Debunking the Hunter-Gatherer Workout: ALR's Response

Blog Node Page Template

September 11, 2012
By Jim Sallis
Debunking the Hunter-Gatherer Workout: ALR's Response

In an op-ed titled Debunking the Hunter-Gatherer Workout in the Sunday New York Times on August 26, 2012, Herman Pontzer from Hunter College comments on a study that he published this July in PLos One.

As someone who has devoted his career to researching and promoting physical activity and preventing obesity, I object to Dr. Pontzer's conclusion that physical activity has nothing to do with obesity. This is a claim that surfaces regularly, so it needs to be addressed repeatedly and thoughtfully. A key point of rebuttal is that, based on their evaluation of the literature, the Institute of Medicine, Centers for Disease Control and Prevention, US Surgeon General, International Obesity Task Force, and the World Health Organization, among other groups, recommend physical activity as part of obesity prevention. However, let's take a look at the Pontzer op-ed and study.

The study was an interesting one. They studied the Hadza people of Tanzania who live a traditional hunter-gatherer lifestyle with little or no intrusion of technology. The investigators used the high-tech method of doubly-labeled water to assess energy expenditure, and compared findings to a sample of "Westerners". I have 3 main objections to the study and op-ed.

First, the results don't add up. The Hadza from Tanzania were on average 65 pounds lighter than the Western sample, and they were much more active, with the Hadza men walking 15-20 miles a day.  How could they conclude there is no connection between physical activity and obesity? Dr. Pontzer compared total energy expenditure, which was actually a little higher in the Western sample. But overweight people expend much more energy than thin people walking the same distance. As pointed out by my colleague Steve Blair, the authors should have compared energy expenditure per kg of body weight, which would have produced a more accurate and obvious conclusion that the Hadza were much more active.

Second, they concluded diet alone is responsible for obesity, but their study had no measure of diet.  Where is the basis for that conclusion?

Third, they made a sweeping conclusion that "inactivity is not the source of the obesity epidemic." They dismissed previous research that shows active people gain less weight. When they say "we are not going to Jazzercise our way out of the obesity epidemic" many people will hear "I don't have to exercise anymore." Did the authors consider the effects of their confusing message to the public, most of whose health is at risk from inactivity?

Because of the prominent placement of this op-ed in the Sunday New York Times, it is likely thousands of people are questioning whether they should bother being active as part of their quest to control their weight.  If some of those people stop exercising because of this op-ed, that would clearly be a negative health outcome because inactive lifestyles are the fourth leading cause of death in the US and world. Did the authors consider the effects of their confusing message to the public, and are they prepared to take responsibility for discouraging people be active?


Great response. Shame on the NYTimes for not accepting it!

Hi Jim
Thanks for the email, and I enjoyed the blog piece. To respond to your three points: first, it's true that the Hadza are smaller, on average, than the Western sample. However 1. The Westerners of similar size expended the same number of calories each day (see our Fig 1) and 2. There was no effect of lifestyle after controlling for body size across groups. Your suggestion that we simply analyze TEE/mass is intuitively appealing but misleading. We've known since the 1930's that energy expenditure increases with body size in a power- law manner, typically as Energy/time = k Mass^0.75. This is Kleibers Law, which I'm sure you're familiar with. A consequence of this is that smaller organisms (and smaller individuals within a species) will always tend to have higher ratios of TEE/mass. This is expected, and is evident within the Western sample as well. The higher TEE/mass ratio in the Hadza is exactly what we expect, and in fact what we see, in any human sample. They fit the Western pattern. As a side note, given the well known scaling of TEE with mass^0.75, I'm always surprised that many human energetics studies focus on TEE/mass when that ratio has a known relationship with mass.

Your explanation of greater TEE in Westerners as a consequence of greater fat mass is also rejected by our results. We included fat mass in our analyses, and still found no difference in TEE between Hadza and Westerners.

Second, while we didn't measure calorie intake (but see our supplementary Fig S2) we take the view that weight gain must ultimately stem from energy imbalance, with intake > expenditure. If expenditures are the same, then differences in weight are most likely due to differences in intake.

Third, for the record, the NY Times approached us and asked for a column. We agreed to summarize the work, including our interpretation. I note that in both the NY Times piece and in the PLoS paper we state that exercise is still very important for health. We feel we were honest with respect to the data.

Finally I'd note that a fair read of the available literature on energy expenditure and weight (some of which is cited in the PLoS paper) shows very little evidence that greater expenditure is protective against obesity. We are not the first to look at the data and conclude that diet plays a larger role than exercise in obesity. There is a large and growing body of literature showing that diet has a much larger effect on weight loss than exercise in intervention studies, and that TEE is remarkably similar across very different lifestyles.

Again, thanks for your interest in the study. All best,

Add comment

(If you're a human, don't change the following field)
Your first name.
(If you're a human, don't change the following field)
Your first name.
(If you're a human, don't change the following field)
Your first name.

Filtered HTML

  • Allowed HTML tags: <a> <em> <strong> <cite> <blockquote> <code> <ul> <ol> <li> <dl> <dt> <dd>
  • Lines and paragraphs break automatically.

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
To prevent automated spam submissions leave this field empty.
By submitting this form, you accept the Mollom privacy policy.