Frequency Distribution of Indoor Dust Lead 193
53. Frequency Distribution of Indoor Dust Lead 194
at Each Traffic Density
54- Indoor Dust Lead vs. Traffic Density 195
55. Frequency Distribution of Windowsill Wipe Lead 196
56. Frequency Distribution of Windowsill Wipe 196
Lead at Each Traffic Density
xiv
-------
LIST OF FIGURES CONT'D
Page
57. Windowsill Wipe Lead vs Traffic Density 197
58. Frequency Distribution of Hand-wipe Lead 199
59. Frequency Distribution of Hand-wipe Lead 200
at Each Traffic Density
60, Hand-wipe Lead vs. Traffic Density 201
61. Venous Blood Lead vs Capillary Blood Lead " 203
Samples from the Same Participants
62. Frequency Distribution of Mean Blood Lead 205
63. Frequency Distribution of Mean Blood Lead 205
for Each Age and Sex Group
64. Frequency Distribution of Mean Blood Lead 206
at Each Traffic Density
65. Paint Lead vs Traffic Density 209
66. Frequency Distribution of Mean Blood Lead 210
Levels of Participants Exposed to
Paint Lead Below and Above 4.0
mg/cm2 in their Homes
67. Blood Lead Levels vs Traffic Density for 215
Each Age and Sex Group
68. Frequency Distribution of FEP 219
69. Frequency Distribution of FEP at Each 220
Traffic Density
70. Frequency Distribution of FEP for Each 221
Age and Sex Group
71. Frequency Distribution of HCT 222
72. Frequency Distribution of HCT at Each 222
Traffic Density
73. Frequency Distribution of HCT for Each 223
Age and Sex Group
74. Frequency Distribution of Carbon Monoxide 225
in Blood
xv
-------
LIST OF FIGURES CONT'D
Page
75. Frequency Distribution of the Variable Blood 227
Lead in Three Smoking Groups of Adult Females
and the Results of ANOVA among these Groups
76. Frequency Distribution of the Variable Blood 228
Lead in Three Smoking Groups of Adult Males
tne the Results of ANOVA among these Groups
xvi
-------
I. INTRODUCTION
MOBILE EMISSIONS
Review of Related Lead Studies
Contributions of lead from various sources in the
environment (air, water, soil, house dust, food) to accumu-
lations in people have been the subject of many scientific
investigations. These have been well summarized in the Envir-
onmental Protection Agency's air quality criteria document
for lead, now in its third draft^). The investigation
reported herein is centered on the contribution of lead to
populations from ambient air sources, primarily from auto-
mobile emissions; accordingly, discussion of the literature
in this report will be restricted to these subjects.
Approximately 88 percent of the total atmospheric
emissions of lead in 1975 were from automobile use with
the remainder coming from lead-using industries primarily
smelters. There has been concern regarding the contri-
bution of lead from mobile sources to the levels of lead in
people, especially those people living near heavily traveled
roads. Lead originates from automotive sources via the
burning of gasoline which contains tetraalkyl lead as
an antiknock agent. The quantity of lead utilized
-------
in the United States has decreased somehwat during the last several
years due to reduced gasoline lead content and to the introduction of
the catalytic converter which requires nonleaded fuel.
In a previous program conducted by Southwest Research
Institute(2), it was shown that traffic policemen and parking
garage attendants had elevated blood leads levels as compared
with control groups of participants. The blood lead levels
were as follows: traffic policemen - 23.1 yg per 100 ml
as compared to their controls - 18.4 yg; and parking garage
attendants 28.3 yg per 100 ml as compared to their controls -
21.3 yg. The elevations in blood lead were attributed to
exposure to lead primarily from mobile emissions. The control
groups of participants were selected to match as closely
as possible the positive group of participants for age, sex,
smoking habits, and socioeconomic parameters. In general,
the study participants did match closely regarding these
parameters; however, the traffic policemen tended to have
less eduation than their controls and the parking garage
attendants were slightly younger than their controls. A
second portion of the study involved adult females living
near freeways and a control group of adult females living
away from freeways. The results showed no significant dif-
ferences in the blood lead of the two female populations,
the values being 12.9 (near freeways) as compared with
11.9 yg per 100 ml. Blood lead values for males were con-
sistently higher than for females and black participants
-------
tended to have higher blood leads than did white partici-
pants. Air lead measurements were not performed as a part
of this study.
Galke, et al. (^)determined the blood levels of lead
in 187 children, age one to five years living in Charleston,
South Carolina. The relationship of lead in soil, paint
and air was examined with regards to traffic densities
and levels of lead in the blood of the children. It was
found that the arithmetic mean blood lead level was related
to both the lead in soil and automobile traffic.
Caprio, et al.(4) examined blood lead levels of children
relative to how close they lived to a major traffic artery.
The study, reported in 1974, included some 5,226
children living in Newark, New Jersey. Over 57 percent
of the children living within 100 feet of a roadway had
blood levels in excess of 40 tag per 100 ml. Lower levels
of lead were seen for those children living further away
from major roadways. The authors concluded that residential
proximity to high traffic densities can contribute substan-
tially to lead absorption in children. Their findings indicated
that residential areas located immediately adjacent to an
urban highway exhibited higher rates of excessive blood
lead absorption in children than did households beyond 200 feet
from the street. This study did not consider lead from other
sources. Other studies of the contribution of mobile emissions to
blood leads indicate that these higher blood lead values are not
-------
usually seen as a result of lead from automobile emissions
alone. Contribution of lead from sources such as lead in
paint might have been involved in this investigation.
Daines, et al, ^ ' studied black females living near
a heavily trafficked highway in New Jersey. The study popula-
tion in this investigation lived in houses on streets paral-
leling a major highway at three distances from it. Air
lead levels were measured as were blood lead values. Mean
annual air lead concentrations ranged from 4.6 yg per cubic
meter to 2.24 yg per cubic meter from the closest to the
furthermost distances. Mean blood lead levels of the three
study groups of women in order of increasing distances from
the freeway were 23.1, 17.4, and 17.6 yg per 100 ml. Measure-
ment of lead was also performed in the air inside and outside
of the homes, in those homes with and without air conditioning,
and during times of winter and summer. The results showed
levels of lead indoors were reduced approximately 50 percent
in the wintertime when the windows were closed. Approximately
the same percentage reduction in lead from outside to inside
air was seen in homes with air conditioning units operated
during the summertime. It was noted that the quality of
the air conditioners utilized in these homes was poor.
Thomas, et al.(6) examined blood lead levels in 50
adults who lived at least three years within 250 feet of
a major freeway in Los Angeles as compared with 50 other
participants who had lived for similar periods near the
-------
Pacific Ocean. The participants living close to the Pacific
Ocean in this particular area were upwind (prevailing wind)
of significant mobile emissions. Thomas et al. reported
mean blood levels of 22.7 yg/lOOml for males and 16.7 yg/lOOml
for females while their controls were 16.0 yg per 100 ml
for men and 9.9 yg per 100ml for women. The results show
significantly higher levels of lead in the population living
near the freeway as compared with their control groups of
participants. The authors concluded that the differences
observed were consistent with coastal-inland atmospheric
and blood lead gradients in the Los Angeles basin and that
the effect of residential proximity to a freeway was not
demonstrated. The results seen in this paper do not support
the conclusion of the authors.
Waldron (7) reported on the mean blood lead levels of
41 males and 51 females living within 800 meters of a highway
interchange. In this study, blood lead levels were measured
in these participants prior to the opening of the highway
interchange, and similar measurements were made following
the opening of the freeway- The lead levels were 14.41
for males and 10.93 yg per 100ml for females during the
baseline period, 18.95 and 14.93 approximately one year
after interchange was open and two years after opening the
values were 23.73 and 19.21 yg/lOOml.
/ Q \
Jones, et al. v ' studied taxi drivers for blood lead
value using a calculation of the driver's relative lead
-------
exposure as a result of the night versus day shift and
the total mileage driven. The mean blood lead for the 50
London taxi drivers was 28 yg per 100 ml. These authors
found no statistically significant differences in the blood
lead levels related to their index of exposure.
A study performed in California by this laboratory
examined the relationship of lead from mobile sources present
in two communities, one in Los Angeles and the other in
Lancaster (9,10,11)^ Tne study area in Los Angeles was
located on the downwind side of the San Diego Freeway (traffic
density of more than 200,000 cars per day) and the second
was in the city of Lancaster, California located in a high
desert area with a population of approximately 50,000.
The two areas were selected to represent a relatively high
and a relatively low area of exposure to mobile emissions.
In each of the two areas, environmental measurements were
made for lead in ambient air, soil and tap water. In ad-
dition, measurements of lead in paint of selected res-
idences were made in the Los Angeles area. Participants
from these two areas were examined for lead in blood, urine,
hair and feces. Lead in ambient air in the Los Angeles
site averaged 6.3 yg per cubic meter while in the control
area (Lancaster), the average was 0.6 yg per cubic meter.
Considerably higher levels of lead in soil were found in
the Los Angeles site (1913.6 yg/g) than in Lancaster (66.9 yg/g)
There was also a sharp dropoff of soil lead values at the
-------
Los Angeles site when samples were examined with regard to dis-
tances downwind of the San Diego Freeway out to 300 feet. There
were no differences in levels of lead in tap water for these
sites. There were no painted surfaces with high levels of lead
within residences in the Los Angeles study area.
Participants from each of the two areas of study included
three primary age groups: children, young adults and elderly
of both sexes. Lead levels of the Los Angeles participants were
significantly higher than those of corresponding age and sex
groups of the Lancaster participants for blood, hair and urine.
Blood lead levels for all males and females were 19.3 and 14.2
yg/lOOml, respectively, in Los Angeles and were 11.8 and 9.6 yg/
100ml in Lancaster. The lead levels in feces of the Los Angeles
participants were about the same or less than the average lead
concentration of the Lancaster participants. Fecal lead measure-
ments have been shown to be useful indicators of the consumption
of lead in the diet. This study indicated that the levels of
lead in the participants living near a significant source of
mobile emissions of lead had substantially higher levels of lead
in blood, urine and hair than did their control group of par-
ticipants. It was also concluded that these differences were
the result of exposure to lead present in air.
/ 1 Q \
Snee v performed an analysis of nine epidemiology
studies on male and female adult populations, four epide-
miology studies of children, and four clinical studies of
adult humans. He calculated the blood lead-air lead relation-
ship for adults and found it to be approximately 1:1. He
-------
stated that this implies an exposure to an additional one
microgram of lead per cubic meter of air can result in an
increase of approximately 1 ug lead per 100 ml in blood.
This calculated relationship of air lead values to blood
lead is extremely important in the establishment of a standard
for lead in ambient air. Using similar calculations of
the blood lead to air lead relationship Bridbord (13) re-
ported the ratio of air lead to blood lead was approximately
1:2 rather than the 1:1 calculated by Snee.
In the air quality criteria document for lead, a summary
of the data on mobile emissions states that automobiles
produce sufficient emissions of lead to increase air and
nearby soil concentrations as well as to increase blood
lead concentrations of both children and adults. The problem
of accumulation of lead in these populations is of greater
importance when the residences are located within 100 feet
of the roadway.t1)
General Objective of this Research Project
This report describes the results of an epidemiologic
study on the impact of automobile emissions on blood lead
levels of persons of similar socioeconomic status, resident
in an urbanized area. The residents to be studied were
to live on streets with traffic densities that vary from
less than 1,000 cars per day to approximately 25,000 cars
per day. Previous studies have examined the impact on blood
lead levels of lead emissions from mobile sources for traffic
-------
densities up to 250,000 cars per day. Unlike the above-mentioned
studies, this one examined neighborhoods which probably
represent a major portion of the population in the United
States, that is, residents who live on streets with traffic
densities between 1,000 cars per day and 25,000 cars per
day. The general objective was to evaluate the contribution
of airborne sources of lead, primarily from automobile emissions,
to the environments near roadways and the contribution of
this source of lead to residents living nearby. The data
from this study should help to contribute to the body of
knowledge necessary for the Environmental Protection Agency
to assess the types of control measures needed, if any,
on avoiding possible health effects of exposed populations
in the United States.
Specific Objectives
The specific objectives of this study were to collect
the necessary data such that the following relationship
could be examined:
1. Traffic counts versus blood lead.
2. Traffic counts versus FEP levels.
3. Traffic counts versus ambient air lead levels.
4. Traffic counts versus soil lead.
5. Traffic counts versus house dust lead.
6. Traffic counts versus hand wipe lead.
7. Traffic counts versus windowsill wipe lead.
-------
These relationships would be examined for traffic
counts of 1,000 cars per day or less and up to approximately
25,000 cars per day with multiple points in between. In
addition to these primary relationships, the following mini-
studies were to be accomplished:
1. Determination of the effects of speed limits,
intersections, and distances from roadway on
correlation #3.
2. Determination of the quantities of lead present in
different particle sizes as a function of distances
from the roadway.
General Description of Project
This study was designed to minimize to the extent
possible the confounding variables on the contribution of
mobile emissions of lead to levels in the environment and
exposed populations. A major metropolitan area was selected
without significant sources of air emissions of lead other
than from mobile sources. The site selected was Dallas,
Texas which does not have any significant sources of lead
other than two small stationary sources (smelter and battery
manufacture) present on the south side of the city. Study
areas were selected to avoid possible contribution of lead
from these two secondary sources.
This investigation was composed of two primary components:
10
-------
1. Examination of the relationship between traffic
density and the level of lead in associated ambient
air.
2. Examination of the relationship of blood lead and
other household lead levels with traffic densities
on adjacent streets.
The two parts of the study were performed in the same general
area of north and northwest Dallas, Texas.
1. Ambient Air Lead Study
The air sampling program was to be performed in several
locations within the same metropolitan area in which study
participants lived. Air lead measurements were to be made
using high volume samplers with sufficient numbers of 24
hour samples collected to provide for an accurate assessment
of the air lead levels associated with different traffic
densities. Mini-studies were conducted to examine the rela-
tionship of air lead values and traffic densities at varying
distances from the roadway, and for in block as well as
between block variation in air lead levels for the same
traffic densities. Additional air samples were to be collected
to provide an estimation of indoor/ outdoor differences
of air lead levels at selected residences. The mini-studies
were also to include examination of various sizes of lead
containing particles as related to distances from roadways
11
-------
at selected traffic densities and evaluation of the effects
of speed limits on the relationship of ambient air lead
to traffic densities.
2. Exaininati£n__of_^ig_Relatjx>-nshlp Between
TrafficDejisi.ty_and Household Lead_Levels_
This part of the study was designed to examine
the relationship of traffic densities and the levels of
lead present in the immediate environment of selected
households, and blood lead and hand wipe lead level in
household residents. The study was designed such that par-
ticipants from residential units only at the ground level
and within 100 feet of the center of the roadway would be
selected. Additional restrictions included that the house-
hold would not be within 300 feet of any crossing roadway
or within 500 feet of a major roadway. The study partici-
pants to be selected were pre-school age, 25 to 45 years,
and over 60 years of age. For each age group, the intent
was to examine those participants who spent a major portion
of their time at home. Thus, for the middle age group,
the objective was to recruit women that worked at home and
in the elderly category, those individuals both male and
female that were retired or had jobs within their home.
The program included the development and design of a suitable
questionnaire for collecting the necessary demographic char-
acteristics of the study participants. The information
12
-------
was to include not only age, sex, length of time of resi-
dence, smoking habits, but also whether or not they made
use of homemade or craft pottery for culinary purposes and
any history of occupational exposure to lead.
For each of the households involved in this
study, measurements were to be made of the lead in house
dust and in the soil near the front stoop. In addition,
measurements were to be made of the closest and most distant
point of the living space to the center of the roadway.
A traffic count would also be obtained near each residence.
Paint surfaces of the residences were to be screened for
lead content to preclude the involvement of leaded paint
as a possible source of lead for young children.
All blood samples collected from the study par-
ticipants were to be analyzed for lead, free erthrocyte
protoporphyrin (FEP) and hematocrits. Hand wipes were also
to be collected from each pre-school child for lead measure-
ments. The measurements for both lead and FEP were to be
made with a rigid quality control program to include some
duplicate analysis with the Center of Disease Control.
Additional environmental monitoring was to include measure-
ment of lead in paint, in windowsill wipes and in the
tap water of each residence.
13
-------
II. METHODS
A. Site Selection
1. General Site Requirements
To accomplish the objectives of this study, a
site was required at which populations are in residence
on urban streets having traffic densities in the range
of <1/000 to >25,000 cars per day. The population living
on the urban streets must include a spectrum of ages of
both sexes from preschool ages to over 60 years. In order
to minimize interferences from extraneous parameters, the
study was designed for middle economic class neighborhoods
which are primarily white in ethnic makeup. Further, the
areas selected should exhibit a low background in ambient
air lead levels and in other identifiable lead sources
(such as drinking water).
A most critical ingredient for site selection
was the presence of sufficient members of populations living
on streets with higher traffic densities: >15,000 cars per
day. In many cases, perhaps the preponderance of cases,
urban streets with greater than 15,000 cars per day become
commercialized and have few or no residences directly on
the street front. The following characteristics were con-
sidered in determining the appropriateness of each proposed
study site: ambient lead levels; the numbers of persons living
on streets with higher traffic densities; and the age, ethnic,
and economic structure of these streets and areas.
14
-------
2. Study Sites Considered
Two metropolitan areas were considered as pos-
sible site locations for the study: San Antonio and Dallas,
Texas. Each of these cities has ambient air conditions
which are conducive to studying air lead from automotive
sources due to the lack of heavy, polluting industry. Traffic
densities and populations of both of the cities lead the
study team to estimate that both should be able to supply
the necessary populations-at-risk in the appropriate age,
race, sex, economic, and traffic-density mixes. The specific
study design called for 480 white, middle economic class
participants living in residences on streets selected for
the study, with 120 participants at each of four separate
traffic density levels:
Traffic Density Site Number of
(1000 cars/day) Designation Participants
<1000
8-14
14-20
>20
Site
Site
Site
Site
1
2
3
4
120
120
120
120
To meet the requirements of the study for 120
participants at each traffic density level, it was estimated
that a minimum of 200-300 candidate residences must be
identified for each traffic density site. To qualify,
candidate residences must be single-family dwellings or
15
-------
duplexes which lie within 100 feet of the center of the
roadway but not within 100 yards of any traffic signal
or stop sign on the roadway and must be located only on
streets with 30-45 mph speed limits. In addition, a
preference was indicated for houses which face the road-
way and not on the corner of an intersection with a
side street.
It should be noted that, in the original study
design, inclusion of 5 traffic density levels was con-
sidered: <1,000; 7,500-12,500; 12,500-17,500; 17,500-
22,500; >22,500 cars per day. These study design criteria
were submitted to the Office of Management and Budget
(OMB) with the standard information regarding requests
for permission to use specific questionnaire forms.
The OMB indicated that the design number of participants
(480) would be more properly applied to four rather than
five traffic density levels. Thus, the study design
criteria were changed to the four site designations
listed.
A preliminary siting study was performed at
San Antonio to determine if that city could support
the study with adequate numbers of residences qualified
for the study. Census tract data from the 1970 census
16
-------
of population were used to determine areas of the city
which met the basic ethnic and economic design, i.e.,
areas predominently white, middle class. From the
census data, areas indicating 70% or greater white,
non-Spanish residents were selected for on-site in-
spection. The areas selected were then inspected by
driving through the potential neighborhoods and count-
ing candidate residences. By direct inspection, the
candidate areas were judged to be acceptable or not
acceptable economically. Those exhibiting the run-down
appearance of a poverty-level neighborhood or the
extremely affluent appearance of a well-to-do neigh-
borhood were omitted from further consideration.
Due to the problems encountered in San
Antonio, a preliminary siting study was initiated in
Dallas to determine the possibility of that city
providing sufficient numbers of candidate residences,
particularly at the higher traffic densities: >15,000
cars per day. Data from the 1970 census of the popu-
lation in that city were obtained and reviewed and data
regarding estimates of current traffic densities on
main thoroughfares were obtained through the traffic
department of that city. Candidate areas meeting ethnic
17
-------
and economic requirements (70% or greater white, non-
Spanish) were selected and on-site inspections were
conducted to determine the numbers of candidate
residences available on the thoroughfares with higher
traffic densities in these areas. Results of the
preliminary siting study in Dallas indicated that more
than adequate numbers of residences would be available
in that city and its surrounding urbanized suburban
areas.
Contacts were then made with the Institute
of Urban Studies at Southern Methodist University in
Dallas and preliminary arrangements were made for a
field office and an on-site coordinator for the study
based at that University. The City of Dallas was con-
tacted and it was determined that the air pollution
control group and the traffic department of that city
were very interested in assisting SwRI to conduct a
traffic lead study. Based on these findings, Dallas
was selected as the study site. Pertinent data re-
garding justification for the change of site are
presented in Appendix A.
3. Description of the Study Site
The general site selected for this study
is the Dallas Metropolitan Area, located in north central
18
-------
Texas. The area has a mild, somewhat dry climate and
has sufficient population to support the study; 1.5
million persons reported in the 1970 census of the
population. It is a highly industrialized and com-
mercialized area with little or no heavy polluting
industries. The industrial-commercial makeup of the
area is typified by light and sophisticated industries
such as electronics, aircraft, merchandizing, and
financial institutions. An abundance of heavily
trafficked, multi-lane thoroughfares exists in the
city, with residences located immediately along many
thoroughfares. The traffic system in that city has
historically been designed around these multi-lane
thoroughfares and a network of such arteries exists
across all of the metropolitan area. A map of the
metropolitan area showing the traffic artery system
is shown in Figure 1.
For use in this study, a set of thorough-
fares was selected by use of data from the 1970 census
of the population and data regarding current estimated
traffic densities obtained from the traffic department
of the city. At the outset, the study area was des-
igned to include major portions of the north central
19
-------
Figure 1, Traffic Artery Map of the Dallas Metro Area.
-------
and northwest metropolitan areas, with some sections
in the southwestern portion of the metro area. In-
cluded in the initial design area, in addition to the
City of Dallas, were portions of the cities of High-
land Park, University Park, Richardson, and Garland.
These cities are either surrounded by the greater city
of Dallas or are immediately adjacent to Dallas city
limits. Review of the selected areas with City of
Dallas and with EPA Region VI personnel revealed the
location of a battery reclamation factory in the south
central portion of the city. The plant was a known
emitter of particulate lead and its location was such
that a possible interference was established with the
potential study areas selected in the southwest por-
tion of the city. The southwest area was eliminated
from use of the study because of potential contamina-
tion by the battery reclamation factory.
The loss of the southwest area of Dallas
presented a handicap to the study efforts. One street
in that area, Illinois Avenue, had been identified
to contain more than 140 residences on portions of
the thoroughfare with traffic densities greater than
20,000 cars per day. The loss of Illinois Avenue
21
-------
required the expansion of the study area to include
the cities of Grand Prairie and Arlington and more
areas in the city of Dallas in the east central and
northeast. Detailed on-site inspection were performed
for candidate streets in the added areas; a number of
residences qualifying for the study were counted and
a catalog was prepared showing numbers of residences
versus traffic counts for all candidate streets. A
map showing the principal areas of the metro area
which served as study sites is shown in Figure 2.
4. Authorization of Study by Local
Governments
Officials of the local governments in each
municipality included in the siting analysis were then
contacted regarding their selection as candidate study
sites. Included were:
City of Dallas
Department of Health
Department of Consumer Affairs
Department of Environmental Health
& Conservation
Town of Highland Park
City of University Park
22
-------
Figure 2. Principal Study Areas within the Dallas Mstro Area
NJ
-------
City of Garland
City of Richardson
City of Grand Prairie
City of Arlington
For each municipality, a letter was prepared which
provided general information regarding the traffic-lead
study and requested permission for performing house-
hold surveys, recruiting volunteer participants, col-
lecting biological samples, and performing traffic counts.
Visits were made to each municipality requesting a
briefing and more detailed information regarding specifics
of the study were provided.
Letters of permission to proceed were
received from each of the municipalities, including
each of the three departments of the City of Dallas.
A great deal of interest was indicated in the results
of the study from a number of the communities and
copies of our documentation upon completion of the
study, were requested. Letters of permission from each
of the communities are presented in Appendix B.
24
-------
B. Project Staff
1. Overall Project
a. Roles
Staffing of the project called for a
number of significant roles, headed by the Project Man-
ager. Organization of these roles is shown in a project
organization chart shown below.
Recruitment
Manager
Recruitment
Survey
Team
Biological
Manager
Biolc
and HOL
Samplin
Project
Manager
On-Site
Coordinator
1
Environmental
Manager
I
gical Environmental
isehold Sampling
g Team Team
Chemical
Analysis
Manager
Chem
labor
Tea
ical
atory
m
1
Data
Manager
1
Data
Entry
Personne
Traffic
Counting
Team
Project Organization Chart
25
-------
The Project Manager had total technical
and administrative responsibility for the conduct of the
study.
The Recruitment Manager was responsible
for development of survey materials, establishment of the
on-site-field office, and conduct of the household surveys.
The on-site coordinator and the Household Survey Team re-
ported directly to the Recruitment Manager during the survey
activity period.
The On-site Coordinator was responsible
for recruitment of the Survey Team, the Biological and
Household Sampling Team, and the Environmental Monitoring
Team, for management of the household survey activities,
and for coordinating activities during the biological and
environmental sampling period.
The Biological Manager had the overall
responsibility for the field sample collection operations.
His special duties were to direct the activities of the
biological collection teams in the collection and process-
ing of the biological and household samples.
The Environmental Manager had the responsi-
bility for directing the personnel involved in the air
sampling, traffic counting, paint analysis, and soil sampl-
ing operations. The principal duties of the environmental
manager were to establish daily sites for the air samplers
26
-------
and to coordinate the traffic counting activities with the
city traffic department involved. These duties also included
the maintenance and repair of the air samplers and traffic
counters. His activities in regard to the paint and soil
sampling involved monitoring the daily activities of the
personnel involved and to investigate any problems which
occurred in these areas.
The Chemical Analysis Manager was responsi-
ble for the laboratory analysis of all samples collected
and for the preparation of all data obtained from these
analyses for submission to data analysis.
The Data Manager was responsible for entry
of all data into computer processing format and for main-
tenance and analysis of the data obtained.
b. Personnel
With the exception of the On-site-coordinator,
Ms. Linda Johnson, a graduate student at Southern Methodist
University, all principal personnel involved in the study
are staff of the Department of Environmental Sciences, Divi-
sion of Chemistry and Chemical Engineering, Southwest Research
Institute.
The Project Manager and principal investigator
for the study reported herein was Dr. D. E. Johnson, Director,
Department of Environmental Sciences.
The remaining principal personnel are listed
as follows:
27
-------
Role Personnel
Recruitment Manager R. J. Prevost
Senior Research Analyst
Biological Manager
and
Chemical Analyses Manager J. B. Tillery
Senior Research Chemist
Environmental Manager J. M. Hosenfeld
Research Scientist
Data Manager K. T. Kimball
Research Statistician
2. Field Operations
a. Field Office
An on-site field office was established
on the campus of Southern Methodist University at the In-
stitute of Urban and Environmental Studies. An arrangement
was made with the Director, Mr. Bennett Miller, for the
use of office space, telephones, and a graduate student
to serve as on-site-coordinator of study activities. For
the purpose of recruitment of the on-site-coordinator and
for definition of the support required for the coordinator,
a qualifications list (Figure 3) was generated and provided
to the Institute for Urban and Environmental Studies.
A qualified student was located and es-
tablished as the on-site-coordinator for the study: Ms.
Linda Johnson, a graduate student in public administration.
Throughout the remainder of study activities in Dallas,
Ms. Johnson served as the principal coordinator of all
study activities in recruitment and field sampling and
28
-------
Figure 3.
Qualifications for On-Site Coordinator and Sample Task Descriptions
1. Title: On-Site Coordinator
2. Availability: 20 Hours Per Week
3. Rate: $5. 20 per hour (including SMU overhead rate)
4. Duration: 3-6 months
5. Office Space to store printed materials
6. Telephone: Available during normal work hours or receptionist/
secretary available to take messages for coordination
in absence of coordinator.
7. Transportation for Coordinator: Must have car; reimbursed @ $. 15/mile
8. Sample Tasks:
a. Making appointments
b. Meeting with officials to coordinate study activities and to col-
lect information
c. Picking up materials
d. Delivery of materials
e. Site surveying - counting residences in specific areas; streets
with specific traffic densities
f. Locating and interviewing survey team of 10 members
g. Help set up traffic counting equipment
h. Help set up air sampling equipment
i. Coordination of household survey of 1000 households
j. Record traffic counts
k. Collect filters from air sampling equipment
1. Help collect soil, dust, and blood samples from selected
households.
served as the point of contact with municipal agencies
and the general public.
A survey team was recruited through SMU
and through manpower placement firms by the on-site coordi-
nator. A set of recruitment and training materials was
29
-------
prepared by the project team with the help of the coordi-
nator and a candidate team was assembled for training.
A set of instructions (Figure 4) was distributed to the
survey team members and details of the study were explained.
The importance of recruitment of volunteers was stressed
in the training sessions. Persons with previous experience
in survey work were selected, where possible, and detailed
instructions in the survey procedures to follow were pro-
vided in the training session.
Survey team members were hired through
the offices of SMU or through local temporary manpower firms.
No team members were hired directly by Southwest Research
Institute. As an incentive, the survey workers were offered
payment based on the number of forms completed and delivered
to the coordinator. Two separate stages of survey activities
were performed using these methods. In the first state,
the workers were offered $2.50 per household form completed,
and $2.00 per participant form completed. A minimum of
$3.00 per hour worked up to 40 hours was established as
a minimum payment. Most workers, however, were able to
better this figure substantially. Because of difficulties
in recruitment of volunteers, the amount paid per completed
form was raised to $5.00 per participant form in the second
stage of survey activities. Instruction materials used
in training survey workers during the second stage are shown
30
-------
in Figure 4.
FIGURE 4.
DALLAS TRAFFIC LEAD STUDY INSTRUCTIONS
FOR PERFORMING HOUSEHOLD SURVEYS
Schedule of Activities
Friday, May 21 Training Session
Saturday, May 22 Survey Begins
Tuesday, May 25 Turn in Completed Forms and Review Progress
Thursday, May 27 - Turn in Completed Forms and Review Progress
Saturday, May 29 Survey Complete
Tuesday, June 1 All Forms must be turned in by this Date
Rate of Pay
$5. 00 per participant form completed
$2. 50 per household form completed
If not enough forms are completed to yield $3. 00 per hour worked, then $3. 00
per hour will be paid. However, a good worker will be able to complete enough
forms to always make more than this minimum rate.
Work Assignment
You will be assigned approximately 60 residences by address, along one or more
busy streets. These residences have been carefully selected to meet certain re-
quirements of the study. The work assignment is designed to take a normal 40
hour work week. You should have no trouble completing the assignment in one
week of work. You may work any hours and days which you choose. Some work-
ers have better success on weekends. You must complete the assignment and
turn in all forms on or before Tuesday, June 1.
Early Completion
If you complete your assignment early, so much the better. You will be paid for
the number of forms delivered. A goal should be to have all work completed by-
Friday, May 29. The sooner you complete your work, the sooner we can process
the paper so you can be paid.
Each survey worker was provided with a
specific assignment area for which he was responsible.
Approximately 60 residences were located in each assignment,
and the worker was asked to complete at least 50 household
forms within one week.
31
-------
b. Sample Collection
(1) Biological Collection Team
(a) Recruitment
The biological collection teams
were composed of two individuals. One person was required
to be a laboratory technician with recent experience in
drawing blood from children. The other team member was
not required to have any special technical skills but a
friendly attitude and outgoing personality were qualities
necessary for the job activities. These people were all
hired through a temporary employment agency. Some of the
laboratory technicians hired for this project were already
working in a clinic or hospital (night shift or part .time)
specifically drawing blood from young children. The other
technicians were not presently working at their profession
but had recently been so. The longest time any laboratory
technician had been away from this type of work was 6 weeks.
All of the laboratory technicians were female and judged
mature enough to perform the required tasks without direct
supervision.
The second member of each
biological collection team was a young woman between 18 and
22 years old. Most, if not all, were undergraduate students
at local universities. These women were selected for this
project based upon their demonstrated maturity and personality.
32
-------
(b) Training
The week prior to the actual
collection of samples, a training session for the biological
collection teams was held on the SMU campus. The main purpose
of the training session was to familiarize the laboratory
technicians with the care necessary in taking blood samples
for Pb analysis. (14) since contamination (especially with
the finger-prick technique) is a major concern in the collec-
tion of blood samples for Pb analysis, special emphasis
was placed upon the modifications of the routine collection
procedures to minimize this problem. Other points covered
during the training session included: (1) what information
was to be collected and how it was to be recorded, (2) how
and where to collect the household samples, (3) procedure
for taking handwipe samples from children, and, (4) procedures
to follow during medical emergencies.
It was necessary that at least
one team member have transportation available. Time was
allotted during the training session for everyone to meet
and to pair off as teams such that the transportation problem
would be solved and there would not be any personality
conflicts between team members.
(2) Environmental Sampling Team
(a) Recruitment
Collection of environmental
samples required two individuals for the air sampling and
33
-------
traffic counting, one person for the paint analysis and
one person for the soil sampling. No specific technical
Skill was required for these individuals but a steady job
record along with demonstrated maturity were essential
qualities necessary.
(b) Training
Team members were trained in
the tasks they were to perform by the environmental manager.
For the most part, this consisted of on-the-job training
during the first week of field sampling. The individuals
selected for these tasks were all male college students
between 20 and 24 years old.
The most difficult job assign-
ment, i.e., the task requiring some degree of technical
know-how was the operation of the portable x-ray analyzer
used to measure Pb content of paint in the participant's
homes. The individual selected for this particular job
received several hours of instruction and observation with
the instrument before he was judged qualified to obtain
accurate data with the instrument.
Instruction, demonstration,
and observation by the environmental manager and soil chemist
(subcontractor) were used to qualify the individual collecting
the soil samples.
34
-------
C. Determination of the Relationship Between Air Lead
Levels and Traffic Flow Characteristics
1. Design
a. General Outline
The determination of air lead levels and
traffic flow can be correlated using a variety of mini-
studies. While each ministudy as described below was complete
in itself, their contribution to correlating the traffic
and air lead values should not be underestimated.
The basic study involved a continuous twenty-
five day sampling period during which traffic and air lead
values were recorded. Streets with various traffic
densities were monitored that included rates of less than
1,000 vehicles/day; 5,000; 10,000; 15,000; 20,000; and
25,000+ vehicles/day. Supplementing this basic study
was the determination of the effects of speed limits, inter-
sections and distances from the roadway. How the particle
size varied as a function of distance from the roadway was
also examined. Another ministudy determined the accuracy
of collection times less than 24 hours. Studies were also
performed to compare indoor and outdoor lead values of
streets with 10,000 and 25,000 vehicles/day- In contrast
to the above studies which used active collectors such as
high volume samplers, one ministudy used a passive collector
for settleable particulates.
35
-------
In defining the lead particulate
pattern from highways, it was important to eliminate or
reduce as many confounding variables as was practical and
so all of the air samples collected in this study had a
number of common points between them. Figure 5 illustrates
these common points. With the exception of those samples
Figure 5. General Criteria for Placement of High
Volume Particulate Sampler.
>25,000 cars/day
<25,000 cars/day
>150m
HI VOL SAMPLER
>90m
TRAFFIC
COUNTER
15m
t
WIND
DIRECTION
I
taken for the distance and intersection studies, all
samplers were placed in the middle of the block and 15
meters from the roadway. The samplers were approximately
1 meter above the ground on the same plane as the roadway;
and unless the wind was parallel to the roadway, the
samplers were placed on the downwind side. The selec-
36
-------
tion of sites for air sampling was also based on the cri-
teria of speed limit, i.e., between 30-45 mph, and distance
from another major roadway, i.e., no closer than 150 meters.
A major roadway for this study was defined as 25,000 cars/day
or more. However, for those cross streets having less than
25,000 cars/day the minimum distance was 90 meters. During
the period that the samplers were collecting particulates,
meteorology data was also being collected. All sites chosen
for air lead sampling were distant from any other known
sources of lead. Sampling was performed only during periods
of dry weather and with the exception of the 25-day study,
only on weekdays.
b. Traffic Density Mini-study
During the time period that biochemical
monitoring of the participants took place, an in-depth
air sampling program was also performed. Begun approxima-
tely one week before the biochemical monitoring, air samples
were collected at six different traffic densities for 25
days. Traffic counts were conducted at each site where
air sampling occurred. Streets in each of the following
traffic densities were monitored: 1,000; 5,000; 10,000;
15,000; 20,000; and 25,000 cars/day. Rather than sampling
at just one site for each traffic density, three streets
in each of the six densities were monitored. The streets
were chosen in different sections of the city within the
f
37
-------
boundaries of the study area. The angle of the street in
relation to the wind direction was also varied. Because
of this wind direction variation, a set of three streets
with the same traffic density was not monitored at the same
time. While the monitoring of the streets for a specific
traffic density was not purposefully randomized, the task
of locating streets as other streets were being sampled
created its own randomization. Each of the three streets
studied for a given traffic density was monitored for 8
consecutive days plus an additional day at one site to meet
the 25-day sampling requirement. Each street thus monitored
had at least one weekend of sampling. Therefore, for the
entire 25-day study at each of the six traffic densities
having three streets each, a total of 150 air samples/
traffic counts were collected.
c. Replicate High Volume Ministudy
A large number of air samples were collected
throughout the air monitoring studies of this project.
These samples were collected using single high volume
samplers deployed in the various sampling schemes described
previously and also following this subsection. Since single
samplers were used, it was important to determine the
accuracy and precision that any one sampler might exhibit.
A study was thus performed to determine these variables.
38
-------
Two high volume samplers were placed about
one meter apart from one another. These samplers were
placed at one of the locations sampled during the twenty-
five day study and followed the common point criteria men-
tioned above. Simultaneous air sampling using the same
two samplers took place for 10 consecutive days. Traffic
counting was also performed during these sampling runs.
d. Distance from Road Mini-study
Sites selected for this study required an
area that was unobstructed for a minimum of 40m back
from the roadway on the downwind side. (Figure 6). Four
Figure 6. Distance from Road Mini-Study.
30m
15m-
7.6 m.
1.5 m.
HI VOL SAMPLER
TRAFFIC
nCOUNTER
TRAFFIC
COUNTER
WIND
DIRECTION
39
-------
samplers were set up at 1.5, 7.6, 15 and 30m from the
roadway in the middle of the block. A traffic count
was initiated and all samplers were turned on simultaneously.
Two 24-hour sampling periods were recorded for each of the
following traffic densities: 5,000, 15,000, and 25,000+
cars/day.
e. Particle Size as a Function of
Distance from Road Mini-study
These samples were collected at the same site
as the distance from the roadway study (25,000+cars/day
only). An Andersen high-volume particle-sizing collection
head was used to collect and separate the particles. This
collection device separated the particulates into five
aerodynamic size ranges, by particle diameter: 7 microns
or larger; 3.3 to 7 microns; 2 to 3.3 microns; 1.1 to 2.0
microns and 0.01 to 1.1 microns. Samples were taken at
each of the four distances as mentioned in the distance
study using the Andersen sampling head.
f. Intersection Mini-study
Five different intersections with varying
traffic densities were studied. These intersections were:
25,000 vs 10,000; 25,000 vs 5,000; 15,000 vs 1,000; 10,000
vs 5,000; and 10,000 vs 1,000 cars/day streets. Each of
these combinations was studied for four weekdays. One
sampler was set up on the downwind corner 15 meters from
40
-------
both roadways (Figure 7). The second sampler was set up
at mid-block on the principal roadway and each roadway had
its own traffic counter. With the aid of radio communica-
tions, the samplers were started simultaneously and the
traffic count taken.
Figure 7. Intersection Ministudy
TRAFFIC
COUNTER
HI VOL
15m"
15m
TRAFFIC
n COUNTER
WIND
DIRECTION
HI VOL
15m
With the addition of corner homes to the
study to increase the potential number of participants, an
expanded intersection study was initiated. This dealt with
the effect that side streets of less than 1,000 cars/day
had on the lead levels of the principal street. Samples were
set up on the corner and midblock of the principal street as
41
-------
described above and run for 24 hours. At the end of this time
period the samplers were physically interchanged and the second
sample collected. The samplers were switched to prevent
any inherent bias in the collection. Traffic counts also
were recorded on each street. Samples were taken at in-
tersections of less than 1,000 cars/day vs 1,000; 8-14,000
vs 1,000; 14-20,000 vs 1,000; and 20,000+ vs 1,000 cars/day.
Only four-corner intersections having stop signs on the
side streets were selected.
g. Speed Limit Mini-study
Two streets were chosen that had
approximately the same number of cars per day. However,
one street had a speed limit of 30 MPH while the other
had a 45 MPH speed limit. Air samples and traffic counts
were collected at each site for five days.
h. Indoor vs. Outdoor Mini-study
Ten residences of participants selected
for biochemical monitoring were also selected for indoor
air sampling. Five residences were chosen from a 25,000+
car/day street and five from 10,000 car/day street. The
indoor sampler was usually placed in the living room. These
samplers were of the same type as used outside, although
the ones that were cleanest and with the best appearance
were placed inside the home. If the homeowner objected
to the living room placement, then another room such as
a bedroom was chosen that was the same distance from the
42
-------
roadway as the living room. The second sampler was placed
outside of the home. Each sampler was started simultaneously
and a sample collected for 10-12 hours during the daytime
(usually from 8-9 a.m. to 6-7 p.m.). Flow rates for both
indoor and outdoor samplers were approximately 50 cubic feet
per minute. Two days of sampling were conducted at each
of the selected residences.
i. Dustfall Mini-study
Ten locations were selected throughout
the study area to collect settleable particulates for 28
days. The number of locations assigned to specific sectors
of the study area approximated the density of participants
anticipated for that sector. Maps with the ten locations
are illustrated in Figures 8, 9, and 10.
43
-------
CITY OF DALLAS
WESTERN SECTION
SCALE: 1" = 0.6. MILES
* * HIVOL SAMPLERS
= DUSTFALL COLLECTORS
8
44
-------
CITY OF DALLAS
NORTH CENTRAL SECTION
SCALE: 1" = 0.6 MILES
* = HIVOL SAMPLERS
=DUSTFALLCOLLECTORS
"igure 9
45
-------
DALLAS FORT WORTH TURNPIKE
CITY OF ARLINGTON
SCALE: V = 075 MILES
= DUSTFALL COLLECTOR
Figure 1C
46
-------
j. Collection Time Less Than
24 Hours Mini-study
Twenty-four samples were collected under
the common point specifications described above except that
they were collected for time intervals less than 24 hours.
Time intervals of 1, 2, 4, 12 hours were used at traffic
densities of 10,000, 15,000 and 25,000 cars/day. The col-
lection scheme consisted of four samples placed side by
side and equidistant from the roadway (Figure 11). All
four samplers were started at the same time as was the
Figure 11. Collection Times Less than 24 Hours
Ministudy.
HI VOL SAMPLERS
1234
Fol [ol fol [o\
t
15m
1
TRAFFIC
n COUNTER
WIND
DIRECTION
traffic counter, usually between 8 and 9 A.M. When the one
hour time interval came up the traffic count was noted and
sampler 1 was turned off and the filter removed. At the
next time interval of two hours, the cumulative traffic
47
-------
count was noted and sampler 2 was turned off. This sequence
continued until the last collection was terminated 12 hours
later. The following day, the sequence was repeated at
the same location. Two sampling days were also performed
at each of the remaining two traffic densities.
2. Sample Collection
a. Traffic Volume
A successful traffic counting operation
was vital to the purpose of this contract. To insure that
accurate and meaningful counts were obtained, coordination
efforts were set up and maintained with the traffic and
safety offices of the following governmental agencies
located in the State of Texas:
City of Arlington
City of Dallas
City of Garland
City of Grand Prairie
City of Highland Park
City of Richardson
City of San Antonio
City of University Park
Texas Department of Highways
The purpose of the coordination efforts
was twofold. The necessity of informing these different
agencies of the anticipated traffic counting operations
48
-------
that would be conducted in their jurisdictional areas is
obvious. Permission to conduct the surveys was then
necessary before any type of activity cound be performed.
At this point, it should be mentioned
that the data gathered under this contract would not have
been possible without the excellent help and advice of the
governmental agencies listed above. Instructions as to
the proper placement of the traffic counting devices, loan
of counting equipment, availability of previous traffic count
data, donations of detailed city maps, all of these items
contributed by the agencies listed above played an important
role. Our gratitude to them is humbly offered.
Following contact with the various
cities on whose streets the counting would be performed,
a review of all published traffic count data was made to
identify potential streets and areas for the counting survey.
The streets were categorized based on their traffic count,
i.e., less than 1,000 cars per day; 5,000; 10,000; 15,000;
20,000; and 25,000 cars/day and above. With the streets
thus pigeon-holed, a culling of the unacceptable streets
was based on availability of potential volunteer residences.
A visual on-site survey of the tentatively selected streets
was then made and the list of potential streets was further
refined.
The apparatus used for traffic counting
was mainly the Traficounter Junior (Streeter Amet, Grayslake,
-------
Illinois) although others such as Models RCH and MR from
the same manufacturer were also used, but to a lesser extent.
The Traficounter Junior is a digital counter which records
axle count, while the MR and RCH counters print the axle
count hourly on a paper tape. Other than this output dif-
ference, the operation and set-up of all counters was the
same.
The streets for which traffic counting
was required were inspected to determine proper placement
of the counters. Instructions received from the Texas
Highway Department, training by the City of San Antonio,
and installation instructions from the equipment manufact-
urer provided the basis for traffic counter placement.
Basically, a minimum distance of 50 meters was required
from any cross street or intersection. Both ends of the
road tube were attached to the pavement with concrete nails.
One end of the tube was plugged and the other end was at-
tached to the counter. A two-lane street was counted with
one road tube and one counter since the counter was capable
of an accurate count over both lanes even with the opposite
traffic flow. On the streets included in the study, there
were no three-lane streets. Four lanes of traffic required
two counters, one for each direction of travel and where
possible, the counters were placed on a center median and
the road tube was extended to the outside lanes. The same
50
-------
set-up was used for 6 lanes of traffic with one counter
for each direction of travel.
After the road tubes were in place
and the counter turned on, a validation procedure was
begun. For each counter, fifty cars were counted that
passed over the road tube. This figure was compared with
the value of the mechanical counter.
On a two lane street with one lane of
traffic in each direction, occasionally two cars would
activate the road tube simultaneously. The result was
that three axles were counted instead of four as occurred
when two cars crossed the road tube independently- Similar
missed counts occurred when two or more lanes of traffic
were flowing in the same direction. In this case when
two cars were traveling side by side and crossed the road
tube only two axles were counted instead of four. A cor-
rection factor was developed for each traffic count that
would take into account these simultaneous crossings. This
correction factor was based on the ratio of cars counted
mechanically versus the visual count. The twenty-four hour
count was then increased to reflect this ratio. A similar
correction factor was also developed for multi-axled vehicles
such as dump trucks or somi-trailer trucks. During the set
up of the traffic counter(s) on a street, a visual count
was made of multi-axled vehicles. This count, which usually
51
-------
lasted for an hour was then applied to the final traffic
count in a manner similar to the simultaneous crossing
factor. However, the truck factor decreased the total
mechanical count because while the total axle count in-
creases, only one vehicle passed over the road tube. With
the exception of those streets over 15,000 vehicles/days,
the truck correction factor decreased the twenty-four hour
count on the average of 5 to 20 vehicles. On streets above
15,000 vehicles/day, the total count decreased an average
of 15 to 35 vehicles.
At each location where traffic counting
was performed, a data sheet was completed (Figure 12).
The form had spaces for the counter machine number, set-up
time and count, removal time and count, day, date, visual
vs machine count, and a place for remarks. The set-up and
removal time interval was maintained as close as possible
to twenty-four hours with a one-hour difference being the
exception. In the majority of cases, the equipment per-
formed as it was designed. However, occasional problems
included loose or broken road tubes, and tubes removed from
the counter by vandals.
b. Air Particulates
The collection of air samples for the
various studies described above was performed to define
the lead particulate patterns near streets with varying
52
-------
Figure 12.
AIR SAMPLING i TRAFFIC COUNTING
Dallas, Tuxas - 1976
Project 01-4a94
Street
Sec Uo
Direction
No.
>I Sam ale r No .
Rotam eter (cfm)
Hour Dav Date
val
D
ige
volume of air sampled
Total X Calibratic
Factor
Traffic Counter
Machine No .
Counter Readings
Hour
Day
Date
Removal
Set Up
Total
Visual/Machine Count:
Axle factor:
24 hour weekday total =
vehicles
Remarks:
53
-------
traffic densities. For each air sample that was collected,
a traffic count was recorded for that same collection in-
terval.
The equipment used to collect the
air samples were high-volume particulate samplers of standard
design. Equipment from three different manufacturers
(Staplex, BGI and General Metal Works) was used on this study.
All high-volume samples used in this study used the same flow
rate, 50-60 cfm. Basically, each consisted of a blower with an
8" x 10" holder for the filter. A weatherproof unit composed of
wood or aluminum, depending on the manufacturer, protected the
filter and the motor from the weather. Prior to sampling opera-
tions in the field, each hi-vol sampler was calibrated with
a series of resistance plates. A secondary calibration
curve was then developed for each sampler based on the primary
calibration curve for the resistance plates. Each sampler
was thus indirectly comparable to other samples used in the
study since all were calibrated using a common source.
Additional comparisons were made when duplicate simultaneous
samples were taken using hi-vols placed next to one another.
Glass fiber filters, Type A (20.3 x 25.4 cm),
without an organic binder (Gelman Instrument Co.) were used
for the particulate collection media. A light table was
used for visual inspection of each filter for thickness
variation and pinholes. Filters passing this test were
then sequentially numbered and separately placed in clean
54
-------
polyethylene bags and sealed. Those filters used for the particle
size distribution study were conditioned at a constant temperature
and humidity until a constant weight was achieved and then the
filters placed in separate bags and sealed.
c. Dustfall
Ten locations were selected throughout the study
area to collect settleable particulates. Locations and times when
hi-vol samplers were operating nearby were avoided. The open-top
collectors were set out for 28 days and were patterned after ASTM
designation D 1739-70. Specifically the collection apparatus
consisted of an acid-washed polyethylene bucket with a polyvinyl-
chloride extension tube. The dimensions were such that the height
of the bucket/extension tube was three times the diameter of the
opening in the extension tube. The device was mounted atop a
2.5m pole and secured with guy wires. A maximum 30-degree angle
was observed from the top of the collector to the nearest obstacles
such as trees or houses. On top of the collection tube was a bird
ring. It was placed so that if birds attempted to land on top of
the tube, the ring prevented this from happening. Thus, any possible
contamination from birds was prevented.
Following the 28-day collection period, the
device was removed from its perch and covered. The sample was
transferred to an acid-washed polyethylene bottle with multiple
deionized water rinses. The sample was labelled and frozen.
55
-------
3. Sample Analysis and Quality Control Procedures
a. Development of Analytical Methodology
Studies were performed to evaluate the analytical
methods for determining the lead concentration in air particulate
samples collected on 20.3 x 24.4 cm glass fiber filters.
A single 2.5 x 20.3 cm strip was used from each
filter (9.8% of total surface area) for lead analysis. To determine
the variability of lead concentration over the total filter, seven
2.5 x 20.3cm strips were cut from a single glass fiber filter sample
and analyzed according to the procedure for air particulate samples.
The results are given in Table 1 under precision of data.
Table 1. Analytical Parameters of Environmental
Samples
Sample Matrix
Sensitivity
Detection Limit '
Linear Range
Recovery
Precision:
n
mean
std. dev .
RSD
\ * i
Air Particulate
0.019 (ig/m3
0.0 21 fig/m
3.38 jj-g/rn
96.3%
7
1.44 ug/m3
0.19 >ig/m3
13.2%
Outdoor Dust '
0.0006 ^g/cm
0.0058 ug/cm
0.070 (ig/cm
96.2%
7
0.035 jig/cm2
0.003 ng/cm2
8.3%
Soil10'
0.95 pg/g
1.02 |jLg/g
150 pg/g
90.0%
4
5.90 (jg/g
0.44 (ig/g
7.4%
(1) see text for definition
(2) does not amply max inum linear range
(3) based upon average recovery of low Pb spite in sample
matrix
(4) calculations based on 2000 m3 sample
(5) surface area of 179.07 cm2
(6) based on 5g sample
56
-------
At a mean lead concentration of 1.44 yg/m3, the relative
standard deviation was 13.2%.
While performing the field studies,
several air samplers were run simultaneously in pairs to
obtain some indication of the variation due to the air
samplers. Section IIIA2 summarizes the results of this
study. There was no significant variation in the lead
concentration of filters collected simultaneously.
The digestion time of air filters
was investigated to determine if this would be a critical
step in the procedure. Replicate samples were digested
for 3 hours and for 24 hours according to the methodology.
The 3-hour samples' average lead concentration was 6.6 yg/m3
(RSD = 8.2%) and the 24-hour samples' average lead concentra-
tion was slightly higher at 7.2 yg/m3 (RSD = 8.6%). This
slight difference between the 3-hour and 24-hour digestions
was not significant.
Table 1 summarizes the analytical para-
meters for air particulate samples.
The detection limit is defined as that
quantity of lead which will give a signal 2X the standard
deviation of a series of spiked samples whose lead signal
is distinctly above the background signal. All values have
been converted into the appropriate unics for each sample
matrix.
57
-------
Sensitivity is defined as that quantity
of lead in the digested sample matrix which will give 1% absorp-
tion.
In summary, these investigations show
that low temperature ashing of the glass fiber filters prior
to digestion was not necessary. The difference between
the ashed and non-ashed filters was within the 13.2 varia-
tion recorded for different samples from the same filter.
