University of Southern California Dissertations and Theses

Occupational exposure to extremely low frequency electromagnetic fields as a potential risk factor for Alzheimer's disease

(USC Thesis Other)

Occupational exposure to extremely low frequency electromagnetic fields as a potential risk factor for Alzheimer's disease

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content OCCUPATIONAL EXPOSURE TO EXTREMELY LOW FREQUENCY
ELECTROMAGNETIC FIELDS AS A POTENTIAL RISK
FACTOR FOR ALZHEIMER'S DISEASE
by
Meleana Elizabeth Dunn
A Thesis Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(BIOMETRY)
May 1995
Copyright 1995 Meleana Elizabeth Dunn
UNIVERSITY O F SOU TH ERN CALIFORNIA
TH E GRADUATE SCHOOL
UNIVERSITY PARK
LOS ANGELES. CALIFORNIA 9 0 0 0 7
This thesis, written by
Meleana Elizabeth Dunn
under the direction of h?. V . Thesis Comm ittee,
and approved by all its members, has been p re
sented to and accepted by the Dean of The
Graduate School, in partial fulfillment of the
requirements fo r the degree of
Master of Science
c. . ' • f W f f T i *
Dean
Tiatp February 23t 1995
.........
ii
To the
memory of my
grandfather,
Robert Francis Mclnerny
ACKNOWLEDGEMENTS
I would like to express my appreciation to Dr.
Eugene Sobel, PhD, my committee chair. I would also
like to thank my committee members: Dr. Wendy Mack,
PhD, for her invaluable suggestions and advice during
the drafting and refining of the manuscript, and Dr.
Victor Henderson, M.D., for sharing his clinical
expertise.
I am grateful to Dr. Stanley P. Azen, PhD, for his
guidance, encouragement, and support throughout my
graduate school experience.
I am indebted to the patients, and their families,
of Rancho Los Amigos Medical Center who participated in
this study.
iv
TABLE OF CONTENTS
Page
DEDICATION ii
ACKNOWLEDGEMENTS iii
LIST OF TABLES V
CHAPTER
I. Introduction 1
II. Methods 11
III. Results 18
IV. Discussion 39
REFERENCES 47
APPENDIX A 50
V
LIST OF TABLES
Table Title
1 Diagnoses of Subjects Included in
Control Group
2 Descriptive Statistics for AD Cases and
Controls
3 Descriptive Statistics for AD Cases and
Controls
4 Univariate Analysis of EMF and Other
Risk Factors in Alzheimer's Disease
Study
5 Univariate Analysis of EMF and Other
Risk Factors in Alzheimer's Disease
Study
6 Univariate Analysis of EMF and Other
Risk Factors in Alzheimer's Disease
Study Males Only
7 Univariate Analysis of EMF and Other
Risk Factors in Alzheimer's Disease
Study Males Only
8 Univariate Analysis of EMF and Other
Risk Factors in Alzheimer's Disease
Study Females Only
9 Univariate Analysis of EMF and Other
Risk Factors in Alzheimer's Disease
Study Females Only
10 Multivariate Analysis of EMF and Other
Risk Factors in Alzheimer's Disease
Study
11 Multivariate Analysis of EMF and Other
Risk Factors in Alzheimer's Disease
Study
12 Multivariate Analysis of EMF and Other
Risk Factors in Alzheimer's Disease
Study Males Only
Page
14
19
21
22
24
25
26
28
29
30
31
33
Multivariate Analysis of EMF and Other
Risk Factors in Alzheimer's Disease
Study Males Only
Multivariate Analysis of EMF and Other
Risk Factors in Alzheimer's Disease
Study Females Only
Multivariate Analysis of EMF and Other
Risk Factors in Alzheimer's Disease
Study Females Only
Final Multivariate Models of EMF and
Other Risk Factors in Alzheimer's
Disease Study
High/Medium EMF Exposure Occupations by
Sex
1
Chapter I
Introduction
The first description of what is now known as
Alzheimer's disease (AD) came from Alois Alzheimer in
1907. Alzheimer observed microscopic lesions in the
brain of a demented patient, including neuritic plaques
and a large number of neurofibrillary changes in the
nerve cell population. Since that time extensive
research has been done on this disease, but only a
modest understanding of the underlying structural basis
has been realized (Alafuzoff, 1992). This
neurodegenerative disorder affects an estimated 7% of
the elderly population in the United States (Edwards et
al., 1991). Health care costs for patients with
Alzheimer's disease in the United States is over $82
billion per year (Ernst and Hay, 1994, Koga, 1994).
Prevalence
The prevalence rate of AD in the Framingham study
was 22.9 per 1000 persons over 60; 11.7 per 1000 men
and 30.1 per 1000 women (Bachman et al., 1992). When
broken down into age subgroups, the prevalence rates
(per 1000) were the following: 5 for subjects ages 65-
74, 41 for subjects ages 75-84, and 131 for subjects
over 84 (Bachman et al., 1992). The East Boston study
2
found markedly higher rates (per 1000) for these age
subgroups: 30 for ages 65-74, 187 for ages 75-84, and
472 for ages over 84 (Bachman et al., 1992). The
differences in prevalence between these two studies are
probably due to methodological differences (Bachman et
al., 1992). Primarily, the Framingham study only
classified patients with moderate or severe dementia as
demented. In contrast, the East Boston study
classified all levels of dementia, including mild
cases, as demented (Bachman et al., 1992).
Diagnostic Criteria
According to the National Institute of
Neurological and Communicative Disorders and Stroke
(NINCDS) and the Alzheimer's Disease and Related
Disorders Association (ADRDA) Work Group's report, the
criteria for the clinical diagnosis of Probable
Alzheimer's disease include:
3
1) Dementia established by clinical
examination and documented by the Mini-
Mental Test, Blessed Dementia Scale, or
some similar examination, and confirmed
by neuropsychological tests AND
2) Deficits in two or more areas of
cognition AND
3) Progressive worsening of memory and
other cognitive functions AND
4) No disturbance of consciousness AND
5) Onset between ages 40 and 90, most
often after age 65 AND
6) Absence of systemic disorders or
other brain diseases that in and of
themselves could account for the
progressive deficits in memory and
cognition. (McKhann et al., 1984)
Histopathologic confirmation is reguired to make a
diagnosis of definite Alzheimer's disease (McKhann et
al., 1984, Friedland, 1993).
The histopathological markers for AD include
neurofibrillary tangles (NFTs) and neuritic plaques
surrounding a central amyloid core (Alafuzoff, 1992).
The one neuropathologic abnormality required for the
diagnosis of definite AD is a sufficient number of
plaques (Khachaturian, 1985). Depending on the age of
the patient, there are minimum numbers of plaques, in 1
square mm of brain tissue, required for a diagnosis of
AD (Khachaturian, 1985). In their absence, the
diagnosis of definite AD is usually not made even in
4
the presence of atrophy and a large number of NFTs
(Blass, 1993). NFTs are one of a number of
cytoskeletal abnormalities characteristic of AD. They
are not, however, specific for AD and are found in
other degenerative brain disorders.
Clinical Features of AD
There are several clinical features of AD. The
most conspicuous early symptom of the disease is the
gradual onset of progressive memory loss.
Characteristic of the early stages of disease is the
relative conservation of motor and sensory abilities,
with memory impairment and orientation functions being
isolated difficulties. Even in later stages, social
skills may be remarkably preserved (Friedland, 1993).
These behavioral and cognitive features, as well as
extrapyramidal dysfunctions, can vary greatly in
different patients that are in a similar stage of
disease. Differences have also been noted between
early versus late age at onset of disease. Some
studies suggest that early-onset patients (i.e., onset
before age 65 years) may have a more profound language
disorder, a more rapid disease course, and a more
frequent positive family history for the disease
(Friedland, 1993).
5
In the later stages of the disease, all
intellectual capacities are lost. Movement becomes
difficult, and urinary and fecal incontinence develop.
Most patients die of pneumonia or other incidental
illnesses. Behavioral "noncognitive" features of AD
are common and include depression, personality changes,
apathy, agitation, irritability, hallucinations, and
delusions. Delusions and psychosis are often observed,
usually involving fears of persecution (Friedland,
1993). Psychotic symptoms are seen in 30%-50% of
Alzheimer patients.
Risk Factors
Several potential risk factors for AD have been
identified. The most significant of these is age
(Blass, 1993). Numerous studies have found head trauma
to be a risk factor although this finding is by no
means universal (Blass, 1993, Van Duijn, Stijnen, and
Hofman, 1991). There is little evidence that an
environmental toxin contributes to the development of
