If you follow the mass-media coverage of nutrition science, it seems that dietary recommendations can change almost on a whim. One day a food is healthful, and the next day new research says it's not. In reality, dietary recommendations should be based on the entire body of existing evidence rather than on one study with alarming headlines.

Nutrition experts and policymakers are tackling this challenge, navigating the often-convoluted and contradictory body of science in order to define and communicate the principal recommendations for a healthy diet — namely, the Dietary Guidelines for Americans (DGA) — to the public. However, the process has unearthed several concerns that could ultimately negatively impact the recommendations put forth.

What are the DGA, and why should you care? The DGA, administered by the U.S. Department of Agriculture and the Department of Health and Human Services, are intended to provide information and advice to the public for building a healthy diet, based on a scientific review of the most recent evidence. Additionally, they are the cornerstone for all federal nutrition and education programs in the U.S. — from school meals to the Agriculture Department's Supplemental Nutrition Assistance Program. Because the reach and impact of the DGA are so expansive, it is critical that the ultimate recommendations are based solely on sound science and that they lead to better health outcomes for the American public.

Among the problematic issues with the current DGA deliberations is an apparent shift from strictly science-based to more politically motivated recommendations. While the freedom exists for the Dietary Guidelines Advisory Committee — a group of 15 health and nutrition experts charged with reviewing the evidence on diet and health — to explore food and nutrition topics that they deem important and scientifically relevant, their mandate explicitly states that the advisory committee's responsibilities "do not include translating the recommendations into policy or into communication and outreach documents or programs." By including recommendations to mandate the nutrition labeling of added sugars, implement purchasing standards for the Supplemental Nutrition Assistance Program and the Special Supplemental Nutrition Program for Women, Infants and Children, and by recommending economic and taxing policies to encourage the production and consumption of healthy foods, the 2015 committee veered from this requirement.

For the DGA process to work as intended and meaningfully improve public health, final recommendations must be reflective of a rigorous scientific process and the best available nutrition science. As I contend in an article I recently published in Nutrition Journal, generating conclusions using Nutrition Evidence Library reviews is the ideal way to accomplish this. This process uses a strict hierarchy of evidence and a rigorous grading scheme, and it is widely acknowledged as a highly credible, transparent and objective process that minimizes bias. In past years, including 2010, when I served on it, the committee relied on the Nutrition Evidence Library to answer the great majority of research questions. However, the 2015 committee used the library to address only 27 percent of its research questions and instead relied on less-rigorous research methods such as meta-analyses and consensus reports to inform the majority of its conclusions.

Furthermore, this practice allows for a greater dependence on observational research. While this type of research is important to inform future studies, it also has critical limitations, including the introduction of bias. Because nutrition science is constantly evolving, policymakers should make every effort to ensure that dietary guidance is based on evidence that will remain relevant and accurate over the long term.

As a dietitian who is also a farmer and an educator, I consider myself an expert on food, nutrition and eating. Serving on the 2010 committee allowed me to understand firsthand the challenge of examining the broad body of evidence on food and nutrition topics to help form dietary recommendations. However, I also understand that, just as we cannot base our dietary recommendations on the latest study to make headlines, it would likewise be inappropriate for weak science, incomplete evidence reviews or political motivations to shape our national food and nutrition policy. We must call for the final 2015 DGA policy to adhere to the original intent to provide science-based advice on health-promoting diets. Otherwise, we may unintentionally steer consumers toward unhealthful diets and violate the most basic tenet of nutrition: "Do no harm."

Joanne Slavin is a registered dietitian and professor in the Department of Food Science and Nutrition at the University of Minnesota.