Ultraprocessed Food Consumption and Pre-Farmgate Greenhouse Gas Emissions Among United States Adults from 2007–2010

Abstract

Background

Changes to the food system can have consequences for human and planetary health, and one recent change is increased consumption of ultraprocessed foods (UPFs). Although correlations between UPF intake and human health have been evaluated in the United States, little is known about the association of UPF intake with pre-farmgate greenhouse gas emissions (GHGEs), which account for the majority of GHGE across the food life cycle.

Objectives

We used a nationally representative survey to evaluate the association between UPF consumption and pre-farmgate GHGE among United States adults.

Methods

Data were from the National Health and Nutrition Examination Survey, 2007–2010. Participants were divided into quintiles based on proportion of grams from UPF using mean of 2-d dietary recall. The database of Food Recall Impacts on the Environment for Nutrition and Dietary Studies was used to estimate the pre-farmgate GHGE (in kilograms of carbon dioxide equivalents [kg CO₂-eq]) of foods. Multivariate linear regression models were used to test the association between quintiles of UPF consumption and pre-farmgate GHGE. Models were progressively adjusted for sociodemographic characteristics, total caloric intake, and red and processed meat intake.

Results

Consuming a greater proportion of grams from UPF was associated with higher pre-farmgate GHGE in unadjusted analyses. However, when the analyses were adjusted for total energy, we saw the opposite trend, such that consuming a higher proportion of grams from UPF was associated with lower pre-farmgate emissions: emissions for the highest quintile of UPF consumption were 4.47 (95% confidence interval [95% CI]: 4.27, 4.67) kg CO₂-eq, compared with 4.85 (95% CI: 4.64, 5.05) kg CO₂-eq for the highest quintile (P_trend = 0.003).

Conclusions

Although we find that diets with a greater proportion of grams from UPF have lower pre-farmgate GHGE, our analyses show an opposite trend when they are not adjusted for total energy intake, demonstrating the need for caution when analyzing the relationship of UPF intake with GHGE.