Background:​ This study aimed to investigate the potential association of exposure to metformin therapy with the risk of hip fracture in adult patients with type II diabetes. We included patients with diabetes who were registered in the 2010 sample cohort database of the National Health Insurance Service in South Korea.

Methods:​ The patients who had been prescribed continuous oral metformin therapy for a 1-year period in 2010 were defined as the metformin group, while those who were not prescribed metformin during the same period were classified as the control group. The primary endpoint of this study was the development of hip fracture between January 2011 and December 2015.

Results:​ A total of 64,878 patients (31,300 patients in the metformin group and 32,439 patients in the control group) were included in this study. Among those, 1655 patients (2.6%) had experienced a hip fracture. After a propensity score matching, a total of 37,378 patients (18,689 patients in each group) were included in the analysis. Using a time-dependent Cox regression analysis on the propensity score-matched cohort, the exposure to metformin was not significantly associated with the development of hip fracture compared to the control group (hazard ratio:​ 1.00, 95% confidence interval:​ 0.86 to 1.16;​ P = 0.985). Similar results were observed using sensitivity analysis of a multivariable time-dependent Cox regression model of the entire cohort (hazard ratio:​ 0.78, 95% confidence interval:​ 0.36 to 1.69;​ P = 0.525).

Conclusions:​ This population-based cohort study in South Korea showed that there was no significant association between the exposure to metformin therapy and hip fracture in patients with type II diabetes mellitus.