Point-of-Care Hemoglobin A1c Testing: An Evidence-Based Analysis

The increasing prevalence of diabetes in Ontario means that there will be growing demand for hemoglobin A1c (HbA1c) testing to monitor glycemic control for the management of this chronic disease. Testing HbA1c where patients receive their diabetes care may improve system efficiency if the results from point-of-care HbA1c testing are comparable to those from laboratory HbA1c measurements.

To review the correlation between point-of-care HbA1c testing and laboratory HbA1c measurement in patients with diabetes in clinical settings.

The literature search included studies published between January 2003 and June 2013. Search terms included glycohemoglobin, hemoglobin A1c, point of care, and diabetes.

Studies were included if participants had diabetes; if they compared point-of-care HbA1c devices (licensed by Health Canada and available in Canada) with laboratory HbA1c measurement (reference method); if they performed point-of-care HbA1c testing using capillary blood samples (finger pricks) and laboratory HbA1c measurement using venous blood samples within 7 days; and if they reported a correlation coefficient between point-of-care HbA1c and laboratory HbA1c results.

Three point-of-care HbA1c devices were reviewed in this analysis: Bayer's A1cNow+, Bio-Rad's In2it, and Siemens' DCA Vantage. Five observational studies met the inclusion criteria. The pooled results showed a positive correlation between point-of-care HbA1c testing and laboratory HbA1c measurement (correlation coefficient, 0.967; 95% confidence interval, 0.960-0.973).

Outcomes were limited to the correlation coefficient, as this was a commonly reported measure of analytical performance in the literature. Results should be interpreted with caution due to risk of bias related to selection of participants, reference standards, and the multiple steps involved in POC HbA1c testing.

Moderate quality evidence showed a positive correlation between point-of-care HbA1c testing and laboratory HbA1c measurement. Five observational studies compared 3 point-of-care HbA1c devices with laboratory HbA1c assays, and all reported strong correlation between the 2 tests.