Background. Rates of dementia may vary among ethnoracial groups. Any real a
nd substantial such difference would merit serious attention by health plan
ners, clinicians and those seeking to advance our understanding of the etio
logy of this group of disorders.
Methods. Randomly selected elderly persons from each of three ethnoracial g
roups (Latinos, African-Americans, non-Latino Whites) residing in a geograp
hic area of northern Manhattan, in New York City were screened for dementia
and assessed with respect to functioning in dairy tasks and other qualitie
s of life. Systematic samples of each group were clinically evaluated for p
resence and subtype of dementia. Subjects were reassessed at an average of
18 months following the baseline interview.
Results. Age-specific prevalence of dementia was found to be higher in Lati
nos and African-Americans than in non-Latino Whites; incidence rates were c
onsistent with this finding. Ethnoracial groups did not vary in the proport
ion of dementias diagnosed as Alzheimer's disease. Prevalence differences b
etween ethnoracial groups remained consistent as diagnostic criteria were v
aried in breadth and when the possible mislabelling of depression was taken
into account. However, level of education was strongly associated with rat
es of dementia and, when age and education were simultaneously controlled,
the ethnoracial differences in rates were not consistently found.
Conclusions. Planning for the wide range of services necessary for care of
those suffering from dementia should rake into account ethnoracial differen
ces in rates. The higher rates found in Latino and African-American groups,
relative to non-Latino Whites, are associated with clear and substantial f
unctional dependencies and hence have important implications for qualities
of life and service needs. Copyright (C) 1999 John Wiley & Sons, Ltd.