Digestion of the air filters was complete
within 3 hours and longer digestion times were not needed.
Since there could also be variation in
the collection process, high-vol air samplers were run in
pairs on several occasions to determine the range of varia-
tion due to the samplers themselves. There were no signi-
ficant variations noted for any of these paired samplers.
The analytical methodology used for
air particulate lead analysis was simple and required mini-
mum sample handling. This allowed for less contamination
of the samples and better data. The time involved in evap-
orating off the digestion acids was rather lengthy, but
the addition of a flow of N2 reduced this time considerably.
For samples with high lead concentrations,
the method was easily adaptable to dilutions to maintain
the lead concentration in the linear working range of the
AAS.
58
-------
b. Air Particulate
The procedure of Thompson, Morgan and
Purdue (l->) was modified and used to analyze the air par-
ticulate samples for lead.
The 20.3 x 25.4 cm glass-fiber filter
was removed from its polyethylene bag and placed on a poly-
ethylene sheet. A plexiglas template and stainless steel
surgical knife were used to cut a 2.5 x 20.3 cm strip from
the filter. The remainder of the filter was returned to
its polyethylene bag.
The strip of filter removed was carefully
cut into 1-cm lengths and placed into a 2.5 x 8.5 cm Pyrex
extraction thimble using Teflon-coated forceps. Eighty mil-
liliters of digestion acid (16 ml of redistilled HC1 and
64 ml of redistilled HNC>3) was added to the specially con-
structed boiling flask (Figure 13) and the extraction thimble
carefully lowered into its neck. The flask was then connected
to the condenser unit and heat was applied. Once the acids
began to reflux and wash over the filter strips, the tem-
perature was adjusted such that a continuous refluxing occurred.
Refluxing was continued for 3 hours before the heat was
removed and the digestion acids allowed to cool. Several
milliliters of 0.1 NHNO^ were poured through the top of the
condenser and allowed to drain into the boiling flask.
59
-------
Figure 13. Apparatus for the Acid Digestion of Air, Window Sill Wipe,
and Hand Wipe Samples
Dust Cover
Water Out
Water In
34/45 Groundglass Joint
A Ahlin-type condenser
B Extraction vessel with course glass disk
C Specially constructed boiling flask
D Hot plate
The boiling flask was removed from the
condenser unit and the extraction thimble removed. The
boiling flask was placed on a hot plate (30QOC) and a flow
of N2 added to assist in evaporating the digestion acids.
Once the volume was reduced to several milliliters (never
60
-------
to dryness) the flask was removed from the hot plate, cooled,
and quantitatively transferred to a graduated centrifuge
tube (15 ml). The sample volume was made to 5.0 ml with
deionized water before the sample was centrifuged for 30
minutes at 2000 RPM.
A portion of the filtrate was decanted
into a polyethylene vial (5-ml) taking care not to disturb
the precipitate present. One ml of this solution was pipetted
into a 10-ml volumetric flask and made to volume with de-
ionized water. This diluted solution was then analyzed
for Pb by aspiration into an air-acetylene flame AAS.
Table 2 gives the analytical parameters use for the AAS
determination of air and other sample matrices.
Table 2. Ana.lyti.cal Parameters lor Atomic
Absorption Spectrophotometric
Lead Analysis
[nsfrument
Pa ran1 ^^ ; r
ENVIRONMENTAL
Air
Soil
Outdoor
Dust
HOUSEHOLD | BIOLOGICAL
Water
Indoor
Dust
WmdowSdl| Veneous
Wipe ; Siood
Capillary 1 Hand-
Blood 1 wipe
233. 3 am
Slit
~ o u r c ? L, ^ r r e n t
Atomizatton F'.anie Flame Flame Flameless FlameLesa Flame Flameless Flameless^' Flame
'.l^t:or. Factor 10
() Flame = air/acetylene
Flamclesa = grapnue tube furnace (HGA-2000).
Ul F'-ar.ieless = graphite tube furnace (IL-455).
61
-------
Air particulate and outdoor dust samples
(soil also) were quantitated for Pb content by analyzing
a series of appropriately spiked sample matrices using the
"method of additions" technique to establish an analytical
curve (peak height vs Pb concentration). Analytical curves
were calculated on a Hewlett-Packard Model 9810A program-
mable calculator using a least-square regression program
to obtain the best fit to the data points. Figures 14 and
15 are typical analytical curves used to determine the Pb
content of air and outdoor dust samples, respectively. Spiked
sample standards were routinely analyzed with every 15 to
20 samples. This allowed a continuous upgrading of the
analytical curve used to quantitate the samples.
Air samples were diluted 1:10 to keep
their concentration within the linear range of the instrument.
Quality control samples consisted of spiked
samples of the appropriate matrix analyzed routinely with
the unknown samples. At least 4 quality controls were used
to determine a daily recovery factor which was applied to
the unknown samples.
c. Dustfall
Outdoor dust samples were thawed and
transferred to a 250-ml Vycor beaker. The sample container
was rinsed with several milliliters of concentrated HN03
and these rinsings added to the sample. Then, 20 ml of
62
-------
concentrated HN03 were added to the sample and it was placed
on a hot plate (150°C). A raised watchglass was placed
on the beaker to keep out contamination while the same was
digested-evaporated. The sample was never allowed to boil
Figure 14. Analytical Curve for Lead in Air Particulate
300-
15
63
-------
Figure 15.. Analytical Curve for Lead in Outdoor Dust
rH
0)
0)
cu
180 L_
160
140
1ZO
100
80
60
40
20
5.0
J_
10.0
15.0
Jig
and when 1 to 3 ml remained, it was removed from the hot
plate and allowed to cool.
64
-------
The sample was quantitatively transferred
to a graduated centrifuge tube (12-ml) and made to 5 ml
with deionized water. After centrifuging for 30 minutes
at 3000 RPM, the supernatant was decanted into a small vial
without disturbing the silica precipitate. This solution
was then analyzed for Pb using flame AAS.
Quantitation and quality control of the
outdoor dust samples has been described above.
D. Determine the Relationship of
Blood Level and Traffic
1. Description of the Study
a. Study Design
The study described herein was designed
to determine the relationship between blood lead and traffic
density. To that purpose, a set of 442 volunteer partici-
pants were recruited who resided on streets with traffic
densities varying from less than 1,000 cars per day to greater
than 20,000 cars per day. Only persons who routinely were
to be found at home and who did not have occupations which
routinely took them away from their residence were selected
to participate in the study. From each participant, blood
samples were obtained as were other samples from the residence
(dust, soil, water) and these samples were chemically an-
alyzed regarding the amount of lead present. Traffic den-
sities were measured for the street on which each partici-
pating residence was located. A detailed statistical analysis
65
-------
was then performed on the numerical results of chemical
analysis and traffic counts to determine the relationship
between these parameters.
b. Data to be Collected
(1) Demographic
The study was designed around a
population of participants who can be generally character-
ized as white, middle class, who routinely spend their time
at home most of the time and who do not have unusual exposure
to lead. To obtain demographic information on selected
households and on the residents of selected households,
appropriate questionnaire forms were designed. The study
design was to collect demographic information with question-
naire forms on 1000 households and on 480 individuals re-
cruited and selected as voluntary paid participants.
(2) Environment a].
On each street and residence selected
for the study, study design included collection of a set
of environmental data. This included obtaining traffic
counts on each street selected and collection of a set of
environmental samples at each residence: soil samples, tap
water, house dust, and windowsill wipes. In addition a
measurement of the amount of lead in surface paint and a
handwipe(from pre-school children) was obtained from par-
ticipating residences.
66
-------
(3) Biological
From each study participant the study
design included collection of a blood sample for analysis
of lead content. For adults and larger preschool children,
this was accomplished with a venipuncture. For smaller
preschool children, the required blood sample was collected
with a finger prick.
2. Data Collection Procedures
a. Demographic
For use in the household surveys, a house-
hold questionnaire form was developed in two parts, one
directed to information on households, including all persons
in the household collectively, and a second part directed
to information on individuals who might serve as study par-
ticipants. Appropriate questions were formulated to obtain
the information sought on households and on individuals.
This questionnaire as well as supporting material was then
sent to OMB for approval (see Appendix C).
Clearance for development of a new form
was obtained from OMB, and an OMB number was assigned (0MB-
158-575022 with an expiration date of February 1977). A
number of changes and improvements to the draft form shown
in Appendix C were made by EPA and OMB in the process of
obtaining OMB clearance. These are reflected in the final
format of the form, shown in two parts: Traffic Lead House-
67
-------
hold Questionnaire (Figure 16) and Traffic Lead Indivi-
dual Questionnaire (Figure 17).
INTERVIEWER USE ONLY
Interviewer ID *
Household ID tt
Estimated Traffic Level I
OMB #168-375022
OFFICE USE ONLY
Expiration Date February, 1977
'p*i i i i 1 D
Cols. 1234 5
TRAFFIC LEAD
HOUSEHOLD QUESTIONNAIRE
NAME:
Last Name
ADDRESS: Street_
City
Middle Initial
Zip Code_
Talephone_
How long have you and your household lived at this address?.
Does any member of your household routinely spend a portion of most
days away from the home?
Yes No Total Number
If Yes, specify by name_
Then, are there any other members of your household who routinely spend
most of their time at home?
Yes ___ No Total Number
What are the ages, sexes, and family position jf these members?
Name
Age
Sex
Position
Total Number Meeting Criteria
During the time in which your household has lived at this address, has
any household member been employed as:
(a) garage mechanic
(b) road maintenance worker
(c) solderer or welder
(d) shipyard worker
(e) battery reclamation plant worker
(0 electrical parts assembler
(g) plumber
(h) ore smelter worker
Yes
No
If Yes, complete the appropriate sections of the matrix.
6. Are any of the following articles used in storing, preparing, or serving
food in your household?
(1) Unglazed Mexican type pottery
(2) Glazed Mexican type pottery
(3) Hand painted china flatware
(4) Any combination of (he above
(5) None of the above
9-28
29-43
49-50
D"-
Name of
Member
No. Years
Employed
Dates of
Employment
Total
No.
54
55
56
57
58
59
60
61
Figure 16. Traffic Lead Household Questionnaire.
68
-------
10.
11.
12.
13,
Has any member oi" your household been screened for excess lead
absorption?
(1) Yes (2) No
Has any member of your household been diagnosed as having excess
lead absorption?
(1) Yes (2) No
If Yes, specify members by name
9. Is your home cooled with any of the following appliances?
(1) Central air conditioning (5) Ceiling exhaust fan
(2) Window air conditioning (6) Other
(3) Evaporative cooler
(4) Window fan
What type of structure is your house? (Sl'~o or more of exterior
surface)
(I) Sc'id brick, concrete, or rock (6) Composition siding
(2) Brick or rock veneer (7) Wood frame
(3) Stucco (8) Other
(4) Asbestos shingle
(5) Aluminum siding
Wlut is the approximate age of your house? .
_ years
What is the highest educational level completed by your head of
household?
(1) less than 8 th grade (5) trade or vocational schoot
(2) Sih grade beyond high school
(3) lugh school - incomplete (6> college (4 years) - incomplete
(4) high-.chool-complete {7) college (4 years) - complete
(8) post graduate
Would you or any of your family members participate in a health survey as
a paid volunteer?
Yes No
If yes, specify members by name ^
Total Number of Participant Forms Collected
D
D
64
65
70
INTERVIEWER NOTE AND RECORD
14. Width of road bv number of lanes.
IS. Is street a divided highway?
(I) Yes (2) No
16. Estimated distance ot' residence from center of roadway m leet.
17. Is residence at an intersection?
(1) Yes (2) No
18. Is residence facing primary street?
(1) Yes (2) No_
n
n
D
n
71
72
73-75
76
77
19. On which side of street is house located?
ll) East 12) West (3) North (4) South
D
Interviewer's Initials.
H 80
Figure 16. Traffic Lead Household Questionnaire (Cont'd.)
69
-------
INTERVIEWER USE ONLY
Interviewer ID &
Household ID *
OMB # 158-S75022
OFFICE USE ONLY
Expiration Date February, 1977
ID #|_
Cols. 1
Mill
2345
TRAFFIC LEAD
INDIVIDUAL QUESTIONNAIRE
NAME:
Last Name
6-20
Sex: (I) Male_
Date of Birth:
First Name
21-32
(2) Female
Middle Initial
33
DOB
Month
Day
Year
How many years have you lived in this city?
4. How many years have you lived at this address?
ACE IN YEARS
years
years
5. In your hobbies do you either make or use any of the following items?
(I) Homemade pottery (5) Other known contacts with lead
(2) Lead soldiers Specify
(3) Hand painted china (6) Any combination of the above
(4) Solder (7) None of the above
b. On the average how many hours a week do you spend away from home?
hours.
7 On the average how many hours do you spend riding in cars or buses
around town? hours.
8. Have you ever been diagnosed as being anemic?
tl) Yes (2) No
9. Have you experienced in the last month any of the following?
(I) Fever over 10! degrees (4) Any combination of the above
(2) Hospitalization (5) None of the above
(3) Traumatic injury
1 0. Have you ever smoked as many as five packs of cigarettes, that is. as many
as 1 00 cigarettes during your entire life?
(I) Yes _ (7) No _
II Do you now smoke cigarettes''
(I) Yes _
NO
6-33
47
48-50
51-53
D
D
54
55
56
57
Figure 17. Traffic Lead Individual Questionnaire
70
-------
12. Ifyouarea current or an e \-cigarette smoker:
a. How many cigarettes do (did) you smoke per day?
(I) Less than 1/2 pack per day ( 1-5 Cigarettes per davl
(2) About 1/2 pack per day (6-14 cigarettes per day)
(3) About 1 pack per Jay I 15-25 cigarettes per Jay)
(4) About 1-1/2 packs per day (26-34 cigarettes per day)
( 5) About 2 packs per day (35 or more cigarettes per day)
(9) N/A
b. How old were you when you first started smoking? years
(99) N/A
c. How old were you when you last gave up smoking, it you no
longer smoke'.' years
(09) N/A
13. If individual is a child: (Preschool Ould!
a. Is the child involved in my oi the following activmes'
( I) Play or nursery school
I 2) Day care away from its hume
(3) Routinely stays away Horn home either at relatives or with Iriends
(4) Any combination of Hie above
(5) None ol the above
14
n
m
5^-60
b. What is I he child'; usual play site''
At home indoors
At home outdoors near the hou.-e
\t home outdoors near the street
Elsewhere
Number of Hours Per Day
Number of Hours Per Day
Number of Hours Per Day
Number of Hours Per Day
It individual is an adult. ( 20 years old or over)
j Winch of the following best describes your activity pattern?
(I) Employed outside the home 5) Full-time student
(2) Employed inside the home !6) Rclired
(3l I'nemployed (71 Other
N/A
14) Housewife C'l
For all lobs held in the last year please -.tale lor each:
I 1) Nature of company, business or agency
(2) Specific job performed
(3) Number of years spent on job years
(4) On the job were you ever exposed lo lead?
(ll Yes (2) No
It >'ei. ^pe^il\ details
61-62
64-65
66-67
68-60
70-71
72
n
YOU HAVE COMPLETED THE QUESTIONNAIRE.
THANK YOU
Interviewer's Initials.
IMNO\VOI2345(.
74
so
Figure 17. Traffic Lead Individual Questionnaire (Cont.'d)
71
-------
b. Household Data Recording and Scheduling
(1) Data Recording
This project required that a large
amount of sample information be gathered from the partici-
pants and their homes. To minimize duplication and centralize
this information, a "Participants and Household Checklist"
was made for each household participating in this study-
Figure 18 is an example of the form used.
Figure 18.
NEIGHBORHOOD LEAD STUDY
Participants &c Household Checklist
01-4294
Appnt. Time:
Day: ~
Area:
Prtcpnt
Prtcpnt
Prtcont
Prtcpnt
I. Informed Consent
Blood
Handwipe
(children only)
Date Collected
II. Window sill wipe
Tap water
Soil
Indoor dust
Outdoor dust
HE. Lead in Paint
Rhenium filter count
Lead filter count
Results
1st- sample-Znd
1st- sample- Znd
1st- sample- 2nd
1st- sample- Znd
Date Collected
Start Date
Indoor
Outdoor
mg/cm
17-20
IV Distar-ce from
Roadway to:
House front
ft;
2Z-Z5
House rear
Date
mg/cm-
ft.
28-30
33-33
V. CadmiurnParticipants ?
72
-------
Information collected by the bio-
logical collection teams, the paint analyzer, and on the
soil collected were all recorded on this one sheet for each
household.
Item I on this checklist provided
space for up to four participants at each household to
include 1st and 2nd blood samples, hand-wipe samples, and
date collected.
Item II provided information on the
household samples collected.
Item III was for recording x-ray
analyses data on paint measurements.
Item IV provided space for measure-
ment of house location in relationship to the street where
traffic densities were measured.
Item V was included for informa-
tional purposes on another EPA project (EPA Contract 68-02-
1725).
Also, a specific area on the check-
list was used to indicate that the second set of validation
forms had been completed.
Prior to each day of biological and
environmental sample collection, these checklist sheets
were organized in each team's notebook according to the
households that team was scheduled to visit. Also included
73
-------
with each household checklist was the necessary sample
labels for all samples to be collected at that location
at that specific time.
Once the biological collection teams
had collected the first samples from each household, the
checklist sheets were placed into the paint analyzer's
notebook. After the paint analysis had been completed
at each residence, the checklist sheets were placed into
the soil collector's notebook. The checklist sheets were
stored in a master notebook following the entry of the soil
data until the biological collection teams needed them for
the collection of the second blood samples.
After all samples at each household
had been completed, the checklist sheets were returned
to the master notebook to prevent them from being lost
during shipment to the San Antonio laboratories.
These checklist sheets provided
a convenient means of accounting for which and what type
of samples had been collected at each household.
Similar checklist sheets were main-
tained on the traffic counting and air sampling operations
as described under the appropriate heading.
(2) Scheduling
Once the household surveys were complete
(March 26, 1976), the SMU on-site coordinator began scheduling
74
-------
participants for collection of biological and household
samples. This was accomplished using the following protocol:
(a) Participant information sheets
were arranged according to general location within the city.
(b) Next, the sheets were grouped
by street name, then by address.
(c) One to three schedulers would
then begin telephoning the participant in a specific area
to arrange a time for the biological collection teams to
come by.
(d) As the appointments were made
they were listed on worksheets giving the date, time, name,
and address of the participant(s). These worksheets were
made up daily for each biological collection team.
(e) At the end of each day the bio-
logical manager would pick up the worksheets and arrange
the checklist sheets in each biological collection team's
notebook according to the schedule. The appropriate computer
printed labels were attached to each checklist and the note-
books were ready for the following day's sampling activities.
(f) The day before the scheduled
appointment, a second telephone call was made to remind the
participants about the appointment.
In the first week of scheduling,
30 minutes per household was alloted for sample collection.
75
-------
This was later reduced to 20 minutes because of the effici-
ency and experience of the biological collection teams.
Because the study could not be con-
fined to one area of the city as initially planned, scheduling
of participants became a critical factor to the success
of the project. The primary thought behind the scheduling
protocol was to minimize the travel time of the collection
teams between appointments. A self-contained "mobile lab-
oratory" (16 ft. motor home) was used to further reduce
travel time of the collection teams. Each day of sample
collection, the mobile laboratory would be parked at some
central location within the area being sampled. Usually
this was a church or school parking lot. This allowed the
biological collection teams to bring their samples in for
processing and provided a convenient place to wait between
appointment (if necessary). Also, the biological manager
was located at the mobile laboratory and this gave him better
control of the sample collection efforts. Frequent contact
was made between the SMU on-site coordinator and the bio-
logical manager at the mobile laboratory to accommodate
any last minute schedule changes.
After the second week of sampling,
the biological collection teams were familiar with the
different sampling areas and the scheduling was routine
enough that a centralized location was used to coordinate
the activities of the collection teams rather than the mobile
laboratory.
76
-------
Scheduling of paint analysis and
soil collection was usually made one to two days before
the actual collection. This was handled by the individual
performing the collection or analysis.
c. Environmental
(1) Traffic Counts
Traffic was counted to categorize
participant homes into sites. Each street on which a par-
ticipant's residence was located had a traffic count taken.
If a number of participants lived within a given block and
there was no major cross street between residences then
one count was taken to represent all those particular re-
sidences. For certain streets, th-en, multiple counts were
required because of the intervention of major side streets
or controlled intersections. Participant homes located
on corners had both the principal street counted as well
as the side street.
(2) Soil
At each residence of participants
in this study, a soil sample was collected. The sample
was collected at the front of the house near the front
door. Surface soil was collected from flower beds or similar
exposed areas. In selecting an area for sampling, an effort
was made to stay away from those areas which appeared to
have sand or top soil recently applied. If the area to
be sampled was next to the house, a check of the area for
77
-------
paint chips was made. If no chips were seen, then a sample
was taken at least 0.3m away from the house. This minimum
distance was maintained to avoid any possible contamination
that might have occurred if the house had been previously
painted with lead-based paint. The soil sample was collected
with a stainless steel trowel. The sample was then trans-
ferred to an acid-washed 250-ml polyethylene bottle for
Pb analysis. A computer-printed label designating soil
from that particular residence was attached to the bottle.
A second sample was also collected for the determination
of soil characteristics.
(3) Tap Water
Water samples were collected in
474-ml polyethylene containers by the biological collection
team during the first visit to each household. The samples
were collected from the cold water tap in the kitchen.
The water was allowed to run for approximately 1 minute.
Then the polyethylene container was rinsed 3 times before
being filled. The appropriate label was attached and within
3 hours of collection the water samples were returned to
the mobile laboratory. The samples were acidified (approxi-
mately 1%) with HN03 (reagent grade) and stored at room
temperature until shipment. Water samples were frozen and
packed in dry ice to minimize leakage during shipment to
the San Antonio laboratories.
78
-------
(4) Housedust
These samples were collected in a
23.0 x 31.2 cm plastic tray over a period of 28 days.
The trays were placed by a member of the biological collec-
tion team during their first visit to the household. Loca-
tion of where the tray was placed and the date was recorded
on the checklist sheet. Ideally the trays were to be placed
in an area where air current would not affect them or children
have access to them. Usually the top of the refrigerator
was the place where most dust trays were placed.
Instructions were given to the house-
hold participants about the dust trays. The trays were
not to be disturbed for any reason and a staff member would
call in about 28 days to make arrangements to pick up the
tray.
Collection of the dust trays was
carried out by one member of the biological collection
team. Usually appointments were made by areas to reduce
the travel time required to collect them. This caused
some trays not to be collected exactly on the 28th day. The
exact number of days for each dust sample is given in Appendix E.
The contents of the tray were quan-
titatively rinsed into a 474-ml polyethylene bottle with
deionized water. The appropriate computer-generated label
was attached and the sample stored at room temperature
79
-------
until shipment to the San Antonio laboratories. To mini-
mize leakage during shipment, all samples were frozen and
packed in dry ice.
(5) Windowsill Wipes
These samples were collected in a
room nearest the street being counted. The biological
collection team collected these samples from each household.
Commercially available "Wash'n Dri Towelettes" were selected
for this particular sampling after several brands were
analyzed for Pb content prior to the field sampling. This
particular brand had the lowest Pb content of any brand
tested.
Initially, a plexiglas template 7.7
x 45.5 cm was used as a reference in collection of windowsill
wipes. This proved unsuitable because many windowsills
were too small to cover all of the template. Each collec-
tion team was supplied with a ruler and the actual area
of the windowsill sampled was measured and recorded on the
checklist sheet. The procedure used to wipe the windowsill
was standardized so each team would collect the samples
the same way. After wiping the windowsill by the prescribed
method, the handy wipe was placed in a self-sealing poly-
ethylene bag and the appropriate label attached.
These samples required no further
processing in the field. They were packed and shipped at
ambient temperature.
80
-------
(6) Paint
A description of the x-ray analyzer
used for paint analysis is given later (see sample analysis).
During the first visit of the biological collection team
to each household, the participants were told to expect
a call from the paint analyzer to set up a time to survey
their home (usually within 2 to 3 days).
Normally, the time required to complete
the Pb paint survey for each household was approximately
20 to 25 minutes. At each household, two rooms were anal-
yzed and at least two separate readings were performed on
the exterior paint of the house. These readings were immed-
iately recorded on the checklist sheet and converted into
mg Pb/cm^ with the appropriate calibration graphs.
Verification of the calibration of
the x-ray analyzer was performed at least twice each day.
Every night the instrument was connected to a charger unit
so the batteries would be fully charged for each day's sampling,
(7) Hand-wipes
Each child participating in this
study had one hand-wipe sample collected from him/her during
the first or second visit of the biological collection team
to the home. Collection of handwipe samples from children
recently bathed were postponed until the second visit.
81
-------
The hand-wipe sample was collected
just prior to taking the blood sample from the child. If
a capillary blood sample was to be collected, then the hand-
wipe became an integral part of the blood collection pro-
cedure (see Biological Samples).
One "Wash *n Dri Towelette" was used
to thoroughly wipe both hands of the child. This "towelette"
was then placed in a polyethylene bag and the proper label
attached. Prior to taking this hand-wipe, the laboratory
technician would use a "towelette" to clean her hand to
reduce the possibility of contamination.
(8) Validation
A validation of information regarding
each residence was performed during the collection of bio-
logical samples. A validation form (Figure 19) was used
to collect pertinent data on each residence to be checked
against the household questionnaire data. All households
selected for the study met the established criteria.
A validation of information regarding
each participant was performed during the collection of
participant samples. A validation form (Figure 20) was
used to collect pertinent data on each participant to be
checked against the participant questionnaire data. All
participants selected for collection of biological samples
met the established criteria.
82
-------
Figurel9. Household Validation Form
Traffic Lead Validation
Informant's Name:
Address:_
Phone $~
Family Position:_
Household Queaiions-Circle Responses-Far right responses disqualify household.
1. Do you live in a. corner residence?
No Yes
2. Do you live in d single family dwelling or a duplex?
Yes No
3. Is your residence 100 feet or less from the street?
Yes No
4. Does your residence face the street?
Yes No
5. Do you live on the ground floor?
Yes No
6. Do you live within 300 feet of a traffic signal or a stop sign?
No Yes Household Eligible-Yes No
d. Biological
(1) General
The biological collection teams collected
two blood samples from each participant approximately one
week apart. These were venous blood samples collected
from trie antecubital vein using a 10 ml Vacutainer (minimal Pb) .
From previous experience, collecting venous blood from small
children was not always possible and many times this would
bias the data in the younger age groups. To assure sufficient
83
-------
Figure 20. Traffic Lead Validation
Individual Questions Circle Responses-Far right responses disqualify individual.
1. Name of Member:
2. How long have you lived at this address?
3 months or more Less than 3 months
3. Sex: Male
Female
4. Age in years:
5. If individual is a. child:
Age: 1 thru 5 years Less than 1 year 8c
more than 5 years
Does your child spend more than 30 hours a week away from home?
No Yes
If yes:
How many hours does your child spend away from home?
Where does your child spend those hours?
6. If individual is a female:
Age: 20 thru 49 years Less than 20 years k
more than 49 years
On the average, how many hours a week do you spend away from home?
Less than 30 hours per week 30 hours or more per week
7. If the individual is 50 years old or more:
On the average, how many hours a week do you spend away from home?
Less than 30 hours per week 30 hours or more per week
Are you retired from your work or work in your home?
Yes NO
8. I M N O W O Unk.
9. Date of Validation:
10. Interviewer's Initials
Individual Eligible Yes No
data points in the younger age groups, an alternate type
of blood sampling procedure would be used in those diffi-
cult cases. Capillary blood taken by the "finger-prick"
method was the alternate procedure chosen because it was
relatively easy to perform on a reluctant child and it was
less traumatic than the venipuncture technique.
84
-------
Because the "finger-prick" technique
limits the volume (lOOul) of blood taken for analysis and
the procedure is more liable to contamination problems(-^)
every effort was made to convince the child and parent(s)
to allow the venous blood be taken.
Fewer venous blood samples were
taken from children on this project than we initially an-
ticipated. The reason for this may be the fact that pro-
viding an alternate procedure, which in the mind of the
parent is less traumatic to the child, may counteract any
argument for the venous blood sample that the laboratory
technician could make.
(2) Venous
Venous blood was taken from partici-
pants by the routine venipuncture technique commonly used
in clinical laboratories. The procedure was modified in
that dilute HNO3 (3ml/liter) rinse of the puncture site
was included to reduce the possibility of Pb contamination
from the skin. The blood was collected in 10ml Vacutainer
(Becton-Dickinson Co., Rutherford, N.J.) using the minimal
Pb type (L3200XF313) containing 143 USP units of sodium
heparin. Previous tests with this type of Vacutainer in-
dicated the Pb content to be approximately 0.1 yg per tub
Once the blood had been drawn, properly
labeled, and thoroughly mixed, it was immediately placed
85
-------
in a small styrofoam cooler containing wet ice to maintain
it at a chilled temperature. The blood samples remained
in this cooler from 30 minutes up to 3 hours depending upon
the sampling schedule of the biological collection teams.
At the centralized location (mobile laboratory) one of the
10ml Vacutainers from each participant was opened and an
aliquot taken for hematocrit determination. The unopened
Vacutainers were properly marked to be used for the blood-
Pb determination at the San Antonio laboratories. The Vacu-
tainer which was opened was also sent along with the unopened
one as a reserve.
(3) Fingerprick
Capillary blood samples were collected
from children using the recommendations of Bratzel and Reed
Usually, the child would sit in the
parent's lap during the blood drawing to maintain control
and to relieve the stress associated with the procedure.
The child's arm would be fully extended and held at the
elbow by the parent. The laboratory technician would grasp
the child's hand and wash it very thoroughly using a "Wash1
Dri towelette." Next, the 3rd or 4th finger would be held
in such a way that the terminal digits were exposed and
under complete control of the technician. A vigorous scrub
was then made of the exposed finger using a gauze pad soaked
in Phisohex soap. This would be followed by another gauze
86
-------
pad soaked in deionized water. Another gauze pad soaked
in dilute HNO^(3ml/liter) would be used to scrub the finger
tip followed by another deionized water rinse and a final
rinse with an isopropyl alcohol soaked pad. A sterile micro-
lance was used to puncture the skin and the first few drops
of blood were allowed to flow freely.
A 100 yl capillary blood-collection
tube (ESA, Inc., Burlington, MA) was then used to collect
the blood. Care was taken not to contaminate the puncture
site or the capillary tube. Once the tube was filled to
the mark, the laboratory technician would tilt the tube
so the blood would not run out either end and hand it to
the other team member to seal with polyethylene end caps.
If possible, 3 or 4 such samples were collected from the
same puncture site along with two hematocrit tubes.
These samples were placed into a
polyethylene bag and the appropriate label attached. They
were stored in the styrofoam ice chest until the collection
team returned to the mobile laboratory.
Immediately upon receipt of the
samples at the mobile laboratory, the hematocrits were de-
termined and the samples stored in the refrigerator until
shipment to the San Antonio laboratories.
Following each day of sample collec-
tion, all blood samples were packed in styrofoam mailers
containing dry ice and shipped to the San Antonio laboratories
-------
by air express (counter to counter delivery). Usually the
time from collection of the blood until it was analyzed
for Pb averaged less than 24 hours.
3. Sample Analysis Procedures
a. General
(1) Instrumentation
All analyses were performed on either
a Perkin-Elmer Model 503 Atomic Absorption Spectrophotometer
(AAS) or a Perkin-Elmer Model 306 AAS. The Model 306 AAS
is modified (Perkin-Elmer Modification Kit 040-0286) to
reduce "stray light" from reaching the photomultiplier tube
during operation of the flameless sampling devices.
Both AAS units are equipped with
a Deuterium-Arc background corrector which corrects for
non-specific absorption. The background corrector was
routinely used on all analyses.
Absorption peaks were recorded on
a Perkin-Elmer Model 056 Recorder with a lOmv range.
Flameless analyses were performed
with the following graphite tube furnaces: (1) a Perkin-
Elmer HGA-2100 with the Model 503 AAS, (2) a Perkin-Elmer
HGA-2000 with the Model 306 AAS, and (3) an Instrumentation
Laboratory IL-455 with the Model 306 AAS. Flame analyses
on both AAS units were by air-acetylene flames using a
single-slot, 10-cm Burner Head (Perkin-Elmer Model 303-0418).
88
-------
(2) Reagents
All regents used for the preparation
and analysis of the samples on this contract were of analy-
tical grade or better.
b. General Quality Control
A major problem in trace metal analysis
is contamination of glassware, reagents, and samples with
the metal(s) being analyzed. Minimizing this problem re-
quires an extensive control program involving glassware
cleaning, protection, and quality control measures.
(1) Cleaning
All glassware and polyethylene con-
tainers that come in contact with samples or reagents are
cleaned by the following procedure: Items are washed
thoroughly with a laboratory detergent (Alconox, Inc., New
York) in tap water. The clean glassware is rinsed with
deionized water and placed in an acid vat containing HNC>3(1:1)
and allowed to soak for 6 to 18 hours. Clean polyethylene
containers are also placed in the acid vat but are removed
after 4 to 6 hours. After acid soaking, the items are rinsed
thoroughly with deionized water and placed in a drying oven
until dry. The dry items are placed in a dust-free area
and allowed to cool. Polyethylene containers are capped
and sealed in polyethylene bags until ready for use. Glass-
ware is returned to its proper container (see below) and
stored in glassware cabinets until ready for use.
89
-------
(2) Protection
All glassware items are kept in poly-
ethylene containers to minimize exposure to dust in the
laboratory. Each container is numbered and contains one
type of glassware (i.e., watchglasses, 4-ml volumetric flask,
etc.). When all the glassware in a container has been used,
it is returned to that container and carried through the
washing procedure (see above) as a unit. While the glassware
is being washed, the container is also washed. Once the
glassware has completed the wash cycle and dried, it is
returned to the proper container. Several items (3 to 7)
of glassware are removed from the container at this time
for quality control checks.
(3) Quality Control
The number of each container of glass-
ware going through the wash cycle is entered into a log
book. Other information, such as name of technician per-
forming washing procedure, type of glassware, length of
acid-soaking, etc., are also recorded in this log book.
This allows the glassware removed for quality control checks
to be identified with a particular set of glassware being
used in the laboratory. The glassware removed for quality
control purposes is checked by rinsing with a known volume
of 0.1N HN03 and comparing with the same acid that has not
been used for rinsing. Normally one metal (Pb) is used
for quality control checks but other metals may also be
90
-------
included if needed. The graphite furnace (AAS) is used for
quality control analysis. Glassware which shows a signi-
ficant difference between the used and unused rinsing acids
metal content is referred back to the container number from
which it came. That container is then returned to the wash-
room and the wash cycle repeated on all of its glassware.
c. Soil
Soil samples were analyzed for Pb using
a modification of the preferential leach procedure Smith
and Window'-*-"'used on dried sediment samples. Table 1 sum-
marizes the analytical parameter for soil samples. The
soil sample was dried at 65°C overnight in an oven. The
dried soil was ground to powder using a mortar and pestle
and sieved through a 250 micron stainless steel screen.
Then 5 g of the powdered sample was weighted into a 125-ml
Erlenmeyer flask and 50 ml of the leach solution {acetic
acid: hydroxylamine HCL (7:3)} was added. A polyethylene
stopper was used to seal the flask. The sample was then
placed on a mechanical shaker for 12 to 18 hours (overnight).
The leach solution was filtered through
a glass-fiber filter (9cm) which had previously been rinsed
with the leach solution. The filtrate was collected in
a 50-ml polyethylene bottle and lead was determined on this
solution using flameless AAS.
91
-------
Soil samples were quantitated for Pb con-
tent by the procedure used for air particulate and outdoor
dust. Figure 21 represents a typical analytical curve for
soil-Pb. Quality control samples were spiked soil samples
which were handled according to the procedures detailed for
air particulate samples.
Figure 21. Analytical Curve for Lead in Soil
180,
50
100
\ig Pb / g m
92
J
150
-------
d.
Water
Acidified water samples were analyzed
directly by flameless AAS. No digestion or concentration
was necessary on these samples. Table 3 summarizes the
analytical parameters for water samples. Water samples
Table 3.. Analytical Parameters of Environmental Samples
Sample Matrix
Sensitivity * '
Window Sill Wipes
0.6 ng/cm
(4)
Detection Limit 5.0 ng/cm2
Linear Range
Recovery
Precision
n
mean
std. dev.
RSD
110 ng/cm2
109.5%
6
26.0 ng/cm2
2.0 ng/cm2
8.7 %
Water
94.4%
25.0%
Indoor Dust
(5)
0.04 ng/ml 0.075 ng/cm'
0.2 ng/ml 1.0 ng/cm2
4.0 ng/ml 7.0 ng/cm2
95.6%
0.4 ng/ml 8.7 ng/cm2
0.1 ng/ml 0.7 ng/cm2
8.0%
(1) see text for definition
(2) does not imply maximum linear range
(3) based upon average recovery of low Pb spike in sample matrix
(4) surface area used to calculate = 350.4 cm2
(5) surface area = 717.6 cm2
were quantitated by the "method of additions" described
earlier for air particulate samples. Figure 22 represents
a typical analytical curve. Quality control consisted of
Pb spiked water samples routinely analyzed with the unknown
93
-------
samples. A recovery factor was determined from these quality
controls and applied to the water samples.
Figure 22. Analytical Curve for Lead in Water
180 r-
160 L_
1.5
3,0 4.5 6.0
ng Pb/ml (ppb)
7.5
e. House Dust
The same procedure used to analyze the
outdoor dust samples was used to determine the Pb content of
94
-------
the indoor dust samples. Table 3 summarizes the analy-
tical parameters for house dust samples. The Pb concentra-
tion was determined by flameless AAS. Indoor dust samples
were measured by the same procedures used for water samples,
Figure 23 is a typical analytical curve. Quality controls
and recoveries were the same as described for water.
Figure 23. Analytical Curve for Lead in Indoor Dust
5.0 6.0 7.0
95
-------
f. Windowsill Wipes
The windowsill wipe samples which were
collected on "Wash 'N Dri" towelettes were analyzed by the
procedure used for the particulate air samples. Table 3
summarizes the analytical parameters for windowsill wipes.
The towelette was carefully removed from the polyethylene
bag and placed into the 2.5 x 8.5 cm extraction thimble.
Eighty milliliters of the digestion acid were added and
the apparatus assembled as previously described (Figure
13). The final volume of the sample was brought to 5.0
ml with deionized water prior to centrifuging. An aliquot
of this final volume was analyzed for Pb by flame AAS.
Quantitation of these samples was by the procedure given
for water samples above. Figure 24 is a typical example
of an analytical curve. Quality controls and recoveries
were the same as described above for water samples.
g. Paint X-Ray Fluorescence
Each home where participants of this study
resided was analyzed for lead in wall paint. An x-ray fluor-
escence analyzer, Model 700, manufactured by Columbia Scien-
tific Industries, Austin, Texas, was used for this determina-
tion. This instrument employed a 3 me Cd10^ source with
a lead and rhenium filter. Standardization of the instru-
ment was obtained using lead standards borrowed from Columbia
Scientific. These standards consisted of various concentra-
tions of lead ranging from 0 to 6.89 mg/cm2 in polyethylene
96
-------
Figure 24. Analytical Curve for Lead in Window,Sill Wipes
20
wafers. By a combination of these standards with various
substrates (1/4" masonite, 1/2" sheetrock, 3/4" and 1-1/2"
wood, aggregate and brick), a calibration curve was obtained
for any combination of substrate material and/or lead level
97
-------
encountered in participants' homes. Figures 25 and 26
illustrate these calibration curves.
Figure 25.. X-Ray Fluorescence Analyzer Rhenium Filter Calibration Curve
ug / cm Pb
The commonly accepted value for excessive
lead in paint is difficult to measure in situ, as the cor-
responding paint thickness is undefined. For this study,
the unit mg/cm2Pb was used which corresponds to the measuring
geometry of the x-ray fluorescence detector. The relation-
ship of paint thickness and mg/cirf Pb corresponding to 1%
Pb is 1 to 4 mg/cm2Pb. The instrumentation used had a sen-
sitivity of + 0.1 to 0.3 mg/cm2pb.
Two different painted surfaces, both in-
side and outside, of the participants' homes were analyzed.
98
-------
Figure 26. X-Kay Fluorescence Anily/
Lead Filter Calibration Curve
12.49
100
(i g I c m P b
On the inside of the home, measurements were taken in the
room closest to the street such as a living room and, also,
in the child's bedroom if applicable. The outside measure-
ments were taken on either the front door or door jamb and
the garage door. Condition of the painted surface analyzed
was noted as was the general condition of all painted sur-
faces in the home. The exterior composition (brick, siding,
etc.) was also noted. Twice a day, usually in the morning
and afternoon, a 100% Pb standard was analyzed. This
99
-------
measurement gave a visual indication of the condition of
the instrument and also provided a decay factor for the
radioactive source. All readings were subsequently ad-
justed to reflect the condition of the source at the time
of calibration.
h. Hand-Wipe Samples
Hand-wipe samples, which were collected
on "Wash 'N Dri" towelettes, were analyzed by the procedure
outlined above for windowsill wipe samples.
Hand-wipe samples from children were quan-
titated by the "method of additions" as described for the
water samples. Figure 27 illustrates a typical analytical
curve for hand-wipe samples. Quality controls for the hand-
wipes were also similar to those of the water samples.
i. Blood
(1) Venous Blood
Development of_ Analytical Methodology
Several studies were performed to
determine what parameters would affect the Pb content of blood
samples collected in 10ml minimal Pb Vacutainers (sodium
heparin anticoagulant).
One important question which needed
addressing was whether the blood needed to be frozen, re-
frigerated or maintained at ambient temperatures between
collection and analysis. Contact with several investiga-
tors running large-scale blood-Pb screening programs indicated
100
-------
Figure 27. Analytical Curve for Lead in Hand Wipes
160,-
I
2.5
5.0
10.0
15.0
they either maintained the blood at ambient temperature
or refrigerated it for shipment. Most investigators would
refrigerate the blood once it arrived at the laboratory
and analyze it within 3 to 5 days.
101
-------
When the study site shifted to
Dallas, plans were made such that blood samples collected
would be refrigerated from the time of collection until
Pb analyses were performed. Also these plans made provi-
sions for the blood-Pb determinations to be done within
24 hours of collection.
A set (3 vacutainers) of bovine blood
samples from the Center for Disease Control (CDC) Blood-Pb
evaluation program were analyzed in duplicate for Pb on
three different occasions over a period of 20 days. The
blood was refrigerated during shipment (wet ice) from CDC
and maintained in a cold room (40°C) during the 20-day period.
Table 4 summarizes the data from this study. The average
variation from the CDC Reference Lab value of the low Pb
Table 4. Effect of Refrigeration on Blood-Pb
Values of Whole Blood (CDC Bovine Blood)
Sample Day No. 1 Day No. 10 Day No. 20 Mean &
ID yg/100 ml yg/100 ml yg/100 ml Std. Dev.
1 38.9
2 80.2
3 60.3
41.0
88.5
67.3
38.3
86.7
63.9
39.4 + 1.4
85.1 + 4.4
63.8 + 3.5
sample over 20 days was 14.7% while the high Pb sample
varied only an average of 2.9%. Refrigeration (at 40°C)
of whole blood samples is adequate protection to get accurate
blood-Pb concentrations from venous blood samples up to
20 days.
102
-------
Bovine blood samples from the CDC
program were collected in Vacutainers containing EDTA as
an anticoagulant whereas the Vacutainers used for this study
contained sodium heparin. A comparison was made to deter-
mine if this would bias the blood-Pb data. Freshly drawn
human blood was collected in both types of Vacutainers from
the same individuals at the same time. Blood-Pb analysis
of these bloods indicated the Vacutainers containing the
EDTA gave results which were 21.9% lower than the blood
from the Vacutainer containing the sodium heparin. Adding
Q
CaCl to the blood with EDTA gave blood-Pb values the same
as obtained from the sodium-heparin blood.
This effect of the anticoagulant
would not affect the venous blood-Pb data from the Dallas
study site since all samples were collected in Vacutainers
containing sodium heparin and the spiked bloods used to
quantitate these samples also contained sodium heparin.
Another study was performed to eval-
uate the effect different kinds of blood would have upon
the blood-Pb values. The purpose here was to investigate
bovine blood as a source for spiking to quantitate the human
blood samples. This would be less expensive than having
to purchase human blood for spiking standards. Both bovine
and human blood samples were spiked with known amounts of
Pb to establish calibration curves (concentration vs peak
103
-------
heights). The slope of the human blood calibration curve
was 1.365 while the slope of the bovine blood calibration
curve was 1.692. This represents an approximate 24% dif-
ference which favors the human blood. Therefore, it was
necessary to match the kind of blood used for spiking stand-
ards with the kind of blood being analyzed for Pb. All
venous blood samples collected in Dallas were quantitated
using human blood purchased from a local blood bank. All
CDC blood samples were quantitated using bovine blood (anti-
coagulant-EDTA) purchased from a local veterinarian.
Table 5 gives the analytical para-
meters for venous blood using this methodology.
Table 5. Analytical Parameters of Biological Samples
Sample Matrix
Sensitivity* '
Detection Limit'1'
Linear Range ' '
R ecovery (~> 1
Precision:
n
mean
std. dev.
RSD
Venous Bloodv '
0.5 fig/100 ml
1.4 fig/100 ml
80 fig/100 ml
101.0%
7
5.8 fig/100 ml
0.3 fig/100ml
5.2%
Capillary Blood1 '
1 .2 fig/100 ml
9.2 fig/100 ml
80 |j.g/100 ml
112.8%(6)
7
46.3 fig/100ml
4.6 fig/100ml
9.9%
Hand Wipes
0.11 ug
1.60 fig
12.5 fig
103.8%
6
9.21 fig
0.80 fig
8.7%
(1) see text for definition
(2) does not imply maximum linear range
(3) 0.5 ml of whole blood
(4) 0.1 ml (100 fil) of whole blood
(5) based upon average recovery of low Pb spike in sample matrix
(6) recovery based upon CDC bovine blood analyzed over period of 11 days
104
-------
Analysis Procedures
Venous blood taken from the antecu-
bital vein of the participants was analyzed for Pb within
24 hours of collection. The samples were refrigerated (but
not frozen) from collection until they were prepared for
Pb analysis in the laboratory- Lead determinations were
performed on the blood from the unopened vacutainer (see
sample collection). The blood from the other vacutainer
was transferred to a 30-cc polyethylene bottle and frozen
(0°C) for future analysis if needed.
The method of blood-Pb determina-
tions used on the venous blood was patterned after that
of Hwang, Ullucci, and Mokeler,(17) and Kubasik and Volosin.
A 500yl aliquot of whole blood was pipetted from the 10-ml
vacutainer into a 5-ml screw-cap extraction tube. Then,
500yl of Trizma buffer solution (pH 7.0) (Sigma Chemical
Co., St. Louis, Mo.) was added, followed by SOOyl of a
chelating-hemolyzing solution consisting of 2% ammonium
pyrrolidine-dithiocarbamate (Aldrich Chemical Co., Milwaukee,
Wisconsin) in a 2% solution of Triton X-100 surfactant (J. T.
Baker Co., Phillipsburg, Pa.).
The sample was shaken to mix the
reagents and then allowed to stand 10 to 15 minutes to en-
sure complete hernolysis of the blood. To extract the
chelated Pb, SOOyl of methyl isobutyl ketone (Eastman Kodak,
105
-------
Rochester, N.Y.) was added and the sample vigorously shaken
for 5 minutes. The sample was centrifuged for 10 minutes
at 2500 RPM and the organic layer removed for Pb determina-
tion by graphite furnace AAS.
Venous blood-Pb concentrations were
determined by the "method of additions" using human blood
(purchased from local blood bank) spiked at 3 or 4 different
levels with Pb standards. From these spiked samples, an
analytical curve was calculated on a Hewlett-Packard 9810A
programmable calculator using a least-squares regression
program to obtain the best fit to the data points. Figure
28 represents typical analytical curves for venous blood.
The unknown blood samples' Pb con-
centrations were determined using the slope of the analy-
tical curve and the peak height of the sample less the peak
height of the reagent blank. After every 15 to 20 samples
were analyzed, a series of the spiked blood standards was
analyzed to allow a continuous upgrading of the analytical
curve.
Daily quality controls for the venous
blood samples were the spiked human blood used to establish
the analytical curve for Pb quantitation. The Pb concentra-
tion of this human blood used for quality controls was veri-
fied against CDC bovine blood samples of known Pb content.
106
-------
Figure 28. Analytical Curve for Lead in Venous Blood
-P
^
tji
rH
160
140
1ZO
100
80
60
40
20
5.5 |j.g/100 ml
20
10
_L
0 10 20
lag Pb/100 ml
30
40
In summary, the extraction procedure
used for Pb determinations in venous blood is relatively
simple procedure that requires a minimum quantity of blood
and because it is an extraction procedure avoids many of
the matrix interferences commonly associated with blood-Pb
107
-------
analysis. The procedural steps are kept to a minimum thereby
avoiding contamination and improving the accuracy and pre-
cision of the data.
It is important that the same anti-
congulant is used on both the blood spiking standards and
the unknown blood samples. Blood samples which contain
EDTA as the anticoagulant give lower blood-Pb values than
blood samples preserved with sodium heparin. The reason
for the lower results with EDTA could be the ability of
the Pb to form complexes with the EDTA which would be more
favored than the pyrollidinedithiocarbamate complex used
to extract the Pb.
It is important to match the kind
(species) of blood being analyzed for Pb with the spiked
blood used to quantitate it. Extraction of bovine blood
and human blood showed a matrix effect which was not
compensated for by the extraction process.
The conditions which are used to pre-
serve the whole blood samples until they are analyzed
can affect the blood-Pb data. The studies performed
to determine the best method of handling blood once it
was collected until it was analyzed indicated that
refrigeration (40°c) will maintain the blood in a condition
acceptable for blood-Pb analysis up to 20 days. The var-
iation between duplicate blood samples collected from the
same individual over a two-week period will also give
108
-------
information as to the effectiveness of the preservation
technique and the analytical methodology used. A large
variation between duplicates could be one indication that
the preservation method is not adequate since the normal
blood-Pb values would not be expected to vary over a very
wide range within a short period of time. Microclotting
of the whole blood caused by inadequate preservation tech-
niques, would give highly variable results both on in-
dividual samples and on duplicates. There were no signi-
ficant differences between the duplicate venous blood-Pb
values for the subjects in this study.
The studies undertaken prior to
collection and analyzing blood from the study participants
helped to eliminate some variables which might have other-
wise confounded the blood-Pb data and made it more difficult
to interpret.
(2) Capillary Blood
Development of Analytical Methodology
Previous studies involving blood
collection from young children indicated an alternate, less
traumatic procedure should be available in those instances
where child and/or parent object to venous blood being taken.
There are a number of screening
methods based upon taking capillary blood by finger-prick
(19-22)
and analyzing by a microanalytical technique using AAS.x '
109
-------
These methods have generally been for screening purposes
where blood-Pb concentrations below 40 ug/lOOml were not
quantitated. Our purpose was to adapt one of these micro-
techniques to analyze blood with Pb concentration less than
30 ug/100ml.
After considering the different tech-
niques available for collecting the capillary blood for
Pb analysis, we decided the 100 yl capillary tubes would
provide the best method considering the field-sampling
and shipping conditions they would be exposed to. Precal-
ibrated capillary tubes are now made specifically for blood-
Pb determinations, i.e., minimal Pb contamination.
The original methodology development
involved diluting the blood to 2.5 ml rather than to 5.0ml.
This dilution was not sufficient to reduce the matrix effect
caused by the blood. At this dilution the background cor-
rector (D2arc) could not compensate for all the matrix in-
terferences present when the sample was atomized in the
graphite furnace. Increasing the ashing temperature to
remove more of the matrix effect resulted in losses of Pb.
Increasing the dilution to 5ml im-
proved the data but there were still some matrix effects
not being corrected. At this point a chemical solution
was tried since the instrument parameters (ashing tempera-
ture, time, etc.) had reached their maximum effect. A
110
-------
"Keeper" element was introduced into the furnace with the
sample to retain the Pb at the higher ashing temperatures
of the furnace. A lOyl injection of a 50ppm Ni solution
placed on top of the sample injection allowed ashing tem-
perature of 600°C to 650°C without losses of Pb. This ef-
fectively removed all matrix effects.
Since the capillary blood samples
would be taken by laboratory technicians without direct
supervision and under conditions unfamiliar to the technician,
the possibility of collecting more or less than the 100 yl
was considered.
By using the diameter of the capillary
tube and measuring the length of blood collected, the volume
of blood in the capillary tube could be determined. To
determine if this was a valid assumption, capillary tubes
were fitted to various size Eppendorf pipets using a modi-
fied pipet tip. Known quantities of CDC blood were drawn
into the capillaries. The length of the blood in the cap-
illaries was measured and the blood was then analyzed by
the capillary blood procedure. Table 6 gives the results
of this study. The variation due to calculating the blood
volume ranged from 1.2% to -14.3%. Most of the blood-Pb
values determined by calculating the volume of blood were
less than the CDC Reference Laboratories value for the blood.
All were within the +15% variation from the mean value
54.5 yg/100 ml).
Ill
-------
Tablo 6. Determining Blood Volume in Capillary
Tube by Indirect Measurement
Sample
ID
80-1
80-2
80-S
90-1
90-2
90-S
95-1
95-2
95-S
100-1
100-2
100-S
110-1
110-2
110-S
120-1
120-2
120-S
Measured
Volume
^
80
80
90
90
95
95
100
100
110
110
120
120
Total
Pb-Conc.
Length of Calculated
Blood, mm Volume, ul
47
47
53
53
57
57
59
58
65
66
71
71
.0
.5
.5
.5
.0
.0
.0
.5
.5
.0
.5
.0
78.
79.
89.
89.
95.
95.
98.
97.
