Hi,

I'm a grad student with a moderate statistics background having a bit of trouble figuring out which binary statistics model to use. I have browsed this forum and found some helpful info, but not specifically answering my question.

I have a dataset with repeated measures of patient lab values at various dates. Each patient experiences eventA only once in their lifetime. There are multiple dates of lab values captured before eventA and a single set of values at the date eventA occurs. I am developing a model to predict the outcome (eventA) using the lab value(s) as predictors.

My question is would this be considered repeated measures, even though eventA is once per lifetime & there are multiple measures of lab values for each patient? I have learned that for logistic regression one of the assumptions is that each observation must be independent from each other/not a repeated measure. If it doesn't satisfy this assumption, then we have to use xtlogit, pa?

If anyone can shed some light to this question, your help would be greatly appreciated.