“LCHF dropped a1c 3.5% in 3 weeks which is crazy (no meds)”



Check out this new update from Dr. Ted Naiman. Note that the A1c test measures average blood sugar over 2-3 months. So these massive drops in just a few weeks are truly impressive and indicate that the blood sugar levels rapidly more or less normalized.

Not eating carbs is a super effective treatment in type 2 diabetes.


How to eat LCHF
Losing Weight and Reversing Diabetes – Maureen Brenner
The 2 big lies of type 2 diabetes – Dr. Jason Fung
The Key to Obesity – Dr. Jason Fung
 How to Do an LCHF Diet – Dr. Eric Westman
Diet & Diabetes – How to Normalize Your Blood Sugar
The Top 5 Tips For Weight Loss


New Study: Improved Diabetes Control With a Low-Carb Diet

“Low Carb vs. High Carb – My Surprising 24-Day Diabetes Diet Battle”

Insight: Beating Diabetes with Dr. Michael Mosley – Full Episode Online


  1. Bob Niland
    That's actually pretty interesting, and suggests that not only is HbA1c a moving average, but also a weighted average, with more recent glycation getting more weight, and thus strongly responsive to a major diet change.

    By the way, does DietDoctor have a target range for ideal A1c?

    Reply: #2
  2. Dr. Andreas Eenfeldt, MD Team Diet Doctor
    I don't really have an ideal range to give you. Personally I'm at about 5.0 % (US) or 33 mmol/mol (EU) but as long as you're significantly below the normal maximum you should be fine.
    Reply: #3
  3. Bob Niland
    re: Personally I'm at about 5.0 % (US) or 33 mmol/mol (EU)…

    I would have guessed that, but didn't want to bias the answer.☺

    re: …but as long as you're significantly below the normal maximum you should be fine.

    As I've written before (not sure about here), it's actually difficult to find a chart of A1c vs. all-cause-mortality, but here's one:
    Although observational, the low data points are probably not confounded by diet, because it's not easy to get below 6.5% unless one is on a low carb diet.

    The curve, by the way, is U-shaped. Mortality heads back up down below about 4%, as going that low usually involves some unrelated pathologies. So what's the sweet spot? Somewhere under 5%, I'd guess, but 5.0 may suffice. Here are some targets from other programs:
    8.0% ADA “less stringent”
    7.0% ADA “reasonable goal”
    6.5% ADA “more stringent”
    5.2% Perlmutter (Grain Brain)
    5.0% Davis (Wheat Belly)
    4.7% Bernstein (Diabetes Solution)
    “More stringent” may be ADA-speak for “our grain and sugar industry sponsors don't want you to even be aware of the possibility of going lower”.

    Replies: #4, #11
  4. murray
    It may well be the case that blood sugar below 4% does not cause increased mortality, it's just that the sample size at that low range is skewed by people with illnesses that cause abnormally low blood sugar or have disorders that cause abnormally short red blood cell life, such that the HbA1C test is a misleading metric of average blood sugar. The data presented do not disambiguate these possibilities.
    Reply: #5
  5. Bob Niland
    re: It may well be the case that blood sugar below 4% does not cause increased mortality, …

    That could well be. I personally would be nervous about going below 4.5% without first understanding what's happening in parallel with Fasting BGs and Postprandial BGs.

    re: …such that the HbA1C test is a misleading metric of average blood sugar.

    It apparently can produce persistently idiosyncratic results even in the ranges of interest to LCHF eaters. RBC lifespan may be a factor.

    re: The data presented do not disambiguate these possibilities.

    HbA1c in the context of keto and various LCHF diets really needs some RCT study, but by the time it ever happens, we'll probably be able to predict the results from the growing army of anecdotes.

    Reply: #6
  6. murray
    Bob I'm at the stage of life where the decision battle is going to be fought between armies of anecdotes, within view of the ivory towers of science. Can't wait 30 years for definitive scientific proof, assuming that ever comes. The available science and theory are valuable but necessarily incomplete guides, so one has to make decisions now and has to triangulate from numerous perspectives: nutrition, metabolism, ketone signalling, mTOR, hominid evolution, ethnography, clinical experience, physicians who see the light after getting sick from following standard advice, etc.

    My morning fasting glucose the past six months has generally been 3.8-4.2 mmol/L (over 90% of the time) so I would not be fussed about HbA1c at 4.0%. My HbA1c was 4.5% when measured about a year and a half ago, but my morning blood sugar then was a little higher then (4.4-4.7 mmol/L range, about 75% of the time).

  7. BobM
    Will your doctors allow you to get an HbA1c done? Mine would not. When I asked my GP about getting an HbA1c done (since my fasting blood sugar was 103, which I thought was high -- this was almost three years ago), she said that we test too much in the US. Then she wanted to put me on a statin! (I guess testing too much is bad, but taking a drug for the rest of your life that's shown very little benefit and much detriment is OK.)

