Abstract
We investigated the association between bisphosphonate and denosumab use and risk of hip fracture in Norway. These drugs protect against fractures in clinical trials, but their population-level effect is unknown. Our results showed lowered risk of hip fracture for treated women. Treatment of high-risk individuals could prevent future hip fractures. PURPOSE: To investigate whether bisphosphonates and denosumab reduced the risk of first-time hip fracture in Norwegian women when adjusting for a medication-based comorbidity index. METHODS: Norwegian women aged 50-89 in 2005-2016 were included. The Norwegian prescription database (NorPD) supplied data on exposures to bisphosphonates, denosumab, and other drugs for the calculation of the Rx-Risk Comorbidity Index. Information on all hip fractures treated in hospitals in Norway was available. Flexible parametric survival analysis was used with age as time scale and with time-varying exposure to bisphosphonates and denosumab. Individuals were followed until hip fracture or censoring (death, emigration, age 90 years), or 31 December 2016, whichever occurred first. Rx-Risk score was included as a time-varying covariate. Other covariates were marital status, education, and time-varying use of bisphosphonates or denosumab with other indications than osteoporosis. RESULTS: Of 1,044,661 women 77,755 (7.2%) were ever-exposed to bisphosphonate and 4483 (0.4%) to denosumab. The fully adjusted hazard ratios (HR) were 0.95 (95% confidence interval (CI): 0.91-0.99) for bisphosphonate use and 0.60 (95% CI: 0.47-0.76) for denosumab use. Bisphosphonate treatment gave a significantly reduced risk of hip fracture compared with the population after 3 years and denosumab after 6 months. Fracture risk was lowest in denosumab users who had previously used bisphosphonate: HR 0.42 (95% CI: 0.29-0.61) compared with the unexposed population. CONCLUSIONS: In population-wide real-world data, women exposed to bisphosphonates and denosumab had a lower hip fracture risk than the unexposed population after adjusting for comorbidity. Treatment duration and treatment history impacted fracture risk.