08 August 2008

Take Care of The Black Box - Exercise for Health, Not Weight Loss

I have written enough for now about how eating affects your physical well-being. Next up, exercise.

Many people believe that exercise is a great way to lose weight and keep trim. This fits in with our societal beliefs that obesity is caused by eating to much and not exercising enough, or as Taubes calls it, "gluttony and sloth." That is generally not true, and research backs it up. This recent study conducted by the University of Pittsburgh shows little difference between the exercise groups on a low fat diet ('Weight loss did not differ among the randomized groups at 6 months' (8%-10% of initial body weight) or 24 months' (5% of initial body weight) follow-up.") and the initial weight loss comes back on the same trajectory regardless of amount of exercise. Their conclusion was that 275 mins per week of vigorous exercise in addition to calorie reduction is "important in allowing overweight women to sustain a weight loss of more than 10%." They didn't mention that the women who actually ended up 10% lower weight were also restricting their Calories more than the others. This is a typical case of cognitive dissonance, seen in research all the time.

Here is a pretty good analysis of the study. She has access to the paper and data that I don't. Be careful though. She also says that long term weight loss doesn't happen at all, because of the "setpoint." She's wrong about that. Here's an analysis of another study that shows that physical activity correlates with higher weight. It doesn't address the composition of the weight however (fat vs. muscle).

This is a discussion of another study that shows that society has increased its level of activity, even while obesity rates have been increasing. It puts another nail in the coffin of sloth.

So why doesn't exercise work for fat or weight loss? There are a few ways to look at it. First, and most simply, burning significant calories through exercise takes lots of time and work. An hour of heavy aerobic work by a medium sized person might burn 600 Calories or so above your basal metabolic rate. One hamburger undoes all that.

The other more basic issue is that when you burn those calories through exercise, you are using the fuel in your body. Your cells still need fuel and to the extent you exceed the fat burning capacity of your system, you will feel hunger. Fundamentally, exercise does not blunt hunger like a proper diet can, so it becomes much less automatic and prone to failure. Some people ramp up the exercise and "willpower" their way into additional weight loss, but if you exceed the capacity of your body to burn its own fat, the weight loss will come from muscle instead of fat, and your metabolism will slow down because it senses starvation.

A Thought Experiment - Some Weight Loss Scenarios

You have a guy. He's in ok cardiovascular shape, carrying a little extra fat. Let's say he's are 6' tall, 200 lbs and 20% body fat. His daily energy need is 2,500 Calories. He decides to lose weight. What will be the outcome of the different approaches?

Note: In physics class you learned that a calorie is the amount of heat needed to change the temperature of 1 ml of water by 1 degree centigrade. In nutrition, a Calorie is actually a kilo-calorie (kcal). I have generally tried to follow the convention that capital-C Calorie is a kcal.

Diet-only Approach
Cut carbs, get plenty of protein, restrict calories to 1500 calories per day.

Exercise-only Approach
Eat normal, standard American diet (65% carbs, 20% protein, 15% fat), exercise 600 calories beyond BMR per day.

Intense Exercise and Diet Approach
Cut back on calories as in the diet-only approach. Do lots of aerobics and resistance training (600 calories per day) to enhance the calorie deficit.

Diet-Only Approach
The subject will have a caloric deficit of about 1,000 Calories per day. This will result in a weight loss equivalent of 7,000 Calories or 2 lbs of fat per week. He is carrying about 40 pounds of fat initially, so he will be able to burn a maximum of 1,240 Calories of fat from his body per day. Assuming his low-carbohydrate approach minimizes insulin response, the majority of the calories should be from fat. As he loses weight his ability to burn fat reduces, so he will have to run less of a deficit. Once his body adapts to fat burning, he will likely not feel terribly hungry most of the time.

