Since January, significant shifts in science -- and the funding, policy and regulation that impact it -- as related to food have abounded. This includes the departure of a National Institutes of Health expert conducting intervention studies for clues to the mechanisms underlying effects of ultraprocessed foods on health.
While other global scientists continue to pursue questions about mechanisms and other aspects of these foods, IAFNS (Institute for the Advancement of Food & Nutrition Sciences) is taking a long-view approach to ensure that classification of foods by processing and formulation (or “processing level” as it is commonly termed) is science-based.
To that end, the IAFNS Food Classification Working Group held a cross-sector workshop on April 15, to begin the process of soliciting feedback on a set of draft “Science-Based Principles for Food Classification Focused on Processing and Formulation in Support of Public Health” (supported by USDA AFRI grant 1033399).
The intention of these principles is to delineate the factors needed for robust food classification schemes that incorporate processing of food as well as the formulation of food products. In addition, such factors can be useful in the design of food classification schemes and to ensure reliable, reproducible findings optimally useful for policy decision-making.
Over several months prior, a writing team comprised of academic and government scientists identified key issues with current food classification schemes as articulated in the public domain literature and reports. They then leveraged these “issues” to generate a set of 10 draft principles aligned with several key identified themes:
· Consistency/reliability and objectivity of groupings
· Plausible mechanisms for an impact on health
· Science of processing
· Science of formulation
· Considering the state-of-the-science
· Utility for research applications
After the writing team presented its rationale for the principles to the April workshop audience, scientists with expertise in food classification, sensory science and nutrition, epidemiology and intervention study design were asked to provide feedback. These comments, in addition to those from the larger audience, ranged from support to criticisms.
It would be challenging to capture the multitude of well-constructed comments in one go, but some highlights are noted below:
+ Food processing is a continuum: Food processing starts with foods created from ingredients, subjected to various processes, purchased in a store and then likely subjected to additional processing at home. The last leg of this journey is often forgotten and at the same time is difficult to capture as there exist little data on how consumers treat foods or ingredients at home. However, there is potential for health impact if e.g., households fry at high temperatures or re-use oils.
+ Alignment on terminology is fundamental: Definitions are never all that exciting, but the goal here is for people to understand each other. Participants noted that the term “processing level” is not well defined (and was meaningless to some) and can be conflated with formulation. Instead, “unit processing” was suggested as generally accepted terminology in food science.
Ingredients are those items consumers see on the food label. However, formulation is much more specific and complex and is not typically evident to the consumer. This points to the need for cross-disciplinary exchange to be sure we are discussing the same thing.
+‘Purpose’ of processing is a concern for some stakeholders: “Purpose” here refers to processes and ingredients “designed to create highly profitable (low-cost ingredients, long shelf-life, emphatic branding), convenient (ready-to-(h)eat or to drink), tasteful alternatives to all other Nova food groups and to freshly prepared dishes and meals" (Montiero et al. 2019).
These elements were considered to represent a philosophical view with variables that are difficult to measure and quantify. For the IAFNS Principles, which are focused on the impacts of processing and formulation in individual foods on health, the concept of “purpose” in this context was considered out of scope.
+ Who are “authoritative or competent bodies”? There was interest in looking to “authoritative or competent bodies” for guidance or updated reviews of science to reduce bias and promote alignment. Participants discussed the reality that identification of what an authoritative body is may be in the eye of the beholder. It may depend on whether the approach of a specific body is considered transparent, comprehensive or state-of-the-art.
Other terms were explored, to include “consulting scientific evidence generated using best practices” or “incorporating relevant expertise.”
+ Context for the scheme validation is key: Typically, a classification scheme or diet quality tool is established based on specific datasets. For example, the Tufts Food Compass is based on data from the U.S. food supply and validated using U.S. population-level dietary intake and health data. Similarly, the Healthy Eating Index (HEI) is formulated based on the U.S. Dietary Guidelines for Americans.
Participants discussed that it would not make sense to apply the Food Compass algorithm to a different country and that the HEI can only assess adherence to U.S. dietary guidance. The same context considerations for both development and validation of any food classification scheme would apply.
+ It is currently not possible to align with the principles: Participants noted that the principles may be aspirational. Given the current state of science the principles may not be realistic to achieve. The principles spell out needs for understanding why processing or formulation are relevant to health, with at minimum some biological plausibility. But there are many processes, ingredients, as well as the intersection of the two, all with likely different biological impacts. In addition, data on food ingredients and quantities is not widely available.
It was discussed that the principles are a framework or benchmark, requiring users to articulate why research is unable to align or the current data gaps that require additional work to make these linkages.
What next?
The statements are currently under revision based on the feedback received – with a plan to submit a paper to a peer-reviewed journal this year.
Mobilization and dissemination of this output will require significant effort, and IAFNS will be sharing the principles at upcoming conferences. Current plans include sessions at the International Union of Nutritional Sciences (IUNS) meeting in Paris and the Food and Nutrition Conference and Expo (FNCE) in Nashville.
These principles will remain open to feedback and iteration – in recognition of science as an evolving enterprise. Impact will only come from implementation. We hope to link with potential users across sectors to use, test, evaluate and improve upon what has been created.
As the science and policy landscape continues to evolve, we hope the principles can serve as science-based anchors across research and decision-making to enable food classification by processing and formulation that supports health and wellness.
Marie Latulippe, MS, MBA, RDN, is director of science programs at the Institute for the Advancement of Food & Nutrition Sciences (iafns.org).
