Related Experiment Videos
Simplification of the HOSPITAL score for predicting 30-day readmissions.
Carole E Aubert1, Jeffrey L Schnipper2,3, Mark V Williams4
1Department of General Internal Medicine, Bern University Hospital, University of Bern, Bern, Switzerland.
BMJ Quality & Safety
|April 19, 2017
Summary
A simplified HOSPITAL score effectively predicts 30-day patient readmissions. This new version maintains accuracy while being easier to use, offering a practical alternative for healthcare settings.
Area of Science:
- Health Services Research
- Clinical Informatics
- Patient Safety
Background:
- The HOSPITAL score is validated for identifying high-risk patients needing transition care.
- Simplifying the HOSPITAL score could enhance its usability without compromising performance.
Purpose of the Study:
- To validate a simplified HOSPITAL score for predicting 30-day patient readmissions.
- To assess the performance of the simplified score in terms of accuracy, discrimination, and calibration.
Main Methods:
- Retrospective study across 9 hospitals in 4 countries (2011).
- Included consecutively discharged medical patients, excluding those deceased or transferred.
- Simplified the HOSPITAL score by modifying criteria and categorizing patients into two risk groups.
Main Results:
- The 30-day potentially avoidable readmission rate was 9.7%.
- The simplified HOSPITAL score demonstrated very good overall accuracy (Brier score 0.08) and good discriminatory power (C-statistic 0.69).
- Excellent calibration was observed between expected and observed risks in the two defined risk categories.
Conclusions:
- The simplified HOSPITAL score performs well in predicting 30-day readmissions.
- Prognostic accuracy is comparable to the original score, with improved ease of use.
- This simplified score presents a viable alternative to the original HOSPITAL score, adaptable to various clinical settings.