Identifying Preventable 30-Day Hospital Readmissions

Abstract This chapter discusses the HOSPITAL score’s ability to predict potentially preventable 30-day readmissions among hospitalized medical patients. A study at a 750-bed academic medical center in Boston evaluated 9212 patient admissions, categorizing readmissions as avoidable or unavoidable through a computerized algorithm. Seven significant variables formed the prediction model, enabling clinicians to identify low-, intermediate-, and high-risk patients for readmission. The HOSPITAL score showed moderate discriminatory power, with a C-statistic of 0.69, and can help clinicians assess readmission risk. Finally, the chapter emphasizes the need for ongoing research to refine interventions for preventing readmissions and improving postdischarge care, as best practices continue to evolve.