[Diagnostic agreement between two glycosylated a1b hemoglobin methods in Primary Care]

Several methods are available for measuring glycosylated A1c hemoglobin (HbA1c), all rapid methods for point of care use in a clinical or laboratory setting. This study attempts to compare the diagnostic agreement between two methods for detection of HbA1c.

A descriptive cross-sectional study of diagnostic agreement was carried out in the Los Carmenes Health Centre. Two groups of patients -with and without type 2 diabetes- were consecutively included. A method for point-of-care use in a Primary Care Clinic setting (DCA(TM) Systems Siemens(®)) was compared with a laboratory test (chromatographic analysis). An analysis was made of the mean concentration of HbA1c, the agreement between methods, using the intra-class correlation coefficient (CCLA1) and the Bland-Altman method.

A total of 102 patients were included, 62 diabetic (60.8%) and 40 non-diabetic (39.2%). The overall mean HbA1c was 6.46% (SD=0.88) in the analysis using capillary blood in the clinic with the DCA™ system, and 6.44% (SD=0.86) using the laboratory test (P>.05). The degree of agreement between the two tests was 0.975 (95% CI: 0.963-0.983). The mean of the differences between the results of the two assessed tests was 0.024 (SD=0.27). The percentage of points outside the limits of optimal agreement, as defined in the Bland-Altman graph, was 2.5%.

Diagnostic agreement between a method for point-of-care use in a Primary Health Care Clinic and a laboratory test was very high. Detection at the point-of-care allows a quick and simple assessment of HbA1c.