For calculating sample size in clinical research, I have encountered different ways. For example, using lognormal distributions or calculating using ICC. I am confused which method to apply. Could anyone point to the some references to calculate sample size pertaining to clinical research?
$\begingroup$
$\endgroup$
4
-
3$\begingroup$ Please provide more details about your own study, as this question is so broad and vague that it could require a book to answer. What type of clinical measurement or outcome is of interest, how many and what types of predictor values do you want to examine, and so forth. $\endgroup$– EdMCommented May 22, 2016 at 2:51
-
$\begingroup$ Thanks! So I am interested in gathering sample size to measure reliability and validity of clinical questionnaire between 2 sets of population. For finding sample size for this situation calculating upper and lower limit of ICC. In another study where we were comparing operative costs between two subgroups, they were being compared using imbalanced and with a lognormal distribution of costs with coefficient of variation. I am not sure when to apply which method. If you can recommend a book, thats fine too $\endgroup$– Div DTCommented May 22, 2016 at 14:00
-
1$\begingroup$ This page has many links to texts on statistics of clinical studies. It's best to consult with a local statistician before you start, as an experienced statistician will more quickly recognize the issues that are most critical for your particular study. Otherwise you may spend a lot of time reading about matters that aren't of immediate importance. That can be good for your education, but bad for your efficiency and the timely, successful completion of your study. $\endgroup$– EdMCommented May 22, 2016 at 23:26
-
$\begingroup$ Thank you so much EdM. This is exactly what I was looking for. $\endgroup$– Div DTCommented May 24, 2016 at 19:14
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
2
This is a broad question and there are books on the subject. Here is one that many have found accessible.
See also the link Good text on Clinical Trials? provided in a comment.
answered Jul 14, 2019 at 10:09
-
$\begingroup$ Does it make sense to close this question as a duplicate of stats.stackexchange.com/q/91630/1352 and migrate your answer there? It looks to me like that one could be a reasonable canonical thread on sample size calculation in clinical trials (based on a short search for threads with the "references" and "sample-size" tags). $\endgroup$ Commented Oct 1, 2022 at 9:17
-
$\begingroup$ @StephanKolassa: Done $\endgroup$ Commented Oct 1, 2022 at 17:56