Protein Decoys, Alcohol, and Energy Intake: Testing a Mechanistic-Ecological Model

Abstract

Ultra-processed diets are associated with excessive energy intake, but studies examining the role of alcohol have produced conflicting results, despite the high energy density of ethanol. We constructed a mechanistic-ecological model to explain this inconsistency and tested its predictions using population dietary data. The model builds on experimentally established mechanisms to predict that, whereas alcohol will contribute ethanol calories irrespective of diet, its impact on total macronutrient energy intake varies depending on dietary pattern. Alcohol consumption stimulates FGF21, a hormone that increases savory (umami) preference and reduces sweet preference. In minimally processed food environments—where umami flavor is a relatively reliable indicator of protein in foods—umami-seeking should direct consumers toward satiating, high-protein foods, with limited effect on macronutrient energy intake. In contrast, on diets rich in ultra-processed savory foods and/or high-fat unprocessed meats, the correlation between umami flavor characteristics and protein is broken by umami-flavored low-protein foods (protein decoys), so alcohol instead drives dietary protein dilution and, through attenuated protein satiation, increased food and macronutrient energy intake (protein leverage). Mixture analysis paired with nutritional geometry confirmed these predictions in population data, suggesting the model may generalize to ecological contexts. Our results potentially explain why alcohol has variable effects on energy balance across dietary patterns and suggest that it exacerbates the obesogenic effects of ultra-processed diets—a potentially important interaction given their global rise.