No, what I mean is that evidence comes in different forms, a randomised controlled trial of masks is unlikely to happen, so the evidence has been built up through observational studies, epidemiology etc. You can very much base your initial approach on it if the intervention has minimal harm...