30-day readmission score after cardiac surgery

Abstract

Background

Hospital readmissions are costly and potentially avoidable. This study aimed to develop and validate a clinical score to predict early readmission.

Methods and results

Retrospective analysis of prospectively gathered data over a 10-year period of all patients undergoing cardiac surgery at the Buenos Aires Cardiovascular Institute (n = 5148). Thirty-day readmission was defined as all cause readmission. The scoring model was developed in a randomly selected test set (n = 2529) and was validated on other 2567 patients. Its predictive accuracy was compared by their area under the receiver operating characteristic curve (ROC). P = 0.05 was used for all test. The observed overall frequency of 30-day readmission in the entire sample was 11.7%. The final predictive score ranges between 0 and 6 points. The 30-day readmission rate at each score level in the validation set fell within the 95% confidence intervals of the corresponding frequency in the test set. This demonstrates the validity of the predictive model at all levels of the early readmission score. The frequency of early readmission across all levels in the test set ranged between 4.7 and 24.9%. The area under the ROC curve for the score in the test was 0.663 (95% CI 0.631 to 0.695) and was similar to that in the validation set 0.639.

Conclusions

The present study provides a clinical score to predict early readmission after open-heart surgery and validates that score in a comparable population, which can help in planning future interventions to avoid unnecessary readmissions.