I am working on the sample size determination for cross-over study. The aim of this study is to measure reliability of the measurements of outcome variable (continuous). For example, There will be 50 -150 clinicians in total, but probably, say, 10% will agree to participate in the study. Each clinician will get a chance to look at each patient and estimate the mortality rate two times at different time periods. (Repeated measures).
I really looking for the sample size needed to determine the reliability measures of mortality rates between the clinicians and within clinicians.
Thus I need to determine how many clinicians would be needed, and how many patients would be allocated to each clinician, and how many times?