    So, I have to pay for my own HbA1c tests. They run about $70, so I don't do them that often.

    Reply: #8
  8. Bob Niland
    re: Will your doctors allow you to get an HbA1c done? … So, I have to pay for my own HbA1c tests. They run about $70, so I don't do them that often.

    You can buy a tester, good for 2 tests, for half that, or US$17 per test, on Amazon. Kits with more strips are available, and may be insurance-reimbursible, but you might not be able to use up all the strips before they expire (± gaming the expiry with refrigeration).

    Don’t know about other brands, but the A1CNow testers are matched to the consumables, and ALL of it expires at the same time (at least a year out on a freshly manufactured unit). You can’t buy more test strips separately. Cartons are dated. On line, look for some statement about expiration date (and use a site with credible recourse if the product turns out to be well into or past expiry).

    Shelf life is extended with refrigeration. The product is typically not shipped refrigerated, however (at least ours wasn’t).

    Bayer exited this business a couple of years ago. Any Bayer kits still on the market are entirely expired.

    re: …she said that we test too much in the US. Then she wanted to put me on a statin!

    You need a real doctor, but you have probably figured that out.

  9. BobM
    Thanks, Bob (from another Bob). The test price also includes fasting blood glucose. I had no idea they had A1c tests you could buy. I'll try those out.

    Recently, I had my "regular" doctor give me a cholesterol and glucose test on Monday after about 12 hours of fasting. I paid for a better cholesterol test, blood sugar, and HbA1c test, which I had done on Friday after 4.5 days of fasting. I want to see what the differences are. Unfortunately, while I'll know fasting blood sugar differences, I will not know HbA1c differences. I'm still awaiting the results of the second sets of tests.

    As for the doctor, it's difficult to find one who believes blood sugar/insulin resistance is bad and cholesterol is meaningless (except as being a marker for insulin resistance), as I do. I've not met one personally.

    Reply: #10
  10. Bob Niland
    re: I paid for a better cholesterol test,…

    Standard lipid panel, or advanced lipoproteins?

    On the standard panel, the only measures I consider useful are HDL and TG (and since we're mooting target values, I'd proposed over 60 mg/dL for HDL and under 60 mg/Dl for TG).

    LDL-C and TC are generally useless, unless at extreme values. The ancient Friedewald equation falls apart entirely for low carb. At most, these metrics might suggest: "get an actual lipoprotein test, like LDL-P".

    Further, if you are actively losing weight, any lipid/lipoprotein or even BG tests are suspect.

  11. chris c
    Look at EPIC-Norfolk (Kay-Tee Khaw) and the five year followup, and a huge New Zealand study (Elley et al.) which both show a pretty linear relationship between A1c and cardiovascular risk, starting at "normal" numbers, below 5%.

    Other studies show a correlation between A1c and microvascular risk, and postprandial glucose spikes and macrovascular risk.

    Some diabetics who test their BG show an A1c which is consistently high, or low, compared to their spot readings. Fructosamine (if available) generally correlates better to the spot readings. So the relationship isn't simple - but close enough for jazz - "normal" numbers correlate to "normal" levels of risk.

  12. Kimberley
    A lot of great questions with no answers. Hmmm. That’s concerning to me. I know from experience, being on a higher carb diet and low fat, following Lindora years ago, it worked. Challenging since it seemed very limited in recipes and taste, but it worked long term. I was about 50 carbs per day. I ate their protein bars that had net carbs of about about 12 if I recall. Worried about doing a high fat diet just because my lifestyle seasonally, does not allow my to run to a bathroom in the case of diarrhea, which many youtubers warned about. And for 4 months out of the year, I might need to walk 6-15 miles a day in my profession. Can anyone touch on the issues with that? I could be miles from a bathroom at any given point. It’s a bit concerning and the restricted carb diet seems a bit challenging and difficult....at times.
  13. Tim
    Why is it so difficult to find a simple chart that shows A1C ranges. I've heard that 5.5 or 5.6 or below is non-diabetic so above that is apparently pre-diabetic BUT how high does that second range go before you're declared type II?
  14. Tim
    It would appear there is quite a lag between questions and answers and that's a bit troubling to me. I'm currently on the trial membership and will have to decide in two days whether or not to start paying. Really need to know someone is keeping tabs on the Q & A here because I consider that an important part of the membership
    Reply: #15
  15. Kristin Parker Team Diet Doctor
    Hi Tim! I am on the Diet Doctor team specifically to help answer questions here on the website. I just joined the team recently in fact. The Mayo Clinic has excellent information about the lab values of A1C and what constitutes Type 2 Diabetes - https://www.mayoclinic.org/tests-procedures/a1c-test/about/pac-20384643

Leave a reply

Reply to comment #0 by

Older posts