109
110
119
118
Mean
SD
CV %
6823
5194
5639
5639
4232
4232
7714
9344
.653
.490
.698
.861
in Blood
|ig/100ml
51.
53.
52.
58.
52.
55.
47.
46.
46.
46.
46.
46.
50.
47.
49.
49.
52.
50.
50.
3.
6.
96
55
75
01
30
15
30
41
86
56
96
70
49
79
14
10
07
88
25
35
7%
% Difference
from Extract
Blood Value
-3.2%
+ 1.2%
-14.0%
-14.3%
-9.8%
-6.6%
-7.8%
As a precaution, all capillary blood
samples were measured and the length of blood recorded
prior to analysis. Calculation of the blood-Pb values
using the measured volume and 100 yl did not change any
of the data significantly.
Variations in the instrument para-
meters were determined to be a serious problem. The degrading
112
-------
of the graphite tube seemed to be a constantly changing
parameter that would affect the blood-Pb data if not com-
pensated for. Analyzing spiked blood standards and CDC
bovine blood routinely with every 5 to 10 capillary blood
samples eliminated this problem. Table 7 summarizes the
results of the CDC bovine blood analyzed as quality control
over a period of 8 consecutive days.
Table 7- CDC Bovine Blood as Quality Control for
Capillary B'cod-Pb Analysis
Blood-Pb
Number Days (xg/100ml
0 43.5
1 44.5
2 47.0
3 45.8
4 39.9
5 54.4
6 49.1
7 44.8
mean 46 .1
st.dev. 4.3
RSD 9.3%
113
-------
Analysis Procedures
Capillary blood taken by the finger-
prick technique was refrigerated from the time of collec-
tion until Pb analyses were performed in the laboratory.
At least one and in some cases two of the extra blood
capillary-tube samples were frozen for later Pb determina-
tion if the necessary information could not be obtained
from the capillary tube analyzed immediately upon receipt
at the laboratory.
Lead analysis of the capillary blood
samples was performed by a procedure similar to that of
Norral and Butler(23)using the graphite furnace technique
(AAS) to improve sensitivity. The micro-capillary tube
containing the blood sample was fitted to a modified pipet
tip attached to a 200-yl Eppendorf pipet (Brinkman Instru-
ments, Inc., Westbury, N. Y.). The blood was expelled from
the microcapillary tube by slowly depressing the pipet
plunger. This blood was expelled into a 5-ml volumetric
flask containing 1.0 ml of 0.02% Triton X-100 solution.
The tip of the microcapillary tube was placed below the
surface of the X-100 solution while ejecting the blood
sample. Then, approximately 100 to 150 yl of the X-100
was drawn into the capillary tube several times to rinse
all the blood into the vial. Deionized water was used to
make the volume to 5ml and the sample was thoroughly mixed.
114
-------
Lead determination was made on this diluted blood sample
by injecting 20 yl into a preprogrammed graphite tube furnace
(IL 455 model) AAS, followed by lOyl of 50ppm Ni as described earlier.
Quantitation of the capillary blood
followed the same "method of additions" procedure outlined
above for venous blood. Figure 29 represents a typical
analytical curve plotted out.
For capillary blood, the daily quality
controls were the spiked human blood used for quantitation
of the samples. This human blood's Pb content was estab-
lished by comparison with CDC bovine blood-Pb samples of
known Pb concentration.
In summary, adapting a finger-prick
blood sampling procedure to quantitate blood-Pb in a non-
exposed population is subject to many hazards because of
the quantity of blood available to work with. The extremely
small volume (100 yl) of blood makes this methodology more
susceptible to contamination since the Pb levels analyzed
are in the ppb range. Due to the absolute Pb concentration
analyzed and the increased possibility of contamination,
the data will be more variable than the venous blood procedure,
Using the very sensitive graphite
furnace AAS technique has allowed this micro method of blood-
Pb analysis to succeed. Increases in the sophistication
of the commercially available heated graphite furnace units
115
-------
Figure 29 . Analytical Curve for Lead in Capillary Blood
tn
H
0)
ffi
s
90 -
80
70
60
50
m m
40
30
20
10
6. 0 jig/100 ml
40 20 0 ZO 40
jig Pb/100 ml
60
80
has moved the finger-prick method of blood-Pb analysis from
a semiquantitative procedure to a fully quantitative one.
This is important in that recent evidence indicates children
may suffer from subacute Pb intoxication and not have a
116
-------
blood-Pb level high enough to measure accurately by the
standard screening techniques. For these children it is
critical to have a quantitative measure of their blood-Pb
concentrations which can be used to evaluate their treat-
ment programs. Since most clinical laboratories are now
equipped with AAS, this makes the finger-prick technique
a rapid and economical means of monitoring these children.
Also for epidemiological studies
involving a blood-Pb monitoring the micromethod of analyzing
capillary blood is both economical and provides better
data with less trauma to the subjects being monitored.
Because commercial instrumentation
has developed to where ppb and ppt levels of metals can
be quantitated the general approach has been to dilute the
sample to where matrix effects are not a problem. This
i
requires very careful collection techniques in taking blood
and also in preparing the sample to prevent excessive amounts
of contamination. Recently, chemical methods have become
popular in reducing these matrix effects to avoid the hazards
associated with trying to analyze diluted samples in the
ppt range.
Studies using 50 ppm Ni solution
along with the sample aliquot in the graphite furnace have
illustrated the feasibility of this technique to retain
Pb while higher ashing temperature are used to remove
117
-------
matrix interferences.
Calculating the volume of blood by
actually measuring the length of blood in the capillary
tube provides a means of compensating for the collection
problem which may be encountered under field conditions.
Studies indicate this will provide accurate results even
when the actual quantity of blood taken is more or less
than lOOyl.
(3) Hematocrits
Hematocrits were determined on all
blood samples at the time of collection (see Sample Collection).
Only one of the two 10ml vacutainers containing venous blood
was opened so an aliquot of blood could be taken for hem-
atocrit determination. This vacutainer was marked so it
would not be used for the initial blood-Pb determination
at the laboratory. It was retained (frozen) as a reserve
sample.
A Clay-Adams hematocrit centrifuge
was used to spin the collected capillary blood tubes and
to measure their hematocrit value.
(4) Free Erythrocyte Protoporphyrin
Blood samples from each participant
were analyzed for Free Erythrocyte Protoporphyrin (FEP).
The method used was based on that described by Piomelli. (24) (25)
118
-------
Blood that was collected from the
participants was analyzed for FEP the day after collection.
The blood was allowed to come to room temperature before
pipeting 20yl into an acid-washed test tube. While the
tube was agitated on a Vortex mixer, 50yl of a 5% "celite
in 0.9% saline solution was added. Two milliliters of
a 4:1 ethyl acetate:acetic acid mixture was added and the
tube agitated for 10 seconds. Following centrifugation
at 2500 RPM for 5 minutes, the supernatant was decanted
into a clean cuvette. To the supernatant, 4 ml of 1.5N
HC1 was added, agitated, arid then allowed to stand in
a dark area for 30 minutes. This allowed the HCl/ethyl
acetate interface to develop. The ethyl acetate layer
was then aspirated with care so that none of the HCl layer
was removed. The sample was then read in a filter fluoro-
meter (Turner Model 430) with the excitation wave-length
set at 405 my and the emission wavelength at 610 my.
Preparation of a standard curve for
each run was based on the method of standard addition.
One participant s sample from each day's run was selected
and used for spiking. Triplicates of 20* yl of blood were
pipetted into five sets of test tubes. These represented
samples with no spike added (the blank) and samples to which
50, 100, 150 and 200 yl of a 1 yg/ml coproporphyrin I
standard were added. The spiking solution was added to
'-119
-------
the blood already in the tube as it was agitated. The tubes
were covered and set aside for 1 hour. After this setting
time, 50 yl of the celite solution was added. The blank
and standards were then extracted and analyzed along with
the blood samples for that day. The amount of FEP in the
samples was computed based on the factor obtained from the
least squares method, then corrected for the hematocrit
of the sample and reported as yg FEP/100 ml RBC.
The source for spiking the blood
was coproporphyrin I obtained from Sigma Chemical (COP-I-5).
To preweighed vials with 5 yg/vial, 5 ml of 1.5N HC1 was
added and then placed in a boiling water bath for 5 minutes.
The vial was removed from the bath and then placed in a
light-tight box to cool to room temperature. The light-
sensitive contents were withdrawn as quickly as possible
and the remaining portion returned to the darkened area.
In addition to the other reagents
used, which had the highest purity obtainable, the ethyl
acetate used in this procedure was cleaned prior to use.
Approximately 500 ml of ethyl acetate was poured into a
separatory funnel. To this approximately 200 ml of a 10
N NaOH solution was added and the two mixtures shaken for
about 5 minutes. The NaOH was decanted and two deionized
water washes followed. The cleaned ethyl acetate was then
stored in an acid washed brown bottle until used.
120
-------
(5) Carbon Monoxide In the Blood
The determination of carbon monoxide in the
blood was performed based on a modification of the method reported
i r~)f.\
by Collisorij et al. Carbon monoxide was released from
the blood sample by the addition of sterox and potassium
ferricyanide into a sample loop previously purged with helium
carrier gas. The released gases were then swept onto a
molecular sieve column and the separated CO was then detected
by gas chromatography.
(6) CDC Blood-Lead Proficiency
Testing Program
Our laboratory began participating
in this evaluation program to verify the accuracy of the
blood-lead extraction procedure (venous blood) used on
this study.
Each month, 3 bovine blood samples
were received in our laboratory from CDC in Atlanta, Georgia.
The blood samples were to contain a low, high, and medium
concentration of Pb obtained from cows fed various quanti-
ties of lead acetate with their feed. The medium Pb con-
centration of these samples was to be approximately 40 yg
Pb/100 ml but this was actually closer to the average low
concentration. The Pb concentrations of the blood were
to be determined and reported by seven reference laboratories.
Other laboratories participating in the program would try
-------
to come within +15% of the reference laboratories' reported
mean Pb concentrations of each sample. Monthly reports
were issued summarizing the results of reference labora-
tories and participating laboratories.
All CDC bovine blood samples were
analyzed for Pb using the procedure outlined earlier for
venous blood and capillary blood samples.
These bovine blood samples were used
to calibrate the spiked blood used daily to quantitate
the venous blood samples (see venous blood analysis).
Also the capillary blood samples (finger-pricks) were
quantitated against these reference blood samples on a
daily basis (see capillary blood analysis).
4. Statistical Procedures
The statistical procedure was to characterize
the environment with respect to airborne lead in resi-
dential areas with traffic densities from less than 1000
to greater than 25,000 cars per day, and also to charac-
terize and compare differences, if any, in blood lead
levels among people who live on these streets. The en-
vironmental parameters that were measured for lead were
air, soil, indoor and outdoor dust, windowsill wipes,
and hand-wipes. Lead in tap water and house paint was
also measured and used as a screening variable. The
122
-------
effects of traffic density, distance from street, inter-
sections, particle size, speed limits and shorter collec-
tion times were also considered.
In general, the blood and environmental vari-
ables were tested either for differences among traffic
sites using analysis of variance (ANOVA) or for a rela-
tionship with individual traffic counts using regression
analysis or, in some cases, using both methods. When
ANOVA was used, the variables were assigned to one of four
traffic density sites: <1,000 (site 1); 1,000-13,500 (site
2); 13,500-19,500 (site 3); and >19,500 cars/day (site 4).
These traffic classes were slightly modified from the
more general classes discussed earlier to conform to
natural grouping in the traffic counts. Histograms were
made for each variable and transformations were applied
if needed to meet the assumptions for the ANOVA. In all
cases which were transformed, base 10 logarithms were used.
When the assumptions could not be met, nonparametric tests
were substituted. The same methods were applied in the
smaller studies using t-tests, paired t-tests, and two-
way ANOVA or their nonparametric analogues.
Regression analysis was used to define the
relationship of some variables to actual traffic counts
123
-------
over the range <1,000 to >25,000 cars/day. Two procedures
were used in interpretation of the regression analyses:
scatter plots were examined and R2 (the amount of variation
in the dependent variable explained by the regression) was
evaluated. Scatter plots of X and Y variables and residuals
were generated in each analysis and were examined for non-
linear relationships or possible transformations to improve
the fit. When no transformations were suggested by the data,
and R2 was very small, further interpretation of the regres-
sion equation and the significance of the regression coef-
ficient was not attempted.
The two blood lead measurements from each
participant were handled in two ways. For the ANOVA, the
duplicate measurements were arranged in a two-way mixed
model design with Samples 1 and 2 as random effects and the
four traffic densities as fixed effects. Variation between
Samples 1 and 2 was then evaluated for bias in every sampling
group and accurate estimates of the variances at each level
could be obtained. The advantage of this method is that
the alternative (averaging the two blood samples) signifi-
cantly depressed the sample variance (P <0.001). For the
regression analyses, however, the intuitively correct ap-
proached is to pair a blood lead value for an individual
with the traffic count at his residence; therefore, the
average blood lead for each individual was used.
124
-------
The raw data for all variables are listed in
the appendices (D-I) or included in the text space (for
the smaller studies). Zeros were used when no measurable
lead could be detected. Blanks were used when no data
was collected.
5. Participant Recruitment
a. General Approach
Volunteer participants who met a set
of age, occupation, race, sex, and residence requirements
were required for successful conduction of this study. The
general approach to participant recruitment in this study
was through use of a house-to-house survey in which volun-
teer participants were recruited. A household survey form
was designed for use in the survey which solicited infor-
mation regarding health and other characteristics of house-
holds along the selected streets. The household health
questionnaire also served to determine if one or more
members of a candidate household were eligible for inclu-
sion in the study as volunteers.
To obtain background information on per-
sons determined to be eligible for the study, a second
portion of the household survey form was designed. With
this individual questionnaire, information regarding per-
tinent characteristics of the individual was obtained for
later use in selecting study participants from those indiv-
duals volunteering for the study and completing an individual
125
-------
questionnaire.
The household surveys, including adminis-
tration of the individual questionnaires, were performed
by survey workers going house-to-house along streets se-
lected for study by the project team. The procedure in-
corporated at each doorway included the following:
approach doorway of house, ring doorbell
or knock
survey worker introduces self and explains
survey
concept of a public health survey regarding
traffic-produced lead is stressed
permission to administer household survey
is determined
household questionnaire is administered
eligibility of members of household is
determined
if no member is eligible, survey worker
thanks resident for information
if one or more member is eligible, survey
worker determines if any would offer their
services as paid volunteers and explains
payment for biological samples
individual questionnaire is administered
to persons who volunteer
volunteers are told they will be notified,
if selected, and arrangements will be made
for sample collection
volunteer families are provided copy of
information sheet on study (shown in Figure
30) .
survey worker thanks resident for information
126
-------
Figure 30.
GENERAL INFORMATION FOR PARTICIPANTS OF
TRAFFIC LEAD PUBLIC HEALTH SURVEY
This public health survey is being conducted by Southwest Research Institute for
the Environmental Protection Agency. The object of this survey is to determine
if residents living near heavily trafficked streets are exposed to undue amounts
of lead. The health survey will be conducted in three stages. The first is a
house-to-house survey of selected residences, using a questionnaire form approved
by the EPA and specifically designed for use on this study. The second stage in-
volves collection of water, soil, and dust samples from residences and blood sam-
ples from individuals who volunteer to participate in the survey. All participants
who provide samples to the study will be paid $25 for their services. In the third
stage, air samples will be taken in the vicinity of residences and traffic will be
counted.
As a volunteer paid participant, the following samples will be collected:
From Residence From Participant
Soil sample(one cup) Questionnaire form
Water sample(one cup) Consent form
Dust sample (window sill wipe) Handwipe from children
Dust sample! srrvt.ll tray) Two blood samples
Examination of sar:ace inside and -One week apart
outside of house for lead -Venipuncture for adults
-Finger prick for small children
Each Participant %vill receive:
(1) Results of screening for lead in paint surfaces of their residence
(2) Summary of results of the overall survey
(3) Direct information if any abnormal blood lead levels are indicated in the
participant
(-<) S25 paid to each participant for their help
Schedule of Activities
(1) You will receive a letter to notify you of your1 selection as a participant
(2) You will be called to confirm your selection and to schedule a visit of the
study team
(3) The study team will visit your home three times to collect samples and to pay
you for your participation
(4) You will receive a letter indicating study results and thanking you for your
help in this important study.
b. Eligibility Requirements
Participant eligibility requirements may
be summarized as follows:
persons who live in residences on streets
selected for study
127
-------
residence at address for 6 months or more
normal occupation does not take individual
away from residence
preschool children (males & females-toddler
or older) females who work at home (ages
20-49) older persons (males & females-ages
50+)
In addition, only persons who are white, non-Spanish, were
to be selected as participants in the study. To optimize
the recruitment efforts required to solicit participants
meeting all requirements, streets were selected for use
in this study which census data indicate are 70% or more
white, non-Spanish. In the recruitment activities, no
effort was made at the survey worker level to bias the
number of volunteers racially, i.e. all races were accepted
as eligible volunteers. Only persons meeting all require-
ments, including race, were later selected as participants
for collection of biological samples.
The distinction regarding persons eligible,
volunteers, and participants should be noted at this time.
Persons are regarded as eligible for the study if study
criteria regarding age, occupation, sex, ethnic background,
and residence are met. Eligible persons who volunteer
are termed volunteers. From those who volunteered, a set
of participants was selected and biological samples were
collected from these participants. In the study recruit-
ment activities, volunteers were sought without reference
to the ethnic requirement (white, non-Spanish); i.e. persons
128
-------
of all races were recruited as volunteers. From those
who volunteered, only persons who met all requirements,
including ethnic, were selected as participants. Partici-
pants who delivered all required biological samples to
the study team were each given $25 as compensation. The
$25 is intended as partial compensation for the services
these participants provided to the public health survey.
This compensation provides an additional incentive for
persons who would be interested in serving the study, but
who otherwise might "let his neighbor" be bothered with
giving blood samples and other samples.
c. First Stage of Household Surveys
In the first stage of survey activities,
20 survey workers were hired, trained, and released to
perform surveys. The assignment for each worker consisted
of a designated street (or streets) between two specified
intersections. The worker was provided with an estimate
of the number of qualified residences in his assignment and
he was given strict criteria for the residences qualified
for survey:
single family dwellings or duplexes
within ]QO feet of roadway
no houses within 100 yards of traffic
signal or stop sign
no corner houses
only houses which face roadway.
129
-------
The design criteria for this study was
based on a compensation of $15 to participants selected
for sample collection. As per the design criteria, the
survey workers made this offer to eligible persons in their
efforts to recruit them as volunteers.
After completion of nearly 1,000 house-
hold surveys, it became apparent that the design estimate
of one participant for every two households was incorrect.
The actual results were closer to one qualified participant
for every four households. Examination of the possible
alternatives indicated that a second stage of survey activi-
ties would be required, and that some changes in techniques
would be necessary if the survey was to be a success inso-
far as recruitment of volunteers was concerned,
d. Second Stage of Household Surveys
A second stage of survey activities was
designed to increase the success rate of volunteer recruit-
ment. For the higher traffic density levels, an immedi-
ate problem was seen: most of the available residences
in Sites 3 and 4 had been surveyed in the first stage of
activities. In order to perform additional surveys, addi-
tional qualified residences would be required. To this
purpose, it was proposed that the restrictions regarding
usage of corner houses and only houses facing the main
roadways be lifted. The Project Officer agreed to the
lifting of these restrictions and thereby provided a sig-
130
-------
nificant new set of residences qualified for survey in
Sites 3 and 4.
Because the percentage of volunteers
per those eligible was viewed to be lower than expected
in the first stage of recruitment activities, additional
incentives were introduced for both the survey worker and
the candidate volunteers. The amount paid to each survey
worker in the stage two activities was raised to $5.00
for each complete participant form delivered to the coor-
dinator. Additionally, the amount offered to the volunteer
for compensation (if selected as a participant) was raised
from the design value of $15 to the amount of $25 for
each study participant providing a complete set of samples
for the study.
The method of assignment of survey areas
to each survey worker was also changed. A complete list
of addresses of qualified residences was compiled by driving
through all remaining candidate areas. Each survey worker
was given a specific set of approximately 60 addresses of
qualified residences. Fifteen survey workers were recruited
and trained for the second set of survey activities and
were each given the assignment of 60 addresses. The neces-
sity for recruitment of preschool children was stressed to
the workers since the first stage of survey activities had
failed in this one group. When qualified preschoolers were
131
-------
determined to be present at a residence, the survey workers
were asked to be particularly persuasive in recruiting that
family as volunteers.
Two additional techniques were introduced
in the second stage of survey activities to increase the
success rate regarding recruitment of volunteers. A letter
designed to introduce the survey worker was placed in the
doorway of residences on a day preceding the actual survey-
ing of the household. A copy of the letter, signed by the
Project Director, is shown in Figure 31. Analysis of reactions
to the household survey form revealed negative feelings regarding
whether or not anyone was at home during daytime hours, as
though they were worried about potential theft. Due to this
noticed uneasiness regarding the form, the sequencing of
questions was rearranged so that Questions 2 and 3 regarding
routine of spending time at home or away from home were
asked later in the interview. Also, a brief set of opinion
questions were introduced in a sampling of cases to determine
if these could soften the questionnaire and generate more
response in those being interviewed. Questions on the sample
opinion survey included the following:
1. Have you received a letter recently informing
you about a public health survey?
2. Are you concerned about air pollution in your
neighborhood?
3. Are you aware that automobile exhaust often
contains lead?
132
-------
Figure 31. LETTER OF INTRODUCTION
SOUTHWEST RESEARCH INSTITUTE
8500 CULEBBA HOAO POST OFFICE OHAWEB 28510 . SAN ANTONIO TEXAS 7328J
May 17, 1976
As you may know, the Environmental Protection Agency is
conducting a public health survey in the Dallas Metro Area. This
survey is being performed by Southwest Research Institute with
the help of the Center for Urban and Environmental Studies at SMU.
The subject of the survey is air pollution resulting from automobile
traffic. As you are probably aware, automobile exhaust often con-
tains lead. The object of the study is to determine if undue amounts
of lead are present in residents living near heavily trafficked streets.
We need your help to accomplish this survey. A member of
our survey team will be contacting you in the next few days to ask
you a. few questions about air pollution in your neighborhood and to
discuss our health study.
Interested persons can participate as paid volunteers to aid
the study. Each participant will be paid $25 for providing blood sam-
ples and samples of water, soil, and dust from their homes. In
addition, air in the vicinity of participating homes will be sampled
and analyzed for the presence of pollutants.
Please remember that public health surveys can only be accom-
plished with the help of interested citizens such as yourself. You can
help us determine the pollution level in your neighborhood. Our inter-
viewer will be contacting you in a few days. If you would like to con-
tact us, call Linda Johnson at SMU, 692-2532.
Sincerely,
Dr. Donald E. Johnson
ANTONIO, HOUSTON, CORPUS CHRISTI, T £ x A S . AND WASHINGTON. O.C
133
-------
Which of the following places do you consider
air pollution to be more of a problem for you
and your family:
a) In the vicinity of your home?
b) Away from your home, i.e., work, while
shopping, driving, etc.?
For which age group do you consider air pollu-
tion to be more of a problem:
a)
b)
c)
Children?
Adults?
Older?
The sample opinion surveys were introduced at the beginning
of the survey for approximately 200 residences. The accep-
tance rate of the survey workers was viewed to be much
higher by the survey workers involved with the sample
opinion survey. Results of the sample survey are shown
in Table 8.
Table 8. Sample Opinion Survey re Air Pollution Concern
1. Received Letter:
2. Concerned:
3 . Aware of Lead:
4. More Problem:
5 . Problem Group:
Yes
No
Yes
No
Yes
No
Home
Away
Children
Adults
Older
% Total
71
29
76
24
93
7
48
52
44
14
42
134
-------
Survey activities were continued by
repeatedly going into areas already surveyed to work the
few residences not reached during earlier rounds. The
survey activities were terminated when the resulting number
of qualified participants was approximately equal to the
redefined criteria of 40 participants per category.
e. Informed Consent
The use of individuals as participants
in a public health study such as that described here re-
quires the complete disclosure of information regarding
the objective of the study, the use of individuals for
provision of study samples, and any risks or potential
of harm, if any, to the individuals as a result of their
participation in the study. An Informed Consent form
was specially designed for use in the study and is present
in Figure 32.
During the initial visit to the resi-
dence of each participant, the details of the study and
any risks to the participant were explained. The Informed
Consent form was provided and a signature was obtained
before initiation of sample collection.
135
-------
32. Volunteer's Informal Consent
SOUTHWEST RESEARCH INSTITUTE
CULEBRA ROAD SAN ANTDNlO, TEXAS 78228
VOLUNTEER'S INFORMED CONSENT
I,
residing at
hereby acknowledge and certify to the following:
i. That I hereby volunteer and consent to participate as a human
test subject in an experiment identified as 'Epidemiologic Study of the
Effects of Exposure to Automobile Traffic on the Blood Lead Levels of
Persons in Selected Age Groups' which is designed to determine the extent
of exposure to environmental pollutants.
Z. That I have been given, in my opinion, an adequate explanation
of the nature, duration and purpose of the experiment, the means by which
the experiment will be conducted and any possible inconveniences, hazards,
discomforts, risks, and adverse effects on my health which could result
from my participation therein;
3. That I have been informed of all appropriate alternative procedures,
if any exist, that might be advantageous to me;
4. That I understand my questions concerning procedures which
affect me will be answered fully and promptly;
5. That I understand that I have the right to withdraw my consent
and to discontinue participation in this experiment at any time without prejudice
regardless of the status of the experiment and regardless of the effect of such
withdrawal on the objectives and results which the experiment is designed to
achieve; and I also understand that my participation in the experiment may
be terminated at any time by the investigator in charge of the project or
the physician supervising the project regardless of my wishes to the matter;
6. That I hereby understand and agree that the samples collected
from me will be analyzed for lead, FEP, hematocrit and carbon monoxide
and that these are the only tests that will be made on these samples and
that no medicinal compounds will be analysed.
136
-------
Figure 32. Volunteer's Informed Consent (cont'd)
7. That I attained the age of years on my last birthday
which was ^^^^ , and that I
am executing this Volunteer's Informed Consent as my free act and deed.
Executed , 19
Volunteer
Signature of Person Informing Volunteer
and Obtaining Volunteer's Consent
Executed in the presence of:
Signature of Witness
(If volunteer is a minor, parent or guardian must complete the following.)
On behalf of , I hereby
(insert volunteer's name)
consent to and approve his/her executing the foregoing consent and participating
in the above-identified experiment as a human test subject.
Signature of Parent/Guardian
Date: .
Executed in the presence of:
Signature of Witness
137
-------
III. RESULTS
A. Determination of the Relationship Between Air
Lead Levels and Traffic Flow Characteristics"
1. Results of Traffic Density Mini-study
The objective of this study was to examine the
relationship of air lead to traffic counts. Air lead was
measured concurrently with traffic counts for several 24-
hour periods at each of 17 locations. These locations were
selected to cover the range of less than 1,000 to greater
than 25,000 cars per day. The air lead concentrations and
traffic counts obtained are presented in Table 9. For
purposes of comparison and analysis, the data are also grouped
into four traffic densities and the mean, standard error
(SE), and number of days of observation (sample size) are shown.
Table 9. Mean Air Lead Concentrations and Traffic Counts at
each Location and Traffic Density.
Air Pb
Mean 1 SE *
Days
M>rli,e
oat.! Mcaa
Mlm.,,-1
Bai IIL- a M ridge
UlulltieW
Miduiv KillM
1'rj.rie Creek
Gates
Inwood
Lo^urs Lane
Moukin^bl rd
Royal
Forest
Jim Miller
Marsh
Coit
N. \V. llvfcy.
Sue 1 1, 000
Site 2 1, 000
Site 3 1 3, 500
Site 4 -19. 500
0. 53
0. hO
0.89
0. 63
1. 07
1. 09
0.83
0. 91
1. 21
1. 48
1.20
1. 10
0. 93
0..15
1.10
1.25
0
0
0
0
0
0
0
0
0
0.
0.
0.
0.
0.
0.
1 . 8b 0.
cars /day
- 13, 500 cars
073
099
103
097
131
109
145
1 10
1 14
1 30
21 3
104
111
097
313
228
/day
3
5
7
9
10
14
16
17
17,
17
18,
20,
23,
37,
117 -
31o -
)7I
578 1
24t>±
1Z9 1
340 I
761 '-
480 t
128
042 -
102 !
740 I
004-'
001 J
311 =
531 !
7
4
10
257
51 D
295
433
509
1245
1182
1912
LOI4
343
841
2175
1042
10
7 Site l«
8
9
7
7 Site 2
8
16
1 1
1
8
7 "
9
9
10
6 Site 4
19
All Sites
- 19. 500 cars/day
cars/day
Air Pb
Mean *SE
u. 67 0.059
0. 89 t Q. 058
Mean Traffu
Level
1.07 * 0.060 16,737
1. 54 ! 0. 148 30, 1)85
1. 06 r 0. 049
** Standard Error
The data for mean air lead vs mean traffic count
for each location has been plotted and is shown in Figure 33,
Over all sites, the mean air lead was 1.06 yg/m3 and the
138
-------
mean traffic count was 17,330 cars per day. The mean air
lead vs mean traffic density is also plotted for the four
traffic density sites. An apparent relationship between
traffic density and air lead is readily detected in Figure 33
ri qure 33.
Hit
'.EAD
LEVEL
u9/m3)
AIR LEAD LLVELS BY TRAFFIC COUNT
r»0.65911 to.0263 X
= mean for each street
A= mean for each traffic site
12 Lfi 20 24 2B
Traffic Density (thousand cars/day)
A total of 152 air lead measurements with corresponding
traffic counts were taken and are listed in Appendix D.
Using the method proposed by Mickey et al.,' ' outlier
cases were tested with regression analysis. One outlier
(4.93 yg/m3 - NW Hwy.) was rejected with the result that
the variable air lead was normally distributed (with outlier
included, skewness = 2.38, P = 0.02; without outlier,
skewness = 1.20, P = 0.23). The frequency distribution
of the variable air lead is given in Figure 34. Airlead
139
-------
Figure 34. Frequency Distribution of the Variable Air Lead.
ANl) COM'-MllAriOfr, iVHICI' FOLLOW EXCLUDE VALUES LISTED AHOVE
'1 10POINTS
>. 1W>
'). AIM)
?. "550)
2. 4iH1) **
2.25PO**
2. UH1) **
I . 9W)***
I . Tjiid) «**
I . 6Sfl) * ********
I . Si)H) ********
I .3c>Pi) **********
I .2iHM **************
I. )S0)'1 **************
.}. ''Cl^) * ** * ***********************
H. 750) ******* *********
}. V]0) * ******* **********
1. 1')P)) *** ***** ****
'I. 3l)PI) *******
I. I 50)*
.). -.111(1)*
-'). I "id)
;>;om> MHANS ART lya-ioiTD w »."s ir IIIPY COINCID^ rnni *'s, N's o-
i . -;i M
!' . 4 ? 7
(J) The values 4.93 and 3 . 56 were excluded so that narrower class intervals could be used.
values were regressed on traffic counts with the result
that 32.3% of the variation in air lead was explained.
The regression equation given below was significant at
P = 1.2 x 10"11.
Y = 0.6598 + 0.0263X
where
X = 24-hour traffic count/1,000
and
Y = air lead (yg/m3)
140
-------
Examination of scatter plots of X and Y and residuals also
indicated that the variable air lead was normally distri-
buted and independent.
Further examination of the scatter diagrams in-
dicated that most of the increase of Y(air lead) with X
(traffic count) was in the 30,000 to 40,000 cars/day range.
To verify this, air lead was regressed on traffic counts/
1,000 with the highest location (NW Hwy.) omitted. This
location included almost all of the traffic counts above
30,000 cars/day. The result was that only 15.1% of the
variation in air lead was explained. The regression equa-
tion for air lead at these traffic densities was Y = 0.7059
+ 0.0214X. There could be a nonlinear relationship, but
the range of traffic counts prohibits its estimation (i.e.,
plots do not give any suggested transformation).
Conclusions: We conclude that on streets with
traffic densities less than 25,000 cars/day, there is a
small but significant relationship between air lead and
traffic counts. Furthermore, as soon as traffic levels
are increased to the range of 30,000 to 40,000 cars/day,
the relationship improves markedly- The relationship is
estimated by the equation Y = 0.6598 + 0.263X, where X is
thousands of cars per day and Y is air lead, (yg/m3).
2. Results of Replicate Hi-vol Mini-study
The objective of this study was to determine the
141
-------
reproducibility of air sampling data from replicate samples,
For this purpose, replicate air samples were taken each
day for 10 days at one location (NW Hwy) by placing two
air samplers side by side for the 24-hour period. The lead
concentrations of the 10 pairs of samples, their means,
and standard errors are listed below. The air lead levels
for the replicates on each day are shown in Figure 35.
AIR
LEAD
LtVELS
3.0
2.0
1.0
FIGURE 35. REPLICATE AIR SAMPLERS
2 3 4 5 6 7 8
DAY MUHDER
in
142
-------
A paired comparisons t-test showed that there was no sig-
nificant difference between lead concentrations in the two
air samples taken each day (t = 0.73, P = 0.48); however,
there was considerable variation over days for each air
sampler. This can be explained by the fact that these samples
include the outlier value (4.93 yg/m3) discussed in the
traffic density mini-study above. Means and standard errors
were calculated omitting the outlier and its replicate with
the result that the standard errors were similar to those
in Table 13. The standard deviation divided by the mean
or coefficient of variation (CV) which is a relative
measure of variation was calculated for these two sets
of samples as well as for the sixteen locations in Table
13 (one location had only one sample). The CV's ranged
from 0.26 to 0.61 and these two samples had midrange CV
values (0.41 and 0.43). Using a paired t-test the air
samples from the remaining nine days were also not signi-
ficantly different (t = 0.13, P = 0.90).
Air Lead (yg/m3)
Air Sampler 1 Air Sampler 2
*4.93 4.11
2.33 2.42
2.06 2.09
3.56 3.35
1.51 1.26
1.91 1-55
2.32 2.38
0.93 1.09
0.82 0.92
1.89 2.30
143
-------
Mean ± SE 2.23 ±0.388 2.15 ± 0.320
N 10 10
**Mean ±SE 1.93±0.262 1.93±0.274
9 9
* - Outlier (P «0.001)
** = Outlier excluded
3. Results of Particle Size Mini-study
The proportion of lead particles which are in
the respirable range in suspended air lead at heights above
the ground breathable by adults and children is important to
the relationship between blood lead and an airborne lead
source. It has been estimated that one-half to two-thirds
of particulate lead emissions are in the respirable range
(0.1 y to 0.5 y) , (-*) an(j that a density decline gradient
in airborne lead concentration exists up to 200 feet from
a highway.( Using <1.0 y as the upper size limit for
respirable particles, analyses were made to determine the
proportion of suspended lead measured 1 meter above the
ground in each of five particle size classes; whether these
proportions changed over distances of 5 feet (1.5m), 25 feet
(7.6ra), 50 feet (15m), and 100 feet (30m) from the street;
and the density decline gradient, if any, in respirable and
non-respirable size ranges. This study was made in two
cities (San Antonio, Texas and Dallas, Texas) on streets
selected to have traffic densities of approximately 25,000
cars/day. Both sets of data are presented in Tables 10 and 11
144
-------
Table 10 shows the air lead concentrations
in each particle size class for one air sample taken at each of
five distances from the street in San Antonio. The upper half
of the table shows the lead concentrations (yg/m3), and
the lower half shows the lead concentration expressed
as a proportion of the whole sample. The proportions in
each size class remain approximately the same as the distance
from the street increases. Approximately 50% of the lead
in the sample is in the respirable range regardless of the
distance from the street. The proportion of lead in each
fraction vs. distance from the street is plotted in Figure 36.
Table 10- Lead Concentration (Ug/ma) in Five Particle Size
Ranges at Four Distances from the Street
San Antonio, Texas
Distance
(Feet)
5
25
50
100
5
25
50
100
Mean
SE
>
0.
0.
0.
0.
p
0.
0.
0.
0,
0.
0.
7. 0
8175
4417
2882
612Q
3
0
0
0
0
roportion
185
101
153
244
171
0299
0
0
0
0
0
0
Particle Size (u)
.3-6.9 2.0-3.2
. 5216
.6624
. 1939
. 2749
of Lead
. 118
. 152
. 103
. 110
. 121
. 0109
0
0
0
0
in
0
0
0
0
0
0
. 4664
. 4053
.2830
. 2142
Each Size
. 106
. 094
. 151
. 085
. 109
. 0146
1.
0.
0.
0
0.
0-1. 9
3360
3112
1 123
2479
<
2
2.
1.
1.
1. 0
2775
5472
0003
1593
Total
4. 4190
4. 3718
1. 8777
2. 5092
Range
0.
0.
0.
0.
0.
0.
076
071
060
099
077
0082
0.
0.
0.
0.
0.
0.
515
583
533
462
523
0250
1. 000
1. 001
1. 000
1. 000
145
-------
FIGURE 36
PROPORTION OF TOTAL LEAD FOU!!D IN EACH PARTICLE SIZE FRACTION
vs. DISTANCE FROr< STREET
.4-
Proportion of
Lead in Eacii
Particle Size
Range
.3
5 25 50
DISTANCE FROM STREET (Feet)
>7.n
3.3-6.P
1.0-1 .9
2.0-3.3
ion
Table 11 shows the same information from the Dallas study.
The two filters from the respirable range at 50 and 100
feet were lost and the results for the distances 5 feet
and 25 feet are given. Again, more than half of the lead
is in the respirable range with the remainder distributed
146
-------
Table 11- Lead Concentration (L'g/m3) and Proportion in
Five Particle Size Ranges at Four Distances from the Street.
Dallas, Texas
Distance
(Feet)
5
25
50
100
>7. 0
Particle Sizes (u)
3.3^7.0 3.0-3.2 1.0-1.9 <1. 0
0.2224 0.2073 0.1295
0. 1290 0. 0999 0. 0707
0. 1163 0. 1074 0.0667
0.0871 0.1147 0.0382
Total
0.1209 0.5461 1.2262
0.0811 1.6987 2.0794
0.0646
0.0616
Proportion of Lead in Each Size Range
5
25
50
100
Mean
0.
0.
0.
181
062
122
0.
0.
0.
169
048
109
0. 106
0. 034
0. 070
0.099 0.445
0.039 0.817
0. 069 0631
1. 000
1. 000
among the other four size classes. The total air lead
concentrations vs. distance from street for San Antonio
and Dallas are plotted in Figure 37.
In Table 12, the lead concentrations at each
distance are expressed as the percentage of the lead con-
centration at 5 feet from the street for the nonrespirable
and respirable size ranges. There is no perceptible decline
in the first 25 feet; in fact, the Dallas sample at 25 feet
had a higher lead concentration. (This variation is not
unexpected in a sample size of two.) At 50 feet and beyond,
the lead concentration drops to approximately one-half the
147
-------
Ai r lead
FIGURE 37
A:R LEAD C;'!.':CE:ITRATIO;: vs. nirT/>'!CE FPW STREET
'Dallas
_ - ~"ran Antonio
25 50
DISTANCE FROM STREET (Feet)
TOO
Table 12. Concentration of Lead Suspended in Air at Increasing
Distances Expressed at % of Lead at 5 Ft. from Street.
i 1. 0 u < 1. 0 U
Distance (Ft. ) Nonres pirable Pb Respirable Pb
Total Pb
San Antonio
5
Z5
50
100
Dallas
5
25
50
100
100%
85. 2
41. 0
63. 0
100%
56. 0
52.2
44. 3
100% 100%
111.8 98. 9
43.9 42.5
50.9 56.8
100% 100%
311.1 169.6
---
148
-------
concentration at 5 feet and is fairly consistent at this
level in these three samples.
Conclusions: Approximately 50% of the airborne
lead concentration is in the respirable range (< 1.0 y)
at distances from 5-100 feet from the street. In the non-
respirable and respirable ranges, lead concentrations did
not decline in the first 25 feet from the street but were
approximately one-half the street level at 50 feet and
at 100 feet.
4. Results of Distance from Road Mini-study
In the previous section, a density decline
gradient in air lead concentrations at distances up to
200 feet from the street was discussed. Air samples and
traffic counts were taken for two days at three traffic
densities at four distances from the street: 5 feet (1.5m),
25 feet (7.6m), 50 feet (15m), and 100 feet (30m). The
air lead concentrations in yg/m , and also expressed as
a percentage of the air lead at 5 feet from the street,
are given in Table 13. At every traffic locality there
is a decline in air lead concentration with increasing
distance from the street. Air lead levels vs distance
from street for each locality are plotted in Figure 38.
The decline seems to be more rapid at the higher traffic
counts (32, 761 and 34, 645), but the sample size for
this study is not large enough to make general statements
regarding decline rate and air lead levels.
149
-------
Distance
(Feet) Actual Traffic Counts (Cars/Da^
5
25
50
TOO
9360 '1538
2.15 1. 04
1.35 0. 86
1.17 0. 59
0.74
16, 886
1. 38
1.65
1.22
0. 47
18,
1
1
1
0
1Z3
. 82
. 37
. 02
.99
Table 13. Concentration of Lead Suspended in Air (Ug/ma at Four
Distances from the Street for Two Days at Three Traffic Densities.
I, 761 34,645
3. 24 3. 14
1. 48 1. 85
1.22 1.46
1.09 1.02
Concentration of Lead Suspended in Air Expressed as the
Percentage of Lead at 5 Ft. From Street
Actual Traffic Counts (Cars/Day)
5
25
50
100
rapidly in the first 50 feet from the street to levels of
about 55% of the street concentration. At 100 feet from
the street, the lead levels were less than 40% of the
concentration at the street.
5. Results of Intersection Mini-Study
Two different intersection studies were done.
The first study measured air lead for four days at each
of five combinations of block and intersecting street traffic
densities: 10,000/1,000; 10,000/5,000; 15,000/1,000;
150
9360 9588
100% 100%
62.8 82.7
54.4 56.7
34.4
Conclusions :
16, 886 18, 123
100% 100%
119.6 75.3
88.4 56.0
34. 1 54. 4
32,761 34645
100% 100%
45.7 58.9
37.7 .46.5
33.6 32.5
Ave. %
100%
74. 2%
56.6%
37.8%
Air lead concentration declines
-------
A1r Laad
Concentration
(up/n-3)
1-
38
AIR LEAP CONCENTRATION-: AS f FU?!CTIP»! "F DISTANCE
FROM P.OAf AMD TRAFFTC OFNSTTY
3.588
16,886
25 50
DISTANCE FROM ROAD (FEET)
100
25,000/5,000; ana 25,000/10,000 cars/day. The purpose of
this study was to determine whether air lead levels were
higher at intersections than at neighboring midblock loca-
tions. The intersecting streets were selected to add from
1,000 to 10,000 cars/day to the midblock traffic levels.
At the time this study was designed, we did not expect to
recruit participants for human studies from homes near
intersections.
The second study was designed when it
became necessary to recruit participants from corner homes
(discussed in Section III B 1). The purpose of this study
151
-------
was to determine whether those participants who lived near
1,000 car/day intersecting streets were exposed to addi-
tional air lead. Four traffic density combinations were
used: 1,000/1,000; 10,000/1,000; 15,000/1,000; and 25,000/
1,000 cars/day- For clarity, we will refer to this study
as the corner home study.
Air lead levels (ug/m3) with actual traffic
counts at the five traffic densities in the intersection
study are given in Table 14. In Figure 39, mid-block ys
intersection air lead levels were plotted to examine the
correspondence between these locations within a block and to observe
whether these air lead levels were related to increasing traffic density.
Each point on the plot is a letter which defines its traffic density group;
A = 10,000/1,000
B = 10,000/5,000
C = 15,000/1,000
D = 25,000/5,000
E = 25,000/10,000 cars/day-
The line drawn at 45° merely indicates which of the two
air leads determining a point is higher. Those points below
the line have higher air lead at midblock than at inter-
sections; and those above the line have higher air lead
at intersections. The graph implies an approximate 1:1
correspondence between air lead at blocks and intersections
152
-------
Figure 39. Intersection Study: Air Lead Levels at Intersection
and Midblock Locations. Traffic Densities: A = 10,000; B = 10,000/5,000;
C = 15,000/1,000; D = 25,000/5,000; E = 25,000/10,000 cars/day.
an
o
fl
32
0 D
1 2 3 if 5 6 7
Air Lead at Midblock (ug/m1)
for traffic densities A-D(less than 30,000 cars/day). In
addition, 50% of the points at these traffic densities
fall below the 45° line indicating air lead is not higher
at these intersections. There seems to be little differ-
entiation between air lead levels at traffic densities less
than 20,000 cars/day as indicated by the lack of clustering
of groups A, B, and C. This agrees with our findings in
153
-------
traffic density mini-study (section III A 1) where the
relationship between air lead and traffic density increased
substantially only after traffic levels rose above 25,000
cars/day. Points labelled D and E on the graph are at
traffic levels above 30,000 cars/day and indeed have higher
air lead levels. The graph also implies that above 30,000
cars/day the relationship between midblock and intersection
air lead changes; however, three of the midblock air lead
levels at density E seem unusually low compared to other
midblock air leads at similar traffic densities (see Table
14). As in the traffic density mini-study, the data are
scarce at levels above 25,000 cars/day and few conclusions
about trends can be drawn. The statistical significance
of differences in air lead demonstrated in this graph was
tested using a two level nested ANOVA with the result that
significant differences were found among traffic densities
(P = 5.0 x 10" ) and between midblock and intersection
locations (P = 1.1 x 10~4). The variances of the ten groups
were homogeneous.
Air lead and traffic density levels for the
corner home study are listed in Table 15 and are plotted
in Figure 40. Again a line is drawn at 45° to assist visual
differentiation between air lead levels which were greater
at midblock (below the line) and those greater at inter-
ections (above the line). Air lead levels were greater
154
-------
Table 14. Intersection Study: Air Lead Concentrations at
Intersections and Midblock Locations.
Air Lead (ug/r.3)
Block
1.44
1.05
.98
1.96
2.61
1.61
1.88
1.39
3.23
1.87
.90
3.69
4. 15
3.47
4. 47
3.88
4.02
1.63
1.31
1.99
Intersection Block Intersecting Street Traffic Density
1.65
2.02
1.09
1.91
2. 18
2.56
2.98
1. 44
2.46
2. 19
1.25
2.73
3.48
2. 19
3.46
2.70
5.77
4.30
5.04
6.80
7,794
9,971
9,970
9,804
10,960
10,994
11, 446
12,259
12,698
12,858
14, 114
14,252
22,818
23, 189
23,429
23,431
23,607
27,7-40
28,226
30, 199
261
1,632
1,803
1.697
5,671
6,908
6, 184
6,781
461
437
427
433
7,646
7,819
8, 105
7,901
7.918
10,337
10,239
11,220
A
10,000/1,000
B
10, 000/5, 000
C
15.000/1,000
D
25,000/5,000
E
25,000/10, 000
Mean Air Lead (yg/m3)
Block
A 1.36
B 1.87
C 2.42
D 3.99
E 2.24
Intersection
1.67
2.29
2. 15
2.96
5.48
at intersections at 8 of the 14 pairs of locations. There
seems to be no clustering of traffic density levels A, B,
C, or D. For this reason and the fact that two density
D samples were lost, a paired comparison t-test was used
to test for differences between air lead levels at inter-
sections and midblock locations. Variances were tested
155
-------
Table 15. Corner Home Study; Air Lead Levels (yg/m3)
at Intersections and Midblock Locations
Air Lead (Ug/m
Block Intersection
.22
.89
1.19
.12
1.64
. 81
1. 06
1.30
.73
1.42
.58
.63
1.19
1.42
.41
.33
.86
.98
1.01
2.15
1.17
2.29
.99
1.72
.71
.58
.90
1.07
Block Intersecting Street Traffic Density
121
146
390
393
3,980
8,651
11,266
12,712
11,467
12,411
16,567
16,629
21,324
23,883
120
140
292
245
199
728
1,229
996
200
146
831
952
249
246
A
1,000/1,000
B
10,000/1,000
C
15,000/1,000
D
25,000/1,000
Mean Air Lead (ug/m3 )
Block
Intersection
A 0.61 0.65
B 1.20 1.66
C 0.84 i.OO
D 1.31 0.99
Overall Mean
0.94 1.08
Standard Error of Mean
0.122 0.158
and were homogeneous. No significant difference was found
between air lead levels at these locations at the 0.05
level (t = 0.90, P = 0.39).
Conclusions: Two intersections studies were
done. In the first, significant increases in air lead
levels were found among traffic densities which ranged
from 8,000 to 40,000 cars/day. In the same study air lead
levels were significantly higher at intersections than at
in midblock locations primarily due to differences at the
156
-------
Figaro 40. Comer Home Study: Mr Le.id Levels at Midblock and Intersections
locations. Traffic Densities: A = 1,000/1,000; B 10,000/1,000; C = 15,0000/1,000;
D = 25,000/1,000 cars/day.
1.0 2-0
Air Lead at MidbJock (uq/m't
3. 0
highest traffic density. The second study tested for an
increase in air lead levels for participants in this study
who lived near 1,000 car/day intersecting streets. No
significant differences in these midblock and 1,000 car/day
intersections locations were found.
157
-------
6. Results of Speed Limit Mini-study
The objective of this study was to determine the effect of
traffic speed on the amount of air lead for a given traffic leve. Air lead
concentrations (vig/m ) were measured in two speed zones: 30 MPH and 45 MPH.
Air samples and traffic counts were taken for five days each on a 30 MPH
and on a 45 MPH street which had similar traffic densities. Although an
attempt was made to locate two streets with the same traffic density, the
mean traffic count on the 45 MPH street (9369 cars/day) was 19% higher than
the mean for the 30 MPH street' (7879 cars/day) . We would expect lower air
lead levels at the faster traffic speed due to the shorter time that the
vehicle is present to emit lead. Higher traffic counts at the faster speed
could cause elevation of air lead levels thereby reducing the difference in
air lead between the two speed zones. This effect, if it exists, will make
the difference in air lead over speed zones more difficult to detect. The
air lead levels and associated traffic counts and means are given in Table 16.
Table 16. Air Lead Concentrations (ug/m?) at Two Speed Limits
(30 and 45 MPH).
30 MPH
45 MPH
Air Lead
(UB / m3 )
0 . 8 '-1 8
0.650
0. 844
0. ^47
1. 860
Traffic
Count
7433
7665
8570
8054
7672
Air Lead
( "s/nV3)
0. 827
0. 424
0. 623
0. 598
0. 219
Traffic
Count
9853
9624
8994
9198
9176
Mean 1. 040
7879
0.538
9369
Air Lead (ug/m3)
30 MPH
45 MPH
Geometric
Mean
0.97
0.49
Li
0.60
0.26
LZ
1.58
0.93
158
-------
A two fold difference can be seen between the
mean values for air lead, with the higher level being
associated to the 30 MPH site. It will be noticed that
one value at the 30 MPH site is significantly higher than
the others. Even without this data point (1.860), the mean
value for the 30 MPH site is still 0.835 yg/m3 vs 0.538
for the 45 MPH site. Comparison of geometric means also
shows a factor of two between the two sites.
To examine the statistical significance of these
results, a paired t-test was used with log transformation
to achieve homogeneity of variance. No significant difference
was found between the air lead levels at the two speed
zones (t = 1.31, P = 0.07), although a two fold difference
can be seen between the geometric means (0.97 and 0.49).
The large variation in these samples is reflected in the
size of the confidence limits plotted in Figure 41.
Conclusions: A large apparent difference was
noticed between air lead levels on 30 MPH and 45 MPH
streets with the lower speed having nearly twice the air
lead of the higher speed. However, the statistical con-
fidence was not established in this result, perhaps due
to the limited sample size.
159
-------
Figure 41
Air Lead Concentrations vs Two Speed Limits
2.0
1.5
T31.0
O
l-l
H
<
0.5
L 1
-i- L 2
L 2
-L L .1
30 mph
45 mph
7. Results of Indoor vs outdoor Air Lead
Mini-study
It was of interest to know whether a significant
portion of the lead in outside air near the street was
present inside homes. Air lead measurements were made
both inside and outside the same house simultaneously for
a 10-12 hour period. As these measurements were made using
standard HI vol samples run between 50 and 60 cfm the distinct
possibility exists for the sampler to have acted as a vacuum and
therefore produced artiflcally low air lead levels. Sixteen
of these pairs of measurements were made at two different
traffic densities (9 at 10,000 cars/day, and 7 at 20,000
cars/day). These data with means and confidence limits are
given below and are plotted in Figure 42.
160
-------
Indoor vs Outdoor Air Lead (yg/m3)
10,000 cars/day
20,000 cars/day
Indoor
*0
0
0
0
0
0
0
0
0
.34
.65
.30
.07
.09
.10
.08
.34
.19
Outdoor
0.
3.
0.
1.
1.
0.
1.
0.
87
17
01
69
70
12
81
32
74
Indoor
0
0
0
0
0
0
0
.12
.23
.27
.07
.12
.25
.55
Outdoor
1.
2.
0.
1.
2.
2.
3.
70
17
98
92
31
46
87
Arithmetic
Mean 0.24
Geometric
Mean
0.18
0.10
1.16
0.92
0.50
N
0.34 1.68
9 9
0.23
0.19
0.10
0.36
2.20
2.05
1.40
3.01
7
* Evaporative Cooler
161
-------
12VJ
FIGURE 42
INDOOR vs. OUTDOOR AIR LEAD CONCENTRATION:
AT TUO TRAFFIC DEIJHTIES
3.Or
2.5
2.0-
.0"
I
95%
CONFIDENCE
LIMITS
indoor outdoor indoor outdoor
JO,000 20,000
cars/day cars/day
All of these houses had central or window air conditioning units
with one exception (marked with an asterick) which had an evaporative
cooler. A two-way ANOVA (fixed effects) was used to test the effects of
indoor vs outdoor air lead and 10,000 vs 20,000 cars/day.