Exercise-Only Approach

The subject will have a caloric deficit of about 600 Calories per day. This will result in a weight loss equivalent of 4,200 Calories or 1.2 lbs of fat per week. He is carrying about 40 pounds of fat initially, so he can support up to 1,240 Calories of fat from his body per day. One troubling aspect of this approach is that the realtively high carbohydrate content of his diet will likely induce an insulin response and thus inhibit fat-burning. Even though the net calorie restriction is less than the diet-only approach, the subject will feel hungry more and will have to "willpower" his way through to losing weight. With this type of approach, weight gain is highly likely if his routine is interrupted.

Exercise and Diet Approach
The subject will have a caloric deficit of about 1,600 Calories per day. This will result in a weight loss of 10,200 calories per week, or almost 3 lbs of fat. There is a potential pitfall here though. He will be exceeding his body's ability to use its own fat stores by about 360 Calories per day. His low-carb approach will allow for use of his fat as fuel, but once that reaches its limit, he will cannibalize muscle for fuel. This will cause hunger, and likely big drops in his metabolic rate. He will have a lack of energy and his physical capacity will decline.

What Does This Tell Us?
The bottom line is that fat-loss is limited by the body's ability to use internal fat as fuel. Fat loss can occur through either exercise or diet. Just don't push it too hard. Resistance exercise will help maintain muscle strength and also signal your body to send protein to the muscles that have been worked. If the combination of what you eat, plus what you burn from your fat stores is equal to or greater than your caloric needs your muscle loss (or catabolism) will be minimized.

On a low carbohydrate diet, your ability to replenish fuel for your muscles will likely be inhibited in the case of high-intensity or longer medium intensity exercise, so you may wish to increase carbs somewhat for those purposes. I think of it as "buying carbs" with high-intensity exercise. If you don't get sufficient carbs before high-intensity exercise you may bonk (run out of muscle fuel). Here's another interesting perspective on the bonk. With low intensity exercise (walking or light aerobics), you should be able to use fat for the majority of your energy needs.

This link will allow you to look at your own numbers. you will need to export the spreadsheet to your own computer.

But Lots of People Use Aerobics to Keep Trim

We all know people who eat whatever they want, then maintain a super-active lifestyle and so never put on any weight. So those people can argue that exercise is what keeps them thin. But the method contains the seeds of failure. Eventually, most people will get injured or sick. When that happen, people with poor diets and lots of activity will tend to put weight on rapidly. They are swimming upstream with their approach to weight loss and maintenance. You can make progress, but if they stop paddling for a second, they lose everything and then some.

Net calories matter, but it's only part of the story for weight loss. Spark of Reason blog has a great write-up on conservation of energy.

To lose or maintain weight, get your diet right first as spelled out in my previous posts here, here, and here. Do exercise to build and maintain muscle and fitness. If you are going to make one change in your life to affect your weight and fitness, fix your diet first.

Why Exercise Then?

Exercise carries many benefits not related to weight loss. Regardless of the weight issue, exercise has a number of benefits including: physical appearance, maintenance of muscle, confidence/self esteem, ability to be active, improved blood pressure, plus chemical advantages, such as increased High Density Lipoproteins (HDLs). I will cover that in future posts.

Addendum 8/15/08: Exercise, especially resistance exercise, might cause weight to increase (by increasing muscle mass). It's good weight though. Even though your BMI may go up, you will also be decreasing your percent body fat.


  1. Good post, though I wonder about the relationship between dietary carbohydrate and the "bonk" you describe. I've been trying to track down whether or not gluconeogenesis will replenish glycogen stores in the absence of dietary carbs, given enough dietary protein. There's excess capacity, I believe, above the baseline requirements for glucose. The question is just whether the metabolic pathway actually goes there. Since metabolism books seem to always be written in the context of modern carbohydrate-heavy diets, it's been hard to track down.

    From an evolutionary standpoint, it would seem advantageous to have this ability. I would guess that making a large animal lunch/avoiding becoming lunch to a large animal would have constituted intense physical activity. If sources of dietary carbohydrate were scarce (perhaps during Ice Ages), having a metabolic process to keep glycogen topped off would have been advantageous. A more modern example would be the Inuit following their traditional hunter-gatherer lifestyle.

