HaRP – Heart failure Readmission Prediction

Unscheduled readmissions create a burden for healthcare with increased cost and a negative impact on quality of care. Being able to predict the risk of unscheduled readmission following a discharge would be very helpful to initiate relevant interventions to optimize care and improve clinical and health outcomes.

Heart failure (HF) patients represent a large patient group where unscheduled readmission confers significant costs. A reliable risk assessment at discharge could trigger appropriate interventions to reduce the risk for readmission to a lower total cost and increased care quality for the patients. In a recent study from Halmstad University and Region Halland, a 30-day readmission prediction method was developed for HF patients, using a mixture of electronic health record (EHR) data available at discharge. The study demonstrated a high enough accuracy for the tool to be of clinical relevance.

The main goal of this project will be towards completion, calibration and practical implementation in practice of the 30-day readmission prediction model as a clinical decision support system (CDSS) for clinicians in several selected clinical end points for hospitalized HF patients at the time of discharge.