Can anyone find the distribution for a1c in diabetics across the U.S. for T1s and T2s?
What percentage of the population achieves an a1c below 7 for both T1 and T2?
I’m also wondering if these statistics have changed with the introduction of AID systems.
Do we have any stats that show a1c alongside treatment methodology (like pump, AID system, or MI)?
According to MS Copilot:
“ The most comprehensive source is the T1D Exchange Registry, which tracks thousands of individuals with type 1 diabetes across the U.S. Although exact percentages for every range are not always published, here’s the breakdown based on recent registry data and clinical studies:
• Below 7% (ADA target for most adults)
About 21% of adults achieve this goal. For youth, it’s even lower (around 17% meet their age-adjusted target)
• 7% to 8%
Roughly 25–30% of adults fall into this range, based on registry distributions and published data.
• 8% to 9%
Approximately 20–25% of adults are in this category
• Above 9%
Around 25–30% of adults have HbA1c levels above 9%
Context
• The average HbA1c for adults in the registry is about 8.4%
• These figures come from people receiving care at endocrinology clinics, so the general population may have even worse outcomes.”
For Type 2, from AI again so might or might not be correct:
“ Here’s what the data shows for Type 2 diabetes in U.S. adults regarding HbA1c distribution:
Estimated Percentages by HbA1c Range
Based on NHANES data (2017–2023) and Healthy People 2030 objectives:
• Below 7% (ADA target)
About 43–54% of adults with diabetes achieved this before 2020, but it dropped to ~43.5% in 2021–2023
• 7% to 8%
Roughly 25–30% fall into this range, inferred from NHANES distribution and mean HbA1c trends (average HbA1c rose to 7.6%)
• 8% to 9%
Estimated 10–15%, based on intermediate control levels reported in national surveys
• Above 9%
About 18% of adults with diagnosed diabetes had HbA1c >9% “
alright. Thank you. Maybe ADA or Joslin has some more detailed data sources.
I’ve got a diabetic who is asking the following question.
“Serious question.
If you are routinely hitting 6.5 A1c on MDI using Humulin R and N and your yearly out of pocket for T1D is about 1,000.00 without insurance, what would justify adopting a pump?“
Part of how I hope to respond is with some data demonstrating what a small percentage of the population at large HAS an a1c less than or equal to 6.5. I’ve seen this data before, but I can’t pull it out of my ass and there may not be a super recent dataset.
He’s primarily interested in T1 diabetics. He interested in financial/physiological justification for the existence of insulin pumps. While we could each write an essay on that, what will be primarily relevant to him is to just showing what a small proportion of the population that is (that has an a1c less than or equal to 6.5).
Well as I am T2, but am on a pump, I give credit to the pump for keeping me under HbA1c of 7% when I was on high doses of prednisone. I have been mid 6% and my 12/08/2025 wasw 5.6%.
In my opinion it was not be economical for this person to use a pump- the initial outlay for the, the cost of the required type of insulin and infusion sets would run much higher than $85.00 per month probably even with insurance.
I have been in the 5’s before with MDI, using Humalog, Lantus, and NPH.
But I don’t think I have used Humulin since the 80’s, so I don’t really have an answer for that.
I think doing it with $1,000 out-of-pocket without insurance would be extremely difficult, because you can’t even cover CGM with that amount of money.
I thought of atleast one of the answers below before posting the question to the evil ChatGPT. My thought initially when reading the questions were that not every 6.5% A1C is created equally, and the response from ChatGPT included that, and more. I agree mostly with the responses given for why a pump may be justified even for someone meeting all the criteria laid out in the original post.
ChatGPT response:
If someone is:
Consistently ~6.5 A1c
Not having frequent severe lows
Has predictable routines
Is comfortable with R/N timing and peaks
And is spending ~$1,000/year without insurance
They are objectively doing very well by any historical or clinical standard. Many endocrinologists would quietly admit that this is better control than a large percentage of pump users.
So the burden of proof is on the pump, not the patient.
A1c can hide:
Frequent highs + frequent lows averaging out
Overnight lows
Large post-meal spikes
A pump (especially with CGM integration) can be justified if:
You’re seeing wide glucose swings (roller coasters)
Dawn phenomenon is difficult to manage with N
Nighttime hypoglycemia is common or unpredictable
If time-in-range improves meaningfully without increasing hypoglycemia, that’s a valid reason.
This is often the strongest justification.
Examples:
Variable schedule (shift work, travel, irregular meals)
Needing flexibility in meal timing without pre-planning insulin hours ahead
Exercise that unpredictably drops glucose
Burnout from rigid timing of R/N
If a pump reduces mental load, stress, or fear—even with the same A1c—that matters.
Even with a good A1c, if:
You’ve had severe lows
You don’t feel lows reliably
You live alone or sleep through lows
Pump + CGM (especially hybrid closed loop) can dramatically reduce risk, which is a strong medical justification.
Pumps allow:
Variable basal rates hour-by-hour
Temporary basal reductions
Micro-boluses
If someone has:
Dawn phenomenon
Nighttime insulin needs that differ from daytime
Illness-related variability
A pump may give smoother control with less effort.
Some people adopt pumps not because MDI is failing now, but because:
Insulin sensitivity is changing with age
Life is becoming less predictable
They want access to evolving closed-loop tech
That’s a personal choice, not a medical necessity—but still valid.
