Source:, Published online June 14, 2010 in the Annals of Internal Medicine

Read the current Defeat Diabetes® E-Lerts™ Newsletter

This website is certified by Health On the Net Foundation. Click to verify.
This site complies with the HONcode standard for trustworthy health information:
verify here.

Defeat Diabetes
150 153rd Ave,
Suite 300

Madeira Beach, FL 33708

HbA1c Variation by Race Weakens Its Exclusive Diabetes Diagnostic Power

Posted: Sunday, June 27, 2010

African Americans have higher levels of glycated hemoglobin (HbA1c) than whites given the same blood glucose concentrations, and the difference is greater as glucose levels go up. Findings could affect the way physicians use hemoglobin A1c to test for diabetes in blacks.

There is a gap in mean HbA1c levels between blacks and whites who are normoglycemic by conventional blood glucose criteria, and it's wider among prediabetes patients and wider still among those with outright diabetes.

Dr. David C. Ziemer (Emory University, Atlanta, GA) and colleagues reported the findings, from retrospective cross-sectional analyses of two distinct populations, which were independent of age, sex, body-mass index, blood pressure, and education.

Their study looked at 1,581 non-Hispanic black and white adult participants in the Screening for Impaired Glucose Tolerance (SIGT) study recruited from January 2005 to March 2008 and 1,967 non-Hispanic blacks and whites older than 40 years who entered the Third National Health and Nutrition Examination Survey (NHANES 3) from 1988 to 1994. People in both cohorts with previously diagnosed diabetes were excluded from the analyses.

"Although the SIGT population was self-selected and the NHANES 3 population was a randomized, stratified sample of Americans, and the populations also differed in time and HbA1c-measurement methodology, the black-white differences in HbA1c level were remarkably similar," according to the authors.

In January 2010, the American Diabetes Association (ADA) published a statement that defined HbA1c >6.5% as diagnostic for diabetes and levels of 5.7% to 6.4% as "prediabetes" and a marker of elevated risk for both diabetes and cardiovascular disease.

Ziemer writes that, the current findings support a number of prior studies that show HbA1c levels to be affected by race "and call into question whether unadjusted HbA1c values accurately reflect glycemic differences between black and white Americans."

They also identify a large subgroup of people in the US, African Americans, for whom the cutoff values do not necessarily apply. "Ethnic groups, in particular African Americans, are often labeled as having poor glucose control based on the HbA1c numbers," he said. The current study argues against reliance on those numbers alone for making a diagnosis of diabetes.

In an accompanying editorial, Dr. William H. Herman (University of Michigan, Ann Arbor) and Dr. Robert M. Cohen (University of Cincinnati, OH) agree that the group's report "highlights the limitations inherent in any new measure of average glycemia and the need to characterize the sources and effect of the variation. In light of observed variation, clinicians should consider the potential peril of the ADA's recommendation to dichotomize HbA1c levels for diagnosis (6.5% [diabetic]). The challenge is to evaluate the prevalence of this variation and determine its physiologic basis and clinical impact."

Blacks overall in the SIGT study had 0.20 of a percentage point higher HbA1c concentration than whites; the difference reached 0.29 of a point in NHANES 3 (p<0.001 for both differences).

The spread was narrower among those who were normoglycemic according to conventional glucose-tolerance-test (GTT) results and wider among those with GTT-defined diabetes. In adjusted analyses, blacks had nearly a one-half absolute percentage point higher mean HbA1c concentration than did whites. The difference approached a third of a point among those with GTT-defined prediabetes.

Increase in Mean Hba1c Levels in Blacks Compared with Whites, in Absolute Percentage Points, in Two Populations by Glucose-Tolerance-Test Results
Degree of glucose tolerance     
SIGT study
*Adjusted for age, sex, body-mass index, blood pressure, education level, and fasting and two-hour oral glucose-tolerance-test results
SIGT=Screening for Impaired Glucose Tolerance study; NHANES 3=Third National Health and Nutrition Examination Survey
"Our findings suggest that physiologic or genetic factors may contribute to racial differences in HbA1c. Regardless of the mechanism involved, our data indicate that assessments of glucose control at the population level should be made with caution when based solely on HbA1c levels," the authors write.
Ziemer emphasized that the caveats apply at the patient level, too. He said the study points to the limitations of using standardized HbA1c values as cutoffs for diagnosing diabetes and argues for using a variety of diagnostic criteria. "One of the study's cautions is that HbA1c is an imperfect test. When you're seeing someone in your office, not only should you look at their HbA1c, you should look at their glucose readings from home and from the office."
   1. Ziemer DC, Kolm P, Weintraub WS, et al. Glucose-independent, black-white differences in hemoglobin A1c levels: a cross-sectional analysis of 2 studies. Ann Intern Med 2010; 152:770-777. Abstract
   2. Executive summary: Standards of medical care in diabetes -- 2010. Diabetes Care 2010; 33 (suppl 1):S4-S10. Abstract
   3. Herman WH, Cohen RM. Hemoglobin A1c: Teaching a new dog old tricks. Ann Intern Med 2010; 152:815-817. Abstract
Join us on Facebook

Send your unopened, unexpired diabetes testing supplies to:

Defeat Diabetes Foundation
150 153rd Ave, Suite 300
Madeira Beach, FL 33708


DDF advertisement

 Friendly Banner

Friendly Banner
Analyze nutrition content by portion
DDF advertisement