Excess short- and long-term mortality risk following any and hip fractures: a prospective matched cohort study using exposure density sampling

Evidence on the magnitude and duration of short- and long-term mortality following osteoporotic fractures was mixed. We aimed to examine the short- and long-term mortality risk after any and hip fractures. We conducted an "exposure density sampling" dynamically matched cohort study based on 363,884 adults from the UK Biobank, with a 1:4 (case: control) matching on age, sex, and frailty status. A piecewise Cox proportional hazards model was used to estimate mortality risk in different time periods from 30 days to 10 years after fractures. For any fracture, the nested analysis cohort included 19,163 cases and 67,522 matched controls, with a mean (SD) age of 58.3 (7.8) years and 59.5% of the cohort being women. During follow-up, men had a higher post-fracture mortality rate than women (15.85 vs. 8.58 per 1,000 person-years). After adjustment, mortality risk peaked within the first 30 days following any fractures (HR = 23.31, 95% CI, 17.08-31.83), gradually declining but remaining elevated for up to 10 years (1.44, 1.08-1.91). For hip fracture, the analysis comprised 3,114 cases and 12,070 matched controls. A similar temporal pattern was observed, with the highest mortality risk occurring within the first 30 days (23.67, 12.17-46.01), followed by a gradual attenuation, though the high risk persisted for up to 10 years (2.15, 1.15-4.00). The above associations were consistently observed across various characteristics of the participants. This study underscores the significant association between fracture incidence and both short- and long-term mortality, highlighting the potential value in comprehensive fracture management strategies - from prevention to long-term care - though future research is needed to establish causality and evaluate effective interventions.