PB84-128305
Evaluation of the Methods Used to
Determine Potential Health Risks Associated with
Organic Comtaminants in the Great Lakes Basin
Minnesota Univ., Minneapolis
Prepared for
Environmental Research Lab.-Duluth, MN
Jan 84
U.S. DEPARTMENT OF COMMERCE
National Technical Information Service
-------
Pb84-128305
EPA-600/3-34-002
January 1984
EVALUATION OF THE METHODS USED TO DETERMINE POTENTIAL
HEALTH RISKS ASSOCIATED WITH ORGANIC COMTAMINANTS IN
THE GREAT LAKES BASIN
by
Leonard M. Schuman, M.D., M.S., Professor
Conrad P. Straub, Ph.D., Professor Emeritus
Jack S. Mandel, Ph.D., M.P.H., Associate Professor
Stephan Norsted, M.P.H., Research Fellow
J. Michael Sprafka, M.S., Research Fellow
Division of Epidemiology
Division of Human Health and the Environment
School of Public Health
University of Minnesota, Minneapolis, MM 55455
EPA Grant 806282
EPA Project Officer
W. R. Swain
Large Lakes Research Station
Grosse lie, MI 48138
ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
DULUTH, MM 55804
-------
TECHNICAL REPORT DATA
(ftegg rtad Instructions on the rertnt btfon computing)
\. REPORT f>O.
FPA-600/3-84-Q02
2.
3. RECIPIENT'S ACCESSION NO.
P8?
4. TITLE AND SU8TITLI
Evaluation of the Methods Used to Determine Potential
Health Risks Associated with Organic Contaminants in
the Great Lakes Basin
S. REPORT OATS
.lanuarv 1984
A. PERFORMING ORGANIZATION CODE
7. AUTMOR(S)
L.M. Schuman, C.P. Straub, J.S. Mandel, S. Norsted and
J.M. Sprafka
. PERFORMING ORGANIZATION REPORT NO.
13. PERFORMING ORGANIZATION NAME AND ADDRESS
School of Public Health
University of Minnesota
Minneapolis, MN 55455
10. PROGRAM ELEMENTNO.
ll.CONTRACT/GR A N TNO.
806282
12. SPONSORING AGENCY NAME AND AOOAESS
Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Duluth, MN 55804
13, TYPE OP REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/600/03
18. SUPPLEMENTARY NOTES
18. ABSTRACT
These results suggest that "lake-bordering" populations (i.e., white populations)
experience higher rates of mortality due to stomach and esophageal cancers as compared
to "non-lake bordering" counties. This trend is consistent when the potential
confounding factor of large urban centers is removed.
7.
KEY WQROS ANO DOCUMENT ANALYSIS
DESCRIPTORS
b.lOENTIFIERS/OPEN ENDED TERMS
COSATl Field/Croup
J8. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
16. SECURITY CLASS ITUl Ripofl)
UNCLASSIFIED
21. NO. OF PAGES
41 f
M. SECURITY CLASS (T*itpogtf
UNCLASSIFIED.
23. PRIGS
EPA
' 1220-1 (»•». 4-77) ontvtoua ZOITIOW ID OUBOLSTB
-------
ATTENTION
PORTIONS OF THIS REPORT ARE NOT LEGIBLE
HOWEVER, IT IS THE BEST REPRODUCTION
AVAILABLE FROM THE COPY SENT TO NTIS.
-------
NOTICE
This document has been reviewed in accordance with
U.S. Environmental Protection Agency policy and
approved for publication. Mention of trade names
or commercial products does not constitute endorse-
ment or recommendation for use.
-------
TABLE OF CONTENTS
List of Tables vii
List of Figures ix
Acknowledgements x
Polychlorinated Biphenyls in the Environment
I. Introduction 1
II. Physical and Chemical Characteristics of PCBs 7
III. Sources and Translocation of PCBs in the Environment 16
Health Effects of PCBs
I. Animal Models 20
II. Human Exposures to PCBs and Other Organic Compounds 29
A. General 29
B. Dietary Exposures 32
C. Exposure via Breast Milk 35
D. Community Surveys 40
E. Occupational Studies 44
F. Teratogenic Research 46
III. Summary 47
Analyses of Morbidity and Mortality Patterns of Populations
Living in the Eight Great Lakes States 50
I. Analysis 1 51
A. Introduction 51
B. Methods 52
C. Results 53
D. Discussion 54
II. Analysis 2 54
A. Introduction 54
B. Methods 55
C. Results _ 56
i i i
-------
III. Analysis 3 56
A. Introduction 56
B. Methods 57
C. Results 57
D. Discussion 59
IV. Analysis 4 59
A. Introduction 59
B. Methods 60
C. Results 60
D. Discussion 62
V. Analysis 5 62
A. Introduction 62
B. Methods 63
C. Results 74
1. Urban and Rural Analysis 74
2. "Rural" Analysis 76
D. Discussion 79
A Pilot Study to Determine the Feasibility of an Epidemiologic
Investigation Among Commercial Fishermen of the Affects of
Polychlorinated Biphenyls on Health
I. Background and Rationale 85
II. Aims and Objectives 89
A. General 89
III. Methods 89
A. Study Protocols 89
B. Identification and Selection of a Study Population 90
C. Questionnaire 95
D. Protocol I Procedure 96
IV
-------
E. Protocol II Procedures 98
F. Protocol III Procedures 98
G. Timetable for Pilot Project Initiation 100
H. Validation Procedures 102
I. Tracing Procedures 104
J. Coding 108
K. Data Analysis 109
IV. Pilot Study Results 111
A. Protocol I 111
B. Protocol II 114
C. Protocol III 117
D. Discussion 1
V. Cost Analysis for Protocol Implementation 1
A. Protocol I 1
B. Protocol II 1
C. Protocol III 1
D. Validation 1
-E. Tracing 1
Conclusions 1
Methodological Approaches for Continued Research 1
Bibliography 1
Addendum 1
Appendix I Sates- and Ratios for Thirty-five Cancer Sites
Appendix II Study Questionnaires and Survey Instruments
Appendix III Pilot Study and Survey Materials
-------
Appendix IV Proposed Set of Coding Instructions
Appendix V Proposed Set of Analytical Procedures
Appendix VI Progress Reports
VI
-------
List of Tables
page
Table 1 Molecular Composition of Some Aroclors (part A) 4
Table 2 Molecular Composition, of Some Aroclors (part 3) 5
Table 3 Physicochemical Characteristics for Several PCB Compounds... 8
Table 4 Solubility of Chlorobiphenyls and Isomers in Water 10
Table 5 Vaporization Rates of Several Aroclors. 11
Table 6 Summary of the Health Effects of PCBs on Animal Models... .21-23
Table 7 Summary of the Health Effects of PCBs on Animal Models... 24
Table 8 Summary of the Health Effects of PCBs on Animal Models... 24a
Table 9 Summarization of the Toxic Effects of PCBs on Aquatic
Animals 26-28
Table 10 Levels of Organochlorine Residues in Human Adipose
Tissues 31
Table 11 Percent of U.S. Breast Milk Samples With Levels of
Chlorinated Hydrocarbon Pesticides or Their Metabolites
at £51 ppb by Geographical Region (n = 1,436) 37
Table 12 Organochlorine Pesticide Residue Levels in Whole Milk
Samples from Women Residing in Arkansas and
Mississippi (n = 57) 39
Table 13 Percent of Lake Bordering and Non-Lake Bordering Counites
Having Morbidity/Mortality Rates Discrepant by > 1
Standard Deviation From State Means 58
Table 14 Means, Standard Deviations, and Comparisons of Mean
Age-Adjusted Cause-Sex-Race-Specific Mortality Rates
per 100,000 for Rural "Lake Bordering" and Rural "Non-
Lake Bordering" Counties of the Great Lakes Basin 61
Table 15 Counties Designated to the 1st, 2nd, and 3rd Orders by
State 64-70
Table 16 Summary Population Figures for the Great Lakes Basin by
Order, Sex, and Race (1960) 71
Table 17 Summary Population Figures for the Great Lakes Basin by
Order, Sex, and Race (1960) Counties with urban centers
containing populations > 100,000 are omitted 72
-------
Table 18 Cancer Sites Demonstrating INCREASING Average Mortality
Rates With Proximity to the Great Lakes Basin 80
Table 19 Rural Analysis - Omitting Counties With Population
Centers >100,000
Cancer Sites Demonstrating INCREASING Average Mortality
Ratas With Proximity to the Great Lakes Basin...; 81
Table 20 Cancer Sites Demonstrating DECREASING Average Mortality
Rates With Proximity to the Great Lakes Basin 82
Table 21 Rural Analysis - Omitting Counties "With Population
Centers 100,000
Cancer Sites Demonstrating DECREASING Average Mortality
Rates With Proximity to the Great Lakes Basin 83
Table 22 Levels of PCBs in Canadian Fish Species 86
Table 23 Sources of Information Regarding Commercial Fishermen 91
Table 24 Number of Commercial Fishermen in the Great Lakes Basin
by Lake and State 93
Table 25 Status of Protocol 1 Participants 112
Table 26 Status of Protocol II Participants 116
Table 27 Status .of Protocol III Participants 119
Table 28 Preliminary Results for Protocols I and II 122
Table 29 Historical Perspective of Commercial Fishing Records 148
Table 30 Breakdown of Study Eligibles, Respondents and Survey
Response Rates "... 152
Table 31 Distribution of Study Eligibles, Respondents and
Response Rates by State .153
Table 32 Preliminary Results for Great Lakes Study 155
VI 1 1
-------
List of Figures
page
Figure 1 Stereochemistry of PCBs 2
Figure 2 Simulated Movement of PCBs Through the Environment 18
Figure 3 Flow Chart of Protocol I Procedures 99
Figure 4 Flow Chart of Protocol III Procedures 101
Figure 5 Protocol Implementation 103
Figure 6 Flow Chart of Validation Procedures 105
Figure 7 Flow Chart of Tracing Procedures 107
-------
Acknowledgements
The assistance of the following in the project is gratefully acknowledged:
Division of Epidemiology
Deb Englehard
Ann Berry
Joan Maronde
Pat Mingee
Irwin Pollack
Virginia Sykes
Division of Biometry
James Boen
Laurie Reinhardt
-------
Polychlorinated Biphenyls in the Environment
I. Introduction
Polychlorinated biphenyls (PCBs) consist of a group of com-
pounds produced by the complete or partial chlorination of a biphenyl
molecule. These compounds belong to a family of chlorinated hydro-
carbons in which chlorine has replaced hydrogen in the molecular
structure (54 ). The stereochemistry of a PCB molecule is demonstrated
in Figure 1. The molecule consists of two benzene rings joined at
their apices with the potential for chlorine substitution on the
remaining ten sites on the rings. The number of theoretically possible
isomers is 209, but far fewer have been found in the environment (54 ).
This may result from manufacturing processes producing specific molecu-
lar compositions eliminating the undesirable chlorinated species and/or
environmental metabolism and degradation.
PCBs were discovered in 1881, but industrial applications and
commercial production were discovered and undertaken in 1929 (50 ). The
Monsanto company was the sole producer of PCBs in the United States and
marketed its products under the trade name Aroclor. Other PCB producers
are found in Germany, France, Italy, Japan, Russia, and Czechoslavakia
and market their products under a variety of trade names ( 54).
Aroclor PCB mixtures are identified by a four digit number with
the first tiwo digits representing the type of molecule (12 - chlorinated.
biphenyl) and the latter two the average percentage of chlorine by
weight in the mixture. For example, Aroclor 1248 represents a PCB mix-
ture with approximately 48 percent chlorine by weight. Tables 1 and 2
depict the molecular composition of several Aroclor products. A compar-
ison of the reported values for Aroclors 1242 and 1254 in Tables 1 and 2
-------
Figure 1
Sterozhenistrv of FCBs
Numbering system for the biphenyl molecule
4'
2'
exarole
Cl
ci
Cl
Cl
3, 4, 4', 5' - tetrachlorobiphenyl
-------
demonstrates variability in the percentages of chlorinated species.
This may be attributable to differences in quantitative analysis and/or
the products themselves. In addition, the variability in total and
isomeric PCB percentages may be compounded when evaluating the analyti-
cal results of PCBs extracted from environmental matrices. In 1970,
when these widespread industrial compounds were recognized as a potential
problem, Monsanto restricted the domestic sales of Aroclors to uses in
closed electrical and hydraulic systems which effectively cut in half
the annual U.S. sales of PCBs. Veith reported data obtained from
initial monitoring of PCBs in waters around the Green Bay, Wisconsin
region indicating that PCB levels decreased sharply in 1971 suggesting
that the self-imposed limitations on PCB production by Monsanto had
immediate effects (112).
The production of PCBs by Monsanto stoppei in August, 1977 and
the sales of inventory stock ceased two months later. The complete ban
on PCBs came into effect on July 1, 1979 with ;he enactment of the
Toxic Substances Control Act (PL-94-46) ( 1 ). However, over the years,
approximately 1.25 billion pounds of PCBs have been purchased by U.S.
industries. Of that amount,it is estimated that 55 million pounds of
PCBs have been destroyed by incineration or environmental degradation,
290 million pounds have been disposed in landfills and are assumed to
have retained their toxic properties, and 150 million pounds are "free"
in the environment (i.e., air, water, sediments, and animal tissues).
The remaining 755 million pounds purchased by U.S. industries are
currently in use (29). In addition, several foreign companies still
produce PCBs without any restrictions on their use and U.S. companies
are free to Import these products. Estimates by the Environmental
3
-------
Table 1
Molecular Composition of Some Aroclors
A. Chlorobiphenyl composition
% Aroclor product
C H Cl
129
C H Cl
128 2
C H Cl
127 3
C H Cl
126 k
C H Cl
1 2 5 S
C H Cl
1 2 W 6
C H Cl
1 2 S 7
C H Cl
122 6
C HC1
12 9
1242
3
13
28
30
22
4
-
-
_
1248 1254
-
2 _
18
40 11
36 49
4 34
6
-
_ _
1260
-
-
-
-
12
38
41
8
1
Hutzincer, 0. et al.. The Chemistry of PCBs. CRC Press,
Cleveland (1974). p. 23
-------
Table 2
Molecular Composition of Some Aroclors
B. Chlorobiphenyl composition 1221 1016 1242 1254
C H
12 10
C H Cl
12 9
C H Cl
128 2
C H Cl
127 3
C H Cl
126 t
C H Cl
125 S
C H Cl
1 2 H 6
C H Cl
123 7
C H Cl
122 8
11
51
32
4
2
<0.5
ND*
ND
ND
<0.1
1
20
57
21
1
<0.1
ND
ND
<0.1
1
16
49
25
8
1
<0.1
ND
<0.1
<0.1
0.5
1
21
48
23
6
ND
* ND - Not Detected ( 0.01%)
Hutzinger, 0. et al. The Chemistry of PC3s. CPC Press
Cleveland (1974). p. 23.
-------
Protection Agency indicate chat the United States has been importing
500,000 pounds of PCBs per year as of 1976 and an unknown, amount .of
PCB-containing products ( 1 ).
In order to evaluate the potential environmental effects of PCBs
sampling surveys must produce quantitative information regarding PCB
concentrations in various environmental strata (i.e., air, water,
sediments, and food products). These data, characterizing PCB behavior
in the environment, may help to predict transportation routes and
identify high risk exposure groups in human populations. The purpose
of this section is to review the physical and chemical characteristics
of PCBs and the mechanisms regarding introduction and transport through
the environment. In addition, a review of the potential health effects
of PCBs on animal models and humans will be presented.
-------
II. Physical and Chemical Characteristics of PCBs
The physical and chemical properties of PCB isomers and mix-
tures vary considerably when these materials are tested in the laboratory.
Chlorine substitution markedly affects the three physicochemical char-
acteristics important in the evaluation of the potential environmental
impact and these include solubility, vaporization rates, and the
capacity to biodegrade. The solubility and volatility of a particular
PGB compound represent potential mechanisms for introduction into the
environment and will determine the ease of distribution throughout
ecosystems. Biodegradation represents a natural mechanism for the
potential redistribution or removal of these materials from the environ-
ment.
The solubility of PCBs in water is low; however, reported values
vary according to the isomeric structure, PCB type, and the composition
of the aqueous test solutions. Dexter and Paulou analyzed several PCB
isomers and found that their solubilities were five times greater when
tested in sea water as compared to distilled water. This suggests that
the potential for environmental contamination may be greater in ocean
and estuary environments than fresh water environments (28 ).
In general it can be stated that as chlorine content increases,
the solubility and volatility decreases. Table 3 depicts the
differences associated with chlorine content, as evidenced by PCB
type, and several physicochemical parameters (i.e., water solubility,
vapor pressure, theoretical half-life) (29 ). These findings indi-
cate that the association between increases in chlorine content and
predicted decreases in solubility and volatility is not consistent
with respect to Aroclor 1254. In addition, the values predicted for
7
-------
Table 3
Physicochemical Characteristics for Several PCB Compounds
Theoretical Half-life
Solubility/ Vapor Pressure/ for Vaporization
PCB Type mg/1 mm Hg From 1 Meter H2 0 Column
1242 0.24 4.06 x lO"** 5.96 hours
1248 5.4 x 10~2 4.94 x 10~ 58.3 minutes
1254 1.2 x 10~2 7.71 x lo"" 1.2 minutes
1260 2.7 x 10~2 4.05 x 10~5 28.8 minutes
Durfee, 3. L. et al. PCBs In the United States Industrial Use
arid Environmental Distributions. Versar, Inc. EPA 560/6-76-005
(1976). p. 43.
8
-------
the vaporization half-life of these Aroclors from a one meter column of
water demonstrate inconsistency regarding PCS type (i.e., chlorine
content). The solubility variations between individual PCB isomers
are evident from data reported by Durfee in Table 4 . Increasing
chlorine substitution decreases the water solubility. The vaporization
rates of several.Aroclors are shown in Table 5 demonstrating that as
chlorine content increases the vaporization rate decreases.
In the environment, however, PCBs are subject to many influences
which may alter standard laboratory classifications. As chlorine
content increases in PCB molecules the solubility and volatility
decrease while the sorption capacity increases (15 ). The particular
matrix to which these molecules sorb will determine their availability
in the environment. For example, the strength of adsorption in sedi-
ments and soils demonstrates a positive correlation with the concentra-
tions of humic acid, ilite clay, and Del Monte sand in that order (95 )•
In addition, Hague and his associates report that vapor losses from a
sand surface are significantly greater than losses observed from an
organic soil ( 43 ). Pierce and his associates concluded that suspended
humic particulates may be an important transportation mechanism through
the water column ( 96 ). Hence, similar PCB molecules sorbed to
different environmental substrates may have different solubility and
volatility characteristics and may move through the environment at
different rates. Evaluation of these variables ±s essential in order
to assess the potential environmental impact of PCBs.
The degradation of PCBs in the environment by two major routes
has been described in the literature. These include microbial activity
and photochemical reactions. The microbial degradation of PCBs involves
9
-------
Table 4
Solubility of Chlorobiphenyls and Isomers in Water
Compound
Monochlorobiphenyls
2-
2-
4-
Dichlorobiphenyls
2, 4 -
2, 2' -
2, 4' -
4, 4' -
Trichlorobiphenyls
2, 4, 4' -
21, 3, 4
Tetrachlorobiphenyls
2,
2,
2,
2,
2,
2,
3,
o
2
2
O
3
3
3
, 5,
, 3,
/ 3,
, 4,
, 4,
, 4'
• 4,
5'
3'
5'
4'
4'
, 5
41
-
-
-
-
-
-
-
' -
Pentachlorobiphenyls
2, 2', 3, 4, 5' -
2, 2', 4, 5, 5' -
Hexachlorobiohenyl
2, 2', 4", 4', 5,
Octachlorobiphenyl
•> 51 T -31 4
•£ I • £ I Jf J / *> f
Decachlorobiphenyl
4, .4' - Dichlorobiphenyl
+ Tween 80 0.1%
-i- Tween 80 1%
+• Humic acid extract
i_
Solubility rng/1
5.9
3.5
1.19
1.40
1.50
1.88
0.08
0.085
0.078
0.046
0.034
0.170
0.068
0.058
0.041
0.175
0.022
0.031
0.0088
0.0070
5.9
>10.0
0.07
Durfee, R. L. et al. PCBs in the United States Industrial Use
and Environmental Distributions. Versar, Inc. EPA 560/6-76-005
(1976) ?. 47.
10
-------
Table 5
Vaporization Rates of Several Aroclors
(Surface area:
12.28 cm2)
1221
1232
1242
1248
1254
1262
1260
1270 (Deca)
Wt. Loss,
g
0.5125
0.2572
0.0995
0.00448
0.0156
0.0039
0.0026
0.0015
Exposure at
100°C, hr.
24
24
24
24
24
24
24
24
Vaporization rate
g/cm2/hr
0.00174
0.000874
0.00038
0.000152
0.000053
0.000013
0.000009
0.000005
Durfee, R. L. et al. PCBs in the United States Industrial Use and
Environmental Distribution. Versar, Inc. EPA 560/6-76-005
(1976) p. 45.
11
-------
hydroxylation and oxidation reactions which convert-aromatic compounds
into ring-fission, substrates (25 ). These reactions may follow dif-
ferent metabolic routes in aerobic environments as compared to anaerobic
environments which may ultimately affect the rates of microbial degrada-
tion and the availability of these contaminants to higher organisms.
Rogoff reported that soil pseudomonads appear very active in the
degradation of PCB compounds. He demonstrated that cleavage of the
ring is preceded by hydroxylation necessary to provide two hydroxyl
groups on the aromatic ring. These usually occur in the ortho position
and the ring is cleaved across a bond adjacent to one of the carbon
atoms which bears a group (102).
Furukawa.and his associates studied the effects of chlorine sub-
stitution on the degradability of PCBs using species of Alcaligenes and
Actinobacter. They concluded that degradation decreased as chlorine
substitution increased. PCB isomers containing more than four chlorine
atoms were less susceptible to degradation. Preferential ring fission
of the PCB molecules occurred with non-chlorinated or lesser chlorinated
rings. They also noted that positional variations in the isomers
affected the rate of degradation. Those PCBs containing all the chlorine
atoms on a single ring were generally degraded faster than when the
same number of chlorine atoms were substituted on both rings (37 ).
Ahmend and Focht report data which indicate that some species of
Achromobacter are capable of oxidizing PCB isomers with two to five
chlorine atoms. However, the bacteria are unable to dehalogenate any
of these chlorinated biphenyls as noted by the absence of chloride in all
the supernatants. They conclude that increasing the chlorine substitu-
tion renders the biphenyl molecule more resistant to microbial attack
12
-------
( 3 ). In subsequent studies by these investigators they succeeded in
isolating two species of Achromobacter from sewage effluent using bi-
phenyl and p-chlorobiphenyl as sole carbon sources ( 2 ).
Sayler and his associates report that both PCB-degrading bacteria
and PCBs were found in higher concentrations in estuarine waters and
sediments compared with marine samples. Their data indicated a greater
correlation between PCS concentrations and numbers of PCB-degrading
bacteria in areas of urbanization as opposed to areas distant from land.
Their results suggest that the assessment of existing PCB-degrading
bacteria in environmental samples may be used as indicators for potential
PCS contamination. They caution, however, that the number of PCB-
degrading microorganisms in a given sample cannot be directly correlated
with the concentration of PCBs in that sample (106).
It may be assumed from these studies that certain microbial species
have the capacity to degrade these complex molecules but, in the natural
environment, will preferentially attack simpler energy forms. If situ-
ations occur in the environment where simple carbon sources are exhausted,
microbial species could, theoretically, develop pathways- to metabolize
more complex molecules like PCBs. However, the extent of microbial PCB
degradation by natural microbial populations remains undetermined.
Photodecomposition is considered to be another potential mechanism
for PCB degradation because of its similarities with chlorinated hydro-
carbon pesticides (.i.e., DDT) and the mechanisms associated with their
degradation (.104). However, Durf ee _e_t -al_., conclude that the probability
of UV disassociation of chlorobiphenyls appears significantly less than for
the chlorinated pesticides ( 29).
Hutzinger and his associates investigated the photochemical behavior
13
-------
of 2, 2', 4, 4', 6, 6* hexachlorobiphenyl in organic" solvents. Results
Indicate a loss of chlorine, rearrangement, condensation, and the for-
mation of new compounds. They caution that photochemical production of
new compounds may lead to further complications in residue analyses
with the possible appearance of chlorobiphenyls in environmental samples
that are not found in commercial samples. Additional studies by Hutzinger
et al., indicated that higher chlorinated biphenyls appear to degrade
faster than those with lower chlorine content when exposed to a radiation
source ( 53,54 )•
These results may have limited applicability to an environmental
situation since these experiments were conducted in organic solvent
systems (i.e., hexane) rather than water. Additional experimental
strategies such as irradiation of chlorobiphenyls in the gas and solid
phases as well as the relationship of positional isomerism and photo-
chemical activity must be examined as potential mechanisms of PCB
degradation in the environment.
The environmental behavior of PCBs is also influenced by their
solubility in lipids and fats -resulting in a tendency to bioaccumulate
in organisms progressively higher in the food chain (118). As a con-
sequence PCB routes through the environment may be subject to specific
metabolic mechanisms which, in turn, are subject to the solubility
differences of PCBs in the lipids of these organisms (88 ).
Biological concentration is a process whereby an organism demon-
strates a greater concentration of a particular chemical factor as
compared to the surrounding substrate or various fo.od sources. This
phenomenon is influenced by such factors as the amounts of .chemical
present in the diet or surrounding environment, the chemical and
14
-------
physical forms of the contaminant, the feeding and behavior traits of
the animal species in question, the degree of assimilation through
cellular barriers, and the extent of retention in the tissue (41 ).
Nisbet and Sarofim report that the more highly chlorinated isomers
of PCBs are retained more efficiently; probably the result of metabolism
and differential excretion ( 88 ). Their preliminary conclusions were
that the more highly chlorinated isomers are not significantly differ-
entiated as they pass through the food chain up to fish and birds.
They concluded that differential metabolism is the primary mechanism in-
the environmental differentiation of isomers. Risebrough and his associ-
ates believe that current data suggest that the amounts and kinds of
lipids may affect the capacity for retention of PCBs, modifying trophic
accumulation predicted by the classical food chain concentration theory
(100 )•
Several strategies have been proposed to study the bioconcentration
phenomenon and these include self-containing ecosystems, dynamic circu-
lating systems, and an equilibrium-type ecosystem. Self-containing
ecosystems are limited because of maintenance considerations and expense
but offer a totally simulated but contained ecosystem. Dynamic circulating
systems are those in which the contaminant is introduced into the system
at a prescribed rate while equilibrium ecosystems are those in which the
contaminant is present in the environment and the organisms are intro-
duced into the system ( 95 ). However, any model used to evaluate
bioconcentration mechanisms has to incorporate and adjust for variations
in eating habits, lipid content, age, sex, and size, all of which affect
PCS levels in higher organisms.
15
-------
III. Sources and Translocatlon of PCgs in the Environment
Environmental contamination with PCBs may be the result of
both point and non-point sources. Point sources include vaporization
or leaking from transistors and capacitors, municipal- and industrial-
effluent discharges, leachates from landfills and dumps, and the incin-
eration practices of manufacturers and municipalities. Non-point
sources include land application of municipal sludges and chemicals
followed by runoff, and the atmosphere ( 85 ).
Recent evidence suggests that a major source of PCBs may be the
degassing of sanitary landfills and industrial dumps. Investigators
have determined a difference of three orders of magnitude between
ambient air and vent gas PCS concentrations ( 46 ). Gaffney^demonstrated
that chlorobiphenyls can be produced in the effluents of municipal
wastewater treatment plants which practice chlorination and receive.
industrial wastes containing biphenyls (i.e., textile industries) ( 38).
The experimental technique, however, was one of "super-chlorination"
so the results may not be applicable to municipalities conducting
normal treatment operations. Lawrence and Tosine report that localized
areas of heavy industry may be responsible for trace organic contaminants
in sewerage systems. These contaminants become associated with the
suspended solid content of the sewage and are incorporated in the
sludges following primary and secondary treatment (75 ). Land applica-
tion, of these sludges..may increase the potential for environmental
contamination following runoff.
Maugh has investigated natural sources of chlorobiphenyls and
hypothesizes that a common metabolite of plants and animals, dichloro-
benzophenone, may be degraded to a dichlorobiphenyl compound by photo-
16
-------
lysis ( 81). However, reaction rates are very slow and the exact
chemical pathways are as yet undetermined. In addition, most PCBs in
the environment have been identified as penta- and hexachlorobiphenyls
which suggests that PCBs come from artificial (anthropogenic) rather
than natural sources ( 88 )• Figure 2 demonstrates the potential
mechanisms by which PCBs are introduced and transported throughout
various ecosystems.
It is generally accepted that the atmosphere is the major route of
entry and subsequent transport through the environment as evidenced by
its ubiquitousness ( 88 ). Once in the atmosphere PCBs are distributed
between the gaseous and particulate phases and are transported through-
out the environment via the prevailing air masses. Most authors agree
that atmospheric transportation occurs more frequently in the vapor-
aerosol state rather than the particulate state ( 16,45,46,85 ).
Harvey and Steinhauer hypothesize that during the atmospheric
lifetime of chlorinated hydrocarbons the vapor-aerosol and particulate
states will equilibrate and exchange many times depending on the condi-
tions ( 45 ). Both dry and wet deposition are potential mechanisms for
loading into aquatic and terrestial ecosystems.
Water transport is considered to be more localized because of the
low solubility of PCBs. PCS discharges into the aqueous environment
will, for the most part, be associated with particulate matter and will
settle out as sludges or become adsorbed to the bottom muds. However,
uptake by zooplankton and fish and subsequent excretion may provide a
transportation route through the aquatic ecosystem via fecal pellets.
Elder and Fowler proposed that PCBs are carried to the sediment by
rapidly sinking particles (i.e., fecal pellets) and these make a sig-
17
-------
Figure 2
SIMULATED MOVKMBNT OF 1'Cllo THROUGH TUB ENVIRONMENT
Mnvf»nu:nt In
00
Monu far. taring
(spills, leaks, breaks)
Capacitor processing
nnd disposal
Municipal and industrial
effluents
Incineration practices
of industries and
municipalities
Water
HovcmonC Out
Volatility (aeration)
Rainout/fallout
Air
Land and
Landfills
Leachate
Rainout/fallout
Photo-
decorepo-
sition
Volatility (vent gases)
Runoff
Dredging
Adsorption
Desorption
Aquatic
Sediments
Uptake/
Bioconcentration
Aquatic Biota
Clearance (fecal pellets)
Biodegradation
Higher Organisms
-------
nifleant contribution to the vertical transport of these substances in
aquatic environments. They reported data demonstrating PCB concentra-
tions in feces (dry weight) ranging from 3.5 to 21 times higher than
that in organisms which formed the feces ( 31 ).
In conjunction with the fecal transport hypothesis consideration
must be given to the metabolic actions by aquatic organisms (i.e. , fish)
on PCB contaminants. Metabolic functions followed by excretion may
alter PCB structure which, in turn, may influence the toxicity or avail-
ability of these materials to other organisms in the aquatic ecosystems.
In addition, these metabolized products may alter PCB structure compli-
cating quantitation.
19
-------
Health Effects of PCBs-
I. Animal Models
Several excellent reports have been written which review the
health effects of PCBs in experimental animal models. These include the
IARC Monograph on the Evaluation of the Carcinogenic Risk of Chemicals
to Humans ( 55 ), The Health Effects of PCBs with Particular Emphasis
on Human.High Risk Groups C 21 ), and the Final Report of the Subcom-
mittee on the Health Effects of Polychlorinated Biphenyls and Poly-
brominated Biphenyls (33 ). These reports are briefly summarized in
Tables 6-8 respectively.
In general, the principal organs affected by long-term .exposure to
commercial mixtures of PCBs in animals are the gastro-intestinal tract,
the liver, and the lymphatic system (21). Reported pathological effects
include carcinogenic and tumorigenic potential, immunosuppression, embryo-
toxicity, and reproductive dysfunctions (see Tables 6-8 ). In addition,
PCBs have several sublethal effects such as microsomal enzyme induction,
clastogenic activity, and modified cell development.
Thompson e_t al_. conclude that Aroclor 1254 is a wide spectrum
inducer of mutagenic activity based on their studies involving the
effects of PCBs on rat-liver homogenate (59 preparations). In addition,
they reported that the hydroxylations of the number 2 and 4 chlorines
proceed via the areiie oxide intermediate pathway which has been impli-
cated in carcinogenesis and mutagenesis 0-11) •
Hargraves and Allen investigated the in vitro binding of tetra-
chlorobiphenyl (.TCB) to rat liver microsomes and reported a protein(s)
in the induced system capable of binding a TCB metabolite. They con-
clude that their results support the concept of a metabolic activation
2Q
-------
TABLE 6
Summary of the Health Effects of PCBs on Animal Models
1) Carcinogenesis
Mice:
hepatocellular carcinoma
hepatoma
Rats:
adenofibrosis of the liver
cholangiofibrosis
hepatocellular carcinoma
neoplastic liver nodule
2) Toxic Effects
Rats:
liver hypertrophy, marked fatty infiltration and degeneration
of parenchymal cells
Chickens:
subcutaneous edema, ascites, hydropericardium
• loss of fat
marked involution of the thymus
atrophy of the spleen
increased liver weight
congestion, mild necrosis and marked infiltrations of the liver
widespread hemorrhage and focal necrosis in the kidney
Rabbits (skin application):
weight loss
hyperkeratosis of the skin
liver-cell atrophy
kidney cell degeneration
atrophy of the thymus
lymphopenia
Rabbits (oral):
liver hypertrophy
atrophy of the uterus
Monkeys:
acne
swelling of upper eyelids
loss of eye lashes
alopecia
subcutaneous edema
gastritis
ulceration
hypoproteinemia
anemia 21
-------
Table 6 (Continued)
3) Immune Effects
Guinea Pigs:
lymphopenia, thymic, and splenic atrophy
decrease in number of circulating lymphocytes
Rabbits:
lymphopenia, thymic, and splenic atrophy
decrease in number of circulating lymphocytes
Monkeys:
lymphopenia, thymic, and splenic atrophy
decrease in number of circulating lymphocytes
Duck:
increased hepatitis susceptibility
4) Endocrine Effects
Eats:
estrogenic effect
Nonhuman primates:
increased level of urinary ketosteroid-s
prolonged menstrual cycles with increased bleeding
5) Eabryotoxicity, Teratogenicity, and Reproductive Effects
Rats:
increased mortality of the offspring
reduced mating performance
reduced litter size
decreased survival of the offspring
decreased maternal weight gain
decreased fetal weight
learning disability
Chickens:
reduced hatchability
abnormalities in the embryo
Rabbits:
fetotoxic abortions and stillbirths
increased maternal deaths
22
-------
Table 6 (Continued)
Monkeys:
facial acne and edema
swelling of the eyelids
loss of facial hair
hyperpigmentation of the skin
gastritis
keritinization of the hair follicles
early abortions
small offspring
reference: IARC Monographs on the Evaluation of the Carcinogenic
Risk of Cehmicals to Humans. Volume 18, (1978).
23
-------
TABLE 7
Summary of the Health Effects of PCBs on_Animal Models
a) Health Effects of PCB: Animal Studies
1. Effects on Liver and Spleen:
weight increase, fatty degeneration, and necrosis of the liver in
birds and a variety of mammals
increased activities of some drug metabolizing enzymes
reduced vitamin A storage in the liver
altered lipid metabolism
depletion of lymphatic nodules in the spleen in pheasants
2. Carcinogenesis and Tumorigenesis:
hepatic adenofibrosis in rats
biliary epithelial hyperplasia in rats
induction of nodular hyperplasias in rats
hepatocellular carcinomas in mice
3. Effects in Adrenal Gland:
morphological alteration in the zona fasciculata of adrenal gland in rats
increased levels of cor.ticosterone in rats
4. Effects on the Reproductive System:
decreased mating indices in rats
number of young delivered, number surviving to weaning, both decreased
increased number of stillborns
decreased egg production in chickens
decreased egg shell thickness
decreased reproductive capacities in pheasants
5. Effects in Chromosomes:
chromosomal aberrations in Ring Dove Embryos
6. Immunosuppression:
reduction in lymphoid tissue, and the presence of amyloid or amyloid-like
material in the liver of chicks
lymphopenia in rabbits
decreased number of antibody-forming cells after stimulation of lymphoid
system in guinea pigs.
Reference: "The Health Effects of PCBs with Particular Emphasis on Human High
Risk Groups," Edward J. Calabrese, Reviews on Environmental Health
Vol. 2(4), 1978.
24
-------
TABLE 8
Summary of the Health Effects of PCBs on Animal Models
Animal Toxicology
1) Reproduction depression:
mink
rhesus monkey
rata
2) Hepatic porphyria:
chicken
rabbit
Japanese quail
rats
3) Liver damage and tumors including hepatocellular carcinoma:
rat
mouse
primate
4) Bone marrow depression:
primate
5) Gastric mucosa damage:
rat
dog
primate
6) Atrophy of the thymus:
rabbit
7) Fluid accumulation:
chickens
finches
primates
Reference: Final Report of the Subcommittee on the Health Effects of Poly-
chlorinated Biphenyls and Polybrominated Biphenyls, Dept. of
HEW, Washington, D.C., July, 1976.
24 a
-------
of microsomes resulting from PCB insult ( 44 }. Subsequent studies by
Stadnicki je£ al. evaluated the potential clastog«nic effects of TCB and
its phenolic and arene oxide intermediates. They report that 2, 2', 5,
5', TCB-3-4 epoxide was more potent in causing DNA single-strand breaks
than either 2, 2', 5, 5', TCB or a mixture of 3-hydroxy or 4-hydroxy-2,
2', 5, 5' TCB. They conclude that PCB epoxide intermediates may poten-
tiate carcinogenesis ( 108).
Ohnishi and Noda evaluated the effects of Kanechlor 400 on dissoci-
ated conjunctival cells in vitro and reported an increase in the mitotic
times of epithelial cells exposed to PCBs. Greater cellular damage was
observed as PCB concentrations were increased, infrastructure changes
of epithelial cells included an increase in the number of vacuoles,
lysosome-like particles, myelin-like figures, and a dilute rough endo-
plasmic reticulum filled with secretion in the cytoplasm ( 89 ).
Hoopingarner and his associates investigated the effects of PCBs
on Chinese hamster cells and primary human lymphocyte cells. In general,
there was a steady drop in cell numbers as the percentage of chlorine in
the PCB mixture decreased for both types of cells. They reported no
apparent effects on chromosomal integrity as measured by cytological
evidence ( 49).
In addition to the health effects of PCBs on experimental animal
models, several investigators have evaluated the toxicity of PCBs on
fish and other aquatic animals. A summary of these results appears
in Table 9 , Of particular interest is the "bioconcentration factor"
which has been reported by several researchers. Essentially, this
"bioconcentration factor" is determined by measuring the contaminant
concentration in water or food sources and the resulting contaminant
25
-------
TABLE 9
Summarization of the Toxic Effects of PCBs on Aquatic Animals
Animal Model:
Fathead Minnow
Experimental Methodology:
Flow-through bioassays were conducted to determine both acute and chronic
effects of the larvae and adults, as well as the bioconcentration of the
mixture of FCBs in the fish.
Experimental Results and Conclusions:
30-day LC-50 for newly hatched larvae was 4.7 mg/1 for Aroclor 1248
30-day LC-50 for newly hatched larvae was 3.3 mg/1 for Aroclor 1260
reproduction occurred at or below 3 mg/1 1248 and at or below 2.1 mg/1 1260_
bioconcentration factor in adult females at 25°C was approximately 1.2 X lO
for Aroclor 1248 and 2.7 X 105 for Aroclor 1260
females accumulated about twice as much as males
the chromatograms of 1248 and 1260 residue were essentially identical to
.the standard after a 200 day exposure. Subtle changes occurred in the
first and second peaks (Defoe, 1978)
Animal Model:
Fathead Minnow
Experimental Methodology:
Continuous-flow bioassays were conducted to determine safe levels of
Aroclor 1242, 1248, and 1254 for the fathead minnow
Experimental Results and Conclusions:
96-hour LC-50 for newly hatched larvae was 7.7 mg/1 for Aroclor 1254
96-hour LC-50 for newly hatched larvae was 15.0 mg/1 for Aroclor 1242
96-hour LC-50 for 3 month old fatheads was 300.0 mg/1 for Aroclor 1242
reproduction occurred at or below 1.8 mg/1 1254 and at or below 5.4 mg/1
1242
males accumulated more 1254 and 1242 than females after a period of
eight months (Nebeker et.al., 1974)
Animal Model:
Flagfish
Experimental Methodology:
continuous-flow bioassays were conducted to determine safe levels of
Aroclor 1248 for the flagfish (Jordanella floridae)
Experimental Results and Conclusions:
no survival at or above 5.1 mg/1 and did not grow well above 2,2 mg/1
(Nebeker, 1974)
26
-------
-TABLE 9 (Continued)
Model:
Daphnia
Experimental Methodology:
Continuous-flow and static bioassays were conducted at 18°C with survival
and reproduction as measures of relative toxicity of Aroclors 1221, 1232,
1242, 1254, 1260, 1262, and 1268
Three PCB-mixture bioassays were also conducted
Experimental Results and Conclusions:
3 week LC-50 for Daphnia under static conditions was 25 mg/1 for Aroclor 1248
3 week LC-50 for Daphnia under continuous-flow conditions was 1.3 mg/1 for
Aroclor 1254 (Nebeker and Puglisi, 1974}
Animal Model:
Gaaaaarus
Experimental Methodology:
Continuous-flow and static bioassays were conducted at 18°C with survival
and reproduction as measures of relative toxicity of Aroclors 1221, 1332,
1242, 1254, 1260, 1262, and 1268
Three PCB-mixture bioassays were also conducted
Experimental Results and Conclusions:
96-hour LC-50 for Garmnarus under continuous-flow conditions was 73 mg/1 for
Aroclor 1242
96-hour LC-50 for Gammams under continuous-flow conditions was 20 mg/1 for
Aroclor 1248
60-day survival percentage was 52% at 8.7 mg/1 1242 and 53% at 5.1 mg/1 1248
(Nebeker and Puglisi, 1974)
Aniaal Model:
Tanytarsus
Experimental Methodology:
Continuous-flow and static bioassays were conducted at 18°C with survival
and reproduction as measures of relative uoxicity of Aroclors 1221, 1232,
1242, 1248, 1254, 1260, 1262, and 1268.
Three PCB-mixture bioassays were also conducted
Experimental Results and Conclusions:
emergence did not occur above 5.1 mg/1 1248 or 3.5 mg/1 1254
3 week LC-50 for Tanytarsus larvae under continuous-flow conditions was
0.65 mg/1 for Aroclor 1254
3 week LC-50 for Tanytarsus pupae under continuous-flow conditions was
0.45 mg/1 for Aroclor 1254 (Nebeker and Puglisi, 1974)
27
-------
TABLE 9 (Continued)
Animal Model:
Coho Salmon
Experimental Methodology:
Yearling coho salmon were fed PCB diets and mixed PCB-mirex diets to
determine the effects on hepatosomatic indices, liver lipids, carcass
lipids, and bioaccumulation
Experimental Results and Conclusions:
hepatosomatic indices were increased in salmon fed PCB diets and a mixed
PCB-mirex diet and were decreased in salmon fed Mirex diets compared to
controls
liver lipids in PCB-fed fish were significantly increased and carcass
lipids significantly decreased
carcass PCB levels in Coho fed high PCB diet (500 mg/kg) were 3.2 fold
higher than Coho fed the low PCB diet (50 mg/kg)
carcass PCB levels in Coho fed a mixed PCB-mirex diet were significantly
higher than fish fed low PCB diet (Leatherland ejt.al., 1979)
Animal Model:
Ciscoe
Experimental Methodology:
Continuous-flow bioassays were conducted to measure the acute toxicity
of arsenic and PCBs, singly and in combination for ciscoe fry (Coregonus)
Experimental Results and Conclusions:
96-hour LC-50 for ciscoe fry was>10 mg/1 for a mixture of 1248, 1254 and 1260
96-hour LC-50 for ciscoe fry was 3,5 mg/1 for the PCB-arsenic mixture
the difference in PCB concentration between male and female C.hoyi was highly
significant (joale ciscoe -* 2.3 mg/g)
(female ciscoe-»1.2 mg/g)
(Passlno and Kramer, 1980)
Animal Model:
Br.ook Trout
Experimental Methodology:
Continuous-flow bioassays were conducted to determine the effects of 1 mg/1
and lower concentrations of Aroclor 1254 on the life cycle of the brook trout
Experimental Results and Conclusions:
no adverse effects observed on survival and growth during 71 weeks of expo-
sure or on their progeny exposed to 90 days of Aroclor 1254
body levels reached an apparent steady state bioconcentration factor 1 X
(Snarski and Puglisi, 1976)
28
-------
concentration in fish tissues. The concentration of PCBs in the
tissues of experimental fish models (i.e., fathead minnow, brook trout)
is approximately 1.0 X 10 greater than that in the water (26). Jarvinen
and his associates evaluated the bioconcentration factor of other
chlorinated hydrocarbons (i.e., endrin, DDT) and reported similar results
to those reported for PCBs (58). This suggests the potential for bio-
accumulation in natural aquatic ecosystems given sources of PCS contam-
ination. It is interesting to note the conclusions from a study by
Weininger which indicate that direct uptake (.i.e., from water to fish
tissue) accounts for only 2 - 3% of the total PCB accumulation by adult
lake trout in Lake Michigan. He suggests that food, primarily the
alewife, is the major contributing source of PCB body burden in lake
trout (115). In addition, he concludes that a large proportion (>50%)
of the PCB burden in Lake Michigan lake trout has been cycled through
the sediments suggesting that potential'public health problems could
persist for years. The human health implications of the behavior of
PCBs in the aquatic environment, resulting from a demonstrated ability
to bioaccumulate in fish, have not been adequately evaluated.
II. Human Exposures to PCBs and Other Organic Compounds
A. General
The results of surveys and experiments which have examined
human exposures to selected organic chemicals are presented below.
Polychlorinated biphenyls (PCBs) serve as the principal model for
organic compounds throughout this review.
Information on the level of PCBs in the general population of the
29
-------
United States has been released by the Human Monitoring Survey (HMS).
Founded in 1967, the purpose of this organization has been to examine
the exposure of the general population to pesticides. Residues from
these compounds are measured in adipose tissue samples acquired from
post-mortem examinations and therapeutic surgeries. Results from the
HMS reported by Yobs in 1972 (117) state that 31.1% of 637 samples
acquired contained one or more parts per million (ppm) of PCBs. In a
comparison of gas chromatography with thin-layer chromatography by
the HMS (98 ), both methods provided similar results indicating that
41-45% of the U.S. population have PCS levels of 1.0 ppm or more with
isomers from Aroclors 1254, 1260, 1262, and 1268. A summary of "other"
organochlorine residue levels for the years 1970-1974 inclusive are
listed in Table 10 ( 74 ). The authors stated that residues of alpha-
benzene hexachloride, llndane, and mlrex were detected in adipose
tissues infrequently (<2.05%). Residues of delta-BHC, aldrin, hepta-
chlor, and endrin were not detected in the samples.
The following conclusions may be drawn from the HMS data:
a. Residues of various organochlorines are present
in quantifiable amounts in the general population
of the United States.
b. Between 30-50% of the general population have
adipose tissues with >1.0 ppm PCBs. The most
common isomers encountered are penta-, hexa, and
heptachlorobiphenyl compounds (73 ).
c. HMS data for 1973 and 1974 indicated an increased
percentage of tissues with trace levels of PCBs
while the limit of detection (1.0 ppm) remained
constant (73),
30
-------
TABLE 10
Levels of Organochlorine Residues in Human Adipose Tissues
Compound
Beta-Benzene
llexachlor Ide
Total DDT
Equivalent
Dieldcin
lleptachlor
Epoxide
Oxychlordane
Survey
Years
1970-
1974
1970-
1974
1970-
1974
1970-
1974
1970-
1974
Average
Sample
Size
1307
1387
1387
1387
1302
Average Percent of
Samples With
Detectable Residues
97.7
99.9
90.4
95.0
96.4
Average Geo-
metric Mean
in ppm.
0.28
6.72
0.18
0.086
0.12
Comments on Data
Set Averaged
A trend was demonstrated
toward a reduction in
concentration over time
but not in the frequency
of occurrence
Same as Beta-Benzene
llexachloride
The geometric means and
percent of samples
which were positive were,
relatively consistent
across the survey year
Same as Oieldrin
Same as Dieldrin
Source: Kutz et.al.(1977)
-------
The sources of exposure to PCBs for the general population include
foods, ambient air, ambient waters, contaminated soils, and occupational
exposures.
B. Dietary Exposures
Estimates of PCB consumption from market-basket samples
indicate an average adult ingestion of 5-10Ug per day ( 21). Variabil-
ity in PCB consumption may result from special diets or accidental con-
tamination of food products. Vegetarians may experience a lower exposure
to PCBs as they limit fresh fish consumption. Accidental contaminations
of human food stuffs are exemplified by the leakage of PCBs into
"Shredded Wheat." from packaging materials ( 66 ), the introduction of
contaminated poultry feed into Holly Farms livestock ( 66 ), and the
"Yusho"" (oil disease) incident ( 71).
The Yusho epidemic represents one of the largest case studies of
acute toxic effects of PCBs in humans. The following description is
a summary of the author's findings. In Fukuoka-ken, Japan a local
rice oil-producing company developed a leak in a heat-exchanger. The
oil was contaminated with 2,000-3,000 ppm Kanechlor 400 (48% chlorinated
biphenyl). The average amount of PCBs ingested by subsequent patients
has been estimated at 2 grams. The approximate minimum dose was 0.5
grams. The mean blood level among Yusho patients was 0.7 parts per
billion Cppb). The predominant initial symptoms experienced by 136 of
the Yusho patients were increased eye discharge, swelling of upper
eyelids, acniform eruptions (chloracne), follicular accentuation, and
increased skin pigmentation. Common, late-appearing symptoms included
those noted above as well as weakness, sweating of palms, and brown
32
-------
pigmentation of the nails. Children born to patients and wives of
patients had a high incidence of grayish, dark=-brown skin, gingiva, and
nails. Increased eye discharge, low birth weight, and infantile jaundice
•were also noted. Of the 13 pregnancies reported, two were stillborn.
At a three-year follow-up half of the patients had clinically improved,
over 10% had worsened, with the remaining patients demonstrating no
change. The conclusion which may be drawn from the follow-up period is
that clinical recovery is not common.
A summary of the clinical findings is presented in the reference by
Kuratsune (70). His group found that severe "Yusho" patients experienced
an increase in leucocytes, total lipids, triglycerides, serum lipoprotein
plc< ratio, Cu, serum globulin oC, and alkaline phosphatase. A decrease
was found in red blood cells, hemoglobin, and Fe. The cases with charac-
teristic "Yusho" acne had higher levels of cholesterol and stearic and oleic
acids than those with normal acne. The pathological responses observed
in the Yusho epidemic could be explained by the presence of chlorinated
dibenzofurans in the rice oil. Chlorinated dibenzofurans are a contam-
inant of PCBs and are known to be highly toxic. Blood PCS levels as
high as in Yusho patients but without adverse health effects
have been reported in occupationally exposed workers (33 ). Therefore,
conclusions cannot be drawn between the physiologic responses attributable
to either of the two compounds.
Current foods which may provide high PCB exposure to the general
population are fish and human breast milk (33 ). Results from the Food
and Drug Administration (FDA) Total Diet Study (1971-1975) indicate
that PCBs are no longer detected in food categories other.than in meat-
fish-poultry composites. The presence of PCBs in this category is
33
-------
attributed to fish. It is estimated that 93% of the U.S. population
consumes fish. The average fish eater consumes 15 pounds of fish per
year, which provides an approximate PCB consumption of 5-10 micrograms/
day. The freshwater fish species most commonly consumed are trout,
bass, catfish, and pike. A large fraction of fish consumed in the
U.S. diet are unclassified breaded products (e.g., fish sticks). PCB
measurements have typically not been initiated in this fraction (24 ).
Further difficulties in describing PCB exposure via fish consumption
are due to: (1) the inconsistent monitoring of PCB levels in fish; (2)
most surveys having been based on small sample sizes; (3) PCB analyses
usually being done on whole fish and not restricted to the edible por-
tions; (A) fish being acquired for consumption without passing through
commercial sectors where monitoring may be undertaken; (5) variations
in PCB concentrations in individual fish by species, age, and geograph-
ical area.
A .study by Humphrey ( 52 ) and others reported on the serum PCB
levels of Lake Michigan sport fish eaters. The authors found these
sport fishermen consume an average of 24-25 pounds of fish/person/year.
A highly significant correlation was found between the amount of Lake
Michigan fish consumed and blood PCB concentrations. Those fishermen
who consumed less than 6 pounds/year had a mean blood value of 0.02 ppm.
Those who ate more than one meal per week, or 24 pounds per year, had
an average blood value of 0.073 ppm. This latter group had an estimated
average intake of 1.7ydg/kg body weight/day of PCBs. The FDA recom-
mends that PCB intake not exceed 1 Mg/kg body weight/day. The authors
found that blood PCB levels did not decrease significantly when fish
consumption was eliminated for up to nine months, nor did the levels
34
-------
alter appreciably from year to year. A correlation between blood PCS
levels and a pathological condition was not found in the high exposure
group. The authors conclude that while Lake Michigan fish contribute
to blood PCB levels in humans, the dose received has not produc'ed a
detectable health effect.
C. Exposure via Human Breast Milk
The other food, noted above, which may provide high
exposure to PCBs is human breast milk. From 1974 to 1977 Kodama
and Ota ( 65 ) studied blood PCB levels in volunteer women and. their
newborns at a hospital in Alchi Perfecture, Japan. The following
samples were acquired by the authors: (1) maternal blood at 8 and 4
months prepartum; at delivery, and 1, 3, 5, and 7 months postpartum;
(2) mother's milk at 1, 3, 5, and 7 months postpartum; (3) cord blood
at delivery; (4) newborn infant blood at 3 months, 1 year, 2 years,
and 3 years after birth. The results of the study indicate that maternal
blood PCB levels increase with gestation and thereafter decrease to
general population levels by 5 months postpartum. Maternal blood at
delivery has a significantly higher PCB level than cord blood. When
considering cord blood as newborn blood, PCB blood levels in breast-fed
infants increased with ingestion of human milk. At 3 months postpartum
the blood levels in the infants exceeded that of their mothers. The
peak blood PCB level was reached at one year of age and decreased there-
after. Those infants who were bottle-fed had consistently low levels
over the same time period. The authors conclude that the quantity of
PCBs passed from mother to child is greater in lactation than through
the placenta.
In the "National Study of Chlorinated Hydrocarbon Insecticide
35
-------
Residues in Human Milk, U.S.A." (105 ) breast milk samples from 1436
women were obtained. Ihe percent of samples positive for dieldrin,
heptachlor, and oxychlordane by geographical region are presented in
Table 11 . Eighty percent of all samples collected had detectable
levels of dieldrin. Oxychlordane, a metabolite of- chlordane, was
detectable in 74% of the samples. Heptachlor was found in less than
2% of the samples, yet its metabolite heptachlor epoxide was found in
63% of the samples. The authors note that after adjusting for fat
content in the samples, the mean residue levels for the three compounds
in Table 11 were 164.2 ppb for dieldrin, 91.4 ppb for heptachlor epoxide,
and 95.8 ppb for oxychlordane. It was quite common for a single subject
to have relatively high values for more than one compound. The trend
for higher mean values in the southeastern U.S. was attributed to more
extensive use of home pesticides and termite control.
Studies which support the above findings of organic contamination
of human breast milk were conducted in Michigan, Arkansas, and Mississip-
pi. The Michigan study (116) was conducted in 1977 and 1978. Breast
milk samples from 1,057 nursing mothers were analyzed for PCS residues.
PCB levels ranged from trace amounts to 5.1 ppm, with all 1,057 samples
positive Cfat weight basis). The mean PCS level was 1.5 ppm. PCB
levels of 1-2 ppm were found in 49.5% of the samples, 2-3 ppm in 17.4%
of those sampled, and 6.14% of the women had greater than 3 ppm. Approx-
imately one-half of the women had PCB levels equal to or greater than
the present FDA tolerance limit for cow's milk. If an infant were breast-
fed by a woman having the mean PCB level (1.5 ppm) in this sample, the
child would have an estimated body burden of 0.89 ppm of PCBs. The
infant's PCB body burden would increase steadily with the consumption
36
-------
TABLE 11
Percent of U.S. Breast Milk Samples With Levels of Chlorinated Hydrocarbon Insecticides or
Their Metabolites at >51 ppb by Geographical Region (N = 1,436)
Geographical Region
Compound
Dieldrin
Oxychlordane
Heptachlor
Epoxide
Northeast
%
55
54
43
Southeast
%
84
72
56
Midwest
%
74
57
57
Southwest
%
76
54
25
Northwest
7,
69
52
24
Total U.S.
%
73
58
42
Source: Savage et. al. (1981)
-------
of breast milk for the entire period of breast-feeding. The authors
stress that the women in this study may not be representative of all
Michigan women as they were not randomly selected. They further state
that the PCB levels found in several counties in the western part of
the lower peninsula as well as in three counties of the upper penin-
sula exceeded 2 ppm. It is speculated that these levels may reflect a
higher dietary intake of PCB-contaminated fish from the Great Lakes.
Fifty-seven women residing in Arkansas and Mississippi contributed
whole milk samples for organochlorine pesticide residue analysis. The
results from this survey are presented in Table 12 ( 74 ).. The percent
of samples positive range from 14.1 (trans-nonachlor) to 100.0 (total
DDT equivalent; p, p'-DDT; p, p'-DDE). The arithmetic mean concentra-
tion in ppm ranged from 0.01 for many pesticides to 0.34 for total DDT
equivalent. The discrepancy in values of the percentages positive be-
tween Tables 11 and 12 for the respective compounds may be due to
differences in analytical techniques, sample size, properties of the
geographical areas surveyed, and a multitude of demographic variables.
Yet, both tables are consistent in the finding that significant propor-
tions of the women surveyed have contaminated breast milk.
A study of Dutch mothers and their infants ( 30 ) provides evidence
which challenges a number of the conclusions drawn above. Pregnant
women were divided into four groups: CD those not on a slimming diet
who breast fed; (2) those not on a slimming diet who bottle-fed; (3)
those on a slimming diet who breast fed; (4) those on a slimming diet
who bottle-fed. Maternal blood samples were acquired as early in
pregnancy as possible and, at parturition, from the umbilical cord and
placenta. Post-natal blood samples at 10 days, 6 weeks , and 3 months
38
-------
TABLE 12
Organochlorine Pesticide Residue Levels in Whole Milk Samples
from Women Residing in Arkansas and Mississippi (n=57)
Pesticide
Total DDT equivalent*
pf p1 - DDT
p, p' - DDE
ft - BHC
Dieldrin
Heptachlor expoxide
Oxychlordane
trans-nonachlor
Percent of Samples
Positive for Pesticide
100.0
100.0
100.0
36.8
28.1
25.1
45.6
14.1
Arithmetic Mean of
Concentration in ppm.
0.34
0.09
0.22
<0.01
<0,01
<0.01
<0.01
<0.01
Source: Kutz et.al. (1977)
-------
were acquired from the infant and mother. Milk samples were taken daily
fXQJn breast-feeders. Data are not presented on organochlorine levels in
human milk. The results of this study indicate that higher organochlo-
rine concentrations are not found in the blood of breast than of bottle-
fed infanta. Further, there are no differences between the organochlo-
rine concentrations in the blood of breast than of bottle feeding mothers.
Between 12 and 21% of the daily intake of dieldrin by mothers was
eliminated in their milk and ingested by the infants. The corresponding
range for total DDT was 36 to 61%. The ranges were based upon the
infants' consumption level during the first 3 months postpartum. It
is interesting to note that the blood concentrations in slimming and
non-slimming mothers were very similar except for those for dieldrin.
As would be expected from the lipid solubility of organic compounds,
dieldrin concentrations were higher in the slimming diet mothers.
Several investigators who have reviewed the health risks of organic
compounds in human breast milk have concluded that no obvious pathology
has resulted from breast feedings by the non-occupationally or acutely
exposed mother (13, 101, 62). In an article on PCBs in human breast
milk (82) the author states that unless a woman has a definite history
of PCS exposure, she should be encouraged to breast-feed her infant.
Kendrick (62) presents a more cautious appeal for the weighing of
potential risks against known benefits in breast feeding.
D. Community Surveys
Testimony to the ubiquitous nature of organic compounds has
been provided by their prevalence in the adipose tissue of the
general population and human breast milk. Community based studies
have been initiated to assess the general health effects of these pol-
40
-------
lutants. The residents of Triana, Alabama were studied in 1979 since
they were known to consume fish contaminated with DDT and PCBs (69 ).
Serum PCB levels were determined for 458 subjects. The geometric mean
serum level was 17.2y^g/L. A positive association was found between
PCB levels and age after controlling for sex, local fish consumption,
obesity, serum cholesterol level, and alcohol consumption. PCB levels
were also positively associated with gamma-glutamyl transpeptidase
level, serum cholesterol level, and blood pressure. This latter posi-
tive association was independent of age, sex, body mass index, and social
class. The rate for borderline and definite hypertension was 30% higher
for the study group than would be expected from national rates for a
demogra;3hically similar population. PCB exposure was associated with
changes in liver function tests. Breast-fed children did not have sig-
nificar.:ly higher PCB levels than bottle-fed children, although, a
positive trend was demonstrated between breast-feeding and DDT. The
authors conclude that the association found between PCB levels and
blood pressure, liver function, and cholesterol concentrations warrant
further investigation since the PCB levels in this population overlap
those found in other communities.
The effects posed by DDT exposure on the health of the Triana
residents are discussed in a second paper (68 )• The national geometric
mean total DDT level in serum was 1.5 ng/ml; the respective geometric
mean for the 499 Triana residents was 76.2 ng/ml. The source of human
exposure was contaminated local fish. From 1947 until 1971, a DDT
plant 10 km from Triana deposited several thousand.tons of DDT waste
into a local tributary.
An average of 86.7% of the total DDT detected was of the metabolite
41
-------
DDE. Acute health effects from DDT exposure were not found in the
predominantly black (86.9%) study population. Positive associations
with DDT levels were found with serura triglycerides, alcohol consumption,
cigarette smoking, and liver enzyme induction. The authors state that
their most striking finding is an increase in serum DDT levels with age
without reaching a steady-state level. Age was the single most power-
ful predictor of DDT level (log transformation). The rate of increase
in DDT levels in women did not change with hormonal status. The authors
further emphasize that the increase in average serum DDT levels in the
elderly suggest age-related changes in absorption, excretion, or storage
between serum and adipose tissues.
In 1976, the sewage sludge used by Bloomington, Indiana residents
for garden fertilizer was found to contain a mean PCB concentration of
479.1 ppm (34 ). The PCBs had been discharged into the city sewerage
system by a local electrical manufacturing firm. To evaluate possible
health effects from this exposure, three study groups were selected by
the Monroe County Board of Health and the CDC: (1) occupationally
exposed workers, (2) the workers' families, and (3) the residents who
had utilized the sewage sludge in their garden. The mean serum PCB
level was highest for the worders (71.7 ppb), second highest for the
workers' families Q3.6 ppb), third highest in community residents not
utilizing sludge in their gardens (23.8 ppb), and lowest to. those res-
idents using the contaminated sludge (17.6 ppb). The sample sizes for
each of these groups are 18, 19, 29, and 91 respectively. Correlations
were not found between serum PCB levels and the number of years of
contaminated sludge utilization, the total number of pounds used, or the
length of the interval since last use. The authors did find that plasma
42
-------
triglyceride concentrations and serum gamma-glutamyl transpeptidase
increased significantly with increases in serum PCB levels. These
associations were found to be independent of age.
One of the most recent community studies evaluating the effects
of numerous toxic wastes on human health was conducted in the Love
Canal area (57 ). When cancer rates in the Love Canal residents were
compared to data from the New York Cancer Registry, no evidence was
found for an increased risk in the disposal area. A higher rate of
respiratory cancer was noted, but appeared to be part of a high rate
for the entire city of Niagara Falls. The authors acknowledge that the
study design could not assess the influence of such confounding variables
as socioeconomic status, smoking, air pollution, and population migra-
tion.
The community studies reviewed above have provided evidence that
organic compounds have been found in residential areas. The source of
the contamination ha s often been identified as industrial wastes. While
PCB levels have been documented in human serum, the above studies have
not demonstrated significant pathological responses. A paucity of
studies have been performed on individuals occupationally exposed to
PCBs. It is hypothesized that these groups experience a greater
exposure potential than the general population, and would therefore
more readily manifest any possible toxic effects. One of the earlier
studies examined the level of PCB residues in the plasma and hair of
refuse workers (42 )• Detectable plasma PCB levels were found in 81%
(32/37) of the refuse burners and only 11% (6/54) of the control popu.-
lation. Scalp hair was found to have no value in estimating PCB
body burdens. The increased plasma PCB levels in the refuse workers
43
-------
were hypothesized to be due to refuse incineration.
