Using the A1C for diagnosis
Among the benefits of using the A1C test to diagnose diabetes or prediabetes is that a blood sample for the test can be obtained at any time. It is not necessary to fast before having the test done, nor does the test re-quire drinking a glucose solution as part of the test. In addition, A1C level is unaffected by stress or illness at the time of the blood draw.
The A1C test may be less useful for diagnosing Type 1 diabetes than Type 2 because people with Type 1 diabetes often have acute symptoms and very high blood glucose levels when they are diagnosed. A random plasma glucose test, which also doesn’t require fasting or drinking glucose, can easily pick this up.
The A1C test is not recommended for diagnosing gestational diabetes, because a woman with gestational diabetes would likely not have an elevated A1C, even when her OGTT result is diagnostic of diabetes. This is because the A1C test gives an estimate of average blood glucose level over the past two to three months, and the OGTT is done when a woman would typically start having high blood glucose during pregnancy, not three months after she might start having high blood glucose. Waiting for her A1C to rise would delay necessary treatment.
The A1C test is also not recommended for diagnosing previously undiagnosed Type 1 or Type 2 diabetes during pregnancy, because changes in the rate of red blood cell turnover that occur during pregnancy would make the test inaccurate.
There are also some medical conditions affecting the blood that make A1C testing inaccurate. For people who have these conditions, either the FPG or the OGTT must be used for diagnosis (and different tests must be used for evaluating long-term blood glucose control).
When the A1C test is used to test for diabetes, the ADA’s International Expert Committee has proposed 6.5% as the level at which diabetes is diagnosed. (Diagnosis should be confirmed with a repeat A1C test unless symptoms of diabetes and blood glucose levels over 200 mg/dl are present.) The committee has also proposed the range of 6% to 6.5% as the level of highest risk for progression to diabetes, indicating that preventive measures might be in order. However, when a person has known risk factors for diabetes, an A1C level lower than 6% might also call for preventive measures.
These recommended cutoffs are based on research showing the usual range of A1C levels found among people who do not have diabetes, those with prediabetes, and those with diabetes. However, researchers have already raised concerns that these cutoffs underdiagnose diabetes in the elderly population as well as in certain ethnic and minority groups. The cutoffs may also not be right for diagnosing diabetes in nonwestern countries. Also, in some countries, the higher cost of A1C testing and/or the lack of standardization among laboratories may make the A1C test less useful as a screening tool.
Research will no doubt continue as use of the A1C test for diagnostic purposes becomes more widespread, and more will be learned about how best to use it in different populations.
Refining the diagnosis
The vast majority of cases of diabetes are either Type 1 diabetes, Type 2 diabetes, or gestational diabetes. All are diagnosed on the basis of high blood glucose, but none of the tests used for diagnosis discriminates between types of diabetes. For that, physicians use other criteria, such as symptoms, whether a woman is pregnant, and personal characteristics such as age, body-mass index, family history, and the presence of certain other health conditions.
In many if not most cases, that’s enough to make a correct diagnosis. But sometimes it’s not clear what type of diabetes a person has. In those cases, a health-care provider may order further tests, since a misdiagnosis can lead to suboptimal treatment.