EPA-600/1-76-029
September 1976
ANALYSIS OF BLOOD, HAIR, URINE, AND DUST SAMPLES
FOR HEAVY METALS
By
Anna M. Yoakum
Stewart Laboratories, Inc.
5815 Middlebrook Pike
Knoxville, TN 37921
Contract No. 68-02-2266
Project Officer
Dr. Carl Hayes
Pooulation Studies Division
Health Effects Research Laboratory
Research Triangle Park, N.C. 27711
...UN AGENCY
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
HEALTH EFFECTS RESEARCH LABORATORY
RESEARCH TRIANGLE PARK, N.C. 27711
-------
DISCLAIMER
This report has been reviewed by the Health Effects Research
Laboratory, U.S. Environmental Protection Agency, and approved for
publication. Approval does not signify that the contents necessarily
reflect the views and policies of the U.S. Environmental Protection
Agency, nor does mention of trade names or commercial products
constitute endorsement or recommendation for use.
-------
FOREWORD
The many benefits of our modern, developing, industrial society are
accompanied by certain hazards. Careful assessment of the relative risk
of existing and new man-made environmental hazards is necessary for the
establishment of sound regulatory policy. These regulations serve to
enhance the quality of our environment in order to promote the public
health and welfare and the productive capacity of our Nation's population.
The Health Effects Research Laboratory, Research Triangle Park
conducts a coordinated environmental health research program in toxicology,
epidemiology, and clinical studies using human volunteer subjects. These
studies address problems in air pollution, non-ionizing radiation,
environmental carcinogenesis and the toxicology of pesticides as well as
other chemical pollutants. The Laboratory develops and revises air quality
criteria documents on pollutants for which national ambient air quality
standards exist or are proposed, provides the data for registration of new
pesticides or proposed suspension of those already in use, conducts research
on hazardous and toxic materials, and is preparing the health basis for
non-ionizing radiation standards. Direct support to the regulatory function
of the Agency is provided in the form of expert testimony and preparation of
affidavits as well as expert advice to the Administrator to assure the
adequacy of health care and surveillance of persons having suffered imminent
and substantial endangerment of their health.
The chemical analyses provided under this contract support a
collaborative survey by the Center for Disease Control and the Environmental
Protection Agency to assess metal absorption in children living in the
vicinity of primary non-ferrous smelters. The results of the overall survey
will be reported separately.
H. Knelson, M.D.
iDirector,
Health Effects Research Laboratory
-------
ABSTRACT
Communities from ten states in the United States and two
cities in Mexico were studied. The communities were chosen for
their proximity to primary non-ferrous smelter industries.
Three lead and five zinc smelter areas were sampled for blood,
hair, and dust. Urine, blood, hair, and dust were collected from
fourteen copper smelter sites and four control cities.
Samples were analyzed for arsenic, lead, cadmium, copper and
zinc.
IV
-------
TABLE OF CONTENTS
Page
I. Introduction 1
II. Summary 2
III. Conclusions 4
IV. Recommendations 5
V. Discussion of Analytical Approach and Results 11
A. Preliminary Treatment for Sample Preparation 11
B. Special Research Studies 44
C. Analytical Methodology and Discussion 46
D. Review of Quality Control Program 51
E. Assessment of Analytical Data 85
-------
I. INTRODUCTION
Documentation of excessive absorption of heavy metals by children
living near nonferrous metal smelters in the U. S. A. and Canada
activated a nationwide study to determine the distribution of several
heavy metals around primary lead, zinc, and copper smelters. Pre-
school children were selected as test subjects because signsof absorp-
tion of heavy metals are more likely to appear with them than with
older children or adults. The purpose of this project was to analyze
tissue and dust samples collected from test subjects living within
a two mile radius of each smelter so that heavy metal absorption could
be evaluated.
Control sites were selected to provide background reference data as
to the existing level of these heavy metals in areas not associated
with nonferrous smelters.
-------
II. SUMMARY
The study sncompassed collections from ten states in the U. S. A.
and two cities in Mexico. Tissue and dust samples from preschool
children and their homes were differentiated for analysis in accord
with the type smelter or control being evaluated.
Three lead and five zinc smelter areas were sampled for blood, hair,
and dust. Urine, blood, hair, and dust were collected from fourteen
copper smelter sites and four control cities.
All dust samples were analyzed for As, Pb, Cd, Cu, and Zn. Hair
samples from lead and zinc smelter sites were analyzed for Cd and Pb;
while copper smelter sites and control areas received As, Pb, and Cd
determinations. Urine collections (copper smelter sites and control
areas) were measured for specific gravity and As. Lead and FEP were
analyzed in blood from all three smelter types. Additionally, Cd
was included for lead smelter sites; Cd and Zn for zinc smelter sites;
and Cu and Zn for copper smelter sites. Control bloods were tested
for all five blood parameters.
Internal and external quality control programs were integrated with
sample analyses. Accuracy and precision data derived from these pro-
grams are reported and analyzed in depth.
Analyses for each collection site were reported in progressive order
in accord with a numeric field assignment for individual subjects from
-------
each participating family. The laboratory analyst and date for each
determination, including quality controls, was shown within this
format.
-------
III. CONCLUSIONS
The quality of the analytical results produced during this
study has been evaluated in terms of uncertainty measurements
which represent at least 95 percent confidence interval based on
measurement error and variability between samples. The
detrimental effects contributed by the samples themselves were
successfully overcome in all sample types except urine. Overall
accuracy and precision, expressed as relative percent, for each sample
category is as follows: dust, + 3%, hair, + 5%, blood, + 10%,
and urine, +49%.
-------
IV. RECOMMENDATIONS
Based on the experience gained in the performance of this contract,
the following recommendations are suggested for the improvement of
future projects of a similar nature:
A. Dust - The smear technique employed in the collection of the dust
samples has two significant shortcomings.
A sample of this type does not lend itself to any standard
analytical quality control program. Since the entire sample
cannot be removed, blind sample splits and recovery studies
cannot be conducted.
Data obtained can be reported only in terms of micrograms of
acid soluble metal per towelette (or per sample). Since the
actual amount of sample collected varied widely, the ability
to express the data as micrograms of metal per gram of dust would
have allowed for an exact comparison of analytical data between
collection sites, rather than the relative comparison which results
from the method employed.
From a collection standpoint, dust is, perhaps, the most difficult
sample type addressed in this project. Paint and other materials
heavily laden with metals may be preferentially dissolved by the
organic liquids in the towelettes. The abrasive nature of the
sampling procedure contributes to the incorporation of these con-
-------
taminants in the samples, and there is no procedure for obtain-
ing an initial weight of the dust as it is collected.
A recommended dust collection technique, which the contractor
has used, involves dusting a surface of sufficient size to render
a sample adequate for complete analysis. A small camel's hair
brush is used to sweep the sample into a small whirlpak. This
disallows external contamination from the hands of field personnel,
surface finishes, and/or the implement used for collection. The
brush must, however, be thoroughly cleaned after each use.
B. Urine - In order to combat the tendency for urine samples to
preferentially precipitate with time lapse and temperature varia-
tions, a change in the collection protocol would be necessary.
Specific gravity measurements should be performed in the field
as each sample is collected.
Since time lapse and temperature variations occur during storage
and transport to the laboratory, urine samples should be shaken
and aliquoted in the field as soon after collection as possible.
If blind splits are to be made, this should also be performed at
that time. This procedure would allow for homogeneity of each
sample, and it would assure uniform analytical handling.
C. Blood - Since clotted blood does not lend itself to uniform sample
aliquoting techniques, two avenues are open as approaches to pre-
clusion of this problem. First, sample splits could easily be
made from the syringe while the blood is fresh and not clotted.
-------
Secondly, a more uniform mixing with an anticoagulant could be
applied. The former would, at this time, appear to be the safest
and simplest.
All blood samples should be ejected from the syringe into containers
appropriate for shipment. Inverted, leaky syringes held together
with various types of tape do not constitute proper shipping vessels.
D. Hair - The extreme variation of metals content in hair indicates
that small samples are not representative of the overall hair of
the subject. Thus, a concerted effort should be made to collect
as large a sample as possible from each subject.
Zip-loc bags provide excellent transport containers; however, they
should be locked carefully as soon as they receive a sample of hair.
A secure seal allows for transport in a condition isolated from
external contamination. Each hair sample should be handled with
gloves and completely enclosed within the plastic bag.
Human scalp hair is widely used for assessing environmental ex-
posure to metals. Trace metal content of hair is, likewise,
reported to be an indicator of deficiency conditions in both humans
and test animals. Because of these considerations and because of
the ready availability of hair as a biopsy tissue, scalp hair lends
itself well to environmental monitoring of humans for trace metal
exposure.
A number of problems are apparent which could have significant
influence on the interpretation of data for hair as a valid
-------
biological indicator for environmental exposure. Findings from
a series of pilot research studies conducted by the contractor,
as in-house satellite investigations, clearly indicate the need to
conduct a more closely controlled investigation in the area of
trace metal analysis of hair. There is, thus, a need for the
design of an analytical study to define, investigate, and establish
the parameters which can adversely influence the interpretation of
hair data.
The recommended research should study, for a period of at least
two years, the trace metal content of scalp hair samples collected
under carefully controlled conditions from a healthy population
living in an area free from major point source atmospheric metal
contamination. All pertinent background information relating to
the sampling population should be made available so that relation-
ships between trace metal contents and personal covariants can be
evaluated. This requires an extensive, well-administered question-
naire. Participation in the program should be limited to those
subjects who are willing and cooperate fully in providing all
pertinent background information to be included in the study;
therefore, the selection of the sampling population should be care-
fully controlled to assure that all vital areas are included in the
study.
Since factors other than environmental exposure can play a signifi-
cant role in trace metal concentrations and distributions, the study
should address itself to the following potential factors:
-------
(1) Seasonal variations—i.e., hair growth rate and
frequency of hair care.
(2) Hair care—i.e., types and frequency of use of
chemical products and their contribution to
elemental alteration of hair.
(3) Inherent variabilities—i.e., physical and physio-
logical hair characteristics and metals content along
single strands of hair.
(4) Time effects—i.e., storage time for washed versus
unwashed hair prior to analysis.
(5) Participant characteristics—i.e., race, medical
treatment, residence location, smoking habits, diet,
sex, age, etc.
E. General Sampling Protocol - Consistency in adherence to a sampling
protocol at each collection site is essential from a quality con-
trol standpoint. Certain basic guidelines should be followed in
order to obtain data which can be compared directly and correlated
with known varying parameters. These parameters should have been
primarily limited to 1) geography, baseline conditions, and
environmental exposure characteristics and 2) inherent biological
uptake of heavy metals.
In order to produce data which represent a real profile of these
conditions, one should properly and legibly code and permanently
-------
label each sample as it is collected. Each individual sample
should be treated precisely the same as all other samples of
that particular type. The protocol should be realistic in terms
of maximum time required to perform each field procedure. For
example, frozen samples shipped over long distances should arrive,
at the laboratory, in the same condition as those shipped from
proximal collection sites.
Provided the recommended protocols for field sample procedures,
as discussed for each sample type, are followed, all collections
could be shipped directly to participating laboratories on an
individual basis. This would eliminate variables such as time
differentials, thawing of samples, inhomogenous aliquots, and
misplacement of partial batches.
Also, uniformity of sample sizes submitted for analysis should be
held at a reasonable tolerance level. All containers to be used
in a project of this type should be procured from single line suppliers
and, preferably, from a given manufacturing batch or lot.
Shipment schedules for submission of samples for analysis should be
planned in advance and modes of transport carefully investigated.
A continuous rapport and conscientious use of quality control among
persons responsible for field sampling, laboratory analyses, and
project management are imperative in the obtainment of valid and
useful analytical data.
10
-------
V. DISCUSSION OF ANALYTICAL APPROACH AND RESULTS
A. Preliminary Treatment for Sample Preparation - This project included
receipt of human tissue and house dust from twenty-four sampling
sites. Of these sites, three were near lead smelters, five near
zinc smelters, twelve near copper smelters, and four were from con-
trol areas (Table 1). Because of deviations from the prescribed
sampling protocol, a laboratory screening of those samples not
meeting the minimum requirements for analysis was established. The
criteria on which samples were rejected are shown in Table 2. The
acceptable size limits were prescribed in the contract; however,
additional measurements were required in the laboratory in order to
differentiate samples which were very near the limit and could not
be visually determined. The various procedures for handling these
samples will be discussed on an individual basis for each sample
type, and the rejections made for each collection site will be
enumerated and related to their cause.
1. Procedures for Sample Screening—Each shipment contained
samples from several collection sites and, therefore, required
a preliminary sorting and assignment to an organized sequence
for analysis. Once this process had been achieved, each
sample type was individually screened for analytical accept-
ability.
a. Blood—The tubes were checked for cracks and/or probable
contamination, and notations were recorded when appropriate.
