04-02-2020, 03:05 PM
I'd be curious of there's any statistical power there with such a small n. I know you can have it if the effect size is large enough. It seems like more than 4/62 patients should normally progress to severe symptoms, but how much more depends on definitions.