Predictors of hospital readmission in a cohort of 236 elderly discharged after surgical repair of hip fracture: one-year follow-up

Abstract

Few studies have investigated predictors for hospital readmission after hip fracture repair. In a prospective cohort study we evaluated factors associated with early (within 3 months) and late (between 3-12 months), single and multiple hospital readmission in 236 hip-fractured older adults admitted to an orthopedic unit. Baseline patient characteristics and hospital course (functional and cognitive status, comorbidity, type of fracture, time to surgery, in-hospital stay, complications) were recorded. Hospital readmission over 12 months and ICD-9 principal diagnosis were ascertained from administrative sources. Functional status at the end of the rehabilitation program was assessed by telephone interviews. Seventy-one patients (30.1%) were readmitted to hospital within twelve months of discharge and 22 (9.3%) had two or more readmission. The total number of readmissions was 105, 43 (41%) occurred in the first three months. The most common readmission causes were cardiac, infectious and cerebrovascular; surgical complication accounted for 5.7%. Patients with a single readmission, like those with multiple readmissions, were sicker (CIRS-CI subscore 4.0+/-1.8 vs 3.2+/-1.6, p=0.010) and more functionally impaired at the end of rehabilitation (2 months' Katz index 2.1+/-2 vs 2.9+/-2.3, p=0.007) than controls. In a multiple logistic regression model, comorbidity and functional status at the end of rehabilitation were the only factors associated with the risk of readmission. Subjects at high risk of readmission can be reliably assessed, since few significant variables were associated with rehospitalization. Subgroups of patients with an elevated risk of rehospitalization after hip fracture may be the target for strategies to reduce the burden of excessive hospital use and improve overall outcomes.