All samples which were not whole blood or could not be
11
-------
H
C/3
C
Id
Z
a.
en
^f
S
|
co
^
Id
os
^
O
z
M
^J
§
U3
Q
, T
Z
o
(d
t— 1
w
o
tn
Q
to
^
H-l
z
Q
M
1
c
S3
nJ
Z
O
Id
Id
O
Id
0£
O
Id
NJ
|
0
Id
M
Id
CJ
Id
03
Ed
C
8
Z
O
H
^
o
B
ooo o o o o o cccoococoooo
p^O-f m 4 0 cc CN &>cM-*<*r^ino^fMfn<-
2£ *H c*J Z F^ £
* UJ W P. X H ...
£ i—t*H EH r^S »cn-HtU 00 < - CD 3-HX--H
CO * tfl .4-1^ . iH f-3 *C » C A- (fii-i»
i— H * u^ Qj w tfl tj i— ( c "U^c^^c v ff O-i r*~.
D>I 4J&.CdE^ •OEtU'-Ui-' AJO-OCCOi-H
>_iXO J-iMCi-icD >^CC1-'OP3OC'HD.33I-I
!U*Hi-i cococoE nji-fo— ic:ore,^:oonr3
xcao CQUSO-< xs:E<j ,-H i— t f^.
\D CO lA Qv
^ c^ o -H
m \o ft o
i— i
r^ \o m \c
9^ 00 00 O
< > i-J U
t/3 A. < Z
£ O
O Z U
* X
Cv3 . (y OJ
< OJ 3 Z
•H cr
* r-l l-i "
•O -H CD tn
M > D OJ
O 5^ CTrH
y_l ^j trt QC
ffl Q) T-H O
CD GH < Z
i-H f-t c-t •-!
1-. kJ J-. J-
J_) 4-J .U -tJ
C C C C
O 0 O O
u u u u
12
-------
TABLE 2. Criteria for Rejection of Samples
from the Project.
Sample Tissue Type
Blood
Hair
Urine
Dust
Criteria for Rejection
< 1 ml
not complete whole blood
indeterminant code identification
< 0.2 gm
indeterminant code identification
< 25 ml
indeterminant code identification
samples not contracted
taken on materials other than
towelettes
indeterminant code identification
13
-------
identified from labels were counted and discarded.
After thawing, the blood volumes were individually compared
visually against a reference 1 ml blood sample. Those
samples determined to be <1 ml were counted and discarded;
however, because of excessive clotting and the large number
of samples which were close to the 1 ml volume, some samples
were processed and analyzed for FEP and later discarded for
further analysis. This was done after the blood was trans-
ferred to a graduated cylinder for volumetric measurement.
FEP determinations were made prior to this transfer because
of the prescribed time requirement in the method. All
blood samples which had been rejected were reassembled and
enumeration checks were finalized.
b. Hair—Since hair and dust were frequently shipped in bags
together, a separation and verification of code assign-
ments of these samples was necessary prior to screening.
Samples obviously too small and/or with unidentifiable
labels were rejected. Improperly sealed bags were noted
as were containers other than Zip-Iocs. A second screen-
ing of hair was attempted by weighing borderline samples
on a top loading balance. If the weights were between
0.20 and 0.25 gm, the samples were reentered and coded for
analysis. Some of these samples fell below the 0.2 gm
limit when reweighed on an analytical balance after they
had been cleaned and dried. These final discards were
added to all of the hair previously rejected, and proper
notations were recorded.
14
-------
c> Urine—Some shipments contained urine from lead and zinc
smelter sites, which were not to be included for analysis
under this contract; therefore, these samples were rejected.
Each of the samples included for analysis was compared,
after thawing, to a reference urine of a 25 ml volume. All
samples failing to meet this criterion were enumerated and
rejected.
d. Dust—After separation from hair samples (see Hair above),
dust was observed for adherence to sampling protocols and
analytical preferences. Since only ten samples per collec-
tion site were to be chosen, properly sealed bags with no
evidence of contamination received top consideration.
Apparent sample size and legibility of label codes were
then utilized as secondary criteria. One entire shipment
of dust from Corpus Christi, Texas, was discarded because
the samples had been taken on cotton balls and blanks were
unavailable. Rejected dust samples were enumerated after
all these selection criteria had been imposed.
2. Sample Conditions and Handling Techniques—The sample conditions
and the various procedures for handling these samples are pre-
sented on an individual basis for each sample type according
to collection site. A summary of samples received versus samples
analyzed is given in Table 1.
a. Lead Smelters—The lead smelter samples were to be analyzed
for: FEP, Pb, and Cd in blood; Pb and Cd in hair; and Pb,
15
-------
Cd, Zn, Cu, and As in dust. Of the 187 bloods received,
162 were completely analyzed; of the 208 hair samples,
126 were completely analyzed; and of the 158 dust samples,
the prescribed 30 were appropriate for complete analysis.
The following is a resume of the condition of sample ship-
ments and preliminary analytical treatment of samples
received from each lead smelter.
16
-------
HERCULANEUM, MO
General - Samples received 9/18/75 with no instructions
therefore all samples initially coded
Blood - All thawed; 28 badly clotted therefore duplicate
FEP's run; condition required special mixing for
Cd; 11 run for FEP only
Hair - Zip-Iocs sealed; 36 rejects
Urine - Received 108 not to be analyzed—all were coded
Dust - Zip-Iocs not sealed; 77 screened
17
-------
BIXBY, MO
General - Samples received 9/25/75
Blood - Partially thawed; 2 badly clotted therefore
duplicate FEP's run; condition required special
mixing for Cd
Hair - Zip-Iocs and sealed; 39 rejects
Urine - Samples received but not to be analyzed
Dust - Zip-Iocs and sealed; 30 screened
18
-------
GLOVER, MO
General - Samples received 9/25/75
Blood - Partially thawed; 1 badly clotted therefore
duplicate FEP run; condition required special
mixing for Cd
Hair - Zip-Iocs and sealed; 7 rejects
Urine - Samples received but not to be analyzed
Dust - Zip-Iocs and sealed; 21 screened
19
-------
b. Zinc Smelters - The zinc smelter samples were to be analyzed
for FEP, Pb, Cd, and Zn in blood; Pb and Cd in hair; and
Pb, Cd, Zn, Cu, and As in dust. Total analyses were com-
pleted for 370 of the 415 bloods received, 297 of the 445
hair samples, and 30 of the prescribed 50 from a total of
241 dust samples. A resume of each zinc smelter follows.
20
-------
BARTLESVILLE, OK
General - Samples received 9/18/75 with no instructions
therefore all samples initially coded
Blood - All thawed; all in inverted syringes; 13 with
white adhesive tape, 59 with black electrical
tape, 18 with scotch tape; 3 badly clotted
therefore duplicate FEP's run; condition
required special mixing for Cd; 3 for FEP only
Hair - Alligator sandwich bags with twist ties; 15 rejects
Urine - Received 75 not to be analyzed—all were coded
Dust - Alligator sandwich bags with twist ties, towelettes
very dry; 55 screened
21
-------
CORPUS CHRISTI, TX
General - Samples received 11/18/75 and 11/20/75; all
samples had to be re-coded from information
received from CDC
Blood - Partially or completely thawed; received 51
samples, 38 of which were rejects because they had
been spun down and only plasma separates were
shipped; condition required special mixing for
Cd on the 13 whole blood samples
Hair - Zip-Iocs sealed; 35 rejects
Urine - Samples received but not to be analyzed
Dust - Zip-Iocs sealed but all 46 were rejects because
they had been taken on cotton balls instead of
towelettes
22
-------
MONACA, PA
General - Samples received 11/18/75, 12/5/75, and 12/16/75
Blood - First and second shipments partially thawed—third
shipment (4) not iced at all; 31 badly clotted
therefore duplicate FEP's run; condition required
special mixing for Cd; third shipment (4) received
after coding, causing handling and data processing
problems because of skips in numerical sequence
Hair - Zip-Iocs and sealed; 13 rejects
Urine - Samples received but not to be analyzed
Dust - No samples received
23
-------
PALMERTON, PA
General - Samples received 1/9/76
Blood - Frozen; 2 badly clotted therefore duplicate
FEP's run; condition required special mixing
for Cd
Hair - Zip-Iocs sealed; 33 rejects
Urine - Samples received but not to be analyzed
Dust - Zip-Iocs sealed; 68 screened
24
-------
AMARILLO, TX
General - Samples received 1/9/76
Blood - Frozen; 1 badly clotted therefore duplicate
FEP run; condition required special mixing for
Cd
Hair - Zip-Iocs mostly sealed; 52 rejects
Urine - Samples received but not to be analyzed
Dust - Zip-Iocs mostly sealed; 42 screened
25
-------
c. Copper Smelters - The copper smelter samples were to be analyzed
for: FEP, Pb, Zn, and Cu in blood; Pb, Cd, and As in hair;
specific gravity and As in urine; and Pb, Cd, Zn, Cu, and
As in dust. Of the 1046 bloods received, 1026 were completely
analyzed; of the 1156 hair samples, 951 were analyzed; of the
977 urines, 793 were analyzed; and of the 812 dust, the pre-
scribed 120 were appropriate for complete analysis. A resume
of each copper smelter follows.
26
-------
HAYDEN, AZ
General - Samples received 9/18/75 with no instructions
therefore all samples initially coded; all samples
completely mixed with MI
Blood - Partially thawed; labels extremely difficult
to read; 12 badly clotted therefore duplicate
FEP's run
Hair - Zip-Iocs mostly not sealed; 4 rejects
Urine - Partially thawed; 8 rejects; precipitated
Dust - Zip-Iocs mostly not sealed; 59 screened
27
-------
MIAMI, AZ
General - Samples received 9/18/75 with no instructions
therefore all samples initially coded; all samples
completely mixed with HA
Blood - Partially thawed; labels difficult to read; 18
badly clotted therefore duplicate FEP's run
Hair - Zip-Iocs mostly not sealed; 17 rejects
Urine - Partially thawed; 9 rejects; precipitated
Dust - Zip-Iocs mostly not sealed; 62 screened
28
-------
MORENCI, AZ
General - Samples received 9/25/75
Blood - Partially thawed; 7 badly clotted therefore
duplicate FEP's run
Hair - Zip-Iocs sealed; 8 rejects
Urine - Completely thawed; 25 rejects; heavily precipitated
Dust - Zip-Iocs sealed; 64 screened
29
-------
AJO, AZ
General - Samples received 10/15/75 and 10/30/75
Blood - Partially thawed; 8 badly clotted therefore
duplicate FEP's run
Hair - Zip-Iocs sealed; 15 rejects
Urine - Partially thawed; 9 rejects; precipitated; 6
samples received too late to run with scheduled
batch
Dust - Zip-Iocs sealed; 59 screened
30
-------
ANACONDA, MT
General - Samples received 10/30/75
Blood - Thawed; 3 badly clotted therefore duplicate
FEP's run
Hair - Zip-Iocs mostly sealed
Urine - Partially thawed; 33 rejects; precipitated
Dust - Zip-Iocs mostly sealed; 47 screened
31
-------
McGILL, NV
General - Samples received 10/30/75
Blood - Thawed; 4 badly clotted therefore duplicate
FEP's run
Hair - Zip-Iocs mostly sealed; 17 rejects
Urine - Partially thawed; bad leakage in shipping
container; 4 rejects; precipitated
Dust - Zip-Iocs mostly sealed; 35 screened
32
-------
SAN MANUEL, AZ
General - Samples received 10/30/75
Blood - Thawed; 14 badly clotted therefore duplicate
FEP's run; 29 samples sent in inverted syringes,
untaped; tubes cracked and bad leakage on samples
and container; codes very difficult to read
Hair - Zip-Iocs mostly sealed; 97 rejects
Urine - Thawed; slight leakage in container; 32 rejects;
heavily precipitated
Dust - Zip-Iocs mostly sealed; 69 screened
33
-------
WHITE PINES, MI
General - Samples received 11/18/75
Blood - Partially thawed; 1 badly clotted therefore
duplicate FEP run; 1 FEP only
Hair - Zip-Iocs partially sealed or not sealed; bags
contaminated with blood; 5 rejects
Urine - Partially thawed; codes very difficult to read;
9 rejects; precipitated
Dust - Zip-Iocs partially sealed or not sealed; bags
contaminated with blood; 74 screened
34
-------
COPPER HILL, TN
General - Samples received 11/18/75, 11/20/75, and 12/16/75
Blood - Partially or completely thawed; 12 badly clotted
therefore duplicate FEP's run
Hair - Zip-Iocs sealed; 4 rejects
Urine - Partially or completely thawed; some codes very
difficult to read; 19 rejects; heavily precipitated;
15 samples received too late to run with scheduled
batch
Dust - Zip-Iocs sealed; 62 screened
35
-------
DOUGLAS, AZ
General - Samples received 11/18/75 and 11/20/75
Blood - Partially thawed; 7 badly clotted therefore
duplicate FEP's run
Hair - Zip-Iocs sealed; 6 rejects
Urine - Partially thawed; 13 rejects; precipitated
Dust - Zip-Iocs sealed; 53 screened
36
-------
AQUA PRIETA, MEXICO
General - Samples received 11/18/75 and 11/20/75
Blood - Partially or completely thawed; 11 badly clotted
therefore duplicate FEP's run
Hair - Zip-Iocs partially sealed or not sealed; 31
rejects
Urine - Partially or completely thawed; 13 rejects;
heavily precipitated
Dust - Zip-Iocs partially sealed or not sealed; 44 screened
37
-------
HURLEY, NM
General - Samples received 1/9/76
Blood - Frozen
Hair - Zip-Iocs sealed; 1 reject
Urine - Frozen; codes very difficult to read; 19 rejects;
precipitated
Dust - Zip-Iocs sealed; 64 screened
38
-------
d. Control Areas - The control samples were to be analyzed
for: FEP, Pb, Cd, Zn, and Cu in blood; Pb, Cd, and As
in hair; specific gravity and As in urine; and Pb, Cd, Zn,
Cu, and As in dust. Complete analyses were achieved on
368 of the 374 bloods, 229 of the 426 hair samples, 291
of the 373 urines, and the 40 prescribed dust samples from
a total of 308. The following is a resume of each control
area:
39
-------
SAFFORD, AZ
General - Samples received 12/5/75 and 12/16/75
Blood - First shipment frozen or partially thawed; second
shipment (10) not iced at all; 11 badly clotted
therefore duplicate FEP's run; condition required
special mixing for Cd; second shipment (10)
received after coding, causing handling and data
processing problems because of skips in numerical
sequence
Hair - Zip-Iocs sealed; 56 rejects
Urine - Partially thawed; 35 rejects; precipitated
Dust - Zip-Iocs sealed; 65 screened
40
-------
PERRYVILLE, MO
General - Samples received 12/5/75 and 12/16/75
Blood - First shipment frozen or partially thawed; second
shipment (9) not iced at all; 5 badly clotted
therefore duplicate FEP's run; conditions required
special mixing for Cd; second shipment (9) received
after coding, causing handling and data processing
problems because of skips in numerical sequence
Hair - Zip-Iocs sealed; 16 rejects
Urine - Partially thawed; 18 rejects; precipitated
Dust - Zip-Iocs sealed; 66 screened
41
-------
ALBUQUERQUE, NM
General - Samples received 12/5/75 and 12/16/75; all samples
not properly labeled—lacked child code, sent
coding sheets to CDC for corrections
Blood - First shipment frozen or partially thawed; second
shipment (9) not iced at all; 7 badly clotted
therefore duplicate FEP's run; condition required
special mixing for Cd; second shipment (9) received
after coding, causing handling and data processing
problems because of skips in numerical sequence
Hair - Zip-Iocs not sealed; 78 rejects
Urine - Thawed; 22 rejects; heavily precipitated
Dust - Zip-Iocs not sealed; 68 screened
42
-------
NOGALES, MEXICO
General - Samples received 12/5/75
Blood - Frozen; 15 badly clotted therefore duplicate
FEP's run; condition required special mixing for
Cd
Hair - Zip-Iocs mostly sealed; 47 rejects
Urine - Frozen; 7 rejects; precipitated
Dust - Zip-Iocs mostly sealed; 69 screened
43
-------
B. Special Research Studies - The deviations of the actual samples
received from the Government specification as set forth in the
RFP were of such magnitude that it was necessary to conduct two
special research studies before a final analysis scheme could be
established.
