I don’t have studies showing outcomes correlated with numbers, but this article seems to be a nice primer:
Average Glucose, Standard Deviation, CV, and Blood Sugar Variability | diaTribe
The articles states that non-diabetic SD is under 20, but also explains the use of coefficient of variation (CV) rather than SD, since SD is dependent on average blood glucose.
Thanks for posting the diaTribe link, @JamesIgoe. I’m sure I read it back in 2018 when it was posted but it’s nice to revisit websites with lots of good data.
My philosophy for treating my diabetes is to aim for normal non-diabetic metabolically healthy targets. It’s not always easy to find these numbers published. Wouldn’t it be nice if, as contributors to Dexcom’s massive database, we could access the aggregate database to reveal these numbers?
I personally do not like aiming for the diabetic numbers since I know I can do better. Seeing non-diabetic numbers like those found at @JamesIgoe’s link posted above allows me to confirm and possibly recalibrate my targets.
The fact that the normal age groups posted SDs that ranged from 15-18 mg/dL confirms my goal of shooting for < 20 mg/dL. Goals work their magic when they exert an influence on your daily behavior and choices. It’s that reaching that can lead to improvements. If you find that the goals posted for diabetics are easily attained, don’t hesitate to tighten your goals to numbers that make you stretch a little.
Thanks @JamesIgoe I hadn’t seen that yet and it has a wonderful explanation!
I don’t find the GMI or eA1c on my Dexcom Clarity report, do I need to go online for that info?
There are a lot of reports on Clarity, I think that GMI should be on the AGP, but it is on the Overview report. Here’s an image from the internet.
BTW - I can see it on computer or phone, and that is not my report, thank goodness! 40% in range.
I am one of the lucky ones! My numbers are usually pretty close. There was one time the lab was much higher than we were looking at with the Dexcom, so we redid the test and it came back much better. Remember just because it’s lab, doesn’t mean it always right. When it is questionable or a really important test result, you can always ask for another test.
And having goals are great and the way to go but only way to go. But only if they are your goals. My time in range and my standard deviation are going to different from the next person with diabetes. And these goals are a moving target. My targets when I first got a CGM are completely different from where they are now.
So when I see my doctor and we review CGM reports, we decide if it is time to lower goals. You have to keep striving for a little better. Some visits I am higher but that just makes me work harder to get back into line.
There are no hard numbers here and everyone is different when dealing with lab results versus CGM results. Good luck with your goal setting!
@Bob10 It’s on the summary page of the Clarity report or the overview on the reports.
International Consensus on Time in Range (TIR) should be the starting point for TIR, then individualize. The AGP is the report to use with TIR. There are actually two equations for converting between HbA1C and BG. The older one was based on DCCT data. The newer one (GMI) is associated with AGP reports.
I had understood 180 as the post-prandial limit for 1-2 hours after eating (which my past Docs ignored/glossed over with result DPM and Retinopathy). There was also a limit or goal for post prandial increase over prior eating. I asked my endo at the why the difference and was advise it because insulin allows control for post-prandial while oral meds don’t.
The danger for people with diabetes is successfully satisfying this international consensus TIR, declaring victory, and concluding that no further efforts are needed. This international time in range consensus benefits people in poor control but should never be considered as the ultimate target; it’s only part of the way there.
As I’ve said many times before, if you meet the 70-180 TIR at 90+%, you should adjust that 180 limit downwards incrementally to create better metabolic health. Tightening the range that forms your TIR exerts a powerful psychological cue that will help you meet that improved goal.
I whole heartedly agree. If the endo community could just tell us what BG can cause micro scale damage, then the better target would be known. I use 140 as the upper end of my target range which is the top of the range for non-diabetics.
I have only recently started using Clarity.
I doubt the accuracy of the GMI. I had my A1c drawn this morning and it was 4.8, but my GMI is 5.8. Using several labs I have only had one A1c as high as 5.8 in 17 yrs. I am trusting my lab report. I have been working hard to have very few readings over 140 or lower than 60.
I looked at my receiver when blood was drawn and the glucose figures matched which pleased me.
I never understood the appearance of the GMI number. Dexcom used to display an “estimated A1c” that corresponded to the average blood glucose. For some reason, possibly regulatory, the eA1c fell out of favor and group of doctors did a study and started promoting the GMI number. Dexcom and other CGM manufacturers adopted the GMI and that’s what we see today.
My GMI number does not correlate well with my average glucose either. My GMI is usually about 0.7% above what an A1c would report. The GMI and A1c have no real utility to me. I know some people like the A1c since they’ve been using that number for many years. I personally prefer time-in-range, average BG, and glucose variability as measured by standard deviation and coefficient of variation % of mean.
My last point of care A1c was my lowest ever and I think it made my provider nervous. I’m thinking about politely refusing another A1c and suggesting that the above CGM measure be used instead. I think the A1c has lost meaning as actionable data.
A1C is based related to how long your RBCs are exposed to glucose. If your RBCs have shorter than “normal” lifespan, then it would not correlate well.
Or the cgm value may be off from true bg value.
Or both.
Yes, my RBC’s probably have a shorter than normal life span. Thanks for the info MM1.
There is also your night time glucose vs daytime.
Rbc’s are mostly made at night while we sleep so your night time sugars will skew your results if they are different from your daytime sugars.
