Risk factors for hip fractures occurring in the hospital

Risk factors for in-hospital hip fractures among patients aged ≥65 years were evaluated in a population-based case-control study conducted in the Canadian province of Saskatchewan from 1983 through 1985. There were 129 cases with confirmed in-hospital hip fractures and 234 controls, who were those controls from a previous study in this setting hospitalized on their assigned index date. Study variables were abstracted from the hospital chart. Six factors were independently associated with a significantly increased risk of in-hospital hip fracture: impaired vision (odds ratio = 1.97, 95% confidence interval (Cl) 1.18-3.30), assisted ambulation (odds ratio = 2.12, 95% Cl 1.25-3.59), confusion (odds ratio = 2.48, 95% Cl 1.37-4.48), psychotropic drug use (odds ratio = 2.02, 95% Cl 1.22-3.33), lowest weight tertile (odds ratio = 2.86, 95% Cl 1.38-5.92), and prior in-hospital fall (odds ratio = 2.71, 95% Cl 1.52- 4.82). The risk increased substantially with the number of factors present, from an odds ratio of 4.08 (95% Cl 1.56-10.67) for one factor (reference group, no factors) to 82.84 (95% Cl 18.6-368.7) for four or more factors (ρ <0.001, test for trend). These data underscore the multifactorial etiology of in-hospital hip fracture and suggest that prevention programs need to target multiple risk factors. Am J Epidemiol 1994;140:830-8.