Table 1 Percentage of simulated pilot studies, estimating differing true standardized effect sizes, in which a main trial would be favoured when the observed point estimate exceeded the minimum important difference (0.25). One hundred thousand simulations were performed for each scenario [22]
From: Should treatment effects be estimated in pilot and feasibility studies?
True standardized effect size | ||||
|---|---|---|---|---|
Below MID | Above MID | |||
0.00 | 0.20 | 0.50 | 0.80 | |
Pilot studies (%) indicating progression to main trial | 19 | 43 | 81 | 97 |