Low-carbohydrate diet proven to be very effective in type 1 diabetics (just as you’d expect)

A fundamental problem with diabetes is the usually-elevated levels of blood sugar (glucose) that go with it. These glucose molecules can react with proteins in the body, damaging them. This process – known as ‘glycation’ – is at the heart of diabetic complications such as cataracts, nerve damage, kidney disease and circulatory problems.

There is also some evidence that high blood glucose levels can impair the uptake of cholesterol into the brain. And while many have done their level best to convince us that cholesterol kills people, it actually performs vital functions in the body and brain.

For example, cholesterol is required in the brain as an antioxidant, an electrical insulator (in order to prevent ion leakage), as a structural scaffold for the ‘neural network’, and a functional component of all membranes. Cholesterol is also utilized in the wrapping and synaptic delivery of brain chemicals (neurotransmitters), and also plays an important role in the formation and functioning of the ‘synapses’ (the tiny gaps between nerve cells that regulate cell-to-cell communication. You can read more about this and the potentially damaging effect glucose has on the brain here.

I was interested to read about a recent study in which the relationship between diabetes/blood sugar control and mental functioning was assessed in adults over a 9-year period [1]. Basically, mental functioning was best in people without diabetes, and worst in people who were diabetic at the start of the study. Those who developed diabetes during the study were, overall, somewhere in between. Also, it was found that poorer control of blood sugar levels led to worse outcomes.

Now, this is an epidemiological study, which means we cannot be assured that diabetes and worsened blood sugar control and diabetes cause ‘cognitive decline’. However, we do also have a plausible mechanism (see above) which might account for this association.

So, what options exist for individuals seeking to control blood sugar levels? One approach might beto avoid eating foods tat disrupt blood sugar levels, including the ‘wholesome wholegrains’ and ‘essential starches’ some health professionals and health agencies insist diabetics should eat. You can read more about this (in my view) nonsense advice here.

One of the arguments that some use to justify the presence of grains and other starches in the diet of diabetics, is that we ‘need’ these foods for blood sugar regulation. Actually we don’t. Plus, it also stands to reason of course that that less we eat of sugar-disruptive foods, the less insulin will be required, and the better blood sugar control will be.

Most diabetics have what is termed ‘type 2 diabetes’ which is usually caused by insulin not working as it should. Some type 2 diabetics take insulin to control their condition, but most do not. In the case of type 1 diabetes, insufficient insulin is being secreted, and these individuals must take insulin to keep blood sugar levels from running out of control. Insulin lowers blood sugar, so it’s often said that starchy carbohydrate is particularly important to type 1 diabetics to ‘balance’ the insulin they’re injecting. But how about this: if a type 1 diabetic eats less carb, they will require less insulin.

So now we have the potential for good blood sugar control and lower insulin requirements. And, as has been noted, this approach is likely to help prevent highs and lows in blood sugar which are common in type 1 diabetic who base their diet on starchy carbs.

In my experience, when one discusses this approach and the logic to it with a type 1 diabetic, they get it in an instant. I’ve met many type 1 diabetics who have worked this out for themselves. And yet, many health professionals and the charity Diabetes UK remain intransigent.

For what it’s worth, some Swedish researchers recently published a study in which a low carbohydrate diet was tested in a group of type 1 diabetics [2]. The study subjects were asked to limit their carbohydrate intake to no more than 75 grams a day. Blood sugar control was assessed with a test known as the HbA1c, which provides a measure of overall control in the preceding three months or so. The HbA1c is usually expressed as a percentage, with less than 5 per cent generally being taken to show very good blood sugar control (typically seen in non-diabetics). The average HbA1c in the study subjects was 7.6 at the start of the study. The HbA1c was retested at 3 months and 4 years.

As with all things, some stuck with the advice regarding carbohydrate restriction, and some did not. Those who did not comply with the advice saw no significant change in their HbA1c levels over time. On the other hand, the subjects who went with the advice saw their HbA1c levels drop to an average of 6.0 per cent. Clinically, this would be seen as a very significant drop clinically, and signal generally much improved blood sugar control.

