PRINT EDITION

PUBLIC HEALTH | November 22, 2007

Junk Science, Junk Food

    

A letter from the editor

WILLIAM PATRICK

“And if there were ever a cautionary tale about trying to make one-size-fits-all health recommendations, it is the story Taubes relates. ”
Gary Taubes is a man on a mission, a crusader against what he calls “junk science.” Even before reviews of his new book Good Calories, Bad Calories appeared in print, the Grain Foods Foundation began circulating a press release attacking his latest critique, which is that the nutrition establishment set the nation 180 degrees off course when it endorsed carbohydrates and decried dietary fat 30 years ago.

Taubes reinforces his position with perhaps the most detailed analysis of the literature ever assembled. So his opponents, such as the Grain Foods people, line up members of the establishment to say, “No, you’re wrong.” But of course they say he’s wrong—they are members of the establishment that he said was wrong in the first place. And the circular argument goes around and around. 
 
This is the kind of trouble that ensues when scientists yield to pressure—from politics, from financial considerations, from the media, even from the desire to serve public health—and make pronouncements that go beyond what has been proven. And if there were ever a cautionary tale about trying to make one-size-fits-all health recommendations, it is the story Taubes relates. Academics in general, and scientists in particular, are often criticized for making “academic” distinctions, accompanied by highly qualified statements filled with “on the other hands.”  
 
But in the 1970s, researchers were asked to deliver truth on demand. The forum was the Senate Select Committee on Nutrition, which had been assembled to address the question of malnutrition in America. In the last glimmering of the Committee’s existence, its staff decided that they should also address the question of “over-nutrition.” “We need to say something,” was the subtext. “So tell us what to eat to ward off obesity, heart disease, and cancer. Now.” 
 
The problem was that no one had the answer. The data were confusing, inconclusive, and contradictory. But that counts for nothing when set against the motivation to save the lives of large numbers of people. Taubes calls this the “Jurassic Park effect.” If you watch the first movie in that series, every 10 minutes or so, some character is saying, “But people are dying out there!” The scientists were admonished by claims like, “I have to feed my family today. Just give us a guess.”
 
At that point, the first objective of science—to arrive at the truth—was sacrificed to the desire to be helpful. Afterward, the motivation shifted more dramatically, and the effort was no longer a disinterested search and an attempt to refute every hypothesis—including the leading contender—but rather a search for consensus, followed by an effort to justify the recommendation and all the decisions that would follow from it.
 
And here the virtues of a simple, compelling story became paramount. In the case of obesity research, that simple story was “energy in, energy out.” If you burn more calories than you take in, you lose weight—even though every middle-aged person in the land can testify that this formula does not work beyond a few unpleasant weeks. In terms of cardiovascular health, the simple story was that if you eat fatty foods that increase your cholesterol level, you will be at greater risk for heart attack. The fact that neither of these propositions could be proven conclusively made no difference. And the simple equation of “calories in, calories out” sounded convincing—if you ignored fat regulation within the cells—and thus the prevailing wisdom regarding America’s health was set in place by journalists, politicians, and congressional staffers.
 
One of the great truths of any complex enterprise, including science, is that God is in the details. Perhaps a more personalized approach to health and healing can lead to greater humility, honesty, and efficacy in scientific pronouncements.