Metformin use is associated with a lower risk of osteoporosis/vertebral fracture in Taiwanese patients with type 2 diabetes mellitus

Objectives To investigate the metformin effect on the risk of osteoporosis (OS) and/or vertebral fracture (VF). Methods We enrolled 14 611 pairs of metformin ever and never users matched on propensity score (PS) from Taiwan’s National Health Insurance database. All patients had new-onset type 2 diabetes mellitus (T2DM) during 1999–2005 and were free from OS and/or any fracture at the start of follow-up on January 1, 2006. They were followed up until December 31, 2011 for the incidence of OS/VF. Cox regression incorporated with the inverse probability of treatment weighting using PS was used in the main analyses. Results New-onset OS/VF was diagnosed in 1757 never users (median follow-up 5.0 years) and 1143 ever users (median follow-up 5.3 years). The respective incidence rates were 2870.97 and 1713.20 per 100 000 person-years. Two-thirds of the incident cases had OS without VF and the other third had VF. In main analyses, the hazard ratio for ever vs never users was 0.592 (95% CI: 0.550–0.638). In either sex, a dose–response pattern was noted and metformin therapy > 2 years was consistently associated with a lower risk. The protective effect attenuated with increasing age but remained significant in patients aged ≥ 80 years. In sensitivity analyses, metformin significantly reduced the risk of both OS and VF (with or without a prior OS) by 30–40%. Additional analyses showed a null association for other antidiabetic drugs, but significant interactions between metformin and insulin, sulfonylurea and pioglitazone, respectively, were noted. Conclusion Metformin use is associated with a lower risk of OS/VF.