OBJECTIVE: To evaluate nonsteroidal anti-inflammatory drug (NSAID) utilization among commercially insured members in a managed care environment. METHODS: This was a cross-sectional study of administrative pharmacy claims data from eleven health plans between July 1, 1994 and June 30, 1997. Data were used to determine the prevalence of prescription NSAID use. Analyses were performed on two groups: those with and those without a rheumatoid arthritis or osteoarthritis diagnosis, occurring < 90 days prior to their first NSAID claim during the study period. Members were required to be continuously enrolled with a drug benefit during the study period (N=149,216). Acute NSAID utilization (defined as <3 prescriptions during the study period), chronic NSAID utilization (defined as > 3 NSAID prescriptions during the study period), and average days supply were analyzed. RESULTS: The prevalence of NSAID use among continuously enrolled health plan members, of all ages, was 32% during the three-year study period. Among members with arthritis, 37% were considered acute NSAID users and 63% were considered chronic users. A 30 to 34 days supply of NSAIDs was most common for members with arthritis. Among members without arthritis, 71% were considered acute NSAID users and 29% were considered chronic users. A 10 to 15 days supply was most common for members without a recent arthritis diagnosis. CONCLUSION: The majority of NSAID users in the managed care population used NSAIDs on an acute basis. Consequently, use of the more costly, new class of NSAIDS, COX-2 inhibitors, which may reduce the incidence of gastrointestinal (GI) bleeding among those at increased risk of GI irritation due to chronic NSAID exposure, may not be warranted for most NSAID users.