The authors make the point that individuals adjusted their insulin downwards as carbohydrate was restricted. So, what we have is much better blood sugar control, lower risk of complications, and less need for medication by adopting a lower-carbohydrate diet.

This is a great study, I think, and the reason I use the words ‘for what it’s worth’ above is because the results are exactly what practitioners with experience of this approach in clinical practice and what individuals who have tried this approach for themselves would expect. But at least now, we have a nice, long-term study proving the value of carbohydrate restriction in type 1 diabetes. And the level of restriction was not even ‘extreme’. Feel free, if appropriate, to introduce your health care provider to this evidence. Their response to it might tell you a lot about their capacity to think for themselves.

Diabetics should discuss any contemplated changes to their diet and medication regime with their doctor first.


1. Yaffe K, et al. Diabetes, Glucose Control, and 9-Year Cognitive Decline Among Non-Demented Older Adults Without Dementia. Archives of Neurology, June 18, 2012

2. Nielsen JV, et al. Low carbohydrate diet in type 1 diabetes, long-term improvement and adherence: a clinical audit. Diabetology and Metabolic Syndrome 2012;4:23

28 Responses to Low-carbohydrate diet proven to be very effective in type 1 diabetics (just as you’d expect)

  1. helen 22 June 2012 at 3:57 pm #

    Dr Bernstein’s diabetes solution proved this years ago – he actually is a type one diabetic so he knows it works !!!!
    and low carb is how diabetics ate once upon a time……………..seems the old wisdom was best.

  2. John Walker 22 June 2012 at 4:48 pm #

    Another part of the problem I suspect, is that many ‘experts’ relate diabetes to obesity; yet of course, not all obese people are diabetic; (although type 2 I suppose is a possibility if one is obese). By the same token, not all diabetics are obese. If this was acknowledged we might get to a point where diabetics are told to eat what they should eat, and not what some experts ‘think’ they should eat. More scandalous is the misinformation that is pushed forward in advertising, by GPs, and fitness experts, all based on myths that have been pounded in over the last 40years or so. Fat might cause atherosclerosis in rabbits, but as rabbits, by nature, are herbivores, I would say there is every chance that fat would do them harm. What is ridiculous is the transferring of those findings to humans! Ludicrous.

  3. Ian Day 22 June 2012 at 6:25 pm #

    This is from a T1 contributing to the late lamented “Sweet” magazine.


  4. Dr John Briffa 22 June 2012 at 6:46 pm #


    “Diabetics should discuss any contemplated changes to their diet and medication regime with their doctor first”, but in my experience, you may as well speak with your Aunty for all the good they are.

    I take your point. My comment was a disclaimer.

  5. Jennifer Eloff 22 June 2012 at 8:10 pm #

    Very nice, in-depth article, Dr. Briffa.

  6. John 23 June 2012 at 1:28 am #

    Thanks Doc. Now, can you convince my endo?

  7. David 23 June 2012 at 1:56 am #

    John, you say “Diabetics should discuss any contemplated changes to their diet and medication regime with their doctor first”, but in my experience, you may as well speak with your Aunty for all the good they are. The utter twaddle that the Nutritionist at my hospital, where I attend the Diabetes clinic, is a case in point. It’s all “whole grains, and cereals [as long as you can’y see sugar glistening on the cereal it’s good to eat], plenty of fruit and vegetables and eat low fat”. What utter nonsense! When I mentioned low GI and GL, I was told that “this is all new and still to be proved” (what the hell?).
    Recently, after only 6 months on my Paleo diet [no wheat-based food at all, no potatoes, Pasta or Rice and definitely no bread, but plenty of vegetables, a few berries (but no regular fruit such as apples, oranges, bananas) and about 80g of protein a day plus enough good, saturated animal fat to sate the appetite (avoiding trans fats and all vegetable oils except Olive Oil)], my GP was really please with weight loss, liver, kidney and thyroid function, cholesterol levels and LDL/HDL ratios and told me to keep eating that low fat diet. When I told him what I was doing, you could see the concern written all over his “Government and Food-Industry, Brain-washed” face.
    I ended up arguing with the Diabetic nurse you’d thought that I’d gone mad and didn’t agree with my approach, despite the results. I feel great and have halved my insulin doses.
    So, talking with the Food-Industry, brainwashed medical profession is not something that I would advise. They would never let you down this path to begin with. They help perpetuate to problems, along with those well-meaning fools at Diabetes UK, that some diabetics experience.
    I would recommend reading “Primal Body, Primal Mind” by Nora T Gedgaudas, “Marks Daily Apple” Web Blog and your Web site for better advise!

