The purpose of this study was to compare contemporary risk of hip fracture in type 1 and type 2 diabetes with the nondiabetic population. Using a national diabetes database, we identified those with type 1 and type 2 diabetes who were aged 20 to 84 years and alive anytime from January 1, 2005 to December 31, 2007. All hospitalized events for hip fracture in 2005 to 2007 for diabetes patients were linked and compared with general population counts. Age- and calendar-year-adjusted incidence rate ratios were calculated by diabetes type and sex. One hundred five hip fractures occurred in 21,033 people (59,585 person-years) with type 1 diabetes; 1421 in 180,841 people (462,120 person-years) with type 2 diabetes; and 11,733 hip fractures over 10,980,599 person-years in the nondiabetic population (3.66 million people). Those with type 1 diabetes had substantially elevated risks of hip fracture compared with the general population incidence risk ratio (IRR) of 3.28 (95% confidence interval [CI] 2.52-4.26) in men and 3.54 (CI 2.75-4.57) in women. The IRR was greater at younger ages, but absolute risk difference was greatest at older ages. In type 2 diabetes, there was no elevation in risk among men (IRR 0.97 [CI 0.92-1.02]) and the increase in risk in women was small (IRR 1.05 [CI 1.01-1.10]). There remains a substantial elevation relative risk of hip fracture in people with type 1 diabetes, but the relative risk is much lower than in earlier studies. In contrast, there is currently little elevation in overall hip fracture risk with type 2 diabetes, but this may mask elevations in risk in particular subgroups of type 2 diabetes patients with different body mass indexes, diabetes duration, or drug exposure.
Bibliographical note© 2014 American Society for Bone and Mineral Research.
Acknowledgments: The authors acknowledge the work of the Scottish Diabetes Research Network–Epidemiology Group, and in particular Dr Colin Fischbacher of National Services Scotland.
Authors' roles: EJH, HMC, and SJL wrote the manuscript; SJL analyzed the data; HMC specified the analysis; SFA, SC, GPL, HCL, RSL, JM, DP, SP, and SHW were involved in data generation and edited the manuscript. HMC accepts responsibility for the integrity of the data analysis. This work was supported by the Wellcome Trust through the Scottish Health Informatics Programme (SHIP) Grant (WT086113).
- Aged, 80 and over
- Diabetes Complications/epidemiology
- Diabetes Mellitus, Type 1/epidemiology
- Diabetes Mellitus, Type 2/epidemiology
- Hip Fractures/epidemiology
- Middle Aged
- Retrospective Studies