1. Urine Precipitate Problems—Essentially, all urine samples
contained a precipitate. Since a number of arsenic com-
pounds are extremely volatile, heating the sample to dis-
solve the precipitate was not a feasible approach. The
best solution to the problem would have been to prepare the
entire sample, as received, for analysis. However, this
solution to the problem was not economically feasible.
The extreme variability in sample size would have required
custom adjustments of the amount of digestion acid required
for dissolution, as well as the final sample volume. These
adjustments would, therefore, create preparation procedures
unique for each individual urine sample. Additionally, data
retrieval, key-punching, and verification involvement would
have increased by more than a factor of three.
The only practical approach to the problem was to try to
remove a representative aliquot from each sample. One of
the larger urine samples from a lead smelter area was shaken
vigorously and divided into four aliquots of 25 ml each and
one final sample of 19 ml. All were analyzed for arsenic.
Results: As = 34 ppb + 6 ppb (or 3.4 + 0.6 yg/100 ml)
(95% Confidence Interval)
Coefficient of Variation = 8.82%
44
-------
Based on these data, this aliquoting approach was adopted
as a compromise preparation procedure. Actual experience
with the samples in the contract revealed that some samples
produced a relatively homogeneous sample after vigorous shaking
while others still contained rather large precipitated aggre-
gates.
2. Dust Collection Background Study—The RFP specifications
called for dust samples of at least 100 mg. In actuality,
the dust samples were collected by a technique which does
not allow for an actual measurement of the weight of the dust
collected. A dust smear was collected from the top of a door
facing using a moist, disposable paper towel, size. 14 X 20 cm
(44 in2) impregnated with 20% denatured alcohol and 1:750
benzalkonium chloride. The towelette most frequently used
was Wash 'n Dri from Canaan Products, Inc., Canaan, Conn.
Because of the inconsistency in the manner by which the samples
were taken and because of the varying mass of the dust collected,
the only analysis avenue open was to employ the sample prepara-
tion procedure used by Vostal and associates in their study.
Actual blank towelettes from the lots used to sample the smelters
were not available for background determinations. Consequently,
background data were obtained from one box of 26 towelettes
Vostal, J. Tares, E. Sayre, J. W., and Charney, E.,
Environmental Health Prospectives, May, 1974, p 71.
45
-------
supplied to the Contractor by CDC. The results obtained for
the towelette background contribution are contained in Table 3.
Appropriate background corrections were applied to all dust
data. Data obtained by the above described method are
representative of the 0.1 N HCL soluble content of the samples.
No relationship can be established between this value and the
total metal content of the dust.
C. Analytical Methodology and Discussion - A number of modifications in
the analytical approach and methodology proposed by the Contractor
for the fulfillment of this contract were required because of the
size and condition of the samples actually received for analysis.
This section discusses these modifications and the analytical methods
actually employed.
1. FEP Analysis of Blood—The original RFP specified that blood
samples would be maintained on ice in the dark from time of
collection until an aliquot was taken for FEP determination at
the contracting laboratory. Based on the assumption that the
samples would be handled in the prescribed manner, the method
chosen by the Contractor for the FEP determination was the
method of Joselow which measures the fluorescence of zinc
protoporphyrin. When the samples arrived, it was learned that
many had been collected for three months or more and stored in
a frozen state for varying periods of time. It was, therefore,
necessary to change from the method of Joselow to the method of
Sassa; and measure all erythrocyte porphyrins. The analytical
46
-------
TABLE 3. Towelette Background Data
Concentrations are expressed as total micrograms per towelette
Sample No.
R 3214
R 3215
R 3216
R 3217
R 3218
R 3219
R 3220
R 3221
R 3222
R 3223
Mean Value
Lead
<0.25
<0.25
<0.25
<0.25
<0.25
<0.25
<0.25
<0.25
<0.25
<0.25
<0.25
Cadmium
<0.025
<0.025
<0.025
<0.025
<0.025
<0.025
<0.025
<0.025
<0.025
<0.025
<0.025
Arsenic
<0.025
<0.025
<0.025
<0.025
<0.025
<0.025
<0.025
<0.025
<0.025
<0.025
<0.025
Copper
1.75
1.25
1.63
1.81
1.88
1.40
1.50
1.38
1.63
1.88
1.61
Zinc
3.25
3.00
4.50
1.00
4.25
4.13
11.8
3.50
4.75
3.25
4.34
47
-------
method of Sassa is the referee method employed by CDC for FEP
o
determinations.
2. Analysis of Blood for Heavy Metals—The original sample prepara-
tion proposed by the Contractor called for the samples to be
lyophilized and for the organic matrix to be destroyed by oxygen-
flask combustion. Prior to this step, however, a microsample
(50 yl) was to be removed for lead analysis by direct flameless
atomic absorption spectroscopy. The presence of many macroclotted
blood samples made it impossible to withdraw valid microsamples
for the blood lead analyses. To overcome this problem, the
proposed sample preparation scheme for blood was replaced with
a wet, acid-oxidation preparation of the total sample (NIOSH
Methods P & CAM 101 and 139).
The acid digest solution was then analyzed by conventional
atomic absorption for copper and zinc and the lead was deter-
mined by direct flameless atomic absorption (using a graphite
furnace with multi-linear temperature programming and simultane-
ous background correction).
At this time, a request was received from the Project Officer
to add an analysis for cadmium on selected blood samples. Pre-
liminary investigation revealed that the blank correction for
the acid-digestion sample preparation was too high for cadmium
to give reliable analytical data. The following analytical
procedure was found to give acceptable results in all cases
except for those with excessive macroclots. Blood samples, in
o
Granick, S., Sassa, S., Granick, J. L. , Levere, R. D., and Kappas,
A. (1972) Proc. Nat. Acad. Sci. USA. 69, 9, 2381-2385-
Aft
-------
their original collection containers, were placed on a vortex
mixer for two minutes. A 250 yl sample was withdrawn with an
Eppendorf pipette and diluted to five ml with distilled water.
An equal aliquot of sample and 1% ammonium sulfate solution
(Cd free) were placed in a pyrolytic graphite tube and analyzed
by flameless atomic absorption.
3. Analysis of Hair for Heavy Metals—All hair samples were washed
with agitation for 30 minutes in a non-ionic detergent
(7-X-O-Matic) and thoroughly rinsed with distilled deionized
water. Samples were then dried (vacuum oven 60° C, 0.5 atm.)
and weighed. The original proposal called for the organic matrix
to be destroyed by oxygen-flask combustion. This, however, was
replaced with the acid-oxidation preparation technique being
used for blood and urine since the results obtained by both
techniques are comparable. Conventional atomic absorption was
employed for the analysis of cadmium and lead in hair while the
gaseous hydride atomic absorption method was used for arsenic.
4. Analysis of Urine for Arsenic—The method originally proposed
and actually used for the analysis of arsenic in urine is the
NIOSH Method No. P & CAM 139. The samples are ashed with a
mixture of nitric, perchloric, and sulfuric acids to destroy the
organic matrix. The ash is treated with ammonium oxalate to
remove traces of nitric acid and the solution is analyzed by
atomic absorption of the gaseous hydride (arsine)„
5. Analysis erf House Dust—The Contractor originally proposed to
49
-------
follow the sample preparation procedure for house dust given
in the RFP as the recommended method. Specifically, the "as
received" sample was to be weighed and then sieved through a
0.5 mm screen while being shaken for five minutes at 260
oscillations per minute. The sieved portion was then to be
weighed and extracted with quartz distilled nitric acid (6N)
at 50° C for 30 minutes. The extract was then to be filtered
and analyzed for Pb, Cd, Zn, and Cu by atomic absorption.
An aliquot of the filtered HNC>3 acid extract was to be treated
to remove all traces of nitric acid and then analyzed for arsenic
by the gaseous hydride-atomic absorption technique.
Additionally, on a selected number of house dust samples, an
extract was to be made on a portion of the unsieved sample and
a comparative analysis was to be made on both sieved and unsieved
portions.
As stated in the previous section of this report, the dust samples
were collected on moist, commercial towelettes. The sample prepara-
tion method prescribed for use in the RFP is applicable only to
dust free from the collection matrix. Consequently, the entire
sample, towelette plus dust, was soaked at room temperature for
16 hours in 20 ml of 0.1 N HCL. The eluates were decanted, final
volume was adjusted to 25 ml, and analyses were performed for the
elements of interest—conventional atomic absorption for Pb, Cd,
Zn, Cu, and atomic absorption of a gaseous hydride for As.
In summary, the employed analytical methodology represents
50
-------
current, state-or-the-art approaches which are basically sound
from a technical standpoint. Documentation of the accuracy and
precision of the methods are contained in the section of this
report which deals with the internal analytical quality control
program. The limits of detection for the various elements in
the appropriate analysis solutions are contained in Table 4.
D. Review of Quality Control Program - This contract required two
quality control programs—a documented internal analytical quality
control program, not to exceed 10% of the total effort, and an
external control program, not to exceed 1%. All quality control
efforts were in addition to the contract samples.
The internal quality control program encompassed blind split sample
analyses for hair, blood, and urine; blind random analyses of
standard reference samples; recovery and precision studies from
two large, composite samples (blood and hair); and the analysis of
standard sample splits with two reference laboratories.
The external control program was to consist of blind split sample
analyses, blind known samples to be analyzed concurrently with
contract batches of samples, and duplicate determinations to be
performed by reference laboratories.
Each 50th blood and urine sample was split into two parts by the
Contractor and one of the aliquots of each was analyzed by the
the Contractor. The other portion was delivered to the EPA Project
Officer for analysis in EPA1s laboratory. The Contractor was not
required to split hair samples for duplicate analysis as a part of
the Scope of Work of the contract.
51
-------
TABLE 4. Limits of Detection
Concentration units are micrograms per liter (parts per billion)
Element Detection Limit, ppb
Arsenic 1.
Copper 10.
Zinc 10.
Lead
Conventional A.A. 10.
Flameless A.A. 0.5
Cadmium
Conventional A.A. 1.
