This thesis aims to provide practical insights into the Linked Data approach to epidemiology. The insights are based on the preliminary results of a real-world study at the Pediatric Hospital of the Medical University of Warsaw which was concerned with investigating various medical and social risk factors on COVID-19 severity in children and SARS-CoV-2 spread dynamics. The thesis delivers results regarding the influence of various comorbidities on COVID-19 severity, based on a case of 90 patients of the Hospital, and establishes that metabolic disorders had a statistically significant influence on the hospitalization length in the observed sample (alpha = 0.05). It discusses the role of Linked Data in obtaining that result and presents ways to reuse that approach to enable similar studies. An overview of the theoretical framework for the paradigm is presented and its practical implications are discussed. Additionally, the thesis provides a technical report of the implemented data processing pipeline.