Drinking alcohol may lead people to overconsume savoury ultra-processed foods, according to new research from the University of Sydney’s Charles Perkins Centre, with researchers suggesting this may contribute to excess energy intake and weight gain.

Published in Obesity Reviews, the study found that alcohol can trigger biological pathways that increase levels of the hormone FGF21, which is linked to protein appetite and is associated with a preference for savoury flavours. When activated, this system can shift cravings toward salty, umami-flavoured foods.

Traditionally, the body associated these flavours with protein-rich foods such as meat. However, modern food environments can disrupt these sensory signals, with umami flavours now also found in foods that are not high in protein. 

The study found these readily available, artificially flavoured savoury foods can act as “protein decoys”, effectively tricking the protein hunger system into seeking foods that taste like protein but do not deliver it. As a result, people may eat more of these foods to satisfy the signal, leading to higher overall intake of fats, carbohydrates, and total energy.

“Many people will recognise the experience of having a few drinks and suddenly craving something salty, like chips, French fries, pizza or other savoury foods. Now we have a better understanding of the hormonal dynamic at play, which may be driving overconsumption of ultra-processed foods,” said Dr Amanda Grech, lead author of the study from the Charles Perkins Centre.

“Cravings for a packet of chips with a drink (sometimes referred to as the "aperitif effect"), a pizza at the end of a big night out, or a fry-up the next morning may be driven by the way alcohol alters the body's regulation of appetite, particularly for protein,” said senior author, Professor David Raubenheimer from the Charles Perkins Centre.

"Our study suggests that when dietary protein is diluted, people compensate by eating more overall to satisfy the increased protein appetite induced by alcohol. In this way, alcohol may contribute to overeating particularly when ultra-processed, low-protein savoury foods are readily available,” he said. 

When it came to sweets, alcohol consumption had the opposite effect, instead reducing intake of sugary foods, another known effect of the FGF21 hormone.

Researchers say these findings help explain why alcohol’s effects on weight gain appear to differ depending on the surrounding dietary environment.

“An important finding is that alcohol has different effects on total energy intake depending on the dietary environment, particularly whether diets are dominated by minimally processed foods or ultra-processed savoury foods. It is not simply a matter of the calories in alcohol itself,” said co-author Professor Stephen Simpson from the Charles Perkins Centre.

To help manage these cravings, researchers recommend keeping satisfying whole-food snacks on hand.

“If you choose to drink, it's worth being mindful of this hormonal interplay,” said Professor Raubenheimer. “Having protein-rich whole foods readily available can help steer you away from ultra-processed foods. Think roasted chickpeas, smoked salmon, lean cold meats, prawns, or oysters.”

How the research worked

The researchers analysed Australian national dietary survey data to examine food intake patterns in relation to alcohol consumption.

They found that people ate more savoury foods on drinking days than on non-drinking days. Each standard drink was associated with increased savoury food intake and reduced sweet food intake, and the effect on energy intake was especially pronounced on low protein ultra-processed foods.

While previous studies have shown mixed results for the relationship between alcohol and weight gain, the authors say their mechanistic approach helps reconcile these inconsistencies by explaining how diet context modifies alcohol’s effects, mediated by the actions of the protein hunger hormone, FGF21. 

Research 
Grech, et al, Protein decoys, alcohol, and energy intake: testing a mechanistic – ecological model, (2026). DOI: 10.1111/obr.70138

Declaration
The authors declare no conflicts of interest. The research was funded by The National Health and Medical Research Council, Australia, Nutrition and Complexity Program Grant, grant/award number GNT1149976.