Acculturation and Alzheimer Disease Risk Among JapaneseAmerican Elderly: The Kame Study

1
Acculturation and Alzheimer Disease Risk Among
Japanese-American Elderly The Kame Study
aViet Q. Nguyen bPaul K. Crane, MD, MPH
aKeerthi Arani bLaura E. Gibbons, PhD
cKristoffer Rhoads, PhD dSusan M. McCurry, PhD
eAmy R. Borenstein, PhD fEric B. Larson, MD,
MPH a University of Washington School of
Medicine, Seattle, WA b Department of General
Internal Medicine, c Department of Psychiatry and
Behavioral Sciences, d Department of
Psychosocial and Environmental Health, University
of Washington, Seattle, WA eDepartment of
Epidemiology and Biostatistics, College of Public
Health, University of South Florida, Tampa, FL f
Center for Health Studies, Group Health
Cooperative of Puget Sound, Seattle, WA

• We considered several covariates as potential
  confounders such as age, gender, and
  apolipoprotein E-?4 alleles. Each potential
  covariate was entered into a Cox proportional
  hazards model9 with acculturation level.
  Candidate confounders that changed the
  coefficient by more than 15 were chosen for
  inclusion in the final model. The two
  confounders that fit this criterion were income
  and education.
• Subjects contributed time to the Cox
  proportional hazards model until they died,
  developed dementia, dropped out of the study, or
  reached the end of the study.

• Objective
• To examine the relationship between acculturation
  to Western society and the risk of incident
  Alzheimer Disease and dementia in a population of
  elderly Japanese-Americans.

• Background
• There has been increased interest in
  cross-cultural studies of dementia and Alzheimer
  Disease (AD) in genetically similar populations
  in order to untangle possible environmental risk
  factors. Early studies in Japan and the U.S. led
  to the development of the Ni-Hon-Sea Project,
  linked prospective studies of incident dementia
  in Japanese or Japanese-American elders.
• The Kame Project examined 3,045 eligible
  Japanese Americans in King County, Washington and
  examined the prevalence of dementia while
  recording possible risk factors1.
• Graves showed that over a period of two years,
  Japanese-Americans who maintain traditional
  Japanese values and a Japanese lifestyle may have
  a decreased risk of cognitive decline2.
• Our investigation expands upon this research by
  examining dementia outcomes of the Kame Project
  participants over a longer follow-up period and
  employing a new approach to the measurement of
  acculturation3, which we define as the extent by
  which individuals moving into a new host society
  adapt to a new set of cultural practices.

• Results
• Individuals in the acculturation quartiles had
  different demographic characteristics (Table 1).
  Specifically, those in the lowest acculturation
  quartile comprised most of the individuals over
  the age of 80 (76 out of 139), were mostly Issei
  or Kibei (371 out of 391), were mostly female
  (69), had the highest percentage of individuals
  with fewer than 11 years of education (43), and
  had the highest percentage of individuals with
  less than 15000 of annual income (36).
• Vascular risk factors did not vary as much
  across individuals in the different acculturation
  quartiles.
• The association between acculturation and
  Alzheimers Disease risk is summarized in Table
  2. In unadjusted models, the relative hazard
  associated with higher levels of acculturation
  was 0.36 (0.20-0.67). With adjustments for
  demographic characteristics, the relative hazard
  associated with higher levels of acculturation
  was 0.41 (0.22-0.76).

• Discussion
• Our results show a nearly 60 decreased risk for
  AD in more acculturated elderly
  Japanese-Americans compared with those who were
  less acculturated.
• Several possibilities may explain this result,
  such as low income leading to poorer access to
  health care, low education and unipolar
  acculturation leading to poorer cognitive
  reserve, and the stress of living away from
  ones native country. However, all of these are
  highly speculative and need to be examined
  further.
• Limitations to our study include its
  observational nature, its unipolar
  conceptualization of acculturation, and missing
  data in the covariates. Also, the specific
  nature of the exposure causing an increased risk
  for AD is unknown.

Conclusion Japanese-Americans who were in the
highest acculturation group appear to have a
decreased risk for acquiring Alzheimer Disease
(Table 2). These results are inconsistent with
Graves conclusion that acculturation to Western
society is correlated with cognitive decline in
elderly Japanese-Americans despite different
exclusion criteria and a longer follow-up period
so a similar study to ours should be replicated
to reaffirm our results. The reasons for this
result can only be speculated, and there are
likely a combination of factors that account for
this protective effect of acculturation on the
risk of AD. It is known that rates of AD are
higher in the United States than in Japan, so if
environmental factors contribute to the risk of
AD in Japanese-Americans, investigation should be
done to see if this holds true for other
population groups.

• Study Design
• From 3,045 eligible participants, an original
  cohort of 1,985 participated in the baseline
  evaluation of the original Kame study. The
  criteria for acceptance into this cohort included
  being over the age of 65 and at least 50
  Japanese ancestry.
• From this cohort, 127 participants lacked
  sufficient data to measure acculturation and a
  further 236 participants were either diagnosed
  with dementia at baseline or were non-demented
  but opted not to continue beyond the initial
  screen. 6 participants were diagnosed with
  prevalent Parkinsons Disease. The remaining
  1,616 participants were followed biennially.
• Baseline screening for dementia included the
  Cognitive Abilities Screening Instrument (CASI)4,
  and those scoring full physical and neuropsychological evaluation
  by a study neurologist or geriatrician. For
  reference, the cutoff score on the Honolulu-Asia
  Aging Study5 was AD was based on DSM-III6 criteria by consensus
  agreement by a committee of physicians.
• Subsequent biennial evaluations for cognitive
  decline were performed with the CASI. Scores prompted the same diagnostic workup. Diagnosis of
  AD was based on NINCDS-ADRDA7 criteria.
• An acculturation score was determined based on
  questionnaire responses regarding childhood
  experiences and exposures, adult exposures to
  Japan and the U.S., adult use of Japanese
  language, and cultural preferences.
• Data from twenty acculturation items was combined
  using item response theory, using PARSCALE 3.1
  (Scientific Software International, 2003).
  Expectation a posteriori scoring was used for the
  graded response model8. Acculturation scores
  were then divided into quartiles for analyses.
  Final models compared the bottom two quartiles to
  the top two quartiles.

Funding for this project was provided by an
Investigator Initiated Research Grant from the
Alzheimers Association (P. Crane, PI). Funding
for the original Kame dataset was supported by
National Institute on Aging grant no. AG09769 (E.
Larson, PI).