New trial muddies the water about diet, exercise, and diabetes

Howard LeWine, M.D.

Chief Medical Editor, Harvard Health Publishing

Long-awaited results from a nearly 10-year trial exploring the effect of changes in diet and exercise among people with diabetes weren’t what most people expected. The Look AHEAD trial found that intensive efforts to lose weight by eating less and exercising more didn’t provide any more protection against heart disease—a common co-traveler with diabetes—than standard diabetes support and education. The spin from some media reports is that weight loss doesn’t reduce heart disease risk among people with type 2 diabetes, but I think that’s the wrong interpretation.

In the Look AHEAD trial, researchers recruited more than 5,000 men and women with type 2 diabetes. All were overweight. Half were assigned to a program aimed at losing weight by exercising and cutting calories. People in this group were asked to eat between 1,200 and 1,800 calories a day and to exercise for at least 175 minutes a week. Their goal was to lose at least 7% of their starting weight, and maintain that weight loss. The other half of the volunteers met three times a year for group counseling sessions that focused on the importance of lifestyle changes like more exercise, a better diet, and greater social support to control their diabetes.

Both groups lost weight and did a pretty good job keeping it off. Those in the intensive-change group lost a little more weight (about 18 pounds) than those in the comparison group (about 14 pounds). After almost 10 years, the rates of heart attacks, strokes, heart-related deaths, and hospitalizations for chest pain were the same in both groups.

The headline only tells part of the story

There are several ways to explain why the intensive intervention didn’t seem to do any better than standard care, at least for heart disease. Here’s the “obvious” one: Weight loss clearly helps lower blood sugar, but it may not play an important role in heart attack and stroke risk.

I think a closer look at the Look AHEAD participants, and other findings, tell a more important story:

  • Both groups lost weight at the end of the 9.6 years of the study. The average difference between the two groups was 2.5% of starting weight. It’s possible that weight loss matters for type 2 diabetes and heart disease, but greater weight loss is needed.
  • Fewer people in the intensive intervention group needed medications to lower blood pressure and cholesterol, probably because their diets and exercise did that for them. More people in the usual care group were treated with such medications, which have been shown to decrease the risk of heart attack and stroke.
  • The volunteers generally had their diabetes under control. One measure of diabetes control is a blood test called hemoglobin A1c. The goal for people with type 2 diabetes is a reading under 7.0. Those in the intensive change group had an average of 7.2; those in the comparison group, 7.3. Greater benefits may be in store for lifestyle interventions among people with poorly controlled diabetes.
  • People in the intensive lifestyle intervention group were instructed to eat a standard low-calorie diet that was relatively high in carbohydrates and lower in fat. But such diets aren’t best for cardiovascular health. Urging the intensive change volunteers to adopt a Mediterranean-style diet may have had a greater payoff.

Lifestyle still matters, a lot

People with diabetes are two to four times more likely to have a heart attack or stroke than those without diabetes. And deaths from heart conditions and stroke are the leading causes of death and disability among people with diabetes.

Lifestyle changes have been emphasized as the mainstay of diabetes treatment, especially for people with type 2 diabetes. Exercising more, losing weight, and adopting a healthier diet are clearly good for lowering blood sugar, which translates to benefits around the body.

The results of the Look AHEAD trial don’t contradict the value of lifestyle changes. People in that group improved their blood sugar with fewer drugs, saving an estimated $600 per year. Those in the intensive change group were 30% less likely to have developed chronic kidney disease, a feared complication of diabetes. They also had less self-reported vision problems (diabetes is a leading cause of vision loss), less depression, and reported feeling better.

The Look AHEAD results reinforce for me that diabetes care needs to be tailored to the individual. It is not a one-size-fits-all recipe. It should go beyond blood sugar control and weight loss, and give equal priority to preventing all of the many complications that follow in the wake of diabetes.

Related Information: Healthy Eating for Type 2 Diabetes


  1. alana

    Great words you have shared which is very useful for me and who have suffered from diabetes.

