Figure 2

Illustration of minimum detectable effect size. The minimum detectable effect size (MDES) is the minimum difference between groups that yields a statistically significant result. The power of the study is calculated using the clinically relevant effect size (CRES). The density of the left curve to the right of the MDES represents the type I error, i.e. 2.5% (one-sided) in this case. The density of the middle curve to the right of the MDES represents the power for the MDES to be detected, i.e. 50%. The density of the right curve to the right of the MDES represents the power for the CRES to be detected, i.e. the power of such a study, which is 98% in this example. CRES, clinically relevant effect size; MDES, minimum detectable effect size.