Fig. 3From: An adaptive two-arm clinical trial using early endpoints to inform decision making: design for a study of sub-acromial spacers for repair of rotator cuff tendon tearsDesign characteristics for two early looks. Estimated probabilities of stopping for futility and efficacy at the first and second looks, expected sample size (ESS) and overall study power, for effect sizes in range 0 to 10 for a\(\alpha ^{*}_{L}=(0.08,0.24,0.975)\), b\(\alpha ^{*}_{L}=(0.16,0.48,0.975)\), c\(\alpha ^{*}_{L}=(0.24,0.72,0.975)\) and d\(\alpha ^{*}_{L}=(0.32,0.96,0.975)\). Here \(\alpha ^{*}_{U}(1)=0\) and \(\alpha ^{*}_{U}(2)=0.001\), ρ=0.5; other settings are as in Table 1Back to article page