A subpopulation of Michigan residents were exposed to polybrominated
biphenyls (PBBs) in 1973 by an accidental contamination of animal feed
supplement ( 19 ). Several of the residents subsequently complained of
memory loss. To test the hypothesis that PBB exposure may affect memory,
twenty-five chemical workers who manufactured PBBs were given learning
and memory tests. PBBs were found in adipose tissue samples of the
workers, yet the mean scores on all memory tests were normal. The
memory dysfunction of the Michigan residents was attributed to a psycho-
logical dysfunction and not to PBB body burden.
E. Occupational Studies
Workers who have been exposed to PCBs while manufacturing
capacitors have shown the following health effects: (1) 14%
of 243 employees demonstrated a reduced forced vital capacity. The mean
number of years employed in this group was 15 (114 ), (2) for those who
had been employed for more than 10 years, the most common symptoms re-
ported involved dermatologic and CNS dysfunctions (34 ), (3) the latter
group of workers had a functional capacity of the cytochrome P-450 system
different from the non-exposed workers. Overall, the investigators were
impressed by the paucity of abnormalities found in the physical examina-
tions ( 34 ) •
In an occupational retrospective cohort study of PCB-exposed workers,
both the "all cause" and cancer mortality rates were lower than expected
(18 ). The corresponding values were 163 observed total deaths versus
182.4 expected total deaths and 39 observed cancer deaths versus 43.8
expected cancer deaths. Elevated rectal and liver cancer mortalities
were noted, but were not statistically significant. An increase in
44
-------
cirrhosis of the liver was found in one of the plants examined. Mor-
tality due to rectal cancer and cirrhosis of the liver increased with
an increase in latency, but these trends were difficult to evaluate due
to the small number of reported deaths. The increase in liver cancer
is noteworthy since it agrees with the findings of animal dietary models.
The two studies referenced below have examined the risk of hydro-
carbon-exposed workers fathering children who subsequently develop
malignancies. Positive results were found in a Quebec study of 386
children who died of malignancies prior to 5 years of age. A significant
excess (relative odds of 2:1) of fathers were found to work in hydrocarbon-
related occupations (32 ). Negative results were found in a review of
the Finnish Cancer Registry ( 40 ) •
The dennatologic lesions found in the Yusho incident have been
reported in occupational groups . The dermal sores have been known
to last for months after removal of the workers from the occupational
exposure. Systemic effects including nausea, lassitude, anorexia,
digestive disturbances, impotence, and hematuria have also been
reported (21).
The occupational studies listed above have found workers to be
exposed to various organic compounds and at risk of experiencing a wide
range of symptoms. A number of methodologic problems may be found in
these studies. A major problem in many occupational studies is adequate
documentation of the amount of exposure to a given worker via inhalation,
ingestion, and absorption. Secondly, many of the industries provide
exposure to other compounds which may act in a synergistic, antagonistic,
or additive manner to the chemical under study. The healthy-worker-
45
-------
effect and small sample size may have minimized adverse trends. In
conclusion, the occupational studies performed to date do not allow a
definitive statement as to the health risks of long-term low-level
exposure to organic compounds.
F. Teratogenic Research
The unborn child is at risk of experiencing the toxic effects
of organic compounds. Many organic chemicals have documented mutagenic
and teratogenic effects. The placental transfer of compounds decreases
with increased molecular weight, increasing electrical charge, and
decreasing lipid solubility. The susceptibility of the fetus
to teratogenic effects depends upon: (1) the properties of the agent
administered; (2) the time of administration of the agent, with the
most critical period being that of organ differentiation; (3) the size
of the dose received; (4) the number of doses given; (5) the route of
administration to the mother; and (6) other factors, such as the general
health of the mother ( 72 ). The lack of certain liver microsomal
enzyme systems may predispose the infant to the toxic effects of sub-
stances which are not oxidized and excreted as rapidly as normal. For
example, a fetus is more susceptible to the effects of alcohols and
phenols if the glucuronidation system is impaired ( 21 ). Conditions
associated with an incomplete or improper glucuronide conjugation system
are Gilbert's syndrome and the Crigler and Najjar syndrome.
A two-year study of mothers with children having central-nervous-
system defects was conducted examining the percent of mothers known to
have had exposure to organic solvents ( 47 ). The results of the study
indicated that significantly more "case-mothers" than "control-mothers"
46
-------
had been exposed to organic solvents during the first trimester of
pregnancy.
Reviews on the risks of pregnancy posed by environmental contam-
inants have been provided by a number of sources ( 79,72,33,35,21).
The discussions by these authors on the teratogenic properties of
biphehyls center on the findings of the animal studies and the "Yusho"
incident presented elsewhere in this review.
III. Summary
Environmental studies have shown PCBs to be extremely stable
compounds in the environment. Their lipophilic properties allow bio-
accumulation in aquatic foodchains with man, or more accurately the
nursing infant, at the top of the trophic scale. Testimony to the
ubiquitous nature of PCBs is found in the results of the Human Monitoring
Survey. A significant proportion of the U.S. general population sampled
have detectable amounts of PCBs in their adipose tissues. Exposure
routes for the general population include ambient air, ambient waters,
contaminated soils, occupational sources, and diet. The latter two
categories may provide the most concentrated exposures. Workers in
PCB-manufacturing industries have experienced dermatological, central-
nervous-system, and diffuse symptomatologies. The study designs which
have provided these results have not evaluated the long-term, low-level
toxic effects of PCBs.
Dietary studies conclude that the two food items which may provide
high concentrations of PCBs are fish and human breast milk. The physio-
logical consequences of ingesting high quantities of PCBs are best
47
-------
exemplified by the "Yusho" incident. Their patients demonstrated acni-
form lesions, skin pigmentation, and increased eye discharge. Those
women exposed during pregnancy had "small-for-date" children with "cola"-
colored skin and gingiva. Their infants were noted to have a high prev-
alence of neonatal jaundice and increased ocular discharge ( 71 ). It
must be noted that the rice oil was also contaminated with polychlorinated
dibenzofurans, which are known to have toxic properties.
The primary source of PCBs in the U.S. diet is fish acquired from
contaminated areas. The study by Humphrey and others ( 52 ) has docu-
mented that Michigan sport fish consumers eat approximately 10 pounds
more per person per year than the average U.S. citizen. The findings also
demonstrated a positive correlation between serum PCB levels and the
amount of fresh fish consumed. It may be hypothesized that the PCB
concentrations in human breast milk increase with the total body burden
of the nursing mother. Therefore, a mother who consumes large quantities
of contaminated fish may present a greater exposure potential to the
breast-fed infant.
The articles reviewed indicate that consumers of fish from contam-
inated waters and infants breast-fed by those consumers warrant further
study. The physiological consequences of long-term, low-level exposure
to PCBs in these populations are currently unknown. Therefore the efforts
of the Great Lakes project have focused on data gathering and quantitative
analysis to determine whehter contact with contaminants in the Great
Lakes has been detrimental to local populations. Initial data evaluations
were conducted on State Vital Statistics records. As these documents
do not indicate relative exposure potentials, a pilot study was conducted
48
-------
to identify fish consumers. Summaries of the data evaluations and
pilot project are presented in the following sections.
49
-------
Analyses of Morbidity and Mortality Patterns of
Populations Living in the Eight Great Lakes States
Five quantitative analyses have been performed by the Great Lakes
Project Staff. The purpose of these data evaluations has been to compare,
by geographical area,specific disease and death rates in the Great Lakes
Basin. The morbidity and mortality parameters examined were those likely
to reflect human exposure to organic compounds. Reviews were based on
cancer death rates, congenital anomaly rates, and fetal, neonatal, and
infant death rates. Vital Statistics data were not available on these
indices for all Great Lakes states. Pennsylvania and New York have col-
lated their records in a manner which is not readily accessible.
The geographical unit which served as a basis for comparison was
the county. The counties within each state were separated into three
"Orders." Order 1 counties were those which contain shoreline on one or
more Great Lakes. Order 2 consists of counties which are adjacent to
Order 1 counties. All other counties in the eight Great Lakes states
were considered as Order 3. Comparisons of mortality and morbidity rates
were made between Order 1 counties and the respective state mean rates,
and between Order 1 counties and those of Order 2 and 3. The contrasting
of rates by geographical proximity to the Great Lakes was based on the
hypothesis of a "graded exposure potential." That is, populations in
proximity to contaminated areas of the Great Lakes are more likely to be
exposed to contaminants than populations residing at greater distances.
In summary, the results of these studies demonstrate a trend toward
increased esophageal, stomach and "other gastro-intestinal" cancer
mortality in Order 1 counties when compared to respective state mean
rates. It may be hypothesized that the detrimental health effects
50
-------
of exposure to organic contaminants in the Great Lakes may be masked by
pollutants associated with urbanization in cities adjacent to the lakes.
When counties with large urban centers were removed from the above
analysis, a greater percentage of Order 1 counties, continued to experience
an excess of esophageal, stomach,and bronchus, trachea and lung combined, as
well as all-cancer mortality over the respective state mean rates.*
The comparisons between Order L and Order 2 mean county cancer mor-
tality rates demonstrated excesses in white males and females for the site of
esophagus (31% and 26%, respectively). White males also demonstrated excesses
for the sites of breast (46%) and thyroid gland (42%). A 68% excess in mor-
tality due to cancer of the endocrine organs was found in white females. The
county comparisons between Orders 1 and 3 found an excess of esophagal cancer
for Order 1 white males (44%) and females (32%). The respective values for
stomach cancer mortality excesses were 35% and 24%.
In addition, there were no striking discrepancies between the fetal
death rate, the neonatal death rate, and the percent of live births with
congenital anomalies, among "lake-bordering" and "non-lake bordering" counties.
For each of the five analyses an introduction, description of methods,
summary of results, and discussion section is presented below.
I. Analysis 1.
A. Introduction
The first step in attempting to ascertain an adverse human health
effect from contamination of the Great Lakes required the utilization of
* Results are for white males only.
51
-------
existing morbidity/mortality data sets acquired from state and federal
agencies. County vital statistics regarding fetal, neonatal, and infant
death rates and congenital anomaly rates were obtained for the states of
Minnesota, Wisconsin, Illinois, Indiana, Michigan, and Ohio for every
fifth year from 1950 to 1975 and for the year 1977. County data for
aite, race, sex, and age-adjusted cancer mortality rates were acquired
from the National Cancer Institute's publication "U.S. Cancer Mortality:
1950-1969." ( 80)
The hypothesis for this study was that populations residing in areas
immediately adjacent to the Great Lakes.experienced a greater exposure
potential to lake contaminants than populations residing in areas more
distant from the lakes. This hypothesis was based on the following assump-
tions:
1. Most"lake bordering"communities have increased industrial
activity compared to most non-lake bordering communities.
2. Individuals living in these "lake bordering" communities were
more likely to be occupationally exposed to those pollutants
being discharged into the air and lakes.
3. Close proximity to the Lakes would lead to greater exposure
via ambient air, soils, and water.
4. Fishermen living in these lake-adjacent communities were more
likely to fish these lakes and may have had higher exposures
due to consumption of their catch.
The analytic evaluations performed on the above mentioned data sets have
incorporated the concept of a graded exposure potential.
B. Methods
The initial evaluation of the above data sets was performed in
52
-------
the following manner:
1. Counties within .each of the Great Lakes states were designated
as either bordering the Great Lakes (hereafter termed "lake
bordering") or not bordering the Great Lakes (hereafter termed
"non-lake bordering").
2. Within each state, a mean rate for "lake bordering" counties
was calculated for each of the mortality and morbidity parameters
listed above. For example, the mean infant mortality rate for
"lake bordering" counties in Minnesota was calculated by summing
the infant mortality rates for each of the Minnesota "lake bor-
dering" counties and dividing this- sum by the number of "lake
bordering" counties in Minnesota.
3. Correspondingly, a mean rate for "non-lake bordering" counties
was claculated for each mortality/morbidity index within each
state.
4. For each of the health parameters of interest, a mean state rate
was calculated by summing the respective rates for all counties
within the state and dividing this sum by the -total number of
state counties. The standard error of each state mean was calcu-
lated.
A description of the data sources, methods of analysis, and results of
this phase of the Great Lakes Project may be found in the October 16, 1978-
June 15, 1979 Progress Report to the Environmental Protection Agency. A
copy of this report, is included as Appendix VI.
C. Results
An excess of esophageal, stomach, and "other gastro-intestinal"
cancer mortality was noted in "lake bordering" counties. However, tests
53
-------
of statistical significance were not calculated for observed differences.
The comparisons of infant, neonatal, and fetal death rates for the six
states noted above were inconclusive as they were not adjusted for maternal
age nor stratified by sex or race.
D. Discussion
Many of the "lake bordering" counties have large urban centers.
The excess esophageal, stomach, and "other gastro-intestinal" cancer mor-
tality in these areas may be due to factors intrinsic to urbanization
rather than potential exposure to toxic substances in the lakes. The
possible confounding influence of urbanization is addressed in a later
analysis.
II. Analysis 2.
A.. Introduction
The amount of organic compounds found in industrial and waste-
water treatment plant effluents are monitored by state governments. A
summary of these records are found in the "Inventory of Major Municipal
and Industrial Point Source Dischargers in the Great Lakes Basin." (56)
Knowing the levels of contaminants being released into set geographical
areas provides an approximation to the exposure potential of the respective
communities. It was hypothesized that those counties with point-source
dischargers of hazardous materials in compliance with state effluent
requirements may have different cancer rates than those counties whose
dischargers fail to comply with state requirements. The following
analysis was initiated to determine if correlations exist between the
recorded levels of organic compounds in industrial and wastewater treat-
ment plant effluents and the cancer mortality rates for the respective
county.
54
-------
B. Methods
For each of the eight Great Lakes states the five respective
counties with the highest cancer mortality rates in 1950-1969 for either
esophageal, stomach, lung, and all neoplasms were designated as list num-
ber I. The communities listed in the inventory above were allocated to
the following four categories in list number II:
1. communities with at least one industry measuring organic com-
pounds (i.e., phenols) prior to discharge and complying with
state effluent requriements, and in which the wastewater
treatment plants complied with state effluent requirements on
the basis of phosphorus (P), biological oxygen demand (BOD),
and suspended solids (SS).
2. communities with at least one industry measuring organic com-
pounds (i.e., phenols) prior to discharge and failing to comply
with state effluent requirements, and in which a wastewater
treatment plant failed to comply with state effluent require-
ments (on the basis of P, BOD, and SS).
3._ communities with at least one industry measuring organic com-
pounds (i.e., phenols) prior to discharge and failing to comply
with state effluent requirements, and in which the wastewater
treatment plants complied with state effluent requirements (on
the basis of P, BOD, and SS) .
4. communities with at least one industry measuring organic com-
pounds (i.e., phenols) prior to discharge and complied with
state effluent requirements, and in which a wastewater treatment
plant failed to comply with state effluent requirements (on the
basis of P, BOD, and SS).
55
-------
5. The counties of high cancer mortality in list I were matched
geographically to the four categories of dischargers in list II.
For example, county X has one of the five highest rates of esoph-
ageal cancer in Ohio. List II was then examined to see if county
X had a discharger. If yes, then the name of the county and the
category in which the discharger appeared was noted in a third
list.
6. Counts of the number of counties represented in each of the four
categories of dischargers were performed.
C. Results
The results indicated that counties with non-compliant communi-
ties experienced higher stomach and kidney cancer mortality rates in white
males and high "nose, auxiliary sinus, etc." cancer mortality rates in non-
white males. Counties with compliant communities had elevated "nose,
auxiliary sinus, etc." cancer mortality rates in white males and females.
A full account of the analytical methods and results are presented in the
"Preliminary Report of Epidemiological and Environmental Data" dated April
4, 1980 by the University of Minnesota Division of Epidemiology. A copy
is included in Appendix VI.
III. Analysis 3.
A. Introduction
As the first analysis did not address the possible confounding
influence of urban pollutants in the examination of mean fetal, neonatal,
infant, and all sites cancer death rates for "lake bordering" versus "non-
lake bordering" counties, an examination of the same data sets was initiated
omitting all counties within the Great Lakes states having population cen-
ters greater than 100,000 inhabitants.
56
-------
B. Methods
A straightforward approach to utilizing the prior results was to
identify the percent of rural "lake bordering" counties versus rural "non-
lake bordering" counties, for the Great Lakes Basin, which demonstrated
either one standard deviation excess or one standard deviation- deficit
from respective state means for 1977 white male esophageal, stomach, lung,
and all cancer mortality rates. An identical evaluation was performed
for the percent live births with congenital anomalies, fertility rates,
neonatal death rates, and fetal death rates when data were available.
A total of 37 counties having population centers of greater than
100,000 inhabitants were omitted from the analysis leaving 72 "rural" lake
bordering and 538 "rural" non-lake bordering counties in the states of
Illinois, Indiana, Michigan, Minnesota, New York, Ohio, Pennsylvania, and
Wisconsin.
C. Results
Table 13shows the percentages of rural "lake bordering" counties
having at least one standard deviation higher mortality/morbidity and those
with at least one standard deviation lower mortality/morbidity than their
respective state mean rates. A larger percent of "lake bordering" counties
had rates in excess of one or more standard deviations (as compared to the
respective state means) than did "non-lake bordering" counties for the 1977
white male cancer sites of esophagus^ stomach, bronchus, trachea and lung, and
all-sites combined. The corresponding values are 16.67%, 13.9%, 8.3%, and 9.72%
versus3.s%, 3.9%,.3.53%, and 3.16% respectively. The percent of counties
with rates for "Live Births With Congenital Anomalies", demonstrated an
opposite trend with a greater percentage of "non-lake bordering" counties
having rates equal to or greater than one standard deviation than "lake
57
-------
TABLE 13
Percent of Lake Bordering and Non-Lake Bordering Counties Having Morbidity/Mortality
Rates Discrepant by £l Standard Deviation From State Means
Percent of 72 "lake
bordering" -counties
having rates 2 1
standard deviation
higher than their
respective state
mean rates
Percent of 72 "lake
bordering counties
having races a 1
standard deviation
Lower than their
respective state
mean rates
Percent of 538 "non-
lake bordering"
counties having rates
^ 1 standard devia-
tion higher than
their respective
state mean rates
Percent of 538 "non-
lake bordering".
counties having rates
> 1 standard devia-
tion lower than their
respective state
munii mcua
Cancer Sites
Bronchus,
Trachea, All
Esophagus Stomach & Lung Cancers
16.67
4.17
3.5
7.8
13.9
4.17
3.9
6.7
8.3
6:9
3.53
6.51
9.72
4.17
3.16
7.06
Reproductive Parameters
Z of Live Births Fetal
Fertility with Congenital Neonatal Daath
Rate Anomalies Daarh Rate Rate
%
*
*
*
3.64
7.27
7.43
6.86
4.69
9.38
6.25
8.56
7.81
7.81
6.0
10.4
* rtata unavailable
-------
bordering" counties. The percent of counties with rates above one standard
deviation and below one standard deviation for "lake bordering" as compared
with "non-lake bordering" are: 3.64% above and 7.27% below, and 7.43% above
and 6.86% below one standard deviation, respectively. The percentage values
for the "Neonatal Death Rate" are 4.69 and 9.38, and 6.25 and 8.56, respec-
tively. The pattern of percentages for the "Fetal Death Rate" is the same
as the pattern for the cancer sites. The corresponding values are: 7.81
and 7.81, and 6.0 and 10.4. The fertility rates for the eight states
were not analyzed due to inconsistent reporting between states.
D. Discussion
The results of this analysis indicate that when comparing
counties without population centers of 100,000 inhabitants to the state
mortality/morbidity rates, the "lake bordering" counties have a greater
probability of experiencing extreme rates in those outcomes of morbidity/
mortality reviewed than do "non-lake bordering" counties. For example, a
rural "lake bordering" county is 5.46 times as likely as a rural "non-
lake bordering" county to have a 1977 white male esophageal cancer rate one
standard deviation above its respective state mean. Hypotheses drawn from
this third anlaysis should be guarded. The cancer rates are for white
males in 1977 and may not be indicative of cancer mortality patterns of
other years, races, or women. The available state vital statistics data re-
garding reproductive parameters were limited to a few states, therefore, results
are based on a small number of reporting counties.
IV. Analysis 4.
A. Introduction
Since the results cited above warranted further investigation,
the analytical strategies were refined so that comparisons could be made
59
-------
between the rural "lake bordering" versus the rural "non-lake bordering"
county age-adjusted cancer mortality rates for specific anatomical sites.
B. Methods
For each of the eight states in the Great Lakes Basin an average
age-adjusted site-sex-race specific cancer mortality rate per 100,000 was
computed for the rural "lake bordering" counties and the rural "non-lake
bordering" counties. The mean of the eight averaged age-adjusted site
specific cancer mortality rates per 100,000 people was calculated to pro-
duce a summary rate for "lake bordering" counties of the Great Lakes Basin.
This procedure was followed to provide similar summary values for "non-
lake bordering" counties.
C. Results
Table 14 presents the summary mean rates, their standard deviations,
and the differences and ratios between means for the rural "lake bordering"
and the rural "non-lake bordering" counties.
The ratio of "lake bordering" to "non-lake bordering" mean age-adjusted
cancer rates (hereafter termed ratio) of the esophagus for white males is
1.468. The corresponding ratio for white females is 1.0374 and .5134 for
both non-white sexes. The ratios for stomach cancer rates are 1.2513 for
white males, 1.1116 for white females, and 1.0759 for non-whites of both
sexes. The corresponding ratios for bronchus, trachea, and lung cancers
are 1.1521 for white males, .9979 for white females, and 1.0691 for non-
whites of both sexes. The corresponding ratios for the category of all
malignant neoplasms are 1.09777, 1.0339, and 1.0734 respectively.
The ratio for whites demonstrating the largest relative disparity
(47%) between "lake bordering" and "non-lake bordering" counties is for
esophageal cancer in white males (1.468). The second largest relative
60
-------
TABLE M
Means, Standard DoviationB, and Comparisons, of Mean Age-Adjusted Causo-Sox-Raca-Speciftc
Mortality Ratos Per 100,000 for Rural "Lake Bordering" and Rural
"Non-Lake Bordering" Counties of the Great Lakes Basin.
^.in rural "L.ike Bordering" mortality
rate fnr alt 8 statfts. (fl)
SlJndarJ Deviation
Me .in rur.it ":lr-n take Bordering" Bortality
: r.itc (or all 8 stales. CO
II
•', StanjArd 'deviation of summary aeon
difference of: (I - N)
1 ,r
Ratio of: "/-j
Eaupliageol Cancer
Hliltc
H
4.44
.1161
2.99
.4649
1.6523
1.468
F
.796
.3003
.768
.1424
.0287
Both
Sexes
J.24
.8336
3.756
.5652
1.481
1.0374 j 1.39«1
Don
White
Both
Sexes
6.08
3.8835
11.836
19.725
-5.759
.5134
Stoasch Cancer
White
M
18.55
3.2833
14.824
2.3426
3.725
1.2513
F
8.41
1.6585
7.561
1.3468
.8436
1.1116
Both
Sexes
26.97
4.548
22.385
3.6332
4.584
1.2048
Han
White
Both
SOXCB
2.4.29
12.958
22.579
9.8951
1.714
1.0759
Bronchus, Trachea t Lung Cancer
White
H
33.28
5.487
28.888
6.008
4.391
1.1521
T
5.08
.9448
5.094
.6288
-.0108
.9979
Roth
Sexes
38.36
6.2895
33.982
6.5547
Non
Ul.lt e
Bath
Sexes
12.59
25.8836
10.482
12.6605
4.383 1 2.1851
1.129 1.0691
All Neoplssas
White
H
170.15
7.8171
IJ5.184
9.0131
15.167
1.0977
; Both
' 1 S
-------
disparity (252) is for stomach cancer in while males. The remaining
ratios have values approaching unity. The cancer ratios for non-whites
should be viewed sceptically as they are based upon very small numbers of deaths.
D. Discussion
These results indicate that white populations of "lake bordering"
counties experience larger averaged cancer rates for every category examined,
(except for bronchus', trachea and lung cancers in females) than did white
populations of "non-lake bordering" counties. The methodological approach
of comparing average cancer rates, as opposed to the percentage of counties
with rates greater than one standard deviation above respective state means;
ia a more direct approach in examining the amount of difference in the
mortality experiences of the two geographical areas.
The results of the third and fourth analyses suggest that when counties
with large urban centers are removed from consideration,specific sites show higher
cancer mortality in those counties which border the Great Lakes than in
those counties which do not border the Great Lakes. The findings of the
four analyses were consistent in revealing elevated esophageal and stom-
ach cancer mortality rates in "lake bordering" counties.
V. Analysis 5.
A. Introduction
A fifth evaluation of the 1550-1969 county cancer mortality rates
was undertaken to determine if the above disparities would be demonstrated
under a decidedly more complex and costly analytical method. Animal
toxicity studies have demonstrated that many types of cancers may result
from exposure to organic chemicals. Therefore, it was decided to examine
rigorously all thirty-five cancer sites reported in the NIH county 1950-
1969 cancer mortality data (.80) . The analytical procedure encompassed the
62
-------
following methods:
B. Methods
1. All age-adjusted site-race-sex-specific cancer rates for each
county in the eight Great Lakes States were extracted from the
NIH 1950-1969 county cancer mortality data set ©0).
2. The counties of the eight Great Lakes states were divided into
three "Orders." First Order counties are those counties which
are adjacent to the Great Lakes ("lake bordering counties").
Second Order counties are those counties adjacent to these "lake
bordering"counties. Third Order counties are considered to be
all remaining counties of the eight Great Lakes states. The
listing of state counties by Order is presented in Table 15.
3. The 1960 population census figures were obtained for each of the
three Orders. Population counts within each Order were totaled
for white males, white females, non-white males, and non-white
females. (See Tables 16and 17). The population counts for 1960
were used in a later step as weighting factors for the refinement
of summary cancer rates. The 1960 census figures were utilized
as this year represents the midpoint of the twenty year data
collection period for the NIH 1950-1969 county cancer mortality
data set ( 80 ).
4. The age-adjusted site-race-sex specific county cancer mortality
rates were summed within each Order and then divided by the total
number of counties within the respective Order to yield average
age-adjusted site-race-sex specific mortality, rates for each of
the three Orders.
5. "Risk" ratios were then calculated comparing Orders one to two,
63
-------
TABLE 15
Counties Designated as the 1st, 2nd, and 3rd Orders by State
1st Order = Lake Bordering Counties
2nd Order * Adjacent to Lake Bordering Counties
3rd Order = Inland Counties (remaining counties)
Pennsylvania
1st
Erie
2nd
Crawford
Warren
1st Order =• 1
2nd Order » 2
3rd Order = 64
Total number of counties » 67
Counties omitted for "rural" analysis
1st 2nd
•Erie —
1st Order = 0
2nd Order = 2
3rd Order = 60
3rd
(All remaining)
Illinois
1st
Lake
Cook
3rd
Allegheny
Philadelphia
Lehigh
Lackawanna
Total number of counties
62
1st Order
2nd Order
3rd Order
2nd
Will
Dupage
Kane
McHenry
2
4
96
3rd
(All remaining)
Total number of counties * 102
64
-------
TABLE 15 (Continued)
Counties omitted for "rural" analysis
1st 2nd 3rd
Cook — Winnebago
Peoria
1st Order » 1
2nd Order « 4
3rd Order - 94
Total number of counties - 99
Indiana
1st 2nd 3rd
Lake St. Joseph (All remaining)
Porter Starke
La Porte Jasper
Newton
1st Order - 3
2nd Order - 4
3rd Order » 85
Total number of counties « 92
Counties omitted for "rural" analysis
1st 2nd 3rd
Lake St. Joseph Allen
Marion
Vanderburgh
1st Order « 2
2nd Order « 3
3rd Order - 82
Total number of counties = 87
65
-------
TABLE 15 (Continued)
Michigan
1st 2nd 3rd
Berrien Cass (All remaining)
Van Buren Kalamazoo
Allegan Barry
Ottawa Kent
Muskegon Newaygo
Oceana .Lake
Mason Wexford
Manistee Kalkaska
Benzie Otsego
Leelanau Montmorency
Grand Traverse Oscoda
Antrim Ogemaw
Charlevoix Gladwin
Emmet Midland
Cheboygan Saginaw
Presque Isle Genesee
Alpena Lapeer
Alcoma Oakland
IOSCQ Washtenaw
Arenac Lenawee
Tuacola Iron
Huron Dickinson
Bay
Sanilac
St. Glair
Macomb
Wayne
Monroe
Gogebic
Ontonagon
Hbughton
Keweenaw
Baraga
Marquette
Alger
Luce
Chippewa
Mackinac
Schoolcraft
Delta
Menominee
1st Order = 41
2nd Order = 22
3rd Order - 20
Total number of counties - 83
66
-------
TABLE 15 (Continued)
Counties omitted for "rural" analysis
1st 2nd 3rd
Macomb Genesee Ingham
Wayne Kent
1st Order • 39
2nd Order » 20
3rd Order - 19
Total number of counties « 78
Minnesota
1st 2nd 3rd
Cook Koochiching (All remaining)
Lake Itasca
St. Louis Aitkin
Carlton
1st Order - 3
2nd. Order - 4
3rd Order - 80
Total number of counties « 87
Counties omitted for "rural" analysis
1st 2nd 3rd
St. Louis — Hennepin
Ramsey
1st Order - 2
2nd Order « 4
3rd Order •* 78
Total number of counties » 84
67
-------
TABLE 15 (Continued)
New York
13C 2nd 3rd
St. Laurence Franklin (All remaining)
Jefferson Herkimer
Oswego Lewis
Cayuga Oneida
Wayne Madison
Monroe Onondaga
Niagara Cortland
Erie Seneca
Chautaugua Ontario
Orleans Livingston
Genesee
Wyoming
Cattaraugus
Tompkins
Hamilton
1st Order » 10
2nd Order - 15
3rd Order - 33
*Total number of counties =• 58
Counties omitted for "rural" analysis
1st 2nd 3rd
Erie Onandaga Albany
Monroe New York
Westchester
1st Order - 8
2nd Order » 14
3rd Order - 30
Total number of counties » 52
* The "U.S. Cancer Mortality by County 1950-1969" includes Bronx, Kings,
Queens, and Richmond counties as New York. All these counties will be
omitted in the "rural" analysis.
68
-------
TABLE 15 (Continued)
Ohio
1st 2nd 3rd
Lucas Fulton (All remaining)
Ottawa Henry
Erie Wood
Lorain Seneca
Cuyahoga Huron
Lake Ashland
Ashtabula Medina
Sandusky Summit
Trumbull
Geauga
1st Order » 8
2nd Order - 10
3rd Order - 70
Total number of counties » 88
Counties omitted for "rural" analysis
1st 2nd 3rd
Lucas Summit Hamilton
Cuyahoga Mahoning
Montgomery
Stark
Franklin
1st Order » 6
2nd Order « 9
3rd Order - 65
Total number of counties * 80
69
-------
TABLE 15 (Continued)
Wisconsin
1st 2nd 3rd
Douglas Burnett (All remaining)
Bayfield Washburn
Ashland Sawyer
Iron Price
Marinette Vilas
*Menominee Florence
Door Forest
Brown Langlade
Rewaunee Calumet
Manitowoc Fond du Lac
Sheboygan Outagamie
Ozaukee Washington
Milwaukee Wau kes ha
Racine Walworth
Renosha
1st Order =• 15
2nd Order = 14
3rd Order =• 41
Total number of counties = 70
Counties omitted for "rural" analysis
1st 2nd 3rd
Milwaukee — Dane
1st Order = 14
2nd Order - 14
3rd Order = 40
Total number of counties - 68
*Note: Menominee County did not exist in 1960. In: "U.S. Cancer Mor-
tality by County 1950-1969," Menorainee includes Oconto and
Shawano counties. For purposes of this analysis Menominee
will be defined as lake Bordering with a population comprised
of Oconto and Shawano counties.
70
-------
TABLE 16
SUMMARY POPULATION FIGURES FOR THE- GREAT LAKES BASIN BY ORDER, SEX, AND RACE (1960)
State
1st Order
White
Male Female
2nd Order
White
Male Female
3rd Order
White
Male Female
PA
MN
WI
MI
IL
IN
NY
OH
119,286
123,469
858,142
1,947,589
2,224,186
291,740
1,161,664
1,174,091
Total:
Male + Female -
PA
MN
WI
MI
IL
IN
NY
OH
Total
:
1
124,271
122,868
880,945
1,970,115
2,297,648
284,230
1,217,220
1,225,393
7,900,167 8,122,690
16,022,857
Nonwhite
Male Female
3
1
39
285
432
45
59
156
,024
'Male -t- Female «
Totals:
PA
MN
WI
MI 2,
IL 2,
IN
NY 1,
OH JLj.
Total: 8,
Male +
Female **
Male and
1st
Male
122,751
124,653
897,671
232,741
656,995
337,117
221,628
330,983
924,539
,465
,184
,529
,152
,809
,377
,964
,892
,372 1
3,660
1,146
40,254
300,536
468,738
47,312
61,155
167,526
,090,327
59,727
48,766
267,546
1,092,906
386,717
136,211
628,895
518,444
62,639
45,797
267,636
1,115,276
392,361
136,232
642,894
532,039
3,139,212 3,194,874
6,334,086
Nonwhite
Male Female
2,
60,
10,
7,
15,
28,
124,
2,114,699
Female by
Order
Female
2,
2,
1,
1',
9,
18,137,
127,
124,
921,
270,
766,
331,
278,
392,
213,
556
931
014
199
651
386
542
375
919
017
614
912
122
232
768
007
239
092
986
248.,
558
815
2,166
61,396
7,686
7,419
14,022
29,120
123,182
168
4,914,866
1,499,258
792,511
479,927
1,824,784
1,737,558
5,630,805
2,683,591
5,173,215
1,531,445
792,123
480,052
1,884,556
1,802,583
6,005,593
2,776,140
19,563,300 20,445,707
40,009,007
Nonwhite
Male Female
411,893
19,373
4,662
17,062
73,602
80,405
626,672
203,118
1,437,227
3,011
Order and State
2nd
Male
60,341
49,678
269,668
1,153,138
397,485
143,218
644,134
564,536
3,264,198
Order
Female
63,
46,
269,
1,176,
400,
143,
656,
561,
3,318,
197
631
802
672
047
651
916
159
056
6,582,254
3rd
Male
5,326,759
1,518,631
797,173
496,989
1,898,386
1,818,403
6,257,477
2,886,709
21,000,527
Order
Female
5,618,387
1,550,276
796,264
- 493,003
1,961,859
1,888,567
6,723,774
2,988,091
22,020,221
445,172
18,831
4,141
12,951
77,303
8 5, '984
718,181
211,951
1,574,514
,741
State
Population
Totals
11,319,
3,413,
3,951,
7,823,
10,081,
4,662,
16,782,
9,706,
67,740,
366
864
777
194
158
498
304
397
558
43,020,748
71
-------
TABLE 17
SUMMARY POPULATION FIGURES FOR THE GREAT LAKES BASIN BY ORDER, SEX, AND RACE (1960)
Counties with urban centers containing populations >100,000 are omitted
State
1st Order
White
Male Female
2nd Order
White
Male Female
3rd Order
White
Male Female
PA
MN
•wi
MI
IL
IN
NY
OH
Total:
Male +
PA
MN
WI
MI
IL
IN
NY
OH
Total:
Male +
Totals:
PA
MN
WI
MI
IL
IN
NY
OH
otal: 2,
Female *
Female »
0
8,799
384,322
695,625
147,938
75,954
410,795
297,147
2,020,580
4,035
0
110
6,608
21,737
6,401
2,861
8,830
11,191
57,738
112
0
8,078
385,501
692,888
133,023
74,375
421,326
299,504
2,014,695
,275
0
92
6,398
21,041
6,294
2,200
7,977
11,234
55,236
,974
59,727
48,766
267,546
755,820
386,717
24,526
428,824
286,631
2,258,557 2
4,543,
614
912
2,122
34,300
10,768
40
8,183
7,734
64,673
125,
62,639
45,797
267,636
767,188
392,361
23,654
434,031
291,950
,285,256
813
558
815
2,166
35,002
7,686
35
6,984
7,811
61,057
730
3,264,038
905,024
683,775
379,688
1,639,041
1,267,409
1,958,806
1,519,389
11,617,170
23,533
91,388
6,004
3,114
12,849
64,395
22,522
63,116
47,247
310,635
621
Male and Female by Order and State
1st
Male
0
8,909
390,930
717,362
154,339
78,815
419,625
308,338
078,318
Order
Female
0
8,170
391,899
713,929
139,317
76,575
429,303
310,738
2,069,931
2nd
Male
60,341
49,678
269,668
790,120
397,485
24,566
'437,007
294,365
2,323,230
Order
Female
63,197
46,612
269,802
802,190
400,047
23,689
441,015
299,761
3rd
Male
3,355,426
911,028
686,889
392,537
1,703,436
1,289,931
2,021,922
1,566,636
2,346,313 11,927,805
Order
Female
3,496,554
892,500
684,453
386,159
1,758,000
1,321,482
2,095,528
1,593,176
12,227,852
3,403,683
886,921
681,517
377,182
1,690,666
1,299,659
2,029,110
1,547,862
11,916,600
,770
92,871
5,579
2,936
8,977
67,334
21,823
66,418
45,314
311,252
,887
State
Population
Totals
6,975,518
1,916,897
2,693,641
3,802,297
4,552,624
2,815,058
5,844,400
4^73,014
32,973,449
Male +
Female
4,148,249
4,669,543
24,155,657
72
-------
one to three, and two to three. A "risk" ratio was defined as the
division of an averaged age-adjusted site-race-sex specific cancer
mortality rate for a given Order divided by the corresponding
average rate "of a different Order.
6. Each average age-adjusted site-race-sex specific cancer mortality
rate was then multiplied by the census figure £o*-the respective
race, sex, and Order. This product was then divided by the total
population figure for the respective sex and race of the eight
Great Lakes states. This quotient represents a population weight-
ed average age-adjusted site-race-sex specific cancer mortality
rate by Order.
7. "Weighted'risk" ratios were calculated comparing the average rate
for each of the Orders to the overall rate for the eight Great
Lakes .states. A "weighted risk" ratio was defined as the division
of the average age-adjusted site-race-sex specific cancer mortality
rate for a given Order by the sum of the three weighted average
age-adjusted site-race-sex specific cancer mortality rates for a
given Order.
(The calculations and data derived from steps 4 through 7 will hereafter
be termed the "urban and rural" analysis.)
8. Steps 4 through 7 were repeated omitting all counties within
the eight Great Lakes states having population centers of 100,000
or more inhabitants.
(The calculations and data derived from step 8 will hereafter be termed the
"rural" analysis.)
9. The average age-adjusted site-race-sex specific cancer mortality
73
-------
rates for the "rural" and "urban and rural" analyses were examined
for disparities of increasing or decreasing mortality rates by
degree of geographical contiguity to the Great Lakes, (e.g.,
Which cancer sites have higher mortality rates in the first Order
counties than second Order counties, which in turn have higher
rates than the third Order counties?)
C. Results
The rates computed for each of the thirty-five sites for both
the "urban and rural" and "rural" analyses are presented in Appendix I,
A summary of Appendix I is given below. Only the most extreme ratio
values for non-white populations will be presented as most non-white rates
were based on relatively small numbers of deaths.
1. "Urban and Rural" Analysis
"Risk" Ratios: Order 1 cour.ties had a 31% excess of esophageal cancer
mortality in white males and a 2(>/i excess in white females over Order 2
counties. A 68% excess in mortality due to cancer of other endocrine
organs for white females and a 46% excess for breast cancer mortality in
white males was noted in Order 1 over Order 2. The "risk" ratio of Order
1 to Order 2 county nasopharyngeal cancer mortality rates was 3.66 for
non-white males and 7.74 for non-white females. The corresponding ratios
for esophageal cancer in non-whites was 1.64 and 0.04 respectively. Non-
white rate ratios for Order 1 to Order 2 counties were 5.36 for thyroid
cancer in males and 8.59 for brain and other nervous system cancers in
females.
A "risk" ratio of 40.14 in non-white females for esophageal cancer
3) All rates, ratios, and percentages quoted in this analysis are in
respect to mortality and not incidence.
74
-------
mortality was noted when comparing Order 2 to Order 3 counties. White males
experienced 42% higher recorded thyroid gland cancer mortality in Order 1
than Order 2.
The "risk" ratios comparing rates in Order 1 to Order 3 counties indi-
cated an excess of esophageal cancer mortality for white males of 44%; for
white females, 32%; non-white males, 41%; and 58% for non-white females.
The corresponding values for stomach cancer were 35%, 24%, 79%, and 33%
respectively. The ratio of rates for Order 1 to Order 3 counties for lip
cancer mortality in non-white females was 5.33, 5.73 for "other" skin can-
cers in non-white males, and 8.05 for lymphosarcoma* in non-white males.
"Weighted Risk" Ratios: The ratio of the rates in Order 1 counties
to the sum of all the weighted rates for esophageal cancer mortality were
1.28 for white males, 1.21 for white females, and 1.22 for non-white males.
The corresponding ratios for stomach cancer were 1.22, 1.16, and 1.34
respectively. White females demonstrated a risk of 0.1116 for mortality
due to cancer of the nose, nasal cavities, middle ear, and accessory
sinuses. The "weighted risk" ratio for lymphosarcoma* in non-white males
was 2.08. Non-white females had a ratio of 2.56 for cancers of the con-
nective tissues.
The mortality ratio of rates in Order 2 counties to the sum of all
the weighted rates for non-white males was 2.21 for rectal cancer, 5.64
for breast cancer, 3.22 for cancers of the eye, 5,06 for bone* cancers, and
3.88 for Hodgkin'^s disease. The corresponding ratios for non-white females-
were 2.62 for salivary gland cancers, 13.58 for esophageal cancer, and
2.07 for melanomas of the skin.
* See the NIH publication (80 ) for a complete listing of cancers included
in this category.
75
-------
The ratio of rates ±a Order 3 counties to the sum of all the weighted
rates for non-white male tongue and mouth cancers was 1.46. Non-white
females demonstrated a ratio of 4.50 for cancers of the connective tissue.
2. "Rural" Analysis
"Risk"Ratios: White males experienced 27% more salivary gland cancer,
26% more esophageal cancer, and 48% more breast cancer mortality in Order
1 counties than in Order 2 counties. Order 1 county white females experi-
enced proportionately greater mortality than Order 2 county white females
for salivary gland cancer (20%), esophageal cancer (27%), nose and middle
ear* cancers (30%), and cancer of the endocrine organs (73%). The esoph-
ageal "risk" ratios for non-whites was 1.78 for males and 0.03 for females.
The corresponding non-white ratios for cancer mortality due to brain and
other nervous system cancers were 2.02 and 10.52. The percent of excess
mortality in Order 1 over Order 2 counties for non-white males was 393%
for nasopharyngeal cancer, 188% for other skin cancers* 102% for brain
*
and other nervous system cancers, 152% for thyroid cancers, 279% for lympho-
sarcoma and 122% for I.C.D. codes not listed.* Non-white females also
experienced proportionately greater mortality in Order 1 than Order 2
counties for the cancer sites of pancreas (332%), nose and middle ear*
(120%), brain and other nervous system* (951%), Hodgkin's disease (114%),
and multiple myeloma (222%).
The "risk" ratios comparing rates in Order 2 to Order 3 counties have
the following values: for white males the thyroid cancer mortality ratio
was 1.44; non-white male mortality ratios were 1.91 for rectal cancer, 8.55
for breast cancer, 2.16 for bladder cancer, 2.27 for other skin cancers',
* See the NIH publication ( 80) for a complete listing of cancers included
in this category.
76
-------
3.29 for cancer of the eye, 5.38 for bone cancer, and 5.59 for Hodgkin's
disease. The corresponding ratios for non-white females included 2.13
for salivary gland cancers, 47.124 for esophageal cancer, 2.31 for stomach
cancer, and 2.58 for cancer of the cervix uteri.
In the comparison of Order 1 to Order 3 counties, excess cancer mor-
tality was noted for the sites of esophagus (39%), stomach (36%), larynx
(31%), and thyroid gland (33%) for white males. The corresponding percent-
ages for white females were esophagus (33%), stomach (24%), and endocrine
organs (42%). Non-white males and females had respective values for excess
esophageal cancer mortality of 42% and 22%. For stomach cancer mortality
for non-whites the excesses were 86% and 33% respectively. The non-white
male additionally had excesses for the sites of nasopharynx (203%), other
skin (555%) and lymphosarcoma (906%). Non-white females showed excess
cancer mortality for the sites of pancreas (201%), nose and middle ear*
(175%), and brain and other nervous system cancers (388%).
"Weighted Risk Ratios": The ratio of the rate in Order 1 counties, to
the sum of all the weighted rates for esophageal cancer mortality was 1.30
for white males and 1.27 for white females. White males demonstrated
elevated risks in the first Order counties for the sites of stomach (1..27),
larynx (1.25), and thyroid (1.20). A risk ratio of 1.38 was noted for
white female mortality from cancer of the endocrine organs. Non-white
males had elevated risk ratios for the sites of nasopharynx (2.50), "other"
skin (3.39), and lymphosarcoma (4,10). The sites of nose and middle ear*
and brain and other nervous system* had weighted risk ratios of 2.18 and
3.42, respectively, for non-white females.
* See the NIH publication (80) for a complete listing of cancers included
in this category.
77
-------
The "weighted risk" ratios comparing site rates in Order 2 to the sum
of all the weighted rates showed minimal differences from "1.0" for white
males and females. The largest departure from "1.0" was 1.30 for white
male thyroid cancer mortality. Non-white males had risk ratios of 4.03
for breast cancer mortality, 2.7 for mortality due to eye cancer-, 3.4 for
bone cancer mortality and 3.24 for mortality due to Hodgkin's disease. Non-
white females had elevated risk ratios for the sites of salivary glands
(2.00) and esophagus (6.19).
The ratios of site rates in Order 3 counties to the sum of all the
weighted rates for white populations of the Great Lakes states did not
differ markedly from unity. However, a 26% excess in non-white males and
a 25% excess in non-white females was noted for tongue and mouth* cancer
mortality. A 36% excess was also shown for thyroid gland cancer mortality
for non-white males.
The populations living in the Great Lakes basin would be expected to
experience varying degrees of exposure to lake contaminants. Exposure
potentials are based upon many factors including occupational exposure,
drinking water sources, landfill contamination, contamination of ambient
air, and dietary habits. Those individuals who may be exposed to contam-
inants at levels which may produce disease may comprise a small subset of
the entire population. Therefore, the excess cancer mortality contributed
by the "high risk" groups may not dramatically increase the mortality
rates in one Order of counties over those of another Order. All average
rates were examined by site for evidence of increasing or decreasing mor-
tality by relative geographic proximity to the Great Lakes. Tables 18andl9
* See the NIH publication ( 80 ) for a complete listing of cancers included
in this category.
78
-------
demonstrate increasing average mortality rates with increased proximity
to the Great Lakes for the "urban and rural" and "rural" analysis respec-
tively.
Tables 20 and 21 list by race and sex those cancer sites which have
average cancer mortality rates which decrease in magnitude as one approaches
the Great Lakes. In summary, twenty-nine race and sex specific sites were
shown to have increased with proximity to the Great Lakes and twenty decreased
in the "urban and rural" analysis. In the "rural" analysis, twenty-seven
sites increased and twenty-three decreased with proximity to the Great Lakes.
D. Discussion
In the "urban and rural" analysis comparing Order 1 counties to
those of the other Orders, white males experienced the highest mortality
from esophageal, stomach, breast, and thyroid cancers when .examining the
weighted and unweighted risk ratios. The demonstration of excess mortality
in these sites for first Order.counties waa consistent with the "rural"
analysis. White females had the greatest excess mortality for cancers of
the esophagus, stomach, and endocrine organs* in the "urban and rural"
analysis in Order 1 counties. This pattern was also found in the "rural"
analysis. The excesses in rates noted for the first Order county white
population were typically between 25% and 50% over those of the other Order
counties. A "risk" ratio of two or greater was not found in the thirty-
five-sites examined.
The non-white populations of the Great Lakes basin demonstrated ele-
vated risks for many sites in Order 1 counties. For example, non-white fe-
males had a risk ratio of 10.06 for lymphosarcoma when comparing Order 1
* See the NIH publication (80) for a complete listing of cancers included
in this category.
79
-------
TABLE 18
Cancer Sites Demonstrating INCREASING Average Mortality Rates
With Proximity to the Great Lakes Basin
Site Race Sex
Lip m F
Nasopharynx W F
Tongue, Mouth W M
W F
Esophagus W M
W F
Stomach W M
W F
NW M
Large intestine w M
Liver NW M
NW F
Pancreas W M
W F
Nose, Middle ear NW F
Larynx W M
Trachea, Bronchus, Lung W M
Breast W F
Corpus uteri NW F
Bladder W F
Other skin NW M
Endocrine organs w M
Lympho sarcoma .W M
W F
NW M
Multiple myeloma w F
Malignant neoplasms W M
W F
NW M
80
-------
TABLE 19
Rural Analysis - Omitting Counties With Population Centers }100,000
Cancer Sites Demonstrating INCREASING Average Mortality Rates
With Proximity to the Great Lakes Basin
Site Race Sex
Nasopharynx W F
Tongue, Mouth W M
W F
Esophagus W M
W F
Stomach W M
W F
NW M
Large intestine W M
Pancreas W M
W F
Larynx W M
Trachea, Bronchus, Lung W M
Breast W F
Corpus uteri NW F
Bladder W F
Other skin NW M
Endocrine organs W M
Lymphosarcoma W M
NW M
All neoplasms W M
W F
NW M
Liver NW M
NW F
Nose, Middle ear NW M
NW F
81
-------
TABLE 20
Cancer Sites Demonstrating DECREASING Average Mortality Rates
With Proximity to the Great Lakes Basin
Site Race Sex
Lip W F
Tongue, Mouth W M
m F
Pancreas NW M
Trachea, Bronchus, Lung W F
Corpus uteri W F
Ovary, Fallopian tube NW F
Kidney NW F
Bladder NW F
Skin melanoma W M
W F
Other skin W M
W F
NW F
Eye W M
Bone W F
NW F
Hodgkin's Disease W F
Leukemia W M
W F
82
-------
TABLE 21
Rural Analysis - Omitting Counties With Population centers 100,000
Cancer Sites Demonstrating DECREASING Average Mortality Rates
With Proximity to the Great Lakes Basin
Site Race Sex
Lip W F
Tongue, Mouth NW M
NW F
Large intestine W F
Pancreas NW M
Larynx NW F
Trachea, Bronchus, Lung W F
Corpus uteri W F
Bladder NW F
Skin melanoma NW M
W M
W F
Other skin W M
W F
NW F
Eye W M
Bone W F
NW F
Leukemia W M
W F
Ovary, Fallopian tubes NW F
Kidney NW F
Hodgkin's disease W F
83
-------
counties with the Order 3 counties in the rural analysis. The Order 1
county average rate was based on twenty-five reported deaths over the twen-
ty year period. The point of the example is that many of the extreme rates
reported here for non-whites are based on very small numbers of reported
deaths. Each ratio reported in Appendix I should be evaluated indepen-
dently by a critical review of both the numerator and denominator data.
The accuracy and consistency of the reporting of cancer deaths in the non-
white populations between 1950 and 1969 (93) are subject to question.
Conclusions drawn from the non-white ratios would be highly speculative
at best. The reader is cautioned since incidence, as opposed to mortality,
data are typically evaluated in studies of disease etiology.
In conclusion, the fifth analysis supports prior analyses which
indicate that white populations in first Order counties experienced a
higher rate of mortality due to esophageal and stomach cancers than in
second and/or third Order counties. These excesses remained when large
urban centers with, their possible confounding factors were removed from
the analysis. Furthermore, these disparities persisted when a second
analysis was performed which weighted each rate by the respective popula-
tion count. Excess mortality due to cancers of the breast in white males
and endocrine organs* in white females in first Order counties were also
noted in this analysis.
* See the NIH data set (80) for a complete listing of cancers included
in this category.
84
-------
A Pilot Study to Determine the Feasibility "of an Epidemiologic
Investigation Among Commercial Fishermen of the Effects
of Polychlorinated Biphenyls on Health
I. Background and Rationale
A critical concern with respect to the presence of polychlorinated
biphenyls (PCBs) in the environment is the introduction of these
compounds into the human food chain. However, there is limited
information regarding PCB exposure to the general population via food
consumption. Jelink and Corneliussen report that for the period
between 1969 and 1975, there have been significant decreases in the
PCB levels in all food commodities with the exception of fish (60).
Data from the U.S. Fish and Wildlife Service. (U.S.F.W.S.) regarding
PCB concentrations in fish species sampled from the Great Lakes indicate
that the average PCB concentrations for several fish species exceed
recommended F.D.A. levels* for human consumption. Average PCB con-
centrations were noticeably higher in fish species taken from Lake
Michigan than those taken from other lakes. It must be noted however,
that these averages are based upon small sample sizes and no trends
could be established from year to year.
Graham reported on the levels of PCB in the edible portion of
commercial fish species from the Canadian waters of the Great Lakes.
His results indicate that several commercial species have mean PCB
concentrations greater than 1 ppm. Table 22 summarizes his findings.
* Current F.D.A. tolerance levels for PCBs in fish = 5 ppm
Recommended F.D.A. tolerance levels for PCBs in fish = 2 ppm
85
-------
TABLE 22
Levels of PCBs in Canadian Fish Species
Mean PCB
Fish Species Concentration (ppm)
lake herring 1.17
lake trout 2.02
chub 2.09
carp 1.55
sucker 1.33
coho salmon 5.11
Erie alewife 1.22
carp 1.27
yellow pickerel 1.16
white bass 1.26
catfish 2.04
coho salmon 3.14
Ontario yellow perch 1.23
smelt 4.16
white perch 1.84
carp 1.69
rock bass 1.76
eel 17.14
coho salmon 4.97
Graham, J.M. "Levels of PCBs in Canadian Fish Species." in: National
Conference on Polychlorinated Biphenyls, November 17-21, Chicago,
Illinois. E.P.A. 560/6-75-004, March 1976.
86
-------
The State of Michigan has collected data regarding PCS concentrations
in fish taken from Lake Superior. Results indicate that lake trout have the
greatest PCS body burdens and exceed recommended F.D.A. levels in many in-
stances. It is interesting to note that the sampling stations used to collect
Lake Superior fish species are adjacent to counties in the upper peninsula of
Michigan which are relatively free of industries normally associated with PCS
discharges. This suggests several possible explanations including:
1) Fish may spend part of their life cycle in highly polluted waters
and then migrate to other portions of the lake.
2) Non-point sources of PCBs (e.g., atmosphere, water column, bottom
sediments) have a much greater impact on fish body burdens than
previously expected.
3) There are unidentified sources of PCBs resulting in increased ex-
posure to fish in these areas.
4) Fish taken from waters adjacent to heavily industrial areas may
have much greater PCS body burdens.
The Region V office of the U.S.E.P.A. is currently evaluating data re-
garding PCS levels in fish taken from Lake Michigan. Preliminary results in-
dicate the fish from southern Lake Michigan have greater PCB concentrations
than those from northern Lake Michigan. This is not surprising due .to the
magnitude of industrialization around southern Lake Michigan waters.
Humphrey and his associates measured the PCB levels in persons consuming
sport fish caught from Lake Michigan. These authors conclude that there could
be long-term accumulation of PCBs in individuals consuming large amounts of
fish (52). However, at the time of their studies, no differences in health
status could be observed between participants (i.e., fish consumers) and
controls. This does not negate the possibility of latent effects resulting
87
-------
from chronic exposure which might be demonstrated at a later date.
In summary, it appears that the scope of PCB contamination of the
food supply has narrowed to the point where fresh water fish are the primary
source of PCB exposure in the diet of fish-consuming individuals. According.
to Jelink and Corneliussen the average daily dietary intake of fresh water
fish of the U.S. citizen is low (60 ). However, their reports do not consider
population sub-groups which may consume significantly larger quantities of fish
than the general population and, therefore, may have a potentially greater ex-
posure to PCBs.
88
-------
II. Alms and Objectives
'A. General
The intent of this pilot study was to evaluate three research proto-
cols to determine the effectiveness and feasibility of an epidemiologic
investigation of commercial fishermen. Several issues were addressed,
including:
1. Could the requisite study participants be found?
2. Would the participants respond to the study instruments (i.e.,
questionnaires?)
3. Would one test protocol produce better responses than another?
4. Is the .information obtained from the study instruments accurate?
5. Could verification of the participant responses be obtained?
6. What is the distribution of fish consumption (both amount and
type) in this cohort?
III. Methods
A. Study Protocols
Three detailed protocols were developed for the collection of infor-
mation regarding the fishing practices and health status of commercial
fishermen. These included a telephone-survey protocol (Protocol I),
a protocol involving a pre-mailing of a set of questions with a sub-
sequent telephone-survey (Protocol II), and a mailing of a questionnaire
protocol (Protocol III).
These protocols were tested using questionnaires and study corre-
spondence developed by the Great Lakes project staff. Protocol effective-
ness was determined by evaluating which method:
1. Produced the greatest proportion of questions answered in the
correct format.
2. Produced the greatest participation rate.
3. Produced the most reliable responses as determined through validation.
89
-------
B, Identification and Selection of a Study Population
Since the purpose of this study was to identify a population in the
Great Lakes basin with potentially high PCS exposure and determine the
feasibility of an epidemiologic investigation of that "high risk" popu-
lation, several possible study populations were identified including
sport fishermen, commercial fishermen, and sub-samples of the general
population from areas contaminated with PCBs or other organics. Based on
the availability of licensing records and the recognition of the problems
associated with determining PCB exposure in the general population,
commercial fishermen were selected as the initial study population.
Available commercial fishing licenses were requested fron the states
of Minnesota, Wisconsin, Michigan, Illinois, Indiana, Ohio, Pennsylvania,
and New York. In addition to.the state listings other sources included
commercial fishing organizations and publication membership lists which
were especially useful in identifying retired commercial fishermen. In
general, the most frequently contacted sources of information were the
respective state licensing departments. Table 23 illustrates the kinds of
information obtained from each state regarding licensed commercial fisher-
men.
A potential problem with the use of current lists of commercial fisher-
men is the omission of individuals who fish intermittently and purchase licenses
on a less than yearly basis (as compared with a full-time fisherman who
purchases a license every year) and recent retirees. State departments were
willing to send current lists of commercial fishermen but expressed con-
cern over the feasibility of abstracting past records. Great Lakes study
members visited the Departments of Natural Resources in the states of
Indiana and Wisconsin to evaluate the content of commercial fishing records.
90
-------
TABLE 23
Sources of Information Regarding Commercial Fishermen
State
New York
Illinois
Indiana
Pennsylvania
Minnesota
Ohio
Michigan
Wisconsin
Number of
Sources of information regarding individuals
commercial fishermen with licenses
1. New York State Eepartment of Environmental Conservation
a) 1980 list of commercial fishermen—Lake Erie 25
—Lake Ontario 25
2. State tepartaent of Natural Resources
a) 1980 list of commercial fishermen—inland waters 382
1. State Department of Natural Resources
a) 1975-1981 list of commercial fishermen—Lake Michigan 66
b) 1980 list of commercial fishermen—inland waters 239.
1. Pennsylvania Fish Commission Division of Fisheries
a) 1979 list of commercial fishermen—Lake Erie 42
1. lepartment of Natural Resources
a) 1979 list of commercial fishermen—Lake Superior 89
2. Claude Ver Euin, Editor of The F. isherman
a) Publication list of former commercial fishermen (retired?) 25
1. Department of Natural Resources
a) 1980 list of commercial fishermen—Lake Erie 134
1. Department of Natural Resources
a) 1980 list of commercial fishermen—Lakes Michigan,
Huron, Superior
2. Claude Ver Bain, Editor of The Fisherman
a) Publication listing of commercial fishermen 170
1. Department of Natural Resources
a) 1974, 1976-1981 list of commercial fishermen in
Lakes Superior, Michigan 449
total 1,646
less - Inland Waters 621
Great Lakes 1,025
Compiled by the Great Lakes Project Staff
91
-------
An additional 58 individuals or 90% of the state cohort were abstracted
from Indiana files and 202 individuals or 45% of the state cohort were ab-
stracted from Wisconsin files. These results suggest that current license
holders do not accurately reflect the number of individuals who have commer-
cially fished in the recent past and that all state licensing departments should
be contacted in person to abstract all available records. In general, license
records are maintained from 3 to 5 years. However, some states (e.g., Wisconsin)
maintain records for up to twenty years.
For the purposes of the pilot study it was decided that the individuals ob-
tained from the various Information sources mentioned in Table 23 were suffi-
cient to test study protocols and instruments.
A. total of 1,025 individuals were identified as licensed Great Lakes
commercial fishermen between the years 1974 to 1981; The majority of commercial
fishermen are found in the state of Wisconsin and fish Lake Michigan. In
addition to the Great Lakes commercial fishermen, there were large contingents of
inland commercial fishermen (i.e., individuals who fish non-Great Lake associated
rivers and lakes in the states of Illinois and Indiana). These individuals totaled
621 but were not included in the pilot study. The reasoning behind this omission
stemmed from the fact that their socio-economic status and racial composition
may be significantly different from that of Great Lakes commercial fishermen. It
was reported that a significant number of these individuals are black and the
majority come from a low socio-economic background. However, their incorporation
into a large-scale study should be considered, possibly with separate analytical
procedures.
The study protocols and instruments were pilot tested using a randomly
selected sub-sample (n * 75) of Great Lakes commercial fishermen. Individuals were
stratified according to lake and state. The number of individuals to be used in
the pilot study from each lake/state stratum was arbitrarily selected to insure
92
-------
TABLE 24
Number of Commercial Fishermen in the
Great Lakes Basin by Lake and State
Lake
States
adjacent
to the lake
Number of fishermen
from each state who
fish the lake
Number
of fishermen
randomly selected
by state
Total
number of
fishermen to
be studied
by lake
vo
CO
Superior
Michigan
Huron
Erie
Ontario
^otals
Minnesota
Wisconsin
Michigan
Wisconsin
Michigan
Indiana
Illinois
Michigan
Ohio
Pennsylvania
New York
Michigan
New York
114
44
25
424
76
66
0
47
134
42
25
3
25
12
6
6
12
6
6
0
6
3
3
0
24
24
9
12
1,025
75
75
-------
a representative sample of commercial fishermen from the entire Great Lakes
Basin region. Table 24 describes the number of individuals available by lake
and state. Included in this table is an arbitrary number of randomly-selected
individuals to be used in the pilot study.
The sub-sample of Great Lakes commercial fishermen was divided and randomly
assigned to each of three study protocols with an equal distribution of individ-
uals among each protocol. The number of individuals assigned to each protocol was
dependent on the number of individuals arbitrarily selected from each state. The
following diagram illustrates this procedure for Lake Superior as an example:
Protocol
Minnesota
12 individuals
LAKE.SUPERIOR
I II III
(4) (4) (4)
I
(2)
Wisconsin
6 individuals
II III
(2) (2)
Michigan
6 individuals
I II III
(2) (2) (2)
94
-------
A similar procedure was used for each lake resulting in a total of 25
individuals selected per protocol for the purpose of the pilot study. The
majority of fishermen selected in this manner came from the states of
Minnesota, Wisconsin, and Michigan. There were no individuals from the state
of Illinois since available records indicated that these individuals were
inland commercial fishermen.
C. Questionnaire
The questionnaire developed for this pilot project was designed to assess
the fish consumption practices, behavioral patternstand health status of
commercial fishermen and their families. There were two forms of this
questionnaire; one for the mailed questionnaire format (Protocol III) > an<* one
for the telephone survey (Protocol I) and the telephone survey with a
pre-mailed set of questions (Protocol II). In addition, the questions were
modified for a proxy interview. There were no major differences among these
questionnaires. In some instances the sentence structure has been rearranged
to accomodate slight differences in subject approach. (See Appendix II ).
Specific information requested from commercial fishermen included the
following topics:
1. Demographic characteristics (i.e., age, race, sex)
2. Occupational histories
3. The fishing practices and consumption patterns of study participants
and their family members
4. The smoking histories of study participants (i.e., cigarette, cigar,
and pipe usage)
5. The drinking histories of study participants (i.e.,•liquor, wine, and
beer usage)
95
-------
6. The prevalence of selected diseases among study participants and their
family members
7. The pregnancy histories of female participants and the wives of male
participants
8. The prevalence of birth defects among family members of study participants
9. A participant check list of symptoms relevant to PCB exposure (and other
organic chemicals e.g., prior exposure to DDT)
10. The menstrual histories of female participants and the wives of males
participants
D. Protocol I Procedure
Twenty-five subjects were randomly allocated to Protocol I by the procedures
described in the preceding section. A file folder was made for each individual
and labeled with the individual's name, address, telephone number, and protocol
number. Materials mailed to each subject included an introductory letter, a
medical consent form, and a stamped,return-addressed envelope to return the
consent form, (see Appendix III).
To facilitate telephone interview scheduling, these materials were mailed
to five different subjects per day so that by the end of a five-day period all
twenty-five subjects had been mailed a Protocol I packet. Subjects were
telephoned seven days after the individual mailing dates of the study materials
(see Section G).
