Retrospective Accuracy Study of Point-of-Care Hemoglobin A1c in Field Conditions

[摘要] To the Editor:Diabetes mellitus (DM)1 is a high-burden disease that relies on laboratory testing for both diagnosis and subsequent monitoring. The American Diabetes Association guidelines include use of CLIA-waived point-of-care (POC) hemoglobin A1c (Hb A1c) testing for DM monitoring but not diagnosis. Recently, a POC device manufacturer applied for US Food and Drug Administration (FDA) CLIA-waived status for diagnosis of DM, spurring conversation regarding POC Hb A1c accuracy (1). This is particularly relevant for testing in waived sites where proficiency testing is not a regulatory requirement and testing is typically performed by nonlaboratorians.While laboratory studies have demonstrated reasonable POC accuracy for some methods (2), routine patient testing may result in greater variability (3). Furthermore, the FDA strategic plan includes postmarket surveillance of medical devices through mining of electronic medical records (EMRs) (4). We therefore queried our EMR to characterize infield POC Hb A1c accuracy at our institution's ambulatory sites (primarily with CLIA certificates of waiver or provider-performed microscopy) by comparison to central laboratory values (CLVs). Using DETECT (Data Mining EMRs to Evaluate Coincident Testing) (5), we analyzed coincident samples where routine testing for POC (DCA Vantage®, Siemens) and central …

[学科分类] 过敏症与临床免疫学