University of Southern California Dissertations and Theses

Head injury and dementia: A co-twin control study of Swedish twins

(USC Thesis Other)

Head injury and dementia: A co-twin control study of Swedish twins

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content HEAD INJURY AND DEMENTIA: A CO-TWIN CONTROL
STUDY OF SWEDISH TWINS
by
Michael Gabriel Crowe
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 ARTS
(PSYCHOLOGY)
December 2000
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 1407907
®
UMI
UMI Microform 1407907
Copyright 2002 by ProQuest Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company
300 North Zeeb Road
P.O. Box 1346
Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UNIVERSITY O F S O U T H E R N CALIFORNIA
THE GRADUATE SCHOOL
UNIVERSITY PARK
LOS ANGELES. CALIFORNIA 8 0 0 0 7
This thesis, written by
M i cU ^ e-l /v c\ ____ _____
under the direction of huJk— Thesis Committee,
and approved by all its members, has been pre
sented to and accepted by the Dean of The
Graduate School, in partial fulfillment of the
requirements for the degree of
.c 4 » lc g y ^
Tint* December 18, 2000
THESIS COMMITTEE
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Acknowledgement
Work supported by National Institutes of Health Grants No. R01-AG08724, AG04563,
AG10175 and AG08861.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table of Contents
Acknowledgement........................................................................ii
List of Tables...............................................................................iv
Abstract........................................................................................ vi
Introduction................................................................................... 1
Methods........................................................................................15
Results......................................................................................... 24
Discussion................................................................................... 37
References................................................................................... 45
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
iv
List of Tables
Table Page
1 Demographics............................................................................................. 18
2 Frequency of risk factors for probands and partners, for all
dementias combined and for Alzheimer’s alone, using combined
data from both twin samples.....................................................................24
3 Frequency of risk factors for probands and partners, for all
dementias combined and for Alzheimer’s alone, Study of
Dementia in Swedish Twins.....................................................................25
4 Frequency of risk factors for probands and partners, for all
dementias combined and for Alzheimer’s alone,
OCTO-twin study..................................................................................... 26
5 Risk estimates based on pairs who were dementia and
exposure discordant, all dementias, combined data from
both twin samples..................................................................................... 27
6 Risk estimates based on pairs who were dementia and
exposure discordant, all dementias, Study of Dementia in
Swedish Twins...........................................................................................27
7 Risk estimates based on pairs who were dementia and
exposure discordant, all dementias, OCTO-twin study......................... 28
8 Risk estimates based on pairs who were Alzheimer’s and
exposure discordant, combined data from both twin samples...............28
9 Risk estimates based on pairs who were Alzheimer’s and
exposure discordant, Study of Dementia in Swedish Twins.................29
10 Risk estimates based on pairs who were Alzheimer’s and
exposure discordant, OCTO-twin study..................................................29
11 Risk estimates, based on pairs who were dementia and
exposure discordant, all dementias, combined data from both twin
samples, stratified by gender and family history of dementia...............31
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
V
List of Tables Continued
Table Page
12 Risk estimates, based on pairs who were dementia and
exposure discordant, all dementias, Study of Dementia in
Swedish Twins, stratified by gender and family
history of dementia................................................................................... 32
13 Risk estimates, based on pairs who were dementia and
exposure discordant, all dementias, OCTO-twin study,
stratified by gender and family history of dementia...............................33
14 Risk estimates, based on pairs who were Alzheimer’s and
exposure discordant, combined data from both twin samples,
stratified by gender and family history of dementia...............................34
15 Risk estimates, based on pairs who were Alzheimer’s and
exposure discordant, Study of Dementia in Swedish Twins,
stratified by gender and family history of dementia...............................35
16 Risk estimates, based on pairs who were Alzheimer’s and
exposure discordant, OCTO-twin study, stratified by gender
and family history of dementia.................................................................36
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
vi
Abstract
History of head injury was examined as a risk factor for dementia and
Alzheimer’s disease using 167 same-sex twin pairs discordant for dementia in a co-twin
control design. In addition to head injury with loss of consciousness, this study examined
milder head injury, participation in contact sports, workplace use of vibrating equipment,
and APOE-e4, a genetic risk factor. APOE-e4 did appear to be a risk factor for
Alzheimer’s disease, as expected. None of the other factors examined were associated
with a significant increase in the risk of Alzheimer’s disease or dementia. The relative
risk for head injury with loss of consciousness was 1.2 (95 % Cl 0.5-2.8) for dementia
and 1.0 (95% Cl 0.3-3.4) for Alzheimer’s. History of playing contact sports was found to
have a significant inverse relationship to dementia for males. Results indicate that an
association between head injury and Alzheimer’s disease is not found when cases and
controls are well-matched.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1
Introduction
Alzheimer’s disease is a tremendous public health problem. It is the major cause
of dementia in older adults. Although a variety of prevalence estimates for dementia and
for Alzheimer’s disease have been reported, the finding that prevalence increases
dramatically after age 60 has been found consistently. For example, a meta-analysis by
Ritchie, Kildea, and Robine (1992) found the prevalence of dementia to double every 5.7
years between the ages of 60 and 90. Conservative rates estimate that about 2.5 million
Americans currently have dementia (Graves & Kukull, 1994). If Alzheimer’s disease
accounts for around 60% of all dementias, then approximately 1.5 million of these have
Alzheimer’s (Graves & Kukull, 1994).
Increasing longevity suggests that the prevalence of Alzheimer’s disease will
continue to rise over the next century. Of the entire population, the 85 and older age
group is both the fastest growing group of people and is the one at most risk for
developing Alzheimer’s disease (Khachaturian & Radebaugh, 1998).
The emotional and physical costs of Alzheimer’s disease to individuals and
families are incalculable, while the financial cost to society is close to $100 billion
annually (Rice, 1991). It is believed that roughly half of the elderly living in nursing
homes have Alzheimer’s disease, contributing substantially to this exorbitant cost (U.S.
Congress, 1987).
With no known cure for Alzheimer’s disease, much research has focused on
finding ways to shift the incidence curve to the later ages. It has been hypothesized that
by delaying onset of the disease by 5 years, the number of people with the disease can be
reduced by 50% (Khachaturian & Radebaugh, 1998). In the search to find ways to delay
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2
onset, studies of estrogen replacement therapy (Henderson, Paganini-Hill, Emanuel,
Dunn, & Buckwalter, 1994) and nonsteroidal antiinflammatory drugs (Breitner et al.,
1994) have been promising. Drugs recently approved for the treatment of Alzheimer’s
are thought to be more effective if initiated as early as possible (Khachaturian &
Radebaugh, 1998).
The task of initiating treatments for Alzheimer’s at an early stage, however, is not
as simple as it may seem. The degenerative process of Alzheimer’s disease may begin as
long as 40 years before cognitive difficulties are detected, so it is likely that too much
damage has occurred by the time of detection for treatments to slow the process
(Khachaturian & Radebaugh, 1998). Since detection is never as early as the disease
begins, preventive strategies to delay onset are warranted. Additionally, if it can be
established that some people are at greater risk based on certain risk profiles, then
preventive strategies to delay onset may be tailored to them.
Developing risk profiles, however, may still be far away. Understanding the
mechanism through which risk factors influence the development of Alzheimer’s and
interact with other risk factors is likely even farther away. It is now widely believed that
the etiology of Alzheimer’s disease is influenced by multiple genes and environmental
risk factors. These multiple risk factors may act independently or synergistically,
complicating the interpretation of any single risk factor. This obviates the need for
studies that take multiple risk factors into account.
Much of what we do know about these risk factors comes from case-control
studies. Family history of dementia is one of few risk factors that has received a great
deal of support. Family history of dementia has been thought of as a marker of genetic
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3
susceptibility (Mayeux et al, 1993). APOE is the most important susceptibility gene for
Alzheimer’s yet found. Environmental risk factors with the most consistent support are
history of head injury, low education, and poor socioeconomic status; these have been
categorized as “probable” risk factors for Alzheimer’s disease (Mortimer, 1995). Other
possible risk factors include history of depression and hypothyroidism (Mortimer, 1995),
although depression may instead be a prodromal change associated with the dementing
illness (Wetherell, Gatz, Johansson, & Pedersen, 1999).
Head Injury
Head injury is often cited as one of the few empirically validated environmental
risk factors for Alzheimer’s disease. Since 1982, more than 35 studies have been
published on the relationship between head injury and Alzheimer’s disease or dementia.
Unfortunately, this body of literature has been plagued by inconsistencies in both
methodologies and findings. Contrary to popular claims, no clear picture has emerged as
to whether head injury is a significant risk factor for Alzheimer’s disease.
Five studies have found a significant positive association between head injury
with loss of consciousness and Alzheimer’s disease (Graves et al., 1990; Guo et al., 2000;
Mayeux et al., 1993; O’Meara et al., 1997; Schofield et al., 1997). Only one of these
studies used a matched-pair case control design (Graves et al., 1990). In addition to these
five studies, two studies found significant risk for men only (Plassman et al., 2000; Salib
& Hillier, 1997), another found significant risk when head injury occurred within ten
years of dementia onset (van Duijn et al., 1992), and one more found elevated risk for
loss of consciousness greater than five minutes (Schofield et al., 1997). Three other
studies found significant risk due to head injury, but either the severity of head injury was
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4
unknown (Heyman et al., 1984; Kondo & Yamashita, 1990), or injuries were not
restricted to loss of consciousness (Rasmusson et al., 1995). This evidence would be
more compelling if not for the fact that nine studies have reported relative risks of 1.0 or
less for head injury (Breteler, de Groot, van Romunde, & Hofman, 1995; Fratiglioni,
Ahlbom, Viitanen, & Winblad B., 1993; Katzman et al., 1989; Launer et al., 1999; Li et
al., 1992; Mehta et al., 1999; Mendez et al., 1992; Soininen & Heinonen, 1982; Williams
et al., 1991).
The odds ratios reported in these studies have varied widely, ranging from 0.3
(Fratiglioni et al., 1993) to 18.0 (Kondo & Yamashita, 1990). This variability appears to
be due, at least in part, to problems in methodological design. Studies that found the
largest odds ratios of 18.0 (Kondo & Yamashita, 1990), 13.8 (Rasmusson et al., 1995),
and 9.9 (Guo et al., 2000) did not pair-match their cases and controls for age or gender.
Similarly, studies that found the smallest odds ratios of 0.3 (Fratiglioni et al., 1993) and
0.6 (Soininen & Heinonen, 1982) did not use a matched-pair design.
Studies that used a matched-pair design showed much less variability in their
reported odds ratios, ranging from 1.3 (Broe et al., 1990) to 6.0 (Chandra, Philipose, Bell,
Lazaroff, & Schoenberg, 1987). In fact, of the nine published studies that used a
matched-pair design, six reported odds ratios between 1.0 and 2.0 (Amaducci et al., 1986;
Broe et al., 1990; Chandra, Kokmen, Schoenberg, & Beard, 1989; Hofman et al., 1989;
Kokmen, Chandra, & Schoenberg, 1988; van Duijin et al., 1992). This suggests that
studies choosing to statistically control instead of match for variables such as age and
gender have wildly different estimates of relative risk, while pair-matched studies are
