COVID ARDS interim analysis, page-5

The speed of enrollment is completely irrelevant here. The interim analysis at 90 patients was chosen as it would be the first point to have a sufficiently powered analysis, without comprising the integrity of the data. In simple terms, the data will not be skewed to one severity of treatment group, moderate or severe.

I also don't understand your reasoning that the speed of enrollment will cause more moderate ARDS patients in the first place. Are you inferring that the speed of enrollment was slow? If I were to make assumptions as to why enrollment was slow, and forgive me if I've misunderstood your reasoning, I would assume that the severe ARDS patients were prioritised. Severe ARDS is drastically less prevalent and is associated with longer ICU stays. This would make more sense to have impacted enrollment time. In any case, the speed of enrollment is exactly what was communicated and planned, so trying to link enrollment speed with a potential skewed data set is a bit pointless, especially when you logic is backwards.