Time to Base Nutrition Policy on Science
A recent article in the Washington Post details five nutrition “facts” we used to be believe. It ends by saying something that you rarely read but is entirely accurate: “In fact, we don’t have a lot of answers about nutrition, which is considered a relatively new science.” But to listen to the U.S. Department of Agriculture (USDA), the Food and Drug Administration (FDA), and popular food activists, you would think the only issue is that Americans just aren’t listening.
The real problems don’t start with consumers, they start with scientific and economic shortcuts. The consequences of bad policies are dire: poor nutrition is linked to chronic diseases such as diabetes, cancer, and heart disease. Unfortunately, one out of every two adults suffer from one or more preventable chronic diseases.
But the federal officials who are charged with making nutrition policies continue to make poorly informed decisions. In 2009, the USDA instituted a program that excluded white potatoes from the Special Supplemental Program for Women, Infants and Children (WIC), presumably to address obesity. They did this despite the fact that many Americans have shortfalls in potassium and potatoes are a great source of this nutrient. Five years later, they finally asked the Institute of Medicine whether this was a good move.
Predictably, one of the Institute’s findings was that “Intakes of … potassium … among low-income women, fall short of current nutrient intake recommendations.” The program may have slightly affected obesity in children, although it is not clear that it has anything to do with potatoes.
The FDA, meanwhile, just finalized its regulations to put calorie labeling in restaurants, theaters, and grocery stores. The rule was initially finalized in 2014 but put on hold by the current administration. Studies are mixed as to whether or not posting calories will do just a little bit or no good whatsoever. But science is not the reason this rule is going forward.
The National Restaurant Association supported the rule, originally as part of the Affordable Care Act, because there was a growing “patchwork” of local and state laws requiring it. This is a perfect example of how not to make scientifically based health policies. In letting the rule go, the Commissioner announced that the rule would institute “predictable, uniform federal standards,” precisely what industry needed. Again, the real problem was with that we did not pay attention to the first adopters, who demonstrated that the information wouldn’t help with obesity.
At least the five nutrition facts cited earlier were originally based on some science. One myth in the article was that “All fat is bad.” But it was only in 2010 when the Dietary Guidelines committee stopped recommending limits on total fats, although they still recommended reducing saturated fat. The original myth was about total fat, but recently multiple studies have found that polyunsaturated fats (and possibly monounsaturated fats) found in foods such as walnuts, salmon, and soybean oil are now considered good for you.
Even more recently, a 2014 study published in the Annals of Internal Medicine was unable to conclude that even saturated fats caused heart disease. Moreover, it remains unclear whether unsaturated fats are good for you. These are still controversial findings, and, without clear scientific backing, policymakers should proceed with caution.
More specifically, public health policy must always be preceded by both sound science and cost-benefit analysis.
Sound scientific evidence must be present for a positive public health benefit to be amply demonstrated. Had there been more research to indicate what manufacturers might do to replace animal fats in the 1980s, activists might not have campaigned so hard against trans fats. As for cost-benefit analysis: “Trying” out public policies, such as nutrition labeling, without credible analysis showing that benefits exceed costs, removes public resources that can be better spent addressing public policies that do pass such a test.
These problems are exacerbated in the case of the new science of nutrition. For diet and disease relationships, dietary guidance and nutrition policy based on memory-based recall data have been shown by professor Edward Archer to be “pseudoscience and inadmissible.” These data, which underpin most of the advice from the Dietary Guidelines, ask consumers to remember what and how much they ate in the last 24 hours. Unfortunately, well over half of consumers do not report eating enough to stay alive. If the data that go into diet-disease relationships are flawed, then the correlations between dietary choices and disease may be wrong. This means that much of the current advice and policies may be wrong.
Taking shortcuts to policy without sound science and cost-benefit analysis leads to policy failures — resulting in poorer health and declining faith in nutrition policies.
Richard Williams is a senior affiliated scholar with the Mercatus Center at George Mason University and a former director for social sciences with the FDA’s Center for Food Safety and Applied Nutrition.