EPA-R1-73-005



December 1972
Environmental  Health  Effects  Research  Series

                                         v.*.•.•.• *^^v^£x'X*!%v* •«ff^*!%*!v**.'K**v%%*»*!%.v •*•" •*«*"••••••"•*" v!*>%*?!. v%*«%"«R*i"X*i%*i.»!«•!. v%.x v










-------

-------
                                            EPA-R1-73-005
                 A SURVEY
                   OF  AIR
            AND  POPULATION
                LEAD  LEVELS
                      IN
SELECTED AMERICAN COMMUNITIES
                      by

           Lloyd B . Tepper and Linda S . Levin

           Department of Environmental Health
                Kettering Laboratory
               University of Cincinnati
                College of Medicine
                Cincinnati,  Ohio 45219

               Contract No.  PH 22-68-28
              Program Element No. 1H1099
        EPA Project Officer: Dr. Robert J. M. Horton

                Office of the Director
          National Environmental Research Center
        Research Triangle Park, North Carolina 27711
                   Prepared for

          OFFICE OF RESEARCH AND MONITORING
        U.S. ENVIRONMENTAL PROTECTION AGENCY
              WASHINGTON, D.C. 20460

                  December 1972

-------
This report has been reviewed by the Environmental Protection Agency and




approved for publication.  Approval does not signify that the contents




necessarily reflect the views and policies of the Agency, nor does




mention of  trade names or commercial products constitute endorsement




or recommendation for use.
The investigations reported here were  supported jointly by  the




Environmental  Protection Agency (Contract No. PH 22-68-28),  the




American Petroleum Institute, and the International Lead  Zinc  Research




Organization, Inc.

-------
                          TABLE OF CONTENTS
LIST OF FIGURES	    iv
LIST OF TABLES	    iv
SUMMARY 	     1
PREFACE 	     2
PURPOSE 	     2
PARTICIPATING ORGANIZATIONS AND ACKNOWLEDGEMENTS  	     3
METHODS AND PROCEDURES  	     3
   Aerometric Sampling  	     4
      Location of Sampling Sites  	     4
      Sampling Equipment  	     4
      Sampling Methods  	     4
      Analytical  Methods  	     4
   Meteorological  Considerations  	    11
   Biological Studies 	    12
      Population Selection  	    12
      Population Lead Levels  	    21
      Estimation of Alimentary Lead Intake  	    22
OBSERVATIONS AND DISCUSSION 	    23
   Aerometric Data  	    23
   Dlood Lead Concentrations	    31
   Urban-Suburban Blood Lead Levels 	    32
   Male-Female Clood Lead Differentials 	    60
   Fecal Lead Levels	    61
   Results of Analysis of Pennypack Data  	    ^
   Results of Analysis of Alpine County Data  	    64
   Results of Analysis of Policemen Data	    65
REFERENCES	    67
APPENDIX 1  	    68
APPENDIX 2	    69
APPENDIX 3	    71
APPENDIX 4	    72
APPENDIX 5	    73

-------
                             LIST  OF  FIGURES
Figure

   1
   2
   3
   4
   5
   6
   7
   8
   9
  10
  11
  12
  13
  14
  15
  16
  17
  18
  19
  20
Sampling Sites
Sampling Sites
Sampling Sites
Sampling Sites
Sampling Sites
Sampling Sites
Sampling Sites
Sampling Sites
Meteorological
Meteorological
Meteorological
Meteorological
Meteorological
Meteorological
Meteorological
  Los Angeles ,
  Philadelphia
  Cincinnati
  Los Alamos   .
  Houston .  . ,
  Chicago .  . ,
  Washington   ,
  New York  . .
Factors
Factors
Factors
Factors
Factors
Factors
Factors
                         Cincinnati
                         Los Angeles  .
                         Philadelphia
                         New York  .  ,
                         Chicago .  .  .
                         Houston .  .  ,
                         Washington  	
Lead-in-Air Data (1959-71)  National Air Sampling  Network
     (Symbols Indicate 1961-62 and 1968-69 Population and
     Air Lead Level  Studies.)   	
Atmospheric Lead Levels by Month for Each Site 	
Distribution of Blood Lead Concentrations  	
Mean blood Lead Levels and n5% Confidence Limits  by Age
Blood Lead Levels and Corresponding Mean Air Lead Levels
                                         (5)
Page

   6
   6
   7
   7
   8
   8
   9
   9
  13
  14
  15
  16
  17
  18
  19
                                                               31
                                                               33
                                                               43
                                                               52
                                                               60
                             LIST OF TABLES

Table
   1    Comparison of Analytical  Results from Duplicate Analyses
             of Wedges from a Single Filter  	
   2    Population Groups Sampled  	
   3    Monthly Mean Concentrations of Lead in Los Angeles,  Phila-
             delphia, Cincinnati, and Los Alamos,  1968-69  	
   4    Yearly Means and Confidence Intervals of Air Pb Concentra-
             tions for Each Station  	
   5    Annual Means of Monthly Average Concentrations  of Lead and
             Particulate Matter  	
   6    Freouency Tables Showing Distribution of Clood  Pb Levels  in
             Okeana, Philadelphia, and Pasadens  	
   7    Geometric Means and Confidence Intervals of Dlood Lead
             Levels Based on a Single Determination from Each
             Subject 	
   8    Analyses of Functional Relationship Between Blood Lead and
             Age for Non-Smokers 	
   9    Analysis of Variation Due to Smoking Status (Females) and
             Location  	
 lOa    Relationship Between Mean Blood Lead Levels and Degree of
             Urbanization in Paired Locations  	
                                                             Page

                                                               10
                                                               20

                                                               24

                                                               27

                                                               29

                                                               45


                                                               48

                                                               50

                                                               54

                                                               55
                                   IV

-------
Table                                                               Page


 lOb    F-Ratios Testing the Difference Between Urban and Suburban
             Lead Levels in the Total  Sample (Including Previous
             Smokers)	    55
  11    Geometric Means of Blood Lead  Levels by Location and Smok-
             ing Status	    56
  12    Further Analyses of Relationship Between Blood Lead and
             Smoking Status 	    56
  13    Geometric Means of Blood Lead  Levels for Los Alamos Males
             and Females by Smoking Status  	    57
  14    Air and Blood Lead Concentrations (1968-71)	    58
  15    Air and Blood Lead Concentrations (1968-71)  (lion-Smokers
             Only)  	    59
  16    Freouency Tables Showing Distribution of Hematocrit Values.    62
  17    Analyses of Functional  Relationship Between  Blood Lead and
             Hematocrit for Non-Smokers 	    63

-------

-------
                                       1

                                    SUMMARY


     The concentration of lead in the ambient atmosphere was determined at 59
sampling sites in eight American communities during the period 1968-71.  Nineteen
sampling sites had existed in a similar survey in 1961-62.   At 14 of these sites
the lead-in-air value was found to be higher in the current study than in 1961-62.
The observed annual mean atmospheric concentration of lead  varied from 0.14,ug/m3
(Los Alamos) to 4.55 jug/m^ (Downtown Los Angeles).  Higher  lead values were
associated with urbanization.

     The concentration of lead in the blood of specific well-defined populations
was determined.  Such populations lived in geographic proximity to specific air
sampling sites.  In the three metropolitan areas from which both urban and sub-
urban population groups had been obtained, the mean blood lead levels were signifi-
cantly higher in the former.   In two of the three areas the blood lead level  was
higher in urban smokers and non-smokers than in the corresponding suburban pop-
ulations, classifi-d by smoking habits.  At each location the concentration of
lead in the blood of smokers  was greater than that in the blood of non-smokers.
The magnitude of the observed urban-suburban difference (for populations com-
parable in smoking habits) ranged from 0.9,ug/100 gms to 4.5 jug/100 gms.  It is
probable that these observations partially reflect lead absorption from ambient
atmospheres differing in lead concentration.  There was no  significant concord-
ance between the ranking by site of mean air lead levels and that of the mean
blood lead levels prevalent in the related populations.

     The observation that urban levels of blood lead are higher than suburban
levels, but that air concentrations of lead are not clearly reflected in blood
lead levels on a general national basis, suggests that factors other than the
atmospheric lead level are of relatively greater importance in determining the
blood lead levels in population groups.

     No relationship was established between age of participant and the blood
lead level.  In husband-wife  pairs, presumably exposed to similar diets and
atmospheres, the males had significantly higher blood lead  levels than did the
females.  This difference could not be attributed to smoking habits or to
hematocrit levels.   The possibility that the difference was due to dissimilar
quantities of foods in the respective diets of men and women was not examined.

     Studies of dietary lead  levels showed them to be generally lower than com-
monly reported in the literature, lOO^ug/day being a closer approximation in
the population studies than the widely quoted 300

-------
                                      2

                                   PREFACE
     In 1958, the Surgeon General of the Public Health Service appointed a
committee to provide advice relative to the health implications of the  use
of lead anti-knock compounds in gasoline formulations.  The committee concluded
that an increase in the tetraethyl  lead content of gasoline from 3.0 to 4.0
milliliters per gallon would not pose a hazard to the public health from air
pollution.  It was recommended that studies be conducted to provide more defin-
itive data on levels and trends of atmospheric lead levels in selected  urban
locations and on the blood levels of lead in selected population groups'.

     In response to these recommendations the Surgeon General formed a  Work-
ing Group on Lead Contamination to develop and execute a suitable program of
studies.  This Group included representatives of the petroleum, automotive,
and fuel additive industries, the Public Health Service, the California State
Health Department, and the University of Cincinnati.  With the guidance and
support of the Working Group air was sampled over a 12-month period at  a num-
ber of locations in Cincinnati, Los Angeles, and Philadelphia.  Blood and urine
samples were obtained from selected population groups in these cities.

     The findings based upon these extensive observations were reported in
1965 in an official Public Health Service document:  "Survey of Lead in the
Atmosphere of Three Urban Communities", Publication No.  999-AP-12.2  This
report presented average, maximum,  seasonal, and diurnal  atmospheric lead
data.  These showed, among other things, that the higher lead levels were
associated with areas of high vehicular density, the fall and winter months,
and early morning peak traffic.  The study of lead in the blood showed  a
tendency toward higher concentrations in groups of persons as they varied from
rural to center-city areas in their places of residence  and work.   A number
of factors appeared to be related to these observations:   geographical  region,
tobacco smoking, occupational exposures.  The Working Group also evaluated  the
influence of inter!aboratory variation, since the required assays  were  performed
by several collaborating organizations.

     It was a primary purpose of the investigation to establish a  baseline for
the lead content of the atmosphere and in selected population groups.   Those
responsible for the work anticipated a subsequent study, which would reflect
changes in these lead levels which might be attributed to various  relevant
factors:  vehicle density, lead content of fuels, mechanical  condition  of ve-
hicles, local  climatology, and population patterns.

     The work reported here summarizes studies conducted in 1968-72 to  examine
those changes, if any, which had occurred during the interval since 1962.  The
investigations were conducted by the Department of Environmental  Health, Ket-
tering Laboratory, University of Cincinnati  with the financial support  of the
Environmental  Protection Agency; the American Petroleum  Institute; and  the  In-
ternational  Lead Zinc Research Organization, Inc.  The course of the study was
guided by the Subcommittee for the Surveillance of Air and Population Lead
Levels, which reflects the public health interests of government and industry.
The Subcommittee is a working unit of the Lead Liaison Committee,  an advisory
group to the Environmental  Protection Agency.

                                   PURPOSE

     The purpose of the investigative program reported here was to:

     1)  Determine whether or not a change in the concentration of airborne
         lead had occurred after an interval of 8 years, at a series of refer-
         ence locations designated in 1961-62.

-------
      2)   Determine the ambient atmospheric levels of lead at selected additional
          sites in the cities studied in 1961-62 and in other communities.

      3)   Examine the extent to which levels of lead in the blood of selected
          population qroups reflect exposure to lead at various levels in com-
          munity atmospheres.

      4)   Determine whether or not a change in the blood levels of lead had oc-
          curred after an interval of 8 years in selected population groups.


              PARTICIPATING ORGANIZATIONS AND ACKNOWLEDGEMENTS

      The  Subcommittee for the Surveillance of Air and Population Lead Levels,
which provided guidance during the study, included representatives from the
American  Petroleum Institute, Automobile Manufacturers Association, E.I. du  Pont
de Nemours and Company, Ethyl Corporation, International  Lead Zinc Research  Or-
ganization, and the Environmental Protection Agency.   Names of individual  repre-
sentatives appear in the appendix.  Environmental  studies were conducted with
the cooperation of local organizations in each area:   Chicago Department of
Air Pollution Control, District of Columbia Department of Public Health, Houston
Department of Public Health, Los Alamos Scientific Laboratory, Los Angeles
County Air Pollution Control District, New York City  Department of Air Resources,
Philadelphia Health Department, and City of Vernon Health Department.  Private
citizens, commercial  establishments, and public institutions generously permitted
sampling  stations to be installed and operated on their property.

     The  collection of biological materials was made  possible through the  coop-
eration of church and public service organizations.

     The  design and execution of this study depended  upon scientific and tech-
nical  competence in many areas.  Names of principal  participants appear in the
appendix.  This study is based upon their individual  and collective contribu-
tions.

      Financial  support for the investigations reported here was provided  jointly
by the American Petroleum Institute, the International  Lead Zinc Research  Or-
ganization, and the Environmental Protection Agency  (Contract PH 22-68-28).


                           METHODS AND PROCEDURES

     The  program of aerometric and biological  studies  was conducted in eight
regions reflecting  a  range of geographic and climatological  characteristics.
The selection of Cincinnati, Philadelphia, and Los Angeles was based on the
fact that extensive sampling had been conducted in these  areas during a 12-
month  period in  1961-62, and that comparisons  could now be made between cur-
rent observations  and those obtained some 8-9  years earlier.   Investigations
of a similar nature were included in the current study in the additional  areas
of New York City,  Metropolitan Washington (D.C.),  Greater Chicago, Houston,
and Los Alamos,  New Mexico.   New York was selected because of its  high vehicle

-------
and population density and Washington because of its  lack  of  industrialization.
Chicago and Houston were added as  examples  of large  "northern"  and  "southern"
cities, respectively.   Los Alamos  represented an area with a  well defined  popu-
lation of over 15,000 persons with minimal  exposure  to  atmospheric  metals.