“Pumps are modern / standard of care”
Pressure from clinicians or forums
A belief that pumps automatically lower A1c
Chasing perfection when current control is already excellent
Assuming more tech = better outcomes
Technology can increase burden if it doesn’t solve a real problem.
If someone is:
Paying ~$1,000/year now
Facing $5k–$10k/year on a pump without strong insurance
Then the bar should be very high. There’s nothing irresponsible or “behind the times” about declining that.
For someone:
Stable at ~6.5 A1c
On R/N
Low cost
Low complication burden
A pump is justified only if it clearly improves safety, flexibility, or quality of life—not just numbers on a chart.
Otherwise, staying on MDI isn’t settling. It’s choosing what already works.
It seems possible (but not guaranteed) that his A1c will drop lower than 6.5 on a pump but there’s more to think about besides A1c. It’s possible to have an A1c of 6.5 and still have a concerning number of lows. If that is the case with MDI a pump might be worth a try.
The main advantage of a pump over MDI is varying the basal rate to conveniently reduce the number of lows…by varying basal rate to “dial it in”, by suspending insulin for exercise, or by using an AID system to do that for you.
Oh yeah, I think he’s well controlled.
He’s building a home in Costa Rica where insulin is basically free.
He might want me to justify my use of an insulin pump. I’m not 100% sure what he’s getting at, but it has something to do with justification for a pump, given the expense. He believes them to be unnecessary. Of course, ‘unnecessary’ is relative. I just tried to give him the basics bout a pump.
He has problems going low. He uses 70/30 and says he crashed a car when he tried Lantus. That might make any changes a little freaky. I think that has to do with his beta cells spontaneously booting up and kicking out a bunch of insulin. That’s a little bit of a weird situation that I don’t, personally, encounter. He takes a very small amount of insulin for TDD, but he has been diabetic since almost infancy. He says he could almost survive without basal and be fine, as long as he didn’t eat.
This seems fantastical….i guess being very insulin sensitive may help limit need for large amounts of basal insulin, but even if a person with T1D does not eat at all, their liver will still release glucose into the bloodstream. Without basal insulin, blood sugar rises and ketones will form.
I didn’t believe him either.
Then he showed me the Joslin study for the 50 year medalists. https://joslin.org/research/medalist-program-study See the ‘Results’ section for the following excerpt:
He’s been T1D since 6 months of age and he still makes insulin. He recently had a C-peptide test because his BG was going low all the time.
Does he have something other than T1D? Lol. This is the first I’m learning of T1Ds producing insulin AFTER honeymoon ends…except in clinical trials that never see the light of day. Interesting.
Sounds like a good reason to get on a pump. 
That’s a little different scenario.
My assumption was that it involved being in the U.S., since you referenced A1C’s “across the U.S.”
I have no idea what 'betes supplies cost in Costa Rica, so I have no way of supplying any useful information on that scenario.
I stand by my original thought though, that doing it in the U.S. - with insulin and syringe and CGM and test strip costs what they are - it would be tough to do for under 1g.
Even at the low rate of $30 per vial for the cheap insulin (that’s the Walmart price for the cheap knockoff stuff), if you get only 1 vial of humulin per month and only 1 vial of NPH per month, that’s already $720 per year for just the insulin! You only have $280 left.
I mean, disregard the A1C of less than 6.5, or any A1C measure. Doing it for less than 1 grand per year without insurance in the U.S. is hard, regardless of what your A1C is!
Diagnoses at 6 months or earlier makes it likely that he is not a Type 1 autoimmune diabetic but has Monogenetic diabetes, in particular Neonatal Monogenetic DM.
If so he has Beta cells, but they don’t function properly. The treatment can be insulin or the use of sulfonylureas, drugs that stimulate the Beta cells to secrete more insulin.
Monogenetic Neonatal and MODY are pretty rare and are often misdiagnosed. It is caused by a single gene variant.
I think it was over on DUDiabetes that a member finally got insurance approval for the test that showed he has MODY.
Of course I am no kind of doctor and even if I were, I could not diagose your friend without the test. It just sounds like Monogenetic DM to me. Today if a child goes hyperglycemic at 6 months or earlier they are routinely tested.
Interesting you guys. I’ll have to run this by him. For some reason I’m really stuck on it because I had just never heard all this before. He’s like a rare, exotic specimen that I have never encountered in the wild.
I get it and I don’t have a 100% direct place that I’m going with this line of questioning.
I just feel like his perspective and experience is different than mine (and a lot of our community), so I’m just kinda interested and exploring it and trying to understand how and why. Rare animals are interesting. He’s different than us in notable ways.
You can get tested for NDM for free: Genetic Testing for Neonatal Diabetes // Diabetes Genes
I’m coming in late to this surprising thread and just wanted to thank you and @ClaudnDaye for adding in things I didn’t know. I can’t add anything useful to the thread as I think the AI in @ClaudnDaye’s post nailed my view too.
@mohe0001 I really appreciate the original prompt. I also used to think that data can change people’s perspectives – and I also wonder about whether pumps are better than pens.
But data no llonger matters much in debates like this. You could have the most convincing numbers and your friend would remain sceptical.
That said, I don’t think there are convincing numbers on relative efficacy in any event.
Let us know what else you find out …
