I would like to perform an interrupted time series analysis to look at the impact of the pandemic on cancer incidence using individual level data. I plan to use a negative binomial segmented regression approach.
I am concerned that there may be clustering by treatment facility as specific hospitals offer different treatments and specialize in specific diseases. How can I account for this clustering in my analysis? Some have suggested using a sandwich estimator but I am skeptical that this may be an oversimplification.
