Tips for Effectively Utilizing Real-World Data

Abstract

In general, randomized controlled trials (RCTs) provide us the best evidence of a treatment. However, it is almost impossible to conduct an RCT in areas such as rare diseases, intractable diseases and medical devices. Moreover, in some post-marketing circumstances, evidence from RCTs is not well generalizable to other patient populations. Therefore, the real-world evidence of the treatment in broad patient groups is required to support decision-making in clinical practice and public health policy. There are several ways to utilize real-world data; data sources for a prevalence/marketing survey, a feasibility survey of a clinical trial, a survey of possible patients to be enrolled in the clinical trial, a survey for designing the study protocol, a post-marketing surveillance study, and an external control group. There were some examples of demonstrating evidence of efficacy of an intervention by combining clinical trial and patient registry data. This paper explores tips for effectively utilizing real-world data, e. g. use not a historical control but concurrent control, adjust for confounding by indication, and identify the real-world data sources when developing a trial protocol. The new rigorous study designs utilizing real-world data will be enhanced. In the current situation, patient cohort-based trials may provide best evidence as the synthesis of a thesis (RCT evidence) and its antithesis (real-world evidence).