Background/Purpose: To estimate the incremental cost-effectiveness of etanercept plus methotrexate versus a triple regimen of disease-modifying anti rheumatic drugs (methotrexate, sulfasalazine, and hydroxychloroquine) over 24 weeks and 48 weeks in patients with active rheumatoid arthritis despite methotrexate therapy (RACAT).

Methods: In this double blind, noninferiority trial 353 patients were randomized to etanercept plus methotrexate or a triple regimen. After 24 weeks of treatment patients not achieving a DAS28 improvement of 1.2 were switched in a blinded fashion to the other therapy. Quality Adjusted Life Years (QALYs) were estimated using US societal values from the EQ-5D instrument which was measured every 24 weeks. Costs of drugs, hospitalizations, procedures, tests, visits and lost productivity were prospectively tracked and monetized from a societal perspective in 2014 US dollars. Incremental cost-effectiveness ratios were calculated using standard procedures assuming an intent-to-treat analysis, with missing data analyzed using multiple imputation and uncertainty assessed using bootstrapping.

Results: Both strategies showed significant improvements in EQ-5D, with etanercept providing marginally more accumulated QALYs (0.358 vs 0.354 over 24 weeks, and 0.742 vs 0.726 over 48 weeks for etanercept and triple regimen strategies respectively).  The etanercept strategy accumulated substantially higher drug costs even considering the switches between treatements at 24 weeks ($11,286 vs $369 cumulative costs from 0 to 24 weeks, and $19,625 vs $3,721 cumulative costs from 0 to 48 weeks for etanercept and triple regimen respectively).  The differences in other health care and productivity costs across strategies were negligible.  The resultant incremental cost-effectiveness ratios for etanercept vs. triple regimen were $2.7 million/QALY (95%CI 0.87 to ∞) gained over 24 weeks and $0.95 million/QALY (95%CI 0.41 to ∞) over 48 weeks.

Conclusion: This economic evaluation based on a prospective tracking of resource use and QALY measurement in a blinded, randomized trial demonstrates that the additional costs associated with using etanercept prior to a triple regimen does not provide good value for money at generally acceptable willingness to pay thresholds. A limitation of the study is its short time frame. However, even when considering the long-term perspective, since the incremental benefits are so small, even under the most optimistic scenarios imaginable, etanercept has only a small probability of being cost-effective compared to triple therapy. Given the opportunity cost associated with all health care spending, adapting a triple regimen prior to etanercept would free up scarce health dollars for use on alternative health care interventions that provide greater health benefits.