Measurement of glycated haemoglobin (HbA(1c)) has become an established procedure in the long-term assessment of glycaemic control in diabetic patients.(1) The availability of current results at the time of consultation is important, and real time within-clinic measurements of HbA(1c) meets this need. A novel, portable, cartridge-based immunoassay system (the DCA 2000, Bayer Diagnostics, Basingstoke, UK) has been found to give reliable analytical results(2) and to be useful in a variety of in-clinic settings.(3) In the measurement of HbA(1c) by immuno-assay, glycated variants of haemoglobin may be included,(4) or excluded,(5) depending on the region of the N-terminal beta chain recognized by the antibody used. Exclusion gives rise to falsely low values and misinterpretation of results is likely in patients whose variant status is unrecognized. This is most likely to occur in subjects who are heterozygotes for HbAS or HbAC, since only relatively minor pathology is associated with these conditions and recognition of variant status is frequently fortuitous. In patients with sickle cell disease (HbSS, HbSC, HbSDPunjab, HbSOArab and HbS beta(0)thalassaemia) measurement of HbA(1c) is of doubtful value, because of the reduced variable red cell life span in such patients. In this study we compared the DCA 2000 immunoassay system with two automated highperformance liquid chromatography (HPLC) systems, the Diamat and the Glycomat in monitoring patients with haemoglobin variants.