What Do We Really Know About Losing Weight?

By Megan McArdle

Tara Parker-Pope had an article in the New York Times Magazine a week or so ago pointing out that we don't know any very good way to lose weight and keep it off if you are very heavy.  There are obese people who lose weight and keep it off, but they do so by superhuman attention to their diets and energy output.  If they let up for even a short while, the weight piles back on:


During the first years after her weight loss, Bridge tried to test the limits of how much she could eat. She used exercise to justify eating more. The death of her mother in 2009 consumed her attention; she lost focus and slowly regained 30 pounds. She has decided to try to maintain this higher weight of 195, which is still 135 pounds fewer than her heaviest weight.

"It doesn't take a lot of variance from my current maintenance for me to pop on another two or three pounds," she says. "It's been a real struggle to stay at this weight, but it's worth it, it's good for me, it makes me feel better. But my body would put on weight almost instantaneously if I ever let up."

So she never lets up. Since October 2006 she has weighed herself every morning and recorded the result in a weight diary. She even carries a scale with her when she travels. In the past six years, she made only one exception to this routine: a two-week, no-weigh vacation in Hawaii.

She also weighs everything in the kitchen. She knows that lettuce is about 5 calories a cup, while flour is about 400. If she goes out to dinner, she conducts a Web search first to look at the menu and calculate calories to help her decide what to order. She avoids anything with sugar or white flour, which she calls her "gateway drugs" for cravings and overeating. She has also found that drinking copious amounts of water seems to help; she carries a 20-ounce water bottle and fills it five times a day. She writes down everything she eats. At night, she transfers all the information to an electronic record. Adam also keeps track but prefers to keep his record with pencil and paper.

"That transfer process is really important; it's my accountability," she says. "It comes up with the total number of calories I've eaten today and the amount of protein. I do a little bit of self-analysis every night."

Bridge and her husband each sought the help of therapists, and in her sessions, Janice learned that she had a tendency to eat when she was bored or stressed. "We are very much aware of how our culture taught us to use food for all kinds of reasons that aren't related to its nutritive value," Bridge says.

Bridge supports her careful diet with an equally rigorous regimen of physical activity. She exercises from 100 to 120 minutes a day, six or seven days a week, often by riding her bicycle to the gym, where she takes a water-aerobics class. She also works out on an elliptical trainer at home and uses a recumbent bike to "walk" the dog, who loves to run alongside the low, three-wheeled machine. She enjoys gardening as a hobby but allows herself to count it as exercise on only those occasions when she needs to "garden vigorously." Adam is also a committed exerciser, riding his bike at least two hours a day, five days a week.

Janice Bridge has used years of her exercise and diet data to calculate her own personal fuel efficiency. She knows that her body burns about three calories a minute during gardening, about four calories a minute on the recumbent bike and during water aerobics and about five a minute when she zips around town on her regular bike.

"Practically anyone will tell you someone biking is going to burn 11 calories a minute," she says. "That's not my body. I know it because of the statistics I've kept."

Based on metabolism data she collected from the weight-loss clinic and her own calculations, she has discovered that to keep her current weight of 195 pounds, she can eat 2,000 calories a day as long as she burns 500 calories in exercise. She avoids junk food, bread and pasta and many dairy products and tries to make sure nearly a third of her calories come from protein. The Bridges will occasionally share a dessert, or eat an individual portion of Ben and Jerry's ice cream, so they know exactly how many calories they are ingesting. Because she knows errors can creep in, either because a rainy day cuts exercise short or a mismeasured snack portion adds hidden calories, she allows herself only 1,800 daily calories of food. (The average estimate for a similarly active woman of her age and size is about 2,300 calories.)

Just talking to Bridge about the effort required to maintain her weight is exhausting. I find her story inspiring, but it also makes me wonder whether I have what it takes to be thin. I have tried on several occasions (and as recently as a couple weeks ago) to keep a daily diary of my eating and exercise habits, but it's easy to let it slide. I can't quite imagine how I would ever make time to weigh and measure food when some days it's all I can do to get dinner on the table between finishing my work and carting my daughter to dance class or volleyball practice. And while I enjoy exercising for 30- or 40-minute stretches, I also learned from six months of marathon training that devoting one to two hours a day to exercise takes an impossible toll on my family life.

