Why carbs and exercise are not the answers to reverse type 2 diabetes

Food pyramid

Several years back, the monumental task of recommending an optimal diet for people with type 2 diabetes was assigned to Dr. Richard Kahn, then the chief medical and scientific officer of the American Diabetes Association (ADA). Like any good scientist, he began by reviewing the available published data.

When you look at the literature, whoa is it weak. It is so weak”, he said. But that was not an answer that the ADA could give.

People demanded dietary advice. So, without any convincing evidence to guide him one way or the other, Dr. Kahn went with the generic advice to eat a low-fat, high-carbohydrate diet. This was the same general diet advice given to public at large.

The United States Department of Agriculture’s food pyramid would guide food choices. The foods that formed the base of the pyramid, the ones to be eaten preferentially were grains and other refined carbohydrates. These are the exact foods that caused the greatest increase in blood glucose. This was also the precise recommended diet that failed to halt obesity and type 2 diabetes epidemics in generations of Americans.

Let’s juxtapose these two incontrovertible facts together.

  1. Type 2 diabetes is characterized by high blood glucose.
  2. Refined carbohydrates raise blood glucose the most.

Type 2 diabetes and carbs

People with type 2 diabetes should eat the very foods that raise blood glucose the most? Illogical is the only word that comes to mind. This happened, not just in the United States, but around the world. The British Diabetes Association, European Association for the Study of Diabetes (EASD), Canadian Diabetes Association, American Heart Association, National Cholesterol Education Panel recommend fairly similar diets keeping carbohydrates at 50-60% of total calories and dietary fat at less than thirty percent.

The 2008 American Diabetes Association position statement on nutrition advised that “Dietary strategies including reduced calories and reduced intake of dietary fat, can reduce the risk for developing diabetes and are therefore recommended”. The logic is hard to follow. Dietary fat does not raise blood glucose. Reducing fat to emphasize carbohydrates, known to raise blood glucose could protect against diabetes?

It further advised that “intake of sucrose and sucrose-containing foods by people with diabetes does not need to be restricted”. Eating sugar was OK for people with type 2 diabetes? This could not realistically be expected to lower blood glucose, and the proof came soon enough.

The 2012 Treatment Options for Type 2 Diabetes in Adolescents and Youths (TODAY) randomized study reduced caloric intake to 1200-1500 calories per day of a low-fat diet. Despite this massive effort, blood glucose was not improved. This classic ‘Eat Less, Move More’ strategy failed yet again.

A comprehensive review in 2013 concluded that several different types of diets did in fact provide better glycemic control. Specifically, four were found beneficial – a low-carbohydrate, low glycemic-index, Mediterranean, and high-protein diet. All four diets are bound by a single commonality – a reduction in dietary carbohydrates, and specifically, not a reduction in dietary fat, saturated or otherwise.

Furthermore, the 2006 Women’s Health Initiative, the largest randomized dietary study ever undertaken, reported that reducing fat did not improve cardiovascular outcomes. Almost 50,000 women followed this low-fat, calorie-reduced diet for over 8 years. Daily caloric intake was reduced by over 350. Yet the rates of heart disease and stroke did not improve whatsoever. Neither did this calorie-reduced diet provide any weight loss. Despite good compliance, the weight difference at the end of the study was less than ¼ pounds despite years of caloric restriction. There were absolutely no tangible benefits to long-term compliance to a low-fat diet.


In people with type 2 diabetes, the story was the same. The Action for Health in Diabetes (LookAHEAD) studied a low fat diet in conjunction with increased exercise. Eating only 1200-1800 calories per day with less than 30% from fat, and 175 minutes of moderate intensity physical activity, this was the recommendation of every diabetes association in the world. Would it reduce heart disease as promised?

Hardly. In 2012, the trial was stopped early due to futility after 9.6 years of high hopes. There was no chance of showing cardiovascular benefits. A low-fat calorie-reduced diet had failed yet again.


Lifestyle interventions, typically a combination of diet and exercise, are universally acknowledged as the mainstay of type 2 diabetes treatments. These two stalwarts are often portrayed as equally beneficial and why not?