highly consistent and much closer to an odds ratio of 1.0.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5
One would expect that the rate of head injury reported in control subjects would
be somewhat consistent, however, this rate has also varied widely, ranging from 1%
(Chandra et al., 1999; Guo et al., 2000; Li et al., 1992) to more than 10% (Canadian
Study of Health and Aging, 1994; Mehta et al., 1999; Mendez et al., 1992; Mortimer,
French, Hutton, & Schuman, 1985; Soininen & Heinonen, 1982). It is interesting that for
all of the studies that found a significant association between head injury and
Alzheimer’s disease, rate of head injury in controls was no more than 6%. Studies that
found significant results may have underestimated head injury in control subjects. In
fact, these rates are substantially lower than the lifetime cumulative incidence of head
injuries reported by Annegers and colleagues (1980) of 20% for males and 8% for
females from Minnesota. In Taiwan, estimates were even higher, with a cumulative
incidence of 28% for males and 14% for females (Hung, Chiu, Tsai, Laporte, & Shih,
1990). An underestimation of head injury for controls would be consistent with the
suggestion that a differential recall bias may account for the observed association
between Alzheimer’s disease and head injury. Proxy informants may be more likely to
recall head injuries for demented than intact study participants.
In an attempt to obtain a more accurate and bias-free estimate of head injury,
some studies have used medical records or hospital admission records. All but one
(Plassman et al., 2000) of these studies have failed to find a significant positive
association between head injury and Alzheimer’s disease (Chandra et al., 1987; Kokmen
et al., 1988; Mendez et al., 1992). Using medical records, Mendez and colleagues found
a history of head injury for 18% of control subjects (1992). This was the highest rate of
head injury reported for controls in any of the published studies of head injury and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Alzheimer’s disease and was closer to the rates found in epidemiological studies of head
injury. However, another study using medical records found history of head injury in
only 1% of controls (Chandra et al., 1989) and, similarly, a review of hospital admissions
yielded a rate of 2% for controls (Kokmen et al., 1988). These latter two studies are
consistent with findings that hospital-based head injury data may be incomplete and may
contain substantial biases (Fife, 1987). It has been found that, among those who
sustained head injury leading to disability, those not admitted to hospital included one-
half of those with three to seven days of bed disability and one-third of those with more
than seven days of bed disability (Fife, 1987). Thus, use of medical records may be even
less complete than retrospective proxy or self-reports of history of head injury. However,
these studies still eliminate any selection bias or differential recall bias.
Selection bias can arise from the use of prevalent dementia cases. For example,
use of prevalent cases may lead to associations that relate to predictors of survival rather
than to the risk of Alzheimer’s or dementia (Fratiglioni et al., 1993). Two studies have
used both a hospital and a community control sample, and both found a stronger
association between head injury and Alzheimer’s disease when comparing cases to
hospital rather than community controls (Amaducci et al., 1986; Mortimer et al., 1985).
These studies highlight the importance of careful selection of controls.
Potential problems having to do with survival, selection bias, and differential
recall bias can be reduced by incidence studies. In these studies, all participants are
selected in the same manner and risk factor information is collected prior to dementia
onset. These studies have predominantly failed to confirm any association between head
injury and Alzheimer’s disease or dementia. After a post-hoc restriction of head injuries
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7
to those involving loss of consciousness for more than five minutes, one incidence study
found a significant relationship between head injury and Alzheimer’s disease (Schofield
et al., 1997). Otherwise, four incidence studies have failed to support head injury as a
risk factor for Alzheimer’s disease (Katzman et al., 1989; Launer et al., 1999; Mehta et
al., 1999; Williams et al., 1991). The contrast between prevalence and incidence designs
would tend to support the idea that various biases are at work in the studies using
prevalent cases.
Another oft cited problem with the head injury literature is the lack of statistical
power in many of the studies. To increase power, data were pooled for seven pair-
matched case-control studies of head injury with loss of consciousness in a meta-analysis
(Mortimer et al., 1991). Head injury was associated with a significant increase in risk of
Alzheimer’s disease and an odds ratio of 1.8 was found. However, if the studies included
had any consistent directional bias, this bias would have been inflated by the meta
analysis and could have resulted in an overestimation of the association between head
injury and Alzheimer’s.
In fact, there does appear to be a directional bias in favor of finding larger odds
ratios for head injury in pair-matched studies with smaller sample sizes. All four pair-
matched studies with sample sizes of less than 150 pairs found odds ratios of 2.0 or more
for head injury. In contrast, studies that used more than 150 matched pairs have been
remarkably consistent in the odds ratios they have reported. Four of five studies found
the same odds ratio of 1.3 (Broe et al., 1990; Chandra et al., 1989; Hofman et al., 1989;
van Duijin et al., 1992), while the fifth found an odds ratio of 1.4 (Kokmen et al., 1988).
None of the larger sample odds ratios were statistically significant. This is curious
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
because one of the five pair-matched studies with smaller sample sizes found a
statistically significant association (Graves et al., 1990). If the problem with these studies
was truly a lack of power, it would be expected that the larger studies would be able to
detect a significant association and not vice-versa.
Most studies have looked at head injury dichotomously, with a loss of
consciousness as the criterion for defining head injury as a risk factor for Alzheimer’s
disease. Studies that have looked at different levels of severity have found every possible
result: a positive association between increasing severity of head injury and risk of
Alzheimer’s (Mayeux et al., 1995; Schofield et al., 1997), no relationship between head
injury severity and risk of Alzheimer’s (Williams et al., 1991), and greater risk for head
injury without loss of consciousness (Graves et al., 1990; Rasmusson et al., 1995). One
unmatched case-control study examined history of head injury with or without loss of
consciousness and found significant risk for Alzheimer’s disease only in males (Salib &
Hillier, 1997). The poor design of the latter study and variability of results in the former
studies necessitates further research on milder head injuries that do not involve a loss of
consciousness.
While the existence of a relationship between head injury and Alzheimer’s
disease is very much debatable, there are undeniable physiological similarities between
head injury and Alzheimer’s neuropathology, primarily involving the neurofibrillary
tangles and senile plaques found in the brains of people affected by Alzheimer’s disease.
Roberts et al. (1991) found B-amyloid protein deposits and neurofibrillary tangles in the
brains of professional and amateur boxers who were demented (with dementia pugilistica,
or “punch-drunk” syndrome) before death and in individuals who died following severe
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9
head injury. It is estimated that 30-50 percent of patients with fatal head injuries show B-
amyloid deposition in the brain, interpreted as an acute-phase response to neuronal injury
(Graham, Gentleman, Lynch, & Roberts, 1995; Roberts et al., 1991; Roberts, Gentleman,
Lynch, Murray, Landon, & Graham, 1994). These observations are consistent with the
theory that B-amyloid deposition is involved in the pathogenesis of Alzheimer’s disease
because this deposition is believed to be a main step in a neurotoxic cascade leading to
senile neuritic plaques and neuronal death (Selkoe, 1993).
A chronic inflammatory process has also been suggested as playing a role in the
pathogenesis of Alzheimer’s disease (Gordon, 1993; Kalaria, 1993; McGeer & Rogers,
1992). There was evidence of an inflammatory response detected in B-amyloid protein
deposits in the brains of Alzheimer’s patients, which may be involved in neuronal death
(McGeer, Rogers, & McGeer, 1994). Inflammation following head injury may have
similar effects (O’Meara et al., 1997). Also, the use of nonsteroidal antiinflammatory
drugs has been linked to delayed age of onset and lowered rate of decline for Alzheimer’s
disease (Breitner et al., 1994; Breitner et al., 1995; Rogers et al., 1993). These findings
lend further support to the belief that head injury and Alzheimer’s disease have very
similar pathogenic mechanisms.
There is little human data on the neuropathological consequences of milder head
injuries. However, there is evidence that neuropsychological abnormalities following
mild head injury may remain for weeks to months afterward (Barth et al., 1983; Bohnen,
Jolles, & Twijnstra, 1992; Gronwall & Wrightson, 1974). One study has found that
evoked potential abnormalities following mild head injury can last more than six weeks
(Montgomery, Fenton, McClelland, MacFlynn, & Rutherford, 1991). These findings
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10
have been interpreted as suggesting that mild to moderate head injury can cause lasting
structural brain changes (Schofield et al., 1997).
There have been several studies of neuropsychological consequences of milder
head injuries resulting from playing contact sports. Neuropsychological impairment has
been observed in retired and active amateur and professional soccer players ( Matser,
Kessels, Lezak, Jordan & Troost, 1999; Matser, Kessels, Jordan, Lezak, & Troost, 1998;
Sortland & Tysvaer, 1989; Tysvaer & Lochen, 1991; Tysvaer & Storli, 1981; Tysvaer,
Storli, & Bachen, 1989), as well as in American college football players (Collins et al.,
1999). However, no studies have found an association between playing contact sports
and subsequent risk of dementia.
APOE-e4
One of the strongest and most widely studied genetic risk factors for Alzheimer’s
disease is the APOE-e4 allele. Every person inherits two APOE genes, or alleles, one
from the mother and one from the father, which are found on chromosome 19. There are
three major variations of the APOE gene, known as the e2, e3, and e4 alleles. The e3
allele is the most common, with a frequency of approximately 78% in the population.
The e4 allele has a frequency of approximately 15%, while e2 is the least common, with
an allele frequency of around 7% (Roses, 1998). However, these estimates of allele
frequency have been found to vary according to locality and ethnicity (Reynolds,
Wetherell, & Gatz, 1999).
The elevated risk for Alzheimer’s disease associated with the APOE-e4 allele has
been established through case-control studies, finding approximately 30% to 40% greater
allelic frequencies in Alzheimer’s disease cases compared with controls (Reynolds et al.,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
11
1999). There is a dosage effect. The risk associated with APOE-e4 homozygosity (two
e4 alleles) is approximately 10-fold compared to those with no e4 (Mayeux et al., 1995).
In contrast, the APOE-e2 allele is associated with lowered risk for Alzheimer’s disease
(Roses, 1998).
A previous study of APOE-e4 and Alzheimer’s disease used 13 dizygotic twin
pairs from the Study of Dementia in Swedish Twins and 13 sibling pairs discordant for
Alzheimer’s disease. Allelic frequencies of APOE-e4 were found to be significantly
greater in probands than cognitively intact siblings (Lannfelt et al., 1995).
In addition to susceptibility to Alzheimer’s disease, APOE-e4 has also been
linked to age of onset and rate of decline. Although some studies have found APOE-e4
to be related to an earlier age of onset (e.g. Benjamin et al., 1996; Hyman et al., 1996),
not all studies have arrived at this conclusion (e.g. Farrer et al., 1995; Mayeux et al.,
1993). Similarly, studies relating APOE-e4 to rate of decline have been inconclusive.
Further research is warranted in making the connection between APOE and these
Alzheimer’s disease variables.
While research on APOE has been fruitful in understanding Alzheimer’s disease,
the majority of cases of Alzheimer’s are not attributable solely to APOE status. Factors
other than APOE-e4 must be examined because there are some people with no APOE-e4
allele who develop Alzheimer’s, and there are some people with an APOE-e4 allele who
never develop Alzheimer’s disease. This observation leads to a search for other genes,
for environmental risk factors, and also for interactions between genes and environmental
risk factors.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
12
Multifactorial Findings
The relationship between Alzheimer’s disease and head injury may be moderated
by APOE genotype. A case-control study by Mayeux et al. (1995) found such a
relationship between APOE and head injury with loss of consciousness. In a community
based sample of 236 elderly, Mayeux et al. (1995) found the effect of head injury on the