AD. Aluminum was once considered as one such possible
toxin, but this research has been quite inconsistent
(Blass, 1993) .
Other distinctive characteristics (or risk
factors) of AD patients have been found, including a
lower educational level and a larger proportion of
6
females as compared to non-AD persons of a similar age.
In two recently completed population-based studies it
was found that lower educational attainment was
inversely related to the prevalence of clinically
diagnosed probable AD (Mortimer and Graves, 1993). The
odds ratios (ORs) for having AD with a low education
level (defined as fewer than 6 years in one study and
fewer than 7 in another), were 7.17 and 4.65
respectively (Mortimer and Graves, 1993).
Several epidemiologic studies have found a
female:male ratio of 2:1 in AD patients (Blass, 1993).
In one study, which used the Mini-Mental State
Examination to measure cognitive deficiencies,
significantly greater deficits were found for women
with AD than for men with AD when age, duration of
symptoms, education, and family history of dementia
were considered (Buckwalter, 1993). In another study,
a significant difference between men and women, in the
prevalence of probable AD, was found for individuals
aged 75 and older. The female:male prevalence ratio
was 2.8 for probable AD (Bachman et al., 1992). There
was not a significant difference in male and female
prevalence in the age groups under age 75 (Bachman et
al., 1992).
7
Researchers have yet to uncover any underlying
cause of AD. It is unlikely that AD is strictly a
product of environmental influences. More likely, it
involves a complex interweaving of heredity and the
environment. In fact, molecular genetic abnormalities
have been identified in a number of families in whom AD
is inherited (familial AD) (Blass, 1993). A
susceptibility locus, particularly in later-onset
familial AD, has been identified on 19ql3.2 (Pericak-
Vance et al., 1991). The gene for apolipoprotein E
(APOE) is located in this region (Blass, 1993). APOE
has three common alleles: APOE-' 2, APOE-' 3, and APOE-c 4
(Corder et al., 1993). A statistically significant
association has been demonstrated between the APOE i . 4
allele and both late-onset familial AD and sporadic AD
(Saunders et al., 1993). The allele frequency of APOE-
e 4 in patients with AD has been reported to be three
times higher than in age-matched unrelated controls
(0.50 ± 0.6 vs 0.16 ± 0.03, p=0.014) (Ernst and Hay,
1994). In one study, a total of 80% of familial and
64% of sporadic AD late onset cases were found to have
at least one APOE-< 4 compared to 31% of control
subjects (Corder et al., 1993). Another study also
found a significantly higher APOE-' 4 allele relative
frequency in an AD group (late-onset sporadic) compared
8
with a control group (0.35 among AD versus 0.13 among
controls) (Tsai et al., 1994). In a sample of 42
families, the AP0E-C 4 allele in a double dose was
nearly always related to the occurrence of AD by age
80. This suggests that APOE-c 4 gene dose, as well as
presence, may be a major risk factor for late onset AD
and that APOE-c 4 may be directly relevant to the
etiology of AD (Corder et al., 1993).
Within families in which early onset AD appears to
be inherited in an autosomal dominant fashion, age of
onset tends to be consistent, with patients in these
families experiencing a relatively early onset and more
severe course of disease (Blass, 1993). In a study of
late onset familial AD, age at onset was related to
APOE-c 4 gene dose. Each additional APOE-c 4 allele
shifted onset to younger age: mean onset was 84.3
[standard error of the mean (SEM) 1.3] years in
subjects who did not have APOE-c4; 75.5 (SEM, 1.0)
years in subjects with one APOE-i 4; and 68.4 (SEM,
1.2) years in subjects with two APOE-i 4 alleles (Corder
et al., 1993) .
Differences in i 4 frequencies in different
nationalities have been found with Caucasians typically
ranging from p(4=0.l3 to pr4=0.16, with an overall
average frequency of p(4=0.15 (n=5,805). Swedes
9
(pe4=0.21) and Finns (pe4=0.24) have the highest
reported frequencies (Yu et al., 1994). The extension
of data from the late-onset familial AD families to the
population of apparently sporadic AD patients suggests
an important functional role for the t4 allele in the
pathogenesis of AD. The e4 allele may be viewed as a
susceptibility gene or risk factor for AD that can be
tested in properly constructed epidemiological studies
(Saunders et al., 1993).
Since the t 4 allele is not associated with
sporadic disease in about one third of the cases and
may only serve to increase susceptibility, it is
important to investigate what environmental factors may
work to promote disease development in more or less
susceptible individuals. In this study we examine a
potential environmental risk factor, occupational
Extremely Low Frequency Electromagnetic Field (ELF EMF)
exposure. ELF electromagnetic fields (50-60 Hz) is
used in most households and businesses. In a study by
Sobel et al., ELF EMF exposure appeared to act as a
risk factor for AD (Sobel et al., 1994). Although the
exact mechanism of its effect is not yet known,
mechanistic hypotheses involve the effect of EMF on
cellular calcium, and its subsequent influence on tau
proteins. Specific frequency/intensity EMF stimuli to
10
various types of cells has been shown to cause cellular
Ca2+ flux across the cell membrane (Goodman et al.,
1992, Walleczek, 1992, Goldberg and Creasey, 1991).
Excessive rises of Ca2+ have been shown to cause
changes in tau similar to those seen in AD NFT's
(Mattson et al., 1993). NFT's contain paired helical
filaments (PHF). The PHF's are formed from abnormal
aggregations of over-phosphorylated tau proteins
(Blass, 1993, Katzman and Jackson, 1991). In normal
brain, tau binds to and stabilizes microtubules. By
contrast, PHF-tau is unable to bind to microtubules
(Strittmatter et al., 1994). An in vitro study of
chick brain tissue found that holding the EMF field
strength constant and varying the frequency revealed
Ca2+ efflux enhancement in two regions- 15 Hz and 45-
105 Hz (Cleary, 1993).
11
Chapter II
Methods
The study subjects were selected from 990 patients
examined for memory loss at the University of Southern
California affiliated Alzheimer's Disease Diagnostic
and Treatment Center (ADDTC) at Rancho Los Amigos
Medical Center in Downey, California. All patients
received a comprehensive diagnostic evaluation that
included history, physical and mental status
examinations, and laboratory tests, including complete
blood cell count, electrolytes, hepatic and renal
chemistry studies, thyroid functions, B2 and folic
acid, and syphilis serologic tests. Most patients also
received a neuroimaging study of the brain (computed
tomography or magnetic resonance imaging). Patient
historical information was obtained in an interview
with the caregiver of the patient, usually a spouse or
offspring (Chui et al., 1994). Of these patients, 361
were diagnosed as having probable AD according to
NINCDS-ADRDA criteria (McKhann et al., 1984). 25
patients diagnosed as having probable AD fulfilled the
criteria for definite AD, resulting from
histopathologic confirmation (Chui et al., 1994). Two
AD study samples were chosen: one with only subjects
who were at least 65 years of age at the first visit to
the Center (Study Sample #1), and the other, a subset
of the first, with only subjects who were at least 65
years of age at the onset of dementia (Study Sample
#2). The study was limited to people with a later age
at onset of disease because there is some speculation
that earlier onset may be influenced more by dominant
genetic rather than environmental factors. Because age
of onset is reported to the clinician by the caregiver
of the patient and the gradual onset of AD makes a
precise onset age difficult to pinpoint, there is
variability in reporting age at onset. For this reason
age at first visit was used as an indicator of later
disease onset. However, in the patient population,
there is some variability in the time between the
"true" onset of disease and the time that people first
seek medical evaluation. For this reason, Study Sample
#2, a subset of #1, was chosen. (This subset was
limited to people with a reported age of onset of at
least 65 years of age.) Because people come into the
clinic after disease onset, all subjects in sample #2
were also at least 65 years of age at their first
visit.
In both samples, subjects with vascular dementia,
possible AD, Down's Syndrome, or a mixed etiology were
