Proficiency testing of Hb A1c: a 4-year experience in Taiwan and the Asian Pacific region.

BACKGROUND The correlation between hemoglobin A(1c) (Hb A(1c)) and risk for complications in diabetic patients heightens the need to measure Hb A(1c) with accuracy. We evaluated the current performance for measuring Hb A(1c) in the Asian and Pacific region by examining data submitted by laboratories participating in the Taiwan proficiency-testing program. METHODS Five fresh-pooled blood samples were sent to participating laboratories twice each year. The results were evaluated against target values assigned by the National Glycohemoglobin Standardization Program network laboratories; a passing criterion of +/-7% of the target value was used. Measurement uncertainty at Hb A(1c) concentrations of 7.0% and 8.0% were determined. RESULTS A total of 276 laboratories from 11 countries took part in the Hb A(1c) survey. At the Hb A(1c) concentrations tested method-specific interlaboratory imprecision (CVs) were 1.1%-13.9% in 2005, 1.3%-10.1% in 2006, 1.2%-8.2% in 2007, and 1.1%-6.1% in 2008. Differences between target values and median values from the commonly used methods ranged from -0.24% to 0.22% Hb A(1c) in 2008. In 2005 83% of laboratories passed the survey, and in 2008 93% passed. At 7.0% Hb A(1c), measurement uncertainty was on average 0.49% Hb A(1c). CONCLUSIONS The use of accuracy-based proficiency testing with stringent quality criteria has improved the performance of Hb A(1c) testing in the Asian and Pacific laboratories during the 4 years of assessment.