The disease 'diabetes' is based on an agreement and we have nothing to lose by changing that agreement in one based on HbA1c instead of an isolated glucose measurement. We discuss the advantages and disadvantages of this proposal and conclude that the advantages have the upper hand. Clinical chemists are pleading for a combination of HbA1c and glucose measurements. However, this would add little value in practice. In another comment, Rutten opposes the proposal. His argument is based, among other things, on the concepts of sensitivity and specificity, from which can be concluded that the 'agreement aspect' of the diabetes mellitus diagnosis does not seem to be easily accepted by everyone. We ourselves believe that, despite the disadvantages and the doubts, our patients will be better off with the HbA1c-based classification system. We understand the hesitation and are inviting discussion, which will hopefully lead to a unanimous position.