162
-------
Prior to a log-,g transformation, the variances of the four
groups were heterogeneous; therefore, log transformed data were
used for the ANOVA. There was no significant difference in the
traffic densities (P = 0.10) nor was the interaction term sig-
nificant (P = 0.14). The difference between indoor and outdoor
air lead levels was highly significant (F = 61.70, P = 1.5 X 10~'
The sampling variation is often large in small
samples such as these. For example, the standard deviation for
these data is 1.11, roughly twice 0.61, the standard deviation
of the 152 air lead measurements used in the traffic density
study- For this reason, differences among means which seem
large may not be significant, i.e., the difference between
outdoor air lead means, 1.16 and 2.20. Referring to Table 9, we
see that air lead means at traffic densities similar to 20,000
cars/day vary from 0.85 to 1.25 ug/m with 6-10 replicates per
mean. Mean air lead for these 25 samples was 1.05 yg/m with
mean traffic count 20,076 cars/day. Interpretation of a 2X
difference in outdoor air lead in this mini-study would be in-
valid in view of the large variation here and the findings of
the larger air lead/traffic density mini-study (IIIA1).
Conclusions: There were highly significant dif-
ferences between outdoor and indoor air lead with outdoor air
lead 5 to 10 times higher. There were no differences between
indoor samples taken on a 10,000 cars/day street and those
taken on a 20,000 cars/day street. Outdoor lead samples
at the 10,000 vs 20,000 cars/day were very close to but not
significantly different at the 0.05 level.
163
-------
8. Results of Indoor vs. Outdoor
Dustfall Mini-study
Outdoor dust samples (28-day dustfall) were taken at 10
locations and each sample was paired with an indoor dust sample either
from the same residence or within the same area. The lead concentrations
of these pairs of samples and actual traffic counts and their means and
standard errors are given in Table 17 and are plotted in Figure 43. One
outdoor dust lead (0.7252) was tested as an outlier using Dixon's r-,0
Table 17. Outdoor Dustfall Lead Concentrations (ug/cma) from Ten
Locations with Corresponding Indoor Dustfall Lead Concentrations
and Traffic Counts.
Traffic Count
474
6654
10637
15156
16219
16381
17452
20483
20928
31542
Outdoor Dust
Pb (ug/cma)
0. 2000
0. 0827
0. 0865
0. 0945
0. 0811
0. 1586
0. 0913
0. 0392
0. 7252-
0. 2121
Indoor Dust
Pb (i£_/cm9
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0026
0092
0025
0604
0016
0128
0104
0034
0047
0071
Mean.SE Meanj.SE
0. 121£f±0. 01779 0. 0122"±0. 0062
N=9 N=9
- Rejected as outlier (P<<0.005)
(28)
test, found to be significantly higher than the other sam-
ples (r1Q = 0.80, P « 0.005), and rejected. If this outlier
had not been rejected, the results of this analysis would have
been the same but with a lower level of significance, because
retaining it makes the variance considerably higher. The
lead concentrations of the 9 remaining pairs of samples
164
-------
FIGUJC 43
LEAD IN OUf-T (INDOOR AND OUTDOOR) vs. TRAFFIC r>E?lMTv
Lead
Concentration
Dust Samples
15000 2nnni
TRAFFIC DENSITY
were conpared using Wilcoxon's signed ranks test and were found to be sig-
nificantly different (P<0.005). Indoor vs. outdoor dust lead is plotted
in Figure 44. The mean outdoor dust lead (0.1218 yg/cm2) was approximately
10 times the mean for indoor dust lead (0.0122 yg/cm2). When outdoor dust
lead was compared with the mean of all 268 indoor dust lead samples dis-
cussed in section III B 3 (0.0082 yg/cm2), the ratio was approximately 15:1.
Different collection methods were used: indoor dust samples
with a tray and outdoor samples with a bucket. Both are described in the
Methods section. The reason for this difference was to avoid possible
effects of wind outside. Rain had no effect on outdoor.samples since rain-
water was simply held in the bucket (see Section III C 2).
Conclusions: Outdoor dust lead concentrations from 9 locations
were found to be significantly higher (at least 10X) than indoor dust lead
concentrations in adjacent or nearby residences.
165
-------
FI'MiE 44. IIIDOOR vs. OUTDOOR flUVT LF./IO COUCENTRATIOns AT » MtTCI'ED LOCATIONS
I'-d'-.rr rt'ist Lead
CT.cir.trMtirr.
(un.'c.i-)
.10 .15 .20 .25
miTiywp OUST LE'.n cn'iCENTR'TtoN (uo/ci"2)
9. Results of Collection Times Less
24 Hours Mini-study
Twenty-four hour collection times for air lead
and traffic density have been used for most of the studies
described in this report. This mini-study was designed to
determine whether shorter sampling times could be used to
measure air lead as efficiently as 24 hour sampling periods.
Air samples and traffic counts were taken for two days each
at three traffic densities for following shorter time periods:
1 hour, 2 hours, 4 hours, and 12 hours. One location was
used for each traffic density (10,000, 15,000 and 25,000
cars/day). Twenty-four hour samples from the same street
which were begun either on the same day, the same day of
166
-------
the week, or a similar day of the week were used.(Midweek
days were not considered similar to Mondays, Fridays, or
weekends.) The number of hours sampled, date sampling began,
traffic counts, and air lead levels are listed in Table 18.
Table 18.
Collection Times Less Than 24 Hours: Air Lead and Traffic Counts
for Five Collection Times at Three Traffic Levels.
Traffic Count Range of
i Hours Date Traffic Projected Air Lead Air Lead
Sampled Began Count to 24 Hr. (uglm3) for each time Estimated Traffic Level
1
' 1
2
' 2
4
' 4
12
'12
24
'24
1
1
2
2
4
4
12
12
24
24
1
1
2
2
4
4
12
12
24
24
9/24
9/25
9/24
9/25
9/24
9/25
9/24
9/25
9/24
9/25
9/29
9/30
9/29
9/30
9/29
9/30
9/29
9/30
10/1
9/23
10/17
10/24
10/17
10/24
10/17
10/24
10/17
10/24
10/21
10/22
1.
2,
1,
7,
7,
9,
9,
1,
1,
2,
3,
9,
9,
12,
14,
1,
1,
2,
2,
5,
5,
17,
io,
22,
23,
483
139
110
653
244
908
624
453
970
971
534
671
158
338
941
086
004
422
693
252
374
282
671
718
396
949
314
903
313
429
11
4
13
7
13
11
15
14
9
9
14
16
13
16
17
13
18
18
12
14
32
30
32
32
35
35
34
33
22
23
,592
,536
,320
,396
,464
,448
,248
,916
,970
,971
,016
,104
,896
,056
,646
,516
,003
,384
,698
,252
,976
,768
,052
,616
,376
,694
,628
, 806
,318
,429
2.
1.
0.
0.
0.
0.
1.
1.
0.
1.
3.
2.
1.
1.
1.
0.
1.
1.
3.
3.
2.
6.
2.
7.
1.
7.
1.
Q ,
4.
4.
23
61
74
01
91
65
31
35
93
OS
19
30
40
24
40
76
77
19
23
69
00
91
33
28
83
23
29
92
15
47
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
4.
4.
5.
5.
0.
62
27
26
04
07
39
16
64
58
46
91
95
35
63
32
10,000 cars/day
15,000 cars/day
25,000 cars/day
Shows effect of sampling began at 10 A.M.
Others began at 3-9 A.M.
Also listed are all of the traffic counts adjusted to a
24 hour time period and the range (or difference in this
case) between the two air lead samples for each time period.
On a given day the 1, 2, 4, and 12 hour samplers were begun
at the same time (8-9 A.M.). On the following day or the
same day one week later, the samplers were begun again at
the same time. There is an exception at the 10,000 car/day
167
-------
location where samplers were begun on 9/25 at 10 A.M. in-
stead of 8-9 A.M. As a result the 1, 2, and 4 hour traffic
counts are depressed by 60%, 41%, and 15%, respectively,
when compared to the corresponding traffic counts on 9/24.
Comparison of the two 12 hour counts shows that the dif-
ference has diminished to 2% after this length of sampling
time, and there is no difference after the 24 hour sampling
period. Thus longer sampling minimizes the effects of the
exact time when sampling began. Air lead levels did not
seem to be affected by the late starting time; however,
the results of the other mini-studies show that air lead
is not especially sensitive to differences in low traffic
counts. The difference in starting time would probably
have caused a difference in air lead at a 30,000 car/day
location.
Traffic counts were overestimated when sampled
for less than 24 hours as indicated by the lower 24 hour
counts in the column for traffic counts adjusted to 24 hour
time periods. This is expected since the shorter time periods
were during the daytime.
The range in air lead decreases as sampling time
increases at all three traffic locations (see Table 18).
This implies that longer sampling periods have lower var-
iances although two samples are not adequate for drawing
conclusions about variances. At the 25,000 car/day location,
168
-------
the range or difference between duplicate times is unusually
large. Comparison of air lead levels and sampling dates
shows that for times 1, 2, 4, and 12 hours, air lead levels
on 10/17 are 2.00, 2.33, 1.88, and 1.29 and on 10/24 are
6.91, 7.28, 7.23, and 6.92 yg/m3. Traffic counts for du-
plicate samples are similar. The consistently lower air
lead levels on 10/17 are probably the result of some meteoro-
logical condition, i.e., a change in wind direction, rain,
or air stagnation. It is not possible to determine the
exact cause at this time. The air lead levels vs. traffic
counts (adjusted to 24 hour level) are plotted in Figure
45 with each of the three locations encircled. The dif-
ferences in dates 10/17 and 10/24 at the 25,000 car/day
Injure !'.». Ml lu.i.l U-vi'ls v;;Ti,»flu UriMly toi I ixii mil- -I i
h-vi-Li, ( - lO.uilO, * IS,mill; 2'j, '10(1 ,\tr:> 'd.r, ) .
. Illl,-.' Vl ll I 1
(,0'>u I il, 'Hin I 1 , nun | ft ,
.'.l-lli'in '1 i ,il I i .' 'ui.i ' H 'I ' l
ii, nun l-l, nut)
169
-------
location are prominent. The smaller ranges of the 24 hour
samples as opposed to the 1, 2, 4, and 12 hour samples are
seen in the 10,000 and 15,000 car/day locations. The low
outlying point at 2280 cars/day reflects the 10 A.M. start-
up time. No relationships such as higher or lower estimates
of air lead associated with length of sampling times can
be seen.
Conclusions; Shorter collection times tend to
overestimate traffic volume and introduce more variability
in air lead measurements. The range in air lead measure-
ments increased with shorter sampling times at all traffic
densities. No relationship between air lead levels and
length of collection times was seen.
B. Determination of the Relationship Between
Blood Lead Levels and Traffic Density
1. Results of Recruitment Activities
a. Number of Subjects
(1) Original Design
The original design criteria for
participants included the following numbers of qualified
volunteers in each age, sex, and traffic density category:
Traffic Density
Site 1 Site 2 Site 3 Site 4 Total
Preschoolers
Male 30 30 30 30 120
Female 30 30 30 30 120
Females that 30 30 30 30 120
work at home
(20-49 yrs)
Older persons 30 30 30 30 120
(50+ yrs.)
Male & Female
170
-------
(2) Modified Design
Even with additional incentives
and techniques, the results of recruitment of volunteers
was less than completely successful in the second stage
of survey activities. This is particularly true regarding
the recruitment of preschoolers.
Because of the severe problems of
meeting the required numbers of preschoolers, it was pro-
posed to the Project Officer that the number be revised
to the following specifications:
Traffic Density
Site 1 Site 2 Site 3 Site 4 Total
Preschoolers
Male & Female 40 40 40 40 160
Females that
work at home
(20-49 yrs.) 40 40 40 40 160
Older Persons
(50+ yrs.)
Male & Female 40 40 40 40 160
480
The Project Officer agreed to this change as being acceptable
and within reasonable probability of attainment, seeing the
results at that time.
(3) Volunteers Obtained
Results of the survey recruitment
activities are presented in Table 19. A total of 1850
171
-------
TABLE 19.
RECRUITMENT RESULTS
Parameter
No. HHQ Forms
Total No. Residents
Ave. No. Residents/HH
Total No. Eligible*
Total No. Volunteers
Ave. % of Volunteers
(of those eligible)
Traffic Density
Site 1
545
1452
2.66
523
201
38.4
Site 2
485
1283
2.64
533
213
40.0
Site 3
409
1026
2.51
242
119
49.2
Site 4
411
1130
2.75
301
184
61.1
Total
1850
4891
2.64
1599
717
44.8
* Number eligible obtained from Household Questionnaire form; no
reference to race on HHQ form.
household surveys were performed for households with 4891 residents. A
total of 1599 residents were encountered who were eligible regarding age,
occupation, sex, and residence of which 717 volunteers were recruited. Of
the 1850 residences, 986 or 53% of the total reported one or more persons
eligible for the study (without regard to race). From the 986 residences
reporting one or more eligible person, 476 or 48% of the total provided
one or more volunteers for the study. The total of 717 volunteers from the
986 residences reporting one or more eligible persons yields a rate of .72
volunteers per residence with one or more eligible persons. This response
rate is considered to be very good for a residential study involving free
living (rather than institutional or isolated) populations.
The Household Questionnaire Form used in this study makes no
reference to race or ethnic background. In the recruitment of volunteers,
no reference was made to any requirement for race or ethnic background. The
neighborhoods selected for the study were selected so as to maximize the
172
-------
potential for finding white, non-Spanish volunteers. The
number of volunteers obtained, shown in Table 19, include
some who were later ineligible due to the requirements for
participants to be white, non-Spanish. From these numbers,
the average percent of volunteers of those listed as eligible
is 44.8%. It is expected that this number would be higher
if the statistics accounted for only white, non-Spanish.
Because no reference is made to race in the HHQ form, and
the number of eligible persons is determined with use of that
form, it is not possible to calculate the precise number.
A second factor is also reflected
in the statistics presented in Table 19 regarding the per-
cent of volunteers. The volunteer rate was significantly
higher in the second stage of recruitment due to the in-
creased efforts directed at recruitment and additional re-
cruitment techniques applied. The second stage of recruit-
ment was concentrated on the higher traffic densities
(Sites 3 and 4) and the percent volunteering reached greater
than 60% as opposed to 40% for Sites 1 and 2.
An analysis has been performed to
compare household characteristics for volunteers versus
persons eligible who did not volunteer for this study.
Results are presented below. Five pertinent characteristics
of the households were used in the comparison: (1) Median
education level of the household head, (2) Occurrence of
173
-------
Comparison of Household Characteristics:
Volunteers vs. Eligible Non-Volunteers
Household
Characteristics
Median Education
Level of House-
hold Head
Occurrence of
Lead Screening
Presence of
Air Conditioning
Median Age of
Structure
Median Length
of Residence
Households of
Eligible Persons
Who Volunteered
(Col 53>0,Col 70>0)
Level 6 - Partial
College
1.9%
94.7%
21 years
7 years
Households of
Eligible Persons
Who Refused
Col 53>0,Col 70=0)
Level 6 - Partial
College
1.8%
95.5%
22 years
11 years
lead screening for one or more members of the family previous
to this study, (3) Presence of air conditioning (central
or window unit) in the home, (4) Median Age of the structure,
and (5) Median length of residence at the address. For four
of the five characteristics, the volunteers are remarkably
similar to those refusing to volunteer. On the fifth charac-
teristic, length of the residence at the address, a sizeable
difference is noticed in the median values: 7 years for
volunteers, 11 years for eligible non-volunteers. This
difference may be interpreted to mean that those who were
174
-------
the more established residents (having lived in the neigh-
borhood longer) may be more reluctant to volunteer than
those with less length of residence in the neighborhood
who are more mobile by definition and perhaps more open to
change or approach by a study such as this. The conclusion
of this comparison is that little or no bias is indicated
due to any differences in characteristics of volunteers
versus those eligible who did not volunteer.
In summary, the average number of
eligible persons per household encountered was 0.86 and
the average number of persons volunteering per household
was 0.39, or approximately one for every 2.5 residences.
The average number of volunteers per residence with one or
more eligible persons is 0.72, and is considered high. The
average number of eligible preschoolers per household was
0.10 and the average number of preschool volunteers per
household was 0.067, or approximately one for every 15
residences. The low number of preschoolers is considered
to be at least in part due to the high traffic densities
involved in the study. A strong bias exists for families
with small children not to locate on busy thoroughfares.
Characteristics of the eligible population who volunteered
are very similar to characteristics of eligible non-volunteers
175
-------
b. Participant Selection
(1) Number of Participants
Accomplished
The number of participants for which
samples have been obtained is presented in Table 20 by age,
sex, and traffic level. Of the persons volunteering for the
study, only those meeting all criteria, including the
ethnic criterion of white, non-Spanish were included as
participants. The numbers presented in Table 20 accounts
for dropout of volunteers who were selected but who sub-
sequently were unable to participate (refused, moved, etc.).
Table 20. Number of Participants by Age, Sex, and Traffic Level
Traffic
Level
1
2
3
4
Total
Total
130
113
117
82
442
M
35
35
28
22
120
F
95
78
Sy
60
322
1-8 yrs
M
20
17
6
11
54
F
24
15
19
9
67
19-49 yrs
M
0
0
0
0
0
F
50
41
48
39
178
50 yrs.
M
15
18
22
11
66
F
21
22
22
12
77
176
-------
Ot the total 442 participants selected
for the study, the number of preschool participants totaled
121, females 13-49 years totaled 178, and older persons
totaled 143. These numbers compare to the design criterion
of 160 persons in each category. It is apparent that re-
cruitment of the younger and older participants was less
successful than the middle group. The worst statistics
were obtained for the highest traffic density level. Site
4, where a total of 20 preschoolers and 23 older persons
were recruited. The design criterion was 40 persons in
each of these categories.
(3) Selection Criteria
From the volunteers recruited, a
set of participants was selected which best met all the
study criteria for age, sex, race, economic level, occu-
pation, langth of residence, and traffic density- For
each of these parameters a set of guidelines was established
for minimum restrictions and for selection of bias where
more than sufficient candidate volunteers were available
177
-------
for a given category of participants.
(a) Age
The age groupings for partici-
pants were adjusted to miximize usage of the volunteers
available for selection. The minimum age for older persons
was shifted from 60 years to 50 years due to insufficient
members of volunteers in the 60 and over group. A small
number of females who work at home were selected with
ages less than 20 years due to multiple members of family
being selected.
(b) Sex
For the preschool children,
the design criteria included separate categories for
males and for females. Because of the difficulties encoun-
tered in recruitment, the separate categories for males
and females were combined to a single category and volun-
teers were selected on the basis of age rather than age
and sex. For Site 1, sufficient volunteers were available
to select equal numbers of males and females.
178
-------
(c) Race
Of those persons volunteering
for the study, only white, non-Spanish persons were selected
to participate in the study.
(d) Economic Level
Study volunteers were obtained
from areas of the city designated as middle class by use
of census records and by direct observation of the apparent
economic level of study neighborhoods.
(e) Occupation
The occupation of volunteers
was restricted to those who were routinely occupied at
home. For persons partially occupied away from home, a
guideline cut off level of 20 hours per week was estab-
lished and applied to candidate volunteers. If the candidate
volunteers spent more than 20 hours a week (half-time of
a normal 40 hour week) at an occupation (work, school,
nursery, etc.) away from home, the volunteer was not accepted
as a participant.
(f) Length of Residence
A minimum of 6 months residence
was required of volunteers selected as participants.
179
-------
(g) Traffic Density
Candidate volunteers were select-
ed for each traffic density site on the basis of estimated
traffic densities determined from maps and existing traffic
records. Traffic on each of the streets from which volunteers
were selected was counted during the study activities.
The traffic counting activities and results are documented
in Section IV. D. These actual traffic counts were used
to replace the estimated counts used to initially assign
candidate volunteers to specific traffic density sites.
Thus, the final assignment of study participants to specific
traffic density levels is based on measured rather than
estimated traffic counts.
(h) Excess Volunteers
Where an excess of volunteers
was available, selection of participants was based on two
premises. First, for preschoolers and persons over 50
years, an equal number of males and females was desired.
Secondly, in all age classes, an even distribution of ages
was desired; i.e., equal representation of all ages within
the category. Participants were selected to meet these
goals as much as possible from the volunteers available.
For preschoolers, no surplus
was available from the volunteers. All qualified volunteers
were selected. For females in the middle group, all with
preschool children enrolled as participants were selected
180
-------
first. Selection of the remainder was based on an even
distribution of ages. In the older group, no excess of
males was available. No excess of older group females was
available in Site 4; all qualified volunteers were selected.
In Sites 1,2, and 3, older females were selected on the
basis of the best distribution of ages.
(3) Participating Households at.
Intersections
An analysis of the number of par-
ticipating households at intersections is presented in Table
21. Of the total of 280 participating households, 55 were
at intersections or a total of 19.6% of all residences.
Table 21. Number of Participating Households at Inter-
sections
Parameter
Total Number
of Households
Contacted
Total Number
of Partici-
pating House-
holds
Total Number
of Partici-
pating House-
holds at Inter
sections
5 Participat-
ing House-
holds at
Intersections
Traffic Densitv
Site 1
545
76
5
6.6
Site 2
485
70
7
10.0
Site 3
409
79
29
36.6
Site 4
411
55
14
25.4
Total
1850
280
55
19 .6
181
-------
The number at intersections varied considerably from lowest
6.6%, Site 1) to highest (36.6%, Site 3). A significant
difference is also noted between the lower two traffic
density sites (average of 8.2%) and the higher two traffic
density sites (average 32.0%). This difference is explained
in that more recruiting occurred in the higher two sites
during stage 2 of the study. It was also taken into account
that location on an intersection with very low traffic density
is less important for thoroughfares with very high traffic
densities (Sites 3 and 4) than for streets with less traffic
density (Sites 1 and 2).
2. Description of the Study Participants
a. General Description
In the study reported herein, the par-
ticipants who were selected and who actually provided the
study with biological and other environmental samples can
be generally characterized as white, middle class residents
of a very urbanized community. Participating families have
heads of household who are educated to the level of partial
college education, as a median value. Of the adult parti-
cipants, 32% are classified as smokers.
b. Paint Lead Restriction
Participants were selected without regard
to paint lead at the outset. Biological and other samples
were collected from all participants. The measurement for
182
-------
paint lead was analyzed after sample collection activities
were complete. Participants with paint lead measurements
in excess of 1% lead (1-4 mg/cm2) were excluded from statis-
tical analysis if the lead levels for these individuals
were found to be elevated. Following this criterion, data
were eliminated for four female participants and no male
participants.
c. Demographic Characteristics
The demographic characteristics of selected
participants is shown in Table 22. These characteristic
patterns are representative of the volunteers excluded
from the study as well as those selected as participants.
The excluded volunteers were of essentially the same dem-
ographic structure as the participating volunteers. From
Table 22 it may be seen that for the characteristics of
number of persons routinely at home, median education, median
length of residence, and median hours per week away from
home, the participant characteristics are remarkably con-
Table 22. Participant Demographic Characteristics
; , Persons Routinely
Traffic . :iumoer or ; i Smokers ' at home
-evel ; omct;?rs . lAdults) ' median
! I 23 26.4 ; 2
2 33 39.3:2
3 32 j 34.3 ' 2
4 13 ' 23.1 ' 2
r^tai. 10: 32.5 j 2
avo.
] .9
L.8
1.8
1 . 6
1.8
Median
Education
6
5
*
6
6
Median
Length of
Residence
6 yrs .
6 yrs .
6 yrs .
5 yrs.
6 yrs .
Median hrs/week
away form home
10
10
10
10
10
What is -re highest educational level completed by your head of
housencld?
\ i) less than 3th grade (5) trade or vocational school
(2; 3th grade beyond high scnool
\2" iigh sor.oo I-incomplete ; 5} college-incomplete
4) r.ich school-complete (7) col Lege (-4 years) -complete
! 8) post graduate
183
-------
sistent for the four traffic density sites. The percentage
of smokers does vary significantly between the highest
39.3%, Site 2) and the lowest (26.4%, Site 1). Sites 1-4
are sufficiently similar in demographic characteristics
to serve as comparison groups for use in the study.
3. Environitiental__Data
a. Soil
Soil lead concentrations (ug/g) from
outside 277 residences were examined to determine their
relationship to traffic density. The frequency distribu-
tions of soil lead over all sites and within the four traffic
densities (sites) are shown in Figures 46 and 47. Soil
lead concentrations were significantly skewed to the right
both in the overall distribution (skewness = 39.6) and in
the four sites (skewness = 9.5, 20.4, 5.2, and 6.8 for sites
1-4, respectively). The critical value for skewness at
P = 0.05 is 1.96. After a base 10 logarithmic transforma-
tion, the data were not significantly skewed (overall
skewness = -0.85, skewness at sites = 0.13, -0.45, -0.33,
0.92). A single classification ANOVA was used to test for
differences among the four sites using a log transformation
to meet the assumptions for this test. There were signi-
ficant differences in soil lead among the four sites at
P = 0.008, F = 4.0. Multiple comparisons tests (Student-
Newman-Keuls or SNK procedure) were done to determine at
184
-------
Figuru 46. Frequency Distribution of Soil Lead.
1-1 ). 1.Vi>
M 1. 1.1(1)
/"> )..1'.)(1>
/ ' t.ll'lll) *
"i- 1 i. lilt))
i'\ !..)ilH) +
*> ' 'i. 1(1(1)
'i.V1. .)<)<*)*
1'1 I. 1'1H) *****
M i. 1.10)**
-1") l.-1'.ICI) ****
i')... lllO) ***
I? l.'IO'l) ***V
'! )..HH1) ********
.-"1 i. 'HUD ***********
?''.). KJO) **************
I ft i.(K)O) **+***+* + ******************< * **
1^ ;.OH0) '! ******* A-*******************-******-******-******
;i i.')H(') t*-**-v-***-* v********* ****** *+**** tvt-v**********-***1* *********/ H
^ J..HH1) *-*r>VA"**tA-t **************************:*********************/ I
1. '1 ) ',-*' A--1-*
-4 J. 01)11)
- ( 1. HIM)
J-JOIJP T-^NS \r?'- DTNfrrrn nv '"i ir ni^y concin:- .vim *'s, ri'.s 'inii-nwi
MAXIMUM
UNIMUH
I27.56H
IIW.424
2~l 1.
Fiquru 47. Fruqucncy DisLihution of Soil Lc.id dt cai.-h Yrnftlu
Sill: I Sill: ' Sill 1
0-M..U1W)
rt.l J.DD15)
/c> I.-UIM)
'>!.). t hill I
'>.-! I..MB)
. I.-IU) *
..)»())*
..M(»>*
. -MM) «*
K4) *****
****
M"«
..HIM)
./IfrllD
i1 MI.AHS HIIR Di-nrirtn nr
HI-. Ml III.V/4
:i. .)(». l-"3.i-H
I /'J.u.n)
M1XIMIM ;iw.nvn
UN IMUM 6. lkl;t
mi r cniMcinH KITH *'s,
I
-------
which sites soil concentrations were different. The SNK
procedure is a posteriori test which delimits sets of means
which are not significantly different but are within a
larger set of means which are significantly different.(29)
The four sites are listed below in increasing size of the
mean (column 5), and nonsignificant sets of sites are under-
lined. The back-transformed means, their confidence limits
(LI and L^), and sample sizes (N) are also given. The rela-
tionship between these means is shown graphically in Figure 48,
Figure 48. Soil Lead vs. Traffic Density.
150'
'00-
Soi1 UvJ
Concentration
50
9.3°' Con'i lence Lirits
Site 1
Site 2
Site 3
Site 4
Geometric
Mean
73.62
92.26
110.92
105.93
Site :
L
59.
76.
94.
87.
i
74
21
44
92
Site 2
TRAFFIC
L2
90.73
111.69
130.27
127.62
Site 3
DENSITY
N
75
68
79
55
Site <:
Nonsignificant
Sets of Means
1234
-
186
-------
The means for soil lead in Sites 3 and 4 (>13,500
cars/day are significantly higher than in Sites 1 and 2 (<13,500 cars/day);
and Sites 2, 3 and 4 (>8,000 cars/day) are significantly higher than Site 1
<600 cars/day) . Although the mean for Site 3 is higher than the mean for
Site 4, the difference is not significant. When the means are ranked in
increasing order (as shown above), they are also in order of increasing
traffic density (with the exception of Sites 3 and 4 in which soil lead
concentrations are approximately the same).
The relationship between soil lead and actual traffic
counts for the residences can be further described by the regression
equation: y = 117.60 + 0.80 X, where y = soil lead (yg/g) and X = traffic
count/1000. This regression equation must be interpreted with caution;
however, since only 0.5% of the variation soil lead has been explained
(R2 = 0.0049). When R2 is small (<0.04), the variance around the re-
gression is almost as large as the original variance of y (soil lead) and
interpretation of the regression is inadvisable. ' ' Examination of
scatter plots and residuals did not suggest any transformations for this
range of traffic counts. The soil lead concentrations and corresponding
traffic counts are listed in Appendix E.
A study was performed to determine how
selected soil characteristics would effect the availability
of lead deposited on soil from auto emissions in the study
areas. This study was conducted by an independent soil
chemist and the detailed report is given in Appendix J.
187
-------
The soil characteristics examined were
texture, clay minerology, organic content, and pH. The
texture of the soils varied but a majority contained a clay
content which would favor retention of divalent Pb deposited
from auto emissions. This clay texture makes the soils
less permeable to rain water and would minimize the amount
of downward transportation of Pb from the top soils. This
would also favor runoff from the top soil layers during
heavy thunderstorms.
A majority (80%) of the soils had moderately
high to high organic contents. Due to the complexing and
chelating agents normally present in organic matter, it
is reasonable to believe these soils would favor retention
of Pb.
Most of the soil was within a pH range
of 7.0 to 8.0. This along with the high carbonate and
sulfate ion content of the soils in this area favor the
formation of insoluble Pb carbonates, phosphates and sulfates.
These compounds are not likely to be leached from the soils
thereby favoring retention of the Pb. About 20% of the
soils examined showed signs of alteration of the natural
texture of the soils. This is thought to be the mixing
of fluvial sands to improve the plant growth characteristics
of the soils. The addition of this sand does not favor Pb
retention on these particular soils.
188
-------
The soil samples studied in the Dallas
county-Arlington area indicated retention of Pb deposited
on them would be favored and leaching by environmental
factors (i.e., rainfall) would be minimal.
Conclusions; There were significant dif-
ferences in soil lead concentrations among the four sites
with Sites 1 and 2 (lower traffic density) significantly
lower than Sites 3 and 4 (higher traffic density). No
linear (or nonlinear) relationship could be described for
these data. In the study of soil characteristics, the
majority of the soils were high in clay content and in
organic matter. These types of soils have a high potential
for adsorption and storage of lead and formation of rela-
tively insoluble precipitates of lead. The clay soils re-
tain lead deposited on and adsorbed by them; however, low
permeability of the clay may reduce initial infiltration
during heavy rainfall.
b. Tap Water
Tap water samples were taken from 271
residences for determination of lead content (yg/ml.).
Because drinking water is a source of lead, it was necessary
to determine the exact levels to which the participants
of this study were exposed and whether these lead levels
varied over the four traffic density sites. The frequency
distributions for the variable water lead over all sites
189
-------
and for each site are shown in Figures 49 and 50. The data
are very skewed to the right because of the large number
Figure 49. Frequency Distribution of 'i'np Water Lead.
i U)"0 HIT'S
.) IH17)
1105)
HIM) **
MHD*
11)2 ) +
imp) *****
oil,?)*
idl ) **
'111 I ) ***********
Mill ) ** + ************* + >* V*******
.HI I ) .", *+ v***t*********fr *****************
Hid) ***>****
.HI I ).", ** v***t*********** ****************
KH1) * ** t* ^** * t'VV*^***-^ ********* v **-**+*** t:-V* ************* *-****R A
'I'll) H-H(-**1rJ-**-t '*'***'* t ******* t*^+*t ** + + ****************** KM
ICC)
"1.00177
? n.
'1.!1dV,0
U. II
of very low values (103 of the total 271 were below the
detection limit of 0.0002 yg/ml and are recorded as zeroes).
Transformations of the data did not correct the skewness;
therefore, a nonparametric test (Kruskal-Wallis one way
ANOVA) was used to determine whether there were differences
in water lead among traffic sites. The results (Table 23)
indicate that lead content is similar in Sites 1, 2, and
4; and the significant difference is caused by low lead
content in Site 3. Therefore, no adjustments were made
for additional sources of lead in water. Means and 95%
190
-------
Figure 50. Frequency Distribution of Tap Water Lead at each Traffic Density.
SITE 1 SITE 2 Sll K i
SHE 4
1.-IH7)
).'.M6)
I.D06)
J . 006) *
1.M06)
). .Mb)
1..H-M)
l./KM)* *
1.CK14) *
J.MM3)
J.0I13)
'!. ll^)
). 002) ** **
.1.1)02)*
^. 0'.1 I ) **
.1.00 I ) **** **
H. ',101 )********* *****
J.0HI ) W ************ »(***
1.000) ******************* ********************
-). 000) *********************24 *********************3I
CJI'IOUI' MEANS ARE DENOTED BY M-"5 IF THEY COINCIDE HITH *"5
'(HAN 0.00052 0.00049
"i i)EV. 0.00067 0.00100
i'l 75.000 67.000
M\XIMUM 0.00430 .5.00580
MINIMUM 0.0 0.0
*
*
*******
********
i( ******* *************2 8
*********************"! 2
, N'S OTHERWISE
0.00030
0.00042
77.^00
0.00280
").0
*
*
*
****
*****
«******
*****************
****************
0.00057
0.00097
52.000
0.00620
0.0
confidence limits at each site and over all sites are shown
in Figure 51. The mean lead concentration in tap water
from all sites was 0.00046 ± 0.000047 pig/ml.
Table 23. Test for Differences Among Sites in H20 Lead Concentration (ug/ml)
Using Kruskal-Wallis Test
Site
Site
Sice
Site
1
2
3
4
N
75
67
77
52
Rank
Sum
11280.
8249 .
9571.
7-755.
Mean
5
0
0
5
0.
0.
0,
0.
,00052
.00049
.00030
.00057
SE
0.
0.
0.
0.
of Mean
0000778
0001224
0000474
0001349
Kruskal-Wallis test statistic = 7.99*, Prob. = 0.045
All Sices: Mean ± SE = O.C0046 ± 0.000047 pg/ml
191
-------
Figure 51. Tap Water Lead Concentration vs. Traffic Density.
.0010
.0008
.0006
Tai: Vata- Lsad
^or. 23'" tT£ti c11.
':'*'.'"' I
.0004
.0002
Site 1 Site 2 Site ?
TRAFFIC DENSITY
All-sites
Conclusions: Lead content in drinking
water was similar in Sites 1, 2, and 4 and lower in Site
3. No adjustments were made for an additional source of
lead via drinking water.
192
-------
c.
Indoor Dust
Indoor dust samples (28-day dustfall) were
taken at 268 residences. In some cases, the dust tray
could not be collected on the 28th day; therefore, all dust
lead values were adjusted to a 28-day base. The adjusted
dust lead concentrations (yg/cm2) and the number of collec-
tion days are given in Appendix E. Frequency distributions
of the variable indoor dust lead over all traffic densities
(Figure 52) and for each site (Figure 5.3) were examined.
The mean indoor dust was highest in Site 2 and a relation-
ship with increasing traffic density is not apparent. Re-
gression analysis explained only 0.7% of the variation in indoor dust lead
(R2 = 0.0065); y = 0.007Q + 0.0001 X, where y = indoor dust lead (yg/cm2)
Figure 52. Frequency Distribution of Indoor Dust Lead.
Ml POINTS
'1M4)
.'(W) *
176)
,572) *
:i68)
164)**
..16(1)*
.'S6>
.148) *
ir
I I 1.1)
>* V-V-V** u -*
: IT***
t******* V******* | | H
iif.-.j or !:n n,Y '""5 II7 T'lf-'Y
//IT1! *T>, N'S OTII^'WIS^
" \ X
\:I 0.0082
1|-'V. 0.0112
I 268.0000
I"!'" 0.0802
r"ii'i o.o
193
-------
Figure 53. Frequency Distribution of Indoor Dust Lead it each Traffic Density.
SITE . SITE 2 "SITE 1 SITE 4
^ i nro I NTS
J.HSH)
.
l.H7ci>
J.H72)
J..164)
O.H6H)
I.J4B)
.J.J16)
J.J2H1*
1.1)2'!)**
O..I2B)
1.H Irt)***
I..112)**
1. -I i)H) ***********
}.'H)4) M **********
OTOlli''MEANS *RE DENOTED BY M'S IF THEY COINCIDE KITH *>S, N'S OTHERWISE
4FAN 0.0051 0.0102 O.OOM 0.0077
S 1EV 0.0054 0.0144 0.134 0-0079
-I 73.0000 67.0000 76.0000 52.0000
HXIMIIM 0.0285 0.0736 0.0802 0.0488
HNII1UM 0.0 0.0010 0.0 0.0
and X = traffic count/1000. Scatter plots did not suggest
any transformations. The means and 95% confidence limits for
each site and all sites are shown in Figure 54.
Conclusions: No detectable relationship
between traffic density and lead concentration in indoor
dust samples was found.
d. Windowsill Wipes
Windowsill wipes were taken at 258 resi-
dences and the lead content of each was measured. Frequency
distributions of the variable windowsill wipe lead over
all sites and at each site are shown in Figures 55 and 56.
Two outliers were rejected: 1.9540 in Site 2 and 4.6857
194
-------
Figure 54. Indoor Dust Lead vs Traffic Density
.015
.010
.005
Site 1 Site 2 Site 3 Site 4
Traffic Density
in Site 3 which were 5.5 aiii 13.3 standard deviations above
the overall mean (P« .001). Means and confidence limits
of each site and over all sites are shown in Figure 57.
195
-------
Figure 55. Frequency Distribution of Windowsill Wipe Lead
HXCIJIDF.n
VAMII-S(l)
* \Anui. AT IONS Ann oo'iPUTArinr^ wurcn FOLLOW EXCLUDF VALUES I.ISTEO
i .nco)
1.0511)
i. ^iici) **
.1. ri50)*
i.a1"'^)**
0.750)
1.7(1(1)*
l.rt(IO)*
H.550) **
"1.500) *
1.450) **
1.4MO)**
1.350) ****
'1.25(5) ****
'4.2HO) *********
>'1. I 5(1) ***************
1. U10) M **********************
'
- ldOO)
-.1:****** ******** ***************************************** I C14
-1 HIM)
GRo:jp '(FVJS AR^ npNornn nv M'S IF THEY coiMcinp wini *'s, N'S OTHERWISE
'/RAN ',t.W>2
0. I 55
56.0t10
M.9I I
MNIMU'l ;.>:
(1) Excluded values arr 1.9540 (Site 2) and 4.6857 (Site 3).
Figure lj\t. Froquoncy Distribution of Vrindowaill v/ipo Load at each Traffic Density.
SiTE I SITE 2 SITE 1 SITE 4
..I. .
TABULATIONS AND COMPUTATIONS WHICH FOLLOW EXCLUDE VALUES I ISTFO AROVF
1.1150)
I.. 1110)
1.950)
I.9H0)
?. 1)50)
M. HUB)
I). 75B>
1.7(10)
H.A50)
H.450)
'I.. 1(10)
.1.2H0)* **** , M
.1.1110) ******* JJ".. **..,.*
). 050) M ******************* ********************* *************+*******p>1 **********
-i'.0((10) *************»*******T5 ******************** **********-t^*.*A.h.t.*.b.fc,hin j.^j.^.j.x^.j.j.
-J.J50)
-«.IH0)
GROUP MEANS ARE DENOTEO M M'S IF THEY COINCIDE KITH 'S, N'S OTHFBXISF:
MEAN 0.052 a.||2 a.103 0.105
S. JEV. 0.072 n.179 0.168 H.179
'.',.. 6*-"?i; 6I.B00 74.01)0 52.HIM)
MAXIMUM 0.414 ^.B98 0.871 JI.9II
(1) 1.9540, (2) 4.6857
196
-------
Figure 57. Windowsill Wipe Lead vs.Traffic Density.
.20
Window 5111
Dust Lead
Concentration
.10-
.05-
Site 1 Site 2 Site 3
TRAFFIC DEIISITY
Site 4 All-Sites
Examination of the means r_.reals no relationship between window-
sill lead and traffic density. Scatter plots did not indicate
any appropriate transformations, and only 1% of the variation
was explained by regression of windowsill lead on traffic counts
(Y = 0.0718 + 0.0016 X, where Y = windowsill lead and X =
traffic count/1000). Mean lead content in windowsill wipe lead
concentrations with corresponding traffic counts are given in
Appendix E.
Conclusions: No significant relationship could
be found between lead in windowsill wipes and traffic density.
1:17
-------
e. Hand-wipes
The lead measured in hand-wipes from 122
children was examined with respect to traffic counts and
blood lead (see Appendix F). The rationale for hand-wipe
analysis is that small children playing outside will come
in more direct contact with lead in soil and outside dust
and can transfer this lead into their bodies by putting
their hands into their mouths. The frequency distribution
of the variable hand-wipe lead over all sites and for each
site are shown in Figures 58 and 59. Hand-wipe lead values
were regressed on traffic counts with the results that only
a small amount of variation (6.2%) in hand-wipe lead was ex-
plained (y = 7.35 + 0.20 X, where y - hand-wipe lead and X =
traffic count/1000). The relationship between blood lead and
hand-wipe lead was also examined but only 0.3% of the varia-
tion in blocd lead was explained by lead on children's hands.
The means and confidence limits for hand-wipe lead (yg/cm2)
over all sites and at each traffic density site are given in
Figure 60; the mean hand-wipe lead concentration over all
sites was 9.3 + 0.65 yg/cm2.
Conclusions: No substantial relationships
were found between lead in hand-wipe samples and traffic
counts or between blood lead and hand-wipe lead.
198
-------
Figure 58. Frequency Distribution of Hand-wipe Lead
lJ: MINTS
-} I .000)
-I::. HH'.i)
.>. -v.
,';))
CT>
-u
AXIMIW
RY
9.379
7. I 73
I2I.
41.410
IF THiiY
:?v- ,vii!^ *'s, N'S op:i-r?wi
-------
Figure 59. Frequency Distribution of Hand-wipe Lead at each Traffic Density.
SU'R 2
'>n>: 3
SITR -1
.IIUPOINI'S
!}l .
4 i
4-5. IHW)
3-S.-IHH)
27.
I 2.i)lM) ******
V./1HM)M******
">. HUB) **************
l.HOH) **************
I.*! )
-3. -l
*
**
**
*
****
M****
**********
*******
***
**
M*
**********
*****
J")f)IJI> MRM'JS ARI7
7. 898
A.7fjy
45. (WM
W.Viit
I.6H1
fiY M'S IF TIIRY COINCIOP rtlTH *'S, N'S OTHFRWISF
9.177 9.405
'\XI.MUM
21.240
2.77H
**
*
H***
+*
****
****
I2.99B
10.114
2PI.0HCI
41.410
T.M.1PI
200
-------
I'lgurc 60. Ilntvl-Wi|
vs Traffic 'Tensity
20 -
15
Hand-wipe
Concentration
dig/cm")
Sit.o 1 Site 2 511: 3
TRAFFIC DENSITY
Ml,-
-------
fingerprick and the remaining 81 were venipuncture samples.
For comparison of the two methods, both types were collec-
ted from some participants. Eight participants allowed
both fingerprick and venipuncture samples to be taken on
the same day, and the blood lead levels were determined.
Venous blood lead vs fingerprick blood lead levels for each
of these participants is plotted in Figure 61. The slope
of the fitted line is 1.24. Using these 8 matched pairs,
the difference between the methods was examined using a
paired comparisons t-test and was not significant (t=1.80,
P-0.11). With log transformed data, the variances from
these two groups of 8 were homogeneous. This means that
the blood lead levels from fingerprick samples were not
significantly higher or lower than matched venipuncture
samples. However, the fingerprick method was significantly
more variable as indicated by a F test comparing the var-
iances of fingerprick samples (n = 154) and venipuncture
samples (n = 81) from all children (F = 2.09, P <.001).
This relationship is approximately the same in the smaller
sex-age-site groups of children. Exclusion of the more
variable fingerprick samples would cause sample size in
most sex-site groups for children to be inadequate for
interpretation of the analysis; therefore, the alternatives
were to either combine the methods or omit the venipuncture
samples for children using only fingerprick samples. The
second alternative would also result in small samples in
202
-------
Figure 61. Venous Blood Lead vs Capillary Blood Lead
Samples from the Same Participant.
a
o
o
35 -
30 .-
25 ..
20
.J cc
E 3-
u
15 ..
10 ..
5 ..
O
^ 10 15 2D 25 30
VENOUS BLOOD
jig/100 ml
some groups; i.e., Site 4. Since both methods were used
in every sex and site group (for children) in approximately
the proportions mentioned above (2 fingerprick: 1 venous),
it was decided to use the combined methods for analysis.
203
-------
The site, sex, age, and blood analyses for each participant
are given in Appendix G. Finger prick blood lead concentra-
tions are listed in Appendix H. Two blood samples were taken
one week apart from each participant. Evaluation of the
paired venous blood samples using a paired t-test on log
transformed data indicated there was no significant difference
between the two samples (mean difference = 0.14 yg/100 ml,
t = 0.85, P = 0.39); the same analysis performed on finger-
prick blood lead samples showed no significant difference
in samples 1 and 2 (mean difference = 0.56, t = 0.21, P =
0.84). The two samples from each participant were therefore
averaged for most of the statistical analyses. The frequency
distribution of mean blood lead over all participants excluding
those eliminated from analysis due to other lead exposures (see
page 208) Is shown in Figure 62. The frequency distribution
of the mean blood lead for each age-sex group and for each
traffic density are shown in Figures 63 and 64, with the same
exclusions as above. The overall frequency distribution was
significantly skewed to the right (P = 5 x 10~^) as were
most of the smaller age-sex and site groups. A log trans-
formation caused the data to be normally distributed. Blood
samples 1 and 2 were not averaged for the ANOVA as in the
other analyses. In the ANOVA, differences between samples
1 and 2 were examined in each analysis. For this reason, the
frequency distribution for all blood samples (ri = 708) was
204
-------
Figure 62. Frequency Distribution of Mean Blood Lead..
t i.tHNI)*
M.11M0)*
II . IMOT)
2/..K1I1)
2l>. 0(1(1)**
24.MH)***
23.000)*
P.'./Mfl) ******
? 1.1-100) ***
2.1./100)****
I J. 0(10) **********
I i.000) ***********
I /..Xlf1)) ****************
I 6. /WO) *****************
I5.;1H0) *********************
I 4. ''(10) *******************************
I 3.1100) *********************************
I 2.''00)M **********************************
I 1.0(10) ***************************************************
I 1."H1H) **************************************
J. .H1H) ***************************************
!J. (1410) **************************************
7.'^t10) ********************
3. :)UO) *******************
b.'KK-l) *********
1..IOH)**
O'JTIP HF7\NS \RE D^IOl'I-D BY M-'S IF rMRY COIMCIDE iVIi'H *'S, N'S OTHRI7WISK
MFM'I 12.193
S. ni^V. 't.ft?'!
I 4 I 6. PI -.If)
JM Ti.(VM
J^ 4. 29H
Figure 63. Frequency Diltribution of Mean Blood Lead foe each Age and Sex iroup.
MALES LI' 1 MALES Of 49 F6MM.FS LT 9 FR"*LB!7 19-49 PEMALFS GT49
MIDPOINTS
37.5HH)
3 I.HMO
27..11)11)*
24.'.(1)01* **
27!min>** ****
t )». S'H*) *** * *****
H.,JM(1)*** ****** *****
I ft. 5|W) ******* **** M*****
15.'1Mtf) '1 ******* ********** *******
12. MO)***** ************* *******
1.1.500)******* ******* *****
j'.'KlO)*** ***** *
l.'j'IH)
MEANS ARC OBJOfEn RY «'S IF THEY COHClnE rtlTH -'S, N'S nTtiCTWISF
MEAN 15.424 12.929 lrt.714 UI.HA'i 1B.D34
S. OKV. 4.a^6 1.14ft 5.515 1.1W4 3.J47
.1 52.:)un A4.,):w 44.mil) lA5.rfirf 7I.DOH
MAXIMUM 27.52H 19..'75 31.HHM 2I.5H5 IV. DHb
MINIMUM 7.ftTi) 4.4U) 7.25rf 4.ft2H 4. 2-JM
205
-------
examined and found to be normally distributed after log
transformation. The distribution of all fingerprick blood
lead was also normal after log transformation.
Fi'jura 64. Frequency distribution of mean blood lead at each traffic density,
SITE I SITE 2 SI IT. 3 SITF
12. ilnnl*
II.. HIM)
2 /.Hl)0)
2-S.MMH)
21. MM)
1 "5.
I 1 ,
I t.
I 2.I
) ******* Ir*
) ****
HUPI) *
.IH11)
1 MF:n<; ARE UPIOTRB nv n"; IF THEY COINCIDE ivirn *">, N'S
12 27A
1 5HJ
I 1.559
1.94rt
I 12.HUM
28.rt(M
4.2011
76.ROD
25.lb«
The distribution of traffic counts for
the 280 households of the participants is shown below:
There is no overlap among the sites because these traffic
counts were used to assign households to sites. The mean
206
-------
traffic count, number of residences in a site (N), and
minimum and maximum traffic counts are given for each site
and over all sites.
Frequency Distribution of Traffic Counts for Households for Participants by Sites
MX; uju
* i 't i H UM
ARI; il^NOl HI
76
5V9
U,3ft(i
'280
37,H5H
155
WITH *'S, N'S
J, l-j?
70
I 1,117?
5,9 IH
I"., IP I
I .9^7
79
I9.MC10
25,964
rt,7!4
I9,57fl
Conclusions: The venipuncture and finger-
prick blood samples for children were combined for statis-
tical analysis. Normal distribution of the variable blood leac
can be achieved using a logarithmic transformation.
(2) Screening Variables
Tap water and house paint were samples
at each residence and measured for lead content so that these
potential sources of lead would not confound the results of
the blood lead analyses. Lead in tap water was discussed in
207
-------
Section III B 3 b and no adjustments will be made for additional sources
of lead in water.
Lead measurements were made on 985 painted
surfaces in the 280 households (two indoor and two outdoor at each house)
and are listed in Appendix I. The means, their standard errors, and
sample sizes for indoor paint lead and outdoor paint lead for each of the
four traffic density sites are given in Table 24; means and confidence
limits are plotted in Figure 65. A measurement of 1.0 to 4.0 mg Pb/cm2
paint (depending on the thickness of the paint) corresponds to 1% lead
content in paint. Using the highest paint lead value obtained, forty-
four percent (124/280) of the residences had a paint lead value over
1.0 mg/cm2. Eight percent (23/280) had a paint lead value over 4.0 mg/cm2
therefore, this level was used as a screening level for excessive lead
in paint. The blood lead levels of participant residing in a house with
any paint value 4.0 mg/cm2 was examined (39 participants, 12 of these 39
were children) . Figure 66 shows the frequency distribution of blood lead
of participants residing in houses with paint lead above and below 4.0
mg/cm2. Blood lead of participants who had used pottery for food containe
Table 24. Paint Lead Concentration (mg/cm2): Means, their Standard Errors and Sample
Sizes lor Indoor and Outdoor Paint Lead at each Traffic Site.
Indoor Paint Lead Outdoor Paint Lead
Site I Mean -SE 0.21 ±0.041 i 10 ±0 167
X 152 125
Sl'-e ~ MeaniSE 0.22 ±0.070 1.04 ±0.148
N 138 108
Si--- Mean-SE 0.44 ±0.137 1 22 == 0 209
153 121
Mean^SE 0.10 ±0.026 0.84*0.167
>' 108 80
208
-------
Figure 65. Paint Lead vs Traffic Density
1.75-,
1.50-
1.25-
1.00-
.75.
Paint
Lead
Concentra-
tion
(mg/cm2)
indoor .50-
1 Outdoor
.25-
Site 1
Site 2
Site 3
Site 4
209
-------
Figure 66. frequency Diatribition of Mean Blood Lead Levels of Participants
exposed to Paint Lead bolov and above 4.0 mg/cnz in Lheir homes.
LOPUNT HIPUNT
MIDPOINTS
n.000)
32.«)i>0>*
31. Until
3.).0»0>*
2s>.Ufr)M)*
2-1.1)1)1))*
27.HHH)
24..]0i)>***
21. DOHI*
Zi.MM) ****
2 I .H00) ***
2j..Jl!0>*»**
IM.IHI«>*****
13.01)0) *****
I 7.H»«) *****
H.flHB) *
I 1.000) *
I 2. «Hfl)M
t l.klWft)
?.t!H0> *****.**
8. PtOH *********
/.H«H> ********
S.VH10) ********
5. ao») *******
****
*****
uDOUP MFANS ARE DS^nrRD BY 'i'S If THEY COINCIDE iH FH *'S, N'S OTHER.IISR
«E*N 12.155 12.140
S. JEV. 4.5U 5.061
N 310.000 tiA.iWO
MAXIMUM 31.551) .13.1100
>4IMIMUM 4.2VO 4.A20
(47), had hobbies or occupations involving lead (44 and
9) , and children who often played near the street (13) were
also examined. Because there was some overlap, the total
number of participants examined was 127. Each log (blood
lead) for each of these participants was compared to the
mean for that age-sex-site group. Only those which were
higher than the mean were tested using Dixon's test for
extreme values: BX = (xn - x)/a(28), where x = the blood
lead measurement, x = the mean for that age-sex-site group,
and s - an independent estimate of the standard deviation.