risk of Alzheimer’s disease to be dependent upon APOE. Only those individuals with
APOE genotypes containing one or more copies of the e4 allele were at increased risk
following head injury in comparison with persons with no head injury and no APOE-e4.
Risk of Alzheimer’s disease was increased 10-fold by having both APOE-e4 and a
history of traumatic head injury (adjusting for age and education, odds ratio (OR) = 10.2,
95% confidence interval (Cl) 1.2-89.0), compared to a 2-fold increase in risk of
Alzheimer’s disease with APOE-e4 alone (OR = 2.0, 95% Cl 1.1-3.5). Head injury
without an APOE-e4 allele did not increase risk in this study (OR = 1.0, 95% Cl 0.3-3.2).
In contrast, O’Meara and her colleagues (1997) obtained different results in their
examination of head injury and APOE. Using a sample of 349 Alzheimer’s patients and
342 controls from an HMO in Seattle, no significant variation was found in the head
injury-Alzheimer’s disease risk relation by APOE-e4 genotype. Head injury with loss of
consciousness was associated with increased risk of Alzheimer’s disease (OR = 2.1, 95%
Cl 1.1-3.8), while APOE-e4 (OR = 4.1, 95% Cl 2.9-6.0) was found to be an “independent
risk factor which neither modified nor confounded the association” (O’Meara et al.,
1997). Mehta et al. (1999) also failed to replicate the findings of the Mayeux et al.
(1995) study.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
In a study of APOE-e4 genotype and functional outcome in traumatic head injury
survivors, there was a strong association found between the APOE-e4 allele and clinical
outcome (Friedman et al., 1999). After controlling for age and time of unconsciousness,
the OR of a suboptimal outcome (fair or unfavorable) was 13.9 (95 % Cl 1.4-134.0).
Another study looking at APOE-e4 genotype and head injuries in boxers found evidence
that the APOE-e4 allele may be associated with increased severity of chronic neurologic
deficits (Jordan et al., 1997).
If, in fact, APOE-e4 does modify the effect of head injury on the production of
Alzheimer’s disease, there is biological evidence that would help to explain this
interaction. Products of the APOE gene found in the brain may be involved in neuronal
branching (Nathan, Bellosta, & Sanan, 1994) and are seen in characteristic Alzheimer’s
plaques and tangles (Han et al., 1994; Namba, Tomonaga, Kawasaki, Otomo, & Ikeda,
1991). In addition, APOE-e4 has been found to bind readily with B-amyloid protein in
vitro (Wisniewski, Golabek, Matsubara, Ghiso, & Frangione, 1993).
Family history of dementia and gender may also modify the association between
head injury and Alzheimer’s disease. A stronger association between head injury and
Alzheimer’s has been found for those without a positive family history of dementia, but
this association was not significantly different from those with a family history of
dementia (Mortimer et al., 1991). Other studies have found no evidence that family
history modifies the risk due to head injury (Launer et al., 1999; Van Duijn et al., 1992).
Studies looking at gender have found history of head injury to increase risk for AD only
in men (Mortimer et al., 1991; O’Meara et al., 1997; Salib & Hillier, 1997) or only in
women (Mayeux et al., 1993).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14
Twin Studies
Given the complexity of understanding risk factors and their role in Alzheimer’s
disease and dementia etiology, twin studies can contribute to sorting out this puzzle. The
co-twin control method is a matched pair analysis that takes twin pairs who are
discordant for the disease to find whether the twin who was exposed to a given risk factor
is more often the diseased twin. The major advantage of this design is that matching of
pairs is much better than for any other type of case-control study. Identical twins share
100% of their genes and fraternal twins share 50% of their segregating genes, thus this
design controls for some genetics. In addition, cases and controls are identical with
respect to age, gender (for-like sexed pairs), and family history of dementia, in addition
to having similar environmental histories. To date, there has been only one twin study to
look at head injury and Alzheimer’s disease (Nee & Lippa, 1999). Using a sample of 10
twin pairs discordant for Alzheimer’s disease, there was no association between head
injury and Alzheimer’s disease.
The present study utilized the co-twin control method in order to examine history
of head injury with and without loss of consciousness and APOE-e4 genotype as risk
factors for dementia in general and for Alzheimer’s disease, specifically. History of
participation in contact sports and workplace use of vibrating equipment were also
examined as risk factors. Whether family history of dementia or gender modified the
estimated relative risks for these factors was examined through stratified analyses. This
is the first large twin study of head injury and dementia. While there have been a few
matched-pair studies of comparable sample size, none have attempted to match for family
history of dementia. Also, these studies were not able to obtain exact matching for age.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
15
Methods
Sample
This study used data from the Study of Dementia in Swedish Twins (Gatz et al.,
1997), which includes all demented twins and their partners assessed from the Swedish
Adoption/Twin Study of Aging, or SATSA (Pedersen et al., 1991). Data from the
OCTO-Twin Study (Berg et al., 1992) were also used, again taking all pairs from OCTO-
Twin in which one or both was demented. Procedures parallel to those used in the Study
of Dementia in Swedish Twins were used to collect risk factor information and obtain
diagnoses for the OCTO-Twin dementia study.
The SATSA sample was taken from the Swedish Twin Registry, a population-
based registry of more than 25,000 same-sex twin pairs bom between 1886 and 1958.
Those eligible for SATSA included all 961 twin pairs from the registry who were reared
apart and a matched group of 961 pairs, also from the registry, who were reared together
and matched for sex, year of birth, and county of birth to the reared-apart twins
(Pedersen, Friberg, Floderus-Myrhed, & McCleam, 1984). The SATSA sample has been
shown to have similar health and sociological characteristics compared to a cross section
of Swedish cohorts. The OCTO-twin study also consists of people taken from the
Swedish Twin Registry. All twin pairs in the registry who were both alive in 1992 and
born before 1913 were included in the OCTO-twin study.
The Study of Dementia in Swedish Twins included all SATSA members
suspected of dementia and their twin partners. Anyone in the SATSA sample born 1935
or earlier and still alive in 1987 was part of the dementia study (N=1978). Either the
Mini-Mental State Examination (MMSE; Folstein, Folstein, & McHugh, 1975) or a
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
16
telephone screening protocol (Gatz, Reynolds, Nikolic, Lowe, Karel, & Pedersen, 1995)
was used to identify dementia suspects.
Those suspected of dementia from the screening and their twin partners
underwent a thorough diagnostic assessment that paralleled the procedures from
Consortium to Establish a Registry for Alzheimer’s Disease (CERAD; Morris et al.,
1989) for physical and neurological evaluations, laboratory tests, neuropsychological
testing, and neuroimaging. A mobile assessment team, comprised of a psychologist and a
nurse, conducted an informant interview involving memory and cognition and performed
a neuropsychological test battery for both the suspected twins, or probands, and their
partners. The neuropsychological battery, administered by a psychologist, tested learning
and memory, verbal performance, visuospatial skills, attention, motor performance, and
executive functioning. A physician visited for purposes of medical and neurological
evaluation (Gatz et al., 1997).
A consensus diagnostic conference, attended by the nurse, psychologist, and
physician trained in geriatric medicine, determined the final clinical diagnosis. The chair
of this conference was blind to zygosity and to any information other than that presented
at the conference in order to minimize bias of diagnostic outcome. During the
assessment, if it was learned that the individual was not demented but instead had some
other condition (e.g. psychiatric disorder or mental retardation), the evaluation was
discontinued and the pair was excluded (Gatz et al., 1997).
Clinical dementia diagnoses were based on convergent criteria from the DSM-III-
R criteria for dementia syndrome and vascular dementia, NINCDS-ADRDA (McKhann
et al., 1984) criteria for probable and possible Alzheimer’s disease, and NINDS-AIREN
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
17
(Roman et al., 1993) criteria for vascular dementia. Longitudinal follow-up and autopsy
have been used to confirm these diagnoses. Autopsy confirmation of diagnosis is
available for 24 cases from the Study of Dementia in Swedish Twins. There was
concurrence between clinical diagnosis and neuropathological results for 21 of these 24
cases (Gatz, Reynolds, John, Johansson, Mortimer, & Pedersen, under review). This is
within the range of 80-90% that has been reported for rate of confirmation of clinical
diagnosis by autopsy neuropathology (Roses, 1998). For two of these cases, the
neuropathology supported a diagnosis of Alzheimer’s disease, while the clinical
diagnosis was vascular dementia for one case and dementia not specified for the other.
For the third case disagreement, the neuropathology supported a diagnosis of mixed
dementia, while the clinical diagnosis was vascular dementia. There were no false-
positive Alzheimer’s disease diagnoses.
Age of dementia onset was the earliest age at which dementia symptoms become
apparent. Onset was established by asking the informant for a description of memory and
other problems, followed by the age at which these changes were observed (Bayles,
1991). Medical records were used to crosscheck the information when available. Pairs
were classified as concordant for Alzheimer’s disease if the proband’s partner had
Alzheimer’s disease or mixed Alzheimer’s and vascular pathology or unspecified
dementia. Pairs concordant for dementia were excluded from the analyses because the
co-twin control design is based on discordant pairs. Twin pairs in which the partner had
dementia due to a specified cause (e.g. hydrocephalus, Parkinson’s disease) were also
excluded from the sample.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
18
A pair was classified as discordant if the partner remained cognitively intact for a
certain period of time after the proband’s dementia onset. This period of time ranged
from 0 to 6 years (see Gatz, Pedersen, Crowe, & Fiske (2000) for a discussion of
definitions of discordancy), and was determined by a sliding scale based on the survival
curves developed by Posner, Pedersen, and Gatz (1999). The sliding scale required
survival for progressively fewer years as age of onset increased. The partner was simply
required to be alive and non-demented at the time that the proband was diagnosed at the
upper end of the sliding scale, in pairs where the proband’s age of onset was 75 or older.
Discordant Pairs
A total of 167 twin pairs were discordant for dementia. Descriptive information
for these 334 individuals is shown in Table 1. Of these 167 pairs, 102 were discordant
for Alzheimer’s disease.
Table 1: Demographics
Combined data Study of Dementia OCTO-Twin
in Swedish Twins
Pairs % Pairs % Pairs %
Sex
Male 53 31.7 23 33.3 30 30.6
Female 114 68.3 46 66.7 68 69.4
Zygosity
Monozygotic 54 32.3 23 33.3 31 31.6
Dizygotic 113 67.7 46 66.7 67 68.4
Family history
Positive 51 42.9 24 55.8 27 35.5
Negative 68 57.1 19 44.2 49 64.5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
19
Risk Factors
Risk factor information, including history of head injury, history of participation
in contact sports, occupational history, and family history of dementia was collected
through a detailed structured interview. Informants (usually spouse or adult offspring)
for both probands and intact partners, as well as self-report interviews of intact partners,
provided this information. If available, medical records were also used to provide
additional risk factor information. Information on workplace use of vibrating equipment
and APOE genotype was available only for the Study of Dementia in Swedish Twins
sample.
For the purposes of this study, head injuries were examined at three levels: l)Any
head injury; 2)mild head injury; and 3)head injury with loss of consciousness. Mild head
injury was defined as a report of head injury with concussion but no loss of
consciousness. Risk factor informants were asked whether the individual ever had a
severe head-injury including concussion. Subsequent information was collected on age
of head injury, loss of consciousness, memory loss, and hospitalization. This information
was collected for up to three head injury episodes. Of the total of 167 discordant pairs,
131 twin pairs, 56 from the Study of Dementia in Swedish Twins and 75 from OCTO-
twin, had complete head injury information. Of the 102 twin pairs discordant for
Alzheimer’s disease, 84 twin pairs, 43 from the Study of Dementia in Swedish Twins and
41 from OCTO-twin had complete head injury information.
Medical records were used to cross check history of head injury for some of the
people in the Study of Dementia in Swedish Twins. For intact twin partners who self-
reported head injury information, there was 100% agreement with medical records
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
20
(N=21). For demented probands who had information gathered from an informant, there
was 93% agreement with medical records (N=29). There was only one case where the
