—To determine prevalence of dementia and its subtypes in Japanese-American men and compare these findings with rates reported for populations in Japan and elsewhere.
Design and Setting.
—The Honolulu Heart Program is a prospective populationbased study of cardiovascular disease established in 1965. Prevalence estimates were computed from cases identified at the 1991 to 1993 examination. Cognitive performance was assessed using standardized methods, instruments, and diagnostic criteria.
—Subjects were 3734 Japanese-American men (80% of surviving cohort) aged 71 through 93 years, living in the community or in institutions.
Main Outcome Measures.
—Age-specific, age-standardized, and cohort prevalence estimates were computed for dementia (all cause) defined by 2 sets of diagnostic criteria and 4 levels of severity. Prevalence levels for Alzheimer disease and vascular dementia were also estimated.
—Dementia prevalence by Diagnostic and Statistical Manual of Mental Disorders, Third Edition, Revised ranged from 2.1% in men aged 71 through 74 years to 33.4% in men aged 85 through 93 years. Age-standardized prevalence was 7.6%. Prevalence estimates for the cohort were 9.3% for dementia (all cause), 5.4% for Alzheimer disease (primary or contributing), and 4.2% for vascular dementia (primary or contributing). More than 1 possible cause was found in 26% of cases. The Alzheimer disease/vascular dementia ratio was 1.5 for cases attributed primarily to Alzheimer disease or vascular dementia.
—Prevalence of Alzheimer disease in older Japanese-American men in Hawaii appears to be higher than in Japan but similar to European-ancestry populations. Prevalence of vascular dementia appears to be only slightly lower than in Japan, but higher than in European-ancestry populations. Further cross-national research with emphasis on standardized diagnostic methods is needed.