search

adayim / consort

Create CONSORT diagrams for clinical studies.
Other
28 stars 4 forks source link

Update consrot_diagram.Rmd #3

Closed dryzliang closed 2 years ago

dryzliang commented 2 years ago

the column should be trialno rather than arm. The number shown in the diagram didn't make sense if it was arm.

adayim commented 2 years ago

Well, this doesn't make any difference. It counts the number of non-missing values. Any variables with desired counts would work. But this change makes more sense in the situation where the example is given. Thank you for the pull request, I will merge it.