Hip fracture has a high mortality rate, but the actual level of long-term excess mortality and its impact on population-wide mortality remains controversial. The present prospective study investigated short- and long-term excess mortality after hip fractures with adjustment of other risk factors. We calculated the population attributable risk proportion (PARP) to assess the impact of each risk factor on excess mortality. We recruited 217 elders with hip fractures and 215 age- and sex-matched patients without fractures from the geriatric department of the same hospital. The mean follow-up time was 46.1 months (range: 35 to 57 months). We recorded data on 55 covariates, including baseline details about health, function, and bone mineral density. We used the multivariate Cox proportional hazards model to analyze hazard ratios (HRs) of short-term (<12 months follow-up) and long-term (≧12 months follow-up) excess mortality for each covariate and calculated their PARP. Patients with hip fractures had a higher short-term mortality than non-fractured patients, and the long-term excess mortality associated with hip fracture remained high. The significant risk factors for short-term mortality were hip fracture, comorbidities, and lower (below cutoff) Mini Mental State Examination score with HRs of 2.4, 2.3, and 2.3, respectively. Their PARPs were 44.7%, 38.1%, and 34.3%, respectively. The significant risk factors for long-term mortality were hip fracture (HR: 2.7; PARP: 48.0%), lower T-score (HR: 3.3; PARP: 36.2%), lower body mass index (HR: 2.5; PARP: 42.8%), comorbidities (HR: 2.1; PARP: 34.8%), difficulty in activities of daily living (HR: 1.9; PARP: 31.8%), and smoking (HR: 2.5; PARP: 19.2%). After comprehensive adjustment, hip fracture was a significant risk factor and contributed the most to long-term as well as short-term excess mortality. Its adequate prevention and treatment should be targeted.

Background: Osteoporotic hip fractures with a significant morbidity and excess mortality among the elderly have imposed huge health and economic burdens on societies worldwide. In this age- and sex-matched case control study, we examined the risk factors of hip fractures and assessed the fracture risk by conditional logistic regression (CLR) and ensemble artificial neural network (ANN). The performances of these two classifiers were compared. Methods: The study population consisted of 217 pairs (149 women and 68 men) of fractures and controls with an age older than 60 years. All the participants were interviewed with the same standardized questionnaire including questions on 66 risk factors in 12 categories. Univariate CLR analysis was initially conducted to examine the unadjusted odds ratio of all potential risk factors. The significant risk factors were then tested by multivariate analyses. For fracture risk assessment, the participants were randomly divided into modeling and testing datasets for 10-fold cross validation analyses. The predicting models built by CLR and ANN in modeling datasets were applied to testing datasets for generalization study. The performances, including discrimination and calibration, were compared with non-parametric Wilcoxon tests. Results: In univariate CLR analyses, 16 variables achieved significant level, and six of them remained significant in multivariate analyses, including low T score, low BMI, low MMSE score, milk intake, walking difficulty, and significant fall at home. For discrimination, ANN outperformed CLR in both 16- and 6-variable analyses in modeling and testing datasets (p?<?0.005). For calibration, ANN outperformed CLR only in 16-variable analyses in modeling and testing datasets (p?=?0.013 and 0.047, respectively). Conclusions: The risk factors of hip fracture are more personal than environmental. With adequate model construction, ANN may outperform CLR in both discrimination and calibration. ANN seems to have not been developed to its full potential and efforts should be made to improve its performance.