RESUMEN
The Canadian Study of Health and Aging (CSHA) provided a population-based estimate of the prevalence of dementia of 8% for those aged 65 and older. Other studies have produced both higher and lower prevalence estimates. Factors that may contribute to these differences include: the use of or the reliance on neuropsychological testing, the consideration of functional impairment as a criterion for dementia and the inclusion of the category of cognitive impairment without dementia in the diagnostic classification. We examined the impact of these methodological factors by reanalyzing the CSHA database for those individuals who completed neuropsychological testing. If the diagnosis of dementia required only impaired neuropsychological test performance, there was an increased prevalence of dementia relative to the clinical consensus diagnosis, but including the requirement of functional impairment for dementia reduced this discrepancy. The findings illustrate the need for clear operationalization of diagnostic criteria for cognitive impairment and dementia in neuroepidemiological studies.