RESEARCH TRIANGLE INSTITUTE
November 1, 1976
Monthly Technical Progress Report No. 1
September 28 - October 31, 1976
EPIDEMIOLOGIC STUDY CONDUCTED IN POPULATIONS LIVING AROUND
NONFERROUS SMELTERS TO DETERMINE BODY TISSUE BURDENS OF
SELECTED NONFERROUS METALS
EPA Contract No. 68-02-2442
RTI Project No. 31U-1372
Robert W. Handy
Research Triangle Institute
Chemistry and Life Sciences Division
Post Office Box 12194
Research Triangle Park, N.C. 27709
Prepared for
Environmental Protection Agency
Research Triangle Park, N.C. 27711
ESEARCH TRIANGLE PARK, NORTH CAROLINA 27709
-------
Table of Contents
Page
I. INTRODUCTION l
II. DESCRIPTION OF OVERALL PROGRESS 2
A. Field Operations 2
B. Sampling Methodology 3
C. Chemical Analysis 6
D. Quality Control 10
E. Sample Packaging and Shipment 11
III. POTENTIAL PROBLEM AREAS. . . ' 13
A. Field Operations. . . . •. 13
B. Chemical Analysis 13
IV. DESCRIPTION OF FUTURE WORK . . . . 15
A. Field Operations 15
B. Sampling Methodology 15
C. Chemical Analysis 15
-------
I. INTRODUCTION
The work started during the report period was directed toward
development .of a Detailed Work Plan. This report will be due at the
end of the next report period. The areas of work covered by this re-
port include the field operations, sampling methodology and chemical
analysis aspects of this program.
-------
II. DESCRIPTION OF'OVERALL PROGRESS
A. Field Operations
During October, work has begun on the detailed work plan and the
data collection instruments. In preparation for the development of the
detailed work plan, the Survey Director has performed a limited litera-
ture review and obtained various materials from and through the Environ-
mental Protection Agency (EPA) relative to various methods of collecting
the necessary environmental and biological samples, including two measure-
ments that are not yet part of the Scope of Work - paint and feces. RTI
tentatively envisions three data collection instruments - a Household
Screening Form, which would elicit eligible and cooperative households
and participants; a Participant Consent Form, which would describe the
study .and the participant's role; and a Questionnaire, which would
obtain household and individual information on demographic variables,
residence histories, and potential special exposure situations.
During October, RTI learned the six performance sites for this re-
search effort, which are as follow:
1. . Ajo, Arizona
2. Anaconda, Montana
3. Bartlesville, Oklahoma
4. Bixby, Missouri
5. Herculaneum, Missouri
6. Palmerton, Pennsylvania
not necessarily in that order. In addition, as the result of a meeting
at EPA, Research Triangle Park, on October 28, 1976, involving EPA
Project Officer Dr. Warren Galke, Dr. Carl Hayes of EPA, and RTI Survey
-------
Director Benjamin Harris, RTI is considering two possible sites for the
pretest effort, either Hayden, Arizona, or Corpus Christi, Texas, probably
in that order of preference. Among the various considerations which will
influence the ultimate decision are the extent of various levels of
local cooperation and the logistics of testing the study protocol and
data collection procedures.
B. Sampling Methodology
Sampling methodology in the RTI proposal is described as a strati-
fied area sample in which a sequential sample of persons is to be selected
for participation in the study survey. Also, a probability sample of
sites and days is to be selected for the purpose of obtaining high-
volume air samples. Selective sampling is to be used to collect samples
of tap water, house dust, and soil because random sampling for these
data was deemed impractical. During October, meetings were held with the
project officer and, later, with RTI staff to outline the initial planning.
RTI was notified during the last week of October of the identity of
six survey sites and the probable pretest site. The identity of these
sites enabled the sampling center to initiate activities in the following
areas:
- preparation of the detailed work plan;
- ordering of selected mapping materials;
- preliminary investigation of site populations to ascertain the
probable amount of stratification that will be involved and,
hence the approximate number of electric drops that will be
needed to conduct the air sampling.
-------
The sites were characterized according to population, migration, and age
characteristics (see table 1). Also, other neighboring communities
(within 10 miles) that could be identified from extant RTI materials
were listed. This preliminary investigation revealed the mapping materials
of the Census of Population will be of very limited value for this study
because of the relatively low-density population areas involved. This
limitation relates to the lack of detail and the small number of Census
areas delineated for the sites; the limitation was anticipated, however,
and the sampling frame will be developed on-site by updating whatever
mapping exists for the .particular site.
The minimum sized stratum for each site can be anticipated by
viewing the smallest cohort proportion and making allowances for possi-
ble nonresponse and occupational exposure. For example, for Ajo, Arizona
approximately 7 percent of the population is male and 60 or older.