In the event that the materials packet was returned to the University because
the subject was deceased or the address was incorrect, attempts were made to
1) locate an appropriate proxy for an interview and retrieve the death certi-
ficate for the original subject or 2) obtain the correct address for the subject
and request an interview. If these procedures failed, the subject was con-
sidered lost to follow-up and his/her folder was placed in the "complete" file.
96
-------
At the time of subject contact it was determined whether or not the individual
wished to participate in the study. The outcomes regarding subject participation
included:
1) Subject could not be interviewed at the time of the call.
2) Subject was deceased.
3) Subject refused to participate in the study.
4) Subject agreed to be interviewed at the time of the call.
If the time of the call was inconvenient for the subject a date and time was
established for a return call and the subject's folder was placed in the "pending"
file with the corresponding date for retrieval.
If the subject was deceased, it was determined whether or not the contacted
individual was appropriate for a proxy interview. If the contacted individual was
appropriate as a proxy, an interview was requested. If the proxy subject refused,
additional next of kin were contacted for a proxy interview. If no next of kin'were
available as proxies, the subject's folder was labeled "complete without an inter-
view" and a request for a death certificate was sent to the appropriate state vital
statistics--off ice.
If the subject refused to be interviewed his/her folder was placed in an
"initial refusal" file for a three-week waiting period. "Following this period the
subject was called again and asked to participate in the study. If the subject
refused a second request,his/her folder was placed in the "complete" file as a
refusal.
If the subject agreed to an interview either on the initial or second request
the subject was interviewed. Following completion of the interview the subject
was asked to sign and return the medical consent form. If the consent form was
97
-------
not returned within a period of ten days, a second letter requesting the
signed medical consent form was mailed to the subject. If the consent form
was not returned after another ten day period the subject was called to
elicit his/her cooperation in returning, the consent form. All efforts were
made to obtain the signed medical consent form for study participants; however,
in those instances where these attempts were unsuccessful, the subject's
folder was placed in the "complete"file under the "no validation" category.
Upon completion of the interview and acquisition of a signed medical
consent form, the study subject's folder was turned over to the validation
section for the purpose of verifying the participant's medical history. A
detailed flow-chart regarding Protocol I procedures appears in Figure 3
E. Protocol II Procedures
Protocol II follows a similar format to that of Protocol I (see Figure 31 ,
the only difference being the contents of the mailed packet for Protocol II
which consisted of an introductory letter, a stamped, return-addressed en-
velope, a medical consent form, and,in addition, a list of questions regarding
fish consumption practices, behavioral patterns, and general health (see
Appendix V . Furthermore, the mailing of this packet was completed during the
week of telephone contact and interviewing for Protocol I. This scheduling
system prevented the development of an interview backlog and facilitated
the completion of both protocols (see Section G).
F. Protocol III Procedures
Twenty-five subjects were randomly allocated to Protocol m by the pro-
cedures described in an earlier section. A file folder was made for each
individual and labeled with the individual's name, address, telephone number,
and protocol number. A packet consisting of an introductory letter, a medical
98
-------
S
r
oade.
cct raiu
>>col I.
J^ly 'allocated to
A file Colder it
( 1UUH6 J
rllOTOCOL I
Aeedical consent ton and the
introductory letter *r« cuiltd
to tin lujbect.
Hie mbjecc (older U placed in
the pending file dated on* weak
io advance of filing date.
U
pacV«t
returned to
tht Univeriicyt
VtS
JL
Sulijm foltlnr ie ruh"»vct( (toe
panJlng Ctl* on th« uppropriatfi
U«te and tlia ititiject Is eattqj.
TBS"
iulijret le IntnrvEttweil fliiJ the
Inlnrvtrw corapUtoJ. (Proxy
tntcrvltu noted tf npprapct*lft
nut] iutij*et folitur li UliotflJ
duccmed nnd cant to Y«Hd«tlon
for dcnih CartJflcatc rggueit.)
tltte and U«o eBUfcltihcrf lor »
rcttttn c*lt and subject loldir
I* f»t«c«d in peiitHnc file "Ith
eorrcipoftdlng d«c« for r*tr|«w*t.
Sulijtct J«
(f contact:
fite for pro
I
t returned du» to
of tuUj«cC*
5*rntl to tracing with foliar frtui
(h< pending Ii!«.
Det*n»|a« ntst of kin for
proxy Interview.
to r
1
0, . i
f in tol.lir labeltd "Loit
atloy-«y.H
f
Subject J«c«a««dl Mo pro«y
int«r«lcu available. Libel
tubjcel folder "teceaicd" and
«nd 10 validation fcr death
certificate «ei«e»t. •'
Qimtluniulrt l< placet! In aul«ji>ci
CoMor labeled "Interview «pU«a'
end put In pending file d«tv*4 Irn
i)n)T« ia advuitr.* of flllnj called.
ainaa ta° interview
i
Give «,ve>tlonaelre to toJinj
Indicating "No Validation."
flace In {elder labelled
"R.lu.ol."
-------
consent form, a stamped,return-addressed envelope, and a questionnaire was
mailed to the subject (see Appendix III).
In the event that the correspondence packet was returned to the
University because the subject was deceased or the address was incorrect,
attempts were made to 1) locate an appropriate proxy for an interview and
to retrieve the death certificate for the original subject or 2) obtain
the correct address for the subject and mail the individual a Protocol III
packet. If these procedures failed, the subject was considered lost to
follow-up and his/her folder was placed in the "complete" file.
One month following the initial mailing, those individuals who had not
responded received a second Protocol III packet. Individuals not responding
to the second mailing were considered refusals and their folder was .placed-in
the "complete" file in the refusal category.
Study subjects responding to the initial or second request for partici-
pation were screened to determine if they had returned their medical consent
forms with the questionnaires. Those individuals who failed to return their
consent forms 'were mailed a second request for the medical consent form. In
those instances where these attempts were unsuccessful, the subject's folder
was placed in the "complete" file under the "no validation" category.
Upon acquisition of a completed questionnaire and medical consent form the
study subject's folder was turned over to validation for the purposes of
verifying the participant's medical history. A detailed flow-chart regarding
Protocol III procedures appears in Figure 4.
G. Timetable for Pilot Project Initiation
A timetable was developed for the initiation of subject contact and
interviewing for all three study protocols.
100
-------
5uli)«?r c r Amlomty el loc .il eit
to Protocol t!l. A f 11-'
folder ia mdiic.
\t>
rtcune 4
PROTOCOL III
See validation (low rtiort.
Subject U ml led » medical
Consent fora. i,uc»u lonnaiie,
and c-tvn* letter.
YKS
«!T I'lbjcci lolOcr it pi need
itt I lit- pvn
Subject la Bailed « aecond
rtoueit lor .edicat con.r.lt
loot.
fattens.n« next of kla for
proxy interview.
NO
/
lias
joct
rec«lv«d two
nllinga of Proto~\
col 111 ajt«ri«U7
Subject tJtce/tBfd. No proxy
interview avcilcbla. tobrt
• ubjcec folder d*>ce
-------
The initial mailing of Protocol III materials took place during the week
of June 1 to 5, 1981. Materials for five of the twenty-five subjects were
mailed each business day. All twenty-five subjects were mailed materials by
the end of the five-day period. After a period of one month, those subjects
not returning their materials were sent a second Protocol III packet during
the period June 29 to July 3, 1981.
The initial mailing of Protocol I materials took place during the period
June 8 to 12, 1981. Materials for five of the twenty-five subjects were
mailed each business day. All twenty-five subjects were mailed materials by
the end of the five day period. The corresponding phone contacts and interviews
took place seven days following the dates of mailing of Protocol I materials
for each subject. The contacts and interviews were begun on the. respective
days of June 15 to 19, 1981.
The procedure for Protocol II was similar to that of Protocol I with the
mailing of Protocol II materials occurring during the period June 15 to 19,
1981 and the contact and interviewing via telephone occurring seven days
following the mailings of Protocol II materials to each study subject. The
schedule in Figure 5 outlines the implementation of the pilot project.
H. Validation Procedures
Following completion of the interview and the acquisition of the signed
consent form the subject's folder was sent to the validation section for
purposes of verifying the medical history of the study participant.
A validation form (see Appendix III) was completed for each subject and
contained information regarding the name, address, and phone number of the study
participant, the name,address, and phone number of the medical sources, the
date and reason for contacting medical sources, and a calendar noting the dates
102
-------
FIGURE 5.
PROTOCOL IMPLEMENTATION
Heek of June 1 to June 5
Initiate Protocol III (Mailed Questionnaire)
6/1/81 6/2/81
Monday Tuesday
Mail materials to: ,5 subjects 5 subjects
Subject folder placed in pending file for one month
Week of June 8 to June 12
Initiate Protocol I (Telephone Interview)
6/8/81
Monday
Mail materials to: 5 subjects
Week of June 15 to June 19
Begin telephone interviews for Protocol I
Initiate Protocol II
6/15/81
Monday
*5 participants
#5 subjects
Week of June 22 to June 26
6/9/81
Tuesday
5 subjects
6/16/81
Tuesday
5 participants
5 subjects
6/3/81
Wednesday
5 subjects
6/10/81
Wednesday
5 subjects
6/17/81
Wednesday
5 participants
5 subjects
6/4/81
Thursday
5 subjects
6/5/81
Friday
5 subjects
6/11/81
Thursday
5 subjects
6/12/81
Friday
5 subjects
6/18/81
Thursday
6/19/81
Friday
5 participants 5 participants
5 subjects 5 subjects
Complete telephone interviews of Protocol I particij* •.•;!..••
Begin telephone interviews of Protocol II participants using similar procedures as outlined above
Week of June 29 to July 3
Begin status check on Protocol III participants and initiate second mailing for those individuals failing
to respond to the first request
Complete telephone interviews of Protocol I participants failing to interview in the proceeding week
Complete telephone interviews of Protocol II participants
Telephone Interview of Protocol I Subjects
Mail Introductory Packages to Protocol II Subjects
-------
and status of request/acquisition of medical records.
A request for medical records, a photocopy of the medical consent
form, and an abstract form were mailed to all medical sources named by the
study participant (see Appendix III). The subject's folder was placed in the
validation file dated two weeks in advance of the filing date. If the medical
records were not received after a two week period, a second request for medical
records was initiated. After three non-responses via the postal system,
medical sources were contacted by phone to obtain their cooperation. If these
methods failed to validate study participant's medical histories, his/her
folder was placed in the "complete" file with the validation complications noted
for coding and data analysis.
In instances where the study subject was deceased, a death certificate
request letter and abstract form were sent to the appropriate state vital
statistics office, (see AppendixHI). The subject's folder was placed in the
validation file dated two weeks in advance of the filing date. If the death
certificate was not received within a period of two weeks, a second request
was initiated. After three non-responses from the state vital statistics
office, the subject's folder was placed in the "complete" file and the un-
availability of the death certificate was noted.
Following the successful acquisition of all validation materials medical
histories will be verified and coded for analytical purposes.
A detailed flow-chart describing the validation procedures appears in
Figure 6,
I. Tracing Procedures
Tracing procedures were intitiated in those instances where the study
materials were returned to the University indicating that the study subject
was no longer located at the address given on the license.
104
-------
Figure t>
Validation Procedures-
O
Ul
in «dv4nc* of filing
t folder
. Ko Ifcslh
itnt to t
Ub*l*d "Ho V»l(cUt Un.
|1a»th crrt t flr^t •
Iro« •!•!« hcntlh
with »*om (orb letter
foldrr vlilcli inclu
«:M ionna lie and «
J ic« I content fon> (tota
Medic*! *o»rc«i
«n ^Uon« to alicc co
co.-aplotcb *
aliJjcion carJ for *tI
condlt i ani
tout, Inyurtng
Sjuci (olJ.'t £on(»[n* ft
fora.
I..WT subject ta\d«, "K«
L«tt«ct fora will be
nt» il»ti»l tvo
• Uv«iic« at tlm
In
Vtllilclor coepletel »«l »*lt>
e tfcand roi|ur«t let uedic&l
r«corJ« vlth «b>(V»et fora.
-------
Initial tracing procedures included the use of directory assistance,
reverse directories (Polks and Coles), and the post office to inquire about
prior addresses of the study subjects.
If these procedures failed to locate the subject, a search request to the
state department of motor vehicles was initiated. This procedure either
resulted in a known address for the subject or indicated that the subject
did not hold a driver's license in that state.
Advanced tracing procedures included contacting crew members and other
commercial fishermen, contacting local fishing ports, contacting commercial
fishing organizations, requesting information regarding state boat regis-
trations, and contacting Great Lakes fish wholesalers.
If these methods failed to locate the study subject, he/she was placed
in the "complete" file noting that the individual was lost to follow-up. A
detailed flow-chart of tracing procedures appears in Figure 7.
106
-------
1 Fil-s returned from post
of [ice on current resins-nee
indicating study subject
is no lor.y?r at the
address listed.
FIGURE 7
TRACING PROCEDURES
File sent to tracing
Change subject's tile
status to: "Lost to
Follow-up."
I
YES
Initial, iincing:
tracer utilizes
1) Directory assistance
2) Polks and Coles
3) Post office o£ prior
addtess
YES
Placo new address in
"SIR." File and enter
subject's status in
calendar file for
immediate contact.
Assign folder to a tracer
and log accordingly by date.
Further methods include:
1) Calls to crew members
listed on state license
2) Calls to local potts
3) Calls to local fishing
associations
4) State boat registrations
5) Ftesh water fiah
wholesalers
NO
•YES
Is
subject
located at a
known addiess?
V
O
\
/
Send a search request to
the respective DMV
f
J
\
* i
•iO
1
Subject license returned
to DMV due to subject
acquiring a license in a
new state.
-------
J. Coding
This section outlines the coding proposals for the Great Lakes pilot
project questionnaire. Testing of the coding and data analysis proposals
would involve the use of computer statistical packages to determine an
efficient analytical procedure applicable to a large-scale study.
The general instructions regarding the Great Lakes pilot study coding
proposal include:
1) placing a "9" in all coding spaces for a response of "dont know"
(e.g., Hi).
2) placing an "8" in all coding spaces for a response of "refused
to answer" or "not answered" (e.g., _8 £ 8).
3) placing a "0" in a coding.space when an answer requires less coding
space than allotted. The coding spaces to the left would be filled
with zeros (e.g., three spaces provided and the answer given is
9 = £ £ 9) •
The proposed set of coding instructions rOr the pilot study questionnaire
is presented in Appendix IV.
108
-------
K. Data Analysis
This, section outlines the data analysis proposal for the Great Lakes
pilot project. This proposal in conjunction with the coding proposal
outlined in the preceeding section should be tested prior to implementation
of a large-scale study. Testing would involve the use of computer
statistical packages to determine an efficient analytical procedure
applicable to a large-scale study.
Participant demographics will be used to develop a descriptive
profile of this occupational cohort. Fishing histories (i.e., the number
of years fished and location where participant fished most often) will be
evaluated in conjunction with reported fish consumption patterns to deter-
mine any discrepancies between these categories. That is, responses in one
category can be used to verify the responses in the other. In addition,
these responses will be used to determine the number of individuals with
the greatest potential PCB/organic contaminant exposure. Based on the
results of this evaluation, study subjects will be stratified into four
exposure groups. High, moderate, low, and none. Individuals classified
as "no exposure" will be used as an internal control group.
The information obtained from the survey regarding morbidity/mortality
within these four categories of exposure will be used to determine fre-
quencies. These frequencies will be used to calculate an odds ratio to
determine potential health differences among the four exposure categories.
In addition, these analyses will be performed for family members (i.e.,
spouse and children) when their exposure (i.e., fish consumption patterns)
are known to be similar to that of the study subject. . Separate odds ratios
will be calculated for stud}1- subjects in each exposure group to evaluate the
potential confounding effects of smoking and drinking behavior on health.
109
-------
The frequencies of birth defects and pregnancy complications among
family members and spouses (wives) of the study participants will be evaluated.
Exposure in these individuals will be based on two criteria including 1) the
exposure category of the study subject and 2) the similarities of fish con-
sumption practices between study subjects and their family members. These
frequencies will be used to calculate an odds ratio to determine potential re-
productive dysfunctions among the exposure categories.
Menstrual histories of female participants and the spouse of male
participants were requested and these individuals will be stratified according
to their potential exposure. Menstrual dysfunction frequencies will be
calculated for each exposure category and used to calculate odds ratios.
The hypothesis that long histories of fish consumption (especially fish
taken from areas of known environmental contamination) may increase PCB/organic
contaminants exposure and result in demonstrable human health effects is not a
realistic outcome of this pilot study. The purpose of the analytical section
of this pilot study is to test the feasibility of these proposals. Appendix V
contains a proposed set of analytical procedures and methods for the pilot
study questionnaire.
110
-------
IV. Pilot Study Results
A. Protocol I
The objective of the protocol I procedure was to evaluate the
effectiveness of a telephone interview format among a cohort .of commercial
fishermen. A correspondence packet containing a medical consent form, an
introductory letter, and a stamped, return-addressed envelope was mailed to
the study subject. After seven days the subject was contacted by phone and
and interview was requested, (see Section III-D).
Twenty individuals or 80% of the protocol I cohort were contacted by
phone and eighteen or 90% of these individuals consented to an interview.
Four subjects were never home at the time of the call and repeated efforts
to contact these individuals failed. One subject had an unpublished phone
number and efforts to contact this individual through the mail failed to
elicit a- response.
The current status of protocol I participants is summarized in Table 25.
Of the eighteen or 72% of the total cohort who granted an interview, four
have failed to return their signed medical consent form and one has refused
to return the signed medical consent form. Efforts are continuing to 1)
obtain the consent forms of the four subjects (29, 39, 45, and 46) who have
not returned their signed consent form or 2) obtain a response indicating a
refusal to cooperate with this aspect of the study procedure. Thirteen
individuals or 52% of the total cohort are complete or partially complete for
validation procedures.
The average time for an interview was approximately thirty-five minutes
with a range of approximately twenty to fifty minutes. In addition, no study
111
-------
subject
study
number
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
TABLE 25
Status of Protocol I Participants
subject
status
interview complete - validation procedures are partially complete
interview complete - validation procedures are complete
subject not home at time of call
interview complete - signed consent form has not been returned
refusal
interview complete - validation procedures are complete
subject nothome at time of call
interview complete - validation procedures are partially complete
interview complete - validation procedures are partially complete
interview complete - validation procedures are partially complete
subject not home at time of call
interview complete - validation procedures are complete
interview complete - validation procedures are complete
interview complete - signed consent form has not been returned
interview complete - validation procedures are partially complete
unpublished phone number
interview complete - validation procedures are partially complete
interview complete - validation procedures are partially complete
interview complete - validation procedures are partially complete
interview complete - signed consent form has not been returned for
wife (subject divorced)
interview complete - signed consent form has not been returned
interview complete - subject refuses to return signed consent form
interview complete - validation procedures are complete
refusal
subject not home at time of call
112
-------
materials were returned due to 1) an incorrect mailing address or 2) the
death of the study participant. This suggests that information regarding commer-
cial fishermen obtained from state files, publication lists, and fishing
organizations was accurate with respect to address and reflected a living cohort.
Several problems were encountered with the telephone interview format in-
cluding:
1) scheduling interviews with the study subject and spouse
2) subjects who were never at home at the time of call
3) subjects who were without a phone
4) subjects who had unlisted phone numbers
Despite these difficulties the addresses and living status were verified
for all protocol I participants.
In those instances where interview times were difficult to obtain, either
the subject or the spouse was requested to complete the entire interview.
This situation occurred in three of the eighteen respondents. In 'one instance,
the study subject completed the entire interview and in two instances the spouses
of the study subject completed the entire interview. In eleven of eighteen
respondents both the subject and spouse participated in the interview. Four of
the eighteen respondents were living alone (i.e., single, widowed, separated/
divorced) and the questions regarding reproductive histories and dysfunctions
were not applicable.
In those instances where subjects did not have a phone or had an unlisted
phone number, a letter was sent requesting their participation along with a card
requesting a number where they could be reached. Of the two individuals which fit
this category one responded and granted an interview.
Four study subjects were never home at the time of call despite consistent
efforts to reach them. A possible explanation for their absence could be that
113
-------
they are living on their fishing vessel during the seasonal months. This
suggests that a large scale study should be undertaken during the off-season
when fishermen are more likely to spend time at their listed residence.
In general, the study subjects in protocol I responded well when asked to
participate in the survey. There appeared to be no difficulties in interpreting
the survey questions on the part of the study subjects with the exception of
questions 149 and 150. These questions ask for the average length of the
menstrual cycle and the average length of the "bleeding" period, and were
consistently misinterpreted by respondents despite explanations by the inter-
viewer. In addition, a potential problem may exist with the validity of
question 26 regarding the pounds of fish consumed in a year. Many respondents
hesitated when answering this question and needed to be coaxed into providing
an answer. However, this answer will be cross-checked with the responses to
questions 19 and 22 regarding the number of fish meals consumed per week or per
month to provide a valid fish consumption index for data analysis purposes.
B. Protocol II
The objective of the protocol II procedure was to evaluate the effectiveness
of a telephone interview preceded by a mailed set of questions identical to
those asked during the telephone survey. A correspondence packet containing
a letter of introduction, a medical consent form, a stamped, return-addressed
envelope and a set of questions identical to those on the questionnaire was
mailed to the study subject. After seven days the subject was contacted by
phone and an interview was requested.(see Section III-E).
The purpose of the mailed set of questions identical to those asked in the
interview was to evaluate whether this protocol produced more complete and
valid responses on the part of the study subject given the fact that they would
familiarize themselves with the questions prior to the interview.
114
-------
Twenty-two individuals Or 88% of the protocol II cohort were contacted by
phone and 19 or 86% of these individuals consented, to an interview. Two sub-
jects had unpublished phone numbers and efforts to contact these individuals
faile'd to elicit a response. One individual had recently moved to California and
had not established a new residence at the time of this study.
The current status of protocol II participants is summarized in Table 26
Of the 19 or 76% of the total cohort who granted an interview five have failed to
return their signed medical consent form and one has refused to return the signed
medical consent form. Efforts are continuing to 1) obtain the consent forms.of the
five subjects (51, 57, 59, 63 and 75) who have not returned their signed con-
sent forms or 2) obtain a response indicating a refusal to cooperate with
this aspect of the study procedure. Twelve or 48% of the total cohort are
complete or partially, complete to validation procedures.
No study materials were returned due to 1) an incorrect mailing address
or 2) the death of the study participant; In addition, the addresses and
living status were verified for all protocol II subjects with the exception
of subject 70 who was in transit to an out of state residence at the time of
the study. In those instances where interview times were difficult to obtain,
either the subject or their spouse was requested to complete the entire inter-
view. This situation occurred in eight of the nineteen respondents. In five
instances the subject completed the entire interview and in three instances
the wife completed the entire interview. In seven of nineteen respondents
both the subject and spouse participated in the interview. Three of nineteen
respondents were single (i.e., never married) and the questions regarding
reproductive histories and dysfunctions were not applicable. One respondent
refused to let their spouse be interviewed, and as a result, no information was
obtained regarding reproductive histories and dysfunctions for this subject:.
115
-------
TABLE 26
Scatus of Protocol II Participants
..subject
study
number
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
.subject
status
interview complete - signed consent form has not been returned
interview complete - validation procedures are complete
interview complete - validation procedures are partially complete
interview complete - validation procedures are complete
refusal
interview complete - validation procedures are partially complete
interview complete - signed consent form has not been returned
refusal
interview complete - signed consent form has not been returned
refusal
interview complete - validation procedures are partially complete
unpublished phone number
interview complete - signed consent form has not been returned
interview complete - subject refuses to return .signed consent form
unpublished phone number
interview complete - validation procedures are partially complete
interview complete - validation procedures are partially complete
interview complete - validation procedures are partially complete
interview complete - validation procedures are partially complete
directory assistance does not have a listing
interview complete - validation procedures are partially complete
interview complete - validation procedures are partially complete
interview complete - validation procedures are partially complete
interview complete - validation procedures are partially complete
interview complete - signed consent form has not been returned
116
-------
The results and problems experienced in protocol II are similar to
those experienced and discussed in protocol I. Methods used to remedy these
problem areas were identical. The major difference between these protocols
was the potential impact of the pre-mailed set of questions in protocol II
which was to aid in the subject's development of responses prior to an inter-
view on the telephone. The effectiveness of this protocol, as compared to
protocol I, will be thoroughly evaluated following the completion of the vali-
dation procedures. However, several potential problem areas were identified
including:
1) the number of questions was too large
2) the specificity of the questions may have been offensive to several
participants
3) the study subjects did not use the set of questions for the purposes
intended
In general, the study subjects in protocol II responded well to the
survey questionnaire. In addition, family members (i.e., spouses) responded well
when asked to participate in the survey. Similar problems with questions 26,
149, and 150 were encountered among the protocol II participants. Future
application of this protocol to a cohort of commercial fishermen should con-
sider modification of the pre-mailed set of questions. Both a reduction in the
number of questions and the specificity of those questions appear warranted.
However, the comparison of validation results for protocols I and II may alter
this observation.
C. Protocol III
The objective of the protocol III procedure was to evaluate the effective-
ness of a mailed questionnaire format among a cohort of commercial fishermen.
A correspondence packet containing a medical -consent form, a stamped, return-
117
-------
addressed envelope, an introductory letter, and a questionnaire was mailed to
the subject. A. second, identical packet was mailed in the event that the study
materials were not returned following a one month time period, (see Section
III-F).
Six individuals responded to the initial mailing for a response rate of
24%. Of these six individuals, one failed to return the signed medical records
request form.
An additional five individuals or 20% responded to the second mailing. Of
these five individuals one returned a refusal message, one responded saying
that he was not nor had he ever been a commercial fisherman, and two failed co
return the signed medical request consent form.
Fourteen individuals or 56% of the study subjects failed to respond in
any manner to the mailed questionnaire format presented in section III-F.
The current status of protocol III participants is summarized in Table 27
Of the eleven individuals or 44% who responded to the mailed questionnaire
format seven or 28% are complete or partially complete for validation procedures
Two or 8% or complete without consent forms, and efforts are continuing to
1) obtain the consent forms of subjects 3 and 17 or 2) obtain a response indi-
cating a refusal to cooperate with this aspect of the study procedure.
The response time for the return.of the study materials for protocol III
ranged between twelve to fifteen days for both mailings. These results indicate
that the one month waiting period was an appropriate length of time for parti-
cipant response. In addition, only one study packet was returned due to an
incorrect mailing address and subsequent efforts to obtain the correct address
were successful. No study materials were returned due to the death of a study
participant. This suggests that information -regarding commercial fishermen ob-
tained from state files, publication lists, and fishing organizations was
118
-------
TABLE 27
Status of Protocol III Participants
subject
study
number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
subject
status
No response following two mailings
responded on the second mailing indicating that he was never a
commercial fisherman
responded on the second mailing - signed consent form has not
been returned
responded on the second mailing - validation procedures are
partially complete
No response following two mailings
No response following two mailings
responded on the first mailing - validation procedures are complete
responded on the second mailing - refusal
responded on the first mailing - validation procedures are complete
No response following two mailings
No response following two mailings
No response following two mailings
responded on the first mailing- - validation procedures are complete
No: response following two mailings
No response following two mailings
No response.following two mailings
responded on the first mailing - signed consent form has not
been'returned
responded on the first mailing - validation procedures are
partially complete
responded on the second mailing - validation procedures are
partially complete
No response following two mailings
No response following two mailings
No response following two mailings
No response following two mailings
responded on the first mailing - validation procedures are complete
No response following two mailings
119
-------
accurate with respect to address and reflected a living cohort.
A major problem experienced with the mailed questionnaire format was
the incorrect or incomplete responses obtained from five of the nine study
subjects who signed and returned the questionnaire. It is difficult to assess
whether this was attributable to questionnaire length, questionnaire mis-
interpretation, or individual inconsistency. However, four individuals com-
pleted the questionnaire in its proper context suggesting that the difficul-
ties with incomplete responses are most likely due to the length of the
questionnaire and/or inconsistency on the part of the study subject.
In general, the participants correctly completed the questions regarding
fishing practices and consumption patterns and their behavioral patterns (i.e.,
smoking, drinking). Incomplete or incorrect responses were most often en-
countered with the questions regarding familial medical histories.
Future application of this protocol procedure to a larger cohort should
consider several modifications including:
1) a reduction of the waiting period for a participant response from
one month to two to three weeks,
2) structural changes of the questionnaire to facilitate participant
response.
3) telephone contact with study subject to encourage cooperation with
the survey in the event that two mailings failed to elicit a response.
D. Discussion
A significant factor regarding the appropriateness of a commercial
120
-------
fishermen cohort to assess the potential health impacts of organic
contaminants in the environment is the distribution of fish consumption,
and thus potential exposure to organic contaminants, among this population.
A preliminary analysis was performed based upon the combined responses
of protocols I and II to evaluate the number of fish meals per months, the
number of years consuming fish with this frequency, the number of pounds
of fish consumed per year, and the types of fish most often consumed.
Table 28 summarizes these preliminary results.
In general, the combined responses of individuals from protocols I
and II suggest that commercial fishermen eat fairly large quantities of
fish and, for the most part, have been doing so for a good many years.
Approximately 14% of the respondents consume between 0-3 fish meals per
month, 42% consume between 4-7 fish meals per month and 44% consume > 8
fish meals per month. The fish species most frequently mentioned as being
consumed were perch, followed by lake whifefish, lake herring, and lake
trout.
It is interesting to note the distribution of study subjects from
protocols I and II regarding the pounds of fish consumed per year; the
majority of which report £.50 Ibs/year. These results appear to be in-
consistent with the results regarding the number of fish meals consumed
per month. A possible explanation is that some individuals consume large
quantities (i.e., >2 Ibs) of fish at one sitting compared to other in-
dividuals who consume small quantities (i.e., < *slb). To resolve this
potential inconsistency an additional question regarding the average
amount of fish consumed per meal should be considered for inclusion in a
large-scale study.
These preliminary results suggest that there is a good distribution
121
-------
TABLE 28
PHELiHIHMIY nesuyrs ron PROTOCOLS i MID_II
Distribution o< study
subjects frow Protocols
1 and II regarding the
number of flflh feeala
eonauaed per. eontn
Distribution of study
subjecte Iron Protocols
I and II regarding the
number of yeara ffah
uero consuued with this
frequency
Distribution of etuoy
aubjeete ftos Protocols
I and II regarding the
nunber of pounds of fish
conauoicd per year
Typed of floh eott
often mentioned ae
consumed by respondents
frou Protocols I and It
mmber of Huh nuetber of
eeele per aonth study subjects
0-1 3
2-3 2
4-i 12
6-7 3
B-« J
10 12
niraber of years
fish conaitneii with number of
thia frequency atudy subjects
0-J 3
6-10 2
11-1) 1
16-20 3
11-30 7
31 21
nu«ber of pounds
of flali consumed number of
per year study subjects
0-2) 10
26-50 «
51-7S &
76-100 3
101 S
fish species
Perch
Lake- uhlttf Uh
Luke herring
Lake trout
Roughflsh
Sxelt
Salmon, Valleys
Chubs
Northern Plk«.
Burbot. Pickerel
*nuiaber of
study
-------
of fish consumers among this cohort ranging from 0-1 meals per month
to acre than 10 meals per month. In addition, there is a good distribution
of fish consumers based on yearly consumption amounts ranging from less
than 25 Ibs/year to greater than 100 Ibs/year.
Humphrey and his associates measured the PCB levels in persons
consuming sport fish and reported that the most frequently recorded
quantity of fish actually consumed was in the 25-35 pounds per year
range. However, the number of.-participants reporting a consumption
greater than 35 pounds per year was low (52). Approximately 252 of
Humphrey's cohort consumed £30 pounds of fish per year. A comparison
with preliminary pilot results indicates that approximately 50Z of the
commercial fishermen cohort consume > 50 pounds of fish per year.
These findings suggest that a population sub-group consuming considerably
greater quantities of fish can be identified and are willing.to par-
ticipate in an epidemiologic investigation.
123
-------
I. Cost Analysis for Protocol Implementation
A. Protocol I
The cost per protocol I participant can be approximated by determining
the cost of individual study instruments and procedures. The following is a
list with approximate costs of study materials and procedures used in protocol I.
letter or introduction $ .05
medical consent form .05
return-addressed, stamped envelope .23
mailing envelope and postage . 23
long distance phone call(interview)
$.38 x 35 minutes 13.50
second request for consent form .05
medical consent form .05
return-addressed,stamped envelope .23
mailing envelope and postage .23
Total $14.62
Based upon the price of a long distance phone call to New York for an
average interview time of thirty-five minutes the estimated phone costs are
$13.50 per study subject. In addition, the costs of mailing both an introductory
study packet as well as a second request for medical consent forms are included.
(On the average there was one additional request for the medical consent form per
study subject). The total cost for interviewing and obtaining returned study
materials from protocol I participants is approximately $15.00 per subject.
Not included in this approximation are secretarial and interviewer costs.
Considering these additional expenses the cose per protocol I participant is
approximately $20.00. However, it must be kept in mind that these are estimated
124
-------
costs based on average participant's response times. In general, telephone costs
(the major expense of this protocol) will vary according to subject residence and
length of interview time.
B. Protocol II
The cost per protocol II participant can be approximated by determining the
cost of individual study instruments and procedures. The following is a list with
approximate costs of study materials and procedures used in protocol II.
letter of introduction $ .05
medical consent form .05
a set of questions (5 pages) .25
return-addressed, stamped envelope .23
mailing envelope and postage .72
long distance phone call
$.38 x 35 minutes 13.50
.second request for consent form .05
medical consent form .05
return-addressed, stamped envelope .23
mailing envelope and postage .23
Total $15.36
The expense figures for protocol II are similar to those of protocol I.
However, the mailing costs, the cost for an enclosed set of questions, and a slight
increase in secretarial costs increase the overall cost per protocol II partici-
pant. Considering these additional expenses the cost per protocol II participant
is approximately $22.00. The variables associated with this cost approximation
are similar to those discussed for protocol I.
125
-------
C. Protocol III
The cost per protocol III participant can be approximated by determining
the cost of individual study instruments and procedures. The following is a list
with approximate costs of study materials used in protocol III.
letter of introduction $ .05
medical consent form .05
questionnaire (23 pages) 1.15
return-addressed, stamped envelope 1.45
mailing envelope and postage 1.62
second request for consent form .05
medical consent form .05
return-addressed, stamped envelope .23
mailing envelope and postage .23
Total $ 4.88
The cost of a second mailing to 76% of the protocol III subjects averages out
to be approximately $3.28 per participant. la addition, there was an average of
one additional request for the medical consent form per study subject.
The cost for the mailed questionnaire format is approximately $8.00 per study
subject. Considering additional secretarial expenses, involving increases in time
and typing, the total costs are approximately. $10.00 per study subject.
In summary, the costs associated with protocols I and II are very similar
(approximately $20.00 per study subject) with, protocol II being slightly more
expensive. The cost associated with protocol III is cheaper by about half. However,
it appears that both protocols I and II are more cost-effective in that they achieve
a much higher rate of response than protocol III. In addition, the information
obtained by protocols I and II is much more complete than that obtained in protocol
III. These results suggest that the use of either protocol I or II would be more
126
-------
cost-effective in a large-scale study and less likely to be biased by non—response
and incomplete information.
The cost-effectiveness of the added expense of protocol II will be determined
following validation. If a significant percentage of responses obtained from
protocol II participants are more accurate than those obtained from protocol I, then
the added expense of protocol II procedures may be warranted in a larger study.
Evaluation of the accuracy of responses from protocols I and II, as determined through
preliminary validation, indicate that these differences are slight. Therefore, pro-
tocol I appears to be the most efficient and cost-effective.
The costs associated with validation and tracing procedures are the same re-
gardless of the protocol and are discussed in the following section.
D. Validation
The cost per study participant for validating medical histories is based upon
the following list of study instruments.
a letter requesting medical records $.05
a copy of the signed consent form .05
a medical records abstracting form
(4 pages) .20
a return-addressed, stamped envelope .40
a mailing envelope and postage .23
On the average, two requests were sent to each medical source. In addition,
the cost associated with obtaining the addresses of medical sources involved an
average of three long-distance calls per source given by the study subject. These
include.:
1 call to directory assistance for the phone number
1 call to a related medical source for information $2.07
1 call to actual medical source for an address $l-r-31
127
-------
There was an average of three medical sources contacted per study subject.
As a result the cost associated with the validation of the medical histories is
approximately $16.00 per study subject.
Not included in this approximation are secretarial and tracer costs. Con-
sidering these additional expenses, (using an average of one hour to obtain the
mailing address of the medical sources mentioned by the study subjects), the total
cost is approximately $21.00 per study subject.
E. Tracing
The cost associated with tracing individuals was minimum in this pilot study
since most of the addresses were current. In several instances, phone calls were
made to verify the addresses of those individuals with unlisted phone numbers, no
phones, or who were not at home at the time of the call. This cost amounted to an
average of approximately $1.00 per study subject.
The cost associated with tracing medical sources (i.e., deceased physicians)
is included in the validation expenses per study subject.
128
-------
Conclusions
A pilot study was initiated to determine the feasibility of an epidemiologic
investigation, among commercial fishermen, of the association of PCBs on health.
Commercial fishermen were selected as a potential "high risk" population because
of their fish consumption habits and the availability of licensing records from
which a cohort could be extracted. The purpose of this pilot study was to evaluate
three research protocols to determine their effectiveness as epidemiologic tools
and the appropriateness of commercial fishermen as a cohort.
The pilot study has provided the following conclusions:
1. The addresses of all study subjects in Protocols I and II were verified.
The format of Protocol III was structured such that the verification of
subject location was not applicable.
2. The response rates for Protocols I and II were similar (72% and 76%
of the total cohort respectively).
3. The response rate for Protocol III was 44% after two mailings.
4. Protocols I and II were more effective in producing:
a) answers in the correct format,
b) the highest response rates, and
c) the most accurate information.
5. The differences between Protocols I and II regarding accuracy of infor-
mation are slight. Therefore, Protocol I appears to be the most efficient
and cost-effective of the three protocols tested.
6. The fish consumption patterns among this cohort, obtained by compiling
the information from Protocols I and II, indicates several trends.
These include:
a) a good distribution of fish consumption per month among the cohort,
b) a good distribution of the number of years fish have been consumed
with this frequency among the cohort,
129
-------
c) a good distribution of the quantity of fish consumed per year among
the cohort, and
d) the preferred consumption of several target fish species from the Great
Lakes among this cohort.
These results suggest that commercial fishermen were appropriate as a study
cohort and that Protocol I would be the most effective and cost-efficient method
of epidemiological ascertainment.
In addition to the pilot study an analysis of the morbidity and mortality patterns
of the Great Lakes populations was conducted. State vital statistics regarding
county fetal, neonatal, and infant death rates and congenital anomaly rates were
examined for the states of Minnesota, Wisconsin, Illinois, Indiana, Michigan, and
Ohio for every fifth year from 1950 to 1975 and the year 1977. Furthermore, an
evaluation of the county site, race, sex, and age-adjusted cancer mortality rates
from the National Cancer Institute's publication "U.S. Cancer Mortality: 1950 - 1969"
was conducted (80).
These analyses have generated the following conclusions:
1. There were no significant trends regarding the percent differences for
neonatal death rates and fetal death rates among "lake bordering " and
'non-lake bordering" counties having rates ^L 1 standard deviation higher
than their respective state means.
2. There was a slight trend regarding the difference between percent of
live births with congenital anomalies among "lake bordering" and "non-lake
bordering" counties having rates _2l 1 standard deviation higher than their
respective state means. This trend favored the "non-lake bordering"
counties.
3. The fertility rates for the eight states were not analyzed due to incon-
sistent reporting between states.
130
-------
4. An evaluation of the cancer rates of the counties of the Great Lakes
states, stratified according to proximity to the lake, indicated an
increasing trend with proximity to the Lakes for esophageal and stomach
cancers. These trends are still apparent after counties with population
centers 2t 100,000 have been excluded from the analysis.
These results suggest that "Lake-bordering"'populations (i.e., white populations)
experience higher rates of mortality due to stomach and esophageal cancers as com-
pared to 'hon-lake bordering1'counties. This trend is consistent when the potential
confounding factor of large urban centers is removed.
There were no apparent trends regarding the fetal death rate, the neonatal death
rate, and the percent of live births with congenital anomalies, among "lake-bordering"
and "non-lake bordering" counties.
131
-------
Methodological Approaches for Continued Research
The purpose of this study has been to determine whether it is
feasible to ascertain analytically any adverse human health effects from
organic contaminants released into the Great Lakes basin. The results
of the feasibility study indicate-that current epidemiologic method-
ologies are available for detection of human pathologic responses to
organic chemicals. This conclusion is based upon the following findings:
1. Federal and State agencies responsible for monitoring environ-
mental indices have reported detectable levels of organic
contaminants in the ambient air, waters, soils, lake sediments,
and fish speciea of the Great Lakes basin.
2. Animal research studies have demonstrated that organic chemicals
are fat soluble and bioaccumulate in aquatic food chains. An-
imal dietary studies have shown that ingestion of contaminated
foodstuffs may produce a variety of pathologic states.
3. Human epidemiologic studies of acute toxic exposures to organic
chemicals have reinforced the findings of animal dietary studies.
4. Existing data sets Ce-g., the Public Health Service publication
0 80 ) titled: U.S. Cancer Mortality by County: 1950-1969)
may be utilized to define patterns of morbidity and mortality
experienced by human-populations of the Great Lakes basin.
5. The Great Lakes Study Pilot Project on Commercial Fishermen
has data indicating that this occupational cohort consumes
target fish, species which have been implicated as providing
an exposure potential to consumers. Furthermore, some fisher-
men have reported the consumption of large quantities of
target species and may constitute a "high-risk" population for
132
-------
disease.
Therefore, full scale studies are warranted in addressing the
issues of whether the organic contaminants released into the Great Lakes
basin are causing a demonstrable adverse human health effect. To discern
the possible complex "agent-host-environment" interactions involved in
this process, multiple approaches will be necessary utilizing the ana-
lytical techniques from many disciplines. The methodologies utilized in
the feasibility study have been incorporated in the following multi-
faceted protocol:
Analysis of the 1950-1969 County Cancer Mortality Rates:
The purpose of refining the evaluation of the county 1950-1969
cancer mortality data set ( 80 ) is to identify t^e variables respon-
sible for the discrepancies noted in analysis five. The procedures to
be added to analysis five are as follows:
1. The standard errors for each'me'an age-adjusted site-race-sex
specific cancer mortality rate would be calculated for the
three Orders.
2. Confidence intervals would be examined comparing the rates
of Order 1 against Orders 2 and 3.
3. For those Order 1 sites whose mean age-adjusted site-race-sex
specific cancer mortality rate confidence intervals lie outside
the corresponding confidence intervals for the other Orders,
data will be gathered on those factors which may be considered
as confounding variables. Demographic information acquired
may include mean income for each county and the percentage
representation by race, religion, industry, and agriculture.
Additional data will include the types of occupations located
133
-------
within a county which may provide exposure to other agents
suspected to cause the disease under consideration. Statistical
techniques (e.g., regression analysis) will be implemented to
discern the influence each possible confounding variable may
have on the observed differences between rates.
4. Environmental monitoring has indicated that the Great Lakes
vary in the degree to which they are contaminated by organic
chemicals. A second analysis would be conducted which strati-
fied the counties by lake. The purpose of this second
evaluation would be to determine if the trends noted for the
Order 1 counties were due to a subset of Order 1 counties
associated with a particular lake.
5. In analysis five all counties which were not of the first or
second Order were considered third Order counties. In r.ha re-
analysis only those counties which are geographically .adjacent
to the second Order counties, and not of 'the first Order,
would be considered third Order counties.
The incorporation of these revisions into the protocol for analysis
five would provide two benefits. First, it would aid in defining those
factors which have produced the differences noted in cancer mortality
among the counties. Secondly, the utility of a data bank such as the
county 1950-1969 cancer mortality data (80 ) will be evaluated for its
usefulness in generating testable hypotheses regarding disease etiology.
A Comparative Morbidity/Mortality Survey of Great Lakes Counties with
Varied Levels of Organic Contamination:
Two Great Lakes counties would be selected for comprehensive reviews
of environmental and human health parameters. One county known from
134
-------
state and federal monitoring programs to be extensively contaminated
by organic compounds would be selected for comparison to a county known
to have minimal contamination. The counties would be comparable on the
demographic variables listed below.
Upon selection of the study counties, thorough environmental char-
acterizations would be initiated. The following 'information would be
evaluated:
1. The data files of the regional Environmental Protection Agency
(EPA) offices would be reviewed regarding surface water dis-
chargers. The drinking water sources would be assessed in
terms of location and mineral analysis.
2. Environmental evaluations would be conducted on ambient waters,
sediments, ambient air, and landfills.
3. A market basket analysis would be requested from the Department
of Public Health on foodstuffs consumed in the communities.
4. Industries and agricultures would be reviewed in terms of the
types of health risks posed to the community and their employees.
The human health parameters to be reviewed by county include:
1. age and cause-specific death rates
2. infant, neonatal, and fetal death rates
3. percent of live births with congenital anomalies
4. birth rates and fertility ratios
5. percent of hospital admissions and out-patient services by
diagnostic category
6. medical care utilization rates
Demographic data would be obtained for both counties regarding:
1. population distribution by age, race, and sex
135
-------
2. average income and education
3. percent of the population employed in various occupations
4. the percent of inhabitants living in urban and rural areas
5. population densities
The data on human health parameters will be examined for disparities
between the two counties. If the difference in rates cannot be explained
by known, risk factors or confounding variables, the routes of exposure
for possible etiologic agents will be discerned.
This investigation would require a field staff for record abstract-
ing and patient interviews. Technicians would be required for additional
environmental sampling and laboratory analysis. The justification for
this study is two-fold. First, the routes of agent-host-environment
interaction need to be identified. Second, it could allow a statement
as to whether existing levels of organic contaminants are associated
with adverse health effects in the populations studied.
Health Survey of Great Lakes Fishermen
The pilot study of Great Lakes commercial fishermen has provided
evidence that some fishermen eat large quantities of fish species known
to have levels of organic contaminants which may exceed Food and Drug
Administration (FDA) limits. The pilot project has developed and pre-
tested material which can be utilized in large scale studies.
The fishermen's health study would be implemented in two phases.
The first phase would focus on interviewing all currently licensed
commercial fishermen in the Great Lakes (U.S. license holders only).
The names and addresses of the registered commercial fishermen would
be obtained from licensing offices of state governments. The second
component of the study would be undertaken with the cooperation of the
136
-------
Wisconsin Department of Natural Resources (DNR). Preliminary discussions
have been held with the DNR for a joint survey of Wisconsin sport fish-
ermen. In 1982 a stratified* one-percent random sample of licensed
sport fishermen would be drawn from DNR files. Each individual would
be mailed a postcard which has a series of questions on fishing practices
and fish consumption patterns. Upon return of the completed postcard,
a second stratified random sample will be initiated based upon levels of
fish consumption. Those sport fishermen selected in the second sample
and all licensed commercial fishermen will comprise the cohort for the
"Health Study of Great Lakes Fishermen."
Survey interviewers would obtain the fishermen's phone numbers from
"directory assistance" and call to establish a telephone interview. The
methods for tracing, subject contact, interviewing, validation, and
data analysis have been outlined in protocol I of the Great Lakes com-
mercial Fishermen Pilot Project.
A subset of the study cohort will be chosen for a survey on dietary
habits and practices. Project staff will conduct in-person dietary
interviews with the families of fishermen. Fish samples will be taken
for laboratory analysis. A seven day dietary record form will be given
to family members to complete and return to the study office. This
phase of the study is necessary to document potential exposure to organ-
ic contaminants and as a validity check on the questionnaire. It is
understood that the present documentation of the ingestion of contaminated
fish does not assume that fish eaten at other times were contaminated.
Current clinical effects such as a cancer would require exposure to the
etiologic agent prior to the sampling.
*Stratification would be based on the type of sport fishing license
purchased in 1982.
137
-------
The data acquired from the "Health Study of Great Lakes Fishermen"
will address the following issues:
1. Is the fish consumed by Great Lakes fishermen contaminated by
organic chemicals? If so, do the levels of contamination lie
within F.D.A. limits?
2. Are the methods of fish preparation (i.e., broiling or smoking)
contributing to the level of contaminants in the fish consumed?
3. Is the information reported in the pilot study questionnaire
consistent with other dietary instruments?
4. Are individuals who consume large quantities of fish at a
greater risk of disease than low or non-consumers? If so,
what clinical effects are manifested in fish consumers?
5. If high fish consumption is a risk factor for disease, are the
agent-host-environment interactions defined? Are these inter-
actions biologically plausible? If yes, which link in the
chain of transmission could be removed for preventive measures?
Each of the three proposed studies addresses an important aspect of
the issue under consideration. The first study examines the utility of.
existing data sources for the identification of mortality and morbidity
patterns. The second study will focus on the health characteristics and
trends of communities with varied levels of contamination. The last
study will discern the health status of a possibly "high-risk" group.
There is a paucity of information in the scientific community on
the human health effects of chronic low level exposures to environmental
pollutants. The three-faceted approach outlined above can help to
increase our understanding of this vital issue.
138
-------
GREAT LAKES STUDY - BIBLIOGRAPHY
1) ABC's OF PCB's. Public Information Report. Wis-SG-76-125 University of
Wisconsin Sea Grant Program (1976)
2) Ahmed, M. and Focht,D.D. DEGRADATION OF POLYCHLORINATED BIPHENYLS BY TWO
SPECIES OF ACHROMOBACTER. Can. J. Micro. _19: 47-52 (1973)
3) Ahmed, M. and Focht.D.D. OXIDATION OF POLYCHLORINATED BIPHENYLS BY
ACHROMOBACTER, PCB Bull. Environ. Contain, and Toxicol. K): 70-72 (1973)
4) Allen, J.L. and Norfaeck,D.H. POLYCHLORINATED BIPHENYL AND TRIPHENYL-INDUCED
GASTRIC MUCOSAL HYPERPLASIA IN PRIMATES. Science 179: 498-499 (1973)
5) Allen, J.R. _et al. BIOLOGICAL EFFECTS OF POLYCHORINATED BIPHENYLS AND
TRIPHENYLS ON THE SUB-HUMAN PRIMATE ENVIRONMENT. Environ. Res. 6:344-354 (1973)
6) Allen, J.R. et al. RESIDUAL EFFECTS OF SHORT-TERM, LOW-LEVEL EXPOSURE OF
NON-HUMAN PRIMATES TO POLYCHLORINATED BIPHENYLS, Toxicol. Applied Pharm.
J30: 440-451 (1974)
7) Allen, J.R. and Barsotti, D.A. THE.EFFECTS OF TRANSPLACENTA! AND MAMMARY
MOVEMENT OF PCB's ON INFANT RHESUS 'MONKEYS. Toxicology 6: 331-340 (1976)
8) Allen, J.R. _et al. RESPONSES OF RATS EXPOSED TO PDLYCHLORINATED..BIPHENYLS FOR
FIFTY-TWO WEEKS. I. COMPARISON OF TISSUE LEVELS OF PCB AND BIOLOGICAL
CHANGES. Arch, of Environ. Contam. 4: 404-419 (1976)
9) Allen, J.R. _et al. MORPHOLOGICAL CHANGES IN MONKEYS CONSUMING A DIET
CONTAINING LOW LEVELS OF 2,3,7,8- Tetrachlorodibenzo-P-Dioxin.
Fd. Cosmet. Toxicol. L5: 401-410 (1977)
10) Allen, J.R. et al. RESIDUAL EFFECTS OF POLYCHLORINATED BIPHENYLS ON ADULT
NON-HUMAN PRIMATES AND THEIR OFFSPRING. J. Toxicol. Environ. Health 6.:
55-66 (1980)
11) Andrews, E.J. PCB DIET. Science 180: 255-257 (1973)
12) Bahn, A.K. et al. MELANOMA AFTER EXPOSURE TO PCB's. NEJM 295: 450 (1976)
13) Barr, M. ENVIRONMENTAL CONTAMINATION OF HUMAN BREAST MILK. Am.J.Pub.Health
_7J.: 124-126 (1981)
14) Barsotti, D.A. _et al REPRODUCTIVE DYSFUNCTION IN RHESUS MONKEYS EXPOSED
TO LOW LEVELS OF POLYCHLORINATED BIPHENYLS (Aroclor 1248) Fd. Cosmet.
Toxicol. _U: 99-103 (1976)
15) Baughman, G.L. and Lassiter, R.R. "PREDICTION OF ENVIRONMENTAL POLLUTANT
CONCENTRATION." IN: Estimating the Hazard of Chemical Substances to
Aquatic Life. A.S.J.M. STP 657, 1978.
16) Bidleman, J.R. and Olney, C.F. HIGH VOLUME COLLECTION OF ATMOSPHERIC
POLYCHLORINATED BIPHENYLS. Bull. Environ. Contam. and Toxic. XI; 442-250 (1974)
139
-------
17) Bltman, J. et al. BIOLOGICAL EFFECTS OF POLYCHLORINATED BIPHENYLS IN RATS AND
QUAIL. Environ. Health Perspect. _1: 145-149 (1972)
18) Brown, D.P. and Jones, M. MORTALITY AND INDUSTRIAL HYGIENE STUDY OF WORKERS
EXPOSED TO POLYCHLORINATED BIPHENYLS. Arch. Environ. Health 3£: 120-129 (1981)
19) Brown, G.G. et al. MEMORY PERFORMANCE OF CHEMICAL WORKERS EXPOSED TO POLY-
BROMINATED BIPHENYLS. Science 212: 1413-1415 (1981)
20) Budiansky, S. IS CANCER ENVIRONMENTALLY CAUSED? Environ.Science, and Tech.
L5: 615 (1981)
21) Calabrese, E.J. and Sorenson.A.J. THE HEALTH EFFECTS OF PCBs WITH PARTICULAR
EMPHASIS ON HUMAN HIGH RISK GROUPS. Reviews on Environ. Health_2£ 285-304 (1977)
22) Calabrese, E.J. ANIMAL MODEL OF HUMAN DISEASE,INCREASED SENSITIVITY TO
POLYCHLORINATED BIPHENYLS. Am. J. of Path. 91: 405-407 (1978)
23) Collins, W.T. et al. EFFECT OF POLYCHLORINATED BIPHENYL (PCS) ON .THE THYROID
GLAND OF RATS. Am. J. of Path. 89: 119-130 (1977)
24) Cordle, F. at al. HUMAN EXPOSURE TO POLYCHLORINATED BIPHENYLS AND POLYBROMINATED
BIPHENYLS. Environ. Health Perspect. .24: 157-172 (1978)
25) Dagley, S. CATABOLISM OF AROMATIC COMPOUNDS BY MICROORGANISMS. Adv. in Microb.
Physio. JK 1-46 (1971)
2';) Defoe, D.L. .et al. EFFECTS OF AROCLOR 1248 and 1260 ON THE FATHEAD MINNOW
(Pimephales promelas). Journal of the Fisheries Research Board of Canada
3_5: 997-1002 (1978)
27) DeLong, R.L. et al. PREMATURE BIRTHS IN CALIFORNIA SEA LIONS: ASSOCIATION WITH
HIGH ORGANIOCHLORLNE P.OLLUTANT RESIDUE LEVELS. Science 181; 1168-1169 (1973)
28) Dexter, R.N. and Paulou, S.P. MASS SOLUBILITY AND AQUEOUS ACTIVITY COEFFICIENTS
OF STABLE ORGANIC CHEMICALS IN THE MARINE ENVIRONMENT: POLYCHLORINATED
BIPHENYLS. Marine Chemistry 6: 41-53 (1978)
29) Durfee, R.L. et al. PCBs IN THE UNITED STATES INDUSTRIAL USE AND ENVIRONMENTAL
DISTRIBUTIONS. Versar.Inc. EPA 560/6-76-005 Final Report (1976)
30) Eckenhausen, F.W. et al. ORGANOCHLORINE PESTICIDE CONCENTRATIONS IN PERINATAL
SAMPLES FROM MOTHERS AND BABIES. Arch, of Environ. Health. 2£: 81-92 (1981)
31) Elder, D.L. andFowler, S.W. POLYCHLORINATED BIPHENYLS: PENETRATION INTO THE
DEEP OCEAN BY ZOOPLANKTON FECAL PELLET TRANSPORT. Science 197; 459-460 (1977)
32) Fabia, J. and Thuy, J.D. OCCUPATION OF FATHER AT TIME OF BIRTH OF CHILDREN
DYING OF MALIGNANT DISEASE. Br. J. of Prev. Soc. Med. 28: 98-100 (1974)
33) FINAL REPORT OF THE SUBCOMMITTEE ON THE HEALTH EFFECTS OF POLYCHLORINATED
BIPHENYLS AND POLYBROMINATED BIPHENYLS. Dept. of Health Education and
Welfare, Washington D.C. (1976)
140
-------
34) Fischbein, A. e_t al. CLINICAL FINDINGS AMONG PCS-EXPOSED CAPACITOR MANU-
FACTURING WORKERS. Ann. New York Acad.Sci. 320; 703-715 (1979)
35) Fraumeni, J.F. CHEMICALS IN HUMAN TERATOGENESES AND TRANSPLACENTAL
CARCINOGENESIS. Pediatrics 53: 807-812 (1974)
36) Fries, G.F. POLYCHLORINATED BIPHENYL RESIDUES IN MILK OF ENVIRONMENTALLY
AND EXPERIMENTALLY CONTAMINATED COWS. Environ. Health Persp. 1: 55-59 (1972)
37) Furukawa, K. et al. EFFECT OF CHLORINE SUBSTITUTION ON THE BIODEGRADABILITY OF
POLYCHLORINATED BIPHENYLS. Appl. and Environ. Micro. 35.: 223-227 (1978)
38) Gaffney, P.E. CHLOROBIPHENYLS AND PCB's: FORMATION DURING CHLORINATION
J.W.P.C.F. ^9: 401-404 (1977)
39) Graham, J.M. "LEVELS OF PCB's IN CANADIAN COMMERCIAL FISH SPECIES", IN:
National Conference on Polychlorinated Biphenyls. November 19-21,
Chicago, Illinois, EPA 560/6-75-004, March, 1976.
40) Hakulinen, T. .et al. CANCER IN THE OFFSPRING OF FATHERS IN HYDROCARBON-
RELATED OCCUPATIONS. Br. J. of Prev. Soc. Med. _30: 138-140 (1976)
41) Halter, M.T. and Johnson, H.E. "A MODEL SYSTEM TO STUDY THE DESORPTION
AND BIOLOGICAL AVAILABILITY OF PCBs IN HYDROSOILS," IN: Aquatic
Toxicology and Hazard Evaluation A.S.T.M. STP634, F.L. Mayer and J.L.
Hamelink (ed's) 178-195 (1977)
42) Hammer, D.I. et al. POLYCHLORINATED BIPHENYL RESIDUES IN THE PLASMA AND
HAIR OF REFUSE WORKERS. Env. Health Perspect. 1: 83 (1972)
43) Hague, R. et al. AQUEOUS SOLUBILITY ADSORPTION AND VAPOR BEHAVIOR OF
POLYCHLORINATE BIPHENYL AROCLOR 1254, Env. Sci. and Tech. 8: 139-142
(1974)
44) Hargraves, W.A. and Allen, J.R. THE IN VITRO BINDING OF 2,2',5,5' - TETRA-
CHLOROBIPHENYL METABOLITES TO RAT LIVER MICROSOMAL PROTEINS. Research
Cotnm. Chem. Path, and Pharm. 25; 33-52 (1979)
45) Harvey, G.R. and Steinhauer, W.G. ATMOSPHERIC TRANSPORT OF POLYCHLORO-
BIPHENYLS TO THE NORTH ATLANTIC, Atmospheric Environment _8: 777-782
(1974)
46) Rolled, G.J. POLYCHLORINATED BIPHENYLS (PCBs) IN THE LAKE SUPERIOR
ECO-SYSTEM: ATMOSPHERIC DEPOSITION AND ACCUMULATION IN THE BOTTOM
SEDIMENTS. Ph.D. thesis, University of Minnesota, (1979)
47) Holmberg, P.C. CENTRAL-NERVOUS-SYSTEM DEFECTS IN CHILDREN BORN TO
MOTHERS EXPOSED TO ORGANIC SOLVENTS DURING PREGNANCY. Lancet _2:
177-179 (1979)
48) Hoppenbrouwers, T. et al. SEASONAL RELATIONSHIP OF SUDDEN INFANT DEATH
SYNDROME AND ENVIRONMENTAL POLLUTANTS. Am. J. of Epidemiol. 113:
623-635 (1981)
141
-------
49) Hoppingarner, R.- et al. POLYCHLORINATED BIPHENYL INTERACTIONS WITH TISSUE
CULTURE CELLS. Env. Health Perspect. _!: 155-158 (1972)
50) Horowitz, W. and Howard, J.W. SAMPLING METHODS FOR TRACE ORGANIC
ANALYSIS IN FOODS. National Bureau of Standard Special Publication
519, Proceedings of the Materials Research Symposium (1978)
51) Hsu, I.C. et al. METABOLIC FATE OF H 2,5,2',5' TETRACHLOROBIPHENYL IN
INFANT NONHUMAN PRIMATES. Bull. Environ. Contam. Toxicol. 14: 233-
240 (1975)
52) Humphrey, H.E.B. EVALUATION OF CHANGES OF THE LEVEL OF POLYCHLOROATED
BIPHENYLS (PCBs) IN HUMAN TISSUE. Final Report on FDA Contract
223-73-2209, Michigan Department of Public Health (1978)
53) Hutzinger, 0. .et al. "PHOTOCHEMICAL DEGRADATION OF •'CHLOROBIPHENYLS (PCBs)
IN: Environ. Health Perspect. (NIH) 72-218 no. 1 (1972)
54) Hutzinger, 0. et al. THE CHEMISTRY OF PCBs. CRC Press, Cleveland (1974)
55) IARC MONOGRAPHS ON THE EVALUATION OF THE CARCINOGENIC RISK OF CHEMICALS
TO HUMANS. Volume Jji, (1978)
56) INVENTORY OF MAJOR MUNICIPAL AND INDUSTRIAL POINT SOURCE DISCHARGERS IN
THE GREAT LAKES BASIN, Great Lakes Water Quality Board, July, 19.78
57) Janerich, D.T. et al. CANCER INCIDENCE IN THE LOVE CANAL AREA. Science
212: 1404-1407 (1981)
58) Jarvinen, A.W. et al. LONG TERM TOXIC EFFECTS OF DDT FOOD AND WATER
EXPOSURE ON FATHEAD MINNOWS (Pimephales promelas) J. Fish Res. Board
.34: 2089-2103 (1977)
59) Jarvinen, A.W. and Tyo, R.M. TOXICITY TO FATHEAD MINNOWS OF ENDRIN IN
FOOD AND WATER. Arch. Environ. Contam. Toxicol. ]_'• 409-421 (1978)
60) Jelin'ck, C.F. and Corneliussen, P.E. LEVELS OF PCBs IN THE U.S. FOOD
SUPPLY. National Conference on Polychlorinated Biphenyls, November
19-21, Chicago, Illinois, EPA 560/6-75-004, March, 1976.
61) Kaiser, T.E. et al. ORGANOCHLORINE PESTICIDE, PCB, AND PBB RESIDUES
AND NECROPSY DATA FOR BALD EAGLES FROM 29 STATES - 1975-1977.
Pest. Mont. J. 13: 145-149 (1980)
62) Kendrick, E. TESTING FOR ENVIRONMENTAL CONTAMINANTS IN HUMAN MILK.
Pediatrics jb6: 470-471 (1980)
63) Kimbrough, R.D. et al. ADENOFIBROSIS IN THE RAT LIVER. Arch. Environ.
Health 27: 390-395 (1973)
64) Kimbrough, R.D. "PATHOLOGICAL FINDINGS ASSOCIATED WITH CHRONIC EXPOSURE
TO PCBs." IN: National Conference on Polychlorinated Biphenyls.
EPA 68-01-2928. Washington, D.C. (1976) p 30-34
142
-------
65) Kodama, H. and Ota, H. TRANSFER OF POLYCHLORINATED BIPHENYLS TO INFANTS-
FROM THEIR MOTHERS. Arch. Environ. Health 35:-95-100 (1980)
66) Kolbyne, A.C. FOOD EXPOSURES TO POLYCHLORINATED BIPHENYLS. Environ.
Health Perspect. 1: 85-88 (1972)
67) Roller, L.D. ENHANCED POLYCHLORINATED BIPHENYL LESIONS IN MOLONEY
LEUKEMIA VIRUS-INFECTED MICE. .Clinical Toxicol. 11: 107-116 ,(1977)
68) Kreiss, K. et al. CROSS-SECTIONAL STUDY OF A COMMUNITY WITH EXCEPTIONAL
EXPOSURE TO DDT. J.A.M.A. 245; 1926-1930 (1981)
69) Kreiss, K. _et al. ASSOCIATION OF BLOOD PRESSURE AND POLYCHLORINATED
BIPHENYL LEVELS. J.A.M.A. 245: 2505-2509 (1981)
70) Kuratsune, M. AN ABSTRACT OF RESULTS OF LABORATORY EXAMINATION OF
PATIENTS WITH YUSHO AND OF ANIMAL EXPERIMENTS. Environ. Health Perspect.