Bridge concedes that having grown children and being retired make it easier to focus on her weight. "I don't know if I could have done this when I had three kids living at home," she says. "We know how unusual we are. It's pretty easy to get angry with the amount of work and dedication it takes to keep this weight off. But the alternative is to not keep the weight off. "
This is not to stay thin, by the way: this woman is 5 foot 5 and weighs 195.  This is what she has to do in order to remain merely very noticeably overweight rather than seriously obese.  And though this news seems to be surprising lots of people, it shouldn't; we have known for years (actually many years) that people's metabolisms and even psychologies change when you put them on severely restrictive diets; that obese people who diet down to "normal" weight act like starving skinny people; and that the success rate of diets in studies is somewhere comfortably under 5%. (Hard to say exactly how far under, because these studies tend to have high dropout rates, and it's reasonable to believe that the people most likely to drop out are probably the ones having the hardest time with the diet.)  The article was a near-perfect rewrite of Gina Kolata's Rethinking Thin, and also of large portions of The Obesity Myth by Paul Campos.

This is not to criticize Parker-Pope--it was a great article!  And a sorely needed corrective to the vast amount of "That fatty should put the Cheetos down" sniggering that fat people are subjected to.  But it's maddening that so many people found this so surprising.  Almost all popular writing about obesity seems to be based on introspection about the writer's experience losing an unwanted 10-30 pounds, augmented by statistics grabbed from other newspaper articles, or the executive summaries of studies they haven't read.  There's almost zero awareness that almost no one loses more than 15-30 pounds and keeps it off long term--and that the reason this is so hard is that it requires constant mental and physical effort to maintain that weight loss that is, for most people, incompatible with anything approaching a normal life.

This lack of awareness is by no means confined to the general public, or opinion journalists.  It is incredibly prevalent among public health "professionals", such as the folks in Georgia who think that the reason that kids in that state are fat is that we haven't made it sufficiently clear that the rest of us think they're revolting.

However, my colleague has provided a bright spot of commentary. As he so often does, he's asked an interesting and fresh question: is this finding merely a result of the fact that they're studying crash dieting?

But I thought her basic conclusion--that weight loss is essentially impossible for majority of humans--was underserved by her reliance on studies that looked a lot like crash diets. 550 calories a day, much of it coming from shakes sounds insane, as does 30 pounds over eight weeks. It isn't shocking that someone on that diet would gain it all back, and then some. In another case the subjects were on a liquid diet of 800 calories a day.

I'm not a scientist, but I have lost roughly a quarter of myself. I've done it at a glacial pace--almost eight years. So glacial in fact that I wouldn't even call it a "diet.": I've gained some in that time, but never yo-yoed back to the heights of my girth. The pattern has been more like lose lot, gain a some, lose some gain a little, lose a lot etc.

Obviously I wish this had happened faster and smoother. But the upshot of taking the long way is that I've learned a lot about how to negotiate world where, at almost every step, cheap high calorie food is at the ready. You can't get that understanding in a lab and you're unlikely to get if your trying to burn of 3-4 pounds a week. That sounds like masochism.
It's a really interesting question, on which I have a few thoughts.  

The first is simply that I don't think that it would be possible to study weight loss at the pace he mentions.

First of all, to really study intake and outtake, you need to have people in the lab, where you can control how much they take in and use up.  (Otherwise, you may not be proving that "this diet doesn't work", but rather, that "the 30 people in this study didn't follow the diet").  You can't do that for eight years.  It's hard enough to do it for eight months, or for that matter, eight weeks.

Second of all, the kind of alternative diet he offers (gradually cut out more and more unhealthy foods) is virtually impossible to specify.  I'm not saying it can't be done by a person.  But it's nearly impossible to describe in a way that constitutes a research protocol.  You can tell someone to cut out Cheetos and the can of coke they eat every day, but if they make that up by taking an extra serving of rice at dinner, you'll see no effect.

Now, you can control for that by having people do very precise journaling, weighing all their food and writing every single thing down.  That will show you if they're making up the lost calories . . . but you're not going to get people to do it consistently for eight years.  You might get some people to do it consistently, but most of them would eventually drop out (or provide you with spotty, useless records).   In the end, the best that could tell you is that the kind of people who are willing (and have enough time) to spend years obsessively journaling their food can keep weight off.  

But beyond that, even if you could study it, I'm skeptical about translating the results into the real world.  Realistically, how many people would sign up for a diet that was going to take them eight years to get close to their target weight?

And the final question is whether Ta-Nehisi has actually reset his set point range, or just migrated from the top to the bottom of it.  Most people have a range of somewhere between 15-30 pounds (it gets larger as you get taller) that they can stay in pretty easily.  At the bottom you're exercising somewhat more control over what you eat than at the top, but it's not really unpleasant; you just have to pass up some of the dessert, cheese, and junk food that you'd like to eat.

