CONTEXT: 3-Hydroxy-3-methylglutaryl coenzyme A reductase inhibitors (statins) have been associated with a decreased risk for Alzheimer's disease (AD). OBJECTIVE: To evaluate the association between statin use and AD adjusted for comorbid medical conditions. DESIGN: A nested case-control study. PATIENTS: Patients at the Veterans Affairs Medical Center in Birmingham, Ala., USA with a new diagnosis of AD (cases) between 1997 and 2001 (n = 309) and age-matched non-AD controls (n = 3,088). MAIN OUTCOME MEASURE: Odds ratio for association between AD and statin use. RESULTS: Statin users had a 39% lower risk of AD relative to nonstatin users (odds ratio 0.61, 95% confidence interval 0.42-0.87). This association appeared to be modified by the presence of certain chronic medical conditions (i.e., hypertension, ischemic heart disease and cerebrovascular disease) in that the reduced risk was observed among those with these diseases, whereas no association was observed among those without any of these conditions. CONCLUSIONS: In this study, following adjustment for confounding factors, a statistically significant inverse association between statin use and AD was observed. The results lend support to looking at AD outcomes in trials of statins to further evaluate their possible beneficial effects.