Updated 23 November 2016

Accuracy of standard diabetes test improved

Researchers say a new approach will enable many patients and their doctors to do a better job controlling blood sugar levels and reduce the long-term risks of heart attack, stroke, blindness and kidney failure.


Industry-funded researchers say they've developed a way to improve the accuracy of a standard diabetes test.

Reduced risks

"We think our approach will enable many patients and their doctors to do a better job controlling blood sugar levels and reduce the long-term risks of heart attack, stroke, blindness and kidney failure" associated with diabetes, said Dr John Higgins, associate professor of systems biology at Harvard Medical School in Boston.

At issue is the HbA1c test, also known as the A1c test, which is used to diagnose diabetes. It also identifies people with prediabetes and provides insight into how well blood sugar is controlled over a three-month period among those monitoring their disease.

Read: Marriage may help diabetics keep weight off

The A1c test "measures how much sugar a person's blood cells have soaked up since the time the cells were produced," Higgins said.

"Before the test was available, patients and clinicians only knew what a person's current blood sugar level was. But effective treatment of diabetes depends on knowing what the blood sugar level has been since the previous checkup," Higgins explained. "The HbA1c test provided the first available estimate of a patient's blood sugar level over the past several weeks."

Rarely inaccurate

For millions of diabetics worldwide, the A1c test forms the basis of their treatment. In the United States alone, more than 29 million Americans have diabetes, according to the US Centres for Disease Control and Prevention.

The test can be inaccurate, however. How much that matters is up to debate.

Higgins said the errors are significant. But another specialist, Dr Joel Zonszein, said the test is rarely inaccurate and "is a good test for the great majority" of patients.

"Patients with diabetes can be properly monitored and treated with the tools we have," said Zonszein, director of clinical diabetes at Albert Einstein College of Medicine's University Hospital in New York City.

"In my experience, the main issue is that individuals with diabetes don't often check their A1c values," said Zonszein, who wasn't involved with the new research.

Read: New diabetes drug approved only for weight loss

For the new study, Higgins and his colleagues used an advanced mathematical formula, or algorithm, to analyze blood sugar levels through the HbA1c test.

Improved monitoring and treatment

This enabled the scientists to account for variations in the age of blood cells among different people, Higgins said. Hemoglobin in red blood cells accumulates sugar over time, and is a major reason for differences in test results, he said.

In more than 200 patients included in the study, Higgins said the new approach reduced significant errors from about one in three to about one in 10. These were errors large enough to affect treatment decisions, he said.

Since people with diabetes often get A1c tests every three months, Higgins said the new approach could improve their monitoring and treatment.

Higgins declined to estimate how much it would cost to add the new calculations to existing tests. But he anticipates the extra expense would be less than the cost of the A1c test itself. And in defence of any higher pricing, he added, "diabetes becomes very expensive if blood sugar levels are not well-controlled."

The study was funded by the US National Institutes of Health and Abbott Diagnostics, a company that develops laboratory medical tests. The study authors, including Higgins, are listed as inventors on a patent application linked to the findings.

Read: Implantable sensor measures blood sugar levels

What's next?

Higgins said the researchers are looking for partnerships that would allow labs to use the algorithm to improve HbA1c testing.

Zonszein said the study findings appear valid, although the algorithm "has not been challenged and/or compared with other possible mathematical models".

For now, however, "this is research, and it is not a practical model that needs to be implemented," he said.

The study appears in Science Translational Medicine.

Read more:

New diabetes drug promising

Diabetes drug improves memory

Diabetes drug, weight loss linked 


Ask the Expert

Diabetes expert

Dr. May currently works as a fulltime endocrinologist and has been in private practice since 2004. He has a variety of interests, predominantly obesity and diabetes, but also sees patients with osteoporosis, thyroid disorders, men's health disorders, pituitary and adrenal disorders, polycystic ovaries, and disorders of growth. He is a leading member of several obesity and diabetes societies and runs a trial centre for new drugs.

Still have a question?

Get free advice from our panel of experts

The information provided does not constitute a diagnosis of your condition. You should consult a medical practitioner or other appropriate health care professional for a physical exmanication, diagnosis and formal advice. Health24 and the expert accept no responsibility or liability for any damage or personal harm you may suffer resulting from making use of this content.

* You must accept our condition

Forum Rules