Approximately 5/.07 = 71 persons will be needed to produce the requisite
sample of 5 in this cohort if all participate and none have had occupa-
tional exposure. Therefore, the 71 plus additional persons to account
for possible nonresponse and occupational exposure suggests that the
minimum stratum size may be approximately 125. What actually constitutes
reasonable allowance for these two sample losses, however, will be
ascertained from local sources and from site experience. This minimum
strata size, the relatively small site populations, and the increasing
strata areas as the distance from the source increases, together, sug- .
gest that fewer than eight strata will be feasible for some sites. The
actual number of strata used will depend on several factors including
what constitutes the "downwind population."
-------
Table 1
Selected Population Characteristics of Survey Sites
(a)
Net migration
Age groups, percent
Smelter
State town
Arizona Ajo
Hayden
(test site)
Missouri Bixby
Herculaneum
Montana Anaconda
Oklahoma Bartlesville
Pennsylvania Palmer ton
Neighboring
towns
Rowood
Winkleman
Hayden Jc.
Christmas
Bass
Buick
Viburnum
Festus
Crystal City
Horine
Povely
Gregson
Warmsprings
Dewey
Tuxedo Park
Population, since 1960, . .
number percent 0-5 6-17 20-35 60
5,881 -16.6 15 25 19 14
1,283 -27.1 - - -
974 . . •
-
<1,000
- .
520
1,885 6.7
7,530
3,898
517
9,771 -18.9 13 22 18 19
29,683 6.4 13 20 20 17 "
3., 958.
5,620 - 5.4 • . 11 17 20 22
(a)
Based on the 1970 Census of Population.
-------
From A to 6 air samplers will be operating virtually throughout the
survey period, and the requisite number of electric drops is expected to
range from 8 to 18 for each site depending on the number of strata.
Table 2 demonstrates a typical allocation of days (1-19) for each of 8
strata. Constraints on the allocation require that no more than 2
strata are designated for monitoring on a particular day, and that no
stratum will be assigned more than three monitors on any single day.
The procedure is as follows:
(1) Randomly select 3 days from 1 to 19 without replacement
and assign these to stratum 1; two locations proximate to the
first two household-sampling units will be monitored on each
of these 3 days;
(2) Repeat the procedure for stratum 2;
(3) Repeat the procedure for the remaining strata except that days
already assigned twice are ineligible, and the number of days
to be assigned to each stratum increases for strata that are
more distant from the smelter.
C. Chemical Analysis
Control tap water, urine and hair collections have been analyzed
for Pb, Zn, Cd and Mn to determine optimum atomic absorption operating
conditions. Variables such as drying, charring and atomizing times and
temperatures, slit setting and wavelength have been investigated to
determine the effect on linear calibration range and sensitivity limits.
Deuterium background correction was employed throughout.
A study is being conducted to determine the relative effectiveness
of hollow cathode and electrodeless lamps as light sources for Pb, Zn
and Cd analysis.
-------
Table 2
Random Assignment of Days for High Volume Air Samples
Day
— — Electric drops
Saratum _ 1 2 34 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 required number
1 xxx
2 x x x
3 xxx x
4 x xx x
5 x x x xx
6 x xx x x
7 x • x xx. x x
8 x x x x x x x
2
2
2
2
2
2
3
3
Each x represents 2 sampling locations for each stratum-day.
-------
Elemental concentration in control collections have been measured
by the method of standard additions.
Precision assessments are being made on all of the following analy-
tical procedures.
1. Tap Water
Analysis of tap water samples has been carried out by adding 50 yl
cone. HNO- to a 10 ml sample and injecting directly into the graphite
furnace. Calibration is accomplished with aqueous standard solutions.
2. Urine
Urine was diluted with 0.5% HNO« and injected directly into the
graphite .furnace. Preliminary studies have shown that sediment forma-
tion is promoted by freezing. Specimens held for periods of time up to
1 month at 5° or room temperature formed little or no precipitate.
Urine calibration solutions are prepared by adding known amounts of.
metal to a 0.5% HNO diluted control urine solution.
3. Scalp Hair
Collections of both male and female hair (barbershop clippings)
have been subjected to a rinsing procedure and a variety of different
washing techniques (Prell shampoo, Snoop, dilute acid). Acid dissolu-
tion methods have been investigated with respect to heating times and
acid concentration.
4. Air Particulates
Glass fiber filters have been evaluated for metal background con-
centrations and their contribution in the analysis of normal or expected
filter loadings. Two grades of 8" x 10" filters are available and have
been investigated: regular Type A-E and spectrograde Type A (both sup-
plied by Gelmah). Reduction of blank values by filter prewashing is
being evaluated.
