Skip to main content
. 2014 Apr 2;15:103. doi: 10.1186/1745-6215-15-103

Figure 3.

Figure 3

Improvement in the number of patients to recruit for ARD and BSD as compared to PGD under scenarios IIIA-C. Positive improvement indicates smaller sample sizes and thus higher efficiency. The sample size requirements for the PGD were 900, 692 and 236 for scenario IIIA-C, respectively. ARD, active run-in design; BSD, baseline selection design; PGD, parallel group design.