What is the general approach for handling data where a period effect appears to be present. So I have a bunch of studies some of which are cross over designs, where the patient is exposed to a drug more than once and EDA suggests some adaptation (inductions/inhibition etc…). Unfortunately since this is a mix of studies analyzed post-hoc the timing of the occasions is not consistent across studies so just making a ordinal variable for occasion won’t tell the whole story.
As a PK person I am doing this in a non-linear mixed effect regression analysis.
Any thoughts on this from my more knowledgable members of the site?