[Validity of the five-word screening test for Alzheimer's disease in a population based study].
INTRODUCTION: In general medicine lack of time impairs screening for Alzheimer's disease (AD). The five word test (FWT) enables rapid assessment of verbal episodic memory in accordance with Grober and Buschke neuropsychological concept. The main steps of the FWT are: induce specific semantic processing, control of encoding to avoid attention deficits, free and cued recall. Cued recall helps to distinguish a recall impairment from storage impairment which is evocative of AD. OBJECTIVE: Evaluate FWT total score (sum of free and cued recalls), FWT total weighed score which give a higher coefficient for free recalls than cued recalls and present the ability of these two scores for AD screening. METHOD: Evaluation performed with 4116 subjects (of whom 73 MA) aged from 65 years and more, randomly selected in two French towns for the "Three Cities" Study, a population-based cohort. RESULTS: The total score was more specific than sensitive with a maximal sensitivity (Se) at 63 percent with specificity (Sp) at 91.1 percent. The total weighed score significantly increased Se (83.6 percent) with control of specificity (84.9 percent) and positive predictive value (9.1 percent). CONCLUSION: The FWT allows quick screening of patients for whom further neuropsychological evaluation is needed to diagnose AD. The ability of is simple test to screen for AD is improved by a simple weighting procedure: the total weighted score.