Flavorings, colors and sweeteners emerge as UPF markers linked to depression risk

By Dr. Sanchari Sinha Dutta, Ph.D.Reviewed by Lauren HardakerSep 1 2026

By tracking dozens of ingredients associated with ultra-processed foods, researchers investigated whether some components of the modern diet may matter more for mental health than others.

Study: Relationship between 39 dietary markers of ultra-processed food and incident depression: evidence from the prospective UK Biobank cohort study. Image Credit: White bear studio / Shutterstock

Higher estimated dietary exposure to certain markers of ultra-processed foods, including flavoring agents, coloring agents, and sweeteners, was dose-dependently associated with a higher risk of incident depression, according to a new study published in the European Journal of Nutrition.

Background

Ultra-processed foods (UPFs) are convenient, ready-to-eat products that contain high amounts of fat, salt, and sugar and low amounts of dietary fiber, protein, and essential micronutrients. Although high UPF intake has been linked to adverse health outcomes, UPF consumption has increased significantly worldwide, with the US and UK recording the highest rates.

Several studies have linked UPF intake to increased risk of obesity, cardiovascular disease, and type 2 diabetes. Recent evidence has also highlighted a significant association between UPF intake and depression risk.

The NOVA classification, the most widely adopted food classification system, designates UPFs as group 4. However, a major limitation in current UPF research is the lack of standardized chemical or structural metrics for classifying food items, which forces researchers to rely on published food lists and can lead to misclassification due to the vague definition of UPFs.

The use of pre-defined markers of UPFs is a promising approach to address the current limitation in UPF research. These markers include cosmetic additives, such as flavoring agents, coloring agents, and sweeteners, and non-culinary ingredients, such as fructose, modified oil, and isolated protein. Food items containing at least one of these markers are classified as UPFs.

Given the potential utility of these markers for the systematic identification of UPFs, researchers at Justus-Liebig University of Giessen, Germany, analyzed the associations between ten marker categories and 39 specific markers and the risk of depression.

Key findings

The researchers analyzed data from 171,818 UK Biobank participants, who had completed at least one 24-hour dietary recall questionnaire. Estimated exposure to ten marker categories and 39 specific markers was assessed as the percentage of total food intake represented by foods likely to contain each marker.

A total of 5,424 cases of incident depression were identified during an average follow-up period of 10.7 years.

Among 10 marker categories, flavoring agent, coloring agent, and sweetener showed a positive, dose-dependent association with incident depression. Specifically, compared with the modeled intake level associated with the lowest depression risk, at 5% of total food intake, having 40% of total food intake represented by foods likely to contain flavoring was associated with a 46% higher risk of incident depression; for coloring agents, 20% versus 3%, the corresponding lowest-risk intake level, was associated with a 49% higher risk; and for sweeteners, 20% versus 0%, the corresponding lowest-risk intake level, was associated with a 45% higher risk of incident depression.

Among the remaining categories, protein source and fiber showed a small but significant non-linear association with incident depression, with the highest estimated risks among non-consumers and no evidence of harm at higher intake levels. No significant associations were detected for flavor enhancer, processing aid, modified starch, varieties of sugar, or modified oil. The principal associations for flavoring, coloring agents, and sweeteners remained significant across all sensitivity analyses.

Study significance

The study identifies three marker categories of UPFs, including flavoring agents, coloring agents, and sweeteners, which are significantly associated with an increased risk of developing depression. Notably, these three marker categories are also associated with increased risk of all-cause mortality, as previously reported by the current study researchers.

A recent UK market analysis reveals that flavoring agents and coloring agents are present in 58.4% and 26.5% of all UPFs, respectively. Previous studies have implicated flavoring agents in weight gain and obesity and coloring agents in adverse health outcomes, including attention deficit hyperactivity disorder and autism spectrum disorder in children and adolescents.

Regarding health risks associated with artificial sweetener intake, previous studies involving diverse populations have also reported an increased risk of depression and irritable mood. Since younger adults are more likely to consume artificially sweetened beverages, future studies should assess the potential long-term effects of this greater sweetener exposure on mental health and depression.

In addition to UPF marker categories, the study identifies fructose as a specific UPF marker that showed a dose-dependent association with incident depression. In animals, a high-fat, high-fructose diet has been found to induce metabolic abnormalities and depression- and anxiety-like behaviors, providing some biological plausibility for the observed association.

Although it is still unclear how the intake of specific UPF markers is associated with depression risk, some potential mechanisms include direct effects of these markers on brain function and neurotransmitter signaling, which may potentially contribute to the pathophysiology of depression. These markers may also be indirectly linked to the observed risk through obesity, metabolic dysfunction, and alterations in the gut microbiota. Future animal and human intervention studies are required to comprehensively understand the involvement of these pathways.

Major strengths of this study include its analysis of predefined UPF marker categories and specific markers within a large, well-characterized prospective cohort, with a 10.7-year mean follow-up and modeling that captures non-linear associations. The study also identified a U-shaped association between overall UPF intake and depression risk, with the lowest modeled risk at about 16% of total food intake. This modeled pattern should not be interpreted as evidence that consuming some UPFs is protective or that avoiding UPFs causes depression. The authors argue that the pattern supports the notion that only part of the marker categories and specific markers are positively and dose-dependently associated with depression risk. Overall, these findings support refining current UPF definitions by focusing on specific UPF markers that show the strongest association with adverse health outcomes.

The study relied on self-reported dietary intake data, which were not independently validated and may contain measurement error, while residual confounding also remains possible. Importantly, the dietary questionnaire did not directly capture ingredient lists, so marker exposure was estimated based on the probability that foods contained each marker. Furthermore, the possibility of reverse causation cannot be fully excluded, as people with depression may rely more on ready-to-eat meals rather than preparing meals due to listlessness. Healthy-volunteer bias may also limit the generalizability of the findings. Future studies should address these factors for more conclusive causal interpretations.