Blinding, Placebo Effect, and GRADE Methodology—What Is Relevant?

Abstract

Placebo-controlled trials are commonly performed in medication studies. Due to ethical and practical concerns, this type of double blinding with the use of a "sham" procedure is not often seen or ethical in surgical research. 1 Non-placebocontrolled trials may run the risk of overestimating efficacy of procedural interventions by the nature of their design. In this issue of JAMA Surgery, Rajkumar and colleagues 2 recognized this problem and performed a systematic review and meta-regression to compare all available placebo-controlled trials on procedural interventions with their unblinded (non-placebo-controlled) counterparts. Although the systematic review by Rajkumar et al suffers from major clinical heterogeneity, significant larger standardized effect sizes are found for non-placebocontrolled trials than placebo-controlled trials of procedural interventions for exercise-related, quality of life, and health care professional-assessed end points. They point out the need for placebo-controlled trials for these types of end points, thereby reducing the potential for effect overestimation. Conversely, no significant placebo effect was found for patientreported end points, blood pressure, mortality, pain, and recurrent bleeding events.The implementation of placebo-controlled trials for exercise-related, quality of life, and health care professional-