My a1 pc is also much lower than the GMI suggests. Around 0.8 higher than my A1c
Marilyn, I find that Clarity GMI is often 0.5 higher than my lab A1C.
Keep in mind that GMI is based on average bg, and that there’s a very well documented amount of scatter between average bg and A1c. See typical scatter plot below.
Upthread I talked about the emergence of the GMI number. Here’s a report that gives the all the details of how that came about and why it replaced the “estimated A1c” formerly used in CGM reports.
I find the inclusion of at least three T1D members of the group that published this report a positive thing.
Glucose Management Indicator (GMI): A New Term for Estimating A1C From Continuous Glucose Monitoring
As I stated earlier, I don’t find the GMI, or for that matter the A1c, actual numbers that mean much to me when I have time in range, glucose variability, and average BG indicating the quality of my glycemia.
Aw come on Terry, TIR and Variability take near-obsession to keep in line with a non-diabetic individual. A1C/GMI is a wonderful vanity, well-accepted, statistic that is much easier to keep low, even while ebbing and flowing between dangerous highs and lows on a regular basis. It is really not even a bad target for the general diabetic population considering how out of control most of that population resides in and considers themselves in OK control.
@CJ114 – Obsession or near-obsession is, of course, a relative term. But I think we can accept that repeated behavior that produces no additional benefit can be considered obsessive. Repeatedly locking your house door before you can finally leave your house is obsessive. Locking your door once or twice is understandable. Doing it eight or twelve times is obsessive.
Please forgive my digression here as I’ve been accused of this by medical professionals more than once. Sometimes it’s explicitly phrased but at other times more subtle in its expression but certainly present just the same.
Poking my finger 15 times per day, as I did for years before CGM was available, provided many opportunities for me to observe this in my doctors and other medical professionals. The irony to me was the fact that none of them lived with diabetes!
Spending more attention on blood glucose tactics than the average person with diabetes is not obsessive from my perspective. Why? Because that extra attention can produce measurable increments of performance that produce perceptible improvements in quality of life – less metabolic drama, more calm.
What I’m trying to say is that there’s a concrete reward for trading more of your time and attention aimed at keeping your blood glucose in a tighter range with less volatility. The tactic is not the end but the means for winning a better glucose metabolism.
I’ve thought about this topic a lot over the years and I’ve wondered why some people with diabetes would be motivated to color the level of my diabetes tactics as obsessive. I speak generally here; this is not specifically about you or others in this discussion!
I think that pasting that label on behavior that they’ve personally considered and for one reason or another they chose to not adopt provides them with a rationalization that justifies their active choice. In other words, it’s not about me and my actions, but rather about them and the choices they’ve made. Again, this is a general observation I’ve made, not specific to any individual.
The A1c and GMI numbers are coarse measures of metabolic control and you do acknowledge the weakness of these numbers in catching excessive swinging of blood glucose. As a coarse indicator of metabolism (> 7% A1c, for example), I agree that it can indicate to a diabetic and his/her doctor that action should be taken.
By the way, from my point of view, the level of effort I put into my diabetes tactics is not as much as you might think. I’ve found that diabetes often accrues a metabolic inertia. Keeping blood sugar levels in a tighter range leads to a momentum that tends to keep it in that range.
The body seems to have metabolic memory of events and it incorporates those events into its circadian memory. Have you ever had a severe low blood glucose and found that almost exactly 24 hours later your metabolism wants to repeat that event even absent the triggers that originally produced it? It’s a circadian aftershock of sorts.
I believe the same homeostasis is in play when good blood sugar levels happen. Twenty-four hours later, that prior good glucose trace exerts some force to repeat the event of a day earlier. Of course, you can add levels of food, exercise, or insulin that can obscure and overwhelm that effect.
Sorry for the long answer to a short question but your comment raised a topic that I’ve given much thought. If you’ve read this far, thank-you for humoring me!
In many ways, it depends on one’s perspective.
The biggest benefits to fitness can be had at some moderate range and moderate intensity, plus some tougher workouts, but the biggest bang for the buck is just getting off the couch. Going beyond moderate fitness brings some benefits, but the cost/benefit ratio isn’t the same, so doubling the work effort only brings incremental benefits. Going into high-volume territory is the place where people win medals, but the effort is extremely high and the personal costs can be significant.
This is not a criticism of your focus, but one can see that the biggest benefits accrue to the ones who simply reach good control. Not that this is optimal, just that that’s the point where the worst ravages from T1D are avoided. Going further will presumably bring benefits, but I imagine some feel, and are told, that good is enough. There is certainly a push to get even lower, with tighter control, but the evidence is not as strong, and even good control is tough enough to achieve for some, let alone an optimal, non-diabetic level of variability.
I’m restarting a publication search, to see if anything has changed, and who knows, there might be enough proof for even tighter control, so bear with while I accumulate studies here…
Clinical Implications of Glucose Variability: Chronic Complications of Diabetes (nih.gov)
Glucose Variability; Does It Matter? | Endocrine Reviews | Oxford Academic (oup.com)
Glucose Variability in Diabetic Pregnancy | Diabetes Technology & Therapeutics (liebertpub.com)
Glucose variability, blood pressure and arterial stiffness in type 1 diabetes - ScienceDirect
Type 2 Only Studies