                             Aerometric Sampling

Location of Sampling Sites

     To permit an accurate appraisal  of changes  which may  have  occurred  since
1961-62 in atmospheric levels of lead,  stations  which existed in  the  study  con-
ducted at that time were duplicated in  the  current investigation.   This  proce-
dure was followed primarily for comparison  purposes;  the suitability  or  epi-
demiological significance of these individual  sites  was not reconsidered.   A
comprehensive photographic record  of the instrument  installations in  1961-62
permitted accurate duplication of  station position and  orientation.   Certain
limited changes were made in the distribution of stations  in  the  current study
as opposed to their distribution in 1961-62:

     Cincinnati:   Two stations (#34,35) approximately 2 miles apart and  24  miles
northwest of Downtown Cincinnati were established on  farms in the town of  Okeana,
Ohio, near the Indiana line.  The  purpose of  these stations was to  provide  a
rural "background" level for atmospheric lead in a locality with  a  population
available for sampling.

     Philadelphia:  The station at the  Philadelphia  Airport (#10) was omitted
because of small  biological  significance and  major changes in the composition
of aviation fuels since 1962.  Station  #13  at 15th and  Market was moved  one
block east because of construction at the original site.   An  additional  instru-
ment (#18) was installed on the 3rd floor balcony of an apartment on  Ritten-
house Square so as to record representative exposure  of a  center-city population.
Two additional stations (#19,20) were installed  in the  suburban area  of  Ardmore-
Wynnewood, where a population was  to be sampled.

     Los Angeles:  Station #2 was  moved approximately 1/4  mile  north  because of
operational problems at the original  location.

     In the 5 regions where long-term sampling of this  type had not been con-
ducted previously, sites were selected  on the basis  of  relevance  to population
exposure, accessibility, and the area to be covered.  In communities  in  which
biological sampling was to be conducted, emphasis was placed  on selecting  an
instrument location which would reflect, insofar as  possible, the usual  human
exposure in that general neighborhood.   In  some  locations, elevations of the
sampling instruments did not approximate those of the breathing zones of the
study population.  For particulates of respirable size, however,  the  evidence
is that differences in elevation,  such  as existed in  this  study,  influence  air
lead concentrations only insofar as such elevations  reflect distance  from  the
lead source.

     It is essential to emphasize  that all  stations,  although representing  cer-
tain types of neighborhoods, elevations, and  proximities to major roadways,
were selected arbitrarily.  Data from a specific site have meaning  for that
site only, and pooling or "averaging" for an  entire  city has  limited  validity.

-------
Alternative sites could have been selected to yield other "averages" for each
city.  It would therefore not be judicious to make unqualified inference from
the findings of a number of stations in a city or other area to that city or
area; for example, the findings from eight selected locations in Los Angeles
cannot be generalized to the "situation in Los Angeles".  Pooled data for a
city nay have validity in certain kinds of comparisons over a period of time
for the specific sites included in the pool.  These data, however, cannot be
precisely compared with pools based upon other combinations of sites arbitrarily
selected.

     The location of sampling sites in the various regions is shown in Figures
1-8.

Sampling Equipment

     The equipment utilized in this study has been described'^).   in essence
it consists of a filter holder for 106-mm Millipore filters (Type WS),  a vane
pump and motor unit drawing a nominal  free air volume of 5.9 cfm, a cooling fan,
and suitable electrical and air connections.  The housing is constructed of
plywood and is designed to permit necessary ventilation of the equipment while
attenuating objectionable pump noise.   Initial and final air flows through each
filter are determined from orifice meter measurements.

Sampling Methods

     Sampling units were operated over a 12-month period at each  site.   Filters
were changed in most locations on Monday, Wednesday,  and Friday.   Because of
previous experience in Los Angeles with filter obstruction in the course of 48-
hour weekday exposures, tandem units with automatic switch-over  permitted the
collection of 24-hour samples in parts of the City with relatively high  traffic
density or industrialization.  No problems of filter  obstruction  were encountered.

     Filters were numbered,  conditioned in a controlled atmosphere, and  weighed
in Cincinnati prior to exposure.   Shipments of filters, each in  a plastic Detri
dish, were made to monitors  in each city.  Monitors recorded on  each dish ori-
fice meter measurements before and after exposure, the duration  of each  exposure,
and notes (if any) about station operation.   Exposed  filters were returned by
mail  to Cincinnati in their  original  dishes.   Upon receipt,  the  filters  were
again conditioned and weighed in the determination of total  particulate  weight.
Calibration curves for each  orifice meter permitted the determination of volume
of air which had passed through each filter.   A glass  template was used  to divide
each filter into ten equal wedges suitable for analysis.

Analytical  Methods

     Lead contained in particulate matter on the filter wedges was determined
in 1961-62 by the dithizone  colorimetric technique.^)  The  three laboratories
cooperating in the study at  that time  modified the method slightly to suit the
needs  of each.   Interlaboratory comparisons  were conducted to standardize the
analytical  bases upon which  inferences  were drawn.

     The developments in atomic absorption spectrometry (AAS)  since 1961-62  in-
dicated that greater efficiency in filter analysis might be  achieved in  the


                                                Text continued on  page 10

-------
(N

 OJ


 a
•H
fe
      0)
     •a
      ra
     ,—1
     •H
      fcfi
      s
     •H
     t—(
      o,
      E
      ra
                                          &
                                          a
                                                        a
                                                       •a
                                                        c
                                                                           m
                                                                           o
                                                                           c
                                                                           ci
                                                                           •a
                                                                 €
                                                                 h
                                                                 01
                                                                                  m
                                                                  C
                                                                  0
                                                                 •D
                                                                                           I
                                                                                           4J
                                                                                           H
                                                                                           a
                                                                                           a.
       m
       o
       c
       u
      •a
01
3
C
01
>
<

•a
o


18

IK
0
.p
18

0
,Q

J
iH
O


C
o
0
I
01
fi
0
SB

O
18
a.

c
E:
01
0.




c
o
•H
4J
18
O
o

M
<
H
o.
i

•o
18
O
ci
f7

HI
fl

(Q
^
01
c

18


1
10

18
>
U

4J
C
18

&

4J
C
1

18
01

H
10
g-
in

01
to
3
•g


•p
•a
OS
                                                                                                  o
                                                                                                  .-I
                                                                                            C
                                                                                            01
                                                                                           •a
 c
 01
•o
                                                                                                                                  O
                                                                                                                                  z
     w
     0)
     rH
     41
     bo
     c
 2    '
 3   tn
 tuo  w
                                                                           CJ  Ifl
                                                                          s  o
                                                                                     &
                                                                                    D
                                                                                           111
                                                                                           p:
          O

          f.
          4-1
          U
          O
          c

          (U
          rH

          1   •
              91

          \  01

             4->
          >,

          QJ  d)

          n)  c

          •H ^:


          2^
          D,  O
          &
          (0  (U
              (A

-------
   10
 .3
    O.

    CO
   w
                           01
                           0
   ra
   O
8
o

I
01
f-l CO
   -p
                          IT)
                          i-l
                               &
                               ID
                                    &
                                    o
    a
   •H
    O
    C
   •H
 • u
PI


2  1

U) -P
•H -H
fe co

    bfi
    a


   I
                               £

                                I
                                     8
                                    •o
                                o
                                CJ
                                         !
                   I
                   1

-------
     o
     bJJ
     OJ
     O
<£>  tn
     0)
 0)  -P
•H   bfi
fa   C
      P.


      TO
     W
                                              •3
                                              o
                                              f
                                              u
                                              CO
                                              id
                                              P.
                                              HI
              8
             a

       8      §
       •g
       CO

       01
ndb
&     3
                                                            B
                                                            11
                                                            CO
              &
                                                           •H
                                                            H
                                                           0)
O*

X

C
                                                                               •g
                                                                               co
                                                                                a
                                                                                K
•s
 id
             •g
             10

             ft
              tn
                                        I
                                        •3
                     m


                     a
Jj
B
11
•O
•H
0]
U
a


tj
o
X
•r4
£
•H
4J
B
U
•O
•^
m
K


"8
X
•H
£
id -i
"4 J->
U B
Vl V
oi 10
§ ~4
i (0
0 V
0 *
•H
4i
B
11
•O
^-1
n
V
K


•§
5
£
•H
4J
S
1

«
                                       o
                                       VO
     o
    •p
     to
in   i

 0   in
•H CQ
     a

     ca
     ra
U
i
.u
h
id
a.
o
Q

11
U
•H
&•

1

•0

D:

B
HI
•H
rH
3
O



,B
4J
i-l
id
1)
S

1

X
^1
id
(X
C
c



£














4J
fi

+J
M
Q
tx
S

X
a

,13


V
n
•H
b
*
o

i


o
4J


Q















•-I
id
n
•a
0)
f.
4J
id
U





E
•3

O
U
4-1
CO

10 1J
id £
rp u
U Vi
4J id

co a
sidence
v
a
u
4J
id
>
•H
U
O,



u
>
v4
VI
Q

O


4J
id
                                        8
                                        •D
                                   V
                                  •a
                  I

-------
 00   I

 a)   ra
 f-i   a>

 u> -H
 •H  W
 PN
      tJO
      c
      a
      E
      co
     w



rH
O
o
£
u
CO

O1
•H
S

g
0
Ll
01

1

i
03
&
O
HJ
10
Ll
0
.£3
n


V
11
Ll



4J
Ul
rH
rH
1

g
11
rH
X


rH
__
U
Ul

rC
D>
•H
E

C
O
W
o
^
1

01
c
(U
01
o


at
S1

rH
0
C

rH



to
1
C

rH
!
(0


c
Ll
O

Li
0
B
1

0
O>
n
H
•H
* &
U ID
•H 01
S C
U tt
11 Ul
O Q

rH
8
o
Ul

o
c
S

U)

1

I
H
III
H

C
O
V
id
U]
&
o
Ul
u
3
i!
O

c
11
H
rC
O
£

1

I
rH
II)
H

C
<0
•p
n)
U)
Schoo.'

g
rH
id
u

,C
4J
g
ul

1

c
o
£
^
n
id

£
ft


|

CO


rH
n
Q

i



o>
2
111
rd

£
0,
                                    •8
                                    X
      •o

       X
•H    ^H
S    £
                                                 U)
                                                 OJ
                                                 a
 C

•O     -O

 in     X
 O     -rl
a     £
              c
              (U
             •o
             •rl
              U)
              HI
             a.
                                                                               S
                                                                              •a
                                                 c
                                                 01
                                                 •o
     C
     o
     4-1
     hD
     C
     •H
     x:
     en
     CO
     Sin
     o>
•H  +->
     M
     ft
     E
     CO
    W
-p
c
i
4J
Ll
rt
a
o
Q



rH
ITJ
11


1

C
o


c
-H
Ll
<

rH
ID
a
•H
O
•H
C
3
£

1

11
3
C
11


CTi
a c
C -H

•H rH
•a -H
C 3
M m
4J
m
4>
Ul
a
5


i

V
3
C
1
to
u
id


u

I

X
?;
IT)



0)
rH
"8
0
s















c
id
rH
Ll
V
N
4-1
•rH
w
id
lie
+J
+J
11
Ll
rH
Id
"71





o

3
O
K

id
•H
c
•H
Ll
TH

Q
"c
u
D

rH
•rH
£
o

1

TJ
10
o
K

rH
rH
-H
X

Ll

•H



I

fO
*o
CO
0)
•p
m
                                    c
                                    n
                                   •a
 C
 1>
•c
                                              c
                                              o
                                             •a
                                                                                            n
                                                                                            V
                                                                                           ce
                      21 ol
                      Si*1

-------
                                      10
current study with this method.   Consequently  this  procedure was  adopted with
special attention to an examination of the  relationship  between analytical  re-
sults obtained by these two methods:  dithizone and AAS.   Since a primary pur-
pose of the entire study was the determination of changes  in air  lead  concen-
trations over a time interval,  it was  essential  to  assure  that demonstrated
differences were not due to variations in technique.

     Methodological  checks were  of two types:   1)  assays of wedges  from the
same filter by dithizone and AAS; and  2) assays of retained wedges  from the
1961-62 study by current analytical techniques (AAS).^   In each type of evalu-
ation it is apparent that wedges from  the same filter do not contain the same
amount of lead; i.e., there is  inter-wedge  variation.  The magnitude of inter-
wedge variation was  examined as  a part of the  consideration of variation between
methods.  It was determined that the precision with which  the quantity of lead
on a filter may be inferred by  the analysis of one  10% wedge is inconsistent
from one filter to the next.  Presumably this  inconsistency reflects to some
extent particle size and influence on  some  filters  of deposited heavy  particles
which appear on only a few segments of the  filter.   In the estimation  of lead
on a filter from one 10% wedge,  the 95% confidence  limits  may be  as narrow  as
97-103% of the mean  for the entire disc or  as  wide  as 67-150%, depending upon
whether one uses the lower or upper estimate of interwedge variance.   In a
general pool of analyses, however, it  was determined that  the 95% confidence
limits for the lead  content of  a filter disc as estimated  from a  10% segment
is 85 to 120% of the mean.  For  most stations  approximately 150 10% wedges
were collected over  the course of 12 months; the site average is  based upon
150 analyses.

     The relative response of the atomic absorption method as compared with
that based upon dithizone colorimetry  was found to  vary  between  0.92  and 0.98.
This observation approximates the 0.95 average value which had been obtained
from lead determinations on filters artificially spiked with lead (95% confi-
dence limits 0.93-0.97).

     To construct a  direct relationship between assays in  1962 and  in  1969-71,
retained wedges of filters which had been exposed,  cut,  and assayed at the
earlier date were examined in the current study.   There  is of course variation
between wedges from  the same filter.  A variation between  1962 and  1969-71
methodologies exists in addition to the interwedge  variation.  Table 1 sum-
marizes a comparison of analyses of single  10% wedges in 1969 with  single 10%
wedges from the same filters in  1962.

          Table 1.  Comparison of Analytical Results From  Duplicate
                       Analyses  of Wedges From a Single  Filter.
City
Av. ratio (%)
1969/1962
No. of
filters
No. filters
with ratio
< 1
No filters
with ratio
> 1
z-Value
Los Angeles         103.47          39          16              19          1.28
Cincinnati            90.86          40          30               6         -6.29
Philadelphia         99.61          40          18              20         -0.21

-------
                                      11
 The  magnitude of  the  z-value reflects the degree of difference between the num-
 ber  of filters with ratios less than and more than one.  The 5% critical  value
 is _ 2.68.   It is  therefore concluded that the 1969 air lead concentration deter-
 minations  in Cincinnati are systematically significantly lower than those obtained
 in 1962 (z  = -6.29),  the average ratio being 90.86%.  Interpretation of the air
 lead concentrations at the Cincinnati sites in 1969 must give this observation
 appropriate weight.   For Los Angeles and Philadelphia no significant differences
 were found  between the 1962 and 1969 measurements.