-------
5. X-Ray Fluorescence Lead Paint Analyzer
Mr. John Rhodes, Columbia Scientific Industries (CSI), Austin,
Texas, was contacted for information on portable XRF analyzers. CSI has
developed new procedures and technology for the detection and measure-
ment of lead in paint under contract to HUD. CSI supplies a Model 700
unit which has the capability of scanning surfaces in situ for lead
levels. Since the device is not a high resolution instrument, scattering
from the wall material interferes with the direct determination of lead.
These interferences must be measured and appropriate corrections made.
CSI has devised a method for overcoming much of this difficulty. An
initial reading is made with no filter (lead plus background) followed
by a second reading with a tungsten filter (background only). Both
readings may be referred to a graph to give a lead concentration value
2
(mg/cm ) - if the nature of the backing material is known. However, CSI
has prepared a nomograph which allows one to read the corrected Pb
content without prior knowledge of the backing. The nomograph procedure
is clearly the measurement method of choice. The cost of the Model 700
is $4,750. Mr. Rhodes made the comment that the unit should be operated
by a trained individual. On the other hand, he claimed that the improved,
direct-reading Model 750 (see below) is easily operated by semiskilled
personnel (i.e., field interviewers).
A new Model 750 portable lead in paint film analyzer has been
developed by Columbia Scientific Industries, under contract with the
HUD. This portable device is capable of rapid, in situ measurement of
lead content in a variety of painted surfaces, irrespective of the
substrate or wall material. The unit provides a direct readout of lead
2
content with a detection limit of 0.1-0.3 mg/cm . This instrument is
-------
10
claimed to be the only one with adequate sensitivity, accuracy and
applicability to meet the needs of HUD and other lead control programs.
Mr. Rhodes stated that the Model 750 will not be commercially
available for 5-6 months; only 4 experimental units exist at this time.
The cost of the Model 750 Analyzer will be in the $5,000-$10,000.
D. Quality Control
1. Calibration of Hi-Volume Samples
The quality assurance group at EPA, RTF is calibrating our top-
loading high volume Orifice Calibrator, Sierra Instruments against a
primary standard R wots meter (positive displacement meter). Hi-Vols
equipped with a pressure gauge for reading flow rates will be purchases
for this program (Fisher Scientific). The. flow rate gauge will be
calibrated against the Orifice Calibrator. The resistance to flow
across the collection media will be simulated by means such as adding a
second and/or third filter to the unit.
An andot device is also available at EPA for final validation of
Hi-Vol flow rate measurements and will be used as required.
2. Working Reference Materials-Air Particulates
. The quality assurance group at EPA, RTP has available glass fiber
filter strips (3/4" wide) containing known amounts of arsenic and lead.
These reference materials are available in 9 different concentration
levels. They are submitted and anlyzed fluid. Results are then forwarded
to EPA. Oh request, similar materials would be made available for the
four other metals. The preparation of these "standards" would be consid-
ered by EPA if the need for them was long term.
We will request the presently available arsenic and lead standards
to validate our .analytical methodology. The evaluation will be done
-------
11
directly with the Project Officer. These materials will also form the
basis of our air particulate QC program.
3. Working Reference Materials-Tap Water
The .quality assurance branch of EPA, Cincinnati has quality control
water samples available to laboratories involved in monitoring water
quality. Samples "certified" for trace metal content are available free
of charge. They are prepared as concentrates and are supplied in three
different concentration ranges. We have ordered a set of these QC
samples for method validation purposes. These materials will become a
part of our routine QC program.
E. Sample Packaging and Shipment
The following summary describes our proposed packaging and shipping
protocol.
All samples will be chilled to 0°-5° (not frozen) after collection
and will be shipped in this condition via Federal Express. It is also
proposed that samples on arrival at RTI be stored under refrigeration.
High density polyethylene containers (120 ml) will be used for tap
water, urine and soil samples. Different types of caps are being investi-
gated to provide a leak-proof seal, without contamination. Shipment by
Federal Express eliminates the possibility of sample contamination due
to pressure gradients in transit.
r
Zip-lock plastic containers will be used for air, hair and dust
samples. Provisions will be made to insure a tight seal to prevent
sample loss and contamination. The plastic bags will be shipped in a
styrofoam-lined box.
Exception will be made in the case of fecal samples. They will be
shipped frozen in 250 or 500 ml plastic containers.
-------
12
Plastic containers will be shipped in styroform blocks in which
depressions are cut to exactly hold each container. Our experience with
such shippers has resulted in no container damage and a minimum of
sample temperature change during transit.
Blood samples will be shipped to RTI in specially designed 8-tube
vacutainer shippers. Blood samples will be chilled (not frozen).