Flameless A.A. 0.02
52
-------
This section of the final report deals with the internal analytical
quality control program. Because of the importance of laboratory
analyses and the resulting actions which they produce, a program to
insure the reliability of the data is essential. Data from a valid
analytical quality control program provide an assessment and measure-
ment of the precision and accuracy of the analytical results. In
addition, a properly designed and conducted program will identify
any segment of the total effort which is "out of control."
1. Accuracy—Accuracy refers to the degree of difference between
observed and known, or actual values. Two approaches were
used to establish the accuracy of the analytical data—recovery
studies and the analysis of standard reference materials.
a. Recovery Studies—These studies were conducted on an
actual sample and not on reference standards. At the
outset of this project, a pint of blood was secured from
the local blood bank. The entire sample was lyophilized
and homogenized to form a large, stable blood composite
for use in the internal quality control program. Known
amounts of copper, zinc, and lead were added to an aliquot
of the control blood sample. Using the composite lyophilized
blood sample, with known concentrations of 17.5 yg/g zinc,
4.60 yg/g copper, and 0.79 yg/g lead, recoveries were 98%,
99%, and 97% respectively. Individual results of this
study are found in Table 5.
b. Standard Reference Materials (SRM) Analysis—Two standard sam-
ples were used to provide a measure of the control of accuracy
53
-------
TABLE 5. Recovery Study Data
Concentration units are micrograms per gram lyophilized whole blood
Copper
Sample No.
R 3007
R 3008
R 3009
R 3010
R 3011
Sample
Concentration
4.
4.
4.
4.
60
60
60
60
4.60
Added Spike Sample Plus Spike Percent Recovery
,0
.0
,0
.0
3.0
7.78
7.35
7.65
7.45
7.36
Average Recovery
102.
97.
101.
98.
97.
99.%
Sample No.
R 3012
R 3013
R 3014
R 3015
R 3016
Sample
Concentration
17.
17.
17.
17.5
17.5
.5
.5
.5
Zinc
Added Spike Sample Plus Spike Percent Recovery
10.0
10.0
10.0
10.0
10.0
26.1
27.0
28.3
25.4
28.0
Average Recovery
95.
97.
103.
92.
103.
98.%
Lead
Sample No.
R 3017
R 3018
R 3019
R 3020
R 3021
Sample
Concentration
0.79
0.79
0.79
0.79
0.79
Added Spike Sample Plus Spike Percent Recovery
1.0
1.0
1.0
1.68
1.88
1.70
1.64
1.73
Average Recovery
94.
105.
95.
92.
97.
97.%
54
-------
during the analysis stage of the contract. These two
samples were Lyophilized Bovine Liver (SRM 1577) from
the National Bureau of Standards and Dried Animal Whole
Blood (Code No. A-2/1974) from the International Atomic
Energy Commission.
Prior to the analysis of any samples from the contract,
eight aliquots from each of these reference materials were
prepared for analysis in order to verify the proposed
analytical methodology.
The bovine liver sample was analyzed first. Results of
these tests are contained in Table 6 . These data show
no significant difference between the actual analysis
mean and the certified value for cadmium, lead, and zinc.
However, a relative error of +3.59% was observed in the
case of copper. Subsequent investigation traced the bias
to the master copper standard employed for the atomic
absorption analysis. This problem was corrected, and the
animal blood standard samples were analyzed. Results for
these samples are found in Table 7. All of the elements
analyzed, including copper, showed no significant difference
between the actual analysis mean and the recommended values
for the standard.
In order to provide a measure of accuracy from analysis
batch to batch, a blind sample of one of these standard
reference samples was analyzed with the samples from each
55
-------
TABLE 6. Analysis of Standard Reference Material 1577 Bovine Liver
Source: National Bureau of Standards
Concentrations are expressed as micrograms per gram (ppm)
Sample No. Copper Zinc Lead Cadmium
R 3030 199. 137. 0.33 0.27
R 3031 209. 135. 0.25 0.31
R 3032 201. 131. 0.31 0.28
R 3033 204. 128. 0.35 0.25
R 3034 192. 128. 0.35 0.26
R 3035 188. 133. 0.36 0.29
R 3036 186. 130. 0.36 0.27
R 3037 197. 138. 0.36 0.26
Mean Value
(95% Confidence
Interval) 197. + 6.6 133. + 3.3 0.33 + 0.03 0.27 + 0.02
Coefficient of
Variation 4.02% 2.93% 11.4% 7.02%
Standard Values
(95% Confidence
Interval) 193. + 10 130. + 10 0.34 + 0.08 0.27 + 0.04
56
-------
TABLE 7. Analysis of Dried Animal Whole Blood, Reference Material
(Code No. A-2/1974)
Source: International Atomic Energy Commission
Concentrations are expressed as micrograms per gram (ppm)
Sample No.
R 3022
R 3023
R 3024
R 3025
R 3026
R 3027
R 3028
R 3029
Copper
50.5
51.6
44.3
45.0
51.5
47.2
44.1
43.9
Zinc
88,
90.
89,
86,
88,
90,
91,
88.7
Lead
0.86
0.89
0.99
0.90
0.96
0.90
0.97
0.96
Mean Value
(95% Confidence
Interval)
Coefficient of
Variation
Standard Values
(95% Confidence
Interval)
47.3 + 2.9
7.26%
45. + 6
89.2 + 1.2
1.57%
89. + 9
0.93 + 0.04
5.03%
0.97 + 0.22
57
-------
smelter and/or control site. These data are tabulated
in Tables 8 and 9. An examination of these data shows
no significant deviation from the standard value of the
reference samples.
Based on both the recovery studies and the actual analysis of
these standard reference materials, it can be stated that at
the 95% confidence interval the overall accuracy of the analytical
data is 98% or better.
2. Precision—Precision is a measure of mutual agreement among
individual measurements of the same property under prescribed
similar conditions. In an analytical quality control program,
precision is determined, not on reference standards, but by
the use of actual samples which incorporate the inherent matrix
variables associated with the sample type under investigation.
a. Quality Control Charts—Analytical quality control charts
visibly represent the continuing validity of routine ana-
lytical data and the performance of individual analysts
within a given laboratory. Limits for these control charts
are calculated using the repeatability standard deviation
for the control sample. Upper and lower control limits
(UCL and LCL) are defined as + 3 standard deviations while
upper and lower warning limits (UWL and LWL) are set at
+ 2 standard deviations. For any given period, if all points
on the control chart are within the control limits and some-
what randomly distributed about the mean value, it can be
58
-------
TABLE 8. Analysis of IAIA Animal Blood (Code No. A-2/1974)
Sample No.
GF 6129
GF 6131
GF 5079
GF 5080
GF 5861
GF 6040
GF 6240
GF 6385
GF 7164
GF 7363
Batch No.
2
2
3
3
4
4
5
5
6
6
Copper
,5
.1
.4
50.
46.
44.
42.8
47.9
51.2
48.
42.
44.6
43.0
,7
.3
Zinc
89.0
94.5
86.3
85.2
88.6
89.0
91.8
93.4
82.9
85.6
Lead
0.86
0.96
0.89
0.90
0.99
0.97
0.90
0.96
0.93
0.99
Mean Value
(95% Confidence
Interval)
Coefficient of
Variation
Standard Values
(95% Confidence
Interval)
46.2 + 2.3
7.06%
45. + 6
88.6+2.7 0.94+0.03
4.23%
89. + 9
4.87%
0.97 + 0.22
59
-------
TABLE 9. Analysis of NBS Bovine Liver (SRM 1577)
Sample No. Batch No.
GF 5582 2
GF 5583 2
GF 5824 3
GF 5893 3
GF 5811 4
GF 5938 4
GF 6217 5
GF 6359 5
GF 7072 6
GF 7310 6
GF 7473 6
Mean Value
(95% Confidence
Interval)
Coefficient of
Variation
Certified Values
(95% Confidence
Interval)
Copper
201.
194.
190.
194.
189.
193.
195.
201.
203.
189.
194.
195. + 3.3
2.51%
193. + 10
Zinc
135.
128.
133.
130.
136.
128.
128.
130.
126.
134.
131.
2.51%
130. + 10
Lead
0.33
0.35
0.25
0.36
0.31
0.36
0.35
0.36
0.30
0.39
0.40
131. + 2.2 0.34 + 0.03
12.5%
0.34 + 0.08
60
-------
assumed that the precision of the analyses is consistent
with the precision of the internal quality control sample.
The quality control sample used to generate the quality
control charts relating to blood analyses was an aliquot
of the lyophilized blood composite used for the. recovery
study. A fresh urine composite sample was used for the
arsenic control. Representative quality control charts
generated through the use of these samples are illustrated
by Figures 1-10. Typical variability in precision among
analysts is demonstrated by a comparison of Figures 1-3
and 4-6. These charts profile the precision of the ana-
lytical methodology employed for the analysis of lead,
zinc, and copper in blood and arsenic in urine. Comparable
charts were also developed for the analysis of arsenic,
lead, and cadmium in hair and cadmium in blood.
A summary of the statistical profile generated by the
control chart data is contained in Table 10. The following
measures apply:
Standard Deviation (s)
the variance.
The square root of
s =
-x2
n-l
where x is the mean
of the samples.
61
-------
Coefficient of Variation (V) or Percent
Relative Standard Deviation (% RSD).
100 s
where x is the mean and s
, the standard deviation of
the sample.
x
95% Confidence Interval. The interval expressed by
x + t(v,F=.975)s/^n, where v = n-1 and where t
is taken from tables of the t-distribution, is
sufficiently wide to grant that the mean of the whole
population has a 95% chance of falling in the interval.
Such tables may be found in most of the CRC mathematics
publications.
The average coefficient of variation for the four elements
is 3.91%. The range exhibited by V is 1.15% to 5.98%.
b. Composite Hair Analysis—Two composite hair samples were
a part of the internal quality control program. In
order to establish the inherent variability of a given
hair sample, a special study was designed using one of
these composites. The entire sample (^ 10 grams) was
washed and dried according to routine standard procedures.
The sample was then cut into one-quarter inch lengths
and mixed thoroughly. Ten aliquots of the dried hair,
weighing 0.5 gm each, were removed and analyzed individually.
The remainder of the hair was initially processed as one
sample; then, after dissolution, ten aliquots were removed
from the prepared sample solution and analyzed.
62
-------
r-
en
:z:
en
Kl
o:
a:
a
en
UJ
j—
en
UJ
C£
Z3
LT3
r--
t—
ui
m
m
to
c.
tH
m
tH
bi
7
63
-------
in
>-
_i
nn
:z:
en
M
o: —
nu
I— Lul
cr i—
cc cc
A
cu
n:
PsI
ui
o:
en
U1
m
ui
(H
tR
PJ
tH
m
7
64
-------
m
on
>-
__i
EE
Z
cn
ta
NJ
o:
a
a:
nn
nu
ct:
czi
m
Ld
o:
za
ua
Lrl
r—
I
in
i
UJ
i—
nc
Lrt
m
m
ui
te
m
fc
rvi
t
tn
03
tK
T
tR
tH
7
65
-------
in
>-
_i
en
2:
nn
CD
o: -
uj _L
D-
EL
o:
DC
ui
i—
nc
LJ
CC
Z3
in
>
u
rn
ui
is
ui
m
3*
UI
a-
tH
ru
tH
66
tH
7
-------
• *
h-
in
>-
_i
cc
• • jr1
^-
tn
t±s
CU
L=J
I
_I
fcD
fc "jr
Z.
UI
r^< I^~-
Ul 1
0- Hsl
I— 1 —
u- —
1=1 1
^ rvi
a:
a
^ w
1 t
1 — 1 —
o: EC
nr ^a
in
i^1
_j
a
a:
t—
z:
t=3
• i
v^*
UI
UJ
a:
•...<•<
. i
ua
!
1 1 <-,*-»*•
ui m -
ui ui u
CD
3-
UJ
a-
tH
tH
ru
67
-------
m
nn
o-
CU
a
_i
fcn
a:
D-
a
tn
a
a:
LD
UJ
in
U1
ui
rn
ui
ui
u
CD
3-
•=»
tH
m
tH
rvi
tH
tH
EB
tH
7
68
-------
r--
r—
i_n
>-
_i
EC
CC
UJ
IZ:
ct:
. i
z: m
UJ —•
in i
o: —
nn —
nc
nu
o:
ni
en
en
UJ
ui
CK
m
in
Lo-
m
rn
m
rvi
rv
rv
ui
rvi
n
tH
rvi
69
-------
LTl
>-
_1
EC
z:
en
nt:
Lrl
ui i
nt: —
tn —
cc
n ••
nn
cu
m
UJ
o:
in
rvi
•at
tn
rvi
in
rvi
ut
rvj
tH
m
ru
7
70
-------
LTI
>-
—I
CC
nn
«^« un
— r-
:z: i
i_d ru
LJI i
cr rvi
cc —
b_ UJ
»—
H- nc
CK i±»
CC
a
a:
a
m
UJ
in
n
in
*
ru
ui
m
u
ry
tn
T
T
71
-------
PJ
DC
rvi
h—
ui
cc
nc
d
ta ui
r--
z i
— nn
Psl
i±a I
en HJ
LjJ —
nt: ••
CZJ UJ
EC
o:
ui
o:
lu-
tH
ti
t
7
72
-------
TABLE 10. Summary of Control Chart Data
Copper (ppm) Zinc (ppm) Lead (ppb) Arsenic (ppi
Mean 4.96 17.5 46.8 2.94
Standard Deviation abs. 0.17 0.20 2.80 0.15
Uncertainty Limits
of Mean abs. + 0.07 + 0.08 + 1.3 + 0.04
(95% Confidence Interval) rel %+ 1.31 +0.43 +2.69 +1.24
Coefficient of Variation 3.44% 1.15% 5.98% 5.06%
73
-------
A comparison of the precision measurements between the
4.49 gm dissolved composite (Table 11 ) and the ten
individual 0.5 gm dry hair aliquots (Table 12) clearly
differentiates the reproducibility of the analytical
instrumentation, as defined by the former, from the
inherent elemental variability in hair, as shown by the
latter. These data conclusively demonstrate the signifi-
cance and magnitude of this variation in trace metal
concentrations when small individual hair samples are
compared analytically. A comparative summary of these
statistical profiles is found in Table 13.
