We reviewed the findings of the Canadian Study of Health and Aging in
the context of studies published between January 1986 and June 1993 th
at documented dementia and Alzheimer's disease prevalence. Studies wer
e identified using a MEDLINE literature search. Additional references
were selected from the bibliography of identified articles Most report
s of all types of dementia prevalence are within a narrow range for ea
ch of the age groups 65+, 75+ and 85+ years. By contrast, two recent r
eports on the prevalence of Alzheimer's disease have reported much hig
her estimates (10.3% and 15.3%) in the elderly (65+ years). A variety
of threats to both validity and generalizability of the estimates are
present in all studies. In community studies which employed clinical i
nterviews most subjects were only mildly affected, the natural history
of impairment of this group requires further study if the consequence
s of these findings are to be understood. There is important variabili
ty in the definition of the functional consequences of cognitive impai
rment in the elderly which affects both the diagnosis and staging of d
ementia.