A SAS ® Macro Approach to Simultaneous Superiority and Noninferiority Testing in Clinical Trials Analysis

Analysis of clinical trial data involves a variety of statistical techniques depending on the study design, sample size, type of data collected and how the data are distributed. Two types of trials, superiority and non-inferiority, involve specific hypothesis testing. In a superiority trial, demonstration that the test drug is superior to the control drug, on average, is the objective. Here, the null hypothesis tests that the difference between the test drug and control is zero. The alternative hypothesis tests that the difference between the test drug and control is greater than zero. Non-inferiority trials are commonly undertaken where the test drug and control are expected to be approximately equally effective. In this case, hypothesis testing looks for either greater effectiveness of the test drug over the control or equivalency of the test and control drug. Many articles exist that discuss superiority and non-inferiority testing. This paper focuses on a SAS macro-based approach to implementing hypothesis testing for superiority and non-inferiority trials in a closed 2-stage simultaneous testing procedure.