Moreira,Mafalda Barata Cardoso2026-02-122026-02-122025http://hdl.handle.net/10400.5/117030Trabalho de projeto de mestrado, Bioinformática e Biologia Computacional, 2025, Universidade de Lisboa, Faculdade de CiênciasPrimary Health Care (PHC) presents the initial point of contact between citizens and the healthcare system. It focuses primarily on routine and preventive services, playing a vital role in disease prevention, health promotion, and ongoing treatment. In contrast, hospital healthcare is more specialized and involves intensive or urgent interventions. Emergency Departments (EDs), in particular, are critical in addressing acute health conditions, as they assess and treat urgent, traumatic, or life­threatening situations. The ageing population and the rising prevalence of chronic diseases have increased healthcare needs and negatively impacted access to PHC. As a result, many patients seek faster and more convenient care at EDs – even in non­urgent situations. This inappropriate use of emergency services negatively impacts patient care and strains the Portuguese National Health Service (SNS), hindering its ability to manage true emergencies effectively. Therefore, greater integration between PHC and EDs is essential. This study aimed to identify and better understand the profile of patients who seek emergent care, particularly those classified as non­urgent. Two complementary analyses were conducted: (1) a descriptive analysis using Power BI to characterize the population using EDs, and (2) the development of a Machine Learning (ML) clustering model to identify patterns in patient profiles. The findings revealed minimal differences between urgent and non­urgent patients. Most individuals seeking emergent care were adults enrolled in PHC, assigned a family doctor, and with multiple comorbidities, specially related to circulatory or mental health conditions. Overall, the results highlight the need for stronger coordination between PHC and ED services. Future work should focus on deeper and more robust data exploration to enhance patient stratification and resource planning.application/pdfengPrimary Health CareEmergency DepartmentsPatient ProfilesMachine LearningClusteringmaster thesis204173426