1: 129-136 (1972)
71) Kuratsune, M. et al. EPIDEMIOLOGIC STUDY ON YUSHO, A POISONING CAUSED '.
BY INGESTION OF RICE OIL CONTANIMATED WITH A COMMERCIAL BRAND OF
POLYCHLORINATED BIPHENYLS. Environ. Health Perspect. 1: 119-128
(1972)
72) Kurzel, R.B. and Cetrulo, C.L. THE EFFECTS OF ENVIRONMENTAL POLLUTANTS
ON HUMAN REPRODUCTION, INCLUDING BIRTH DEFECTS. Environ. Sci. and
Tech. 15: 626-640 (1981)
73) Rutz, F.W. and Strassman, S.C. RESIDUES OF POLYCHLORINATED BIPHENYLS
IN THE GENERAL POPULATION OF THE UNITED STATES. Proceedings of the
National Conference on Polychlorinated Biphenyls. EPA 560/6-75-004,
p.139-143 (1975)
74) Kutz, F.W. et al. "SURVEY OF PESTICIDE RESIDUES AND THEIR METABOLITES
IN HUMANS" IN: Pesticide Management and Insecticide Resistance,
Academic Press, Inc. p.523-539 (1977)
75) Lawrence, J. and Tosine, H.M. ADSORPTION OF POLYCHLORINATED BIPHENYLS
FROM AQUEOUS SOLUTIONS AND SEWAGE. Environ. Sci. and Tech. 10:
381-383 (1975)
76) Leatherland, J.F. _et al. EFFECT OF DIETARY MIREX AND PCBs ON HEPATOMATIC
INDEX LIVER LIPID, CARCASS LIPID, AND PCB AND MIREX BIOACCUMULATION IN
YEARLING COHO SALMON, ONCORHYNCHUS, KISUTCH. Biochem. Physiol. 63:
243-246 (1979)
77) Loose, L.D. .et al. POLYCHLORINATED BIPHENYL AND HEXACHLOROBENZENE
INDUCED HUMORAL IMMUNOSUPPRESSION. J. of Reticulo. Soc. 22: 253-271
(1977)
78) Loose, L.D. .et al. IMPAIRED HOST RESISTANCE TO ENDOTOXIN AND MALARIA
IN POLYCHLORINATED BIPHENYL AND HEXACHLOROBENZENE-TREATED MICE.
Infect, and Imnun. 20: 30-35 (1978)
143
-------
79) Longo, L.D. ENVIRONMENTAL POLLUTION AND PREGNANCY: RISKS AND UNCERTAIN-
TIES FOR THE FETUS AND INFANT. Am. J. Obstet. and Gynecol. 137:
162-173 (1980)
80) Mason, T.J. and McKay, F.W. U.S. CANCER MORTALITY BY COUNTY: 1950-1969.
U.S. Department of Health, Education and Welfare. UHEW Publication
No. (NIH) 34-615, National Cancer Institute,•Bethesda, Maryland
81) Maugh, T.H. DDT: AN UNRECOGNIZED SOURCE OF POLYCHLORINATED BIPHENYLS.
Science 180: 578-579 (1973)
82) Miller, R.W. jet al. PCBs IN BREAST MILK. Pediatrics 62: 407 (1978)
83) Monograph: PRELIMINARY REPORT OF EPIDEMIOLOGYCAL AND ENVIRONMENTAL DATA.
Contract number EPA/R806282 (Division of Epidemiology, Division of
Environmental Health, University of Minnesota) April 4, 1980.
84) MORBIDITY AND MORTALITY WEEKLY REPORT, Center for Disease Control:
Polychlorinated Biphenyl exposure - Indiana 27: 99-100 (1978)
85) 'Murphy, T.J. and Rzeszutko, C.P. POLYCHLORINATED BIPHENYLS IN PRECIPITA-
TION IN THE LAKE MICHIGAN BASIN. EPA Grant 803915 Environmental
Research Laboratory, Office of Research and Development, U.S. EPA,
Duluth, MN (1977)
86) Nebeker, A.V. et. al. EFFECT OF POLYCHLORINATED BIPHENYL COMPOUNDS ON
SURVIVAL AND REPORDUCTION OF THE FATHEAD MINNOW AND FLAGFISH. Trans.
of the Am. Fish Soc. 103: 562-568 (1974)
87) Nebeker, A.V. and Puglisi, F.A. EFFECT OF POLYCHLORINATED BIPHENYLS
(PCBs) ON SURVIVAL AND REPRODUCTION OF DAPHNIA, GAMMARUS, AND TARRYTARSUS.
Trans. Am. Fish Soc. 103: 722-728 (1974)
88) Nisbet, C.T. and Sarofim, A.F. RATES AND ROUTES OF TRANSPORT OF PCBs IN
THE ENVIRONMENT. Environ. Health Perspect. JL: 21-28 (1972)
89) Ohnishi, Y. and Noda, S. EFFECTS OF POLYCHLORINATED BIPHENYLS (PCBs)
ON DISSOCIATED CONJUNCTIVAL CELLS IN VITRO. Jap. J. of Opthalmol.
21: 496-506 (1977)
90) Parrki, M.G. at al. LONG-TERM EFFECTS OF SINGLE AND COMBINED DOSES OF
DDT AND PCS ON DRUG-METABOLIZING ENZYMES IN RAT LIVER. J. of Toxicol.
and Environ. Health 3: 903-911 (1977)
91) Passino, D.R.M. and Kramer, J.M. TOXICITY OF ARSENIC AND PCBs TO FRY OF
DEEPWATER CISCOES (Coregonus). Bull. Environ. Contam. Toxicol. 24:
527-534 (1980)
92) Peakall, D.B. et al. EMBRYONIC MORTALITY AND CHROMOSOMAL ALTERATIONS
CAUSED BY AROCLOR 1254 IN RING DOVES. Environ. Health Perspect.
1: 103-104 (1972)
93) Percy, C. et al. ACCURACY OF CANCER DEATH CERTIFICATES AND ITS EFFECT
ON CANCER MORTALITY STATISTICS. Am. J. Pub. Health 71: 242-250 (1981)
144
-------
94) Perry, T.W.'et al. "DIETARY AROCLOR 1254 IN THE-MILK OF LACTATING CATTLE."
IN: Trace Substances in Environmental Health Education. Hamphill, D.T.
University of Missouri, Columbia, Missouri (1979) p.271-273
95) PESTICIDES IK AQUATIC ENVIRONMENTS. Mohammed Abduhl Quddus Khan (ed.)
Plenum Press, New York and London, Proceedings of a Symposium for
the International Congress of Entomology held in Washington, D.C. (1976)
96) Pierce, R.H. Jr., et al. P,PT - DDT ADSORPTION TO SUSPENDED PARTICULATE
MATTER IN SEA WATER. Geochimica et Cosmochimica Acta ^8: 1061-1073
(1974)
97) POLYCHLORINATED BIPHENYLS. National Academy of Science, Washington, D.C.
(1979)
98) Price, H.A. and Welch, R.L. OCCURRENCE OF POLYCHLORINATED BIPHENYLS IN
HUMANS. Environ. Health Perspect. JL: 73-78 (1972)
99) Renterghem, R.V. and Devlaminck, L. POLYCHLORINATED BIPHENYL COMPOUNDS IN
MILK AND DAIRY PRODUCTS. Z. Lebensm. Unters. Forsch. 170: 346-348 (1980)
100) Risebrough, R.W. et al. PESTICIDES: TRANSATLANTIC MOVEMENTS IN THE
NORTHEAST TRADES. Science 159; 1233-1236 (1968)
101) Rogan, W.J. _et al. POLLUTANTS IN BREAST MILK. N.E.J.M. 302: 1450-1453
(1980)
102) Rogoff, M.H. OXIDATION OF AROMATIC COMPOUNDS BY BACTERIA. Adv. in Appl.
Micro. 2: 193-222 (1961)
•103) Rozman, K. _et al. CHRONIC LOW DOSE EXPOSURE OF RHESUS MONKEYS TO HEXA_
CHLOROBENZENE (HCB). Chemosphere : 177-184 (1978)
104) Safe, S. and Hutzinger, 0. POLYCHLORINATED BIPHENYLS: PHOTOLYSIS OF
2,4,6,2',4',6', - HEXACHLOROBIPHENYL. Nature 232: 641-642 (1971)
105) Savage, E.P. et al. NATIONAL STUDY OF CHLORINATED HYDROCARBON INSECTICIDE
RESIDUE IN HUMAN MILK, USA. Am. J. Epidemiol. 113 413-422 (1981)
106) Sayler, G.S. et al. POLYCHLORINATED BIPHENYL (PCS) DEGRADING BACTERIA
AND PCB IN ESTUARINE AND MARINE ENVIRONMENTS. Estuarine and Coastal
Marine Science 6: 553-568 (1978)
107) Snarski, V.M. and Puglisi, F.A. EFFECTS OF AROCLOR 1254 ON BROOK TROUT,
Salvelinus fontinalis. Ecological Research Series EPA-666/3-76-112,
December (1976)
108) Stadnickl, S.S. et al. DNA SINGLE STRAND BREAKS CAUSED BY 2,2',5,5' -
TETRA CHLOROBIPHENYL AND ITS METABOLITES. Res. Commun. in Chem. and
Pharm. _24: 313-327 (1979)
109) Stotz, I.J. and Greichus, Y.A. THE EFFECTS OF A POLYCHLORINATED BIPHENYL
(AROCLOR 1254) ON THE-WHITE PELICAN: ULTRASTRUCTURE OF HEPATOCYTES.
Bull. Environ. Contact. Toxicol : 319-325 (1979)
145
-------
110) Sullivan, P.M. and Barlow, S.M. CONGENITAL MALFORMATIONS AND OTHER
REPRODUCTIVE HAZARDS FROM ENVIRONMENTAL CHEMICALS. Proc. R. Soc. Lond.
205: 91-110 (1979)
111) Thompson, C.Z. _et al. EFFECT OF AROCLOR 1254 DOSE UPON S9 ACTIVATION OF
AROMATIC AMINE BACTERIAL MUTAGENS. Mutation Res. 21~- 361-366 (1980)
112) Veith, G.D. RECENT FLUCTUATIONS OF CHLOROBIPHENYLS (PCBs) IN THE GREEN
BAY WISCONSIN REGION. Environ. Health Perspect. 1: 51-54 (1972)
113) Vos, J.G. TOXICOLOGY OF PCBs FOR MAMMALS AND FOR BIRDS. Environ. Health
Perspect. 1: 105-117 (1972)
114) Warshaw, R. et al. DECREASE IN VITAL CAPACITY IN PCB-EXPOSED WORKERS IN
A CAPACITOR MANUFACTURING FACILITY. Ann. N.Y. Acad. Sci. 320;-
277-283 (1979)
115) Weininger, D. ACCUMULATION OF PCBs BY LAKE TROUT IN LAKE MICHIGAN. PhD
Thesis, University of Wisconsin (1978)
116) Wickizer, T.M. _et al. POLYCHLORINATED BIPHENYL CONTAMINATION OF NURSING
MOTHERS' MILK IN MICHIGAN. Am. J. Pub. Health 71^: 132-137 (1981)
117) Yobs, A.R. LEVELS OF POLYCHLORINATED BIPHENYLS IN ADIPOSE TISSUE OF THE
GENERAL POPULATION OF THE NATION. Environ. Health Perspect. J.: 79-81
(1972)
118) Zar, H. INVESTIGATION OF CHLORINATED AND NON-CHLORINATED COMPOUNDS IN
THE LOWER FOX RIVER WATERSHED. EPA 905/3-78-004 (1978.)
146
-------
Addendum
Survey of Large Cohort of Commercial Fishermen
A. Introduction
With the demonstration of the apparently more cost-effective
telephone interview approach (Protocol I) in the pilot feasibility
study, it was decided to proceed with a study of a large commercial
fishermen cohort using an improved questionnaire (Appendix II) based on
experiences with the pilot questionnaires. This was possible with the
time remaining in the basic contract.
There follows a detailed description of this survey and its
preliminary results. It should be noted that the completion of this
initiated study, in accordance with the analytic proposals in
Appendices IV and V, is contingent upon renewal funding. In addition,
the proposed coding instructions in Appendix IV apply to the questionnaire
used in the survey of the large cohort of commercial fishermen, whereas
the analytic proposals in Appendix V apply to the questionnaires used
in the small-scale pilot study.
To evaluate the potential size of a cohort of commercial fishermen
it was decided to compile a list based upon all available records in
the Great Lakes states. This procedure would hopefully provide
additional information regarding cohort size.
147
-------
B. General Description of the Cohort
During January and February of 1982 the study supervisor traveled
to the states of Indiana, Ohio, Pennsylvania, and Wisconsin to obtain
listings of potential commercial fishermen. Listings from Illinois,
Michigan, Minnesota and New York could be mailed to the Division and thus
visits were not required.
DNR and license bureau records varied in their completeness, but
all eight states were able to provide listings back to 1975. In many
states, roster entries included company names such as Steve Phalen
Fisheries and H and G Fish Company. These entries pose a problem as
any number of fishermen and crew members can be employed under this
blanket company license and will not appear as individual license holders.
Table 29 describes the availability of records and approximate number
of commercial fishermen by state.
Table 29. Historical Perspective of Commercial Fishing Records
State
Ohio
Illinois*
Indiana
Minnesota
Michigan
Wisconsin
Pennsylvania
New York
Availability Approximate Number
of of
Records Commercial Fishermen
back to 1968
back to 1975
back to 1974
back to 1943
back to 1971
back to 1971
back to 1961
back to 1971
Dunkirk Station
back to 1971
Cane Vincent Station
Total
200
5
80
250
300
300
200
150
86
1,571
+ 75* (Illini
1,646
*Have requested remaining records back to 1970; DNR official stated
that he would guess approx. 75 commercial fishermen held licenses
from 1970-1975.
148
-------
The results of this investigation indicate that there are
approximately 1,600 independent commercial fishermen from the eight
Great Lakes States dating back as far as 1968. However, for initiation
of this survey the lists detailed on page 91 of Report II were used
to generate study participants. It can be assumed, therefore, that
the use of the lists dating back to 1968 would provide another 600
individuals.
For purposes of this survey it was decided to restrict study
eligibility to individuals possessing a commercial fishing license
from any of the eight Great Lakes States and fishing Great Lakes waters.
As a result of eligibility requirements, individuals employed under a
company license, inland waters commercial fishermen and crew members
of license holders were not included in the survey. The use of these
individuals in an expanded study would substantially increase cohort size.
C. Survey Implementation
Five Senior Survey interviewers were hired and began interviewing
November 13, 1981 after a three day training session conducted by the
project supervisor. General principles of interviewing, field pro-
cedures, and question-by-question specifications were reviewed during
the first training session, followed by role plays, mock interviews and
the distribution of assignments during the second session.
Interviewers were given 10-20 assignments each week and the study
supervisor met with each interviewer once a week to distribute new
assignments and review the previous week's completed interviews.
Detailed information regarding study materials and procedures can be
found in Appendix III.
149
-------
Each completed questionnaire was edited by the study supervisor
for errors, omissions and contradictory and/or confusing data. When
this occurred, the supervisor or interviewer re-contacted the respondent
to obtain the correct information. Not only did this editing process
produce as thorough and complete an interview as possible, it also
provided the study supervisor with an accurate ongoing evaluation of
the interviewer's work.
Tracing was not a major problem in the Great Lakes Fishermen
Health Study as the lists of potential commercial fishermen were obtained
in 1980. Respondents who had moved from the last known address were
usually located with few major problems since the commercial fishermen
tended to be an informed, close-knit community. In addition to identi-
fying respondents who had moved, fellow fishermen were invaluable in
supplying information about vacationing respondents and confirming or
negating the commercial fishing status of unavailable-respondents.
Due to time constraints no attempts were made to validate the
respondents' medical history although consent forms were requested from
each participant. The focus of this survey was to collect complete
information on as many fishermen and their families as possible via the
Great Lakes Fishermen Health Survey (see Appendix II).
D. Preliminary Survey Results
From an original roster of 1,025 licenses 71 were eliminated
because they were owned by companies. Of the remaining 954 individuals
with licenses, 241 were ineligible because they did not fish commercially.
The reasons why individuals possessed a commercial licences but did
not fish include:
150
-------
1. The individual had taken out a license, but never used'it
due to the prohibitive cost of equipment and DNR regulations.
2. The individual held the license in name only while another
family member(s) did the actual fishing.
3. Depending on the state or particular area, owners of charter
•boat services, hatcheries, and bait and tackle shops were
either required to purchase a license or held a license
"just in case".
4. The individual fished on inland rivers and lakes only.
Thirty one individuals were deceased and another 41 were unavailable
during the time of interviewing due to winter vacations and/or requested
disconnected telephones. Repeated attempts to contact the unavailable
individuals were made throughout the duration of the interviewing phase.
Fifty one individuals initially refused to participate in the
study. Of these 51, 10 were converted into complete interviews
achieving a refusal conversion rate of 19.6%, leaving a total of 41
refusals. Six individuals could not be located through tracing efforts.
In summary, there was an original roster of 1,025 licenses which
provided 675 eligible study participants. Of the 675, interviews were
obtained on 587 individuals for a response rate of 87%. Out of 587
respondents, 483 fishermen had wives (wives who were also fisherwomen
are included in this category), 75 had no wives (i.e., wives were
deceased, divorced over 10 years or single, never married) and 29 (6%)
of the respondents wives refused to participate in the study. The
response rate among wives of study participants was 94%.
151
-------
Table 30 presents the general results of the survey and Table 31
presents the distribution of eligibles, respondents and response rates
by state.
Table 30. Breakdown of Study Eligibles, Respondents and
Survey Response Rates
Original Sample
Company license
Ineligible
Deceased
Other*
1,025
71
241
31
7
675 Study Eligibles
Refusals
Unavailables
Lost to follow-up**
41
41
6
587
Study Respondents
Response Rate = 87%
Fishermen w/o wives
Refusals (wives)
75
29
483
Study Respondents (wives)
Response Rate = 94%
*0ther represents duplicate names
**Lost individuals could not be located through tracing efforts
152
-------
Table 31. Distribution of Study Eligibles, Respondents
and Response Rates by State*
Original Sample
Ineligibles
Deceased
Other
POSSIBLE
INTERVIEWS
Refusals
Unavailables
Lost
COMPLETED
INTERVIEWS
IL
5
0
0
0
5
1
2
0
2
IN
63
19
1
0
43
3
3
1
36
MI
147
39
1
3
104
5
11
1
87
MN
101
16
6
0
79
6
2
1
70
NY
49
18
0
0
31
1
4
0
26
OH
no
45
0
2
63
4
6
1
52
PA
41
18
0
0
23
1
1
0
21
WI
438
86
23
2
327
20
12
2
293
Response Rate 40%
0.3%
84%
84% 89% 84%
83% 91% 90%
% Contribution
to total cohort
6.1% 14.8% 11.9% 4.4% 8.9% 3.6% 49.9%
*Fishermen only, wives not included in this distribution
153
-------
the responses of survey participants regarding fish consumption
practices, number of years fish have been consumed with this frequency
and preferred fish species is presented in Table 32. In general, the
responses confirm the results of the pilot study in that commercial
fishermen and their families consume large quantities of fish and hava
been doing so for many years. In addition, the preferred fish speciea
are those which have demonstrated elevated levels of PCE contamination
(see Report Z).
These findings support those from the pilot study and suggest
that a population of commercial fishermen may provide valuable in-
formation regarding potential health risks associated with ingestion
of organic contaminants via fish.
It is recommended that the study be expanded to include individuals
employed by commercial fishing companies and crew members of commercial
fishermen as eligible study participants. Finally, the data collected
should be thoroughly analyzed in accordance with proposals in Appendices
IV and V to evaluate potential associations between fish consumption
and health among fishermen and their families.
154
-------
Table 32. Preliminary Results for Great Lakes Study
Distribution of study
the number of fioh
Number of fish
meals consumed
Hever
1-2 times/week
3-5 times/week
6-7 times/week
<1 time/month
1 time/month
2-3 tinea/month
respondents regarding
meals consumed
Number of Study
Respondents %
9 1.5
340 57.9
76 12.9
12 2.0
29 4.9
40 6.8
81 13.8
Distribution of study respondents
regarding the number of years fish
were consumed with this frequency
Number of years
fish consumed with
this frequency
0-5
6-10
11 - 15
16 - 20
21 - 30
31+
31+
Number of study
respondents %
44 7.5
76 12.9
39 6.6
39 6.6
90 15.3
299 50.9
Types of fish most often mentioned
as consumed by study respondents
*Number of study
Fish species respondents
Perch 344
Lake Whitefiah 204
Lake Trout 143
Lake Herring 71 ^
Smelt 59 3
Chub 57
Walleye 45
Catfish 33
Salmon 24
Burbot 22
*Number* do not total 587
because individuals mentioned
more than one species.
-------
APPENDIX I
Rates and Ratios for Thirty-five Cancer Sites
for "Urban and Rural" and "Rural"
Analyses in the Great Lakes Basin
-------
CANCER SITE: LiP
Urban and Rural Counties
|~T Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.
White White Non-white Non-white
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
»TM-k" ti f. Averaged 1st order rate.
RISK raulO Ot ~ : T : , I
Averaged 2nd order rate
"rta" i- i i - -f AveraSed 1st order rate.
KISK ratio or . _ •- :
Averaged 3rd order rate
„„,.,„ . , j. Averaged 2nd order rate
Averaged 3rd order rate
Sum of weighted rates:
"W-lnhtsd risk" ratio of Avera8ed lst order rate-
weignueu risK xauxu o± T^ ^T : , .. .
& Sum of weighted rates
"uMoi -r---i ^-rcir" rar-i- nf Ave^aged 2nd order rate.
weisnted risK. ratio or _ , ?—. . :
5 Sum of weighted rates
"U-lal -tsd ri-k" rati- of Averased 3rd order rate
Weighted risk ratio of s^ Q£ weighted rates .
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
Males Females Hales Females
0.3491
0.3910
0.3744
0.8928
0.9324
1.0443
0.3696
0.9445
1.0579
1.0130
0.0110
0.0141
0.0287
0.7801
a. 0124
0.0000
0.2742
0.3833 0.0452.
0.4878
0.0227 0.1573
0.4846
0.6211
1.2643
X
0.0788
1.7437
0.0064
0.0050
0.0012
1,2800
5.3333
4.1667
0.0034
1.8823
1.4706
0.3529
-------
CANCER SITE: Salivary Glands
C2I Urban and Rural Counties
|~"7 Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"tl-k" ratio of Avera*ed 1st order rate
ZW.SK LCLUJ.O OI7 -"- . - r^^ 7 •
Averaged 2nd order rate
"rijk" rar-r- f Avera*ed lst order rate-
KJ.SK ratio 01 , _ , . :
Averaged 3rd order rate
"riik" rnri- -' Avera8ed 2nd order rate
Risk ratio o. Averaged 3rd Qrder rate.
Sum of weighted rates:
"l^iEhtnd risk" ratio of Avera§ed lsc order rate-
wexgnt-cu Z.LSK rauio OL T ^ ~ .
Sum of weighted rates
"u til r-J I-' k" --Li- -f AveraS£d 2nd order race
wexgciueu rxSK rauu.o or ^ ,. ; ;
6 Sum of weighted rates
"U -Ijl r^af r-f^t" vjri- of AveraSed 3rd order rat£.
we-Lgtiteu risK. ratio 01 „ **, :
0 Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Males Females Males Females
0.4229
0.3397
0.3991
1.2A49
1.0596
0.8512
0.3991
1.0596
0.8512
1.0000
i
t
0.2318
0.1971
0.2210
1.1755
0.0967
0.0186
0.2127
5.1989
I
1.0484 0.4546
0.8918 0.0874
0.2214 0.1574
1.0470
0.8902
0.9982
0.6143
0.1182
1.3513
0.0830
0.6525
0.3129
0.1272
0.2653
2.0860
0.2380
0.3487
2.6155
1.3147
-------
CANCER SITE: Nasopharynx
-Urban and Rural Counties
Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
White White Non-white Non-white
i
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"ri-L." ii- of Avera*ed lst order rate-
IV.LSK ratio OJL ~ _- 0 . . • i
Averaged 2nd order rate !
,,_. ,,. . _f Averaged 1st order rate.
Risk ratio of Avgraged 3rd Qrder tate-
„-..,,, , f Averaged 2nd order rate.
Risk ratio of Averaged 3rd Qrder rat£.
Sum of weighted rates:
...... . ._,„ . , Averaged 1st order rate.
Weighted risk ratio of Sum £f weighted rate£ .
..„ . . . . , „ . , Averaged 2nd order rate.
Weignted rxsk ratio of Sum Qf weighted rates •
.... . . .,",, r Averaged 3rd order rate
Weighted risk ratio of Sum Qf W£ighted rates •
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
Males Females Males Females
0.2744
0.3223
0.2758
0.8514
0.9953
1.1690
0.2802
0.9793
1.1502
0.9843
0.1072
0.0952
0.0829
1.1260
0.2337
0.0639
0.0814
3.6573
1.2931 : 2.8710
i
1.1484 ; 0.7850
i
0.0904 ' 0.1408
1.1858
1.0531
0.9170
X
1.6598
0.453-8
0.5871
0.0387
0.0050
0.0280
7.7400
1.3821
0.1786
0.0312
1.2404
0.1603
0.8974
-------
CANCER SITE: Tongue and
C2J Urban and Rural Counties
Q Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
nr -f_,k.. r^r±Q nf Aver*8ed lst order rate.
K_Lstc ratio or - n ~ . •
Averaged 2nd order rate
"Risk" ratio of Averaged 1st order rate;
Averaged 3rd order rate
„->,,, ... - Averaged 2nd order rate.
Risk ratio or Averaged 3rd Qrder rate-
Sum of weighted rates:
"u ii.1 r J riLA" ratio of AveraS£d 1st order rate.
weignteu risK. ratio or " .
0 Sum of weighted rates
"u i-i i J l^" rar-fo of Avera2fid 2nd order rate
Weighted j isk ratio of gum of weighted rates •
"u ill I r-f-L" rsri- of Avera8ed 3rd order rate.
Weighted risk ratio of Sum Q. WQighte(j rates •
Trends
The averaged first order rate is greater
Chan the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White
Males
3.8479
3.3003
3.0066
1 1660
•I, * J»W V *•/
1.2798
1.0976
3.2539
1.1825
1.0143
0.9240
X
|
White
Females
0.9223
0.8795
0.7318
1 0487
X * V*TW *
1.2603
1.2018
0.7954
1.1595
1.1057
0.9200
X
Non-white
Males
1.1997
1.8376
4.4046
0 6529
\J * \J J imj
0.2724
0.4172
3.0113
0.3984
0.6102
1.4627
Non-white
Females
0.4874
0.5698
1.6990
0 8554
\J * Uj^^r
0.2869
0.3354
1.1753
0.4147
0.4848
1.4456
-------
CANCER SITE: Es°PhaSus
[33 Urban and Rural Counties
j~"T Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
'
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
n»,.f_iii tJ ,e Averaged 1st order rate.
Averaged 2nd order rate
„-,-,,. , _ .f Averaged 1st order rate.
K1SK. ratlO 01 T . _ .,
Averaged 3rd order rate
,,-, . ,it _p Averaged 2nd order rate.
Averaged 3rd order rate'
Sum of weighted rates:
it,. _.f-, , • in _ ^j- r Averaged 1st order rate.
weigriteu risfc ratio ot ^ .. . r , '•
Sum of weighted rates
,,UA, ., , ._,.,i . . _f Averaged 2nd order rate.
Weighted risk ratio of SUQ Q£ weight£;d rates •
"•J Jii i i i • L" i- i i f Averased 3rd order rate
Heiaut?u risic ratio or zr^ ^^ ^ , , .
* Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White
Males
4.2989
3.2781
2.9805
1.3114
1.4423
1.0998
3.3514
1.2827
0.9781
00 ftf\ i
.8893
K
White
Females
0.9902
0.7857
0.7503
1.2603
1.3199
1.0473
0.8152
1.2147
0.9638
Or\f\f\i
.9204
X
Non-white
Males
8.2348
5.0238
5.8199
1 6392
-*• • \J^J y 6m
1 4149
-± • *T J."+ J
0.8632
6.7378
1.2222
0.7456
0.8626
Non-white
Females
1.0127
25.7817
0.6422
O^QOfl
• J?4o
1 5769
J> * — * * U 7
40.1459
1.8978
0.5336
13.5850
0.3384
-------
CANCER SITE: Stomach
DQ Urban and Uural Counties
D Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis»)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Ili-k" ratio of Avera8£d 1st order rate.
Averaged 2nd order rate
HTJ-I i-ii ^. - Averaged 1st order rate.
Averaged 3rd order rate'
»« i n ^j f Averaged 2nd order rate
Averaged 3rd order rate' :
Sum of weighted rates:
Mr, -L - j _, i n ^j c Averaged 1st order rate
Sum of weighted rates
•|T, . . , . , „ . c Averaged 2nd order rate
weignteu nstc rauio or r ^s 7 rr j . .
6 Sum or weighted rates
"u-r-l i l i-i^" i-arl- -f Avera^ed 3rd order rate.
Weighted risk ratio of Sum Q{ we±ghted rates .
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Hales Females Males Females
19.3606
16.6168
7.1509
1.1651
1.3525
1.1609
15.8535
1.2212
1.0481
0.5772
X
9.2993
7.6511
4.8292
1.2154
23.9864
19.9937
7.4590
1.1997
1.2395 1.7868
1.0198 1.4894
7.9769 17.8695
1.1658
0.9592
0.6054
X
.
i
i
1.3423
1.1189
0.4174
X
8.8318
14.5268
14.5268
0.6080
1.3334
2.1933
7.8361
1.1271
1.8538
0.4773
-------
CANCER SITE: Large Intestine
33 Urban and Rural Counties
O Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
1
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
„„_, _, „ ^ _ _c Averaged 1st order rate.
Averaged 2nd order rate
„ ,„ . . - Averaged 1st order rate.
Averaged 3rd order rate*
„... ,,, rl. - Averaged 2nd order rate.
KISK ratio ot , , _ , . :
Averaged 3rd order rate
Sum of weighted rates:
H, , j , ^ j ., -i ii _-_•- -e Averaged 1st order rate.
Weisnt&u risk ratio oz ^ ' » . , , :
0 Sum or weignted rates
"TJ • i , j rl-L" rnri- -f Avera2Qd 2nd order rate.
weisnteu nsK racxo or _ - "; . , .
6 Sum of weighted rates
it,, . , j • , ii ^j c Averaged 3rd order rate
weighted nsic ratxo OE r ~i . . , :
6 Sum of weighted rates
Trends
The averaged first order rate is greater
than che second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Males Females Males Females
16.5203
16.0048
15.8063
1.0322
1.0452
1.0125
16.0110
1.0318
0.9996
0.9872
X
16.5682
16.5117
16.9205
1.0034
9.7122
11.2259
11.0775
0.8652
0.9792 0.8767
0.9758 1.0134
16.7893 10.5440
0.9868
0.9835
1.0078
0.9211
1.0647
1.0506
15.9118
9.1557
15.3798
1.7379
1.0346
0.5953
15.3129
1.0391
0.5979
1.0044
-------
CANCER SITE:
Rectum
E2J Urban a-nd Rural Counties
Q Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Risk" ratio of f^aged 1st order rate.
Averaged 2nd order rate
"•04 -I."
Risk" ratio of
Averaged 1st order rate.
Averaged 3rd order rate'
„_, , „ . , Averaged 2nd order rate
Risk ratio of &i ••; , . •: T~:
Averaged 3rd order rate
Sum of weighted rates-:
"Weighted risk" ratio of Averaged 1st order rate;
Sum of weighted rates
"Weighted risk" ratio of Averaged 2nd order rate
° Sum of weighted rates
"Weighted risk" ratio of Averaged 3rd order rate;
Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Males Females Kales Females
7.7925
7.8417
7.2972
0.9937
1.0679
1.0746
7.4809
1.0416
4.8053
5.0198
4.7094
0.9573
4.7652
1.0084
1.0482 | 1.0534
0.9754
0.9883
2.2489
9.1949
5.0750
0.2446
1.0204 ! 0.4431
1.0659 1.8117
4.1549
0.5413
2.2130
1.2214
3.1272
1.1791
3.1700
1.8191
0.9865
0.5423
3.0892
1.0123
0.3817
1.0262
-------
CANCER SITE:
Liver
rT[ Urban and Rural Counties
CJ Rural Counties (counties having urban
centers with. 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Risk" ratio of Averted 1st order rate;
Averaged 2nd order rate
" -"
of
" "
Risk" ratio of
Averaged 1st order rate.
Averaged 3rd order rate'
Averaged 2nd order rate.
Averaged 3rd order rate'
Sum of weighted rates:
"Weiahted risk" ratio of Avera8ed 1st order rate,
weighted risk ratio or Sum Q£ weighted rates
"Weighted .risk"
of
"Weighted risk" ratio of
Averaged 2nd order rate
Sum of weighted rates
Averaged 3rd order rate.
Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White
Males
5.1953
4.6395
4.8028
1.1198
1.0817
0.9660
4.8874
1.0630
0.9493
0.9827
White
Females
6.5479
6.0074
6.2763
1.0900
1.0433
0.9672
6.3187
1.0363
0.9335
0.9933
Non-whi te
Males
6.6440
5.8985
5.6348
1.1264
1.1791
1.0468
6.0473
1.0987
0.9754
0.9318
Non-white
Females
5.9107
5.0079
4.8715
1.1803
1.2133
1.0280
5.2839
1.1186
0.9478
0.9220
-------
CANCER SITE:
Pancreas
D8 Urban and Sural Counties
Q Rural Counties (counties having urban
Centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"ttl-k" rati- tf AveraSed la* order rate.
AxSix zawZO Ox . . ^ - , •
Averaged 2nd order rate
"ri£k" nMti- f Averaged 1st order rate.
Averaged 3rd order rate
•Ti-L" L tf £ Averaged 2nd order rate.
Risk ratio of Avaraged 3rd order rate-
Sum of weighted rates:
"u-ltthr-l --1-k" raiij -f Avera8ed lst order rate
weignueu riLSK racxo 01 _ - ; , , t
° Sum of weighted rates
"U \i\ L^J i-I V" raH of Avera2fid 2r>d order rate-
weignteu rxstc ratio on ^ - ^ t T ^i
6 Sum of weighted rates
"U --tahr 1 ri -1" rari^ cf AveraSed 3rd order rate
weiKuuPa risK ratio or „ "_ , , :
0 Sum of weightea rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Males Females Males Females
9.2717
8.9293
8.4645
1.0383
1.0954
1.0549
8.7206
1.0632
1.0239
0.9706
X
6.0605
5.5112
5.4478
1.0997
3.8775
5.7570
7.7707
0.6735
1.1125 0.4990
1.0116 0.7409
6.8162 5.6109
0.8891
0.8085
0.7992
X
t
0.6911
1.0260
1.3849
10.7675
2.9088
3.9151
3.7017
2,7502
0.7430
8.8068
1.1613
0.3303
0.4445
-------
CANCER SITE:
Nose and Middle Ear
23 Urban and Rural Counties
| i Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"ni-k" ratio of Avera§ed lst order rate-
Averaged 2nd order rate
"Risk" ratio of Averaged 1st order rate
Averaged 3rd order rate
„„.-,.„ . ±. -f Averaged 2nd order rate
EtiSr<. ratio or . .. « . , .
Averaged 3rd order rate
Sum of weighted rates:
.IT- in, , , Ll-,« rnLi -f Avera8ed 1st order rate.
6 Sum of weighted rates
"u ijli 1 tl" rir-t- -f AveraSed 2nd order rate.
weigtiteu risK ratio 01 r ™-r ; ,
6 Sum of weighted rates
„,, . . . , „ . ,. Averaged 3rd order rate
weignteu nsic ratio or _ , , . •; .
Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Males Females Males Females
0.3829
0.5282
0.4440
0.7249
0.8624
1.1896
0.4369
0.8764
1.2090
1.0162
0.2280
0.1785
0.2280
0.3576
0.4094
0.2673
r
S
1.2273 0.8735
1.0000 : 1.3378
0.7829 1.5316
0.2051
0.1116
0.8703
1.1116
0.3100
•
1.1535
1.3206
0.8622
9.4382
0.2148
0.1796
2.0400
2.4399
1.1960
0.2823
1.5522
0.7609
0.6362
-------
CANCER SITE: Larynx
Q£j Urban and Rural Counties
l~"t Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Risk" ratio of Averaged 1st order rate.
Averaged 2nd order rate
Risk" ratio of
Averaged 1st order rate.
Averaged 3rd order rate'
„,,.,„ _-_.,„ . Averaged 2nd order rate.
Risk ratio of Avera|ed 3rd order rate=
Sum of weighted rates:
,„. . , , . . „ _. - Averaged 1st order rate
Weighted risk ratio of — °~^ r—r—3 :
* Sum of weighted rates
II,. . , , . , ii j r Averaged 2nd order rate
Weighted risk ratio of •— ^ 7-r—3 :
6 Sum of weighted rates
«„ . , j * , it _., £ Averaged 3rd order rate
"Weighted risk" ratio of Sum gf weighted rates :
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Males Females Hales Females
2.3800
1.8688
1.7748
1.2736
1.3401
! 1.0530
1.9407
0.1546
0.2131
0.1774
0.9516
0.8715
1.2012
0.1752
1.2264 j 0.8824
0.9629 ! 1.2163
0.9145
1.0126
1.8640
1.9599
1.6403
0.9516
1.1370
1.1948
1.7447
1.0689
1.1233
0.9402
0.2067
0.1952
0.9280
1.0589
0.2227
0.2103
0.6136
0.3369
0.3181
1.5124
-------
CANCER
SITE' Trachea, Bronchus and Lung
[33 Urban and Rural Counties
d[ Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
White White Non-white Non-white
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"r-r^k" r^m- of Avera8ed lst order rate
Averaged 2nd order rate
"cr i" -i i - -f Avera§ed lst order rate-
Risk ratio of Averaged 3rd order rate-
"ri-k" AI i f Avera^ed 2nd order ra*e
Risk ratio of Averaged 3rd order rate-
Sum of weighted rates:
"U f-1 I 1 -isk" r-Ll-ia jf Avera!?ed lst order rate -
weigntect nstc ratio or ~_
e Sum of weighted rates
"U f 1 . 1 -Uk" rari- rf Avera8ed 2nd order rate-
Weighted risk ratio or Sum Q£ we±ghted rates -
'"J i ii J l-k" -iti- -f Averased 3rd order rate.
e Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
Males Females Males Females
32.9002
30.6619
28.5293
1.0730
1.1532
1.0747
29.8764
1.1012
1.0263
0.9549
X
i
5.1322
5.1388
5.1546
0.9987
29.6472
21.9218
27.7351
1.3524
0.9956 1.0689
0.9969 0.7904
5.1473 28.2114
j
0.9971
0.9983
1.0014
X
1.0509
0.7771
0.9831
5.7151
3.2147
5.0540
1.7778
1.1308
0.6361
5.2313
1.0925
0.6145
0.9661
-------
CANCER SITE: Breast
ESI Urban and Rural Counties
Q Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
»~, rV, t, f Averaged 1st order rate.
IU.SK ratio oz , — „ . , :
Averaged 2nd order rate
"rj-k" I--I.T f Avera*ed lst order rate
Averaged 3rd order rate
,,r, ,,, r .. f Averaged 2nd order rate.
Risk ratio of Avfiraged 3rd order rafce.
Sum of weighted rates:
"u -TCI -t^J t-lak" rjti- af Avera8ed 1st order rate.
weiCnueu rxsK racjuo or _ - . , . .
ft Sum of weighted rates
"u iui i^l il L11 LI -f Avera2£d 2nd order rate-
& Sum of weighted rates
"U ici t-j rick" i-iti- n- Averaged 3rd order rate
Weighted risk ratio o* Sum flf weighted rates •
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater Chan the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Males Females Males Females
0.2661
0.1841
0.2429
1.4554
1.0955
0.7579
0.2428
1.0960
0.7582
1.0004
25.5191
25.0840
24.1887
1.0173
0.2154
1.7380
0.2499
0.1239
j
1.0550 0.8616
1.0400 6.9520
24.6189 0.3082
1.0366
1.0189
0.9825
X
t
0.6989
5.6392
0.8108
23.0836
27.6612
17.5420
0.8345
1.3159
1.5769
20.1563
1.1452
1.3723
0.8703
-------
CANCER SITE: Cervix Uter±
CXI Urban and Rural Counties
£2 Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"list" ratio of Avera§ed lst order rate-
AlSK ratZLw Oi : r : _ «
Averaged 2nd order rate
»-. ,« -., f Averaged 1st order rate.
Rxsk ratio of Averaged 3rd Qrder rat£-
„-, ,„ - ,-, - -f Averaged 2nd order rate.
Risk ratio of Averaged 3rd Qrder rate-
Sum of weighted rates:
„., , , , , , ii . /- Averaged 1st order rate
5 Sum of weighted rates
ltr, , . , . , ii . ,• Averaged 2nd order rate
Weignted risk ratio of Sum Q£ weightei rates .
.,._... , . ., „ . . . _f Averaged 3rd order rate
Weighted risk ratio of Sum Qf weighted races -
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White
Hales
White
Females
8.1263
7.8110
7.9403
1.0404
1.0235
0.9837
7.9749
1.0190
0.9794
0.9957
Non-white
Males
Non-white
Females
15.3534
23.0495
9.4707
0.6661
1.6211
2.4338
12.3712
1.2411
1.8631
0.7655
-------
CANCER SITE:
Corpus Uteri
CSJ Urban and Rural Counties
|~~? Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
i.r.i.Lii , e Averaged 1st order rate.
Averaged 2nd order rate
"Itifik" ratio of AveraS?ed 1st order rate.
Averaged 3rd order rate
"Risk" ratio of Averaged 2nd order rate; |
Averaged 3rd order rate
Sum of weighted rates:
"XJpichtPd risk" ratio of Avera8ed lst order rate-
Sum of weighted rates
"Up-fahtpJ ri£k" ritio of Averased 2nd order rate
6 Sum of weighted rates
"IkiahttJ riEt" ratio of Avera^ed 3rd order rate
WaigLttJ iisfc latio of gum o£ weighted rates •
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Hales Females Males Females
i
6.5222
6.8524
6.8903
0.9519
0.9466
0.9945
6.7925
0.9603
1.0088
1.0144
X
11.5115
10.6140
9.0597
1.0846
1.2706
1.1716
10.0872
1.1412
1.0522
0.8981
-------
CANCER SITE: Ovary and Fallopian tube
Urban and Rural Counties
Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Risk" ratio of Averaged 1st order rate;
Averaged 2nd order rate
„_. . „ J , Averaged 1st order rate
Risk ratio of ° , , . :
Averaged 3rd order rate
„_. . „ . - Averaged 2nd order rate
RISK ratio or Avera^ed 3rci order rate =
Sum of weighted rates:
"Weighted risk" ratio of
Averaged 1st order rate f
Sum of weighted rates
"Weighted risk" ratio of Averaged 2nd order rate;
Sum of weighted rates
IIT, . , , - , ii . r Averaged 3rd order rate
Weighted risk ratio of — °-z . . :
0 Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White
Males
White
Females
Non-white
Males
Non-white
Females
8.3Q35
8.6594
8.4251
0.9589
0.9856
1.0278
8.4175
0.9865
1.0287
1.0009
4.4672
5.9559
6.3573
0.7501
0.7027
0.9368
5.6004
0.7977
1.0635
1.1351
-------
CANCER SITE:
Urban and Rural Counties
Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Risk" ratio of Averaged 1st order rate;
Averaged 2nd order rate
"Risk" ratio of
"Risk" ratio of
Averaged 1st order rate.
Averaged 3rd order rate*
Averaged 2nd order rate.
Averaged 3rd order rate'
Sum of weighted rates:
... . , , . ,„ __. , Averaged 1st order rate
Weighted risk ratio of — «-= 7-7-—-: :
° Ctvrn s\4- r.T/a-i *»ri *-AJ-* *•«> 4- A»
Sum of weighted rates
„,, . , _. . , it . c Averaged 2nd order rate
Weighted risk ratio of — °-: -r—r—7 :
e Sum of weighted rates
"Weighted risk" ratio of Averaged 3rd order rate;
6 Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Males Females Males Females
18.2989
18.5688
18.3151
0.9855
0.9991
1.0138
18.3369
0.9979
1.0126 I
0.9988
17.2375
15.9260
21.8424
1.0823
0.7892
0.7291
19.7328
0.8735
0.8071
1.1069
-------
CANCER SITE:
Testis
33 Urban and Rural Counties
V~ Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Risk" ratio of Averaged 1st order rate
"Risk" ratio of
Averaged 2nd order rate'
Averaged 1st order rate.
Averaged 3rd order rate'
"iMefc'i ratio of Averaged_. 2nd order rate
Risk ratio of Averaged 3rd Qrder
Sum of weighted rates:
„„ . . . . , „ . , Averaged 1st order rate
'Weighted risk" ratio of '«.•"- ;-.- . r :
0 Sum of weighted rates
it,, . . , _, , 11 . c Averaged 2nd order rate
'Weignted .risk" ratio of Sum gf weighted rates •
"Weighted risk" ratio of Averaged 3rd order rate;
0 Sum of weighted rates
Trends
The averaged first order rate is greater
than che second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the .first order rate.
White
Males
0.9401
0.8991
0.9189
1.0456
1.0231
0.9784
0.9223
1.0193
0.9748
0.9963
White
Females
Non-white
Males
Non-white
Females
0.0831
0.0567
0.1875
j 1.4656
I
! 0.4432
t
• 0.3024
i
I
I
: 0.1399
0.5940
0.4053
1.3402
-------
CANCER SITE: Kidney
C2I Urban and Rural Counties
C~T Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Risk" ratio of Averaged 1st order rate.
Averaged 2nd order rate
Risk" ratio of
Risk" ratio of
Averaged 1st order rate.
Averaged 3rd order rate'
Averaged 2nd order rate.
Averaged 3rd order race'
Sum of weighted rates:
"Weighted risk" ratio of
Averaged 1st order rate
Sum of weighted rates
„,, . , , , , „ . , Averaged 2nd order rate
"Weighted risk" ratio of Sum £f weighte,l rates
„,, . , , . , „ . - Averaged 3rd order rate
Weighted risk ratio of — °-^ . , • •• j,
6 Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White
Males
4.1238
3.6236
3.7703
1.1380
1.0938
0.9611
3.8465
1
1.0721
0.9420
0.9802
i
r
i
White
Females
2.1792
2.0213
2.0801
1.0781
1.0477
0.9718
2.0995
1.0380
0.9627
0.9903
:
|
Non-white
Hales
2.0134
2.3809
2.2870
0.8456
0.8804
1.0411
2.1832
0.9222
1.0905
1.0475
Non- white
Females
0.5323
0.8328
1.7303
0.6392
0.3076
0.4813
1.2221
0.4356
0.6814
1.4158
-------
CANCER SITE: Bladder
Urban and Rural Counties
Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Risk" ratio of Averaged 1st order rate.
Averaged 2nd order rate
"Risk" ratio of Averaged 1st order rate;
Averaged 3rd order rate
"Risk" ratio of Averaged 2nd order rate;
Averaged 3rd oraer rate
Sum of weighted rates:
"Weighted risk" ratio of Averaged 1st order rate;
0 Sum of weighted rates
n.r ., v j j i n _, c Averaged 2nd order rate
Weighted risk ratio of —r °-= j—r—j :
* Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Males Females Males Females
6.8957
7.3321
5.8743
0.9405
1.1739
1.2482
6.2875
1.0967
2.5049
2.2967
2.1449
1.0708
2.2523
1.1122
1.1661 j 1.0198
0.9343
0.9524
4.8102
5.2621
2.6309
1.0906 0.9141
1.1678 • 1.82S3
2.0001
3.6211
1.3284
1.4532
0.7265
1.8569
2.0817
2.6733
0.8920
0.6946
0.7787
2.3279
0.7977
0.8942
1.1484
-------
CANCER SITE: Skln
Urban and Rural Counties
Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Risk" ratio of Averted 1st order rate.
Averaged 2nd order rate
White White Non-white Non-white
Males Females Males Females
Risk" ratio of
"t>4 -I."
Risk" ratio of
Averaged 1st order rate.
Averaged 3rd order rate'
Averaged 2nd order rate.
Averaged 3rd order rate'
Sum of weighted rates:
"Weighted risk" ratio of Averaged 1st order rate;
6 Sum of weighted rates
urr j , _, j , ti ., - Averaged 2nd order rate
"Weighted risk" ratio or - *£ veighted rates !
„,, , , , . , „ . , Averaged 3rd order rate
"Weighted risk" ratio of Sum *£ weighted rates :
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
-than the second order rate which in turn
is greater than the first order rate.
0.9569
1.0196
1.2218
0.9385
0.7832
0.8345
1.1327
0.8448
0.7774
0.9110
0.9804
0.8533
0.7929
0.9292
0.8436
0.9001 j 0.9886
1.0787
1.0639
0.0464
0.0823
0.0702
0.5638
0.6610
1.1724
0.9215 0.0613
0.7569
1.3426
1.1452
0.0103
0.3984
0.3027
0.0258
0.0340
1.3161
0.1926
0.0534
2.0685
1.5716
-------
CANCER SITE:
Other skin
HT] Urban and Rural Counties
1~~ Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
npj v» -t f Averaged 1st order rate_
Averaged 2nd order rate'
"m" ran- -r Averased 1st order rate
Risk ratio of Averaged 3rd Qrder ratg.
"Y- -i" rjftl -r Averaged 2nd order rate
Risk ratio of Averaged 3rd order rate-
Sum of weighted rates:
•i*. „ , _ j j t 11 ^ * Averaged 1st order rate.
Wei2hteu risk ratio or _ _ :™~: t
0 Sum of weighted rates
,,„ • ,. , ,-,-,.. -ir,- -* Averaged 2nd order rate.
weientea nsK ratio 01 • , .
0 Sum of weighted rates
"'.J--t3-r-3 ri^-" -ar-I- -f AveraSed 3rd order rate_
-?eignted rxsk ratio of s^ Qf weighted rates •
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Hales Females Males Females
1.2889
1.3164
1.4384
0.9791
0.8961
0.9152
1.3873
0.9291
0.9489
1.0368
0.5505
0.6241
0.6423
0.8821
2.8011
1.0289
0.4887
2.7224
0.8571 5.7317
0.9717 2.1054
0.6170 1.4306
0.8922
1.0115
1.0410
X
1.9580
0.7192
0.3416
0.1210
0.1402
0.4418
0.8630
0.2739
0.3173
0.3030
0 3993
0,4627
1.4581
-------
CAKCER SITE: Eye
DQ Urban and Rural Counties
O Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Hink" ratio af Averaf?ed 1st order rate-
Averaged 2nd order rate
MTM-I," r,t-t- nf Avera8ed 1st order rate.
XVJ.SK racio 01 . , „ . , . .
Averaged 3rd order rate
"Risk" ratio of AveraBed 2nd order rate.
Averaged 3rd order rate'
Sum of weighted rates:
"Ifclcht-j ri.sk" ratij jf AveraSed 1st order rate.
wexKitLeu rzsK racxo or _. .. , . . .
Sum of weighted rates
"U^ialit^J --i-k" IAII af Avera«ed 2nd order rate
wexgnceu :.j.sic rauxo or r ^ ; . , .
Sum or weighted rates
urT-^-v^ j MJ -i ii -^-.- _£ Averaged 3rd order rate.
Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White
Males
0.1919
0.2052
0.2380
0 93S2
V • .S -J J 4*
0.8063
0.8622
0.2227
0.8617
0.9214
1f\f ft -T
.0687
X
White
Females
0.1871
0.1490
0.1946
1 ?S57
x . t-jj i
0 9615
\J • J \J ^-J
0 7657
\J • i \JJ 1
0.1881
0.9947
0.7921
1.0346
Non-white
Males
0.0089
0.0345
0.0100
A 2?51
\j . j^jjj*
n 8800
\J * (J*J \J\J
•5 4500
•J • ** J W \J
0.0107
0.8318
3.2243
0.9346
Non-white
Females
0.1484
0.0000
0.1363
i nfifis
J. • UUU w
0.1350
1.0993
1.0096
-------
CANCER SITE- Brain and Nervous System
TTf Urban and Rural Counties
l~j Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"TM-I-" - • f Averaged 1st order rate.
Averaged 2nd order rate'
,,r. , „ . - Averaged 1st order rate.
K.ISK rauio or . ^ « T _
Averaged 3rd order rate
„ „ - Averaged 2nd order rate
Risk ratio or Averaged 3rd order race-
Sum of weighted rates:
"17 ' 1 I 1 L-k" vjM- -f AVera8£d lst °rder ratS-
Weighted risk ratio of gum Qf weighted rates •
,,„ . . . .,",,. -f Averaged 2nd order rate.
Weighted risk ratio of Sum Qf weighted rates •
it,, - , , , M j <• Averaged 3rd order rate
Weighted risk ratio of gum Q£ weighted rates •
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Males Females Males Females
4.1597
4.3553
3.9854
0.9551
1.0437
1.0928
4.0683
1.0225
1.0705
2.7555
2.6536
2.6692
2.0993
1.1240
1.6092
1.0384 1.S677
1.0323 1.3046
0.9942 0.0771
2.6897 1.7255
1.0245
0.9866
f
0.9796
I
!
0.9924
1.2166
0.6514
0.9326
4.5234
0.5267
1.0448
8.5882
4.3294
0.5041
2.3823
1.8987
0.2211
0.4386
-------
CANCER SITE: Thyroid gland
CXI Urban and Rural Counties
1• Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"ttiak" ratio o* AveraSed lst order rate-
Averaged 2nd order rate
ti-nj i ii . j f Averaged 1st order rate
K.J.SK ratio 01 — — : — _ , . :
Averaged 3rd order rate
"III ak" r^M- -f Averased 2nd order rate
Averaged 3rd order rate
Sum of weighted rates:
n,, j , n j . Averaged 1st order rate
Weiiznted risR ratio or _ ^ ; — : _ :
Sum of weighted rates
,lr, j ,t ^j r Averaged 2nd order rate
* e Sum of weighted rates
"U^'-Ii ' i-t" Ya.i±- -• Averaaed 3rd order rate.
0 Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Males Females Hales Females
0.4902
0.5173
0.3631
0.9476
1.3500
1.4247
0.4117
1.1907
1.2565
0.8819
0.7535
0.7189
0.7252
1.0481
0.1848
0.0345
0.3715
5.3565
1.0390 0.4974
0.9913 ;' 0.0929
0.7318 0.2813
1.0296
0.9824
0.9910
0.6569
0.1226
1.3206
0.2463
0.1850
0.3964
1.3313
0.6213
0.4667
0.3284
0.7500
0.5633
1.2086
-------
CANCER SITE:
Endocrine organs
Urban and Rural Counties
Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
1
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
••TH-VI' --,,-.-, n? Averaged 1st order rate.
Averaged 2nd order rate
"ri-v" AI • -f Avera^ed 1st order rate.
tvisK racio or . i ~ j .
Averaged 3rd order rate
"•«.; tit j f Averaged 2nd order rate
Averaged 3rd order rate'
Sum of weighted rates:
"U^-td l ri-k" nr-I- -f Avera8ed 1st order rate-
weigncea ristc ratio or „ , . , .
* Sum of weighted rates
-TT r-, , , , .i. .,.-.,. -f Averaged 2nd order rate
weigntea risK. ratio or _ , - , . :
Sum of weighted rates
nr» j. -, , . ,n _.^j f Averaged 3rd order rate
* Sum 01 weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Males Females Males Females
0.2856
0.2768
0.2667
1.0318
1.0709
1.0379
0.2726
1.0477
1.0154
0.9784
X
0.2654
0.1577
0.1948
0.07~75
0.0366
0.0515
1.6829 | 2.1175
1.3624 1.5048
0.8095 0.7107
0.2091 0.0611
>
1.2692
0.7542
0.9316
1.2684
0.5990
0.8429
0.0213
0.0506
0.0473
0.4209
0.4503
1.0698
0.0373
0.5710
1.3566
1.2681
-------
CANCER SITE: Bone
Urban and Rural Counties
Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis. )
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"nni-" rat-la of Avera2ed I" order rate:
.tvisK racjio or , — r — ^ * , :
Averaged 2nd order rate
"in it" rat I -f AvaraSed 1st order race.
K.ISIC ratio or . . . „ . . :
Averaged 3rd order rate
,,ric,r, , _ _f Averaged 2nd order rate
. K.XSK ratio or . - . _ . , :
'-eraged 3rd order rate
Sum of weighted rates:
..,T-i!rhrpd riEk.< tl f Averaged 1st order rate.
weixnueu J^ISK ratio or _ ^ . . , :
Sum of weighted rates
"U -lal rpj rick" r^ti- n.f Avera^ed 2nd order rate
weiftnueu ristc rauxo or ^ ^ t , . :
6 Sum of weighted rates
"u^-taht-arf r-icir" 1-^^^n nf Averaged 3rd order rate.
weisnteu nsK. ratio or — - _ . , , :
& Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
Ts greater than the first order rate.
White White Non-white Non-white
Males Females Males Females
1.3432
1.4098
1.3313
0.9528
1.0089
1.0590
1.3424
1.0006
1.0502
0.9917
.
0.8044
0.8390
0.8824
0.9588
0.6826
6.5224
1.2662
0.1046
0.9116 0.5391
0.9508 5.1512
0.8581 . 1.2890
0.9374
0.9777
1.0283
X
0.5296
5.0600
0.9823
0.2112
0.2964
0.9172
0.7125
0.2303
0.3232
0.6137
0.3441
0.4830
1.4945
-------
CANCER SITE:
Connective tissue
CZJ Urban and Rural Counties
\~~~2 Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Ri-k" rania of Avera8ed lst order rate-
Averaged 2nd order rate
'"M -L-» ran - -f AveraSed 1st order rate.
Averaged 3rd order rate
•Ti-i-" --jit- -if AvaraSed 2nd order rate.
K.J.SK ratio 01 A j •> j j ^ •
Averaged 3rd order rate
Sum of weighted rs.tes:
"B I I r J i-I-t" r-ri- -f AveraSed lst order rate-
6 Sum of weighted rates
"U • it J i-L" I--MJ -f Avera«ed 2nd order rate.
Weighted i isk r^tio of gum Q£ weighted rates •
'"J I i r 1 rl-k" rail- f Averased 3rd order rate.
& Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
White White Non-white Non-white
Males Females Hales Females
0.5325
0.6488
0.6220
0.8207
0.8561
1.0431
0.6017
0.8850
1.0783
1.0337
0.3985
0.5463
0.4597
0.3774
0.7038
0.3767
0.7294 • 0.5362
j
0.8660 ; 1.0019
1.1884 ; 1.8683
0..4528
0.8801
1.2065
1.0152
0.3928
0.'9608
1,7917
0.9590
0.3912
0.2277
0.6883
1.7180
0.5684
0.3308
0.1530
2.5569
1.4482
4.4987
The averaged third order rate is greater
than the second order rate which in turn
is greater than the .first order rate.
-------
:ANCER SITE:
CQ Urban and Rural Counties
If Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
!Risk" ratio of
Averaged 1st order rate.
Averaged 2nd order rate'
„.,..„ „„.,„ ,: Averaged 1st order rate.
Risk ratio of B—•:—T— 1 :
Averaged 3rd order rate
"Risk" ratio of
Averaged 2nd order rate,
Averaged 3rd order rate'
Sum of weighted rates:
"Weighted risk" ratio of Averaged 1st order rate;
Sum of weighted rates
"Weighted risk" ratio of Averted 2nd order rate;
6 Sum of weighted rates
"Weighted risk" ratio of Averaged 3rd order rate:
0 Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
£han the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Males Females Males Females
2.2210
2.1782
2.2463
1.0196
0.9887
0.9697
2.2328
1.0060
1.2231
1.2260
1.3215
1.0202
3.8701
0.7334
0.9976 ! 0.2636
0.9255
0.9277
1.3910
5.2769
1.2867 : 0.9986
0.9947 ! 0.9506
0.9755 I 0.9528
1.0270
1.0216
3.8755
0.7344
0.5785
0.2143
0.7158
2.6995
0.8082
0.2994
0.6399
0.9040
0.3349
1.1186
-------
CANCER SITE: Lymphosarcoma
Si Urban and Rural Counties
133 Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
;'^i-k" ratio of Avera§ed lst order rate-
Averaged 2nd order rate'
„„ , , it . - Averaged 1st order rate
iCisK ratio or , , . i . .
Averaged 3rd -order rate
White White Non-white Non-white
Males Females Males Females
4.9200
4.7684
4.6771
1.0318
1.0519
\
..„.,.. . _ Averaged 2nd order rate
Risk ratio of Averaged 3rd Qrder rate-
Sura of weighted rates:
,,„ t .,.. , ,-,„ r,ri, _f Averaged 1st order rate.
weigntea riSK. ratio or „ - . , ,
s Sum of weighted rates
..„.,, i -i.. - tl- f Averaged 2nd order rate.
Weighted risk ratio of Sua of weighted ratfis -
„., . , . , ,„ .. ±. - Averaged 3rd order rate.
weigntea riSK ratio or. _ - . , ._ ,
e Sum ot weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate-
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
1.0195
4.7492
1.0360
1.0040
0.9848
X
3.2943
3.2386
3.2264
1.0172
11.3366
3.4169
1.4085
3.3178
1.0210 j 8.0487
•
1.0038 '; 2.4259
1
3.2350 | 5.4374
1.0152
0.9980
0.9943
X
2.0849
0.6284.
0.2590
1.1738
1.4381
1.2652
0.8162
0.9278
1.1367
1.2371
0.9488
1.16-25
1.0227
-------
CANCER SITE: Malignant Melanoma
CS Urban and Rural Counties
|~~T Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Risk" ratio of- Averaged 1st order rate.
Averaged 2nd order rate
"Risk" ratio of Averted 1st order rate;
Averaged 3rd order rate
"Risk" ratio of Averaged 2nd order rate;
Averaged 3rd order rate
Sum of weighted rates:
"Weighted risk" ratio of Averaged 1st order rate;
Sum of weighted rates
"Weighted risk" ratio of Averaged 2nd order rate;
Sum of weighted rates
"Weighted risk" ratio of Averaged 3rd order ratE;
Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
.than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Males Females Males Females
1.6870
1.7778
1.7174
0.9489
0.9822
1.0352
1.7157
1.3077
1.1799
1.1641
1.1234
1.0136
1.2024
0.9833 i 1.0876
1.0362 j 0.9813
1.0010
0.9681
1.3904
1.8585
1.5248
1.1083 0.7481
0.9119
1.2188
1.4877
0.9346
1.2492
1.0249
1.8647
0.7126
1.9614
2.6169
0.9507
0.3633
1.8684
0.9980
0.3814
1.0498
-------
CANCER SITE:
Leukemia
223 Urban and Rural Counties
[~ Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Iti-lt" ratio of Avera8ed lst order rate-
Averaged 2nd order rate
HTM-I-H ~=rir< *f Averaged 1st order rate.
K.ISK ratio or ~ : ^r : r .
Averaged 3rd order rate
itn_r i H _-^j ^ Averaged 2nd order rate
&isK ratio or *r : ^ ~. . \
Averaged 3rd order rate
Sum of weighted rates:
,,,, _, ^ _, _^ , ,, tj = Averaged 1st order rate
Weighted rxSK ratio or ^ ^~r : — , , ' :
Sum or weighted rates
'"j -slr^J r'-k" Art- -f Avera§ed 2^d order rate.
neigntea risK ratio or _ - , , , :
Sum or weighted rates
•"J-lalt-d rl k" M f AveraS£d 3rd order rate.
weignupu risK racio or. r ^ ~ T ^ .
Sum of weighted rates
Treads
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the .first order rate.
White
Males
8.5566
8.6695
8.9543
0.9870
0.9556
0.9682
8.8224
0.9699
0.9827
1.0149
X
t
White
Females
5.5781
5.6443
5.8889
0.9883
0.9472
0.9585
5.7848
0.9643
0.9757
.1.0180
K
Non-white
Males
7.1985
8.4316
5.5888
.
0.8527
1.2225
1.4319
6.3637
1.1312
1.3249
0.3782
Non-white
Females
2.7852
2.0768
2.9523
1.3411
0.9434
0.7034
2.8482
0.9779
0.7292
1.0365
-------
CANCER
SITE- ICD'S not listed
P(J Urban and Rural Counties
O Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis . )
White White Non-white Non-white
Hales Females Hales Females
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Hl-k" ratio of Averaged 1st .order rate.
Averaged 2nd order rate
n«.t ,n ^., f Averaged 1st order rate
KISK ratio Ot . — : — r — ; I
Averaged 3rd order rate
\\-r-j in j f Averaged 2nd order rate
Averaged 3rd order rate'
Sum of weighted rates:
i.,,plEhrpJ ri-v.. r r± f Averaged 1st order rate.
weignr.eu nsK rauio 01 -, ,. : , . r
• Sum of weighted rates
"U-ial -r-J ii£k" -nti- Qf Avera8ed 2nd order rate-
weixnceu i ISK ratio ot « ^ .. . , :
Sum or weighted rates
"U -'2l r-J rifik" rati- -f Avera8ed 3rd order rate
6 Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
:than the second order rate which in turn
is greater than the first order rate.
