In this installment of The Practical Scientist we take a look at Bray’s review of Good Calories, Bad Calories. Bray’s review is very in-depth and offers counter points to some of Taubes’ key assumptions. Bray’s review provides a key lesson in separating Fact from Fiction.

Key Points from Good Calories, Bad Calories Gary Taubes; New York: AA Knopf (Bray 2008)

“Good Calories, Bad Calories is a scholarly book that musters the evidence for the case against the high-fat hypothesis for heart disease, cancer and obesity and in favour of the carbohydrate-insulin hypothesis. The bibliography is robust and contains a wealth of information.

The book begins with Mr. William Banting and the diet he published in 1863 as a small pamphlet called ‘A Letter on Corpulence Addressed to the Public’. In this pamphlet, Banting described his dietary success with a low carbohydrate diet. The hostility that Mr. Banting aroused among the ‘medical establishment’ in the 1860s is reminiscent of some of the comments about popular diets that have come from the ‘medical establishment’ in the last half of the 20th century.

The background work on energy expenditure in human beings from Antoine Lavoisier, Hermann Helmholtz, Robert Mayer Carl Voit, Max Pettenkoffer, Max Rubner and Wilbur Atwater is very nicely detailed by Mr. Taubes. In addition to these, there are many other descriptions of scientists and their work that make this book particularly fascinating to read.

Good Calories, Bad Calories is divided into three main parts. The first part is a critique of the Diet-Heart hypothesis, and the idea that dietary fat was the principal culprit in the rising incidence of heart disease during the 20th century. From an analysis of published data and discussion with many leaders, Taubes concludes that the Diet-Heart hypothesis detracted from our understanding of the relation of diet to heart disease. As Taubes sees it, Ancel Keys played the role of major villain in selling the idea

The second part of Good Calories, Bad Calories sets forth the ‘carbohydrate hypothesis.' This hypothesis is Taubes’ basis for explaining the evils of the ‘nutrition transition’ that have afflicted countries moving from their traditional diets to the Western type of high-fat, high-sugar, high-salt diet.

In the third part of Good Calories, Bad Calories, called the ‘Mythology of Obesity’, the author argues that the energy-balance equation does not adequately explain obesity because obese people do not eat more than lean ones, and because they can lose weight eating a large number of calories, provided that the calories are very low in carbohydrates – that is, high in protein and fat.

In the section on ‘The Mythology of Obesity’ and the carbohydrate-insulin hypothesis, there is no mention of doubly labeled water, a sophisticated technique that has allowed us to ‘check’ on the accuracy of self-reports of food intake. Also missing is a discussion of the ‘nutrient’ balance hypothesis. These limitations may change the conclusions that are reached from reading Good Calories, Bad Calories.

Critique of Good Calories, Bad Calories

One summer I admitted a group of four overweight teenagers to our clinical research unit at the Harbor UCLA Medical Center and put them on a 1000-cal diet under direct observation. As expected, all of the girls lost weight and were delighted with the result.

When they left at the end of the summer, one of the girls, who lived only a short distance from the hospital, wanted to come back to see us so she could continue to lose weight. We instructed her on how to keep a food diary. When she returned 2 weeks later, she had gained a few pounds. She showed us her diary, which was very neatly and carefully kept. The average daily food intake was about 300–400 kcal d-1. As she had lost weight eating 1000 kcal d-1 while directly observed in the hospital, we were skeptical of the accuracy of her outpatient recording. We instructed her again on keeping food records. She returned after another 2 weeks, having gained even more weight and with records still showing she only ate 300 kcal d-1 or so. There was an obvious discrepancy, reflecting the difficulty of keeping reliable records.

It was thus clear that this girl was either kidding herself or trying to kid us about how much she was eating. We have subsequently had the opportunity to study weight loss in a number of subjects in both the metabolic unit and then on similar diets in an outpatient setting. Weight loss under observation is about 50% faster than with the equivalent ‘prescribed’ energy deficit in an outpatient setting. The difference is adherence to or compliance with the diet.

Obesity is the result of a prolonged small positive energy surplus with fat storage as the result. An energy deficit produces weight loss and tips the balance in the opposite direction from overeating.”