Risk factor profiles for early and delayed mortality after hip fracture: Analyses of linked Australian Department of Veterans' Affairs databases

Abstract

Introduction One-year mortality after hip fracture may exceed 30% with a very large number of reported risk factors. Determinants of mortality beyond 1 year are rarely described. This study employs multiple data linkages to examine mortality rates, risk factor profiles and age-specific excess mortality at intervals from 30 days to 4 years. Methods Retrospective cohort study of linked administrative datasets describing hospital episodes, residential aged care (RAC) admissions and date of death for 2552 Australian veterans and war widows hospitalised for hip fracture in 2008-09. Associations between time to death and patient age, sex, pre-fracture accommodation, fracture type, treatment options, selected comorbidities and complications were tested in Cox proportional hazards models. Results In a population with mean age of 86.6 years (range 54-100 years), overall death rate was 11% at 30 days, 34% at 1 year, 47% at 2 years and 67% after 4 years. For males hospitalised from RAC 1-year mortality was 72%, contrasting with 19% for females from the community. Risk of death within 1 year was increased by male sex, increasing age, pre-fracture RAC residency, transfer to intensive care and coexistent cancer, cardiac and renal failure, cerebrovascular disease and pressure ulcers. Patients selected for rehabilitation had lower mortality rates. Patterns of determinants for mortality changed over time. Above-expected age-specific mortality was sustained for 4 years except for males 90 years and older. Conclusion Pre-fracture RAC residence was the strongest determinant factor for mortality. Patients selected for rehabilitation had lower mortality rates. The profiles of explanatory variables for death altered with increasing time from the index fracture event.