The standard deviation of all venous blood samples (n = 708)
was used as the independent estimate of a. The analogous
210
-------
fingerprick standard deviation (n = 160) was used for the
fingerprick blood samples. Because these are screening
variables, a very conservative level of a (0.10) was set
which rejected cases greater than 1.3 standard deviations
above the mean. This is a one-tailed test, because the
alternative hypothesis is that the mean of these blood lead
values is higher due to this additional exposure to lead.
Twenty-nine blood samples from 20
of the 127 participants were found to be outliers (P<0.10).
This means that 29 blood lead measurements were greater
than 1.3 standard deviations (one-tailed test) above the
mean for their respective age-sex-site group. From a table
of areas of a normal curve, we see that 9.68% or 84 of the
868 blood samples are expected to be this deviant in a
normally distributed variable and we have examined approxi-
mately 30% of the total participants (127 of 441). There-
fore, more stringent requirements for rejection were sought.
Ten of these participants were found to have significantly
high blood lead in both samples. These 10 were rejected.
Their blood lead measurements and probabilities are shown
in Table 25.
Conclusions; Six screening variables
(tap water, house paint, pottery, hobbies, occupations,
and playsites near street) were examined as additional
sources of lead exposure. The blood lead levels of 127
participants who could have been exposed to lead from
211
-------
Table 25. Results of Test for Extreme Values (Dixon1 s Bi Test) on
Blood Samples of Participants Potentially Exposed to Other Lead Sources
ID # Site Sex Age Exposure13 Meana Bl lc P Bl 2 P _df_
2412 1 M 55 2,3 14.55 25.38 .067 25.21 .070 28
0076 1 F 31 1,2,5 8.98 18.73 .022 27.50 .001 95
0276 1 F 38 2 8.98 17.00 .040 15.69 .063 95
0067 1 F 42 5 8.98 16.42 .049 16.07 .055 95
5397 2 F 54 2 9.97 15.74 .106 19.95 .031 41
9326 2 F 66 2 9.97 19.47 .036 17.58 .062 41
5498 3 F 1 5 15.93 28.50 .093 40.60 .019 30
5626 3 F 31 1,2,3 10.15 17.01 .078 16.23 .099 93
3311 3 F 54 1 8.96 16.25 .053 18.73 .024 42
7821 4 F 36 1 9.30 15.16 .090 19.27 .024 74
a antilog of mean of log y
Exposure Code
1 Paint
2 Hobby
3 Job
4 Playj near street
5 Pottery
c Bl 1 blood sample 1, Bl 2 blood sample 2
one or more of these sources were examined, and 10 partici-
pants were rejected as outliers.
(3) Blood Lead vs Traffic Density
Two way ANOVA was used to test for
differences in blood lead among sites and among samples
(1 and 2) for each age-sex group. The ANOVA was mixed-
model with sites fixed and samples as random effects. Inter-
action and error mean squares were pooled for testing the
212
-------
fixed effects according to the rules proposed by Bancroft(^1).
The results are shown in Table 26. There are significant
differences among sites at P<0.05 in all sex-age groups
except males >49. Wo significant differences between samples
1 and 2 are seen in any groups nor are there any signifi-
cant interactions (sites x samples).
Table 26. Two-Way ANOVA of the Effects of Sites and Samples
on Lcg(Blood Lead) for Each Age-Sex Group
Males < 9
Source
Sites
Samples
Sites^Samp
Error
DF
MS
F Ratio Prob. Nonsignificant Sites
3 0.1034 4.2033 0.0076
1 0.0017 0.0670 0.7963
3 0. 0027 0. 1080 0. 9553
98 0. 0252
3 < 1 < 4 < Z
Males > 49
Sites 3 0.0377
Samples 1 0.0002
Sites*Samp 3 0.0114
Error 121 0.0169
2.2575 0.0850
0.0139 0.9064
0.6736 0.56Q8
3 < 2 < 4 < 1
Females < 9
Sites
Samples
SLtes*Samp
Error
3
1
3
121
0. 0868
0.0601
0.0015
0.0263
3. 3806 0. 0205
2.2891 0.1329
0.0580 0.Q816
4 < 3 < 1 < 2
Females 19-49
Sites 3 0.0807
Samples 1 0.0126
Sites-Samp 3 0.0065
Error 328 0.0201
4.4039 0.0076
0.6270 0.4290
0.3251 0.8072
1 < i < 2 < 3
Females > 49
Sites
Samples
Sites#Samp
Error
3 0.0685 3.3438 0.0211
1 0.0044 0.2111 0.6467
3 0. 0068 0. 3293 0. 8042
136 0.0208
< 4 < 2 < 1
213
-------
The mean of each site (with samples
1 and 2 pooled) were ranked from smallest to largest, and
multiple comparisons (SHK procedure) were used to determine
groups of site means which were not significantly different
at P<0.05. These nonsignificant sets of sites are under-
lined in Table 26. There appears to be no pattern of in-
creasing blood lead with increasing traffic density (site
number) at these traffic levels. The back-trans'formed means
(antilog of mean of log x), their confidence limits (L^
and L2), and sample sizes (N) are given in Table 27. Mean
blood lead vs. traffic density is plotted for each age and
sex group in Figure 67.
The relationship of individual blood
lead values to corresponding actual traffic counts was ex-
amined using regression analysis. The blood lead concen-
. 3!ood Lead Concentrations 'us/100ml): Means, Confidence Limits (L and L ), and
l f-r ..-ich A^e-Sax Grouo at each Site (Computed from Log (Blood Lead) and'Sack-Tramformed).
\lejn
L
12.
; !
'
3.J
13
; 2 .
1 5 .
-7
14
13
16.
4S
i
i
*
"
:o.
J
11
dO
20
32
(,
35
1 ~
n
37
33
-7
20
i8
7b
75
d7
16
14.
18.
34
12
11.
13.
36
13
16.
21
30
10.
q.
10.
70
VJ.
3.
10.
38
48
}0
23
42
08
'I
58
26
23
07
>0
^0
3 5
37
45
11.
9.
14.
12
11.
10.
12.
44
14.
13.
16.
36
10.
Q
10.
'2
i.
1 .
9.
41
67
38
51
56
52
71
do
22
78
05
44
69
6 7
6^
76
15.
12.
17.
22
12.
11.
14.
22
13.
10.
17.
15
9.
8.
9.
73
8.
8.
9.
24
07
94
54
53
08
17
60
77
19
14
47
87
91
03
89
214
-------
Fi
-------
to be significantly higher than either adult female group
(P<0.001) and female children to be significantly higher
than male children at a probability approximately equal
to 0.05 (ts = 1.71 as opposed to critical value = 1.65).
These were one-tailed tests.
Conclusions: Blood lead was significantly
higher in children than in adults for both males and females.
With regard to sex differences in blood lead, female children
were slightly higher than male children, and male adults
were significantly higher than female adults. There were
no differences between middle-age and older female groups.
b. FEP and Hernatocrit
FEP determinations have been shown to
be related to lead exposure and have been proposed as a
simple and reliable prescreening test for undue lead ab-
(25^
sorption especially at levels above 39 yg/100ml.v '
Hematocrits have also been suggested as a less expensive
and more efficient method of detecting lead exposure.
The frequency distributions of the FEP
and HCT are shown over all sites, for each site (ages and
sexes combined), and for each age and sex group (traffic
sites combined) in Figures 68-73. Each FEP and HCT value
represents the average of two samples from one participant.
Regression analysis was used to examine the relationship
218
-------
Table 28. Two-Way ANOVA of the Effects of
Sex and Age on Log(Blood Lead) at Sites 1, 2.. 3, 4, and All-Sites
Site 1
Source
Sex
Age
Sex*Age
Error
SS
0. 0240
0.1047
0.2874
3. 4837
DF
1
1
1
150
MS
0.0240
0. 1047
0.2874
0. 0232
F Ratio
1.0327
4.5074
12. 3767
Prob.
0.3112
0. 0354
0.0006
Site 2
Sex
Age
Sex* Age
Error
0.0427
1. 5161
0.2595
2.7189
1
1
1
134
0. 0427
1. 5161
0.2505
0.0203
2.1048
74.7177
12.7911
0. 1492
0. 0000
0.0005
Site 3
Sex
Age
Sex*Age
Error
0. 0022
0.3624
0.3354
3.0016
1
1
1
129
0.0022
0.3624
0. 3354
0.0233
0.0945
15.5735
14.4137
0. 7590
0.0001
0. 0002
Site 4
Sex
Age
Sex* Age
Error
0. 1854
0.3494
0.0541
1.4726
1
1
1
79
0.1854
0.3494
0.0541
0. 0186
9.9484
18.7444
2.9043
0. 0023
0.0000
0. 0923
All-Sites
Sex
Age
Sex* Age
Error
0.2115
2.3628
0.7658
11.5633
1
1
1
504
0.2115
2.3628
0'. 7658
0.0229
9.2171
102.9859
33.3784
0.0025
0. 0000
0. 0000
Neither sex had consistently higher
blood lead levels over all age groups. Using t-tests, the
age groups were examined individually- Adult males were shown
217
-------
(4) Sex and Age Difference in Blood
Lead
Previous studies have shown that males
as a group maintained higher blood lead levels than females living in
the same area with much the same external lead exposures and that children
had higher levels than adults ^ (32) _ Two-way ANOVA (fixed effects)
were used at each site and for all sites with sexes (male,
female) as rows and ages(<9, >49) as columns to determine
whether the same relationships were true for these data
(Table 28). Because the design must be balanced for this
analysis, the females 19-49 group was not included. The
interaction term (sex X age) was a highly significant
(P<0.001) at sites 1, 2, 3, and all-sites. In Figure 67,
we see the explanation for the significant interaction:
the lead values for male adults were higher than for female
adults; but for children, the females were higher than the
males. In Site 4, the males were higher than females in
both age groups; therefore, the interaction term in the
ANOVA was not significant. In all five ANOVA, the terms
for differences among ages were significant at P<0.05 (in
spite of the slight overlap at Site 1) or at P<0.01 (sites
2, 3, 4, and all-sites). Because the interaction involved
the sexes and not the ages, we can safely generalize that
age groups were significantly different; i.e., blood lead
for children was significantly higher than for adults in
both male and female groups.
216
-------
Figure 68. Frequency Distribution FLIP
ODD)
6 t
-1. .1110)
47.300)*
I i.ihW)*
t }.")',W ***
-Kid)
;H1H)
PK1d)
,1;1;1)
*****
******
***************
********************<*
*****************************
********* t*************************
************ *********************
M ******************* ************* ************
********* *********** * A *************
******************************
*** V*** ** V********** *******
********************
* ** ****** ***
**
*****
I 1
/."i
.)(1!1)
"ill")
)
»F\rlS ARF
BY »"S IF THEY COIHCinE WITH *'S, N'S OTHERWISF
Ml- \fl
S. ih'V.
'I [ i [MUM
7?
U)
between FEP and blood lead resulting in the equation, Y = 71.96 + 0.07 X,
where Y = FEP concentration (yg/100 ml RBC) and X = blood lead (ug/100 ml)
Only 0.01% of the variation in FEP was explained (R2 = 0.0001) by this
regression. This is consistent with the findings of Piomelli (25) who
reported no significant relationship between FEP and blood lead levels
when only the samples with lead levels up to 39 ug/100 ml were considered
(R = 0.02). Although we did not find any blood lead levels above 40
yg/100 ml (see Appendix G) , the blood lead in children was higher than in
adults. Therefore, FEP concentration was regressed on blood lead from
children only with the same result: only 0.3% was explained (Y = 61.07 +
0.42 X, where Y = FEP concentration in yg/100 ml RBC, and X = blood lead
in yg/100 ml) . Examination of scatter plots of X and Y and
219
-------
1'iyure 69. irequency Distribution of FEP at each Traffic Density.
siru: i sm- 2
SITE -1
16 3.1100)*
I6I .000)
lb4.l7ll.1M)
14 /.'1H0)
I .1:-1.000)*
I H.V100)*
I r>.'100>**
i 12.000)**
U)l
9.-(.
9 I .
H t.
. 000) ******
.:100) ********
. ,)00) ***********
.000) *************
7 /.1100)M*********
/.1. 000) *********
63.01H3) ********
56. MM) *******
49.000) ****
42. ',11)0 )*****
35.HOHJ*
2;i.00H>
.000)
.0110)
.000)
,1.0 )*
-7.000)
GDOIJP MEAilS \Ri= il':NnrEri ilY
'Ul.OUB
24.9-1M
J 1. 0;<:i
167. s.i,1
21.
I t.
7.
******
*******
**********
***** *****
M****
************
***********
*******
******
***
**
*
**
J|- illl-Y COINCinR lU'l
/'). I 111
.V. /V i
**
*+*
*****
*******
** **********
***********
!(*******
***********
*************
*******
********
*
**
MXIMILM
6S.64 1
21.HHV
I 44. l
2 I . '
+***
***
*****
******
!f* **********
********
*****
****
*****
*
/H.4il7
2S.I49
ujV.OHH
179. WO
residuals did not indicate that transformations would im-
prove the fit.
The relationship of hematocrits and
blood lead was examined. Only 3.7% of the variation in hema-
tocrits was explained by the regression on blood lead (Y = 38.64 + 0.22 X,
*
where Y = HCT and X = blood lead in yg/100 ml). Scatter plots and plots
of residuals were examined, and further transformations and analyses were
220
-------
Figure 70. Frequency Distribution of FUt1 for Crtch Acjc-3cx Croup.
MAI.KS ! 1' 9 '(M.l-'i 01' 49 l-l"'< M.i-"S LT ') i^'
19-49
H-"Al.i-S GT49
.l.iHU))
I.OHM)
I. KIM)
i. 1(10)
j.'KKD*
I /. 1(1(1) *V*
i j.'l'Kl) **
I I. '11111) *
**
1. OC'dl) **
l.HIKD*
i. )00>*
1. 1 )
l.lJllll)
l.OIH-1)
P MKANS ARE l)l:K)On;0 RY M'S IF THEY COINCIDE iVIFM -*"~<, N'
7(1.929 67.H41
V. .>R.9'1H 2ft. I74 10. 3()S
'I I 7.'mo 64.'100 16*. 00 1
'UN I
-i7.71r>
.>R.9 tM
I 7.'mo
MUM l'.17.'-vM
MUM I.'. .1 10
******* IT* C ******
* k-*** ****** ***+**27 ********
* k
-------
Fiyure 71. Frequency Distribution of IICT.
'I IMI'OIH l'!j
';..'. -HIM) *
') I. !),«))
OOH) ***
MOtt) ****
illllJ)
) ****,*********
**********
) *****************
) *************************
) **********************************************
) **********************************************
) M * * **** * * A i* A** * *** »:****** *********
I***********************************
***************************************
***** ************************************
) **********************
) *** ** *> ** **
) *************
) *****
.OillD *
. mil)
. .HIM)
. "1110)
|' "1F4NS
(I; I)|-N(1|T,|)
M'S Il; THffY COINCIDfT WITH *'S, N'S OTIIKDWISF
111.
fJ 1/V.IV10
MAXIMUM 'il.V.HI
nguro /i. frequency Distribution of IICT at each Traffic Density.
SHE I Ml i. > -iHF I
'I I ..
4 l.i-li
./M.'MIM)
.^ /. .llMt
'ct'ir MI AUS AKK ui-.iiori-[) ny
'! *******
******
*******
(J(IINCII)I: ,11111 *'S, II "i
W.64S
222
-------
riguro 73. Frequency Distribution of I1CV for each Afjo-Gcx Group.
HAI.^S Lf 9 MALES Of 49 I-GJALE--, LT 9 FMM.I-S 19-49
R-MALGS C'l 49
MIDPOINTS
Sl.tWM)
jU.fHH)
49.HMHI
41.111101
41.HHHI
4H.HOH) **
1V.H0M) ******
H.HUH) *****
17. HUM) II******
M.HH0J ***
lr>.HU0) ***
***
*****
)I. UWO) *** **
1J.10M) *
II.H(W)
W.IWP)
JJ.11IIH)
OIIOIII' HI:AMS AI1E UE-iOIKD RV M"~. II THIJY COI'ICIIIE WITH *'S,
MI-Afl
s. nsv
IV.2^7
I.7M
17.(Mil
44.
t'1.171
?.H?9
I A'i.llOH
-IV.IV-lvl
11.,I'M
41.117
.1. MO
oV.il.U
Ta:sLe 29. FEP (L^/ 100 ml RBC 's ) Meana, :heir Standard Errors, and Sample Sizes (N) lor each
Aye - 5*x Group ac each Site.
>.[ean ~SE
N
Mean = SE
N
Mean ^SE
N
Mean ^SE
N
Mean - SE
X
Site 1
34 7-12. OO
1 1
16 . b ~ ~ . ib
n
66.4=3 J0
1 J
34. 9 ^3. 87
>1
78 3=5 44
41
Site 2
59.5=7 05
21
70. 5 = 4. 78
36
60. 1 * 11. 24
10
80. 3 i 3. 64
77
73 7=4 36
38
Site 3
50 3 =19. 69
4
66. a =4. 90
44
65 6 i 14. 5
9
63. 7 ±3. 56
92
68 8 =6 04
41
Site
38. 5
6
76. 5
22
68. 5
4
68. 5
73
72. 3
23
4
= 9. 27
= 9. 14
i26.63
±2. 80
= 7 40
223
-------
Table 30. Hemacoc r-.t |"»)- Means and Their Standard Errors and Simple Sizes (N) for
Each A^e - Sex Group it each Site.
Site i Site 3 Site 4
Mean - 5E
Mean -S
N
Mean =S
N
Females ^ 4li Mean -S
X
3o 6-0. il
25
42. 5 -0. 56
28
37.5 ~0 38
31
40.4 0.34
04
40. 3 i-0. 41
41
37.
33
45.
36
37
28
40.
79
41
40
3 -i-0. 34
3 i-0. 32
0 *0. 43
8 to. 41
9 ±0. 68
37. 0
11
44. 5
40
36. 9
34
40.5
90
40. 8
39
±0.
*0.
±0.
±0.
=0.
65
51
37
35
.59
36.
21
43.
22
36.
18
39.
76
41.
23
9 *0. 34
4 *1.08
4 to. 47
7 ±0.32
0 ±0.51
c. Carbon Monoxide (CO)
Carbon monoxide levels in the blood were
determined on a subsample of the study population. From
the venipuncture blood specimens collected an attempt was
roade to select an approximately equal number of specimens
for males and females, smokers and norsmokers for each of
the four traffic density sites. This resulted In CO deter-
minations being made on 201 blood specimens from 163 persons.
Thirty-eight of these samples were second samples from the
same person. The frequency distribution of the ?01 samples
is shown in Figure 7^- Two peaks can be seen in the
histogram: one at 0.1$ which represents the nonsmokers
and one at 0.8% for the smokers. Before a relationship
between traffic levels and %CC could, be determined, the
smokers were separated from the nonsmokers. CO content in
blood samples from 141 participants who do not smoke was
analyzed for differences among the four traffic density sites
using a Kruskal-Wallis analysis of variance and found to be
not significant (H = 2.237, X2 _05(3) =7.815). The nonpara-
metric test was used after transformations did not correct
224
-------
skewness of the data. The means and their standard errors
and sample sizes are given below:
Mean + SE N
Site 1 0.136 + 0.0107 42
Site 2 0.138 + 0.0125 32
Site 3 0.122 + 0.0088 37
Site 4 0.139 + 0.0114 30
All concentrations (for nonsmokers and
smokers are given in Appendix G.
Conclusions; There was no significant
difference in CO content in the blood of persons who do
not smoke and who live at four different traffic densities.
Figure 74. Frequency Distribution of Carbon Monoxide in Blood.
0. o ) *****
0.10) **************##******#
-------
d. Blood Lead Levels and Cigarette Smoking
Smokers, exsmokers, and nonsmokers were ex-
amined for differences in blood lead levels. Participants
from all traffic densities were pooled for this analysis,
since increasing traffic density had no detectable effect
on blood lead at the traffic levels in this study. Data for
adult males and adult females were examined separately,
however, because their blood lead levels were shown to be
different. Single classification ANOVA was used to test for
differences among mean blood lead levels for the three
groups: smokers, exsmokers, and nonsmokers. The blood
lead value for each participant was the average of two blood
lead measurements. A log transformation was used on the
variable blood lead to meet the assumptions for this test;
therefore, the means discussed below are geometric means.
For the adult females, blood lead levels were
11.2, 9.2, and 8.7 yg/lOOml for smoker, exsmokers, and nonsmokers,
respectively. The differences were highly significant (P =
1.1 x 10 ~8). Multiple comparisons tests (SNK procedure) showed
no significant difference between exsmokers and nonsmokers;
however, blood lead levels in smokers were significantly higher
than in either exsmokers or nonsmokers. The log (base 10)
transformed frequency distribution, means, standard deviations
and sample sizes are presented in Figure 75. The backtransformed
(geometric) mean for each group is given in parentheses.
226
-------
Figure ' frequency DiSLnbatlon c: -
Groups of Adult Fe.Ti.iles j- -
-p "ariabii-' blO'-a Lrad 1-"' Three Smoking
the Pesults -f ASOVA a-no-.g these Groups.
.!.'.»; (8.7)
i. i "n
'. J^. (9.6)
'.Mi
0.000000011
For the adult males, mean blood lead levels
for smokers, exsmokers, and nonsmokers were 12.9, 13.1
and 11.5 yg/100 ml, respectively. Although these means were
not significantly different (P = 0.20), a positive relation-
ship can be seen between blood lead levels in males and
cigarette smoking. Blood lead levels for smokers and ex-
smokers are greater than for nonsmokers. Figure 76 shows
the log transformed frequency distribution for these three
groups.
Conclusions: A highly significant association
between smoking and blood lead levels was seen in adult females
Blood lead levels among female smokers was significantly
greater than for exsmokers and nonsmokers. The relationship
227
-------
Figure 76. Frequency Distribution of the Variable Blood Lead in Three Smoking
Groups of Adult Males and the Results of ANOVA among these Groups.
I? (12.9)
) (13.1)
I , 2H.1
t!. /S44
*****
M+-*
I . !'» i (11.5)
1. I -I.1
Hi A-; '-/)) -L T p :j,\ fin
I.1" i| I .ft1T>
between blood lead levels for the three groups was as
follows: smokers>exsmokers>nonsmokers. In adult males,
differences among groups were not statistically significant;
however, smokers and exsmokers had similar blood lead
levels which were higher than those for nonsmokers.
228
-------
5. Multivariate Analyses
In earlier sections of this report, each variable
was examined either singly or for a relationship with one
other variable. The variation in blood lead is here ex-
amined for a relationship to the combination of all environ-
mental variables and other characteristics of the partici-
pants and their homes obtained through the questionnaire
information. Twenty-five independent variables were
used in a least squares regression analysis with
blood lead as the dependent variable. Separate analyses
were made for children and for adults using the same vari-
ables with the exception of number 22 (smoking for adults
and hand-wipe lead for children). The 25 variables and
regression coefficients for each least squares regression
analysis are listed in Table 31. The analyses estimated
the amounts of variation in blood lead that is accounted
for by the simultaneous effect of the environmental and
other variables. Simple correlation coefficients among all
of the 26 variables for adults and for children are presented
in Table 32.
For children, 21,7% of the variation about the
mean blood lead can be attributed to the linear effects of
the variables listed in Table 31. This result is obtained
using ordinary least squares regression. The regression
coefficient associated with traffic count is negative but not
229
-------
Table 31. Variables and Regression Coefficients for Least Squares
Regression Analysis
Regression Coefficients
Number Name
0
1 Traffic Count
2 Occupational Lead Exposure
3 Glazed Pottery
4 Screened for Lead Absorption
5 Diagnosed as Lead Poisoned
6 Cooking Appliances
7 Age of House
8 Education of Head of Household
9 Paint Lead - Indoor
10 Paint Lead - Indoor
11 Paint Lead - Outdoor
12 Paint Lead - Outdoor
13 Distance from Road
14 Sex
15 Age
16 Length of Residence
17 Exposure to Lead Through Hobby
18 Hours Away from Home
19 Hours in Car/Bus
20 Anemia
21 Sick
22 Smoke (Adult) Handwipe (children)
23 Tapwater
24 Soil
25 Indoor Dust
26 Blood Lead
Children
17.33
0.37
3.95
0.46
1.92
3.36
3.68
0.09
0.91
0.14
0.31
0.16
0.05
0.01
1.90
0.34
0.11
3.72
0.07
0.21
2.86
1.74
0.02
6.51
0.00
42.00
Adults
11.55
- 0.17
- 0.91
0.62
1.98
- 4.25
- 0.86
- 0.01
0.13
- 0.13
0.18
- 0.07
- 0.06
- 0.00
2.76
0.03
- 0.02
- 0.33
0.02
0.02
- 0.68
1.22
- 1.15
- 0.42
0.00
- 4.77
significantly different from zero at the 5% level. A negative
coefficient here indicates that when the other 24 variables
are held constant that blood lead tends to decrease as traffic
count increases. Since the coefficient is not significant
this means that if we exclude the variable traffic count from
the regression we can expect that the resulting R2 will not
decrease significantly from 21.7%.
230
-------
Table 32.
Simple Correlation Coefficients Among All Variables for Children and
Adults.
to
CO
1 '
llil l.liii
1-000
.00)
-.in)
.Ob2
.105
.045
I/I
. 1 JU
-.011
- . oy6
-.OaO
-.025
015
.\2U
-.oy7
.019
.064
-.006
.052
.211
.2)2
-.OH
-.020
.1 JO
-.020
n
1 ooo
. 122
. ib5
- Ij45
"73
-.163
-.000
-.OJ3
-.it;
-.020
-. 166
-.025
-.005
-.031
JO 2
- . 0'J5
- . O'J >
.080
-.12)
.Ibl
.0)1
. JUd
-.114
.158
1.000
-.Ob4
- .052
.9b4
.077
211
.113
- Oil
.001
.0/1
-.'141
- . 033
. 115
.oj;
.072
- . 07b
- 0'J3
-.14)
. iy4
.125
.007
-.127
-.053
1.000
.623
.050
. 14)
.126
- 012
. 1 JO
-.123
-.0)9
all
-.045
-,1)J
-.022
- , Ob9
- . 092
-.055
- . ube
.050
.201
-.053
-.0/4
.m
1 . 000
.m
.'I2b
.j)b
-.329
-. Jbl
-.077
-. 151
. ill
-. J64
-.011
-.114
- 12b
. 139
-.114
-. J55
. Jbb
. J'35
-. )4b
-. 163
.147
1.000
-.005
.503
.047
.004
-.ota
.049
-.073
-.366
.015
.022
.oyi
.005
-.13b
-,43'J
-.250
.041
-.121
-.205
-.039
l.UOO
.009
.064
.259
.301
.408
.040
-.022
.200
-.159
-.246
-.011
-.073
-.261
.107
.025
.045
.124
.110
1.000
.095
-.013
-.002
.Ib7
-.016
- 195
.061
.055
-.OJJ
-.027
-.200
-.331
-.044
.190
-.230
-.234
.044
1.000
-.067
-.039
-.Ob5
-.065
.018
-.066
-.021
-.065
-.001
-.052
-.073
-.031
.090
-.053
-.077
-.024
-.061
-.076
-.037
.044
-.061
-.145
-.026
.067
-.01J
.013
.21D
-.004
-.093
-.055
.233
.out
1.000
.491
.193
.058
.064
.079
-.057
-.062
.117
-.002
.007
.010
-.0)3
-.021
.010
-.059
1.000
.080
-.072
.105
.142
-.023
-.15U
.11U
-.028
-.011
.146
.057
.060
.157
.059
1.000
.061
.04g
.188
-.013
-.082
-.011
-.123
-.109
-.169
.065
-.071
-.125
-.030
1.000
-.135
-.109
-.08)
-.060
.035
.196
-.120
.068
-.011
-.045
.011
-.167
1.000
.325
.045
.249
-.070
-.052
.140
-.050
-.000
-.062
.048
-.128
1.000
.021
.300
-.056
-.075
-.176
.091
.128
-.054
-.044
-.058
1.000
.093
.194
-.024
-.038
-.004
.133
-.075
-.063
.002
1.000
.150
-.058
-.007
-.215
.116
-.188
-.059
-.163
1.000
-.056
.079
-.059
.070
-.004
.051
.088
1.000
.145 1.000
.007 .127 I. 000
-.082 -.02) -.044 1.000
.024 .102 .079 -.133 1.000
.016 .117 .199 -.190 .029 l.OOO
-.177 .058 .169 .Oil .029 .103
(continued)
-------
Table 32. (continued)
M
U>
NJ
Alulls
1.000
-.05)
-.200
-.IXJ/
.015
.028
-.201
.0-10
.009
-.029
-.097
.053
.071
-057
.056
-.012
-.042
.058
-.002
-.044
-.085
-.Ml
.113
.015
-.005
-.006
i
1.000
.015
.17U
.063
.056
.030
-.112
-.047
-.057
.038
-.013
.052
.116
.11?
.129
.071
-.046
-.058
-.031
-.081
-.097
-.019
.045
-.010
.005
1.000
-.065
-.04U
.069
.070
.074
-.001
-.017
-.023
-.076
.002
-.042
-.028
.033
.107
.132
.003
122
.036
.059
-.003
.034
.121
.018
I.CLIO
.740
.036
-.042
-.008
-.027
.091
.067
-.049
.007
.053
.014
-.061
-.063
-.103
-.106
-.044
-.052
-.137
-.006
-.001
.011
.021
1.000
.027
.038
-.052
-.030
.158
.107
-.055
.017
.061
019
-.008
-.047
-.089
-.117
-.039
-.038
-.030
.004
.054
.036
-.066
1.000
-.147
.181
.004
-.067
-.055
.050
.018
.007
.050
-.003
-.060
.040
.061
-.026
.055
-.048
.023
-.133
-.027
-.018
1.000
.050
.121
.236
.248
.267
.129
.110
.112
.298
.048
-.150
-.099
-.050
.011
.025
-.117
.207
.OU9
.021
1.000
.073
-.068
-.082
.062
.069
-.112
-.198
-.090
.083
.082
.053
.042
-.126
.095
.041
.025
-.024
-.065
1.000
.051
-.004
.044
-.092
-.015
-.043
-.034
-.001
-.032
-.058
-.014
-.030
.027
-.002
.003
-.030
-.045
1.000
.070
.060
.023
.000
.005
.079
.010
.018
-.030
-.013
-.002
.020
-.023
.062
-.009
.021
1.000
.359
.072
.041
-.013
.025
-.185
-.133
-.053
.028
-.027
-.015
-.051
.108
.060
-.043
1.000
.152
.041
,028
.060
-.043
-.042
-.004
-.061
.035
-.06L
-.013
.057
.011
.000
1.000
.161
.191
.Ul
.021
-.067
.004
.163
-.058
-.017
-.057
.004
-.022
.060
1.000
.606
.34£
.137
-.043
-.040
-.316
-.113
.008
-.029
-.016
-.002
.388
1.000
.594
-.012
-.052
-.042
-.263
-.152
.129
-.018
-.073
-.051
.268
1.000
.116
-.020
-.120
-.178
-.064
.089
-.061
.012
-.025
.121
1.000
.008
-.002
.007
-.005
-.077
-.011
-.OJ2
-.041
.010
1.000
.408
.013
.029
.053
.039
-.101
.017
.037
1.000
-.029
.073
.024
-.022
-.062
-.057
.047
1.000
.008 1.000
-.004 -.118 1.000
.082 -.018 -.012 1.000
.038 -.040 -.074 -.028 1.000
-.054 -.046 .081 -.005 .061 1.000
-.199 .092 -.215 .048 .068 -.063
1.000
-------
Ridge regression is a biased estimation technique
which in theory can provide regression coefficients that are
closer to their true value on the average than those coefficients
resulting from least squares regression. This method of analysis
yielded the fact that_the sign of the traffic count coefficient
did not change to a positive value for the set of possible ridge
coefficients (ridge trace) computed by Hocking's program. Evident-
ly, the reason that the sign of this coefficient is negative is
not due to the independent variables being highly correlated.
Variable selection methods developed by Hocking
selected the variable traffic count only after many other variables
had been included in the regression equation. In no case was traffic
count selected before 14 of the other 25 variables had been selected.
In this sense the traffic count is an important determinant of blood
lead only when many other more important variables are held constant.
Apparently, when traffic count is measured as in this
study its linear effect on children's blood lead is weak. However,
it seems plausible to assume from this analysis that the suggestion
that blood leads decrease as traffic count increase is a general
conclusion that can be expected to be found in other studies similar
to this one. There may be underlying factors that are specific to
this study situation that are responsible for this coefficient
233
-------
having the wrong sign.
Factor analysis is a multivariate technique used
to probe the variables for underlying factors of central
importance. Principal component extraction of factors was
applied to the 25 independent variables. Each factor explains
a portion of the variance in the correlation matrix of the
25 independent variables. Normally, only the factors explain-
ing the highest portions are considered worthy of further
analysis. However, our goal in using factor analysis is to
attempt to determine if the reason that the regression
coefficient of traffic count is negative is due to spurious
error sources.
We do this in three steps.
First, it should be noted that the regression
coefficient of traffic count and blood lead can be written
as a linear combination of the ith factor loading with the
jth variable (say a. .) and the regression coefficient (say
b^) of the ith factor on blood lead. That is, the regression
coefficient, of traffic count on blood, lead (including as
independent variables all 25 variables) is equal to
25
x
J alj bi
1=1
So our first step is to regress these 25 underlying factors
on blood lead to obtain the a..
Nov. some of these terms are positive and some are
negative. (If the 25 variables had all been unccrrelated then
234
-------
only one term would be nonzero). Also some of these factors
primarily represent true underlying mechanisms of central
interest in this study. That is, mechanisms that can be
expected to influence the data whether this study was conducted
in Dallas or San Francisco. Other factors represent mechanisms
that are unique to this study. We call these common factors
error factors.
Secondly, we proceed on the assumption that factors
whose terms are positive in the above sum are the true factors.
If we add only these terms we can expect to obtain an adjusted
positive regression coefficient. When the last assumption
is completely false this correlation should be 'too big'.
Thirdly, we concentrate our analysis on the 'positive
factors'. If we can identify these factors then we will obtain
assurance that the last assumption holds (at least for those
factors so identified).
Instead of concentrating our efforts on identify-
ing the positive factors we could have considered the negative
factors. However, we did not do this because it was thought
that identifying negative factors correctly would be a much
more difficult process.
Lastly, even though this process may seem sub-
jective, it is not difficult to consider relationships between
variables that could pass undetected by this analysis. For
example, the requirement of orthogonality which simplifies
235
-------
the analysis usually makes the factor identification process
more difficult. Thus, we consider this method to be a quick
and dirty look at the multivariate aspect of this problem.
Considering only the first six factors (listed in Table 33)
Table 33. Correlations with the 25 independent variables
of the six principal component factors that
contribute most to R2 (Data from children only)
Variable
No. I
1 .0298
2 .0679
3 -.1711
4 .0299
5 -.0662
6 -.0390
7 .1232
8 .0159
9 -.4392
10 -.2759
11 .1086
12 .0860
13 .1473
14 .5012
15 .0931
16 .2082
17 -.0232
18 .1108
19 -.1616
20 .4570
21 -.1308
22 .0256
23 -.1247
24 -.0320
25 -.2187
Percentage
Variation
Explained
6.62
II
.0523
.1399
-.1145
.2411
.2589
-.0074
.3555
.0719
-.0234
-.0417
.1950
.2750
.1077
.1676
-.2542
-.1358
-.1538
-.5144
-.0931
.0038
-.1226
. 3277
.0143
.1896
.1166
4.25
III
.1697
-.2458
-.1503
-.3407
.2344
-.0514
-.0887
-.2639
.0143
-.1592
.2367
-.0971
-.1783
.1563
-.0938
-.0428
.2868
-.0146
-.3619
-.3172
-.2442
-.1717
.1666
.2165
.0381
2.15
IV
-.1521
-.1452
.0722
-.1389
.1530
.2393
-.4509
-.1507
.0543
.0911
-.0425
-.0905
.4074
-.0722
.2140
-.2838
-.0499
.0730
.0299
.0968
-.2668
.4381
.0611
.1390
.0403
1.81
V
.2027
.2890
-.1461
.3904
.2629
.0166
-.3858
.0115
.0062
. 0043
-.3594
-.3278
-.3126
.0300
-.1516
-.2698
.1102
-.0199
-.0628
.0603
.0941
.0849
.0907
.0595
-.0659
1 . Id
VI
.0173
-.2559
.0676
-.2163
-.2441
.2591
-.0135
.0768
.0655
.1540
-.0767
-.2371
-.0070
.1970
-.4575
-.3860
-.2252
-.1547
-.0326
.1558
-.0786
-.2058
-.3335
-.0548
.0115
1 1 9
236
-------
that are most highly correlated with blood lead we find that
o P
these factors account for an Rc of 17-1 whereas before an R
of 21.7 was achieved using all the variables. Thus, we see
that most (79% - 17-1/21.7) of the variation explained using
the 25 variables is explained using only these six factors.
Factors II, IV and V are the positive factors of the six
factors. Combined they account for 1.8l + 4.25 + 1.14 =
7.20 of the R2 of 17.85? or 40$ of 17.8.
Factor II is correlated with the age of the house,
hours away from home and handwipes. The blood lead is
higher due to this factor when the chrld spends less time
away from home, lives in an older house and has higher
handwipe lead values.
Factor IV is correlated with age of house, distance
from the road and handwipes. The blood lead is higher on
this factor for a child who lives in a newer house farther
from the road and has higher handwipe lead levels.
Factor V is correlated with screened for lead
absorption, age of house, and both indoor and outdoor paint.
The blood lead is higher on this factor for a child who
lives in a newer house which has low lead in the indoor and
outdoor paint. Neither he nor members of the family in the
house have been screened for excess lead absorption.
Factor II seems to be indicating that the subgroup
of this population of children that have the characteristics
that are correlated with factor II are in a slightly higher
237
-------
exposure category to automobile generated lead. Factor IV,
the most important of the three factors, indicates that
distance from the road to the house is an additional
important variable.
Factor V is difficult to interpret as reflecting
exposure from environmental lead. For this reason this factor is
regarded as an error factor.
In conclusion, there are 3 positive factors that
suggest that blood lead increases when traffic counts increase.
Two of these factors suggest that more careful screening is
required to insure that the children live in houses
of a similar age and a similar distance from the road and spend
only a short amount of time away from home. However, even with
such increased control there is no suggestion that a dramatic
improvement in the R2 will result. This analysis only consid-
ered the effect on the relationship between blood lead and
traffic count caused by these 24 other variables. The inclusion
of additional variables may improve the relationship. For
example, traffic speed limits would be a good variable to
include. More confident conclusions would require a more
in depth analysis.
For adults, 28.6$ of the total variation in
blood lead was explained by regression on the 25 environmental
variables. Using only the first six factors we achieve
an R2 of 21.0, which is 73% (21.0/28.6) of the variation
238
-------
Table 34. Correlations with the 25 independent variables of the six
principal component factors that contribute most to R2
(data from adults only).
Variable
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Percent
Variation
Explained
I
.0076
-.1587
.0405
-.1033
-.1289
.0373
-.3030
.1211
.0130
-.1077
-.1555
-.1460
-.2944
-.4072
- .4481
- .4346
-.0497
.1556
.1444
.2538
.1145
-.0311
.0941
-.0574
-.0173
5.2%
II
-.1876
.2371
.0341
.0183
-.0574
.0953
-.0198
-.3481
-.2379
-.0376
.1315
.1346
.0310
.0229
-.0275
-.0033
.0781
.1343
.2720
-.0614
.5298
-.4834
-.1611
-.0761
-.1718
4.58%
III
.2276
.3788
-.2032
-.0045
.0929
.0118
-.0003
.0611
-.0054
-.0780
-.0277
.0824
-.0999
-.1661.
-.0759
.1277
.2114
-.0086
-.0228
-.0802
.4945
.5968
.0015
.1573
-.0207
4.30_%
IV
-.1453
-.2511
-.0311
.0008
.1412
.2382
-.1343
-.4398
.0837
-.1805
-.0301
.1775
.0730
-.4620
-.0653
.2806
.2726
.1147
-.0241
-.2089
-.2719
.0363
.0669
.1982
-.0474
2.32%
V
-.3840
-.1922
-.0081
.1148
.0913
-.2541
.0907
.1652
.0418
-.4358
.0684
-.0665
.0096
.1636
.0259
-.0817
-.0911
-.0597
.1971
-.2477
.1567
.0756
.4648
.2089
-.2387
2.30%
VI
.1015
.1172
.1075
-.0164
-.0412
.0449
-.0278
-.0560
.5955
-.0319
.0229
.1092
-.3751
.2371
-.0051
-.1403
.1488
.1277
.0214
-.3209
.0466
-.2037
.0454
.1525
.4011
2.29%
239
-------
explained by using all 25 factors. These six factors are
shown in Table 3^. Only factors IV and VI which are the
positive factors will be interpreted. Factor IV is cor-
related with sex and education of head of household. Blood
lead is higher on this factor for adult males living in
households of lower educational levels. Factor VI is cor-
related with indoor paint lead, distance from road, and
indoor dust lead. Blood lead levels were higher on this
factor for those persons exposed to higher indoor dust lead,
indoor paint lead, and living closer to the roadway. The
meaning of an association between paint lead and blood lead
in adults is not clear. Perhaps these paint lead levels are
related to higher dust lead levels inside the home.
In summary, for adults from 2% to 5% of the varia-
ation in blood lead can be explained by factors concerned
with sex, education, indoor dust lead, and distance from
road. The exact nature of these relationships is not clear.
At this low level of correlation, the effect of a variable
on blood lead may be positive when combined with one variable
and negative when combined with another. None of the relation-
ships shown here seem to have a strong effect on blood lead
or clarify the effect of traffic density on blood lead.
Conclusions: No strong relatjonships between
blood lead and environmental variables were seen in the
240
-------
regression analyses for children and for adults. A smell
portion of the variation in blood lead was explained by
factors concerned with these who spent more tine awcy fron
home, sex, education levels, indoor dust lead, ard distance
from the road. No effects of traffic density on blood leac
were clarified for adults or children.
241
-------
IV- DISCUSSION
A. Air Monitoring Study
The principal objective of this air monitoring
effort was to examine the relationship of ambient air
lead levels with concurrent traffic count measurements.
Previous studies have demonstrated that there is a
general relationship between traffic counts and air
lead values obtained immediately adjacent to the road-
way. This study was designed to calibrate the relationship
in the general vicinity of the epidemiologic study as well
as to investigate the importance of other parameters of
traffic exposures from combustion of leaded baseline.
Furthermore, these data may be useful in estimating the air
lead exposures of a number of urban residents in the United
States, since traffic count data are often available for
urban areas. These data are relevant to the total problem
of air lead exposures from automobiles because, as stated
in the Introduction of this report, a very large proportion
of the total population of the United States resides on
streets with traffic densities ranging from less than 1,000
cars per day up to 25,000 cars per day, the traffic densities
covered in this study.
The results of this study indicate that there was
a small but significant relationship between ambient air
242
-------
lead levels and traffic counts. This relationship was
much stronger when the traffic densities reached 30,000
to 40,000 vehicles per day. The ambient air lead levels
ranged from 0.5 at less than 1,000 vehicles per day to
1.9 yg/nr for the higher traffic- density levels. This
study shows that a relationship exists between traffic
counts and air lead over this range but that the rela-
tionship is not a strong one and that other variables
such as the presence of obstructions near the site where
the air samplers were located, meteorological factors
and sampler variability may have obscured the primary
relationship. During the study, placement of the high
volume air samplers was conducted very carefully to
avoid the problems of interference from high shrubs and
buildings; however, thei'r influence could not be com-
pletely avoided. The samplers were always located at
a specified distance from the roadway and were always
located downwind from it relative to the wind direction
for that 24 hour period of monitoring. Air data collected
during the early stages of the project indicated that
placement of a high volume air sampler near a residence
with a large shrub in between the sampler and the road-
way provided a much l..-we:/ level of lead than for corresponding
residences in which no shrub was present In the pi 3rr.Hrv
study samplers vere placed t~ avoid thiy type cf prctl-Ki
243
-------
The high volume air samplers were also located at approximately
3 feet above ground level which is near the breathing zone
for small children. Most ambient air samplers operating in the
United States for regulatory purposes are located between
10 and 20 feet above ground level. Data from this study
may not be directly comparable to data collected at other
sites with samplers located between 10 and 15 feet above
ground level due to the possibility that a vertical gradient
in air lead levels may be present.
Outdoor air lead was measured at four distances from
the street to determine whether air lead levels decline
in the first 100 feet from the roadway. For all distances
measured, 5 to 100 feet from the road, air lead concen-
trations declined rapidly with distance from the street.
At 50 feet, concentrations were about 55 percent of the
street concentrations. For air lead collections from
5 to 100 feet from the street, approximately 50 percent
of the airborne lead was in the respirable range, that is,
less than 1 y and the proportions in each size class
remained approximately the same as the distance from
the street increased.
Two intersection studies were done to determine
whether residents at intersections are exposed to higher
air lead levels than their neighbors at midblock locations
on the same street. One study used only intersections
with small side streets (less than 1,000 cars per day).
The objective of this study was to determine additional
lead exposure, if any, of corner home residents who
244
-------
were participants in the epidemiology portion of thds study
No additional lead exposure was measured for these parti-
cipants. However, in the second study where larger inter-
sections were used, significantly higher lead levels were
found at intersections. This occurred mainly at inter-
sections where primary streets had greater than 25,000
and secondary streets had at least 10,000 cars per day.
Further studies will have to be made to clarify this,
because of the small amount of data collected at the
higher traffic densities.
Air lead levels were measured at two speed limits
(30 MPH and 45 MPH) on streets with approximately the
same traffic density to determine whether air lead levels
are influenced by vehicular speed The air lead levels
at 30 MPH were twice those at 45 MPH, although a statis-
tical difference in the two samples could not be shown,
possibly due to large variation and small sample sizes.
Most of the air lead samples in this study were
taken for a 24 hour period. A study was made of air lead
measurements taken for shorter collection times: one, two,
four and twelve hours. Shorter collection times over-
estimated traffic volume and introduced more variability
in air lead measurements. The range in air lead measure-
ments increased with shorter sampling times at all traffic
densities. No relationship between air lead levels and
length of collection times was seen.
245
-------
Two of the ministudies conducted under the air-
sampling program dealt with the influence of indoor versus
outdoor exposures. These studies were the simultaneous
hi-vol monitoring inside and outside of homes and the dustfall
study. Although the use of hi-vol samplers in the indoor
environment may be somewhat questionable because cf the
possible "cleansing" effect of the high sampling rate the
results obtained in this study are not greatly different
from those obtained in studies which used lower sampling rates.
In this study, indoor/ourdoor air lead samples were
taken at two traffic densities, nine samples at 10,000 cars
daily and seven at 20,000 cars daily. The results showed
a highly significant difference in the levels cf lead indoors
versus outdoors at both the sites. Thus at 10,000 cars per
day, the indoor air lead levels were 0.18 yg per cubic meter
while the mean value for the outside air was 0.92 yg per
cubic meter. The levels cf lead in air indoors at the 20,000
cars per day was 0.20 yg per cubic meter. Therefore, our
results indicate an approximate ten-fold reduction in air
lead levels from outdoor to indoor air. These monitoring
studies were conducted during the summer months in Dallas,
Texas where the temperatures are usually in the 90's.
Questionnaire data indicated that virtually all the houses
in both the epidemiology study as well as the monitoring study
had air conditioning of seme type.
Daines et al. using a low volume sampler in
New Jersey during the summer discovered a 65# reduction in
246
-------
air lead levels when air conditioners were used.
Benson et al. have summarized the indoor/outdoor literature
as indicating particulate lead may be significantly reduced
by efficiently operating air conditioners with filters
in place, but that operation of evaporative cooJers produce
an almost one to one ratio of air lead to indoor lead.
That there might be a problem with the "cleansing"
effect may be seen from the following calculations. The hi-
vol samp lei' was operated at a flow rate of approximately 50
cubic feet per minute usually in the living room of the
selected residences. Both indoor and outdoor monitoring were
performed over the same 12 hour period during the daytime.
A typical room in which the sampler was placed would contain
350 cubic feet of air, and in this study the doors into the
other rooms we're left open, thus allowing free movement of
air within the home. Allowing for no infiltration of air
into the room there would be a complete passage of air through
the fLiter every 7 minutes or 100 times during the monitoring
period. If the living room was 1/4 of the cubic footage of
the home there would be 25 passages v/jthjn the monitoring
day.
Outdoor dust samples were taken at ten locations and
each sample was paired with an indoor dust sample covering
the same time period (28 days). These comparisons of
indoor versus outdoor dust levels were made at traffic
densities from less than 500 cars per day to more than
30,000 cars per day. The mean for the ten outdoor samples
was 0.12 ug per square raster while the mean for indoors
247
-------
was 0.012 yg per square meter. These results are consistent
with the approximate ten-fold reduction of lead outdoors
versus indoors in the air lead measurements. This ten-fold
relationship occurred at each traffic density.
In a study conducted in Southern California there
were two study areas, one near a freeway in Los Angeles
with traffic densities of more than 200,000 cars per day
and the other in a city of 50,000 (Lancaster) in a high
desert area. The mean air lead values were 6.3 yg/m3
for Los Angeles and 0.6 yg/m3 for Lancaster. The range
of air lead means for this study were from 0.5 to 1.9
yg/m3. The residences studied in Los Angeles were within
100 feet of the edge of the freeway and they did not have
air conditioning. During the sampling effort (September
1974), the windows of these residences were open much of
the time. The majority of the residences in Lancaster
utilized evaporative coolers which provide a large ex-
change of outside air with inside air. It has been
concluded(33) that air lead levels present indoors in homes
with evaporative coolers are equal to levels present in
outside air. The outside air lead measurements were
therefore good representations of air lead exposure for
the Southern California studies. As stated earlier,
this probably is not true for the Dallas study; soil
lead levels in the Los Angeles study area ranged from
673 to 3633 yg/m3 and from 43 to 98 yg/m3 in the
Lancaster area. For this study the soil lead values
(34)
ranged from 4 to 730 yg/m3. Yankel et al. reports an extensive
248
-------
study of the relationship of air and soil lead levels
to levels of lead in blood of children living in these
environments. He stated that soil levels in excess of
1000 ug/m3 (30 day average) will result in excessive
percentages of children exceeding 40 pg lead/100 ml
blood. These conditions were present for the Los
Angeles site for our southern California study but not
for this study.
Topsoil characteristics were determined for the
study area near participants' residences. These
characteristics were determined on the same soil samples
in which soil lead measurements were made. The objec-
tive of this effort was to examine the potential of the
type of soil with regards to lead adsorption, retention
and release of these top soils. The majority of the soils
were high in clay content and in organic matter. These
types of soils have a high potential for adsorption and
storage of lead. The chemistry of the soil favors the
formation of relatively insoluble precipitates of lead.
The clay soils should retain lead deposited on and absorbed
by them; however, low permeability of the clay may reduce
initial infiltration during heavy rainfalls.
B. Epidemiology Study of Traffic Density Relative to
Levels of Lead in the Environment and Blood of
Residents
It has been suggested that the distribution of blood
lead levels for any relatively homogeneous population
249
-------
closely follows a log normal distribution ' ' .In
the southern California study performed in this laboratory
(9,10,11)
blood leads were log normal for the adult popula-
tion. Results from this study show that the blood lead
values for adults and children had log normal distributions.
The measurement variance was considerably smaller for
this study than for the one in southern California and was
consistent with the variance in other lead studies. This
was true for venous blood lead measurements but not for
the micromethods used on blood collected from the finger
prick method. The micromethod of blood lead measurement
results in more variability than for the usual macro
method of analysis. Also the procedure for collecting
capillary blood using the fingerprick technique is sub-
ject to more contamination effects at the time of collection
and analysis than the venipuncture procedure.
Examination of the relationship of traffic counts to
levels of lead in the blood of participants revealed no
detectable relationship for traffic counts ranging from
less than 1,000 cars per day to more than 25,000 cars per
day. No significant statistical relationship between-
traffic density and blood lead levels for any age group or
sex was found. Blood lead levels were significantly higher
in children, ranging from 7 to 33 Mg per 100 ml, than in
adult-, which ranged frcn '< to ?1 yg per JOO ml. No
250
-------
relationships were found between traffic counts and measure-
ments of FEP, hematocrits, carbon monoxide levels in blood,
and lead in hand-wipe samples.
A positive relationship between smoking and blood lead
levels was found for both females and males in this study.
Females who smoke had significantly higher blood lead levels
than female exsmokers and nonsmokers. Male smokers and
exsmokers had higher blood lead levels than nonsmokers,
although this difference was not significant.
This study was designed to examine participants at
four basic traffic densities: less than 1,000; 8,000-14,000;
14,000-20,000; and 20,000-30,000 cars per day. The epidemio-
logy study was designed such that there would be some
expected differences in exposure of these residents to levels
of lead in the air as a result of their living on streets of
these traffic densities. The results of the air monitoring
study discussed above indicated that mean air lead levels
ranged from 0.5 yg per cubic meter to a high of 1.9 yg per
cubic meter. This differential in levels of lead in the
air is quite small. Previous studies examining the rela-
tionship of air lead with blood lead levels have been
successful for air lead at 2 yg and above. If there are
significant differences in blood lead over the range in-
vestigated in this study, it is obvious that it could have
been seen only with much larger numbers of participants
than were included in this study.
251
-------
Another factor which may have contributed to the
failure to demonstrate a relationship between blood lead and
traffic court in this study was the difference in exposure
between the indoor and outdoor environments. Since this
study was performed during the hottest time of the year it
may be reasonabJe to suppose that participants may have spent
a significant proportion of the:"r day in the cool environment of
their hone produced by their air conditioners. If the air lead
levels measured within the home are accurate then the observed
result would be entirely predictable.
