An ontology-based system to predict hospital readmission within 30 days

Background: Unplanned readmission is one of the common issues in healthcare organizations. Most of the hospital readmission cases can be prevented and avoided. Owing to the complexity of healthcare data and processes, decision-making becomes uneasy task. Objective: The objectives of this study are to develop an ontology-based system for readmission within 30 days to identify the patient at risk and to score the predicted level of readmission risk. Methods: The proposed ontology-based system uses data of admitted patients to assess the risk factors. In the study, we created two main concepts that are patient profile and readmission risk factors. We focused on three risk factors that are length of stay, number of previous hospitalization, and number of administered medications. According to the patient medical information, a score is associated for each of the above factors. Result: According to the score, the ontology performs the risk assessment on a patient profile and predicts the potential risk level of readmission within 30 days. Conclusion: Predicting readmission probability helps caregivers to provide needed services and prevent readmission. In addition, it allows policymakers to eliminate unnecessary admission. The convenient of web-based healthcare applications and ontology will help patients and their families to prepare and prevent unwanted readmission.