                        Meteorological Considerations

      The previous study (1961-62) demonstrated that storm conditions with strong
 winds  were  associated with low atmospheric lead levels; precipitation alone pro-
 duced  no significant effect nor did flooding, snow cover, or high space heating
 demands on  cold days.

     To determine whether or not unusual meteorological conditions might  have
 influenced  the observations reported here on the atmospheric levels of lead, a
 brief  examination was made of the representativeness of the sampling period.
 Significant deviations from conditions recorded in 1961-62 might have important
 influences  upon the conclusions to be drawn from the current study.

     Meteorological data obtained at the Greater Cincinnati Airport are summar-
 ized in Figure 9.  Precipitation during the sampling period was somewhat  below
 normal, especially in late winter.  Heating requirements were slightly less than
 normal.  Average wind speeds were near normal and higher in the winter than in
 the summer  months.  The overall situation appears to approximate that which ex-
 isted  in 1961-62 and is not inconsistent with the 30-year average.

     Meteorological data obtained at the Los Angeles International Airport are
 summarized  in Figure 10.   January precipitation totalled 9.6 inches, a signifi-
 cant departure from normal.  Average monthly wind speed for the year was  con-
 sistently higher than the 30-year average.   Average wind speed in 1968-69 was
 higher than in 1961-62 except in the month  of May.   Advisory statements of high
 air pollution potential for Los Angeles  were issued for May 10-11  and November
 3-5,  1969.   An increase in total  particulate matter and atmospheric lead  levels
 at these times was observed at most sites.   In 1961-62 the relatively high con-
 centrations of atmospheric lead in the fall  and winter were attributed in part
 to relatively low wind speeds  during these  seasons.   For 1968-69,  atmospheric
 lead  levels during these months was again relatively high as compared to  those
 recorded in spring and summer months,  the differential  being somewhat greater
 than  observed in   the previous  study.   The  average  monthly wind speed in  1968-
 69, however, varied less  than  in  the previous study.

     With respect to Philadelphia,  precipitation  was  generally below normal  and
 the 1961-62 level  except for the  very  wet summer  in  1969.   Heat requirements
 for the two test  periods  and the  30-year average  were consistent.   Wind speed
 in the winter of  1968-69  was  significantly  greater  than the 30-year average;  in  '
 1961-62 wind speed had been less  than  average.   Inspection of atmospheric lead
concentrations  as  recorded 1961-62 and  1968-69 shows  that lower levels  or rela-
 tively small increases were recorded during  these relatively windy winter months
of 1968-69, as  compared with  the  corresponding still  months of 1961-62.   An  .

-------
                                      12
advisory of high air pollution potential in the Philadelphia area for September
23-24, 1969, was not associated with unusually high levels of atmospheric lead,
although total particulate matter increased at all  sites.

     Corresponding meteorological data for Los Alamos are  not published;  however
observations during the test year do not indicate major departures from normal.

     Meteorological data recorded at Central  Park in New York City indicate that
precipitation was approximately half of normal during the  summer of 1970.  Heat-
ing requirements approximated normal levels except for November and December,
which were cold.  The average wind speed for the test period was slightly less
than normal except in December, 1970.

     Precipitation recorded in Chicago during 1970-71 was  generally well  in ex-
cess of normal.  Heating requirements were somewhat below  normal for the  year.
Wind speeds were slightly above the 30-year average in March and October, 1970.

     Corresponding data for Houston show that 14.39 inches of rain fell  in May,
1970, more than three times normal; positive and negative  departures from normal
occurred throughout the year.  Heating requirements approximated the 30-year
average.  Throughout the year the average monthly wind speed was less than nor-
mal by approximately 20-30%.

     Meteorological data obtained at Washington National Airport are summarized
in Figure 15.  Precipitation for the year approximated the normal  rainfall.
High air pollution potential advisories on April 29-30, June 9-12, July 28, and
August 15-16, 1970, were not associated with significant elevations in the air
concentrations of lead.  Heating requirements and wind speeds approximated the
30-year mean.


                             Biological Studies

Population Selection

     A principal objective of this investigation was an evaluation of the rela-
tionship between atmospheric lead levels and the concentrations of lead in the
blood of persons exposed to the atmospheres in question.   While a number  of dif-
ferent population groups have been studied in attempts to  establish this  rela-
tionship, multiple uncontrolled variables associated with  these groups have cor-
fused the appraisal of the actual exposure and the construction of the correct
inference.  In the absence of personal  monitoring devices  one cannot establish
the total integrated exposure over months or years  of police officers, garage
mechanics, "drivers", "commuters", aircraft workers, or other similar groups.
Occupational and wide-ranging geographical factors  are not reflected in air
sampling at fixed locations not clearly related to the population in question.

     For these reasons particular attention was focused in this study on  the
definition of populations with a specific consistent relationship to known air
levels of lead.  Populations sampled in this study were derived from women volun-
teers living within a prescribed region surrounding an air sampling instrument.
Women, in general, currently tend to spend a greater proportion of their  time
at home or in the neighborhood of their home than do men.   Furthermore, women
who are employed at a distance from their homes rarely work in lead-using trades.
For inclusion in the study a continuous residence time of  not less than 5 years


                                                Text continued on  page 20

-------
                                         13
                                                                  W
                                                                  SH
                                                                  O
                                                                  •P
                                                                  O
                                                                  I—I CO
                                                                9> ra d
                                                                <-i o d
                                                                3 -H -H
                                                                & M> o
                                                                •H 0 d
                                                                fe rH -H
                                                                  O O
                                                                  fn
                                                                  O
                                                                  01
                   ro

                  S3HONI
Q
LU
LJ
Q.
CO
Q  L
LJ
CD
<
cr
LU
    yduu 'O33dS QNIM 39VU3AV
                                                  to
                                                  h-
                                                  z
                                                  LU
                                                  S
                                                  LU
                                                  O
                                                  LU
                                                  tr
                                                  <
                                                  LU
                                                      O O O O O O O
                                                      o o o o o o o
                                                      — o o> oo r- (£> in
                                                                        o o o o
                                                                        o o o o
                                                                        
-------
o
LU
a:
o_

                    1   '  '

         ill
                             — O
                S3HONI
                                            14
   ta
   ^
   o
   +j
   u
   OJ W
O  (it 
           •*T-xi	1	1	r
                                il
                                      V)
                                 —  cP
                   ONIM 39W3AV

-------
                                                15
o
LU
o:
Q_
                                                                              w
                                                                              fn
                                                                              O
   CO 03
i-H  fe -H
rH    J3
   r-H C,
0)  CO i-l
?H  O 0)
3  -H T3
M  be oj
•H  O rH
6n  i-H -H
   O JZ
   h &
   O
   ooi^-tDio^t   rooj

                     S3HONI
                                                                UJ    f-
                                                                ^
                                                                UJ
                                                                £

                                                                o
                                                                UJ
                                                                o:
                                                                <
                                                                LJ
                                                                         V
                                                                      i—V
                                                                     o   o   o  o   o   o
                                                                     o   o   o  o   o   o
                                                                     OJ   O   O)  tD   ^-   OJ
                                                                           SXH7 33U93O

-------
                                       16
                                                                   o
                                                                   •M
                                                                   O
                                                                 cj w o
                                                                 fc CJ ><

                                                                 SB ho s
                                                                 •H O 0)
                                                                   O
                                                                   0)
                if)  co  —

            S3H3NI
Q
LU
UJ
CL
CO

Q
z

?


UJ
(T

UJ
       I

^
*
                             CO
     if  CJ


  VOfc/ '
-------
                                                   17
          ~ii    r
O
UJ
a:
CL
                                          Q



                                          2



                                          O
   tn
   SH

   O
   +J

   y
   Ct!
<1> ca  aj
5n O  y
  S-H  -H
  bfi  rj

•H O  o
                                                                               O
                                                                               0)
                                                                               -p

                                                                               I
                           rocvj   —
                                     H<
                                                                LU
                                                                ^
                                                                Z)
                                                                o
                                                                UJ
                                                                a:

                                                                CD
                                                                ui
                                                                T.
                          39VU3AV
                                                                     o  o   o  o   o   o   o
                                                                     o  o   o  o   o   o   o
                                                                     ro  —   CD  h-   10   ro   —
                                                                         SAVQ

-------
 o
 LJ
 or
 Q.

                                                18
              1 - 1 - 1
                      _J	i  .X _\_
                                           O


                                           CO
             to
             i-(
             o
             JJ
             o
             0!
          
                                                                             0)
                   S3H3NI
Q
LJ
LJ
CL
CO
LJ
O
<
(T
LJ
LJ
2
LJ
cr


o
LJ
(£.
                                                                LJ
                                                                I
             'O33dS  ONIM 39VU3AV
    o    o   o   o  o
    o    o   o   o  o
    if)    *fr   ro   CVJ  —

        SAW 33U93O

-------
                                        19
 O
 O
 LU
 o:
 CL
           ~ii
      ^  i
         I   I  I   I
        00 h-  
                                                                  CO W)
                                                                   10
                                                                   CO
                                                                 o
                                                                 a;
Q
LJ
LU
Q_
(O

Q
UJ
CC
LJ
                             (f)
LfdUU '
          CVJ O 00  CD ^ CJ

                      39VU3AV
                                              i-
                                              UJ
                                              5
                                              ID
                                              O
                                              Ld
                                              cc
                                              h-
                                              <
                                              LJ
                                              X
                                                                  33U93CI

-------
                                      20
within this area was required.   The area was defined either by political  bound-
aries ("Pasadena", "Port Washington")  or by the circumference of a  circle one
mile or less in radius drawn from the  reference station.   Homogeneity of  land
use within each area was reasonably consistent.  This study was concerned ex-
clusively with absorption of ambient community atmospheric lead via the respira-
tory system and did not involve pediatric exposures nor exposures of men  with a
variety of possible occupational opportunities for additional  lead  absorption.
The extent to which blood lead levels  in women reflect levels in the general
population was not examined.

     An attempt was made to sample "suburban" and "urban"  populations in  several
of the regions under study.  In some areas this objective  was confounded  by the
fact that highly mobile populations and rapidly changing neighborhoods made it
unlikely that a group of acceptable volunteers could be recruited without ex-
tensive personal solicitation.   In the same region a comparison of  "suburban"
and "urban" populations would tend to  minimize the influence of regional  factors
which might influence regional  biological lead levels.  Climate and sources of
food and water would tend to affect equally populations living in the same gen-
eral metropolitan area.  Populations sampled in the current study are presented
in Table 2.

                      Table 2.   Population Groups Sampled
Community
Philadelphi
Cincinnati
Los Angeles
Los Alamos
Washington
New York
Chicago
Houston
Urban
Station
a Rittenhouse
Square
Pasadena
fWoodly Park \
^Cleveland Pk.j
Greenwich Vil .
Bridgeport
Nw. Houston
18
4, 5
74
85
63
56
N
137
209
224
140
148
204
Suburban
N* Station N
136 Ardmore-
Wynnewood
Okeana
(rural )
193
Los Alamos
219
140 Port Wash.
147 Lombard
191
19, 20 156
34, 35 166
41 , 42 204
88, 89 203
66 208

N*
150
162
191
198
208

*A11 statistical computations performed on populations corrected to exclude those
persons with possible unusual lead exposure and those who were less than 20 or
more than 79 years of age.   See p. 31.
     In addition to the groups of women which participated as described, smaller
studies were conducted in four other populations.

     1)  Husbands of 100 women participants in Los Alamos were sampled to see
whether or not there is a male-female differential in the blood level  of lead.
It was possible to define the occupational  exposures of these men, most of whom
were employed by the Los Alamos Scientific  Laboratory.   The atmosphere at the
Loboratory does not vary from that in the residential areas.  Husband-wife pairs

-------
                                      21


 were  considered  to consume approximately the same foods, although the quantity
 consumed  by  the  husband will, of course, generally exceed that-consumed by the
 wife.   In  the absence of occupational lead exposure and under these circumstances
 it  was  believed  that the degree of difference in male and female blood lead levels
 could be  determined.

      2)   In  1962, blood levels of lead were determined in 120 Cincinnati police
 officers.  In 1971, it was possible to identify 65 of these men still on the
 police  force.  Of these 48 volunteered to provide blood samples.  The purpose
 of  this study was to see whether or not blood lead levels had changed in this
 population over  the course of 10 years.  Such changes as might be observed could
 be  attributed to the effect of aging or to actual differences in absorption over
 the intervening  period.

     3)  A group of 87 men and women residents of the Pennypack Woods section of
 north-eastern Philadelphia was studied in April, 1961, as representative of a
 "suburban" population.  Of these 44 had moved from the neighborhood by April,
 1971; 5 more had died.  Of the remaining 38, it was possible to obtain blood
 samples from 23  persons in 1971.

     4)  The report issued in 1965 included blood lead data from assays performed
 on  37 men and women who had participated in a multiphasic screening program in
 Alpine County, a rural area in the Sierra Nevada.  Most of these persons were
 of  American  Indian  ancestry and had lived in the County for most of their lives.
 Blood samples were obtained in October, 1971, from 39 persons who had lived in
 Alpine County for at least 5 years.   This group did not represent to a signifi-
 cant extent those individuals who had been sampled previously; however they were
 considered to be representative of persons who had been long-term residents of
 Alpine County.

 Population Lead Levels

 Data Collection:   In order to examine the correlation between air and blood lead
 levels, it is necessary to control  those variables which might be expect  d to in
 fluence the concentrations of lead in the blood.  The selection of populations
 permitted residential  specifications as to location and duration.  Through the
 use of an interview, other variables were evaluated:   occupation, hobbies, smok-
 ing habits, sources of food and water, medical  history.   The principal study was
 based exclusively on women, a fact which eliminated differences in blood lead
 due to difference in sex,  and to a significant degree, in occupation and exposure.

 Determination of BJood Lead Levels:   Persons in the several  populations studied
who met the residential  criteria for inclusion and who had been interviewed con-
 tributed blood samples for analysis.  Blood was drawn from the ante-cubital  vein
 into lead-free evacuated tubes of 20 ml  capacity.   Tubes  were stored under re-
 frigeration until the  time of analysis.   Assays.were  performed according to the
standard dithizone method  of Bambach and Burkey'^'  as modified by Cholak and
Burkey.*

     There is no  argument  with the position that an epidemiological  investigation
of this  type  would be  improved by the determination of the total  body burden of
lead in  exposed  populations.   It is  apparent,  however, that  these determinations
 or active clinical  procedures such  as provocative lead excretion  tests with

*Method  available upon request to J.  Cholak, Kettering Laboratory, Cincinnati,
 Ohio  45219.

