The authors report that conduct of non-inferiority trials appears inconsistent and does not follow the recommendations in available statistical guidelines (e.g. CONSORT, FDA guidance for non-inferiority trial design, CHMP guideline on the choice of the non-inferiority margin and the International Conference on Harmonisation guidance on statistical principles for clinical trials). They note that these guidelines are however are not wholly consistent themselves. The authors highlight that failure to clearly describe the methods used and descriptions of and justifications for their design and primary analysis risks misleading conclusions being drawn, with consequent effects on clinical practice.