Construction of a Semi-Naive Model to Predict Early Readmission of COPD Patients by Using Quality Care Information

Le résumé fourni par la source

The quality of inpatient care in emergency rooms has proved a major factor to be taken into account upon hospital admission. Several studies show that early readmission of Chronic Obstructive Pulmonary Disease patients after discharge can be predicted according to information on the assistance quality received during the patient's hospital stay, however, these studies use many variables to be captured, from admission to discharge dates, which makes it difficult, or even impossible, to follow the track and record all the information. We propose an algorithm to construct a Semi-Naive model capable of both selecting and aggregating attributes related to inpatient care quality to result in a selection of very few attributes, while overcoming the performance of other early readmission prediction models.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.