diabetes supplement
Diabetes Testing

The Three Ways to Diagnose Diabetes

For years the usual way that we got diagnosed with diabetes was a fasting plasma glucose test. But there are two other ways that we get the news now.

The newest way is when a doctor told us the results of a glycosylated hemoglobin test, which we usually call simply an A1C test. While a few of us learned that we have diabetes after we had an oral glucose tolerance test, that has always been the least common diagnostic tool for diabetes.


The blue circle is the universal symbol for diabetes

The quickest and easiest is clearly the A1C test. But it has several limitations on its accuracy. In fact, none of these three ways that our doctors diagnose diabetes is perfect.

Because of its advantages, the A1C may now be the most common way we learn that we have this chronic disease. It’s the only one of these three tests that doesn’t require that we have been fasting before hand.

The Standard Test

The standard way that doctors diagnose our diabetes is still a fasting plasma glucose test. Since 1997 the cutoff point separating diabetes from prediabetes has been a level of 126 mg/dl (7.0 mmol/l) or more. Doctors have a lot of experience with this test, which is widely available. But because it requires that we fast for eight hours, which is inconvenient for many people, the A1C test has become more popular.

Another old way of diagnosing diabetes is the oral glucose tolerance test, where we swallow a standard dose of glucose and wait two hours for our blood levels to be checked. While it is more sensitive than the fasting plasma glucose test, it also requires that we are fasting. It is also less inconvenient, more expensive, and less reproducible than the other tests.

A1C Accepted

Finally, in 2009 the A1C became the third accepted test for our doctors to diagnose diabetes. An international expert committee found that it has become at least as accurate and precise as the other two approved ways to diagnose diabetes. The committee established a cut off point of an A1C level of 6.5 percent.

Its advantages include being at least as good at setting this level of high blood glucose at which the prevalence of retinopathy really takes off. It is the “threshold level that results in retinopathy most characteristic of diabetes.” An earlier expert committee study had found that retinopathy is the closest thing that exists for a biological marker of diabetes. The A1C test also has “appreciably superior technical attributes…and is more clinically convenient.”


But none of the three ways to diagnose diabetes is perfect. Probably the biggest problem is that A1C compared with fasting blood glucose tests miss the diagnosis of diabetes for many people. Other limitations of using A1C tests for diabetes diagnosis is that people who had conditions like sickle cell, thalassemia, anemia, or late-stage kidney disease may actually have diabetes even when they test with an A1C level of 6.5 or less. On the other hand, people who are deficient in iron or otherwise have a of turnover in their red blood cells might not have diabetes even when their A1C tests are 6.5 or more.

In addition, unless the tested blood glucose level is more than 200 mg/dl (11.1 mmol/l), a second A1C test is needed to confirm the diagnosis. These tests need to be made with clinical laboratory equipment because point-of-care devices may not be accurate or precise enough to diagnose diabetes.

Getting Personal

Personally, I  found out that I have diabetes more than 21 years ago on the basis of an A1C test. That was, of course, long before this was an accepted diagnosis. But the doctor at the VA Clinic in Santa Barbara, California, who told me that I have diabetes had no doubt that I have the disease and I don’t either. My A1C level was 14.4 and a separate fingerstick test was 312 mg/dl, far above the current cutoff.

How was your diabetes diagnosed?

This article is based on an earlier version of my article published by HealthCentral.

Never Miss An Update

Subscribe to my free newsletter “Diabetes Update”

I send out my newsletter on first of every month. It covers new articles and columns that I have written and important developments in diabetes generally that you may have missed.

Previous Post Next Post

You Might Also Like These Articles

  • RLR at

    I had annual blood tests and T2D was not diagnosed until I made an appt. with my doctor because I was feeling tingling in my feet. This leads me to believe that the 125 fasting B/S level is too high. I’d make the cutoff B/S level 100 max.

    • David Mendosa at

      The level of 125 was set by panels of doctors, not by me, RLR. But I understand your point. The diabetes complication that you have is called diabetic peripheral neuropathy, but in fact you can get it when you have pre-diabetes.

  • Rose at

    David thank you for that link. It was very informative and addressed my concerns.

  • Rose at

    David, the only unreliable factor in this test is the possibility of a diabetic having some hypos during the night (or day) which would even out any spikes and still present the person with a ‘normal’ result. I don’t know how prevalent this is. Do you?

  • Ralph at

    I had about a week of drinking gallons of water a day, urninating frequently and losing about 2 pounds a day. I call my doctor and she said “You’ve just described the symptoms of diabetes, come in today.” My A1c was 11.9 and my blood sugar was 312. Yeah, I have diabetes. At the end of this years I will have been following Dr. Berstein’s solution for 5 years and my A1c is 5.2 – 5.4 with an average fasting blood glucose of 86.

    • David Mendosa at

      Good for you, Ralph! I’ll bet that you are much healthier now in every way too.

  • Mike at

    I had numb feet. My A1C was5.4. My fasting glucose was 122. They told me I am diabetic. My A1C 10 years later is 5.1 and fasting is anywhere from 99-110 consistently. I think all 3 tests are needed since each one gives you different information. My 2 hour OGT was 225! That’s what did me in.

  • ana at

    Under disadvantages:
    Other limitations of using A1C tests for diabetes diagnosis is that people who had conditions like sickle cell, thalassemia, anemia, or late-stage kidney disease may actually have diabetes even when they test with an A1C level of 6.5 or more.

    It should say “less,” correct? or am I misreading this??

    • David Mendosa at

      Thank you, Ana! You are absolutely right and I will get this corrected.

  • ana at

    I think your 6.5 comment on sickle cell should be 6.5 or LESS.
    I was dx with prediabetes (fasting 102) in my late 40s. I think it was due to decades of the low-fat/high-carb diet. Currently fasting is 97 but a1c was higher than I expected (5.9) so I am trying to see if high fiber carbs added back in may (or may not) even things out. Interestingly, I know a few people with fasting of 112-120 and their a1c was 5.4…clearly not an expected outcome.

    • David Mendosa at

      I don’t see a comment on sickle cell, Ana. Where did you see that?