An illustrative statistical analysis of cutoff-based randomized clinical trials

Cutoff-based randomized clinical trials (RCTs) are designed to balance ethical and scientific concerns. Patients scoring below a cutoff score on a baseline measure (i.e. the least severely ill) are assigned to the control-treated group, those scoring above a second cutoff score (i.e. the most severely ill) are assigned to the test-treated group, and those scoring within the interval (i.e. the moderately ill) are randomly assigned. This paper provides a formal illustration on the statistical analysis of cutoff-based RCTs using data from the Xanax Cross-National Collaborative Study. To overcome problems specific to cutoff-based designs, we generally recommend a backward elimination approach that tests interactions before main effects.