Predicting readmissions: poor performance of the LACE index in an older UK population.

INTRODUCTION: interventions to prevent hospital readmission depend on the identification of patients at risk. The LACE index predicts readmission (and death) and is in clinical use internationally. The LACE index was investigated in an older UK population.
METHODS: randomly selected alive-discharge episodes were reviewed. A LACE score was calculated for each patient and assessed using receiver operator characteristic (ROC) curves. A logistic regression model was constructed, compared with the LACE and validated in a separate population.
RESULTS: a total of 507 patients were included with a mean (SD) age of 85 (6.5) years; 17.8% were readmitted and 4.5% died within 30 days. The median LACE score of those readmitted compared with those who were not was 12.5 versus 12 (P = 0.13). The Lace index was only a fair predictor of both 30-day readmission and death with c-statistics of 0.55 and 0.70, respectively. Only the emergency department visit was an independent predictor of readmission, with a c-statistic of 0.61 for readmission. In a validation cohort of 507 cases, the c-statistic of the regression model was 0.57.
CONCLUSION: the LACE index is a poor tool for predicting 30-day readmission in older UK inpatients. The absence of a simple predictive model may limit the benefit of readmission avoidance strategies.