A new diabetes test could be more accurate than those currently used, but can it replace the current “gold standard?”

For more than 400 million people with diabetes around the world, blood testing is a routine part of managing their disease.

Those tests, however, aren’t always accurate for a variety of reasons.

A team of researchers thinks there might be a better way.

The researchers say they have devised a new method for estimating blood sugar levels that can reduce errors by more than 50 percent.

The researchers published their findings today in the journal Science Translational Medicine.

In their study, they combined a mathematical model of hemoglobin glycation in red blood cells with large data sets of patient glucose measurements. It indicated that the age of red blood cells is a major indicator of A1C variation because hemoglobin accumulates more sugar over time.

When they controlled the age of cells and tested it on more than 200 people with diabetes, they say the error rate went from 1 in 3 to 1 in 10.

One of the researchers, Dr. John Higgins, an associate professor at Harvard Medical School, told Healthline these calculations can be used to correct the test results that people with diabetes now get at their regular checkups.

It can also provide an estimate of the A1C result for patients using continuous glucose monitors.

Essentially, it has the potential to be the new gold standard in diabetic testing, Higgins said.

Read more: Prediabetes: To screen or not to screen? »

The current gold standard for diabetes screening is the glycohemoglobin test (HbA1c).

It is a general gauge of diabetes control that specifies an average blood glucose level over a few months. It measures glucose that sticks to hemoglobin inside red blood cells.

On a more daily basis, people with diabetes typically check their blood using meters, which measure blood glucose.

There’s also the fasting blood sugar test, which is often used to diagnose prediabetes and diabetes.

There are a lot of different possibilities with each test.

Temperature and testing equipment are just a few things that can affect blood monitors and their readings, the Mayo Clinic reports.

In addition, a normal fasting blood sugar may not eliminate the possibility of type 2 diabetes. Some people may have a blood glucose test that indicates diabetes while their A1C is normal, or vice versa.

Accuracy is relative when it comes to A1C or even blood glucose tests because an A1C test result can be up to half a percent higher or lower than the actual percentage.

In short, the tests can be confusing and are often used in combination to diagnose or treat diabetes.

Read more: Diabetes treatment without needles »

Higgins said the current test A1C has been a huge advance over what was available before it.

Still, there can be a “pretty significant difference” between A1C and other glucose measurements.

“It’s clear that there’s still quite a bit of room for improvement,” he said.

The medical community has also pointed out flaws in current blood testing methods.

“We have known for a long time that the HgbA1c might not be 100 percent accurate as it is dependent on normal red blood cell age, amount, and morphology,” Dr. Deena Adimoolam, an assistant professor in endocrinology, diabetes, and bone disease at the Icahn School of Medicine at Mount Sinai in New York, told Healthline.

Levels can be falsely elevated or decreased from other conditions, such as chronic kidney disease.

Read more: Living with an artificial pancreas machine »

Adimoolam said that Higgins’ test eliminates the variation of A1C by controlling blood cell age, but there are still other issues in blood sugar testing.

“This [Harvard study method] may be more accurate than the testing we have now, but [it] does not eliminate other red blood cell variations that might lead to a higher or lower HgbA1c,” she said.

“A more accurate way of assessing diabetes control would mean relying on methods that are not so dependent on red blood cells,” she added.