-------
13
III. POTENTIAL PROBLEM AREAS
A. Field Operations
RTI is concerned about the rationale, procedures and additional
expense of collecting the feces (and paint lead) samples from smelter
community subjects in this study. Not including the apparatus to measure
paint lead, the cost for the sample collection alone could be as much or
more than $24,862, based upon the following assumptions:
1. A feces sample is required from each subject in the pretest
and study populations (1,880 persons).
2. A feces sample is collected from each of those 1,880 persons.
3. It is not necessary to go beyond 1,880 persons.
4. Fifteen (15) minutes-of interviewer time are required per
subject to explain the procedure for collecting the feces
sample and obtaining the paint lead measurement (at $3.50/hour).
5. A ten dollar ($10.00) incentive per subject for the feces
sample will be sufficient.
6. One hundred (100) of the portable seats will be sufficient at
an estimated retail unit cost of $5.97 not including the
collection bags, at an estimated retail price of 5 bags for
$.94.
B. Chemical Analysis
1. Blood Clotting
This phenomenon will not only make FEP analysis impossible but will
require the use of more time consuming trace metal workup conditions.
Clotting is clearly undesirable and must be prevented. Implementation
of adquate anticoagulant treatment is absolutely essential.
-------
.14
2. Conversion of Organometallics Into An Assayable Form
Certain organometallics may escape detection by AA analysis. . This
potential analytical problem is of particular concern with arsenic. A
thorough recovery study of organo-arsenites (arsenates) will be conducted
and if required, appropriate sample preparation methods developed for
total metal assay.
It is also possible that other forms of the metals of interest will
resist degradation to assayable forms. More severe workup conditions
will be investigated on real samples to determine the magnitude of this
phenomenon.
3. Air Particulate Collection - Power Drops
An unknown factor in the placement of power drops for the Hi-vol
air samplers is the time lag between the request for such drops and
their installation. Undue delay will result in air sampling operations
continuing after all other smelter site activites are concluded. To
avoid this potential difficulty, air sampling strata will be identified
and requests for drops will be made at the earliest possible time.
-------
15
IV. DESCRIPTION OF FUTURE WORK
A. Field Operations
During November, drafts of the detailed work plan and data collec-
tion instruments will be submitted to the EPA Project Officer for re-
view. Also during November, RTI anticipates a final decision on the
pretest performance site, but will probably reserve this decision until
after the RTI Survey Director attends a meeting tentatively scheduled
for Phoenix the week of November 15. At that time, Mr. Harris plans to
visit both Ajo and Hayden, Arizona, and meet with various state and
local representatives.
B. Sampling Methodology
During November, mapping materials, to the extent they exist, will
be ordered. For the smaller communities and surrounding areas, the
highway departments, chambers of commerce, and banks have proven to be
good sources. For larger communities, the Census maps will also be
ordered. Extent aerial photography is available for most sparsely
populated areas and will be investigated as a possible source of mapping
information. Emphasis will be placed initially on obtaining materials
for Hayden, Arizona, which is a potential pretest site. The sampling
frame, the sample households, and the air-monitoring sites will be
documented sufficiently in advance of the data collection at each site
to allow for review by the. project officer.
Also during November, the work plan will be drafted. This work
plan will be consistent with the proposed methodology but will be as
explicit as possible and will contain more detail than the RTI proposal.
C. Chemical Analysis
Analytical studies already in progress (see Section IIC) will be
continued. In addition the following activities are either being initiated
or are planned for the coming report period.
-------
1. Soil
Soil workup and analysis is being initiated. Soil samples of
different types have been collected from Maryland, Missouri, North
Carolina and Wyoming. These materials will be extracted under a variety
of conditions and analyzed to establish operational parameters and to
determine the best control sample.
2. Blood
Aged blood has been obtained from the Duke Blood Bank and pre-
liminary work has begun on workup and analytical methodology. Our
initial work is based on a procedure that we have employed successfully
in another blood lead program. Blood is diluted with 0.5% HNO_ and
injected directly into the furnace. Calibration standards are prepared
by adding known amounts of metal to a similarly diluted control blood
sample. Estimation of total metal concentration is determined by the
method of standard additions. •
3. Dust and Air Particulates
Workup procedures will be evaluated. A variety of extraction
techniques will be used on a pooled dust collection. Glass fiber
filters will be spiked with known amounts of metal and the effectiveness
of different extraction procedures measured.
4. Free Erythrocyte Protoporphrin (FEP)
FEP calibration and analysis procedures will be investigated. The
procedure used will be the spectrofluorimetric method described by
Granick, et al.
Spiking and recovery experiments are planned for each metal in •
every sample matrix.
A time study will be carried out in which trace metal concentra-
tions, in the analytical solutions (after sample workup) will be deter-
mined and changes in titer noted.
------- |