The authors performed a systematic review to summarize the epidemiologic evidence on the association between use of nonsteroidal antiinflammatory drugs (NSAIDs) and the risk of dementia. A total of 25 case-control and cohort studies that reported an odds ratio/relative risk were included. Study-specific log relative risks were weighted by the inverse of their variances to obtain pooled relative risks and 95% confidence intervals. The authors divided the reports into studies with prevalent dementia cases, studies with incident dementia cases, and studies where cognitive decline was used as the clinical endpoint. The pooled relative risks of the three groups of studies were 0.51 (95% confidence interval (CI): 0.37, 0.70), 0.79 (95% CI: 0.68, 0.92), and 1.23 (95% CI: 0.70, 2.31), respectively. Within these subgroups, heterogeneity was present only in the studies with prevalent cases (p = 0.001). Because the benefit of NSAIDs in preventing dementia or cognitive impairment was 50% in studies with prevalent dementia cases, declined to 20% in studies with incident dementia cases, and was absent in studies where cognitive decline was used as the endpoint, the authors conclude that most of the reported beneficial effects of NSAIDs may result from various forms of bias: recall bias, prescription bias, and publication bias.