-------
                                      22


chelating compounds require experimentation upon humans  in a way which  cannot
be currently justified in epidemiological  studies.   Blood lead levels do,  how-
ever, reflect recent major changes in exposure and  the body burden  of lead,
insofar as there has been long-term exposure at a relatively constant level  so
that a steady-state relationship between blood and  the body burden  exists.   In
the situations under examination the likelihood of  recent major changes in ex-
posure is small.  If the exposure to lead has been  constant over a  period  of
time, it is reasonable to assume that the levels of lead in the blood of these
populations are meaningful indices of body burden for these adult populations.

Estimation of Alimentary Lead Intake

     Since lead levels in man reflect intakes from  both  the respiratory and  gas-
trointestinal systems, it is not prudent to make quantitative inferences about
absorption from the lungs with knowledge reflecting only total lead absorption.
Without standardized alimentary intakes of lead or  knowledge of its level  in
each population, a situation may exist in which differences in blood lead  levels
may be incorrectly attributed to variability in atmospheric lead levels.  It
is also true that a lack of difference in mean blood lead levels between popula-
tions may be due to compensatory variations between sources of absorption  for
those populations.  In other words, high alimentary lead and low respiratory
lead levels in one population may produce the same  blood lead level  which  is ob-
served in a population exposed to high atmospheric  and low food lead levels.  To
infer that different atmospheric lead concentrations have no influence  on  blood
lead levels is obviously incorrect in this situation which omits consideration
of lead intakes in food and water.

     In carefully controlled studies of lead metabolism, it has been possible to
collect accurate duplicate diets of the experimental subjects, for  whom alimentary
lead intakes could then be determined.  This approach did not lend  itself  to the
study reported here, however, because of the cost of duplicate diets for large
numbers of people and the difficulty in obtaining accurate duplicates from in-
experienced people.  It was determined that information  of approximately the same
utility might be derived from measurement of lead excreted in the feces, wh;^ii
contain some 90-92% of lead ingested in the diet.  Accordingly 20 volunteers in
each region were recruited to participate in a ten-day excretory collection.
Urine and feces were collected separately in low-lead plastic containers.  A
measured aliquot of urine and all fecal samples were shipped to Cincinnati for
assay.  To assure total collections only intelligent and motivated  persons were
invited to participate.  A detailed instruction period and daily collections of
materials by a nurse contributed to the completeness of  the collections.

     The manner in which, the metabolic study was designed reflects  certain as-
sumptions which affect its validity.  It has been assumed that persons  who live
in a general metropolitan region derive their food  and water from essentially
equivalent sources, so that measurements of alimentary lead in Port Washington
on Long Island also describe the situation in New York City.   Similarly, measure-
ments in Chicago or Philadelphia suburbs are applicable  as well  to  the  respective
urban area.  There are obviously differences between individuals in their  choice.
of foods, but we believe that the sampling method gave a reliable index for  the
region.  We further believe that the dietary habits of 20 people over the  course
of ten days are representative of the general habits of the parent  population
of 200 people.

-------
                                       23


      It is  also  apparent  that extrapolation of a ten-day sample to an expression
 of annual lead ingestion  is not fully  satisfactory.  It is impractical, however,
 to sample throughout  a year in the absence of compelling scientific reasons and
 participants with  special motivations  and attitudes.  Since the samples used in
 this  study  were  collected during times of the year when an abundance of fresh
 foods was not normally available, however, we believe that the ten-day sample
 does  reflect the general  situation for most of the year.  Only during the sum-
 mer,  when fresh  fruits and vegetables  are more readily obtained, might there be
 major changes in the  food sources of some individuals.  We have permitted the
 assumption, therefore,that the sampling as described does reflect general ali-
 mentary lead intake.

      Analyses of excreta  for lead were conducted by the atomic absorption tech-
 nique; 10%  of the  samples were also analyzed by duplicate dithizone procedures.
 In the case of the fecal  samples, which were received and weighed in plastic
 bags, the entire sample with the bag was ashed for assay.  Lead content of con-
 trol  bags had been determined and found to be only several micrograms per bag.

                          OBSERVATIONS AND DISCUSSION

      The investigative program described in this report was designed to:

      1)  Determine whether or not a change in the concentration of airborne
         lead had  occurred after an interval  of 8 years at a series of refer-
         ence locations designated in 1961-62.

      2)  Determine the ambient atmospheric levels of lead at selected additional
         sites in  the cities studied in 1961-62 and in other communities.

      3)  Examine the extent to which levels  of lead in the blood of selected
         population groups reflect exposure to lead at various levels in  com-
         munity  atmospheres.

      4)  Determine whether or not a change in the blood levels of lead had oc-
         curred  after an interval  of 8 years  in selected population groups.

 Except in the case of the fourth objective, data were developed which are useful
 in answering the questions presented.

                                Aerometric Data

      Aerometric data are presented in  Tables  3 and  4.   With respect to changes
 in atmospheric levels of lead during the  interval  1961-62 to  1968-69,  the reported
 data  (Table 5)  demonstrate higher lead levels during the latter period at 14 of
 the 19 sites which were re-established.  A careful  examination of experimental
methodology tends to exclude the possibility  that this  change is  an artifact re-
flecting changes  in technique.   Current assays of 1961-62 filter  wedges from Los
Angeles and Philadelphia have validated the lead assay  techniques  used during
 the period 1968-71.  Because of the observation that filter wedges  from the  1961-
62 Cincinnati  study, when re-assayed in 1969, yielded  results  approximating  91%
of those obtained in the earlier period,  it is possible  that  increases in at-
mospheric lead levels at the re-established Cincinnati  sites  are  greater  than
may be apparent.   Consequently,  increases  in  the current study can  hardly be


                                                Text continued on page 30

-------
                                                    24
f»
g
ia , Cincinnati, and Los Alamos, 1968-6;
.c
a

Q)
•o
to
rH
•H


«
10
a>
rH
0)
hn
UU
c

cn

^
£
•H
•a
ca
3
 OCMOO
otoeoooincnincom tocMrHOtjit~cocorHtocM coooco
o 	
mto^cococo^in 'tf CMCMincorHrHrHCMCMi— i CM CM I-H co
tofocM'
to ^
ra -H 3 "O
tp  Q)^^ rH c" ^— ? CO
•P bJD "^ ^O cl) CD O CU cj
Q  O
w OL a. -H occC2a
Q J3 -H
rH PXl U
CM
i-H
0
rH
r-t
rH
CM
rH
O
00

0



O)

o



N
O


J5

0


^

o


m

o


to

0






CO
CO


















CO
o
in
o
CO
o
a
CO
o
CO

o



CO

o



CO
o


m

o


TT

O


CO

O


rH

O





*
CO











(O
c
(U
J^*"
o


CO
o
in
0
0
T»<
0
CO
0
CM

o



CO

0



co
0


m

o


rf

o


CO

o


CM

O





*
m
CO











fl3
e
a>
^
0


rH
rH
CO
00

O



00

O



oo
0


,_,

1-1


00

0


to

o


to

o


4-
tn

-------
25



•d
0)
^
C
•rH
•P
C
O
tJ


«

fjQ
m
f$























nd Houston, 1970-'
CO
X
0
i»

^
cu
CD
O
•rl

rj

C
lflj
•c
CO
3

•p
•H
O

rH CD
0 O
CM CM
TJ< in

CO CO
O CM

rH CO
CO CO

m CD
m CD
rH rH
CD CO
rH rH
rH rH
co m
rH rH
CD 00
coin
o'oo
coco
00 CO
, f
Chicago 60 ]
61 ]

in co
rH CO
O O
m o

o oo
Tf CO

in rH
CD 00

rH CM
CM CM
CD rH
iH CM
O iH
O 00
CM iH
O CD
oco
C0£
*s
CM £

s-
o
Q.
cr
-a
'£.
GO

CD
m
m
CD

CD

o

CM
CM
CM
CM
00
iH
CN
0
0
CO
s
CM

S

CD
rH
O
oo

CD
CM

O>

co
CM
CN
CD f-
rH CM
CM CO
00 CN
rH rH
r> o
SS
rH 0
rH 0
rH rH

lO CO
co co
"O
to
o
_l

00 00
CO CM
co o
CO N
, ,
CM CO
MI CN

00 T}<
t*« 00

O CM
CD m
rH CO
rH CO
CM CM
t- 0
r- o
rH (N
CM CD
rH CM
OJrH
rH rH
m CM

5S

CM
00
CD

in
CO

iH
CD

rH
CN
rH
CM
rH
00
rH
CM
rH
rH
rH
un

+-
CQ
0)
•P
rH
CM CO
O CO
CO M1
CD 00
1 rH
i m

CO CM
CO ^

T CO

M1 CD
CO CO
•

rH •rf' CM
CD CD 00

t~ ao o
CD 00 CM

rH O CD
O 00 CD
CN rH CN
CO CO CD
00 00 CD
rH rH CM
CO CD CD
O O ^
N CM CN
t> CM 00
CD t~- rH
rH rH CN
CO CD rH
CD 00 Tji
rH rH CM
CD rH i— 1
oo CD m
rH rH CM
m CD CD
M* CN CD


*
co rf in
00 00 00
o
•1 —
3
0)
CD
i.
C3
CO f
O O)
I-H d
O N t-
O 00 CD
rH O O
£*• ^^ C^
lO CO CD
CD CO Tj*
tO C^J O

00 C(OrHO
CMrHCOTjCQOOCDoi
SCD O CM rH 00 in CD
co CM ir r CD oo CD
rHOrHrHrHOOO
t>rHt>CMrHCn^l
-------
                                                          26
•d
 0)
4J


O
U
CO

3
pa
IS
         O
 be h
 c  -H
•H  CO

 M  
    u
    •H
    ja
  p
         u  fi
         a  o
         c  a
         <0 -H
         •P
         c
    co	









       i—I      4N  H  r-l f-l 4N  r-4  1-4








       O      Wi-tOOl-tOi-l




               Sr-i  oo  o in co  TJ<  to
               oo  co  o o eft  01  c^
    ^   •        .......
       l-l      rH  |H  rH rH r-4  O  ^




       (o      to  m  cn oo 
-------
                                              27

                                          Table 4
       Yearly Means  and  Confidence  Intervals of Air Pb Concentrations For Each Station
                 Area
                  Station    Geometric Mean    95% Confidence Limits
                                                                            (ug/nP)
 Cincinnati
   Dec 1968  -
   Nov 1969
 Philadelphia  Suburban
   Dec  1968 -
   Nov  1969

 Philadelphia  Urban
   Dec  1968 -
   Nov  1969
 Okeana
 Okeana

 Ardmore
 Wynnewood
 Los  Anqeles
   Dec  1968 -
   Nov  1969
                       Rittenhouse Sq.
                       Pasadena
                       Pasadena
 Los Alamos
  Dec 1968 -
  Nov 1969

 Washington, D.C.
  Oct 1970 -
  Sept 1971
New York Suburban
  Apr 1970 -
  Mar 1971
                       Wood ley Park  ~l
                       Cleveland Park]
Port Washington
Port Washington
 30
 31
 32
 33
 34*
 35*

 19*
 20*
                   11
                   12
                   13
                   14
                   15
                   16
                   17
                   18*

                    1
                    2
                    3
                    4*
                    5*
                    6
                    7
                   41*
                   42*
70
71
72
73
74*
75
76
77

88*
89*
1.92
1.45
2.13
0.85
0.31
0.33

1.28
1.01
             1.89
             1 .74
             3.75
             2.18
             1.39
             1.09
             1.08
             1.67
             4.03
             4.34
             3.46
             3.72
               06
               36
               39
               55
             0.14
             0.20
1.10
1.64 (7 mo.)
2.21
1.78
1.19
1.13
2.34
1.08

1.14
1.10
(1.63,
(1.16,
(1.75,
(0.71,
(0.24,
                   (1.63,
                   (1.52,
                   (3.30,
                    1.90.
                    1.19,
                   (0.90,
                   (0.92,
                                                   (0.95,
                                                   (1.36,
                                                   (1.91,
                                                   (1.53,
                                                   (1.07,
                                                   (0.99,
                                                   (2.05,
2.27)
1.83)
2.60)
1.01)
0.41)
 (0.20, 0.40)

 (1.06, 1.55)
 (0.86, 1.19)
         20)
         00)
         26)
         52)
         65)
         33)
         27)
                   (1.42, 1.93)
(3.40,
(3.58,
(2.96,
(3.20,
(2.62,
(1.92,
(2.62,
4.76)
5.26)
4.05)
4.32)
3.57)
2.90)
4.39)
                   (3.71, 5.60)

                   (0.11 t 0.18)
                   (0.16, 0.25)
       1.27)
       2.00)
       2.55)
       2.06)
       1.32)
       1.28)
       2.67)
(0.94,  1.24)

(1.03,  1.27)
(0.98,  1.24)
* blood sampled

-------
                                             28


                                     Table 4 (Continued)

                 Area                  Station      Geometric Mean   95% Confidence Limits
                                                       (uq/mj)(ug/m3)

 New York Urban                            81            1.76               (1.44, 2.15)
  Apr 1970 -                              82           2.04               (1.74, 2.38)
  Mar 1971                                83           1.73               (1.57, 1.90
                                          84           1.71               (1.54, 1.89
                       Greenwich Vil.      85*          2.08               (1.79, 2.43)
  May 1970                                86           1.34               (1.13, 1.58)
  Apr 1971                                87           1.22               (1.06, 1.39)

 Chicago Urban                             60           1.30               (1.17,1.45)
  Mar 1970 -                              61            1.35               (1.23, 1.49)
  Feb 1971                                62           1.53               (1.33, 1.76)
                       Bridgeport         63*          1.76               (1.60,1.93)
                                          64           1.83               (1.67, 2.02)
                                          65           1.57               (1.39, 1.78)
                                          67           1.55               (1.37, 1.76)
                                          68           1.87               (1.51, 2.31)

 Chicago Suburban       Lombard            66*          1.18 (7 mo.)       (0.83, 1.68)
  March -
  Sept. 1970

 Houston                                   50           1.15               (0.92, 1.43)
  Mar 1970 -                              51            1.02               (0.86, 1.21)
      1971                                52            2.13               (1.83, 2.47)
                                          53            2.26               (1.87, 2.73)
                                          54            1.32               (1.16, 1.49)
                                          55            0.87               (0.73, 1.03)
                       NW Houston         56*          0.85               (0.75, 0.96)
* blood sampled