9.8353
9.5067
9.6726
1.0346
1.0168
0^9828
9.6976
1.0142
0.9803
0.9974
9.4042
9.1145
9.4163
10.1293
5.0741
13.7322
i
1.0318
1.9963
0.9981 0.7376
0.9679 0.3695
0.3828 11.8870
1.0023
0.9714
1.0036
i
i
i
i
1
I
i .
1
0.8521
0.4269
1.1552
12.4898
7.6560
1.3590
1.6314
1.3345
0.8180
10.5081
1.1886
0.1286
0.8906
-------
CANCER SITE*
neoplasms
Urban and Rural Counties
\~~r Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Risk" ratio of- Averaged 1st order rate;
Averaged 2nd order rate•
White
Males
Risk" ratio of
Averaged 1st order rate.
Averaged 3rd order rate'
„„. . ,, . , Averaged 2nd order rate
Risk ratio of Averajed ,rd order race:
Sum of weighted rates:
•I,. . . . j , ii • s Averaged 1st order rate
"Weighted risk ratio of — *-: . ... :
sum of weighted rates
,,., . . , . , „ . e Averaged 2nd order rate
"Weighted risk ratio of — °-^ . . ; :
0 Sum of weighted rates
M,. , , • , .1 * * Averaged 3rd order rate
"Weighted risk ratio of.-r °~z :~T—2 ;
0 Sum of weighted rates
Trends
The averaged first order r.ate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the .first order rate.
White
Females
Non-white
Males
Non-white
Females
169. 290.
161.980
154.326
1.0494
1.1015
1.0496
159.155
1.0681
1.0177
0.9697
X
131.514
127.931
127.666
1.0280
1.0301
1.0021
128.677
1.0220
0.9942
0.9921
X
149.829
140.887
127.847
1.0635
1.0869
1.0220
142.739
1.0497
0.9870
0.9657
133.907
148.075
109.866
0.9043
1.2188
1.3478
120.956
1.1071
1.2242
0.9083
-------
CANCER SITE:
. Lip
Urban and Rural Counties
133 Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from anal7sis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Risk" ratio of Averaged 1st order rate.
Averaged 2nd order rate
Risk" ratio of
Averaged 1st order rate.
Averageu 3rd order rate'
"D-! i," •< - Averaged 2nd order rate
Ris* ratio or Averaged 3rd Qrder =
Sum of weighted rates:
"Weighted risk" ratio of Averaged 1st order rate;
& Sum of weighted rates
.... , , ... it . - Averaged 2nd order rate
Weighted risk ratio or — =-? r-r—~, '
6 Sum or weighted rates
,,., . , , . , H . r Averaged 3rd order rate
Weighted risk ratio of — °-^ — j :
° Sum of weignted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
jihan the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Males Females Males Females
0.3513
0.3959
0.3771
0.877
0.932
1.050
0.3765
0.0095
0.0144
0.0292
0.660
0.325
0.493
0.0246
0.9331 ! 0.3862
1.0515 i 0.5854
1.0020
1.1870
0.2849
0.2044
1.3938
-------
CANCER SITE: Saliva^ Glands
Urban and Rural Counties
Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"TM-v" ratin of Avera§ed lst order rate-
AloK L&LJ.O Oi ~ ; r : ; .
Averaged 2nd order rate
n«,_k,i tl. f Averaged 1st order rate.
.\ISK racio or ~ —
Averaged 3rd order rate
ii-Mct-ii ,--,,--!- -,f Averaged 2nd order rate.
Risk ratio or •. • •: — - — :
Averaged 3rd order rate
Sum of weighted rates:
"•J^iffVr-i r-T-t" rarj- -f Averased 1st order rate.
weigncea risic ratio 01 _ ", . , .
Sum of weighted rates
'"J-iB- r^J -1-k" r^r-I- -f AveraSed 2nd order rate-
weix.iceu risK racio 01 t; ^ ; ; : .
6 Sum of weighted rates
...,, isvr-, r,-,i-» raM- -f Averaged 3rd order rate.
iJeigfited ristc ratio or _ "- • . , ,
Sum or weighted rates
Trends
The averaged first order rate is greater
than the second order rate- which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the .first order rate.
White
Males
0.4189
o T?cn
\J • J JVJ.J
0.3952
1.268
1.060
0.836
0.3890
1.0769
0.8491
1.0159
White
Females
0.2349
f) 1 QSQ
u . J.J jy
0.2213
1.200
1.130
0.885
0.2194
1.0106
0.8929
1.0086
Non-whi te
Males
0.0732
0,2094
0.350
0.1600
0.4575
1.3088
Non-white
Females
0.0780
fl fifi?6!
\t • o o / j
0.3231
0.113
0.241
2.128
0.3435
0.2271
2.0014
0.9406
-------
CANCER SITE: Nasopharynx
L~T Urban and Rural Counties
QT] Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Hirik" ratio nf Averaf?ed 1st order rate.
Averaged 2nd order rate
•Tifit" rit • f Averased 1st order rate.
Kistc ratio or — , ~ — . . :
Averaged 3rd order rate
,.r.clrn _,.-,- -f Averaged 2nd order rate.
KISK ratio or . — . _ • , :
Averaged 3rd order rate
Sum of weighted rates:
"Uaiohtad risk" ratio of AveraSed lst order rate-
weignLeQ risic ratio or ^ ^ ;^ :^ ; .
Sum of weighted rates
"U'aiaht^d ri-k" ratij of Avera§ed 2nd order rate
Sum of weighted rates
"U^icl -rM r-f^" ratia -f AveraSed 3rd order rate.
weignteu risK ratio or ~ _ . , , :
0 Sum or weighted rates
•Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
•The averaged third order rate is greater
-than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Hales Females Males Females
0.2483
0.3191
0.2683
0.778
0.925
1.189
0.2730
0.9095
1.1689
0.9828
0.1020
0.0939
0.0815
1.086
1.252
1.152
0.0858
1.1888
1.0944
0.9499
X
0-.1988
0.0403
0.0656
4.933
3.030
0.614
0.0796
2.4975
0.5063
0.8241
0.0232
0.0248
0.935
0.0210
1.1048
1.1810
-------
CANCER SITE: T°ngue and M°Uth
r~T Urban and Rural Counties
QTT Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
t.pj .« fj f Averaged 1st order rate.
Averaged 2nd order rate'
„ ,„ . ... Averaged 1st order rate.
Averaged 3rd order rate'
„.. ,„ . _- Averaged 2nd order rate
Risk ratio of Averaged 3rd Qrder rat(,.
Sum of weighted rates:
"u i -• i -r-i" -t f Averaged 1st order rate.
°a Sum of weighted rates
, . i • i " r,r-f- -f Averaged 2nd order rate.
Weighted risk ratio of Sm Q£ weighted rates -
, i -r k" r-Li- -f Avera8ed 3rd order rate
rte-_2i*teci rxsK racio or _ - , r . .
* Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than, the second order rate which in turn
is greater than the .first order rate.
White White Non-white
Males Females Males
3.5841
3.2118
2.8988
1.116
1.236
1.108
3.034
1.1827
1.0599
0.9566
X
0.9064
0.8769
0,7178
1.034
1.263
1.222
; 0.7636
1.1870
1.1484
0.9400
X
0.6546
1.6276
4.3316
0.402
0.151
0.376
3.4376
0.1904
0.4.735
1.2601
X
Non-white
Females
0.3131
0.5354
1.6891
0.585
0.185
0.317
1.3466
0.2325
0.3976
1.2543
-------
CANCER SITE: Esophagus
CU Urban and Rural Counties
l3{7 Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
White White Non-white Non-white
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"IU-1-" ratio of Averased 1st order rate.
.XVXoK iciUlU OIL — ~ A , , »
Averaged 2nd order rate
"risk" i-ii-i- -f Averased 1st order rate.
KU.SK rauio 01 ~ . _ . . .
Averaged 3rd order rate
'Ti.sU1 i-ari- -f AveraB£d 2nd order rate
AJ.SK. rauiLO ot , « . . — •
Averaged 3rd order rate
Sum of weighted rates:
"Urlahtrd rl-lc11 rati" af Avera8ed lst Order rate -
weignued nstc rauxo OL T ^ ; r . .
Sum of weighted rates
"XJMsl-r- 1 r-Uk" i-jr-I- -f AveraS£d 2nd order rate-
weigncea risK. ratio or ;; ~: ; r~ — ; :
iium of weignted rates
"U^iahfl ri£k" ran- -f AveraSed 3rd order rate.
wexgriutru LJ.SK. raczo or z ^ ^ ; :^
Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
.than the second order rate which in turn
is greater than the first order rate.
Males Females Males Females
4.0135
3.1768
2.8908
1.263
1.388
1.100
3.0741
1.3056
1.0334
0.9404
X
i
i
0.9776
0.7710
0.7367
1.268
1.327
1.046
0.7715
1.2671
0.9994
0.9549
X
717575
4.3694
5.4566
1.775
1.422
0.801
5.6010
1.5484
0.8721
1.0290
0.7111
27.4450
0.5824
0.026
1.221
47.124
4.4352
0.1603
6.1880
0.1313
-------
CANCER SITE:
S totnach
•[ Urban and Rural Counties
123 Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"n-k" ratio of Avera§ed lsc order rate-
Averaged 2nd order rate
'-I-U1 rati- -f Avera§ed lst order rate
KXSK ratxo 01 T^ ; ~ t ,, .
Averaged 3rd order rate
"-Ik" 1-j.ri- -f AveraSed 2nd order rate
Risk ratio of Averaged 3rd OJ_der rat£.
Sum of weighted rates:
,,y .3Vf_l1 r:i_kM ...,- f Averaged 1st order rate.
weixncea rxsic ratio or ., •. • :
* Sum of weighted rates
"U r-'T-l -l-L" r rij -f Avera8ed 2nd order rate-
wej-gnteci rxsK. racio or _ ,. . , 3 • :
0 Sum of weighted rates
"U 'ltd --L" Mti- -f Avera8ed 3rd order rate
wexKuu€*Q rxsic racxo QL „ ,. . , . .
6 Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the .first order rate.
White White Non-white Non-white
Males Females Males Females
19.3141
16.6741
14.2376
1.158
1.356
1.171
15.2290
1.2682
1.0949
0.9349
X
0.2747
7.6514
7.4825
1.212
1.240
1.022
7.7290
1.2000
0.9900
0.9680
X
24.0850
19.6415
12.9416
1.226
1.861
1.518
15.4279
1.5611
1.2731
0.8388
X
8.5759
14.9487
6.4664
0.574
1.326
2.312
7.9503
1.0787
1.8803
0.8134
-------
CANCER
SITE- LarSe Jntestine
Urban and Rural Counties
Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Rink" ratio of Avera8ed 1st order rate.
Averaged 2nd order rate
"Plat11 rani- -f Avera^ed lst order rate-
Averaged 3rd order rate
"ric.k" ,.111- if Averased 2nd order rate.
AISK ratio or ~ — — : r . .
Averaged 3rd order rate
Sum of weighted rates:
"XJ itthr -j ri-k" rati- of Avera8ed lst order rate-
weignceu risic rauio or - e ; , . .
& Sum of weighted rates
"U '-I i I v-I-V" ,-ai-I -f AveraSed 2nd order rate-
weigntea risK ratio or „ . . , . ,
0 Sum or weighted rates
"U U1 i 1 -i-k" r^t-1 - f Avera^ed 3rd order rate.
Weighted risk ratio of Sum of weighted rates •
Trend's
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
"than the second order rate which in turn
TTs greater than the first order rate.
White White Non-white Non-white
Males Females Males Females
16.0514
15.8353
15.6321
i.014
1.027
1.013
15.7143
1.0214
1.0077
0.9948
X
16.4708
16.4945
16.8474
0.998
0.978
0.9.79
16.7508
0.9833
0.9847
1.0058
X
8-. 9 946
10.9660
10.9501
0.820
0.821
1.001
10.6917
0.8413
1.0256
1.0242
16.1056
8.8441
15.4474
1.821
1.043
0.572
14.5894
1.1039
0.6062
1.0588
-------
CANCER SITE:
Rectum
ir Urban and Rural Counties
j_T7 Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"IM-k" ratio of Avera8ed lst order rate -
.IN.ISK caLiu oj. — - - : :
Averaged 2nd order rate
,,.. .,,, ... .f Averaged 1st order rate
Averaged 3rd order rate
"n-k" ~aci- of Averaged 2nd order rate
Averaged 3rd order rate'
Sum of weighted rates:
'"J laii--^ ri-k" r^r-ia -f AveraSed lst order rats-
weigncsa risK racio or ., , , . .
6 Sum of weighted rates
.... •--, i ,-fM-t. raM- -f Averaged 2nd order rate
weigntea risic ratio or „ , . . :
e Sum of weighted rate;;
"U --• • r 1 isk." mti- ^f Averaged 3rd order rate.
weisncpu ristc racio or ^ ^ i \ »
* Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater' than the .first order rate.
White White Non-white Non-white
Males Females Males Females
7.3843
7.7999
7.1967
0.947
1.026
1.084
7.3063
1.0107
1.0676
0.9850
I
4.6759
5.0243
4.6700
0.931
1.001
1.076
4.7207
0.9905
1.0643
0.9893
1.5363
0.5218
4.99.40
0.161
0.308
1.907
5.2092
0.6509
1.8279
0.9587
2.6566
1.4953
3-. 0834
1.777
0.862
0.485
2.8014
0.9483
0.5338
1.1007
-------
CANCER SITE:
Liver
EU Urban and Rural Counties
|*J Rural Counties (counties having urban
centers with. 100,000 inhabitants are
excluded from analysis.)
White White Non-white Non-white
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Ri~k" ti £ Averaged 1st order rate.
Averaged 2nd order rate"
»„_, , n . , ,. Averaged 1st order rate
KJ.SK ratio or . : — ; — : . :
Averaged 3rd order rate
"ri L" - - i • - f AveraSfid 2nd order rate
Averaged 3rd order rate
Sum of weighted rates:
"U-ltl-t-J risk" rati- Jf Averaged lst order rate-
0 Sum of weighted rates
"U i 1-r-l ri 1." ran- cf AveraSed 2nd order rate -
wei:!ntec risK ratio or •* ,
Sum or weighted rates
«r, ^ ^^ j • , .i _j . r Averaged 3rd order rate
weiizhteci nsK ratio 01 — r ~: : ; : — :
Sum or weighted rates
Trend's
The averaged first order rate is greater
than the second order rate which in turn
is greater than che third order rate.
The averaged third order rate is gre'ater
j:han the second order rate which in turn
is greater than the first order rate.
Males Females Males Females
5.1124
4.6267
4.7725
1. 105
1.071
0.969
4.7950
1.0662
0.9649
0.9953
6.5919
6.0335
6.2999
1.092
1.046
0.958
6.2986
1.0466
0.9579
1.0002
6'. 2985
5.7708
5.5268
1.091
1.140
1.044
4.6661
1.1116
1.0185
0.9754
X
6.1691
5.0352
4.8811
1.225
1.264
1.032.
5.0695
1.2169
0.9932
0.9628
-------
CANCER SITE:
Pancreas
~~] Urban and Rural Counties
|Xj Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
„-,,,, .. . M . _f Averaged 1st order rate.
Risk ratio of Averaged 2nd order rate-
ii_. ,n . f Averaged 1st order rate.
" s "a Averaged 3rd order rate'
„. ,. , Averaged 2nd order rate.
&3.SK ratio or Averaged 3rd ordar rate-
Sum of weighted rates:
"U 1 1 i 1 -'-V11 "Ati- c.f Avera§ed lst order rate
Weighted risk ratio of Sum Qf weighted rates .
„ . . .,...,, r- -r Averaged 2nd order rate
Weignted risk ratio of Sum Qf welghted rates .
,.., ... , .,„.__,. Averaged 3rd order rate
Weighted risk ratio of. Sum Qf weighced rates :
Trends
The averaged first order rate is greater
than the second order rate 'which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater' than the .first order rate.
White White Non-white Non-white
Males Temales Males Females
9.1222
8.8440
8.4038
1.031
1.085
1.052
8.5577
1.0660
1.0334
0.9820
X
6.0195
5.4865
5.4293
1.097
1.109
1.010
5.5111
1.0922
0.9955
0.9852
X
2.8699
5.1862
7.5443
0.553
0.380
0.687
6.5689
0.4369
0.7895
1.1485
X
11.4348
2.6449
3.7939
4.323
3.014
0.697
4.6169
0.0874
0.5729
0.8217
-------
CANCER SITE: Nose and Middle Ear
D Urban and Rural Counties
133 Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
uri3kt, tl f Averaged 1st order rate.
Averaged 2nd order rate
•Ti-t" r-c'- f Averased 1st order rate.
Averaged 3rd order rate
"n -t" i--ti- f Averased 2nd order rate.
Risk ratio of Averaged 3rd Qrder rate-
Sum of weighted rates:
"TTplBht&d risk" ratio of Avera8ed lst order rate-
Sum of weighted rates
"U-ial-r-J rl-t" rati- c.f Avera8ed 2nd order rate
wej.gnt.su risic rano 01 z ^e ^ , . .
Sum of weighted rates
"U -1al r 1 r-i^-" rarii cf Avera8ed 3rd order rate -
Weighted risk ratio or s^ Q£ w£ighted rates -
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
Ts greater than the first order rate.
White White Non-white Non-white
Males Females Hales Females
0.3687
0.5410
0.4433
0.682
0.832
1.220
0.4477
0.8235
1.2084
0.9902
0.2280
0.1748
0.2276
1.304
1.002
0.768
0.2202
1.0354
0.7938
1.0336
0-.3242
0.2900
0.2350
1.118
1.380
1.234
0.2551
1.2709
1.1368
0.9212
X
0.4860
0.2206
0.1769
2.203
2.747
1.247
0..2231
2.1784
0.9888
0.7929
-------
CANCER SITE:
Larynx
~~y Urban and Rural Counties
\T. Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
•"•I-k" ratia of Averase
-------
CANCER SITE: Trachea> Bronchus and Lung
CU Urban and Rural Counties
Qj Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
••TM-J," r,tl, af Averaged 1st order rate.
KXSK r&ulO OZ ~ : r • .
Averaged 2nd order rate
,,r. , „ - Averaged 1st order rate
Averaged 3rd order rate'
„_. ,„ . _f Averaged 2nd order rate.
Risk ratio of Averaged 3rd order race-
Sum of weighted rates:
"U d -i k" ri f Avera8ed 1st order rate.
Sum of weighted rates
"U I 1 t 1 i-l -k" rarl- cf AveraSed 2nd order rate-
weiRrmeci rxsK. rncxo 01 r ^ . . . *
6 Sum of weighted rates
"U-iclt^J risk" r-ti- 3f Avera2ed 3rd order rate
tvexxncea rxsK ratio oz — ^ . , t
Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
-ehan the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Males Females Hales Females
31.3235
30.0740
28.0289
1.042
1.118
1.073
28.7382
1.0900
1.0465
0.9753
X
4.9250
5.0627
5.0978
0.973
0.966
0.993
5.0714
0.9-711
0.9912
1.0052
X
26.4475
19.0065
26.6694
1.323
0.992
0.750
25.6432
1.0314
0.7798
1.0400
5.3389
2.9137
4.9340
1.832
1.082
0.590
4.6978
1.1365
0.6202
1.0503
-------
CANCER SITE:
Breast
•r Urban and Xural Counties
HT Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
»r--k" --Jtio -f Avera8ed lst order rate-
RISK ratio on . ; r r , •
Averaged 2nd order rate
,,...,. . . - Averaged 1st order rate.
Risk ratio of Averaged 3rd Qrder rate-
... „ , Averaged 2nd order rate.
Risk ratio of Averaggd 3rd Qrder rat(,.
Sum of weighted rates:
,. . . ....,„,. _f Averaged 1st order rate.
Weighted risk ratio of s^ Qf weighted rates •
...... , . ,„ -,-,- -f Averaged 2nd order rate.
Weighted risk ratio of S(jm Qf weighted rates •
...... . ,.. , . r Averaged 3rd order rate.
Weighted risk ratio of Sum of weighted rates •
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than ;he second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Males Females Males Females
0.2628
0.1778
0.2389
1.478
1.100
0.74-''
0.23:; 3
1.12t4
0.7621
1.0240
i
25.0697
24.9366
24.0034
1.005
1.044
1.039
24.2674
1.0331
1.0276
0.9891
X
0.2047
1.8365
0.2149
o.iii
0.952
8.546
0.4557
0.4492
4.0301
0.4716
22.5271
28.2036
17.0904
0.799
1.318
1.650
19.3798
1.1624
1.4553
0.8819
-------
CANCER SITE:
Cervix Uteri
Urban and Rural Counties
Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Risk" ratio of- Averaged 1st order rate.
Averaged 2nd order rate
Risk" ratio of
""
Risk" ratio of
Averaged 1st order rate.
Averaged 3rd order rate'
Averaged -2nd order rate.
Averaged 3rd order rate'
Sum of weighted rates:
„., . , , . , ,, . .. Averaged 1st order rate
Weighted risk ratio of —= ^ ——3 :
Sum of weighted rates
it,, . . j * i ti . .c Averaged 2nd order rate
Weighted risk ratio of —z f r r- •. '
0 Sum of weighted rates
„., ..,.,!i . , Averaged 3rd order rate
Weighted risk ratio of —r °-z 7—-—-: :
6 Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Males Females Males Females
8.1718
7.7446
7.9294
1.055
1.030
0.977
7.9335
1.0300
0.9762
0.9995
14.5349
23.4777
9.1016
0.619
1.507
2.580
11.8566
1.2259
1.9801
0.7676
-------
CANCER SITE:
Corpus Uteri
r~I Urban and Rural Counties
L*3 Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"n1clr«i riMr, nf Averaged 1st order rate.
Averaged 2nd order rate
'"M-k" ratio af Avera&ed lst order rate-
AXSK rauxo 01 L : r •
Averaged 3rd order rate
..Tr-t.i rl. -f Averaged 2nd order rate.
AISK ratio or , _, _ , ,
Averaged 3rd order rate
Sum of weighted rates:
"U lE'iraJ ri-k" rati- -f Avera*ed lst order rate-
wexKnceu LJ.SK. raua.o or r ^: ; : , «
Sum of weighted rates
"W -nl-t-d ri-k" rari- -f Averaged 2nd order rate_
Sum of weighted rates
"u tr-t d ri-k" r^ri- -f Averased 3rd order rate.
ne.Lx.tu?u nsK rauiuo or _ ,. . r ^
6 Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the .first order rate.
White White Non-white Non-white
Males Females Males Females
6.5272
6.8642
6.9039
0.951
0.945
0.994
6.8515
0.9527
1.0018
1.0076
X
11/6947
10.4828
8.9368
1.116
1 309
1.173
0.5139
1.2292
1.1018
0.9393
-------
CANCER SITE: Ovary and Fall°Pian tubes
EZJ Urban and Rural Counties
pTf Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Ri-t" ratio of Avera8ed lst order rate-
.RISK ratio 01 , : T r ^ •
Averaged 2nd order rate
"n-L" YATI f AveraSed 1st order rate.
RISK racio 01 . : r , , .
Averaged 3rd order rate
•TitL" I-AI-I- -f Avera&ed 2nd order rate
jxxSiC rac.LO or . r r ; . :
Averaged 3rd order rate
Sum of weighted rates:
"Uficht-Ld ri-t" ratia -f Avera8ed lst order rate-
Sum of weighted rates
"U 'a' i J riqk" ratio cf AveraSed 2nd order rate.
0 Sum of weighted rates
"U--tBl-r--1 ritl;" -aria -f Avera8ed 3rd order rate
weignteu risK rario 01 _ . , ,
e Sum or weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order race is greater
than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Males Females Males Females
8.0938
8.5943
8.3943
0.942
0.964
1.024
8.3852
0.9652
1.0249
1.0011
i
4.2110
5.7223
6.2791
0.736
0.671
0.911
5.9324
0.7098
0.9646
1.0584
-------
CANCER SITE:
Prostate
if Drban and Rural Counties
l23 Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Risk" ratio of Averaged 1st order rate;
Averaged 2nd order rate
Risk" ratio of
Risk" ratio of
Averaged 1st order rate.
Averaged 3rd order rate'
Averaged 2nd order rate.
Averaged 3rd order rate'
Sum of weighted rates:
it,, . . . _, i ii ..- c Averaged 1st order rate
Weighted risk ratio of — «-T . , •
0 Sum of weighted
rates
"Weighted risk" ratio of Averaged 2nd order rate
6 Sum of weighted rates
,„, ,, . , . , >. . ,- Averaged 3rd order rate
Weignted risk ratio of —5 =7 • :
0 Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the .first order rate.
White White Non-white Non-white
Males Females Males Females
18.2882
18.5279
18.3156
0.987
0.998
1.012
18.3423
0.9970
1.0101
0.9985
15.0168
14.7194
21.4893
1.V020
0.699
0.685
19.6153
0.7656
0.7504
1.0955
-------
CANCER SITE: Testis
Urban and Rural Counties
Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Risk" ratio of Averaged 1st order rate.
Averaged 2nd order rate
"Risk" ratio of Averaged 1st order rate;
Averaged 3rd order rate
Risk" ratio of
Averaged 2nd order rate.
Averaged 3rd order rate'
Sum of weighted rates:
„,, . . , . , it . * Averaged 1st order rate
Weighted risk ratio of — =^ ——-,
Sum of weighted rates
„,, . . , . . „ . , Averaged 2nd order rate
Weighted risk ratio of -7 °-z —r—- :
e Sum of weighted rates
it,, ., . . • , .i j c Averaged 3rd order rate
Weighted risk ratio of —5 °-: ——
0 Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Males Females Males Females
0.9430
0.8982
0.9220
1.050
1.023
0.974
0.9213
1.0236
0.9749
1.0008
i
0.0286
0.0236
0.1907
1.213
0.150
0.124
0.1441
0.1984
0.1638
1.3232
-------
CANC£R SITE:
Kidney
Urban and Rural Counties
Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"ri-v« ratio of Avera§ed lst order rate-
.RISK rauio t/i. ~ ~ ~ ; . .
Averaged 2nd order rate
,,*... .,„ r1. f Averaged 1st order rate.
Risk .atio of Averagfid 3rd order rate-
"-; L" -JM- if Avera8ed 2nd order rate
Risk ratio of Averaged 3rd order race-
Sum, of weighted rates:
,,„ . ... , ±.,.M M. .f Averaged 1st order rate.
Weignted risk ratio of ^ of weighted rat£S .
..„ . , , . , it . , Averaged 2nd order rate
neiLguteci risic racio or ^ >: ^ r t
6 . Sum of weighted rates
„„ . , , • ,»,--,,, F Averaged 3rd order rate.
Weighted risk ratio of Sum Qf weighted races -
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the .first order rate.
White White Non-white Non-white
Males Females Males Females
4.0716
3.5932
3.7473
1.133
1.086
0.959
3.7666
1.0810
0.9540
0.9949
2.1718
2.0285
2.0740
1.071
1.047
0.978
2.0796
1.0443
0.9754
0.9973
1.5582
2.3489
2.2447
0.663
0.694
1.046
2.1683
0.7185
1.0830
1.0350
0.3425
0.8401
1.7412
0.408
0.196
0.482
1.4318
0:2392
0.5867
1.2161
-------
CANCER SITE: Bladder
d Urban and Rural Counties
|TT Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Hi -It" ratio of Avera8fid lst order rate -
.K1SK raulO Or — ~ . ,
Averaged 2nd order rate
"itr L" id -f Averased 1st order rate.
KiSK ratio ot — j - . . :
Averaged 3rd order rate
"ri^L" iiLi f AverflSed 2nd order rate
Risk ratio of Averaged 3rd order rate-
Sum of weighted rates:
"Uplehr^d ri^" ratio of Avera8ed lst order rate
wexgnceo riLSK rauxo or ~z ^ * . , .
Sum of weighted rates
"U--U1 i J ri-1" ratio c^ Avera^ed 2nd order rate.
wej-Kuteu riSi. r3.Cj.O 0*. ^ ^: ; ; ;
6 Sum of weighted rates
"U lit J ri-k" ratio -f Avera*ed 3rd order rate
° Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Males Females Males Females
6.6565
7.2752
5.7923
0.915
1.150
1.256
6.1129
1.0889
1.1901
0.9476
2.4786
2.27.22
2.1219
1.091
1.168
1.071
2.1874
1.1331
1.0388
0.9700
X
4.7320
5.2706
2.4443
0.898
1.936
2.156
3.1714
1.4921
1.6619
0.7707
1.6780
1.8197
2.6279
0.922
0.638
0.692
2.3898
0.7022
0.7614
1.0996
-------
CANCER SITE:
Skin Melanoma
d Crban and Rural Counties
|1CT Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged' 3rd order rate:
••n-t^k'« ^-in nf Averaged 1st order rate.
rvistc ratio 01 — - - .
Averaged 2nd order rate
""I L-" url- rf Avera&ed lst order rate-
Risk ratio of Averaged 3rd order ratfi.
„ „ . , Averaged 2nd order rate
Risk ratio of Averaged 3rd order rate-
Sum of weighted rates:
"ut.tr -i -Heir" riri-t -if Averaged 1st order rate.
Weighted risk ratio or — - ~ ; r — j :
6 Sum of weighted rates
....... . ,,, , ,f Averaged 2nd order rate
Weigated risk ratio of Suffl Q£ weightfid rat3S -
„ , . . , „ . . . Averaged 3rd order rate.
Weighted risk ratio of ^ Qf weighted rates •
Treads
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the .first order rate.
White White Non-white Non-white
Males Females Males Females
Q.9310
0.9973
1.2169
0.0934
0.765
0.820
1.1493
0.8100
0.8677
1.0588
X
0.7607
0.9125
0.9832
0,834
0.774
0.928
0.9456
0.8045
0.9650
1.0398
X
0.0126
0.0284
0.0651
0.444
0.192
0.432
0.0531
0.2373
0.5348
1.2373
X
0. 3851
0.3058
1.259
0.2776
1.387-2.
1.1016
-------
CANCER SITE:
Other Skin
Urban and Rural Counties
Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
,.rl-kr, rj *. Averaged 1st order rate.
KISK racio o— ~ . ; .
Averaged 2nd order rate
"TM L" AM " Averaged 1st order rate.
K.ISK. ratio or . . „ . . :
Averaged 3rd order rate
"TM^" .in f Avera
-------
CANCER SITE:
Eye
Urban and Rural Counties
Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
White White Non-white Non-white
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
IITM-I.I« -r^nn nf Averaged 1st order rate.
RISK ratio or a — - — - — :
Averaged 2nd order rate
"•neb" -r^i-t^ -,f Averaged 1st order rate.
Risk ratio of - a — : — : — •: -; :
Averaged 3rd order rate
„ ,„ , Averaged 2nd order rate.
Averaged 3rd ordet rate'
Sum of weighted rates:
..„.,. , . , , .. .... £ Averaged 1st order rate
wexguceu rxsK racio ox z - . , .
6 Sum of weighted rates
„„ . , , . , n . £ Averaged 2nd order rate
weisnceu rxsK rsno ox, ,-, £ . T^ . .
6 Sum of w€-.ighted rates
„„ .... , . .,„ (1 f Averaged 3rd order rate
weignteu nsK ratio or „ ,- , , , :
0 Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the .first order rate.
Males Females Males Females
0.1859
0.2100
0.2392
0.0885
0.777
0.878
0.2283
0.8143
0.9198
1.0477
X
0.1879
0.1484
0.1953
1.266
0.962
0.760
0.1878
1.0005
0.7902
1.0399
0.0286
0.0087
3.2S7
0.0106
2.6981
0.8208
0.1579
0.1363
1.158
0.1196
1.3202
1.1396
-------
CANCER
Brain and Nervous system
d Urban and Rural Counties
£23 Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
••THrlr" T-,Mn ne Avera8ed 1st order rate.
Averaged 2nd order rate
"RUU" M -- Avera8ed lst order rate-
RISK rauxo or , - , . .
Averaged 3rd order rate
•Tink" r-iil -f AveraSed 2nd order rate
Averaged 3rd order rate
Sum of weighted rates:
•Ttlnht-d ri-k" ratj- af Avera^ed lst order rate
wexgnueu LJ.SK. LOLL.*.O QL ^ , . *
Sum or weignted rates
"Up-tal r 1 r-Uk" rari- cf Avera?ed 2nd order rate-
weignceo nsjc racio ot „ • • . , . :
Sum of weighted rates
"tfpiehr^d risk" -atla of AveraSed 3rd order rate.
weixLiu&CL rxstc j.auio 01. ~ » ; ; .
Sum or weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
Than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Males Females Males Females
4.0784
4.3283
3.9547
0.942
1.031
1.094
4.0235
1.0136
1.0758
0.9829
2.6748
2.6248
2.6556
1.019
1.007
0.988
2.6536
1.0080
0.9891
1.0008
2.0118
0.9971
1.5742
2.018
1.278
0.633
1.5464
1.3010
0.6448
1.0180
4.9574
0.4713
1.0169
10.518
4.875
0.463
1.4481
3.4234
0.3255
0.7022
-------
CANCER SITE:
«land
~f Urban and Rural Counties
2£3 Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Ri-t" ratio of Avera§ed lst order rate-
.RISK jrauio O£ : r , . •
Averaged 2nd order rate
»-l-k" --jria af Avera8ed 1st order rate.
JxlStC raulO Or ~ : r : ; .
Averaged 3rd order rate
"•M^k" , r-iT -f Avera8ed 2nd order race-
KISK ratio oi . , _ , . :
Averaged 3rd order rate
Sum of weighted rates:
"U lal-r 1 rl -k" r^ri- -f Avera?ed lst order rate-
wexsnted nsic racxo oz r - : . .
6 • Sum of weighted rates
"'J I-li 1 risk." rarl- -f Avera§ed 2nd order rate
nexKutBu nstc racxo or ^ ^ t ; ^ •
5 Sum of weighted rates
"•J;'la'i 1 ri-k" rat-i- -f Avera8ed 3rd order ratG
tveigntea risn ratio or - , . . , . :
0 Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Males Females Males Females
0.4758
0.5146
0.3514
0.925
1.330
1.440
0.3948
1.2052
1.3034
0.9053
0.7330
0.7144
0.7243
1.026
1.012
0.986
0.7240
1.0124
0.9867
1.0004
0.0365
0.0145
0.3741
2.517
0.098
0.039
0.2754
0.1325
0.0526
1.3584
0.1952
0.1366
0,3914
1.429
0.499
0.349
0.3296
0.5922
0.4144
1.1875
-------
CANCER SITE:
Endocrine Organs
f~T Urban and Rural Counties
C*j Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"TM-v" ratio of AveraSed 1st order rate.
.IV.LSK rarxo or ~ : - . ,
Averaged 2nd order rate
"lit a" rji i ; nf Averased lst order rate-
KISK ratio or ~ r . , .
Averaged 3rd order rate
"rii.L" run f Avera&ed 2nd order rate-
Risk ratio of Averaged 3rd order rate-
Sum of weighted rates:
HT, _, , ., i «t • * Averaged 1st order rate
6 Sum or weighted rates
"u tit 1 ritt" ratio cf Avera^ed 2nd order rate.
e Sum or weighted rates
"u i-ti l ,-1-L" i-Ari- of Avera8ed 3rd order rate
wexgntefl risK. ratio 01 _ •*, .i_^j ^
0 Sum or weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Males Females Males Females
0.2758
0.2757
0.2614
1.000
1.055
1.055
0.2653
1.0396
1.0392
0.9853
X
0.2759
0.1591
0.1941
1.734
1.421
0.820
0.1994
1.3836
0.7979
0.9734
0.0440
0.0272
0.0480
1.618
0.917
0.567
0.0444
0.9910
0.6126
1.0811
0.0471
0.0427
._..
1.103
0.0378
1.2460
1.1296
-------
CANCER SITE:
Bone
Urban and Rural Counties
Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Ul ik" ratio of Avera§ed lst order rate-
Averaged 2nd order rate
"ti k" nti- -f Avera§ed lst order rate
XJ.;»K ratio or ~ ; - . . •
Averaged 3rd order rate
""• L" -jti- ^f Avera8ed 2nd order rate
iv3.SK. rauio OL T t ^ i . .
Averaged 3rd order rate
Sum of weighted rates:
"U luht-pd ri-k" ratio of Avera8ed lst order rate -
wcxstiuBu rxsic raui.o oz :; -. . \ . •
6 Sum of weighted rates
.... . , , . .„ ±. f Averaged 2nd order rate.
WGiKnt&u ris.c rac.LO or r .. - . ; , .
6 Sum of weighted rates
"u -ai i -j ri-k." rnri- af Averased 3rd order rate.
Weighted risk ratio of Sum Qf weighted rates •
Trends
The averaged first order rate is greater
than che second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than che second order rate which in turn
is greater than the .first order rate.
White White Non-white Non-white
Males Females Males Females
1.3431
1.4135
1.3269
0.950
1.012
1.065
1.3412
1.0014
1.0539
0.9893
0.8122
0.8450
0.8821
0.961
0.921
0.958
0.8682
0.9355
0.9733
1.0160
X
0.5750
6.9115
1.2838
0.083
0.448
5.384
2.0297
0.2833
3.4052
0.6325
0.1433
0.2245
0.9112
0.638
0.157
0.246
0.7140
0.2007
0.3144
1.2762
-------
CANCER SITE:
Connective Tissue
Urban and Rural Counties
Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Risk" ratio of-
"Risk" ratio of
Sum of weighted
"Weighted risk"
"Weighted risk"
"Weighted risk"
Trends
Averaged 1st order rate.
Averaged 2nd order rate'
Averaged 1st order rate
Averaged 3rd order rate"
Averaged 2nd order rate
Averaged 3rd order rate'
ratesT
- Averaged 1st order rate.
Sum of weighted rates
Averaged 2nd order rate
Sum of weighted rat.js
, Averaged 3rd order rate.
Sum of weighted rates
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater Chan the first order rate.
White White Non-white Non-white
Males Females Hales Females
0.5190
0.6483
0.6191
0.800
0.838
i.047
0.6105
0.8501
1.0619
1.0141
0.3785
0.5478
0.4580
0.691
0.826
1.196
0.4608
0.8214
1.1880
0.9939
0.3568
0.7236
0.3638
0.493
0.981
1.989
0.4166
0.8564
1.7369
0.8732
0.3954
0.2246
0.6983
1.760
0.566
0.322
0.5915
0.8581
0.3797
1.1806
-------
CANCER SITE:
Hodgkin's
!I Urban and Rural Counties
\~X~i Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
..r-.-k.. ran-, Of Averaged 1st order rate.
RISK racio 01 i , r , , .
Averaged 2nd order rate
"•ML" r^i-1- -f AveraSed 1st order rate.
Kistc ratio or — r^ . • , .
Averaged 3rd order rate
"ri-U1 -.111- -r Averased 2nd order rate
Risk ratio of Averaged 3rd order rate-
Sum of weighted rates:
"u fir \ -1-1." rarl- -f Avera8ed 1st order rate.
weigntea rxstc. ratio 01 _ - , .
e Sum of weighted rates
"u r-lr i -ll" i-ari- if Avera?ed 2nd order rate
Weighted risk ratio of Sum Qf weighted rat£S -
,,y . . . • ,» --tl- af Averaged 3rd order rate
Weighted risk ratio of gum Qf weighted rates •
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the .first order rate.
White White Non-white Non-white
Males Females Males Females
2.1772
2.1638
2.2358
1.006
0.984
0.968
2.2182
0.9815
0.9755
1.0079
1.1993
1.2136
1.3145
0.988
0.91-2
0.923
1.2860
0.9326
0.9437
1.0222
X
0.9282
3.9428
0.7054
0.235
1.316
5.589
1.2186
0.7617
3.2355
0.5789
0.4193
0.1964.
0.7189
2.135
0.583
0^273
0.6056
0.6924
0.3243
1.1871
-------
CANCER SITE: Lymphosarcoma
Urban and Rural Counties
Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Ri-k" ratio n£- Avera8ed 1st order rate.
K.XSK rauxo OL T^ : 7 r^ , .
Averaged 2nd order race
"TM ,-!,•• T-zrin nf Avera2ed 1st order rate.
K.ISK ratio or j r — , , •
Averaged 3rd order rate
"111 5k" rri- -' Avera8ed 2nd order rate.
XU.5K ratio Oi. — ~ r ;
Averaged 3rd order rate
Sum of weighted rates:
"W-ial L J -i£t" riti- f Avera^ed 1st order rate
weignteo nsK ratxo or _ ... . , . .
Sum of weighted rates
"TJMal i J r-uk" rati tf Avera^ed 2nd order rate
Weighted risk ratio o£ gum of Wfcightad rates .
"U it) i 1 r--c.i" i-Ail- f AveraRed 3rd order rate
wezznteoi riLSK. ratio or ~ _ ; ; , .
6 Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
.than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Males Females Males Females
4.8484
4.7359
4.6538
1.024
1.042
1.018
4.6902
1.0337
1.0097
0.9922
X
3.2486
3.1958
3.2166
1.016
1.010
0.994
3.2176
1.0096
0.9932
0.9997
12,5699
3.3140
1.2497
3.793
10.058
2.652
3.0673
4.0980
1.0804
0.4074
X
1.0770
1.3418
1.2338
0.803
0.873
1.088
1.2290
0.8763
1.0918
1.0039
-------
CANCER SITE:
ICD's not listed
r~[ Urban and Rural Counties
\X] Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Hi-lc" nrli of Avera«ed lst order rate-
Averaged 2nd order rate
"n-,-k» . . _- Averaged 1st order rate.
KILStC ITati O OT . — , _ , v — *
Averaged 3rd order rate
««,.-,-.-•• V-M- -f Averaged 2nd order rate.
K.ISK ratio oi . . — - , ,
Averaged 3rd order rate
Sum of weighted rates:
"U icli d rick" rait- -f Avera«ed lst order rate-
6 Sum of weighted rates
"U -lul i 1 -i-k" rait f Averaged 2nd order rate.
wei&utecL rxstc ratxo oi ~ ^ ^ ; , .
Sum of weighted rates
"u iiit i i • i" -AH r Avera2ed 3rd order rate
Weighted risk ratio of Sum Qf weighted rates -
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Hales Females Kales Females
9.6478
9.3495
9.5897
1.032
1.006
0.975
9.5629
1.0089
0.9777
1.0028
i
9.3665
9.0721
9.4003
1.032
1.036
0.965
9.3498
1.0018
0.9703
1.0054
0.7299
4.3872
13.7450
2.218
0.708
0.319
11.8121
0.8237
0.3714
1.1636
12.5807
7.3556
9.1978
1.710
1.368
0.800
9.3718
1.3424
0.7849
0.9814
-------
CANCER SITE' ^^ Malignant Neoplasms
Urban and Rural Counties
Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"ri.nl-" rntia of Avera?ed lst order rate-
Averaged 2nd order rate
"rr-k" TAI t f Avera?ed lst ord£r rate.
KJLSK ratio or ~ , ,. . ,
Averaged 3rd order race
,,r, _,.., r,ri- -f Averted 2nd order rate.
Risk ratio of Aver^,ed 3rd order race-
Sum of weighted rates:
"U-iEht-J ri-k" rAtia of Avera8ed lst order rate
Sum of weighted rates
"U lahtiJ rl-k" -id- cf AveraSed 2nd order rate
wexKHLeu nsK tatiLO OL r ^ ~ r~-" , .
* Sum of weighted raues
"u luht^l risk" r^i-i- af AveraSed 3rd order rate
6 Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
.than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Males Females Hales Females
165.7077
160.4561
152.8577
1.033
1.084
1.050
X
130.0746
127.3310
127.1617
1.022
1.023
1.001
127.5475
1.0198
0.9983
0.9970
X
140.1987
135.1143
134.2927
1.038
1.044
1.006
135.2029
1.0370
0.9993
0.9933
131.4023
148.4828
108.0597
0.885
1.216
1.374
116.8480
1.1246
1.2707
0.9248
-------
CANCER SITE:
Leukemia
CJ Urban and Rural Counties
["XT Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"iH-lr" ratia cf AveraSed lst order rate-
KJLSK ratio ot ~ . r ~ .
Averaged 2nd order rate
"Pt-l" r*rt- -f Avera§ed lst order rate-
KISK. ratio or . , « , , •
Averaged 3rd order rate
,,r. ,„ . f Averaged 2nd order rate.
Ki*k ratio of Averaged 3r, ^rder ratg.
Sum of weighted rates:
"•J r -i-i 1 rl-L" rati- of Avera^e
-------
CANCER SITE: ***iplB
Drban and Rural Counties
1*7 Rural Counties (counties having urban
centers with 100,000 inhabitants are
excluded- from analysis.)
Averaged 1st order rate:
Averaged 2nd order rate:
Averaged 3rd order rate:
"Hint" rntia af AveraBed 1st order rate
Averaged 2nd order rate
"-IQL" rarl if Averased 1st order rate.
Averaged 3rd order rate
"IHt,-L" rat! -f Avera8ed 2nd order rate
Averaged 3rd order rate
Sum of weighted rates:
"U -isl tad r^L" r-itia -f Averased 1st order rate.
wei.2tii.eci rxsjc ratio ot ~ ^ ^ : , :
6 Sum of weighted rates
"u r it J L-J L" i -f Avera«ed 2nd order rate-
wexsnccu rxsic rauio ot r ^ ^ : ; •
s Sum of weighted rates
"u--mt-J ri-i" r^rl- 3f Avera&ed 3rd order rate.
weiguceu risK. ratio or ••- , .
6 Sum of weighted rates
Trends
The averaged first order rate is greater
than the second order rate which in turn
is greater than the third order rate.
The averaged third order rate is greater
than the second order rate which in turn
is greater than the first order rate.
White White Non-white Non-white
Kales Females Hales Females
1.6703
1.7654
1.7115
0.946
0.976
1.031
1.7139
0.9746
1.0300
0.9986
1.2923
1.1488
1.1624
1.125
:1.112
0.988
1.1766
1.0983
0.9764
0.9879
1.0636
1.7415
1.4088
0.611
0.755
1.236
1.4124
0.9402
1.2330
0.9974
1.8625
0.5793
1.9602
3.217
0.950
0.296
1.7504
1.0640
0.3310
1.1198
-------
APPENDIX II
Study Questionnaires and Survey Instruments
for the Study of Great Lakes Commercial Fishermen
- Telephone interview questionnaire (pilot) Protocols 1 and 2
- mailed questionnaire (pilot) Protocol 3
- telephone proxy questionnaire (Pilot)
- Great Lakes Fishermen Health Survey - Large Cohort
-------
Telephone Interview (Protocols 1 and 2),
KIMIl:icHKH IIKAl.TIt SIIIIVKY
INSTRUCTIONS I Flenoc enoucr (lie fallowing qucsttong ant
return the questionnaire within five days. Uao the enclosed
Kt raped, self-addressed envelope for Bailing. Tour anewcm
are confidential and will only be used for research purposes.
Your identity will not be associated with the survey results.
If you cannot give un exact answer, provide your best estimate.
I. Whac ie your currcue ccMreisef
(street)
(city) (state) (tip)
2.
1.
4.
5.
«.
•Whet
Utot
What
Uhat
What
la
is
U
ie
lit
your
Dent
your recel
your
your
your
,
1
O-le
)£) White
birthdeteT
Social
narital
nonth
2)C
| Fencle
2)Q Black :
1
day
»
year
!)("") Hispanic *)Q American Indian 5)^") Other
Security number? - - .
status? l)(
"^Slnitle
.arrleJ) 2)C*} Harried 3)f^ Separated
or divorced *)(~} Widowed
7. Do you presently hold a commercial fishing license? ))( )Ho 2)( ) Yes—"-Approximately over wliot years liave you held c license? /
^~/ ^ T7o*~"To~~
1 * ~ ° fc"~"' (go to question 10)
6. Did you hold a coK*erclal (lilting license in the past? I)(y Ho 2)^) Yes—»- Approxlnately over wlutt years did you hold a license? /
froto to
(go to quest loci 10)
-------
9, Btv« you ever been a crew nesber for, or e partner wtth^sn Individual owning K coaaerciel ftehlng license?
t)(_) Ho 2^O *ee-—» Over vhet years have you been e crew oenber or partner /
I from to (go to question; tO)
(If Ho, plena* ctop here end return the questionnaire »6 coon ae pocelble In the addreeeed envelope. Thank you for your cooperation.)
0. Hear what town* do (did) you commercially fish Boat often?
0*. Where It (was) your hooe port located!
City'State
-------
I. 16 cemerclel (Ttcblce year cttinreae t)(~)Ko
full-tine oecstpetioaf
12. Wluit in your curr '—#0»«r vhflt ye«rs tutve you been e pert-tine ficbenua? /
fcos to
(go to queetloa 14)
2 )f")lfee —(.Over ubat years wera you full tteof /
from to
Ower whist JCECR heve you cotmcrc Jelly flched full tUset
__
froo Co
(go to question I
ft. Da you currently
-------
Tu Hi"' In-ill ul yiiif l'iMiw|i'iinl mitt |'ivnm()
Employed
Street
City
State
•fro«
To
.. Do (or did) you confute any of the flah you catch commercially or o» a (port fisbernan? I ) (~) Yee ; 1 )
I. Approximately how eany of your neala contain fish caught (by yourself or a friend) from the Great Latest
1) _ per week (or) 2) _ per month
No (If No, go to queetioq 28)
I. What types of Creat lakes flab do you eat Boat oftent
Saloon Walleye Hough flah
tako Trout BatB
Perch Burbot Chub Lake Whitefiali Smelt Lake Hotting _
ApproxJaately how pany year* have you conauaed Great Lekee fleh with thle frequency? yetre
Northern;Pike
Lake Whitetieli Saelt
Paa flah
Otheri (epeclfy)_
Other Trout
!. Approximately how nany of your ceale contain fish ceughc (by youreelf or & friend) fron watere other than the Great
1) per week (or) 2) per sontb
-------
Whet types off B»tn®«; intend «3tecc" Cieb do you «»t >o»t of Cent
_ ___ Seleoa _________ Kellerc Hough fiuh _ Northern Pika
Perch
Burbot
Chub
Lake Uhiteflch
_ take Trout Bane
Smelt _ Luke Herring
__ J?aa flub
Other I (/specify)
Other Trout
24. Approximately tow cusny ye* re heve you conduced inland w«tes fish with thte frequency!
yeere
25. In conpr.ric.on to your col f> Bow often doer, your wife eat lleM How of tea do your none c./st fieht Botr often do your deughtere one Cleht
tflfel Sons i Daughter*!
i) (___) Ac often f-e ajreelf l)(
2) (tore often then ayeelf
3)QLe6c often then caycclf
*)Q Does not eppty
5) Hever
l)(Jie often e« myeell
2) Q More often th«n oysolf
1) Q Leee often than nyr.clf
*) Q Doec not apply
5) Never
1) As often «e nyfiotf
2) Q More often than cyeelf
3) Q LCCE of tea than
4) Q Doee not apply
5) Never
26. Approximately how eany pounds of ft eh hcvo you consumed pec yeert
(r.asuBd 1 f(eh peel equate *| pound of filth) pounde
11. Of the fleh you c«tcb end consume, whet percentage of the fish ere prepared by eech of (be following eethodeT
1) Broiling 1
2) Pen frying X
3) Siwked X
4) Boiling
5) Patched
6) Other
-------
28. Do you Drcxatly BBok* cigar«tta«T 1)0"° 2) O ***
29, How old uere you when you Eiret began to evoke eigeretteef
30. What la th* «vet«g* otuiber of cigarette* you presently BooKe per d«yT
31. Bow many y««r* b«v« you naked el««r«tt«s wsth title Itcqueacjl
(go to queetioa 34)
2S«. DM you «ofee elgcretcce in the paetT l)QKo 2>OT
(go to question 34)
32. Bow old were you when you fleet bcgefi to Esott
33. BOB Beny years din! you csofcc
($o to quoctloa 34)
-------
24. Do jva presently moke e nlpef I) Q Ha 2)Qtec
Old you moke & pipe tn the peetj
(If to. go to queBtiOa nufibcr 40)
35. HOB old uere you whca you firet begeo to tssokc e plpef_
36. Ohet £c the evercge auzsber of plpafule you pceccntly make pec
-------
«0. Do you preoeocly eawke cigar*? 1)0"° 2)
41. llotr old were you when you began to moke cigar*?
42. What IB the average [timber of clg*c> you preeently onoke per day?
43. Bow nany year* have you aaokeJ cigar* with thio frequency?
J>0e. Did yo«i moke clger* In the peat t)Qlto 2)Qtec
(If No, go to queetlon number 46}
(go to question 46)
44. Botf old were you when you begea to eaoke
clgarel ______
45. Row eany yearn did you esoke
(go to questloa 46)
46. Have you ever chewed tobacco regularly? l)) No 2) tea
47. Have you ever ueed anuff regularly? l)Qlto 2)QTcr,
48. Have you ever smoked non-tobacco products regularly? l)No
-------
4Urd liquor it coaelder«d to Includei Clo| UhUkcy{ 8coceh( Vodtej Eun( Brendy
(A drink - I to 11} OK. of elcohol.)
$. Do you detak Etecd ilquoref J) Ho
Did you eoaeum« hard llre la the peett
SO. Jlow old vere you vheo you fleet bognn to dclnk herd llouoref
SI. .Uhct le the svetege mrabcr of bard liquor drlnfec you eooetme per week?
52. Dou cany yeeie have you coaeuued hard liquor ulth Cblc frequeacyt
Sli How old ware you when you flrct begem to drink herd
llquorf ____.
S4. Uou ncny year* did you COOOUKO herd liquor)
(go to queue loo SS)
(go to queatioa 55)
-------
55. Do you drink becrT D£)HO 2) £) fee -
55c. Did you drink beer la the p«ctT
v
56. Hoy old were you when you f tret b«ge» to drlafc beer?
57. What 1« the everege number of beeta (12 ot. bottlen) you preeently coasuse
pet week? ________
58. How many ye«r* have you consumed beer with thie frequeocyT .
:(go to quoBttoo 61)
59. Bov old were you uhea you first began to drink
60. Ho« ectty ye*re did you coamnse beort _^__
v
(go to question 61)
(go to queetloa 61)
-------
Cl. Do 709 drtob efesetf S)Qt«» ?)
6lE. DW 700 drink wine in the peett t)Qlto 2)QY
4*
(go to o.ue*tlon 67)
62. Sov old tfere you eban you flret began Co
-------
ilave you, your «Het or any of four children ever had sny of this following diseases or conditioner (Flense check appropriate boK.)
67.
68.
69,
70.
71.
72.
7J.
74.
75.
76.
77.
78.
79.
80.
81.
82.
61.
A«tte« t)(jHo 2)QjTfee
y~>k /~*v
Bronchltle l)f~}Ho 2)(~JYea
Emphysema 1 ) Q Ho 2) O YeE
Tuberculosis- TB I)f~)Ho 2)r~)7oB
HononucleoEie-Moao- ^_^ ^_^
Kissing Dlaecee l>OHo 2) (J) Yec
Pneuftonifi OT ) No ^) C ) ^e®
Any other diseaee of
the respiratory
Eyaten
Specifyi l)^~^Ho 2) (^J f 06
Hepatitis or yellou ^^^ ._
jaundice l)f")Ho 2)(jYee
S~\ S~*L
Clrrhoalv Of") Bo 2)r"jYee
Any other liver dleesse _.
Specify: l)()No 2)()Ye6
Spondylitie l)QHo 2)OTee
Cout l)Olto 2>OVee
Rhcunntold Arthritis l)(~)Ho 2)(~)Yee
>^^ ^*N
Osteoarthrltla l)(~JHo 2)^J tee
Any other dlseaeen of
bonea and joints
Specifyi DliMo 2)<^)Tea
High triglyceridea l)(^) Ho 2)(~)lTe«
Menlngltla l)(~)No 2)OIee
8A.
85.
86.
87.
88..
89.
90.
91.
92.
93.
94.
93.
96.
97.
96.
99.
Hypertension or high
blood presoure t)f}Ko
^-^
High cholesterol l)^JNo
Angina pectorle l)(~)'lo
Heart attacli-KI or coronary 1) (_)NO
Stroke, cerebral accident.
(CVA)-heoorrhage( throaboelB,
efabolisia l)C ) "°
Any other heart or circula-
tory dlseaoa Specify) l)^~)Ko
Diabetes 1) [t Wo
Thyroid it Is l)(~}Na
Any other glandular disorder ^^^
Specify! I)l j Ho
Eye dleeasee
Specifyi I) O«o
Paoriasie I)Q~JHo
Eczema I)^JHo
llerpea totter-chlnclcc ^^^
(dennatitla) l)(_J)Ho
Penphigue l)C^Ho
Any other dlaeaaea of ekia ,^.
Specifyi . I) 1 j Ko
Multiple Sclerocia l)(~)Ho
2) C)
^^
»o
2>O
2)Q
2>O
»o
2)C~)
2)(~)
2>O
«o
2)O
2)f*^)
2,O
Tee
-------
tOO. Any other diseases of the
101.
102.
103.
104.
105.
106.
107.
108.
109,
110.
111.
112.
113.
114.
115.
116.
Specify:
Anefflia
Gastritis
Dicers (stomach or. duodenal)
Any other diseases of the
digestive system
Specify: _
Cancer of the breast
Cancer of the stomach
Cancer of the esophagus
Cancer of the mouth/tongue
Cancer of the large
intestine
Cancer of the rectun
Cancer of the trachea,
bronchus or lung
Cancer of the liver or
biliary passage
Cancer of the bladder
or urinary organs
Cancer of the skin
Cancer of the thyroid
Leukemia
l)0Mo
*«-^^
1)0 Ho
l>O«o
1)0 Ho
1)0 No
DO*
1)0 No
1)0 No
1)0 NO
D0No
DO"0
1)0 No
1)0 No
l)0Ho
1)0 Ho
1) 0No
1)0 No
2)0 Yes
2)0 Yes
2)0 Yea
2)0 Yes
Yes
2)0 Yes
2)0 Yes
2)0Ves
2)O Ve8
2)OYe*
2)0 Yes
2>0Yes
2)0 Yeo
2)0 Yes
2)0 Yes
2) O'Ves
117, Other cancer
Specify:
118. Mental retardation
1)0 No 2)0Ye«
-------
For ea4h of ttie diseases you have Ben cloned la your I Rail? Ket.:bcSv. ploo.-.i nail not
Name of disease/
condition
Homes of family members who have the
disease/condition:
Names and addresses of the clinics, hospitals, or
doctors consulted:
Date of diagnosis:
Name of disease/
condition
Names of family members who have the
disease/condition:.
Names and addresses of the clinics, hospitals, or
doctors consulted;
Date of diagnosis:
-------
For (Mich of the diseases you have mentioned In your faatly xeabare, please Call net
Name of disease/
condition
Naaes of family meabers who have the
disease/condition:
Names and addresses of the clinics, hospitals, or
doctors consulted:
bate of diagnosis:
Hame of disease/
condition
Kames of faatly members who have the
disease/condition:.
Names and addresses of the clinics, hospitals, or
doctors consulted:
Date of diagnosis:
-------
of tltc diseases you have ocncioncd in your family Eeatjers, ploaoe tell oe»
of disease/
condition
Names of family members uho have Che
disease/condition:
Names and addresses of the clinlco, hospitals, or
doctors consulted:
Date of diagnosis:
Name of disease/
condition
Names of family members uho have the
disease/condition:
Names and addresses of the clinics, hospitals, or
doctors consulted t
Date at diagnosis:
-------
For each of tlws diseases you tutve mentioned In your foali; rstmbere, pluaeo tall net
Kane of diaeisee/
condition
Names of family members who have the
disease/condition!
Kaaee end eddressee of the cllnlce, hospitals, or
docCori consulted)
bate of diagnosis:
Name of disease/
condition
(lames of faally meabera who have the
d tsease/cond It Ion : .
Names and eddrecses of the cllnlce, hospitals, or
doctors consulted:
Date of diagnosis:
-------
tor'fcacli of the diseases you have acnttoned in your Cosily senberu, please tell met
Name of disease/
Condit ion
Names of faintly mrmbcre who have the
d tscase/cond It ion :
Names and addresses of the clinics, hospitals, or
doctors consulted:
Date of diagnosis!
Name of disease/
condition
Names of family members vho have the
disease/condition:.
Names and addresses of the clinics, hospitals, or
doctors consulted:
Date of diagnosio:
-------
H.IVC fixi tivni li*i( any {>f tlm lot liiwlug
cyeptoce?
119. Severe fever
120. Extreme tiredness
121. Frequent or very bad headaches
122. Problems with uouth or throat
123. Any problems with ears or hearing
Specify:
124. Any unusual difficulty with eyes
or eyesight other than a change
of the prescription of glasses?
Specify:
125. Sudden weakness or heaviness of
arms or legs?
Specify:
126. Numbness in arms or legs?
127. Swelling in arms or legs?
128. Stiffness in joints or bones?
Specify:
129. Pains in joints or bones?
Specify:
130. Spasms of limbs?
131. Spells' of dizziness?
132. Have you fainted or blacked out?
133. Very strong heartbeats?
134. Irregular (fast, slow or
inconstant) heartbeats?
Ho
Tee
... .
II. iw ti I'ljiirnl ly
Uoctf (or did)
thle occurf
(deye, tion.the,
or year*)
II. iu IIIIIK tin
eynptone leecT
(days, aonthe,
or yc/srs)
Did you ccncmlt
a doctor!
No Yea
Doctor's nr,r,e & address
Year oast
recently
experienced
-------
Have you ever li«ii Buy of tlie fulluui
symptoms?
135. Have you hail pain, discomfort,
or trouble In or around your
licnrt ?
136. Itching of the skin?
137. Any unusual discoloration or
eruptions on the skin?
Specify:
138. Any problems with your stoooch
or digestive system?
Specify:
139. Swollen glands in your neck,
armpits or groin?
140. Have you lost 20 or more pounds
in the last five years?
(include dieting)
>8
No
Yes
How frequently
does (or did)
this occurT
(days, nontha,
or years)
X
{low long to
(or did) the
oymptoffls laat't
(iScys,, raontiiBj
or years)
X
Did yoi
a doctc
No
i consult
>r?
Yee
Doctor's naae 6 address
Year most
recently
experienced
-------
'HIE U3T SECTION OF TllE QUESTIONNAIRE RELATES TO YOUR WIFE AND CHILDREN. IF I HAY SPEAK TO YOUB WIFE, SUE MAY BE ABLE TO ANSWER THE FEW QUESTIONS
REMAINING.
1. How nctny netursl children do you have?
How ttuny eicp or adopted children do you hove?
Plcaee tell »c of ell pregnancies onJ outcomes of which you were the father/Bother in the sequence they occureJ.
Prcg.
t
1.
2.
3.
4.
5.
6.
Complication® in
till a pri/giumcy
(describe)
fhif como
Mlncacrlngo
Stillbirth
Live Birth
Other
Infl
ai
M
mts
iK
F
Data of Birth
Live
Birch
Weight
Hospital, clinic or
place of birth
Doctor 'a name
. H«ve any of your natural children died? (do not include Btillblrtha nor step or adopted children)
I) No 2) Yes (If Yes, please complete the following teble.)
I.
2.
3.
4.
Sex
M f
Data of Birth
Date of Death
Cause of Death
City end Stete in which death WEE reported
-------
142. Do you knou of any mental or physical abnormalities at birth (birth defects) in yourself or your natural children.
1)0 No 2)0 Yes (if yes, please complete the following table.)
i
(go to question 143)
1.
2.
3.
4.
(clat ionsMp to you
>clf Son Daughtei
[Check appropriate boxes)
Mving
Dead
Nature of Birth Defect (describe)
-------
141. Ba you O*« <" te«i «n««er qaeetlooe 144 end US)
(£o to question 146)
144. Khist DCtNod of birth control hive you used the Bate during your edult llfeT
uQpltl 2)QlUD J)QdUphrn» 4)Qfo«e or Jelly S)£) rhyths 6)Qcoado- 7)Qother (specify)r
ItS. *t what c0c did you begin ueing birth control rsethodel ________________
1*4. M whit Bge 414 you begta to neaetruite}
Utt, Etvc you (topped Reattruitlng? l)QJ)Ho J)^J)YeC
(1C Ho, go to qucotloa 149)
UT. At «h»t c0e did you stop MmttruttUfct
141. DU the ceototloa occur naturally or due to eurgeryt
2) (_) Surgery ^ Uhct u«c the rcx«oa for the eureeritt ,
Doctor'^ ofi&e *nd addreecT ^^
DCBC tddreee
149. Uh*t le (me) the ever BE e length of your nenetrucl cycle! dayg (Froa flret dey of bleeding Co fir it diy of next period)
ISO. da the everege, how long 1> (w*«) your period! d«yt
lil. In the p«et five yecr«( tut your blood (lou during oenatruetioa !)(_) Increaaedt 2)^J) dccrcoecdt 3) (~) atayed the eiuieT
1S2. Biv« (or did) you experience eny tbnortxl spotting or bleedln| between yout eenatrual cycleaT l)("jNo 2)fjTcc (go to queatlon
(go to ouectloa 154)
liJ. It yee, «ld you concult a phyclclanT ' l)(~) No 2)(")Yea Doctor'* naae t adJreee
^-"^ ^^ K*M«
eddreec
t$4. Have you__ucc<*, or dld^ou use, any taedicationc prescribed by * doctor for ccnetrual Irregulirltiesl
!B » Mama of medication
Queetlone 1-140 have been anawered by
Queatlone ttl-tS) have been anawered.by
Thank you for your cooperation In th|a health eurvey. Pleaae algn the conacnt fora and return the queationnaire In the eAcloaed self-^eddreaeed atanpcd
envelope.