  8. Ian Day 23 June 2012 at 10:18 am #

    You rightly say “And yet, many health professionals and the charity Diabetes UK remain intransigent.” I have tried to discuss my T2 low carb experience with technical folk at DUK – 12 years diabetic, 7-8 years of DUK diet led to crippling complications – change to low carb reversed complications – 4 years low carb & no complications.

    I did get a phone call from one who admitted she was neither diabetic nor a health professional – she offered to send me the DUK documents I had quoted to them. She agreed to pass my submissions to their research dept. I got this “reply” –

    “Thank you for your reply email dated 15 June 2012.
    We are always grateful to receive feedback and will ensure your comments are taken into account when we update our information in light of the new nutritional guidelines.

    Best wishes
    Clinical Advisor
    Policy and Care Improvement team”

  9. Ian Day 23 June 2012 at 10:27 am #

    I did suggest they at least lurked on the http://www.diabetes.co.uk forum to see the results & confusion their diet advice caused, & how a low carb diet together with support from other diabetics helped others. I hope they do.

    I’ve never seen any posts they could have been made by a DUK.

  10. Janknitz 23 June 2012 at 10:53 am #

    We visited my sister in law tonight and had dinner with her and her friends. One was a Type I diabetic. He has a service dog to alert him to low or high blood sugars and tested and took his insulin right there at the table. It was interesting to see him load up his plate with bread, rice, and dessert (he also took meat and veg) while I took only meat and veg. Did I mention he’s a physician.

    Later he contemplated a second brownie, but ultimately concluded it was “not worth the insulin”. The dog was lovely but maybe his blood sugars would be more steady if he didn’t eat all those carbs.

  11. Dr John Briffa 23 June 2012 at 11:14 am #


    …but maybe his blood sugars would be more steady if he didn’t eat all those carbs.

    Almost certainly.

  12. Greg 23 June 2012 at 12:49 pm #

    I’m a type 1 diabetic with very good control – top of the ideal range quoted. And I eat lots of supposedly “good” carbs. It isn’t easy maths-wise to be successful in this way and I do get swings but I can certainly say it is possible. Meanwhile I very much agree with the “carbs are killing us [= happy go lucky Brits] hypothesis” and eat butter, full fat milk (and yes I have good cholesterol too)… Like just about everyone else I like food with carbs. That’s probably my first problem in moving in the low carb direction. The second problem I get no sympathy for whatsoever – I have struggled to get onto and stay in the normal range on the BMI as I lose weight easily and it is tough to put it back on. I am therefore fearful of changing my eating habits in this way even though I would love to have fewer to insulin-carb calculations to do. Any thoughts?

  13. Waterwoman 24 June 2012 at 12:55 am #

    As a type 2 since 2003, and one who has been enjoying a vlc (less than 30 per day) program since 2005, I find the idea of the “subjects reducing the carbs to 75 g per day” almost laughable. Makes you wonder how many they were eating to begin with. There is nothing new about carb restriction, since all diabetics are intolerant of carbs, it shouldn’t be a “lightbulb” moment that reducing them is conducive to better blood glucose control whether you are type 1 or type 2.

    I’m sure the article will be most helpful to those unaware of the carb restricted programs, and be beneficial to their blood sugar health.