  2. Robert Bramel

    Why is calorie restriction still considered at all? Study after study shows that the failure rate of such diets is essentially 100%. A recent HMC Healthbeat posting seems bizarre to me; the thesis is that the human body is just like an automotive engine, with no ability to dispose of excess calories except by turning them into adipose tissue. Their idea is that if a person only reduces the caloric value of what the person eats by 3500 Kcals that the body will be obliged to lose one pound of weight. No allowance is made for body metabolism adaptation.
    We know that most systems in the body have exquisite feedback. An example is the kidneys; they retain or dispose of many things differently, including water, depending upon body needs. Has anyone ever really, scientifically studied whether the GI system absorbs food differently depending on the body’s current needs? The GI system must have any number of ways of not absorbing that which the body doesn’t need. The alternative, that the system must always, always absorb exactly the same components regardless of hunger level, would run counter to any other body system.
    So why do people have so much adipose tissue (get fat)? Many studies demonstrate that adipose tissue grows in the presence of elevated serum insulin and shrinks as insulin decreases. I don’t know what my insulin level was seven years ago when I was 30 pounds overweight and on a typical high carbohydrate diet, but since stopping carbs my weight has come down to a BMI below 25 without any other lifestyle modifications, my non-alcoholic fatty liver has returned to normal, and my fasting glucose has dropped below 100. No drugs, no special regimens required. My insulin sits at 5. Standard insulin reports think getting insulin below 25 should be a goal. Wow!


    i agree with you. good article though. you can’t rely on drugs and procedures. it’s best to know about the products, and if you do use them, it’s best to use products that are natural. we all need to take vitamins every now and again, or sleeping products, stuff to do with stress. etc… it’s just a part of life. it shouldn’t be our one and only solution, but it does help.

  4. N.D.

    I believe the study may be slightly flawed as it does justify the need for good diet and exercise. We already know this… all diabetics know this. So is there a study that shows which is more important – exercise or diet? I am a insulin dependent diabetic. I know by experience that if I do not get my daily exercise, I can feel sluggish after a day or so until I get back to the routine. I also know that if I get a good work-out like an all day bike ride or actively playing volleyball all day or even hiking all day long; that exercise will stay with me for sometimes two days. This means using less insulin for any typical meal or basal rate. This means even though I ate, I am still felling low. So I hope more studies could help to clarify that exercise is the leading factor, along with insulin, to control diabetes. Eating is the necessary factor. We all need the benefits of eating. We cannot just stop eating. The healthy eating will help the heart but the benefits of exercise far outweigh the diet, which all people should be aware of the effects of unhealthy eating – not just diabetics.

  5. Startup Blogs that Accept Guest Posts

    I agree with this.We have to maintaining a perfect diet and have to do exercises daily.

  6. Karen Okamot

    I am surprised that none of these studies use the ideas of Dr Richard Bernstein who is himself a Type 1 diabetic,…keeps his A1c to 4.5 or there abouts, …exercises, has 6 carbs for breakfast and 12 for lunch and 12 for dinner.. His weight is low, he is 79 years old etc etc . I am Type 2 and on insulin; my daughter Type 1 and my 4 year old grandson Type 1. I started following Bernstein’s diet and dropped from an a1c of 7 to 5.6 immediately. I am working to improve this. Dana Carpendar, a low carb cook, has many low carb recipes that follow this 6-12-12. No one will listen. We are carb intolerant and cannot have a lot of carbs. Small units of insulin, with small amounts of carbs means less threats of lows, good control, and GOOD HEALTH!!!!! It also means less weight gain as we take so much less insulin!!!!

  7. Marilyn Gulledge

    I have had chelation therapy for many years. In 2007 I had cardiac by pass surgery. No heart attack. My twin had a heart attack, sustained damage, but is healing very well., but has not chelates. It is expensive and not covered by I durance. I think I skated the heart attack because of the chelation, and vigilance on my part. I don’t think we will ever know. I still chelate.

  8. Ben Joshua

    We agree with you, that we need to consume “right food” and daily exercise. Thanks for sharing this useful information about diabetes. Keep sharing and stay healthy.
    Best Regards

  9. Coby Hunt

    Yes this is very true. Eating better and exercising is always the best way to go about it…but what about people who have a hard time feeling full?

