Close Server: KOPWWW05 | Not logged in

Welcome to Health Care POV | sign in | join
Press Start: Lead an Empowered Life as a Clinical Laboratorian

Glucose or HbA1c..Who you Gonna Call?

Published March 1, 2014 1:28 PM by Glen McDaniel

I have the most instructive conversations with physicians and others outside of our profession. Some day I would like to write a book of such conversations and recommend its use as a discussion point for medical laboratory students. My friend the endocrinologist was gushing to me about how wonderful HbA1c is. “I no longer order glucoses,” she said.


Her logic was that glycosylated hemoglobin gives her a more accurate, realistic, long term view of the patient’s glucose management. The patient does not have to be fasting and she finds that she can “bust” patients who watch their diet meticulously a few days before their visit, hoping she will see what she calls “a good number” suggestive of better glucose management than really does exist.


She wanted to know why “the lab” has not developed point of care HbA1c instruments for use in clinics and medical offices. Why can’t she have that result when the patient is in the office? I was explaining the state of the art technology available when a pathologist joined us.


The endocrinologist recapped the discussion  for the pathologist who immediately furrowed his brow and declared that a glucose is still far superior to a HbA1c. Results from different labs and even from different visits at the same laboratory are not comparable, he said. HbA1c should ideally be used a few times per year for monitoring patient compliance with their medical regimen.


He continued that glucose has ben around for ever, the technology is better and it should always be used as the screen which then triggers a HbA1c. Besides, a physician will only be reimbursed for ordering a HbA1c a couple times a year. I had no way of verifying if that’s true or not, so I did not comment.  But when he launched into a discussion on the relative inaccuracy and imprecision of HbA1c compared to glucose,  and the merits of the oral glucose tolerance test (OGTT) I sort of zoned out to protect my brain.


This pathologist was an anatomical pathologist, not a clinical pathologist, which is an issue for an entirely different discussion. But the conversation and how it progressed got me thinking about several things.


-Clinicians like this endocrinologist are very eager to discuss the use and relative merits of laboratory tests with the experts-whether those are MLSs, pathologists, pharmacists or vendors pushing the newest black box.

- Are pathologists, especially anatomical pathologists, really the right individuals to represent the clinical laboratory as experts on what we do?

-Some research after our talk indicated that fasting glucose variation can run around  8 percent In a single individual day over day, so my pathologist friend was not entirely correct about how accurate/precise glucose determinations are.

-I discovered that the American Diabetes Association (ADA) has been recommending that HbA1c be adopted as an adjunct in the diagnosis of diabetes and prediabetes. Further, agencies like the National Glycohemoglobin Standardization Program (NGSP) has done tremendous work in standardization and controlling coefficient of variation among tests. Consequently, the accuracy and comparability of HbA1c have increased remarkably in recent years.

-Our discussions with clinicians need not be too technical. But we should remind them when they ask about a new test or want a new point of care toy that as scientists we need to look at aspects like ease of performance, specimen requirement, accuracy, clinical utility, and positive or negative correlation with disease or disease risk.  

-People tend to believe us if we speak with authority and have a history of not mis-speaking. 


Because of family history and my own personal medical history I have  a vested interest in how my physician uses and interprets glucose and HbA1c, but more importantly, I want medical laboratorians to be informed, to keep current, and to offer sound scientific guidance to clinicians. Nature abhors a vacuum and if we are absent, all sorts of “experts” will jump in, often offering information which is misleading at best.


I have to address Cyn Walker and say she must have exceptional pathologists. In my 30+ year career in the lab, pathologists have always been in charge but their value is mostly in administrative functions and talking to doctors. It's colleague talking to colleague.

Most of this pathologist to doctor is at our request and using information that we provide. That has been my experience in the 5 or 6 jobs I have had in different organizations.

I  have never known a pathologist who is an expert at technical procedures. Would you ask a pathologist to troubleshoot a Chemistry analyzer, or re=calibrate troublesome enzymes or help in fixing your immunoassay instrument?

We have had pathologists who specialized in Blood Bank, perform bone marrows, identify immature cells etc. But their profession is different than ours. I have always taken the role of advising pathologists on schedules, workload, technical procedures and so on. I even do the research on new tests and tell them about the advantages. Techs did the  leg work on offering anti-Xa, myoglobin and changing our troponin assay recently.

Pathologists will always answer your questions if you ask. But it doesn't mean it's the correct or most current answer.

I also agree we do have to keep up to date. if someone asks about HbA1C we don't want to draw a blank. We don't want to say, "you are the doctor you should know" and we most certainly do not want to confirm the INCORRECT information that it is ONLY used 3-4 times per year to monitor diabetes. That is one use, but not the only use according to RECENT findings and recommendations of professional groups.  Times change and we have to keep up to date.

Part of our problem as a profession is always leaving the thinking part of our profession to others-even those who will give wrong information. We have to do better.

Mary Connor MT (ASCP) April 10, 2014 5:17 PM
Miami FL

Interesting discussion.  First, as a clinician, I have to say it is not strictly correct to say glucose is a  better test than HbA1C. We used to use glycosylated hemoglobin only for monitoring patients and checking up on them 3-4 times a year.

It was like a term paper they had to pass.  Some insurance companies still require we use that protocol (HbA1c x 4) as part of treating every diabetic patient. You do this together with lipids, plus annual eye exam, annual foot exam and all that of course.

