Why Exercise is Not the Best Prescription for Weight Loss

runnerIf you are exercising regularly as part of a weight loss program, then good for you. Studies have proven the value of exercise in reducing cardiovascular disease, preventing and treating diabetes, sustaining cognition, enhancing the immune system, and even reducing the risk of getting certain cancers. But narrowly considered, does exercise really help people lose weight? Not as much as you might think.

Most doctors and health agencies tell people that exercise is a key component of weight loss.  The American Heart Association and the American College of Sports Medicine issued new guidelines in 2007 urging 60 to 90 minutes of physical activity a day.  Add in kids, a full time job, and trimming your fingernails before they become unsightly claws, and you’ve got little time left for sleep.  Is getting that much exercise an evidence-based solution?

A recent study, published this year by PLos One, looked at 464 overweight women, and divided them into four groups. Women in three of the four groups were asked to work out with a personal trainer for 72 minutes, 136 minutes, and 194 minutes per week, respectively, for six months.  Women in the fourth group acted as a control and were told to maintain their usual routines.  All the women were asked not to change their dietary habits.

You would expect that the group exercising for 194 minutes per week lost the most weight, right?  Here’s a table of results, with 4, 8, and 12 KKW groupings reflecting the increasing energy expended by the three exercising groups.

All the groups lost weight that was statistically significant from their baseline – that’s the good news – but the absolute amount was only 2-5 pounds.  Unfortunately for the American Heart Association and College of Sports Medicine, there were no significant differences between groups in terms of amount of weight lost, including the control group that didn’t increase exercise at all (P>0.05 for all between-group comparisons).

The authors found that in those who exercised the most, compensatory behaviors like increased hunger and perhaps fatigue cancelled the benefits of burning more calories.  They also speculated that the control group lost a little weight simply by being more conscious of their health through the filling out of the study’s health forms.

The exercise groups did lose about an inch off their waistlines, but were found to have no statistical difference in body fat composition.  Losing an inch is not a bad thing, and perhaps reflects less abdominal fat stores (which have been associated with diabetes and heart disease among other problems).

This study and others like it help explain why all the fitness crazes, and the 20 billion dollars we spend a year on gym memberships, have not necessarily contributed to a svelte nation of Americans.  Exercise, although excellent for our mental and physical health overall, is not a panacea for achieving weight loss.  It naturally makes us hungry.  An article in Time Magazine about this issue cited another study’s conclusions:

Last year the International Journal of Obesity published a paper by Gortmaker and Kendrin Sonneville of Children’s Hospital Boston noting that “there is a widespread assumption that increasing activity will result in a net reduction in any energy gap” — energy gap being the term scientists use for the difference between the number of calories you use and the number you consume. But Gortmaker and Sonneville found in their 18-month study of 538 students that when kids start to exercise, they end up eating more — not just a little more, but an average of 100 calories more than they had just burned.

We also tend to think of exercise as a grueling, Nike-advertisement-looking-muscle-and-sweat-and-grunting-display of physical prowess.  But other studies are showing similar health benefits from more moderate intensity exercises like walking, and the accumulated activities of an entire day (like taking the stairs instead of the elevator).  Even fidgeting seems to help more than the 194 minutes of traditionally conceived exercise:

Overweight people have a tendency to sit, while lean ones have trouble holding still and spend two hours more a day on their feet, pacing around and fidgeting… The difference translates into about 350 calories a day, enough to produce a weight loss of 30 to 40 pounds in one year without trips to the gym – if only heavy people could act more restless, like thin ones.

I would also add that people living a modern urban lifestyle paradoxically mimic the physical activity levels of our more thin ancestors:

Researchers looked at data from 11,541 survey participants in 11 countries, which included the United States, Lithuania, Brazil, Sweden and Japan. Those individuals who reported living in a city neighborhood with easy access to sidewalks were 15-50 percent more likely to get moderate-to vigorous activity at least five days a week for at least 30 minutes each day.

So in general it would seem that current thought leaders are moving away from regular, vigorous bursts of exercise as a prescription for weight loss.  Exercise is very important for the bones, mind, body, and heart.  But for those who are frustrated and defeated by their exercise program’s failures, in narrow terms of weight loss, it may be helpful to refocus on the input side of the equation rather than the output.  Eating better foods, and less of them, seems to be the key for weight loss, with physical activity being more accurately recommended for overall health.

I’m going to walk at a moderate pace to the celery store now, and then come home and fidget for a good hour. Awesome.


23 thoughts on “Why Exercise is Not the Best Prescription for Weight Loss

  1. emmy

    Thanks for all the information. I have sacraficed a few trees and printed it out so I can take it to the gym with me and read it while I’m not losing weight. …seriously, I’m not losing weight.

  2. Josh MD

    interesting… i often hear of people complaining that despite herculean exercise routines, they are still not losing weight. the danger is that they then abandon exercise completely, losing all the benefits as you’ve mentioned. perhaps tempering expectations and clarifying benefits will be helpful as i continue to recommend exercise to patients.

    great post, thanks for your time organizing these thoughts and introducing the PloS One study in particular.