excluded from the analyses. Patients with the
diagnosis of vascular dementia were excluded from the
analysis due to a suspected sizable overlap with
misdiagnosed AD patients. A few subjects with unknown
diagnoses or a diagnosis of "No Cognitive Impairment"
were also excluded. Cases were defined as subjects
with probable or definite AD. Controls included all
other non-AD demented subjects not previously mentioned
as excluded. (See Table 1 for a list of all diagnoses
included in the control group)
Although other previously mentioned studies used
outpoints of 6 and 7 years to dichotomize level of
subject education, in this study no association was
found between education level and disease group using
those cutpoints. However, exploratory analyses
revealed a significantly higher risk for AD in the
subjects in this study with 12 or fewer years of
education. Therefore, a high level of education was
defined to be greater than 12 years of education.
Subjects' occupations were defined as their
primary occupation throughout life. High/Medium versus
Low/No occupational ELF EMF exposure was determined by
criteria developed by a certified industrial hygienist,
Joseph Bowman, on the basis of his experience with
occupational EMF and a study of occupational exposure
14
Table 1: Diagnoses of Subjects Included in Control Group
Diagnosis N
Age Associated Memory Impairment 10
Alcohol abuse or dependence/Current alcohol use 21
Amnestic syndrome 1
Cerebrovascular disease not meeting ADDTC criteria for 8
Ischemic Vascular Dementia
Central nervous system infection 1
Delirium 5
Depression/depressive mood disorder 10
Diagnosis deferred 2
Drug abuse or dependence 10
Etiology Undetermined 12
Frontal lobe degeneration 1
Head trauma 2 6
Lewy body disease 1
Medication (toxic effect or metabolic derangement)/ 12
metabolic derangement/metabolic disorder/toxin
Normal pressure hydrocephalus 5
Other 18
Parkinson's disease 6
Pick's disease or other frontal temporal syndrome 1
Previous hypoxia/anoxia 1
Progressive supranuclear palsy 1
152
15
to EMF (Sobel et al., 1994). Dressmakers, seamstresses
and tailors were judged to have medium to high exposure
based on direct measurements of sewing machines (Sobel
et al., 1994). The quantitative criteria for defining
occupational exposure levels were the following: high
exposure means an average exposure of at least 10
milligauss (mG) or regular intermittent exposures of at
least 100 mG; medium exposure means an average
exposure of between 2 and 10 mG or intermittent
exposure of at least 10 mG; low exposure means an
average exposure below 2 mG and intermittent exposure
below 10 mG (Sobel et al., 1994). Previous reports
have found different effects of EMF exposure on men and
women (Sobel et al., 1994). For this reason, each
analysis was done for men and women separately as well
as for men and women combined.
Statistical Methods
Separate analyses were done on the two study
samples. Descriptive statistics were computed for
cases and controls in each categorical risk factor
group. The mean and standard deviation of the
continuous risk factors were also computed separately
for cases and controls. Logistic regression was used
to test for the significance of each risk factor (ELF
EMF exposure, sex, education, age at first visit, and
age at onset of dementia) univariately. Logistic
regresssion was then used to assess several
multivariate models. For each risk factor, H0: B=0
was tested with both the likelihood ratio x2 test and
Wald's test. Odds Ratios were estimated as exp(£) with
$ representing the log relative risk derived via
maximum likelihood estimation under the logistic model.
The 95% confidence intervals (C.I.'s) for the odds
ratio estimates were calculated as follows:
Exp( & ± 1. 96[ S. E. (6)] }
The likelihood ratio and Wald's tests evaluate the same
hypothesis but with slightly different methods. For
this reason, they produce similar but not always
identical results. For example, it is possible that
the likelihood ratio test of a covariate would give a
p-value of less than 0.05 but the C.I., which uses the
same standard error as the Wald's test, would include
1.00. The final multivariate models were chosen
according to both clinical and statistical
appropriateness. Statistically, all variables that
were significant at p<0.05 were included as covariates
in the multivariate model. The objective of this was
to determine variables which were statistically
independently related to AD. Since the purpose of this
17
study was to examine the effect of high/medium EMF
exposure on the risk of AD, age, gender, and education,
which are known risk factors for AD, had to be
controlled for.
Chapter III
Results
In Study Sample #1, there were 326 AD cases and
152 other demented controls (Table 2). Ten cases and
five controls were missing information about their
years of education. Six cases and 17 controls were
missing information about their age at onset. Both
male and female cases had higher rates of high/medium
ELF EMF exposure than the controls (Table 2). The
proportion of male cases with high/medium exposure was
slightly higher than the proportion of similarly
exposed female cases. There were three times more
female cases than male cases. There were an equal
number of male and female controls. A majority of male
and female cases and controls had 12 or fewer years of
education (Table 2). The proportions of male and
female cases with 12 or fewer years of education were
higher than the male and female controls (Table 2).
The difference between the proportions of female cases
and controls with lower education is much greater than
the difference between the proportions of male cases
and controls (Table 2). The average age at first visit
was about the same for cases and controls, males and
females. The average age at onset was about the same
19
Table 2: Descriptive Statistics for AD Cases and Controls3
Risk Factor Cases Controls1 3
ELF EMFC
N (%
)
N (%
)
Males
Low Exposure 75 (87.2%) 73 (96.1%)
High/Medium
Exposure
11 (12.8%) 3 ( 3.9%)
Females
Low Exposure 218 (90.8%) 73 (96.1%)
High/Medium
Exposure
22 ( 9.2%) 3 ( 3.9%)
Sex
Male 86 (26.4%) 76 (50.0%)
Female 240 (73.6%) 76 (50.0%)
Education
Males
<12 Years 52 (63.4%) 42 (56.8%)
>12 Years 30 (36.6%) 32 (43.2%)
Females
<12 Years 187 (79.9%) 47 (64.4%)
>12 Years 47 (20.1%) 26 (35.6%)
Age at First
Visit
H
MeantSD
H
Mean±SD
Males
86 7 5.8±6.3 76 77.1±6.8
Females
240 77.7±6.5 76 77 . 7±7 . 6
Age at Onset
Males
86 10.7 ±7.0 69 74.1±7 . 8
Females
234 72.7±6 . 8 66 74.8±8.3
3 Study Sample #1
b Controls include subjects with cognitive impairment due to
varying etiologies, other than AD, as listed in Table la.
c Extremely Low Frequency Electromagnetic Fields
20
for men and women. The controls had a slightly higher
average age at onset than the cases.
In the Study Sample #2, there were 276 cases and
126 controls (Table 3). Nine cases and four controls
were missing information about education. One case was
missing information about her age at first visit.
Except for the slightly higher average age at first
visit and average age at onset Sample #2, the results
of the descriptive statistics for Sample #2 were
similar to those reported for Sample #1.
We provide univariate analytic results, but note
that multivariate analyses are more appropriate. In
the univariate logistic regression analyses for Study
Sample #1, men and women combined ELF EMF exposure,
sex, education, and age at onset were all found to be
statistically significant risk factors at the 0.05 a-
level (Table 4). The OR for high/medium EMF exposure
was 2.74 [95% C.I.=(1.12,6.69)]. AD risk was increased
by high/medium occupational ELF EMF exposure, female
sex, 12 or fewer years of education, and younger age at
onset of dementia. In this, and all subseguent
univariate and multivariate analyses, education is
dichotomized at 12 or fewer years. Although other
studies have used 6 or 7 years, in this study
dichotomizing at these numbers did not show a
Table 3: Descriptive Statistics for AD Cases and Controls3
Risk Factor Cases Controls1 5
ELF EMF°
N m N (%)
Males
Low Exposure 60 (87.0%) 63 (96.9%)
High/Medium
Exposure
9 (13.0%) 2 ( 3.1%)
Females
Low Exposure 190 (91.8%) 58 (95.1%)
High/Medium
Exposure
17 ( 8.2%) 3 ( 4.9%)
Sex
Male 69 (25.0%) 65 (51.6%)
Female 207 (75.0%) 61 (48.4%)
Education
Males
<12 Years 48 (72.7%) 39 (60.9%)
>12 Years 18 (27.3%) 25 (39.1%)
Females
<12 Years 162 (80.6%) 35 (60.3%)
>12 Years 39 (19.4%) 23 (39.7%)
Age at First
Visit
N Mean±SD
N
Mean+SD
Males
69 77.6±5.5 65 77.2±7.0
Females
206 79.0+5.7 61 78.1±7 . 3
