In the randomized clinical trial (RCT), the process of deciding the randomization method and implementing is critically important. Unfortunately, it is not unheard of for problems to arise. In an article (Downs et al 2010 1), it is noted that as well as initial errors of trial design, problems can arise from errors with programming of the randomization or even human error during the course of the trial. Maintaining the rigor of the RCT relies on robust and reliable randomization with no errors. If treatment allocation is inadequately concealed then overestimation of treatment effect can occur, and the ‘randomized’ control trial becomes effectively ‘non-randomized’ – putting the entire study at risk (2).