Internal medicine patients are mostly elderly with multiple comorbidities, usually chronic. The high prevalence of comorbidity and multimorbidity has a significant impact on both positive responses to treatment and the occurrence of adverse events. Clustering is the process of nosography grouping into meaningful associations with some index disease, so that the objects within a cluster have high similarity in comparison with one another. In the decision-making process it is imperative that, in addition to understanding the immediate clinical problems, we are able to explicit all the contextual factors that have to be taken into account for the best outcome of care. Cluster analysis could be leveraged in developing better interventions targeted to improve health outcomes in subgroups of patients.

Data from our study confirm that atrial fibrillation is a common finding in patients hospitalized in Internal Medicine units, and this population is characterized by multiple comorbidities and severe prognosis. Discrepancies exist between recommendations by guidelines and real-world management, owing to the complexity of patients and limits of existing treatment strategies.