FIGURE 2.

Optimized design parameters for single‐stage designs and the expected utility, averaged over the prior. Parameters are $a=(r_1,{\omega }_1)$ for enrichment trials and a = (r₁) for umbrella trials. Results are classified by ${\mu }_1$ and ${\mu }_2$, the prior means of ${\theta }_1$ and ${\theta }_2$, and the prior SD $\psi ={\psi }_1={\psi }_2$. The prior correlation between ${\theta }_1$ and ${\theta }_2$ is fixed at $\rho =0.5$ and the population prevalence of subgroup 1 is assumed to be $\lambda =0.3$ [Colour figure can be viewed at wileyonlinelibrary.com]