Age at Onset
Males
69 73.1±5.3 65 75.0±7.2
Females
207 74.2±5.7 61 75.9±7 . 6
3 Study Sample #2
b Controls include subjects with cognitive impairment due to
varying etiologies, other than AD, as listed in Table la.
c Extremely Low Frequency Electromagnetic Fields
Table 4: Univariate Analysis of EMF and Other Risk Factors in Alzheimer's Disease Study3
Risk Factor Levels OR (95% C.I.)
1
x 1
1
O I
1
1
1
1
P-Valuec P-Valued
EMF Exposure None/Low
High/Medium
1. 00
2 . 74 (1.12-6.69) 5.95 0.01 0.03
Sex Male
Female
1. 00
2 . 79 (1.87-4.17) 25.22 <0.00001 <0.0001
Education >12 Years
<12 Years
1. 00
2 . 02 (1.33-3.08) 10.76 0. 001 0.001
Age at first visit Trendb 0. 99 (0.97-1.02) 0.16 0. 69 0.69
Age at onset Trend5 0.96 (0.93-0.99) 9.35 0. 002 0. 003
3 Study Sample #1
b Test for trend in number of years
c Likelihood Ratio Chi-Square Test
d Wald's Test
23
significant difference in AD risk. Exploratory
analyses in this study revealed the cutpoint of 12
years to produce a significant difference in AD risk.
Age at first visit was not statistically significantly
related to AD diagnosis. The univariate analyses for
Study Sample #2, for men and women combined, yielded
essentially the same results as the sample #1 analyses
(Table 5). The OR for high/medium EMF exposure was
2.52 [95% C.I.=(0.94,6.72)]. The only minor difference
was that EMF exposure was of borderline significance in
sample #2, with the Wald's test giving a P-value
slightly above 0.05, but below 0.06.
In the univariate analyses for the males only in
sample #1 EMF exposure was of borderline significance
while age at onset was a significant risk factor (Table
6). The OR for high/medium EMF exposure was 3.57 [95%
C.I.=(0.96,13.32)]. AD risk increased with high/medium
EMF exposure and younger age at onset of dementia.
Education level and age at first visit were not
significantly related to AD diagnosis. These same
analyses for sample #2 found only EMF exposure to be a
significant risk factor, again with AD risk increasing
with high/medium exposure (Table 7). The OR for
high/medium EMF exposure was 4.73 [95%
Table 5: Univariate Analysis of EMF and Other Risk Factors in Alzheimer's Disease Study3
Risk Factor Levels OR (95% C.I.)
v2 c
K 1
P-Valuec P-Valued
EMF Exposure None/Low
High/Medium
1.00
2.52 (0.94-6.72) 4 . 04 0.04 0.07
Sex Male
Female
1. 00
3 . 20 (2.05-4.98) 26.80 <0.00001 <0.0001
Education >12 Years
<12 Years
1. 00
2 . 39 (1.50-3.81) 13 . 27 0.0003 0.0003
Age at first visit Trendb 1. 03 (0.99-1.06) 2.39 0.12 0.12
Age at onset Trend1 3 0.96 (0.93-1.00) 4 . 85 0.03 0. 03
3 Study Sample #2
b Test for trend in number of years
c Likelihood Ratio Chi-Square Test
d Wald's Test
Table 6: Univariate Analysis of EMF and Other Risk Factors in Alzheimer's Disease Study3
Males only
Risk Factor Levels OR (95% C.I.)
v2 c
A 1
P-Valuec P-Valued
EMF Exposure None/Low
High/Medium
1. 00
3 . 57 (0.96-13.32) 4 . 27 0.04 0. 06
Education >12 Years
<12 Years
1. 00
1.32 (0.69- 2.51) 0 . 72 0.40 0.40
Age at first visit Trendb 0. 97 (0.92- 1.02) 1.77 0.18 0. 19
Age at onset Trendb 0. 94 (0.90- 0.98) 8. 14 0. 004 0.01
a Study Sample #1
b Test for trend in number of years
c Likelihood Ratio Chi-Square
d Wald's Test
Table 7: Univariate Analysis of EMF and Other Risk Factors in Alzheimer's Disease Study3
Males only
Risk Factor Levels OR (95% C.I.)
v2 c
* 1
P-Valuec P-Valued
EMF Exposure None/Low
High/Medium
1. 00
4 .73 (0.98-22.77) 4 . 77 0.03 0.05
Education >12 Years
<12 Years
1. 00
1.71 (0.82- 3.58) 2.05 0.15 0.15
Age at first visit Trendb 1.01 (0.96- 1.07) 0.10 0.76 0.76
Age at onset Trendb 0.96 (0.90- 1.01) 2 . 78 0.10 0.10
3 Study Sample #2
b Test for trend in number of years
c Likelihood Ratio Chi-Square Test
d Wald's Test
27
C.I.=(0.98,22.77)]. Education, age at first visit, and
age at onset were not significantly related to AD risk.
In the univariate analyses for the females only in
sample #1, education and age at onset were
statistically significant risk factors (Table 8). The
OR for high/medium EMF exposure was 2.46 [95%
C.I.=(0.71,8.44)]. AD risk increased with 12 or fewer
years of education and a younger age at onset.
High/medium EMF exposure was associated with a non-
significantly elevated risk for AD. The results were
the same for sample #2, except that age at onset was of
borderline significance (p=0.06; (Table 9). The OR
for high/medium EMF exposure was 1.73 [95%
C.I.=(0.49,6.11)].
A series of multivariate models were analyzed to
test the significance of high/medium EMF exposure after
adjusting for other risk factors. All Sample #1
multivariate models testing whether BEMF=0, for men and
women combined, were statistically significant at the
0.05 a-level (Table 10). These models tested the
significance of EMF exposure after controlling for the
univariately significant risk factors. The same
results were found for Sample #2 except that
high/medium EMF exposure in some of these models was
only borderline significant (Table 11). In all Sample
Table 8: Univariate Analysis of EMF and Other Risk
Females only
Factors in Alzheimer's Disease Study3
Risk Factor Levels OR (95% C.I.)
v 2 c
k 1
P-Valuec P-Valued
EMF Exposure None/Low
High/Medium
1. 00
2 .46 (0.71-8.44) 2 .48 0.12 0.15
Education >12 Years
<12 Years
1.00
2 . 20 (1.24-3.92) 6.97 0.01 0.01
Age at first visit Trendb 1. 00 (0.96-1.04) 0.01 0.94 0.94
Age at onset Trendb 0.96 (0.92-1.00) 4 .46 0. 03 0. 04
a Study Sample #1
b Test for trend in number of years
c Likelihood Ratio Chi-Square Test
d Wald's Test
Table 9: Univariate Analysis of EMF and Other Risk Factors in Alzheimer's Disease Study3
Females only
Risk Factor Levels OR (95% C.I.)
X2lC
P-Valuec P-Valued
EMF Exposure None/Low
High/Medium
1. 00
1.73 (0.49-6.11) 0.81 0.37 0.39
Education >12 Years
<12 Years
1. 00
2.73 (1.45-5.13) 9 . 40 0.002 0. 002
Age at first visit Trend6 1. 03 (0.98-1.08) 1. 19 0.28 0. 28
Age at onset Trend6 0.96 (0.92-1.00) 3 .51 0. 06 0. 06
3 Study Sample #2
b Test for trend in number of years
c Likelihood Ratio Chi-Square Test
d Wald's Test
Table 10: Multivariate Analysis of EMF and Other Risk Factors in Alzheimer's Disease Study3
Odds Ratios
Model No. of Log- x2 EMF- Sex- Education- P-valueb P-valuec
Parameters Likelihood High/Med Female Age Onset <12 Years
1 1 -295.94 2.74
2 1 -286.31
3 1 -271.99
4 1 -283.98
5 2 -283.01 6. 59d 2 . 96
6 2 -268.15 7 . 67e 3 . 66
7 2 -279.75 8.46f 3.90
8 3 -254.65 7 . 829 3 .83
9 3 -268.47 9 . 34h 4 . 26
10 3 -251.83 8.56' 4 . 59
11 4 -239.61 9 . 07' 4 .93
2.79
0.96
2.02
2 .85 0.01 0. 02
0.96 0.01 0. 02
1.99 0. 004 0.01
3 . 12 0.95 0. 01 0.01
2 .77 1.71 0. 002 0.01
0.95 2.41 0.003 0.01
3 . 08 0 . 94 2. 12 0.003 0.01
Study Sample #1
Likelihood Ratio Chi-Square Test of BEMF=0 (controlling for other variables in the model)
Wald's Test of B,
H c
H c
H c
H c
H
c
H
B,
EMF
3EHF
EHF
3EMF
EHF
3EHF
EHF
SEX °»
AGE ONSET- 0 ' 1
= 0/ 1
=0;
=0;
EHF=0 (controlling for other variables in the model)
DF
EDUCATION
SEX,AGE ONSET
SEX,EDUCATION
AGE ONSET,EDUCATION
SEX,AGE ONSET,EDUCATION
DF
DF
1 DF
1 DF
=0; 1
=°;
DF
1 DF
u>
o
Table 11: Multivariate Analysis of EMF and Other Risk Factors in Alzheimer's Disease Study3
Model No. of
Parameters
Log-
Likelihood
X2
EMF-
High/Med
Odds
Sex-
Female
Ratios
Age Onset
Education-
<12 Years
P-valuebP-valuec
1 1 -247.95 2 . 52
2 1 -236.57 3.20
3 1 -247.55 0.96
4 1 -235.31 2.39
5 2 -234.27 4 . 6 ld 2.76 3.26 0.03 0. 05
6 2 -245.61 3 . 89e 2.48 0.96 0. 05 0.07
7 2 -233.14 4 . 33f 2 .84 2.34 0. 04 0. 06
8 3 -231.22 4 . 4 29 2.71 3.41 0.96 0.04 0.05
9 3 -219.91 5. 32h 3.26 3 . 34 2.18 0.02 0. 04
10 3 -230.38 4.11' 2.80 0.96 2.44 0.04 0. 07
11 4 -216.40 5.03' 3.20 3 . 51 0.95 2 .29 0.02 0. 04
Study Sample #2
Likelihood Ratio Chi-Square Test of BEHF=0 (controlling for other variables in the model)
(controlling for other variables in the model)
=0; 1 DF
DF
=0; 1 DF
0; 1 DF
Wald's Test of BEHf=0
b e
B l
b e
b e
b e
B,
EMF
EMF
SEX
AGE ONSET
EDUCATION
SEX.AGE ONSET
SEX,EDUCATION
AGE ONSET.EDUCATION'
SEX.AGE ONSET,EDUCATION
=0; 1 DF
=0; 1
=0;
DF
1 DF
32
#1 multivariate models for males only, high/medium EMF
exposure was statistically significantly elevated
(Table 12). Similar results were found in Sample #2,
but high/medium EMF exposure was borderline significant
(Table 13). Multivariate analyses of Sample #1 for
females only revealed only one model (controlling for
education) in which high/medium EMF exposure was of
borderline significance (Table 14). None of the other
models which controlled for the univariately
significant risk factors were significant. In sample
#2 multivariate analyses for females only, risk
