OBJECTIVES: To estimate the differential cost of using non-steroidal anti-inflammatory drugs (NSAIDs) in a Medicaid population. NSAID use is associated with adverse gastrointestinal (GI) events such as GI distress, symptomatic ulcer, and anemia, which vary in severity and may require symptomatic treatment or hospitalization. The economic burden of these events is represented by the differential cost of treating NSAID-related toxicities. METHODS: Treatment cost and adverse GI event rates were compared for two cohorts (NSAID and non-NSAID users) selected retrospectively from the South Carolina Medicaid population. To be included in the analysis, all subjects must have been at least 18 years old and continuously eligible for 12 months following an index event in 1995. For NSAID users, the index event was at least two NSAID prescriptions or a 60 day supply (n=21,278). For non-NSAID users (comparator group), the index event was at least one prescription during 1995 (n=152,072). Average cost per subject by type of service (physician, other ambulatory, inpatient hospital, outpatient hospital, emergency department, prescription drugs) was compared for the cohorts. Regression analysis was used to evaluate the effect of NSAID use on treatment cost, controlling for age, gender, prior exposure to NSAIDs, use of gastro-protective agents, and adverse GI events. RESULTS: NSAID users had higher physician ($1885 vs. $1488), other ambulatory ($5785 vs. $5060), inpatient hospital ($5966 vs. $3326) and prescription drug ($1012 vs. $738) costs than the non-NSAID group. Regression analysis showed that NSAID use was a significant explanatory variable (p <0.0001), as were age, gender, race, and adverse GI events. CONCLUSIONS: NSAID use is associated with higher average treatment costs and is a significant predictor of modeled treatment costs after controlling for other effects. Substantial cost savings may be realized if NSAID-related GI toxicities and adverse events can be managed or avoided.