[Skip to Content]
[Skip to Content Landing]
Comment & Response
February 20, 2019

Unifying Design and Analysis for Superiority and Noninferiority Trials With Appropriate End Point

Author Affiliations
  • 1Division of Cardiology, Brigham and Women’s Hospital, Boston, Massachusetts
  • 2Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts
  • 3Department of Biostatistics, Harvard University, Boston, Massachusetts
JAMA Surg. 2019;154(5):466-467. doi:10.1001/jamasurg.2018.5852

To the Editor Acuna et al1 claim that, unlike a superiority study, interpretation of noninferiority studies is not straightforward, ie, the result from a noninferiority trial should be classified as noninferior, inconclusive, or not noninferior, but the result from a superiority trial is positive or negative. We respectfully disagree with this statement. In fact, one can design and analyze a study via 95% confidence intervals without prespecifying a superiority or inferiority hypothesis. That is, at the design stage, one may define a prespecified confidence interval width (eg, 10% for a study assessing laparoscopy for surgical resection1) as the required precision for estimating between-group difference and calculate the sample size accordingly. This width can be chosen from clinical considerations and/or feasibility of conducting the study. At the analysis stage, one then uses the observed confidence interval to claim superiority, noninferiority, inconclusive results, or inferiority for the new treatment, depending on where this interval is located, eg, items 1, 3, 5, and 8, respectively, in Figure, A.1 We claim that item 5 is inconclusive, since its confidence interval is very large, indicating insufficient information for decision making. On the other hand, the confidence interval for item 3 is sufficiently narrow for claiming noninferiority or equivalence. This unified approach seems more appealing than that under either the superiority or noninferiority trial hypothesis testing setting.

Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    ×