Real-world differences in denosumab persistence, reinitiation, and switching among cohorts of older adults in Canada and the United States

作者：Kaleen N. Hayes, Selvam R Sendhil, Sulbh Aggarwal, Andrew R. Zullo, Sarah D. Berry, Arman Oganisian, Michael Adegboye, Suzanne M. Cadarette · 发表于：JBMR Plus · 年份：2025 · DOI：10.1093/jbmrpl/ziaf061 · 被引用次数：8 · 研究领域：Pancreatic and Hepatic Oncology Research

Denosumab is an injectable osteoporosis medication administered twice per year. Discontinuation of denosumab can result in rapid rebound fractures, but the evidence is limited on real-world persistence with denosumab. We conducted 2 parallel, population-based cohort studies leveraging (1) healthcare administrative data from Ontario, Canada (ON; 100% population) and (2) a 20% random sample of US Medicare beneficiaries (US). The first denosumab claim (US: 1/2010-12/2019; ON: 1/2012-12/2021) was identified using pharmacy claims (ON) and Medicare Parts D and B claims (US). Patients aged <66 yr, residing in long-term care (LTC), or with implausible data (eg, death before first claim) were excluded. We developed and applied an algorithm that used dosing and days between dispensations to clean denosumab claims. We assumed a days supply of 183 d for each dispensation and defined discontinuation as a 60-d gap in coverage. We estimated initial persistence, reinitiation, and switching to other osteoporosis medications using Kaplan-Meier estimators, censoring on death, disenrollment (US only), LTC admission, or study end (12/31/2022 [ON], 12/31/2020 [US]). We also estimated the monthly proportion of patients with an on-time denosumab dose to explore time trends. We identified 168 339 eligible individuals in ON (mean age = 78 yr; 90% female) and 97 595 in the US (mean age = 77 yr; 90% female). In ON, the median time to denosumab discontinuation was longer (median 2.3 yr [ON] vs 1.7 yr [US...