A CRITIQUE OF CURRENT CRITERIA FOR EARLY DEMENTIA IN EPIDEMIOLOGIC STUDIES

Differences between age-associated memory impairment and mild dementia have yet to be clarified. Attempts to validate the diagnosis of mild dementia have raised considerable difficulties in the absence of a 'go ld standard'. Agreed criteria of mild dementia are still lacking. Prog ressive deterioration of cognition is generally held to be a cardinal feature of dementia. Many of the multiple factors which can impair cog nition and give a spurious impression of mild dementia still present m ajor methodological problems. Agreed diagnostic criteria, use of stand ardized measuring instruments, clinical interview, key informant inqui ry and possibly psychometric testing are highly desirable in order to detect early dementia. For epidemiological surveys reliable screening instruments are important. It is concluded that, at present, there are too many fundamental conceptual and methodological problems to be ove rcome before we can successfully undertake large-scale community epide miological studies of mild dementia.