Exercise improves weight-loss efforts, although its effects are much more modest than most assume. Nevertheless, physical inactivity is an independent risk factor for more than 25 chronic diseases, including type 2 diabetes and cardiovascular disease. Low levels of physical activity in obese subjects are a better predictor of death than cholesterol levels, smoking status or blood pressure. The benefits of exercise extend far beyond simple weight loss. Exercise programs improve blood pressure, cholesterol, blood glucose, insulin sensitivity, strength and balance.

Exercise enhances insulin sensitivity, without involving medications and their potential side effects. Exercise has the added benefit of being low-cost. Trained athletes have consistently lower insulin levels, and these benefits can be maintained for life as demonstrated by studies on Masters’ level athletes. Exercise programs have proven themselves in obese people with type 2 diabetes as well.

Yet results of both aerobic and resistance exercise studies in type 2 diabetes are varied. Some show benefit for A1C, but others do not. Meta-analysis shows significant reduction in A1C, but not in body mass, suggesting that exercise does not need to reduce body weight to have benefits.

Despite all the benefits of exercise, it may surprise you to learn that I think that this is not useful information. Why not? Because everybody already knows this. The benefits of exercise have been extolled relentlessly for the last forty years. I have yet to meet a single person who had not already understood that exercise might help type 2 diabetes and heart disease. If people already know its importance, then what is the point of telling them again?

exerciseThe main problem has always non-compliance. The spirit is willing but the flesh is weak. This is only more of the game of ‘Blame the Victim’. A myriad of issues may deter an exercise program. Obesity itself, joint pain, neuropathy, peripheral vascular disease, back pain, heart disease may all combine to make exercise difficult or even unsafe. Overall, I suspect the biggest issue is lack of results. The benefits are greatly overhyped and exercise doesn’t work nearly as well as advertised. Weight loss is often minimal. This lack of results, despite great effort is demoralizing.

Conceptually, exercise seems an ideal way to burn off the excess ingested calories of glucose. Standard recommendations are to exercise 30 minutes per day, five days per week or 150 minutes per week. At a modest pace, this may only result in daily 150-200 kcal of extra energy expenditure, or 700-1000 kcal per week. This pales in comparison to a total energy intake of 14,000 calories per week. A single day of fasting creates a 2000-calorie deficit, without doing anything!

There are other well-known limitations to exercise. In studies, all exercise programs produce substantially fewer benefits than expected. There are two main mechanisms. First, exercise is known to stimulate appetite. This tendency to eat more after exercise reduces expected weight loss and benefits become self-limiting. Secondly, a formal exercise program tends to decrease non-exercise activity. For example, if you have been doing hard physical labor all day, you are unlikely to come home and run ten kilometers for fun. On the other had, if you’ve been sitting in front of the computer all day, that ten kilometer run might start sounding pretty good. Compensation is a well-described phenomenon in exercise studies.

The main problem

In the end, here’s the main problem. Type 2 diabetes is not a disease that is caused by lack of exercise. The underlying problem is excessive dietary calories, usually glucose and fructose, causing hyperinsulinemia, not lack of exercise. Exercise can only improve insulin resistance of the muscles. It does not improve insulin resistance in the liver. Reversing type 2 diabetes depends upon treating the root cause of the disease, which is dietary in nature.

Imagine that you turn on your bathroom faucet full blast. The sink starts to fill quickly, as the drain is small. Widening the drain slightly is not the solution, because it does not address the underlying problem. The obvious solution is to turn off the faucet.

In type 2 diabetes, a diet full of processed grains and sugar is filling our bodies quickly with glucose and fructose. Widening the ‘drain’ by exercise is minimally effective. The obvious solution is to turn off the faucet. If the underlying cause of the disease is not lack of exercise, increasing it will not address the actual cause of the problem and is only a Band-Aid solution at best.

Dr. Jason Fung


Top posts by Dr. Fung

  1. Longer fasting regimens – 24 hours or more
  2. How to renew your body: Fasting and autophagy
  3. My single best weight loss tip

Videos with Dr. Fung

More with Dr. Fung

All posts by Dr. Fung

Dr. Fung has his own blog at idmprogram.com. He is also active on Twitter.

