Secondary Logo

Journal Logo

Institutional members access full text with Ovid®

Clinical Evaluation of a Rapid A1C Test (A1cNow) for Home Use

Klonoff, David C. MD*; Bergenstal, Richard M. MD; Cole, Thomas G. PhD; Bohannon, Nancy J.V. MD§; Ammirati, Erika B. BS; Blatt, Joel M. PhD; Irvin, Benjamin R. PhD; Stivers, Carole R. PhD; Clark, Annette L. BS

Point of Care: The Journal of Near-Patient Testing & Technology: September 2006 - Volume 5 - Issue 3 - p 116-120
doi: 10.1097/01.poc.0000232578.91019.03
Original Articles

Introduction: A1C can be measured by portable point-of-care methods that might offer advantages compared with conventional sampling of venous blood for eventual laboratory testing.

Methods: In a 2-part study, we compared the performance and ease of use of A1C measurement with a single-use, disposable A1C test (A1cNow) and a venous sample measured by a reference laboratory. Part 1: At 3 sites, 297 untrained subjects self-tested with an A1cNow. Trained medical professionals performed a second A1cNow test on each subject. Venous blood was sent to a National Glycohemoglobin Standardization Program Secondary Reference Laboratory for A1C testing. Untrained and professional A1cNow test results were compared with the reference results and with each other. A quiz and questionnaire evaluated, respectively, subject comprehension of A1cNow's product labeling and opinions on ease of use. Part 2: At a fourth site, trained medical professionals performed an A1cNow test on 30 subjects. Venous blood was sent to the same reference laboratory. Professional A1cNow test results were compared with reference results. The professionals recorded the amounts of time needed for A1cNow testing and reference laboratory testing.

Results: Part 1: For untrained A1cNow versus reference, the slope and y intercept were 0.988 and 0.168, respectively, with r = 0.93 (paired Student t test, P = 0.50). For professional A1cNow versus reference, the slope and y intercept were 0.965 and 0.400, respectively, with r = 0.94 (paired Student t test, P = 0.21). For untrained versus professional A1cNow, with Deming regression, the slope and y intercept were 0.972 and 0.269, respectively, with r = 0.88 (paired Student t test, P = 0.58). Overwhelmingly, subjects responded correctly to quiz questions and favorably to opinion questions about the product's ease of use. Part 2: For professional A1cNow versus reference, the slope and intercept were 0.9504 and +0.28, respectively, with r = 0.95 (paired Student t test, P = 0.85).

Conclusions: Untrained users can operate the A1cNow test with good performance equivalent to that obtained by trained medical professional users.

From the *Dorothy L. and James E. Frank Diabetes Research Institute, Mills Peninsula Health Services, San Mateo, CA; †International Diabetes Center, Minneapolis, MN; ‡Core Laboratory for Clinical Studies, Washington University, St. Louis, MO; §Monteagle Medical Center, San Francisco, CA; ∥Ammirati Regulatory Consultants, Los Altos, CA; and ¶Metrika, Inc, Sunnyvale, CA.

Reprints: David C. Klonoff, MD, Diabetes Research Institute, Mills-Peninsula Health Services, 100 S San Mateo Dr, Room 3124, San Mateo, CA (e-mail:,

Supported by grants received from Metrika, Inc.

© 2006 Lippincott Williams & Wilkins, Inc.