Another study was conducted on a second hair composite.
The composite was divided into thirteen individual 0.5 gm
samples. These samples were submitted to the laboratory
as blind samples in the last three batches of the project.
The analytical results are given in Table 14.
Based on these three studies, the innate variability of
certain elements in hair was clearly demonstrated. This
fact will have a significant bearing on the accurate
interpretation of the blind sample split analyses con-
ducted for this contract. This topic will be discussed
in detail in a later section of this report.
3. Interlaboratory Comparison Analyses
a. Comparative Arsenic Analyses—Concurrent with this contract
effort, the Contractor, was employed by EPA to analyze
-------
TABLE 11. Analytical Data from 10 Aliquots of Composite Hair Sample
(4.49 grams) - Prepared as One Single Sample.
Concentration units are micrograms per gram
Sample No.
BB 007-1
BB 007-2
BB 007-3
BB 007-4
BB 007-5
BB 007-6
BB 007-7
BB 007-8
BB 007-9
BB 007-10
Lead
10.0
10.0
10.0
9.85
9.80
9.90
10.0
10.1
10.0
10.0
Cadmium
45
50
45
45
45
45
45
45
50
1.45
Arsenic
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
Copper
24,
24.
24.4
24.6
24.
24,
24,
24,
24.6
24.3
.3
.6
.3
.3
,3
.3
Zinc
497.
497.
497.
497.
497.
497.
500.
494.
494.
494.
Mean Value abs.
(95% Confidence
Interval)
9.97 + 0.06
Standard Deviation,
abs. 0.09
1.46 + 0.02
0.02
<0.05
24.4 + 0.10
0.14
496. ^
Coefficient of
Variation
0.90%
1.44%
0.58%
0.3*
75
-------
TABLE 12. Analytical Data for Ten Individual 0.5 gram
Aliquots of Composite Hair Sample
Concentration units are micrograms per gram
Sample No.
Z 019
Z 020
Z 021
Z 022
Z 023
Z 024
Z 025
Z 026
027
028
Lead
8.80
10.0
11.3
12.5
10.0
10.0
10.0
17.5
11.3
10.0
Cadmium
0.67
0.67
1.11
1.22
0.67
0.75
1.50
2.13
1.38
0.88
Arsenic
<0.05
<0.05
0.10
<0.05
<0.05
0.10
0.06
<0.05
<0.05
<0.05
Copper
24.9
.1
.1
.1
30.
27.
27,
24.6
25.9
27.1
28.6
25.1
29.1
Zinc
506.
593.
521.
516.
481.
491.
526.
518.
488.
538.
Mean Value abs.
(95% Confidence
Interval)
11.1 + 1.76
Standard Deviation,
abs. 2.46
Coefficient of
Variation
22.1%
1.10 + 0.34
0.48
43.4%
<0.06
27.0 + 1.34 518. + 22.9
1.87
6.95%
32.
6.19%
76
-------
TABLE 13. Summary-Hair Study Precision Data
Mean Value, 95%
Confidence Interval,
Composite
Individual
Lead
9.97 + 0.06
11.1 + 1.76
Standard Deviation
Composite 0.09
Individual 2.46
Coefficient of Variation
Composite 0.90%
Individual 22.1%
Cadmium
1.46 + 0.02
1.10 + 0.34
0.02
0.48
1.44%
43.4%
Copper
24.4 + 0.10
27.0 + 1.34
0.14
1.87
0.58%
6.95%
Zinc
496 + 1.35
518 + 22.9
1.90
32.
0.38%
6.19%
77
-------
TABLE 14. Analysis of Quality Control Hair Composite No. 2
Concentration units are micrograms per gram
ample No. Batch No.
F 5264 4
T 5377 4
tf1 5487 4
IF 5578 4
IF 6727 5
JF 6774 5
iF 6846 5
IF 6920 5
JF 8270 6
JF 8347 6
;F 8389 6
iF 8482 6
^F 8561 6
lean Value abs.
(95% Confidence
Interval)
standard Deviation, abs.
loefficient of Variation
Lead
5.00
12.5
11.3
13.8
11.3
12.5
11.3
15.0
10.0
12.5
11.3
10.0
7.50
11.1 + 1.57
2.6
23.5%
Cadmium
<0.25
0.38
0.25
0.25
0.25
0.38
0.25
0.25
0.75
0.38
0.75
0.25
<0.25
0.36 + 0.11
0.18
51.3%
Arsenic
0.25
0.075
0.35
N. A.*
0.40
0.33
0.28
N. A.*
0.28
0.28
N. A.*
N. A.*
0.25
0.09
32.7%
*N. A. = Not Analyzed. Sample was controlling a zinc smelter
78
-------
arsenic in particulate matter. The analytical methodology
employed in this effort was NIOSH Method P & CAM 139 (arsenic
in urine and air) which is the identical method used for
Contract No. 68-02-2266. In the initial phase of the
aforementioned project, a correlation study was conducted
to compare the NIOSH arsenic method with EPA flameless
atomic absorption technique. Results of these analyses on
sample splits provided and analyzed by EPA are given in
Table 15. The correlation coefficient for these data is 0.9943.
b. Comparative Blood Lead Analyses—On December 30, 1975, the
Contractor received twelve reference blood samples from
the Center for Disease Control (CDC). The Contractor ana-
lyzed these samples by the graphite furnace atomic absorption
technique. In addition, the three samples with the highest
lead values were also analyzed by conventional atomic absorp-
tion.
A comparison of the data obtained by the Contractor with the
CDC standard values is shown in Table 16. Performance
evaluation by CDC was based on the following criteria:
For values ^ 40 yg/100 ml, + 15%
For values < 40 yg/100 ml, + 6 yg
In only one case out of twelve, did the Contractor's data
differ from the CDC value by more than + 6 yg. Sample 5-27
differed by 6.2 yg.
c. Interlaboratory Study Conclusions—Based on the results of
79
-------
TABLE 15. Comparison of Analytical Results of the NIOSH and EPA Arsenic Methods
Concentrations are micrograms Arsenic per milliliter
Contractor Results EPA Results
Sample No. (NIOSH Method P & CAM 139) (Flameless AA)
4 2.00 1.7
7 20.0 19.2
12 15.A 19.5
13 1.56 1.6
14 5.60 4.8
15 6.10 6.5
20 44.0 44.8
80
-------
TABLE 16• Comparative Data - Blood Lead
(Concentrations are expressed as ug Pb/100 ml whole blood)
Contractor Code
CDC Code
Contractor Result
CDC Value
GF 6958
GF 6958*
GF 6959
GF 6960
GF 6961
GF 6962
GF 6963
GF 6964
GF 6965
GF 6965*
GF 6966
GF 6967
GF 6968
GF 6969
GF 6969*
5-29
5-29 Direct Aspiration
5-18
5-36
5-15
5-12
5-27
5-24
5-35
5-35 Direct Aspiration
5-33
5-32
5-30
5-28
5-28 Direct Aspiration
76.6 + 3.6
76.4 + 2.1
12.1 + 2.1
25.5 + 2.0
13.4 + 1.0
22.7 + 0.8
38.0 + 0.9
37.2 + 1.3
55.4 + 1.0
49.1 + 1.5
40.8 + 0.8
56.1 + 0.7
44.9 + 1.0
78.3 + 1.8
81.3 + 2.0
71.8 + 4.6
14.4 + 1.9
26.4 + 2.8
16.3 + 2.2
19.7 + 1.7
31.8 + 1.1
33.2 + 1.7
50.2 + 2.6
37.2 + 2.4
59.4 + 3.6
41.5 + 2.6
79.0 + 3.1
*Direct Aspiration Atomic Absorption Data
A sample of NBS bovine liver was also analyzed with these samples.
are expressed as yg Pb/gram of SRM.
These results
Contractor Code
GF 6970
NBS Code
SRM 1577
Contractor Result
0.33 + 0.04
NBS Certified Value
0.34 + 0.08
81
-------
these two interlaboratory comparisions between the Contractor
and the reference laboratories of EPA and CDC, it can be
concluded that the accuracy data generated by the internal
quality control program of the Contractor provide a valid
assessment of the quality of the analytical data produced
under this contract.
4. Blind Sample Split Analyses—A large portion of the internal
analytical quality control program encompassed blind split
sample analysis of hair, blood, and urine. A measure of the
statistical profile of these data is the coefficient of varia-
tion for the split sample pairs. Complete data for this phase
of the contract effort are appended. Comparative precision
data for the appropriate type of composite control sample and
the mean value for the actual sample splits are contained in
Table 17. If all pertinent criteria for effective quality con-
trol have been satisfied, the Coefficient of Variation should
be approximately equal for the control samples and the actual
sample splits.
An analysis of the data immediately indicates that some phase
of the program is "out of control." The final segment of this
report section is devoted to an assessment of the internal
analytical quality control program and related ramifications.
5. Assessment of Analytical Quality Control Program—A quality
control program in the laboratory has two primary functions.
First, the program should monitor the reliability of the results
82
-------
TABLE 17. Coefficient of Variation Comparison for Control
Samples and Mean of Actual Sample Splits.
V for Mean of Actual
V for Control Samples Sample Splits
Hair
Lead 5.98% 15.2%
Cadmium 6.15% 17.5%
Arsenic 5.06% 14.8%
Blood
Lead 5.98% 22.0%
Copper 3.44% 10.6%
Zinc 1.55% 9.0%
FEP Not Determined 4.51%
Urine
Arsenic 5.06% 24.9%
83
-------
reported. It should continually provide an answer to, "How
good are the submitted results?" This phase may be termed
"measurement of quality." The second function is the control
of quality in order to meet the program requirements for reli-
ability.
The control of analytical performance in the laboratory is
based on the assumption that a "valid sample" has been submitted
for analysis. A "valid sample" implies that the sample was
properly taken, preserved, and delivered to the laboratory in
a condition appropriate for all analytical techniques to be
employed. Further, the validity of controlled analytical per-
formance is dependent upon the use of currently recognized
analytical methods substantiated by the recording and reporting
of subsequent laboratory results in a systematic, uniform, and
permanent fashion. It must be recognized, however, that quality
control begins with the sample collection and does not end until
the resulting data are reported. The laboratory control of
analytical performance is but one essential link in obtaining
reliable data. Each single phase of a quality control effort
will only be as good as the poorest, least controlled area which
has an effect on the results.
A review of the internal analytical quality control program
precision, accuracy, and interlaboratory comparison data on
standard samples revealed that all phases of the analytical per-
formance in the laboratory were "under control." Based on the
data from the composite hair study and visual observations of
84
-------
the blood and urine samples as they were received in the laboratory,
the "out of control" portion of the project was readily identifi-
able as the samples themselves. Analytical quality control is
based on the assumption that a homogeneous sample, free from
obvious external contamination, has been analyzed. All urine
samples received for analysis contained varying amounts of pre-
cipitate, and many of the blood samples contained macroclots.
Hair samples have an inherent variability and cannot be considered
homogeneous tissue samples. In addition, the hair samples were
extremely small; and most split analyses were performed on samples
sizes of "j 0.25 gin.
When all pertinent factors are considered, it can be concluded
that the Coefficients of Variation obtained for the blind sample
splits represent sample inhomogeneity rather than the nonreproduc-
ibility of precision for the analytical methodology. This con-
clusion is in no way detrimental to the internal analytical
quality control program. It does, in fact, demonstrate that the
program was properly designed and conducted since the "out of
control" fraction was isolated and identified.
E. Assessment of Analytical Data - The problem of less than ideal samples
was readily apparent as soon as the first shipment of samples arrived
in the laboratory. Appropriate method modification to minimize the
overall effect on the project were suggested by the Contractor and
approved by the Project Officer. These modifications were designed to
overcome sample shortcomings in all areas except those which were
85
-------
exclusively dependent on sample homogeneity—specifically, blind
sample split precision. An overall assessment of the analytical
data for each sample type will now be presented.
