Abstract
BACKGROUND: Hemoglobin A1c (HbA1c) level measurement using venous blood sample is essential for diabetes management. However, this method is not always readily accessible in primary health care in Indonesia. This study aimed to evaluate the comparability of venous versus capillary blood HbA1c measurement. METHODS: A cross-sectional study was performed, involving subjects with and without type 2 diabetes (T2D). Venous blood samples were analyzed using a high-performance liquid chromatography (HPLC) analyzer. Simultaneously, capillary blood samples from the same subjects were analyzed using a point-of-care testing (POCT) instrument based on capillary electrophoresis (CE). Passing-Bablok regression was performed to assess constant or proportional bias between methods. Bland-Altman test was conducted to evaluate the agreement of HbA1c measurement between venous and capillary. RESULTS: Out of 162 subjects, 129 (79.6%) had diabetes, 9 (5.5%) had newly developed diabetes, and 24 (14.8%) did not have diabetes. Mean venous HbA1c was higher (7.79%) than capillary HbA1c (7.49%). Bland-Altman analysis showed that mean bias for all subjects was -0.30%, with 93.8% of subjects within the limit of agreement. Besides, mean bias between subjects with venous HbA1c < 6.5% and ≥ 6.5% was - 0.14% and -0.38%, respectively. Meanwhile, Passing-Bablok regression showed no constant and proportional bias in subjects with venous HbA1c < 6.5%. However, constant and proportional bias were found in all subjects and in those with venous HbA1c ≥ 6.5%. CONCLUSIONS: Capillary HbA1c results have a good agreement with venous HbA1c. Capillary HbA1c POCT method can be considered as an alternative for monitoring diabetes. HbA1c measurement ≥ 6.5% should be interpreted with caution. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s13340-026-00889-3.