Minimum detectable effect size for main outcomes (accounting for sample
design and clustering)
Given that we do not have a baseline statistics for this research given its novelty, we conducted a few power calculations using the stata command “power”. For example,
* Assuming that the proportion of participants who exhibit either RIC, IA, or SSM is 0.2, we ran the following command: “power twoprop .2, n(400) p(0.8).” The minimum detectable effect size for main outcome = 0.1227.
* Assuming that the proportion of participants who exhibit either RIC, IA, or SSM is 0.35, we ran the following command: “power twoprop .35, n(400) p(0.8).” The minimum detectable effect size for main outcome = 0.1378.
* Assuming that the proportion of participants who exhibit either RIC, IA, or SSM is 0.5, we ran the following command: “power twoprop .5, n(400) p(0.8).” The minimum detectable effect size for main outcome = 0.1383.
Thus, we should be able to detect changes in the proportion of participants who exhibit RIC, IA, or SSM if the difference is greater than 0.14 in the third, sixth, seventh, eighth, ninth, and tenth step above.