Objective: This study investigates the effects of benzodiazepine (BZ) use on cognitive performance in elderly persons in a longitudinal design. Study design and setting: Data were obtained from the Longitudinal Aging Study Amsterdam (LASA), in the Netherlands. 2, 105 respondents (≥62 years of age) were included and had repeated measurements over a period of 9 years. For all BZs the type, dosage, frequency and duration of use was measured. The equivalent of a dose of diazepam was determined with regard to type and dosage and a cumulative dosage was calculated. General cognitive functioning was measured with the Mini-Mental State Examination, information processing speed was measured with the coding task, fluid intelligence with Raven's Coloured Progressive Matrices and episodic memory with the Auditory Verbal Learning Test. Multilevel analyses were used to investigate the relationship between BZ use and cognitive decline. Results: A negative effect of BZ use on cognitive performance was found. However, the effect sizes were very small. Conclusion: This study suggests that both duration and cumulative exposure to BZ has a small negative effect on the long-term cognitive functioning of elderly people in the community.