OBJECTIVE: To evaluate the frequency of upper gastrointestinal (GI) tract adverse events associated with risedronate during two (2-year) randomized, double-blind, parallel-group, placebo-controlled studies.PATIENTS AND METHODS: Male and female patients aged 40 to 80 years with mild to moderate medial-compartment knee osteoarthritis were enrolled. Data were pooled and analyzed for risedronate at 5 mg and at 15 mg once daily and compared with placebo. The results of the once-weekly dosages (35 or 50 mg) were assessed separately.RESULTS: A total of 2483 patients were randomized: 622 to placebo, 628 to risedronate at 5 mg/d, 609 to risedronate at 15 mg/d, 310 to risedronate at 35 mg once weekly, and 314 to risedronate at 50 mg once weekly. During the study, 77% of patients were regular nonsteroidal anti-inflammatory drug (NSAID) and/or analgesic users (defined as those who took medication >= 3 days per week), and 68% were regular NSAID users. The number of upper GI tract adverse events was similar between treatment groups, with no dose-related response: 161 for placebo, 176 for risedronate at 5 mg/d, and 150 for risedronate at 15 mg/d. The time to the first upper GI tract adverse event was similar between treatment groups. There was no difference in the frequency of upper GI tract adverse events in risedronate-treated patients compared with patients who were regular users of NSAIDs or NSAIDs and/or analgesics. Findings were similar for those in the once-weekly risedronate groups.CONCLUSION: The results of this study show that risedronate regimens at 5 mg/d or 15 mg/d as well as once weekly at 35 mg or 50 mg are not associated with an increased frequency of upper GI tract adverse events, even in patients who have an increased risk for such events.