-------
in
 0)

3
       LU to

       «* rl

       R -°
       H? E
       r^  E
       uu  q,
       —i  o
r-  (D

^B
o  §

81
    i
       i§
       O 01

       oo 5

       
•H CO
!H E
-P X.
• CO CM N CM
m o o to to CM cn
rH t^~ in ^ CO 00 00 I I I
CM rH •"J' CM 00 tD r>
rH rH rH




tO (f> If CM Tj< f~ 00 1 1 1
in rH co CM in Tf ^*








N to CM in CM  to i i i
rH rH rH
1 1 1

* * * *
* * * *
to >-H in CM to in to
CO rH CM CM CM 1 1 1
+ 1 + + + + +





o^^*inooo^CiOOh*oorH
OOt^NrHCOOOtOCMO








CJ> in to co o Is- to
CO N CM rH rH 00 00
rH rH CO CM rH O O







* * *
rH W CO TP lO tO t^ 00 d ^5


a
•H
X!
P.
rH
0)
•a
CO
rH
•H
£





O 'J* N O N O
00 to 00 00 O rH
oo TC m to to m
to t** t** m ^ ^j*







rf rH CM  in





m to N oo i i
co to <* CM








m to t- to i i
CM
i i

* *
* *
co co in CM
rH CO CM II
+ + + 1





CM m co m I-H co
Ol TP rH 00 CO CO








o en o t>
t^ o t~ oo
• •••ii
rH rH i-H O







* *
o rH CM oo ^ m




•H
P
03

C
•H
O
a
•H
o
                                                                                                                 m

                                                                                                                 •P
                                                                                                              •a  en
                                                                                                              a>
                                                                                                              i-H  -P
                                                                                                              a c
                                                                                                              E  a)
                                                                                                              in  u
                                                                                                              en  -H
                                                                                                                 =H


                                                                                                              o  &
                                                                                                              i-H  -H
                                                                                                              CQ  CO

                                                                                                              *  *
                                                                                                                 *

-------
                                    30
attributed to the lead assay technique.   Air flow measurements have been based
upon identical procedures during the two test periods.   Insofar as  can be deter-
mined, the sampling and assay procedures during the two periods are directly
comparable.

     One must consider the possibility that the atmospheric lead levels reported
in 1961-62 were inaccurate and too low,  a possibility which would tend to show
that current levels erroneously reflect  a non-existent increase.  Evidence for
this might rest in the fact that in 1961-62 a number of Los Angeles samples had
to be excluded from the site averages.  If the excluded samples represented
periods of relatively high atmospheric lead levels, the apparent monthly or
annual mean lead concentrations at the affected sites would have been too low.
Most of the excluded samples had been collected during the photochemical!  smog
season, from June through November.  Organic constituents in the air altered
the porosity of the filter media in a way which reduced air flow and made
volume measurements unreliable.  There were times in 1961-62 when only one or
two out of eight stations were providing usable data.  Special tandem pump units
and a reduction in smog levels in 1968-69 eliminated this problem of filter
obstruction.

     Most of the high-reading samples in 1961-62 in Los Angeles occurred during
periods of surface inversions associated with low temperatures rather than during
the smog season.  The number of lost samples in Los Angeles in 1961-62 approxi-
mated 3% of the total filters exposed.  While it is possible that the operational
situation in Los Angeles in 1961-62 resulted in loss of certain high-reading
samples, it is quite unlikely that the losses were of sufficient magnitude to
result in the apparent change which is observed in comparison with  1968-69 sam-
ples.  The problem of filter obstruction never constituted an important problem
in other cities.

     An examination of meteorological conditions is relevant to the comparison
of data collected during the two test periods.  Since no meteorological instru-
ments were located at the actual sampling sites, inferences with respect to
climatological conditions must be drawn  from the data collected by  the National
Oceanic and Atmospheric Administration (NOAA) at its reference location in each
city.  These data are valid for precipitation and heating requirements; the
data are somewhat less satisfactory for  wind velocity and direction but have
general validity for the area.  That is, wind conditions at the Los Angeles In-
ternational Airport do not invariably reflect the situation in Pasadena.   How-
ever, over long periods, such as those used in these studies, local deviations
between sampling sites and the NOAA reference station would not be  expected to
differ to an important degree from year  to year.  An examination of climatologi-
cal data during the sampling periods and over a 30-year period does not suggest
that either sampling period is non-representative.

     That air levels of lead measured in 1961-62 and 68-69 are sufficient to
demonstrate a trend at the sampling sites (based upon two points in time) has
been questioned.  Especially in Los Angeles, the National Air Sampling Network
lead data at the reference station show  wide annual deviations, upward and down-
ward, over the course of more than ten years (Figure 16).  While the results
of the study reported here show relatively higher levels of lead in 1968-69
than in 1961-62 at 2 of the 4 Cincinnati sites, at 2 of the 7 Philadelphia sites,
and at all of the 8 Los Anqel^ sites, the NASN data cannot be ignored and de-
serve further consideration

-------
                                      31
         4
         0
             Los Angeles
            •' Cincinnati—'
                  60      62      64     66     68      70      72
                              ANNUAL  MEAN
    Figure 16.   Lead-in-Air Data  (1959-71)  National Air  Sampling Network^)
 (Symbols Indicate 1961-62 and 1968-69  population and Air  Lead Level Studies.)

Note:  The Los  Angeles  NASN site  is  Station #  1  in  this  study.
       The Philadelphia NASN site is Station # 14 in this  study.
       The Cincinnati  NASN site is Station  # 30  in  this  study.

It is relevant  that the NASN system  was  designed for a less specific purpose, that
different collection media and analytical techniques are used, and that the annual
mean at each site is based on approximately 26 samples.

     Information on atmospheric levels  of lead at sites  newly established  m 1968-71
serves to confirm previous observations  which  associate  higher levels with urban pop-
ulation centers and lower levels  with areas of low  population densities.   Data de-
rived from Okeana (Stations #34,35), Los Alamos  (Stations  #41,42), and Alpine County
reflect an almost "background" situation with  respect to atmospheric lead  levels.

     Month-to-month variations in atmospheric  lead  levels  were generally observed.
For the most part highest levels  occurred during July-October while the lowest levels
occurred in February-April.  The  variations were most apparent at the Los  Angeles
sites, where the highest lead levels tended to occur in  October-December and the
lowest in April-July.   Examples of the  monthly variations  are presented in Figure 17.

                     Blood Lead  Concentrations

     Histograms showing the distributions of blood  lead  concentrations in  populations
meeting specific epidemiological  criteria and  living in  areas of dissimilar character
and atmospheric lead levels are presented in Figure 18.  Corresponding frequency
distributions are given in Table  6.   Mean blood  lead values and their 95%  confidence
limits are presented in Table 7.   For epidemiological purposes subjects were excluded
from tables and statistical consideration when they:

-------
                                      32

     1)  were under 20 or more than 79 years  of age;
     2)  when there appeared to be significant occupational  or avocationa'l
         exposure to lead;
     3)  when there was a significant amount  of wild  game in the diet

The inclusion of these subjects in the analyses does  not change significantly
the computed data means for the several population groups.   Appendix 3 presents
the blood lead concentrations with no exclusions from the study population.

     The age distribution was not consistent  for subjects at each area under
consideration (Appendix 4).  Consequently an  age effect might have been apparent
in the mean blood lead levels and in their distributions in  the various groups.
The relationship between age and blood lead levels was examined by means of
regression analyses (Table 8).  These were carried out for each location separ-
ately since patterns of mean levels differed  among locations (Figure 19).   When
a third order model was assumed, the regression of lead on age was shown to
be not significant in each location at an overall  5%  level.   Therefore no  ad-
justment for age has been made in the analyses of blood lead data presented in
this report.  Mean blood lead levels by age for each  location, given in Figure
19, show that there is considerable overlapping of the confidence intervals,
supporting the conclusions of non-significant regressions of lead on age.

     Since the distribution of persons by cigarette smoking  status was not  consis-
tent in each location (Appendix 5), and since cigarette smoking was believed to have
a possible influence on blood lead levels, tests of the differences in mean blood
lead values among locations were carried out  after partitioning the sample  into three
cigarette smoking categories (Tables 9,11,12).  The F-ratio  testing the differences
between the 11 locations is 31,47, which is significant at the 1% level.

     Table 9 also shows the influence of smoking upon blood  lead levels to  be clear-
ly significant (F = 8.27).  The blood lead level of smokers  exceeded that  of non-
smokers and previous smokers (Tables 11,12).   The relationship held true for males
as well as females in the husband-wife pairs  in Los Alamos (Table 13).

     Since the interaction term in Table 9 is not significant (F = 0.81),  one mav
conclude that the significant differences among mean  blood lead levels in  the
various locations exist in each smoking category.   That is,  the average blood
lead values of smokers is significantly different among these eleven locations;
the same holds true for previous smokers and  non-smokers.

                   Urban-Suburban Blood Lead   Levels

     The investigation of whether mean blood  lead levels reflect degree of  ur-
banization was carried out by applying a binomial  test to the data from three
metropolitan areas where an urban and suburban population had been sampled.   In
each location a sample of smokers and non-smokers  was used for this test,  giving
a total of six comparisons between urban and  suburban blood  lead levels (Table lOa).

     In both smoking categories the Philadelphia urban groups ("Rittenhouse") were
higher than the suburban ("Ardmore"); the Chicago urban groups ("Bridgeport") were
higher that the suburban ("Lombard"); and in  New York the urban groups ("Green-
wich Village") were higher than the suburban  ("Port Washington").  The proba-
bility of obtaining 6 relationships in this direction (urban > suburban) out of 6
comparisons is approximately equal to 1/2^ =  1/64 = 0.015, if one assumes either
direction is to be equally likely.  Consequently it is concluded that urban mean

                                                Text  continued on page 49

-------
n
     bj) Q)
        3-P
        -H
        w
     Wl      /-N
    rH  .d   W
N   a)  o   aj
r-l   >  CO  rH
     0)  W   0)
(DM       U)
^      h   C
3  T3  O  <

•H   qj       oi
  O  C
  •H  O

  0)

  Q.^1

  O
  E
                                                                                                                                                              -p
                                                                                                                                                               c
                                                                                                                                                               o
                                                                                                                                                                            o
                                                                                                                                                                            t-
                                                                                                                                                                            Oi
                                                                                                                                                                           00
                                                                                                                                                                           CD
                                                                                                                                                                           Ol
                                                                                                                                                                             I
                                                                                                                                                                            r-f
                                                                                                                                                                            CD

-------
                                                              34
                                                                   in
                                                                   O  00  CD
                                                                                                                                I
                                                                                                                                O
                                                                                                                                00
                                                                                                                                <£>

                                                                                                                                Cl
•o
0)
3
 c
 o
o
3
hO
•H

fcl
a
     •a
     as
                  
-------
                                                                35
                                                                    O
                                                                    CM
                                                                                                             O
                                                                                                            hw
                                                                   O 00  CO
                                                                                         CNJ
                                 00  CO
                                 d  6
                                                                                                                               i
                                                                                                                               O
                                                                                                                               OS
                                                                                                                               CO
                                                                                                                               oo
                                                                                                                               <£
                                                                                                                               OS
•o
d>
c
•H
-P
ti
O
o
     0)
     T3
     CO
CD
                                                                                                                               CM
                                                                                                                               CD
                                                                                                                                I
                                                                                                                               (—(
                                                                                                                               CD
                                                                   O CO  CO

-------
                                                                 36
                    eo
                                                             o
                                                            I-CO
                   o oo  co
                                                    CO  CD
                                                    d  CD
                                                                                                              Q
                                                                                                                                N
                                                                                                                                I
                                                                                                                                O
                                                                                                                               CD
                                                                                                                                I
                                                                                                                               00
•c
0)
c
o
o
5
 CO
 §
•rl
 U
•H
O
              O CO  CO   •<*       CJ
                                                                     CO
                                                                    C\J
                                                                             ,-  00 CO
                                                                                d> o
                                                                                    •*  co
                                                                                    ci  ci
                                                                                                        L°
                                                                                                         -co
                                                                                                                     •p
                                                                                                                     c
                                                                                                            ,0
                  O CO  CO    •t
                                                                    CO
                                                                    eo
                                                                   O CO  CD
                                                                                        CM
                                                                                                                               to
                                                                                                                               I
                                                                                                                              O)
                                                                                                                              I—I
                                                                                                                              •
                                                          ,m/Slf qd

-------
                    m
                    in
                                                               °
                  O 00  CD
                                                      00  CO

                                                      0  O
                                                                      37
 3

 C
•H
O
O
 c
 O
+-!
 CQ
 3

a
                O  00 CD
X!
-p

-------
                                                              38
3
C
•H
4->
C
O
O
N
r-t
    O
    bfi
    eo
    u
    •H
                  (0
                 O 00  CD
                  O
                  (O
                 o  oo  CD
                                                                                                              z
                                                                                                             -o
                                                                                                             -co
                                                                     (O
         i
                    I
                   O
                                                                                                                                 I
                                                                                                                                QO
                                                                                                                                CD
                                                                    O 00  CD
                                                                                                     T—I	1	1	
                                                                                                      00 CD

                                                                                                      O O
                                                                     CO
                                                                     <0
_ ^—
 <
                                                                    O CO  CD
                                                                                                      00 CD
                                                                                                      C) CD

-------
                                                                    39
                    (O
                    (0
•o
0)
3
C
 d
 o
o
£
3
                                                              - Z

                                                               O
                                                                CO
                   O 00 CD
                                         C\J
     U
     •H
I
         I
        o
 I
oo
CO
                                                                       O 00  CD

-------
                                                             40
                O 00  CD
                                                                     O 00  CD
                                                                                                                 Q
T3
0)
 a
 O
O
     C
     o
     -!->
     M
     C
     •H
                O  00 CO
                                                                      (O
                                                                     O  00  CD
                                                                                                                 o
                                                                                                                 -to
00  CD
d  a
                                                                                                                                       I
                                                                                                                                      O
ft)   5t
^   <•>

a
•H
                                                           -o
                O 00  CD
                                                                     CO
                                                                     r^.
„ >
i 	 1 	 1 	 r i —
0 00 CD •*
*~

C\J i- 00
C)

LL
- —3
i 	 \ —
CD O 00
d> ^


CD

                                                                    O  00  CD
                                                                                                                Q
                                                                                                                                       I
                                                                                                                                      oo
                                                                                                                                      <£>
                                                                                                                                      ff)

