Low or minimal doses of aspirin are widely used for prevention of cardiovascular diseases. Aspirin is known to produce severe adverse gastrointestinal effects, such as bleeding and perforation. Less is known about the risk associated with minidose aspirin. Our aim was to assess the possible association of upper gastrointestinal tract bleeding with minidose aspirin therapy. A retrospective controlled design was used. Patients hospitalized for melena or hematemesis between January 1, 2000, and December 31, 2001, were identified by ICD-9 codes, and their clinical findings were compared to these of patients without upper gastrointestinal bleeding hospitalized during the same period and matched for age and sex. Bleeding was attributed to therapy if patients used a nonsteroidal anti-inflammatory drug or aspirin therapy within 30 days before hospitalization. The study group included 318 patients (59% male), and the control group 141 (65% male). Mean ages were 67 ± 19 and 64 ± 19 years, respectively. Study patients had more accompanying diseases, used more medications, and required more blood transfusions than controls (37%, vs. 2% of controls; P < 0.001). Minidose aspirin was used by 28% of the study group and 18% of the controls (P = 0.03). The average dose was 40 ± 86 and 21 ± 55 mg/day, respectively (P = 0.012). Only 26% of the study patients received a gastric protective agent. On multivariate analysis, aspirin consumption was the only independent risk factor for upper gastrointestinal tract bleeding. There appears to be an association between minidose aspirin treatment and hospitalization for upper gastrointestinal tract bleeding. Despite the advanced age of the patients, only one-quarter were treated with gastric protective agent.