OBJECTIVES: To compare dosing stability and dose changes, through a retrospective database analysis, for rheumatoid arthritis (RA) patients taking infliximab over five years. METHODS: We utilized medical and pharmacy claims data from the Medstat MarketScan database. The sample consisted of “new starts” for infliximab users in the commercial and Medicare population: (a) with a diagnosis of Rheumatoid Arthritis (RA) (ICD-9 714.xx), (b) who had at least three documented administrations of infliximab between 1999 and 2005, (c) who had no prescription history for any TNF blocker for a 6 month period prior to the index infusion, and, (d) who were continuously enrolled for a minimum of 365 days after their index infusion. We examined the proportion of infliximab users who showed an increase, decrease, or no change in dose over time in both groups. RESULTS: Medicare infliximab users (n= 729) were older (mean age =73 years) with higher co-morbid scores (mean score = 9.71) than commercial users (n=1903, mean age = 50 years, mean score = 3.83). Following the index infusion, 30.4% of commercial and 17.7% of Medicare users showed no change; 24.6% of commercial and 43.1% of Medicare users showed a decrease; and 44.9% of commercial and 39.2% of Medicare members showed an increase in their dose. For the five-year period, the total average dose change between the first and last infusion was 24mg for the Medicare group and 44mg for the commercial group. Medicare members also had a lower starting dose than commercial members (328mg vs. 366mg). CONCLUSION: This study demonstrates that there is dose stability in RA patients treated with infliximab, as the dose changed on average, by less than 3% per year over the five-year period. Further research is needed to evaluate the impact of dose changes on clinical and economic outcomes.