-------
                                                                   41
                     CO
                     00
                    O CO  (O
                                                T-  op  CO
                                                    o  6
                                                                  o
                                                                   co
0)
3
C
•H
•P
C
O
o
(1)
 o
>•
     (D
     SB
                   o  oo co
                                                                                                                                   i
                                                                                                                                    h-
                                                                                                                                     I
                                                                                                                                    o
                                                                                                                                    CD
                                                                                                                                     I
                                                                                                                                    CD
                                                                     O CO  CD

-------
                                                                     42
T3
0>
3
     0
     f*
                                                                       O>
                                                                       00
                                                                      o oo  co
 I
o
 I
oo
'-D
CD
                                                                      00
                                                                      CO
                                                                     O 00  (D
                                                                                            c\i

-------
                                                  43
TJ
C3 >->
(b hn
••3 o
0
TD r-H
O hi
oo .-i a
rH CQ O

0) 
-------
                             44
TJ
(U
3
fi
•H
•P
C
o
o
oo
I
o
£
2
LU
LU
CC
0
1 I
O
CVJ


LU
O
<
_J
_l
>
1




0
Oo>

2 1
1
0


o
CO 1
rO
||
X
^
1 |
c

                                                       o»
                                                       O
                                                       O
                                                       \
                                                       o>
LU
_J

O
O

3
DO
                         o    °  o    o

              AON3IT03UJ 3AU.V13U%

-------
                                                                            45
 0)
r-l

•8
E-t
 CU

 CU



XI
          O  CU
          O tJ
         rH  (0
         CQ  cn
 o

 c c
 o 
id
cu
id
to
CU
in
O

0)
-P
-P
•rH
CU
iH
O
&
—4
IU

id C
> -rl
1 I /"|
•M U
cu a
-P tJ
C -rH
H S
-Q rH
di id
OH
rHCU
tjl-p
O(-l
t-l
r-5 H
S££££S2coSc£,HrHo'
o o o o o r-



iH
"a-
o o o o o •«*




r~ r~- c r~- r^ co
VO VD 0 V£) VD CO
o o CM o CM in

o o rH ro co CM
rH
u
o
o oo co co criTr
r~~coo^ii'a ro v£> vc CMOO-slCMrH
rHiHCMCMCO^t^J'in A
OOOOrHrHrHrHiHrHiHHrH
                                                                                                                             co
                                                                                                                             co
                                                                                                                                   m
                                                                                                                             ro   VD
                                                                                                                        o   o
                                                                                                                             CM   O
                                                                                                                             m   in
                                                                                                                             CM   OJ
                                                                                                                             ro   us
                                                                                                                        C   X
                                                                                                                        -H   (d
                                                                                                                        g   S

-------
                                                                46
 CO
rH
 0)
 >  c
 Q)  O
r4  -P

X)  C
PJ  *H
    (Q
     £
(J M

rH rH CM r** C*)
O O O VO CM
O O rH H CM VO CO
rH CM


oo in
CM O
O O O O O CM O
rH



CM l> CM «* ^J1 VO
in in vo rH o ••r
O O >H CM CO CM rH
rH CM

0 0
m in
o o o o o r- r>-
rH

"Si
o
O OOOOOQ in  r-
CM oo m ro in
•^ vo o vo •»* o o
CM i-H CM



«* vo in CM
en vo oo in
o ro r» o o o o
CM

o in o o in
O CM O O CM
o " rH o in o 1-1 o
CO rH rH



oo *3* *^ o\ OA in in
r^ o co vo o in in
A
oo ^r o vo CM oo
CM co ^r ^r in m
rH rH rH rH rH rH in
1 1 1 1 1 1
CM 00 *» O VO CN rH
CM CM CO •^••'J1 in A
rH rH rH rH i-H rH
                                                                                                                   o
                                                                                                                    •

                                                                                                                   00
                                                                                                                   V0
                                                                                                                           O\
                                                                                                                           CM
                                                                                                                                  o
                                                                                                                                    »
                                                                                                                                  CO
                                                                                                                              o
                                                                                                                                •
                                                                                                                              co
                                                                                                                                  o
                                                                                                                                    •
                                                                                                                                  en
                                                                                                                                  rH
                                                                                                                                  CN
                                                                                                                           CTi
                                                                                                                           (N
                                                                                                                           CO
                                                                                                                           co
                                                                                                                                   o
                                                                                                                                   CO
                                                                                                                    C      X
                                                                                                                   •H      fti
                                                                                                                   s      a

-------
                                            47
c
o
+J
10
3
O
a


J^
U
•H
*J;
c
0)
(U
VH









U
C
0)
3
cr

M


0)
^
•rH
•p
rH
(U


rJP


























x;
o
•H
s
0)
0

o






rH
'O
t ,
H
0)
-p
c
H




JJ
E
Ti
•H
m
•a
£
o








(U

ft
rH
•H
£>







4)
a
•rH
a
•o
•H
S




vooomcM^oo^'eroooo
• f
rH CM CM rH rH
COOO-'J'CM'^'iHOOOOOOOCM'^OO
ooi-HrHcorrinr~r-CMrHrHooo
CM CM rH i-H









rH VOrHnOOOOrOr-l>£)i-HrH
i^ cor-^J-incM^r^cor-r^
OOOOCMinrHcOCnrHOCMOOO
CM CM i-H i — 1 rH







^^
tn
0
CU GO OO OO ^^ C^ P**1 LO LO 00 ^J* ^5* 0^ CF\ LO ^O

O^ V i~H i~H *~H <™i rH f^l CN O4 CO rO fO
3 A




                                                                                       o   o

                                                                                       o   o
                                                                                            o   c»
                                                                                            ro   o
                                                                                                 (M
                                                                                       o   o
                                                                                       CD   
i-H
rH
00
CM
rH
1
CM
CM
rH
CO
i-H
1
CO
CM
rH
0
rH
1
CO
rH
S
rH
1
0
rH
CM
m
i-H
I
rH
00
rH
1
CM
in
rH
                                                                             oo
                                                                             in

-------
                           Table 7
       Geometric Means  and Confidence Intervals of
              Blood Lead Levels  Based on a
         Single Determination  from Each Subject
(Rank)
5
9
11
7
3
10
4
6
2
8
1
Area
Ok can a
Ardmoro
Rittenhouse
Pasadena
Los Alamos (~)
Washington, D.C.
Port Washington
Greenwich Village
Lombard
Bridgeport
Houston
N
162
150
136
193
191
219
198
140
208
147
191
Geometric M<
(mg/lOOg)
0.0157
0.0180
0.0205
0.0175
0.0149
0.0192
0.0153
0.0166
0.0139
0.0176
0.0125
                                                    (mg/lOOg)
                                                (0.0150, 0.0165)

                                                (0.0171, 0.0190)

                                                (0.0196, 0.0215)

                                                (0.0169, 0.0182)

                                                (0.0144, 0.0155)

                                                (0.0186, 0.0198)

                                                (0.0148, 0.0159)

                                                (0.0159, 0.0173)

                                                (0.0135, 0.0144)

                                                (0.0169. 0.0183)

                                                (0.0120, 0.0130)
All Females
1935
0.0162
(0.0160,  0.0164)
Los Alamos  (M)
  80
0.0172
(0.0163,  0.0182]

-------
                                      49


blood lead levels are significantly higher than those from related suburban
areas (at the 1.5% level), based on data from three paired locations in
Philadelphia, New York, and Chicago.

     Tests of the differences in blood lead levels in urban and suburban pop-
ulations of smokers only and of non-smokers only for each location separately
(i.e., Philadelphia, New York, and Chicago) were carried out by calculating
F-ratios for the comparisons.  The significance tests are made by comparing
these F-ratios.to the critical values specified by the appropriate statistical
test.  On the basis of the judgment that the Tukey statistical test is most
appropriate to the situation under examination, it is inferred that a significant.
urban-suburban gradient for both smokers and non-smokers exists in the Philadelphia
and Chicago areas.  The differential in the New York area was not shown to be clear-
ly significant.

     The relationship between mean blood lead levels and degree of urbanization
was also investigated by testing the significance of F-ratios obtained when the
urban and suburban mean blood lead levels were compared in each of three loca-
tions.  Based on the total sample (smokers, non-smokers, previous smokers), the
urban mean lead value is higher than the suburban in each location and all loca-
tions combined on the basis of the Tukey test (Table lOb).

     Table 14 and Figure 20 present mean blood lead levels and corresponding mean
air lead levels at the sites where both were measured.  The association between
these mean values was measured by means of the Pearson Product Moment Correlation
Coefficient and Kendall's Rank Correlation Coefficient.  The values obtained were
0,412 and 0.354, respectively, for all data.  Corresponding values for non-smokers
only were 0.400 and 0.345.  These values were not significant at the 5% level.
It was therefore concluded that the data do not contradict the hypothesis of no
association between average blood and air lead concentrations under the total en-
vironmental circumstances which exist at the study locations.

     The observations that urban levels of blood lead are significantly higher
than suburban levels, but that air concentration of lead are not clearly reflected
in blood lead levels generally, suggest that other variables are more important
than ambient air lead-levels in determining concentrations of lead in the blood.
The precise nature of the variables, which evidently differ in the several regions
studied, remains undefined.  Presumably alimentary lead intake plays a significant
role; however this was not thoroughly evaluated in the short-term metabolic stuny.
Climatic factors may also be relevant.

     In a given metropolitan area, urban-suburban comparisons tend to minimize
the influence of diet and climate.  That airborne lead contributes to the rela-
tively higher blood lead concentrations in center-city populations would seem
to be the most probable interpretation of this consistent observation.

     Mention should be made of the possibility that persons who live primarily  •
indoors, as do the population groups in this study, are exposed to air levels of
lead which differ from those which exist in the ambient outside neighborhood at-
mosphere.  While this is very probably the case, it is likely that the ratio
between indoor and outdoor air lead concentrations is similar in different com-
munities.  Consequently indoor exposures are directly related to lead levels

                                                  Text continued on page 60

-------
                                           50
                                         Table 8
                    Analyses of Functional Relationship Between Blood
                              Lead and Age for Non-Smokers


Okeana
N = 131


Ardmore
N = 99


Rittenhouse
N = 76


Pasadena
N = 134


Los Alamos (F)
N = 106


Washington, D C.
N = 105


Port Washington
N = 87


Greenwich Village
N = 56

Source of Variation
Due to regression
About regression
Error
Lack of fit
Due to regression
About regression
Error
Lack of fit
Due to regression
About regression
Error
Lack of fit
Due to regression
About regression
Error
Lack of fit
Due to regression
About regression
Error
Lack of fit
Due to regression
About regression
Error
Lack of fit
Due to regression
About regression
Error
Lack of fit
Due to regression
About regression
Error
Lack of fit
df
3
127
125
2
3
95
93
2
3
72
70
2
3
130
128
2
3
102
102
_
3
101
99
2
3
83
81
2
3
52
50
2
SS
0.1346
2.6019
2.5953
0.0066
0.1653
1.7854
1.7795
0.0059
0.0413
0.9678
0.9487
0.0191
0.1072
2.0087
1 . 9656
0.0431
0. 0907
1.2900
1.2900
—
0.0210
0.9962
0.9909
0.0053
0.0454
1.1190
1.0573
0.0617
0.0004
0.6132
0.6088
0.0044
MS
0.0448
0.0205
0.0210
0.0033
0.0551
0.0188
0.0190
0.0029
0.0138
0.0134
0.0140
0.0096
0.0357
0.0155
0.0150
0.0215
0.0302
0.0126
0.0126
*"
0.0070
0 0099
0.0100
0. 0027
0.0151
0.0135
0.0130
0.0309
0.0001
0.0118
0.0120
0.0022
F
2.19 (NS)


0.16
2.93 (NS)


0. 15
1.03 (NS)


0.69
2.30 (NS)


1.43
2.40 (NS)



0.71 (NS)


0.27
1.12 (NS)


2.38
0.01 (NS)


0.18
(NS)   Setting a 5% significance  level  for testing the significance of regression on age
      for all locations,  we must use a 0.5% significance level  in each individual loca-

      tion. The critical valuefor this  test  is  equal  to  4.28. (df=3 ,•*) .

-------
                                           51

Lombard
N = 125


Bridgeport
N = 85


Houston
N = 123


Los Alamos (M)
N = 32

Source of Variation
Due to regression
About regression
Error
Lack of fit
Due to regression
About regression
Error
Lack of fit
Due to regression
About regression
Error
Lack of fit
Due to regression
About regression
Error
Lack of fit
df
3
121
119
2
3
81
79
2
3
119
117
2
3
28
28
"•
SS
0.0029
1.3071
1.2793
0.0278
0.0353
0.6980
0.6557
0.0423
0.0286
1.4867
1.4756
0.0111
0.0443
0.4348
0.4348
~
MS
0.0009
0.0108
0.0110
0.0139
0.0117
0.0086
0.0080
0.0211
0.0095
0.0125
0.0130
0.0056
0.0147
0.0155
0.0155
"~
F
0.08 (NS)


1.26
1.36 (NS)

2.64
0.76 (NS)


0.43
0.95 (NS)



(NS)   Setting a 5% significance level  for testing the significance  of  regression on age
      for all locations,  we must use a 0.5% significance level  in each individual  loca-
      tion. The critical vaJue for this test is equal  to 4.28. (df=3 ,^4 .

-------

§
    25
    20
     15
    25
    20
     10
    25


    20



     15
     10
                           52

                         Figure 19
             OKEANA
RITTENHOUSE
           LOS ALAMOS (F)
                         ARDMORE
LOS ANGELES
                    LOS ALAMOS (M)

                    I	I   I   I    I   I
          20- 30- 40- 50- 60- 70-    20- 30- 4O- 50- 60~ 70-
          29  39 49 59 69  79    29 39  49 59 69 79
                          AGE (Years)
                    Mean Blood Lead Levels (ug/lOOg)
                    and 95% Confidence Limits by Age

-------
                            53
I
51
     25


     20
      15
     10
S   25
s1
^   20
      15
     10
     25


     20


     15
     10
WASHINGTON, D.C.

