The Growing Groupthink Problem in Science and Policy

Government Dietary Guidelines issued jointly by the Departments of Agriculture (USDA) and Health and Human Services (HHS) every five years are meant to provide health providers and policy makers with current and reliable information about dietary patterns impacting human health. Science, they say, is the foundation of these guidelines. And we should trust science, right? Unfortunately, the institutions charged with synthetizing research into “the science” and science-based policy have done little to inspire public faith. Too often, we have seen these institutions choose politics over evidence and not just as a result of industry influence. As the process of creating the dietary guidelines indicates, science often plays second fiddle to dogma and political agendas among those selected as scientific experts.

To ensure the guidelines are based on up-to-date evidence, HHS and USDA convene a committee of advisors to review existing research, usually 15 to 20 prominent members from various fields of expertise. Once they review all relevant evidence, this committee makes recommendations to the government about what, if any, changes they think should be made to the guidelines. The government is under no obligation to implement the committee’s recommendations, a flexibility that has sometimes created controversy. 

That the ultimate decision about the guidelines rests with the agencies opens the door for industry lobbying, with various sectors vying for recommendations favorable to their bottom line. Thus, the government’s choice to diverge from the recommendations of the advisory committee is invariably followed by accusations that the authorities chose to ignore experts in favor of powerful industries. But this discretion also serves as a safeguard against any biases of the committee members, should those biases inappropriately influence the committee’s recommendations. As we’ve seen over the past year, such protections remain vital.

Experts—including scientists—are not perfectly objective. Even those at the highest echelons of their profession are subject to personal biases that can color their interpretation of evidence. Systems like peer review seek to address these vulnerabilities but are insufficient when ideology spreads throughout a field and peer reviewers share the same biases. For whatever reason, nutrition science appears particularly susceptible to this type of groupthink. Take, for example, the 2015 dietary guidelines.

Led by “environmental nutritionist” Angela Tagtow, the advisory committee based its recommendations on some non-dietary considerations, like climate change and agricultural sustainability. This led the committee to favor a plant-based diets as promoting both the health of individuals and the planet. Tagtow, whose consulting firm described its mission as establishing “healthier food systems that are resilient, sustainable, [and] ecologically sound,” had a clear bias, apparently shared among the 15 members of the advisory committee and a growing contingent of nutrition scientists who voiced support for the move.

The meat and dairy industry, unsurprisingly, opposed the recommendations. But they weren’t the only ones. Experts in many fields, even those who share the committee’s interest in mitigating climate change, raised concerns about the appropriateness of using the dietary guidelines to that end, provoking deeper questions regarding the purpose and parameters of the guidelines.

Without a doubt, human food systems impact the environment. Since human health is dependent on the environment, providing information about how food choices might interact with environmental well-being is not without merit. But, as agricultural economist Jayson Lusk wrote, “the trouble comes when a single set of omnibus recommendations attempt to integrate disparate issues like environment, nutrition, and other factors.”

The dietary guidelines were meant to focus on nutrition; hence, none of the 15 members of the advisory committee had a background in agricultural or environmental sciences. They were nutritionists, physicians, and public health specialists, because the purpose of the dietary guidelines is not to answer environmental questions, but rather, what dietary patterns are most likely to promote individual health and reduce disease risk.

That was the conclusion of the Obama administration’s HHS and USDA, which deemed such concerns beyond the scope of the dietary guidelines mandate. Of course, critics ignored valid questions about the purpose of the guidelines and instead painted the government’s rejection of the committee’s recommendations as bowing to corporate lobbies. It was, according to experts like Jeffrey Sachs, then-director of Columbia University’s Earth Institute, a “shameful abnegation of political responsibility.”

As Sachs unintentionally reveals, inclusion of environmental concerns in the dietary guidelines was, at least in part, politically motivated. But the guidelines should not be a platform for any political agenda, no matter how valid or popular. To that end, the government’s rejection of the committee’s sustainability language was not a victory of industry influence over science as critics claimed; it was necessary to protect the guidelines from being infected by politics.

Regardless of how biased the recommendations, there will always be those who view anything less than unquestionable deference to “experts” as science denialism. The most recent example of this occurred this year when the government rejected the dietary guidelines committee’s recommendation to cut the maximum alcohol intake for men in half, from two drinks a day to one. “The Trump administration instead passed on science and yielded to industry lobbying,” critics argued. But the opposite is true.

Decades of research indicate a link between moderate alcohol consumption and better health outcomes, particularly cardiovascular, compared to those who abstain and those who binge drink. Because of the amount and consistency of that evidence, government recommendations on drinking since the 1990s have advised adults to consume no more than one drink a day for women and two drinks a day for men. But, that implies some alcohol consumption can be part of a healthy diet. A growing contingent within public health academia sees this implication as a threat to their political agenda and had hoped to use the dietary guidelines to overcome it.

For a number of years, there has been a push to dispel the idea that moderate drinking may confer health benefits because this widespread belief—backed up by a mountain of evidence—makes it harder for those in public health who want to institute greater controls on how alcohol is sold, advertised, consumed, and taxed. With limited success, they push the idea that no amount of alcohol consumption is “safe.” With even less success, they try to convince the public that this is what the science shows.

To make their case, they point to a limited number of studies linking low and moderate alcohol consumption to slight increases in the risk of certain cancers. Though the evidence is slim and likely dependent on drinking patterns, diet, environmental influences, and a host of other factors, this, they claim, is justification to assert that no level of drinking is safe, even if it may offer some protection against something like heart disease, the number one killer in most industrialized nations.

This philosophy seems to be behind the dietary guidelines advisory committee making the recommendation to halve the alcohol intake limit for men. Of the 20 members on the committee, only one seems to have a background in alcohol research, Dr. Timothy Naimi of Boston University. Niami has made no secret about his personal biases when it comes to alcohol. He is proponent of stricter regulatory controls, higher taxes, has described the scientifically established benefits of moderate drinking as a “tool” used by the industry to avoid regulation, and suggested drinking guidelines “no longer assume any protective effects from low dose [alcohol] consumption.” But, whatever his personal biases, one might assume the other members of the committee would ensure their recommended guidelines on alcohol were based on the strongest available evidence. That is not what happened.

To conclude that men should be advised to halve their alcohol intake, the advisory panel limited their evidence review to only studies published since 2010. Of those 60 studies, half found reduced risk of death associated with low alcohol intake compared to abstinence, while the other half found no relationship. In the entire review there was only one study that examined the potential differences in risk for men who drink two drinks a day versus one drink a day.

It is never advisable to base conclusions on a single study, let alone when that one study contradicts a much larger and longstanding body of evidence. But, that one study said exactly what the committee wanted, so that is the one they paid attention to. In other words, it seems the experts on the dietary guidelines advisory committee cherry-picked the evidence to fit an agenda. As six Harvard researchers wrote in a scathing response, the panel employed a “selective, inconsistent process that appears intended to support a pre-determined and, in our view, unscientific conclusion.”

Whatever one can say about the Trump administration’s overall relationship with science, in this case the critics are wrong: the administration did not ignore science because of politics. By ignoring the committee’s biased alcohol recommendations on alcohol, they actually protected the guidelines from becoming just another platform for some group’s political agenda. That should be seen as a win for science and science-based public policy.