The Fat Trap: Is that really what diets are?

This piece in the New York Times Magazine by Tara Parker-Pope has drawn a lot of interest in the last week. The basic idea is what Pope calls "a sobering reality" - "once we become fat, most of us, despite our best efforts, will probably stay fat." Getting to and maintaining a desired (and reasonably healthy) weight is simply not possible and it's not because of a lack of discipline but because your body burns fewer calories at a given weight if you once weighed more. I don't think we have a good handle on how people who drop from say 300 pounds to 150 do in the long run, but that's an extreme and atypical story - the average American male weighs 17 pounds more than he did in 1960, adjusting for changes in height, so most weight loss regimes are in the 20-30 pound range.

You know we love data at this site, so let's look at Pope's citations:

#1: University of Melbourne, 2009, which Pope calls "not conclusive":

Subjects: 50 obese men and women. Male average weight = 233 lbs; female average weight = 200 lbs.

Diet: 500-550 calories per day for eight weeks. (Specifically, Optifast shakes and two cups of low-starch vegetables.)

Results: dieters lost an average of 30 pounds during the study. During the next 44 weeks, dieters gained an average of 11 pounds and reported feeling "far more hungry and preoccupied with food" than before they lost the weight.

First things first - if the subjects lost 30 pounds in eight weeks, then their overall metabolic rate (the number of calories needed to sustain their weight) was 2400 calories per day during the study. That's roughly what we'd expect for a group of men weighing 233 and women weighing 200 lbs and quite a bit more than we'd expect from people 30 pounds lighter than that. Simple calories in-calories out seems to work here.

However, these people were actually on a ridiculous extreme starvation diet during the experiment, so we would expect their bodies to adjust to the starvation and force their metabolic rate way down. Except it doesn't seem like that happened during the diet. Maybe it happened after the diet was over and that's what pushed their weight up; we don't know. But I think the key takeaway here is that you can't learn much from such an extreme experiment - if you're a 233-lb guy who wants to drop 50 pounds, nobody would advise you to do something like this.

#2: Canadian researchers Claude Bouchard and Angelo Tremblay in the 1990s:

Subjects: 31 pairs of male twins

Diet: Subjects were fed 700 calories more than their metabolic rate and did not exercise for 120 days

Results: Subjects gained between 10 and 29 pounds (expected: 24 pounds.) Each pair of twins had similar weight gains.

Is this really in dispute? Yes, some people can eat more than others without gaining weight.

#3: Institute of Biomedical and Clinical Science at the University of Exeter:

Subjects: Schoolchildren

Diet: "snacks of orange drinks and muffins and then allowed to graze on a buffet of grapes, celery, potato chips and chocolate buttons"

Results: children with a particular gene ate 100 calories more than students who didn't have the gene.

Unfortunately, 65% of people of European descent have this gene (and always have). I don't know what to make of the suggestion that 2/3 of the population is screwed in some respect.

#4: Rudolph Leibel and Michael Rosenbaum at Columbia University:

Subjects: 130 individuals for six months or longer at a stretch and measure body far percentage and metabolism.

Diet: Liquid diets of 800 calories a day until they lose 10 percent of their body weight. Then another round of intensive testing as they try to maintain the new weight.

Results: "The data generated by these experiments suggest that once a person loses about 10 percent of body weight, he or she is metabolically different than a similar-size person who is naturally the same weight."

Pope claims that people who lose weight are at "a disadvantage of about 250 to 400 calories." She uses the example of a 30-year-old woman who came in to the program at 230 pounds and was eating 3000 calories to maintain that weight. She dropped to 190 pounds and now needs 2300 calories per day. Pope then claims that a typical 30-year-old 190 pound woman can consume 2600 calories a day. I don't know where that number comes from, but you can play around with the metabolic rate calculator and see if you agree. A 30-year-old 6'2 man with a desk job weighing 230 pounds can eat 2600 calories a day; there's no way a woman can do that.

I'm not saying the study is wrong. It's just that Pope's data is all jumbled up; when we're talking about studies with exact numerical outcomes, you can't do that. Plus, and not that we needed this too, we're again talking about a starvation diet, which we already don't think has positive effects on people.

#5: Pope interviews a 66-year-old woman, Janice Bridge, who dropped from 330 to 195 pounds:

Diet and Results: Bridge weighed 300 pounds in 2006; she spent 9 months on an 800-calorie diet and dropped to 165 pounds. That implies a metabolic rate of approximately 2500 calories. If she's 5'4" and burns 500 calories a day exercising (both are suggested later in the interview) then that's exactly the weight loss we'd expect. In other words, she experienced no reduction in metabolic rate even though she ate only one-third of the calories she needed every day.

Here we get more into Pope's psychology. Her interview subject drinks a lot of water, weighs herself every day, exercises a lot (at low intensity), and makes every effort to eat healthy and eat reasonable portions, including at restaurants, and avoid foods that cause her to overeat later. Bridge is serious enough about keeping the weight off that she keeps a detailed food diary and tracks how many calories she burns when exercising. The catch seems to be that Bridge's metabolic rate should be 1800 calories per day and has now apparently (and I say "apparently" because this is self-reported) dropped to 1500 calories.

Pope writes that even talking to Bridge is "exhausting" and reveals that she has tried to find the same level of discipline but can't do it. Indeed, that's a recurring theme throughout the piece: Pope is "perplexed about [her] inability to keep weight off" but admits that she can't exercise for more than 30 or 40 minutes at a clip because she's too busy with other things. She tells us that her mother was always on a diet (or cheating on it) and that meals growing up went from healthy to KFC, but she seems to be looking for a biological explanation for everything rather than an explanation that involves "family eating habits."

Indeed, she contemplates getting genetic testing for obesity but if the test came back negative she would assume that's because her obesity gene simply hasn't been identified yet. And even when she finds that people who maintain weight loss exercises regularly, eats properly, weighs themselves regularly, watches less TV than average and make sure not to over-eat on weekends and holidays, she focuses in on one thing: they also probably eat 50-300 fewer calories than people who never gained weight in the first place. Unless, as one of the experts in the piece notes, they didn't gain all that much weight and they weren't overweight for all that long - how much is 'that much' and how long is 'that long'? Nobody knows.

A lot of people seem excited about this piece (1028 comments as I write), but I don't feel like I came away from it convinced of anything other than what we already knew: most other things being equal, some people can eat more than others. I read many of the comments and listened to a couple of interviews with the author and I think she decided on a thesis that would absolve her of most of the responsibility for her own situation irrespective of what she found when she researched the issue. Even though she focuses on starvation diets, numerous commenters tell her that specific diets (e.g. Weight Watchers or Veganism a la Bill Clinton) are effective; she tells each commenter that diets, in general, fail 80-95% of the time; she won't admit that she has no data about the effectiveness of individual strategies. It's ironic that the author of an article that attempts to debunk the notion of one-size-fits-all diets can't admit that different diets might actually work for different people.

Real, lasting fitness has always required discipline - that was apparent as soon as guys started hitting the weight room in high school. I work with someone who used to be an MMA fighter - he works out six days a week, has 8% body fat, and has eaten the same lunch for the last 20 years. (I'm not sure I find it all that unjust that there are also guys at work who don't exercise, eat crap but have the genetics to only be a little bit flabby.) Fitness success is no less elusive than success in other areas and it's self-delusion to look for a purely genetic answer for why it's not easy.