Fig. 5From: Prophylaxis in healthcare workers during a pandemic: a model for a multi-centre international randomised controlled trial using Bayesian analysesStatistical model power, by incidence rate. The simulated type I error rate when hydroxychloroquine (HCQ) has no prophylactic effect (hazard ratio = 1) ranges from less than 1 to 3% for a wide range of attack rates in the no prophylaxis group. If HCQ has a 10% reduction in the risk of acquisition of SARS-CoV-2 infection, the proposed design declares superiority of HCQ in less than 9% of simulated trials across scenariosBack to article page