NEUROPATHOLOGICAL DIAGNOSES IN EPIDEMIOLOGIC STUDIES

This paper discusses the potential interest of including neuropathological examination in epidemiologic studies of Alzheimer's disease (AD). Neuropathological studies could provide estimates of both false-positive and false-negative error rates of clinical diagnosis in prevalence or incidence studies on AD. To reach this objective, large and representative autopsy samples are required. Another objective could be to analyse distributions of the various modifications of the brain tissue according to age. The validity of the neuropathological examination, and particularly the level of inter- and intra-observer agreement, must be further assessed before undertaking large autopsy studies in a population.