While associations of ultra-processed food (UPF) consumption with adverse health outcomes are accruing, its environmental and food biodiversity impacts remain underexplored. This study examines associations between UPF consumption and dietary greenhouse gas emissions (GHGe), land use, and food biodiversity.

Prospective cohort study. Linear mixed models estimated associations between UPF intake (grams/day and kcal/day) and GHGe (kg CO₂-equivalents/day), land use (m2/day), and dietary species richness (DSR). Substitution analyses assessed the impact of replacing UPFs with unprocessed or minimally processed foods.

368,733 participants in the European Prospective Investigation into Cancer and Nutrition (EPIC) study.

Europe

Stronger associations were found for UPF consumption in relation with GHGe and land use compared to unprocessed or minimally processed food consumption. Substituting UPFs with unprocessed or minimally processed foods was associated with lower GHGe (8.9%; 95%CI: -9.0; -8.9) and land use (9.3%; -9.5; -9.2) when considering consumption by gram per day and higher GHGe (2.6%; 95% CI: 2.5: 2.6) and land use (1.2%; 1.0; 1.3) when considering consumption in kilocalories per day. Substituting UPF by unprocessed or minimally processed foods led to negligible differences in DSR, both for consumption in grams (-0.1%; -0.2; -0.1) and kilocalories (1.0%; 1.0; 1.1).

UPF consumption was strongly associated with GHGe and land use as compared to unprocessed or minimally processed food consumption, while associations with food biodiversity were marginal. Substituting UPFs with unprocessed or minimally processed foods resulted in differing directions of associations with environmental impacts, depending on whether substitutions were weight- or calorie-based.

We determined whether dietary species richness (DSR) (i) can be robustly measured using 4-day food intake data, (ii) is dependent on socio-demographic characteristics and (iii) is associated with diet quality.

The National Diet and Nutrition Survey (NDNS) nutrient databank 2018–2019 was expanded to include FoodEx2 food classifications, ingredients, the number and identity of unique species, Nutrient Rich Food 8·3 (NRF 8·3) Index scores and greenhouse gas emissions. Four-day food intake data and socio-demographic variables were used to calculate diet quality and DSR on the food and diet level.

The United Kingdom (UK).

Participants from NDNS 9–11 (2016–2019).

Composite dishes had the highest DSR (median 8 (Q1 = 4, Q3 = 12)), followed by seasoning, sauces and condiments (median 7, (Q1 = 4, Q3 = 10)) and, grains and grain-based products (median 5, (Q1 = 2, Q3 = 7)). Median DSR over 4 days was 49 (Q1 = 43, Q3 = 56; range 14–92), with the first 2 days achieving 80 % of DSR measured over 4 days. DSR was significantly higher in those who were younger, those with a higher household income or those with a lower level of deprivation (all P < 0·001). Higher DSR was associated with a small but significant improvement in nutritional quality (P < 0·001). Also, adherence to dietary guidelines such as fibre, fruits and vegetables and fish was associated with significantly higher DSR (all P < 0·001).

We successfully established DSR based on 4-day food intake data. We also identified opportunities to improve DSR by increasing the consumption of fruits, vegetables, fibre and fish.