Multi-center accuracy assessment of a1cnow+: a disposable system for monitoring hemoglobin a1c
Author Block: KEITH A. MOSKOWITZ, BEATRIZ WALSH, MARCY SHIPWASH, STEPHANIE MIHANE, KARMEN
MERCER, EMILY SUSCHA, MARIA SHAFAI, CORINA LINDKE, CHARLES XIE, JAMES A. ANDERSON, Indianapolis,
IN, Santa Cruz, CA, Thomasville, NC, Denver, CO, Sunnyvale, CA
Abstract:
Background: HbA1c is indicated for diagnoses (Dx) of diabetes using central laboratory (CL) analyzers. Point of
Care (POC) Hb A1c analyzers offer advantages such as real time consultation, but are not indicated for
screening or Dx due to perceived inaccuracy. CAP criteria allow +/- 6 % bias and WHO allows POC when it is
the only option or when a quality assurance (QA) program exists. POC instruments are often CLIA waived thus
not subject to QA proficiency testing. Accordingly, the accuracy of PTS Diagnostics A1CNow+ POC Hb A1c test
was evaluated relative to CL Hb A1c systems.
Methods: Blood from 94 subjects was taken from three US wellness centers: Finger stick blood was used to
test A1CNow+. EDTA venous blood was used for the Roche Cobas Integra and Abbott Architect. Heparin
venous blood was used for the Tosoh G8 comparator. Correlation regression analyses were performed to
determine accuracy and % difference to assess bias. Clinical risk stratification was assessed using Hb A1c
category (cat) cut points of < 5.7; 5.7-6.4; and ≥ 6.5 and if differences resulted in 0, 1, or 2 cat change. Fisher’s
exact test was used to assess differences associated with risk.
Results: In total, slopes were 1.020 and 0.986 for CL and A1cNow+ relative to Tosoh (p = 0.63); r of 0.99 and
0.96, and intercept 0.24 and 0.13, respectively (p = 0.64). Bias was 0.3% for A1cNow+ and 3.8% for CL. Risk
was unchanged in 81.7% of CL and 77.7% of A1CNow+ measures, resulting in non-statistical (p= 0.54) cat 1
differences between the POC and CL methods. There were no cat 2 risk differences regardless of method.
Conclusions: A1CNow+ was at least as accurate in measuring Hb A1c relative to three CL analyzers and well
within the 6% CAP guideline for bias. Risk stratification revealed no differences between the CL and A1CNow+
in classifying the patient state. Ease of use and disposability of A1CNow+ should help assess Hb A1c in
situations where CL analyzers are unavailable and to provide physicians with real time information to better
manage diabetes.
Author Disclosure Information:
K.A. Moskowitz: Employee; Author; PTS Diagnostics. Stock/Shareholder; Author; PTS Diagnostics. B. Walsh:
None. M. Shipwash: None. S. Mihane: None. K. Mercer: Employee; Author; PTS Diagnostics.
Stock/Shareholder; Author; PTS Diagnostics. E. Suscha: Employee; Author; PTS Diagnostics. M. Shafai:
Employee; Author; PTS Diagnostics. C. Lindke: Employee; Author; PTS Diagnostics. C. Xie: Employee; Author;
PTS Diagnostics. Stock/Shareholder; Author; PTS Diagnostics. J.A. Anderson: Employee; Author; PTS
Diagnostics. Stock/Shareholder; Author; PTS Diagnostics.
Category (Complete): 12-A Clinical Therapeutics/New Technology–Glucose Monitoring and Sensing
Presentation Preference (Complete): Oral Preferred
Financial Support (Complete):
* ADA Support: No
Keyword (Complete): A1c ; Accuracy
Payment (Complete): Your credit card order has been processed on Monday 12 December 2016 at 4:44 PM.
Status: Complete