Change from or adjust for baseline for Binary outcomes


New Member
any thoughts on how to estimate change from baseline, or adjust for baseline value
when both the baseline value and the post-intervention value is binary.

Should it be included as a covariate or should we use a logitudinal model?


TS Contributor
Well, what is your study about actually? For example,what is your research question, your research design, sample size, and what is your binary?

With kind regards



Omega Contributor
So there was something present (yes/no), then an intervention, followed by examining the yes/no variables again? So you have four outcomes (stayed no, no-to-yes, yes-to-no, or stayed yes).

Did you randomized the intervention?