I thank the colleagues Grant and Boucher for their interest in my clinical insight article [1]. And I also thank them for giving me the opportunity to better clarify a concept already present in my paper [2]. Prospective and adequately sized randomized clinical trials (RCTs) are the main tool (together with their meta-analysis) for building evidence. Observational studies and even more so retrospective ones cannot provide solid and unquestionable conclusions, especially if the results of the RCTs do not go (and unequivocally) in their same direction.
