In a noninferiority trial, which statement correctly describes the analysis goal?

• A. To demonstrate that the new device is superior to the comparator.
• B. To demonstrate that the new device is not worse than the comparator by more than a pre-specified margin.
• C. To establish equivalence within a zero-margin difference.
• D. To compare safety only, ignoring efficacy.

Answer: (B)

The main idea is that a noninferiority trial is designed to show the new device isn’t unacceptably worse than the standard by a pre-set amount. In the analysis, you look at the difference in outcomes between the new device and the comparator and its confidence interval. If the entire interval stays above the negative of the pre-specified noninferiority margin, you conclude noninferiority. The margin is a clinically justified threshold for how much worse the new device could be while still being acceptable, perhaps for reasons like easier use, cost, or safety trade-offs.

That’s why this choice is correct: it captures the goal of proving the new device isn’t worse by more than the allowed amount.

Why the other ideas don’t fit: noninferiority isn’t about proving the new device is better (superiority), though some trials may test for superiority in a secondary analysis. It isn’t about proving equivalence with a zero-margin difference, which would require demonstrating the difference lies within a symmetric, fixed range around zero. And it isn’t limited to safety outcomes alone—noninferiority typically concerns efficacy or overall clinical outcomes, though safety can be part of the broader assessment.