associated with high/medium EMF exposure was not
significant after controlling for the univariately
significant risk factors (Table 15).
Three final multivariate models were chosen to
best estimate the relative risk of AD associated with
high/medium occupational EMF exposure. For consistency
and because the results for the two study samples were
so similar, all models were chosen from the analyses
done on sample #1. One model is for males, one for
females, and one for both sexes (Table 16). The model
for both sexes includes sex, education, and age at
onset as covariates, in addition to EMF exposure. The
OR for high/medium EMF exposure of 4.93 [95%
C.I.=(1.43,16.95)] was highly significant. All
Table 12: Multivariate Analysis of EMF and Other Risk Factors in Alzheimer's Disease Study3
Males Only
Model No. of
Parameters
Log-
Likelihood
X2
EMF-
High/Med
Odds Ratios
Age Onset
Education-
<12 Years
P-valueb P-Valuec
1 1 -109.85 3 . 57
2 1 -102.43 0.94
3 1 -107.57 1.32
4 2 - 98.34 8. 19d 10. 17 0.94 0.004 0.03
5 2 -104.51 6. lle 5. 39 1. 23 0.01 0.03
6 3 - 94.65 7.81f 9.97 0.93 1.76 0.01 0.03
3 Study Sample #1
b Likelihood Ratio Chi-Square Test of BEMf=0 (controlling for other variables in the model)
c Wald's Test of BEHf=0 (controlling for other variables in the model)
H 0 = B EHF|AGE ONSET- ® ' ^
f H o : B EHF!EDUCAIION= 0 ' 1
B o : B EHF|AGE ONSET,EDUCATION- 0 '
Table 13: Multivariate Analysis of EMF and Other Risk Factors in Alzheimer's Disease Study3
Males Only
Model No. of
Parameters
Log-
Likelihood
X2
EMF-
High/Med
Odds Ratios
Age Onset
Education-
<12 Years
P-valueb P-valuec
1 1 -90.44 4.73
2 1 -91.43 0.96
3 1 -89.07 1.71
4 2 -89.14 4 . 59d 4 . 62 0.96 0.03 0.06
5 2 -86.82 4 . 50e 4 . 59 1.61 0.03 0.06
6 3 -85.86 4 . 30* 4 .49 0.96 1.70 0.04 0.06
Study Sample #2
Likelihood Ratio Chi-Square Test of BEHf=0 (controlling for other variables in the model)
Wald's Test of BEHF=0 (controlling for other variables in the model)
H o : b ehf{age onset = 0 ' 1 D F
H o : B EMF!EDUCAT!ON= 0 ' 1 D F
B o : b e h f ;age onsei,educaiion= b B F
Table 14: Multivariate Analysis of EMF and Other Risk Factors in Alzheimer's Disease Study3
Females Only
Model No. of
Parameters
Log-
Likelihood
X2
EMF-
High/Med
Odds Ratios
Age Onset
Education-
<12 Years
P-valueb P-valuec
1 1 -173.09 2.46
2 1 -155.84 0.96
3 1 -164.91 2.20
4 2 -155.05 1. 58d 2.09 0.96 0.21 0.25
5 2 -163.04 3 . 73e 3 .51 2.20 0.05 0.10
6 3 -143.82 2 . 47f 2 .90 0.95 2 . 53 0.12 0.16
3 Study Sample #1
b Likelihood Ratio Chi-Square Test of BEHf=0 (controlling for other variables in the model)
c Wald's Test of BEHF=0 (controlling for other variables in the model)
H o : b emf;age onset= 0 ' 1 D F
f H o : b emf!Education_0 ' 1
B o : B EHF|AGE ONSET.EDUCAIION-0 ' ^
Table 15: Multivariate Analysis of EMF and Other Risk Factors in Alzheimer's Disease Study3
Females Only
Model No. of
Parameters
Log-
Likelihood
X2
EMF-
High/Med
Odds Ratios
Age Onset
Education-
<12 Years
P-valueb P-valuec
1 1 -143.34 1.73
2 1 -141.99 0.96
3 1 -133.05 2.73
4 2 -141.61 0. 77d 1.71 0.96 0.38 0.41
5 2 -132.38 1. 34e 2.27 2.69 0.25 0.29
6 3 -129.85 1. 27f 2 .25 0.95 2.81 0.26 0.30
3 Study Sample §2
b Likelihood Ratio Chi-Square Test of BEHf=0 (controlling for other variables in the model)
c Wald's Test of BEHF=0 (controlling for other variables in the model)
* Ho: Behf, ageonsei=0; 1 DF
d H o= B EHF[EDUCATION °> 1 D F
H * R —f l * 1 nu
o' D EMF[AGE ONSET.EOUCATION ' x
Table 16: Final Multivariate Models of EMF and Other Risk Factors in Alzheimer's Disease Study3
Risk Factor
EMF- High/Medium Sex- Female Education- <12 Years Age at Onset
OR (95% C.I.) OR (95% C.I.) OR (95% C.I.) OR (95% C.I.)
Both Sexes
4.93 (1.43-16.95) 3.08 (1.97-4.83) 2.12 (1.32-3.39) 0.94 (0.92-0.97)
Males
10.17 (1.27-81.64) 0.94 (0.90-0.98)
Females
2.90 (0.65-13.03) 2.53 (1.37-4.67) 0.95 (0.92-0.99)
3 Study Sample #1
38
covariates are statistically significant univariately
and remain so in the final multivariate model. The
final multivariate model for the males only includes
the covariate age at onset, in addition to EMF
exposure. The OR for high/medium EMF exposure of 10.17
[95% C.I.=(1.27,81.64)] was highly significant. Age at
onset is statistically significant univariately and
remains significant in the multivariate model.
Education, which was not univariately significant, did
not become statistically significant when added to the
multivariate model and so was not included in the final
model. The female only final multivariate model
includes the covariates age at onset, education, as
well as EMF exposure. The OR for high/medium EMF
exposure was 2.90 [95% C.I.=(0.65,13.03)]. Education
and age at onset were univariately and multivariately
significant. The odds ratio for high/medium EMF
exposure was not statistically significant univariately
or in the final multivariate model, although in both
cases it approached significance (p<0.12).
39
Chapter IV
Discussion
In this study, high/medium occupational ELF EMF
exposure was not a statistically significant risk
factor for females, either univariately or
multivariately. However, in the final multivariate
model for Study Sample #1, after controlling for both
education and age at onset, high/medium EMF exposure
had an OR of 2.90 (0.65-13.03) (Table 16). This model
was of borderline statistical significance. Given the
size of the relative risk estimate and the similarity,
in both size and direction, to the results found by
Sobel, et al. (1994), and the relatively small number
of controls, further study of the effects of this
exposure on females would be warranted.
The significance of EMF for males and the lack
thereof for females seems to be due to the higher rate
of high/medium occupational EMF exposure for male cases
as compared to female cases, while the proportion of
controls with high/medium exposure is about the same
for males and females (Tables 2,3). This rate
differential yielded a sizable difference in the OR
estimates for high/medium EMF exposure in males versus
females. The OR for males, after controlling for
40
education and age at onset is 10.17, which is three
times the relative risk for females (Table 16). This
finding suggests that sex may act as an effect modifier
for EMF exposure. However, there are several other
possible explanations for this differential relative
risk by sex. First, it may be due to certain
characteristics of EMF exposure that vary by
occupation. In this study, the majority of female
high/medium EMF exposure occupations involved exposure
to sewing machines, while the male high/medium exposure
occupations were primarily machinists, carpenters, and
electricians (Table 17). Second, it may be that
occupational exposures other than ELF EMF are the true
etiologic agents and these exposures are differentially
present in the male versus female high/medium EMF
exposure occupations. Finally, since occupational
category is a surrogate for directly measured EMF
exposure, we may have differentially misclassified male
versus female dominated occupations by, for example,
getting better information for males and missing much
home sewing.
This analysis partially confirms the results from
a previous study on ELF EMF exposure done by Sobel, et
al (Sobel et al., 1994). In that study, the proportion
of AD cases with high/medium exposure was 9.8% in the
41
Table 17: High/Medium EMF Exposure Occupations by Sex
N
Occupation Male Female
Carpenter 3 0
Machinist/Machine Mechanic/Machine Operator 6 1
Tool and Die Maker/Welding Contractor 2 0
Electrician/Electrical Assembler/Assembler/ 2 4
Electrical Circuit Boards/Wiring and
Soldering
Tailor/Seamstress/Sewer/Dressmaker 1 16
Typist/Clerk Typist/Keypunch 0 4
14 25
42
series-combined analyses (three different study samples
were combined). In the study done here, the proportion
of cases with high/medium EMF exposure was 10.1%, in
Study Sample #1, and 9.4% in Study Sample #2. The
proportion of controls with high/medium exposure in
Sample #1 was 3.9%, and was 4.0% in the Sample #2.
Sobel, et al. found the proportion of controls with
high/medium exposure to be 3.4%.
There are many possible reasons the results are
found to be different. First, the studies were
conducted at different study sites. Over 70% of the
data used in the series-combined analyses from the
study by Sobel, et al. was from subjects living in
another country, Finland. Although the same EMF
exposure classification methods used in this study were
used for the study by Sobel, et al., differences in
diagnostic practices could account for some differences