-------
Mailed Questionnaire (Protocol 3)
HKIIKMHKN IIKAI.TII MIKVKY
INSTRUCTIONS! Please annucc the following questions and
return tlie questionnaire ultliln five days, Use the enclosed
stamped, self-addressed envelope for mailing. Your answers
are confidential end will only be uoed for research purposes.
Your Identity will not be associated with tho survey results.
If you cannot give an exact answer, provide your beat collaiat
I. What IB your current address?
(street)
(city) (state) dip)
. 1, What 1* yout «CK? l)Hale 2) Fcraale
3. What Is your racet J)^~~N White 2)(~") Black 3)f~) Ilinpanic 4)^) Aaerlcen Indian 5)^) Other
4. What IB your birilxIsteT _ / _ / _
nonth day year
5. What 1« your Social Security number? _ - _ - ______J____
6. What Is your n.trltal status? 1)T JSlngto (never married) 2)( J llnrrled 3)f) Eepuratcd or divorced *)( ) Uldowed
7. Du you presently liold a connerclsl fishing license? l>r^)no 2)f J) tee— »• Appro* Irenlely over wl»t yearo have you held e license? /
• ^"^ (ron to
r • (go to question 10)
8. nid you hold « comerelal ftcliing license In the paetl DV^^yMo 2)f) Yea— *- Approximately over v\t»t yoara did you hold ft HccnoeT _ _ /
frotn
to
(go to question 10)
-------
t. Kcve. you ever beea c crew uember for, PC e partner ulth,«n individual owing e cotwercliit finding licence?
t)lfc 2}Qtee—1> Over tthet ;o«re.heve you been fi crew neaber or partner! / (go tp quencion iO)
i
(roe to
*
(If Ho, pleeee stop here end return the queetlponeire en eoon re pooelble in the eddreeced envelope. Thank you for your cocpcrctloo.)
(0. Indicate (rtth ea "K" OB the Cficolced Ecp where you corotctclcUy Clahed cost often. (Tou ccy indicate core ttusn one locctloo).
It. le coBiserctel fCehtag your current
pccupetlooT
12. Wat ie your current full-tiae pccupitlonl
13. Wea coonerciel fishing ever your full tteo occupation?
1 ) C_) Ho ——pOver whet year* heve you been e pert-ttee
fro* tp
(go to question 14)
2 )^~)Yee —t- Over uhat yea re were you full tbtef /
{taf ta (go to question 14)
• Over what years have you eoanercially flehed full tine?
front to
(go to question 14)
14. Do you currently owa c eport flehing llccnceT i)^~^Ho (go to question IS) 2)(~}YcB (go to question 16)
IS. Htve you ever oteatd e eport flehine licencet l)QMo (go to question 17) 2)QYe£ (go to question 16)
16. Indicate with «a "0" on the enclosed cep where you eport fl»h(ed) noat often. (You aay indicate more than one location.)
-------
17. To the best of your fcnouloJge, ptcauc Hot llic nanca and adiircutica of full or putt-time crew mcraiiero? (|iaHt iinJ (iruKu
Kane
Street
City
State
Zip
Employed
From
To
18. Do (or did) you consume any of the fifth you cetch commercially or se e sport fisherman? • )(_) YeB 2^
19. Approximately hov stany of your actlt contain fish caught (by yourself or « friend) from the Great l-afcesT
1) per week (or) 2) pec month
20. How many of those nealo Include the following types of f ieh:
_ Salmon _ Walleye _ Rough fish _ Hot them Pike _ Lake Trout _ Base
_ Perch _ Burbot _ Chub _ Lake Vhltefinh _ Smelt _ iJike Herring'
21. Approximately hou aany years have you consuned Greet Lakes fish with this frequency?
years
M°
""' C° tO 1uascton
_ Pan fl«h
Other I (epectfy)
22. Approxlaately hou nany of your deals contain fish caught (by yourself or a friend) froa waters other than the Great Lttkeet
1) _ per week (or) 2) _ per nonth
23. How nany of those deals contain the following types of fiah?
_ Saloon _ Walleye Rough Klsh _ Northern Pike _ Lake Trout _ Bess _ Pan ri«h
Other Trout
Perch
Burbot
Chub
Lake Uliiteflsh
finelt
Lake Herring
Other Trout
Othert
specify
-------
2*r tpproEtcxtcty host seny yccrc here you contused inland uet$r flch with thle frequency? years
25. How often da your foully tn«ber« eet £l«hf
tttfei Sonei Daughters?
t) ^J Ae of te& ec myeelf 1) (_) JLe often e.e eyeelf I) (~y A
2)Qhore often then eyeelf 2) (_) More often then ayeelf 2) (_) Hore often than eyeelC
3) Q Lece often then tsyeell 3)(^)L«BE often thea ayeelf 3^O L*ee o£Ccn th*'n ^yeelf
J&)("J)Docc not cpply 4)(~JOoe8 not cpply 4)^~^Doee oot ipply
5)Q Hevet 5) Q Hever 5) Q «ever
16, Approiclractely bow rjiny pound e of ft eh have you coneuaed per yeec?
(eeEuset I fleb rstcl ccjuclc % pound of fteh) pounds
27. Uhct perceatsgee of flch you cetch end consume ere prepared by the following stcthodei
1) Broiling X
2) P«n frylne X
3> S«K>feed t
«) Boiling X
5) Poached X
6) Other I
-------
28. Do you procatly (oak* cig«rett«>t I)Q«o 2) Q T««
28». Did you cnok4i
in the peotf
29. Uou old were you whea you first began to coofce cig*i*ttacT
30. What le the nvercge nuaber of clgerettca you presently eaoke per
31. How n«ny jrcere Mve you taoked elgcrettec wnth thle fcequencyT
(go to questIon 3*)
(go to qucotloo 34)
32. Bow old were you when you flicc begcn to evofc* cig«rettA*T
33. BOM anny yeerc did you eroke eigcrettetf
(go to Question 34}
-------
34. Do you presently snoka a pipe? l)Qlto 2)
). How old were-you when you first began to SB oka a pipe?
36. What IR tha average number of pipefuls you presently smoke per day?_
37. How taany years have you smoked & pipe with this frequency?
(go to question 40)
34a. Did you smoke a pipe In the past? l)(~)No 2) (^) Yee
i
(If Ho, go to question number 40)
38. How old were you when you first began to smoke
e pipe?
39. How many years did you smoke a pipe?
(go to question 40)
-------
40. Do you presently smoke cigars? l)(~)No 2)
.40a. Did you smoke cigars In the past 1) ( ) Ho
(If No, go to question number 46)
4.1. How old were you when you began to smoke cigars?
42. What is the average number of cigars you presently smoke per day?
43, How many years have you smoked cigars with thta frequency?
(go to question 46)
.44. How old were you when you began to smoke
cigars?
AS. How many years 414 you smoke cigars?
(go to question 46)
46. Have you ever chewed tobacco regularly? !)(_) No 2) Q~J Yes
47. Have you ever used snuff regularly? l)^^No 2) (J Yes
.48. Have you ever smoked non-tobacco products regularly? 1) f~*\ Ho
Yes
-------
ti-juor tc cooeldered Co includei Cinj Wbt«fce]r[ Scotch; Vodfce| Bust Brendjr
(A dciak •» 1 co l
-------
55. Do yon drink bccrt l)Qno 2) Q Ie« .
56. Hou old vere you when you fleet began to drink heart
57. Uhat Is the average mmber of beers (12 or, bottle*) you presently consume
per veekl ________
58. no« nany years have you coneuned beer with tht» freipiencyt _
(go to question 61)
55*. Did you drink beer to th^ pcctt l)QHo 2) Q Tec
to qu
(|o to queatlon 61)
59. ROM old were you when you flret begta to drink beerI
60. Hou cumy ye«ru did you concuae beerT ______
(go to quectloo 61)
-------
it. Da yoo drtofe tfteef
2)Qf«e-
6le. Bid you drink vine in the peett l)Qlto 2)Qt
4*
(go to question 67)
61, How old were yea whea you fltet begea to drink wtncf _
61. Whet le the Evcc«$e nusbcr of wine drinks you bsve pec ecefef ______
64. lion tsony J«CCE hcve you coasuaed nine with this frequencyT
~~ (go to quecstloa 67)
65. How old were you when you fleet begun to drink wlne'T
66. How o*oy ye*r« did you eooeuee
(go to queBttoo 67)
-------
Have you, your vif«, or ex>y of your children ovor had any of the following diseases or condition*. (Fleece check appropriate box.)
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
,i.
78.
79.
80.
ei.
82.
8).
Asttsui 1)( JHo 2)1 )Yes
V — / '. v_y
Broachltis I)(_)Ho 2)C_)lem
_^
Eaphysea* l)(~)Ho 2)(J)Vc«
Tuberculosis- TB l)f~)Ho 2)(~)te«
Manonucleosis-Mono*
Kissing Disease l)f~)Mo 2)r~)Yes
Pneumonia l)f~)Ho 2)Q~^Vee
Any other disease of
the respiratory
ay 01 en
Speclfyt Of^No 2) ("^ Yea
Hepatitis ot yellow ^^^ ^^^
jaundice I) f") No 2)f}Yee
Cirrhosis "(2)"° 2* O *"
Any other liver diseaae
Specify: l)(_)Ho 2)lJYes
Spondylitls uQ"0 2>Olfee
Cout "O"0 2)OV"
Rheunatoid Arthritis 1)(2) No 2>OTe*
Osteoarthrltls I)("jHo 2)f"j Yes
Any other disease* of
bones and joints
Specify: I)i)Ho 2)^J) Yes
High trlglycerides l)(_~) Ho 2)/*jYes
Meningitis l)Ono 2)(^)Yes
8A.
85.
86.
B7.
88.
89.
90.
91.
92.
93.
94.
93.
96.
97.
98.
99.
Hypertension or high
blood pressure l)f~^No
High cholesterol l)fjNo
Angina pectorin l)(~^)No
Heart attach-Hi or coronary 1) C_JHo
Stroke, ctrcbral accident.
(CVA) -hemorrhage, throobosls,
eabolioa ')C3 "°
Any other heart or circule- ^^^
tory disease Specif yl O("jNo
Diabetes 1) I i Ho
Thyroidltle l)^)No
^~^
Any other glandular disorder _.
Specify, l)fl Mo
Eye diseases
Specify; UtlNo
Psoriasis O^~)HO
Ectena l)ONo
Herpes zoster-shingles ^_
(dermatitis) l)(~^Ho
Pemphigus I)(~)NO
Any other diseases of skin _.
Specify: DlMMo
ffciltlple Sclerosis 1)ON°
2)OTee
2) (~) Yes
^^
2)O™
2)OY"
2)Q Yes
2)0*"
... /~\ y
2)O*««
\—/
2>O1t««
\~s
2)Ote«
' V-/
2)0*"
2)OTe«
2>O »••
2)QY"
2)^~) Ye«
«O'-
-------
100. Any other diseases of the
nervous ay at ess
Specify:
101.
102,
103.
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
115.
116.
Anemia
Gastritis
Ulcers (stomach or duodenal)
Any other diseases of the
digestive system
Specify:
Cancer of the breast
Cancer of the stomach
Cancer of the esophagus
Cancer of the mouth/ tongue
Cancer of the large
Intestine
Cancer of the rectum
Cancer of the trachea.
bronchus or lung
Cancer of the liver or
biliary passage
Cancer of the bladder
or urinary organs
Cancer of the skin
Cancer of the thyroid
Leukemia
^** *>—iS
I) Q»o 2)0 Yes
1)0 No 2)0 Yea
D0 No 2)0Y«s
t)0Ho 2)QYes
x. — f v_«x
E)0No 2)0Ycs
1)0 No 2)0 Yes
1)0 No 2)0 Yen
I)0No 2)0 Yea
,, ^"^
1)0 No 2)0 Yea
1)0 do 2)0Yeo
1)0 Mo 2)0 Yea
1)0 No 2)0 Yes
1)0 No 2)0 Yes
1)0 No 2)0 Yea
l)0Ho 2)0 Yes
1)0 Mo 2)0 Yes
117. Other cancer
Specify;
118. Mental retardation
I)QNo 2)QYes
-------
For each of the diseasea you have checked on the preceedlng two pages please complete the following tables:
Name of disease/
condition
Please list the nanc(e) of family
mcmber(s) with this disease/condition
Flcaoe list the name (a) and addrcssCes) of the
cllnic(s), liospttal(s) or doctor(s) consulted
Please give the date
of diagnosis
Name of disease/
condition
Please list the name(s) of family
member (a) with this disease/condition
Please list the name (a) and addreas(ea) of the
cilnlc(B), hoapltal(a) or doctor(s) consulted
Please give the date
of diagnosis
-------
For each of the disease* you have checked on the preceedlng two pagee pleese complete the following tables!
None of disease/
condition
Please list the n«me(s) of family
aeaber(a) with thie diseaae/conditlon
Please list the naae(s) and adJress(cs) of the
clinlc(e), hospital (o) or doctor(a) consulted
Please give the date
of diagnosis
Name of disease/
condition
Please list the name (a) of family
ueober(s) with this disease/condition
Please list the nane(a) and addresses) of the
cllnlc(s), hospitel(s) or doctor(a) consulted
Please give Hip date
of diagnosis
-------
For each of the diseases you have checked on Che proceeding two pages please complete the following tables!
Name of disease/
condition
Please list the name(s) of family
member (B) with this disease/condition
Please list the naoie(s) and address(cs) of the
clinic(e), hospltal(s) or doctor(s) consulted
Please give the date
of diagnosis
Name of disease/
condition
Please list the name(s) of family
member («) with this disease/condition
Please list the naaie(s) and sdJrcsii(cs) of the
clinlc(B), hoBpltal(s) or doctoc(s) consulted
Please give the date
of diagnosis
-------
For each at the diseases you have checked on the preceedlng two psgea pler,sc complete the following tebledl
Name of disease/
coodlcioa
Please list Che nsme(s) of family
BCjeber(s) with thta disease/condition
Please list the nane(s) and address(es) of the
cllnlc(e), hospital (a) or doctor(s> consulted
Please give the date
of diagnosis
Name of disease/
condition
Please list the name (a) of family
member (s) with this disease/condition
Please list the name (a) and addrese(es) of the
clinlc(o)> hospital(e) or doctor(s) consulted
Please give the date
of dlagnoslo
-------
For each of the diseases you have checked on the proceeding two pages pleaee complete the following tables!
of disease/
condition
Please list the narae(s) of family
member (a) with this disease/condition
Please Hot the namc(B) and addresa(eB) of the
clinic(o), hOBpltal(fi) or doctor(s) consulted
Please give the date
of diagnosis
Name of disease/
condition
Please list the name (a) of family
member (s) with this disease/condition
Please list the name(s) and address(es) of the
clinlc(s), hospltal(a) or doctor(s) consulted
Please give the date
of diagnosis
-------
ll.iv« you over licul any of the following
cyBptaas?
119. Severe fever
120. Extreme tiredness
121. frequent or very bad headaches
122. Problems with mouth or throat
123. Any problems with ears or hearing
Specify:
124. Any unusual difficulty with eyes
or eyesight other than a change
of the prescription of glasses?
Specify:
125. Sudden weakness or heaviness of
arms or legs?
Specify:
126. Numbness in arms or legs?
127. Swelling la arms or legs?
128. Stiffness In joints or bones?
Specify:
129. Pains in joints or bones?
Specify:
130. Spasms of 1 bobs?
131. Spells of dizziness?
132. Have you fainted or blacked out?
133. Very strong heartbeats?
134. Irregular (fast, slow or
Inconstant) heartbeats?
Ho
Yee
- -
._. .
How fri-qurnlly
does (or dlJ)
this occur?
(days, Booths,
or ycnrc)
lluw lunB Jo
(or did) Che
synptosaB lest?
(deya, »onthE,
or yean;)
Did you consult
a doctocf
Ho Tee
- - -
Doctor's na»e 4 odd re si?
Year no at
recently
experienced
-------
Have, yuu ever hail any of tlio following
135. Have you dad pain, discomfort,
or trouble in or around your
heart?
136. Hcliing of tlic skin?
137, Any unusual discoloration or
eruptions on the skin?
Specify:
138. Any problems with your stomach
or digestive system?
Specify:
139. Swollen glands In your neck,
armpits or groin?
140. Have you lost 20 or more pounds
in the last five years?
(Include dieting)
No
Yen
How frequently
does (or did)
this occur?
(days, ttonclia,
or years)
X
How long do
(or did) the
eynptoua last*
(dtye, Bontha,
or years)
Did you
a doctc
No
i consult
rl
Tee
Doctor 'a name & address
Year most
recently
experienced
-------
The following section is to be filled out by the wives of nfile fishermen/or/by fishervooen. If the wife 1* deceased, would the husband complete
thle lection to the best of bis knovledge.
141. How raany natural, children do you hsve?
How cany step or adopted children do you have?
Please list ell of the pretmanclea and outcomes of which you were Che father/mother in the sequence they occurred.
Preg.
t
1.
2.
3.
.4.
5.
6.
Complications in
this pregnancy
(describe)
Outcome
Miscarriage
.Stillbirth
Live Birth
Other
Infants
sex
H F
•Date of Birth
Live
Birth
Height
Hospital, clinic or
piece of birth
Doctor's naae
I4la. Have any of your natural children died? (do not Include stillbirths nor step or adopted children)
1>ON° 2)Qlfes
-------
142. Do you know of any Dental or physical abnormalities at birth (birth defects) in yourself or your natural children.
t)ONo 2)Qvc9 (If Yes, please complete the following table.)
I
(go to question 143)
1.
2.
3.
A.
iclatlonshlp to you
">clf Son Daughte
(Check appropriate boxes)
Uving
Dead
Nature of Birth Defect (describe)
-------
143. Do you use e birth control Bechodl 1>O *° 2)O'*«• r")other (epecify)i
145. At what age did you begin using birth control methods?
146. At what age did you begin to menstruate? _______^_^_________
146a. Have you stopped neostruating? I)("~)NO 2)C~~)^ea
(If Ho, go to^questlon 149)
147. At what age did you etop menstruating?
148. Did the cessation occur naturally or due to eurgery?
1) Q_) Naturally 2)^) Surgery fr Whet was tha reason for the surgery?_
Doctor's name and address?
name aodresH
149. What is (was) the Average length of your oenstrual cycle? dayo (Froo first day of bleeding to first day of next period)
ISO. On the average, how long Is (was) your period? days
151. In the past five years, has your blood flow during menstruation 1) ^"^ increased? 7)^^ decreased? 3) C~} stayed the same?
152. Have (or did) you experience any abnormal spotting or bleeding between your Eenstrucl cycles? O^jNo 2)^") Yes (go to question
(go to question 154)
153. If yes, did you consult & physician? I)(~)NO 2}f"jYes Doctor's name £ address
none sddreee
154. Have you used, or dldyou u&e, any Dedications prescribed by a doctor for aenstrusl irregularities?
I) No 2) Yes > NaBe of medication ...
Questions 1-140 have been answered by
Questions 141-153 bave been answered by
Thank you for your cooperation in this health survey. Please sign the consent fora and return the questionnaire In the enclosed self-addressed stamped
envelope.
-------
Reproduced from
best available copy
TwTN;p--^rr;,J'
$^*^<''sV.#"
<£?ti£^-:L-T> vV.
-------
Proxy Questionnaire
FISHERMEN HEALTH SURVEY
o aid In our understand ing of Mr;/Mra._
Instructions: _ _
fishing habits and health, we ask that you complete the following quest-
ionnaire answering ttie questions as you believe he (she) would. Answer
as many questions. as possible from your knowledge of Mr./Kre.
!»F oinru:Ti;i> BY OFFICE)
If you don't know how they would have answered a question, write "don't
know". Start with question number two.
I'roxy (nlt-rvlca >if Mr./Mrs.
llils proxy Interview Is being completed by:
'.''ill i-; i hi- rrlat lunshlp of the per Bon couplet Ing the Interview to Mr,/Mrs.
I. AJdriHs of p-'rson completing interview?
Street
Relationship
City
State
Zip
ite 2) O Blabk 3)
1. HI., i ; «;•» hls'hcr sex? I)
3. '.M,,, - ...,., hi Whcr race? i)
i>. HlMt u'.is lils/lior l»lrcl«lnte?
month day year
r>. Hh.it ".is his/her So<:i,il Security ituraber?
'.. Wl,.ii w.is Ms/her martial status? I )( — Jsinfcle (never married)
Hispanic
«)'
Awericsn Indian 5) |~~I Other
Married
or divorced
Widowed
7. IH.I tic/Shi? liol.l n current coranerclal fishing license? 1)C3 ^ 2)1 JYes Approximately over what years did he/she hold a license? _ /
frO1" to
S. Ui
-------
ID. I ml liN
a= jitf/jtiu ever s n-eu m.-mlicr for, or ,1 pnrlnor wltli, an Indlvldunl owning a commercial fishing liccnect
I )[^ | il.i Z)| I Yen - ^ Ovot wluit years was lio/sfic a crew »embcr or pnrlnnr? /
I from to
i If UK, |>U':>!ip stoji licrr ami return tlic questionnaire ns soon os poecllilc In the addressed envelope. Thank you for your cooperation.)
ml liNiir wild ,111 "x" on llic enclosed nap where lie/she commercially fished most often. (You rany Indicate more than one location).
-------
"• Was ...rj.ocirl.il fishing hid/her I) |_j
curr.nl dil I-tine occupation?
12. Wliae wait lita/tier current full-time occupation?^
13. Has commercial fiohing ever his/her full-tlae occupation?
2) L_J Yea ~* Over what years was ne/ahc full tl»e? /
from to
I) LjNo—*0ver what yenrs vaa he/she B part-tlBe'fiahernan?
(go to question 14) '
Ifes -*0ver what years did he/she comaerclally fish full tine? /_..-..
from to
from to
(go to question 14)
I'.. OiJ li.-,'jiln- >:urronily own a s|>ort fishing license? 2)f~] Yes t)t] No 15. Old he/she ever own a sport fishing license? 2)
I
|J. Indicst. wltli an "0" on the enclosed raap where he/she sport fished nost often. (You Bay indicate more than one location )
Yes 1)Q **»
(go to question 17)
-------
\l. To tlii- ki'-'jl <>' v kni'wIrilKi', plr:i:ir llstl I In- n:iinr!i nnil nJdrcsM-M uf hlx/hrr lull or pot l-i line crow nrnlii-ruT (|>;i!iL ;nil—-I No (lf "°« 8° to «!"«-•« •"" 2fl )
ft. ,-,n.i..xlm.it.-ly how nnny of his/her meals contained fish caught (by themselves or a friend) from the Great Lakes?
1) |n-r week (of) 2) _____ per moutn
Ai. ll.ivi mnuy "1 ilioso mrnls Inclinlc the following types of flshi
__r..Tln.on Walleye Hough Cisli Northern Pike Lnke Trout Oass ^ I'on fish
(itlior Trout Perch Burbot JChuh Lnke Whlteflsli Smelt Lake Herring
01 lu-i : (r.|n-c I fy)
21. A|.,-....«ira.ncly Imw n.iny years did he/she consume Great Lokca fish with thl« frequency? years
". A,., r...ii,,;,i,.|y Imw in:iny of his/her meals contained fish caught (by themselves or n friend) fro- wotero other than the Great lakes?
'> !•«•«• »cck (»c) per month
Reproduced from
best available copy.
-------
21. How many of those meals Include die following types of fifth)
SaUon Walleye Bough fish
Other Trout
Northern Pike
Perch
Burbot
Chub
Oilier: (specify)
14. Approximately how many yeare did he/she consume inland water fish with this frequency?
2*>. II.>w often illd his family members eat fish?
l>n/L': Sons!
1)1 I As often a» he/she 1)I | As often as he/she
2) 1—I More often thrtn he/she 2)] 1 More often than he/she
1) I I Less often than he/she 3) I | Less, often than he/she"
•'•) I 1 Doc.s not ply 4) | I Does not apply
•/ f.M »«'<'«T 5) O Never
Lake Trout
Lake WMtefish
years
Baas
Pan fish
Smelt
_Lake Herring
Daup.htersi
1)1 L.Aa often as he/she
2)t. 1 Horc often than he/she
3)[_J Less often than he/she
1)(, J Docs not opply
5) 1 ) Never
^6- .i|'|n i>xint.il r.ly how m.iny pniinilo of fish did he consume per year?
(assume: 1 fish wciil equals 'j pound of fleh) . pounds
11. Wh.it percentages of fish he/she caught and consumed were prepared by the following methods:
I) Broiling I
2) I'.in frying Z
'l) Smoko.l I
I.) Hoi line Z
'») I'oochoil . 5!
I.) Otlicr Z
-------
28. W.i-; liu/shi- ••> clu.-irutlc smoker
.•it the time of dc:ilh?
29. How oji! w.is he/she when he/ohe fleet began to smoke cigarettes?
30. What was the average number of cigarettes he/she smoked per day?
31. How aany years did he/she smoke cigarettes with this frequency?
(go to question
W.ir. he/she ever a cigarette smoker? l)j \ Ho 2)| | Yea
I
(1'f No, go to question number 34)
32. How old was tie/she when lie/aha first began to smoke
cigarettes?
33. How many years did he/she omoke cigarettes?
(go to question 3$)
-------
I'l. W.i:i lir/sho a pipe smoker at
I he t l»c of death?
I)
2)f~7lYes
NX
Via. Was l.c/sl.e ever a pipe snokerT l)C~]No
(If Ho, go co question number 40)
35. How old was he/she when he/she first began to smoke a pipe?
36, Hint Js the average number of plpefuls he/she sacked per day?
37.. How many years did he/she smoke a pipe with this frequency?^
(go to question 40}
3Q. How old was he/she when he/she first began to
smoke a pipe?
3$. How cany years did he/she smoke a pipe?
(go to question 40)
-------
Reproduced from
besl available copy.
the tlm,- jf rfr.illi?
'•Oil. W.is he/she ever a cigar smoker? l)( Jj No 2)1 1 Yea
i
(If No, go to question number 46)
How old was lie/she when he/she began to omokc cigars?
What IB the average number of cigurd lie/olio smoked per day?
How many years did he/she smoke cigars with this frequency?
(go to question
How old waa he/she when he/site began to
smoke cigars?
How many years did lie/she smoke cigars?
(go to question 46)
*<•. hi.I liu/slir '-vet chew tobacco regularly? Oi—J No 2){ JYes
47. DiJ he/-:l WIT us,- snuff rry.ularly? 1)1 | No 2)| \ tee
'•'J. '(Mil h<-/>ili'.- OVCT smoke non-tobacco products regularly? 1)J^ | "e 2)|^ jYes
-------
Hard liquor is considered to. IncludeJ Gin, Whiskey, Bourbon, Vodka, Eu»e Brandy, or Scotch
(A drink-1 to l>i oz, of alcohol or a mixed drink contelning this amount.)
*<). w«s lnVslic a liard liquor l)QHo
.li iiiK-r ;it Uwr t li»o of
ilc jili?
f
Yea-
''(If Ho, f.o to question number 55)
50. How old uae he/she when he/she first began to drink hard liquor?
51. Hhat is the average number of hard liquor drinks he/she consumed
per week?
52. How many years did be/she constant hard liquor with this frequency?
(go to question 55)
>>')n . Vas he/she ever a hard liquor consumer? 1)L iHo 2)| J.Ves
S3> Hou old was he/she when he/she first began to drink
hard liquorT
*• lion raany years did he/she consume hard liquor drinks? _____
(go to question 55)
-------
>, ',•!.>:: hc/shr n I.eer I) |
-------
r.l. '..'.is lie/she a wtnc
Jr inker at the tine
ni
-------
Reproduced from J^
besf available copy. UsHjP
Dill (ltr/::hc) , -lih: or 'her PIHMISC
67.
Crt.
69.
70.
71 .
1 -,
/ .. .
73.
7 A .
73.
76,
77.
76.
7-J.
so:
az.
«i.
S', .
s-i.
Hf,.
ASttlln.l
Ri 'oiifhil 1::
ti»|.f.y;.-wi
TMl.orriilo:;l>:-TI>
Itoitoiim: 1 «i)!i 1 s-Hono-
Klsslllf, DIliCJIHC
v
Any tilhur tllncafio of
tin- respiratory
Specify:
Hep.it i rlj: or
yellow j.tnmllce
Cii r hunts
Any o t Itt-r 1 1 vCr
clir.rds.-
.'•PCC i I'y: „„_
Spoitdyl il •:•
C..-,t
Mlii-iiHi.iiuiil An drli is
ll-;|.-...ir\ hfll Is
.-..-.»
Itottt": t
i— ___| HO
nO NO
f— ,.^
DLU No
DC] NO
on NO
DtU No
n CD NO
nCZJ NO
l)C3 No
i) cm NO
DCU No
I) CD No
l)tZ3 No
1)1 I No
his/her children ever
2) IZ3 Yea
2) EU Yes
2)C3 Yea
2)C3 Yes
2) d Yes
1 — 1
2) L~~*J Yes
2) CD Yes
2) CD Yes
2) CH Yes
2) \ Ij Yea
2) L3 Yea
2) CJ Yea
2) CD Yc.3
2) CD Yes
2) CD Yea
2) tZI *«•
2){~~I Yes
2)CU v«
2)Q »«•
2)CU Yes
luwe any
07.
88.
89.
90.
91.
92;
93.
94.
95.
96.
97.
98.
99.
100.
101.
102.
103.
104.
of the following diseases or
Heart attnch-MI or coronary
stroke-cerebral accident
(CVA)
Hemorrhage- thrombosis-
embollsm-ony other disease
of heart
Or citculotory system
Specify:
Diabetes
Thy ro idle Is
Any other glandular disorder
Specify:
Rye diseases
Specify:
Psoriasis
Eczema
Herpes zoster-shingles
(dermatitis)
Pcmphtnr.»8
Any other diseases of skin
specify:
Multiple. Sclerosis
Any other diseases of
nervous uystom
Specify:
Anemia
Gastritis
Ulcer«(stom,ich or duodenal)
Any other dinenne." of Che
digestive system
Specify:
conditions.
l)Cm No
l)f~1 No
1)1 1 Ho
onu NO
1)D No
t)dHb
ntm NO
1)CU No
D CD No
n cm NO
O CD No
on NO
ndJ NO
D 1 No
1) J Ho
nUD NO
1) { 1 No
nl~~t NO
(Please check appropriate box)
2) i 1 Yes
2) QYes
2) {^JYcs
2) CU Yes
2) ["""{Yen
2) CJ Yes
2) £3Dves
2) j /Yes
2) | JYca
2) f~~| Yes
2) f~~l Yrs
2)r— jvcs
2) m Yes
2) LH Yes
2) r~~|Ycs
2)C3^'
2) i | Yes
-------
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
115.
116.
117.
Cancer of the breast
Cancer of the stomach
Cancer of the esophagus
Cancer of the mouth
or tongue
Cancer of the large
intestine
Cancer of the rectum
Cancer of the trachea.
bronchus or lung
Cancer of the liver
or biliary passage
Cancer of the bladder
or urinary organs
Cancer of the skin
Cancer of the thyroid
Leukemia
Other Cancer
Specify:
Dp~) No 2) CU Yc»
0 LTD -NO 2)[
l)QNo 2)1
OtHjNo 2)]
nlHl No 2)
l)CZj No 2)
HI Yea
HjYes
ID Yea
LjY6ff
[YCS
I>C J Mo 2) CHI Yes
l)CHI No 2) 1 jYes
l)I~"t No 2)
nl~l HO 2)
l)CHNo 2)
nCUNo 2)1
~~]Yes
ZU Yes
j Yea
~~| Yes
I'll 1 Un 71 f i Yon
-------
For each of die diseases you have checked on the proceeding two pages please complete the following tables:
Name of disease/
condition
I'lea-x: 1.1 nt the nai'ie(r>) of family
member (n) with lliln tl 1 r;i:n'ic/coiul 11 Ion
Ploncc .Mat. Clio MOIHC(B) and nJdreBs(ea) of tl>o
clinic (ft), lioBpJtnl(n) or Joctor(s) conoiiltetl
Please give tl'C Jntc
of OlngnoBls
Karae of disease/
condition
list the nau
-------
, i i... .11 :..;••!.-» you imvr. chccko! on the prceeeJinn two pngcn please coraplete tlic following
I'lc.inc list the nonc(s) of family
mi-mliri (r.) wlili thin il l.icnsc/cond ttlon
Please list the ooeseCo) and ad
-------
II,.- ,11:
yon tn»vt>. clmcknl «>i> tl>ti
tvo pnacii ulcnoc complete the folloulnR tnlilent
I'lc-Tr.c Vlst llic unmc(r.) of family
mi ml. cr(.':) wllli tlilfi dlncnoc/<*'«vJltiou
ric.iBc list tlic nnmc(p) and ndclroBs(en) of the
cllnlc(s), hoo|)ltnl(B) or doctor (s) consulted
Plcnsc give the Jntc
ol Ulaynosie
rl<-:i-u- 1 IKI the nnn<:(i;) of tnmlly
i»<.-int>rr(r.) with tills it 1 Pcasc/con.) 11 Ion
ric.noc Hat Lite name(B) nn«- f,ivc Clio
-------
i „ I, ..I it,. ,11,.,-.,•!,••. ymi |,ntfi< dtrtTkril I'll llul pr«!t:ufillt»n two |inRrii |.|rnnc c««|ilr'n tlio follnvinR Utitaot
I'lcnno list tlio namc(s) of family
ntvmlx-i (R) wttli tliln d Isonsc/condltlon
Please Hat the nanc(s} and adclreas(ea) of tli«
cUnlc(B), tK>8pltal(a) or doctor (s) consulccd
Please give tlic date
of diagnosis
l'li':i>«- lint the iinmi:(n) of Cninlly
«. tolicr (r.) wJtli this disease/condition
PJcnso Use Clio na«e(n) and nildrcssCos) at the
clinic(s), lioapital(n) or doctor(s) consnlte
-------
I 1:1. H..V iii.iitv .-liil.li.-it .11-1 lir/.ilii. IMV.-Y ( ___ iiiiliirnl clilliiiru) (ind {
iili-ii ur mluplcd children)
rii-.-ir:.- ! l::i :i 1 1 (In- |.r c|.|i.tuc I on nnd nnCconcn of which lic/sho wns the C nl licr/raotliec In the acqucnce they occurred.
I'l vgttaticy :;.M|m:tn c
IIIIMllx' 1
1
?
1
/,
S
6
Ulnc.-irrlnr.c
StlHblrl.li
Live nirth
Other
Infnuta
Sex
H F
Date of Dlrth
Ltve
Dirth
Weight
Iloepitol, Cllnicfor
plnce of birth
Doctor's Nnme
119. ll.-ivi- any of his/her natural children dted? (do not include sttllblrfhs nor step or adopted children)
1) | J Ho Z) [ | Yeo (I£ yee, plcnsc complete the following table.)
..JL
nx
r
n.itc o( nirth
Dote of Death
;
Cause of Death
City and State in which dcnth wan reported
-------
i.-«. Ki >..u tiiuw ut miy niciilitl in (.liyiitcol aliMonunUtiea ot birth (birth defects) in him/her or his/her natural children.
1) | | Ho 2) 1 | Yeo (If yen, pleoee complete the following table.)
(go to question 121)
He I.It
lllciirlf/lict.iylf
lonnjijp
Soil
to lilra/hcr (check approprloce boxes)
D.THP.IltOl
Living oc Jcatl
—
Nature of birth Defect (describe)
-------
Ill 1 t.r
I :i . SevKre fevci
fU. €it»O"i! t 1 redness
1 11. 1
.:«. •
1 iS. A
t,
i oouful ur vcty li^il hi..;tilnchrn
rol»U-n>s ullli uuiiit h uf Itiruut-
I'fCIFT:
' 26 . Auv unir:iia ) (1 1 1 f 1 1:11 1 (.y with ryes or
eye', if.'" otlii'i III. in ,1 cli.mc.e of the
pi c sc r 1 |»l 1 mi i>f p. l.i.ssrn?
Sri C I l-'v :
117. Sudiirit wr;iliiiir!in or lic;ivliiCRfi of
.lin:-! LI ll-|'.:.''
1 -1 . 1
luiiiltii.':;:) In .-il'mrt nr lop,«?
>%J«- I 1 1 IIJ*. Ill .irill!: 1M' 1c|',!>?
HO. S< i 1 f IHT-.S In juliils or Imiicn?
srttirv:
1 \l .
'.lin^ In |(iliil:: or liunr:;?
1 12. S,,.r:i.i-: of 1 Iml.s?
IJ'- Hut l.o/shc: li.'iv spells of dizziness?
It/,. IM'I >iu/';ho f.ilul ur lilack out?
IIS. Vcvv slrouf, hour Ideal K?
Reproduced from
best available copy. ^^^
Vlii|il I
No
—
Yon
li>M 1 | c'|lli'||| |y
Mil L|i In mciifl
(days, monChs,
or years)
lluw IUHK illit
I lir nyin|il umti
Inst? (days,
moot lie or
years)
Did lie/she
consult a doctor?
No Veo
Doctor's namo & aOJrcsa
Ycnr monr/
recently
cxpprl'rnc.
-------
-• ... . ...
ll(*. 1 1 1 » I'.ii f.-ii {f.vcl, Hltiw, nr
in. .Hi-ii .nil ) li«':irl lii-.ie riT
IU. lilil li /sl.c li.ivc pain, discomfort, or
irmililt' In or around Che heart?
MH . 1 1 • li 1 il j; M| 1 IIP r.k t II?
1 >o . ,\nv niiii'iii.'i t il I uc«» lornl 1 «»ii or
« i ii|»i l.«ii:t (>ii (In: nktn?
: ri I'll V:
l'ii|. An-/ (inililtins utth i lie scomnch or
d Ir.ni -:l IVP Kystcni?
:•!'< '-I 1 i :
|.',l . Swi.ll.-ii f.l.in
1 —
i or
1° I
IT) 1
j~< j
w
Ho
.Yen
II. iu ( I 1'ijMi'ilt |y
illil 1 lilii ni-i-iii 7
(days, Months,
or years')
llnw IIHIH illil
( In* iiy«|
-------
U., i',, i i. .,.(,„•.- ., , i Inn In lo In- I Illi-it nut |.y I lit- wile til miilr Htiliriuii'n/or/by ( lnluTuomcii , If tin; wife In ilcccmiwl , Winild you complete tliln in > I Inn
>» i In- I -I nl youi kimwlnlc.r :i hi ml the practices o£ the 1 luliurraau'u wife.
145. U,, K|,t. us.- ;i Mrtli control method? 1)0"° 2)d]Ye8 (If yen, answer questions 144 and 145}
liA. Ulinl iiii-lliii'l "f til rlli run! rnl What una the rcnson for tlic surgery?
Doctor's name 4 address?
: address
tl* Wi.ji 11, («.•.!> I In- .ivi r.-i|-,L- li-u^tli of her menntrunl cncle? days'(From the first day of bleeding to the first day of next period.)
i'.) On t h.- :iw-rjp,e, Imw Innp. Is (w;)s) l(er period? (dnys)
iSi. in.ti.n I..-IKI Mvr y,-..,.;, liuillicr blood flow during ncnst ruat Ion I.C3 Increaned? I (U decrenaed? 3 tZJ stayed the same?
1 Tit Ha.-e (IT illil) r.lie i-xpei Irnrn any .iliiinrm.il spotting or bleeding between her menstrual cycles? l|^ j Ho 2J \ Yea (go to question 153)
(go to question 154)
•'•>• l< yv., .11.1 si,,- imisuK .-, phynlclan?) 1 .Q No 2-.Q ye8 Doctor's name & address
home address
M.I.J ;l« •.:.,!. ,,r .11,1 „;<• ,,nc. niiy «e
-------
Respondent ID No. / / / /_
Date
Day of
Week
Results Codes
Intervj
Refusal
Unavail
ew complet
Time
Result of
Contact
(Enter Code)
Languag
e .... 01 Decease
..... ft? Rdcnnilrl
able 03 Ihcompe
Comments
e problem .... 04 Apointment m
d .05 Call back n<
ent moved .... 06 No answer, n
tent 07 Line busv .
Interviewer
ID No.
) appt . made . 09
o one home. . .10
11
-------
Telephone Interview
Large Cohort
GREAT LAKES FISHERMEN HEALTH SURVEY
Interviewer Name
Interviewer ID No
Respondent ID No / / / /
Date of Outcome __/__/
Interview Time Began a.m./p.m.
Interview Time Ended a.m./p.m.
University of Minnesota
Division of Epidemiology
-------
GREAT LAKES FISHERMEN HEALTH SURVEY
I WOULD LIKE TO BEGIN BY ASKING SOME GENERAL QUESTIONS ABOUT YOURSELF,
1. First, what is your current address?
street
city state zip
2. What is your birthdate? / /
month day year
3. What is your current marital status? I. f) Single (never married)
2. f \ Married
3. ( ) Separated or divorced
Widowed
4. How many children do you have? Sons Daughters
-------
5. Do you presently hold a commercial fishing license?
1. C j No 2. ( ) Yes :> Approximately how long have you held this license?
(PROBE FOR DIFFERENT PERIODS OF TIME)
_
from to
from to
from to
No. of years
GO TO QUESTION 8
6. Did you ever hold a commercial fishing license?
1. \ No. 2. Yes > Approximately how long did you hold this license?
(PROBE FOR DIFFERENT PERIODS OF TIME)
from to
from to
from to
No. of years
GO TO QUESTION 8
7. Have you ever been a crew member for, or a partner with, an individual owning a commercial fishing license?
1. T 'j No 2. C \ Yes :>How long have you been a crew member or partner?
(PROBE FOR DIFFERENT PERIODS OF TIME)
/ / /
No. of years
from to
from to from to
IF NO, STOP THE INTERVIEW. THANK THE RESPONDENT FOR HIS COOPERATION AND EXPLAIN THAT
CURRENTLY WE ARE ONLY INTERVIEWING PEOPLE WHO HAVE COMMERCIALLY FISHED.
— 2—
-------
8. Near what towns do (did) you commercially fish most often?
city state city state
8a. Where is (was) your hailing port?
.city state
9. Is commercial fishing your current full time occupation?
1. ( ) No 2. \~\ Yes •> How long have .you commercially fished full time?
(PROBE FOR DIFFERENT PERIODS OF TIME) N°' °f
from to from to from to
GO TO QUESTION 12
10. What Is your current full time occupation? (PROBE FOR TITLE, DUTIES AND PLACE OF EMPLOYMENT)
title/duties place of employment
-3-
-------
11. Was commercial fishing ever your full time occupation?
1. O No 2. Yes >How long were you full time? (PROBE FOR DIFFERENT PERIODS OF TIME)
No. of years
liow lo.ng have you been a part time commercial fishermen? (PROBE FOR DIFFERENT PERIODS OF TIME)
No. of years from to from to from to
12. To the best of your knowledge, tell me the names and addresses of full or part time crew members.
(Past and present). Employment
name st reet, cit ; , state, zip, plione number No. of years, from - to
-4-
-------
In this section I am going to ask you some questions about your fish consumption habits.
13. Approximately how many times do you eat fish from the Great Lakes?
1. Never 5. < 1 time/month
2. 1-2 times/week 6. 1 time/month
3. 3-5 times/week 7. 2-3 times/month
4. 6-7 times/week
14. Approximately how many years have you consumed Great Lakes fish with this frequency?
No. of years
15. On the average, how many pounds of fish do you consume per meal?
Pounds of fish
16. What types of Great Lakes fish do you eat most often?
Salmon Burbot Northern Pike Smelt Pan Fish
Perch Rough Fish Lake Wliitef ish Bass Other Trout
Walleye Chub Lake Trout Lake Herring Other (specify)
-5-
-------
17. In comparison to yourself, how often does your family eat fish?
Wife;
1. As often as myself 1.
2. More often than myself How often? 2.
3. Less often than myself How often? 3.
4. Does not apply 4.
5. Never 5.
6. Don't know 6.
As often as myself
More often than myself
Less often than myself
Does not apply
Never
Don't know
How often?
How often?
1.
2.
3.
4.
5.
6.
As often as myself
More often than myself
Less often than myself
Does not apply
Never
Don't know
How often?
How often?
1.
•2.
3.
4. _
5. _
6.
As i
Mori
Les:
Doe
Nev'
Don
As often as myself
More often than myself
Less often than myself
Does not apply
How often?
How often?
1.
2.
3.
4.
5.
6.
As often as myself
More often than myself
Less often than myself
Does not apply
Never
Don't know
How often?
How often?
1.
2.
3.
4.
5.
6.
As often as myself
More often than myself
Less often than myself
Does not apply
Never
Don't know
How often?
How often?
-6-
-------
In the next section, I would like to ask you some questions about your tobacco and alcohol use.
18. Do you smoke cigarettes now?
1. ~} No 2. f~~ Yes
19. How old were you when you first began to smoke cigarettes?
20. What is the average number of cigarettes you presently smoke per day?
21. How many .years have you smoked with this frequency?
GO TO QUESTION 25
18a. Did you ever smoke cigarettes?
1. No 2. Yes
GO TO QUESTION 25
22. How old were you when you first began to smoke cigarettes?
23. How many cigarettes did you smoke per day?
24. How many years did you smoke cigarettes?
GO TO QUESTION 25
-7-
-------
25. Do you smoke a pipe now?
1. (~\ No 2. O Yes
Y
25a. Did 'you ever smoke a pipe?
1. (~\ No 2. C Yes
32. Do you. smoke cigars now?
1. (~) No 2. (""") Yes
32a. Did you ever smoke cigara?
1. No 2. Yes
V
GO TO QUESTION 39
26. How old were you when you first began to smoke a pipe?
27. What is the average number of pipefuls you presently smoke per day?
28. How many years have you smoked a pipe with this frequency?
GO TO QUESTION 32
29. How old were you when you first began to smoke a pipe?
30. How many pipefuls did you smoke per day?
31. How many years did you smoke a pipe?
GO TO QUESTION 32
33. How old were you when you began to smoke cigars?
34. What is the average number of cigars you presently smoke per day?
35. How many years have you smoked cigars?
GO TO QUESTION 39
36. How old were you when you began to smoke cigars?
37. How many cigars did you smoke per day?
38. How many years did you smoke cigars?
GO TO QUESTION 39
—8—
-------
39. Have you ever chewed tobacco regularly?
1. \ No 2. Yes
40. Uave you ever used snuff regularly?
1. \ No 2. Yes
HARD LIQUOR IS CONSIDERED TO INCLUDE: Gin, Whiskey, Scotch, Vodka, Rum, Brandy.
(One drink = 1 to l>s ounces of alcohol)
41. Do you drink hard liquors now?
1. (~J No 2. (""") Yes ^
42. How old were you when you- first began to drink hard liquors?
43. What is the average number of hard liquor drinks you have per week?
44. How many years have you consumed hard liquor with this frequency?
GO TO QUESTION 48
M
4la. Did you ever drink hard liquors in the past?
'•O
No 2.
Yes
\
GO TO QUESTION 48
45, How old were you when you first began to drink hard liquor?
46. How many drinks did .you have per week?
47i How many years did you drink hard liquor?
GO TO QUESTION 48
-9-
-------
48. Do you drink beer now?
I. No 2. " Yes -
\
48a. Did you ever drink beer?
1. O No 2. ( } Yea
v t
55. Do you drink wine now?
1. ~\ No 2. Yes
\l
GO TO QUESTION 55a
49. How old were you when you first began to drink beer?
50. What is the average number of beers (12 oz. bottles) you presently drink
per week? __
51. How many years have consumed beer with this frequency?
GO TO QUESTION 55
52. How old were you when you first began to drink beer?
53. How many beers did you drink per week?
54. How many years did you drink beer?
GO TO QUESTION 55
56. How old were you when you first began to drink wine?
57. What is the average number of wine drinks (4-6 oz. glass) you have per
week?
58. How many years have you consumed wine with this frequency?
GO TO QUESTION 62
-10-
-------
55n. Did you ever drink wine?
1. No 2. Yes
GO TO QUESTION 62
59. How old were you when you first began to drink:winet
60. How many wine drinks did you have per week?
61. How many years did you drink wine?
GO TO QUESTION 62
In this section I would like to ask you some medical conditions you or your family may have
had. Please tell me if youf your wife, or children have ever had any of the following
conditions or diseases diagnosed by a doctor.
Disease
62. Asthma
63. Bronchitis
64. Emphysema
65. Tuberculoais-TB
1-No
2-Yes
3-Don't
Know
FnmLly
Member
S-Self
W-Wife
Son
D-Dmighter
Name & Address of Clinics,
Hospitals, or Doctors Consulted
Dote of
Diagnosis
-11-
-------
66.
67.
68.
69.
•70.
71.
72.
73.
74.
75.
76.
77.
78.
Disease
Mononucleosls-Mono-Kissing
Disease
Pneumonia
Any other disease of the
respiratory system
Specify:
Hepatitis or yellow jaundice
Cirrhosis of the liver
Any other liver disease
Specify:
Spondylltis
Gout
Rheumatoid Arthritis
Osteoarthritis
Any other diseases of bones
and joints
Specify:
Meningitis
Hypertension or high blood
pressure
1-No
2-Yes
3-Don't
Know
Family
Member
S-Self
W-Wife
Son
D-Daughter
Name & Address of Clinics,
Hospitals, or Doctors Consulted
Date of
Diagnosis
-12-
-------
79.
80.
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
Disease
High cholesterol
Angina pectoris
Heart attack-Mi or coronary
Stroke, cerebral accident,
(CVA) -hemorrhage, thrombosis,
embolism
Any other heart or circulatory
disease
Specify:
Diabetes
Thyroiditis
Any other glandular disorder
Specify:
Eye disease
Specify:
Psoriasis
Eczema
Herpes zoster-shingles
(dermatitis)
Pemphigus
1-No
2-Yes
3-Don't
Know
Family
Member
~S=Seir
W-Wlfe
Son
D-Daugliter
Name & Address of Clinics,
Hospitals, or Doctors Consulted
Date of
Diagnosis
-13-
-------
Disease
i-No
2-Yes
3-Don.'t
Know
Family
Member
~S-Self~
W-Wtfe
Son
D-Daughter
Name & Address of Clinics, Date of
Hospitals, or Doctors Consulted Diagnosis
92.
93.
94.
95.
96.
97.
98.
99.
99a.
100.
101.
102.
Any other diseases of skin
Specify:
Multiple Sclerosis
Any other diseases of the
nervous system
Specify:
Anemia
Gastritis
Ulcers (stomach or duodenal)
Any other diseases of the
digestive system
Specify:
Have you, your wife, or
children ever had cancer?
If yes, what location?
Leukemia
Hodgkin's Disease
Other Cancer
Specify:
103. Mental retardation
Any other disease
Specify:.
-14-
-------
In this next section I am interested only in symptoms you may have experienced.
you ever had any of the following symptoms?
Have
Symptom
105. Severe fevor
106. Extreme tiredness
107. Frequent or very
bad headaches
108. Problems with
mouth or throat
109. Any problems with
ears or hearing
Specify:
110. Any unusual diff-
iculty with eyes
or eyesight other
than a change of
the prescription
of glasses
Specify:
111. Sudden weakness
or heaviness of
arms or legs?
Specify:
i ,.M~
2-Yes
3-Don't
Know
How often does
(PROBE for times
per day , times
per week, etc. )
How long did the
symptoms last?
(PROBE for hours,
days, weeks,
months or years.)
Did you
conoull
n doctor!
iun
2-Yes
3-Don ' t
Know
Doctor's nnme
ond address
Ycnr most
recently
experienced
-15-
-------
Symptom
112. Numbness in arms
or legs
113. Swelling in arms
or 'legs
114. Stiffness in joints
or bones? Specify:
115. Pains in Joints or
bones? Specify:
116. Spasms of limbs?
117. Spells of dizziness
118. Have you fainted
or blacked out?
119. Irregular, fast or
slow heartbeats?
120. Have you had pain
discomfort, or
trouble in or
around your heart?
i-No
2-Yes
3-Don't
Know
How often does
(did) this occur
(PROBE for times
per day , times
per week, etc.)
How long did the
symptoms last?
(PROBE- for hours
days , weeks ,
months or years)
Did you
consult
a doctor?
1-No
2-Yes
3-Don't
Know
Doctor's name
and address
Year most
recently
experienced
-16-
-------
Symptom
121. Itching of the skin?
122. Any unusual dis-
coloration or
eruptions on the
skin? Specify:
123. Any problems with
your stomach or
digestive system?
Specify:
124. Swollen glands in
your neck, armpits
or groin?
125. Have you lost 20
or more pounds in
the last five years?
(include dieting)
1-Ho
2-Yea
3-Don't
Know
How often does
(did) this occur
(PROBE for times
per day , times
per week, etc.)
How long did the
symptoms last?
(PROBE for hourfe
days, weeks,
months or years)
Did you
consult
a doctor?
1-No
2-Yea
3-Don't
Know
Doctor's numo
and address
Year most:
recently
experienced
126. Finally, what is your race?
1. White
2. Black
3. American Indian
l*. Other:
-17-
-------
The last section of the questionnaire relates to your wife and children. If I could speak to
your wife, she may be able to answer the few questions remaining. Thank you very much for your
time and cooperation.
What is your first name?
127. How many natural children do you have?
128. How many step or adopted children do you have?
129. please tell me about all pregnancies and outcomes of which you were the mother/father in the sequence
they occurred. I am interested in your current marriage only.
Preg.
No.
1
2
3
A
5
6
Complications in
this pregnancy
(describe)
Outcome
Mis-
carriage
Still-
birth
Live
Birth
Other
Infants
sex
M
F
Date of
Birth
Live
Birth
Weight
Hospital, clinic
or place of birth
Doctor's
name
-18-
-------
129a. Have any of your natural children died? (do not include stillbirths nor step or adopted children)
1
GO TO QUESTION 130
Sex
M F
I
2
3
A
Date of Birth
Date of Death
Name of Child
Cause of Death
City and State in
which death was reported
130, Do you know of any mental or physical abnormalities at birth (birth defects) in yourself or your natural
children?
1. C j No 2. () Yes (If yes, please complete the following table)
GO TO QUESTION 131
CHECK APPROPRIATE BOXES
Relationship to you
Self
Son
Daughter
Living
Dead
Nature of Birth Defect (describe)
-19-
-------
131. Do you use a birth control method?
I . No 2 . ~\ Yes - > fiO TO QUESTION 132
131a. Did you ever use a birth control
2.QYes
GO TO QUESTION 134
132. What method of birth control have you used the most during your adult life?
1. r^) pill 5. (~~\ rhythm
2. f~^\ IUD 6. fj condom
3. fj diaphragm 7. f J other (specify)
4. ^ J foam or jelly
133. At. what age did you begin using birth control methods?
134. At what age did you begin to menstruate?
135. Have you stopped menstruating?
1. O No 2. Yes
V
GO TO QUESTION 137
-20-
-------
136. Did the cessation occur naturally or due to surgery?
1. " Naturally 2. f~~\ Surgery
What was tlie reason for the surgery?
Doctor's name and -address
137. Have you taken any medications (since that time)prescribed by a doctor?
1. """ No 2. *~^\ Yes 3» What medications are these?
138. What is (was) the average number of days between periods?
_days
139. On the average, how long is (was) your period?
days
140. In the past five years, has your blood flow during menstruation
increased?
decreased?
stayed the same.
does not apply.
L. Have (did) you experience(d) any abnormal spotting or bleeding between your menstrual cycles?
I. C~^\ No 2. ("} Yes
V
GO TO QUESTION 143
-21-
-------
142. Did you consult a physcian?
1. C j No 2. C j Yes :>• Doctor's name and address:
143. Have you used or did you use any medications prescribed by a doctor for menstrual irregularities?
1. r~\ No 2. *\ Yes > Name of medication
That was my last question. Thank you very much for your time and cooperation.
-22-
-------
FILL IN IMMEDIATELY AFTER COMPLETING THE. INTERVIEW:
I. Questions 1-126 have been answered by
2. Questions 127 - 143 liave been answered by
3. The overall quality of the interview was:
1. Very good
2. Good
3. Fair
4. Poor
4. How reliable do you feel the respondents' answers were? 1. High quality
2. Questionable
3. Generally reliable
4. Unreliable
Interviewer's comments:
-23-
-------
APPENDIX III
Pilot Study and Survey Materials
Great Lakes Study
-------
Contents
1 Introductory letter Protocol I
2 Introductory letter Protocol I - Proxy
3 Introductory letter Protocol II
4 introductory letter Protocol II - Proxy
5 Introductory letter Protocol III
6 Introductory letter Protocol III - Proxy
7 Great Lakes Study Consent Form
8 Great Lakes Study Consent Form Without Medical Records Request
9 Great Lakes Study Consent Form - Proxy
10 Great Lakes Study Consent Form Without Medical Records Request - Proxy
11 Second Request for Consent Form
12 Medical Records Request Letter - Doctors
13 Medical Records Request Letter - Clinics
14 Medical Records Request Letter - Hospitals
15 Second Request for Medical Records
16 Medical Records Abstract Form
17 Validation Form
18 Death Certificate Request Letter
19 Death Certificate Abstract Form
20 Study Participation Request Letter (appropriate for individuals with
no phone or unlisted numbers)
21 Study Participation Return Postcard
22 Set of Questions (appropriate for Protocol II Participants
-------
The following materials were developed for use in the Survey of the
Large Cohort of Commercial Fishermen.
23 Respondent Information Sheet
24 Interviewer Assignment Record
25 Definitions (appropriate for questioning regarding health histories)
26 Survey Manual - Great Lakes Fishermen Health Study
27 Non-interview Report
-------
UNIVERSITY OF MINNESOTA
TWIN CITIES
Division of Epidemiology
School of Public Health
1360 Mayo Memorial Building
420 Delaware Street S.E.
Minneapolis, Minnesota 55455
Dear Fisherman:
The sport and commercial fishing industries of ..the ..Great
Lakes region are of significant economic importance to the Great
Lakes states. The health and"well-being of the members of the
fishing industry are important to their families as well as to
the economy of the Great Lakes Basin.
In 1978 our "Great Lakes Project" was initiated to evaluate
the health of those individuals involved with sport or commercial
fishing in the Great Lakes. It is our understanding that you have
either served on a commercial fishing vessel or have possessed a
commercial fishing license.
In several days an interviewer will contact you -by phone .to
ask you questions regarding your fishing practices and health
status. Your answers will be kept confidential. This is a medi-
cally-related survey and we uphold your right to privacy.
Enclosed with this letter you will find a medical consent
form. The interviewer will explain its purpose and answer any
questions you may have regarding this health survey.
Thank you for you cooperation. Should you have any questions
please call the study office, collect (612) 376-8775.
Sincerely yours,
i. I :
* I /' . :
Leonard M. Schuman, M.D.
Professor and Director
LMSrkb
encl.
HEALTH SCIENCES
-------
UNIVERSITY OF MINNESOTA
TWIN CITIES
Division of Epidemiology
School of Public Health
1360 Mayo Memorial Building
420 Delaware Street S.E.
Minneapolis. Minnesota 55455
The sport and'commercial fishing industries of the Great
Lakes region are of significant economic importance to the Great
Lakes states. The health and well-being of the members of the
fishing industry are important to their families as well as to
the economy o.f the Great Lakes basin.
In 1978 the University of Minnesota initiated a "Great Lakes
Study"-to evaluate the health of those individuals involved with
sport or commercial fishing in the Great Lakes. It is our under-
standing that Mr./Mrs. served on a, commercial fishing
vessel or held a commercial fishing license.
In several days an interviewer will contact you by phone to
ask you questions about the fishing practices and prior health
status of Mr./Mrs. . We will ask that you answer
our questions as you believe Mr./Mrs. _would. Your
answers will be kept confidential. This is a medically-related
survey and we uphold your and Mr./Mrs. 's right to
privacy^
Thank you for your cooperation. Should you have any questions
please call the study office, 'collect (612) 376-8775.
Sincerely yours,
Leonard M. Schuman, M.D.
Professor and Director
LMS:kb
encl..
HEALTH SCIENCES
-------
UNIVERSITY OF MINNESOTA
TWIN CITIES
Division ot Epidemiology
School of Public Health
1360 Mayo Memorial Building
420 Delaware Street S.E.
Minneapolis, Minnesota 55455
Dear Fisherman:
The sport and commercial fishing industries of the Great
Lakes region are of significant economic importance to the' Great.
Lakes states. The health and well-being of the members of the
fishing industry are important to their families as well as to
the economy of the Great Lakes Basin.
In 1978 our "Great Lakes Project" was initiated to evaluate
the health of those individuals involved wi'th sport or-commercial
fishing in the Great Lakes. It is our understanding that you have
either served on a commercial fishing vessel or have possessed a
commercial fishing license.
Enclosed with this letter is a set of questions regarding
your fishing practices and health status. A medical consent form
is also enclosed. In a couple of days an interviewer will con-
tact you by phone and request your answers to this set of -ques-
tions. Your answers will be kept confidential. This is a
medically-related survey and we uphold your right to privacy.
Thank you for your cooperation.- Should you have.any questions
please call the study office, collect (612) 376-8775;
Sincetely yours,
Leonard M. Schuman, M.D.
Professor and Director
LMS:kb
encl.
HEALTH SCIENCES
-------
UNIVERSITY OF MINNESOTA
TWIN CITIES
Division of Epidemiology
School of Public" Health
1360 Mayo Memorial Building
420 Delaware Street S.E.
Minneapolis. Minnesota 55455
The sport and commercial fishing industries of the Great
Lakes region are of significant economic importance to the Great
Lakes states. The health and well-being of the members of the
fishing industry are important to their families as well as to
the economy of the Great Lakes Basin.
In 1978 the University of Minnesota initiated a "Great Lakes
Study" to evaluate the health of those individuals involved with
sport or commercial fishing in the Great Lakes. It is our under-
standing that Mr./Mrs. served on a commercial
vessel or held a commercial fishing license.
Enclosed with this letter is a set of questions regarding
the fishing practices and prior health status of Mr./Mrs.
. In a couple of days an interviewer will con-
tact you by phone to ask you how you believe Mr./Mrs.
would have answered these questions. Your answers will be kept
confidential. This is a medically-related survey and we uphold
your and Mr./Mrs. 's right to privacy.
Thank you for your cooperation. Should you have any questions
plesse call the study office, collect (612) 376-8775.
Sincerely yours,
Leonard M. Schuman, M.D,
Professor and Director
LMSrkb
end.
HEALTH SCIENCES
-------
UNIVERSITY OF MINNESOTA
TWIN CITIES
Division of Epidemiology
School of Public Health
1360 Mayo Memorial Building
420 Delaware Street S.E.
Minneapolis. Minnesota 55455
Dear Fisherman-..
The sport and commercial fishing industries of the Great
Lakes region are of significant economic importance to the Great
Lakes states. The health and well-being of the members of the
fishing industry are important, to their families as well. as to
the economy of the Great Lakes Basin.
In 1978 our "Great Lakes Project" was initiated to evaluate
the health of those individuals involved with sport or commercial
fishing in the Great Lakes. It is our understanding that you have
either served on a commercial fishing vessel or have possessed a
commercial fishing license.
Enclosed with this letter is a questionnaire which asks
questions about your fishing practices and health -status .: We
would like you to fill out the questionnaire to the best of .your
ability. A section at the end of the questionnaire is important
for your wife to complete. Your answers will be kept confidential.
This is a medically-related survey and we uphold your right to
privacy.
On the last page you will find a medical consent form. We
ask you for your signature so as to permit us to ask your doctor (s)
for the exact diagnosis of the conditions you have listed. If
your wife fills out the last section of the questionnaire, we
ask that she sign the medical consent form as well. Regardless
of whether you or your wife sign, complete the questionnaire and
mail it back in the prestamped self-addressed envelope provided.
Thank you for your cooperation. Should you have any questions
please call the study office, collect (612) 376-8775.
Sincerely yours r
Leonard M. Schuman, M.D.
Professor and Director
LMS: te-b
encl.
HEALTH SCIENCES
-------
UNIVERSITY OF MINNESOTA
TWIN CITIES
Division of Epidemiology
School of Public Health
1360 Mayo Memorial Building
420 Delaware Street S.E.
Minneapolis, Minnesota 55455
The sport and commercial fishing industries of the Great
Lakes region are of significant economic importance to the Great
LaKes states. The health and well-being of the members of the
fishing industry are important to their families as well as to
the economy of the Great Lakes Basin.
In 1978 the University of Minnesota initiated a "Great Lakes
f'tudy" to evaluate the health of those individuals involved with
s,-;.->rt or commercial fishing in the Great Lakes. It is our under-
5-randing that Mr. /Mrs. _ served on a commercial fishing
_
or held a commercial fishing license.
Enclosed with this letter is a questionnaire which asks
f.insstions regarding the prior fishing practices and health status
01. Mr. /Mrs. _ .. We ask that you complete the question-
naire, as you think Mr. /Mrs. _ would. Your answers
will be kept confidential. This is a medically-related survey and
we uphold your right to privacy.
Thank you for your cooperation. Should you have any questions
plaase call the study office, collect (612) 376-8775.
