The obesity paradox: Why Coke is promoting a theory that being fat won’t hurt your health

Over the past few years, an unconventional idea has been gaining traction in the scientific literature: Maybe our pursuit of thinness is a big, fat waste of time.

This idea is known as the "obesity paradox," and it's based on studies finding that people who classify as overweight or even moderately obese appear to have better health and mortality outcomes than their normal or thin counterparts. This research suggests that those extra pounds might have protective effects on health, especially when it comes to certain chronic conditions like diabetes, kidney disease, and high blood pressure.

Still, despite the fact that "obesity paradox" science is very much unsettled, it does have one influential champion: Coca-Cola. The world's largest manufacturer of sugary drinks has been promoting the paradox, lavishing its proponents with speaking and consulting fees, and funding researchers who back the idea.

This isn't the first time Coca-Cola has waded into the obesity debates. Back in August, a New York Times investigation found the company had been quietly funding organizations and researchers who blame obesity on too little exercise, rather than on too many calories, a view that might cause people to look askance at soft drinks. Much of Coca-Cola's money went through a nonprofit called the Global Energy Balance Network, which until recently did not disclose its Coke ties.

In September, following that story, Coca-Cola published a list of researchers and health professionals it has paid $2.1 million to over the past five years. The company said these people regularly "collaborate and consult" with them. Coca-Cola also published a list of research institutions and organizations that received $118.6 million for scientific research, as well as to support health and well-being partnerships.

Included on these lists were the leading obesity paradox researchers in the United States.
The researchers have maintained that this funding does not sway their scientific work. But other scientists worry that Coca-Cola's influence could affect which sides of this debate travel the furthest.
"The problem here is there is compelling evidence that the obesity paradox is not a real biological finding at all but just sloppy science," said Andrew Stokes, a professor at Boston University who studies the obesity paradox but does not receive industry funding. Of course, critics of the paradox do not have Coca-Cola's backing behind them. So, Stokes said, "this opposing evidence is not given adequate weight."

The "obesity paradox" started to emerge more than a decade ago. In one of the earliest studies, published in 2003, researchers were puzzled by the fact that heavier patients suffering from heart failure seemed to have better health outcomes than their thinner counterparts. From there, a theory was born. Perhaps having extra weight can be good for you.

But Stokes, along with other researchers, have found a big flaw in the studies on the obesity paradox: They only look at people's weight at one point in time. This is like asking someone who has smoked for her entire life but quit last week about her health, and then classifying her as a "nonsmoker" without including any data about her long-term habits. That, Stokes says, can bias the findings in these studies.

"The paradox disappears," Stokes says he's found in his work, "when we use weight histories to separate out people with weight loss from the normal weight category and restrict the sample to never-smokers. Not only does it disappear but we find that overweight and obesity become significantly positively related to risk of death."

In other words, if you classify the data properly, the evidence is clear: Obesity has all sorts of negative health effects. There's no paradox here.

There may be other causes of the J-shaped relationship between obesity and mortality, which shows an increased risk of death in the slimmest and heaviest people. Frank Sacks, a researcher at the Harvard School of Public Health, said it could be explained by the fact that skinny men and women have "other co-morbidities, other problems, and bad habits."

Many smokers would fall into this category, for example, and there have been studies among the elderly and, again, in heart failure patients showing that really thin people can suffer certain health problems. "But they have lost weight because they're sick," says Sacks. "The weight loss is the result of the disease." In other words, yes, they are thinner, yes, they have a higher morbidity rate, but it's not their low weight that is putting them at a higher risk of death — it's their illness.

Other recent research appears to cut against the paradox. Last year, Stokes published a study looking at more than 10 years of data on American adults between the ages of 50 and 84, accounting for people's longer-term weight histories. This made it possible to break up people who were "normal weight" into two separate groups: those who had maintained a normal weight throughout their lives, and those who were normal weight at the time of the study but had experienced weight loss.

Stokes found that people who had always been a normal weight had an extremely low risk of death, but that the other normal-weight group — that is, people who were formerly obese — had a much higher mortality rate. After redefining the normal-weight category to only include the stable weight individuals, he found much stronger associations between excess weight and mortality.

