In letters published in CMAJ, Robinson and Sohal1 and Tonelli and colleagues,2 like many, agree that A1C testing is a more expensive tool for screening diabetes than blood glucose measurement. This may not be the case and is applicable only if the costs of the reagents used in the laboratory are considered. A simple comparison certainly makes A1C testing appear to be more expensive; however, 2 important factors need to be considered.
First, the associated costs (i.e., phlebotomy, tubes, reporting) of processing a sample are equal for measurement of A1C and glucose and are substantially greater than the cost of reagents. Processing costs vary but they may be 20 times greater than the cost of the reagents for A1C and so the percentage difference between A1C and glucose measurement is relatively small, (e.g., the typical costs of processing any single laboratory sample is about $20.00, while the reagent costs for plasma glucose are about $1.00 and for A1C are about $5.00. Those who raise concerns about the cost of A1C see the 5-fold difference in costs compared with glucose, but the true difference is less than 20% [$21.00 v. $25.00]).3
Second, A1C is much more stable and reproducible than any measure of serum or plasma glucose; and it is much more likely that a definitive diagnosis can be trusted after a single measurement of A1C, unlike the repeated measurements required for glucose.4
After these considerations, A1C may well be a cheaper way of screening for diabetes than using serum or plasma glucose measurements.