Multicenter assessment of a hemoglobin A1c point-of-care device for diagnosis of diabetes mellitus

Philip M. Sobolesky, Breland E. Smith, Amy K. Saenger, Karen Schulz, Fred S. Apple, Mitchell G. Scott, Alan H.B. Wu, Randie R. Little, Robert L. Fitzgerald

Research output: Contribution to journalArticlepeer-review

9 Scopus citations


Objective: A multisite investigation compared the analytical performance of a point-of-care (POC) HbA1c device with multiple commonly used HbA1c laboratory methods and an NGSP (National Glycohemoglobin Standardization Program) reference method. Research Design and Methods: The Afinion AS100 POC device analyzed HbA1c using 618 EDTA whole blood excess patient specimens with clinically indicated HbA1c testing. Results were compared to measurements across five clinical laboratories and the NGSP reference method. Precision was evaluated over 8–10 consecutive days for low-, mid-, and high-range HbA1c specimens at all five sites. Results: Over a wide range of HbA1c values (4.0%–15% HbA1c), 97.1% of the POC results and 94.5% of routine laboratory results fell within the target value of ±6% of the NGSP reference method results. The POC HbA1c results at 6.5% exhibited a total relative bias of −0.6% (−0.04% HbA1c) compared to the reference method while the aggregate of laboratory methods displayed a relative bias of −0.9% (−0.06% HbA1c). The total imprecision of the POC results ranged from 0.74–2.13% CV across the analytic measurement range compared to 0.81–3.23% CV for the routine laboratory methods. Conclusions: The accuracy and precision of the Afinion POC HbA1c method was comparable to the laboratory HbA1c methods supporting the FDA's recent approval of the Afinion HbA1c Dx device for use in the diagnosis of diabetes.

Original languageEnglish
Pages (from-to)18-22
Number of pages5
JournalClinical Biochemistry
StatePublished - Nov 2018


  • Diabetes
  • HbA1c
  • NGSP
  • Point-of-care


Dive into the research topics of 'Multicenter assessment of a hemoglobin A1c point-of-care device for diagnosis of diabetes mellitus'. Together they form a unique fingerprint.

Cite this