I was using the Friends of Cancer Research website today and ran across an article, though from 2019, that's still a good reference. It discusses "external control arm" studies, such as when contemporary external controls are used as a reference, or when administrative controls (based on computer records) are used as a reference.
I am sure this can be good sometimes, but experts (and me) have seen cases from time to time where external control arms produce comparisons of effect that are simply unbelievable. (An example was a prediabetes YMCA study as a single arm study with adminstrative billing records of other Medicare patients used as controls. Alva 2017).
FOCR has a 2019 article on this topic. I gather it was reprinted with permssion from Pink Sheet. Worth reading if you are considering a one-arm study relying for comparison, on propensity-adjusted controls. Pros and Cons discussed.
