Nutritionally adequate and environmentally respectful diets are possible for different diet groups: an optimized study from the NutriNet-Santé cohort

Research has shown that vegetarian diets have a low environmental impact, but few studies have examined the environmental impacts and nutritional adequacy of these diets together, even though vegetarian diets can lead to nutritional issues. Our objective was to optimize and compare 6 types of diets with varying degrees of plant foods (lacto-, ovolacto-, and pescovegetarian diets and diets with low, medium, and high meat content) under nutritional constraints. Consumption data in 30,000 participants were derived from the French NutriNet-Santé cohort using an FFQ. Diets were optimized by a nonlinear algorithm minimizing the diet deviation while meeting multiple constraints at both the individual and population levels: nonincrease of the cost and environmental impacts (as partial ReCiPe accounting for greenhouse gas emissions, cumulative energy demand, and land occupation, distinguishing production methods: organic and conventional), under epidemiologic, nutritional (based on nutrient reference values), and acceptability (according to the diet type) constraints. Optimized diets were successfully identified for each diet type, except that it was impossible to meet the EPA (20:5n–3) + DHA (22:6n–3) requirements in lacto- and ovolactovegetarians. In all cases, meat consumption was redistributed or reduced and the consumption of legumes (including soy-based products), whole grains, and vegetables were increased, whereas some food groups, such as potatoes, fruit juices, and alcoholic beverages, were entirely removed from the diets. The lower environmental impacts (as well as individual indicators) observed for vegetarians could be attained even when nutritional references were reached except for long-chain n–3 (omega-3) fatty acids. A low-meat diet could be considered as a target for the general population in the context of sustainable transitions, although all diets tested can be overall nutritionally adequate (except for n–3 fatty acids) when planned appropriately. This trial was registered at clinicaltrials.gov as NCT03335644.