found in the results of the two studies. If there are
any differences in the etiology of AD between the Finns
and other nationalities, this could be a factor in the
differences found between these studies. The unusually
high frequency of the 4 allele that was reported in
the Finnish population could support the theory that
there is a genetic and environmental interaction (Yu et
al., 1994). International differences in home and
43
business electrical wiring, as well as differences in
work practices, safety and otherwise, could also be
factors. Secondly, the selection of control subjects
in the two studies was different. In the study by
Sobel, et al., non-demented controls made up 83.4% of
the series-combined control group (Sobel et al., 1994).
The two study samples in this study may be subject
to classification biases. Individuals could have been
categorized into the wrong EMF exposure category in
several ways. Each occupation's EMF exposure level was
based on an assessment of that occupation in general,
not on the individual's own experience in that
occupation. If a person was in an occupation in which
exposure was considered "low" but they were exposed to
high amounts of EMF either due to unique aspects of
their job environment or due to uncontrollable proximal
factors, such as a nearby power plant, then their EMF
classification would be inaccurate. In addition,
duration of exposure was not assessed in this study, so
an individual in a "high" exposure occupation for forty
years would have the same classification as an
individual in that same occupation for twenty years.
If a longer duration of EMF exposure increases the risk
of developing AD and if males, on the average, held
these high/medium exposure occupations longer than the
44
females, the risk of AD would be higher for men.
Because this study did not examine duration, it cannot
be said for certain if this is the reason for the
higher OR in men versus women. In future studies it
would be beneficial to document a complete occupational
history, with duration of each position held, for all
subjects.
Further study of EMF exposure in epidemiological
studies is important, however biological studies of how
EMF may cause the pathological changes characteristic
of AD are also vital. Theories about how EMF causes
these changes are only educated guesses. It is
suspected that the changes occur when EMF affects
cellular calcium. Calcium is a major intracellular
messenger which links electrical activity of excitable
membranes to biochemical processes, often through an
interaction with calcium binding proteins, particularly
calmodulin (Crapper McLachlan et al., 1987). The brain
also contains calbindin D28K (CaBP) , a protein
immunologically and physiochemically similar to the
vitamin D dependent calcium binding protein of other
tissues (Crapper McLachlan et al., 1987). Alterations
in concentration of calcium binding proteins such as
calmodulin and CaBP in diseases of the brain could be
expected to affect calcium homeostasis and,
45
consequently, the regulation of a large number of brain
functions (Crapper McLachlan et al., 1987).
Recent work indicates that the calmodulin
messenger RNA (CaM) content in AD affected cerebral
cortex is only 68% of age-matched, non-AD, control
cerebral cortex (Crapper McLachlan et al., 1987). The
reduction in calmodulin mRNA content (and protein
product) could be a homeostatic response to reduced
calcium uptake and cytosolic concentration in brain
tissue analogous to that reported in non-nervous tissue
(Crapper McLachlan et al., 1987). CaBP occurs in high
concentrations in certain neurons including hippocampal
CA1 and neurons of the nucleus basalis, cells which
undergo a selective degeneration in AD (Crapper
McLachlan et al., 1987). It is not yet known whether
calcium binding proteins are directly related to the
primary pathogenic events which initiate AD or are
secondary homeostatic responses (Crapper McLachlan et
al., 1987). However, reduced concentrations of these
important proteins are almost certainly important in
the pathogenesis of functional deficits in neurons and
glia and may contribute to calcium mediated
irreversible cytotoxic events in the latter stages of
the pathological process (Crapper McLachlan et al.,
1987). If the calcium binding proteins are directly
46
related to the primary pathogenic events or contribute
to cytotoxicity in the latter stages, then the effect
of ELF EMF on Ca2+ homeostasis may occur through the
calcium binding proteins.
Alterations in calcium-dependent kinases in AD
have been reported (Mattson et al., 1993). However, a
number of kinases have been reported to induce
abnormalities on tau associated with PHF, including
Ca++/calmodulin-dependent protein kinase, and perhaps
protein kinase C. Which of these kinases or which
combination of kinases plays a significant role in the
formation of PHF in the AD brain is not yet clear
(Blass, 1993).
Biologically, it appears plausible that EMF could
produce the changes in the brain which lead to AD. In
this study, occupational exposure to high/medium ELF
EMF has been shown to increase the risk of developing
AD relative to other dementias. While EMF exposure may
not be the definitive explanation to the question of
what causes AD, further exploration in that direction
would surely be worthwhile.
47
References
Alafuzoff I, The Pathology of Dementias: An
Overview, Acta Neurologica Scandinavica. Suppl.
139: 8-10, 1992.
Bachman DL et al., Prevalence of Dementia and
Probable Senile Dementia of the Alzheimer Type in
the Framingham Study, Neurology. 42: 117-118,
January 1992.
Blass JP, Pathophysiology of the Alzheimer's
Syndrome, Neurology. 43(suppl 4): S26-S30, August
1993 .
Buckwalter JG et al., Gender Differences on a Brief
Measure of Cognitive Functioning in Alzheimer's
Disease, Neurology. 50: 759, July 1993.
Chui HC et al., Extrapyramidal Signs and
Psychiatric Symptoms Predict Faster Cognitive
Decline in Alzheimer's Disease, Archives of
Neurology. 51: 677, July 1994.
Cleary SF, A Review of In Vitro Studies: Low-
Frequency Electromagnetic Fields, American
Industrial Hygiene Association Journal. 54(4):
179, April 1993.
Corder EH et al., Gene Dose of Apolipoprotein E Type 4
Allele and the Risk of Alzheimer's disease in Late
Onset Families, Science, 261: 921-923, August
1993 .
Crapper McLachlan DR et al., Calmodulin and Calbindin
D28k in Alzheimer Disease, Alzheimer Disease and
Associated Disorders. 1(3): 171-178, 1987.
Edwards JK et al., Are There Clinical and
Epidemiological Differences Between Familial and
Non-familial Alzheimer's Disease?, Journal of the
American Geriatrics Society. 39: 477, 1991.
Ernst RL, and Hay JW, The US Economic and Social Costs
of Alzheimer's Disease Revisited, American Journal
of Public Health. 84(8): 1261-1264, August 1994.
48
Friedland RP, Alzheimer's Disease: Clinical
Features and Differential Diagnosis, Neurology,
43(suppl 4): S45-S46, August 1993.
Goldberg RB, and Creasey WA, A Review of Cancer
Induction by Extremely Low Frequency
Electromagnetic Fields. Is There a Plausible
Mechanism?, Medical Hypotheses. 35: 2 68, January
1991.
Goodman R. et al., The Heat Shock HSP70 Gene is
Induced in a Variety of Cell Types Exposed to ELF
EM Fields or AC Current. Does the Cell Respond to
EM Field Exposure as Cellular Stress?, The Annual
Review of Research on Biological Effects of
Electric and Magnetic Fields From the Generation.
Delivery & Use of Electricity. San Diego,
California, November 8-12, 1992.
Katzman R, and Jackson JE, Alzheimer Disease: Basic
and Clinical Advances, Journal of the American
Geriatrics Society. 39: 516-520, 1991.
Khachaturian ZS, Diagnosis of Alzheimer's Disease,
Archives of Neurology. 42: 1097-1105, 1985.
Koga K, Two USC Projects Probe Cause, Cost of
Alzheimer's Disease, University of Southern
California Chronicle. Page 7, September 12, 1994.
Mattson MP et al., Comparison of the Effects of
Elevated Intracellular Aluminum and Calcium Levels
on Neuronal Survival and Tau Immunoreactivity,
Brain Research. 602: 21-22, 1993.
McKhann G et al., Clinical Diagnosis of
Alzheimer's Disease: Report of the NINCDS-ADRDA
Work Group under the auspices of Department of
Health and Human Services Task Force on
Alzheimer's Disease, Neurology. 34: 940-941, July
1984 .
Mortimer JA, and Graves AB, Education and Other
Socioeconomic Determinants of Dementia and
Alzheimer's Disease, Neurology. 43 (Suppl 4):
S39-S44, 1993.
49
Pericak-Vance MA et al., Linkage Studies in Familial
Alzheimer's Disease: Evidence for Chromosome 19
Linkage, American Journal of Human Genetics. 48:
