Treatment patterns and long-term persistence with osteoporosis therapies in women with Medicare fee-for-service (FFS) coverage

Osteoporosis, a chronic disease, requires long-term therapy. In Medicare-insured women, denosumab persistence was higher than oral bisphosphonate persistence over up to 3 years of follow-up. Longer-term persistence was higher among women who persisted in the first year of therapy. Osteoporosis, a chronic, progressive disease, requires long-term therapy; this study assessed long-term persistence with anti-resorptive therapies in postmenopausal women. This retrospective cohort study used administrative claims for women with data in the 100% Medicare osteoporosis sample who initiated (index date) denosumab, oral/intravenous (IV) bisphosphonate, or raloxifene between 2011 and 2014 and who had ≥ 1 year (zoledronic acid: 14 months) of pre-initiation medical/pharmacy coverage (baseline). Persistence was assessed from index date through end of continuous coverage, post-index evidence of censoring events (e.g., incident cancer), death, or end of study (December 31, 2015). The study included 318,419 oral bisphosphonate users (78% alendronate), 145,056 denosumab users, 48,066 IV bisphosphonate users, and 31,400 raloxifene users; mean age ranged from 75.5 years (raloxifene) to 78.5 years (denosumab). In women with at least 36 months of follow-up (denosumab N = 25,107; oral bisphosphonates N = 79,710), more denosumab than oral bisphosphonate initiators were persistent at 1 year (73% vs. 39%), 2 years (50% vs. 25%), and 3 years (38% vs. 17%). Persistence decreased over time for all treatment groups, with denosumab users having the highest persistence in every follow-up time interval at or after 18 months. Women using denosumab, oral bisphosphonates, or raloxifene who persisted in a given year were more likely to remain persistent through the subsequent year. Denosumab users persisted longer with therapy than women using other anti-resorptive medications, including oral bisphosphonates. Early persistence may predict long-term persistence. Overall persistence with osteoporosis medications is suboptimal and may impact fracture risk. Efforts to improve first year persistence are needed.