Abstract / Summary
Capillary blood glucose meters remain widely used in diabetes management despite increasing use of continuous glucose monitoring systems. This study aimed to evaluate the agreement and bias of three capillary blood glucose meters compared with venous serum glucose and to assess their performance across relevant glycemic ranges in routine primary care. A prospective multicentre method-comparison study was conducted across seven primary care centres within one healthcare organisation. Adults with type 2 diabetes undergoing routine blood testing were opportunistically recruited. Capillary glucose was measured consecutively using three glucose meters (Contour Care, GlucoMen Aereo, and FreeStyle Optium Neo), with venous serum glucose obtained during the same clinical encounter as the clinical comparator. Agreement was assessed primarily using Bland–Altman mean bias and 95% limits of agreement (LoA). Mean absolute difference (MAD), Pearson correlation, and the proportion of measurements within tolerance criteria were assessed complementarily. Tolerance was defined as ± 15 mg/dL for venous serum glucose < 100 mg/dL and ± 15% for values ≥ 100 mg/dL. Analyses were stratified into < 100, 100–180, and ≥ 180 mg/dL ranges. A total of 142 measurement episodes were recorded; 131 had complete measurements from all three meters and venous serum glucose and constituted the primary sample. Correlations with venous serum glucose were high ( r = 0.953–0.970). FreeStyle showed the smallest observed overall mean bias (− 2.67 mg/dL; 95% CI − 4.41 to − 0.93), with a MAD of 8.08 mg/dL and 96.9% of measurements within tolerance. Contour and GlucoMen showed positive mean biases of + 10.73 mg/dL (95% CI + 9.10 to + 12.36) and + 6.85 mg/dL (95% CI + 4.82 to + 8.87), respectively, with 80.2% and 82.4% of measurements within tolerance. In the < 100 mg/dL range, mean bias was + 3.34 mg/dL for FreeStyle, + 14.10 mg/dL for Contour, and + 10.83 mg/dL for GlucoMen, with 96.6%, 55.2%, and 69.0% of measurements within tolerance, respectively. All venous serum glucose values in this range were between 73 and 99 mg/dL. The three capillary blood glucose meters showed strong association with venous serum glucose, but high correlation did not imply close agreement or clinical interchangeability. Systematic differences were observed between devices, particularly in the lower glycemic range. Because no venous serum glucose value was below 70 mg/dL, the study does not assess performance in confirmed hypoglycemia. These findings should be interpreted as method-comparison results in routine primary care rather than as evidence of formal analytical accuracy, ISO 15,197 compliance, or clinical interchangeability. Trial registration: Not applicable.