1034-1050, 1991.
Saunders AM et al., Association of Apolipoprotein
E allele e4 with Late-Onset Familial and Sporadic
Alzheimer's Disease, Neurology. 43: 1467-1470,
August 1993.
Sobel E et al., Occupational Exposure to
Electromagnetic Fields is Associated with
Increased Risk for Alzheimer's Disease,
Neurobiology of Aging, 15 (Suppl 1): S74, 1994
(Manuscript under review).
Strittmatter WJ et al., Hypothesis: Microtubule
Instability and Paired Helical Filament Formation
in the Alzheimer Disease Brain Are Related to
Apolipoprotein E Genotype, Experimental Neurology.
125: 167-168, 1994.
Tsai MS et al., Apolipoprotein E: Risk Factor for
Alzheimer Disease, American Journal of Human
Genetics. 54: 646-648, 1994.
Van Duijn CM, Stijnen T, and Hofman A, Risk Factors for
Alzheimer's Disease: Overview of the EURODEM
Collaborative Re-Analysis of Case-Control Studies,
International Journal of Epidemiology. 20(2,
Suppl. 2): S4-S11, 1991.
Walleczek J, Electromagnetic Field Effects on Cells
of the Immune System: The Role of Calcium
Signaling, Federation of the American Societies
for Experimental Biology Journal. 6: 3178-3181,
1992.
Yu CE et al., The Apolipoprotein E/CI/CII Gene
Cluster and Late-Onset Alzheimer Disease, American
Journal of Human Genetics. 54: 637, 1994.
Appendix A: Listing of Individual Occupations by ELF
EMF Exposure Level Group and Disease Group
NO/LOW ELF EMF EXPOSURE
AD
Occupation
Physician
Ford Motor Co.
Manfact. Repres
Bus driver
Plant Supervisr
ForestryResrch
Military-pilot
OwnedMachineShp
RaceTrackEquip
Custodian,Laund
Hardware sales
InstallFloorCov
Truck Driver
Building Contrc
Farmer
Attorney
LA Co Engineer
GlassBlwr Muscn
Teacher/Account
Aircraft indust
Floor business
Engineer
Ice Manufacture
Jr.Coll.Teacher
Milk truck driv
Cabinet Maker
Window displays
Sheet Metal wrk
Post Office
Truck driver
Prof. Chemistry
Aircraft worker
Physician
Sales
Surveyr&Enginer
Clothing Sales
Mech Engineer
Truck driver
Lg.applnc sales
Sheet metal
Banker
Salesman
Dairyman
Navy/postoffice
Brick layer
Car spray paint
Construction
Gardening
Maint. mechanic
Paint sprayer
Accountant
Biochemist
Beautyshop ownr
Accounting
SoldReal Estate
Lumber Business
Postal Ser.Main
Engineer
Engineer
Chrysler emplye
Chrysler auto
Dept Head-Shell
Fireman
Business owner
Hotel Manager
Coffee Blender
Plant Manager
Lab Water Testr
Fire Dept
Physician
Supervisor-GE
Inspctr-Chryslr
Post Office
Saw Mill
Agrcltre/Crpntr
Accountant
Meat Co.
Housewife
Office clerk
Babysitter
Housewife
Sales Departmnt
Housewife
Credit Co Clerk
Cannery worker
Clerk-BowlAlley
Housewife
Laundry Worker
Housewife
Factory worker
Housewife
Realtor
Housewife
Elemtry Teacher
Housekeeper
Factory Worker
Librarian
Real Estate Brk
School Administ
Bookkeeper
Housewife
Preschl teacher
Bookkeeper
Housewife
Postmistress
Library
Chore Worker
Teacher
Housewife
Housewife
Office
Beauty shop
DouglasAircraft
Maintenance
Cleaned Houses
Missionary
Housekeeper
Auto Office
Housewife
Housewife
Cotton Fields
Beauty shop own
Cook-USC Fratrn
Sales
Housewife
Office work
DeptStore Clerk
Match factory
Owner-Restr&Bar
Stenographer
Nun til 40's
Clerk-Optometry
Grocery clerk
Piano Teacher
Factory worker
LVN
Math teacher
Asst Hotel Mgr
Insurance Busin
Clerical
Housewife
Housewife
Housewife
Cafeteria Manag
Teacher
Bookkeeper
Cook
Newspaper typis
Salesperson
Telephone Offic
Housewife/farmr
Cleaning lady
Semi-skilled
Undetermined
Laundry
Steam laundry
Roof co samples
Teacher
Housewife
Housewife
Sales
Rancho
Supvis mailroom
Beautician
Clothing cutter
Housekeeper
Sales-Penneys
Insurance co.
Housewife
Legal typist
Housewife
Hotel reg&cashr
Housewife
Nurses aide
Bookkeeper
Nurses aide
Housewife
CashierSchecker
Housewife
Housewife
Housewife
Secretary
Cosmetics Sales
Beautician
Teacher
Cook
Bookkeeper
Housekeeper
Hosp housekeepr
Asst to mayor
Housewife
Real Estate
Physician
Housewfe-writer
Housewife
Housewife
Housewife
Homemaker
Secretary
Factory work
Housekeeper
Mgr-janitorial
Teacher
Parts departmen
Teacher
Cleaned houses
Reg Nurse
Housewife
Beautician
Housewife
Drs' Office
Sales-auditing
Owned 2 stores
Waitress
LVN
Housewife
Mattel Toys
Housewife
Laundry
Housewife
House Worker
Office Work
Restaurant ownr
Housewife
Dry cleaning
Secretary
Clothing st own
Accounting
Office work
Sales
Shoe factory
Bank accountant
Managed aprtmts
PBX Operator
Teacher
Waitress
Cannery Worker
Housewife
Clerk
Owned business
Housewife
Teacher
Teacher
Factory work
Housewife
Housewife
Housewife
Proofreader
Housewife
Housewife
Mfg TortillaBox
Waitress
Hairdresser
Farm laborer
Wiring supervis
Housewife
Hospital aide
Housewife
Housewife
Housewife
Sales-Clothing
Secretary
Housewife
Teacher
Secretary
Housewife
Custodian
Housewife
Housewife
Housewife
VolunteerDocent
Housewife
Housewife
Market Owner
Cashier
Housewife
Teacher
Telephone Co.
Teacher's Aide
Business Woman
Bookkeeper
Medical Office
Office Worker
Housewife
Domestic Work
Factory
Custodian
Housekeeper
SchoolCustodian
Teacher's aide
AircraftAssembl
InventoryContrl
Clerical
CONTROL
Occupation
Aerospace
Accountant
Construction
VP Western Airl
Farmer
Aerospace prod.
Professor-Law
Social worker
Housewife
Auto Mechanic
Reporter/editor
Sales Manager
Farmer-Minister
Merchant
Auto Mechanic
Telephone
Chef
Truck driver
Kitchen supply
Administration
U.S. Steel
For Texaco
Turbine Operatr
Auto mechanic
Mail carrier
Truck operator
Computer opratr
Farmer
Attorney
Pilot,Bailbonds
Businessman
FireProtectnEng
Master Brewer
Civil Engineer
Lead Business
Merchant Marine
Tool Maker
VP Oil industry
Lumber Co.
Own Energy Co.
Pres, paper co.
Phone Co.
Quality control
Painter
Security Officr
Accountant
Machine Shop
Mechanic
Salesman
Maitenance Engn
EngineerCntrctr
Mortgage broker
Auto Mechanic
Dental Technicn
Engineer
Navy/INS/retird
Engineer/Econom
Steel Estimator
Baker
Meat Packer
Banker
Fireman,painter
Truck mechanic
Teachr/Principl
Minister
Mechanic
54
OilCoShiftWorkr
Factory worker
Race Track
Shipyard worker
Unknown
Truck Driver
Construction
Nun-CollegeProf
LVN
Teacher
Housewife
Housewife
Domestic Work
Secretary
Receptionist
Housewife
Hughs Aircraft
Housewife
Housewife
Lab Tech
SchoolBusDriver
Crossing Guard
Office Work
Librarian
PBX Operator
Housewife
Housewife
Clerical
Housewife
Housewife
Cook
Cafe owner
Accountant
Computer
Clerk
Housewife
Phys. Therapy
Housewife/farm
Colleg registrr
Dept.Motor Veh.
GroceryStoreOwn
Grocery Checker
Housewife
Nurse's aide
Waitress
Housewife
Phone Co.
RegDirSrDayCare
Housewife
Waitress
Housekeeper
Counseling
Product planner
Artist/writer
Housewife
Secretary
Acct &exec secy
School RN
Practical Nurse
Secretary
Bookkeeper
Secretary
G E Factory
Housewife
Teacher
Accounting
Ran grocery str
Order taker
Phone operator
Nurse
School office
Accountant
Secretary
Pilot
Business
Housewife
Piano teacher
Office Work
Milliner
Housewife
HIGH/MEDIUM ELF EMF EXPOSURE
AD
Occupation
Machinist
MachineMechanic
Machinist
Tailor, Painter
Tool & Die Make
Machinist
Carpenter
Carpenter,barbr
Machinist
Electrician
Carpenter
Seamstress
Seamstress
Elect.assembly
Assembler
Seamstress
Keypunch
CONTROL
Occupation
Machine Operatr
Seamstress
Sewing
Sewing factory
Elec circuit bd
Typist
Seamstress
Clerk typist
Seamstress
Seamstress
Housewife/sewin
Wiringssolderng
Seamstress
Seamstress
Seamstress
Sewing Factory
Machine Operatr
Welding contrac
Electrician
Clerk typist
Dressmaker
Sewing
INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI
films the text directly from the original or copy submitted. Thus, some
thesis and dissertation copies are in typewriter face, while others may be
from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the
copy submitted. Broken or indistinct print, colored or poor quality
illustrations and photographs, print bleedthrough, substandard margins,
and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete
manuscript and there are missing pages, these will be noted. Also, if
unauthorized copyright material had to be removed, a note will indicate
the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand comer and
continuing from left to right in equal sections with small overlaps. Each
original is also photographed in one exposure and is included in reduced
form at the back of the book.
Photographs included in the original manuscript have been reproduced
xerographically in this copy. Higher quality 6” x 9” black and white
photographic prints are available for any photographs or illustrations
appearing in this copy for an additional charge. Contact UMI directly to
order.
UMI
A Bell & Howell Information Company
300 North Zeeb Road, Ann Arbor MI 48106-1346 USA
313/761-4700 800/521-0600

