Characterization of Hospital Recurrence: Epidemiological Patterns, Intervention Assessment, and Predictive Modeling with Artificial Intelligence.

• Grant number: 25/09618-4
• Support Opportunities: Regular Research Grants
• Start date: February 01, 2026
• End date: January 31, 2029
• Field of knowledge: Health Sciences - Medicine - Medical Clinics
• Principal Investigator: Antônio Pazin Filho
• Grantee: Antônio Pazin Filho
• Host Institution: Faculdade de Medicina de Ribeirão Preto (FMRP). Universidade de São Paulo (USP). Ribeirão Preto , SP, Brazil
• City of the host institution: Ribeirão Preto
• Associated researchers: Maria Izaura Sedoguti Scudeler Agnollitto ; Mario Sérgio Adolfi Júnior
• Associated scholarship(s): 26/03484-9 - Clinical and healthcare delivery characteristics of recurrent patients in a tertiary hospital: a descriptive and comparative analysis of readmissions and emergency department revisits, BP.IC; 26/03541-2 - Evaluation of the impact of care delivery interventions on hospital recidivism: an interrupted time series analysis in a tertiary hospital, BP.IC

Abstract

Hospital recurrence- understood as the early reuse of healthcare services by the same patient after discharge - is a multifaceted phenomenon that directly contributes to overcrowding in high-complexity hospitals and reflects weaknesses in continuity of care. It encompasses distinct events, such as hospital readmissions and unplanned returns to the emergency department, each with its own clinical, care-related, and operational characteristics. Nevertheless, these events are often treated as a single indicator in research studies, which may compromise epidemiological analysis and limit the development of effective public health policies. This study proposes the use of the term "hospital recidivism" as a broad yet differentiated analytical category capable of capturing the complexity of these events. The study will be conducted at the Emergency Unit of the Hospital das Clínicas, Faculty of Medicine of Ribeirão Preto, University of São Paulo (UE-HCFMRP-USP) - a tertiary hospital that has used electronic health records for 15 years. Using both structured and unstructured data from the unit, we will apply artificial intelligence techniques, such as machine learning and natural language processing (NLP), to characterize recidivism events, assess the impact of institutional interventions on their rates, identify risk factors, and develop predictive models applicable to clinical practice. The results are expected to support improvements in both patient care and healthcare system management, thereby contributing to the better functioning of the health system. (AU)