Assessing equivalence and noninferiority

Jonathan R. Treadwell, Stacey Uhl, Kelley Tipton, Tatyana Shamliyan, Meera Viswanathan, Nancy D. Berkman, Xin Sun, Craig I. Coleman, Adam G. Elshaug, Sonal Singh, Shi Yi Wang, Rema Ramakrishnan

Research output: Contribution to journalArticlepeer-review

Abstract

Objective: For systematic reviews, no guidance exists for what review methods support valid conclusions of equivalence (EQ) and noninferiority (NI). To provide such guidance, we convened a workgroup of 13 experienced systematic reviewers from seven evidence-based practice centers (EPCs) and the Agency for Healthcare Research and Quality (AHRQ). Study Design and Setting: The Lead EPC first performed two methods projects intended to assist the workgroup in clarifying the context, prioritizing the issues, targeting the scope, and summarizing the state of the art. Results: Based on expert opinion, we devised guidance in four areas: 1) Unique risk of bias issues for trials self-identifying as EQ-NI trials; 2) Setting the reviewer's minimum important difference; 3) Analytic foundations for concluding EQ or NI; and 4) Language considerations when concluding EQ or NI. Conclusion: This article summarizes the main recommendations, and the full guidance chapter appears on the AHRQ Web site.

Original languageEnglish (US)
Pages (from-to)1144-1149
Number of pages6
JournalJournal of Clinical Epidemiology
Volume65
Issue number11
DOIs
StatePublished - Nov 2012

Keywords

  • Equivalence
  • Methodology
  • Noninferiority
  • Risk of bias
  • Strength of evidence
  • Systematic reviews

ASJC Scopus subject areas

  • Epidemiology

Fingerprint Dive into the research topics of 'Assessing equivalence and noninferiority'. Together they form a unique fingerprint.

Cite this