1. Hair—With the exception of the samples used in the blind
split quality control program, the total hair sample as received
was analyzed. The uncertainty of the data obtained from those
samples which were analyzed in their entirety (>90%) and not
split for quality control, correlates directly to the accuracy
and precision of the analytical method as previously presented.
The mean value for the blind sample split analyses is the most
accurate representation for these samples. This mean value has
a comparable accuracy to the samples which were analyzed as a
whole.
2. Blood—The situation for blood is identical to that for hair.
3. Dust—All dust samples were analyzed as total sample received.
On an absolute basis, the data are as good as the accuracy and
precision of the methods employed .
4. Urine—The urine analyses are the least precise data in the entire
study. Because of an extreme variation in sample size, it was
not analytically feasible to perform the analyses on the total
sample as received. The nature and amount of precipitate changed
drastically from sample to sample. Every effort was made to
remove a representative aliquot for analysis; however, the magni-
tude of the Coefficient of Variation for the blind sample splits
86
-------
(24.9%) indicates that the difficulty was not overcome in many
cases. While the urine data are less precise on an absolute
basis, it should be pointed out that on a relative basis the data
are still meaningful for collection site comparisons. For
example, the median value for arsenic in urine for site AJ is
9.77 yg/100 ml, for site HA, 3.06 yg/100 ml, and for site PV,
0.49 yg/100 ml. Thus it is apparent that the lack of precision
in the urine data definitely does not preclude assessment of
the relative significance of the analytical results for the
study. It should, however, be pointed out that the validity
of specific gravity measurements taken on precipitated samples
has not been established.
The measures of accuracy formulated in section D of this report
can be applied directly to an evaluation of all data for a given
element analyzed in a specific matrix. A summary treatment of the
analytical means of each type of determination for the four matrices
from various collection sites is shown in Tables 18 through 21.
The accuracy of these values, and the individual data from which
they were generated, is basically equivalent to that achieved
analytically on standard reference materials.
In the past, it has been customary to exclude "less-than" values
from chemical analytical data before performing statistical ana-
lysis. This is regrettable as the less-than values may reflect
the condition of an important subset of the data, and their
omission possibly precludes the likelihood that the finished
87
-------
TABLE 18. Analytical Data Summary - Hair
Concentration units are micrograms per gram
"Range indicated is the mean + 2 standard deviations"
Collection Site
Lead
Number of
Positive Results
Cadmium
Arsenic
AJ
AL
AM
AN
AP
BV
BX
CC
CH
DO
GL
HA
HK
HL
MG
MI
MN
MO
NG
PL
PV
SA
SM
WP
All Sites
19.2 + 0.3
17.5 + 0.3
38.4 + 0.7
25.4 + 0.5
23.5 + 0.4
56.7 + 1.
170. + 3.
45.8 + 0.8
22.8 + 0.4
26.7 + 0.5
51.1 + 0.9
26.4 + 0.5
64.0+1.
20.4 + 0.4
15.9 + 0.3
31.7 + 0.6
20.4 + 0.4
17.6 + 0.3
15.9 + 0.3
44.4 + 0.8
12.6 + 0.2
11.9 + 0.2
19.3 + 0.3
13.9 + 0.3
29.3 + 0.5
2.68 + 0.08
0.69 + 0.02
4.11 + 0.12
1.94 + 0.06
2.22 + 0.06
9.14 + 0.26
7.94 + 0.23
2.78 + 0.08
2.80 + 0.08
1.88 + 0.05
6.35 + 0.18
1.81 + 0.05
4.26 + 0.12
1.65 + 0.05
1.33 + 0.04
1.90 + 0.06
2.13 + 0.06
1.96 + 0.06
1.29 + 0.04
7.40 + 0.21
1.18 + 0.03
1.09 + 0.03
1.26 + 0.04
1.84 ± 0.05
2.95 + 0.08
1.91 + 0.19
0.10 + 0.01
20.3 + 2.1
0.48 + 0.05
0.14 + 0.02
0.43 + 0.04
2.09 + 0.21
0.18 + 0.02
0.37 + 0.04
0.47 + 0.05
0.31 + 0.03
0.41 + 0.04
0.09 + 0.01
0.10 + 0.01
0.23 + 0.02
0.43 + 0.04
2.14 + 0.22
1577
1451
1126
88
-------
TABLE 19. Analytical Data Summary - Blood
Concentrations are yg/100 ml whole blood
"Range indicated is the mean + 2 standard deviations"
Collection
Site
AJ
AL
AM
AN
AP
BV
BX
CC
CH
DO
GL
HA
HK
HL
MG
MI
MN
MO
NG
PL
PV
SA
SM
WP
All Sites
Number of
Positive
Results
22.
26.
28.
26.
35.
26.
33.
28.
18.
30.
22.
30.
26.
34.
19.
27.
25.
23.
37.
35.
21.
23.
23.
17.
27.
FEP
1 +
1 +
8 +
8 +
8 +
5 +
9 +
6 +
5 +
0 +
2 +
4 +
8 +
1 +
4 +
3 +
9 +
8 +
0 +
6 +
0 +
1 +
8 +
3 +
1 +
Lead
1.7
2.0
2.2
2.1
2.8
2.1
2.6
2.2
1.4
2.3
1.7
2.4
2.1
2.7
1.5
2.1
2.0
1.8
2.9
2.7
1.6
1.8
1.9
1.3
2.1
1941
12.5
17.7
22.7
13.4
19.6
28.9
13.7
19.6
16.5
20.5
12.6
21.2
18.8
17.1
+ 2.0
+ 2.8
+ 3.6
+ 2.1
+ 3.1
+ 4.6
+ 2.2
+ 3.1
+ 2.6
+ 3.3
+ 2.0
+ 3.4
+ 3.0
± 2-7
0
0
0
0
0
0
0
Cadmium
.21 + 0.03
.12 + 0.02
.51 + 0.07
.12 + 0.02
.16 + 0.02
.16 + 0.02
.32 + 0.05
9.1 + 1.5
17.3
14.8
13.9
15.3
17.9
16.9
15.2
18.0
18.6
17.5
1921
+ 2.8
+ 2.4
+ 2.2
+ 2.4
+ 2.9
+ 2.7
+ 2.4
+ 2.9
+ 3.0
+ 2.8
0
0
0
0
0
0
.22 + 0.03
.29 + 0.04
.37 + 0.05
.24 + 0.03
.18 + 0.03
.26 + 0.04
844
Co]
105.
99.4
90.3
95.7
97.7
94.6
152.
75.5
96.6
126.
106.
104.
92.7
78.1
106.
106.
103.
1392
Pper Zinc
± 7.
± 7-
+ 6 .
+ 6 .
~t" 6 .
+ 6.
+ 10.
+ 5.
+ 6.
± 8-
± 7.
± 7.
+ 6.
+ 5.
+ 7.
+ 7.
± 7-
372.
368.
371.
372.
320.
552.
295.
346.
339.
521.
383.
352.
424.
351.
340.
352.
362.
372.
346.
362.
374.
379.
1762
± 18
± 18
T J_O
"T" J-O
± 16
± 27
± 15
+ 17
± 17
+ 26
± 19
± 17
± 21
± 17
± 17
+ 17
± 18
+ 18
± 17
± 18
± 18
± 19
89
-------
TABLE 20. Analytical Data Summary - Dust
Concentrations are absolute micrograms per towelette
"Range indicated is the mean + 2 standard deviations"
allection
Site
AJ
AL
AM
AN
AP
BV
BX
CC
CH
DO
GL
HA
HK
HL
MG
MI
MN
MO
NG
PL
PV
SA
SM
WP
U.1 Sites
lumber of
Positive
Results
Lead
44.2 + 0.8
17.7 + 0.4
35.9 + 0.6
45.2 + 0.8
48.4 + 0.9
76.2 + 1.4
293. + 5.3
82.5 + 1.5
142. + 2.6
37.5 + 0.7
56.4 + 1.0
81.0+1.5
44.1 + 0.8
17.4 + 0.3
19.6 + 0.4
22.2 + 0.4
37.4 + 0.7
86.1 + 1.6
25.2 + 0.5
24.1 + 0.4
25.5 + 0.5
50.2 + 0.9
59.8 + 1.1
220
Cadmium
1.73 + 0.05
0.88 + 0.03
1.50 + 0.04
2.22 + 0.06
1.34 + 0.04
8.64 + 0.25
0.49 + 0.01
1.74 + 0.05
2.86 + 0.08
0.71 + 0.02
2.35 + 0.07
1.18 + 0.03
0.61 + 0.02
0.82 + 0.02
0.85 + 0.02
1.51 + 0.04
0.92 + 0.03
10.3 + 0.30
1.23 + 0.04
1.14 + 0.03
0.42 + 0.01
1.51 + 0.04
2.10 + 0.06
220
Arsenic
3.97 + 0.4
0.09 + 0.009
0.50 + 0.05
34.2 + 3.4
1.18 + 0.12
0.77 + 0.08
0.27 + 0.03
0.40 + 0.04
3.67 + 0.37
0.09 + 0.009
8.93 + 0.9
0.12 + 0.01
0.27 + 0.03
0.91 + 0.09
0.56 + 0.06
0.41 + 0.04
1.43 + 0.14
2.35 + 0.24
0.15 + 0.02
0.31 + 0.03
0.41 + 0.04
0.44 + 0.04
2.87 + 0.29
214
Copper
128. +1.5
559. + 6.
423. + 5.
122. + 1.4
123 + 1.4
17.6 + 0.2
2.14 + 0.03
78.1 + 0.9
462. + 5.4
123. + 1.4
709. + 8.
1023 + 12.
408. +5.
56.9 + 0.7
107. +1.2
158. + 1.8
63.2 + 0.7
140. + 1.6
2.73 + 0.03
21.3 + 0.25
35.3 + 0.41
1784. + 21.
295. + 3.
220
Zinc
260. + 2.
96.6 + 0.7
241. + 1.8
111. + 0.8
89.7 + 0.7
546. +4.
114. + 0.9
149. + 1.1
192. + 1.5
43.8 + 0.33
471. + 4.
179. + 1.4
50.2 + 0.38
39.3 + 0.30
110. + 0.83
275. + 2.
98.3 + 0.7
1616. + 12.
60.0 + 0.5
145. + 1.1
40.7 + 0.3
434. +3.
247. +2.
219
90
-------
TABLE 21. Analytical Data Summary - Urine
Concentrations are micrograms As/100 ml
Uncertainty* Calculated Using
Collection
Sites
AJ
AL
AN
AP
CH
DO
HA
HL
MG
MI
MO
NG
PV
SA
SM
WP
All Sites
Mean Coefficient
of Variation
10.8%
30.6%
12.3%
24.7%
10.5%
38.2%
16.5%
33.5%
22.8%
15.4%
3.9%
16.9%
61.9%
43.5%
27.7%
20.7%
24.9%
Mean
Value
11.53
1.33
3.82
2.19
2.19
2.75
3.81
1.19
2.41
2.53
2.28
2.30
1.34
1.90
2.80
2.50
3.07
Actual Site V
+ 2.5
+ 0.8
+ 0.9
± 1-1
+ 0.5
± 2-1
± 1-3
+ 0.5
± 1-1
+ 0.8
+ 0.2
+ 0.8
± 1-7
± 1-7
+ 1.6
+ 1.0
Mean V for Project
+ 5.7
+ 0.7
+ 1-9
± 1-1
+ 1.1
± !-4
± 1-9
+ 0.6
+ 1.2
± !-3
± l-1
+ 1.1
+ 0.7
+ 0.9
± I-*
± I-2
+ 1.5
Number of
Positive Results
1034
Uncertainty measurements indicate tolerance limits of + two standard
deviations.
91
-------
statistics are representative of the population from which the
samples were taken. A technique exists for working with data
having less-thans all of the same value (i.e. truncated data),
but no technique has been established for analyzing data having
variable less-thans.
If all less-thans are considered to be in fact zero, a mean taken
from the data is a low bound on all possible means that may be
taken from the data by any technique. Similarly, if the less—
thans are considered positive at the value stated for the
less-than value, a high bound on the mean may be set. The mean
best fitting the data must lie between these two extremes.
The data in Tables 18-22 have been treated in the customary
manner. A comparison of the means determined from consideration
of only positive results and the means determined by the high
and low bounds on these groups of data in which less-than values
appeared is contained in Table 22. For most of the data from
this study, the calculation of a mean from only positive results
produces very little alteration of the population representation.
