The Primary Outcome Is Positive — Is That Good Enough?

Abstract

There is a natural tendency to simplify the findings of a clini- cal trial into a binary conclusion: “Was there a positive outcome — or not?” In order to address this question with some objectivity, attention is typically focused on whether the prespecified measure of success for the primary T outcome has been met — that is, whether a P value of less than 0.05 has been achieved for the difference in treatments. In reality, a more nuanced interpretation requires a thorough examination of the totality of the evidence, including secondary end points, safety issues, and the size and quality of the trial. In this article, which focuses on the evaluation of “positive” studies — as in our previous article,1 which focused on the appraisal of “negative” studies — our intent is to facilitate a more sophisticated and balanced interpretation of trial evidence. Again, we make our points using examples from trials involving cardiovascular disease (our area of expertise), but the messages can be easily applied to other subject areas.