Reference change value for HbA1c in patients with type 2 diabetes mellitus.

We estimated reference change value and steady state disease biological variation of glycohemoglobin (HbA1c) from serial measurements in 47 clinically stable type 2 diabetes patients whose home-measured capillary glucose was stable throughout the study. Whole blood HbA1c assays were performed by turbidimetric inhibition immunoassay. The analytical imprecision (coefficient of variation) was 7.1% and 5.0% for control materials with HbA1c of 5.1% and 10.7%, respectively (n=152). The sampling interval was 6 months. Patients were classified into three groups (good, acceptable and poor control) according to the American Diabetes Association (ADA) clinical practice recommendations of 1999 based on HbA1c reference interval. Steady state disease biological variation for each control group was 7.9%, 5.4% and 3.9%. HbA1c absolute reference change value was 1.42, 1.50 and 1.37%, or as relative reference change value, 29.1%, 24.4% and 17.8%, respectively. The analytical goal, defined as one-half of the biological variation, was lower than 3.9% for well-controlled patients.