92
-------
TABLE 22. Mean Ranges of Groups Containing Less-than Values
Collection
Site
AJ
AJ
AL
AL
AL
AL
AM
AM
AN
AP
AP
AP
AP
BV
BV
BX
CC
CH
CH
CH
DO
DO
DO
GL
GL
GL
GL
HA
HA
HK
HL
HL
HL
MG
MG
Sample
Type
Hair
Hair
Dust
Hair
Hair
Urine
Blood
Hair
Hair
Hair
Hair
Hair
Urine
Blood
Blood
Dust
Hair
Hair
Hair
Urine
Hair
Hair
Hair
Blood
Dust
Dust
Hair
Hair
Urine
Dust
Hair
Hair
Urine
Hair
Hair
Element
Cd
Pb
As
As
Cd
As
Cd
Cd
Cd
As
Cd
Pb
As
Cd
Pb
As
Cd
As
Cd
As
As
Cd
Pb
Pb
As
Zn
Pb
As
As
As
As
Cd
As
As
Cd
Mean
of
Positives
2.684
19.185
0.090
0.096
0.690
1.328
0.120
4.118
1.943
0.477
2.224
23.450
2.187
0.508
28.909
0.271
2.779
0.144
2.803
2.185
0.430
1.876
26.739
12.559
0.094
43.817
51.161
2.086
3.812
0.117
0.183
1.647
1.193
0.369
1.330
Mean
High
Bound
2.660
18.850
0.077
0.089
0.486
1.231
0.117
3.562
1.868
0.467
2.030
23.202
2.131
0.502
28.624
0.222
2.536
0.139
2.718
2.057
0.426
1.733
25.985
12.091
0.087
39.443
49.636
2.066
3.723
0.108
0.174
1.474
1.135
0.357
0.860
Mean
Low
Bound
2.653
18.739
0.072
0.083
0.390
1.223
0.115
3.501
1.854
0.464
1.983
23.133
2.129
0.501
28.577
0.217
2.486
0.133
2.712
2.051
0.425
1.699
25.895
12.013
0.084
39.436
49.456
2.066
3.721
0.105
0.169
1.437
1.130
0.354
0.679
93
-------
TABLE 22. Mean Ranges of Groups Containing Less-than Values
(con't)
Collection
Site
MG
MI
MN
MO
NG
NG
NG
NG
PL
PV
PV
PV
PV
SA
SA
SA
SA
SM
SM
SM
WP
WP
WP
WP
Sample
Type
Hair
Urine
Blood
Urine
Hair
Hair
Hair
Urine
Hair
Dust
Hair
Hair
Urine
Hair
Hair
Hair
Urine
Hair
Hair
Hair
Hair
Hair
Hair
Urine
Element
Pb
As
Cd
As
As
Cd
Pb
As
Cd
As
As
Cd
As
As
Cd
Pb
As
As
Cd
Pb
As
Cd
Pb
As
Mean
of
Positives
15.888
2.531
0.223
2.281
0.410
1.289
15.922
2.301
7.389
0.154
0.092
1.176
1.340
0.100
0.092
11.861
1.902
0.228
1.262
19.318
0.428
1.844
13.907
2.502
Mean
High
Bound
14.570
2.463
0.220
2.252
0.391
0.878
15.393
2.208
7.294
0.128
0.082
1.049
1.049
0.100
0.972
11.672
1.728
0.189
0.990
15.182
0.419
1.574
13.348
2.462
Mean
Low
Bound
14.197
2.461
0.220
2.251
0.385
0.664
15.209
2.204
7.289
0.123
0.075
1.026
1.025
0.096
0.945
11.633
1.718
0.163
0.811
13.799
0.417
1.525
13.220
2.460
94
-------
'iHHF'LE
4.4
- '.'UHLIVi lOiriPul.
FEP
BLOOD
L 11M F 11 ; I i <
COPP.SRMPLE FP cone. f-u-uti ":,!•• rn[in;)
GF4G01
GF4017
GF4029
GF4041 25.4
GF401;:: 24.5
IJF40TM 25.2
GF4083
GF41 1 3
GF4125
TF4143
GF4161 17.7
T.F4179 22.7
GF4203 14.7
GF4215 16.2 I-:.-:. 1-09
GF4233 1T.9 I'-.-.' *.d^
i;F2c!l 1 13.1 1 "'- - 5- u?
",-;.,' GF2502 75.t." 7^.' 1.43
2-;. =, GF2r:i7 25.i 2:..-j J.H
-•--:. 4 GF2536 24.5 '-.•.- '-• • 2*
:U! GF2547 12.8 12.0 '>.y7
lO.o GF2565
-.? n LF2D78
74.4 CF259U 13.0 1.7 ".19 Hi
-•-. -- GF2307 24. :• 1-, '- 4,20 :"'••
2'I.. 7 GF2851 24.:- "---••' I"-1: ^
!-;,..:: GF2843
GF2880
GF2891
GF2612
GF2t':i
GF26oO
GF2t.71
GF2639
GF2701
96
-------
III - '."'UHLIT i'
FEP
BLOOD
UL SHt'lFLE
\ 421
\ 422
F 423
F 424
F 425
\ 426
F 427
F 428
I 429
F 420
F 4 3 1
F 432
(433
F 434
K435
F 436
F 437
F 423
F 439
F 440
F 546
F 547
F 543
F 549
F 550
F 551
F 552
F 5=1-:
F 551
1 556
F 5 '37
F 559
F 560
F 561
( 562
F 563
F 564
1 565
\ 56'."
F.567
F 563
F 569
F 570
F 571
t 572
FP i:
13.
17.
18.
18.
44.
13.
14.
i -• _
23.
14.
2-, _
16.
15.
i 4 .
- c
20 .
1 3.
'•^ m
24.
1 9 .
2 ,- .
i 7 .
i '"-'• .
J4.
14.
15.
24.
17.
14.
24.
16.
1 8 .
11.
17.
15.
22.
JK .
-! •', m
~.^ ^
19.
25.
22 .
i 4.
,1 '~' .
:OML .
7
1
M
0
1
7
1
•It
ij
Q
cr
=,
2
t
4
r,
1
"l
r
if.
cr
Q
£
=3
^
i_i
:=•
^
o
i>
3
•i
£
I'
1
—
1;
'3
1
0
1.1
:<
UJPP.SRHFLE
GF4S83
CF4S9("1
CF4897
GF48'^9
GF4903
GF491 3
GF4931
G F 4 9 3 3
C1F4933
GF4951
GF4954
GF4969
GF4973
GF4990
GF4998
GF50'J4
G F 5 0 0 7
GF502-
GFc'iM=;2
GF5054
GF5775
GF5737
GF5797
GF5307
GF5317
GF5829
GF5833
GF53 t6
GF5356
GF58T6
GF5836
GF5907
GF5917
GF5927
GF5935
GF5945
GFS ^-^
GF5965
GF59"5
GF5935
GF1:9?5
G F 6 0 0 5
GF6015
GFi025
GF6034
FP C
15.
20,
17.
18.
43.
21 .
13.
1 ki •
ui "' •
15.
21 .
1 7 .
14.
14.
2o .
21 .
1 4 .
3 ^ .
20.
21.
2 3 .
17.
1 c .
j! 2 .
I"' .
16.
1 ?.
£ ~, t
1 7
23.
i — >
19.
1 1 .
1 7 .
15.
22.
24.
">Ci
"• ~*
22.
16.
" Q
14.
26 .
-i'i' .
u'i * i " •; : i '..r
OMC . riLHIl VHt1 IlU l iiil. ". ' '.-ML rP
2 ].i.j 6.9; M'-
4 |::.7 12.5 -";' ;
5 1 ". 1 . 69 ,iH
7 1'';.4 2.74 R!!
6 l:'.^ 0.77 H'J
1 I1 ',9 :.:.4.-: riN
7 1 :. r' 1. 34 Mf
9 L :. 4 4. ':-:.) '1 ',
5 l:. ; 1.4: !!<;
3 r: . i 2.03 •>:
'-. 2 "' l -" . Q l ' i':
0 1 r . : ' ^ . 2 9 ' ! !J
7 1 ., "' 2, 43 ::•
7 I4.i 2.47 n
? 2-:._ 4.11 ',,
5 2'.'' 2.7'3
9 If. * 4 . 5 3 I , :
""• 7> " :• 'ij = t -
4 22.4 :2. 4
7 2 ' 1 - : •' ~ . ^ ^ '
1 2 -:, i 3. 7»
5 I " . -' 1 . 00
5 1": . :" '•':•
2 23.". •'- . 1 6 r
0 1 '-.'"' 1.17 '..- -
5 1-l.u ^.:;~ !i-
7 : ' - ^ 14.". 1 ,
0 1 r , l ' 3' . 7 i
0 l"..9 9.3~
2 23.? 4.3 •'-, <~ ,
;-. I-' I -..-; ;-L,
2 1 '-' . " 3 . 5 -:<
8 11. ": 0
3 17.1 0
6 l;..r '? ;.'
22." L"1
M 2' r" . " ". '"'
1 34. •'- 2. 94
5 21.2 i.71
,-'-'. -O.o
.j
— i ^
6 . . -"' . 6
- L . • i;.
97
-------
i:; - C'UHLITY rriHrPOt.
FEP
BLOOD
[ i.FFFKIEHT
Ft- UJhC. LUPP. SHtlPLE FP COHL.
ijF70
1- j 7 3 ~ \ . ^
( 627 I1'"- >-"•
f C2'~i 20. t'
I -2"'J »4,5
t 630 26 . 1
(631 20.3
^ 632 20. 7
( 633 24. 4
1034 22.9
( Ui 35 2;J . 7
J r" ". t-" 22. '"'
(657 15,6
} t", "-,3 23. T
t 6 : '"' 2 0 , '.4
( 640 23,7
( 6<: 1 lr . 9
f 642 J 3. 4
& K44 23. 4
(645 24.L1
( 6 J 6 3 7 . 3
( 64^ 2'-1. 3
( 643 2-3. 4
t 6 4 "< 24. 0
(630 72.2
( 6f 1 30, 6
r 6:.- 2i\ 3
( 653 22. 4
1- ^ -:*\ "I'"' . '""I
(655 33.3
( 656 30. 1
( t-"57 2 3 . '?
'( 6'",,;: 19.7
\ ^ ",•--! i .: . .-:
( 6 1:. 0 16.7
r 6 6 1 17.2
( 7 o 7 2 3 . 6
r 70" 26. j
( 709 17.3
1 7 1 rt 1 Ci 1 .
GF6045
GF6142
GF6151
CF6161
GF6171
GF6132
GF6192
GF6202
GF6212
GF62 1'3
GF6223
GF62::3
G r" 6 249
LF625?
GF626?
GF6230
GF6290
GF62 ?r
G Ft" 1: 07
GF6317
GF632^
GF6337
GF6347
GF6357
LF6367
u F 6 :• 7 4
CF6324
GF6 3 ?-
.- f~ - I ,', C"
1 . r fi' 4 v ! -•
GF6417
GF642"
GF6435
G F 6 4 - 5
G F 6 4 5 4
GF6463
GF6467
GF70:2
G F 7 0 4 j
G F 7 0 c" 3
G F . ':'* -
33.2
17.3
21. 1
13.5
2o . t-j
1 3 . 6
1 ~J . '-1
24.7
22. 2
19.9
22 . 7
Is! i
23.7
21 .2
24.4
1 '3 . 7
1 :J . i
21.4
>i :' . 7
21.7
41.6
j! 7" • ^
21 .7
24. 0
71.0
30. 1
2 3 . 0
"' !~! M
34.5
j|7 . ^
24. t.
20. "
1 J . 1
17. 5
1 3 . 0
_ C i '
1 !- -i.
7" 1 , ".
1 •".: . '
2. 64
5.67
4. 2'-
»j
1.11
:'. 4e
98
-------
I'll! - OUHL1TY COHlr-'UL
PEP
BLOOD
!'L'L f-F!HE.HT UF
uc I.HMPLE Fr r..jm:. roKP.sflf'iFLE FP com:. HEWI ''HPiHriuM' •. • VTLTP
(71b it'.O GF7122 14.8 14.9 1.21 Fif
1717 I'r'.O GF713J.1 19.0 I1-', i 2.74 MG
t 71'J . ::. J GF7142 20.8 22. n 8.04 NG
(719 I''.L GF7152 41.8 42. '", 2.09 NG
t'72u 29. t" GF71b2 29.3 29.4 O.b-J UG
GF7172
GF7182
GF71 ?2
r 7"4
f 775
\ 7 t
'r 77;
\ 77c.
22.4
'ji- , :^-L jo. M
GF7342 24.0
GF72:.^ 50 !4
14. t"
21.1
41 . u
1794 iJ.t '^--t'l ]:!:.b -;;•.. 2.2i
99
-------
HM hilul,,
LEhli
BLOOD
CH' ;,hHFLE
t 29,."'
L -' q v
r t. r .'
i i'M
!• I:1? 5
\ "' '^ r'
I "' '^ ;-'
J 4 1 9
- k+ " I-"*
I 421
r 4 "-' "
r ** w.
r 4 2' 3
! 42.1
> 42-;
r 427
f 428
( t ^.'r4
[ 420
f 4 3 1
( 432
1 4^4
\ J -• 5
i 426
1" ""+ "'
1- 4 3 8
Pi3 COH'2.
4G. 1
17.7
30. 5
4 !~> . ^
i j. i
27.9
13.1
~> r: i
,: . j i
^ . X'^
1J. ^
- -• r:
•~* i^ ~r
7.04
1 0 . b
~. o7
13. 3
12.5
12.5
23.3
12. 3
*'^ 1
i 9 . D
1 4 . 0
12.8
12.3
1 b . 0
; -i. 4
CO PP. 3RHFLE
GF2502
Li F 2 52 9
LF2542
GF25b4
GF2575
GF2580
GF2598
Gh4G>"4
GF4377
1 j F 4 o 'i :!
