Fracture Risk Associated With Long‐Term Discontinuation of Postmenopausal Osteoporosis Treatment: A French Population‐Based Study Using National Claims Data (Système National des Données de Santé)

Abstract

Objective This retrospective cohort study evaluated the long‐term discontinuation (LTD) of osteoporosis treatments in postmenopausal women in France and associated fracture risk and economic burden. Methods From the national French health care database (Système National des Données de Santé), we included postmenopausal women aged ≥60 years initiating oral bisphosphonates, intravenous bisphosphonates, or denosumab between 2012 and 2015 with three or more (bisphosphonates) or two or more (denosumab) years of continuous treatment thereafter. LTD was defined as no treatment for ≥12 months. Within each treatment cohort (intravenous bisphosphonates, oral bisphosphonates, or denosumab), propensity score matching and Cox proportional hazard regression model were used to compare fracture risk in women with versus without LTD, and mean per capita annualized (direct and indirect) costs were assessed between women with and without LTD. Results Among 81,263, 19,111, and 28,606 eligible women initiating oral bisphosphonates, intravenous bisphosphonates, and denosumab, respectively, LTD incidence was 55.1%, 68.9%, and 42.5%. LTD of oral bisphosphonates was associated with a 12% increase in fracture risk (hazard ratio [HR] 1.12 [95% confidence interval (CI) 1.06–1.19]), and LTD of denosumab was associated with a 92% increase in fracture risk (HR 1.92 [95% CI 1.75–2.08]); no significant increase was observed for intravenous bisphosphonates. LTD increased mean per capita annualized costs by €446.6 for oral bisphosphonates and €747.2 for denosumab. Conclusion We observed high rates of LTD of osteoporosis treatments among postmenopausal women in France, with LTD of oral bisphosphonates increasing fracture risk by 12% and LTD of denosumab almost doubling fracture risk. These data underscore the need for careful LTD decision‐making, especially for denosumab.