Coronary Stent Noninferiority Trial Design Biases Toward Noninferiority

Noninferiority trials can have design features that bias them toward concluding that an investigational treatment strategy is noninferior to an approved one. A noninferiority trial seeks to test whether a particular strategy is not worse than a comparator by more than a prespecified amount. Thus, the size of that amount, the noninferiority margin, takes on great importance. Large margins can lead trials to conclude that two strategies are not different even when the results show a substantial difference.

In a systematic review and meta-analysis, investigators sought to determine how the estimated event rates in coronary stent noninferiority trials affected the noninferiority margins and trial interpretation. They studied 58 trials and found …