Electronic health records-based research in Cardiology: The time has come for pragmatic trials

Randomized clinical trials (RCTs) are the cornerstone of evidence-based medicine, as they minimize bias in the allocation of interventions. However, RCTs performed in a very selective population and overcontrolled conditions may impair the generalizability of results. Moreover, increasing running costs and regulatory complexity compromise the conduct of these studies. The need for pragmatic trial designs, with streamlined procedures and low running costs, will shape the short-term future of research in RCTs. Electronic health records (EHR) are routinely collected as part of the treatment of patients. These provide large amounts of data at no significant cost. The so-called "real-world data" are often used in observational studies with unavoidable bias. However, by combining the randomization of large numbers of patients with the data collected in EHRs, it is possible to answer very relevant clinical questions at a relatively low cost. In this review, we describe how the integration of EHR and randomization is fostering innovative approaches to the conduct of RCTs in Cardiology.