Superiority trials are similar to non-inferiority trials, but instead of trying to show that one drug is no less safe and effective than another, the goal is to show that a drug is actually better than another, usually in terms of safety and/or efficacy but possibly also in terms of cost-effectiveness. In the simplest terms, superiority trials are designed to find a difference between two treatments. These are a type of head-to-head trial, like non-inferiority and equivalence trials. While the objective is clear—showing one thing is better than another instead of showing it’s at least as good—the way that pans out in the design can be a lot trickier.
Deeper dive
As this journal article explores in-depth, “the classification of superiority or non-inferiority trials is not always straightforward.” That article is critical of superiority and non-inferiority articles, so it could be a great place to get ideas of questions to ask the principal investigators or outside commenters regarding the design of a superiority trial you’re covering.
One of the most important aspects of a superiority trial is how the researchers calculate the power they need to detect a difference and the biostatistical methods they use to analyze the data. If there were ever a time to reach out to a biostatistician to get comment on a study, superiority trials (or any head-to-head trial, for that matter) are definitely the time. Since there is an explicit goal inherent in the design of a superiority trial, it’s particularly important to keep a lookout for possible bias in the study design, the interpretation of the data, and the conclusions from the data.