Everyone knows their A1c, but what do you really understand about this ubiquitous test? It’s been one of the most important inventions for T1D management–as well as measuring metabolic health for everyone–and yet, it also comes with misinformation and misunderstanding.
I try to help explain some of these things in my article, HbA1c Tests and T1D: The Good, The Bad and the Ugly
Thanks for drawing our attention to the A1c number. As you have written, it is not as simple a number as we have first believed. With its x.x% format we’re led to believe it is much more precise and accurate than is possible.
First of all, it is based on an assumed life of hemoglobin cells. They may live an average of 120 days, but that varies from person to person. Those with iron deficiency anemia, for instance, will have an A1c higher than their actual average glucose.
A1c is a population statistic and the community of a patients and their clinicians tend to use it as marker of individual glycemic performance.
The most alarming thing about the A1c population statistic is that we, as a community, are not doing so well. Most of us have poor glycemic control and, like the non-diabetic cohort, are mostly overweight and obese. I realize I’m not making friends here, perhaps some reality can help guide us toward a healthier future.
Our clinicians are trying to help. Unfortunately, that help distorts the real glycemic performance that matters. We’re told that 70-180 mg/dL is a good time in range (TIR) yet 70-140 mg/dL is probably a better standard.
Your article admirably brings up glycemic variability (GV). I agree that this is often overlooked by us. If you examine using glycemic variability as a worthwhile measure, you then need to look at food choices. That is definitely not an enjoyable topic!
I’m not bringing up topics that are popular. But I’ve personally found that deemphasizing A1c while focussing on a better TIR (I use 65-120 mg/dL), and better GV (<35 mg/dL is good, < 25 mg/dL is better) are goals that are reachable if one will make the requisite effort.
My analysis can be dismissed as obsessive and unrealistic but I urge taking a second look. I’ve lived with T1D for almost 40 years now and have accumulated some serious battle scars. I’ve only played this “game” reasonably well for the last 12 years.
Diabetes plays for keeps. We often don’t appreciate the gravity of our sustained choices until the heavy consequences appear. Don’t kid yourself!
We can effectively balance the quality of our lives with the tools we have. We can eat better; we can get help with food addiction. We can use CGM statistics like our lives depend on it, because it does.
Sorry for the harsh talk. Sometimes the cold reality can wake us up to enjoy a better tomorrow.
I think there is another objective that needs to be considered as well. The A1C levels and TIR are levels that are at this time generally accepted by the medical community. My endo is always saying that my numbers are great and I can easily run my average BG and A1C higher, which would make my life easier. My BG target is 100, and my A1C is always in the 5.6% to 5.7% range.
The difference is that my endo is coming from a perspective of what is medically considered best practice to keep from getting long-term complications. My personal goals, right or wrong, are to ignore the medical recommendations and keep my BG and A1C closely in line with individuals who are not diabetic.
I know others in tight control on this forum often do the same but do not advertise they are doing that because they risk others trying to shame them that they are running their numbers too low. I would like to see more commenters who use this method post that it is OK to run much lower control as long as hypoglycemic events never or rarely occur.
First of all, I celebrate your attitude. It’s healthy, good for the soul and eminently doable!
The best way I know to concisely communicate that I can run BGs and A1c’s lower without undue hypoglycemia is to publish a concise graph that depicts that. This depiction of glycemia is often viewed as unrealistic for the average diabetic, intimidating, or boastful – in other words counter-productive.
I wonder if the unstated resentment of that position might be the knowledge, often unconscious, that the person has decided to not employ the needed tactics for one reason or another.
I both agree and sympathise with your sentiments here.
Because T1D is such a highly emotional experience, rife with high rates of depression and anxiety (especially among youth), it’s very challenging to communicate your personal experiences that may seem to negatively affect others in the community, even with the best intentions.
Regarding your strategy of glycemic control, I do the same as you, and my A1c is 5.6% with a TIR ~97% since 2019 when I finally get a handle on how to interpret CGM patterns more efficiently. But I’m also keenly aware that what works for me (or you) may not work for others.
All this is covered in my article, You’ve Got Type 1 Diabetes! Let the Fun and Agony Begin, A Journey Through The Three Stages of Self-Management. Stage 1 is where you don’t know anything so you have no choice but to rely on what your medical team tells you. But as you learn and experience your own unique metabolic system, you learn to be “nuanced” about conventional medical recommendations. (A more thoughtful method than just simply “ignoring” them, but I know how you feel.) It is only then that you can truly improve T1D management. Now, how well you can do is less about knowledge and decisions and more about maintaining a determined self-discipline to stick to what you know works.