    I would guess that this mechanism does exist in some mammals, such as large predators. The rate of energy delivery from glucose is higher than that from fat, so for brief high-intensity activity, such as running down and killing prey, a carnivore with the ability to maximize glycogen stores in the absence of a source of dietary carbohydrate would have an advantage.

  2. Thanks for the comment Dave. From my experience, when I lift weights or do high intensity exercise while I'm low carbing, I tend to run out of energy at some point and bonk. I don't know if it's due to low carb or low calories. I sometimes do an energy bar an hour or so before I workout hard. That seems to help, although I don't know if it's a placebo effect or real.

    Some folks swear by the anabolic or metabolic diet, which looks like a cyclic ketogenic diet. High calories, low carb during the week, then a carbo load on Saturday and Sunday. Anabolic diet fans claim it's not actually ketogenic.

    Here are a few links: http://stronglifts.com/anabolic-diet/ and
    http://www.metabolicdiet.com/maurodipasquale.htm" (DiPasquale's website)

    Here's a discussion thread that has been going for over three years on t-nation.
    http://www.t-nation.com/tmagnum/readTopic.do?id=658379 (T-Nation anabolic diet thread).

    I believe the claim is that it stimulates the gluconeogenesis pathway.

    It would definitely be bad to bonk with a saber tooth tiger chasing you. On the other hand it would likely be a very short effort and you would succeed or fail quickly. So I wonder how much topping off of glycogen would be evolutionarily advantageous.

    Cheers, e4e

  3. Thanks for the links, I'll check them out. Here's another related paper that coincidentally was cited on Dr. Eades' blog:


  4. Thanks Dave, That's a great paper. In the conclusion, it talks about keotgenic diets not being appropriate for anaerobic activities such as weight lifting. That is the source of my bench press bonk at the gym the other day.

    From the paper.
    "Therapeutic use of ketogenic diets should not require constraint of most forms of physical labor or recreational activity, with the one caveat that anaerobic (ie, weight lifting or sprint) performance is limited by the low muscle glycogen levels induced by a ketogenic diet, and this would strongly discourage its use under most conditions of competitive athletics."


  5. Yeah, I was wondering about that. I still wonder if they've studied this long enough to know whether there is further adaptation that cranks up gluconeogenesis to fill up glycogen reserves. So I sent Dr. Phinney the question. I'll let you know what I hear.

    I eat pretty low-carb, probably average under 20g per day. Maybe the workouts I do (slow burn) don't get me to the bonk, but I've yet to experience it since going low-carb. I'm signing up for judo which looks pretty intense, so it will be a good experiment.

    I suspect there is a level of activity that would require you to replenish with dietary carbs just because the rate of gluconeogenesis couldn't possibly keep up. Michael Phelps' Olympic schedule is probably an example of this.

  6. Fascinating post! I find the whole subject area so confusing but your post has really helped me. I'm lately starting to understand more why I can gain weight when I expect to lose and lose when I'm sure I have gained.
    I hope to learn a lot more about this. Surely this kind of information should be taught in schools!

  7. Here's another question: the purpose of "training" is presumably to cause muscle adaptation, e.g. increase the number of mitochondria, thus increasing both peak power output and endurance. But is there any adaptation occurring in the training regime before muscle exhaustion? And if not, then does the amount of stored glycogen matter for training?

    This is a different question than athletic performance, e.g. if you want to win the gold medal in power-lifting or swimming, you probably want energy stores to be maximized. Maybe this is the point of the metabolic diet you mentioned (lots of material there - still haven't had a chance to read through the details).


  8. Hi Crazylady, I'm glad the post helped. There is so much misinformation out there about exercise and diet. I'm pretty sure that i am not 100% right either, but i do believe it's a good working model, even though I may not have all the science right.

    Dave, Thanks for the followup post. In addition to the muscular adaptation there is the neural adaptation that is especially strong when first starting to workout. It can contribute to a 20% to 40% increase in strength without any change in cross-sectional area of the muscle. I haven't gotten far enough to really understand the mitochondrial impacts.


  9. This comment has been removed by a blog administrator.

  10. thanks for the information keep sharing.