OBJECTIVES: To characterize dosing patterns and utilization of tumor necrosis factor-alpha (TNF-alpha) inhibitors in rheumatoid arthritis (RA) patients as part of a quality assurance initiative on appropriate use of TNF-alpha inhibitors. METHODS: RA patients given infliximab or etanercept between January 2001 and May 2002 were identified (N = 936) using pharmacy and medical claims. Prescribers of the affected health plan were requested via letter to send copies of identified patients' charts. TNF-alpha inhibitor utilization data were abstracted from available charts. Outcomes measured included TNF-alpha inhibitor dosing, switching, and discontinuation. This retrospective study started prior to the compliance date of the Health Insurance Portability and Accountability Act (HIPAA); nevertheless, it was conducted in a manner that was compliant with HIPAA requirements. RESULTS: Charts were obtained for 457 (49%) patients; 369 were eligible for study enrollment (164 [44%] received infliximab, 205 [56%] received etanercept). Mean first doses were 3.5 mg/kg (infliximab) and 27.9 mg (etanercept). Subsequent doses were 4.2 mg/kg (infliximab) and 25.7 mg (etanercept). Overall, 67.2% of infliximab infusions used no more than the number of vials expected if dosed at 3 mg/kg. The infliximab cohort had a significantly lower switch rate than the etanercept cohort (5.2% versus 15.5%, p = 0.006). Discontinuation (i.e., no documentation of TNF-alpha inhibitor during the 70-day period prior to study end date or a chart order to discontinue TNF-alpha inhibitor) occurred among 63.3% patients receiving etanercept and 40.0% receiving infliximab (p = .0002). CONCLUSIONS: Most patients received acceptable doses of TNF-alpha inhibitor. A minority of patients receiving either medication experienced an increase in dosing. Patients receiving infliximab had lower switch and discontinue rates than patients receiving etanercept; further research is needed to examine reasons for this observation. These results can be used by managed care organizations to estimate pharmacy costs for TNF-alpha inhibitors.