Objective: To provide guidance on how to manage the concepts of equivalence and noninferiority in the context of systematic reviews.
Methods: This guidance was developed by a workgroup consisting of 13 individuals from seven Evidence-based Practice Centers (EPCs) and the Agency for Healthcare Research and Quality, under the leadership of the ECRI EPC. Prior to developing any guidance, the lead EPC also performed two methods projects intended to assist the workgroup. The first was a review of 12 existing guidance documents pertaining to equivalence and noninferiority, all of which were intended for primary researchers. The second project assessed the methodology used for a random sample of 50 recent systematic reviews that concluded equivalence or noninferiority between two or more treatments. Based on the previous experience and knowledge of the workgroup members, as well as insights from the two methods projects, guidance was developed and underwent posting for public comment and peer review.
Results: No guidance documents specifically addressing equivalence and noninferiority in the context of systematic review were identified. The workgroup developed a list of recommendations for four areas. First, how to assess the unique risk of bias for trials self-identifying as equivalence or noninferiority trials. Second, how to set Minimum Important Difference for a systematic review. Third, the analytic foundations for concluding equivalence or noninferiority in a systematic review. Fourth, language considerations when concluding equivalence or noninferiority in a systematic review.
Conclusions: Systematic reviewers need to adopt a consistent and conceptually sound approach to interpreting, concluding, and expressing equivalence or noninferiority in the context of systematic reviews. This paper provides preliminary guidance in that endeavor.