Recently we learned a couple things about HbA1C. First, although HbA1C is theoretically a measure of the glucose level over the last 120 days, it is skewed to the last 30-60 days (let's say 45 day average). So it does provide long term, but still recent, information.

Second, tight glycemic control over time is most important, not just a recent glucose value.

if I know nothing about a patient or if I am monitoring my diabetic patient, I still use FBS mostly, BUT it is a mistake to think HbA1C is not useful. It is extremely useful for me to know : Is this patient really taking his meds daily, is it the correct dose for good glycemic control or should it be adjusted, how can I tell what is average glucose has been between office visits?

The reasoning of the new ADA recommendation  is that suppose you see a new patient whom you suspect, but don't know to be diabetic for sure. Say the patient is not fasting. In that case a HbA1C does in fact give you a MUCH BETTER idea of what their glucose has been over time. It is superior to a glucose in that situation.

Here is another example. Say I drew blood on a patient at a health-fair and ran tests back in the lab. A high RBS would mean less than a high HBA1C.  In reality I would contact and advise any patient with an elevated RBS or HbA1c  to see me in my office or to follow up with their own doctor. But using an accurate lab, a HbA1C of 8% on a new patient is absolutely diagnostic of diabetes!  No question about that.

As doctors we have our preferences and our old habits like anyone else. But we must adopt new practices as we learn more.

Marilyn Carter MD April 7, 2014 9:58 PM
Los Angeles CA

Great comments. Kathy, you are exactly right, of course. There is systematic and random variation which are usually method/analysis-related. Additionally there are inter-and intra individual variation due to physiology, metabolism, when specimen was drawn etc.

The conversation I think illustrates the danger of not understanding or explaining the entire picture. Both tests serve a function, both tests have some variation; but each specialty was seeing one side.  They both assumed "their" test was automatically better.

Cyn: HbA1C does give a  multi-month picture so it should be used as a measure of compliance with treatment and "big picture." But for that same big picture reason, the endocrinologist was partially correct if  she wanted to see what Patient X's glucose has been over time as opposed to just at this moment. It is for that reason that the American Diabetic Association, World Health Organization and some European groups have adopted the Hba1C as an adjunct in diagnosing (not just monitoring) diabetes.

That does not necessarily mean my endocrinologist friend will have her wish of a POCT HbA1c test anytime soon.

Glen McDaniel April 7, 2014 9:21 PM

While I agree that medical technologists should be involved in helping providers to learn about the different options in lab testing, your pathologist is on the right track. He understands both the clinical side that the providers see as well as the testing side that we in the lab see. He is correct about the A1C being an acceptable test only a few times a year. The A1C test is a monitor of treatment and should only be performed quarterly. The glucose test is used for screening and once diabetes is diagnosed and under treatment then the A1C is used to monitor the condition. Point of care testing is an awesome option for immediate answers for providers but the variability in and quality of the finger stick sample can leave alot to be desired. If you check your procedures/package inserts for most of the A1C analyzers you will find that they are only approved for "monitoring" and not for "diagnosing" so if you are using them to diagnose diabetes you are not using the analyzer in an approved method. More power to the technologist that takes the time to gain education and use it in manner to advance quality testing and better patient care.

Cyn Walker, , Lab Manager Central Utah Clinic, PC March 15, 2014 12:54 PM
Provo UT

To quote the blog..."-Some research after our talk indicated that fasting glucose variation can run around  8 percent In a single individual day over day, so my pathologist friend was not entirely correct about how accurate/precise glucose determinations are."  This sentence reinforces a confusion that my students always have because it does not separate the variation within individuals, the variation among individuals, and the variation inherent in our assays (accuracy, precision, etc).  Perhaps the pathologist was referring to glucose assays having good accuracy and precision (I suspect they do).  That is different than the intraindividual variation due to the mere fact that glucose does vary (apparently as much as 8%) because the body mobilizes glucose, it gets used, so it drops a bit, and then the body needs to mobilize it again.  So there is a modest oscillation even when fasting.  But we should not confuse those three different areas of variation.  We cannot do anything about intraindividual variation or interindividual variation.  We can improve analytical variation.

Kathy Doig, Professor March 15, 2014 6:51 AM
E. Lansing MI

As scientists we need to keep up to date on the cutting edge of science. Sometimes we think it is important to know only about the latest equipment or the newest computer system, but not so much about the latest test or the latest disease research. In fact some MLS and MLT dont even keep up on the technology they use.

Befor I moved to TX, I worked in several states as  a traveling tech for a while. You would be surprised at the varying protocols out there for diagnosing diabetes. What is even more frightening are how many doctors still jump from a high random glucose to ordering a GTT.  Have you ever had that test done? I did when I was  pregnant with my youngest and it is not fun.

I was given the standard dose of Glucola, I was not offered the option of doing a FBS followed by a 2hr and I certainly did not have a HBA1c ordered. I did some research on the ADA recommendations and when I printed them off for my doctor, he said " Those are just suggestions and still very experimental."

We are not doctors by any means, but this is our profession and so we ought to at least be reasonably knowledgeable about what we do and how tests are used.

Just my 2 cents worth. Thank you for this article.

Beatrice Gonzales, MLS March 12, 2014 5:34 PM
McAllen TX

leave a comment

To prevent comment spam, please type the code you see below into the code field before submitting your comment. If you cannot read the numbers in the image, reload the page to generate a new one.

Enter the security code below:


About this Blog

Keep Me Updated