Sincerely yours,
Leonard M. Schuman,. M.D,
Professor and Director
LMS:kb
end.
HEALTH SCIENCES
-------
Great Lakes Study
Consent form
Maae:
Current Street Address:
Street
City State Zip
Phone
I {we) hereby give my (our) permission to .be interviewed by
the "Great Lakes Study" of the University of Minnesota.. -.My (our)
involvement in this study is voluntary. My (our) participation
consists of an interview of my (our) fishing practices and/or
health status.
I (we) permit the "Great Lakes Study" to contact medical sources
listed in the interview for the' purpose of reviewing my (our)
medical records. I (we) understand that all information contained
within the interview and the review of medical records will be
keat confidential and used only for research purposes. I (we)
further understand that my (our) names will not be associated with
the results of this survey.
Husband Date
Wife Date
-------
<3reat Lakes Study
Consent form
Name:
Current Street Address:
Street
City State Zip
Phone
I (we) hereby give my (our) permission to be interviewed by
the "Great Lakes Study" of the University of Minnesota. My (our)
involvement in this study is voluntary. My (our) participation
consisits of an interview of my (our) fishing practices and/or
health status.
I (we) understand that all information contained within the
interview will be kept confidential and used only for research
purposes. I (we) further understand that my (our) names will not
be associated with the results of this survey.
.Husoahd Date
Wife Date
-------
Great Lakes Study
Consent focm
Name:
Current Street Address:
Street
City State Zip
Phone
I hereby give my permission to be interviewed by the "Great Lakes
Study" of the University of Minnesota. My involvement in this study is
voluntary. My participation consists of an interview of the fishing
practices and prior health status of Mr./Mrs. . .
I understand that in answering questions about Mr./Mrs.
I will report only that information which-I know to'be true, and'wiH not
be asked to make judgements or opinions about the character of that in-
dividual. I reserve the right tb refuse to answer any question regarding
Mr./Mrs. which I do not believe to be in their best
interest.
I permit the'"Great Lakes Study" to contact the medical-sources
listed in the interview for the purpose of reviewing Mr./Mrs..
medical records. I understand that all information contained within the
interview and the review of medical records will be kept confidential
and used only for research purposes. I further understand that..our names
will not be associated with the results of this survey.
Signature Date
Relationship to Subject
Please return this form in the attached addressed, stamped envelope.
Thank you.
-------
Great Lakes Study
Draft: Proxy Consent Form/ interview.only
Name:
Current Street Address:
Street
City State Zip"
Phone
I hereby give ray permission to be interviewed by the "Great
Lakes Study" of the University of Minnesota. My involvement in
this study is voluntary. My participation consists of an inter-
view of the fishing practices and prior health status of Mr. /Mrs.
_ __. I understand that in answering questions about
Mr. /Mrs. _ . I will report .only that information which
I know to be true and will not be asked to make judgements nor
opinions about the character of that individual. I reserve the
right to refuse to answer any question regarding Mr. /Mrs. _
which I do not believe to be in their best- interest.
Signature Date
Relat.ionshii) to deceased
Please return this form in the attached addressed, stamped
envelope. Thank you.
-------
UNIVERSITY OF MINNESOTA
TWIN CITIES
Division of Epidemiology
School of Public Health
1360 Mayo Memorial Building
420 Delaware Street S.E!
Minneapolis. Minnesota 55455
Thank you for your participation in the health study of
Great Lakes commercial fishermen conducted by the University
of Minnesota. We appreciate your contribution to this research
in the field of preventative medicine.
We are writing to-request .your signed consent form.
(Please find enclosed an .additional consent form copy.if you
have discarded the original.) We ask that you sign; >he.-consent
form and return it in the addressed, stamped envelope. Please
be assured that all information we receive will be. kept- in .the.
strictest confidence. Your name .will not be associated with
the study results in any way.
Thank you for your cooperation.
Sincerely,
Leonard M. Schuman,,M.D.
Professor and Director
LMSrkb
encl.
HEALTH SCIENCES
-------
UNIVERSITY OF MINNESOTA
TWIN CITIES
Division of Epidemiology
School of Public Health
1360 Mayo Memorial Building
420 Delaware Street S.E.
Minneapolis. Minnesota 55455
REP:
The patient named above is participating in a health study
of Great Lakes commercial fishermen conducted by the University
of Minnesota. This patient has given us permission to contact
you about his/her medical history. A copy pf his/her authoriza-
tion is enclosed. . .
In order to classify this patient appropriately, we need
to verify patient conditions with the exact diagnoses. We would
appreciate your assistance in completing the enclosed Medical
Records Abstract form. If you would prefer to send copies of
clinical summaries or medical records, please do so.
We greatly appreciate your time and effort in complying
with our request. Should you have any questions, please call,
collect (612) 376-8775.
Thank you for your cooperation.
Sincerely yours,
Leonard M. Schuman, M.D.
Professor and Director
LMSrkb
encl.
HEALTH SCIENCES
-------
UNIVERSITY OF MINNESOTA
TWIN CITIES
Division of Epidemiology
School of Public Health
1360 Mayo Memorial Building
420 Delaware Street S.E.
Minneapolis, Minnesota 55455
ATTN: Medical Records
REFr
Dear Madam/Sir:
The patient named above is'participating "in a health study
of.. Great Lakes commercial fishermen conducted by the University
of Minnesota. This patient has given us permission to contact
you about his/her .medical history. ...A copy of, his/her authoriza-
tion is enclosed.
In order to classify this patient properly,'we-need diag-
nostic information regarding past and present medical.conditions.
and would appreciate your completing the enclosed .medical .records
abstract form, if you would prefer to send us a summary or copy.
of these medical records-, please do so.
We wish to thank you in advance, for your time and effort in
complying with our request. Should you have any questions please
call the study office, collect (612) 376-8775.
Sincerely yours,
Leonard M. Schuman, M.D,
Professor and Director
LMS:kb
encl.
HEALTH SCIENCES
-------
UNIVERSITY OF MINNESOTA
TWIN CITIES
Division of Epidemiology
School of Public-Health
1360 Mayo Memorial Building
420 Dataware Street S.E.
Minneapolis, Minnesota 55455
ATTN: Medical Records
REF:
Dear Madam/Sir:
The individual named above has indicated that he/she, was a
patient in your hospital. This individual is participating in
a health study of Great Lakes commercial fishermen conducted by
the University of Minnesota and has given us permission to review
his/her medical records. A copy of his/her consent form is en-
closed.
In order to classify this individual appropriately, we need
to verify the patient's hospitalizations and diagnoses at your
institution. We would appreciate your assistance in completing
the enclosed medical records abstract. If you prefer to send
copies of clinical summaries or medical records, please do so.
We greatly appreciate your time and effort in complying
with our request. Should you have any questions, please call
the study office, collect (612)376-8775.
Thank you for your cooperation.
Sincerely yours,
Leonard M. Schuman, M.D.
Professor and Director
LMSrkb
encl.
HEALTH SCIENCES
-------
UNIVERSITY OF MINNESOTA
TWIN CITIES
Division of Epidemiology
School of Public Health
1360 Mayo Memorial Building
420 Delaware Street S.E.
Minneapolis. Minnesota 55455
REF:
Several weeks ago we requested some specific medical infor-
mation regarding one of your patients. In the event that you did
not receive that communication, or misplaced the original copy,
we enclose an additional form for your convenience. . . .
The patient named above 'is participating in a health study
of Great Lakes commercial- fishermen conducted by the University
of Minnesota and has given us permission to contact you regarding
his/her medical history. A copy of his/her authorization is
enclosed.
In order to' classify this patient appropriately/, we .need to
verify patient responses with exact medical records. We would
appreciate your assistance by filling out the attached form..... If
you.would prefer to send clinical summaries or copies--of-records,
please do so.
We appreciate your time and effort in complying with..our
request. If you have already returned the form, please disregard
this letter. Should you have any questions, do not hesitate'to'
.call the study office, collect (612) 376-8775.
Thank you for your cooperation.
Sincerely yours,
Leonard M. Schuman, M.D.
Professor and Director
LMSrkb
encl.
HEALTH SCIENCES ,
-------
Great Lakes Study
Medical Records Abstract
Dr./Mr./Mrs.
has authorized the
University of Minnesota, Department of Epidemiology, to review
their medical records. A copy of the signed consent form is
attached. We ask for your cooperation in verifying diagnoses
made by yourself or your staff in the study participant.
(Obstetricians and Gynecologists please skip the first two
questions and answer questions three, four and five). Thank you
for your cooperation. •
1. If you or any staff members have confirmed or diagnosed any
of the following conditions, please check the corresponding
box and write the exact date of diagnosis.
Disease Name
Condition has been confirmed Date of.
or diagnosed in subject (•} diagnosis
Asthma
Bronchitis
Emphysema
Tuberculosis -TB
Mcnonucleosis - Mono
Pneumonia
Any other respiratory
disease
SPECIFY:
Hepatitis or yellow
jaundice
Cirrhosis
Any other liver
disease
SPECIFY:
Spondylitis
Gout
Rheumatoid arthritis
Osteoarthritis
Any other diseases of
bones and joints
SPECIFY:
High Triglycerides
Meningitis
Hypertension
-
-------
Condition has been confirmed Date of
Disease Name or diagnosed in subject (V) Diagnosis
High Cholesterol
Angina pectoris
Heart attack or MI
Stroke- cerebral accident
or CVA- hemorrhage,
thrombosis, emoblism
Any other heart, or
circulatory diseases
SPECIFY:
Diabetes
Thyroiditis
Any other glandular
disorders
SPECIFY:
Eye diseases
SPECIFY:
Multiple Sclerosis-MS
Any other disease of
the nervous system
SPECIFY:
Anemia
Gastritis
Ulcers (stomach or
duodenal)
Any other disease of
the digestive system
SPECIFIC:'
Psoriasis
Eczema
Herpes Zoster (dermatitis)
Shingles
Pemphigus
Any other skin disease
SPECIFY:
Cancers:
Breast
Stomach
Esophagus
Mouth or Tongue
Large intestine
Rectum
Trachea, Bronchus
or Lung
--
-------
Disease Name
Condition has-been confirmed Date o'f.
or diagnosed in subject ft/} Diagnosis
Cancers :
Liver or Biliary passage
Bladder and Urinary organs
Skin
Thyroid
Leukemia
Other:
Mental Retardation
2. If you or any staff members have been consulted regarding any
of the following symptoms, please check the corresponding box
and write in the exact date(s) of consultation.
Condition
Condition has been Date of
confirmed in subject Diagnosis
Severe fever
Extreme tiredness.
Frequent or very bad headaches
Problems with mouth or throat
Any problems with ears or hearing
SPECIFY:
«ny unusual difficulty with eyes or
eyesight other than a change of the
prescription of glasses
SPECIFY":
Sudden weakness or heaviness of
arms or legs
SPECIFY:
Numbness in arms or legs
Swelling in arms 'or legs
Stiffness in joints or bones
SPECIFY:
Pains in joints or bones
SPECIFY:
Spasms- of- limbs
Spells of dizziness
Fainted or blacked out
Very strong heartbeats
Irregular (fast, slow, or incon-
stant) heartbeats
Pain, discomfort, or trouble in
or around vour heart
Itching of the skin
L . : .
-------
Condition^
Condition has been
confirmed in subject
Date of
Diacnosis
Any unusual discoloration or
erruptions on the skin
SPECIFY:
Any problems with your stomach
or digestive system
SPECIFY:
Swollen glands in neck, armpits
or groin
3. Please list all the pregnancies and outcomes which you or your
staff members provided medical care and assistance.
Pregnancy
Number
1
2
3
Preznancv Outcome
Abortion Miscarriage Stillbirth Live Birth
' I
5
6
Infants
Sex
M F
Date. of Birth
Live ,
Birth '
Weight
4, Has the subject ever consulted you or your:staff members
regarding abnormal spotting or bleeding between her menstrual
cycles?
O
No
Yes
Diagnosis;
Date :
Has the subject ever had surgery on her reproductive organs
performed by you or your staff members?
No
Yes
Diagnosis:
Date:
Type of surgery:^
Date of surgery:
6. Has the subject experienced complications in &tiy of the above
listed pregnancies?
O
No
Yes
Diagnosis':
Date:
Type of treatments^
Etiology;
-------
CONSENT FOHM
CURRENT ADDRESS*
HUOAUUINU
PHONE (I!) :
<0)J
INTERV1EWERS_
NEVER REC'D:_
UPDATE:
OTHERt
(1) CONTACT (Physician,
Clinic or Hospital) t
DATE & REASON FOR CONTACT
REGARD ING i
Mr/Mrs
OTHER INFOR!-SATIOt4£
(1) CONTACT (Physician,
Clinic or Hospital) t
DATE & REASON FOR CONTACT
REGARDING}
Mr/Mrs
OTHER I N FORM AT I ON i
oATfi itfc.rriiRS.SGNT to MEDICAL SOURCES! RETURNS FROM MEDICAL SOURCES j
1st
Mailing
2nJ mailing
6 phone call
Comments from
Phone calls
Abstract
Forms
No
Records
Chart
Copy
Other
-------
UNIVERSITY OF MINNESOTA
TWIN CITIES
Division of Epidemiology
School of Public Health
1360 Mayo Memorial Building
420 Delaware Street S.E.
Minneapolis. Minnesota 55455
Dear Sir:
The Departments of Environmental Health and Epidemiology,
University of Minnesota, are currently involved in a s-tudy-to
evaluate the health status of commercial fishermen in the Great
Lakes Basin. The validation procedure of this study involves
the collection of medical records for living cohort members and
the collection of death certificates for deceased cohort.members.
We would appreciate receiving one-copy of the death-certi-
ficate of the person(s) .listed on the attached sheet. .We are
listing.the full name, social security number, date of birth, and.
date of death to-assist in matching. In addition, we have de-
veloped a coding system for the results of your search (see
attached sheet). Please return the roster with the death certi-
ficate and your notations.
Enclosed you will find a check for the retrieval costs of
the death certificate search. Thank you for your cooperation.
Sincerely yours,
Leonard M. Schuman, M.D.
Professor and Director
LMS:kb
end.
HEALTH SCIENCES
-------
UNIVERSITY OF MINNESOTA
TWIN CITIES
Page 2
Division of Epidemiology
School of PubltcTHealth
1360 Mayo Memorial Building
420 Delaware Street S.E.
Minneapolis, Minnesota 55455
Commercial Fishermen Death Certificates
No. Name
Death Date Birth Date Social Security Number Results
*Code
+ = found
- = not found
I = further search needed
C = comment (add in space below name)
HEALTH SCIENCES
-------
UNIVERSITY OF MINNESOTA
TWIN CITIES
Division of Epidemiology
SchooJ of Public-Health
1360 Mayo Memorial Building
420 Delaware Street S.E.
Minneapolis, Minnesota 55455
The Division of Epidemiology of the University of Minnesota
is conducting a study to determine the general health" patterns, of
families involved in the commercial fishing industry.
We have tried to contact you by phone to ask a few'simple
questions about you and your family's health. -Since we have
been unable to reach you, we ask that you please call us •collect
at:. (612) 376-3775. Or, please write on the enclosed, card a
phone number where we can reach you.
Your contribution is important to the success of this
project. We are willing to talk with you at a time that is
convenient for you. All information that you share 'with us•
will be confidential. Your name will not be associated with
the study results.
We would be glad to answer any questions that you may have
regarding this study.
Thank you for your cooperation.
Sincerely,
Leonard M. Schuman, M.D.
Professor and Director
HEALTH SCIENCES
-------
NAME:
My phone number is (
I prefer not to release my phone
number, but I will call the study
office, collect {612} 376-8775 on
at .
(date) (time)
Thank you for your cooperation in this study.
-------
We have provided the following list of questions for your review
prior to the interviewer's call. We suggest that as you read through
this list you note all necessary names, dates, and addresses requested,
We hope that in providing you with an advanced -listing of the inter-
view questions, you will have time to consider your answers and,gather
information needed. Thank you for your cooperation.
What is your current address?
What is your sex?
What is your race?
What is your birthdate?
What is your Social Security number?
What is your marital status?
Do you presently hold a commercial fishing license?
Oid you hold a commercial fishing license in the past?
Approximately over what years have you held a license?
Have you' ever .been a crew member for, or a partner with,;an
individual owning a commercial fishing license?
Over what years have you been a crew member or par.tner?
Is commercial fishing your current full-time occupation?
Over what years have you commercially fished full-time?
Do you currently own a sport fishing license?
To the best of your knowledge, please tell me .the names and addresses
of full or part-time crew members? (past and present)
Do (or did) you consume any of the fish you catch commercially or. as
a sport fisherman?
Approximately how -many of your meals contain fish caught (by yourself
or a friend) from the Great Lakes?
Approximately how many years have, you consumed Great Lakes'-fish with
this frequency?
Approximately how many of your meals contain fish caught (by yourself
or a friend) from waters other than the Great Lakes?
How often do your family members eat fish?
Approximately how many pounds of fish have you consumed per year?
(assume: 1 fish meal equals % pound of fish)
What cooking methods are used in preparing the fish you consume?
Do you presently smoke cigarettes?
How old were you when you first began to smoke cigarettes?
What is the average number of cigarettes you presently smoke
per day?
How many years have you smoked,cigarettes with this.frequency?
-------
page 2
Did you smoke cigarettes in the past?
How old were you when you first began to smoke cigarettes?
How many years did you smoke cigarettes?
Do you presently smoke a. pipe?
How old were you when you first began to smoke a pipe?
What is the average number of pipefuls you presently smoke per day?
How many years have you smoked a pipe with this frequency?
Did you smoke a pipe in the past?
How old were you when you first began to smoke a pipe?
How many years did you smoke a pipe?
Do you presently smoke cigars?
How old were you when you began to smoke cigars?
What is the average number of cigars you presently smoke per day?
How many years have you smoked cigars with this frequency?
Did you smoke cigars in the past?
How old were you when you began to smoke cigars?
How many years did you smoke cigars?
Do you drink hard liquors?
How old were you when you first began to drink hard liquor?
What is the average number of hard liquor drinks you.presently
consume per week?
How many years have you consumed h'ard liquor with this frequency?
Did you consume hard liquor in the past?
How old were you when you first began to drink hard liquor?
How many years did you consume hard liquor drinks?
Do you drink beer?
How old were you when you first began to drink beer?
What is the average number of beers (12 02. bottles) you presently
consume per week?
How many years have you consumed beer with this frequency?
Did you drink beer in the past?
How old were you when you first began to drink beer?
How many years did you consume beer?
Do you drink wine?
How old were you when you first began to drink wine?
What is the average number of wine drinks you have per ,week?
How maay years have you consumed wine with this frequency?
Did you drink wine in the past?
How old were you when you frist began to drink wine?
How many years did you consume wine?
-------
page 3
Have you, your wife, or any of your children ever had any of the
following diseases or conditions.
Asthma
Bronchitis
Emphys ema
Tuberculosis-TB
Mononucleosis-Mono-
Kissing Desease
Pneumonia
Any other disease of
respiratory system
Specify:
Hepatitis or yellow
jaundice
Cirrhosis
Any other liver disease
Specify: __
Spondylitis
Gout
Rheumatoid Arthritis
Osteoarthritis
Any other diseases of bones
or joints
Specify: _«_^__
High tr iglycerides
Meningitis
Hypertension or high blood
pressure
High cholesterol
Angina pectoris
Heart .attack-Ml or coronary
Stroke-cerebral accident-(CVA)
Any. other diseases of the heart
or circulatory system in the
family
Specify: ._..__
.Diabetes
Thyroiditis
Any other glandular disorder
Specify:
For each disease listed above you will be asked to list the names of
family members who have had the disease, the consulting doctor or
clinic, and .the date of diagnosis.
Eye diseases
Specify:
Psoriasis
Eczema
Herpes zoster-shingles .-(dermatitis)
Pemphigus
Any other diseases of skin
Specify:
Anemia
Gastritis
Ulcers (stomach or duodenal)
Any other diseases of the "digestive-
system
Specif y-:_
Cancer of the breast
Cancer of the stomach
Cancer of the esophagus
Cancer of the mouth or tongue
Cancer of the large intestine
Cancer, of the rectum
Cancer of the trachea., bronchus, lung
Cancer of the liver or biliary passage
Cancer of the bladde-r o-r urinary
organs
Cancer of the skin
Cancer of the thyroid
leukemia
Other cancer
Specify;_
Mental Retardation
How many natural and step or adopted children do you have?
Please list all pregnancies and outcomes of which you were the father/
mother in the sequence .they occurred. For each pregnancy we will ask:
Were there complications?
Was the outcome a miscarriage, stillbir.th," live birth, .or other?
What was the child's sex?
What was the date of. birth?
What was tfie live birth -weight?
-------
Name the hospital, clinic, or place of birth.
What was the doctors name?
Have any of your natural children died? If yes, what was the child'-s
sex, date of birth, date of death, cause of death, and the city and
state in which the death was reported?
Do you know of any mental or physical abnormalities at birth (birth
defects) in yourself or your natural children. If yes, who had the
defect, are they living or dead, and what type of defect did they have?
Have you ever had any of the following symptoms?
Severe fever Have you had spells of dizziness?
Extreme tiredness Have you fainted or blacked out?
Frequent or very bad headaches Very strong heartbeats?
Problems with mouth or throat Irregular (fast, slow, or inconstant)
Any prodlems with ears or hearing heartbeats?
Specify: _,«___ Have you had pain, discomfort, or
Any unusual difficulty "with eyes trouble in or around your ..heart?
or eyesight other than a change Itching of the skin?
of the prescription of glasses? Any unusual discoloration or erup-
Specify: ^__ tions on the skin?
Sudden weakness or heaviness of Specify; ,
arms or legs? Any problems with your stomach or
Specify; •__ .__ digestive system?
Numbness in "arms" or legs ?Specify: .
Swelling in arms or legs? Swollen glands in your neck, armpits
Stiffness in joints or bones? or groin?
Specify: Have you lost 20 or more pounds in
Pains in joints or bones?"~" the last five years? (include
Specify; _ _ ; '_, dieting)
Spasms of limbs1? ' * " ' ~
If you have experienced any of the above symptoms you will be asked
how frequently they occured, how long the symptoms lasted, the year
you most recently experienced the symptoms, and the names of the doctors
you consulted if applicable.
Fisherwomen or the wives of male fishermen will be asked the fol-
lowing questions. If the wife is deceased we will ask that the
husband answer these questions to the best of his knowledge.
Do you use a birth control method?
What method of birth control do you use?
At what age did you begin using birth control methods?
At what age did you begin to menstruate?
Have you stopped menstruating?
At what age did you stop?
Did the cessation occur naturally or due to surgery? If due to surqery,
what was the reason for the surgery and what was the doctor's name?
-------
pa-ge 5
What is (was) the average length of your menstrual cycle? (Frpm frrs-t
day of bleeding to first day of next period)
On the average, how long is (was) your period?
In the past five years, has your blood flow during menstruation increased,
decreased, or stayed the same?
Have (or did) you experience any abnormal spotting1 or .bleeding be.tweten'
your menstrual cycles? If yes, you will be asked to give the. nawe of
the doctor you consulted if applicable.
Have; you used, or did you use, any medications prescribed .by -a doc.to.r'
for menstrual irregularities? If yes, what was the name of the medication,
-------
GREAT LAKES FISHERMEN HEALTH STUDY
Respondent Information Sheet
Respondent ID Number: I I I I
Name:
last first middle
Address:
street
city state zip
Telephone:
area code number
-------
GREAT LAKES FISHERMEN HEALTH STUDY
INTERVIEWER ASSIGNMENT RECORD
Interviewer Name:
Interviewer
Assignments -Interviews Gormen ts
Resoondent ID#
Date Assigned
Date of
Interview
!
Date of
Return
Refusal, Unavailable., etc.
-------
DEFINITIONS
62. Asthma
63. Bronchitis
64. Emphysema
65. Tuberculosis-IB
66. Mononucleosis-Mono-
Kissing Disease
67. Pneumonia
69. Hepatitis or yellow
jaundice
70. Cirrhosis of the liver
72. Sponcylitis
73. Gout
74. Rheumatoid Arthritis
75. Osteoarthritis
77. Meningitis
78. Hypertension or high
blood pressure
79. High cholesterol
80. Angina pectoris
81. Heart attack-Mi or
coronary
A condition marked by"recurrent attacks of breathing
difficulties marked by-wheezing
Inflammation and usually infection of the bronchi,
which is a part of the lung
A lung disease characterized by destruction of lung
tissue causing difficulty in breathing, often secondary
to smoking
A specific infectious disease usually affecting the lung.
A virus disease characterized by severe changes in
the white blood cells causing fatigue, severe sore
throat, and swollen lymph glands
Infection of the lungs
Inflammation of the liver, frequently infectious
A chronic disease of the liver characterized by the
replacement of normal tissue by scar tissue
Inflammation of the spine
A genetic form.of arthritis which affects chiefly men
and is due to high levels of uric acid in the blood
A persistent disease of the joints characterized by
deformity and pain in the joints
A degenerative joint disease associated with wear and
tear of the tissues and with aging
Inflammation of the membrane around the brain and spinal
cord usually caused by infection
Persistently high blood pressure
A high level of cholesterol in the blood
Sudden tightness and pain in the chest occuring during
physical exertion or tension, and subsiding with rest,
caused by disease in the coronary arteries
Death of any muscle tissue of the heart
-------
82. Stroke, cerebral accident,
(CVA)-hemorrhage, thrombosis,
embolism
84. Diabetes
85.. Thyroiditis
88. Psoriasis
39. Eczema
90. Herpes zoster-shingles
(dermatitis)
91. Pemphigus
93. Multiple Sclerosis
95. Anemia
9.6. Gastritis
97. Ulcers (stomach or
duodenal)
Death of brain tissue which results from lack of blood
to a portion of the brain or from a hemorrhage in the
brain
A disease which is characterized-by-high levels of
sugar in the blood
Inflammation of the thyroid gland
A chronic disease of the skin which usually involves
the scalp, elbows, knees, and shins
An inflammation, generally of the skin
An inflammatory disease of nerves caused 'by the virus
of chicken pox and characterized by groups of ..small.
blisters in the skin
A disease characterized by clusters of large
blisters
A disease of the central nervous system. Some symptoms
are lack of coordination, weakness, speech and visual
problems
A condition in which there are low levels of red blood
cells in the body
Persistent inflammation of the stomach
A local cavity in the inside surface of the stomach or
duodenum which usually results-from..-inflammation in
that area
100. Leukemia
101. Hodgkin's Disease
A chronic disease characterized by an abnormal, increase.
in the number of leukocytes in the tissues and often- in
the blood
A disease marked by chronic enlargement of the lymph
nodes, often cervical at the onset and then generalized,
together with enlargement of the spleen and often of the
liver
-------
GREAT LAKES FISHERMEN HEALTH SURVEY
TABLE OF OCNTEMS
!• Intrtjduction to Surveys
What is a Survey 1
Interviewing 2
Ethics of Survey Interviewing ...2
2- Using the Questionnaire
Asking the Questions 4
Maintaining Rapport « 5
Probing 7
Kinds of Probes - 8
The Don't Know Response 10
3. Question by Question Specifications. 13
4. Respondent Letter 21
5. Consent Form 22
-------
INTRODUCTION TO SURVEYS
What is a Survey?
A survey usually involves collecting data from a group
of people selected to accurately represent the population
under study. This group of people is called a sample. People
in the sample are asked a series of questions through the
use of a questionnaire. The answers obtained are put together
in an organized manner so that conclusions can be drawn. This
information is then used in planning, research, and solving
particular problems.
Skillful interviewing procedures are used to ensure full
and accurate information. Careful methods are followed so
that the data gathered from the sample or respondents can
be confidently used to represent the total population. The
use of the sample means that a small number of•respondents
can be selected.to represent the whole population, making
it possible to avoid the expensive and time consuming' proc-
edure of taking a census {a census involves a complete ac-
counting of every person in the population being studied).
Interviewing
The interviewing stage, which is one of the core opera-
tions of any survey, includes the.recruitment and training
-------
of interviewers, the preparation of general and specific
interviewing instructions,field supervision, administration
of interviews, and the validation of interviews.
During the interviewing period, the success of the survey
rests solely in the hands of the interviewer. Researchers
strive to develop the best interview guides possible, but
even the best interview guide is only as good as the inter-
viewer's skill in using it.
Ethics of Survey Interviewing
Persons working in jobs and professions which deal
with the experiences, thoughts, actions and feelings of
people have an ethical responsibility to these people. Sur-
vey research interviewing is one of these occupations, and
interviewers roust, therefore, accept the ethics of the
profession. Just as doctors and lawyers must respect in-
formation about their patients and clients as privileged,
so must the survey interviewer.
The interviewer must often ask questions that one
would not think of asking a close friend; questions which
might be thought of as "too personal." You will find that
the average person is willing to answer these questions,
sometimes offering information which would not .be given even
to a close friend or relative. Your protection of all infor-
mation about respondents gained during the conduct of re-
-------
search is therefore essential.
The main reason research studies can point to success
is confidentiality. Interviewers can, and do, promise the
people who are interviewed that their answers to the., ques-
tions will be kept strictly confidential.
-------
USING THE QUESTIONNAIRE
Asking the Questions
The interviewer's goal is to collect accurate information
through the use of the study questionnaire. Data from study
participants must be collected in a uniform manner. Thus,
all people in a sample must be asked the same questions in
the same way.
The following principles and techniques must be em-
ployed when using the questionnaire:
1. Always remain neutral. The interviewer must be
careful that nothing in words or manner implies crit-
icism, surprise, approval, or disapproval of either
the questions asked or the respondent's answers.
Through a relaxed professional attitude, the inter-
viewer can put the respondent at ease and gain his
confidence. We need the respondent's answers to the
questions with as little influence as possible by the
interviewer. Another interviewer should be able to
obtain the same answers.
2. Ask ALL questions in the order presented in the
questionnaire. Never change the order of the questions
in the questionnaire. The questions follow one an-
other in a logical sequence; to change that sequence
would subvert the intent of the questionnaire.
-------
3. Ask ALL questions exactly as worded. Do not change
even one word in the question that is printed for you.
Many times, the smallest change can affect the whole
meaning of the question. Simply repeat the question
if the need arises. If you do repeat the question,.
read all the words in the question. Even though you
feel that a question could be worded much more simply,
do not improvise on the method of asking the question.
Every word is there for a purpose. In order for all the
interviewer's work to be combined there must be.no
doubt that each respondent heard exactly the same ques-
tion before answering.
Maintaining Rapport
You began your rapport-building process'with your
introduction and it must be continued throughout the
interview. Through your accepting and understanding
behavior and your interest in the respondent,, you can
create a friendly atmosphere in which the respondent can
talk freely and fully.
Occasionally, however, rapport may be broken during
the interview because the respondent finds a particular
question "too personal" for example. If the respondent
feels a question is- too personal, take time to reassure him
that he may speak freely without fear.. This may be done by
-------
restating the confidential nature of the questionnaire and
the anonymous nature of the study. If a respondent refuses
to answer a question after you have reassured him of
confidentiality, do not press him. Record what the
respondent said in refusing to answer the question and
proceed to the next question. The interviewer should not
irritate the respondent and provoke a refusal to complete
the interview.
-------
Probing
Probing is the technique used by the interviewer to
stimulate discussion and obtain more information. -A ques-
tion has been asked and an answer is given. For any number
of reasons, the answer may be inadequate, requiring the inter-
viewer to seek more information. Probing, therefore, has
three major functions: (1) to motivate the respondent to
enlarge, clarify, or explain his answers; (2) to focus the
respondent's answers so that irrelevant and unnecessary
information can be eliminated; and (3) to pinpoint objective
information, such as dates" and names, as accurately as pos-
sible. This must be done, however, without introducing
bias or antagonizing the respondent.
You must fully understand the purpose and meaning of
each question. Once you understand the purpose of a question,
•you will find it much easier to decide if you have a satis-
factory answer or whether you should probe for a clearer and
more complete one.
Probes must alway remain neutral. Remember, probing is
to motivate the respondent to respond more fully -or to focus
his answer without introducing bias. The potential for bias
is great in the use of probes. Under the pressure of the
interviewing situation, the interviewer may..quite uninten-
tionally imply that some answers are more acceptable than
others or may hint that a respondent might want to consider
-------
this or include that in giving a response.
Kinds of Probes
A nun±>er of different neutral probes which appear as
part of a normal conversation can be used to stimulate a
fuller, complete response.
1. An expression of interest and understanding. By comments
such as "uh-huh" or "I see" or "yes", the interviewer in-
dicates that the response has been heard, that it is interes-
ting, and that more is expected.
2. An expectant pause. The simplest way to convey to the re-
spondent that you know he has begun to answer the question,
but has more to say, is to be silent. The pause often
accompanied'by an expectant look or nod of the head al-
lows the respondent time to gather his thoughts.
3. Repeat the question. When the respondent does not seem
to understand the question, misinterprets it, seems unable
to decide, or strays from the subject, it is often useful
to repeat the question. Many respondents, when hearing the
question a second time, realize what kind of answer is needed.
4. Repeating the respondent's reply. Simply repeating what
the respondent has said is often an excellent probe. Hearing
the response just given often stimulates the respondent to
further thought.
5. A neutral question or comment. Neutral questions or -com-
-------
ments are often used to obtain clearer and.fuller responses
Following are some suggestions for probing "techniques.
Probes to Clarify
What do you mean exactly?
What do you mean by ?
Could you please explain that a little?
I don't think I quite understand.
Probes for specificity
What in particular do you have in mind?
Could you be more specific about that?
Tell me about that. .What/who/how/why/when....?
Probes for data specificity
Was tha-t before or after your first hospitaliza.t.ion?
Were you married during that' time?
Who were you working for at the time?
Probes for relevance
I see. Well, let me ask you again..REPEAT EXACT QUESTION.
Would you tell me exactly how you mean that?
-------
10
Probes for completeness
What else?
What else can you think of?
What other reason/things/examples, ect. can you think.of?
The Don't Know Response
The "I don't know" answer can mean a number of things.
For instance:
1. The respondent doesn't understand the question and
says 1 don't know to avoid saying he doesn't under-
stand.
2. The respondent is thinking the question over, and
says I don't know to fill the silence and give him-
self time to think.
3. The respondent may be trying to evade the issue be-
cause he feels uninformed, or is afraid of giving
the wrong answer, or because the question seems too
personal.
4. The respondent may really not know the answer to
a question.
.If the respondent actually doesn't have the information re-
quested of him, this in itself is significant to the study
results. It is the interviewer's responsibility, however,
to make certain this is the case. An expectant pause; a
-------
11
reassuring remark, repeating the question, a neutral question-
all will encourage the respondent to reply.
Examples of Probing
The following are illustrations of probing which will
help you avoid biasing the respondent's answer:
1. Don't ask whether a person means "this or that." This
suggests only one of two answers, even though there may.
be many other possibilities which the respondent is think-.
ing about.
QUESTION; At the present time, what is the average number
of cigarettes you smoke per day?
RESPONSE; Oh, just a few.
IMPROPER PROBE; Would that be three or four?
(You are pushing the respondent to one of two alternatives
when he might mean something else entirely)
PROPER PROBE; And how many cigarettes is that?
2- Don't ask whether the respondent meant a particular thing
by a certain word. This suggests one answer, when he
might actually have another one in mind.
QUESTION:'What is your .occupation at the present time?
RESPONSE: I assist the manager.
-------
12
IMPROPER PROBE; Oh, you're a supervisor?
(The incorrect probe is an attempt to define for the re-
spondent. A neutral probe will give the respondent an
opportunity to tell what is meant.)
PROPER PROBE; Assisting the manager? Could you tell me
a little bit more about that?
-------
13
QUESTION BY QUESTION SPECIFICATIONS
Throughout the questionnaire, there are interviewer
instructions and parts of some questions which are not
meant to be read to the respondent (R). Those instruc-
tions and responses which should not be read are
CAPITALIZED. The numbered responses in mpsjt questions
should not be read to the R. Follow the interviewer's
instructions carefully and record all responses clearly
and legibly.
During an interview, a R may answer "I don't know,"
or he may refuse to answer a question. In this case, these
responses should be recorded as "don't know" or "refuse to
answer" respectively.
-------
14
QUESTION BY QUESTION SPEOFICATIONS
Cl Verify address with respondent.
Q2 Record month, day and year of birth. Do not accept "I can't
remember* as an answer. If R seems uncertain about his date of birth,
suggest that he look at his driver's license. Write out the month;
do not use a numerical value.
Q3 If R responds "single" probe to determine whether he has ever been
married. "Single" may only be a response if R has never been married.
Q4 Record number of children. If R has no children record "0" in the
blank. If R offers additional information regarding step and/or adopted
children, note this in the white space below.
Q5 If no, go to Q6. If yes, record the number of years R has held a
oormercial fishing license. Next, obtain the span of years involved
(i.e. "from 1959 to 1979") and probe for different periods of tiros.
For example, if R has held a carmsrcial fishing license for 20 years
we need to know if he held that license for a consecutive 20 year period
or if instead, he has held a license on and off for a total of 20 years
(i.e.- fron 1950 to 1960; from 1965 to 1970, and from 1974 to 1979).
C6s7 Same specifications as Q5
I? R ANSWERS NO TO Q5, 6, and 7, STOP THE INTERVIEW. TRY TO DETERMINE
IF WE HAVE THE CORRECT RESPONDENT. THANK HIM FOR HIS TIME AND COOPERATION
AND EXPLAIN THAT CURRENTLY WE ARE ONLY INTERVIEWING PEOPLE WHO HAVE
COMMERCIALLY FISHED.
Q8 Ask as written. It is especially important to obtain the state where
these towns are located.
-------
15
Q8a Ask as written. A hailing port is where the fisherman decks his
boat.
Q9 If no, go to Q10. If yes, record the number of years cotmercial
fishing has been his full time occupation. Next, obtain the span
of years involved (i.e. "from 1969 to 1979") and probe for' different
periods of time. For example, if R's full time cccuapticn was
corinercial fishing for 10 years, we need to know if this was a
consecutive 10 year period of if instead, he has been a full time
fisherman on and off for a total_of 10 years (i.e. from 1959 -to 1961;...
from 1971 to 1979).
Q10 Probe for specifics. If R holds a job which does not have a specific
job title, ask for a description of the' type of work. Probe to obtain
a job title which reflects the type of work performed-as accurately
as possible. As a rule, one-word entries are usually inadequate. For
example:
Inad£~iate Specific
Factory worker Electric motor assembler,
forge heater, punch press
operator, spray painter, turret
lathe operator
Labortir Sweeper, porter, janitor, window
washer, hand trucker, stevedore
Foreman Specify the' craft or activity
irtvolved such as foreman carpente:
foreman truck driver,, etc.
Qll Same specifications as Q9.
Q12 Correct spelling and address information are imperative. Nbta.that
employment information requests both the number of years and actual
years involved. Past interviewers report that it was easier.and smoother
to obtain all crew members' names first and then request each individual's
address and work history.
-------
16
Q13 Ask as written. Respondents usually answer with a weekly or
monthly consumption pattern. If not , read the categories to him;
probe to convert the consumption to weekly or monthly patterns.
Q14 Ask as written.
Q15 Ask as written.
Q16 Check all fish that R mentions. Be sure to record any and all other
fish not listed under "Other."
Q17 We are interested in R's family's fish consumption in comparison to
himself. For each family member/ read the choices to R, starting
with his wife. If. R does not have a wife, check #4 "Does not apply."
Blanks have been provided to record each son/daughter's consuoptibn.
Thus, if R has one daughter and two sons, the same information must
be obtained for all three children. If R responds to items $2 or 53,
record how often in the blank.
Q18 If no, go to Q18a. If yes, ask Q19-21 and record R's answers. Probe
for as accurate as possible information from the respondent.
QlSa If no, go to Q25. If yes, ask Q23-24. Me are now interested in past
srroking habits. Again, probe for as accurate as possible information
from the respondent.
Q25-61 Same specifications as Q18-24.
Q62- This section of the questionnaire identifies 32 specific diseases.
104
For each of the diseases, R's response should indicate whether he, his
wife, or his children have been told by a doctor that he or they had
this disease. It is very important that a positive response from R
be based on a doctor's diagnosis and not on his own suspicions. The
diseases are broken up into groups, i.e. lung diseases, liver diseases,
etc.
-------
17.
Starting with Q62, pronounce each disease slowly and clearly.
After each disease, pause long enough for R to respond. If R
says no, record "1" in column one. If R says yes, record "2"
in column one and obtain the following information:
Column 2: Family member with-disease
-------
18
Q127 Record R's answer; natural children only. If R has no natural
children record "0" in the blank.
Q128 Record R's answer; step or adopted children only. If R has no
ste? or adopted children record "0" in the blank.
Q129 In this section we are interested only in R's pregnancies with the
ccnnarcial fisherman. Pregnancies from a'previous raarraige should
not be included. For each pregnancy, in the order they occurred/
record the following:
Column. 2: Briefly describe any complications R experienced
during her pregnar.cy. (e.g. excessive bleeding,
breech birth, etc.) If R experienced no corolications
record "None."
Column 3: Check the appropriate box for .the outcome of the
pregnancy . •
Column 7: Check appropriate box for infant's sex
Column 8: Date of infant's birth
Column 9: Live birth weight
Colunsi 10: Full name and address of hospital, clinic, or other
place of birth. Obatin as specific an address as
possible
Column 11: Record attending doctor's full name
Q129a If R answers no, go to Q130. If R answers yes, obtain the following
information:
Column 1: Record M or F for sex of child
Column 2: Record date of birth
Column 3: Record date of death
Column 4: Record cause of death and city and state where death
was reported
Q130
If R answers no, go to Q131. If R answers yes, obtain the following
-------
19
infonnation for R or R's natural children, checking the appropriate
boxes from left to right. Record R's description of the birth
defect in the last column (i.e. cleft palate, mongoloidism, etc).
Q131 Check appropriate box. If R answers no, proceed to next question.
If R answers yes, go to Q132.
Ql31a If R answers yes, go to Q134. If R answers no, proceed to next
question.
Q132 Check the birth control method most vised during R's adult life.
Q133 Record age at which R began to use birth control methods.
QD4 Record age at which R began to menstruate.
Q135 If R answers no, go to Q137. If R answers yes, proceed to next
question.
Q136 Check appropriate answer. If cessation occurred due to surgery,
obtain reason for surgery and full name .and address of doctor. As
always, obtain as specific as possible information.
Q137 if R answers yes, record medications prescribed by a doctor. 'If-R-
cannot remember the name of the medications, suggest that she look
at the bottles. If she dees not have the bottles, record DK and-
collect any other information possible—reason she takes Dedication,
how often she takes it, etc.
Q138 Ask as written.
Q139 Ask as written.
Q140 Read the question and.choices to R. Check appropriate response.
-------
20
Q141 If R answers no, go to Q143. If R answers.yes, proceed to next
question.
Q142 If R answers no, proceed to next question. If R answers yes, obtain
doctor's full narte and address.
Q143 Same specifications as Q137.
**************************
Page 22: Observations by the Interviewer
These observations are to be recorded by the interviewer immediately
following the interview. While Ql and Q2 are strictly informational, Q3 and
Q4 CT 11 for the interviewer's judgement. If you have difficulty in making
these judgements in the preceded terms, please write out your oonments and
depressions in the Garments section below.
-------
NON-INTERVIEW REPORT
GREAT LAKES FISHERMEN HEALTH STUDY
Respondent Name: Interviewer Name:
I.D.//: Date:
* A Non-Interview Report must be completed if the result of the
contact is "First Refusal", "Second Refusal", "Unavailable",
"Other".
* A Non-Interview Report must be completed if"the result of the
contact is "Respondent Moved" and the assignment is being turned
in to the supervisor.
* A Non-Interview Report must be completed when an incomplete assign--
ment is returned after 12 or more contact attempts.
CIRCLE'APPROPRIATE NUMBER
1. With whom did you speak? (FINAL CALL)
Respondent 1
Other household member.
(SPECIFY) ..2
Neighbor... ;....., 3
Other (SPECIFY) _4.
No one 5
2. What is the result of the contact for this assignment? Please circle
appropriate number.
1 First/Second Refusal (Q3) 3 Respondent Moved (Qll)
2 Unavailable (Q8) 4 Other (Q14)
5 Maximum Calls.... . (Q16)
-------
Q3-Q7 For REFUSALS only. PLEASE CHECK WHETHER THIS IS A FIRST OR SECOND
.REFUSAL
FIRST REFUSAL
SECOND REFUSAL
3. What were the reasons given, for the refusal?
4. What do you think was the primary reason given for the refusal?
5. Was the refusal hostile, firm but NOT hostile, or mild?
Hostile... 1 (End)
Firm 2 (Q.6)
Mild 3 (Q.6)
6. Do you think another interviewer might be able to obtain the interview?
Yes 1 (Q.7)
No 2 (End)
Perhaps 3 (Q.7)
7. What type of interviewer do you think might be able to obtain the
interview?
END
-------
Q8-10 For UNAVAILABLES only.
8. Why was the respondent unavailable? (e.g* no show, disconnected phone,
respondent deceased, etc.)
9. When will the repsondent be available for an interview?
LO. Do you think there were other reasons for the non-interview? (e.g.
a polite refusal) If so, specify:
END
Qll-13 For RESPONDENT MOVED only.
LL. How did you determine that the respondent had moved?
L2. Did you obtain a new address?
13. What is the respondent's new address?
Yes.
No..
.1 (013)
.2 (End)
(Number)
(Street)
(City)
(State)
(Zip)
(Phone)
END
-------
Q14-15 For OTHER only.
14. Explain fully the circumstances involved, (e.g. language problem,
evasive or suspicious, drunk or intoxicated, respondent was never
a commercial fisherman, etc.)
15. Would you suggest any special action for this assignment? If so,
specify:
END
Q16-18 For MAXIMUM CALLS only.
16. How many attempts did you make to reach the respondent?
Number of telephone attempts
17. Did you ever speak to anyone in the respondent's household?
Yes .1 (Q18)
No 2 (End)
18. With whom did you speak and what was (were) the result(s) of your
contacts?
Respondent 1
Other household member....2
(SPECIFY)
Other (SPECIFY) 3
Result:
END
-------
APPENDIX IV
Proposed Set of Coding Instructions
-------
Coding Instructions
Great Lakes Fishermen Health Survey
The response materials for the Great Lakes Fishermen Health Survey
will consist of three records per study subject. The first record will
consist of the non-interview report materials which have a possible total
of 150 characters. The second record will consist of the responses for
the first half of the interview. Specifically, those questions directed
toward the commercial fishermen (licensee). This record has a possible
total of 503 characters. The third record will consist of the responses
for the second half of the interview. Specifically, those questions
directed toward the spouse of the commercial fisherman. This record has
a possible total of 89 characters.
Unless otherwise specified,-.responses indicating a "don't know" answer
should be coded as a -1. Coding space should remain blank in instances
where no information is available (i.e. a blank response space). In instances
where an answer requires less coding space than that provided, the spaces to
the left should be filled with zeros (e.g. three spaces provided, answer is
9 = JL JL JL )•
Partial information, at a minimum, must include a name and city/state of
residence.
-------
Record I
(Non-Interview Report)
.All non-interview reports for each study subject should be coded.
Space is provided on the coding sheet for five non-interview reports per
study subject. Note that the subject ID number is not repeated when more
than- one non-interview report is filed per subject.
1. Study subject ID number.
five digit code
Each study subject will receive a five digit identification number.
2. . Interviewer ID number.
two digit code
Each interviewer will receive a two digit identification number.
3. 'Date of non-interview report.
/ / six digit code
Code exact month, day and the last two digits of the year.
4. (Question 1). Contact status at the time of call.
single digit code
1 - respondent 4 = other
2 = other household member 5 = no one
3 = neighbor
5. (Question 2). Result of the contact for the assignment.
single digit code
1 = first refusal 4 = respondent moved
2 = second refusal 5 = other
3 = unavailable 6 = maximum number of calls
NOTE: If "maximum number of calls" is coded than Questions 16-18
should be coded. Questions 3-15 should be blank.
6. (Question 3). Reason for refusal.
/ / three digit code
First digit corresponds to a reason of "not interested".
Second digit corresponds to a reason of "not available-at the time of call",
Third digit corresponds to "other" reason.
Code "1" if mentioned as a reason for refusal.
Code "0" if not mentioned as a reason for refusal'.
-------
7. (Question 4). Primary reason for refusal.
/ / three digit code
First digit corresponds to a reason of "not interested".
Second digit corresponds to a reason of "not available at the time of call".
Third digit corresponds to an "other" reason.
Code "1" if mentioned as a reason for refusal.
Code "0" if not mentioned as a reason for refusal.
8. (Question 5). Type of refusal.
single digit code
1 = hostile
2 = firm
3 = mild
Code type of refusal indicated by interviewer.
9. (Question 6). Interviewer determination if subject interview is obtainable
by means of another interviewer.
single digit code
1 » no
2 « yes
3 « perhaps
10. (Question 8). Reasons why study subject was unavailable at the time of
call.
.__/__/ / / ' five digit code
First digit corresponds to "not home at time of call".
Second digit corresponds to "study subject has moved".
Third digit corresponds to "study subject is deceased".
Fourth digit corresponds to "phone has been disconnected".
Fifth digit corresponds to "other".
Code "1" 1f determined as a reason why subject was unavailable.
Code "0" if not determined as a reason why subject was unavailable.
NOTE: Question 8 should only be coded when the answer to Question 2 is
"unavailable", (i.e., "3").
11. (Question 12). Was a new address obtained.
single digit code
1 = no
2 *= yes
NOTE: Question 12 should only be coded when the answer to Question 2--fs
"respondent moved", (i.e., 4).
-------
10 (Question 14). "Other" reasons for non-interview report.
two digit code
Code the total number of "other" reasons wrjtten by interviewer.
13. (Question 16). How many attempts did you ma Ice to reach subject.
__- two digit code
Code.number of attempts written by interviewer.
NOTE: Question 16 should only be coded when the answer to Question 2
is "maximum calls", (i.e., 6).
14-. (Question 17). Did interviewer speak to any household member.
single digit code
1 = no
2 = ves
-------
Record I
First Non-Interview Report.
1.
2.
3. / /
4. (Question 1}
5. (Question 2)
6. (Question 3)
7. (Question 4)
8. (Question 5)
9. (Question 6)
10. (Question 8)
11. (Question 12) _
12. (Question 14)
13. (Question 16)
14. (Question 17)
(34 Characters) Total =34 Characters
Second Non-Interview Report.
2,
3, / /
4. (Question 1)
5. (Question 2)
6. (Question 3) _/_/_
7. (Question 4) / /
8. (Question 5) __
9. (Question 6) _^_
10. (Questions) __/__/_/__/_
-------
11. (Question 12) _
13. (Question 16) __
14. (Question 17) _
(29 Characters) Total = 63 Characters
Third Non-Interview Report.
2. __
3. __ / __ /___
4. ._
5. _
6.
7.
8.
9.
10 •
11. _
12. __
13. __
14. _
(29 Characters) Total = 92 Characters
Fourth Non-Interview Report.
2. __
3. __ / __ / __
4. _
5.
7-
S.
o
-------
10- _/_/_/_/_
11- _
12.
13.
14.
(29 Characters) Total = 121 Characters
Fifth Non-Interview Report
2.
3. / /
4. (Question 1)
5. (Question 2)
6. (Question 3} / /
7. (Question 4) / /
8. (Question 5)
9. (Question 6)
10. (Question 8) _/_/_/_/_
11. (Question 12) __
12. (Question 14)
13. (Question 16)
14. (Question 17)
(29 Characters) Total = 150 Characters
-------
•Subject ID number
Name of coder
Date
Record II
Great Lakes Fishermen Health Survey
1. Study subject ID number.
five digit code
Each subject win receive a five digit identification numoer.
2. Interviewer ID number.
two digit code
Each interviewer will receive a two digit identification number.
.3. Length of interview.
three digit code
The length of time required for the interview will be calculated by the
coder. Code the length of time in minutes.
4. (Question 2). Birthdate of respondent.
/ / six digit code
lode month, day, and the last two digits of the year.
5. (Question 3). Marital status of respondent.
single digit code
1 = sinqle (never married)^ 3 = separated or divorced
2 = married 4 = widowed
6. (Question 126). Sex of respondent.
.single digit code
1 = male 2 = female
7. (Question 4). Number of children of study respondent.
/ two digit code
First digit corresponds to the number of sons.
Second digit corresponds to the number of daughters.
8. (Question 5). Does respondent currently own a commercial fishing license.
/ /_ four digit code
First digit corresponds to the direct answer of Question 5.
1 = no 2 = yes
Second digit corresponds to the number of different periods of time the
respondent has held a commercial fishing license.
-------
Third and fourth digits correspond to the total number of years the study
participant has held a license.
NOTE: If the first digit is "1" corresponding to a "no" answer, than the
remaining spaces should be blank.
9. (Question 6). Did respondent ever hold a commercial fishing license.
I / four digit code
Coding instructions identical to those presented for number 8 (Question 5).
10. (Question 7). Did respondent ever crew or be a partner with an individual
owning a commercial~~fishing license.
/ / four digit code
Coding instructions identical to those presented for number 3 (Question 5).
NOTE: If Question 5-7 have been answered "no", than the rest of Record II
and Record III (coding spaces) are blank.
11. (Question 8). Lake and surrounding land masses will be stratified into segments.
/ two digit code
First digit corresponds to the Lake.
1 - Superior 4 = Erie
2 = Michigan 5 = Ontario
3 = Huron
Second digit corresponds to the lake segment.
NOTE: See appendices for lake stratifications.
12. (Question 9). Is commercial fishing respondents current full-time occupation.
/ / four digit code
Coding instructions are identical to those presented for number 8 (Question 5).
13. (Question 10). Current full-time occupation of respondent.
single digit code
1 = Occupation known to have potential exposure to hazardous materials.
2 = Occupation subject to have potential exposure to hazardous materials.
3 - Occupation not known to have potential exposure to hazardous materials.
4 = Occupation unclassifiable.
NOTE: See appendices for classification of occupations.
14. (Question 11). Was commercial fishing ever respondents full-time occupation.
/ / four digit code
Coding instructions are identical to those provide for number 8 (Question 5).
-------
IS. (Question 12). Total number of names/addresses of full- or part-t.ime
crew members.
two digit code
Code- number of names/addresses obtained from respondent.
NOTE: Partial information should be coded as if the information were complete.
16. (Question 13). How many times does respondent consume Great Lakes fish.
single digit code
1 = n.ever 5 = 1 time per month
2 .= 1-2 times per week i - 1 time per month
3..= 3-5 times per week 7 = 2-3 times per month
4.,.= 6-7 .times per week
.17... (Question 14). How many years has respondent consumed Great Lakes fish with
this' fr-equency.
two digit code
Code number of years given by respondent.
18. 'Question 15). How many pounds of fish does study participant consume per
i.ieal.
single digit code
Code response in terms of pounds of fish per meal. The response can be
coded in terms of a fraction of a pound.
19. (Question 16). What types of Great Lakes fish does respondent consume most
often.
_/__/_/_/_/_/_/_/_/_/_/_/_/_/_ fifteen digit code.
0 = not mentioned by respondent
1 = mentioned by respondent
First digit = salmon Ninth digit = lake trout
Second digit = perch Tenth digit = smelt
Third digit = walleye Eleventh digit = bass
Fourth digit = burbot Twelveth digit = lake herring
Fifth digit = rough fish Thirteenth digit = pan fish
Sixth digit = chub Fourteenth digit = other trout
Seventh digit = northern pike Fifteenth digit = other
Eighth digit = lake whitefish
NOTE: All spaces provided should be coded with either a "0" or a "1".
20. (Question 17). How often do the family members of respondent consume fish.
/ / / / / six digit code
-------
First digit corresponds to the consumption patterns of the spouse.
1 = as often as respondent 4 = does not apply
2 = more often than respondent 5 = never
3 = less often than respondent
Second-sixth digits are for additional family members when appropriate.
NOTE: Spaces should be left blank if there are no family members and should
be coded with a "-1" in those instances where a "don't know" is given
as a response.
(Question 18-24). Cigarette smoking history.
21. / two digit code
First digit corresponds to whether or not the respondent smokes cigarettes
currently.
Second digit corresponds to whether or not the respondent smoke cigarettes
in the past.
1 = no
2 = yes
NOTE: Both digits should be coded. If both digits correspond to "no" than
the coding spaces for 21a should remain blank. If either, of the digits
correspond to "yes" than-'-the spaces for 21a should be coded.
21a. / / six digit code
First and second digits correspond to the age at which the respondent began-to
smoke cigarettes.
Third and fourth digits correspond to the average number of cigarettes the
respondent smokes (had smoked) per day.
•Fifth and sixth digits correspond to the number of years the respondnet smoked
cigarettes (with this frequency).
NOTE: Twenty-one a should be blank if 21 is coded _i_/_l_ . Twenty-one a should
be coded if either digit in 21 is coded with a "2".
(Question 25-31). Pipe smoking history.
22. / two digit code
First digit corresponds to whether or not respondent smokes a pipe regularly.
Second digit corresponds to whether or not respondent smoked a pipe in the past..
1 = no
2 a yes
NOTE: Both digits should be coded. If both digits are answered- "no" than, the
coding spaces for 22a should remain blank. If either of the digits-
correspond to "yes", than the spaces for 21a should be coded.-
22a. / / six digit code
First and second digits correspond to the age at which respondent began to smoke
a pipe.
-------
Third and fourth digits correspond to the number of pipefuls the respondent
smokes (had smoked) per day.
Fifth and sixth digits correspond to the number of years the respondent smoked
a pipe (,w.ith this frequency).
.NOTE:. Twenty-two a should be blank if 22 is coded _!_/ 1 . Twenty-two a should
be coded if either digit in 21 is coded with a T".
(Questions 32-38). Cigar smoking history.
23- __/_ tw° digit code
First digit corresponds to whether or not the respondent smokes cigars
currently.
Second digit corresponds to whether or not the respondent smoked cigars in
the past.
1 = no
2 - yes
NOTE: Both digits should be coded. If both digits correspond to a "no" than
the coding spaces for 23a should remain blank. If either of the digits
correspond to "yes" than the spaces for 23a should be coded.
22a. / / six digit code
First and second digits correspond to the age at which the respondent began
to smoke cigars.
Third and fourth digits correspond to the average number of cigars the respon-
dent smokes (had smoked) per day.
Fifth and sixth digits correspond to the number of years the respondent
smoked cigars (with this frequency).
NOTE: Twenty-three a should be blank if 23 is coded _!_/_!_. Twenty-three a
should bfi coded if either digit in 23 is coded with a "2".
24. (Question 39). Has the respondent ever chewed tobacco regularly.
single digit code
T = no
2 = yes
25. (Question 40). Has the respondent ever used snuff regularly.
single digit code
1 = no
2 = yes
(Question 41-47). Liquor drinking history.
26. / two digit code
First digit corresponds to whether or not the respondent drinks liquor
regularly.
-------
Second digit corresponds to whether or not the respondent drank liquor in the
past.
1 = no
2 » yes
NOTE: Both digits should be coded. If both digits correspond to "no", than
the coding spaces for 26a should remain blank. If either of the digits
correspond to "yes" than the spaces for 26a should be coded.
26a. / / six digit code
First and second digits correspond to the age at which the respondent began to
drink liquor.
Third and fourth digits correspond to the average number of hard liquor drinks
the respondent consumes (had consumed) per week.
Fifth and sixth digits correspond to the number of years the respondent drank
hard liquor (with this frequency).
NOTE: Twenty-six a should be blank if 26 is coded _!_/_!_. Twenty-six a should
be coded if either digit in 26 is coded with a "2".
(Questions 48-54). Beer drinking history.
27. / two digit code
First digit corresponds to whether or not the respondent drinks beer currently.
Second digit corresponds 'to whe-ther or not the respondent drank, beer in the
past.
1 = no
2 = yes
NOTE: Both digits should be coded. If both digits correspond to "no", than
the coding spaces for 27a should remain blank. If either of the digits
correspond to "yes8, than the spaces for 27a should be coded.'
27a. / / six digit code
First and second digits correspond to the age"at which respondent.began to
drink beer.
Third and fourth digits correspond to the number of beers the respondent
drinks (had drunk) per week.
Fifth and sixth digits correspond to the number of years the respondent
drank beer (with this frequency).
NOTE: Twenty-seven a should be blank if 27 is coded _1_/J_. Twenty-seven a
should be coded if either digit in 27 is coded with a "2".
(Questions 55-61). Wine drinking history.
28. _/__ two digit code
First digit corresponds to whether or not the respondent drinks wine currently.
Second digit corresponds to whether or not the repsondent drank wine in the
past.
1 = no
2 = yes
-------
f.'OTE: Both digits should be_coded. If both digits correspond to "no" than
the coding spaces for 28a should remain blank. If either of the digits
correspond to "yes", than the spaces for 28a should be coded.
-_^ •__/__ _V__ _ six digit code
First and second digits correspond to the age at which respondent began to
drink wine.
Third and fourth digits correspond to the number of wine drinks the respondent
consumes' (had consumed) per week.
Fifth and sixth digits correspond to the number of years the respondent dranic
winre(with this frequency).
NOTE: Twenty-eight a should be blank if 28 is coded _!_/_!_. ' Twenty-eight a
should be coded if either digit in twenty-eight is coded with a "2".
(Questions 62-105). Familial medical conditions of respondent.
29. _/_/_/_/_ five digit code
-First digit corresponds to the acknowledgement of a disease/condition
variable within the family.
1 = no
2..= yes
NOTE: If the first digit is coded with a "1" than the remaining coding spaces
should remain blank. If the first digit is coded with a "2", than
the remaining digits must be coded.
Second digit corresponds to the family member with the disease/condition.
1 = self 3 = son
2 = spouse 4 = daughter
Third digit corresponds to whether or not information is provided regarding the
name(s) and address(es) of medical personnel and/or service. Partial informa-
tion should be coded as if information were complete.
1 • = no
2 = yes
-1 = don't know (no information given)
Fourth digit corresponds to whether or not information is provided regarding
the date of diagnosis. Partial information should be coded as if information
were complete.
1 = no
2 = yes
-1 = don't know (no information given)
Fifth digit corresponds to whether or not the disease/condition was verified
by medical record.
1 = no
2 = yes
NOTE: The five digit coding sequence (i.e.., _ / _ / _ / _ / _ ) and the instructions
for each of the digits are to be USP^ for Questions 62-104.
-------
(Question 105-125). Symptomatology of respondent.
72. / / /__/ / / eight digit code
First digit corresponds to the acknowledgement of a symptom experienced by
by the respondent.
1 = no
2 = yes
NOTE: If the first digit is coded with a "1" than the remaining coding spaces
should remain blank. If the first digit is coded with a "2", than the
remaining digits must be coded.
Second digit corresponds to whether or not information regarding the frequency
of symptoms was given by the respondent.
1 = no
2 = yes
-1 = don't know
Third digit corresponds to whether or not information regarding- the duration
of symptoms was given by the respondent.
1 = no
2 = yes
-1 = don't know
Fourth digit corresponds to whether or not respondent consulted a physician.
1 a no
2 = yes
NOTE: If the fourth digit is coded with a "1" than the remaining digits should
be blank. If the fourth digit is-coded with a "2", than the fifth
through eighth digits must be coded.
Fifth digit corresponds to whether or not information is provided regarding the
name(s)/address(es) of medical personnel and/or service. Partial information
should be coded as if the information were complete.
1 = no
2 3 yes
-1 = don't know
Sixth and seventh digits correspond to the last two digits of the year .the
symptom was most recently experienced (e.g., 1979 =_7__9_).- Code -1 -1- if
respondent doesn't know.
Eighth digit corresponds to whether or not the symptom was verified by medical
record.
1 = no
2 = yes
NOTE: The eight digit coding sequence (i.e., _/__/__/ /__/__/ ) .and the
instructions for each of the digits are to be used for Questions 105-125,
-------
Record II
1. _____ ____
2, . __
'3. ; ___ _•
(Question ?.} 4- __ / __ / __
(Question 3) 5. _
(Question 126) 6. __
(Question '4) 7. _ / _
(Question 5) 8. __/__/ __
( Qu es t i o ri 5 ) 9 . __/__/ __
(Question 7) 10. __/ _ /; __
(Questions) 11. __/ _
(Question 9) 12. __/__/__
(Question 10) 13.
(Question 11) 14. _/__/ __
( Quest io.n, 12) .15. __
(Question 13) 16. _
(Question 14) 17. _ _
(Question 15) IS. _
(Question 16) 19. _/_/__/_/_/_/_/_/_/_/_/_/_/__/.
(Question 17) 20. __/___/___/__/__/__
21.' /
^
(Question 18-24) <
\
. 22. /
(Question 25-31)
^ .22a. __•/_ /
23. /
(Question 32-38)
" -23a. / /
-------
(question 39)
(Question 40)
(Question 41-47) <
(Question 48-54) <
(Question 55-61 )<
(Question 62}
(Question 63).
