Objective: To systematically review the modeling approaches and quality of economic analyses comparing cycling tumor necrosis factor inhibitors (TNFi) to swapping to a therapy with a different mode of action in patients with rheumatoid arthritis whose initial TNFi failed.
Methods: We searched electronic databases, gray literature, and references of included publications until July 2017. Two reviewers independently screened citations. Reporting quality was assessed using the Consolidated Health Economic Evaluation Reporting Standards (CHEERS) statement. Data regarding modeling methodology were extracted.
Results: We included 7 articles comprising 19 comparisons. Three studies scored ≥16 of 24 on the CHEERS checklist. Most models used a lifetime horizon, took a payer perspective, employed a 6-month cycle length, and measured treatment efficacy in terms of the American College of Rheumatology improvement criteria. We noted possible sources of bias in terms of transparency and study sponsorship. In the cost-utility comparisons, the median incremental cost-effectiveness ratio was US $70,332 per quality-adjusted life-year for swapping versus cycling strategies. Rituximab was more effective and less expensive than TNFi in 7 of 11 comparisons. Abatacept (intravenous) compared to TNFi was less cost-effective than rituximab. Common influential parameters in sensitivity analyses were the rituximab dosing schedule, assumptions regarding disease progression, and the estimation of utilities.
Conclusion: Differences in the design, key assumptions, and model structure chosen had a major impact on the individual study conclusions. Despite the existence of multiple reporting standards, there continues to be a need for more uniformity in the methodology reported in economic evaluations of cycling versus swapping strategies after TNFi in patients with rheumatoid arthritis.
© 2019, American College of Rheumatology.