informant indicated no prior history of head injury and the medical records disagreed.
This disagreement involved a mild head injury, one that did not involve loss of
consciousness.
As a proxy measure of multiple mild head injuries, information on history of
playing contact sports was collected. Informants were asked whether the individual
participated in contact sports, such as boxing, soccer, or hockey. As another proxy
measure of mild head injury, informants were asked whether the individual ever worked
with tools or heavy equipment that resulted in vibration or shaking, such as a
jackhammer. A total of 133 pairs discordant for dementia and 85 pairs discordant for
Alzheimer’s had complete information on contact sports. Information on workplace use
of vibrating equipment was collected for 58 pairs discordant for dementia and 44 pairs
discordant for Alzheimer’s disease from the Study of Dementia in Swedish Twins.
Family history of dementia was classified as positive if at least one first-degree
relative had dementia. If it was known that no first-degree relatives had dementia, family
history of dementia was classified as negative. For the Study of Dementia in Swedish
Twins, but not for OCTO-Twin, there was also a more stringent measure of family
history, in which cases were not classified as negative unless it was known that at least
one parent had survived to age 65 dementia-free. Because a comparable measure was not
available from OCTO-Twin, the coding that was comparable for both was used.
Standard procedures for preparing DNA from peripheral blood were used, and APOE
genotypes were determined as previously described by Wenham,, Price, & Blundell
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
21
(1991). Only dizygotic pairs were informative for the APOE analysis, as monozygotic
pairs would be identical for APOE.
Analyses
The main method of analysis was the co-twin control method. By using only twin
pairs discordant for the disorder, this type of case-control matching controlled for age,
gender, and family history of dementia. Thus, statistical adjustment to control for these
variables was not necessary. The odds ratio was used to estimate relative risks
attributable to the factors under investigation. This measure of association is the standard
estimate of relative risk for rare diseases, such as Alzheimer’s disease.
The odds ratio represents the likelihood of disease in an exposed versus an
unexposed group (Schlesselman, 1982). The reference number for the odds ratio is 1.0,
and represents no association between risk factor and disease. Size and direction of
associations are determined by the extent to which they differ from the 1.0 reference
point. Odds ratios greater than 1.0 are indicative of positive associations, while those less
than 1.0 are negative associations. If the 95% confidence interval for an odds ratio does
not include 1.0, then the association is statistically significant with a 5% chance that the
association is due to Type I error. Effect size is explicit because the odds ratio tells how
many times more (or less) likely the disease occurs in the exposed group. For example,
an odds ratio of 5.0 would indicate that the risk of disease in those exposed is five times
greater than risk of disease in those not exposed. To determine proportion of risk
attributable to a risk factor for a disease population, the etiologic fraction may be
calculated. This proportion is a function of the relative risk and the proportion of
exposed individuals (Schlesselman, 1982). Using the etiological fraction, one can
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
22
estimate what percentage of the cases of disease would be eliminated by removal of the
risk factor.
To obtain the maximum likelihood estimate of the odds ratio for the pair-matched
sample, only pairs discordant for both disease and exposure are used. The maximum
likelihood estimate of the odds ratio was calculated by dividing the number of discordant
pairs in which the proband had a history of exposure and the partner did not by the
number of pairs in which the partner had a history of exposure and the proband did not.
The Mantel-Haenszel method was used to calculate statistical significance and
confidence intervals (Schlesselman, 1982).
Separate analyses were performed for all head injuries combined, mild head
injury, and head injury with loss of consciousness, as well as for participation in contact
sports, workplace use of vibrating equipment, and APOE-e4. Data were stratified by sex
and family history of dementia to examine their effects on these associations.
Odds ratios were calculated both for Alzheimer’s disease and all dementias
combined in order to examine risk specificity. In addition, separate analyses were done
for data from the Study of Dementia in Swedish Twins and data from the OCTO-twin
study. While both are samples from the Swedish Twin Registry, the Study of Dementia
in Swedish Twins and the OCTO-twin study are comprised of different subsets of the
twin register. The two samples were very similar with respect to gender and zygosity
proportions (Table 1), but differed in proportion of family history of dementia and
dementia diagnoses (Tables 2-4). In the Study of Dementia in Swedish Twins,
Alzheimer’s disease accounted for 77% of those with dementia; Alzheimer’s disease
accounted for 55% of all dementias in the OCTO-twin study.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
23
The power to detect a statistically significant odds ratio depends on sample size,
size of the odds ratio, and the base rate of exposure among controls. With a type I error
fixed at 0.05, a type II error fixed at 0.10, and an exposure rate of 0.10 among the
controls, the smallest odds ratio that would be detected in a total sample of 170 matched
pairs would be 3.0 (Broe et al., 1990). This sample consisted of 167 matched pairs with
an exposure base rate of 0.12 for head injury with loss of consciousness. Thus, odds
ratios smaller than 3.0 would not be expected to reach statistical significance. As a
measure of effect size, which has implications for clinical significance, the etiological
fraction was calculated.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
24
Results
Frequencies of risk factors for twin pairs discordant for dementia and
Alzheimer’s disease are shown for combined data (Table 2), for the Study of Dementia in
Swedish Twins only (Table 3), and for the OCTO-twins only (Table 4). Report of any
head injury was more frequent for the intact partners than for the probands in the SATSA
sample and slightly more frequent for the probands than for the intact partners in the
OCTO sample. Rates of head injury including loss of consciousness were very similar
for the two samples. There was a higher frequency in reporting of contact sports for the
intact partners compared to probands from SATSA, while rates were similar for proband
and intact partner in OCTO.
Table 2: Frequency of risk factors for probands and partners, for all dementias combined
and for Alzheimer’s alone, using combined data from both twin samples
All dementias Alzheimer’s
Proband Intact twin Proband Intact twin
Any head injury
Yes (n (%))
No (n (%))
31 (23.7)
100 (76.3)
34 (26.0)
97 (74.0)
15 (17.8)
69 (82.1)
22 (26.2)
62 (73.8)
Mild injury (n (%)) 16(12.2) 19(14.5) 6(7.1) 14(16.7)
With loss of
consciousness (n (%)) 15(11.5) 15 (11.5) 9(10.7) 8 (9.5)
Contact sports
Yes (n (%))
No (n (%))
11 (8.3)
122 (91.7)
17(12.8)
116(87.2)
4 (4.7)
81 (95.3)
9(10.6)
76 (89.4)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
25
Table 3: Frequency of risk factors for probands and partners, for all dementias combined
and for Alzheimer’s alone. Study of Dementia in Swedish Twins
All dementias
Proband Intact twin
Alzheimer’s
Proband Intact twin
Any head injury
Yes (n (%))
No (n (%))
Mild injury (n (%))
With loss of
consciousness (n (%))
Contact sports
Yes (n (%))
No (n (%))
Vibrating equipment
Yes (n (%))
No (n (%))
APOE-e4 (DZ pairs)
Yes (n (%))
No (n (%))
10(17.9)
46 (82.1)
3 (5.4)
7(12.5)
3 (5.5)
52 (94.6)
7(12.1)
51 (87.9)
8(38.1)
13(61.9)
17(30.4)
39 (69.6)
10(17.9)
7(12.5)
9 (16.4)
46 (83.6)
7(12.1)
51 (87.9)
3(14.3)
18(85.7)
7 (16.3) 13 (30.2)
36(83.7) 30(69.8)
2(4.7) 8(18.6)
5(11.6) 5(11.6)
2(4.5) 7(11.4)
40 (95.5) 35 (88.6)
7(15.9) 5(11.4)
37 (84.1) 39(88.6)
7 (35.0) 2 (10.0)
13(62.0) 18 (90.0)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 4: Frequency of risk factors for probands and partners, for all dementias combined
and for Alzheimer’s alone. OCTO-twin study
All dementias Alzheimer’s
Proband Intact twin Proband Intact twin
Any head injury
Yes (n (%))
No (n (%))
21 (28.0)
54 (72.0)
17 (22.7)
58 (77.3)
8(19.5)
33 (80.5)
9 (22.0)
32 (78.0)
Mild injury (n (%)) 13(17.3) 9(12.0) 4 (9.8) 6(14.7)
With loss of
consciousness (n (%)) 8(10.7) 8 (10.7) 4 (9.8) 3 (7.3)
Contact sports
Yes (n (%))
No (n (%))
8 (10.3)
70 (89.7)
8(10.3)
70 (89.7)
2 (4.7)
41 (95.3)
2 (4.7)
41 (95.3)
Odds ratios for all types of dementias together are shown for the combined data
(Table 5), the Study of Dementia in Swedish Twins only (Table 6), and the OCTO-twins
only (Table 7). For the combined sample, using all head injuries together and mild head
injury, there was essentially no association or a very small inverse relationship. A
nonsignificant positive association was found for head injury with loss of consciousness.
Playing contact sports was also found to have an inverse association to dementia,
although not statistically significant. In the Study of Dementia in Swedish Twins sample,
there were statistically significant inverse relationships found for both mild head injury
(OR = 0.2, 95% Cl 0.1-0.9) and having played contact sports (OR = 0.1, 95% Cl 0.02-
0.9). Head injury with loss of consciousness showed no association with dementia.
Occupational exposure to vibrating equipment was not a significant risk factor. In the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
27
OCTO-twin sample, head injury, both mild and with loss of consciousness, showed a
positive association with dementia, although not statistically significant, while contact
sports was not related to dementia.
Table 5: Risk estimates based on pairs who were dementia and exposure discordant, all
dementias, combined data from both twin samples TOR = odds ratio')
Number of discordant twin pairs
OR 95%CI Proband yes / Intact no Proband no / Intact yes
Any head injury 22 25 0.9 0.5— 1.6
Mild head injury 10 15 0.7 0.3-1.5
With loss o f consciousness 12 10 1.2 0.5-2.8
Contact sports 5 11 0.5 0.2-1.3
Table 6: Risk estimates based on pairs who were dementia and exposure discordant, all
dementias. Study of Dementia in Swedish Twins (OR = odds ratio)
Number of discordant twin pairs
OR 95%CI Proband yes / Intact no Proband no / Intact yes
Any head injury 7 14 0.5 0.2-1.2
Mild head injury 2 9 0.2 0.1-0.9
With loss o f consciousness 5 5 1.0 0.2-4.2
Contact sports 1 7 0.1 0.02-0.9
Vibrating equipment 3 3 1.0 0.2-6.3
At least one APOE-e4 5 0
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
28
Table 7: Risk estimates based on pairs who were dementia and exposure discordant, all
dementias. OCTO-twin study TOR = odds ratio)
Number of discordant twin pairs
OR 95%CI Proband yes / Intact no Proband no / Intact yes
Any head injury 15 11 1.3 0.6-3.0
Mild head injury 8 6 1.3 0.5-3.8
With loss of consciousness 7 5 1.4 0.4^1.4
Contact sports 4 4 1.0 0.2-1.3
Odds ratios for Alzheimer’s disease are reported for the combined data (Table 8),
the Study of Dementia in Swedish Twins only (Table 9), and the OCTO-twins only
(Table 10). The findings were similar to results using all dementias together. Again,
especially in the Study of Dementia in Swedish Twins sample, there was an inverse
association for mild head injury and an inverse relationship, although nonsignificant, for
playing contact sports.
Table 8: Risk estimates based on pairs who were Alzheimer’s and exposure discordant,
combined data from both twin samples (OR = odds ratio)
Number of discordant twin pairs
OR 95%CI Proband yes / Intact no Proband no / Intact yes
Any head injury 11 18 0.6 0.3-1.3
Mild head injury 4 11 0.4 0. 1 - 1.1
With loss of consciousness 7 7 1.0 0.3-3.4
Contact sports 2 7 0.3 0.1-1.3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
29
Table 9: Risk estimates based on pairs who were Alzheimer’s and exposure discordant.
Study of Dementia in Swedish Twins (OR = odds ratio)
Number of discordant twin pairs
Proband yes / Intact no Proband no / Intact yes OR 95%CI
Any head injury 5 11 0.5 0.2-1.3
Mild head injury 1 7 0.1 0.02-0.9
With loss of consciousness 4 4 1.0 0.2-5.0
Contact sports 1 6 0.2 0.03-1.1
Vibrating equipment 3 1 3.0 0.3-25.8
At least one APOE-e4 5 0
Table 10: Risk estimates based on pairs who were Alzheimer’s and exposure discordant.
OCTO-twin study TOR = odds ratio")
Number of discordant twin pairs
Proband yes / Intact no Proband no / Intact yes OR 95%CI
Any head injury 6 7 0.9 0.3-2.5
Mild head injury 3 4 0.8 0.2-3.3
With loss of consciousness 3 3 1.0 0.2-6.3
Contact sports 1 1 1.0 0.04-24.5
Odds ratios estimating the relative risk of APOE-e4 were calculated using data
from the Study of Dementia in Swedish Twins. These odds ratios are found in Table 6,
for risk of all dementias, and Table 9 for risk of Alzheimer’s disease only. None of these
findings were statistically significant; however, the data show a clear pattern. There were
5 disease discordant pairs (in all of which the proband had Alzheimer’s disease) also
discordant for having at least one APOE-e4 allele. The intact partner did not have an
APOE-e4 allele in any of these pairs. Thus, the odds ratio was 5/0, showing a strong
positive association between APOE-e4 and Alzheimer’s disease. The computational
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