(Question 64)
(Question 65)
(Question 66)
(Question 67)
(Question 68)
(Question 69)
(Question 70)
(Question 71)
(Question 72)
(Question 73)
(Question 74)
(Question 75)
(Question 76)
(Question 77)
(Question 78)
(Question 79)
(Question 80)
(Question 81)
(Question 82)
24. _
25. _
/>* 25< _/_
X
\25a . / /
\27a / /
/^ 28. __/__
29. _/__/__/__/_
30 • __/_/_/__/_
31. _/__/__/__/_
32 • _/__/_/„/_
33. __/__/__/__/_
34 • _/_/_/„/_
35 • _/_/__/_/_
36 • _/_y_-/__/_
37 • -_*L_/__/.__/_
38. _/_/__/__/_
39. _/_/_/_/_
40 - ^/__/_jL_/_
41 . __/__/_/„/__
«. __/„/„/„/_
43 • __/_/__/_/_
44 . _J_J_J_I
45 • __/_/„/_/__
46. _/__/_/_/_
47 . _/__/__/__/_
48 . _/__/_/_/_
49. (Iff
-------
(Question 83)
(Question 84)
(Question 85)
(Question .8.6)
(Question-87)
(Question 88)
..(Question 89)
(Question 90)
(Question. 91.)
(Question 92.)
(Question 9.3)
(Question 94)
(Question 95)
(Question 96)
(Question 97)
(Question 98)
(Question 99)
(Question 1.00)
(Question 101)
(Question 102)
(Question 103)
(Question 104)
(Question 105)
(Question 106)
(Question 107) 74. _/_/_/_/__/__/__/_
(Question 108) 75. _/__/_/_/__/__/ /.
(Question 109) 76.
(Question 110) 77,
(Question 111) 73.
73. /_/ /_/__/_/__/
_/_/__/_/_/_/_/_
_/__/__/_ /_/_/_/_
-------
(Question 112) 79. _/_/__/_/__/_/_/_
(Question 113) 80. __/_/__/__/_/__/__/_
(Question 114) 81. _/__/_/_/_/__/_/__
(Question 115) 82. _/_/__/_/_/_/_/_
(Question 116) 83. __/_/__/„/__/„/__/_
(Question 117) 84. - _/__/_/_/_/_„/_/„
(Question 118) 85. _/_/__/_/_/_/__/_
(Question 119) 86. _/__/__/__/_/__/„/_
(Question 120) 87. _/„/__/_/_/__/__/_
(Question 121-} 88. _/_/„/„/_/__/_./__
(question 122) 89. _/__/__/_/_/__/__/_
(Question 123) 90. _/_/„/_/_/„_/_/__
(Question 124) 91. _/__/_/_/__/__/__/„
(Question 125) 92. _J „! _J _J _J _J _J __
Total = 503 Characters
-------
.Record III
Great Lakes Fishermen Health Survey
93. (Question. 127). Number of natural children of the respondent.
two digit code
Code number of children given by respondent.
94. (Question 128). Number of step or adopted children of the respondent.
two digit code
Code number of children given by respondent.
95-100. (Question 129). Information regarding pregnancies and pregnancy outcomes
of respondent.
_/_/_/_/_/__/_/__/ nine digtt code
First digit corresponds to the order of pregnancies of the respondent.
(e.g., Third pregnancy = 3 ).
Second digit corresponds to whether or not there were any complications
during this pregnancy.
1 - no
2 = yes
Third digit corresponds to the outcome of the pregnancy.
1 = miscarriage 3 = live birth
2 = stillbirth 4 = other (abortion)
NOTE: If the" third digit is coded "1" or "4" than the fourth,.fifth, and
sixth digits should be blank.
Fourth digit corresponds to the infants sex.
1 = male
2 = female
Fifth digit corresponds to whether or not a date of birth is provided by
the respondent.
1 = no
2 * yes
Sixth digit corresponds to whether or not a live birth weight is provided
by the respondent.
1 - no
2 = yes
NOTE: The third digit must be coded with a "2". If the third digit is
coded with a number other than a "2" than this space should be
blank.
-------
Seventh digit corresponds to whether or not information is provided
regarding the name(s)/address(es) of hospital or place of birth.
1 = no
2 - yes
-1 = don't know
Eighth digit corresponds to whether or not information is provided
regarding the name(s) of the attending physician(s).
1 = no
2 = yes
-1 =• don't know
Ninth digit corresponds to whether or not the pregnancy and outcome was
verified by medical record.
1 = no-
2 = yes
NOTE: There are spaces provided for six pregnancies on the coding sheet.
If more than six pregnancies, code the first six pregnancies 1-isted
on the questionnaire.
101. (Question 129a).
: /__/_ three digit code
First digit corresponds to whether or not any of the respondents natural
children have died.
1 = no
2 = yes
NOTE: If the first digit is coded "1" than the second and third digits
should be blank. If the first digit is coded with a "2" than the
second and third digits must be coded.
Second digit corresponds to whether or not any information regarding the
child's sex, date of birth, date of death, name, cause of death, and
the city and state in which the death was reported.
I = no
2 =• yes
NOTE: Partial information should be considered as a "yes" answer. The
most important pieces o.f information are the name of the child and
the city and state in which the death was reported.
Third digit corresponds to whether or not the death certificate was
obtained from the respective state agency.
1 = no
2 = yes
-------
102. (Question 130).
/ two digit code
First.digit corresponds to whether or not the .respondent.or. any of the
respondent's natural children have birth defects.
1 = no
2 = yes
NOTE: If the first digit is coded with a "1" than the second digit should
remain blank. If the first digit is coded with a "2" than the
second digit must be coded.
Second digit corresponds to whether or not information is provided regard-
ing the relationship to the respondent, the current status (i.e. ..living
or dead) of the individual, and the nature of the birth defect.
1 = no
2 * yes
NOTE: Partial information should be considered as a "yes" response.
The most important piece of information is the nature of the birth
defect.
103. (Questions 131-131a).
/ two digit code
First digit corresponds to whether or not the respondent uses a birtti
control method in the past.
1 = no
2 - yes
NOTE:' Both spaces should be coded.
104. (Question 132). What method of birth control was most frequently used
during the respondent's adult life.
single digit code
1 = pill 5 - rhythm
2 = I.U.D. 6 = condom
3 = diaphram 7 = two of the above
4 = foam or jelly 8 = other
105. (Question 133). Age at which' respondent began to use birth control methods.
/ two digit code
Code age given by respondent.
105. (Question 134). At what age did the respondent begin to menstruate.
/ two digit code
-------
107. (Questions 135-137).
_/__/__/_/_/ six digit code
First digit corresponds to whether or not the respondent has stopped
menstruating.
1 = no
2 = yes
NOTE: If the first digit is coded "1" than the second, third and fourth
digits should remain blank. If the first digit is coded "2" than
the remaining digits should be coded.
Second digit corresponds to whether the menstrual cessation was natural
or due to surgery.
1 = natural
2 = surgery
NOTE: If the second digit is coded with a "1" than the third digit should
be blank. If the second digit is coded with a "2" than the third
digit should be coded.
Third digit corresponds to whether or not information is given regarding
the reason for surgery and the attending physician's name and address.
1 = no
2 = yes
-1 = don't know
NOTE: Partial information should be considered as a "yes" response.
Fourth digit corresponds to whether or not the respondent to any medica-
tions since menstruation began.
1 = no
2 = yes
Fifth digit corresponds to whether.or not information is provided regarding
the names of the medication.
1 = no
2 = yes
NOTE: Partial information should be coded as a "yes" response.
Sixth digit corresponds to whether or not surgical cessation of menses was
verified by medical record.
1 = no
2 = yes
108. (Question 138). What was the average number of days between the respondent's
periods.
two digit code
Code average number of days.
-------
109. (Question 139). What was the average length of the respondent's period.
two digit code
Code average number of days.
ITO. (Question 140). Has;respondent's blood flow during menstruation changed .
in the last five years.
single digit code
1 = increased 3 = stayed the same
2 = decreased 4 = does not apply
.111. (Question 141).
__/ /__/__ four digit code
First digit corresponds to whether or not the respondent experience any
•spotting or bleeding between menstrual cycles.
1 f. no
2 = yes
HOTE: If the first digit is coded with a "1" than the remaining spaces
should be blank. If the first digit is coded as a "2" than the
second digit must be coded.
Second digit corresponds to whether or not the respondent consulted a
.physician.
1 = no
2 = yes
NOTE: If the second digit is coded "1" than the remaining spaces should
be blank. If the second digit is coded with a "2" than the
remaining digits must be coded.
Third digit corresponds to whether or not information'is provided regarding
the name(s) and address(es) of physicians consulted for this problem.
1 = no
2 = yes
NOTE: Partial information should be considered as a "yes" response.
Fourth digit corresponds to whether or not respondent's condition was
verified by medical record.
1 = no
2 * yes
112. (Question 143).
/ two digit code
First digit corresponds to whether or not the respondent used medications for
menstrual irregularities.
1 = no
2 = yes
-------
Second digit corresponds to whether or not information is provided by
respondent regarding the name(s) of the medication(s).
i
1 = no
2 =* yes
-1 = don't know
NOTE: Second digit should be blank if the first digit is coded with a "1".
113. (Questions 1-126). Were answered by:
single digit code
1 = study subject
2 = spouse of study subject
3 = other
114. (Questions 127-143). Were answered by:
single digit code
1 = study subject
2 = spouse of study subject
3 = other
-------
Record III
(Question 127) 93.
(Ques.tiQn.128) 94.
95.
(Question 129)
.J-J-J-J-J-J-J-
129a)
130)
131-131a) 103.
(Question
(Question
(Question
(Question
(Question
(Question
(Question
(Question
(Question
(Question
(Question 141-142) 111. __/_/_/_
/_/_/_/_/_/_/_/,
/__/_
132}
133)
134)
135-137)
138)
139)
140)
104.
105.
106.
107.
108.
109.
110.
(Question 143)
112.
113.
114.
Total = 89 Characters
-------
CODES TO BE USED FOR OCCUPATIONS (QUESTION 10)
01 = Professional
02 = Officials and Managers
03 = Sales Workers
04 = Clerical Workers
05 = Craftsmen
06 = Foremen
07 - Operatives — non transport
09 =» Operatives - transport equipment
10 =» Laborers - unskilled
11 = Unemployed
12 = Disabled
13 = Retired
88 = Not Reported
14 = Student
15 = Military
16 = Farm Managers or Farmers
17 = Service Workers (except Private Household Workers)
18 = Private Household Workers
U.S. Bureau of Census: 1970 Census of Population .
Classified Index of Industries and Occupations.
U.S. Government Printing Office, Washington, D.C., 1971
-------
CODES FOR CAUSES OF. DEATH (QUESTION I29a)
'0-1- = Diseases of heart
02 = Malignant neoplasms (tumor, leukemia, carcinoma)
03 = Cerebrovascular diseases
04 = Accidents
'05 = Influenza and pneumonia
06.= Tuberculosis, all forms
07 = Diabetes mellitus
08 = Bronchitis, emphysema, and asthma
09 = Cirrhosis of liver
10 = Suicide
11 = Congenigal anomalies
12 = Homicide, War
13 = Nephritis and nephrosis
14 = Peptic ulcer, Hemorrhaging
15 = Other Vascular Diseases
16 = Old age
17 = Other.- Alcoholism
99 = Unknown
-------
CODES TO BE USED FOR BIRTH DEFECTS (QUESTION 130)
1 = Central nervous system -
l=Anencephaly 3=Hydrocyphalus 5=Encephalocele
2=Spina Bifida 4=Microencephaly 6=not specified
2 = Craniofacial -
l=Cleft palate 3=Congenital cataract 5=not specified
2=Cleft lip + palate 4=Anopthalmus
3 = Cardiovascular -
l=Transposition 3=Coarctation 5 not specified
2=Tetralogy 4=Ventricular septal
defect
4 = Gastrointestinal -
l=Tracheo-esophageal atresias 4=Pyloric stenosis
2=Small bowel atresias 5=Diaphramatic hernia
3=Anorectal atresias 6=0mphalocele
7=not specified
5 = Genitourinary - .
l=Exstrophy of bladder 3=Hypospadias
2=Septic kidney disease 4=not specified
6 = Musculoskeletal
l=Club foot 4=Polydactyly
2=Reduction deformities 5=Syndactyly
3=Dislocated hips 6=not specified
7 = Chcomosomal -
l=Down Syndrome 3=Trisomy E
2=Trisomy O 4=not specified
Codes adapted from - Center for Disease Control: Congenital
Malformations, Surveillance, July 1978 - June 1979, issued
July 1980.
-------
APPENDIX V
Proposed Set of Analytical Procedures
(appropriate for pilot questionnaires only)
-------
Ql. Check answer against data file and correct data file if
necessary.
Q2. Categorical response will be used to stratify participants
in analysis. (e.g., search males for prostate cancer.) A
frequency count to describe the cohort will be performed.
Q3. Categorical response will be used to stratify participants
in analysis. (e.g.? search Blacks for heart disease.) A
frequency count to describe the cohort will be performed.
Q4. Birthdate will be utilized for:
a) determining age distribution of the cohort
b) determining age of the participant
c) to determine the age of the participants at the given
life events of Qll, Q13, Q67, to Q118
Q5. Answer will be utilized to obtain medical records and death
certificates when applicable.
6. Categorical response will be used to stratify participants
in analysis. (e.g., search marrieds for lung cancers.) A
frequency count to describe the cohort will be performed.
Q7. Answer will be used to"determine years of possible exposure
potential. Reliability will be determined by comparison with
Qll, Q12, and Q13 and available state records.
Q8. Answer will be used to determine years of possible exposure
potential. Reliability will be determined by comparison with
Qll, Q12, and Q13 and available state records.
Q9. Answer will be used to.determine years of possible exposure
potential. Reliability will be determined by comparison with
Qll, Q12, and Q13 and available state records.
NOTE: Q7, Q8, and Q9 will serve as exclusion criteria. If
the answer is "no" for the three questions participants will
be excluded from the analyses.
QIC. Each lake will be stratified into segments. Each segment
will be environmentally characterized according to the degree
of contamination with PCBs and other organics. Characteriza-
tion will be based upon available environmental data regard-
ing 'organics in water, sediments, and fish. Fishing location
will be used to determine exposure potential.
QlOa. Answer will be used in conjunction with Q10 to determine
expsure potential.
NOTE: QlOa is not found in Protocol III.
Qll. Answer will be used to determine years of possible exposure
potential. Reliability will be determined by comparison -with
Q7, Q8, and Q9 and available state records.
-------
Q12. Answer will be used to determine if a possible confounding
exposure is occurring through a secondary occupation.
Q13. Answer will be used to determine years-of possible exposure
potential. Reliability will be determined by comparison with
Q7, Q8, and Q9 and available state records.
Q14. Answer will be used to determine if sport fishing and poten-
tial sport fish consumption may be an additional route of
exposure. A frequency distribution on this variable will be
•performed.
Q15. Answer will be used to determine if sport fishing and poten-
tial sport fish consumption may be an additional route of
exposure. A frequency distribution on this variable will be
performed.
Q16. Answer will be used to determine in which states and what
types of waterbodies study participants sport fish.
Q17. .Answer(s) will'be:
a) added to the data file for compiling a future cohort
b) a frequency distribution will be compiled regarding
the number of names listed per questionnaire.
Q18. Answer is a dichotomous variable that will be used to:
a) serve as an exclusion criterion for Q19 to Q27
b) determine a frequency distribution for the percent of
this cohort reporting a possible exposure
Q19. -Answer will provide a frequency distribution pn the number
of meals consumed containing Great Lakes fish. Participants
will then be stratified on this variable to determine odds
ratios for Q67 to Q118, and Q119 to Q140.
Q20:. Answer will be used to determine the relative frequency of
consumption for each type of fish from the Great Lakes.
Participants will be stratified according to specific fish
species consumed.
Q21. A frequency distribution will be compiled. This variable
will then be stratified and used as a criterion for grouping
participants by years of possible exposure for comparison
against Q67 to Q118, and Q119 to Q149.
Q22. Answer will provide a frequency distribution on the number
of meals consumed containing non-Great Lakes fish. Partici-
pants will then be stratified on this variable for determining
ratios for Q67 to-Q118, and Q119 to Q140. The answer will
pertain to inland water fish meals as opposed to Great Lakes.
in addition, this variable will be used to determine total
potential exposure via fish consumption from any source.
-------
Q23. Answer will be used to determine the relative frequency of
consumption for each type of fish from inland waters. Par-
ticipants will be stratified according to specific fish species
consumed. In addition, this variable will be used to determine
total consumption of specific fish species from any source.
Q24. A frequency distribution will be compiled. This variable will
then be stratified and used as a criterion for grouping par-
ticipants by years of possible exposure for comparison against
Q67 to Q118, and Q119 to Q140. In addition, this variable
will be used to determine total potential exposure via fish
consumption.
Q25. Frequency counts will be made separately for wives, sons, and
daughters. Stratifications performed in Q19 and Q22 will be
compared with the three frequency counts to determine potential
family member exposures via fish consumption.
Q26.. A frequency distribution will be calculated for the entire
cohort. Reliability will be determined by comparisons with
Q19 and Q22. Subjects will then be stratified by this vari-
agle for morbidity analysis with Q67 to Q118, and Q119 to Q140.
Q27. A frequency distribution will be calculated for the entire
cohort as well as for the various levels of stratification in
Q20 and Q23.
Q28. Answer will be used to r-^tablish a frequency count. Risk
ratios will be calculat^-n regarding the presence or absence
of this variable and the orevalence of a morbid condition,
listed in Q67 to Q118, and Q119 to Q140.
Q29. This answer will be used to determine years of exposure to
cigarettes (i.e., smoking history).. A frequency distribution
will be calculated.
Q30. This answer will be used to establish a frequency count. Risk
ratios will be calculated regarding high, medium, and low levels
of consumption and the prevalence of a morbid condition listed
in Q67 to Q118, and Q119 to Q140.
Q31. This answer will be used to establish a frequency count. Risk
ratios will be calculated regarding the number of years smoked
and the prevalence of a morbid condition listed in Q67 to^QHS,
and Q119 to Q140.
Q32. This question is to aid the participant in answering Q33. A
frequency distribution will be calculated.
Q33. This answer will be used to establish a frequency count. Risk
ratios will be calculated regarding the number of years smoked
and the prevalence of a morbid condition listed in Q67 to.QHS,
and Q119 to Q140.
-------
Q34. This answer will be used to establish a frequency count. Risk
ratios will be calculated regarding the presence or absence of
this variableJand the prevalence of a -morbid condition, Isited
in Q67 to Q118, and Q119 to Q140.
Q34a. This answer will be used to establish a frequency count. Risk
ratios will be calculated regarding the presence or absence of
this variable and the prevalence of a morbid condition, listed
in Q67 to Q118, and Q119 to Q140.
Q35. This answer will be used to determine years of exposure to
pipes (i.e., smoking history). A frequency of distribution'
•will 'be calculated.
Q36. This answer will be used to establish a frequency count. Risk
.ratios will be calculated regarding high, medium, and low levels
of consumption and the prevalence of a morbid condition listed
in Q67' to Q118 and Q119 to Q140.
Q37. This answer will be used to establish a frequency count. Risk
ratios will be calculated regarding the number "of years smoked
and the prevalence of a morbid condition listed in Q67to Q118,
and Q119 to Q140.
Q38. This question is to aid the participant in answering Q39. A
frequency distribution will be calculated.
.Q39, This answer will be used to establish a frequency count. Risk
ratios will be claculated regarding.the number of years smoked
and the prevalence of a morbid condition listed in Q67 to Q118,
and Q119 to Q140.
Q40. 'This answer will- be used to establisn a rrequency count. RISK
ratios will -be calculated -regarding the presence or absence of
this variable and the prevalence of a morbid condition, listed
in Q67 to Q118, and Q119 to Q140.
Q40a. This answer will be used to establish a frequency count. Risk
ratios will be calculated regarding the presence or absence of
this variable and the prevalence of a morbid condition, listed
in Q67 to Q118, and Q119 to Q140.
Q41. This answer will be used to determine years of exposure to cigars
(i.e., smoking history). A frequency distribution will be
calculated.
Q42.. This anser will be used to establish a frequency count. Risk
ratios will be calculated regarding high, medium, and low levels
of consumption and the prevalence of a morbid condition listed
in Q67 to Q118, and Q119 to Q140.
Q43. This answer will be used to establish a frequency count. Risk
ratios will be calculated regarding the number of years smoked
.and the prevalence-of a morbid condition listed in Q67 to Q118,
and Q119 to Q140.
-------
Q44. This question is to aid the participant in answering Q45. A
frequency distribution will be calculated.
Q45. This answer will be used to establish a frequency count. Risk
ratios will be claculated regarding the number of years smoked
and the prevalence of a morbid condition listed in Q67 to Q118,
and Q119 to Q140.
Q46. This answer will be used to establish a frequency count. . Risk
ratios will be calculated regarding the presence or absence of
this variable and the prevalence of a morbid condition, listed
in Q67 to Q118, and Q119 to Q140.
Q47. This answer will be used to establish a frequency count. Risk
ratios will be calculated regarding the presence or absence of
this variable and the prevalence of a morbid condition, listed
in Q67 to Q118, and Q119 to Q140.
Q48. This answer will be used to establish a frequency count. Risk
ratios will be calculated regarding the presence or absence of
this variable and the prevalence of a morbid condition, listed
in Q67 to Q118, and Q119 to Q140.
Q49. This answer will be used to establish a frequency count. Ris.k
ratios will be calculated regarding the presence or absence of
this variabel and the prevalence of a morbid condition, listed
in Q67 to Q118, and Q119 to Q140.
Q49a. This answer will be used to establish a frequency count. Risk
ratios will be calculated regarding the presence or absence of
this variable and the prevalence of a morbid condition, listed
in Q67 to Q118, and Q119 to Q140.
Q50. This answer will be used to determine years of exposure to hard
liquor. A frequency distribution will be calculated.
Q51. This answer will be used to establish a frequency count. Risk
ratios will be calculated regarding high, medium, and low.levels
of consumption and .the prevalence of a morbid condition listed
in Q67 to Q118, and Q119 to Q140.
Q52. This answer will be used to establish a frequency count. Risk
ratios will be calculated regarding the number of years drunk
and the prevalence of a morbid condition listed in Q67 to Q118>
and Q119 to Q140.
Q53. This question is to aid the participant in answering Q54. A
frequency distribution will be calculated.
Q54. This answer will be used to establish a frequency count. Risk
ratios will be calculated regarding the number of years drunk
and the prevalence of a morbid condition listed in Q67 to Q118,
and Q119 to Q140.
Q55. This answer will be used to establish a frequency count. Risk
ratios will be calculated regarding the presence or absence of
this variable and the prevalence of a morbid condition, listed
in Q67 to Q118, and Q119 to Q140.
-------
Q56. This answer will be used to determine years of exposure to
beer (i.e., drinking history). A frequency distribution will
be calculated.
Q57. This answer will be used to establish a frequency count. Risk
ratios will be calculated regarding' high, medium, and low levels
of consumption and the prevalence of a morbid condition listed
in Q67 to Q118, -and Q119 to Q140.
Q58. This answer will be used to establish a frequency \count., Risk
ratios will be calculated regarding the number of years drunk
and the prevalence of a morbid condition listed in Q67 to Q118,
and Q119 to Q140.
Q.6'0.. This answer will be used to establish a frequency count. Risk
ratios will be calculated regarding the number of years smoked
and the prevalence of a morbid condition listed in Q67 to Q118',
and Q119 to Q140.
•Q61. This answer will be used to establish a frequency count. Risk
ratios will be calculated regarding the presence or-absence of
this variable and the prevalence of a morbid condition, listed
in Q67 to Q118, and Q119 to Q140.
•Q61a.. This answer will be used to establish a frequency count. Risk
ratios will be calculated regarding the presence or absence of
this variable and the prevalence of a morbid condition, listed
in Q67 to Q 118, and Q119 to Q140.
Q62. This answer will be used to determine years'of exposure to
wine (i. e. >• drinking history) . A frequency•distribution will
be calculated.
Q63. 'This answer will be used to establish a frequency count. .Risk .
ratios will be calculated regarding high, medium, and low levels
of consumption and the nrevalence of a morbid condition listed
in Q67 to Q118, and Q119 to Q140.
Q64. This answer will be used to establish a frequency count. Risk
ratios will be calculated regarding the number of years drunk
and the prevalence of a morbid condition listed inQ67 to Q118,
and Q119~ to Q140.
Q65. This question is to aid the participant in answering Q66. A
frequency distribution will be calculated.
Q66. This answer will be used to establish a frequency count. Risk
ratios will be calculated regarding the number of years drunk
and the prevalence of a morbid condition listed in Q67 to Q118,
and Q119 to Q140.
-------
Q67 to Q118.
Q119 to:Q140,
Q141,
"A"
2
A frequency distribution will be compiled for each disease
by the number of:
a) Fishermen who have the disease
b) Wives who have the disease
c) Husband and wives how have the disease
d) Children per family who have the disease
e) Fishermen and one or more children who have the disease
f) Wife and one or more children who have the disease
g) Husband, wife, and one or more children who have the
disease
Risk ratios of disease prevalence versus (Q18, Q19, Q20, Q21, Q22,
Q23, Q24, Q25, Q26, Q27, Q28, Q28a, Q34, Q34a, Q40, Q40a, Q49,
Q49a, Q55, Q55a, Q61, Q61a) will be calculated.
( } these questions will hereafter be refered to as set
Risk ratios of disease prevalence versus (Q30, Q36, Q42, Q51,
Q57, Q63, Q31, Q37, Q43, Q52, Q64, Q33, Q39, Q45, Q54, Q60, Q66)
will be calculated.
2
( ) these questions will hereafter be refered to as-set "B"
All information provided for these questions will be added to
the date file including validation—confirmation or denial.
A frequency distribution will be compiled for each symptom oy.
the number of:
a) Participants who respond Yes
b) Frequency "of occurrence
c) Length of symptoms
d) Participants who consulted a Doctor
Risk ratios of symptom prevalence will be calculated against
question sets "A" and "B".
All information provided for these questions will be added to
the date file including validation—confirmation or denial.
A frequency distribution will be compiled for:
a) Number of natural children
b) Number of step or adopted children
c) Number of miscarriages
d) Number of stillbirths
e) Number of livebirths
f) Number of other (e.g., abortions)
g) Number of male children
h) Number of female children
i) Live birth weights
Risk ratios for the frequency of the above events will be
calculated against set "A" and "B".
All information "provided for these questions will be added to
the data file, including validation—confirmation or denial.
-------
Q141a. A frequency distribution will be compiled for:
a) Number of participants answering Yes
b) Percent of deaths which are male
c) Percent- of deaths which are female
d) Average number of deaths for those who answer."Yes"
e) Duration of life
f) Causes of death
Risk ratios for those participants who answered "Yes" to this
question will be compared against those who answered "No" to.
this question for set "A" and "B".
All information provided for these questions will be added to
the data file, including validation—confirmation or denial.
.Q142. A frequency distribution will be compiled for:
a) N-umber of participants answering Yes
b) Number of defects listed among those who. answer Yes
c) Type of defect in self
d) Type of' defect in sons, and percent living.
e) Type of' de'feot in .daughters., and percent living
Risk ratios for the frequency of the above events will be
calculated against sets "A" and "B".
Q143. A frequency distribution will be calculated for the number of
women using birth control methods.
Q144. The percent of women using each method will be determined.
Q145. A frequency distribution'of age at initiation of birth, control
use. will be calculated.
Q146. The frequency distribution,of age at menarche will be calculated.
Q14-7.. A frequency distribution will be calculated:
a) For the entire cohort
b). By stratification for Q19, Q22, Q25, and Q26
c) By stratification for set A and B questions if sufficient
numbers of participants respond to this question.
Q148. A percentage rating of natural versus surgical cessation of
menstruation will be calculated. A frequency distribution will
be calculated from validation reports regarding types of surgery.
Q149. A frequency distribution will be calculated:
a) For the entire cohort
b) By stratification for Q19, Q22, Q25, and Q26
c) 3y stratification for set A and B questions if sufficient
numbers of participants respond to this question.
Q150. A frequency distribution will be calculated:
a) For the entire cohort
b) By stratification for Q19, Q22, Q25, and Q26
c) By stratification for set A and B questions if sufficient
numbers of participants respond to this question.
-------
Q151. A frequency distribution will be calculated:
a) For the entire cohort
b) By stratification for Q19, Q22, Q25, and Q26
c) By stratification for set A and B questions if sufficient
numbers of participants respond to this question
Q152. A frequency distribution will be calculated:
a) For the entire cohort
b) By stratification for Q19, Q22, Q25, and Q26
c) By stratification for set A and B questions if sufficient
numbers of participants respond to this question
Q153. A frequency count will be computed.-
Q154. A frequency count will be computed.
-------
APPENDIX VI
Progress Reports
1. October 16, 1978 - June 15, 1979
2. April 4, 1980 (Preliminary Report)
-------
Assessment of Potential Health Risks Associated with Organic Contaminants
Feasibility Study - Great Lakes Basin
Progress Report
October 16, 1978 - June 15, 1979
Leonard M. Schuman, M.D.
Conrad P. Straub, Ph.D.
Principal Investigators
1. Introduction
In this progress report we summarize the activities undertaken by the
Divisions of Epidemiology and Environmental Health, School of Public Health,
University of Minnesota, Minneapolis, Minnesota, under Contract/Grant No.
EPA/R806282-01-0, during the period October 16, 1978 through June 15, 1979.
During this period 12 students were employed to gather and assist in
the evaluation of information pertinent to the study. These included five .
graduate students from the Division of Epidemiology under a young physician
acting as project coordinator, and seven graduate students from the Division
of Environmental Health with a scientist acting as project coordinator.
Throughout the study period, close liaison has been maintained between
the two study groups with frequent meetings between the two project coordinators,
between the two project coordinators and the principal investigators, and
between the principal investigators and the total staff involved in the study.
This progress report consists of the following sections: 1) a report
summarizing the findings of the epidemiological study group, 2) a summary report
of the findings of the environmental study group, and 3) a series of questions
which will be the basis for the continuation of the present study during
year two.
References have not been included with this summary statement, but will
be properly documented in preparation of the overall project report.
-------
-2-
Specific directions to be taken during year two will also be defined.
2. Epidemiologic.Studies
A literature review was undertaken to identify epidemiologic studies
of the ..effects of organic pollutants on human health. The two clases of
organic, pollutants of special interest were halogenated and polycyclic
aromatic hydrocarbons. Studies of primary interest are those which associate
chronic disease effects (morbidity and mortality).with exposure to organic
pollutants.and include mutagenic, teratogenic, and/or carcinogenic effects.
Other than, the reported Yusho incident involving exposure to polychlorinated-
biphenyls (PGBs) in.-Japan, and the Michigan and Indiana studies, few
.epidemlologic studies associating PCB exposure to'human disease have been
reported. To broaden the approach of the possible effects of PCB and other
organic pollutants on human health, reported results of animal.studies were
evaluated to provide clues for hypothesis building of effects on man. Much
support could be found in the literature on the mutagenic, carcinogenic,
and teratogenic effects of organic pollutants on animals. One obvious
problem is 'the translation of these results to human exposure experience.
With the virtual nonexistence of human epidemiologic studies in this
area, our epidemiology group explored sources which could provide basic
data .on general and cause specific mortality; infant, neonatal, and perinatal
mortality; fertility; congenital malformation, etc., for the development of
needed studies in this field.
State health departments contacted were Minnesota, Wisconsin,
Michigan, Ohio, Indiana, and Illinois. From these states, rates were obtained
for infant, neonatal, and fetal deaths. These rates were inconclusive
because of lack of adjustment for maternal' age. From some states, congenital
anomaly rates for live births were obtained but not for all years of interest.
It must be kept in mind that such rates are grossly deficient because of under-
reporting, even of those anomalies detectable at birth, and because of the
-------
-3-
difficulty of ascertaining many anomalies at birth. Site specific cancer
rates by county were available from Michigan and Wisconsin; other states
reported only cancer deaths or reported deaths collectively by system, i.e.,
respiratory system, gastrointestinal system, urinary system, etc. Site
specific rates could probably be accessed by a review of death certificates.
This is planned.
The National Center for Health Statistics lists cancer deaths by
system but not by site. Only raw numbers were available. Rates were
calculated using populations denominators from the Census Bureau. There
were no age, race, or sex breakdowns.
County rates were desired to compare rates for lake-bordering as
opposed to non-lake bordering counties. Use of date currently available
for this comparison presents several problems including lack of adjustment
for urban/rural differences, different sources of water supply, differences
in length of residence, dietary habits, etc.
From the National Cancer Institute, twenty year (1950-1969) summaries
were obtained-of site-specific cancer mortality by county. These data were
age-adjusted using as a standard the entire 1960 U.S. population. The data
also identified white and non-white rates. Because the specific year data
were not available, it is not possible to examine trends of mortality over
time. For such data, we are planning to contact the National Cancer
Institute and the National Center for Health Statistics to ascertain sex,
age, and race-specific data by county and year, and the cost thereof.
A preliminary examination of the twenty-year data from the National
Cancer Institute indicates a possible excess of stomach, esophageal, and
other gastro-intestinal cancers in the lake-bordering counties as compared.
to non-lake bordering counties. However, because of urban/rural differences
-------
-4-
mentioned before, and other confounding factors, additional data are needed
before conclusions may be drawn.
3. Environmental'Health Aspects
Much of the available literature was reviewed, particularly that
concerned with polychlorinated biphenyl (PCB) concentrations in. air,
in water and wastewater, in sediments and soils, and in the aquatic
environment; with PBC intake by fresh water and marine organisms including
fish; with the degradation of PCBs in the various environments; and vith
the effects of exposure on animals and humans. A general summary of this
literature review follows.
PCS is a ubiquitous contaminant and is encountered in all environmental
media. Some of the isozners have been identified in animal and human tissues.
The-mechanisms of transport have been indicated and do vary with the region
studied. In the Great Lakes area, transport by air (rainfall and/or
dry deposition) is the primary source of these materials, whereas in other
locations direct discharge from industrial Or other sources may be the
major contributor. Because these substances are generally insoluble in
water.they adsorb on to particulates and eventually deposit on'the bottom
of lakes or other bodies of water as sediment or deposit directly on the
soil. Generally, concentrations in water are very low, in.many instances
at or just above detectable levels. High concentrations are encountered in
sediments and extensive biomagnification occurs through the aquatic food
chain. Volatilization and biodegradation of the less chlorinated PCBs have
reduced these levels in the sediments and have resulted in higher concentrations
of these in the air and in air particulates. As a result, the more heavily
chlorinated compounds are retained in the sediments, taken up by micro-
organisms and retained in lipids for long periods of time. They are not
-------
-5—
readily degraded nor metabolized.
Differences in results have been reported by many investigators and
these differences will have to be reconciled. In addition, there are marked
differences in uptake among various species of organisms exposed, and there
are differences in response between so-called "pure" PCBs and those encountered
in the environment.
In recent years attempts have been made to reduce the discharge of PCBs
from controlled sources, but these procedures have not reduced levels in the
Great Lakes or other bodies of water. These water bodies have become sinks for
the deposition of large amounts of these substances with their retention in
sediments. As a result, with minor degradation, they will serve as sources
for the long-term, continued contamination of organisms associated with these
environments. Attempts must be made to identify these sources, point or non-
point, and to control them at their sites of production. Control measures are
available for the destruction and degradation of some of these isomers. In
other instances, more degradable isomers can be substituted for those currently
used.
For the evaluation of possible health effects of exposure to PCB, we have
the results of the Japanese Yusho exposure incident studies, and information
on the effects of industrial exposure to PCBs during production and use to
draw upon.
Questions have been raised as to whether the effects observed are due to
PCB itself or to the presence of certain impurities associated with these
materials. One impurity identified in many of the PCB isomers in commercial
4 6
use is polychlorodibenzofuran, which is reported to be 10 to 10 times more
toxic than PCB. The levels of these impurities must .be determined,
particularly since the major, route of exposure, other than occupational
exposure, appears to be related to the consumption of fish taken from the
-------
-6-
Great Lakes or other containingted bodies of water. Of significance, is whether
ingestion of the mixtures of PCS isomers present in fish tissue can induce
health effects directly, as a result'of synergistic responses due to mixtures
of isomers, or related to the presence of impurities, such as the chlorinated
dibenzofurans., or other degradation products.
4. Questions Identified as a Result of Our Studies
From the studies carried out, numerous questions have arisen, some of which
we hope to address during the coming year. These questions are identified
below. In the epidemiology area the questions include:
1. Methods of .accessing data not directly available from existing sources:
a. county age-, sex-, and race- adjusted site-specific cancer
death rates
b. county maternal age-, race— adjusted infant, neonatal, and
fetal death rates
c. county malformation rates
d. county-fertility rates
2. Methodological problems of ascertaining trends in:
a. cancer mortality
b. infant, neonatal, and fetal mortality
c. malformation morbidity
d. fertility morbidity
3. Ascertainment of the appropriateness of existing methodology for
teratogenic, mutagenic, and carcinogenic relationships to our study
4. Proper selection of high and low exposure contrast communities
for retrospective studies
5. Possibility of doing rigourous, single community prospective studies
6. Appropriateness and feasibility of using bacterial and/or animal
models to ascertain carcinogenic, mutagenic, and teratogenic effects
-------
- 7 -
of the organic pollutants in lake water, j.n food chains, and in
air at specific sites, i.e., in an individual community
In the environmental health area the questions that have arisen include:
1. Significance of the polychlorodibenzofuran impurities vis a vis
effects currently attributable to various PCB isomers
2. Levels of exposure to occupational workers having contact with
PCBs, its various isomers, or its derivatives, and health effects
attributable to these exposures
3. Relevance of occupational exposures of the workers to exposure
of his family
4. Determination of the sources responsible for total exposure to
populations from environmental sources, their identity, and
significance as exposure pathways
5. Identification of specific methodologies that can be used to reduce
and/or control of the levels of PCB released to the environment
6. Evaluation of the methodologies used to collect samples, their
effectiveness for providing information on actual environmental
levels, and possible development of standardized procedures
7. Crititcal examination of reported data from a statistical point
of view to assess the value of the findings reported
8. Determination of whether chronic toxicity is related to the
metabolism of PCB and its intermediates or to the highly
chlorinated stored PCBs
9. Study of the individual congeners, both those metabolized and
those stored by man, is urgent because of their demonstrated
carcinogenic potential
10. Evaluation of the long-term effect of release of PCBs from stored
bottom sediment on the biota of these aquatic systems
-------
- 8 -
It is hoped that some of these questions can be addressed during year
two of the project to identify methodologies -that will permit, if possible,
a feasibility study, to relate environmental exposure levels to human effects.
-------
Preliminary Report of
Epidemiological and Environmental Data
Contract //EPA/R806282
Assessment of Potential Health Risks Associated with
Organic Contaminants in the Great Lakes Basin
Feasibility Study
University of Minnesota
Division of Epidemiology
Division of Environmental Health
April 4, 1980
-------
-1-
1. Surgiary
Several geographic areas in the Great Lakes Basin were selected
and matched to existing vital statistics data. These areas were
identified by reported poiat-source discharge violations of individual
state effluent requirements by industries and municipal wastewater treat-
ment.plants. Point sources are associated with local areas of environmental
contamination; however, the degree of environmental contamination (as
measured by water pollution parameters) cannot predict the rates and routes
of human exposure and ultimate health effects. Furthermore, current
environmental data may not be completely indicative of past environmental
conditions in a local area. However, current data may be viewed as a rough
approximation of past' conditions in situations where persistent contaminants
have been discharged to the environment (PCBs and like organic compounds).
Hence, the use of current environmental data (1970 - 1979) to characterize
the previous 20 year period must be looked upon with great caution,
particularly when comparing it with 20 year (1950-1969) cancer mortality
rates,
The use of point source dischargers of potentially toxic -materials
as indicators of environmental contamination cannot entirely account for
differences in the cancer, fetal, neonatal, and infant mortality rates
observed between lake and non-lake counties. These differences may be
associated with differences in the smoking, drinking, and dietary habits
of the populations being studied, their ethnic and socio-economic status,
and the degree of industrialization of the geographic areas in which
they live and/or work. For example, differences in stomach cancer mortality
rates may be associated with certain ethnic groups rather than environ-
mental pollutants. Similarly, lung cancer differences may be far more
-------
-2-
associated with cigarette consumption patterns than with exposure to
environmental pollutants. Furthermore, the state vital statistics data
are not adjusted for age, sex, and race differences in each county
within each state. Therefore, some observed differences between lake
and non-lake counties may be attributed to differences in the age, sex,
or race composition of the counties.
It is apparent that the inadequacies of a county's epidemiological
and environmental data have made evaluation of the association between
riortality rates and environmental contaminant levels difficult. However,
valuable experience has been gained during t.his initial accumulation
and evaluation of existing data, particularly in identifying•specific
areas to be addressed in future efforts and in the need to assess con-
founding and interacting risk factors.
-------
II. Morbidity/Mortal!try Data Collection and Analysis
An important initial step in determining-the potential health
effects of certain ubiquitous organic pollutants in the Great Lakes
Basin was to investigate the availability of existing morbidity and
mortality data from various data sources. It was apparent from the
literature that PCBs (and like organic compounds) are potential
carcinogenic and teratogenic agents. Thus, the investigation focused
on site-specific cancer mortality rates, fetal, neonatal, and infant
death rates, and congenital anomaly rates (which can be expressed as
either death rates or as a percentage of live births).
Since these types of organic pollutants are quite persistent
in the environment and have been used.for over thirty years by various
industries, data back to 1950 were sought. However, the voluminous
amount of data involved over this thirty year period for eight states
(Minnesota, Wisconsin, Illinois, Indiana, Michigan, Ohio, Pennsylvania,
and New York) warranted a limited examination of these data. Specifically,
state vital statistics were thoroughly examined for Minnesota, Wisconsin,
Illinois, "Indiana, Michigan, and Ohio for every fifth year from 1950
plus the. most.current year (usually 1977).
The hypothesis was posed that areas adjacent to the Great Lakes have
a greater exposure potential to these pollutants than those areas more distant
from the Lakes based on the following assumptions:
1. Most lake bordering communities have increased industrial
activity compared to most non-lake bordering communities.
2. Individuals living in these adjacent communities are more
likely to be occupationally exposed to those pollutants being
discharged into the air and the Lakes.
-------
3. Close proximity to the Lakes would lead to greater exposure
via ambient air, soils, and water.
A. Fishermen living in these lake-adjacent communities are
more likely to fish these lakes and may have higher
exposures due to consumption of their catch.
Based on this hypothesis, it was decided to examine (for camparative
purposes) the morbidity/mortality data for lake and non-lake communities.
Since the smallest geographical area for which data are readily
available in the country, the investigations centered on gathering
data by county and comparing lake counties to non-lake counties
within each state.
The first source investigated was the vital statistics record for
each individual state. Rates by county for site-specific cancer
mortality; fetal, neonatal, and infant mortality; and congenital
anomalies were requested from each State vital statistics office for
the following years: 1950, 1955, 1960, 1965, 1970, 1975, and 1977.
The data available are indicated below:
1. All states have data available by county on fetal, neonatal,
and infant death rates. However, in most cases, these are
not race or sex specific.
2. All states have rates for deaths due to congenital anomalies by
county but these are not age, sex, or race specific. Data on
congenital anomalies as a percent of live births by county
-------
—5—
are available from selected states for selected years. These data,
again, are -not sex or race specific.
3. Most states do not have site-specific cancer mortality rates by
county. For the states that do, the data are generally not age-
race- or sex-specific.
(For a.more detailed account of the data available from each state,
refer to the cover page which precedes each individual state's
vital statistics).
The yearly publications of the National Center for Health Statistics
(NCHS) were also examined for relevant information. These data are
comparable to the data obtained from each individual state. The NCHS
does publish fetal, neonatal, and infant death rates by county but they
are not race-or sex-specific. They also publish death rates for congenital
anomalies by county-. Site-specific cancer mortality data are not published
by county by the NCHS.
The best source of cancer mortality data by county was found in the
National Cancer Institute publication entitled: "U.S.. Cancer Mortality:
1950 - 1969*'. This publication contains 20-year summary rates of site-
specific cancer mortality by county. These rates are sex-and race-specific
and age-adjusted. The drawbacks of these data are: 1) since the rates
are a 20-year summary, time trends cannot be examined and 2) there are
no similar data that are more recent.
Lake bordering and non-lake bordering counties were compared using the
following information: 1) site-specific cancer mortality rates from the
NCI 20-year data and from state vital statistics and 2} fetal, neonatal,
and infant death ratios from state vital statistics data. The fertility
and congenital malformation rates were also compared when these types of
-------
-6-
data were available from the state's vital statistics.
Mean values for all cancer death rates and fetal, neonatal, and infant
death rates of lake-bordering counties were compared to the respective
mean values of non-lake bordering counties within each state. Those
parameters having higher mean values in lake-bordering counties than non-
lake bordering counties are noted for each state, (see cover sheet attached
to state data). No tests of statistical significance were calculated for
these observed differences.
Maps were prepared for each state depicting the counties with the highest
and lowest rates (within that particular state) for selected parameters.
The parameters used included the NCI 20-year cancer mortality data for
cancers of the stomach, lung, esophagus and all neoplasms. These particular
cancer sites were chosen because their mean mortality rates appeared to be
consistently higher for lake-bordering counties than for non-lake bordering
counties in all states. Additional parameters include the fetal and neonatal
death rates for the years 1970, 1975, 1977, percent of live births with
congenital anomalies for the years 1970, 1975, 1977, (only available for
Wisconsin, Michigan, and Minnesota) and fertility rates for the years 1960,
1970 (only available for Ohio, Wisconsin, and Minnesota). The means and
standard deviations were calculated for all available parameters within
each state. Those counties with rates significantly Qf 2S.D.)
different from the state mean are identified with an asterisk on the mans.
-------
—7—
III. Selection of Study Populations Based uponJSnyironmental Factors
Upon completion of an extensive scientific literature review of
PCBs (and like organic compounds), it was apparent that these substances
are ubiquitous environmental contaminants. Much of the data, however,
clearly demonstrate that there are regional differences and that the
effects are consistent across all media (e.g., water, sediments, fish,
and birds), generally showing greater concentrations in highly developed
areas and areas of industrial activity. Thus, it was not surprising
that the Great Lakes Basin area (the most industrialized area in the
United States), based on the literature review, appeared to be experiencing
a higher degree of environmental contamination with these potentially
toxic materials as compared to other regions of the country.
To elaborate further on the degree of environmental contamination
the Great Lakes Basin has sustained from PCBs (and like organic compounds),
it was quite obvious that our data base had to be expanded to include
specific quantitative and qualitative data and information of specific
areas within the Basin. Thus, various state, federal, and international
environmental agencies (with some type of jurisdiction in the Basin) were
consulted concerning the existence and availability of specific
environmental data. Each agency was requested to provide qualitative and
quantitative data on organic contaminants which the agency -had identified
in point sources (waste water discharges), atmospheric sources, waste-
water sludges, runoff, sediments, ambient water, drinking water, landfill
leachates and vent gases, and fish caught within the Basin.
The agencies and organizations contacted responded favorably to
the request but little meaningful data were actually received. Generally
-------
-8-
it was apparent that considerable environmental data do exist but are
not readily available for review or compilation. (Most states store such
data in vertical files.) However, a series of documents were received
from the International Joint Commission (IJC) which were compilations of
data solicited from various Canadian and American jurisdictions within the
Great Lakes Basin (1,2,3,4). It should be noted that the IJC documents
include data that some state officials have identified as also being
available from their agencies; therefore, there may be duplications of
data when multiple data sources are used.
The IJC documents were thoroughly reviewed and it was decided that the
data contained therein could be used to select (on the basis of various
criteria) specific areas or communities within the Basin that may be
experiencing a high degree of PCS (or like organic compound) contamination.
It was then assumed that people living in these communities could have
potentially higher degrees of exposure to these contaminants in their
immediate environment. Thus, this attempt permitted an initial approach
at integrating environmental data which identified groups 'receiving potentially
higher exposures with morbidity/mortality data. A number of approaches were
taken and will be discussed.
The focus of this investigation is to identify study populations which can
be used to evaluate the potential health risks associated with organic
contaminants in the environment of the Great Lakes Basin. Using data fron
the International Joint Commission Great Lakes Water Quality Board concerning
designated "problem areas", high PCB concentrations in Great Lakes sediments,
and point source dischargers, several communities were selected for further
environmental and epidemiological characterization.
Table 1 lists "problem areas" based on an evaluation of the data
-------
-9-
contained in the "Great Lakes Water Quality 1978 Report". These "problem
areas" were selected because of .reported violations of water quality
standard with respect .to organics (i.e., .phenols, chlorobenzenes) and/or.
PC3s. Point sources were implicated as the cause of these problem areas
but were not recorded in this table.
Other communities were selected as potential sites' for.environmental
and epidemiological evaluation using existing sediment data from the "Status
Report on Organic and Heavy Metal Contaminants in the Lakes Erie, Michigan,
(2 3)
Huron, and Superior Basins". A ' 'Table 2 lists committees adjacent to the
Great--Lakes where PCB bottom sediment concentrations' ^mg/kg have been
Identified. These high levels are thought to be associated with point
source dischargers but were not so identified in this report.
-------
-10-
TABLE 1
Followiag an examination of the data obtained in the "Great Lakes
Water Quality 1978 Annual Report" the following geographical areas were
identified as possible sites for future epidemiologlcal and environmental
evaluation. These identifications were based upon violations of objectives
or standards with respect to organics (i.e., Phenol) and/or PCBs. Both
industries and wastewater treatment plants are identified as being
potential sources for these discharged substances. The following table lists
the problem areas and the associated lake.
LAKE PROBLEM AREA
Superior Thunder Bay, Jackfish Bay **
Huron St. Marys River, Spanish River **
Michigan Waukegan Harbor, Indiana Harbor
Erie and St. Clair St. Glair River, Detroit River, Cleveland
area,* Black River, Rouge River, and
Ecorse River
Ontario Buffalo River, Upper Niagra River, Lower
Niagara River, Mississauga - Clarkson area, **
Grass River
*Areas where water quality objectives have not been achieved because remedial
programs are not yet completed.
**Canadian jurisdiction
-------
-11-
Table.2
List .of. communities adjacent to the Great Lakes where PCB* concentrations
equal to or greater than 1 mg/kg have been identified in the bottom
sediments.
Lake Basin
Location
PCB, mg/kg
Level
"Michigan
Erie
Huron
Superior
Ontario
Waukegan, IL
Indiana Harbor, IN
Fox River, WI
Escanaba, MI
Manistique, KE
Milwaukee, WT
Cuyahoga River, OH
Ashtabula, OH
Fairport, OH
Cleveland, OH
No data presented
above 1 mg/kg.
No data presented
above 1 mg/kg.
Hamilton, ONT.
United States locations
were below 1 mg/kg
0.1 to 16,400
.04 to 25.7
0.67 to 11.56
1.6.
10.2, 3.2, 25.5H
6.4
0.29 to 2.20
<0.1 'to .1.10
0.7 to 1.10.
<0.01 to 2.30
1.3 to 10
+ Combined Aroclor 1254 + 1242
* Total PCB
-------
-12-
Point source dischargers can account for higher local levels of
contaminants as compared to those levels found in the surrounding environment.
These sources may influence the exposure potential and effect the health
status of surrounding communities. Point sources include combined storm
sewers, wastewater treatment plants, industries, and power plants. The
data contained in an "Inventory of Major Municipal and Industrial Point
Source Dischargers in the Great Lakes Basin" were used to identify sources
(4)
of potentially hazardous organic compounds in the Great Lakes Basin, and
select and compare several geographic areas or possible sites for a
cooperative epidemlologic and environmental study. The areas selected are
identified based on the following criteria:
1. communities in which at least one industry measured organics (i.e.,
phenols) prior to discharge and complied with state- effluent
requirements, and in which the wastewater .treatment plants complied
with state effluent requirements [on the basis of phosphorus (P),
biological oxygen demand (BOD), and suspended solids (SS)]. (Tbl 3)
2. communities in which at least one industry measured organics (i.e.,
phenols) prior to discharge and failed to comply with state effluent
requirements, and in which a wastewater treatment plant failed to
comply with state effluent requirements (on the basis of P, BOD,
and SS). Table 4.
3. communities in which at least one industry measured organics (i.e.,
phenols) prior to discharge and failed to comply with state effluent
requirements, and in which the wastewater treatment plants complied
with state effluent requirements (on the basis of P, BOD, and SS).
Table 5.
-------
-13-
4. communities in which at least one industry measured organics
(i.e., phenols) prior to discharge and complied with, state effluent
requirements, and in which a wastewater treatment plant failed to
comply with state effluent requirements (on the basis of P, BOD,
and SS). Table 6.
-------
-14-
TABLE 3
Communities in which at least one industry measured organics (i.e.., phenols)
prior to discharge and complied with state effluent requirements, and in which
the wastewater treatment plants complied with state effluent requirements
(on the basis of P, BOD, and/or SS).
Allen
Porter
Gratiot
Saginaw
St. Clair
Erie
Niagara
S t. Lawrenc
S t. Lawrenc:
City
Fort Wayne, Ind.
Chesterton, Ind.
Alma, Mich.
Saginaw, Mich.
Port Huron, Mich.
Lackawanna, NY
N. Tonawanda, NY
Messena, NY
Ogdensburg, NY
Cancer Mortalitv Rates
high esophagus (6.7), stonach
(18.8), lung (47.8) all neo-
plasms (207.0)
high esophagus (5.5)
* Identified as being high from preliminary reported information for a
a particular county in a specific state.
Qlndicates mortality rates (deaths/100,000) within a specific county
for respective states.
-------
-15-
TABLE 4
Communities in which at least one industry measured organics (i.e. phenols,)
prior to discharge and failed to comply with state effluent requirements,
arid in which a wastewater treatment failed to comply with state effluent
requirements (on the basis of P, BOD, and/or SS).
City Cancer Mortality Rates
E. Chicago, Ind. high esophagus (5.4), stomach
(21.9), lung (47.9), all
neoplasms (198.4)
Kalamazoo Kalamazoo, Mich.
Ottawa Holland, Mich.
Wayne Wyandotte, Mich. high esophagus (6.4), lung
(47.3), all neoplasms (209.2)
Allen Lima, OH
Cuyahoga Cleveland, OH high esophagus (7.5), stomach
(20.7) lung (45.8), all
neoplasms (211.9)
Lorain. Lorain, OH high esophagus (5.6), stomach
(20.2), all neoplasms (189.7)
*See Table 1
-------
-16-
TABLE 5
Communities in which at least one industry measured organics
(i.e., phenols) prior to discharge and failed to comply with State
effluent requirements and in which the wastewater treatment plants
complied with state effluent requirements (on the basis of P, BOD,
and/or SS).
County City Cancer Mortality Rate*
Niagara Lockport, NY high esophagus (5.5)
*See Table 1
-------
TABLE 6
Cocsaunities. in whicn at least one industry measured organics
(i.e.;.phenols) prior to discharge and complied with State effluent re-
quirements, and in which a wastewater treatment plant failed to comply-
with State, effluent requirements (on the basis of P, BOD and/or SS)..
lake
Wayne.
Erie
Monroe
Niagara
Cuyahoga
Lucas
*See Table 1
**Areas where
City
E. Chicago, IL
Gary, IN
Alpena
Bay
Gene ss e
Ottawa
Wayne
Alpena, MI
Bay City, MI
Flint, MI
Holland, MI
Detroit, MI
Trenton, MI**
Tonawanda, NY**
Rochester, NY
Niagara Falls, NY
Cleveland, OH**
Toledo, OH
Cancer Mortality Rate*
high esophagus (5.A), stomach
(21.9) lung (47.9), all
neoplasms (198.A)
high esophagus (5.4), stomach
(21.9), lung (47.9) all
neoplasms (198.4).
high lung (43.9), all neoplasms
(196.4)
high esophagus (6.4), lung (47.3),
all neoplasms (209.2)
high esophagus (6.4), lung (47.3),
all neoplasms (209.2)
high esophagus (6.7), stomach
(18.8), lung (47.9), all neoplasms
(207.0)
high esophagus (5.5)
high stomach (21.7)
high esophagus (7.5), lung (45.8)
stomach (20.7), all neoplasms (211.9)
high esophagus (5.7), lung (45.6),
all neoplasms (196.4)
there are 2 or more WTPS, some in compliance others not.
-------
-18-
Following preparation of these city lists, the state maps
indicating high and low county cancer mortality rates were used to
determine if these cities were located in counties with unusually
high cancer mortality rates. Blanks for cancer mortality rates do
not imply an absence of cancer, but that the rate for the county
containing the specific community was not among the five highest in
the state for any of the cancer sites of esophagus, stomach, and
lung and/or all neoplasms. These results indicate a possible associ-
ation between point sources of hazardous materials, their compliance
status with respect to state effluent requirements, and cancer mortality
rates.
It was hypothesized that those counties with point-source dis-
chargers of hazardous materials in- compliance with state effluent re-
quirements may have different cancer rates than those counties with
point source dischargers of hazardous materials failing to comply
with state effluent requirement. A two tailed T-test was used to
evaluate the level of significance for differences observed between
cancer mortality rates in counties containing cities listed in Table 3
(i.e., communities in which at least one industry measured organics
prior to discharge and complied with state effluent requirements and
in which the wastewater treatment plants complied with state effluent
requirements) and counties containing cities listed in Table 4 (i.e.,
communities, in which at least one industry measured organic prior to
discharge and failed to comply with state effluent requirements and
in which a wastewater treatment plant failed to comply with state
effluent requirements).
The mean cancer rates fox counties listed in Table 3 (total com-
pliance) were compared to the mean cancer rates for counties listed
-------
-19-
in-Table 4 (total non-compliance). The cancer mortality rates for
counties containing the cities listed in Table 4 were significantly
greater than the cancer rates in counties containing cities listed
in Table 3 at the p•» 0.05 level for the following sites:
white males stomach (0.012)*
kidney (0.029)
non-white males nose, auxiliary sinuses, etc.
(0.011)
and at the p.« 0*10 level for the following sites:
white males nasopharynx (0.080)
pancreas (0.075)
all other (0.092)
white females stomach (0.097)
pancreas (0.098)
non-white males stomach (0.100)
However, the cancer mortality rates for counties containing the
cities listed in Table 3 (total compliance) were significantly greater
than the cancer rates in counties containing cities listed in Table 4
(total non-compliance) at the p = 0.05 level for the following sites:
white males nose, auziliary sinuses, etc.
(0.041)
white females nose, auziliary sinuses, etc.
(0.032)
*() Indicates p value
-------
-20-
non-white males salivary (0.049)
malanoma (0.032)
lymphoma, etc. (0.030)
all other (0.013)
and at the p * 0.10 level for the following sites:
non-white males biliary and liver (0.059)
non-white females salivary (0.099)
biliary and liver (0.099)
other skin (0.059)
-------
-21-
Bibliography
1. "Great Lakes Water Quality 1978 Annual Report." International
Joint Conmission (IJC), Great Lakes Water Quality Board, July 1979.
2, "Great Lakes Water Quali.ty - Appendix E: Status Report on Organic
and Heavy Metal Contaminants in the Lakes. Erie, Michigan, Huron,
and Superior Basins." IJC, Great Lakes Water Quality Board,
July 1973.
3* "Great Lakes Water Quality - Appendix E: Status Report on the
Persistent Toxic Pollutants in the Lake Ontario Basin." IJC,
Great Lakes Water Quality Board, December 1976.
4. "Inventory of Major Municipal and Industrial Point Source. Dis-
chargers in the Great Lakes Basin." IJC, Great Lakes Water Quality
Board, July 1978.
-------
-22-
IV. Intense Environmental Characterization: Pilot Study to Determine
the Availability of Existing Environmental Data
It was apparent that the preliminary selection of potential
study counties was based on very broad and non-specific data.
However, these counties were selected on the premise that detailed
quantitative and qualitative data on specific contaminant concen-
trations in various environmental strata could be obtained and
evaluated. Thus, a complete environmental characterization
could be developed for each individual lake county, with the
intent of ultimately determining morbidity and mortality rates
and potential routes of exposure of the indigenous populations
to specific contaminants. A similar approach could be taken
•for counties chosen as controls (i.e., non-lake counties).
In order to accomplish these objectives, a pilot study to
determine the availability of existing data and to evaluate the
usefulness of these data in assessing possible exposure levels
was established. Two counties were selected within the same
state to determine the kinds of data available locally. Additional
data were obtained by contacting State and Federal agencies who
were responsible for environmental monitoring in these counties.
These data would provide insight on some aspects of the prevailing
environmental conditions. Furthermore, deficiencies in the
available data could be identified and the need for additional
monitoring could be evaluated.
The state environmental agency was contacted first, to request
access to their file's, which contained environmental information
for these counties. The following were requested: 1) surface
-------
-23-
water dischargers (wastewater): NPDES permits, effluent and
sludge analysis, for specific potentially toxic materials (PCBs),
sludge handling and disposal practices, and sludge disposal sites-
2) public drinking water supplies: source (ground or surface),
treatment provided, general water quality parameters, and specific
analyses for potentially toxic materials (PCBs, chlorinated
pesticides) of both the finished and raw water; 3) sanitary
landfills: location, leachate analysis, types of materials
disposed, and vent gas analysis; 4) ambient air: location of
monitoring stations, specific parameters, and location of in-
cinerators;. 5) special environmental monitoring surveys for
PCBs, pesticides, etc.: lake and river sediments, ambient water,
fish flesh, and soil. Other agencies in the state were also
contacted and relevant data were requested. These include:
flesh analysis of fish caught within the waters of each county,
and market-basket analysis of food products consumed.
The USEPA, Revion V office was contacted for data pertaining
to environmental concentrations of organic contaminants in these
counties available through their computer information service
(STORED.
Studies conducted by NIOSH evaluating the health hazards
associated with PCB usage in industry were requested to evaluate
potential worker exposure in similar industries located in the
study counties.
The following summary characterizes the types of data col-
lected from these data sources.
1. State environmental agency data and information.
-------
-24-
a. Agenny personnel have identified as suspect seven land-
fills in one county and two in the other .as receiving
PCBs (wastewater treatment sludges and industrial oils
and solvents). The monitoring of leachates and aquifers
beneath landfill sites will be initiated in 1980; very
few specific results are available at this time.
b. 27 wastewater dischargers (i.e., industrial and municipal
wastewater treatment plants - WTPs) in one county and
four in the other were identified as potential sources
of PCBs or other similar synthetic organic compounds.
Two large WTPs in one county reported detectable
levels of specific synthetic organic compounds (PCB
and DDT) and a phthalate ester in their effluents.
c. Public water supply information.indicated that six areas
in one county obtain drinking water from surface supplies.
The remainder of the county population obtains drinking
water from groundwater sources. In the other county
drinking water is obtained exclusively from ground sources.
Pesticide levels in surface water supplies are
available for one county. These date are pursuant to
the provisions of the safe drinking water act. However,
they are reported as meeting the requirements rather
than reporting the actual concentrations found. No
PCB data exist for groundwater supplies.
d. Surface water quality data are available for one county
and included measurements for dieldrin, DDT (total),
and PCBs in three locations (two rivers and an adjacent
-------
-25-
lake). Surface water quality data for pesticides or
PCBs are not available for the other county
e. Two environmental surveys have been carried out in one
of the counties. These included analyses of fish,
ambient water, and sediments for PCBs and other synthetic
organic compounds. No such surveys were performed in
the other county.
f. Fish analyses in the one county included PCB, and
chlorinated pesticides concentrations in lake trout and
chinook salmon. No data were reported for the other
county on PCBs or ai\y other contaminants.
2. USEPA "STOKET" computer information
a. Surface water quality data are available for both counties;
these data do not include measurements for PCBs, pesticides
or other organics. However, phenolic and oil and grease
concentrations are measured at several monitoring
stations in both counties.
b. Data on analysis of sediments are sparse. There are only
two monitoring stations in one county, and four in the
other. Results are available from analysis of only one
sample collected at each monitoring station.
c. Analysis of fish collected in both counties were available.
Data are not, however, consistent throughout the computer
print-out. Critical variables not reported include:
species, age, sex, weight, time of year tested, portion
tested and fat content.
d. There is no information on PCB (or other synthetic
-------
-26-
organic compounds) concentrations in air, landfill
leachates or vent gases.
3. State Departments of Conservation and Public Health information.
a. Both agencies have generated copious quantities of data
pertaining to fish flesh analyses for PCBs, heavy metals,
and chlorinated pesticides. A large variability exists,
however, in the time of year samples, number per sample,
and type of sample (i.e., whole, fillet, etc.). For one
county data are available for yellow perch, lake trout,
coho salmon, and bloaters. In the other county, only
carp were surveyed consistently for PCBs and chlorinated
pesticides.
4. Local health departments were contacted but their data were
similar, if not identical, to data obtained from the state
agencies.
5. Research data from universities in the vicinity of the counties
were contacted. However, to date no data have been obtained
from these sources.
The location of specific data sources are plotted on large-
scale county maps. These are useful in locating dischargers of
various contaminants and their potential for the contamination of
fish, water, sediments, etc. and the location of possible sources
of exposure to local population groups. It should be pointed
out that these data, for the most part, covered concentrations
measured and found in the last five to ten years. However, since
these contaminants (PCBs and chlorinated organic pesticides)
are highly persistent, and have been in the environment at various
-------
-27-
levels since their industrial use, it may be possible to extrapo-
late these findings over a longer period of past exposure. Finally,
when specific study counties are eventually selected the experience
learned in this pilot study should provide a useful approach to
the characterization of study county environmenta.
------- |