GF4390
GF4897
GF4C99
GF4903
GF4913
GF4931
GF4933
GF4933
GF4951
GF4?54
CF49tL9
CF4978
LM'^y
GF49^8
GF5004
GF500"
CF5024
L ' •
PB COHL. tlL'iil '.
37. 3 4i. 7
1 '-' "' 1 ' "l
1 '-'.-' J. ' . '
3b. 4 ;:j. i
14.2 i r . l -,
12.8 U . '-•
9.07 1 •:•.=.
14.1 1 : , .2
£.77 4, -"4
1 * . 5 ! :' , „'
9. 22 1 ! , ^
14.3 10.8
16.7 1 j . 3
1 1 . '3 :< . 4 ':'•
13.1 It,
ll.o 7 . . ' -
5.32 11.2.
.3 .13 n . 2
9.36 H'.'-
1 4 . 0 1 i: . ."
3.98 Ki.-J
15.9 2 i . 0
: 5 . r; 17..-
14.1 1.1
12.9 1 . . 8
1 4 . - ' ~:,,~
17." 16.."
15.1 1 * . 7
1 i'h f L f bUT U!7
h|. in riON' '. ' Sl'ILT
14.3 Hi
2.32 Hi
12.5 rir
13. 2' Kt
11.4 HI
72.0 Hi
5.29 t\\
64.9 hli
:. j . J. '"i i ^
25.1 dh
46. 3 nr-
-• 0 .4 r i '
35. 9 HM
37.3 H i •
'-' -r • -• " *
"7.? 11-
-f'.L MG
! 5 . 2 " G
^ i . :• r > j
34. i nij
1 6 . "' '" . l
^- ' ^ r. :
'-' . 1 • ; i ,
13.-
r" . 0 3 M
< 1 .-• i
( 4'"O
( 45-t
145= 2 2.2
r 456 21.0
r ^t J
t 446
J4--1 !:":. 6 ijF261 2
', ,i i '; S GF2615
GF2622
100
-------
RPF'EHDIII - OURLITY
OC SflMPLE Ft1 Li'iMC. iIOPP. SRt'lPLE
LEriD
BLOOD
i nL i [ i . :. ,, i
F'B COHL. HLiltl '/up' lM I Ii 'it '.
GF27':2 25.4 2-' 2 t..'^0 HI
CFJ:7.S(i 22. f 2 1. 'i 11.5 ,h
GF27St" 22.5 JM,] lb..-J iJR
. GF2787 25.4 2?.;' 12.'^ hH
f 504 2i.2 CF2S04 2'?. 2 2t. .£ lo.l ;:'.•'
CF28LJb •'?.'? 29. "' 5;:'. 3 t1 '
'jF^':r' L b I i . j! 1 M . J •' • '- •'
G r 2 8 2 U ;. 7 . '? 5 _ . -
•IF2S42 55.4 5''..- -jr . ,-
GF284t' l'r.7 i -. . L: 38.2
4 1: . 2 32,: 47,?
H.I
12.'?
11.'?
GF2S51
GF409'?
IJF4102
GF4178
GF418'.:i
101
-------
DI:: - OIJHLITV ojufKu.
LEflD
BLOOD
U'CFFICIEIIT OF
DC 'iFlNPl.E FB COhL. COPP. SHllPLE FB COMC. I'lLntl V hi'1 IhT IuH ' '.
1 J. r-:
2 > 1 _ .
n. K
1 £• i.r C1
\ •"<-"•?
\ 570
L571
' c-( "" I-
t 57 j
I1 *J i 'j'
f 57'--
' ~ '-'. f i
r 551
, ^r^
f ci1- "'
t c< '", 4
t o 1 7
t 6 1 8.
t 0 L 'f
,• 62r'"i
t oil
t 622
f 623
t o24
r if- 2 5
i 6 "' '-•
I 627
1 623
t 62-
t <-, - o
1 L" 11
> o ;2
! 63:
t t" :' 4
1 if. '•• J
\ 6 ' -"
r rf. i 7
f tot1
t b :.".
> b'4'J
f L. M I
I f,4j
t c"43
1 I-, -L t;
t if.'tS
t C4t"
t >': t7
! 646
I 4 . :'.:
^..i. ."
2 0 . 'r1
"-' . cr
21:. 4
21.3
1 -T . 1
2 0 . 0
14, i
17.0
19.7
19.4
; '.; ^
•L 2 , 9
1 7 . j
•J.o7
ij cr "
1 1 . t"
2 i . t
17.9
i 3 . 0
1 ,"' . f)
i -' '*
1 j . -'
9 . 3 '3
i 0 . if.
i ^ * J
j :i . J
1 9 . 1
[ t , ^
1 :: . '"-1
'5. 4U
14.7
1 1 . 0
12. 2
11.9
1 1 . j
': . 9 .3
i 7 . 1
1 0 . j
1 _' . 4
1 4 . 0
, 1 • .J
I'l . 1
' :. . J
GF5995
GF60C1 5
GF6015
i j F 6 0 2 5
GF6034
GF6045
GF419?
GF42CU
G F 4 2 0 6
GF4219
GF4231
GF4241
GF4247
GFJ936
GF3997
CF4007
GF-1017
GF4027
GK4033
GF4046
CF4050
GF-40'fi;.
GF4070
GFt.142
GF61-.1-
GFblbl
G Ft 171
G r 6132
GF6192
C F 1 2 0 2
Li f* b - 1 2
CF6218
GF622.3
Li 1~ C ul - C'
GF6249
GFo259
CF62t "'
IjH- .. ':• k.1
1 j F 6 2 :; 0
1 j r c J '.^
'',,-"'-"3117
|.,Ft"317
i " F t" !' 2 T"
G tr 6 3 5 7
i",Ki-"347
12
20
13
^! j-
21
^!4
9.
1 1
-" .
9 .
14
14
i r'
'-4
14
4.
10
~*
13
i :•
1 3
t', .
12
;4
"•
;I; _
1 ;'
1 0
1 '.
--
1 =
4 .
l „;
3 .
12
20
i if
1 i
22
1 f
1 -"'
_ .
•. ~-_
'_ -^
.4
. 1
, 2
, 9
. 1
. T
63
. ^
C5
4 1
.2
• t
1
t"7
• ->
i .•'
. i
17
~^
• 2
.-*
9 ~-l
, \
. 4
4"
L ^
. 1
. -•
• ^
~" r
~"
T 1
' :
, j_
.
,
. ,
102
-------
fiprtiiJ.ii::
ZINC
BLOOD
L Mt F- h i 1. iLh I
i.ii' ;.hf1PL£ IN iJJHC. LUTF. SHHPLE IN COtlL. MUiil '/Ml'1 IFiTluN '
J
f
f
r
f
r
r
^
r
f
L
r
r
j
V
f
t
t
r
r
r
>
t
f
r
I
>
,
r
h
41'^ 1 i ' - 1 .
4..0 33? .
421 441.
1 4 2 4 .-' b .
422 271..
424 3>35.
42"-" 34:3.
4 2 o 3 "J '- .
4 " " 40'.
423 3c.:.:.
429 293:.
-13-3 UC|.
41 i 2:34.
412 13: 1 .
4 3 3 4 1 0 .
434 353.
•»3c >;2,
4 ; 7 -lc"2 .
423 42-i.
439 704.
440 444.
-4b 40 7.
447" 4 1 4 .
44* 405.
4-r9 tCl.
451 .. >;.
452 TOS.
453 • > " 7 .
454 4 3 !;' .
455 41":.
4 S r", :; ; : 3 .
4571 Mi.iS.
4?:-: -.27.
453 V'O-.
4 7 >J :> 3 I .
•' ' 1 r > 7 '." *
-T c" 2 j -1 2 .
433 '3:30.
|-
G
G
G
G
G
G
i..
G
G
G
G
G
G
G
G
G
^
G
G
G
C
U
G
G
G
i^
G
G
,-
G
!_]
j
G
G
b
G
f.,
F4
F4
F4
F4
F4
F4
F4
F 4
F4
F4
F4
F4
F4
F4
F4
1-4
r-*
-- cr
r j
i* tr
f" J
F5
F5
F 5
h 4
F2
r 2
r 2
F2
F2
F2
F2
F2
F2
F2
F2
F2
K2
f 4
F2
F 2
HK
C1 i
S3
a 9
S9
fiQ
9 0
9 1
1=1 3
93
93
93
95
9o
'^ '
'-;'->
0 >"i
0 0
02
05
05
to
6 0
6 1
b !
t">2
n3
-:5
b5
6 3
f- 7^
be
b'r
70
~* r1
7" C'
7*:
"^ ^
4
^*
3
o
"
9
I'
1
:
;;;
1
4
?
a
ij
4
r
4
-,
4
^
Q
3
Cj
7
1
5
0
'j
9
-
9
4
^
0
3 i j .
342.
3 o 7' .
5b4.
422.
3 3 0 .
I 2' 0 .
1 3' ; .
4 -< L"l
;77.
4 r-, ,
-- ;' 2 .
2 1 T1 .
-b4.
-'^- '
444.
-> 3 c" .
-L':6.
'- 39.
*, -1 1
I b 3 .
„ ~ 4
274.
-2:9.
^44 .
12".
_ ~ ':* .
420.
: 2 -* .
: i b .
: - o .
3 ~ 9 .
4-2 .
45: .
591.
-^ i. ~°
—~ 5 "^
1.71
4 : .
J4' , 4. ^'f:
40, 3
liH
RiJ
103
-------
:: - nijHLiTY
LEflD
BLOOD
C0~h"l:!'.: IB
0,: SflMFlE PB COHC. COPP.SfiMPLE PBCOMC. HLHM ^'iflTIOl
14>1 GF7354 22.0 13.' ,1.1
17. «5 GF, 'Jb5
,,.ul GF7375
--;. i GF7335
GF739j
CF740t"
GF7416
GF742* IS- !"• ' ^'t ;-l
. , , GF7476 17. i i--'-- :_,-_ :'';
18'.:. GF7441 22.0 ^-- '-•'
10.4 GF-4?2 20.1 ic-:-: :;•: '^
: -i-i GF74el a.'^ -'- - : - :. \ ~
te'.^ ^7471 12.1 l>.» =:- ^
, -o= i 4 4 'jF, 4c J 1' • -' L -'•.. ;: . , .
-^ ;i;f GF74^2 I'?.? I'-''
104
-------
|V Shi'11 Lb
U 11 i - I.'UHLI T, C UH i r UL
LEflB
BLOOD
COFFFIC IENT Of-
i.OPP.SflMPLE PE: CONC. HFnt ' 'i iP IHT IC'H' '. •
GFt":?7 12.0
GFb3t7 It". 1
Glrt"';r4 20.7
r,Fii:84 11.5
GFt'::l?4 lO.b
1 =•". j G F 6 4 <-_ 7
,-11 . '? G F 7" 0 3 2
1Kb GF'7043
CF7074
r, F 7 0':' 4
GF70'?4
105
: l-J . 4
H - GFF.42^ 1-.J i4- -" i'3"
il;; '.F643= 13.0 r. ; 10.0
1^.5 G F14 -15
- .:* i',F6454
: ^ ^ .
14.1
-------
TECHNICAL REPORT DATA
fl'luiisc read In^lritctwm on the reverse before completing)
1 REPORT NO |2
EPA-600/1 -76-029 1
4. TITLE AND SUBTITLE
ANALYSIS OF BLOOD, HAIR, URINE, AND DUST SAMPLES
FOR HEAVY METALS
7 AUTHOR(S)
Anna M. Yoakum
9 PERFORMING ORGANIZATION NAME AND ADDRESS
Stewart Laboratories, Inc.
5815 Middlebrook Pike
Knoxville, TN 37921
12, SPONSORING AGENCY NAME AND ADDRESS
Health Effects Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Rpqparrh THanglp Park, N r P7711
15. SUPPLEMENTARY NOTES
3. RECIPIENT'S ACCESSI ON" NO.
5 REPORT DATE
September 1976
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
1AA601
11. CONTRACT/GRANT NO.
68-02-2266
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA-ORD
16. ABSTRACT
Communities from ten states in the United States and two cities in Mexico
were studied. The communities were chosen for their proximity to primary
non-ferrous smelter industries.
Three lead and five zinc smelter areas were sampled for blood, hair, and
dust. Urine, blood, hair, and dust were collected from fourteen copper smelter
sites and four control cities.
Samples were analyzed for arsenic, lead, cadmium, copper and zinc.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS b.lDENTIFI
arsenic sampling
lead Ulood
cadmium .
-r IE?
smelters
13 DISTRIBUTION STATEMENT 19 SECURI
DTI TAT TO PIIPI TP UNLL
KLLLMoL 1U rUbLIU 2o SECURI
UNCL
ERS/OPEN ENDED TERMS C. COSATI Field/Group
06, A, F
TY CLASS (This, Report/ 21 NO. OF PAGES
ASSIFIED 12°
TY CLASS (Thispage) 22 PRICE
ASSIFIED
EPA Form 2220-1 (9-73)
106
------- |