252
-------
V. CONCLUSIONS
1. Examination of the relationship of traffic counts from >1,000 to
<25,000 cars per day with the concentrations of lead in the blood
of residents on these streets revealed no detectable relationship.
2. No significant relationships were found between traffic counts and
FEP, hematocrits, carbon monoxide levels in blood, and lead in
hand-wipes.
3. There was a small but significant relationship between levels of
lead in the air and traffic densities. Mean air lead values for
each street ranged from 0.5 to 1.9 yg/m^ from Site 1 to Site 4.
4. There were significantly higher levels of lead in soil near resi-
dences in Sites 3 and 4 than in Sites 1 and 2. Soil characteristics
in the study site areas favor retention of lead.
5. No significant relationships were found between traffic counts and
the level of lead in indoor dustfall and window-sill wipes.
6. Comparison of lead levels in air and dustfall inside selected resi-
dences with similar measurements taken outside revealed tenfold
higher levels of lead indoors.
7. The lack of an association between traffic counts and blood lead
levels can be understood in terms of the
253
-------
lead, exposures generated by the studied traffic volumes.
The air lead levels observed at the maximum traffic-
densities (<2 yg/rr.-') have not been previously shown to
result in increased blood leads and neither have the
soil Jead levels observed. Furthermore, the use cf air-
conditioning may have contributed to the lack of an
association by further reducing the effective air lead
exposure.
8. The levels of lead in the populations studied are
similar to values seen in other middle class sub-
burban populations without occupational lead expo-
sures and not living near an expressway.
9. Higher levels of lead in blood were found in
children than in adults and higher blood leads were
seen in adult males than in adult females. In
children, females had slightly higher levels of
lead in blood than did males.
10. Approximately 50% of the airborne lead was in the
respirable range at distances from 5-100 feet from
the street. The ratio of particle sizes of lead
did not change with distances from the street (5-
100 feet). Air lead concentrations decline rapidly
in the first 50 feet from the street.
11. There were no increases in lead levels at intersections.
12. Air lead levels in 30 MPH speed zones were approximately
twice air lead levels in 45 MPH zones; however, this
difference was not statistically significant.
13. No significant relationship was found between lead
levels at peak traffic hours and traffic counts.
254
-------
14. A significant association between blood lead and
smoking was seen. Female smokers had significantly
higher blood lead levels than exsmokers and nonsmokers
In males, smokers and exsmokers had higher blood
lead levels than nonsmokers, although this difference
was not significant.
15. Lead contamination of blood samples is more critical
with the capillary blood (finger-prick) collection
than with the venous blood collection.
16. The accuracy and precision of the capillary blood
methodology is more variable than the venous blood
methodology.
255
-------
VI. RECOMMENDATIONS
1. To provide the data needed to accurately evaluate
the effects of automotive emissions of lead on the
population and the environment, more studies of this
type are needed. This type of data is essential
in making regulatory decisions regarding lead in
gasoline.
2. Efforts should be made to evaluate the levels of
lead present indoors in air and dust. Indoor lead
levels of residential homes, particularly those
using air conditioning units appear to be much
lower than in outdoor areas. More information is
needed regarding how and where people spend their
time at home. Factors such as the amount of time
spent indoors vs the amount spent outdoors need to
be addressed. Sampling should be done to better
quantitate the indoor vs outdoor environments in
various types of residential areas.
3. Evaluation of alternate dispersion mechanisms for
lead emissions from automobiles should be done to
better identify what happens to it in the environ-
ment .
4. Analytical methods used to quantitate the amount
of lead in people and the environment need to be
standardized so that improved comparisons and
256
-------
evaluations can be made -
257
-------
REFERENCES
1. Air Quality Criteria for Lead, Environmental Protection
Agency, Draft 3, August, 1977.
2. Johnson, D.E., J.B. Tillery, J.M. Hosenfeld and J.W. Register.
"Development of Analytical Techniques to Measure Human Exposure
to Fuel Additives." EPA Contract 68-02-0595, March 1974.
3. Galke, W.A., D.I. Hammer, J.L. Keil and W.W. Lawrence. "Environ-
mental Determinants of Lead Burdens in Children." Proceedings of
the International Conference on Heavy Metals in the Environment,
October, 1975.
4. Caprio, R.J., H.L. Margulis, and M.M. Joselo. "Lead Absorbtion
in Children and Its Relationship to Urban Traffic Densities."
Arch. Environ. Health 28:195, 1974.
5. Daines, R.H. , D.W. Smith, A. Feliciano and J.R. Troup. "Air
Levels of Lead Inside and Outside of Homes." Industrial Medi-
cine & Surgery 41:26, 1972.
6. Thomas, H.V. , B.K. Milmore, G.A. Herdbreder and B.A. Kogan. "Blood
Levels in Persons Living Near Freeways." Arch. Environ. Health
15:695, 1967.
7. Waldron, H.A. "Lead Levels in Blood of Residents Near M6-A38 (M)
interchange, Birmingham." Nature 253:345, 1975.
8. Jones, R.D., B.T. Commins and A.A. Cernik. "Blood Lead and Cor-
boxy Hemoglobin Levels in London Taxi Drivers." Lancet 2:302, 1972.
9. Johnson, D.E., J.B. Tillery and R.J. Prevost. "Levels of Platinum,
Palladium and Lead in Populations of Southern California."
Environmental Health Prospectives 12:27, 1975.
10. Tillery, J.B. and D.E. Johnson. "Determination of Platinum,
Palladium and Lead in Biological Samples by Atomic Absorption
Spectrophotometry." Environmental Health Prospectives 12:27. 1975.
11. Johnson, D.E., R.J. Prevost, J.B. Tillery, D.E. Camann, and
J.M. Hosenfeld. "Determine Requirements for Obtaining Baseline
Levels of Platinum and Palladium in Human Tissues." EPA Contract
68-02-1274, 1975.
12. Snee, R.D. "Evaluation of Studies of the Relation-
ship Between Blood Lead and Air Lead." Technical
258
-------
Report PLMR-8-77 Petroleum Laboratory, E. I.
duPont deNemours and Company, April 7, 1977.
13. Bridbord, K.C. Commission of the European
Communities Proceedings of the International
Symposium on Environmental Health Aspects of Lead.
Amsterdam, 1972.
14. Bratzel, Jr., M.P- and A.J. Reed. Microsampling
for bloodlead analysis. Clin. Chem. 20, 2, 217-221
(1974) .
15. Thompson, R.J., G.B. Morgan and L.J. Purdue. Analysis
of selected elements in atmospheric particulate matter
by atomic absorption. Air Quality Instrumentation,
Vol. 1, Ed. J.W. Scales, Instrument Society of
America Pittsburgh, Pa., 178-188 (1972).
16. Smith, Jr., R.G. and H.L. Windom. Analytical hand-
book for the determination of arsenic, cadmium,
cobalt, copper, iron, lead, manganese, mercury,
nickel, silver and zinc in the marine and estuarine
environments. Georgia Marine Science Center, Uni-
versity Systems of Georgia, Skidaway Island, Ga.,
Technical Report Series No. 72-6, p 49-50 (1972).
17. Hwang, J.Y., P.A. Ullucci, and C.J. Mokeler. Direct
flameless atomic absorption determination of lead
in blood. Anal. Chem. 45, 4, 795-798 (1973).
18. Kubasik, N.P. and M.T. Volosin. Use of the carbon
rod atomizer for direct analysis of lead in blood.
Clin. Chem. 20, 2, 300-301, (1974).
19. Delves, H.T. A micro-sampling method for the rapid
determination of lead in blood by atomic absorption
spectrophotometry. Analyst (London) 95, 431(1970).
20. Olsen, E.D. and P.L. Jatlow. An improved delves
cup atomic absorption procedure for determination
of lead in blood and urine. Clin. Chem., 18, 1312
(1972) .
21. Crisler, J.P., N.T. Lao, L.C. Tang, B.A. Serrano,
and A. Chields. A microsampling method for the
determination of blood lead. Miorcchem. J., 18,
77 (1973) .
22. Cernik, AA. and M.H.P. Sayer. Determination of lead
259
-------
in capillary blood using a paper punched disc
atomic absorption technique. Brit, J. Ind. Med.,
18, 392(1971).
23. Norral, E. and L.R.P. Butler. The determination
of lead in blood by atomic absorption with the
high temperature graphite tube. Anal. Chim. Acta.}
58, 47-56(1972).
24. Piomelli, S. A micromethod for free erythrocyte
porphyrins: The FEP test. J. Lab. Clin. Med., 81,
6, 932-940(1973).
25. Piomelli, S., B. Davidow, V.F. Guinee, Young, P.,
and G. Gay. The FEP (Free Erythrocyte Porphyrins)
test: a screening method for lead poisoning.
Pediatrics, 51, 2, 244-259, (1973).
26. Collison, A., F. Rodhey, and D. O'Neal. Determina-
tion of carbon monoxide in blood by gas chromato-
graphy. Clin. Chem., 14, 162-171(1968).
27. Mickey, M.R., O.J. Dunn, and V. Clark. Note on the
use of stepwise regression in detecting outliers.
Computers and Biomedical Research. 7:105-111(1967).
28. Dixon, W.J. Analysis of extreme values. Ann. Math.
Stat. 21:488-506(1950).
29. Sokal, R.R. and F.J. Rohlf. Biometry: the principles
and practice of statistics in biological research.
W.H. Freeman and Co., San Francisco, p 776(1969).
30. Dunn, O.J. and V.A. Clark. Applied statistics:
analysis of variance and regression. J. Wiley &
Sons, New York,(1974).
31. Bancroft, T.A. Analysis and inference for incom-
pletely specified models involving the use of pre-
liminary test(s) of significance. Biometrics, 20:
427-442(1964).
32. Johnson, D.E., J.B. Tillery, and R.J. Prevost. Trace
metals in occupationally and nonoccupationally
exposed individuals. Env. Health Persp. 2tf:151
(1975) .
33. Benson, F.D., J.J. Henderson, and D.E. Caldwell.
Indoor-outdoor air pollution relationships. EPA
Publication AP 112. August 1972.
260
-------
34. Yankel, A.J., I.H. van Lindern, and S.D. Walter. "Tiie Silver Valley
Lead Study. "The relationship of childhood lead
poisoning and Environmental Exposure." Journal
Air Pollution Control Association. 27:763(1977).
35. Tepper, L.B. and L.S. Levin. "A survey of arid
population lead levels in selected American com-
munities." EPA-Rl-005, U.S. Environmental Protec-
tion Agency, Washington B.C. 1972.
36. Azar, A., R.D. Snee and K. Hapibi. "An epidemic-
logical approach to community air lead exposure
using personal air samplers." Environmental Quality
and Safety, Supplement Volume 2 - Lead 254-288,
(1975).
261
-------
APPENDIX A
JUSTIFICATION FOR CHANGE OF STUDY SITE TO DALLAS
262
-------
JUSTIFICATION FOR CHANGE OF STUDY SITE TO DALLAS
San Antonio, Texas, was selected at the outset of this study as the
study site after preliminary studies of traffic densities and general
population characteristics. Thoroughfares with adequate traffic densities
were identified for each of the five major categories of traffic density for
which the study is designed: 1) 1,000 or less cars per day,
2) 7,500-12,500 cars per day, 3) 12, 500-17 , 500 car s per day,
4) 17,500-22,500 cars per day, and 5) 22,500 and more cars per day.
Demographic information from census records were used to identify
expected population characteristics. From the limited pre-analysis, it
was concluded that San Antonio cauld meet the study requirement for
adequate number of participant populations residing on thoroughfares with
appropriate traffic densities.
An in-depth site analysis was performed in San Antonio as the first
step in Contract 68-02-2227. Table 1 presents a summary of the results of
that site analysis. The residences indicated in Table 1 are located on
thoroughfares or residential streets in areas of the city appropriate for
study due to economic and ethnic requirements: areas which are primarily
white and middle class.
Residences which are not at an intersection are desired for the study,
This restriction would eliminate all corner houses. To meet the minimum
study requirement of 200 or more residences in each traffic density level,
corner houses would have to be considered in San Antonio for four of the
263
-------
Not on Corner
300
120
64
61
77
On Corner
--
159
156
122
35
Total
300
279
220
183
112
five traffic density classes, as seen from Table 1. For 1000 or less
cars/day, there are-more than adequate numbers. The 300 shown are
Table 1.
Summary of Site Analysis for Selected Areas of San Antonio
Traffic Density Number of Residences Within 100 ft. of Roadway
1 or less
7.5-12.5
12.5-17.5
17.5-22.5
22.5 or more
but a sampling of those available. For all other traffic density levels,
corner houses would have to be considered. For the highest two levels,
even consideration of. corner houses would not provide the 200 residences
required.
Review of the results of the in-depth site analysis has indicated that
San Antonio will not provide the study with the desired numbers of
residences for the purposes of recruitment of study participants. Three
principal factors are seen as the cause of this. First, much of the traffic
in San Antonio is carried by the freeway network, which is quite extensive,
rather than by non-freeway, mul+i-lane traffic arteries. This limits the
area of the city which can provide potential for study. A major freeway
264
-------
is presently under construction in the central northside area of the city.
Until it is completed, the existing multi-lane thoroughfares are carrying
the traffic load. Sufficient traffic densities were found to exist in that
area of the city to much greater extent than any other area. This,
coupled with economic and ethnic restrictions regarding the candidate
populations, provided the basis for limiting the study to the central
northside area. Second, much of the portions of thoroughfares with the
highest two levels of traffic density has been developed commercially
such that few residences are present in many of the best areas considering
the traffic densities. Third, large numbers of the residences found on the
thoroughfares are corner houses which face the side street rather than
the primary thoroughfare. Use of these residences in the study is not
desired due to possible error introduced by traffic from the side street.
A preliminary siting study in Dallas, Texas, has indicated that the
potential exists for finding adequate numbers of residences appropriate
for study in that city. The city has sufficient population to support a study
of this type, with 1.5 million people reported in the metropolitan area in
the 1970 census. The current estimate of population in the combined
Dallas/Fort Worth metropolitan area is approximately Z.5 million persons.
Though highly industrialized and commercialized with light and sophisticated
industries such as electronics, aircraft, and merchandizing, little or no
heavy or polluting industry exists in the area. Data on air quality indicate
there are no point sources of lead (telephone contact with Mr. Fred
Barnes, Chief, Air Pollution Control, City of Dallas, Texas).
265
-------
An abundance of heavily trafficked, multi-lane thoroughfares exist in all
parts of the city: some in excess of 30,000 cars per day. The traffic
system in Dallas has historically been designed around such multi-lane
thoroughfares, and a network of such arteries spreads across the city.
Figure 1 shows a map of the metropolitan area of Dallas, Texas.
Results of preliminary counting of residences on a sampling of
thoroughfares in Dallas are presented in Table 2. Only single family
residences which face the roadway were counted. Thus, corner houses
facing side streets are not included in the count. Corner houses facing
the primary thoroughfare were not excluded from the count. It is
estimated that no more than 20% of the residences counted were corner
houses and, thus, 80% or more would be appropriate for this study.
The streets shown are but a sampling of those available throughout the
city and were selected from areas estimated to have acceptable economic
and ethnic characteristics. From Table 2, it is apparent that the minimum
study requirement of 200 or more residences in each traffic density class
should be easily obtained in that city.
Because working in Dallas, rather than San Antonio, will require
significant efforts away from the facilities of Southwest Research Institute,
there will be need for establishing a base for activities in that city.
Preliminary contacts have been made with personnel of the Center for
Urban and Environmental Studies at Southern Methodist University.
Preliminary arrangements have been made to devote a graduate student
part-time to this project to serve as on-site coordinator for study activities
266
-------
in Dallas. Through the office of the graduate student, the University
will provide office space and telephone facilities for utilization by the
study team while on location in Dallas.
Contacts have also been made with the air pollution control
group of the City of Dallas, and they are very interested in assisting in
the study proposed for their city. The traffic department of Dallas has
also indicated that they will assist in the study.
267
-------
TABLE 2. SINGLE FAMILY RESIDENCES: DALLAS
Number of Residences Within 100 Feet of Roadway
Which Face the Roadway
Street
Mockingbird Lane
Jupiter Road
Garland Road
Forest Lane
Marsh Avenue
Lemon Avenue
Inwood Road
Gaston Avenue
Illinois Avenue
Buckner Blvd.
East Grand Avenue
Hampton Road
TOTAL
12.5-17.5
Thous. cars/day
184
60
120
16
23
50
17.5-22.5
thous. cars/day
127
29
10
27
78
32
69
22.5 -
thous. cars/day
68
10
29
32
146
453
372
285
80% of Total
362
297
228
268
-------
Figure 1.
Expressways and Major Thoroughfares
Dallas, Texas
-------
APPENDIX B
LETTERS OF PERMISSION TO PROCEED FROM LOCAL GOVERNMENTS
270
-------
CITY OF DALU\S
March 17, 1976
R. John Prevost
Senior Research Analyst
Department of Social Sciences
Southwest Research Institute
Post Office Drawer 28510
San Antonio, Texas 78284
Dear Mr. Prevost:
In response to your letter of March 4, we certainly
have no objection to your conducting the proposed
study of blood lead levels in people exposed to
particularly high levels of air pollution by auto-
mobiles in Dallas. We would appreciate receiving
a copy of whatever information you develop.
Sincerely,
Berry, MVDV,-^M .f . H.
Director of Public Healtl(
ELB:tg
271
DEPART MCNT OF PUt'UC HEALTH
-------
CITY OF DALLAS
March 10, 1976
Mr. R. John Prevost
Senior Research Analyst
Department of Social Sciences
Southwest Research Institute
Post Office Drawer 28510
San Antonio, Texas 78284
Dear Mr. Prevost:
This is to acknowledge our telephone conversation this
date dealing with your plan to conduct a public health
survey in Dallas under contract with the U. S. Environmental
Protection Agency.
I do appreciate your informing me that you will have
representatives conducting a door-to-door survey using
an EPA questionnaire. Since your proposed survey as
described will involve neither charitable solicitation
nor home solicitation as defined in existing City ordinances,
no permit of any type from this department will be required.
Please contact me if you need further information.
Sincerely,
Charles H. Vincent, Director
Department of Consumer Affairs
CHV/jec
272
DEPARTMENT OF CONSUMER AFFAIRS OlYHALL P'L1. *S TEXAS 7SZOI rELEPHOIvf 214 740-9711
-------
CITY OF DALLAS
March 10, 1976
Mr. R. John Prevost
Senior Research Analyst
Department of Social Sciences
Southwest Research Institute
8500 Culebra Road
P. 0. Drawer 28510
San Antonio, Texas 78284
Dear Mr. Prevost:
We thank you for sending us your plans March 3
on lead surveys in Dallas. Our department has
no objections to these plans.
Please let us know if we can be of assistance,
and keep us informed of the results of this
survey -
Very truly yours,
^ _
George"1\7. Ilintgen, P.E.
Environmental Research Coordinator
Environmental Health and Conservation
GWH : wg
273
ENVIHONMtNT AL HfALTH AND CONSERVATION
-------
MAYOR
Ashley H. Priddy
MAYOR PRO TEM
Richard L. Jones
COUNCILMEN
George Rather Jones
Robert F. See
Ralph W. Smith. Jr.
W. L. Todd, Jr.
JUDGE
Pat A. Robertson
THE TOWN OF
4700 DREXEL DRIVE, DALLAS 75205
Telephone 521-4161
March 23, 1976
SECRETARY AND
TREASURER
Jerry T.Btll
TOWN ATTORNEY
H. Lou Morrison, ],
CHIEF OF POLICE
AND FIRE
r. H. Gardntt
Mr. R. John Prevost
Southwest Research Institute
8500 Culebra Road
P. O. Drawer 28510
San Antonio, Texas 78284
Dear Mr. Prevost:
This will acknowledge your letter of March 19 in which you indicate
the selection of Mockingbird Lane in Highland Park for a portion of
your surveys.
The Town attempts to control the activities of solicitors of all types
through the Police Department. For this reason, and because of the
possibility of traffic surveys, you are asked to contact the Police
Department prior to beginning any activity of the type described.
Sincerely yours.
JDHrmrn
CC: Chief W. H. Gardner
I. D. Hancock, Jr. ,
Town Administrator
274
-------
,^-r -T^^F^^TT-"'- 7
! :': !,-: La./.!..;'.'- * k- .
n. 'f.'-i" ; :-:v '. " '-< ! "-=-" - " -^
w>.-'^' V"l*:''-'-:''':'"': N i:NL-i'- - ;-£*
ssoo
E. WILSON GERMANY
Usyor
FRED N PEEK
PETER S. CHANTILIS
IELAND D. NELSON
Cilv Manager - tier*
.O. a OX SOO5
75205
March 9, 1976
Mr. R. John Prevost
Senior Research Analyst, Department of Social Sciences
Southwest Research Institute
8500 Culebra Road
P. O. Drawer 28510
San Antonio, Texas 78284
Dear Mr. Prevost:
I have your letter of March 4, 1976, in which you indicate that you are
going to conduct a study in the Dallas metropolitan area under an EPA
contract. More definitely, you are going to study and examine the re-
lationship between traffic density and lead in the air and blood of exposed
populations. Included in that sample study will be a portion of Univer-
sity Park.
In response to your request to use a portion of the City of University
Park for your study, I would say that we would have no objection what-
soever. In fact, if we can help y^u in any way in your study, please let
us know. We would be interested in the results of your study.
Sincerely yours,
Leland Nelson
City Manager
LN:dh
cc: Mr. Jim Murphy, City Engineer
Mr. J. D. Brown, Chief of Police
275
-------
L/Ou RAINCi
MAY HP
IOMKhELE
MAW "PO TEM
ROM P i HPAND
Dt I Of-1 A LE#l->
CHAPLK:G.CLACK
r.iM NICHOLSON
Uw.VfilC GFNTSCH
VLPti'Jti £ . EMORY
JOHNIE I' I.UfjHALL
Post Office Box 189 / Garland, Texas 75040 CHAS.E.D;CKWOPTH
CIT'r MANAU P
City of Garland
March 30, 1976
Mr. R, John Prevost
Senior Research Analyst
Southwest Research Institute
P. 0. Drawer 28510
San Antonio, Texas 78284
Dear Mr. Prevost:
The purpose of this letter is to confirm the results of our
visit Monday, March 29. The Southwest Research Institute is
authorized to conduct the study to examine the relationship
between traffic densities and lead in the ambiant air and
blood in the exposed populations on Walnut Street in Garland.
As we discussed, you agreed to supply the City of Garland with
the names of all project researchers assigned to Garland, the
locations to which they will be assigned, and the location of
traffic densities and air quality measuring devices. This in-
formation should be supplied to Mr. Bill Cox, City Sanitarian.
Should further information be required, please advise.
Sincerely,
Donald E. Paschal, Jr,
Community Services Administrator
DEP/hc
cc: Linda Johnson
276
-------
THE CITY OF
RICHARDSON
RICHARDSON. TEXAS
May 2.6, 1976
"'TH A
'"< COUNCIL
iiYMONO D. NOAH
(Mayor)
"M J. EWBANK
!Voyor Pro Tern)
--OMAS G. HARDY, JR.
-3WARD D. HERN
= ETER L. ROLLOSSON
.:SEPH E. DUGGAN, JR.
VARTHA E. RITTER
= :S HUGHEY
1 ~i Manager
Mr. R. John Prevost
Senior Research Analyst
Department of Social Sciences
Southwest Research Institute
Post Office Drawer 28510
San Antonio, Texas 7828^
Subject: EPA Contract 68-02-2227
Dear Mr. Prevost:
This letter is to advise that Richardson will participate in the
above cited EPA study at the following locations.
1. Belt Line Road from Central Expressway west to Coit Road.
2. Coit Road from Spring Valley Road north to Arapaho Road.
If further locations are considered in your survey, please advise
this office prior to commencement of survey work.
Yours very truly,
Bob Hughey
BH:br
cc: Mr. Ted C. Willis
Assistant City Manager
cc: Mr. Kenneth R. Yarbrough
Chief of Police
277
'THE ELECTRONICS CITY OF THE WORLD" A CITY OF FINE CHURCHES, SCHOOLS, HOMES AND INDUSTRY
P 0. BOX 309 RICHARDSON, TEXAS 75080 AREA CODE 214 235-8331
-------
THE CITY OF
RICHARDSON
RICHARDSON, TEXAS
May 27, 1976
UNCIL
ID D. NOAH
tor)
EWBANK
' Pro Tern)
G. HARDY, JR.
5 D. HERN
L. ROLLOSSON
E. DUGGAN, JR.
k E. RITTER
IGHEY
lager
Mr. R. John Prevost
Senior Research Analyst
Department of Social Sciences
Southwest Research Institute
Post Office Drawer 28510
Subject: EPA Contract 68-02-2227
Dear Mr. Prevost:
This letter is confirming our telephone conversation today whereby
Mr. Hughey is granting permission to extend the boundaries of your
EPA study from Coit Road at Spring Valley Road north to Campbell
Road to include the nine residences between Arapaho Road and
Melrose Drive.
Yours very truly,
(Mrs.) Barbara A. Rusch
Secretary to Mr. Hughey
/BR
cc:
Mr. Ted C. Willis
Assistant City Manager
278
THE ELECTRONICSCITY OF THE WORLD" A CITY OF FINE CHURCHES, SCHOOLS, HOMES AND INDUSTRY
P 0. BOX 309 RICHARDSON, TEXAS 75030 AREA CODE 214 235-8331
-------
pn,i' n on r'.n't
\ii\k iJ i uAlitit
ALL AMERICA CITY
i 1
np
! !/
March 22, 19?6
R. John Prevost
Dept. of Social Sciences
Southwest Research Institute
P. 0. Box 28^10
San Antonio, TX 7325k
Dear Mr. Prevost:
I have been asked tc answer your letter dated March YJ. 1976. We
will be happy to cooperate in a_.y way possible in your public health
survey. Please feel free to call on me.
Yours truly,
. ''# . ( '''<'/..','(
''Burl Cockrell, F:. S.
Director of Environmental Health
BC ''j'ni
cc: Kenneth Burr, Chief of Police
279
CITY OF GRAND PRAIRIE, P.O. BOX 11, GRAND PRAIRIE, TEXAS 7f,050 TEL. (AC 214) 2G3-5221
-------
v i - i *.-
Oty of /irhng tonTexas
Box 231
Zip Code 76010
Arlington Phone
(817)275-3271
Dallas Phone
(214)262-4660
Mr. R. John Prevost
Senior Research Analyst
Department of Social Sciences
Southwest Research Institute
P.O. Drawer 28510
San Antonio, TX 78284
Dear Mr. Prevost:
In regard to your correspondence of March 17, 1976, the City of
Arlington has no objections to the lead study work to be conducted
in the City by the Southwest Research Institute.
In this connection, please inform me of the proposed locations for
automatic traffic counting, and particulate air sampling devices within
the City at least one week prior to their planned installation. City
Public Safety, and Traffic and Transportation personnel will review the
locations to determine their adequacy from a safety standpoint. You
will be notified only if the locations proposed are not satisfactory.
Enclosed for your information is correspondence sent to your Project
Coordinator at SMU.
Sincerely,
Ross B. Calhoun
City Manager
cc: Linda Johnson, Project Coordinator
SMU
enclosure
280
Tori J V.v-V'j-ift M.!u». Council S J Slo\ ill M ivor Pn>T, n Df '"i G Aii'\,i"Jt". CiroV' W SnuVr H iroM P.irh'r'.m K i ill Snollon M. n
-------
Giy of A-lrngtonTexas
Box 231
Zip Code 7G010
Arlington Pnonu
(817)275-32/1
Dallas Phone
(214)262-4660
March 29, 1976
TO WHOM IT MAY CONCERN:
This letter is to certify that the Southwest Research Institute
has the permission of the City of Arlington to conduct a house-
to-house public health survey.
This survey is part of a study by the Institute to examine re-
lationships between traffic densities, and lead in the ambient
air and lead in the blood of exposed populations.
The study is being conducted by the Institute for the U.S.
Environmental Protection Agency.
Thank you for cooperating with the survey team.
Sincerely,
Ross B. Calhoun
City Manager
/ss
281
lomj Vanctajriff, Mjyor. Carnal S J Stovjil. M:-i/0'Pro-Tom. Dr R G Alexander. C.vo'yn W Snider H iroM P iMuison Rjiph Sh.'iion Mirth,, V\i'ker
-------
APPENDIX C
JUSTIFICATION FOR HOUSEHOLD HEALTH SURVEY FOR LEAD
282
-------
JUSTIFICATION OF HOUSEHOLD HEALTH SURVEY FOR LEAD
A. Supporting Statement Justification
1. This new questionnaire is required to obtain information on
a population on human volunteers living in a major metropolitan area
(San Antonio, Texas). The primary objective of this program is.to
determine body burden levels of lead in populations of a major metropolitan
area without occupational exposure to lead. The amount of lead found in
blood of the human participants, lead in soil, lead in house dust, lead
in hand wipes, lead in tap water, and lead in ambient air will be correlated
with traffic densities. No suitable questionnaire forms of this type are
currently available within the Environmental Protection Agency or known
from other agencies which meet the design requirements of this study.
This questionnaire is to be utilized on EPA contract number 68-02-22.2,1
entitled "Epidemiologic Study of the Effect of Exposure to Automobile
Traffic on the Blood Levels of Persons in Selected Age Groups". A copy
of the Work Plan for this program is attached.
It has been shown that the levels of lead in body burdens of
populations are related to factors such as age, sex, race, cigarette
smoking, and use of certain types of lead-containing articles to prepare
or serve food. This questionnaire form would provide the necessary
information to assist in the final statistical evaluation of collected data
that would guide the EPA in its assessment of potential health effects
associated with body burdens of lead. This questionnaire will solicit
283
-------
personal information from some 1Z50 individuals living within the study
area with regard to age, sex, smoking habits, and occupation. From
a maximum of 1250 subjects completing the questionnaire, 440 will be
selected to provide the study with the following three age groups:
preschool, 1-6 years; adult, 20-59 years; retired, 60 years and over.
Each of these individuals would then be sampled twice for blood. These
samples would be analyzed for concentrations of lead. The mean
concentrations of each of the three age groups will be calculated along
with standard deviations and confidence limits for the means. These data
will then be correlated with traffic density levels to determine the
relationship. Additional statistical correlations will be made to determine
the possible relationship between concentrations of lead and the covariate
information collected on the questionnaire.
This project will be conducted as a contract with SwRI under
the technical direction of Dr. Donald E. Johnson. The attached Work
Plan from Southwest Research Institute provides more details on the
plans for this survey.
2. The data collected using this survey instrument and the
follow-up statistical analysis will be utilized by the Environmental
Protection Agency to assess the potential health effects associated with
body burdens of lead as related to automobile traffic.
284
-------
3. There are some preliminary data in the literature which
indicate that certain parameters such as age, sex, ethnic origin,
smoking history, etc. , are associated with the body burdens of lead.
This in-depth survey of human subjects and comparisons to traffic density
will provide more definitive information about these relationships.
B . Description of Survey Plan
1. The survey design will be aimed at determining the body
burdens of lead of populations living within 100 feet of streets carrying
traffic densities which vary from less than 1000 cars per day to greater
than 25, 000 cars per day. It is estimated that the total population living
within 100 feet of such streets within the United States is significantly
over one million.
2. The survey is designed to initially contact and survey
1,000 households, ZOO at streets carrying each of the following approximate
traffic densities: less than 1000 cars per day, 10, 000 cars per day,
15, 000 cars per day, 20, 000 cars per day, and 25, 000 or more cars per
day. Each household will be contacted by a personal interviewer. A set
of general information will be obtained from the person interviewed. If
appropriate age and occupational status are determined, the person will
be asked questions 12 and 13 of the Household Health Survey for Lead:
"Would you participate in a health survey as a paid volunteer ? Other
members of household?" It is anticipated that approximately 440 subjects
who respond to this questionnaire will be chosen, on the basis of age,
285
-------
occupation, and location categories, to participate as paid volunteers.
Each subject will be sent a letter announcing that he or she has been
selected for the survey; a time and place will be specified as to when a
team will visit the subject's house to accomplish the blood and soil
samples.
A pre-test of the questionnaire has been conducted. This
pre-test was conducted with five representative individuals from the staff
of Southwest Research Institute, and no difficulties were encountered.
The question of non-responders for this particular survey does not
appear to represent a problem, since the burden for the individuals at
households to respond to this is minimal and number of non-responders
is expected to be very low.
3. The statistical design of the project will come primarily
from Mr. David Camann with Southwest Research Institute. The
Environmental Protection Agency's statistician,
, has reviewed this protocol.
4. Name of the Contractor: Southwest Research Institute.
Contractor's role: the primary role is to collect information and provide
a final report to include statistical evaluation of data and conclusions
regarding the body burdens of lead found and the correlation to traffic
density.
Southwest Research Institute guarantees confidentiality of
the collected data to the subjects surveyed. No direct reference to the
286
-------
collected data using the subject's name or address will be made. The
compilation of subject's name and address will be maintained in a
confidential file and will not be directly related to any of the collected
data. Each subject surveyed will be assigned a code number, and the
coded number with the individual's name will be maintained only in the
files of Dr. Donald E. Johnson, principal investigator, at Southwest
Research Institute.
C. Time Schedule for Data Collection and Publication
The contract term is for twelve months with an additional 60 days
for review of the draft final report. The final report should be completed
September, 1976. Monthly progress reports will also be submitted. It
is estimated that the elapsed time between the completion of data collection
and the issuance of the first published report will be six months .
D. Consultations Outside the Agency
Southwest Research Institute has consulted with the City of San
Antonio for assistance and coordination in the conduct of the survey.
E. Estimation of Respondent Reporting Burden
The estimation of respondent reporting burden for the Household
Health Survey is ten minutes, with a maximum of 20 minutes. This
estimate is based on a preliminary pre-test survey of randomly selected
Southwest Research Institute employees and should be the amount of time
necessary for even a large household to complete.
287
-------
F. Sensitive Questions
Questions 7 and 8 of the General Information Questions ask
questions regarding diagnosed incidence of lead poisoning and
approximate age of the home. This information is essential in order
to provide an assessment of the background of lead poisoning and potential
for lead poisoning due to pica. Questions 10 and 11 of the General
Information and 5, 6, and 7 of the Individual Information Questions provide
data regarding the occupation of the individual. These questions are
essential to establish that the individual is not occupationally exposed to
lead and that he generally spends his day in the vicinity of his home.
Questions 8 and 9 of the Individual Information establish the individual's
smoking history. This is essential because this study is concerned with
air-borne pollutants, and cigarette smoking affects the absorption rate
of such chemicals. Other questions refer to some of the socio-economic
background necessary to make valid statistical comparisons with other
areas.
288
-------
QUESTIONNAIRE 4Z94-1
HOUSEHOLD HEALTH SURVEY FOR LEAD
Name
Last Name, First Name M. I.
Address: Street_
City_
Zip Code
Telephone
General Information Questions
1. How many persons reside in your household?
2. For each person in your household, including yourself, please indicate the age and sex,
beginning with the oldest and proceeding to the youngest:
Age
Sex
3. Households with minor children present:
Do both parents reside in household?
4. What educational level has been completed by the head of your household?
1 - Less than 8th grade 5 - College - Incomplete
2 - 8th grade 6 - College - Complete
3 - High School - Incomplete 7 - Graduate School
4 - High School Completed
5. Do you cool your home with any of the following appliances?
1 - Central air conditioning 4 - Window fan
2 - Window air conditioner 5 - Ceiling exhaust fan
3 - Evaporative cooler 6 Other
6. Are any of the following articles used in preparing or serving meals in your household:
1 - Unglazed pottery made in Mexico 3 - Hand painted flatware
2 - Glazed pottery made in Mexico 4 - None of these
7. "Has any member of your household been diagnosed for lead poisoning?
If so, which member or members?
8. What is the approximate age of your home? years
9. .Are you ever aware of a smell o.r odor from automobile traffic?
Inside house
Outside house
10. Do any members of your household have occupations which normally take them away from
home during their work hours?
If so, which members?
11. Do any members of your household have occupations which normally do not take them a\vcv
irom home?
li so, which members?
12. Would you participate in a health survey as a paid volunteer ? (P^iid $15 for blood, dust,
and soil samples)
13. Other members of household?
289
-------
Individual Information Questions
1. Date of Birth Month Day Year
2. Sex: 1 - Male 2 - Female
3. What is your marital status?
1 Single 4 - Divorced
2 - Married 5 - Widowed
3 - Separated 6 - Other
4. How many years have you lived in San Antonio? years
At your present address? years
5. What is your usual occupation? (Please specify)
Are you presently occupied in this manner?
6. Does your occupation usually take you away from home?
7. Which of these best describes your present occupational status?
1 - Employed fulltime (including self employed)outside home
2. Employed part-time outside home
3 - Employed inside home
4 Unemployed
5 Housewife
6 - Student
7 Play/Nursery School
8 - Pre-school
9 - Retired
8. Have you ever smoked as many as five packs of cigarettes, that is, as many as 100
cigarettes during your entire life?
Do you now smoke cigarettes?
9. If you are a current or an ex-cigarette smoker:
a. How many cigarettes do (did) you smoke per day
l--Less than 1/2 pack per.day (1-5 cigarettes per day)
2--About 1/2 pack per day (6-14 cigarettes per day)
3--About 1 pack per day (15-25 cigarettes per day)
4--About 1-1/2 packs per day (26-34) cigarettes per day)
5--About 2 packs per day (35 or more cigarettes per day)
b. How old were you when you first started smoking? Years
c. How old were you when you last gave up smoking, if you no longer smoke? Yeai
YOU HAVE COMPLETED THE QUESTIONNAIRE-THANK YOU
Intei-vjewer Note and Record
1. Site Number
2- Distance to center of primary roadway
3. Distance to other nearest roadway ^~~~
4. IMNOWO 123456
290
8/7/75
-------
iTANDARD FORM NO. 83
JFFICE Of MANAGEMENT
.NO BUDGET
CLEARANCE REQUEST AMD NOTICE OF ACTION
(Under Federal Reports Act and Bureau of the Budget Circular No. A-40, as amended)
FOR O.M.B. USE
MPORTANT Submit the required number of copies of SF-83, together
with the material for which approval is requested to:
READ INSTRUCTIONS BEFORE COMPLETING FORM
CLEARANCE OFFICER
OFFICE OF MANAGEMENT AND BUDGET
WASHINGTON, D.C. 20503
^ PART A - REQUEST BY FEDERAL AGENCY FOR CLEARANCE
Items marked with asterisk may be omitted for preliminary plans or recordkeeping requirements
, SEND "NOTICE OF ACTION" TO: Name and mailing address
2. Bureau and division or office originating
request
3. Name(s), title(s), and telephone numbers of
person(s) who can best answer questions
regarding request.
IFORM OR
DOCUMENT
IDENTIFI-
CATION
4. Title of form or document submitted
HOUSEHOLD HEALTH SURVEY FOR LEAD
"5. Agency Form Number(s)
6. Type of form or document
i Q Application
2 fj] Program evaluation
3 Q Other management
report
4[^Statistical survey
or report
51 | Preliminary plan
or contract
6 O Recordkeeping
requirement
7 Q Other - Specify-
7. Current (or former) O.M.B. clearance
Number
Expiration date
8. Requested expiration date
June 1976
9. Type of request
i [Xj New
z[ | Revision
3 I I Extension
(No change)
4I | Reinstatement
*10. Frequency of use
i Q Single time s D Quarterly
2 (Xj On occasion 6 O Semi-annually
3 Q Weekly 7 O Annually
4 D Monthly e D Other <** i
11. Related forms or documents (Give O.M.B. number. Enclose in
parentheses any to be replaced.)
12. Catalog of Federal Domestic Assistance program number (if applicable)
COLLECTION
AND
RESPONDENTS
*13a. Collection method
( Check as many as apply)
] Personal interview
Other - Describe
*13b. Collected by -
4 Q Agency
y[XI Contractor
6 I I Other - Describe -
14a. Type of respondents involved
(Check predominant one)
i [X] individuals or households
2 | | Business firms (non-farm)
3 Q Farms
4 Q Government agencies
5 Q Other - Describe ,
*15. Summary of estimated
respondent burden
a. Estimated number of
respondents
14b. Brief description of respondents
(i.e., "households in 50 largest
SMSA's; "retail grocery stores")
b. If sample, approximate
number in universe
c. Reports filed annually by
each respondent (item 10)
d.Total annual responses (a X c)
e. Estimated average number of
man-hours required per response
f. Estimated TOTAL MAN-HOURS
of respondent burden (d X e)
Number
1250
over
one million
onee
1250
1/3
416
AUTHORITY
AND CONFI-
DENTIALITY
*16a.ls report form -
i Q Voluntary?
2 S3 Required to obtain benefit?
Mandatory? Cite statute
*16b. Does your agency pledge
confidentiality?
i [^ Yes 2QNo
CONSULTA-
TIONS OUT-
SIDE AGENCY
17. In developing the report form or other documents, were
consultations held with individuals or organizations
outside your agency?
Yes - If "yes," identify persons and describe outcome in
SUPPORTING STATEMENT. (See instructions)
NO
CERTIFICATION BY AUTHORIZED OFFICIALS SUBMITTING REQUEST - We certify that the form or other document submitted (or approval is
necessary for the proper performance of this agency's functions, that the information requested is not available from any other source, to the best
of our knowledge, and that the request is consistent with applicable O.M.B. and agency policy directives. Signature and title of:
Approving official for agency
Date
Agency clearance officer or other agency official
Date
-------
APPENDIX D
AppendixD contains the air lead concentrations and corres
ponding traffic counts. The abbreviated variables and their units
are explained below.
Traffic Class Estimated traffic class
Street Name of street
Date Date on which 24-hour count was started
Traffic Count Actual traffic count
Air Pb Air lead concentration ((a/m^
292
-------
APPENDIX D
Traffic
Class
01000
01000
01000
01000
01000
0J000
01000
0J000
01000
.01000
01000
,01000
01000
01000
01000
01000
01000
01000
01000
01000
01000
.01000
01000
01000
01000
05000
05000
05000
05000
05000
05000
05000
05000
05000
05000
05000
05000
05000
05000
05000
Street
MIMOSA
M I MOS A
MIMOSA
MIMOSA
MIMOSA
M I MOS A
MIMOSA
MIMOSA
COSTA MCSA
COSTA MESA
COSTA MPSA
COSTA MESA
COSTA MRS A
COSTA MESA
COSTA MPSA
MY9TICH
MYRTICC
MYRTICE
MYRTICE
MYRTICE
MYRTICE
MYRTICE
MYRTICE
MYRTICE
MYRTICE
BLUFFVIEW
BLUFF VI Erf
BLUFFVIEW
BLUFFVIEW
BLUFFVIEW
BLUFFVIEW
BLUFFVIEW
MIDWAY HILLS
MIDWAY MILLS
MIOWAY HILLS
MIDWAY HILLS
M 10 WAY HILLS
MIDWAY HILLS
MIDWAY HILLS
MIDWAY HILLS
Date
05/26/7*
?5/27/76
05/28/76
05/29/76
06/02/76
06/^3/76
06/^4/76
06/05/76
06/05/7*
06/06/76
06/07/76
."16/08/76
06/09/76
^6/1(^/76
(36/1 1/76
26/15/76
06/16/76
'16/17/76
06/18/76
06/19/76
06/20/76
06/21/76
^6/22/76
06/23/76
06/24/76
05/28/76
05/29/76
05/3CV76
05/31/76
06/^1 /76
06/02/76
06/03/76
06/^3/76
06/^5/76
^ 6/06/7 6
^6/^7/76
,36/^8/76
^6/^9/76
;'56/hV76
:»6/1 1/76
Fr =i f £ \ r
Co-int
336
164
375
358
3^5
4-17
343
421
32 (-1
.3-11
3 15
325
327
321
314
69
4 1
18
26
06
J6
12
33
21
146
4947
3 6 "17
3193
5977
*>56R
6146
62-P0
6577
7344
6634
-! 1 6 f'
7555
VT'-ll
5338
Air Pb
.65
1 .22
1 .v3
.35
.74
.89
1 ,f)5
1.16
.46
.32
.33
.65
.56
1 .05
.36
.61
.46
.4 i
.41
.71
. 89
.43
.33
.22
.89
1 .50
.61
.60
1 . 02
1.21
1 . 2 -:J
1.33
1.32
.82
.74
1 .53
1.25
.92
1.08
.62
293
-------
APPENDIX D (CONTINUED)
05000
05000
05000
05000
05000
05000
0 50 rA 0
05000
05000
10000
10000
10000
10000
10000
10000
10000
10000
10000
10000
10000
I0000
10000
100:50
10000
10000
10000
10000
10000
10000
1 0000
10000
10000
10000
1 5000
1 5000
1500;'
1 500 }
\ 50 0 ;i
15000
TVr'Cc;
RARN'FS
R ARM ES
RARNFS
RAINES
RARNFS
BARMF'S
RARNFS
RA'A'ES
PRA I PI
PRA I R!
PRA IR I
p P A I o i
PRA IP I
PRAIPI
PRA IP I
PRA IP!
OATCS
OATHS
OATFS
OATHS
OATFS
OATFS
OATnS
OATFS
OAT-?
OATHS
OATFS
OATFS
OATPS
OATFS
OAT^S
OATF.S
r-i-.'oon
IN..1 001
I'J.;oOf;
IN (001
I,'! vOOO
L>,: A'Onn
RPIDO1"
nR IOGC
nR Ll'li:
RR I10r-
RR ID OH
^.R lOO^
RR IDGE
PR [OOP
R.RlDOi;
~ CRFFK
E CRFFK
F CRCCK
F CRFFK
F CRFFK
p CR-:FIC
F CRFF!<
F CPFFK
o^t°
06/17/76
0V 1 8/76
06/ 1 9/76
OV20/76
06/21/76
06/22/76
06/23/76
06/24/76
06/25/76
06/04/76
06/05/76
06/06/76
06/07/76
06/08/76
06/09/76
0 VI 0/7 6
06/1 1/76
06/1 1/76
06/12/76
0VI3/76
06/14/76
06/15/76
06/16/76
06/17/76
06/18/76
06/29/76
06/30/76
07/01/76
07/03/76
07/04/76
07/05/76
07/06/76
07/07/76
05/26/76
05/2c
-------
APPENDIX D (CONTINUED)
Traffic
Class
1 5000
15000
15000
15000
15000
1 5000
1 5000
15000
15000
15000
15000
1 5000
15000
15000
15000
15000
1 5000
15000
15000
1500^
15000
15000
15000
20000
200^0
20000
20000
20000
20000
20000
20000
20000
20000
2VW*
2'/J . ,"j ^
H943 .62
M4'iA |.63
13 66 '3 ..74
M 6 7 -' .53
1 7^ M .37
191^2 .97
13475 .94
16 P.I .54
M9!^ I.0.?
1 7221 1 .42
H P P 1 . P
16V55 .67
15569 .53
R '252 .69
-VJ431 .76
P36T 1 . |'J
1 >.} ^, .V4
P 1 '?": 1 . . ;;
H.^'1 i.";-ri
PW^ 1.62
1 5 12 3 1 . -!- 3
21 177 .62
2n 3 '"-3 1 . 6 J
I>1|4 .31
19625 l.-.)o
2 i.^9 .97
2 '6 -i- 1 .M,-
196o^ 1.42
2PM 1.19
R3r.3 1.42
2-U7- 1.16
16 M- l.j?
17 '. '-. .25
ispii ?.-.: 3
P534 .77
06/-17/76
295
-------
APPENDIX D (CONTINUED)
Traffic
Class
20000
20000
20000
20000
20000
20000
20000
20000
25000
25000
25000
25000
25000
25000
25000
25000
25000
25000
25000
25000
25000
25000
25000
25000
25000
25000
25000
25000
25000
25000
25000
25000
25000
Street
MOCKINGBIRD
MOCKINGBIRD
MOCKINGBIRD
MOCKINGBIRD
'{OCKINOBIRD
MOCKINGBIRD
MOCKINGBIRD
MOCKINGBIRD
N. W. HWY
N. W. HWY
N. W. HWY
N. W. HWY
N. ;v. HWY
N. W. HWY
N. W. HWY
N. W. HWY
N. W. HWY
N. W. HWY
N. W. HWY
N. W. HWY
N. W. HWY
N. 'ft. HWY
N. W. HWY
N. W. HWY
N. W. HWY
N. A!. HWY
N. W. HWY
CO IT
CO IT
COIT
COIT
COIT
COIT
Date
06/24/76
06/25/76
06/26/76
06/27/76
06/28/76
06/29/76
06/30/76
07/01/76
05/27/76
05/28/76
05/29/76
05/30/76
05/31/76
06/02/76
06/03/76
,36/04/76
06/05/76
06/06/76
06/07/76
06/04/76
06/05/76
06/06/76
06/07/76
06/08/76
06/09/76
06/10/76
06/1 1/76
06/22/76
07/02/76
07/04/76
07/05/76
07/06/76
07/07/76
fr^flc
Count
13790
21 197
10877
1 3878
17902
18495
16857
13345
42529
45805
.33762
33700
42800
41079
42899
41602
3^846
33716
38950
4038 1
3801 8
31477
32835
31217
.35137
3^345
31 998
.31542
26355
20483
20481
24572
164.30
Air Pb
1.37
1.37
.73
.93
1 .20
1.94
1.01
.97
1.34
4.93*
2.33
2.06
3.56
1.51
1.91
2.32
.93
.82
1.89
2.22
.95
.88
1.51
.69
.74
.53
.22
.67
.38
.63
1 .61
.84
2.40
Rejected as outlier (P « .001)
296
-------
APPENDIX E
The variables tap water lead, soil lead, indoor
dust lead, windowsill wipe lead, and traffic for each
household are given in Appendix E. The abbreviated
captions and units are given below.
Hsld. ID
Site
H20
Soil
# Days
Idust
Wsill
Traffic
Household ID
Traffic density site
Lead concentration in tap water
(yg/ml)
Lead concentration in soil (ug/g)
Actual number of sampling days
for indoor dust
Lead concentration in indoor
dust (ug/cm2)adjusted to
to a 28 day base
Lead concentration in windowsill
wipe dust (ug/cm2)
Actual traffic count
297
-------
APPENDIX E
Hsld.
.ID Site
H20
Soil #Days Idust WslH Traffic
150.0
2435
5115
5880
7240
0005
0010
0015
0,025
0050
0055
0060
0065
0070
0075
0080
0090
0100
0J05
0120
0130
0135
0140
0J45
0J50
0155
0160
0165
0170
0180
0250
0255
0260
0265
0270
0275
1395
.1400
1405
1410
0.0009
0.0000
0.0005
0.0002
0.0004
0.0003
0.0002
0.0000
0.0000
0.^)004
0.0017
0.0003
0.0009
0.0021
0..0003
0.0005
0.0002
0.0000
0.0000
0.0008
0.0003
0.0012
0.0007
0.0005
0.0012
0.0009
0.0000
0.0003
0.0000
0.0016
0.0006
0.0000
0.0013
0.0003
0.0010
0.0003
0.0000
0.0020
0.0000
69.78
6. 10
38.61
36.58
40.64
39.62
54.87
45.72
75.19
1 18.88
127.01
49.79
49.79
424.47
481 .07
86.37
184.92
81.29
95.51
168.67
29.47
28.43
22.34
44.69
47.73
133.03
63.98
140. 14
76. 17
82.26
157.41
60.93
481.07
61 .94
1 10.69
77. 18
27
27
28
27
27
27
27
27
27
31
27
27
27
27
27
29
28
47
27
27
27
27
27
27
28
27
28
28
28
27
28
27
27
0.0035
0.0055
0.0080
0.001 1
0.0024
0.0059
0.0(327
0.0046
0.0015
0.0033
0.0000
0.0048
0.0062
0.0030
0.0038
0.0027
0.0031
0.0032
0.0146
0.0029
0.0062
0.0022
0.001 .1
0.0054
0.0013
0.0122
0.0113
0.0235
0.0164
0.0024
0.0065
0.0016
0.0012
0.0838
0.3143
0. 1 162
0.0104
0.0000
0.0093
0.0000
0.0461
0.0246
0.0050
0.01 16
0.0061
0.0183
0. 0049
0.0954
0.01 15
0.0126
0. 0400
0.0545
0,0127
0.0403
0.0926
0. 1349
0.041 1
0. 1282
0.0332
0.0384
0.0421
0. 0300
0.0152
0.0000
0. 1 172
0.0563
320
320
194
599
558
558
186
596
596
596
599
599
529
529
537
346
346
336
336
336
336
336
336
.346
336
472
596
474
155
346
471
.155
202
155
155
298
-------
APPENDIX E (CONTINUED^
Hsld.
10
Soil
I dust
Wsill
1415 1
1420
1 42 5
1515
1520
15.30
15.35
1540
1550
1560
1565
1840
1845
1850
1855
1860
187(5
1875
1390
1895
1900
1905
1915
1925
2145
2 1 60
2165
217.3
21 75
2200
2205
24 1 ,)
2440
2460
2465
2470
2480
2485
2490
2500
0.^006
0.0000
0. 0^02
'0.0004
0.0002
0.0000
0.0000
0.0005
0.0000
0.0'^ 00
0.^002
0.0043
0.0000
0.^014
0.0005
0.0008
0.0000
0.0006
0.0008
0.0000
0.0010
0.0000
0.0004
0 . 00 0 0
0.0013
0.0009
0.0007
0.0000
0.0007
0.0006
1.0002
0. 0005
0.0r*00
<). ?OH2
0. 1000
0.0000
0.0007
0. -5009
',). 00(52
8 7 . 3 3
47.73
1 17.80
1 13.82
73 ).09
8. 1 3
33.56
96.61
103.32
2 OH. 40
73.22
65.09
95.60
1 15.93
43.81
27.46
14.74
29.49
I 0 . I 7
29.49
17.29
43.31
45.77
243. 83
2 9 1.21
46.49
63.63
65.70
1 0 1.03
17 6., 38
51.54
26'\34
26.23
3'?. 5.51
271 .67
147.57
51.5 6
37.93
37.40
27
27
27
23
29
27
27
23
27
38
27
23
27
27
27
27
27
27
27
27
27
28
27
27
27
27
27
27
27
pp.
