Like should be compared with like

BMJ 1995; 310 doi: https://doi.org/10.1136/bmj.310.6981.740c (Published 18 March 1995) Cite this as: BMJ 1995;310:740
  1. David J Goldie,
  2. Andar Gunneberg
  1. Consultant chemical pathologist Senior registrar Department of Clinical Chemistry, Lewis Laboratory, Southmead Hospital, Bristol BS10 5NB

    EDITOR,—Eric S Kilpatrick and colleagues conclude that measurement of haemoglobin A1c (HbA1c) concentration is to be preferred to measurement of haemoglobin A1 (HbA1 concentration on the grounds that the former classified a higher proportion of diabetic patients as having poorly controlled disease.1 But the basis of this classification—poor control being equated with a glycated haemoglobin concentration >5 SD from the reference population mean—is flawed because the coefficient of variation of the reference population mean is not the same for the methods of measurement compared. Inevitably, if (as is the case) the SD is much smaller for HbA1c than for HbA1 measured by electrophoresis but the means differ proportionately much less, any criterion of poor control based on the same number of SDs above the mean …

    View Full Text

    Log in

    Log in through your institution


    * For online subscription