A prospective cohort study comparing two predictor models for 30-day emergency readmission in the elderly.

Abstract

Aim: To undertake a prospective study of the efficacy of two models (LACE and BOOST) in predicting unplanned hospital readmission. Methods: Data were collected from a single centre prospectively over a continuous 30-day period on all patients over 75 years old admitted to the acute medical unit. The primary outcome was the area under the curve for both models. Results: Area under the curve were calculated for both tools with BOOST score 0.667 (95% CI: 0.559-0.775, p=0.005) and C-statistic for LACE index 0.685 (95% CI: 0.579-0.792, p=0.002). Conclusion: In this prospective study, both the BOOST and LACE scores were found to be significant predictive models of hospital readmission. Recent hospitalisation was found to be the most significant contributing factor. Key Words: Elderly, prediction, readmission