  14. Andrew DiMino 24 June 2012 at 1:50 am #

    “Diabetics should discuss any contemplated changes to their diet and medication regime with their doctor first” Yes we have a responsibility to let our doctors know that we are intelligent and we can read and can research better options for our needs for our bodies. And even if our doctors don’t agree with what we try to do to take better control of our health, we need them to know what we do and to note it in their records so that they will have reference points to when and how we improved ESPECIALLY if they didn’t help.

    I also wanted to let you know we’ve added a permanent link to your site from our “CarbSmart Friends” section and we’ve introduced your article to our readers at http://www.carbsmart.com/will-a-low-carb-diet-help-a-type-1-diabetic.html.

    It was great meeting you and talking to you on the Low Carb Cruise.

  15. Steve_A 24 June 2012 at 5:15 pm #

    John, what is the best dietary advice for someone middle-aged who is a type 1 diabetic, but who has had a recent heart attack. The medical advice is to focus on a conventional low-fat diet to avoid further heart attacks, but what is likely to be his ‘optimum’ health diet.

  16. Reijo Laatikainen 24 June 2012 at 8:18 pm #

    It’s interesting that there is no such things as ‘limitations of the study’ discussion in the whole paper. No side effects like night time hypoglycemia were reported. They cite Skeaff&Miller meta-analysis totally wrong and much more. This belongs to your basket ‘shocking Swedish science’ .Very interesting and provocative observations on own patients anyway.

  17. kate 24 June 2012 at 10:43 pm #

    Read the study. What I take away from it is the statement :
    ‘The model described here may be an option for 10-20% of the patients with type 1 diabetes. ‘

    Worth trying, I guess, given that very strong endorsement!

  18. Wanda Mariano 25 June 2012 at 5:44 pm #

    “Worth trying, I guess, given that very strong endorsement!” – Even if it’s 1%, it would be worth the shot as long as there’s a chance to beat the disease.

  19. jake3_14 26 June 2012 at 2:17 am #

    To Andrew DiMino,

    You suggest that “…And even if our doctors don’t agree with what we try to do to take better control of our health, we need them to know what we do and to note it in.” That’s a great way to get labeled a non-compliant patient, which will result in you getting declined for any private life, disability, or health insurance you apply for when the underwriter looks at your medical records. If you don’t know for certain that your doctor will be receptive to learning that you eat a LCHF diet, just be happy along with him/her at the test results, tell them that you’ve cut out all processed foods and sugar, and ask if you can reduce your medications (if necessary).

  20. Ian 28 June 2012 at 10:44 pm #

    I was just wondering if you’ve read yesterday’s article on the front page of the Daily Express:

  21. jake3_14 29 June 2012 at 12:59 am #

    I found this illuminating post commenting on the “high” CRP found in the low-carb group in the study:

    This CRP catastrophe it total hogwash… The CRP data shows that CRP fell in all three diets from 1.75 to 0.76 for low gI; to 0.78 for low fat; to 0.87 for LC. However they say this about CRP:

    Variables with skewed distribution were log-transformed for analysis. One variable with extreme skew (CRP) was rank transformed for analysis.

    I dont know what rank transformed means, but it sounds like statistical gobbledy-gook. Extreme skew I think refers to the low fat data not being as “good” as the low GI. Maybe LC was compared to low GI, which was “better” than low fat for CRP? Or maybe the scatter of data points? But, anyway, the LC diet did not cause CRP to increase as was said in these stupid reports. It was cut in half.

    And Ludwig, who has been a bigtime anti-LCer for years, in his analysis admits that CRP “TENDED” to be higher. When he says tended, he means it wasn’t statistically significant. In other words, it is statistically the same as low fat and low GI.

    However, CRP tended to be higher with the very low-carbohydrate diet (median [95% CI], 0.78 [0.38-1.92] mg/L for low-fat diet; 0.76 [0.50-2.20] mg/L for low–glycemic index diet; and 0.87 [0.57-2.69] mg/L for very low-carbohydrate diet; P for trend by glycemic load = .05).

    The p for the trend by glycemic load (whatever that means) was .05, barely significant. But in the results table, the overall p for CRP was 0.13, nowhere near statistical significance.

