A recent study led by researchers from Mass General Brigham has provided an essential resource for consumers and policymakers to better understand the extent of food processing in offerings from major grocery stores across the United States. The study, which analyzed products from three major grocery retailers—Target, Whole Foods, and Walmart—introduces a system that scores the degree of food processing, helping to highlight trends in food quality and offering a novel approach to tackling the growing challenge of nutrition and public health.
Published in Nature Food, the research is part of an ongoing effort to make food information more accessible and manageable for consumers. The initiative came in response to the overwhelming amount of conflicting messages about what people should eat in the face of increasingly complex food options. The study’s goal is to offer practical insights that allow individuals to make more informed dietary choices without needing to wade through an overload of complicated information.
Dr. Giulia Menichetti, Ph.D., a principal investigator from the Channing Division of Network Medicine at Brigham and Women’s Hospital and corresponding author of the study, shared her insight into the purpose of the research. “There are a lot of mixed messages about what a person should eat. Our work aims to create a sort of translator to help people look at food information in a more digestible way,” she said. The study’s key aim was to reduce the complexity surrounding food choices by introducing a processing score for different foods. This system helps take the guesswork out of choosing healthy foods, empowering people to make better dietary choices.
The Growing Popularity of Ultra-Processed Foods
Over the past few decades, grocery shelves across the world have been flooded with foods that are heavily processed, including foods that fall into the “ultra-processed” category. These ultra-processed foods often contain additives, preservatives, artificial flavors, and a high concentration of unhealthy components like sugar, salt, and fat. Although the convenience and low cost of these foods have contributed to their popularity, research indicates that a diet rich in ultra-processed foods may be a major driver of chronic health issues like obesity, diabetes, and cardiovascular disease. Additionally, these foods are often stripped of essential nutrients, which further contributes to the declining quality of modern diets.
Despite increasing awareness of the risks of consuming too many ultra-processed foods, there has been little ability for consumers to easily differentiate between processed and unprocessed foods on grocery store shelves. In fact, very little data exists about the nature of food processing and its presence across real-life food markets. The study aimed to fill this gap, providing a clearer view of food processing levels for shoppers as they navigate grocery stores.
Dr. Menichetti pointed out that when people hear about the dangers of ultra-processed foods, many ask, “What are the rules? How can we apply this knowledge?” She emphasized that research efforts like theirs are aimed at offering accessible tools for people to adopt healthier eating behaviors. “We are building tools to help people implement changes to their diet based on information currently available about food processing. Given the challenging task of transforming eating behaviors, we want to nudge them to eat something that is within what they currently want but a less-processed option,” she explained.
The Development of GroceryDB and Processing Scores
To conduct their analysis, the researchers used an algorithm-based tool that can evaluate and score foods based on the degree to which they have been processed. This tool analyzes data such as ingredient lists, nutrition facts, and prices from over 50,000 food items available at three large grocery retailers: Target, Whole Foods, and Walmart. From this, they developed GroceryDB, a comprehensive database cataloging food products’ degrees of processing. The data has been made accessible to the public through the TrueFood website, which offers consumers the opportunity to browse foods, view their processing scores, and access detailed nutritional information.
The scoring system utilized by the researchers is based on their proprietary FPro algorithm, introduced in a 2023 paper in Nature Communications. This algorithm leverages machine learning techniques to assess the processing level of food products and produces a processing score. Foods are scored on a scale, with higher scores indicating more highly processed (or ultra-processed) products.
Whole Foods was found to offer the most minimally processed options among the three grocery chains, but the researchers were still surprised by the prevalence of highly and ultra-processed foods across all three stores. In fact, the vast majority of the food found in these stores, even Whole Foods, was ultra-processed, with only limited minimally processed or whole food offerings in some categories.
Insights into Specific Food Categories
The study also uncovered clear patterns in the types of processed foods available at these grocery retailers. One prominent trend revealed that certain categories of food offered minimal variation in terms of the degree of processing between stores. For instance, the cereals sold at Whole Foods were available in a broad range of processing levels, from minimally processed options to highly processed choices. However, both Target and Walmart only offered cereals that scored highly for processing, primarily ultra-processed varieties. This pattern was consistent in multiple categories, including soups, stews, milk and milk substitutes, yogurt, and cookies.
Even though consumers may expect a wide variety of products, it became clear that the diversity of food choices within each category often translated to identical options in terms of processing levels. This suggests that, despite the quantity of available products, grocery stores can limit nutritional diversity by favoring ultra-processed items across most categories, leaving shoppers with fewer options for healthier, minimally processed alternatives.
Limitations and Future Directions
While the information provided by the GroceryDB and the TrueFood website is useful, the researchers cautioned that the database only reflects a snapshot of food available from the three retailers and is therefore limited in scope. The data used in this study represents only one point in time, and the foods offered in stores can fluctuate significantly based on factors like location, seasonality, and sales trends.
To make this tool even more powerful, the researchers plan to expand the database to reflect not only more stores but also a wider range of geographic locations. “We hope to eventually incorporate geolocation data, allowing us to monitor food options in different regions of the country,” said Dr. Menichetti. They also plan to explore how the variability in available foods might relate to social determinants of health, such as socioeconomic status and access to nutritious foods.
The Promise of Large-Scale Data Tools
One of the overarching goals of this research is to transform how we approach food processing information. By applying artificial intelligence (AI) and machine learning to nutrition science, this study demonstrated the potential for automating and scaling up food assessments. Dr. Menichetti emphasized that most current nutrition studies rely heavily on manual curation, which limits their scope and potential impact. “Our study shows that artificial intelligence and data science can be used to scale up,” she noted.
Ultimately, the team hopes their research lays the foundation for a broader initiative to improve public health through enhanced food information systems. Dr. Menichetti highlighted the potential to create data-driven tools that promote global nutrition security, ensuring people everywhere can access healthier food options in their daily routines. By combining comprehensive, open-access data with advanced technologies, the goal is to foster healthier diets for people everywhere while advocating for improved global access to quality food options.
Conclusion
This study is a significant step toward demystifying the processed food landscape, providing a valuable resource for consumers and policymakers who are navigating the complex world of modern food offerings. As ultra-processed foods continue to dominate grocery store shelves, tools like GroceryDB and the TrueFood website empower individuals to make more informed choices. The researchers hope their work will eventually lead to broad changes that support better public health by shifting consumer behavior and policy decisions toward healthier, minimally processed food options. In doing so, they aim to lay the groundwork for a future where everyone has access to nutritious, quality foods.
Reference: Prevalence of processed foods in major US grocery stores, Nature Food (2025). DOI: 10.1038/s43016-024-01095-7. www.nature.com/articles/s43016-024-01095-7