  3. Rob

    As a practicing GP in England, this has been my observation over the last 20yrs of dealing with diabetes and requests for help in weight management.
    I totally agree that there are lots of very good health benefits to exercise but weight loss as a primary outcome is not one of them. I agree with Josh that many abandon exercise as they would any of the fad diets.
    I therefore don’t dwell on exercise when taking through weight loss regimens. This ‘gentle’ approach often blossoms into proper exercise as they begin to feel healthier from some weight loss and a healthier lifestyle

  4. Alden Diet

    In fact, exercise can be anything from sweeping and mopping the floor to dancing around the house with your children. Alden Diet.

  5. george.w

    I’ve been exercising regularly for five years, and have lost little weight. But even so, and despite other physical problems, I can move around better, fill up my car’s gas tank once a month, and recently learned to ride a unicycle. Seems our monomania for weight loss ignores strength gain and how it might contribute to personal freedom.

  6. PalMD

    Yeah, I’ve never been convinced that exercise is the key—an essential, healthy behavior, yes, but without calorie restriction, you don’t lose weight. Ugh.

  7. robin andrea

    It’s hard to counteract the impact of sedentary lives and poor dietary habits. That we have to create an entire industry for exercise, which for most of human history just meant the activity of life itself, means we have separated out activity from our normal routine of living. We would all be healthier if our lives required activity (that is really one of the weirdest sentences I’ve ever written). We have learned to live (albeit poorly) with very little demand on our bodies. I watch sea lions and pelicans everyday, and I am amazed by the amount of energy it takes to get food. We should all be so challenged, but not by something foreign called “exercise” but by life itself.

  8. Greg P

    I wouldn’t argue with these observations and data collection, but I think exercise is far more important than weight loss. The mistake that is made is to think that someone with a normal BMI is healthier than someone overweight or obese. I think it’s one of the problems with some approaches to bariatrics, where weight loss per se is the measure of success.

    Mild exercise, with weight loss, probably reduces the risk of things like DVT, but doesn’t do much if anything for your cardiovascular health. One thing you see as you watch runners (the more common ones, not the professionals or those obsessed with it) is that there is a wide range of BMIs for those who seem to be able to tolerate exercise.

    There are, of course, negative aspects of exercise, such as the cumulative effect of jogging on knees and hips, but there are plenty of ways to stimulating your cardiovascular system other than running.

  9. Finn

    If the recommended level of exercise is 60-90 minutes per day, why were the study subjects divided into groups that did less than 30, 20, or 10 minutes of exercise per day? I mean seriously, less than 10 minutes of exercise a day? What’s the point? I get more exercise than that walking from my desk to the coffee pot and the restroom.

  10. julie

    Finn-walking from your desk to bathroom isn’t exercise, it’s considered NEAT (non-exercise activity thermogenesis), and a hypothesis I’ve been hearing recently, is that this often goes down as exercise goes up (two hours at the gym is more likely to make me more sedentary than usual the rest of the day). This NEAT can be a lot of calories-compare somebody cleaning house vs reading blogs or watching tv. It adds up to a lot, day after day. I don’t think this is responsible for our obesity epidemic, that’s more likely due (in my opinion) the huge amounts of crappy food we eat, but may contribute to it.

  11. Isis the Scientist

    Unfortunately for the American Heart Association and College of Sports Medicine, there were no significant differences between groups in terms of amount of weight lost, including the control group that didn’t increase exercise at all …

    But, none of the groups exercised in accordance with the AHA or ACSM’s guidelines. The AHA’s recommendation of 60 minutes most days of the week is 300 minutes. The highest activity group here is 194 minutes. So, it’s more unfortunate for the people in the study that they weren’t given a program in line with the current recommendations. They also were not counseled in proprer nutrition. I might be going out on a limb here, but I am going to guess that the people in that study didn’t get obese eating apples and skinless chicken breasts.

    The purpose of this study was to determine whether there are compensatory mechanisms that might hinder weight loss with exercise. The conclusion is, yes, there are. So, rather than writing off the benefits of exercise, how do we mitigate the other factors that can sabotage results?

  12. drcharles Post author

    I’m not sure anyone reads the comments down this far, but Dr. Isis posted a response on her blog about this issue, and it got her all mad. Since she seems to think I’m “writing off the benefits of exercise” I’d like to post my response to her on my blog as well:

    Hi Dr. I,

    Just a few comments to perhaps lessen your ire, and clarify things.
    Not sure if you read my whole post, but I also speculated that a decrease in waist circumference might be good independent of the narrow focus on absolute weight loss, since increased abdominal fat stores are associated with insulin resistance and even some forms of cancer.

