Rethinking Evidence-Based Medicine in the perioperative safety scenario.

Abstract

EditordWe read with great interest the recent Editorial by Webster inwhich he invites one to think beyond the traditional randomised controlled trial (RCT) in patient safety studies. We work on patient safety in the perioperative setting and have some additional comments regarding Evidence-Based Medicine (EBM). Scientific output and evidence in the field of patient safety has increased dramatically in recent years. Research on the efficacy of the interventions to decrease unnecessary risks related with patient safety has some particularities, clarified by Brown and colleagues: First, patient safety interventions are often complex interventions that require carefully planned evaluation and development. Sometimes interventions and outcome evaluations are difficult to blind. Second, patient safety interventions are often delivered to and implemented with groups of subjects (or clusters i.e. surgical patients, hospitals, certain populations) rather than at an individual level, in an equally complex environment, such as health organisations. Third, patient safety interventions are often expected to do more good than harm, implying that professional equipoise may be absent. Therefore, traditional study designs such as a parallel RCT may not be ethically acceptable. Finally, if patient safety research refers to interventions to prevent harm, countable outcomes are rare events and in many times difficult to assess. They require large multicentre studies and collaboration with increased cost and logistics. Prospective RCTs are considered the research design of choice to evaluate the efficacy of health interventions providing the most robust evidence. Nevertheless, interventions in patient safety are not suitable to be studied using only this approach. In patient safety, studies on interventions (using randomisation) are less common than large observational population studies of the incidence and causes of medical errors, and the number of RCTs and systematic reviews of randomised trials published in specialised journals are scarce. At a first glance, this could be interpreted as a lack of strong evidence, but as Leape and colleagues suggest in