Improving diets is one of the world’s biggest public health and environmental challenges. More than 2.5 billion adults are currently overweight or obese, 88% of countries with available data face a severe burden of two or more types of malnutrition, and food systems account for a quarter to a third of global greenhouse gas emissions.
Published in natural foodthis study introduces MATILDA (Micro-Macro Assessment Tool for Identifying Low-Impact Dietary Behaviors), a global database that combines individual dietary surveys with country-level food supply chain and population data. Covering 165 countries, it provides a globally harmonized picture of how the nutritional quality and environmental impacts of diets vary within populations, giving policymakers a stronger evidence base than national averages alone.
Looking beyond the national average reveals just how vastly different eating habits vary from country to country. The highest-impact diets produced nearly twice as much greenhouse gas emissions as the lowest-impact diets.
In the 165 countries studied, every 5% improvement in diet quality measured using the Alternative Healthy Eating Index (AHEI) was associated with a 7.7% reduction in diet-related greenhouse gas emissions, along with reduced water and land use. Diets containing higher proportions of plant-based foods were also consistently associated with lower environmental impact.
People’s dietary patterns are shaped by far more factors than individual choices. Where they live, whether in town or in the countryside, how long they have spent in education, how much money they can afford, and broader social and environmental circumstances all come into play. Recognizing these differences creates an opportunity to develop food policies that better reflect the realities of people’s food choices.
Lead author Dr. Oliver Taherzadeh from Leiden University’s Institute of Environmental Sciences (CML) said:
“MATILDA connects what’s right in front of us with impacts on the global environment. We do this by linking group-level dietary patterns to macroeconomic models that capture impacts within and across national borders.To date, macroeconomic food system models have overlooked national-level differences in food consumption patterns, Integrating data into country case studies has been fragmented. We seek to fill this gap so that dietary food modeling and otherwise hidden vulnerable consumer groups can be accurately identified, making sustainable diet policies more targeted, equitable and effective. ”
MATILDA developer Dr. Hongyi Cai from Leiden University’s Institute of Environmental Sciences (CML) said:
“There is no one-size-fits-all approach when it comes to sustainable and healthy eating. When we benchmark populations against their actual diets, the impacts on food systems become clearer, health inequalities become clearer, and diverse needs become more apparent.”
People do not make food choices in isolation. What we eat depends on where we live, what we can buy, and what is available. We are entering an era in which food policy becomes more evidence-based. Rather than treating entire populations as if they were eating the same thing, we can now identify where interventions are most likely to make a difference, while ensuring that vulnerable populations are not left behind. ”
Professor Paul Behrens, co-author, Oxford Martin School, University of Oxford
This paper discusses a range of possible policy responses, including public information campaigns, supermarket incentives, and other measures to facilitate healthier, lower-impact choices. Broader approaches such as taxation and regulation also play an important role. The authors conclude that one-size-fits-all interventions are not sufficient because dietary patterns and nutritional needs vary within the population.
sauce:
Reference magazines:
Taherzadeh, O. others. (2026). The environmental impact and quality of diets vary by sociodemographic characteristics. natural foods. DOI: 10.1038/s43016-026-01389-y. https://www.nature.com/articles/s43016-026-01389-y