UMI Number: 1378412
Copyright 1995 by
Dunn, Meleana Elizabeth
All rights reserved.
UMI Microform 1378412
Copyright 1996, by UMI Company. All rights reserved.
This microform edition is protected against unauthorized
copying under Title 17, United States Code.
UMI
300 North Zeeb Road
Ann Arbor, MI 48103

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The relationship between fatty acid composition of subcutaneous adipose tissue and the risk of proliferateive benign breast disease and breast cancer

PDF

Rates of cognitive decline using logitudinal neuropsychological measures in Alzheimer's disease

PDF

Polymorphism of CYP2E1 gene and the risk of lung cancer among African-Americans and Caucasians in Los Angeles County

PDF

The Effect Of Non-Sampling Measurement Errors On Relative Risk Estimates Of Coronary Heart Disease In Framingham Women

PDF

Familiality and environmental risk factors of peptic ulcer: A twin study

PDF

Auditory brainstem responses (ABR): variable effects of click polarity on auditory brainstem response, analyses of narrow-band ABR's, explanations

PDF

Cognitive dysfunction and occupations with likely significant magnetic field exposure: A cross-sectional study of elderly Mexican Americans

PDF

A computational model of NMDA receptor dependent and independent long-term potentiation in hippocampal pyramidal neurons

$High-dose chemotherapy followed by autologous hematiopoietic progenitor cell transplantation for relapsed or refractory Hodgkin's disease$

PDF

High-dose chemotherapy followed by autologous hematiopoietic progenitor cell transplantation for relapsed or refractory Hodgkin's disease

PDF

A model for integrating occupational therapy procedures

PDF

A kinetic model of AMPA and NMDA receptors

PDF

Comparison of gene expression of SCG10 and Stathmin/p19 in aging rat brain: an in situ hybridization study

PDF

The effects of dependence among sites in phylogeny reconstruction

PDF

Rationalizing risk: sexual behavior of gay male couples

PDF

The application of Sarbin's theory of emotions as narrative emplotments to stories of two men diagnosed with cancer

PDF

Fine motor skills of two- to three-year-old drug exposed children

PDF

The Hall Canyon pluton: implications for pluton emplacement and for the Mesozoic history of the west-central Panamint Mountains

PDF

A physiologic model of granulopoiesis

PDF

Assistive technology for clients with visual impairments of the Alaska Division of Vocation Rehabilitation: a qualitative evaluation

PDF

The relationship of stress to strain in the damage regime for a brittle solid under compression

Asset Metadata

Creator Dunn, Meleana Elizabeth (author)

Core Title Occupational exposure to extremely low frequency electromagnetic fields as a potential risk factor for Alzheimer's disease

School Graduate School

Degree Master of Science

Degree Program Biometry

Degree Conferral Date 1995-05

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag biology, neuroscience,gerontology,health sciences, public health,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Sobel, Eugene (committee chair), Henderson, Victor (committee member), Mack, Wendy Jean (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-8162

Unique identifier UC11357699

Identifier 1378412.pdf (filename),usctheses-c18-8162 (legacy record id)

Legacy Identifier 1378412-0.pdf

Dmrecord 8162

Document Type Thesis

Rights Dunn, Meleana Elizabeth

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA