The Return of Counter-Conventional Wisdom

Diet and exercise: they were supposed to be the answer to all that ails America’s obesity and health care cost problem.

Signs of this Utopian vision are everywhere.  From entire government departments encouraging healthy lifestyles through fitness, sports and nutrition, government websites that encourage “healthy lifestyles,” and entire community efforts to partner with health care organizations to fight obesity with the hope of cutting health care costs.

What if, believe it or not, when it comes to people with Type II diabetes, diet and exercise don’t affect the incidence of heart attack, stroke, or hospital admission for angina or even the incidence of death?

Suddenly, all health care cost savings bets are off.  Suddenly, we have to re-tool, re-think our approach, understand and appreciate the limitation of lifestyle interventions to alter peoples’ medical destiny.  Suddenly we have to come to grips with a the reality that weight loss and exercise won’t affect outcomes in certain patients.  Suddenly, there is a sad reality that patients might note be able to affect their insurance premiums by enrolling in diet and exercise classes after all.

These thoughts are so disruptive to our most basic “healthy lifestyle” mantra that few can fathom such a situation.  Nor would any members of the ever-beauty-and-weight-conscious main stream media be likely to report such a finding if it came to pass.

And yet, that is exactly what has happened.

The Look AHEAD trial studied 5145 adults with type 2 diabetes who had a body mass index (BMI) > 25.  The purpose of the study was to compare the incidence of  nonfatal myocardial infarction (heart attack), nonfatal stroke, death, or hospitalization for angina between diabetics who received a rigourous weight loss and exercise program with education to just an educational approach alone.   Interestingly, the study failed to show any effect of weight loss and exercise over simple education about the disease in the incidence of these “macrovascular” endpoints.  In fact, the study was stopped early.

So disturbing were these findings to our basic understanding of disease prevention that the principle investigator recently appeared on Medscape in print and in video format to reassure the physician community:  “I can tell you from the outset that we were successful.”

And yet, they were not: they did not affect the indicence of stroke, heart attack, death, or admission to a hospital for angina in overweight Type II diabetic patients one bit, even after 11 years of trying.

You see, it is uncomfortable to sit with the reality that exercise and fitness might not be as helpful as we had hoped at altering certain health care outcomes.  So we ignore these trials.  We don’t report them in main stream media because we don’t like to feel uncomfortable with the realization that there’s much we still don’t know or understand about exercise and weight loss at affecting health outcomes in medicine.

Yet there is so much to learn from trials like this BECAUSE they fly in the face of conventional wisdom.

Maybe we should stop pouring money into fitness rooms and health clubs and promote other intellectual or spiritual pursuits instead.   Maybe we should reconsider the benefits of exercise and weight loss as psychologic more than physical.  Maybe we should de-fund all those government programs set up to promote exercise and fitness as our path to health care cost-savings.

Or at the very least, we should just eat some humble pie, stop fooling ourselves, and understand the limitations of lifestyle interventions like weight loss and exercise to improve medical outcomes or to reduce health care costs in America.

But be careful.

Saying the truth is sure to get you banned from main stream media.


1.Look AHEAD Research Group, Wing RR. Long-term effects of a lifestyle intervention on weight and cardiovascular risk factors in individuals with type 2 diabetes mellitus: four-year results of the Look AHEAD trial. Arch Intern Med. 2010;170:1566-1575. (Pubmed).

Westby G. Fisher, MD, (aka Dr. Wes) is a board certified internist, cardiologist and cardiac electrophysiologist practicing at NorthShore University HealthSystem in Evanston, IL. He is also a Clinical Associate Professor of Medicine at the University of Chicago’s Pritzker School of Medicine. He blogs at Dr.Wes, where this post originally appeared.

8 replies »

  1. You are gave an ADD response here.
    1. These pts were on average 100 kg, nearly 60 yrs. old. 2. Consider the exercise that they did- actually not well known because they have only questionnaires & accelerometer data. They were encouraged to exercise for “more than 10 minutes”. 3. Compare that with the Erickson study of the effect of exercise (group walking) on memory and size of the hippocampus in a group of older but healthier non-diabetic people, PNAS 108: 3017-22, 2011. These older people started group walking for 10 min and built up to 40 min walks three times a week. They had significant increases in the size of the hippocampus, not maintained if they stop all exercise. These obese diabetics exercised alone without supervision and never built up to sustained periods of exercise. The Puget Sound dementia treatment study had older people working at 75% to 85% of heart rate reserve for 45 to 60 min/day, 4 d/wk. This study asked too little. 5. The control group did remarkably well- their treadmill exercise scores improved- you don’t see that with ordinary 100 kg diabetics- the controls changed their life style (less so) and narrowed the difference between exptl & control groups. Mortality in both groups was much lower than anticipated before the study began. Both groups benefited. 6. Now look at the mobility data for the ILI group (Rejeski, N Engl J Med. 2012; 366: 1209–1217, mobility improved in the exptl group and number of medications used and cost of meds was less than controls- that’s very significant (Redmon, Diabetes Care 33:1153–1158, 2010).
    I did two migraine trials – both control & exptl groups improved markedly because someone met with them regularly, i.e. just being in a clinical trial is very helpful when bad habits are a major factor. Drug company was mad because there was no statistically significant difference- that’s what you have here.
    Accelerometers, gyms and that stuff are distractions- you need group exercise building up to progressively longer periods of walking, cycling or dancing. These people were far along and way too heavy- would have expected mobility to decrease and medication use to keep going up- not what was found. If these patients are followed long enough and they continue the ILI there will in time be a significant difference in mortality. The time to start the diet/exercise program is within a year of the diagnosis of diabetes and they must build up to longer exercise bouts- forget the accelerometers. It usually won’t work without group support- this is one area where church members often have real advantages.
    This study asked too little, too late, but they still had beneficial effects.
    The time to indoctrinate people into the benefits of exercise is in elementary school; dancing is the best form of exercise for most people and the easiest to sustain – social benefits.

  2. “Look at the people who come into your office. Now tell me with a straight face that you’re doing them harm by telling them to stop smoking, lose a little weight as needed and get some exercise.”

    You might be.

    What if these people have tried and failed to lose weight – or have yo-yo’ed with their weight to the point that their health is damaged due to the weight loss/gain cycle? Look at the results of the LookAHEAD trial – participants averaged only a 5% weight loss, even with a regimented diet/exercise plan.

    So what if weight loss is not a realistic suggestion for your patient? What if you tell her to lose weight, and what she hears is “Don’t come back until you’ve lost weight”? What if by shaming fat patients, you are driving them away from healthcare.

    This is not a hypothetical. It really happens.

    I’m not saying no doctor should ever suggest lifestyle changes to patients – but doing so automatically and thoughtlessly, without attention to the whole picture, is as likely to be harmful as helpful.

  3. Dr. Wes,

    What was the macronutrient composition of the diet (% protein/carb/fat)? Might that have played a role?

  4. “you’re doing them harm by telling them to stop smoking, lose a little weight as needed and get some exercise”

    Complete straw man, and you know it.

    What Dr. Wes was saying was that “entire government departments encouraging healthy lifestyles through fitness, sports and nutrition, government websites that encourage “healthy lifestyles,” and entire community efforts to partner with health care organizations to fight obesity”
    is a self-perpetuating industry that may not be giving us much return on bucks that could be better used elsewhere.

  5. You had me with you for a moment and then you lost me.

    You’ve overthinking this one massively.

    The correct response to your argument is, that’s all well and fine and good but so what?

    Look at the people who come into your office. Now tell me with a straight face that you’re doing them harm by telling them to stop smoking, lose a little weight as needed and get some exercise. A cardiologist is saying this???

    Go home and get some sleep ..

  6. The study showed that both groups had lower than expected negative outcomes from Diabetes. So maybe the takeaway is that better adherence to medication compliance and other things that come from an education program are key. Also, don’t bold that “it was stopped early”. It was stopped after 11 years, your post makes it seem like it was stopped because it was dangerous somehow. The study also showed other health benefits from losing weight like higher quality of life and lower incidence of sleep apnea.

    Finally, separate studies have shown that weight loss helps people who are at risk of developing diabetes from developing diabetes. So most of what we think we know is still true.

    Another post from “Dr. Wes” where he’s trying to be too clever by half and contrarian for its own sake, but with little to actually say.

  7. “We don’t report them in main stream media because we don’t like to feel uncomfortable with the realization that there’s much we still don’t know or understand about exercise and weight loss at affecting health outcomes in medicine.”

    Funny because I heard about these results in the mainstream media (I don’t know what the link to the Dr. Oz homepage at the end of the OP is supposed to indicate – a more specific link would help).

    I agree that the reults are surprising and counterintuitive, but a clearer perspective is provided here:

  8. Perhaps the point is not that exercise and weight loss are ineffective, but that our ability (as physicians) to get people to exercise and lose weight is limited.