In another newly published study in the journal Obesity, Stokes and a colleague at the University of Pennsylvania examined data from more than 30,000 participants in the National Health and Nutrition Examination Survey between 1988 and 2011, and found the same thing: When you controlled for weight history (and, separately, smoking), the obesity paradox went away.

Now, this doesn't definitively debunk the obesity paradox. More evidence could emerge later. "We cannot rule out that there might be health advantages associated with overweight and obesity," says Stokes, who only receives funding from the National Institutes of Health. "But our data suggest that any benefits are small in comparison to the harms associated with being overweight or obese."

There are some researchers who continue to pursue the notion that extra weight may have some beneficial health effects. New Orleans cardiologist Carl Lavie is arguably the most prominent voice in the obesity paradox debate, having popularized the idea in his 2014 book The Obesity Paradox: When Thinner Means Sicker and Heavier Means Healthier. He’s written what are considered the most influential papers on the obesity paradox, and appeared in numerous popular news articles extolling the idea.

Lavie's main message is that we have exaggerated the importance of the obesity epidemic. As he put it concisely in a video promoting his book, "Body fat is not always the devil."

Lavie does, however, concede that his critics have a point — and calls for further research in the area. "Stokes is correct that most studies do not account for non-purposeful weight loss prior to study entry, which would be associated with a poor prognosis," he told me by email. "There still is substantial work needed to explain the obesity paradox, and currently except for underweight who generally have the worst prognosis, I am not recommending any to increase weight, although it would be worth many improving their fitness and their muscular strength."

While many researchers remain skeptical of his ideas, Coca-Cola appears to be on board. The company paid for Lavie's consulting work, as well as travel and honoraria to lecture on the obesity paradox. It has also funded webinars by him on the subject. And while Lavie mentions his relationship with Coke in his scientific papers, that disclosure does not appear on his obesity paradox website — nor is it mentioned in the many news articles he's cited in.

Lavie isn't alone. Other key "obesity paradox" researchers in the US — such as exercise scientists Steven Blair and Timothy Church — have scientific relationships with Coke, receiving millions of dollars' worth of unrestricted educational and research grants.

Blair was a major focus of the Times's investigation, since he's the vice president of the Global Energy Balance Network. Church, meanwhile, is connected to Louisiana State University's Pennington research lab, the center that received the most Coke research funding over the past five years ($6.7 million). He's been an author on Coke-funded studies out of that center, suggesting exercise — not sugary food or drinks — is the biggest contributor to obesity. (That view, too, is not shared by many health and obesity researchers. There's overwhelming evidence that suggests exercise has only a small impact on weight when compared with calorie intake.)

In an email last week, Lavie told me that Coca-Cola had no influence on his research agenda. He arrived at his views on the obesity paradox independently of the soda giant, he said, and Coke hasn't directly funded his obesity paradox research, only the related speaking and consulting work. Lavie also pointed out that other money from Coca-Cola for science has been through unrestricted research grants, "so Coca-Cola has no control over the conduct of the studies or the data."

Still, the fact that Coca-Cola funds him means that his views can gain a wider audience. The funding also raises questions about which researchers Coca-Cola chooses to support and why, and whether industry covertly influences the direction of their research.

There's suggestive evidence that Coca-Cola does have the ability to influence science in this area. In one of the largest reviews to date on the science of sugary drinks, researchers found that studies reporting a financial conflict of interest were much more likely to conclude that the effects of soda on weight gain and obesity were unclear. By contrast, studies that had no conflict mostly reported soda was a potential risk factor.

"Since Coca-Cola has a well-established interest in promoting activity as the best way to prevent obesity — not what you eat or drink — it is understandable why the company would fund a researcher who shares this opinion," said Marion Nestle, a New York University professor who's studied Coca-Cola's influence for her most recent book, Soda Politics.
"It’s not that industry-funded investigators are bought," she added. "It’s that the influence of food-industry funding is unconscious, unintentional, and unrecognized."