I   I   I    I   I
GREENWICH  VILLAGE

 I   I   I    I   I   I
             BRIDGEPORT
                  1   1    1   1
                                   PORT WASHINGTON

                                   I    I   I    I   I   I
                                1—  i
                                   HOUSTON
           20- 30- 40-50- 60-70-   20- 30" 40-50-60-70-
           29 39 49 59  69 79    29  39 49  59 69 79
                           AGE (Years)
                        Figure 19 (Continued)

-------
                                  54
                              Table 9
            Analysis of Variation Due to Smoking Status
                      (Females)  and Location
  Source                df        SS          MS

Smoking

Location

Interaction              20      0.2120     0.0106      0.81

Error                  1893     24.8060     0.0131

** Significant at the 1% level.
2
10
0.2167
4.1244
0.1083
0.4124
8.27**
31.47**

-------
                                      55
                                   Table lOa
                  Relationship Between Mean Blood Lead Levels
                and Degree of Urbanization in Paired Locations
                                    Smokers
           Comparison

Rittenhouse vs Ardmore
    (N=40)       (N=29)

Greenwich Village vs Port Washington
    (N=47)             (N=69)

Bridgeport vs Lombard
    (N=54       (N=99)
Rittenhouse vs Ardmore
    (N=77)       (N=99)

Greenwich Village vs Port Washington
    (N=56)             (N=87)
Bridgeport vs Lombard
    (N=85)      (N=125)
F-ratio
6.65*
on 1.67
26.09*
Non-Smokers
8.29*
on 1.31
31.77*
Direction
Urban >
Urban >
Urban >

Urban >
Urban -
Urban ->
Suburban
Suburban
Suburban

Suburban
Suburban
Suburban
                                   Table lOb
      F-Ratios Testing the Difference Between Urban and Suburban
     Lead Levels in the Total Sample (Including Previous Smokers)
Rittenhouse vs Ardmore
    (N=133)      (N=150)

Greenwich Village vs Port Washington
    (N=139)            (N=197)

Bridgeport vs Lombard
    (N=146)     (N=207)
Total:   Urban vs Suburban
(based on Philadelphia, New York, and Chicago)

*Significant at the 5% level  (Tukey Test)

         Note:
                                  F-ratio

                                   18.18*


                                    5.98*


                                   65.43*

                                   71.56*
        Urban


        Urban


        Urban
Suburban


Suburban


Suburban
        Urban >  Suburban
Critical Values (5%) for Judging Significance
    of Comparisons in Tables lOa and lOb
"Standard" F
    3.84
            Tukey Critical  Value
                    4.55
Scheffe Critical  Value
         18.30

-------


56
Table 11




GEOMETRIC MEANS OF BLOOD LEAD LEVELS
BY LOCATION AND SMOKING STATUS
(mg/lOOg)
Non-Smokers Previous Smokers
Okeana
Ardmore
Rittenhouse
Pasadena
Los Alamos (F)
Washington D. C.
Port Washington
Greenwich Village
Lombard
Bridgeport
Houston
All Females
Los Alamos (M)

0.0156
0.0180
0.0202
0.0172
0.0144
0.0181
0.0149
0.0158
0.0136
0.0167
0.0126
0.0158
(1127)
0.0165
Table 12
0.0139
0.0177
0.0200
0.0179
0.0154
0.0205
0.0150
0.0167
0.0146
0.0166
0.0120
0.0162
(304)
0.0168

Smokers
0.0171
0.0181
0.0214
0.0183
0.0160
0.0201
0.0162
0.0173
0.0146
0.0191
0.0124
0.0173
(495)
0.0193

Further Analyses of Relationship Between
Blood Lead and Smoking Status
Comparison
F
Direction

Smokers vs. Non-Smokers        40.01**  Smokers > Non-Smo"kers




Smokers vs. Prev. Smokers      10.37**  Smokers > Prev. Smokers




Prev. Smokers vs. Non-Smokers   2.57    Prev. Smokers > Non-Smokers




** Significant at the 1% level.

-------
                                       57
Los Alamos (F)

Los Alamos (M)
                                      Table 13

                      Geometric Means of Blood Lead Levels
                        for  Los Alamos Males and Females by
                                 Smoking Status
Non-Smokers
   0.0144
   0.0165
Previous Smokers      Smokers
     0.0154           0.0160
     0.0168           0.0193
  All
0.0149
 (191)

0.0172
  (80)
                Analysis of Variation due to Sex (Los Alamos data)
         Source          df        SS          MS           F

     Sex

     Smoking Status

     Interaction

     Error
1
2
2
263
0.1822
0.1093
0.0167
2.9402
0.1822
0.0546
0.0084
0.0112
16.30**
4.89**
0.75

                Correlation Coefficients between Los Alamos Couples
          Smoking  Status
             Same
             Differing
            No.           Correlation Coefficient
           Couples       (95% Confidence Interval)

             36                   0.353*
                              (0.027, 0.610)

             43                 0.265 (NS)
                              (-0.038, 0.581)
   * Significant  at  5% level.
   **Significant  at  1% level.

-------
                              58


                         Table 14


     Air  and Blood Lead  Concentrations  (1968-71)
                       (All Subjects)

                                       Geometric Means
Air
Rank
1
2
3
4
5
6
7
8
9
10
11
Area
Los Alamos
Okeana
Houston
Port Washington
Ardmore
Lombard
Washington , D. C .
Ri ttenhouse
Bridgeport
Greenwich Village
Pasadena
0.
0.
0.
1.
1.
1.
1 .
1.
1.
2 _
3.
Air
Gug/m3)
17
32
85
13
15
J8 (7 mo. )
19
67
76
08
39

Blood
(mg/lOOg. )
0.0149 (F)
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0157
0125
0153
0180
0139
0192
0205
0176
0166
0175
Blood
Rank
3
5
1
4
9
2
10
11
8
6
7
    Relationship Between Means of Air and Blood Lead (Females)
   Based on Pearson Product Moment Correlation Coefficient (r)
          and Kendall's Rank Correlation Coefficient (T)

                       r  = 0.412 (NS)
                       T  -= 0.354 (NS)
(F)  Females
(NS) Not significantly  different from 0 at  the 5% level.

-------
                                59
                           Table 15


      Air and  Blood  Lead Concentrations  (1968-71)
                      (Non-Smokers Only)
                                     Geometric Means
Air
Rank
1
2
3
4
5
6
7
8
9
10
11
Area
Los Alamos
Okeana
Houston
Port Washington
Ardmore
Lombard
Washington, D.C.
Rittenhouse
Bridgeport
Greenwich Village
Pasadena
Air
(ug/m3)
0.17
0.32
0.85
1.13
1. 15
1.18 (7 mo. )
1.19
1.67
1.76
2.08
3.39
Blood
(mg/lOOg. )
0.0144 (F)
0.0156
0.0126
0.0149
0.0180
0.0136
0.0181
0.0202
0.0167
0.0158
0.0172
Blood
Rank
3
5
1
4
9
2
10
11
7
6
8
      Relationship Between Means  of Air and Blood Lead  (Females)
     Based on Pearson Product Moment Correlation Coefficient  (r)
           and Kendall's Rank Correlation Coefficient  (T)

                         r  = 0.400 (NS)

                         T  = 0.345 (NS)
 (F)  Females
(NS)  Not significantly different from 0 at the 5% level.

-------
                                      60
measured outdoors.  It is not immediately apparent that certain groups spend
relatively greater amounts of time outdoors.  If this point were significant,
presumably the Pasadena population would experience relatively greater outdoor
exposure.  Accordingly we do not feel  that a consideration of indoor-outdoor
time or air conditioning alters the principal  inferences drawn from this study.
                                  Figure 20
                     Blood Lead Levels and Corresponding
                            Mean Air Lead Levels
           0.03

           0.02
         i
            O.Gl
             0
                        •?if)
                              •c:
                              c;
^

.o
C;
BLOOD LEAD LEVELS
AS A FUNCTION OF
AIR LEAD LEVELS
() - number of people
                       ;


"S
1 J
-t 1 ~^
1.0
<£
!
2.0


3.0

- 1
4.0
                            LEAD IN AIR, jjg/m3
                   Male-Female Blood Lead Differentials

     The examination of blood lead levels in husband-wife pairs in Los Alamos
revealed a significant difference between males and females (Table 13).  The
basis for this difference remains to be established.  We believe that it cannot
be attributed to occupational factors, for the work of each individual was char-
acterized and conducted under comprehensive and detailed industrial  hygiene
supervision.   It is generally true that men smoke more than do women; however,
the blood lead levels in males remained higher than in female smokers and non-
smokers .

     Since men  have generally higher hematocrit levels than do women and since
lead is transported in association with red blood cells, the possibility exists
that higher blood lead concentrations in men reflect a higher level  of circu-
lating red cells.  An associated question related to whether or not relatively
higher hematocrits in women of Los Alamos (elevation:  7,300 feet) might be re-
sponsible for their blood lead levels being higher than would be the case at sea

-------
                                      61


 level.  The anticipated gradient in hematocrits, Los Alamos males > Los Alamos
 females > Pasadena females, was observed (Table 16).  The analysis of variance
 testing the significance of the difference among the hematocrit values of these
 three groups was significant at the 1% level  (F = 80.26).  The comparison be-
 tween Pasadena and Los Alamos females also was significant at the 1% level,
 (F = 11.54), the mean value of Los Alamos females being higher than that of
 Pasadena females.

     The functional relationship between blood lead and hematocrit level was
 examined within each of the three groups separately (non-smokers only) to see
 if an "adjustment" could be made in the blood lead values which would make it
 possible to obtain better comparisons among the three groups (Table 16).  In
 Los Alamos, the regressions were not significant; therefore it was concluded
 that no significant association between blood lead determinations and hemato-
 crit values could be found among Los Alamos males or females.  In Pasadena a
 significant second-order regression was found which is difficult to interpret
 as far as the effect on blood lead values is concerned.  For this reason no
 "adjustment" for hematocrit level  was made on the blood lead values of this
 area.

     Since the mean blood lead value for Los Alamos is lower than that of Pasa-
 dena and since there was no significant (linear) correlation between blood lead
 and hematocrit level, the relative size of the mean blood lead levels in Pasadena
 and Los Alamos(F). would not be effected by an "adjustment" due to the average
 hematocrit levels of these two populations.

     It is also worth mentioning that although both men ^and women of the studied
 Los Alamos couples consumed diets  presumably similar in composition, the men
 consumed a generally greater quantity of food.  It is therefore likely that the
 alimentary exposure of men to lead may be greater than that of women.

                                Fecal  Lead Levels

     The analysis of fecal  lead levels, (Table 17)  conducted to appraise one of
 these variables, viz., alimentary  lead intake, did  not demonstrate high or low
 levels compensating for respiratory exposures so as to obscure effects of the
 latter.   That is, the failure of the Pasadena group to show a "high" blood lead
 level was  not due to a low alimentary lead intake.   Conversely the failure to
observe "low" blood lead levels in Okeana or Los Alamos was not due to a com-
pensatory  high food lead level.   It is of interest  that the Houston fecal  level
of lead was  relatively high while  the blood lead level  was relatively low.   This
observation  was not further investigated  in this study.

     In each area the alimentary level  of lead appeared to be much lower than
the 300 jug/day commonly quoted as  the mean lead intake.   This higher number is
based upon  studies  of men,  who very probably ingested larger amounts of food
than  did the women  in this  study.   Nevertheless, the somewhat lower food intakes
by women cannot account for the discrepancy.   It is possible that current im-
proved food  manufacturing  practices are associated  with reduced lead levels  in
foods.


                                                Text continued on  page  64

-------
                             62

                         Table 16
  Frequency Tables Showing Distribution of Hematocrit Values

                   % Relative Frequency
Hematocrit
Interval
<34
34-35
36-37
38-39
40-41
42-43
44-45
46-47
48-49
S50
Hematocrit
Location
Pasadena (F)
Los Alamos (F)
Los Alamos (M)
Pasadena
(F)
0
4.9
4.4
21.2
29.1
18.7
15.8
2.5
3.0
0.5
Mean Values and
Los Alamos
(F)
1.0
1.0
4.5
11.9
17.8
38.1
16.3
6.4
2.0
1.0
Los Alamos
(M)
0
0
0
2.0
1.0
17.3
34.7
24.5
15.3
5.1










Their 95% Conference Limits
N Arithmetic Mean
203
202
98
40.97
41.97
45.55
95% Confidence Limits
(40.54, 41
(41.55, 42
(45.03, 46
.40)
.39)
.07)
Analysis of Variance Testing Differences in Hematocrit Levels

     Source      df       SS_          MS         F

     Groups      2      1418.62     709.31     80.26**

     Error      500     4418.94       8.84

        Further Comparison Between Hematocrit Levels

     Comparison           F_                Direction

   Pasadena  (F)          11.54**       Los  Alamos >Pasadena
        vs
   Los  Alamos  (F)

   ** Significant  at the -1%  level

-------
                                      63
                                  Table 17

                 Analyses of Functional Relationship Between
                  Blood Lead and Hematocrit for Non-Smokers
Pasadena (F)
  N = 131
  R - 0.26
Los
  N
  R
Los
  N
  R
Alamos
= 104
= 0.18
Alamos
= 32
= 0.44
(F)
(M)
 *Significant at 5%
**Significant at 1%
Due to regression         3
    Cubic                 1
    Quadratic             1
    Linear                1
About regression        127

Due to regression         3
About regression        100

Due to regression         3
About regression         28
                level
                level
                                                0.1393
                                                0.0008
                                                0.1381
                                                0.0004
                                                  9346
                                                    1
0.0435
1.3027

0.0933
0.3858
0.0464
0.0008
0.1381
0.0004
0.0152

0.0145
0.0130

0.0311
0.0138
                                                              3.05*
                                                              0.05
                                                              9.08**
                                                              0.02
1.11 (NS)


2.25 (NS)
                                  Table 18
               Geometric Averages  of Fecal  Lead Ten  Day Totals
                        Area

                 Okeana*
                 Ardmore
                 Los  Angeles
                 Los  Alamos  (F)
                 Washington, D.C.
                 Port Washington
                 Lombard
                 Houston
                                     20
                                     20
                                     20
                                     20
                                     20
                                     20
                                     20
                                     20
                                              Fecal  Lead
                                               (mg/10d.)'
                                          0.852
                                            139
                                            147
                                          0.973
                                          1.168
                                          0.967
                                          0.880
                                          1 .505
                 *Total  for donor #0112  adjusted  to  account  for  two
                  unusually high  values.

