Outcomes after hip fractures sustained in hospital: A Propensity-Score Matched Cohort Study

Abstract Aims: The aim of this study was to compare outcomes following hip fracture for patients who sustained their fracture whilst in hospital (inpatients) with those who sustained their fracture in the community (outpatients). Patients and Methods: Data on all hip fracture admissions aged 65 years or over between 1st May 2007 and 31st March 2018 was analysed from a prospectively collected hip fracture database. Patient demographics, co-morbidities, and discharge information were analysed. Outcome measures included mortality (inpatient, 30-day and one year), surgical site infection (SSI) rate and mean length of stay (LOS). Baseline characteristics were used to generate propensity-match scores for each patient, with inpatients matched to outpatients in a 1:1 ratio. Outcomes were compared after matching. Results: 7,592 patients were included in the study. 338 were identified as having an inpatient hip fracture. There was a significantly greater level of comorbidity in the inpatient group at baseline. After propensity-score matching, there were 229 patients in the inpatient group and 222 in the outpatient group, with no significant difference in baseline co-morbidities. In this propensity score matched cohort, 30-day mortality was significantly higher in the inpatient group (16%) compared to the outpatient group (10%), P=0.049. 1-year mortality was also significantly higher in the inpatient group (44%) compared to the outpatient group (34%), P=0.03. There was no significant difference in inpatient mortality, mean LOS and SSI rates between the two groups. Conclusion: Patients who suffer a hip fracture whilst in hospital have significantly poorer outcomes than those who suffer a hip fracture whilst an outpatient, even after adjusting for co-morbidities. Dedicated guidelines are needed for this particularly vulnerable group.