Stage 3 is when you finally come to peace with that, and no longer regard it as anguish, “work,” struggles, and many other stresses. Some (most?) may never get there, but one thing I realized after I turned 60: My real goal is to be able to pick up my grandchildren and play with them one day. I don’t want to just watch them from a bedside with tubes keeping me alive. That’ll be a long way off, as my son is only just starting college, so maintaining a non-diabetic health pattern is essential.
I keep my A1c in the 4’s although my last one was 5.1 due to stress. I have had non diabetic A1c’s for the last 19 yrs. I have had type 1 diabetes for 64 yrs. My TIR is in the 90’s. I tend to run lower numbers without any problem. This is hard work though!
I don’t want to have neuropathy and I want to keep my arteries as clear as I can. I have been in what you refer to as stage 3 for about 30 yrs now and I am doing well. I don’t really worry about being unhealthy in my70’s and 80’s, but I do wonder what my life will be like in the future. My internist only writes my prescriptions, and I make all the decisions about my diabetic care. I hope that my mind will stay clear enough to give my many daily injections in the future. My husband and I try not to worry too much, but being in our 70’s we have realized that our bodies are no longer young. We are very happy to be alive and relatively healthy.
The people here who are talking about tight control are all older. I was 50 before I really tightened my control. For many years I ate whatever I wanted and didn’t worry too much about my diabetes. When the A1c test first became widely used, I started paying much more attention to what I ate. When I read Dr. Bernstein’s book, I aimed for lower A1c’s. I don’t know how I got through my teenage and college years without developing problems with my eyes
I don’t have a lot of years left and I want to be as healthy as possible, so I keep my control tight.
I’m sorry you’ve felt shamed. I’m also baffled. The most frequent posters on this forum often seem to report unusually low A1Cs and extraordinarily high TIR. Maybe that feeling has come from elsewhere and you’ve attributed it to tudiabetes?
The people here who are talking about tight control are all older. I was 50 before I really tightened my control.
True, and there are many factors that lead to this. Generally speaking, people move into their senior years with a lot more wisdom about “life,” by becoming more accepting of things they cannot change. They are more aware of age and the struggles that come with it. Many see their parents die or become dependent. Their objectives change from “building” (family, wealth, status, career, etc.) and enter a phase of “preservation” – to keep the things you have and become more focused on quality of life.
T1Ds in the same position have also become so familiar with the disease that few experiences are novel anymore. You can almost predict what’s going to happen, so it’s easier (so to speak) to deal with. It’s not that challenges aren’t there, but more that you are better able to anticipate them, and consequently avert them. Older T1Ds are therefore more proactive in their daily routines–you know what mornings are like. You know what bolus you’ll need when you eat at that restaurant, etc. Kids don’t know those things yet.
For all those reasons, that’s why we’re in tighter control.
Not to worry, I have very thick skin, probably from all the insulin injections, LOL.
I stay in the high 5 range, occasionally I’ll hit 6%. I think the data shows not a lot of difference in health between 4.9 and 5.9. But there is a lot of difference once you go past 6.5%.
My TIR is generally around 95% and never under 90%.
My pump does a lot of the heavy lifting, but also I exercise a lot and I limit fat and processed carbs. Otherwise it’s not that hard to get into say 5.8%. I’m happy here.
For years I was around 7% thinking that was pretty good. Now I know better.
I don’t need to live into my 90s but I want to live into my 80s and still be functional and able. I’ve been type 1 for 36 years. About 30 of those years were marginally controlled.
Once one’s A1c is below 6% and there aren’t hypos or wide glucose variability (CV% < 30), then you start to think more like a non-diabetic for health issues. This is where cardio, lipids, and many other biomarkers are important to monitor, but I suspect you already know that. 
contents removed by original author
I don’t really follow what you’re saying… I just published a different article on the science of CGM accuracy, and thought you might like it. There’s no advertising… it’s the same platform where you read the article about A1c…
anyway, feel free to ignore if you like.
My estimated a1c on my dexcom ap is usually very close to what I get from my a1c test.
Lots I could respond to, but I’ll mention a couple.
Typos: Goldie-Lox should be Goldilocks unless you are making some point that escapes me.
And marco is obviously intended to be macro.
You mention the assumption that HbA1c is limited by the lifetime of RBCs, though you properly don’t give it much weight. In fact, the 90-day idea is based on the assumption that conversion from HbA to HbA1c is irreversible. But all chemical reactions are reversible, the only question being how fast the forward and reverse reactions occur. AFAIK, the only measurements for HbA<->HbA1c were those done by Henrik Mortensen in the 1980s. The following is from the misc.health.diabetes FAQ, written in the mid or late 1990s, after Charlie Coughran discovered the Mortensen paper. I would change some statements now, especially due to the rise of CGM, but here it is in the original.