27
27
27
?7
27
30
27
23
:\i
21
0.0030
0.0010
0.0041
0.001 3
0 . 0 06 1
0.0027
0.0000
0.0089
0.009 1
0.0027
0.0037
0.0030
0.0/57 1
0.0155
0.0029
0.0000
0.0022
0.0026
0.0026
0.0024
0.0012
(1.0022
0.0285
0.001 3
0 . 0 ^ \ 2
0 . 0 1 1 8
0.0029
0.0 '3 50
0.0000
0.0038
ii.0.524
0.0056
0.0033
0.0013
0.0225
0.0J20
f 1.0 '06 3
0 . 0 02 7
0.0059
0.0143
3.0248
0.047'}
0.0574
0.0356
0.0742
0.01 56
0.2126
0.02 13
0. 10.35
0.0242
0.0082
0.4133
0. 0148
0. 044 1
0.0064
0.0070
0.01 57
0. 1375
0.0539
0.0369
0. 0.304
0.0000
0.0503
0.0165
0. 0206
0.0097
0.0087
0.0227
0.0451
0. 2273
0.0071
0. (52 19
0.07)59
0.0602
343
255
343
571
571
571
571
571
571
57 1
571
320
194
194
194
335
335
.3.35
320
320
320
32 0
335
320
336
173
173
173
471
471
471
479
259
259
259
300
300
474
474
364
299
-------
APPENDIX E (CONTINUED)
H.sld.
ID
2505
3105
31 15
3210
3220
3225
3230
3425
3435
3665
3670
3675
3695
3870
3875
3900
3905
5380
5390
5395
55/5
5590
5645
5660
5665
5670
5675
5685
5700
5725
7185
7265
7290
90 1 0
901 5
9020
9070
9075
9090
9095
Site
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
H20
0.0000
0.0035
0.0022
0.0005
0.0058
0.0002
0.-1000
0.0010
:3.0003
0.0041
0.0003
0.0002
0.0000
0.0000
0.0000
0.0000
0.0000
0.0010
0.0000
0.0000
0.0000
0.0005
0. W02
0.001 2
0.0020
0.0000
0.75000
0.0^13
0.0000
0.0002
0.0004
0.0009
0.0000
0.^006
0.0000
0.0008
0.0004
0.0000
Soil :
56.60
97.22
73.69
47.07
143.27
74.71
104. 38
1 10.52
61 .40
148.38
103.48
1 1 1 .55
1 35.08
56.23
1 90.67
26.0.5
1 1 1 .65
53.03
214. 14
28.77
95.57
55.63
31 .55
32.48
26.91
34. 33
105.57
173.24
51 . 44
39 . 3 1
37.39
230. 34
312.67
167.56
156.97
59.71
145.42
91 .49
27
27
27
27
27
27
27
27
27
27
27
27
28
27
29
29
27
28
28
28
29
27
29
27
27
27
27
29
27
27
27
28
27
27
27
27
26
29
29
Irfust
0.004 1
0.0043
0.0187
0.0 157
0.0:167
0.0066
0.0055
0.0095
0.0027
0.0030
0 . 0 1 1 1
0.0030
0.0086
.1.0.32 1
0.0092
0.001 1
0.0 59 1
0.0025
0.0016
0.0024
0.0039
0.0020
0. 0-332
0 .001 0
0.0102
0.0 126
0 . 0 08 0
0.0043
0 . 0 '52 0
0.0059
0.0068
0.0030
0.0073
0.4712
0.0041
0.0028
0.0 132
°'i .001 0
Wsill
0 . 0 1 49
0.0231
0.0312
0.0139
0. 1728
0. 3.309
0.0087
0.0102
0.0390
0.0264
0.4468
0.0506
0.01 55
0.0093
0.0867
0.0124
0. 1749
0.2145
0.2357
0.0031
0.0193
0.0126
0.0296
0.0242
1.9540
0.0303
0.2196
0.0337
0.0012
0.0102
0. 1906
0. 0159
0. 2072
0.0819
0.0261
0.0479
0.0298
Traffic
364
9538
6577
1 1765
1 1 765
1 1 76 5
1 I7-S5
8651
6654
12514
12514
12514
12514
1 1467
1 1467
1 '32W
10200
130 00
1 3000
1 3(W J
10637
1241 1
9362
9362
1 1765
12411
10637
12514
1 1765
1241 1
9098
12331
1 2 3.3 1
7154
7154
6776
7154
7154
7154
7154
300
-------
APPENDIX E (CONTINUED)
Hs 1 d.
in
9100
9 105
91 15
9125
9130
9195
9205
9210
9215
9220
9225
9230
9235
9280
9285
9290
9295
9300
9305
9310
9315
9325
9340
9345
9360
9365
9370
9380
9390
9485
9490
3080
3090
3100
3310
3335
3650
3660
4020
4025
Sites
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
H20
0.0001
0.0000
0.0000
;).0000
0.0002
0.0003
0.0000
0.0000
0.0000
0.0000
0.0000
0.0002
0.0002
0.0000
0.0000
0.0003
0.0004
0.0000
0.0000
0.0002
0.0000
0.0000
0.0000
0.0002
0.0007
0.0002
0.0004
0.0003
0.00150
0.0010
0.0004
0.0004
0.0008
0.0000
0."|004
0.0007
0.0000
0.0003
0.0004
Soil t
235.30
494. 17
124.71
58.74
1 10.74
73.78
102.56
540.84
72.57
122.83
307.49
145. 14
84. 19
200.30
196.43
163.05
57.09
83.22
64.83
109.34
106.44
49.84
144. 17
3.87
77.44
3 1 .77
iflv.21
140.98
169.78
41.70
298.84
1 37. 1 3
63.41
91 .08
38/1.35
345. V2
226.36
1 19.73
104.97
8! .86
£nRv^
27
27
27
27
27
28
27
34
28
28
28
28
28
27
27
27
27
27
27
27
29
29
21
37
27
27
27
20
27
27
28
27
27
27
27
28
27
31
28
28
Id'ist
0.0178
(\ 0.590
0.0 (Ml
v',.0123
0.0 ^72
0.001 1
0. 0403
0.0,138
0.0314
0.0012
0.0043
0.0083
0. 0:52 4
,'.0736
0.0100
0.0037
0.0054
0.0204
0.0138
0.0304
0.01531
Pi.0-:5?9
d.0598
0 . 0 02 5
0.15022
0.0135
0.0624
0.0056
0.0048
0.0038
0.0050
0.001 2
r'. 0(3 17
0.0362
3.0802
0 . 0 ' 5 0
0.0624
'.'1.1,1 553
0.0036
(1.0,352
rtstll
0.029 1
0.02R6
0.0249
0. 1 190
0.4858
0.3855
0.0350
0. 0330
0.8182
3.3286
0.3980
0.0217
0.0318
0.0125
0.0279
0.0130
0. 1 128
0.0196
0.0247
0.0529
0. 0986
0.07 16
0.0439
0.0466
0.0282
0. 1563
0.0140
0.0174
?!.02 27
0. 0523
0.0342
0.0147
0. 0509
0.0653
0.0092
Tr =\ f f t c
7154
1 154
71 -54
7154
7 1 54
8209
1 708
1708
1708
1338
1338
1338
1 1 338
11817
1 1 494
8197
13072
1 3072
8464
r;708
8464
9456
9456
9456
9456
1 1708
1 1 708
13072
9456
1 1 494
5918
16219
16219
16219
13800
1 3300
16128
16128
17931
17931
301
-------
APL>ENT>_X £ (CONlir.JU.LD)
Hsld.
ID Site
H20
Soil #Day.s Must Wsill Traffic
5080
5085
5090
5095
5100
5105
5130
5! 35
5165
5240
5365
5370
5400
5405
5440
5445
5455
5460
5465
5^70
5430
54915
5495
55*15
5510
5520
5525
5540
5545
5550
5555
5565
5585
5595
5605
5625
5635
5630
5690
5695
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
0.0000
0.0003
0.0005
0.0002
0.0004
0.0004
0.M010P!
0.0003
0.0000
0.0005
0.0000
0. 00 H0
0.0000
0.0003
0.00 13
0.0PI02
0.0003
0.0002
0.0000
0.0009
0.0009
0.0000
(/). n^0r/i
0.0004
0.0000
0.0004
0.1^007
0.000Q
0.0004
0.0005
0. Mf/HKI
l>. 0(^00
0.000"}
0.0000
'4.(V02
0.0002
0.0008
'.0009
33. 15
30. 70
1 12.59
73. 50
31 .64
9! . 18
236.95
1^7.21
137.70
7R. 16
153.52
141.43
124.68
45.39
39. 16
7,3. 32
2 5 3 . '0 1
1 87.80
58.74
1 14.82
1 30.84
70. 31
69.43
1 10.36
84.55
62.68
236.03
236.07
16.79
61.33
222.03
67. 74
2 39.61
77.95
337.87
1 16.56
I 3. 91
62. 1 7
7' 1.3 3
105.57
27
27
28
27
27
28
27
27
26
27
29
28
27
28
28
27
30
28
28
28
28
28
28
27
23
27
28
23
27
39
23
28
28
23
27
27
31
27
27
0. 0.579
0.001 5
'.,.0100
0.0030
',-1.0037
0.0027
0.0097
0.0030
0.0073
0.0055
0.0024
0 .0127
0.0150
0.0,512
0.0026
0.0046
0.0047
0.0000
0.0037
0.0104
0.001 8
0. 003!
0.0337
0.0019
0.0023
0.0.563
0.0097
0.0052
0.0032
0.0035
0.0053
0.0233
0.0038
0.0330
0.0041
0.0(532
0.0063
0.0020
0. 0974
0.0333
0.0, ')80
0. 1608
0.15097
0.0180
0. 0142
0.2062
0.2792
0. 1411
0.0065
0.0203
0.0563
0. 0385
0.01 77
0. 0336
0.0171
0.0331
0.0256
0. 1234
0.4252
0. 1405
0.0714
0.2477
0.0364
0. 1041
0. 0404
4.6857
0.0746
0.0739
0.0407
0.0105
0.3451
0.0391
0.0071
0.0471
0.0:500
13790
1 879 1
18790
13790
13790
1 3790
14000
14000
13245
13835}
15257
15257
15570
1 5570
16832
16882
15442
15442
18245
13245
17452
17452
1 8543
17000
1 5000
13300
14000
190 '50
13245
17000
1 3800
16500
1 8245
16381
17119
13354
18354
1 3354
18543
17119
302
-------
APPENDIX E (CONTINUED)
Hsld.
in
5705
5715
5730
5735
5740
5745
5755
5760
5765
5770
5775
5780
5785
5850
7075
7 I 00
7340
7385
7580
7605
7620
7630
7655
7705
7720
7745
7775
7850
0470
0475
3700
5225
5230
5360
54 1 0
5415
5485
7170
7130
7220
Site
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
H2D
0.000 0
0.0005
0.0000
0.0000
0.000?
0.0000
0.0000
','.'1000
0.ft00ft
0.0003
0.^00 0
0.0000
0 . 00 0 0
0.0002
0.0002
0.0005
0.0002
0.0000
0.0003
0. 0005
0.0<7103
0.0000
0.0000
0.0000
0.0003
0.0000
0.0028
0.0000
0.0009
0.0000
0.ftft05
0.0003
0.0000
0.0(7100
0.0000
0.0006
0.0013
0.0004
0.0025
0.0003
80 il -
65.87
93.84
160.6?
31.21
27.07
109. 18
1 !">.09
434.4 1
76.70
39.33
79.40
1 33.84
347. 53
170.54
1 1- ! . 8 9
387. 13
33.2'1
1 78.3?
492. 97
203.40
203.40
1 46 . 2 2
132. 16
1 36. 84
68 . 60
401 .34
33.27
55. 13
54. M
1 86.65
191. 3 6
1 60.96
47.45
76. 30
58.7 >r
56.07
40.05
1 17.92
95. 37
260.22
frnny.
28
33
23
28
23
23
30
27
28
27
27
29
28
23
27
23
28
28
27
27
27
27
27
29
28
27
27
27
27
35
28
27
28
23
28
27
27
PR
3 1
lH,,*t
0.01 'f:l
0.0445
11.0080
0 .016 0
0.0063
0.0123
0.0047
0.0022
0 . PHI Q
0.0199
0.0160
'1.0080
0.01^3
0.0,170
0.0017
01.0153
0.0128
0.0240
0.0036
0.00! 3
0.007 1
0.0044
0.0057
0.006 1
0.0045
0.001 7
0.0094
0.0080
'..0033
0.0157
ft. 00 3 4
ft. 006 7
0.00 17
< * ft 1 T I
' I I \
0 . '-; ft i 7
'- 1.003 5
ft .0 } I ft
0.0065
,-lL Tr,
ft 03^7
0.2357
0.01 32
0.5395
0.01 1 3
0.0483
0. 0248
O. 0469
0.0193
0.0289
0.2812
'A. 0208
0. 01 05
0.0157
0.0ft9!
0.0175
0.2517
O. 0875
0.0230
0.2223
0.0322
0.5377
0. ftf^ft ft
0.0108
0.0302
0.0043
0.0053
0.7161
0. 1 126
ft. 0 ft 42
0.5263
0.2164
'". 1 580
0.0043
0. 1258
0 . ft4 1 I
O. 7964
0.0583
-0. 0O4 1
=if ^i r
7119
1515 6
37-M
37/10
37-lfl
370,1
377M
3700
37T0
1 3700
3 7'i 0
3700
37-1 1
6536
3790
3790
6331
633 1
8790
3790
3634
3684
3684
3424
4500
3634
3790
8216
7047
3354
37350
^089J
^0362
20923
20923
20928
2.0890
20483
21 340
2.1432
303
-------
APPENDIX E (CONTINUED)
l-hld.
ID Site
H20
Sni. 1
IdMSt
Wsill Traffic
7225
727H
7275
7390
7395
74! 5
7515
7520
7525
7545
7550
7585
7610
7640
7645
7650
7660
7665
7670
7675
7680
7635
7690
7700
771 5
7725
7735
7750
7760
7765
7780
7785
7790
7795
7800
7805
7810
781 5
7820
7825
7830
7835
7840
7845
9465
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
0.0062
0.0004
0.0003
0 . 00 0 0
0.00(53
0.0000
0.0000
0.0002
0.0004
0.0009
0.0004
0.0002
0.0003
0.0009
0.0006
0.0002
0.0000
0.0009
0.0000
0.0021
0.0000
0.0004
0.0007
0.0003
0. "1(7)02
J.0002
0.0006
0.0000
0.0013
0.0945
0.001 1
0.0000
0.0000
0.0007
0.0000
0.0000
0.0007
0.0002
0.°>009
0.0008
0.0000
0.0000
0.0013
86.28
214.37
1 60. 10
127.51
175.44
78.61
71 .90
1 17.92
222.63
53.43
22 1 .67
1 57.47
100.29
620. 1 9
233.06
76.86
136.84
76.86
79.67
30.93
79.67
29.06
92.79
312.74
60.34
64.64
169.2 1
471 .77
262. 14
65.59
101 .72
52.29
1 53.06
50.89
173.01
112.18
212.03
26.62
54. 10
8?. 70
349.34
106.47
55. 1 3
56.09
39.85
27
27
27
32
25
27
32
27
28
27
32
27
27
32
27
29
27
27
31
27
23
28
27
27
27
27
27
27
29
27
27
27
28
29
28
27
27
27
27
27
27
27
28
0.0113
0.0035
(1.0133
0 . 0 1 1 7
0.0074
0.0000
0.0332
0.0073
0.0142
0.0027
0.0488
0.0045
0.0033
0.0016
0.0130
0 . 0 02 5
0 . 0 1 5 0
0.0099
^.0070
0.0030
0 . 0 0 1 7
0.0027
0.0074
0.0056
0.0J32
0.0262
0.0057
0.01 44
0.0032
0.0057
0.0023
0.004 1
0.0224
0.0036
0.0022
0.0040
I1. 01 17
0.0057
0.0093
0.0075
0.0074
0.0030
0.0038
0.0949
0.0643
0.0246
0. 0057
0.0641
0.0044
0. 1065
0.0161
0.0238
0.0147
0.21 18
0. 03 15
0.0837
0. 1 169
0.051 3
0. 9109
0.01 32
0.3727
0.0185
0.0036
0. 1710
0. 1531
0. 1206
0.0063
0.0094
0.0269
0.0358
0. 0602
0.0189
0.0180
0.0090
0.0592
0. 1 392
0.0426
0. 0601
0.0174
0.0662
0.0149
0. 0245
0.0000
0.0175
0. 0365
20432
20432
20432
23835
23885
22670
37850
37850
37850
37350
37850
20432
37850
27537
37850
20fl fl0
31542
31542
31542
19578
19578
31542
20141
20572
20483
21300
20432
21300
2 1 300
20432
31542
3 1 542
31542
20928
20432
20928
20141
31542
31542
31542
31542
31542
31 542
20432
20362
304
-------
APPENDIX F
Appendix F contains the variables handwipe lead, blood lead,
and traffic counts for each participant (children only). The units are
given below.
Site
Traffic Count
Part. ID
Handwipe Pb
Blood Pb 1
Traffic density site
Actual traffic count
Participant ID
Lead concentration in handwipe sample (|Jg/cm )
Lead concentration in blood sample 1 (^lg/100 ml)
305
-------
APPENDIX F
Traffic
te Count
320
194
599
558
553
558
186
^6
596
D96
399
1 599
1 ^29
529
346
336
336
336
336
336
346
346
336
1 336
1 472
T96
506
346
346
202
67 1
1 571
I-- -i t
5/ 1
i . , .
' 57 1
1 194
1 3 3 5
I_
320
1 ^20
1 173
I.
17 *
1 471
Part.
ID
0007
0017
0026
0052
0053
0059
006 1
0066
0072
;i077
,109 1
0:^92
,'103
0107
0132
0 1 52
0156
0 1 6 1
0 1 6 7
0163
0172
0173
0 1 82
3 M 3
0252
0257
0253
0271
02 72
402
53 1
532
561
562
352
362
392
393
2 1 72
2173
2201
H^rr^w i o°
Pb
10.82
9. 22
2.22
7. 37
3.84
8. 42
9.27
3.35
11.37
7.73
1.83
7. 1 1
6. 12
5.60
4.25
13. 35
1 . 64
7. 10
2.29
9. 79
6.54
1 . 6<3
5. 19
7.50
5. 38
1 1.53
10.71
8.45
3.36
6.57
3.34
14. 31
4.62
39.73
5. 39
17.371
3.84
1 . 92
26. 30
5. 21
3. 34
Rlor><-f
Pb 1
6.40
17.47
30. 90
36.00
17.81
1 3.70
19.60
3 1 . 3P!
10.40
15. 12
19.70
5.40
28.20
5.40
14. 10
9. 10
19.60
R. 30
1 3.90
15.10
15.93
11 .06
12.00
8.10
17.00
16.08
15.80
16. 30
7.25
1 3.29
7.50
1 1 . 93
1 1 .93
9. 36
14.86
1 4.R0
10.41
18.40
20. R^
17.60
306
-------
APPENDIX F (CONTINUED)
Site
e
1
1
1
1
2
2
3
2
2
2
2
2
4
4
4
4
4
3
3
3
3
3
3
3
3
3
3
3
2
2
3
2
2
3
3
T
3
3
3
4
4
f r =if f i c
Count
471
474
364
364
1 1765
1 1765
16128
12514
12514
1251 4
12514
1 1467
20390
20928
20028
20923
200? 3
7452
3548
7000
7000
5'. * 0 0
8245
3300
650,1
3245
638 1
7119
P3.62
1 I 765
13354
2514
1 7*5
370T
37i'0
3700
3700
3700
6536
20433
20432
D^rt .
ID
2206
2437
2501
2502
3226
3227
3662
3667
36/2
3673
3697
3876
5227
5362
54 1 2
541 3
5417
5493
5497
5507
5508
5512
5547
5557
5567
5586
5597
5606
5662
5667
5681
5686
6701
5732
o757
5771
5776
5731
5352
7171
7272
U^n,-fwi Qra
°b
5.72
12.04
6. 82
5. 54
11.01
1 2. 46
12.^0
5.57
10.57
1 1 .69
10. 14
6. 54
9.23
4. 1 3
4. 42
6.72
6.65
7. 30
3.33
10.08
1 9 - 03
17. P6
23.30
1 3.29
9. 70
19. 73
6. ?3
4.23
4. 13
8. 68
4. ?5
6. 86
2.99
4. }1
4.90
3.03
7. 40
1 8 . 6 3
7. 34
12.70
3. 45
nl^n8
Pb 1
3 1 . 00
1 M. 03
11.10
1 4 . 4 71
1 7 . 06
24.^5
12.90
1 4.40
1 7 . 06
1 2.38
1 3.48
16.^8
15.80
15.95
20.80
9.50
21 .*0
R. 20
17. f0
16. 10
24.20
1 ^.64
15.61
5.40
12.70
1 9.^0
19.70
* . 10
15.90
? 3.4f/l
10.52
1 "> . 44
13. 35
1 3.70
14. 40
50.5'?
1 7 . 00
1 ^. 70
1-4.46
17.1'*
1 1 . 60
307
-------
APPENDIX F (CONTINUED)
Site
2
4
4
4
4
3
4
4
4
4
4
4
4
.4
3
2
2
2
2
2
2
2
2
2
2
2
2
n
£.
2
2
4
3
3
2
2
2
2
2
3
i'ra f f i c.
Count
12331
21835
27537
37850
20572
14500
2 1 300
2 1300
31542
31542
20928
2.4928
20141
31542
13216
7154
7154
1 1 333
1 1333
11317
3197
3197
1 3072
1 3(372
13072
9708
9708
1 1 708
13072
9456
20362
17047
17047
1 1 494
5918
5918
59 18
5913
17452
Part.
ID
7291
7392
7642
7647
7702
7722
7752
7753
7787
7783
7797
7807
7312
7817
7351
9071
9131
9221
9237
9282
9292
9293
9297
9302
9303
9312
931 3
9372
9382
9392
9466
9471
9472
9437
9492
9493
9494
9496
5492
4qn --
-------
APPENDIX G
The variables listed in Appendix G pertain to identification
of the participant and the blood analyses. Abbreviated captions are
explained below and units are given for each variable.
Hsld. ID
Site
Traffic Count
DOW
Part. ID
Sex
Age
Blood Pb 1
Blood Pb 2
HCT 1
HCT 2
FEP 1
FEP 2
CO 1
CO2
Smok. code
Paint
- Household ID
- Traffic density site
- Cars/day
- Day of week on which the traffic count was made
- Participant ID
- Sex of the participant (1 = Male; 2 = Female)
- Age of the participant
- Blood lead measurement (|_lg lead/100 ml blood) from
blood sample 1
- Blood lead measurement (|_lg lead/100 ml blood) from
blood sample 2
- Hematocrit value from blood sample 1
- Hematocrit value from blood sample 2
- FEP value from blood sample 1 (p.g/100 ml RBC's)
- FEP value from blood sample 2 (|_lg/100 ml RBC's)
- % carbon monoxide in blood sample 1 (ml/100 ml)
- % carbon monoxide in blood sample 2 (ml/100 ml)
- Smoking Code # packs/day
0 0 (does not smoke)
1 £1/2
2 1
3 1 1/2
4 2
- Highest of the four paint values for the household (mg/cm )
309
-------
APPENDIX G
u>
\->
o
Hsld. Traffic
ID Site Count
1530
1 H90
2brtH
t)075
0105
01 55
0'125
HI50
Ml 60
0165
0255
I860
1890
22C15
0v155
>H590
24.-S5
25M0
01 70
1560
0120
2480
1425
2470
1410
1 550
1515
2160
1540
2165
571
320
364
596
529
336
599
336
3.36
336
596
335
320
471
558
599
474
.364
346
571
5.37
300
34.3
300
155
571
571
178
571
I7B
OOW
3
3
5
5
5
2
4
2
2
2
5
4
3
5
4
4
5
5
2
3
5
3
3
3
2
3
3
2
3
2
Pirt.
ID SRX Aoo
15.31
1892
2501
0077
'-1107
0156
0(126
0152
0161
0167
M258
IH62
IH93
2206
0058
0092
2487
2502
0172
1 56 ->
0122
2481
U26
2471
141 1
1551
1517
2162
1541
2167
HI
01
01
02
02
02
03
03
H.3
03
03
03
03
03
05
05
05
05
06
06
60
60
61
61
65
65
66
67
6t5
72
TlnnH
Pb 1
7.50
11.10
15. 12
5.40
19.60
30.90
9. 10
8 . 30
1 3 . 90
15.80
14.80
10.41
31 .'10
I7.8I
5.40
I 0.08
I 4. 40
I 5. 98
9.36
12.95
24.52
12.24
I 3. 88
I2.5I
14.07
1 4. 64
I I. 22
18.73
15.73
Tloo-1
Pb 2
1 4 . 4 'A
9.50
10.60
22.9£1
22.50
8.80
12.90
7.30
1 .ftPI
4.30
M.20
1.12
3.50
7.12
29. tin
10.39
7.9M
IH. 17
8.76
12.94
1 1.25
12.42
14.47
15.00
16.00
12.17
10.13
16.47
16.75
HCT 1
36
35
38
42
38
39
37
45
41
46
44
48
43
39
43
45
37
MCI ?
37
36
3i1
32
33
35
35
35
36
33
4.1
38
39
37
37
37
39
37
41
33
44
41
45
41
.39
1/1
48
4:-)
FEP 1
48
80
IM
7.3
70
98
107
My
43
2 '/i
87
21
56
53
146
83
FRP 2 CO 1 CD 2
121 .1
177 .1 .1
47 .1 .1
ML .2
R9
I cm .1 .3
114 .3 .8
44
67 .1 .2
81 .2
51 .8
71
82 .2 .2
47 .1 .2
43 .1 .3
Co.-!*? 9 -Tint
.2
2. 5
. -)
1. i
. .)
I. 6
I . I
I . 5
2.7
. )
I'!. I
.2
'1 2.5
6. >
(1 . ;1
. .1
'1 . .1
. .)
(1 I.I
.9
:1 1 . 4
1 . 1
.2
v'i . ,i
..; 1 . 5
1 .5
3.2
-------
APPENDIX G (CONTINUED)
Hslr).
ID Sltf
M030
2465
0140
24/4-)
0.)5k4
0 1 in.'
00yii
0 I 30
0 I 80
!H155
0250
I41HJ
0270
I ('. 50
2 I /W
MM 5
IK1I5
0"160
0065
Ml 65
Ml 7U
0 1 r)0
0255
I5i'5
22lH1
(1070
2170
1 560
0 1 55
1 520
Traffic
Count
599
259
336
259
558
529
599
.346
336
5 58
472
2M2
346
194
178
320
194
186
596
336
346
3.36
596
571
471
596
178
571
3.36
571
now
4
2
2
2
4
5
4
2
2
4
3
3
2
4
2
3
4
4
5
2
2
2
5
3
5
5
2
3
2
3
Part.
in 5
0081
246/
0142
2441
0052
i) 1 (1 3
!)09 1
.1 1 .32
/)I82
0059
0252
1402
0271
1852
217?
0110 /
0.1 1 7
0:-16 1
0r;>66
0168
0173
018.3
0257
15.32
22/11
007 ">
217.3
1561
0157
1521
i
. \)
. ..t
.5
. >
1. 1
2. /
1.1. 1
0 .2
*. )
1 '> . I)
. .)
. y
1 . 6
i 1 . '-)
-------
APPENDIX G (CONTINUED)
OJ
H-1
hJ
llsld.
ID Slt(
0265
I85H
00 1 5
0250
0 1 50
1400
1 530
1890
02 7 H
005;')
0 1 <10
0255
I860
0I60
2205
0025
0I30
H I 70
2 I 70
0005
0055
(II 65
I875
2485
0I0H
0I05
0I45
0060
22-40
I 845
Traffic
? Count f
1 55
I94
I94
472
336
202
57!
320
.346
558
.3.36
596
335
336
47!
599
346
.346
I 78
320
558
336
335
474
529
529
336
I 86
47!
I 94
~\f)i.'l
2
4
4
3
2
.3
3
3
2
4
2
5
4
2
5
4
2
2
2
3
4
2
4
5
5
5
2
4
5
4
Part.
in 5
0267
1 85 1
00 I 6
025I
0 I 5 1
I 40 I
1 53 3
I 89 1
0272
005 I
0I8I
0256
I 36 1
0I62
2207
0027
0I3I
0I7I
2I7I
0006
0056
0I66
I 376
2486
0I0I
0I06
HI 46
0062
2202
I 846
5ex
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2-
2'
2
2
2
2
2
2
2
2
2
a.ne
23
23
25
25
26
26
26
26
27
28
28
28
23
29
29
30
30
.30
39
3
3
3
3
3
33
33
33
34
34
35
Blood
Pb I
I 7 . 00
5. 18
I 3.55
1 0.59
9.28
5. 18
9. in
5. 18
7.25
10. I 8
1 0.4 I
14.25
5.7I
8.06
10.87
3.75
I 1. 69
8.09
I6.0I
6.94
6.48
I5.9I
I 2. 00
I I . 1 0
9.94
6.22
6.0 1
II. 10
3.44
8 1 oor|
Pn 2 h
15.69
4.59
10.36
12.06
8.12
6.94
9.29
5.00
16.99
10.0-1
10.1.3
12.57
7.26
7.54
8.99
1 1 . 38
8. II
10.34
6.24
1 0 . 85
7.75
5.25
13.73
10.64
12.95
9.90
5.83
5. II
10.53
or i H
35
35
42
34
41
39
46
40
34
43
42
35
43
44
44
4M
37
43
41
45
44
43
47
47
41
4.3
38
40
46
41
CT 2 t
41
33
11
38
.39
37
38
4 t
.37
38
42
43
43
43
34
42
38
43
41
37
.39
* 42
3")
45
41
41
39
33
35
4'-l
~P.n 1
199
103
Q'J
97
62
107
55
1 19
128
4
127
79
151
58
19
89
51
49
1 19
3;i
85
118
1 f*3
47
IH
96
4:5
55
2'W
S:n
FRP 2 00 1 CO 2 C
'fl 76
68
105
68
94
131 . i!
l?8 I.I
95
96
100
87 .9
48
85
72 .1
66
66
70
64
IP5 . 1
61
124
101 .2
197
75 .2
118
67
58
128
93 .0
ok.
o.i « "alnt
.2
.5
.9
. ')
1.6
Ci 1 . /
3 .2
2 . '.)
2. 1
1.4
1 2.7
UK 1
.2
0 2.7
6.5
I.I
.7
1. 1
;'( . *)
4.5
. ;:)
0 . 'J
.2
0 .,)
. 5
. .)
1. 1
. H
0 3.0
1.4
-------
APPENDIX G (CONTINUED)
U)
I'
U)
Hslcl.
IU Sttp
1915
2 5' -10
iK)70
2I7'5
(-I26H
1 56 5'
IH95
1195
H 0 1 ()
1515
249')
IH7D
0115
1405
1425
2505
0140
142(1
2460
1415
1905
24,-Hl
0120
1410
190(1
1925
2410
IH4rJ
2 1 6(1
IH55
Traffic
Count
115
164
596
471
474
571
120
155
.120
571
474
1.15
146
155
141
164
1.16
255
259
14.1
120
100
517
155
.120
32H
479
120
178
194
DOW
4
5
5
5
5
3
3
2
3
3
5
4
2
2
3
5
2
.1
2
.1
3
3
5
2
3
3
5
3
2
4
P.irt.
10
!9I6
2501
Ml \
2176
0261
1566
IH96
1 196
HOI 1
15.16
2491
1871
01 16
1406
1427
2 506
0141
1421
2461
1416
1906
24b2
0121
1412
1901
1927
241 1
IH4I
2161
1 856
Sex
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
Aqe
35
36
37
40
42
42
42
43
44
44
44
47
49
49
49
50
52
53
53
54
55
55
56
56
56
brt
59
61
66
67
Ph 1
9. 10
5.12
n.56
10. IH
7.20
H.06
5 . '16
11.96
7.54
11.53
6 . fl 1
1 1.72
9.H9
6. 12 '
6.75
10. in
16.69
9. R9
14.57
1 1.98
9. 11
15.96
17.1 1
6.4H
9.84
9.R9
16. 19
I'.j.'i?
rt.79
1.1.41
llood
Ph 2
7.71
4.50
9.71
1 1 .64
7.90
0.50
6.42
11.57
7 . 5'^)
10.90
5.5M
I2.6H
H.44
6.27
10.13
I(}.0'1
1 .1 . 6!1
9.H4
13.^9
1 W . 99
7.05
17.75
17.00
7.96
10.25
18.69
1 0 . 99
24.52
9.66
HCT 1
44
17
41
45
14
4f1
43
42
41
42
39
47
42
40
41
43
43
42
40
44
4«i
17
41
44
/16
1'i
41
42
42
'iOT 2
19
3 a
43
17
16
H
19
1 1
11
1H
17
45
41
4U
42
4.1
4?
41
.11
45
1'1
11
3 j
'1 1
41
17
4?
1.1
16
4 1
FEP I
I 14
54
18
20
5d
50
61
78
62
77
65
77
74
1 f6
129
59
97
47
rbi-t
li-M
1 19
54
in
0
H2
7d
24
r)H
IB
I6>J
F.P2
71
73
79
89
57
121
97
1 12
52
34
lnl
96
92
95
73
IMH
50
75
06
54
05
4:1
UK)
R9
1(53
1(11
HI
123
99
CO 1 CO 2
.1
.0
.2
. 1
. 1
1.2
.7
.2
. 1
. 1
1 . .1
.9
'^TOk .
Co^ Pnlnt
.,
(\ )
12! j
(1 4. 4
1. 1
. '-1
1. ti
?.7
1.4
D 2.1
.1 . .)
1.4
. /
2. 1
0 .2
2 ..!
.9
1.6
1 . :i
. .)
:i . 2
;) . .)
1.1
0 I . '-j
?. .'V
I. I
1 . v>
. /
. /
I.I
-------
APPENDIX G (CONTINUED)
OJ
rlsl-1.
If) S
2165
1515
2440
2145
1925
5660
92»>(>
9.:) 7;)
9290
3665
36 7 M
3875
93.10
3225
36/0
9291)
9295
93 1 1)
5685
7290
9485
3695
3220
3425
9230
9345
3I«5
9325
5380
"
ite
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
traffic.
Count r
178
571
259
259
116
320
9362
lid 17
7154
H 197
12514
12514
1 1467
13072
1 1 765
12514
8197
1 3472
9708
12514
12331
1 1494
12514
1 1765
8651
1 1 338
9456
9588
9456
1 3'MW
10W
2
3
2
2
3
3
2
2
3
5
3
3
3
2
3
2
3
4
4
3
3
3
Part.
in se>
2166 ;
isi^ ;
2466 ;
2442 ;
2U6 ;
1926 ;
5662
9282
9M7I
9292
3667
3671
3376
9102
3227
3672
9293
9297
9312
5696
7291
9487
3697
3221
3427
9231
9346
3106
9127
5381
A cj f*
> 6/
1 ^O
? 6V
> 77
> 79
> 84
HI
(11
02
cv?
;*3
(13
(11
i)3
04
04
H4
04
04
05
(15
05
06
51
57
59
59
62
63
68
ninor)
Ph 1
9.012
7. 01
9.H2
7.63
9.94
9.83
15.90
19.60
10.61
14.40
14.40
12.88
16.88
13.113
24.85
1 7 . 06
13. 18
35.84
13.1)3
12.44
23.40
18. 15
13.48
14.64
12.50
1(1.50
8.58
17.36.
1 3. 84
7. (15
Ph 2 \
9 .rid
7.20
9.75
9.82
8. 3D
7.48
13. SCI
14.60
13.57
16. W
14.50
13.2:4
15.61
14.14
27.21
17. Ml
16.30
19.20
15. (-10
16.65
16. hi
15.98
13.57
16.43
15.28
7.58
14.14
14.04
13.37
7.12
or i i
41
4;>
3f
45
43
35
34
37
33
38
37
38
41
39
33
39
38
38
40
33
36
41
45
43
44
4(1
48
46
42
rrr ">. H
41
42
44
15
i;
43
35
35
17
14
39
37
4-)
33
38
33
37
35
37
36
33
38
45
43
44
42
46
45
43
RP 1 F
75
6H
52
3d
81
77
14
08
67
53
28
95
B4
86
43
44
102
1 02
85
89
7;>
1 19
99
59
63
"VI
CP 2 CO 1 00 2 0
67
116 .2
124 .1
97
1 31 . 2
13
70
81
51
1 12
38 .1 .1
118 .1
29
112
85
72 . 1
6'.i
73 . 1
95 . 1
95 .7
26 .1 .>
3K.
oi'l" I'qint
0 3.2
'1 . ''
o i! i
. .)
3.3
1 ..-i
1.4
.2
1.4
?.. 1
. )
. J
1.4
1.4
2.6
.1 R.I
0 I.I
.2
4.7
.2
i< ,2
1.5
il 2.3
ci .;1
1 1 . '>
," a ^J
-------
APPENDIX G (CONTINUED)
LJ
M
Ul
HsU. Traffic
in 31 1
43
44
44
33
.37
38
39
35
37
39
38
38
.38
35
41
3*!
33
39
38
37
46
37
Snoi: .
ICT 2 F:-:'> I FRP 2 CO I CO 2 Co
43
41
37
34
37
38
3-S
37
.31
36
41
37
41
33
3'J
35
34
4^S
33
40
3
87
93
45
7;-)
25
|i'"9
76
""7
37
1 10
42
48
1 14
57
7(1
57
89
35
102
81 1 . 3
56
4fi
5M .«3 .6
8.3 .2 .1
I2PI . 1
1 18
37 ' 1.7
1 '1 1 .3 . 2
72 . 1
63
ri
. 1
96
58 . 1
96
88
91 .5
95
17
3
«
2.
2 2.
il
^
m
;,1
1
'1 1.
B
o^
1.
3.
1 1.
2.
3.
">
5.
6.
2.
.1
.
f
>_
.
i.
2 1.
.
> ^
5
0
1
5
i,
T
-',
7
.J
6
/
1
6
ij
3
1
( >
2
7
9
1
t)
n
;>
1
i,
'<
i
/
1
-------
APPENDIX G (CONTINUED)
U)
in <
5/K0
3 6ft 5
39M0
94, J5
5675
9295
9100
566 b
92 9i !
.1225
91.170
9I3H
93911
91ft5
y020
3671!
56 H 5
9115
3695
9.190
9170
9205
50 70
9I0C-)
92IK
9.3 1 0
5590
9235
39,15
5575
Jtte
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
Traffic
Co'int
1 1 765
12514
IU200
1 1494
10637
13072
1.3072
1 1 765
til 97
1 1765
7154
7154
13072
1 1 70H
6776
12514
12514
7 1 54
12514
9456
1 1 708
1 1 70H
1241 1
7154
1 1 708
97(18
1241 1
11338
1 M^00
10637
now-
3
2
2
1
3
3
3
3
3
3
2"
3
3
3
2
2
6
3
2
2
6
4
2
Part.
in
57M2
3666
3901
94H.1
5676
9296
9301
5*66
9291
3224
9:172
9132
938 1
9166
9C12 I
16 7 1
5687
9116
3696
9391
9171
9206
5671
9101
921 1
931 1
5591
92.16
39H6
5576
Sex
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
Ane
24
26
26
26
27
27
27
28
2H
29
29
29
29
.10
31
32
32
32
31
.11
14
37
.19
3M
3 a
3H
39
39
40
41
TlorH
Pi) 1
S.90
12.29
4.66
12.99
7.97
10.61
1 H.50
1 1 . 30
11.14
11.15
21.62
7.40
6.12
9.01
9.32
1 :i . 6 1
(1.71
19.29
15. IB
14.64
17.7.1
1 1 . 99
14.92
12.76
'). 11
1 1.99
>i.67
ninorl
Ph 2
9 . 89
9.Wi
4.78
IP). 15
7.96
8.12
15.^6
5.15
8.4 |
12.42
I3.H6
11.27
1 7 . 58
7.54
6.98
7.54
12.66
12.71
7.26
23.9.3
11.27
13.57
10. 17
9.42
14.72
1 3.70
10.14
1 1 . 04
9.i*>6
HCT 1
.14
39
34
37
43
4. ''I
41
45
37
48
41
5"
42
38
40
41
39
46
44
44
3H
41
41
47
41
42
36
42
MCT 2
14
4'-)
1 \
35
42
41
4'1
41
46
17
41
41
42
4-1
16
17
42
41
41
43
42
44
15
45
19
42
41
33
41
41
FK? I
1 12
122
1 19
36
92
6?»
9.1
61
59
6K
80
44
65
72
0
/14
80
91
67
62
62
I-.1.3
71
89
46
77
,1 |
107
FFtP 2 CO 1 CO 2 Co
105 .3
74 .2
47
A4
1 '17
75
°?
68
1 17
90
193
1.32
49
106
102
68
RB
112
89
54
H3
84 . 1
46
71
97 .1
43
72
69
.2
1.4
1. 1
1. 1
4.7
3. j
6.9
i. :j
,1 . ,)
2.'1>
1. 1
1 8. 3
. .)
5. 7
.5
1.9
-------
APPENDIX G (CONTINUED)
ilslr).
if) s
566 0
5/25
3210
90 1 5
9195
94b5
934H
3425
9235
5645
9.195
9215
9090
9345
9O 1 0
53PO
7IH5
31 4)5
5380
36 /^
9230
9360
9225
3230
9 3 05
9315
9125
5/75
5555
5515
1
tte
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
raffle
Count r
9362
1241 1
11765
7154
82M9
1 1494
9456
0651
1 1494
9362
7154
1 1 /0H
7154
9456
7 1 54
1 30H0
909H
96HH
1 3,)0tf
12514
1 1 13H
9456
1 1 330
1 1 /65
SJ464
9464
/I54
13700
1 3800
IH245
)OW
2
6
3
3
2
3
4
2
2
3
2
3
3
3
3
4
3
4
3
2
2
3
3
3
5
Part.
in <
5661
5726
3211
9(1 1 6
9196
94H6
934 1
3426
9286
5646
9096
9216
9091
9347
9012
5391
7 Id 6
3107
53
37
3 J
4 I
43
3 1
3/
33
3/
F«=P 1 1
46
97
93
71
164
77
39
42
63
og
Pi 3
55
1 13
53
75
58
H2
f»6
')
1. 1
.-1
. )
0 2.3
4 . j
1 . . i
0 3. )
; ) . . '>
;) .0
2. 1
3.5
. 5
6.2
'i . 2
1 .2
1 1 . 6
. /
. ->
2.4
n I.I
-------
APPENDIX G (CONTINUED)
00
ID Sttf?
5490
56)0,'l
H245
.3684
3 )OO
7452
16219
18790
1 5570
I68B2
17931
I6H82
1 5 1 56
IB 790
18245
1 36.S4
16128
18790
18790
18790
15570
16219
16381
Iti245
1.3700
16381
17*147
17452
6
3
3
5
4
3
6
3
5
4
4
4
4
5
5
4
6
5
5
5
4
.3
4
5
3
4
6
Port.
ID SOY A<7<>
5492
5681
5781
5242
5166
76.31
331 ?
5481
3 1 0 1
50,-i 1
540 1
5441
4025
5446-
5717
5 1 06
5466
76? .J
3651
5096
5102
7076
5406
308 1
7387
f)5
06
05
53
54
56
59
59
60
63
64
65
66
67
67
6H
68
6H
70
70
77
77
78
8:)
81
5586 2 01
5771 2 01
5597 2 02
9471 2 iI2
5493 2 H3
D 1 onH
Ph 1
14.60
I-/I.52
13.70
1 0. (19
12. 38
8.67
9 . 0 1
9. 1 3
I5.H7
12.93
12. 32
5.43
15.57
7.R4
10. 50
1 1.27
1 1.45
17. -38
9.90
1 1 . 50
1 1.50
12.68
10.45
1 1.99
10.74
19.60
39.50
19.7:)
19.27
3. 2W
PI nod
Ph 2 IICT 1 '
16.10
H.29
14.91
1(1.39
12.43
9.9I
I0.6I
1. 3. 55
15.86
13.79
12.57
3.34
12.64
8.42
15.57
13.61
13.73
17.44
?I.C-t2
13.31
12.58
15.25
18.06
9.27
8.42
17.10
17.70
1 1 . 60
15.69
12.8:1
.4(1
.34
37
47
43
43
41
4>>
47
5"!
45
44
4d
4M
45
44
4.3
47
48
48
43
41
35
45
40
36
35
.34
37
39
\Cf 2 1
4'1
33
47
4?
44
41
49
4.)
\?
43
45
4 1
42
12
46
43
45
41
47
-1,1
.31
33
37
39
rink.
:9J I FRP 2 CO 1 CO 2 Cn.(i i'-ilnt
5-
62
5.3
44
53
ny
1 °r>
69
6(-'
.6"
;y
6/
46
67
65
M4
75
126
0?
46
34
1 :!6
66
1 17
91
r;
73
69
73
1 3
56
0
85
7.3
85
110
0
'.t
107
65
28
17
.1 .1
1.6 1.9
.1 .1
.2 . 3
. !
. 1
.2
.2 .2
.1 .1
I.I \."t
.1 .1
.1 .1
.1 .1
.2
.2
.2
.2
.1 .1
.2
. 1
.2
3.ii
0 1 . )
2.3
2 .7
0 .'
n 12. 5
1 J. 5
'" 12.5
" . 2
.2
4.6
0 I.I
1. 1
r< I.I
:! . 7
0 .2
I.I
n 4 . 4
0 2.5
C* ; J
0 . /)
. ,1
1 . 6
1 3. 1
0 I.I
2.3
. )
. 2
(I . .)
3. 0
-------
APPENDIX G (CONTINUED)
Ms Id.
5505
56.15
5/55
3660
7/20
/'J50
V470
5505
55 1 1)
5565
5/30
5(150
54; 5
5 570
5 7/0
5-)4l)
7100
55 1 J
5525
5550
5565
5/60
366-1
54/0
5595
/j IW
5 4 -t 5
5->5"j
5/05
5 755
1
)tf>
.,
3
}
3
3
3
3
3
3
3
3
3
3
-)
1
3
3
3
3
3
3
3
3
3
3
3
3
3
)
3
nff Ic
17, tOU
1 /I 19
13700
16128
14500
1 H2 1 6
17047
1 /i'JM;1
I5.WO
I6511H
1 3700
I65H6
18548
15257
13/011
I 9000
I o/yn
I'.H.W
I40H0
1700,1
I 6500
13/00
16128
U5245
16381
1 879H
15442
I3-W0
1 /I 19
13700
or.'
3
3
6
3
2
3
5
3
3
"3
2
5
3
6
5
4
5
3
3
P^rt.
in r
^V'jOT
S -^(^^
575 /
3662
7722
735 1
94/2
5508
551 2
556 /
5752
535?
549 /
53/1
57/2
5541
7101
551 1
552 S
5551
5566
5761
3661
5471
559 -S
758 1
5^56
5556
57.16
5756
«Y
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
81 nod
AQ P '""h 1
03 1 6 . I H
03 6. 10
03
',14
C'1
(51
04 ,
05 ,=
i'5
t'5
05
05
C'6
19
22
23
23
4.40
2.90
7.4,1
1.91
7. ovi
4.20
3.64
2.70
3.7!i
4.46
/ . I 0
I. 74
I. 27
0.52
0.56
26 14.43
26 10.80
26 6.24
?.(> '3.67
26 11.66
27 14.68
2H 12.44
28 11.53
28 7.02
29 H.H.3
30 14.77
30 l;).4.-5
31 /.05
Blood
Pli 2 MCT I HOT 2
33.91-1
12^00
1 1 .9'-1
1O.80
I 3. 00
I5.?0
13.36
I 4 . I C1
9.9'5
15.4)1
14.90
13.11
9 . 00
1 1 .H5
10.75
IH.9-1
12.09
11.79
9 . 66
6 . f 19
3.37
14.27
7.07
13.1 3
1 5 . 69
7.72
9.13
7.12
9.6;)
H . 1 6
3ti
3.3
3 H
4'"!
36
37
36
37
36
34
35
33
45
4»
37
4?
41
41
36
30
4!
37
43
39
44
40
39
40
35
41
] )
f5
37
.j )
41
4 1
33
57
36
3 J
31
) j
43
4 5
36
42
44
41
37
4 I
33
42
33
4 5
"i 7
3 i
4.1
'.3
-FP 1
46
72
"" 5
53 '
66
I ia
45
37
97
34
18
104
9.3
55
39
55
')')
I 3
.66
59
^ no!:.
FTP 2 cn I CO :» !>rin :^Tlr>t
I24
16
3O
II5
92
26 .9
83
67 .2
25
64
50
73
0
87 .1 .1
62
42
105
122
55
68 . ?
48
3. 2
3.,)
1..-.
.2. 4
. /
1.3
. )
3. ;
. j
1 '' . 5
1. 1
. .)
. 2
6. 3
.? . .1
. /
1
. .1
. 2
5. 5
l?.5
12. ->
0 2 . 4
. /
. ^
.?. '
. 5
.). ;
1 1.6
1.6
-------
APPENDIX G (CONTINUED)
Hsld.
IU SI
5460
4020
5490
5545
57/5
3335
5535
56 HO
56 VO
5/45
55:0 5
94 70
9475
57t;0
5135
5495
7775
5520
5740
56 ^5
57 30
56 3 5
5765
7705
5365
5(350
57ii5
3090
7745
5735
te
3
3
3
3
3
3
3
.3
3
3
3
3
3
3
3
.3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
Traffic
Co'int
15442
17931
17452
IB245
I3700
1 3300
18245
IB 354
1 B54B
13700
1 7000
17047
IH354
I.37M0
I4'J00
I854B
K5790
1 3 i:-)0
1 3 700
1 7 1 1 9
13700
IB 354
13700
IB424
15257
16586
1.3700
16219
I36H4
13700
DOW
4
6
5
3
3
5
5
3
3
3
5
5
5
2
.3
3
3
4
3
3
3
/I
3
Part.
I!)
5461
402 1
5491
5546
5777
3336
55H7
5682
5691
5746
5506
9473
9476
57B2
5136
5496
7776
552 1
574 1
5696
57.31
5636
5766
7706
5366
5B5 1
5786
.3091
7746
5736
SRX
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
Aqp
32
33
33
33
33
.34
34
34
34
35
36
36
36
37
38
38
38
39
39
40
4rt
41
41
42
44
44
45
47
47
49
ninod
Pb 1
1 1 . 50
10.50
10.98
1 1.91
9.27
7 . 28
B.67
7.74
7.7.3
1 0. T)
9.50
7.24
1.27
5.74
'1.52
1.27
9.66
9.95
0. 29
7.58
9.37
11.27
7.97
7.02
10. 14
16.37
25.41
5.84
6.53
"1 rind
Pb 2
10.64
8.08
9.6.3
6.76
9.18
4.51
7.36
9.13
8.95
R.I 7
9.46
15. 4.6
12.57
13.1 "I
9.82
1 1.91
16.99
14.40
8.37
9.93
10.39
11.61
5.53
7.72
10.54
12.67
14.77
6.02
6.86
MCI' 1
43
38
44
40
40
48
41
39
36
.37
43
42
42
43
41
42
3 P.
43
3M
41
41
42
3;>
41
39
41
46
45
37
'iCT 2
41
37
41
35
43
'50
36
4 1
37
4.)
43
4'-l
43
45
46
41
H
44
.3 )
3>
43
41
45
4?
42
43
37
29
FKP 1
53
80
51
4.3
22
29
31
7,i
|0n
~l'\
79
71
62
92
54
I'M
1 IB
71
8 1
19;
41
95
73
62
45
129
144
79
0
FF.P 2 CO 1
69
1 02
87
0
67 1.6
*7
47 . 1
III .1
99
69
28
36
|4 1.2
45
45
185
25
16
48 . 1
26
73 . 1
74
40
121
50
0
0
45
71
S-no-: .
CO 2 Code i-'iint
1
. 5
.7 1 3 . )
1. 1
.2
1.4 2 . H
2. 3
n I . ;>,
) 1 . 6
. .-'
1.2
.;-) 'i ..i
. '
.2 2. 3
.6 2 2.5
.9 1 .2
. 0
1.4
.'I
:! 2.3
1.4
0 . 2
. v)
. -J
I
. 0
.0
1.4 4 1.4
4.5
. i)
-------
APPENDIX G (CONTINUED)
oo
to
HsH. '1
n sitfi
5 1 30
5240
7720
7340
7.385
4025
54H0
7655
7620
5.)90
5105
5715
3,)h.-)
509 '3
5445
7605
5465
5085
5405
5 1 00
7075
54IO
78 15
/ J90
/ M0
-i 160
5.11-3
7/50
7/85
7640
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
r^f'fr
Count HOW
1 4000
1 3300
14500
16381
16381
7931
7452
3684
3684
8/90
8/90
15156
16219
18/90
168/12
18790
18245
18790
15570
18790
18790
20928
31542
23885
20572
20928
20928
2 1 300
31542
27537
5
3
4
4
4
6
4
4
5
5
3
5
4
5
5
5
4
5
5
4
3
4
4
4
/i
5
3
Pnrt.
5131
5241
7721
7341
7386
4.127
5482
7656
7621
509 1
5li)7
5/16
3082
509 7
5447
7606
5467
5086
5407
5101
7077
5112
7817
7392
7702
5362
5417
7753
7788
7642
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
Aae
5?
