No, you start off with "we assume it does nothing". "Clinical tests" is a wide range of studies, and initially may indeed be 'simple' placebo controlled studie

No, you start off with "we assume it does nothing". "Clinical tests" is a wide range of studies, and initially may indeed be 'simple' placebo controlled studies. If an effect is seen, then testing will be done with 'various degrees' to determine a dose response, etc.