30
convention when there is a zero cell is to add a constant to both off-diagonal cells, in
order to calculate an odds ratio and confidence intervals. The usual constant is 0.5.
However, adding 0.5 would result in the misleading odds ratio of 11.0. Therefore, no OR
or Cl are provided. With respect to testing for the interaction between APOE-e4 and
head injury, there was only one disease-discordant pair in which the proband had a
history of head injury and an APOE-e4 allele and the partner had neither (OR = 1/0).
Table 11 presents the odds ratios resulting from stratification by gender and
fam ily history of dementia, for all dementias together using the combined data. For the
head injury variables, there were positive associations for females and inverse
associations for males, all nonsignificant. There was also a nonsignificant positive
association for playing contact sports for females and a significant inverse association for
males (OR= 0.2, 95% Cl 0.1-0.9). For any head injury and playing contact sports, there
were nonsignificant inverse associations for those with positive family history of
dementia; for negative family history, there was a nonsignificant positive association for
any head injury and a nonsignificant inverse association for playing contact sports.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
31
Table 11: Risk estimates, based on pairs who were dementia and exposure discordant, all
dementias, combined data from both twin samples, stratified by gender and family history
of dementia (OR = odds ratio')
Number of discordant twin pairs
Proband yes / Intact no Proband no / Intact yes OR 95%CI
Any head injury
Male 6 13 0.5 0.2-1.2
Female 16 12 1.3 0.6-2.8
Mild head injury
Male 2 7 0.3 0.1-1.2
Female 8 8 1.0 0.3-3.1
With loss of consciousness
Male 4 6 0.7 0.2-2.3
Female 8 4 2.0 0.6-6.5
Contact sports
Male 2 9 0.2 0.1-0.9
Female 3 2 1.5 0.3-8.9
Any head injury
Family history (+) 7 14 0.5 0.2-1.2
Family history (-) 14 9 1.6 0.7-3.6
Contact sports
Family history (+) 1 4 0.3 0.03-1.9
Family history (-) 3 6 2.0 0.1— 1.9
Odds ratios from stratification by gender and family history of dementia, for all
dementias together, are shown for the Study of Dementia in Swedish Twins sample
(Table 12) and the OCTO-twin sample (Table 13). Trends were similar for both samples,
with odds ratios of head injuries and playing contact sports higher for women than for
men (although confidence intervals overlapped considerably between them). Data from
the Study of Dementia in Swedish Twins revealed a significant inverse relationship for
those with any head injury who also had a family history of dementia (OR=0.1, 95% Cl
0.02-0.5). This was not true for the OCTO-twin sample: there was no association for
those with head injury and family history of dementia. In both samples, there was a
large, but nonsignificant inverse association for playing contact sports in males.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
32
Table 12: Risk estimates, based on pairs who were dementia and exposure discordant, all
dementias. Study of Dementia in Swedish Twins, stratified by gender and family history
of dementia (OR = odds ratio')
Number of discordant twin pairs
Proband yes / Intact no Proband no / Intact yes OR 95%CI
Any head injury
Male 3 8 0.4 0.1-1.3
Female 4 6 0.7 0.2-2.3
Mild head injury
Male 1 4 0.3 0.03-1.9
Female 1 5 0.2 0.03-1.4
With loss o f consciousness
Male 2 4 0.5 0.1-2.6
Female 3 1 3.0 0.3-25.8
Contact sports
Male 1 5 0.2 0.03-1.4
Female 0 2
Any head injury
Family history (+) 1 10 0.1 0.02-0.5
Family history (-) 5 2 2.5 0.5-12.2
Contact sports
Family history (+) 1 2 0.5 0.05-5.3
Family history (-) 0 4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
33
Table 13: Risk estimates, based on pairs who were dementia and exposure discordant, all
dementias. OCTO-twin study, stratified by gender and family history of dementia (OR =
odds ratio)
Number of discordant twin pairs
Proband yes / Intact no Proband no / Intact yes OR 95%CI
Any head injury
Male 3 5 0.6 0.1-2.5
Female 12 6 2.0 0.8-5.2
Mild head injury
Male 1 3 0.3 0.04-2.9
Female 7 3 2.3 0.6-8.7
With loss o f consciousness
Male 2 2 1.0 0.1-9.6
Female 5 3 1.7 0.4-6.9
Contact sports
Male 1 4 0.3 0.03-1.9
Female 3 0
Any head injury
Family history (+) 6 4 1.5 0.4-5.3
Family history (-) 9 7 1.3 0.5-3.4
Contact sports
Family history (+)
Family history (-)
0
3
2
2 1.5 0.3-8.9
Odds ratios from stratification by gender and family history of dementia, for
Alzheimer’s disease only are shown for the combined data (Table 14), for the Study of
Dementia in Swedish Twins sample (Table 15), and the for OCTO-twin sample (Table
16). The sample sizes were smaller for these analyses and none of the odds ratios
reached statistical significance. Again, odds ratios for head injury variables tended to be
higher for females. Odds ratios from stratification by family history of dementia were
calculated using a small number of pairs; they showed considerable overlap in confidence
intervals compared to analyses with all dementias combined.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
34
Table 14: Risk estimates, based on pairs who were Alzheimer’s and exposure discordant,
combined data from both twin samples, stratified by gender and family history of
dementia (OR = odds ratio)
Number of discordant twin pairs
Proband yes / Intact no Proband no / Intact yes OR 95%CI
Any head injury
Male 3 8 0.4 0.1-1.3
Female 8 10 0.8 0.3-2.0
Mild head injury
Male 1 5 0.2 0.03-1.4
Female 3 6 0.5 0.1-1.9
With loss o f consciousness
Male 2 3 0.7 0.1-3.9
Female 5 4 1.3 0.3-4.6
Contact sports
Male 2 5 0.4 0.1-1.9
Female 0 2
Any head injury
Family history (+) 5 10 0.5 0.2-1.4
Family history (-) 5 7 0.7 0.2-2.2
Contact sports
Family history (+) 1 3 0.3 0.04-2.9
Family history (-) 0 3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
35
Table 15: Risk estimates, based on pairs who were Alzheimer’s and exposure discordant.
Study of Dementia in Swedish Twins, stratified by gender and family history of dementia
(OR = odds ratio')
Number of discordant twin pairs
Proband yes / Intact no Proband no / Intact yes OR 95%CI
Any head injury
Male 2 6 0.3 0.1-1.5
Female 3 5 0.6 0.1-2.5
Mild head injury
Male 0 3
Female 1 4 0.3 0.03-1.9
With loss o f consciousness
Male 2 3 0.7 0.1-3.9
Female 2 1 2.0 0.2-21.0
Contact sports
Male 1 4 0.3 0.03-1.9
Female 0 2
Any head injury
Family history (+) 1 8 0.1 0.02-0.7
Family history (-) 3 2 1.5 0.3-8.9
Contact sports
Family history (+) 1 2 0.5 0.05-5.3
Family history (-) 0 3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
36
Table 16: Risk estimates, based on pairs who were Alzheimer’s and exposure discordant.
OCTO-twin study, stratified by gender and family history of dementia (OR = odds ratio)
Number of discordant twin pairs
Proband yes / Intact no Proband no / Intact yes OR 95%CI
Any head injury
Male 1 2 0.5 0.1-5.3
Female 5 5 1.0 0.2-4.2
Mild head injury
Male 1 2 0.5 0.1-5.3
Female 2 2 1.0 0.1-9.6
With loss of consciousness
Male 0 0
Female 3 3 1.0 0.2-6.3
Contact sports
Male 1 1 1.0 0.04-24.5
Female 0 0
Any head injury
Family history (+) 4 2 2.0 0.4-10.6
Family history (-) 2 5 0.4 0.3-8.9
Contact sports
Family history (+) 0 1
Family history (-) 0 0
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
37
Discussion
This study did not find a significant association between head injury and
dementia or Alzheimer’s disease. Participation in contact sports was found to be
protective for dementia in men. These two main findings are discussed below, along with
the results for APOE-e4 and workplace use of vibrating equipment.
Head Injury
For head injury with loss of consciousness, odds ratios of 1.2 (95 % Cl 0.5-2.8)
for dementia and 1.0 (95% Cl 0.3-3.4) for Alzheimer’s disease were found. There was
no detectable increase in risk due to this type of head injury. These findings are
consistent with incidence studies of Alzheimer’s disease and dementia (Katzman et al.,
1989; Launer et al., 1999; Mehta et al., 1999; Williams et al., 1991) and other pair-
matched case-control studies that used more than 150 pairs (Broe et al., 1990; Chandra et
al., 1989; Hofman et al., 1989; Kokmen et al., 1988).
Similarly, milder head injuries that did not involve a loss of consciousness were
not associated with increased risk. Odds ratios of 0.7 (95% Cl 0.3-1.5) for dementia and
0.4 (95% Cl 0.1-1.1) for Alzheimer’s disease were found in this study. These odds ratios
were not statistically significant. They can be interpreted either as failing to confirm a
positive association or as signaling that there may be an inverse relationship.
The odds ratio of 1.0 found for head injury with loss of consciousness and
Alzheimer’s disease, as well as the inverse associations when looking at milder head
injuries were unexpected. There is no plausible reason to suspect that head injuries
would be protective for dementia. Biological studies have focused on the similarities
between head injury and Alzheimer’s disease pathology (e.g. Roberts et al., 1991).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
38
Furthermore, blind speculation seems inappropriate in the absence of statistical
significance. However, interestingly, many other studies have reported odds ratios of 1.0
or less for head injury (Breteler et al., 1995; Fratiglioni et al., 1993; Katzman et al., 1989;
Launer et al., 1999; Li et al., 1992; Mehta et al., 1999; Mendez et al., 1992; Soininen &
Heinonen, 1982; Williams et al., 1991). Future studies of head injury and dementia
neuropathology should perhaps focus on finding differences, rather than similiarities
between these conditions. This, too, could lead to important advances in understanding
the etiology of Alzheimer’s disease and other dementias.
If the odds ratio found in this study for head injury with loss of consciousness and
dementia is an accurate estimate, a much larger sample would be needed to find statistical
significance. This may not be worth the time or expense, due to the small amount of risk
that can be attributed to head injury. With an estimated relative risk of 1.2 and an 11.5%
rate of head injury with loss of consciousness, the etiological fraction for head injury in
this study is 0.02. That is, about 2% of dementia cases could be prevented if head
injuries with loss of consciousness were completely eliminated.
Participation in Contact Sports and Workplace Use of Vibrating Equipment
Participation in contact sports showed a nonsignificant negative association to
dementia (OR=0.5, 95% Cl 0.2-1.3) and Alzheimer’s disease (OR=0.3, 95% Cl 0.1-1.3).
However, this protective effect was significant for dementia in men only (OR=0.2, 95%
Cl 0.1-0.9). The odds ratio found for women was 1.5 (95% Cl 0.3-8.9). Thus, playing
contact sports was found to be protective for dementia, and gender modified this
association. One previous study (Broe et al., 1990) looked at risk for Alzheimer’s disease
due to history of boxing, wrestling, or rugby and reported an odds ratio of 1.07 (95% Cl
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
39
0.5-2.2). The only other study that has looked at contact sports and Alzheimer’s disease
(Canadian Study of Health and Aging, 1994) reported an odds ratio of 1.5 (95% CI=0.4-
6.0). These studies did not find an inverse relationship. However, neither of these
studies looked at dementia in general, neither reported the correlation between head
injury and contact sports, and neither performed separate analyses for men and women.
None of the people who had a history of playing contact sports in this sample
reported having had a head injury with loss of consciousness due to playing sports. There
was some overlap between those playing contact sports and those reporting mild head
injury: 38% of those who played contact sports also reported mild head injury; 29% of
those who reported mild head injury also played contact sports. Using the entire sample,
regardless of discordancy, the tetrachoric correlation between mild head injury and
playing contact sports was 0.25.
Studies that have linked contact sports to neuropsychological deficits have used
amateur and professional athletes (Collins et al., 1999; Matser et al., 1999; Matser et al.,
1998; Sortland & Tysvaer, 1989; Tysvaer & Lochen, 1991; Tysvaer & Storli, 1981;
Tysvaer et al., 1989), whereas that is not the case in this study. It is possible that playing
contact sports was a measure of physically active lifestyle in this study. Being more
physically active could lead to greater development of small blood vessels, which would
deliver more oxygen to the brain, thereby improving brain functioning and increasing
cognitive reserve. Consistent with this hypothesis, Broe and colleagues (1990) found a
significant increase in risk of Alzheimer’s due to lack of physical activity (OR=3.5, 95%
Cl 1.3-9.6).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
40
No association was found between workplace use of vibrating equipment and
dementia (OR=1.0, 95% Cl 0.2-6.3). A positive nonsignificant association was found for
the relationship between vibratory equipment and Alzheimer’s disease (OR=3.0, 95% Cl
0.3-25.8). One other study has reported on the association between workplace use of
vibrating equipment and Alzheimer’s disease (Li et al., 1992). They found an identical
odds ratio to that found in this study, 3.0 (95% Cl, 0.5-19.0).
APOE-e4 and Family History of Dementia
As expected, APOE-e4 did behave as a risk factor. Unfortunately, there were
insufficient data to test for the interaction, in order to see whether those with an APOE-e4
allele and had head injury with loss of consciousness would be significantly more at risk
for Alzheimer’s disease, in part due to the fact that APOE genotyping is not yet available
from the OCTO-Twin sample.
While not significant, these data suggested that head injury may be more
