Jane Brody's New York Times column this week talks about a series of articles in The Lancet, a medical journal. Three of the four articles are focused on public health and the obesity pandemic. The third, of greatest interest to me, focuses on quantifying weight changes based on energy in versus energy out. (I'd like to save the world, of course, but how about me first!) Because I am a geek and because I wanted to avoid all chores and other problems for a while, I flopped on the couch this morning and read the actual Lancet article on this quantification.
So guess what? Existing projections of how much weight you should lose by cutting calories or upping exercise are wrong, and overestimate weight loss. Well, duh. We knew that, those of us out here in the trenches of the weight wars. The sophistication they have added to the calculations isn't overwhelming: very fat people lose weight more quickly than others for a given cut in food or increase in exercise, simply because it takes more calories to maintain their body weight (they burn more calories while lying on the couch) and because they burn more calories during aerobic exercise (it takes more calories to move those big bodies). As your weight declines, you lose that effect and so it takes cutting out more calories or upping exercise even more to continue to lose at the same weight.
(They do not include in their model any assumption that exercise leads to greater muscle mass which burns more calories - they include the relative burns of fat versus muscle, and assume the fat goes away as weight loss occurs, but explicitly do not include any increase in absolute muscle mass. They say it's probably not a big deal for aerobic exercise. Very fat people have to have muscles just to move, of course. I know I've added muscle, not just relatively but absolutely, but I do quite a bit of weight training.)
An additional conclusions they reach which I have to quote because I can't follow all of how they got there: when there is a lot of fat to lose, you will lose faster by adding exercise, but after a certain level of exercise has been added, there is more benefit to cutting food than to adding more exercise. Going beyond the math, they suggest it does make sense to "go on a diet" - have a big, artificial restriction in food - at the beginning of a weight loss program, because otherwise their model's predictions of how long it will take to get to goal is very discouraging. Most books I've read of popular diets follow that approach - lose, then taper off towards maintenance.
They spend some time looking at plateaus. They note that many dieters reach a plateau after eight months, and speculate on why. Apparently no in-patient studies go beyond six months, and they say only in-patient studies are accurate for food intake and exercise records - "diaries are notoriously inaccurate". Because there is no change in the math they have calculated during the six months, they don't think there is a sudden change at month seven or eight to account for the plateau. Instead, they note there is a lag between adding food and gaining weight back, and they hypothesize most "free-living" (their term) dieters stop being rigorous in their diets after a few months, but the weight loss continues and so they think they are ok doing that, but it finally catches up with them and they plateau on the scale with no additional cheating going on. This is a second answer to "why doesn't what I do now work the way it used to?".
There is a side-bar on whether diet composition (eg high carb versus low carb) makes a difference. There is a very grudging tone to any discussion about whether low carb works - there really is a prejudice out there - it's not just purely scientific. They say several studies support that low carb leads to greater weight loss, "at least in the short term". They offer some possible reasons for this: (1) physiology at the cell / molecule level means (I think they say just like Atkins et al) that carbs take less energy to turn into fat; (2) eating protein means you feel more energetic and thus burn more calories; and (3) high protein, low carb may lead to more satisfaction with your diet and therefore fewer calories consumed. They go on to say that since diet logs lie there is no way to tell what people are actually eating. They go on to conclude that any effect from composition doesn't matter over the long term (based on what?) and so they say it doesn't matter over the short term either and a calorie is a calorie is a decent rule of thumb. I don't exactly get this.
So how might I use this information? I don't know. The very short takeaway for me is that if I want to lose weight, I have to keep up the exercise, but I really have to cut calories, and even then it will take a long time.
Sigh. I think I actually knew that already.
No comments:
Post a Comment