In the early 1980s, Henrik Mortensen and colleagues at Glostrup University Hospital, in Denmark, measured the reaction rates in vitro. Their results showed the assumption of irreversibility to be untrue. In fact the reverse reaction (HbA1c to HbA and glucose) proceeds at about 1/8 the rate of the forward reaction, which is very far from irreversible. Mortensen et alia also built a biokinetic model based on the measurements, and validated the model by comparing its predictions to actual patients. See references below.
Among other things, Mortensen’s work shows that after a change in average bG level, the HbA1c level restabilizes after about 4 weeks. This has several consequences. Clinically, the most important are these:
First, the HbA1c is an exponentially weighted average of blood glucose levels from the preceding 4 weeks, with the most recent 2 weeks being by far the most important.
Second, measuring HbA1c less often than monthly results in unmonitored gaps between measurements. To use HbA1c as a continuous monitoring tool, you need to check it at least once a month.
Third, it is worthwhile checking the HbA1c of newly diagnosed patients as often as once a week to determine the effectiveness of the newly imposed treatment.
Reference: Mortensen HB, Christophersen C: Glucosylation of human haemoglobin a in red blood cells studied in vitro. Kinetics of the formation and dissociation of haemoglobin A1c. Clinica Chimica Acta 134:317-326, 15 November 1983.
Reference: Mortensen HB, Volund A, Christophersen C: Glucosylation of human haemoglobin A. Dynamic variation in HbA1c described by a biokinetic model. Clinica Chimica Acta 136:75-81, 16 January 1984.
Reference: Mortensen HB, Volund A: Application of a biokinetic model for prediction and assessment of glycated haemoglobins in diabetic patients. Scandinavian Journal of Clinical and Laboratory Investigation 48:595-602, October 1988.
Ed, thanks for the references to the typos. Hence, the value of alert readers.
Your comments regarding the detailed nuances of A1c tests–their weighted towards recency vs. evenly spread among 90 days being one example–are all valid, of course, but get too far into the weeds on what I’d hoped to emphasize: There are good, bad, and ugly uses of the A1c test, each of which are utilized in ways that go beyond the nuances.
Using A1c values among populations, and in longitudinal retrospective studies, are the best uses of the test, and smooth out the technical variances and error ratios that may exist on a smaller basis.
That said, those very variances (which don’t really need to be spelled out in detail for this audience) are why using A1c values for individual care, especially for “established” T1Ds, is highly problematic. Doctors don’t have a lot of time to spend with patients, so it’s important to use it most effectively. Here, reviewing CGM data yields a greater bang for the buck, as it were.
As for a good case for using the A1c for individual monitoring is for the non-T1D population. Here, the A1c is quite useful as a routine blood test to monitor for creeping glucose levels that might hint at emerging T2D: anyone who shows an A1c level >6.0 should be monitored very closely. IMHO, they should be on CGMs, which can be very telling insofar as seeing patterns that hint at significant insulin resistance, such as wild BG swings, especially after eating.
As for your comment that I included above regarding the use of A1c results to monitor the effectiveness of newly diagnosed T1Ds, I still think that the CGM is so far and above superior to A1c tests, that the nuances about A1c technicalities are mostly academic.
Viewing all of this one level up, monitoring T1D is a good tool, but the real work is patient self-management, and that can only be done by the patient themselves as they go through their daily in-the-moment management decisions. That’s a whole different ball of wax.
Which of course was not available in the 1990s, when the misc.health.diabetes FAQ was written. That’s the biggest reason that (as I said) I would write it differently today.
I am also a “tight” mgt person (perhaps aggressive would be a better adjective for me). I’ve been challenged by Endo’s (I refer to those MD’s as Endorks) based on their assumption that anyone Dx’d with any type of D MUST also be Dx’d with 3 co-morbidities: Stupidity, Laziness and a Propensity to Lie (i.e. logbooks). I’ve been told it is “impossible for someone like you to have such good labs” - that sort of thing. Then there is the lecture that the A1C is an average - ergo lots of high bg’s offset bh lots of low bg’s. Fortunately I brought hard copies of my T-Connect reports. Shut that guy right up. When asked why I work so hard, I tell them that I am worth it. Why should I settle for an A1C of 7-ish? I get very angry at the general low opinion of patients by HCP’/s in the diabetes field.