This case-control study investigates whether history of organic solvent exposure is associated with increased risk of Alzheimer's disease. The study base includes about 23,000 persons aged 60 years or more from the local membership of a health maintenance organization in Seattle, Washington, who entered the study between 1987 and 1992. Probable Alzheimer's disease cases (n = 193) who had presented with new dementia symptoms were identified, enrolled, and diagnosed by our Alzheimer's Disease Patient Registry following standardized criteria. Control subjects (n = 243), free of dementia and neurologic disease causing dementia, were selected randomly from the study base and frequency matched to cases for age and sex. Proxy informants provided specific solvent exposure history as well as job descriptions likely to involve solvent use as part of a comprehensive risk factor interview. Kappa statistics indicated substantial agreement for control-control proxy solvent responses. History of exposure to one or more solvent groups (benzene and toluene; phenols and alcohols; ketones; other solvents) yielded an adjusted Alzheimer's disease odds ratio of 2.3 (95 percent confidence interval 1.1-4.7); among males only, it increased to 6.0 (95% confidence interval 2.1-17.2). Thus, past exposure to organic solvents may be associated with onset of Alzheimer's disease.