Association between ten complications and mortality : insights from the Dutch Hip Fracture Audit

This study examined the association between post-admission complications and 120-day mortality using data from a national hip fracture registry. A sub-analysis explored the impact of improved complication registration on these associations.

We conducted a cohort study including patients (1 January 2018 to 31 December 2023) from the Dutch Hip Fracture Audit (DHFA). Ten complications (anaemia, delirium, fall incident, heart failure, pneumonia, pressure ulcer, pulmonary embolism (PE), renal dysfunction, urinary tract infection, wound infection) were assessed. From 2018 to 2019, seven hospitals were recruited for additional data collection for the DHFA to improve and develop quality indicators. A multivariate Cox regression model was used to calculate the hazard ratio (HR) between complications and mortality at 30, 60, and 120 days. Subgroup analysis assessed the effect of better registration.

A total of 86,524 patients were included, with a median age of 81 years (IQR 72 to 87). The overall complication rate was 25.7% (n = 22,210), anaemia was the most common complication (n = 8,027; 9.3%), followed by delirium (n = 7,836; 9.1%). In subgroup analysis, project hospitals had a higher overall complications rate (38.8% vs 27.4%; p < 0.001). After adjustment for confounders, at 30 days, heart failure (HR 2.45 (95% CI 2.15 to 2.75)), delirium (HR 1.28 (95% CI 1.19 to 1.38)), PE (HR 2.62 (95% CI 2.11 to 3.27)), renal dysfunction (HR 2.20 (95% CI 1.96 to 2.47)), and pneumonia (HR 2.35 (95% CI 2.15 to 2.56)) were significantly associated with an increased hazard of death. In the subgroup analysis, fewer complications were associated with an increased hazard of death in the project hospitals in comparison to the non-project hospitals.

Five complications were significantly associated with an increased risk of mortality at 30 and 60 days. These findings highlight the importance of closely monitoring and managing complications. Improved registration practices in project hospitals were linked to fewer complications that showed a relation with mortality. Complications without a clear association with mortality may be more suitable for short-cycle in-hospital feedback rather than a national audit.