consequential for those without a family history of dementia. Consistent with this trend,
a meta-analysis of case-control studies found a stronger positive association between
head injury and Alzheimer’s disease for those without a positive family history of
dementia (Mortimer et al., 1991). However, other studies have found no relation between
family history of dementia and risk due to head injury (Launer et al., 1991; Van Duijn et
al., 1992).
It is possible that the family members of those who were family history positive
lived to an older age, thus giving them a greater chance of becoming demented than
family members of family history negative people whose immediate family members
may have died earlier and had less chance of becoming demented. If this were true, those
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
41
with family history positive may have had a more robust genetic makeup that made the
inverse relationship stronger. The positive association between head injury and dementia
for those who were family history negative is consistent with the idea that environmental
risk factors are more important for those without a pre-existing genetic susceptibility
(Mortimer et al., 1991).
However, the possibility that the family history was negative because first degree
relatives did not survive sufficiently long to become demented makes the interpretation of
negative family history of dementia very difficult. In the sample from the Study of
Dementia in Swedish Twins, family history was coded a second way, in which a negative
family history was assigned only if there was at least one family member known to have
lived to age 65 or older. Repeating the family history analyses with this more rigorous
definition did not change the results. This coding was not available for OCTO-Twin.
However, even with the more rigorous requirements for coding, the confounding of
family history with survival remains for any late onset disease.
Advantages and Limitations
The main limitations of this study and other case-control studies are the problems
of recall bias and sample size. Proband proxy informants may be more likely than
control proxy informants to recall past exposures that may be relevant to the disease. In
the present study, information on many different historical variables was collected and
the proxy informant was kept blind to which of these variables were being examined as
risk factors for Alzheimer’s disease in order to reduce bias. In addition, use of proxy
informants generally leads to greater misclassification of exposure history (Nelson et al.,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
42
1990) because the proxies lack the relevant information. This is especially an issue for
late onset diseases, when proxies are often adult children, as was the case in this study.
Medical records were used to cross check history of head injury for some of the
people in the Study of Dementia in Swedish Twins. Accuracy of head injury reports was
very high for those whose medical records were also available. There was 100%
agreement for intact twin partners who self-reported head injury information, and 93%
agreement for information gathered from an informant.
In the present study, it is possible that rates of head injury were lower in cases
simply because cases relied on proxy data to a greater extent than did controls, and there
was greater underreporting of exposure for cases than for controls. However, by using
multiple sources of information, self-report of head injury was the sole source of
information for only five of the non-demented twin partners (only one of whom reported
a history of head injury).
While the overall sample size for this study was comparatively large, there were
small numbers of discordant pairs who were also discordant with respect to risk factors
due to low baserates of these risk factors. The one previous twin study to look at head
injury and Alzheimer’s disease (Nee & Lippa, 1999) used a sample of 10 discordant twin
pairs. The present study used a much larger sample of 84 twin pairs discordant for
Alzheimer’s disease.
A major advantage of this study is use of co-twin controls, a method that is
especially useful for the study of rare diseases. This is the first substantial study to match
exactly for age. No other study has used pairs matched for family history of dementia.
Another advantage of this study is that we examined milder head injuries, contact sports,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
43
and occupational exposure to vibrating equipment. Most previous studies examining
head injury and Alzheimer’s disease defined head injury dichotomously and did not look
at these other variables.
Determination of discordance was also an improvement in this study over
previous twin studies of Alzheimer’s disease. In previous studies, the unaffected twin
partner must have remained intact for a minimum of three years to be classified as
discordant (e.g. Breitner et al., 1994). However, evidence from survival curves for
monozygotic and dizygotic pairs based on the length of time the partner remained intact
after the proband’s onset of Alzheimer’s disease showed that partners (some of whom
eventually became demented) had over a 90% chance of remaining intact 3 years after the
proband’s onset (Posner et al., 1999). Defining discordance is further complicated by the
fact that a specific age of onset for Alzheimer’s disease is very difficult to pinpoint. With
these issues in mind, this study used more stringent criteria for defining discordance.
Based on the aforementioned survival curves, the unaffected twin partner must have
remained intact for a period of time after the proband’s estimated age of onset ranging
from six years for the youngest old to zero years for the oldest old. These criteria should
have yielded a more accurate sample of twins discordant for Alzheimer’s disease, as well
as more accurate results.
Summary
Head injuries with loss of consciousness do not appear to increase risk of
Alzheimer’s or dementia, nor do less severe head injuries. Playing contact sports may be
protective for dementia in males. At the least, this study suggests that playing contact
sports in a recreational context does not place a person at higher risk for dementia. Since
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
44
this study did not use professional athletes, who would presumably have a greater
frequency of head injuries, these findings are probably not generalizable to that
population. It remains unclear how gender and family history of dementia change the
head injury-dementia association, but there is evidence from this study and others
suggesting that these may be important moderating factors to examine more closely in
future studies.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
45
References
Amaducci, L. A., Fratiglioni, L., Rocca, W. A., Fieschi, C., Livrea, P., Pedone,
D., Bracco, L., Lippi, A., Gandolfo, C., Bino, G., Prencipe, M., Bonatti, M. L., Girotti, F.,
Carella, F., Tavolato, B., Ferla, S., Lenzi, G. L., Carolei, A., Gambi, A., Grigoletto, F., &
Schoenberg, B. S. (1986). Risk factors for clinically diagnosed Alzheimer’s disease: A
case -control study of an Italian population. Neurology, 36. 922-931.
Annegers, J. F., Grabow, J. D., Kurland, J. T., & Laws, E. R. (1980). The
incidence, causes, and secular trends of head trauma in Olmstead County, Minnesota,
1935-1974. Neurology. 30, 912-919.
Barth, J. T., Macciocchi, S. N., Giordani, B., Rimel, R., Jane, J. A., & Boll, T. J.
(1983). Neuropsychological sequelae of minor head injury. Neurosurgery. 13. 529-533.
Bayles, K. A. (1991). Age at onset of Alzheimer’s disease. Archives of
Neurology. 43. 155-159.
Benjamin R., Leake A., McArthur F. K., Candy J. M., Ince P. G., Edwardson J.
A., Torvik A., Morris C. M., & Bjertness E. (1996). Apolipoprotein E genotype and
Alzheimer’s disease in an elderly Norwegian cohort. Neurodegeneration. 5. 43-47.
Berg, S., Johansson, B., Plomin, R., Ahem, F.M., Pedersen, N.L., & McClearn,
G.E. (1992). Origins of variance in the old-old: The first presentation of the OCTO-Twin
Study in Sweden. Behavior Genetics. 22. 708-709.
Bohnen, N., Jolles, J., & Twijnstra, A. (1992). Neuropsychological deficits in
patients with persistent sypmtoms six months after mild head injury. Neurosurgery. 30.
692-696.
Breitner J. C., Gau B. A., Welsh K. A., Plassman B. L., McDonald W. M., Helms
M. J., & Anthony J. C. (1994). Inverse association of anti-inflammatory treatments and
Alzheimer’s disease: initial results of a co-twin control study. Neurology. 44. 1609-1611.
Breitner J. C., Welsh K. A., Helms M. J., Gaskell P. C., Gau B. A., Roses A. D.,
Pericak-Vance M. A., & Saunders A. M. (1995). Delayed onset of Alzheimer’s disease
with nonsteroidal anti-inflammatory and histamine H2 blocking drugs. Neurobiology of
Aging. 16. 523-530.
Breteler, M. B., de Groot, R. M., van Romunde, L. K., & Hofman, A. (1995).
Risk of dementia in patients with Parkinson’s disease, epilipsy, and severe head trauma:
A register-based follow-up study. American Journal of Epidemiology. 142. 1300-1305.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
46
Broe, G. A., Henderson, A. S., Creasey, H., McCusker, E., Korten, A. E., Jorm,
A. F., Longley, W., & Anthony, J. C. (1990). A case-control study of Alzheimer’s disease
in Australia. Neurology. 40. 1698-1707.
The Canadian Study of Health and Aging. (1994). Risk factors for Alzheimer’s
disease in Canada. Neurology. 44. 2073-2080.
Chagani, H. T., Phelps, M. E., & Mazziotta, J. C. (1987). Positron emission
tomography study of the human brain functional development. Annals of Neurology. 22.
487-497.
Chandra, V., Kokmen, E., Schoenberg, B. S., & Beard, M. C. (1989). Head
trauma with loss of consciousness as a risk factor for Alzheimer’s disease. Neurology. 39.
1576-1578.
Chandra, V., Philipose, R. N., Bell, P. A., Lazaroff, A., & Schoenberg, B. S.
(1987). Case-control study of late onset probable Alzheimer’s disease. Neurology. 37.
1295-1300.
Collins, M. W., Grindel, S. H., Lovell, M. R., Dede, D. E., Moser, D. J., Phalin,
B. R., Nogle, S., Wasik, M., Cordry, D., Daugherty, M. K., Sears, S. F., Nicolette, G.,
Indelicato, P., & McKeag, D. B. (1999). Relationship between concussion and
neuropsychological performance in college football players. JAMA. 282. 964-970.
Corsellis, J. N., & Brierly, J. B. (1959). Observations on the pathology of
insidious dementia following head injury. Journal of Mental Science. 105. 714-720.
Farrer L. A., Cupples L. A., Van Duijn C. M., Kurz A., Zimmer R., Muller U.,
Green R. C., Clarke V., Shoffner J., & Wallace D. C. (1995). Apolipoprotein E genotype
in patients with Alzheimer’s disease: implications for the risk of dementia among
relatives. Annals of Neurology. 38. 797-808.
Fife, D. (1987). Head injury with and without hospital admission: Comparisons
of incidence and short-term disability. American Journal of Public Health. 77. 810-812.
Folstein, M.F., Folstein, S.E., & McHugh, P.R. (1975). "Mini-Mental State": A
practical method for grading the cognitive state of patients for the clinician. Journal of
Psychiatric Research. 12. 189-198.
Fratiglioni L., Ahlbom A., Viitanen M., & Winblad B. (1993).Risk factors for
late-onset Alzheimer’s disease: a population based, case-control study. Annals of
Neurology. 33. 258-266.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
47
Friedman, G., Froom, P., Sazbon, L., Grinblatt, I., Sochina, M., Tsenter, J.,
Babaey, S., Yehuda, A. B., & Groswasser, Z. (1999). Apolipoprotein E epsilon 4
genotype predicts a poor outcome in survivors of traumatic brain injury. Neurology. 52.
244-248.
Gatz, M., Kasl-Godley, J. E., & Karel, M. J. (1996) Aging and mental disorders.
In J. E. Birren & K. W. Schaie (Eds.), Handbook of the Psychology of Aging (pp. 365-
382). San Diego, CA: Academic Press.
Gatz, M., Pedersen, N. L., Berg, S., Johansson, B., Johansson, K., Mortimer, J.
A., Posner, S. F., Viitanen, M., Winblad, B., & Ahlbom, A. (1997). Heritability for
Alzheimer's disease: The Study of Dementia in Swedish Twins. Journal of Gerontology:
Medical Sciences. 52A. Ml 17-125.
Gatz, M., Pedersen, N. L., Crowe, M., & Fiske, A. (2000). Defining discordance
in twin studies of risk and protective factors for late life disorders. Twin Research. 3.
159-164.
Gatz, M., Reynolds, C., John, R., Johansson, B., Mortimer, J. A., & Pedersen, N.
L. (2000). Telephone screening to identify potential dementia cases in a population-
based sample of older adults. Manuscript submitted for publication
Gatz, M., Reynolds, C., Nikolic, J., Lowe, B., Karel, M., & Pedersen, N. (1995).
An empirical test of telephone screening to identify potential dementia cases.
International Psychogeriatrics. 7. 429-437.
Gordon, M. N. (1993). Microglia and immune activation in Alzheimer’s disease.
Journal of the Florida Medical Association. 80. 267-270.
Graham D. I., Gentleman S. M., Lynch A., & Roberts G. W. (1995). Distribution
of beta-amyloid protein in the brain following severe head injury. Neuropathology and
Applied Neurobiology. 21. 27-34.
Graves, A. B., & Kukull, W. A. (1994). The epidemiology of dementia. In J.
Morris (Ed.), Handbook of Dementing Illnesses (pp. 23-70). New York: Marcel Dekker,
Inc.
Graves A. B., White E., Koepsell T. D., Reifler B. V., Van Belle G., Larson E. B.,
& Raskind M. (1990). The association between head trauma and Alzheimer’s disease.
American Journal of Epidemiology. 131, 491-501.
Gronwall, D., & Wrightson, P. (1974). Delayed recovery of intellectual function
after minor head injury. Lancet, 2, 605-609.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
48
Guo, Z., Cupples, L. A., Kurz, A., Auerbach, S. H., Volicer, L., Chui, H., Green,
