Background: Use of non-steroidal anti-inflammatory drugs (NSAIDs) is recognized as an important cause of peptic ulcer complications. The aim of this nested case-control study was to identify risk factors for NSAID-related ulcer complications. Methods: Cases were consecutive NSAID users admitted with an ulcer complication (n = 118), and controls were a random sample of all NSAID users without ulcer complication identified by a pharmacoepidemiologic database (n = 540). Results: Ninety-four of 118 cases were interviewed, and 324 of 540 controls answered the questionnaire. Analysis showed no difference between included and non-included subjects. Risk factors for patients at start of NSAID therapy were high age: 60-75 years (odds ratio (OR), 3.5 (95% confidence interval (Cl), 1.8-7.1); >75 years (OR, 8.9 (4.3-18.3)); male sex (OR 1.7 (1.0-3.0)); ulcer history (OR 2.5 (1.2-5.1)); steroid treatment (OR 2.0 (0.8-4.6)); smoking (OR 1.6 (0.9-2.7)); and alcohol use (OR 1.8 (0.9-3.6)). Risk factors for patients receiving NSAID therapy were high age, male sex, ulcer history, smoking, and, furthermore, dyspepsia (OR 2.0 (1.0-4.2)), especially NSAID-related dyspepsia (OR 8.7 (4.0-18.9)). Risk was lower for patients treated more than 3 months. Conclusion: Risk measured from this design can be shown to correlate strongly with the rate difference, a measure that is more clinically relevant than conventional relative risk estimates. Strong risk factors for NSAID-related ulcer complication are high age, male sex, ulcer history, and dyspepsia related to the NSAID therapy. Avoiding NSAID therapy in these high-risk patients, whenever possible, might prevent many adverse events.