Sample size of needed randomised controlled trial as a function of power to detect a 50% reduction in rate of hospitalisation due to any respiratory cause (Type I error=0.05, 1:1 allocation ratio, no dropouts, and using Fisher’s exact test)
Upper Curve ‐ Assuming rate of hospitalisation in control group = 14.9% (Cohen 2005)
Lower Curve ‐ Assuming rate of hospitalisation in control group = 17.5% (Giebels 2008)