This is fundamentally different from what someone who is 5'8 and 100 pounds overweight has to do.  They have to exert rigid discipline.  And that discipline is not a matter of "eating healthy"; it's fundamentally a matter of eating a lot less than you want--of being pretty hungry, and pretty focused on food, every waking moment*.  

If you have a fairly "thin" set point, you can go from the top of your range to the bottom and lose a quarter of your weight (my range is probably slightly higher than that) because the "overweight" portion of your weight at the top of your range is (relatively) small compared to your overall weight.  But if you're 100 pounds overweight, and your "normal" weight is 165, the amount of weight you have to lose is much larger relative to your overall weight--and it almost certainly takes you below that set point range.

Karl Smith, who has done some of the best writing on the web about this, explains it thus:

The problem is that the calorie balance interpertation implies a completely false understanding of what is going on. There is an extent to which geekdom can tolerate this level of nonsense and there is a point at which it must be combated.

I will compare to something I know Kevin gets. The calorie balance logic is equivalent to saying.

Government deficits drain savings. Savings are the engine of growth. Therefore, cutting the deficit immediately is our best shot at growth.
In both cases you are taking an accounting identity

  1. Private Savings - Public Borrowing = Net National Investment
  2. Calories-In - Calories-Out = Calories Contained in the Body
And, treating it as if it were a model of the world.

You have to be aware that public borrowing might effect private savings. In particular if public borrowing stimulates the economy it will increase private income which in turn will increase private savings.

You also need to be aware that Calories-In affects energy and hunger levels which not only feeds back to Calories-Out but also to other Calories-In.

I used to post this thing a lot, but since the blog has new readers it might be worth our while to look at how a properly functioning metabolism responds to a rapid increase in Calories-In



The big question we have is why does this stop working in some people? Just to note, there are many, many other feedback loops that are important as well. I point out this one because it so obvious both that it works in the healthy metabolism and that it fails in the unhealthy one.

You are probably aware of the relationship between diabetes and obesity. It is commonly assumed that obesity causes diabetes. This is in part because even some scientists are fixated on the accounting identity. However, there is a reasonable case that diabetes may cause obesity.

That is, the resistance of the muscles to insulin causes the breakdown in the "sugar rush" response (and other loops) which then breaks down the feedback from calories-in to calories-out.

Now, if it is in fact the case that sugary drinks induce insulin resistance this connection may still hold. However, it is almost certain that the simple minded thinking that in general dropping a 150 calorie item from your diet will not feedback on other metabolic components promotes a fundamental misunderstanding of what's going on.


I think it's certainly possible that very, very slow dieting will reduce set points.  But I also think it's possible that this sort of dieting is just fundamentally different from the kind of dieting that seriously heavy people would have to do.

And that matters, because the seriously healthy people are the ones with health problems.  Let me emphasize that: remember last year when you resolved to "get healthy" and lost twenty pounds?  There's really very, very little evidence that this did anything at all for your health.  

In fact, people with "overweight" BMIs have better mortality rates than people with "normal" or "thin" BMIs.  Going from 185 to 165 is not going to add years to your life.  

Exercising probably will (but there's no evidence that it makes you thinner), and eating better might (though there's also not really much evidence that this--as separate from deliberately restricting your calories--will make you thinner).  But simply losing some weight doesn't have big effects unless you're already really heavy. 

It can have big effects if you're really heavy.  But those people probably need to lose large amounts of weight to cut their mortality risk, not twenty pounds; going from 360 to 330 is fine in its way, but you'll still have all sorts of elevated risks.

So I think a real danger remains that those of us whose set points range from "thin" or "normal" to "overweight" are simply inappropriately extrapolating our experience to obese people whose needs and challenges are fundamentally different.  And unfortunately, I think that's a very hard question to resolve empirically.


*  And yes, this applies to low-carb too; you can keep the weight off if you stay on Atkins/Sugarbusters/Paleo, but just as with other sorts of diets, people mostly don't.  They do great for 6-18 months, and then they start getting cravings for carbs, and eventually they give in, and the weight comes back . . . and the diet's advocates explain that of course paleo works, but not for lazy slobs who inexplicably go off it.  At least from what I've seen, the pattern is really surprisingly un-different from other sorts of diets.

This article available online at:

http://www.theatlantic.com/health/archive/2012/01/what-do-we-really-know-about-losing-weight/250826/