Purpose: Early initiation of anti-osteoporosis medications (AOMs) is recommended for patients on long-term glucocorticoid (GC) therapy. This study aimed to clarify the real-world effectiveness of AOMs against incident hip and vertebral fractures in patients undergoing GC therapy using the nationwide health insurance claims database of Japan (NDBJ).
Methods: Patients aged ≥50 years who were prescribed GC (≥5 mg/day prednisolone or equivalent) for ≥90 days and who were followed up regarding AOM prescription and hip and clinical vertebral fracture incidences for the subsequent 1080 days between 2012 and 2018 were selected from NDBJ. Associations of AOMs prescribed within 90 days since GC therapy initiation with hip or vertebral fracture risk were evaluated by Cox proportional hazards regression using propensity score inverse probability weighting (IPW) for receiving any AOM or individual AOMs.
Results: In total, 96,475 women and 98,385 men were included in the analysis; 38.0 % of women and 27.6 % of men received AOMs. Patients who received any AOM and those who received bisphosphonates or denosumab had a significantly lower risk of hip and clinical vertebral fractures than those who received no AOM in both sexes after propensity score IPW. Teriparatide was associated with an increased risk of both fractures in women and an increased risk of clinical vertebral fractures in men. Selection biases such as confounding by indication might have caused an underestimation of AOMs' protective effects.
Conclusions: Bisphosphonates and denosumab were associated with a lower fracture incidence in patients on long-term GC therapy in real-world settings.
Keywords: Anti-osteoporosis medication; Glucocorticoid-induced osteoporosis; Hip fracture; Nation-wide health insurance claims database study; Retrospective cohort study; Vertebral fracture.
Copyright © 2022 Elsevier Inc. All rights reserved.