Key points are not available for this paper at this time.
Background: On July 18, 2017, a biosimilar health policy was implemented in British Columbia, Canada, mandating that all patients initiating a new biologic medication be prescribed a biosimilar, if available, in order for the cost to be covered under the universal provincial drug plan. Since then, the use of biosimilars has increased drastically. This study adds to the existing knowledge from randomized controlled trials by providing real-world evidence on the overall performance of biosimilars in routine clinical practice, and on the impact of the health policy at the health system level. Objectives: To examine the impact of biosimilar use by comparing healthcare utilization outcomes among new users of biosimilar vs. originator etanercept in inflammatory arthritis using a population-based fuzzy regression discontinuity design. Methods: In this retrospective cohort study using administrative health data for the entire population of British Colombia, Canada, we identified all incident users of etanercept aged ≥ 18, without prior prescription over 6 months, with rheumatoid arthritis (RA), psoriasis or psoriatic arthritis (PsO/PsA), or ankylosing spondylitis (AS) with initiation date between Jan 01, 2014 and Dec 31, 2020. Healthcare utilization over 1 year after etanercept initiation was assessed using five outcomes: (1) physician visits, (2) hospitalizations for any cause, (3) hospitalization for, or complicated by, an infection (based on diagnostic code in any position in hospitalization data), (4) length of hospital stays (summ of all days from admission to discharge date for all hospitalization episodes), (5) emergency room visits. Outcomes were standardized into rates for each patient, calculated by dividing the total number of outcomes accumulated by the follow-up time of each patient, to account for the censoring effect. Using generalized additive models combined with a regression discontinuity method, we compared risks of healthcare utilization in patients initiating etanercept immediately before and immediately after the policy change date (i.e., cut-off point), representing the intention-to-treat (ITT) effect. Additionally, we estimated the complier (new users following the health policy manadate) average causal effect (CACE) of biosimilar use on healthcare utilization outcomes using Poisson regression with the instrumental variable control function estimator method. Results: A total of 3242 incident etanercept users were identified. The proportion of biosimilar users increased significantly from 0.006 (9 out of 1626) to 0.812 (1312 out of 1616) from the pre-policy period to the post-policy period. According to both the ITT and CACE estimates, the biosimilar health policy, and the initiation of biosimilars, had no significant impact on healthcare utilization outcomes (Table 1 & Figure 1). Conclusion: In this population-based longitudinal study taking advantage of a policy change as a natural experiment, no significant impact of a health policy mandating biosimilar use or actual biosimilar use was observed on healthcare utilization among incident users of etanercept for inflammatory arthritis in British Columbia, Canada. Table 1. The intention-to-treat effect of policy change and the complier average causal effect of biosimilar use on healthcare utilization outcomes for incident etanercept users. REFERENCES: NIL. Acknowledgements: We would like to thank the Ministry of Health of British Columbia and Population Data BC for providing access to the administrative data. All inferences, opinions, and conclusions drawn in this publication are those of the authors, and do not reflect the opinions or policies of the Data Stewards. Disclosure of Interests: None declared.
Zhou et al. (Sat,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: