COVID-19 and the Prediction of 30-Day Hospital Readmissions: A Brazilian Retrospective Cohort Study Applying the Hospital Score Before and After the Pandemic.

Identifying preventable readmissions for healthcare systems facing significant financial strain from rehospitalizations is crucial. The Hospital score, recognized for its predictive power in diverse settings, has yet to be extensively validated in low-income countries, particularly amid the challenges posed by the COVID-19 pandemic. This study validates the Hospital score for predicting 30-day readmission in a tertiary private hospital in Southern Brazil, assessing its efficacy in a middle-income country context. We retrospectively analyzed adult medical discharges from Hospital Moinhos de Vento between January 2019 and December 2023. The study focus was evaluating the Hospital score's accuracy in forecasting 30-day readmissions, including a specific assessment of its performance among COVID-19 patients during the pandemic years. Data extraction from electronic health records facilitated a comprehensive analysis of score variables and patient outcomes. Among 69 971 discharges, 2890 patients were readmitted within 30 days, marking a 4.3% readmission rate. The Hospital score demonstrated consistent predictive accuracy, with an overall score accuracy of 0.70 (95% CI: 0.66-0.73; p < 0.01) and 0.72 (95% CI: 0.64-0.75; p < 0.01) among COVID-19 patients. Notably, the pandemic did not significantly impact the score's predictive validity. The Hospital score remains a reliable predictor of 30-day readmissions in a tertiary hospital within a middle-income country, retaining its effectiveness amidst the COVID-19 pandemic. These findings underscore the score's potential as a universal tool for reducing preventable readmissions, highlighting its relevance across healthcare contexts.