    Secondly, the American College of Sports Medicine (cool that you’re a member… I can still dunk a tennis ball)…. the ACSM recommending 60-90 minutes of exercise 5 days a week… that’s pretty hard for most people to budget, and probably leads to a lot of people giving up before they’ve begun. Is that 60-90 minute recommendation evidence-based? Can you point me to the relevant Category A/B evidence the ACSM used? I’d really like to read it as a counterpoint to the PLos One study (which probably falls into Cateogry B evidence). Or was this just a “consensus statement” which as I’m sure you know carries less weight, and is very prone to dogmatic repetition of thought leaders’ conjecture (see deteriorating evidence that very tight glycemic control in diabetics improves outcomes, very counterintuitive, but thoroughly disproven, and thought leaders and most physicians are still dogmatically chasing HBA1c levels below 6.5, to the possible detriment of their patients’ health)

    I would guess that if you asked the authors of the Plos One study if they had required a cohort of the women in their study to cram in an additional 106-256 minutes of exercise per week, they might speculate that the “compensatory mechanisms” (that rendered all four cohorts’ weight loss statistically insignificant when compared with one another) might also be expected to occur. Unfortunately, they might hypothesize that the extra exercise might not break the trend, and would certainly be unsustainable as a lifestyle, and that a successful weight loss strategy includes many other factors including dietary education as you mentioned, but that was outside the scope of this article and my post on it.

    As a clinician, not a researcher, I can tell you that some people are very successful in losing weight, and most do it with diet and exercise. I also see a lot of people who are unhappy, because they are exercising like crazy, trying to eat right, and their weight is not budging. As I mentioned in the very first paragraph and in multiple other places in my post – exercise is wonderful for a lot of reasons. Exercise may not need to be terribly vigorous, but rather the accumulated exertions of an entire day. Did you check out the fidgeting stuff? Interesting. I recommend exercise to just about all my patients without contraindication… but perhaps telling single moms, middle aged professionals with arthritis, and grad students preparing for their oral defenses to find one and half hours, Monday through Friday, for the rest of their lives (?), to dedicate to running on a treadmill as a way to lose weight…. Please show me the concrete evidence, ala this PLos One study, that this is effective and sustainable. The real world lacks your consummate will power!

    What and how much we eat seems a better prime determinant of whether we lose weight or not, with exercise being very important for many reasons and certainly not detrimental, unless people are given the wrong expectations. I guess that’s the bigger question… do you think people will stop exercising if more studies like this fail to reinforce current dogma?

    I wish I could work out for 2 hours every day, and on some days I do. No one is telling people not to.

  13. chairman meow


    At some point a lot of these things have to be decided on an individual level, as your body generally tell you what is good. Basically you should eat plants, and stay active. If you do that, your brain will release some tasty endorphins and let you know you are on the right track.

  14. Matt M

    I, too, have observed that exercise does not result in weight loss. However, it does result in greater strength, lower resting heart rate, and increased energy, in my case. I feel better and stand straighter. As for my belly, I suppose you could say that I have a higher grade of bacon, now.

  15. ERV


    Former power-lifter: If you are completely sedentary, steady-state cardio (50% VO2 Max?? ) is better than nothing. If you are serious about losing weight and changing your physique, you need to lift weights. Not 5 lb dumb bells. Weights.

    Im 5’8″-9″-ish. In 2 years I went from 130, 25% body fat (at least one hour a day on a treadmill/elliptical/bike), to my peak 160 lbs at 13% bf with power-lifting ONLY. Zero cardio.

    I dont have time for power-lifting in grad school, and I have to run the dog (steady state cardio, dirp!), but I make time for heavy lifting, so now I am comfortably at 135, 15% bf.

    Unless you are injured or otherwise completely sedentary, Im not surprised at the results of this paper… Well, I am surprised the weight loss was so bad… ‘Max’ of 4.6 lbs lost in 6 months? Shit… Get those ladies to a real gym with a real training program. God that study is almost cruel.

  16. Sleeve98

    In ’93 I was, on doctor’s orders, Non-Per-Oral (no food or water), for about six weeks. Despite intravenous “hyperalimentation” I lost so much weight I looked like a crack fiend, going from ~180 to roughly 150lbs. No exercise, just a pack-and-a-half a day to smoke, since I couldn’t eat. No, exercise isn’t necessary for weight loss if you’re STARVING!

  17. camilla

    No where in this post does the author note that the study under discussion was limited to sedentary, overweight or obese, postmenopausal women with elevated blood pressure. Importantly, the original paper states “we do not know if the results will apply to other women or men”.

  18. ERV

    Camilla– “sedentary, overweight or obese, postmenopausal women” (or men), are the only people I would have thought this very light ‘exercise’ would work for. They walked, very, very slowly, for 30-90 minutes, instead of doing nothing for 30-90 minutes. 50% VO2 Max. Thats minimally better than nothing.

    If “sedentary, overweight or obese, postmenopausal women” have pathetic weight loss results with this pathetic routine, why do you think healthy adults should do it when they are capable of more intensive, functional programs??

Comments are closed.