Thank you very much!
Some subjects did come to later follow ups, so I guess there could be any reason. I was not involved in the data survey and I didn't see a variable for missing reasons. Still, since it is with Alzheimers patients, some may have had clinical reasons not to come or may have simply fell out of the study for health reasons. But I think that wouldn't be too many, since the sample should consist of early cases. I think I have to look deeper into this.
I had hoped there would be a way to build a template with missing visits, without biasing the whole thing and rendering the results non-interpretable.
Your last point would mean a cross-sectional design with groups and time as UV, if I understand you correctly?
Greetings, Fabian
|