    The whole tone of the article sounds apologetic, and I had to read it twice before I realized it was really pro-LC. Nice comments in it about LC and improvement in components of the metabolic syndrome. The CRP concern is totally made up, and the urinary cortisol excretion – who knows what that means, if it means anything. I’ve been reading about metabolism, inflammation, cardiac risk factors and diets for 10 years and I’ve never seen that referenced before. These guys like Ludwig and Bray will have to retire or die before any real progress will be made in sensible dietary advice.

  22. Andrew DiMino 29 June 2012 at 7:49 am #

    Jake: I am not advocating being combative with our doctors but we need to present other options to them and to not just blindly do what they say if we think we have viable options. And we should consider letting the doctors document in our files what options may be working for us.

  23. Dr. Bill Wilson 2 July 2012 at 2:06 am #

    Another study was published this week showing the a low carb diet beats a high carbohydrate diet when it comes to weight issues and all metabolic parameters.


    We also know that low carb diets are good for your brain. Over time excessive fructose mainly from sugar and HFCS and high glycemic carbohydrates mainly from grains can trigger a form of brain dysfunction we call “sugar-brain”. People with sugar-brain start to crave sweet and starchy food and they develop mild brain dysfunction symptoms. Over time sugar-brain can transition to a serious medical condition called Carbohydrate Associated Reversible Brain syndrome or CARB syndrome. People with CARB syndrome can develop up to 22 brain dysfunction symptoms that interfere with your ability to function.

  24. Erik 1 September 2012 at 12:39 am #

    As an endocrinologist, I found this article independently. Great article. Would be interesting if there was a randomized control trial, but I don’t think that the insulin makers will be sponsoring that!

  25. Liz Smith 5 September 2012 at 8:35 pm #

    I take my diabetic friend shopping every week, she can hardly walk, but likes to go and chose her own food. I’m astounded at how many carbohydrate foods she buys, having a sweet tooth myself, she buys about 5 times the sweet stuff I buy. I’ve asked her if her doctor is ok with all the goodies she buys and says she assumes so. She regularly gets periods where she cannot get her insulin levels stabilized. I’ve printed this off and will give it to her to read, here’s crossing my fingers she will do something about it.

  26. Melissa 7 September 2012 at 5:31 am #

    My now 7 year old daughter was diagnosed with type 1 diabetes over 2 years ago. She eats a low carb diet, lots of bone broth soups, meat and offal from organic animals, organic eggs, organic vegetables and small amount of fruit. We are in Australia, and according to her endocronologist this diet is “lacking carbohydrates”. The diet we have been told that she needs to eat is, weetbix for breakfast, muesli bar and fruit for morning tea, sandwiches for lunch, crackers for afternoon tea, pasta or rice to go with her dinner and cheese and crackers before bed. Her diabetes cannot be controlled on this diet. On the diet she is currently on, she is thriving, she is tall and slim, she is full of energy and her blood sugar is near normal most of the time, her HbA1c is 5.8.

  27. marcsfl 7 September 2012 at 8:26 pm #

    After less than a year of following a low carb and no grain diet, I’m pleased to say that my 20 years of Type 1 diabetes has been re-diagnosed by my Endo as Type 2, and I’ve now reduced my Insulin from 140 units a day to 0. Still taking Metformin (for now). I’d say that this has been a successful n=1 test. Oh, I lost 55 lbs as well, and my Asthma, Sinus Infections, Joint Pain and Memory lapses are also history.

  28. Sally Roman 5 December 2013 at 6:40 pm #

    Hello Dr. Briffa,

    Thank you so much for your very good article.

    We have two Type 1 Diabetic sons who were diagnosed over 5 years ago. Along with two doctors and other researchers, we developed a diet and treatment program. They are still insulin-free, and they have improved pancreatic function. Their last A1cs were 4.9 and 5.0. They are growing and thriving. We help Type 1 Diabetic families all over the world learn healthy diet and lifestyle habits to help them achieve their optimal health.

    Our website is: healthesolutions.com

    Thank you, again, for your valuable insight.

Leave a Reply