Nonsteroidal anti-inflammatory drugs (NSAIDs) are frequently needed for the treatment of patients with arthritis. However, long-term use of such drugs that are cyclooxygenase-2 (COX-2) selective inhibitors has been reported to increase cardiovascular risk as compared with placebo, whereas long-term, randomized controlled trials assessing the risk of traditional NSAIDs versus placebo are lacking. The MEDAL program is designed to provide a precise estimate of the relative cardiovascular event rates with the COX-2 selective inhibitor etoricoxib in comparison to the traditional NSAID diclofenac in patients with osteoarthritis and rheumatoid arthritis. The MEDAL program consists of 3 multinational, randomized, double-blind trials in patients with osteoarthritis and rheumatoid arthritis comparing etoricoxib (60 or 90 mg daily) to diclofenac (150 mg daily). All investigator-reported thrombotic cardiovascular events will be adjudicated by an independent panel of experts blinded to treatment assignment. The primary analysis is a noninferiority comparison of etoricoxib versus diclofenac for confirmed thrombotic cardiovascular events, defined as an upper bound of the 95% CI for a hazard ratio of < 1.30. With the planned 635 observed events from approximately 40,000 patient-years of exposure, using an estimated annual event rate of 1.30% in the control arm, the maximum annual event rate for etoricoxib that would meet the noninferiority criteria would be approximately 1.46%, yielding a hazard ratio of 1.12. A total of 34,701 patients have been enrolled in the MEDAL program. Roughly 13,000 and 10,000 patients will, respectively, have had > or = 18 or > or = 24 months of exposure, with maximum exposure of approximately 40 months. The MEDAL program will help to better define the risk-to-benefit ratio of 2 NSAIDs, that differ in their selectivity for COX-2, notably diclofenac and etoricoxib.