Calculate power for cohort with different levels of exposure

[simoneuv]

Hi!

Does anyone know a formula to calculate power for a cohort study, when you know the approximate sample size, estimated odds ratio for the association between exposure and outcome, and the ratio between exposed and unexposed participants? Also, is it possible to take into account different levels of exposure (i.e. 3 categories rather than "exposure" vs "no exposure").
I've tried using this one http://www.openepi.com/SampleSize/SSCohort.htm, but this one calculates the sample size rather than the power of the study.

Thanks!

 

[hlsmith]

Can you post the list values for what you are trying to do (e.g., odds ratio, prev, sample size etc.). Power can likely be gleaned from a simulation. What is the hypothesis as well?