                     Analysis  of  Variation  of  Fecal  Lead
                       (Ten Day Total) Due  to  Location

                 Source      df         SS^       MS_          £

                 Location      7       0.9133    0.1304   2.00 (P  =  0.08)

                 Error       152       9.9089    0.0651

-------
                                      64
                    Results of Analysis of Pennypack Data


     In April 1961 and'April 1971 blood samples from 23 persons in Pennypack
were obtained.  (Each person was sampled twice:  once in 1961  and once in 1971).
The geometric mean of 1961 blood lead values is 0.0124 mg/lOOg; the geometric
mean of 1971 values is 0.0138.  The ratio of 1971  to 1961 mean values  is  0.0138/
0.0124 = 111/7..  The result of a paired T-test performed to test the significance
of the difference between mean blood lead values in 1961 and 1971  yielded T =
1.764, df = 22.  This value would occur by chance with probability P = 0.09
if in fact the mean lead values in 1961 and 1971 were equal.  It was therefore
concluded that there may be a real difference between them;  more evidence is
needed to make definite conclusions.  Interpretations must take into account
the fact that the sampled participants were ten years older  in the latter study
than they were in the former.

                  Results of Analysis of Alpine County Data

     In 1960 and 1971 blood samples from two independent samples of males and
females from Alpine County, California were measured and recorded.(2,6)   yne
geometric means of each group are as follows:

                 Mai es         Fema 1 es        Sex Unknown        l£ifll

     1960        0.012          0.009            0.013           0.011
                  (16)           (11)             (10)            (37)

     1971        0.0201          0.0159             -             0.0177
                  (18)           (21)                             (39)


     The standard errors of the 1971 mean values are 0.0205  for males, and
0.0223 for females.  No data on the variation of 1960 values are available.

     To determine if there is a difference in the 1971 blood lead values  of
males and females in Alpine County an unpaired T-test was performed.  The re-
sultant T-value was T = 3.393, df = 37.  This value would occur by chance wit'
probability P = 0.002 if in fact the mean lead values of males and females wp'-r
equal.  It was therefore concluded that the difference between them is high1/
significant, that of males being higher.  The ratio of 1971  male to female me*
values is 0.0201/0.0159 = 126.42%.

     Note:  The F-statistic testing the equality of variances  yielded  F  = i.397
(df = 20, 17), P = 0.245.

     To determine if there is a statistically significant difference between
the average 1960 and 1971 blood lead values, 95% confidence  intervals  about the
1971  mean values were found.  These are, for males and females, respectively,
(0.0182, 0.0222) and (0.0143, 0.0176).  Since the 1960 mean  values, 0.012 and
0.009 respectively, do not lie within these intervals, we conclude that  the
1960 and 1971 mean blood lead values are significantly different at the  5%
level; for both males and females, the 1971 mean values are  higher than  those
in 1960.

-------
                                      65
      Continuous air sampling over a 6-month period at a representative site in
 Alpine  County permitted the mean atmospheric lead level to be determined:
 0.03 jLiq/m^.  There is no reason to believe that a higher or lower level  pre-
 vailed  in  1960.

      In each of these 2 situations the higher 1971 blood lead determinations may
 reflect an actual increase or an apparent increase due to differences in method-
 ology used in the two study periods.  Standards are not available which  would
 permit  a longitudinal comparison of methods.  For the Alpine group, the  differ-
 ence  is very likely a reflection of method, for there has been no apparent change
 in the  region which would account for the observation.  Accordingly the  1960
 levels were too low, or the 1971 levels were too high.  Insofar as we can deter-
 mine, from interlaboratory comparisons and from an inspection of data generated
 in association with the original Alpine County survey" and in this study, the
 1971  values are of relatively greater validity.  There is no evidence that the
 regional sources of food have changed during the interval between the studies.
 The possibility that cultural and dietary habits of the 2 sampled populations
 are dissimilar cannot be excluded from consideration.

                    Results of Analysis of Policemen Data

      In March 1962 and March 1971 blood samples from 45 policemen in Cincinnati
 were obtained.  (Each policeman was sampled twice:  once in 1962 and once in
 1971.)  Blood lead values were measured and recorded.  The geometric mean of
 1962 blood lead values is 0.0235 mg/lOOg; the geometric mean of 1971 values is
 0.0179 mg/lOOg.  The ratio of 1971  to 1962 mean values is 0.0179/0.0235  - 76%.
 In order to determine whether the average blood lead value in 1971 is signifi-
 cantly different from that in 1962, a paired T-test was performed.  The  re-
 sultant T-value was T = 8.579, df = 44.  This value would occur by chance with
 probability P = 0.00 (to two decimal  places)  if in fact the mean lead values
 in 1962 and 1971  were equal.  It was therefore concluded that the difference
 between them is highly significant.  Here again an interpretation requires con-
 sideration of the fact that the policemen were older in 1971  than in 1962.  The
 job assignments of a number of policemen had changed over the years; si  e men
 had changed from traffic duty to assignments in police cars.

     The significance of data derived from the studies of Cincinnati police,
 Pennypack residents, and Alpine County residents is unclear.  There are no direct
 standards which would permit comparison of 1961-62 and the later blood lead as-
 says.   Any interpretation of the observed differentials must take into consid-
 eration the fact that the determination of lead concentrations in biological
materials is  a difficult analysis and that experienced laboratories report vari-
able results  when presumably identical  samples are assayed at different  times
or under different circumstances.   Accordingly,  longitudinal  comparisons  of
 blood  lead levels may contain random and systematic errors which complicate in-
 terpretations.   While it would seem desirable to run daily standards,  it  is an
unfortunate fact that blood "standards" are often  of dubious  uniformity  and
 that standards which are introduced into only the  final  step  of the determina-
 tion fail  to  reflect the ashing or  sample preparation steps.

     Experience suggests that day-to-day systematic variability may reduce the
validity of comparison studies when samples  from different groups are assayed
at different  times.   Consequently it  would seem appropriate to intermix  samples

-------
                                      66


from groups to be compared so that samples from all  groups are assayed under
the same conditions.   Such a procedure would dictate that all  samples be col-
lected prior to their intermixing and assay.  The impracticality of this approach
was apparent in this  study.   Insofar as possible, however, urban and suburban
samples from the same metropolitan area were intermixed and assayed simultane-
ously.

-------
                                      67


                                 REFERENCES


1.  Public health aspects  of increasing tetraethyl lead content in motor fuel,
    U.S. Dept.  Health,  Education,  &  Welfare, PHS Pub. No. 712, Washington, 1959.

2.  Survey of lead in the  atmosphere of three urban communities, U.S. Dept. Health,
    Education,  & Welfare,  PHS Pub. No. 999-AP-12, Washington, 1965.

3.  Pierce, J.O., and Meyer, J.H., Technical note.  Sampling and analysis consid-
    erations in evaluating levels  of atmospheric lead, Atmos. Environ. 5:811,
    1971.

4.  Bambach, K. and Burkey,  R.,  Microdetermination of lead by dithizone, Ind.
    Eng. Chem.  Anal., 14:904, 42.

5.  Shearer, S.D., Akland, G.G.,  Fair, D.H., McMullen, T.B.  and Tabor, E.G.,
    Concentrations of particulate  lead in  the ambient air of the United States,
    Division of Atmospheric  Surveillance,  Environmental Protection Agency, Re-
    search Triangle Park,  1972.

6.  Lead in the environment  and  its  effect on humans, Department of Public Health,
    State of California, Berkeley, 1967.

-------
                               68
                           APPENDIX 1
Project Group - University of Cincinnati

Analytical Procedures
     Jacob Cholak, principal
     Roland E. Burkey
     Bernard G. Meiners
     James 0. Pierce
     Bernard E. Saltzman
     Lawrence J. Schafer
     David W. Yeager
Atmospheric Studies
     Lloyd B. Tepper, principal
     Jacob Cholak
     Bernard G. Meiners
     Jozef Svetlik
Data Management and Statistical Evaluation
     Joseph H. Meyer, principal (1968-69)
     Linda S. Levin, principal  (1970-72)
     C. Ralph Buncher, consultant
     Katherine Hayes
     Kathleen Poskus
     Lawrence J. Schafer
Population Studies
     Lloyd B. Tepper, principal
     Eva C. Petering
Subcommittee for the Surveillance of Air and Population Lead Levels
     Robert J.M. Horton, Chairman, Office of Air Programs, Environ-
       mental Protection Agency
     Vincent J. Castrop  (General Motors Corporation),  Automobile
       Manufacturers Association
     Jerome F. Cole, International Lead Zinc Research Organization
       (from 1969)
     Robert E. Eckardt  (Esso Research and Engineering Company),
       American Petroleum Institute  (through 1969)
     Don G. Fowler, International Lead Zinc Research Organization
       (through 1968)
     Harold H. Golz, American Petroleum Institute  (from 1970)
     Howard E. Hesselberg, Ethyl Corporation
     Alden J. Pahnke,  E.I. du Pont de Nemours and Company
     Elbert C. Tabor,  Office of Air Programs, Environmental Pro-
       tection Agency

-------
                              69
                          APPENDIX 2
                        Statistical Notes
1)   Blood lead concentrations (mg/100 g)  were measured to three
    decimal places.  Tables in this study show blood lead measure-
    ments to four decimal places.  This has been done to prevent
    rounding to the third decimal place and hence losing accuracy
    in comparing results.

2)   In the analyses of this report, the data are assumed to be
    normally distributed.  Confidence limits and F-tests obtained
    from analysis of variance techniques depend to varyina dearees
    on the validity of this assumption.  It was sugaested by pre-
    vious studies that using the log transformation on blood lead
    values successfully normalizes the data.  To test this assump-
    tion the Kolmogorov-Smirnov statistic was calculated for each
    location separately.   (Actually a Z-transfornation of this
    statistic valid for large samples was used.)  This statistic
    was used to test the  normality of the logarithm of the blood
    lead values in each area.  The mean values and standard de-
    viations of the normal curves against which the observed data
    were tested were obtained from the observed data in the usual
    manner; that is:
                              X =
EX.
 si
                                     (X.-X)?

                                   ~  N-l --J
              where X.  = logarithm of blood lead.
    The overall a level in these tests has been set at 0.20; each
    individual a level, therefore,  is equal to 0.01.   The table
    below shows resultant Z-values  and their probabilities P(Z).

-------
                              70
        Table of Z-Values Obtained From Kolmogorov-Smirnov
        Tests of Normality of Log^o Lead Concentrations
          Area                        Z_          P (Z) t

          Okeana                    1.055        0.22

          Ardmore                   1.142        0.15

          Rittenhouse               0.695        0.72

          Pasadena                  0.948        0.33

          Los Alamos (F)             1.613        0.01*

          Washington, D.C.          1.384        0.04

          Port Washington           1.180        0.12

          Greenwich Village         1.244        0.09

          Lombard                   1.425        0.03

          Bridgeport                1.290        0.07

          Houston                   1.282        0.07


          Los Alamos (M)             0.696        0.72
     The conclusion from these tests of normality is that there
is one location where the distribution of the logarithm of blood
leads differs significantly from normal; however, since this test
is conservative, one must question normality in those locations
where the Z values have probabilities close to 0.01, namely,
Lombard and Washington, D.C.  In spite of this, there is enough
justification in the table above to assume log normality of the
blood lead data, and for this reason all tables of mean values
present the geometric mean rather than the arithmetic mean, the
geometric mean being the antilog of the arithmetic mean of the
logs.  In addition, all the analyses of variance are done using
log values.


*Significant at 1% level.

tSince two parameters are estimated from the data, it is suggested
 by Siegel, Non-Parametric Statistics, McGraw Hill, 1956, P. 60,
 that this test is a conservative one.

-------
                                   71
                               APPENDIX 3
(Rank)
Area
   5   Okeana




   9   Ardmore




  11   Rittenhouse




   7   Pasadena




   3   Los  Alamos  (F)




  10   Washington,  B.C.




   4   Port Washington




   6   Greenwich Village




   2   Lombard




   8   Bridgeport




   1   Houston






      All  Females







      Los  Alamos  (M)
and Confidence Intervals of
'lOOg) Levels Based on a
nation from Each Subject
L Sample Before Exclusions)
N Geometric Mean 95% Confidence Limits
166
156
137
209
204
224
203
140*
208*
148
204
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0158
0180
0206
0176
0149
0192
0154
0166
0139
0176
0124
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
0150,
0171,
0194,
0168,
0144,
0186,
0149,
0159,
0135,
0169,
0120,
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0166)
0190)
0218)
0184)
0154)
0198)
0159)
0173)
0144)
0183)
0129)
             1999
              101
0.0162
 0.0172
 (0.0160, 0.0164)
(0.0164,  0.0180)
      *No change in N.

-------
                                                                  72



                                                           APPENDIX   4
              0)
              o

              CO
                               CM
                               CO
                                      lO
                                                    Ol     rH
 (0
 o

 CO
 c
 

-------
                                                         73

                                                   APPENDIX  5
                         00     t»    CM
                               CM    T}<
 0
 CO
 3
                               O
                               CM
                                     CO
                     o
                     OJ
                                 CM
•o
a
CO
 a
 CO
 o
•H
.a
CJ
                        CO
                              CM
                                    CO
                                          CM
                                                O    rH
                                       00
                                       O
                                       CM
r<
o
               a
o
Q

~
a
o
P
b£>
a
•H
A
(0
CO
•e
a
•H
w
0)
rH
CO

O
CK
 oo m T)< ooi o
CM CM r-( i-l ^
1 r^








c
0
4^
M
C
•H
J3
tn
ra
^:

•P
t,
£
f-H




CO rH (^ CM CM irtl 00
CM CM CO f-H ^

1 ,_^







cj

P
fc
e
o
+j
bO
e
•H
j-t
*XH
CQ
ca
=E




r-l t^ rH ^f W r-l Oi
CM CM CM CM i-l
CM




             &
CM
 I
O
CM
                            O5
                            CO
                            I
                            O
                            CO
                                  cn
                                       m
                                       o
                                       m
                                              05
                                              to
o
CO
                                                    o>
                                     CO
                                    •P

                                    £

-------
3 S
O> 2_
- -TO O 2

5 5~-c
10 TO a) i/> a.
— 3 3 CD O

<" CT °" S =
~™ sSs
        0.
  o z "H. o>
  3 D TO w
  "O - 0) -,
  ^ -ou
  fo =r -r n
  3 TO mo
  O.SH
  af^i
  — ,-t, I ro
  o - m -
  3 ;
  CD

  CD
   -i _^ lu -,

  ° O ^ ~
  T3 ° "• tn
  J ^ O _
  CD _ — TO

  n &'~^

    3 ? =
    ^ "> o
        a>
                                    z
                                    <
                                  -H 39
                                  2  O
                      c
                      -2
                      -n
                      r~
                      O
                             -< o  o  5
                                    *
 z S « =!

 °!5i
 M   3 J>
                                    n
                                  <
                                  30
O

p  3
>  >
CO   i
1/5  u>
2  Si
                                  00
                                  o
                                  rr>
                                  2
                                  n

-------