R. C., Sadovnick, A. D., Duara, R., DeCarli, C., Johnson, K., Go, R. C., Growdon, J. H.,
Haines, J. L., Kukull, W. A., & Farrer, L. A. (2000). Head injury and risk of AD in the
MIRAGE study. Neurology. 54. 1316-1323.
Haier, R. J. (1993). Cerebral glucose metabolism and intelligence. In P. A.
Vernon (Ed.), Biological Approaches to the Study of Human Intelligence. Norwood, NJ:
Ablex Publishing Corporation.
Han S. H., Hulette C., Saunders A. M., Einstein G., Pericak-Vance M.,
Strittmatter W. J., Roses A. D., & Schmechel D. E. (1994). Apolipoprotein E is present in
hippocampal neurons without neurofibrillary tangles in Alzheimer’s disease and in age-
matched controls. Experimental Neurology. 128. 13-26.
Henderson, V., Paganini-Hill, A., Emanuel, C., Dunn, M., & Buckwalter, J. C.
(1994). Estrogen replacement therapy in older women: Comparisons between
Alzheimer’s disease cases and nondemented controls. Archives of Neurology. 51. 896-
900.
Heyman, A., Wilkinson, W. E., Stafford, J. A., Helms, M. J., Sigmon, A. H., &
Weinberg, T. (1984). Alzheimer’s disease: A study of epidemiological aspects. Annals
of Neurology. 15. 335-341.
Hofman, A., Schulte, W., Tanja, T. A., van Duijn, C. M., Haaxma, R., Lameris,
A. J., Otten, V. M., & Saan, R. J. (1989). History of dementia and Parkinson’s disease in
l st-degree relatives of patients with Alzheimer’s disease. Neurology. 39. 1589-1592.
Hung, C. C., Chiu, W. T., Tsai, J. C., Laporte, R. E., & Shih, C. J. (1991). An
epidemiological study of head injury in Hualien County, Taiwan. Journal of the
Formosan Medical Association. 90. 1227-1233.
Huttenlocher, P. R. (1979). Synaptic density in human frontal cortex:
Developmental changes and effects of aging. Brain Research. 163. 195-205.
Hyman B. T., Gomez-Isla T., Rebeck G. W., Briggs M., Chung H., West H. L.,
Greenberg S., Mui S., Nichols S., Wallace R., & Growdon J. H. (1996). Epidemiological,
clinical, and neuropathological study of apolipoprotein E genotype in Alzheimer’s
disease. Annals of the New York Academy of Sciences. 802. 1-5.
Jordan, B. D., Relkin, N. R., Ravdin, L. D., Jacobs, A. R., Bennett, A., & Gandy,
S. (1997). Apolipoprotein E epsilon 4 associated with chronic traumatic brain injury in
boxing. JAMA. 278. 136-140.
Kalaria, R. N. (1993). The immunopathology of Alzheimer’s disease and some
related disorders. Brain Pathology. 3. 333-347.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
49
Katzman, R. (1993). Education and the prevalence of dementia and Alzheimer’s
disease. Neurology. 43. 13-20.
Katzman, R., Aronson, M., Fuld, P., Kawas, C., Brown, T., Morgenstem, H.,
Frishman, W., Gidez, L., Eder, H., & Ooi, W. L. (1989). Development of dementing
illnesses in an 80-year old volunteer cohort. Annals of Neurology. 25. 317-324.
Khachaturian, Z. S., & Radebaugh T. S. (1998). Alzheimer disease: Where are
we now? Where are we going? Alzheimer Disease and Associated Disorders. 12 (Suppl.
3), 24-28.
Kokmen, E., Chandra, V., & Schoenberg, B. S. (1988). Trends in incidence of
dementing illness in Rochester, Minnesota, in three quinquennial periods, 1960-1974.
Neurology. 38. 975-980.
Kondo, K., & Yamashita, I. (1990). A case-control study of Alzheimer’s disease
in Japan: Association with inactive psychosocial behaviors. In K. Hasegawa & A.
Homma (Eds.), Psychogeriatrics. Biomedical, and Social Advances (pp. 49-53).
Amsterdam: Excerpta Medica.
Lannfelt L., Pedersen N. L., Lilius L., Axelman K., Johansson K., Viitanen M., &
Gatz M. (1995). Apolipoprotein e4 allele in Swedish twins and siblings with Alzheimer’s
disease. Alzheimer Disease and Associated Disorders. 9. 166-169.
Launer, L. J., Andersen, K., Dewey, M. E., Letenneur, L., Ott, A., Amaducci, L.
A., Brayne, C., Copeland, J. R., Dartigues, J. F., Kragh-Sorensen, P., Lobo, A., Martinez-
Lage, J. M., Stijinen, T., & Hofman, A. (1999). Rates and risk factors for dementia and
Alzheimer’s disease: Results from EURODEM pooled analyses. Neurology. 52. 78-84.
Li, G., Shen, Y. C., Li, Y. T., Chen, C. H., Zhau, Y. W., & Silverman, J. M.
(1992). A case-control study of Alzheimer’s disease in China. Neurology. 42. 1481-1488.
Lucassen, P. J., Van Someren, E. J., & Swabb, D. F. (1998). Are active neurons a
better defense against aging in Alzheimer’s disease? Tijdschrift voor Gerontologie en
Geriatrie. 29. 177-184.
Matser, J. T., Kessels, A. G., Jordan, B. D., Lezak, M. D., & Troost, J. (1998).
Chronic traumatic brain injury in professional soccer players. Neurology. 51. 791-796.
Matser, J. T., Kessels, A. G., Lezak, M. D., Jordan, B. D., & Troost, J. (1999).
Neuropsychological impairment in Amateur Soccer Players. JAMA. 282. 971-973.
Mayeux R., Ottman R., Maestre G., Ngai C., Tang M. X., Ginsberg H., Chun M.,
Tycko B., & Shelanski M. (1995). Synergistic effects of traumatic head injury and
apolipoprotein-e4 in patients with Alzheimer’s disease. Neurology. 45. 555-557.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
50
Mayeux R., Ottman R., Tang M. X., Noboa-Bauza L., Marder K., Gurland B., &
Stern Y. (1993). Genetic susceptibility and head injury as risk factors for Alzheimer’s
disease among community-dwelling elderly persons and their first-degree relatives.
Annals of Neurology. 33. 494-501.
McGeer, P. L., & Rogers, J. (1992). Anti-inflammatory agents as a therapeutic
approach to Alzheimer’s disease. Neurology. 42, 447-449.
McGeer, P. L., Rogers, J., & McGeer, E. G. (1994). Neuroimmune mechanisms in
Alzheimer’s disease pathogenesis. Alzheimer Disease and Associated Disorders. 8. 149-
158.
McKhann, G., Drachman, D., Folstein, M., Katzman, R., Price, D., Stadlan, E. M.
(1984). 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. 939-944.
Mehta, K. M., Ott, A., Kalmijn, S., Slooter, A. J., Van Duijn, C. M., Hofman, A.,
& Breteler, M. M. (1999). Head trauma with loss of consciousness and risk of dementia:
The Rotterdam Study. Neurology. 53. 1959-1962.
Mendez, M. F., Underwood, K. L., Zander, B. A., Mastri, A. R., Sung, J. H., &
Frey, W. H. (1992). Risk factors in Alzheimer’s disease: A clinicopathologic study.
Neurology. 42. 770-775.
Montgomery, E. A., Fenton, G. W., McClelland, R. J., MacFlynn, G., &
Rutherford, (1991). The psychobiology of minor head injury. Psychological Medicine.
2L 375-384.
Morris, J. C., Heyman, A., Molls, R. C., Hughes, J. P., van Belle, G., Fillenbaum,
G., Mellits, E. D., Clark, D., & the CERAD investigators (1989). The consortium to
establish a registry for Alzheimer’s disease (CERAD): I. Clinical and neuropsychological
assessment of Alzheimer’s disease. Neurology. 39. 1159-1165.
Mortimer, J. A. (1995). Prospects for prevention of dementia and associated
impairments. In L. A. Bond, S. J. Cutler, & A. Grams (Eds.), Promoting successful and
productive aging (pp. 131-147). Thousand Oaks, CA: Sage Publications.
Mortimer, J. A., French, L. R., Hutton, J. T., & Schuman, L. M. (1985). Head
trauma as a risk factor for Alzheimer’s disease. Neurology. 35. 264-267.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
51
Mortimer J. A., Van Duijn C. M., Chandra V., Fratiglioni L., Graves A. B.,
Heyman A., Jorm A. F., Kokmen E., Kondo K., Rocca W. A., Shalat, S. L., Soininen, H.,
& Hofman, A. (1991). Head trauma as a risk factor for Alzheimer’s disease: a
collaborative re-analysis of case-control studies. International Journal of Epidemiology.
20. (Suppl. 2), 28-35.
Namba Y., Tomonaga M., Kawasaki H., Otomo E., & Ikeda K. (1991).
Apolipoprotein E immunoreactivity in cerebral amyloid deposits and neurofibrillary
tangles in Alzheimer’s disease and kuru plaque amyloid in Creutzfeldt-Jakob disease.
Brain Research. 541. 163-166.
Nathan, B. P., Bellosta, S., & Sanan, D. A. (1994). Differential effects of
apolipoproteins E3 and E4 on neuronal growth in vitro. Science. 264. 850-852.
Nee L. E., & Lippa C. F. (1999). Alzheimer’s disease in 22 twin pairs— 13-year
follow-up: Hormonal, infectious and traumatic factors. Dementia & Geriatric Cognitive
Disorders. 10. 148-151.
Nelson L. M., Longstreth W. T., Jr., Koepsell T. D., & Van Belle G. (1990).
Proxy respondents in epidemiologic research. Epidemiologic Reviews. 12. 71-86.
O'Meara E. S., Kukull W. A., Sheppard L., Bowen J. D., McCormick W. C., Teri
L., Pfanschmidt M., Thompson J. D., Schellenberg G. D., & Larson E. B. (1997). Head
injury and risk of Alzheimer’s disease by Apolipoprotein E genotype. American Journal
of Epidemiology. 146. 373-384.
Pedersen, N. L., Friberg, L., Floderus-Myrhed, B., & McClearn, G. E. (1984).
Swedish early separated twins: Identification and characterization. Acta Geneticae
Medicae et Gemellologiae. 33. 243-250.
Pedersen, N. L., McClearn, G. E., Plomin, R., Nesselroade, J. R., Berg, S., & de
Faire, U. (1991). The Swedish adoption twin study of aging: An update. Acta Geneticae
Medicae et Gemellologiae. 40. 7-20.
Plassman, B. L., Havlik, R. J., Steffens, D. C., Helms, M. J., Newman, T. N.,
Drosdick, D., Phillips, C., Gau, B. A., Welsh-Bohmer, K. A., Burke, J. R., Guralnik, J.
M., & Breitner, J. C. (2000). Documented head injury in early adulthood and risk of
Alzheimer’s disease and other dementias. Neurology. 55. 1158-1166.
Posner, S. F., Pedersen, N. L., & Gatz, M., (1999). Application of life table
analysis to the onset of dementia in a genetically informative design. Neuropsvchiatric
Genetics. 88. 207-210.
Rasmusson, D. X., Brandt, J., Martin, D. B., & Folstein, M. F. (1995). Head
injury as a risk factor for Alzheimer’s disease. Brain Injury. 9. 213-219.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
52
Reynolds, C. A., Wetherell, J. L., & Gatz, M. (1999). Heritability of Alzheimer’s
Disease. In B. Vellas, J. Fitten, and G. Frisoni (Eds.), Research and Practice in
Alzheimer’s Disease. New York: Springer Publishing Company.
Rice, D. (1991). The burden of Alzheimer’s disease. The Gerontologist. 31. 303.
Ritchie, I., Kildea, D., & Robine, J.-M. (1992). The relationship between age and
the prevalence of senile dementia: A meta-analysis of recent data. International Journal
of Epidemiology. 22. 763-769.
Roberts, G. W., Gentleman, S. M., Lynch, A., & Graham, D. I. (1991). Beta A4
amyloid protein deposition in brain after head trauma. Lancet. 338. 1422-1433.
Roberts G. W., Gentleman S. M., Lynch A., Murray L., Landon M., & Graham D.
I. (1994). Beta amyloid protein deposition in the brain after severe head injury. Journal
of Neurology. Neurosurgery, and Psychiatry. 57. 419-425.
Rogers J., Kirby L. C., Hempelman S. R., Berry D. L., McGeer P. L., Kaszniak A.
W., Zalinski J., Cofield M., Mansukhani L., & Willson P. (1993). Clinical trial of
indomethacin in Alzheimer’s disease. Neurology. 43. 1609-1611.
Roman, G. C., Tatemichi, T. K., Erkinjuntti, T., Cummings, J. L., Masden, J. C.,
Garcia, S. H., Amaducci, L., Orgogozo, J. M., Brun, A., & Hofman, A. (1993). Vascular
dementia: Diagnostic criteria for research studies. Report of the NINDS-AIREN
international workshop. Neurology. 43. 256-260.
Roses, A. D. (1998). Genetic testing and Alzheimer disease: the promise.
Alzheimer Disease and Associated Disorders. 12. (Suppl. 3), 3-9.
Salib, E., & Hillier, V. (1997). Elead injury and the risk of Alzheimer’s disease:
A case control study. International Journal of Geriatric Psychiatry. 12. 363-368.
Schlesselman, J. J. (1982). Case-control studies: Design, conduct, analysis. New
York: Oxford University Press.
Schofield, P. W., Tang, M., Marder, K., Bell, K., Dooneief, G., Chun, M., Sano,
M., Stern, Y., & Mayeux, R. (1997). Alzheimer’s disease after remote head injury: An
incidence study. Journal of Neurology. Neurosurgery, and Psychiatry. 62. 119-124.
Selkoe, D. J. (1993). Physiological production of the amyloid beta-amyloid
protein and the mechanism of Alzheimer’s disease. Trends in Neuroscience. 16. 403-409.
Soininen, H., & Heinonen, O. P. (1982). Clinical and etiological aspects of senile
dementia. European Neurology. 21. 401-410.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
53
Sortland, O., & Tysvaer, A. T. (1989). Brain damage in former association
football players: An evaluation by cerebral computed tomography. Neuroradiology, 31.
44-48.
Sternberg, R. J., & Kaufman, J. C. (1998). Human abilities. Annual Review of
Psychology. 49, 479-502.
Tysvaer, A. T., & Lochen, E. A. (1991). Soccer injuries to the brain. American
Journal of Snorts Medicine. 19. 56-60.
Tysvaer, A. T., & Storli, O. (1981). Association football injuries to the brain.
British Journal of Sports Medicine. 15. 163-166.
Tysvaer, A. T., Storli, O., & Bachen, N. I. (1989). Soccer injuries to the brain.
Acta Neurologica Scandinavica. 80. 151-156.
U. S. Congress, Office of Technology Assessment (1987). Losing a Million
Minds: Confronting the Tragedy of Alzheimer’s Disease and Other Dementias (p. 323).
Washington, D. C.: U. S. Government Printing Office.
van Duijn C. M., Tanja T. A., Haaxma R., Schulte W., Saan R. J., Lameris A. J.,
Antonides-Hendriks G., & Hofman A. (1992). Head trauma and the risk of Alzheimer’s
disease. American Journal of Epidemiology. 135. 775-782.
Wenham, P. R., Price, W. H., & Blundell, G. (1991). Apolipoprotein E
genotyping by one-stage PCR. Lancet. 337. 1158-1159.
Wetherell, J. L., Gatz, M., Johansson, B., & Pedersen, N. L. (1999). History of
depression and other psychiatric illness as risk factors for Alzheimer’s disease in a twin
sample. Alzheimer Disease and Associated Disorders. 13. 47-52.
Williams D. B., Annegers J. F., Kokmen E., O'Brien P. C., & Kurland L. T.
(1991). Brain injury and neurologic sequelae: a cohort study of dementia, parkinsonism,
and amyotrophic lateral sclerosis. Neurology. 41. 1554-1557.
Wisniewski T., Golabek A., Matsubara E., Ghiso J., & Frangione B. (1993).
Apolipoprotein E: binding to soluble Alzheimer’s beta-amyloid. Biochemical and
Biophysical Research Communications. 192, 359-365.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Cognitive functioning and dementia following cancer: A Swedish twin study