52
53
54
58
60
6/)
60
62
63
63
63
64
64
64
64
65
66
66
71
75
01
01
02
02
0.3
04
04
(.14
05
Pb I
7.97
7.97
5.2I
7.32
7.58
lb.28
12.38
10.32
6. 31
8.37
8.90
6. 3/
9.01
7.49
7.84
14. 5/
19.79
8.90
7.61
1 1.98
8.20
20. 80
9.50
1 7.32
29. 10
15.95
2 1. 60
15.16
7.00
14.60
Rl OT-)
Pb 2 HOT 1 MOT ?
7.48
4.52
7.38
7.12
10.58
11.36
10.83
5.5.3
7.8.1
10.50
2.21
6.68
7.96
9.46
15.5.4
11.94
8.45
11.27
14.28
8.93
17.40
15.10*
17.20
21.20
19.90
21 .60
1.3. "6
11.20
17.50
39
45
35
38
41
43
4.3.
40
44
36
46
43
49
4.3
38
41
38
47
37
41
39
33
34
35
39
37
37
37
37
37
41
36
41
42
41
42
39
45
35
35
45
4 1
38
45
36
46
39
36
33
36
.37
38
1 )
36
3 i
37
FF'> 1 FRP 2 CD 1 cn 2 O-HO ->qlnt
75
53
43
70
92
81
40
43
78
I 05
64
4H
I 79
68
I 00
59
87
62
4R
I22
I 15
59
39
4
132
31
3 .2
26 . (.1
118
107
5tt .7
75
65 .2
0 .1
102
55 . 1
43
81
100
67 .1 .1
98
24
Pi
0
. 1
. 1
20 .1 .1
.,
!/
. 7
1 1.6
0 I.I
1. 1
12.5
2 .>.'
1.1
J . /)
0 .2
. /
0 1. 1
. /'
1. 1
. /
0 I.I
.7
1.6
. "i
. )
. 2
.5
0 1.4
3. 1
'i . :>
. i
0 1.6
. .1
T. 2
-------
APPENDIX G (CONTINUED)
u>
N)
HslH.
ID Site
7795
7,510
3/tW
7545
7415
765(1
7520
7 IU/)
7395
7550
7275
7610
7525
7750
7645
77H5
7>i05
5225
54IH
727M
9165
7 1 70
775.1
7H25
7/CH1
7cH)5
7640
54 1 ,)
5.36U
72/0
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
frafflc
Count DOW
20928
20 I 4 I
37.-)5iJ
3/J50
22670
2vM3H
37^50
2 I 37)0
2.3805
.17;) 50
20432
3/H5H '
37.350
2 1 .300
37S5M
3 1 542
2092R
20-J90
20928
20432
20.362
2048.3
21 30(1
3 1 542
20572
20928
27537
2092B
2092H
20432
4
5
5
5
5
4
5
3
5
5
5
5
3
4
5
4
.3
5
5
5
.3
4
4
4
4
.3
Part.
10 SRX Aq°
7/97
7812
37v) 1
7546
7417
7652
7522
7IH2
7397
7551
7276
761 1
115
"5
65
65
66
67
69
7(1
72
74
75
77
7527 1 93
7752 2 H 1
7647 2 02
77H7 2 H2
7807 2 C"2
5227 2 03
541.3 2 H3
7272 2 0.3
9466 2 05
7171 2 06
775 1 2 1 9
7026 2 19
77.11 2 24
7SM6 2 24
7641 2 25
5411 2 26
5.361 2 28
7271 2 2'.'
FUnrH
Ph I
I I . 8^
IP). M
I I. 79
I 7. I 7
8.6.3
I3.9C1
8.67
1 2. 38
1(1.21
11.74
12.45
14. 1 I
16.22
I4.9k1
1 1.40
.3i1.9',1
15. R«
9.50
11.60
1 1.27
17. 1:1
9.H3
10.52
9.13
4.95
10.90
4.95
1 1.5.3
9.62
Ph 2 HCT 1 MCT 2
9.21
12.4.3
12.88
13.55
8 . 6fl
1 .1 . 1 3
7.23
2.3.67
9.98
1 5 . Pi 1
1 1.66
15.74
16.97
18.30
10.50
5.56
11.50
1 7 . 70
1 1 . (12
2.3.04
1 1 . 32
3.63
i-i.93
11.66
5.56
12.06
10. 14
.36
39
4*
46
4:.?
5.3
4R
42
32
41
47
42
39
.34
.37
.34
.34
.37
4C1
34
.37
.38
43
.39
37
.34
37
39
43
4;i
37
39
4<)
46
4>1
51
47
4 I
33
11
44
43
4.1
.35
.3-1
35
35
4.)
38
^
.36
33
41
4 )
.37
.35
31)
39
42
41
-I?1 1 1
132
52
38
61
9.3
29
1 «6
59
84
82
\\'-2
1.31
04
5.3
ia
61
4.3
47
83
62
7.3
RP 2 CO 1 CO 2 Co-iR ^nlnt
47
62 . 1
45 .7
73 . 1
r>9 .1 .2
22 .1 .2
04 . .3
42 .2 .1
173 .2
54 . 1
42 ..3 .4
65 . 1
1(17 .1
1 8
31
5.9
61 .3
122 .7
73 1.3
41 .1
, F> >i ^
''"is . i
2 . 5
n . ?
i: 1.4
fi 2. 1
" 2.7
(i . ii
« . '1
0 . 5
2 1.4
I) 8.7
1 .2
:) .)
:) .5
1.6
.V
. .)
..I
1.6
.2
. vl
. !)
1. i
1.6
1 3.9
2- l.t>
. )
2 .3.2
il . 2
2 .9
i1 . -1
-------
APPENDIX G (CONTINUED)
U)
to
Hsld. Traffic
in Sit*; Count OOW
7645
7/95
5415
5485
7715
7785
7810
7815
5225
/5<)5
7(i30
/5I5
7685
//35
9465
7665
767!)
Y7«0
76/0
7335
/390
/ 760
7800
/66v)
76iH1
/69H
7790
784;)
7/25
//65
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
37850
20928
20928
20890
2H483
31542
20141
31542
20890
20432
.31542
37850
31542
20432
20362
31542
195/8
31542
31542
31542
23885
2 1 300
20432
31542
19578
21)141
3 1 542
31542
? 1 300
20432
5
4
4
5
5
3
.3
5
3
.3
5
3
3
5
3
3
3
3
4
5
3
3
3
3
5
3
Part.
ID Sex Aoe
7646
7796
5416
5486
7716
7786
7811
7816
5226
7586
7331
7516
7686
7736
9467
7666
7676
7781
7671
7836
7391
7761
7801
7661
7681
7691
779 1
784 1
7726
7766
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
29
29
30
3
3
3
.3
3.
3
3
33
34
34
34
.34
35
37
37
39
4'1
4l
42
42
46
46
48
48
48
49
49
llood
Ph I
7.26
8.55
9.42
6.59
7.74
9.M
\ 1.45
12. 15
6.27
1 8 . .39
10.32
7.07
7.97
10.85
5. 19
1 1.0.3
11.91
6.35
1 1.45
15.22
9.13
5. 13
10.29
9.84
8.44
10.98
8.67
8.67
12.21
Rlood
Ph 2 HOT 1 !
6.50
4.67
10.50
6.34
4.22
8.94
16.99
10.9 >1
10.65
5.83
19.02
10.8-3
8. 37
6.32
1 1.53
7.10
1 1.66
13.19
5. 86
8.90
12.06
12.17
6.10
9.89
11.31
7.I;1
9.32
1 1.66
15.21
43
39
40
40
40
41
41
3R
44
41
40
42
.38
4!-i
42
41
38
30
41
41
43
38
41
45
44
32
4-')
40
4P!
37
Sn
ICT 2 F'EP 1 FSP 2 CO 1 CO 2 (
4-1
37
39
4 I
4?
43
38
37
41
4.1
.37
39
36
43
41
13
35
4.3
4;)
33
42
37
38
45
4'1
3;i
36
42
41
33
6fi
66
1 f'9
57
55
81
60
63
65
57
70
73
86
1 15
55
77
18
68
0
75
59
.84
74
62
137
57
4.3
57
93
7(1
IH7
66
70 . 1
79
48
79 .6
81 .2
87 .6 I.I
47
66 .2
72 l.t<
52
64 .2
39
83
69
109 .2
24
88
61 .1
76 .5
1 12
79
74
58
115
87
64 .,1
:rHn i'nlnt
.9
?.5
. ''
0 1.4
. .)
. >
2 .2
0 .0
4 1.6
. "3
1 .2
1 . /
. .)
4 .,)
'*'
. 5
. . i
:i .2
. .)
. i;i
M 1 . 4
0 .-J
. il
.2
. 5
"
. 'l
. >1
.2
1 .:-)
-------
I'.
APPENDIX G (CONTINUED)
Us Id.
liJ Si
52.1*1
722H
7650
7H-J5
/225
'1* 1 5
/52v)
7 3V 5
7 55'"
761;)
7 1
-------
APPENDIX H
Fi.ngerpri.ck samples 1 and 2. for each participant are given
in Appendix H. The units are explained below.
Participant ID
Fingerprick 1 Lead concentration in fingerprick
sample 1 ([_ig/100 ml)
Fingerprick 2 Lead concentration in fingerprick
sample 2 (Hg/100 ml)
325
-------
APPENDIX .W
P^rticioant ID Fimerorl ck I r irnrirV-
10.90
'4 '52 16.00
13.70
19.60
ii. 10
10.40
0107 5.4
Ml-12 14.1
0\ s i
167 13.90
/ i x r-i
/1168 15.10
I2-1^
8.10
19-70 |~/;yf,
5.40 ?-).UV
.
19.60 pp 5,,
^ <_)«'
B.10 ,7 ?n
' ' - ' '
9. D0
2 ^98:^
" :r
-'1 V..5H
54 «« 17.40
b417 2?*60 !!-^
w* il:^ -f:
-")4c;^ « :>«. '
r- , ,,-, O . id 0 I 5 V-< *
D497 17 I i 1^.^
5'^'7 Mr, -;-'^
16.10 ,
326
-------
APPENDIX H (CONTINUED)
P^rticioant ID
5 5 5 7
556V
55^,6
5597
5606
5662
5667
57'i!
5732
5/57
5771
5776
5781
71 71
7272
7291
7392
7642
7647
7702
7722
7737
7788
7797
7807
78 1 2
7817
7851
9131
922!
9^32
9292
9293
9297
9303
9^72
9382
9392
Q'72
9-'-9 2
9493
9194
9 -96
5.4
-------
APPENDIX I
The paint lead concentrations, distance from street, and
composition of each household are given in Appendix I . The
abbreviated captions and units are explained below.
Hsld. ID
Site
Inside 1st & 2nd
Outside 1st & 2nd
Dist 1
Dist 2
Composition
Household ID
Traffic density site
Lead concentrations in paint from, two places
inside the house (mg/cm^)
Lead concentrations in paint from two places
o
outside the house (mg/cm )
Distance from the street to the front of
the house (ft)
Distance from the street to back of the
house (ft)
Composition of the house, cross street, speed limit
BK
WD
ST
ASB.SHGL.
AL. SHGL.
AL. SIDING
METAL SHGL.
WD.SHGL.
Brick
Wood
Stucco
Asbestos shingle
Aluminum shingle
Aluminum siding
Metal shingle
Wood shingle
328
-------
APPENDIX I
Hsld. Inside In<;ld» Outside Outside
ID Site 1 st 2 nd 1 st 2 nd Dlst 1 nist 2
0005
0010
0015
0025
0*450
H355
0060
0065
0073
0075
0^J80
0^90
0100
0105
0120
0130
0135
0140
0145
0153
0155
0163
0165
0170
0183
0250
0255
0263
0265
0270
0275
1395
1400
1405
1410
1415
1420
1425
1515
1520
.2
.2
.3
.0
.0
.0
.0
.5
.0
4.3
.3
.3
.0
.0
.0
.2
.0
.7
.7
.2
.0
.0
.7
.0
.2
.0
.3
1. 1
.0
.0
.0
.3
1.7
.0
.7
.0
.0
.0
.0
.3
.0
.0
.0
I.I
.0
.0
.0
.0
.0
.0
.9
.0
.0
.0
.3
.0
.2
.5
.0
.2
.7
.7
.2
.2
.5
.0
.0
.2
.2
.0
.0
2.5
.0
.5
.0
.0
.0
.0
.9
.9
1
1
12
1
1
2
1
2
10
1
3
1
2
1
3
.7
.4
.9
.0
.4
.0
.0
.0
.5
.0
.0
.0
.5
.0
.4
.7
.7
.7
.0
.0
.7
.9
.1
.7
.0
. 1
.9
.2
. 1
.0
.3
.9
.3
.0
.0
.6
.2
.2
.7
4.
,
1.
^
a.
2.
.
.
,
i.
i!
2.
.
.
.
1.
2.
2.
.
I .
1 .
.
1 .
5
0
1
0
6
1
0
7
3
9
1
0
6
3
5
5
0
4
1
7
0
6
5
0
6
40
42
38
79
53
51
43
1 16
82
80
68
73
70
69
90
44
44
44
43
42
47
42
62
64
42
77
85
63
52
54
64
46
63
48
46
49
46
47
47
51
66
74
69
1 19
86
85
1 19
178
1 15
1 13
94
88
98
98
183
66
32
74
77
86
3:4
73
85
9fl
71
10.3
128
100
73
86
103
69
86
74
69
7S
71
75
81
81
Composition
wn*w
BK.
BK. iwn.
BK.
BK.
BK.
BK.
BK
ST.
ST.
BK
BK
BK
BK.
BKSWO.
T^.WD.
RKKWn
/iD^> RK
3 ^
BK&wn
8K*WD.
TK&iVH
BK
8KSWD
BK
BK
iVO.
BK.
BK
BK
'.(O.ABK.
1KS.HD
BK&WD.
8K*WD
8KS.W")
BK.
TKAifl.
.VD&BK
SKKWD.
329
-------
APPENDIX I (CONTINUED)
Hsld. Inside Inside Outside O'jtsHe
ID Site 1 st 2 nd 1 st 2 nd Dlst 1 Hist 2
1530
1535
1540
1553
1560
1565
1843
1345
1850
1855
I860
1873
1875
189^
1895
I9v>0
1905
1915
1925
2145
2160
2165
2173
2175
2200
2205
2410
2440
2463
2465
2470
2480
2485
2490
2500
2505
.2
.3
.0
.2
.5
.0
.0
.5
.5
.0
.2
.5
.0
.9
.2
.2
.0
.0 --
.5
.0
.0
.0
.3
.0
.5
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.3
3105 2 .0
3115 2 .9
-3210 2 .0
3220 2 .0
.0
.0
.0
.0
.a
.0
.0
.2
.0
.0
.2
.2
.2
.2
.7
.0
.0
.7
,2
.0
.0
.0
.0
2.1
.0
.5
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.1
.2
.0
2.1
.5
.0
.9
.7
1.4
.0
1 .4
.2
1.4
.0
2.5
2.7
.9
.0
.0
1.1
.0
.5
.0
.0
2.5
3.0
3.0
.9
.0
.0
.0
.0
.0
.0
.0
.0
.0
.5
.0
.0
.7
.5
.0
.0
.5
.2
1 . t
.2
2.5
3.0
.2
.0
.7
.3
.0
4.4
.0
6.5
.5
.0
.0
.0
.0
.0
1.4
.0
38
47
45
4.3
44
43
40
45
45
47
5">
48
45
44
44
53,
64
44
40
S3
6(5
69
65
57
52
73
75
83
62
48
57
71
59
53
75
60
87
45
43
48
71
72
73
63
68
71
75
75
76
76
76
72
67
76
67
71
87
69
68
I 17
96
1 14
85
89
73
1 17
148
126
95
1 19
87
107
92
89
1 14
103
1 14
34
63
83
Comnosl tlon
RK*WD.
rtO
BKXWD
BKAWD.
WD1BK
^K&wn
Rtr*WD
,yn&RK
WD*8K
3K&WO
.KQ*RK. '
3KKWO
rtnXRK
r*v:*WD
s»c*wn
3K
3KSWD
BK&IVD.
3<
BKKWn
BK
3K.
BK&WD.
BK
BK
3K&WD.
F)K
BK
WD*BK.
3KRWD.
rfD
RK
BK&WT)
HD.
ST.
BK.
BK.
RK
RWWD-.35MPH
R>C-35MPH
330
-------
APPENDIX I (CONTINUED)
Hsld. Inslrie Inside 0-.
ID Site 1 st 2 rrf
:«lri<> Outside
! st 2 nd Dist 1 Hist 2
3225
3230
3425
3435
3665
367(9
3675
3695
3870
3875
3903
3905
5383
5390
5395
5575
5593
5645
5663
5665
5670
5675
5685
5703
5725
7100
7185
7265
7290
7583
7605
9013
9015
9023
9,470
9075
909(3
9095
9100
9105
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
2
2
2
3
3
2
2
2
2
2
2
2
2
2
.5
.2
.2
.0
.3
.0
.0
.0
.5
.3
.3
.0
.0
.0
.0
.0
.5?
.0
.0
.0
.0
.0
.2
.2
.0
.0
.0
.2
.3
.3
.0
.0
.9
.a
.d
.0
.0
.0
.0
.0
.a
.2
.3
.3
.2
.0
.0
.3
.0
.5
.3
.0
.0
.0
.0
.0
.0
.0
.0
.0
.3
.0
.0
.0
.0
.4
.0
.2
.0
.7
.7
.2
.0
.0
. J
.0
.0
.0
.0
.7
.0
.2
.0
.7
J.4
2.1
4.7
.0
2.7
1.3
.0
.0
.0
1 . 1
3.9
.0
.0
.0
.0
.0
.9
I.I
.0
.0
.2
1.4
2.5
.0
2.5
.5
.7
.2
1.8
.0
.0
2.3
.0
.0
1 .8
.0
.7
.5
.0
.5
.0
.5
.0
.0
3.0
.0
.0
.3
1. 1
2.5
.0
43
53
39
51
60
62
60
59
85
J20
41
42
103
64
93
42
21
42
45
52
38
55
53
23
15
83
84
43
63
55
46
59
68
63
39
38
40
39
42
32
66
70
62
82
91
100
89
89
\2"i
175
96
97
130
95
1 19
76
8T
8H
81
92
125
91
91
67
97
128
122
65
96
1 12
74
87
97
98
54
69
79
6?
65
65
RKKWH.-35VPH
'/Wr.-35>'PH,
rtn&RK.
BIT
RK.
BK.
8K8.WO
ST..
RK&WD
RK.
RK.
BK.
RK.
RK*WD.
Rf
RTKWD.-35MPH.
SK-HILLCPf-ST
Rl«'^wn.-35"PH
Rr*WD-35VPH
RK.*ST.*WD.
RK-HILLCPEST
RKKWD.-35MPH
R<
BK-FHDGUSON
BK-PPFSTON
RK
BKXWD.
BK
BK.
BK
RK
BK
BK
RKAWD
RK
RK.
RK.
BK.
RK*WD.
RK*WD.
331
-------
APPENDIX! (CONTINUED)
Hsld.
ID
91 15
9125
9130
9195
9205
9210
9215
9220
9225
9230
9235
92d0
9285
9290
9295
9300
93k)5
93M
9315
9325
9340
9345
9360
9365
9370
9380
939J
9435
9490
3030
3090
3100
3310
3335
3650
3660
4023
4025
5083
5085
InsHe Inside 0;
Site
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
1 st
.0
.3
.0
.0
.0
.9
.0
.0
.0
.0
.0
.5
.0
.0
.0
.0
.0
.0
.0
.2
.2
.5
.0
.0
.0
.2
.0
.0
.0
.0
.0
.3
.0
.a
.0
.2
.0
.2
.5
2 nd
.0
.5
1.6
.0
.9
1 . 1
.0
.0
3.0
.0
.0
.0
.0
2.6
.0
.2
8.3
.2
.0
.9
.0
.0
.0
.0
.2
1.8
.2
.0
3. 1
.0
.0
.0
.0
.0
2.4
.0
.0
.0
.2
Jtslde Outside
I
2
2
3
5
3
1
1
1
3
2
3
1
9
2
1
st
.2
.7
.0
.0
.5
.0
.0
.0
.7
.5
.7
.8
.6
.4
.2
.0
.0
.0
.6
.4
.9
.3
.0
.0
.0
.0
.8
.0
.4
.0
.5
.0
.5
.0
.5
.1
.0
.7
2 nd Dist 1 Dlst 2
1.1
2.5
.0
.4
3.9
6.2
.0
.0
.0
1. 1
.9
.0
.0
.0
.0
1.6
5.6
1 . 1
6.9
.2
2.7
.0
2. 1
.2
.0
.0
.5
.0
42
39
31
43
41
44
73
57
57
60
60
54
32
60
60
60
50
50
50
45
44
57
17
83
42
60
33
53
40
70
66
67
46
46
53
62
36
36
40
45
77
73
65
74
7*
73
1 1 !
85
91
103
|f)4
35
73
90
84
86
33
91
83
78
65
87
57
124
94
84
66
95
69
100
93
103
67
74
84
91
64
64
100
108
Compost t Ion
BKSWD
Bf.
R < & WD .
RK.
ID. 35'^PH.
WO-35«PH
rfOOD*AL.SIDINO-35MPH.
RK&wn
WD.
RK
RKKWD.
WO.-35MPH
WD. 35MPH
RKRWD.
rtD. 35MPH
WD. 35''PH
BK.-35MPH
WD^AL.SHGL J
R<-35Ut>H
WD-35MPH.
WD.-35MPH
WD. 35'fPH
ifD.-35>
-------
APPENDIX I (CONTINUED)
Hsld.
ID Site
5J93
5395
51013
5105
5130
5135
5165
5249
5365
5373
5403
5405
5440
5445
5455
5463
5465
5473
5483
5493
5495
5505
5510
5523
5525
5543
5545
5550
5555
5565
5585
5595
5635
5625
5635
5680
5690
5695
5705
5715
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
Inside Insldf? Outside Outside
1 st 2 nd 1 st 2 nd Dlst 1 ni,t 2
.0
.0
.0
.0
.0
.0
.0
,ij
.0
.0
.0
.3
.0
.5
.2
I.I
.2
5 ^
.3
.0
.3
.0
.0
.2
.0
.3
.0
.9
12.5
.0
.2
.0
8.0
.2
.0
.0
.3
.3
.0
.a
.a
.0
.0
.0
.0
.0
.2
.0
.0
.0
.0
.0
.5
.0
.0
.0
.a
.2
.0
.0
.3
.0
.7
.0
5.3
.0
.0
.0
.0
.0
2.5
.0
.0
.0
2.3
.a
.0
2
6
3
1
1
1
2
1
I
1
2
2
1
1
1
.0
.0
.0
.2
.9
.5
.0
.7
.0
.3
.6
. 1
.7
. 1
.0
.8
.0
.0
.0
.7
.0
.1
.0
.4
.a
.1
.0
.4
.0
.1
.0
.5
.3
.0
.8
.6
.6
.7
.0
.0
.3
.9
.0
.0
1 .5
4.6
1.6
I.I
.0
.0
12.5
3.3
.0
3.2
.0
.9
.9
2.3
3.8
3.0
.3
) . 1
1.6
.7
46
43
46
47
35
44
40
45
49
49
60
62
93
84
58
44
39
34
82
150
35
25
30
25
45
16
50
47
25
21
38
41
58
12
56
23
61
65
26
96
89
87
79
73
77
65
7'?
78
79
83
91
132
1 30
1 16
84
84
64
105
233
94
73
57
73
78
56
10'*
83
71
71
1 11
68
138
57
94
89
8'1
185
62
Con on i It Inn
RK
RK
RK
BK
RK
RK
RK 35«PH
RK. 3"5MPM.
RKKWn.
dD.
RKKWD
RK.
rfD.
RK.
RK.
Rf. 3C5'
-------
APPENDIX I (CONTINUED)
Hsld.
Inside Insidfi Outside Outside
ID Site
5733
5735
5743
5745
5755
5763
5765
5773
5775
5783
5785
5850
7375
7343
7385
7623
7633
7655
7705
7723
7745
7775
7853
9473
9475
3703
5225
5233
5363
5413
5415
5485
7173
7183
7223
7225
7270
7275
7393
7395
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1 st
1.4
.0
.3
.3
.0
12.5
.0
.0
.3
.3
.3
.0
.3
.5
.0
2.1
.3
.3
.3
1.6
.3
.2
.3
.3
.0
1.6
.3
.3
.3
.3
.3
.3
.2
.0
.0
.0
.0
.3
.0
2 nd
.3
.0
H
.0
.2
.0
.0
.0
.2
.2
.0
.0
.0
.3
.3
4.4
.0
.3
.3
.i)
.3
.y
.0
.0
.<3
.0
.0
.3
.0
.0
.5
.2
.5
.0
.0
.0
.2
1.4
.0
1 St
.2
.3
.0
.0
1.6
4.1
.3
.3
.7
.0
.3
.0
1.6
.9
1.6
12.5
.3
.0
.3
4.6
.0
1.3
.0
1.4
.0
.3
.9
.2
.0
.9
1.8
.5
.7
.0
.0
.0
I.I
1.4
2 nd Dlst 1 Dtst 2
.0
.3
.0
.0
2.3
.0
.3
.0
.2
1. 1
2.3
12.5
.9
.7
.0
.9
.3
.5
.3
.0
.0
.0
1.4
.3
.5
1.6
.0
.0
66
71
49
72
50
91
22
15
33
63
73
68
50
47
43
73
45
73
1 15
45
49
54
83
48
61
82
56
69
47
49
53
80
1 13
47
57
57
63
53
47
44
131
Iflfl
133
132
ai
135
84
74
76
92
1(14
I03
131
88
66
1.34
72
1 15
167
92
95
123
125
1 0H
1 32
I3S
9>^
1 18
84
76
79
1 18
158
73
99
104
93
85
69
67
Cnnnosl tion.
BK
BK
RK
BK
RK&wn-BELT LINE
ST.
RK.
RK RH.r f-INE
RK.-RELT LINE
RKKrfD.
RKXASR. SHGL
3K.
BK
wo.
RK&WD.
RK
WD.-
RK
BK
wn.
qr-CCHAR CREST
RK»,',VD. E. MOCKINGBIRD
.vn&^SR. SHGL
RK. PRESTON
RK.-WPRBS CHAPEL
RK.
rtD 4CH1PH
BKKiVn-40MPH
HD.
BK1WD.
RKAASR.SHGL
BK*rtD. 4PMPH
VSB.SHGI.-35MPH
RKAHH. 35VPH
BK
BK
BK
BK
A SB. SHGL
WD.
^
334
-------
APPENDIX I (CONTINUED)
Hsld. Inside Inside Outside OutsHe
ID Site 1 st 2 nd I st 2 nd Rlst 1 Dlst 2
7415
7515
7523
7525
7545
7553
7585
761 i)
7643
7545
7650
7660
7665
7673
7675
7683
7685
7690
7700
7715
7725
7735
7750
7760
7765
7780
7785
7790
7795
7300
7805
7813
7315
7320
7325
73313
7335
7840
7845
9465
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
.0
.2
.0
.5
.0
.9
.0
.0
.0
.5
.0
.0
.5
.0
.0
.5
.0
.0
.0
.0
.0 -~
.0
.0
.0
.0
.2
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.2
.9
.5
.0
.3
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.2
.0
.0
.5
.0
.0
.0
.0
.0
.0
.0
.5
.0
.0
.2
.0
.0
.2
.0
2.7
.0
.0
.0
1.4
3.2
.5
.2
3.2
.9
.0
.0
.0
.0
.2
.0
3.0
.0
.0
.0
I.I
.5
.0
.0
1.4
.0
.0
.2
.0
7.5
.9
.0
.5
.0
2.3
.7
.0
.2
2.1
8.7
.2
.9
.2
.5
.0
.2
.5
2.3
J.6
.0
2.5
.0
3.9
.0
77
57
67
60
81
63
55
122
45
152
107
32
35
35
25
24
33
35
45
33
31
51
35
34
44
33
34
34
41
43
33
59
30
31
35
47
34
30
41
55
1 15
141
93
1 13
1 15
1 14
86
162
37
175
1 32
75
76
67
95)
95
. 95
1 IM
30
1 15
1 1 1
75
74
37
93
74
93
87
35
37
69
' 33
84
82
85
74
9v5
79
85
125
Composition
Rv*WD.
T'&wn.
3WWD.
RKKWT.
RK
RK
Rlf
WO&RK
ASB.KD SHGL
rt'D&RK
>VD.
RK-COIT RD
3K.
RK.
3 K. -A R PA. MS
RK. -ARRAYS
BK-COIT PD
BK PPESTON TO
SK.
RT-MADSH
BK.->'\RSH
RV-w*LNUT
Rfft WD.-WA.BSH 35MPH
wn&RKT-MAPSH-aSf-'PH
RK WALNUT
Rf.-COIT PD
RkT.-COIT PD
Rf COPNEP HOUSE-COIT
ST*WD
.vnitJK-WALNUT
ASR.SHGL
R>C.-PPI=STON
B^-COIT
RK-COIT
RK&wn.
T'-COIT
RK.-HILLCPEST
RK.
RK- WALNUT
RK.40MDH-FOPCST LN.
335
-------
.APPENDIX J
Report to Southwest Research Institute by
Geoderma Consultants - Dallas, Texas
336
-------
STUDY OF SELECTED SOIL CHARACTERISTICS IN
DALLAS COUNTY/ARLINGTON AREA IN RELATION TO
THE RETENTION, RELEASE AND TRANSMISSION OF
AUTOMOBILE EXHAUST LEAD
REPORT TO SOUTHWESTERN RESEARCH INSTITUTE
by
Geoderma Consultants
4810 Cole Avenue
Dallas, Texas 75205
October 26, 1976
337
-------
Abstract
A study was performed to examine soil characteristics
which have the potential to affect lead adsorption, reten-
tion and release from top-soils in the Dallas/Arlington
metroplex area.
In particular, texture, clay mineralogy, organic con-
tent, and soil pH were studied. The majority of soils were
high in clay content and organic matter and had high poten-
tial for adsorption and storage of automobile exhaust if
it were deposited on the soil surface and allowed to infil
trate into the top soil. Soil chemistry is such that rela-
tively insoluble precipitates are likely to be favored.
The slowly permeable native of the clay soils might
reduce the amount of lead infiltrating into the soil and
allow it to be removed by runoff.
Introduction of exotic sand has modified the texture
of 10 to 20% of the soils checked by laboratory procedures
and would reduce lead retention in such soils.
338
-------
Table of Contents
Page
Title Page i
Abstract ........... ii
Table of Contents iii
I. Introduction ........ 1
II. Method 2
III. Results 3
a) S.C.S. Classifications of Soil Textures . 3
b) Laboratory Analysis Check on Grain Size . 5
c) Clay Mineralogy of Selected Soil Samples. 5
d) Organic Content of Soils 7
e) The pH of Soil Samples 8
IV. Conclusions ......... 9
V. List of Tables 11
Table 1. Master List of Soil Conservation
Service Texture Classification. ... 11
Table 2. Classification of Top-Soils Accord-
ing to the S.C.S. Soil Texture Classes. . 23
Table 3. Laboratory Analysis of 50 Soils
for Grain Size, pH, Organic Content and
Clay Mineralogy(10 Samples) .... 24
Table 4. Summary of the Percentage of
Organic Matters in Soils of the Laboratory
Analyzed Sub-Samples ... ... 28
VI. References ... 29
339
-------
I. Introduction
A study to examine levels of lead in the blood of
Dallas/Arlington metroplex residents was conducted under
the auspices of the Southwest Research Institute and fund-
ed by the Environmental Protection Agency. It was hypo-
thesized that top-soils in yards might be a source of lead
absorbed into the blood of the local residents. The ori-
gin of the lead was thought to be automobile exhaust.
A supplementary study was conducted contemporaneous-
ly with the examination of lead levels in residents' blood
and in yard top-soils. The study examined the soil compo-
nents which would affect the retention and release of lead
from the soil and soil characteristics which might deter-
mine potential routes for lead ingestion into the human
bloodstream.
Soil components known to interact with atmospherically
derived lead are clay minerals and a variety of chelating
and complexing compounds. Soil chemistry, in particular
pH and the availability of anion groups such as C0§, SO^
and phosphates, is important in the formation of lead com-
pounds of varying solubility and in affecting the fixation
of 1ead in soil.
Other characteristics such as texture, amount of hu-
mus material and permeability are important in determining
the leaching environment, and the susceptibility of top-
340
-------
soil to erosive forces which remove and transport lead
contaminated particles.
The supplementary study, designed to provide back-
ground information to aid in the interpretation of the
lead analyses, was confined to top-soil environments, 0-6
inches, in residential yards. Samples were collected by
staff of the Southwest Research Institute.
Each sample was classified according to grain-size
distribution (texture) and a large sub-sample was random-
ly chosen for more detailed laboratory analyses of tex-
ture. pH, clay mineralogy and total organic matter.
II. Method
Soil Conservation Service (S.C.S.) mapping of soils
in Dallas County and Arlington is available on 1:20,000
scale aerial photographs. The location of each sample
site was examined and the soil texture described in the
S.C.S. soil series descriptions, noted for the upper six
inches .
A randomly chosen sample of 50 soils were examined
by laboratory procedures to ascertain the reliability of
the S.C.S. classification and mapping.
The same 50 samples were later analyzed for pH and
total organic content. The three major soil textural
classifications found in the Dal 1 as/Ar1ington area were
341
-------
analyzed for clay mineral species using x-ray diffraction
methodology; a total of ten soils were thus examined.
Additional information on soil chemistry was taken
from the S.C.S. soil series descriptions and available
chemical data and used as an aid to interpret the soils
as potential sources and sinks of environmental lead.
III. Results
a) S.C.S. classification of soil textures (Table 1)
More than 50% of the soil samples had a clayey
(C) texture (see tables 1 and 2). For these soils clay
size particles constitute greater than 40% of the inor-
ganic content. Clay size particles are the principle ad-
sorbers of divalent lead (Pb~) and retain lead longer than
other inorganics. The clay textured soils have the great-
est potential for lead storage.
Greater than 33% of the samples have a silty clay
(Sic) texture- These soils also have greater than 40%
clay size particles and a similar high potential to retain
le.ad as the clayey soils.
Clay loam (CL)and sandy clay loams (SCL), greater
27% and 20% clay respectively, have sufficient clay to make
them significant storers of lead but constitute such a small
part of the total sample as to be insignificant.
Sandy loams (SL) have less than 20% clay and are
342
-------
capable of adsorbing far less lead on the mineral surfaces
than the clayier counterparts.
* The majority of the soils have textures and min-
eralogy which make them slowly permeable to rain-water or
artificial sprinkling. Lead deposited in the upper hori-
zons would, after initial wetting of the soil, be subjected
to a slow downward flow of water through miniscule capil
lary spaces and along adsorbed layers coating the clay min-
eral surface. Downward transportation of lead would be
minimized compared to the coarser textured loamy sands or
sandy loams. Indeed ca ions which are usually more solu-
ble and transportable than Pb ions are often incompletely
removed from the top-soils of the clays and silty clays.
However, the surface water runoff from the soils
is greater than for the coarser textured soils and if auto-
mobile lead were confined by fall-out to the very upper
part of the topsoil surface it might be largely washed a-
way across the soil surface during heavy thunderstorms on
even gentle (>1%) slopes.
The clayey and silty clay soils have a certain
degree of cohesiveness when dry and are less susceptible
to wind erosion than the loamier or fine sandy soils. How-
ever, transportation into houses by wet clay adhering to
boots and shoes is obvious.
343
-------
b) Laboratory analysis check on grain size
Fifty of the samples were checked for grain size
by laboratory procedures. Twenty percent had coarser,
sandier textures than allowed by the S.C.S. mapping units
and classification (Table 3).
Two possible reasons exist for this discrepancy.
Soils of limited areal extent can either be overlooked
during field mapping or be too small to include as separ-
ate map units. Alternatively, the soil texture could have
been altered by the addition of exotic sands for purposes
of improving soil drainage or yard fertility, or during
cons truction.
The practice of adding fluvial sands to clayey
or silty clay soils is common on the Blackland Prairie
areas of Dallas County and other parts of North Central
Texas. That it occurred for 20% of the subsample serves
as a warning that a considerable part of the total sample
probably has been modified by sands brought into the gar-
den. This may be significant when interpreting the impor-
tance of on-site adsorption of lead by soil minerals.
c) Clay mineralogy of selected soil samples
Ten subsamples representing the 3 major soil tex-
ture classes were analyzed for clay mineralogy (Table 3).
The clay fraction was divided into coarse (2p to 0.2p) and
344
-------
medium to fine (<0.2y) sizes.
The clay textured soils showed no significant
difference between coarse and medium size fractions. Mont-
morillonite (>40%) was dominant, kaolinite present in
smaller quantities (10-20%). Montmorillonite has a high
cation exchange capacity (CEC 80 to 120 me.) and a great
potential to adsorb and retain lead. Kaolinite (CEC 10
to 20 me.) has a lesser potential to retain lead. Thus
the clayey montmori1lonite soils that formed over the
Austin Chalk and Eagle Ford Shales could be significant
sinks for atmospheric derived lead. Especially if acting
in conjunction with high amounts of soil organics.
The clay mineralogy of the silty clay textured
soils changes from coarse to medium clays. In the coarser
clays, degraded mica (illite?) predominates (>40%) with
kaolinite and montmori1lonite in lesser quantities (10-
20%). The degraded mica has a high cation exchange capa-
city and could be a good potential storer of lead. In the
finer clay fraction montmori11onite is the dominant clay
(>40%) and consequently the potential for storing lead on
the many small clay surfaces is high.
The sandy loam textured soils are predominantly
kaolinite (>40%) in the coarser fractions, with degraded
mica and mixed mica/montmori1lonite interlayered clays
subordinate. Kaolinite, as previously noted has a lesser
345
-------
potential for adsorbing lead. In the finer textured clays
kaolinite gives way to mixed layered clays and these.mica/
montmori1lonite interlayered minerals have a high cation
exchange capacity and potential to adsorb lead. However,
the smaller total clay amount in sandy loam must be con-
s idered.
The predominant adsorbed cation is ca + in most
of the soils derived from the Cretacious formations.
Bittel and Miller (1974) measured the exchange of Pb2 +
against Ca on montmori1lonite , illite and kaolinite.
In these experiments Pb2 + was preferentially adsorbed, re-
sulting in selectivity coefficients of about 2-3.
d) Organic content of soils
The great majority of soils (80%) had organic
contents between 6 and 15% dry weight (Table 4). Sixty
percent of soils had organic contents greater than 9%.
These soils have moderately high to high organic amounts.
Various organically derived compounds and organo-clay com-
plexes are the major retainers of lead along with clay in
the soil. Although the analyses was not designed to anal-
yze for chelates and complexing agents, it is a reasonable
assumption that soils with high percentages of organics
anc clays have a good probability of forming such deri-
vatives. Certainly clay soils with organic matter >9%
346
-------
have much greater potential to adsorb and retain environ-
mental lead than the sandier or loamier textured soil with
low organic contents such as samples #9100 and #9345.
e) The pH of soil samples
The great majority of the laboratory analyzed sub-
samples had pHs between 7.0 and 8.0. This is within the
neutral to moderately alkali range that is expected of top-
soils derived from calcium carbonate rich parent material.
Apparently yard practices, including addition of fertili-
zers and organic matter (?) have not substantially changed
surface soi1 pHs.
Relatively little is presently known about lead
chemistry in soils. Gasoline combustion being the main
source of lead most of the lead will be deposited as solu-
ble halides, lead Chlorobromide, Pb CLBr. Singer and Han-
son (1969) described how excess lead is probably decreased
after deposition on the soil due to the formation of rela-
tively insoluble compounds with carbonates, phosphates and
sulfates. The clayey and silty clay soils derived from
the Austin Chalk and Eagle Ford Shales are high in both
carbonate and sulfate ions. At neutral to moderately
alkaline pHs the formation of the lead carbonate and lead
sulfates would be favored At very high pHs the lead would
be partially released from these compounds. At very low
347
-------
pHs lead would be desorbed from the clay-mineral surfaces.
However, the large reservoirs of ca ions in the soils,
especially on the clay surfaces acts as a buffer against
extremes of pH and favors stabilization and retention of
the lead against leaching and root absorption. The low
hydraulic conductivity (K factor) of the clay and silty-
clay soils would further reduce the leaching of the lead
compound precipitates.
IV. Conclusions
Soil characteristics in the study area appear to be
favorable for the retention of lead and lead compounds
within the soil system. In particular high quantities of
negatively charged clay mineral surfaces for adsorption
of Pb^+, high organic contents to form complexing and
chelating agents,and an ample supply of sulfates and car-
bonates at favorable pHs so that insoluble lead compounds
can form. In addition soil permeability is low due to
the swelling clays in the profile,and the climatic char-
acteristics of heavy rainfalls with associated high water
losses in surface runoff do not favor leaching. The latter
factor might be important in reducing the initial infil-
tration of automobile lead into the soil due to losses in
surface runoff, or preferential movement of water through
dessication cracks to depth within the profile.
348
-------
As high as 20% of the sub-sampled soils had coarser
textures than would be expected from S.C.S. mapping and
classification. This is thought to be due to the practice
of mixing fluvial sands with clayey soils to produce more
favorable textures for plant growth. The sand would have
lesser capability to retain lead within the soil system.
The problem of the exotic sands being contaminated with
lead from non-automobi1e exhaust sources has not been
studied and is unknown at this time.
349
-------
Table 1. Master List of Soil Samples with Soil Conserva-
tion Service Texture Classification
Sample #
2485
2490
0005
0010
1840
1880
1890
1895
1900
1905
1925
0050
0120
1505
1515
1520
1530
1535
1540
1550
1560
1565
2160
2165
2170
2175
2180
2200
2205
Traffic Density
474
474
320
320
320
320
320
320
320
320
320
558
537
571
571
571
571
571
571
571
571
571
178
178
178
178
471
471
471
Texture
C
Sic
C
C
Sic
Sic
Sic
Sic
C
C
Sic
Sic
C
C
C
C
C
C
Sic
C
C
C
Sic
Sic
Sic
Sic
Sic
Sic
Sic
350
-------
Sample
0130
0135
0170
0270
2440
2460
2465
2470
2475
2480
2410
2500
2505
2145
0025
0090
1860
1870
1875
1915
0015
1845
1850
1855
1395
1405
1410
0250
1400
0140
0145
Traffic Density
346
346
346
346
259
259
259
300
300
300
479
364
364
336
599
599
335
335
335
335
194
194
194
194
155
155
155
472
202
336
336
Texture
Sic
C
C
n
\_
c
Sic
Sic
Sic
C
C
Sic
C
C
Sic
Sic
C
Sic
Sic
Sic
Sic
C
C
C
C
C
C
C
C
C
C
C
351
-------
Sample
0150
0155
0160
0165
0180
1415
1425
1420
0060
0065
0070
0275
0255
9220
9225
9295
9300
9380
4020
4025
3210
3220
3225
3230
3900
3905
3310
3335
3870
3875
3285
Traffic Density
336
336
336
336
336
343
343
255
186
596
596
471
596
11338
11338
13072
13072
13072
17931
17931
11765
11765
11765
11765
10200
10200
13800
13800
11467
11467
12944
Texture
C
C
C
C
C
C
C
C
C
C
C
C
C
SL
SL
SL
SL
SL
SL
SCL
Sic
Sic
Sic
C
Sic
Sic
Sic
Sic
CL
Sic
Sic
352
-------
Sample
9305
9315
9485
9310
9325
9340
9345
9360
9390
5620
5685
3650
3660
3665
3670
3675
3695
9480
3105
3425
9490
9495
3435
9010
9015
9020
9070
9075
9090
9095
Traffic Density
8464
8464
11494
9708
9456
9456
9456
9456
9456
12514
12514
16128
16128
12514
12514
12514
12514
12514
9588
8651
5918
5918
6654
7154
7154
6776
7154
7154
7154
7154
Texture
C
C
Sic
SL
C
C
C
C
C
Sic
Sic
C
C
C
C
C
C
C
Sic
C
C
C
C
Sic
Sic
Sic
Sic
Sic
C
Sic
353
-------
Sample
9100
9105
9115
9125
9130
5575
5615
5645
5660
5675
3080
3090
3100
9195
9290
9205
9210
9280
9365
9370
5440
5445
5555
5730
5780
5545
5635
5365
5370
5625
5695
Traffic Density
7154
7154
7154
7154
7154
10637
9362
9362
9362
10637
16219
16219
16219
8209
8197
11708
11708
11817
11708
11708
16882
16882
13800
13700
13700
18245
18354
15257
15257
18354
17119
Texture
Sic
Sic
C
Sic
Sic
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
SL
SCL
C
C
C
C
C
C
SL
C
C
354
-------
Sample
5670
5520
5665
5225
5230
5485
5240
5480
5490
5550
5510
5585
5080
5085
5090
5095
5100
5105
5110
5115
5455
5460
5725
5410
5415
5360
5400
5405
9465
5605
Traffic Density
1244
13800
11765
20890
20362
20890
13800
17452
17452
17000
15000
18245
18790
18790
18790
18790
18790
18790
18790
18790
15442
15442
12411
20928
20928
20928
15570
15570
20362
17119
Texture
C
C
C
C
C
C
C
Sic
Sic
C
SL
Sic
Sic
Sic
Sic
Sic
Sic
Sic
Sic
Sic
Sic
Sic
Sic
SCL
SCL
SL
SL
SL
C
C
355
-------
Sample
5560
5380
5390
5595
5680
5700
5505
9475
5130
5135
5525
5540
9470
5715
5495
5500
5590
5470
5690
5710
5165
5465
5770
5565
7815
7840
7690
7750
7835
Traffic Density
17119
13000
13000
13000
18354
11765
17000
18354
14000
14000
14000
19000
17047
15156
18548
14769
12411
18245
18548
18245
18245
18245
13700
16500
31542
31542
20141
21300
31542
Texture
Sic
SL
SL
SL
C
C
C
C
Sic
Sic
Sic
C
C
C
C
Sic
Sic
C
C
C
C
C
C
C
Sic
Sic
C
C
C
356
-------
Sample
7700
7705
7720
7775
7675
7390
7395
7415
7790
7660
7680
7845
7760
7640
7780
7805
7170
7180
7850
7810
7075
7100
7580
7605
7620
7630
7655
7745
7800
3700
Traffic Density
20572
18424
14500
18790
19578
23885
23885
22670
31542
31542
19578
20432
21300
27537
31542
20928
20483
21300
18216
20141
18790
18790
18790
18790
13684
13684
13684
13684
20432
37850
Texture
Sic
C
SL
Sic
Sic
C
C
SL
C
Sic
Sic
C
C
C
C
Sic
C
C
C
C
Sic
Sic
Sic
Sic
Sic
Sic
Sic
Sic
C
CL
357
-------
Sample
7515
7520
7525
7545
7550
7610
7645
7710
7765
7340
7385
7785
7185
7665
7670
7825
7685
7755
7220
7225
7265
7270
7275
7290
7585
7725
7820
7830
7650
Traffic Density
37850
37850
37850
37850
37850
37850
37850
21336
20432
16381
16381
31542
9098
31542
31542
31542
31542
20928
20432
20432
12331
20432
20432
12331
20432
21300
31542
31542
20000
Texture
CL
C
CL
CL
CL
CL
CL
C
C
LS
SL
Sic
Sic
Sic
Sic
Sic
Sic
SL
C
C
C
C
C
C
Sic
C
C
C
CL
358
-------
Table 2. Classification of Top-Soils According to S.C.S. Soil
Texture Classes
SILTY CLAY SANDY CLAY SANDY CLAY LOAMY
CLAY(C) (Sic) LOAM (SL) LOAM (CL) LOAM (SCL) SAND (LS)
142 93 17
359
-------
Table 3. Laboratory Analysis of Subsample of 50 Soils for Grain
Size, pH, Organic Content and Clay Mineralogy (10 Samples)
#
5130
7675
1395
7705
7170
7800
3695
9090
5590
9115
3210
5550
7700
9205
5460
9315
5115
9100
5605
CLASS
16
16
90
90
9
9
12
16
12
16
9
18
9
16
41
16
16
9
GRAIN
SIZE
Sic
Sic
C
C
C
C
C
C
Sic
C
Sic
C
Sic
C
Sic
SL
Sic
Sic
C
CHECK pH
/ 7.3
/ 7.8
/ 7.2
/ 7.8
/ 7.8
/ 7.6
/ 7.6
SL 7.3
/ 7.4
SL 7.2
/ 7.7
/ 7.9
/ 7.1
/ 7.8
/ 7.1
/ 7.6
/ 7.8
LS 6.8
/ 7.4
ORG
0.7
8.8
15.9
11.8
11.3
11.4
8.2
8.1
11.0
12.0
7.3
11.5
11.7
8.3
9.0
5.2
12.3
2.7
11.3
CLAY MINS CLAY MINS
<.02y >.02p
Ml K3 Ml K2 M2
Q3
Ml K3 Ml K2 M2
Ml K3 Ml K2 Q3
Ml K3 Ml K2 QS
MI K3 MI K3 Q3
MI K3 Mj K2 Q3
Ml K3 Mi3 Mi K2 M2
Ml K3 013 M! K2 013
MCI K? Ki Mi-z Mc-i
J -L J J.
360
-------
#
3650
9345
9390
5455
7790
7835
2440
7630
2460
3675
5495
7075
2480
5585
7850
7185
9325
7745
5480
5660
5665
7100
CLASS
9
9
9
90
9
9
9
16
16
9
9
18
9
85
9
16
9
18
85
9
9
16
GRAIN
SIZE
C
C
C
C
C
C
C
Sic
Sic
C
C
Sic
C
Sic
C
Sic
C
Sic
Sic
C
C
Sic
CHECK pH
/ 7.9
S 7.8
/ 7.7
/ 7.9
/ 7.6
/ 7.6
/ 7.8
/ 7.4
/ 7.2
/ 7.8
/ 7.4
SCL 5.5
/ 7.3
/ 7.0
SC 7.5
/ 7.6
SCL 7.4
/ 7.6
/ 7.2
SC 7.8
/ 7.4
SL 7.5
ORG CLAY MINS CLAY MINS
% < . 02y > . 02(j
13.1
1.3
8.2
9.6 Mj K3 Q3 M! K2 Q13
13.4
11.1
12.3
8.0
9.6
12.7
7.8
12.9
11.7
5.5
6.2
8.8
4.8
9.6
5.3
5.5
11.5
12.4
361
-------
#
5505
7725
5080
5485
7680
3335
9305
5670
5715
CLASS
9
85
16
9
16
16
41
9
9
GRAIN
SIZE
C
Sic
Sic
C
Sic
Sic
SL
C
C
CHECK pH
/ 7.6
/ 7.6
SL 7.4
/ 7.7
/ 7.4
/ 7.2
/ 7.3
/ 7.7
/ 7.8
ORG CLAY MINS CLAY MINS
% <.02y >.02u
10.6
9.6
4.8
11.2
10.5
12.1
4.3
13.0
12.8
362
-------
M = Montmorillonite
K = Kaolinite
Mi = Mica (Illitie)
Me = Mixed layered clays; montmorillonite/mica.
Q = Quartz
1 = > 40% Estimation from
2 = > 20% areas under x-ray diffractogram
3 = < 10% curve.
Note: % will not necessarily sum to 100% due to non-crystalline
allophanic substances within clay fraction.
363
-------
Table 4. Summary of the Percentage of Organic Matter in Soils of the
Laboratory Analyzed Sub-Sample
Percent Organic
Matter >1% >3% >6% >9% >12% >15%
# Samples 2 7 10 19 10
364
-------
References
Bittel, J.E. and Miller, R.J., 1974. Lead, cadmium and
calcium selectivity coefficients on montmori1lonite,
illite and kaolinite. Journal Environmental Quality,
rr _ ^ r f\ ""i r~ T
3 : 250-253.
Lindsay, W.L.
in soils
pp . 41-51
1973. Inorganic reactions of sewage wastes
Chapter 3 of micro-nutrients in Agriculture,
Singer, M.J., and Hanson, L., 1969. Lead accumulations in
soils near highways in the Twin Cities metropolitan
area. Soil Science Society of America Proceedings,
33, pp. 152-153.
365
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/1-78-055
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
EPIDEMIOLOGIC STUDY OF THE EFFECTS OF AUTOMOBILE
TRAFFIC ON BLOOD LEAD LEVELS
5. REPORT DATE
August 1978
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
D. Johnson, R. Prevost, J. Tillery, K. Kimball,
J. Hosenfeld
8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Southwest Research Institute
3600 Yoakum Blvd.
Houston, Texas 77006
10. PROGRAM ELEMENT NO.
1AA601
11. CONTRACT/GRANT NO.
68-02-2227
12. SPONSORING AGENCY NAME AND ADDRESS
Health Effects Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
RTF, NC
14. SPONSORING AGENCY CODE
EPA 600/11
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This study investigated the absorption of lead by persons of different age-sex
groups exposed to automobile emissions of lead at traffic densities from less than
1,000 cars per day to 25,000 cars per day. The relationships between traffic
density and lead in various environmental samples were also examined. A house-to-
house survey based on a strict set of selection criteria was used to recruit study
participants. At each house a series of environmental measurements were taken:
traffic volume, tap water, paint-interior and exterior, housedust and window sill
wipes. Two blood samples were taken a week apart. In the range of traffic
exposures studied no relationship with blood lead levels was observed (maximum mean
air lead < 2.0 yg/m3). A positive relationship between smoking and blood lead
levels was found for both males and females. This relationship was statistically
significant for females but not for males.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Held/Group
lead
blood analysis
toxicity
automobiles
vehicular traffic
epidemiology
environmental surveys
Dallas
Texas
06 F, T
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (This Report)
UNCLAj
3. SECURiT
SSIFIED
20. SECURITY CLASS (Thispage)
UNCLASSIFIED
21. NO. OF PAGES
378
22. PRICE
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION is OBSOLETE
366
------- |