Influence of premature mortality on the link between type 2 diabetes and hip fracture: The fremantle diabetes study

Abstract

Context: Studies of hip fracture complicating diabetes have not considered the effect of premature mortality. Objective: The aim of our study was to determine influence of the competing risk of death on the association between type 2 diabetes and hip fracture. Design: The study was designed as a longitudinal observational study. Setting: The study setting was an urban community. Patients: Participants included 1291 patients with type 2 diabetes (mean age 64.0 years) and 5159 matched residents without diabetes. Main Outcome Measures: Primary outcome measures were incident hip fracture hospitalizations and deaths. Hip fracture risk was assessed using proportional hazards and competing risk regression modeling. Results: During a mean of 14.1 years of follow-up, the incidence rate ratio for first hip fracture hospitalization in participants with vs without diabetes was 1.33 [95% confidence interval (CI), 1.05 to 1.68; P = 0.013]. Type 2 diabetes was associated with a cause-specific hazard ratio (csHR) for hip fracture of 1.50 (95% CI, 1.19 to 1.89; P < 0.001) and a subdistribution hazard ratio (sdHR) of 1.21 (95% CI, 0.96 to 1.52; P = 0.11) after adjustment for age, sex, and comorbidities. In patients with diabetes, significant csHRs for incident hip fracture were male sex (protective), body mass index (protective), insulin use, and renal impairment. These variables, with increasing age, also had significant sdHRs. Conclusions: The diabetes-associated risk of hip fracture is attenuated after allowing for the competing risk of death. Risk factors for hip fracture in diabetes were those in reported in general population studies plus insulin use.