Upper gastrointestinal tract symptoms are common in the elderly and, despite a paucity of data, nonsteroidal antiinflammatory drugs (NSAIDs) are believed to be important risk factors. We aimed to evaluate the association of NSAIDs with dyspepsia and heartburn in a population-based study. An age- and gender-stratified random sample of Olmsted County, Minnesota, Caucasian residents aged 65 years and older was mailed a valid self-report questionnaire; 74% responded (N = 1375). Age- and gender-adjusted (to 1980 US Caucasian population) prevalence rates for NSAID use, dyspepsia (defined as pain located in the upper abdomen or nausea), and heartburn (defined as retrosternal burning pain) were calculated. Logistic regression analysis was used to estimate the association of dyspepsia and heartburn with potential risk factors adjusting for age and gender. The age- and gender-adjusted annual prevalences (per 100) of aspirin and nonaspirin NSAID use were 60.0 (95% CI 57.2, 62.7) and 26.1 (95% CI 23.6, 28.7), respectively. The annual prevalences of dyspepsia and heartburn were 15.0 (95% CI 12.9, 17.0) and 12.9 (95% CI 10.9, 14.8), respectively. Aspirin was associated with dyspepsia and/or heartburn (OR = 1.6, 95% CI 1.2, 2.2) as were nonaspirin NSAIDs (OR = 1.8, 95% CI 1.3, 2.6), but smoking and alcohol were not significant risk factors. Aspirin and nonaspirin NSAIDs are associated with almost a twofold risk of upper gastrointestinal tract symptoms in elderly community subjects.