Determinants of reduced survival following hip fractures in men

Abstract

To assess determinants of poor survival after hip fractures in men, a population-based cohort study was conducted among 131 men in Rochester, MN, who had their first hip fracture during the period from 1978 to 1989, and an equal number of age-matched control men from the community. One hundred nine patients with fractures died during 373 person-years of followup, but only 75 control men died during 742 person-years of observation. The risk of dying increased with the level of comorbidity among hip fracture cases (hazard ratio 3.2; 95% confidence interval, 1.2-8.2), as well as with age (hazard ratio, 1.4 per 10-year increase; 95% confidence ratio, 1.1-1.8) and mental confusion during hospitalization (hazard ratio, 4.2; 95% confidence interval, 2.5-6.9). Discharge to a nursing home and low activity status also were predictors of death in the univariate analysis. Excess mortality among men with hip fractures can be explained best by interaction of the fracture with serious underlying medical conditions.