Utilization Patterns of Disease-Modifying Antirheumatic Drugs for the Treatment of Rheumatoid Arthritis: Rationale for Improving the Harmonization of Coverage Criteria

Utilization patterns of csDMARDs were highly comparable between drug plans overall (in decreasing order: methotrexate, hydroxychloroquine, sulfasalazine, leflunomide, and azathioprine). The proportion of csDMARDs were comparable (e.g., approximately 30% of csDMARD use for methotrexate), although differences in coverage criteria may have resulted in variances in the use of leflunomide. Differences in adjudication of coverage criteria may have resulted in a modest variance in the number of csDMARDs used prior to initiating bDMARDs (i.e., allowing for an “early escape” to bDMARDs for some jurisdictions such as Manitoba and the Atlantic provinces). The mean time to initiate bDMARD therapy (range of 664 to 792 days) revealed a divergence between jurisdictions into 2 groupings whereby Manitoba, Saskatchewan, and the Atlantic provinces drug plans (mean time of 664 to 681 days) saw the initiation of bDMARDs approximately 4 months faster versus other jurisdictions (British Columbia, Alberta, and Ontario, with a mean time of 748 to 792 days), possibly due to their coverage criteria not requiring 3 lines of csDMARDs therapy. Despite differences in the time to initiate bDMARDs, there was no notable difference in the persistence of bDMARDs 6 months after the initiation for any drug plan (61% to 76% range for patients 67 years of age and older). Utilization patterns of bDMARDs was highly comparable between drug plans (i.e., highest use with adalimumab, etanercept, and infliximab), although British Columbia and Manitoba were the only jurisdictions that saw decreasing costs per patient of bDMARDs over time, likely due to a higher uptake of biosimilars or other managed formulary strategies such as tiering.