Rethinking diet advice to prevent diabetes

Physicians and other healthcare professionals often tell their patients with type 2 diabetes to avoid eating too much starch and sugar in order to keep their blood sugar from going too high.

But if the patients follow that advice, they’ll end up eating more fat and more protein, which could increase their risk of cardiovascular and renal complications.

Worse yet, a high fat intake may actually keep the patients diabetic.

It was clear by the early 20th century that diets that include a lot of fat result in impaired glucose tolerance whereas starchy, low-fat diets restore the ability to tolerate glucose. Thus, the low-carbohydrate diet that many patients with type 2 diabetes are told to eat could actually be contributing to their diabetes.

A randomized clinical trial published in 2006 showed that a low-fat diet with carbohydrates based entirely on unrefined plant foods providing 75% of calories outperformed the American Diabetes Association’s standard dietary recommendations for people with type 2 diabetes.

The subjects assigned to the high-carbohydrate diet lost more weight, had better laboratory values (including lower HbA1c and LDL cholesterol), and were more likely to be able to discontinue taking at least one of their prescription medications.

They were also more likely to stick to their diet. Although their food choices were restricted (they could eat nothing but vegetables, fruits, whole grains, and legumes), they could eat as much as they wanted. They didn’t have to count or weigh anything, and they never had to go hungry.

A shift to a low-fat diet based on unrefined starches and vegetables is a promising approach for reversing type 2 diabetes and has also been beneficial in cases of type 1 diabetes.

Not only does this kind of diet promote weight loss, it helps to reverse insulin resistance even before the individual has lost much weight. It also has beneficial effects on blood pressure, cholesterol levels, and other problems that can contribute to the complications of diabetes.

Ponder these points. Maybe you should change your practice.

George Lundberg is a MedPage Today Editor-at-Large and former editor of the Journal of the American Medical Association. Laurie Endicott Thomas is former medical editor and the author of Where Do Gorillas Get Their Protein?

Submit a guest post and be heard on social media’s leading physician voice.

Comments are moderated before they are published. Please read the comment policy.

  • Kathleen Summers

    Thank you for sharing this important message! A carb-based diet of legumes, whole grains, veggies, and fruits works -  and we need to spread the word.

  • Ann Wendel

    I have actually seen the opposite to be true with many of my T1 and T2 patients. For nearly a year now I have been researching and utilizing an eating plan that is gluten, grain and legume free, as well as mostly dairy free. Protein comes from high quality sources (grassfed beef, free range chicken and eggs), vegetables (organic when possible) play a big part, fats come from sources of long n-3/n-6 (avocado, olive oil, grass fed meats, wild caught fish), sources of inflammatory n-6 fats are avoided (corn, soy, safflower oil). This eating plan decreases inflammation (important for T1D and all autoimmune illnesses), increases insulin sensitivity, usually leads to weight loss, and doesn’t require weighing or keeping track of calories. I am presenting at the Ancestral Health Symposium at Harvard in August on The Effects of Paleo Diet on Type 1 Diabetes. I invite you to read about this information and make your own decision about how to counsel your patients.

  • Anonymous

    You are apparently advocating a vegan diet, which has been shown to be lacking in vitamin B12. That’s the first flaw, and it may not show up in a short-term study. Second, there are plenty of studies done in humans showing the opposite results, and the only animals that truly seem to react that badly to fat are mice, which are not at all equivalent to humans in their dietary needs. There just isn’t enough good evidence to advocate a vegan diet as strongly as you seem to be doing, and there is plenty of evidence to contradict it.

  • Elizabeth Diop

    it never ceases to amaze me how often traditional medical advice is proven unsound in research studies.  it just makes me wonder what else we’re doing wrong!

  • dianalane

    The study in question compares a vegan diet to the standard ADA diet, which is also high in carbs.
    Here is a link to a study which looked at the effect low-carb diets on type 2 diabetes:

    Here is link to a blog post with further discussion of this issue:

  • TP Singh

    Very misleading opening statement , No MD dealing with a diabetic patient directly or indirectly encourages patient to eat more Fat or increase Proteins in the diet- unless patient is using very low proteins . Most of patient population is not only eating large amounts of corbohydrates- they are eating poor quality- high glycemic index ones.

    Need to analyse each pt diet and individualize the diet. The most important things which help patient compliance is to include their most enjoyed foods in small portions and concept of small portions depending upon each patient calorie intake. It is a life long change for patient and their families .

    It is always a challenge for vegetarians and vegans as well as ethnic populations.Agree with Ann Wedel

  • Jerome Grossman

    For a lot of T2s eating a low-carb diet to control blood sugars, the fact that they would fare worse on a glucose tolerance test is neither here nor there. If you’re not eating a lot of fast-acting carbs, you don’t NEED to improve your insulin resistance–you manage just fine with what you have.  (The study you cite, from 1940, relates to using an OGTT to differentiate between insulin-sensitive (i.e., deficient) diabetics and insulin-resistant diabetes.  It provides insight to low-carbing T2s who may be asked by their doctors to ingest a much higher level of carbs than normal for several days if contemplating an OGTT test, but it isn’t clear that the study really has any implications for regarding what diabetics should eat on a day-to-day basis.  To cite that study in such a way–to state simply that “starch, low-fat diets restore the ability to tolerate glucose”, without explaining what the study was actually about–is highly misleading.
    Your allusion to possible renal and cardiovascular issues cites to a study that discusses specifically high-protein diets, and says no more with respect to coronary risk than that “[h]igh-protein diets ‘may’ also be associated with increased risk for coronary heart disease due to intakes of saturated fat, cholesterol, and other associated dietary factors.”  That statement was written in 2001.  It could be stated no more concretely today, more than a decade later–in fact, it might have to be couched in even more vague terms.  Subsequent research, such as the meta-analysis by Krauss et. al (, has found no strong evidence of causation between dietary saturated fat and cardiovascular disease.  Moreover, there is no reason to suppose that a low-carb diet must be high protein.  Recommendations by Dr. Richard K. Bernstein, for example, author of The Diabetes Solution and a pioneer in the use of the portable blood glucose monitor and carbohydrate restriction to tightly control blood sugar levels, and by many others–advise that patients AVOID consuming excessive protein.

Most Popular