His book The Obesity Code is available on Amazon.

The Obesity Code

His new book, The Complete Guide to Fasting is also available on Amazon.



  1. Andy Braithwaite
    Great reading.

    Can you clarify the calorie terms in the Exercise section? You switch between kcal and calories but isn't kcal - 1000 calories or 1 Calorie in food intake. I think you should either use the same unit for comparison or correct the case, otherwise the argument makes no sense.

  2. Kenneth Russell
    One thing that is overlooked in this discussion is the relationship between high insulin levels and sedentary behavior. I am 54 years old and I've been overweight, obese, and morbidly obese for most of my life. I did my first diet in my early teenage years (Atkins '72) and I've been a roller coaster dieter ever since. I was overweight at about 200 pounds in my teens and ended up 440 pounds at age 50. There were a lot more ups than downs. Over these past few decades I had several good stabs at "dieting" - mostly low carb, since that is the only diet I could ever stick to for any length of time. I had at least 1/2 a dozen attempts where I lost 50 to 70 pounds. EVERY TIME that I started making good progress with a low carb diet I also ended up starting an exercise routine and I ENJOYED DOING THE EXERCISE. I didn't enjoy the diet as much as I did the exercise. But when the diet faltered, the exercise came to an end. Why would I stop doing something that I enjoyed doing? Logic would tell me that even though I fell off the diet I should continue the exercise that I loved to help keep the weight off. But that never happened. When I wasn't dieting I was gaining weight and not exercising.

    In February 2014 I weighed 440 pounds, I was a type 2 diabetic with chronic lower back pain. I had sciatic nerve pain, too. If I stood for more than 30 minutes my left leg would turn numb (or worse - shards of pain or a burning sensation). Being planted in my recliner chair was my preferred body position. Then I made LCHF a lifestyle change instead of a diet. Within 3 months I lost 50 pounds. I also started getting fidgety from sitting too long. I kept finding reasons to get up out of my chair. I started doing yard work again (instead of having my kids do it). I may have lost some weight, but I still had that bad back. I still had sciatic nerve pain when I was up and about for too long. But still I WANTED TO GET UP AND MOVE. I started a walking routine. For the first few months it hurt. I would end my walk when the pain got to be too much. But over time I lost more weight, my back got better, and eventually the sciatic nerve pain resolved itself too. Walked 3 or 5 times per week became walking everyday. I've been walking daily for over 2 years. I love my walks. I love exercise. Why? Because I still eat low carb. I think that I am hyper sensitive to carbs. When my insulin is high my body fat is not available for use. When my energy lives or dies by what I put in my mouth I end up gaining weight and conserving energy. When my insulin levels are low and I am fat adapted, then I am compelled to get up and move. That is my thoughts about it, anyway. Eating low carb makes me want to exercise. Eating high carb plants me back in that recliner.

  3. Ed
    For the sake of enlightenment, an alternative method and discussion is the Kempner Rice Diet (true low fat diet, <10% cal. from fat, or less than ~200 calories/day), which was successfully used to treat diabetes. The standard so called "low fat diets" aren't even low fat diets when they're typically at 30% cal. from fat.

    We know about Keto, we know about fasting, we know about true low fat diets. The next few years will be about incorporating it all into macronutrient cycling - LCHF for X days cycled with LFHC X days while incorporating fasting, or, for the more healthier folks, even LCHF for lunch and LFHC for dinner.

    Fat and Carbs must be separated for metabolic health. Its why you see traditional societies thrive on either LCHF or LFHC. Its when the 2 are mixed that issues arise.

  4. Cherise
    I agree with Kenneth. My husband and I have had a related argument. He will come home after talking to people in his Jiu-Jitsu class and say, but carbohydrates are needed for energy. I'll respond with think about how we feel after eating pizza or pasta...like we have to take a nap and like we are not going to do anything but watch TV for the rest of the evening. Then, think how you feel after eating low carb, much more productive and energized. I think there is something about the desire to exercise and insulin.

