Figure 2

Conventional versus multiplex trial design. In most conventional randomized controlled trials, the effect of only one intervention is examined. Multiple questions are typically addressed in a serial manner. Thus in the first trial patients 1-1000 are studied to compare red and brown treatments, in the second trial patients 1200 -1600 (green vs. dark green) the third trial patients 1800-2000 (yellow vs. orange) and in the fourth trial patients 2200-3200 (light blue vs. dark blue). The different numbers of included patients result from different power requirements. The gaps in included patients are related with logistical issues. Within the multiplex trial concept as many questions as possible are addressed simultaneously, provided clinical equipoise exists between all treatments (and their combinations) that are examined. For example the three different subtrials may examine the two interventions for blood pressure, pain and fever respectively in patient 2000 through 4000. Although the subtrials use more patients and thus have greater power, a lower overall number of patients is required, underscoring the efficiency of factorial trial design. The continuous nature of the multiplex design also reduces “down-time” (as indicated by the breaks between the four convential trials). Interactions between treatments can only reliably be detected with a factorial design, although the needed sample size to do so requires advanced analysis.