PDF

History of depression, antidepressant treatment, and other psychiatric illness as risk factors for Alzheimer's disease in a twin sample

PDF

Alcohol expectancies and consumption: Age and sex differences

PDF

Depression as a risk factor for dementia

PDF

Hispanic culture, acculturation, and distress among caregivers of dementia patients

PDF

Hedonic aspects of conditioned taste aversion in rats and humans

PDF

Adolescents' social attitudes: Genes and culture?

PDF

Effects of bilateral stimulation and stimulus redundancy on performance in processing nonword letter trigrams

PDF

Frequency and risk factors of poststroke dementia

PDF

Insights into the nature of phonological and surface dyslexia: Evidence from a novel word learning task

PDF

Depression and suicidality in Latino adolescents: A study of acculturation and gender role beliefs

PDF

An examination of affective modulation, psychopathy, and negative schizotypy in college and community samples

PDF

Depression in offspring following severe prenatal stress

PDF

Content analysis of articulated thoughts of chronic worriers

PDF

Articulated thoughts regarding cognitions toward older adults

PDF

Affection and conflict in family relationships

PDF

Dementia caregiving and ethnicity: African American caregivers and the sociocultural stress and coping model

PDF

Depressed children and the social and behavioral attributes of their best friend

PDF

Aging and the use of context and frequency information in ambiguity resolution

PDF

Expectancies for alternative behaviors predict drinking problems: A test of a cognitive reformulation of the matching law

Asset Metadata

Creator Crowe, Michael Gabriel (author)

Core Title Head injury and dementia: A co-twin control study of Swedish twins

School Graduate School

Degree Master of Arts

Degree Program Psychology

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

Tag OAI-PMH Harvest,psychology, clinical,psychology, psychobiology

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Gatz, Margaret (committee chair), [illegible] (committee member), Lavond, David (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c16-39013

Unique identifier UC11341966

Identifier 1407907.pdf (filename),usctheses-c16-39013 (legacy record id)

Legacy Identifier 1407907.pdf

Dmrecord 39013

Document Type Thesis

Rights Crowe, Michael Gabriel

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