    I also agree with Ed. I've thought about this mix of LCHF and LFHC before. I'm not sure how it'll work for everyone and I would also add I at least need to be careful with what I eat on LFHC. I remember binging on pasta and tomato sauce (no fat added) and on low-fat cereals with skim milk and on fat-free/processed sweetened food. I think these would be in the <10% fat category, but maybe the wheat, dairy, sugar or just the processed nature of the food made these bad for me. Maybe, LFHC cycles, for some people, but make sure it is real food and maybe no wheat or dairy. Just a thought.

  5. Jeremy
    This is idiotic. Stop giving advice that could kill someone just because it gets more internet traffic. Diabetes is not a result of high blood sugar. Diabetes is when the beta cells in your pancreas cannot produce insulin, inhibiting your muscles' ability to absorb blood glucose, resulting in high blood sugar. This happens when you repeatedly eat saturated fat and cholesterol rich foods. Eat plants
    Reply: #7
  6. Keven
    No Jeremy, your comment is what's idiotic. Type 1 diabetes is a condition where the pancreas quits making insulin, usually in childhood. The reasons for it are not really well understood.

    Type 2 diabetes is really just severe insulin resistance. You become resistant to something by being exposed to it repeatedly. It is the result of constant carbohydrate ingestion.

    Yes you can use a plant based diet to control and reverse diabetes, but it is still the reduction of processed carbs in the diet that makes it work. Keto can also be done vegan and still proves to be better for diabetics than a high carb vegan diet.

    If all you are going to pay attention to is that which aligns with your own confirmation bias, then you should stick to only posting on those articles and forums that you agree with. You can't make a reasonable argument if you don't look at and try to really understand all the science, use reason to come to your conclusions rather than emotions and in this arena argue purely for the betterment of human health, not pushing the agenda of those that have an issue with killing for meat. Understand that if our ancient ancestors hadn't eaten other animals we would have never developed the mental capacity to have the compassion we do for them.

  7. 3 comments removed
  8. Nathan
    Exercise has direct physiological benefits beyond obesity and diabetes that may benefit the individual at the cellular level. Furthermore, there are other benefits in chronic disease prevention via exercise. Yes, diabetics may be more prone to insulin regulation issues, and therefore diet is vital, yet, exercise should not be neglected and is also imperative. Both may even create a synergistic effect.
  9. Abe
    I respectfully disagree, Nathan, that "exercise ... is also imperative." I was over 200 pounds obese. My D2 drugs weren't working after a decade of worsening diabetes, and I was close to having both feet amputated due to PN. 9 months ago, I started a very LCHF diet. I never exercised and still don't (though I naturally move more since it's easier with less mass). Today, not only am I close to my ideal weight, but I am off ALL my D2 drugs, with monthly check-ups at my doctor showing my blood chemistry in the normal range. My doctor is ecstatic about my changes.

    Of course, I'm just one person. But in my dietetic counseling group sessions at our university medical center, I meet many others who're in a similar boat. For DECADES, we dieted while exercising. The effects for nearly all of us were horrible. We got much hungrier after exercising. Didn't matter that we sought out professional trainers who changed our regimen to match the latest findings (duration, intensity, supplements...). Not to mention the costs. Having a personal trainer for 90 minutes a day, five days a week, for over a year, I lost a very tiny bit of weight (despite eating "healthfully," I was so hungry all the time I ate too much). But it was both financially and psychologically unsustainable for reasons the article (above) mentions.

    Exercise (with dieting) works for SOME people to get rid of excess, debilitating weight. For many of us, the primary objective has to be finding a sustainable way to eat so that other interventions (less stress, better rest, physical activity we enjoy...) become more effective. For me, and for many like me, diet has been a life-saver. And for anyone else who's in my boat, I'm a LCHF vegan living on fewer than 30g of carbohydrates per day. Any more than this (carb grams) and my appetite explodes and I simply cannot control my eating. I was an athlete in college. I love sports and physical fitness. But these aren't effective interventions for many of us--and, ironically, are often counter-effective.

    No offense meant. Have a healthful summer.

Leave a reply

Reply to comment #0 by

Older posts