  10. Arun Mehta

    with all precaution and medication and exercise, why in diabetic patient leg pain persist;please suggest remedy.

  11. RSD

    I learned what is rsd when I was recently diagnosed with it. Given two of my relatives have diabetes, I’m at high risk for contracting it to. I am encouraged to know that living a healthy lifestyle can play a role in good health.

  12. William Anderson, LMHC

    I’m afraid that the story here implies that weight loss doesn’t really help diabetics to reduce their risk of suffering and sudden death. That would be absolutely untrue. Making that impression will discourage overweight diabetics and hurt them. The “intensive group”, the most successful people in this study, only lost an average of 14 pounds (long term). No wonder the improvements were minimal. People who are obese and ill because of it need to lose more to get the improvements they need.

    I am a behavioral therapist who helps overweight people to solve their obesity problem. Most of my clients and patients are 50 to 150 pounds overweight. Most lose 50 to 100 pounds or more and their health improves remarkably. It is not unusual for them to eliminate their symptoms of diabetes, HBP and high cholesterol, and be able to stop taking medications to control those diseases. Their health and quality of life improves remarkably.

    The story leaves the impression that working to lose weight will not really help these people reduce the risk of heart disease, heart attack and stroke. That is just not true. You risk discouraging people who can reduce their suffering and risk of sudden death. You will hurt people unless you correct this false impression.

    William Anderson, LMHC
    Licensed Counselor and Weight Loss Expert
    Author of “The Anderson Method – Secrets of Permanent Weight Loss”

  13. ralphla54

    A new treatment for weight loss also has tremendous impacts on diabetic HbA1c levels

    ADA Poster Presentation Abstract
    Lorcaserin, a selective 5-HT2C agonist, was recently approved for weight management in conjunction with lifestyle modification in obese patients (BMI ≥30) and overweight patients (BMI ≥27) with at least one co-morbidity. In patients without diabetes, proportions achieving ≥5% weight loss and absolute weight loss at Week (W)52 for lorcaserin vs. placebo were 47 vs. 23% and 5.8 vs. 2.5kg respectively (MITT-LOCF). In patients with type 2 diabetes mellitus (T2DM) results were 38 vs.16% and 4.7 vs.1.6kg respectively.

    To limit exposure and maximize benefit the predictive value for >5% W52 weight loss was assessed at W12. Patients not losing at least 5% at W12 (non-responders) should be discontinued.

    Proportions of responders without diabetes lorcaserin vs. placebo were 49.3 vs. 22.6%. W52 weight loss in lorcaserin responders without diabetes was 10.6kg (23 lbs) with 86% and 50% achieving at least 5% and 10% weight loss respectively.

    Proportions of responders with T2DM lorcaserin vs. placebo were 35.9 vs. 11.5%. W52 weight loss in lorcaserin responders with T2DM was 9.3kg (20 lbs) with 71% and 36% achieving 5% and 10% weight loss. W52 reductions in FPG and A1C in lorcaserin responders with T2DM were 29.3mg/dL and 1.2%. W52 reductions in systolic and diastolic BP and heart rate were 3.4mmHg 2.5mmHg and 2.5BPM in lorcaserin responders without diabetes and 2.6mmHg 1.9mmHg and
    3.2BPM in lorcaserin responders with T2DM.

    Achievement of ≥5% weight loss by W12 is a strong predictor of robust one-year lorcaserin responses in weight cardiovascular vital signs and glycemia.

    • Robert Bramel

      How does this study relate to the totality of this special set group? It seems the down-select makes this a fairly small fraction of the total population. First, only those individuals that were sufficiently motivated to begin the study started. What fraction of all contacted individuals began the study? Then, all non-responders at W12 were dropped. What was that fraction of initial participants that were excluded? Then, only 50% of that very select group achieved a ten percent goal even for one year. What would be the fraction of “one year successes” divided by number contacted for the study?

  14. Anthony Breezie

    You cant rely on drugs and procedures for everything. Eating better and exercising more are probably the best things you can do for diabetes. They have really helped me.

  15. Dan

    Thank you for providing a dissenting viewpoint. Most other articles are simple regurgitations of the study, without considering all sides of the story.

Commenting has been closed for this post.