9 seems adequate to get some kind of statistical significance from the results. Additionally if this somehow has a very high mortality rate, due to unseen conse

9 seems adequate to get some kind of statistical significance from the results. Additionally if this somehow has a very high mortality rate, due to unseen consequences, 9 is a lower number than say, 300. In reality though, running medical trials is hugely expensive, the FDA requires all kinds of procedural documentation, as well as documentation of software code etc.