OBJECTIVES: In recent years, methodologies used in outcomes research have advanced. In a field where different disciplines interact frequently, a tool to communicate information clearly and effectively through graphical means has become a necessity. Using data visualization techniques, we present treatment patterns of patients diagnosed with rheumatoid arthritis (RA). METHODS: Using the 100% national Medicare data with Part D information, patients who are over age 65 and have had 2 diagnoses for RA at least 60 days apart were selected for the study. We identified patients who initiated therapy with tumor necrosis factor (TNF) agents. For 2 years after the initiation of TNF, we examined treatment patterns such as switching to another TNF, switching to a non-TNF, and drug discontinuation. Using a processing language, we created a data visualization tool to illustrate changes in treatment patterns after first, second and the third switches. RESULTS: A total of 50,455 RA patients initiated therapy with a TNF. 4.37% of these patients switched to another TNF, 5.81% switched to a non-TNF, 61.52% discontinued therapy and 28.31% continued their initial therapy. Among patients who switched to another TNF, 37.36% remained on the switched therapy, 46.71% discontinued, 8.40% switched to another TNF, and 7.54% switched to a non-TNF. In 2 years after TNF initiation, 7.03% of patients who switched twice after the initial TNF switched a third time to another TNF, while 51.35% continued their second switched drug, 35.68% discontinued their second switched TNF, and 5.95% switched to a non-TNF.  CONCLUSIONS: Treatment patterns can be difficult to present, especially if they were analyzed for subsequent years and switches. Data visualization tools can assist researchers in the effective presentation of these complicated flows.