Background/Purpose: The multi-biomarker disease activity (MBDA) score is a validated test used to assess disease activity for patients with rheumatoid arthritis (RA). How it is used in clinical practice in the U.S. is unclear. We evaluated the likelihood that rheumatologists would add or switch biologic therapies based on the MBDA test result.

Methods: Using previously published methods, we linked results of MBDA tests obtained as part of routine clinical care to 2012-2014 Medicare fee for service claims data for RA patients. We characterized patients as being on a biologic or targeted synthetic DMARD in the 90 days prior to the MBDA test and evaluated biologic/tofacitinib treatment changes in the 90 days following the MBDA test. MBDA test scores were classified as low (<30), moderate (30-44), and high (>44). The unit of analysis was the 90-day interval before and after each MBDA test score. Alternating logistic regression was used to compute odds ratios (OR) to quantify the likelihood that patients made any change (add or switch), accounting for the clustered nature of the data (intervals nested within patients, and patients nested within doctor practices) and physician-level variability, controlling for patient age and sex. Sensitivity analyses used a 6-month interval for outcome ascertainment after the MBDA test.

Results: Using previously validated methods, a total of 27,621 unique RA patients were linked to 44,438 MBDA test scores. For the 27,256 intervals where RA patients were not on biologic therapy when the MBDA score was obtained, a total of 13.2% of patients added a biologic. Patients with high MBDA scores were significantly more likely to add a biologic (Table). For the 17,182 intervals where RA patients were already on a biologic, a total of 19.1% of patients switched or stopped the biologic that they were taking. Patients with lower MBDA scores were significantly more likely to stay on their therapy, whereas those with higher scores were more likely to stop and/or switch biologics.

After adjustment, results from the regression analyses showed that patients with moderate MBDA scores were 1.47 (95% 1.29-1.67)-fold more likely to add or switch biologics, and those with high MBDA scores were 2.54 (95% CI 2.19-2.94)-fold more likely to add or switch biologics. Men (OR=0.90, 95% 0.82-0.98) and older patients (OR=0.92 per 5 year increment, 95% CI 0.91-0.93) were less likely to add or switch therapy, even after controlling for variability between physicians (OR = 1.10, 95% CI 1.02-1.19). These results were robust and ORs were numerically larger when extending the interval to 6 months.

Conclusion: Results from the MBDA score were significantly associated with the likelihood that a physician added or switched biologic therapies, with either type of change being more frequent when the MBDA score was high. Further evaluation of outcomes after switching, conditional on the MBDA score, is warranted.