Repeated Measures

[hlsmith]

I will acknowledge that I dont get a lot of repeated measures data. Someone is working on getting me some data. Simplified, it will consist of a treatment variable, binary(X), and a continuous outcome for subjects across time. People will likely get the same treatment across time (X=0) up until recently where a handful will get a new treatment (X=1). Also the continuous variable will be changing across time (becoming larger due to an underlying disease progression).

Study question, what is the benefit of the new treatment? Can i do some type of difference in differences across time? Any comments would be appreciated!

 

[hlsmith]

BUMP
@GretaGarbo any input. So there is binary time-varying treatment for a handful of subjects. I will find out about data tomorrow and will provide a sample.

 

[noetsi]

Can you do segmented regression. The impact has to be immediate for that to be useful. You look for a change in the regression intercept or curve. It will take covariates which is a nice feature. I have a tome on it if that will help.

Another possibility is intervention analysis (so called) a specialized form of ARIMA. Hays et el wrote a monograph on it I have somewhere. In theory you can look at gradual impact with this, although I doubt you can in practice.