Modern Approaches to Clinical Trials Using SAS: Classical, Adaptive, and Bayesian Methods

10.6 Additional Considerations

10.6.1 More Than Two Treatments and Non-Binary Outcomes

Many RCTs involve more than two treatment arms. For instance, in a phase II dose ranging study, the treatment arms may be a placebo and/or an active control and several doses of an investigational drug. The study objectives may include estimation of various treatment contrasts and testing hypotheses of equality of treatment effects. Frequently, the optimal allocation ratio is unequal across treatment arms [53]. As noted in [33], “…the adaptive randomization light shines brightest in complicated multi-arm settings,” and indeed, optimal RAR designs are expected to provide more substantial gains over equal randomization in trials with more than two treatment arms ...