Some popular drinks were linked to depression and anxiety. Coffee was different

By Dr. Sanchari Sinha Dutta, Ph.D.Reviewed by Lauren HardakerSep 6 2026

From coffee and energy drinks to bubble tea and soft drinks, an analysis of more than 2 million people reveals a surprisingly complex relationship between what we drink and mental health.

Study: Nonalcoholic Beverage Consumption and the Risk of Depression, Anxiety, and Stress: A Systematic Review and Meta-Analysis. Image Credit: SAWITRE INTAYAM / Shutterstock

A systematic review and meta-analysis published in the journal Depression and Anxiety found that higher consumption of some nonalcoholic beverages was associated with greater odds of depression and anxiety, while evidence for an association with stress was less consistent.

Background

Depression, anxiety, and stress are major public health burdens worldwide. About 322 million and 264 million people live with depressive disorders and anxiety disorders, respectively, worldwide. The prevalence of stress-induced symptoms is also rapidly increasing across all age groups.

Emerging evidence suggests that higher intake of some nonalcoholic beverages, such as tea, coffee, sugar-sweetened drinks, energy drinks, and carbonated drinks, may be associated with these mental health outcomes. However, substantial inconsistencies have been observed in these results. Some studies have reported either protective or detrimental effects, while others have reported no significant association.

This systematic review and meta-analysis aimed to clarify how different types of nonalcoholic beverages are associated with depression, anxiety, and stress across populations and to inform targeted public health and clinical strategies.

The researchers systematically screened various electronic databases and identified 59 relevant studies involving 2,050,716 participants. The assessment of publication quality found that most included studies were rated as moderate quality. Overall, these studies were conducted in 22 countries across multiple continents.

Key findings

The meta-analysis of included studies revealed that overall nonalcoholic beverage consumption was positively associated with both clinically diagnosed depression and depressive symptoms. In beverage-specific analyses, sugar-sweetened beverages, bubble tea, carbonated beverages, soft drinks, artificially sweetened beverages, and energy drinks were positively associated with depression. In contrast, higher coffee intake was inversely associated.

The pooled estimates were higher in studies conducted in lower-middle-income countries and in those involving only pediatric populations.

A significant positive association was also found between higher intake of nonalcoholic beverages and anxiety, especially among consumers of bubble tea and carbonated beverages. The association was statistically significant in studies conducted in high-income countries and was strongest among pediatric populations.

Regarding the risk of stress, the main pooled analysis showed a non-significant association with nonalcoholic beverage intake. However, the association became significant after excluding one study from the meta-analysis, which had contributed substantial heterogeneity.

Positive associations were found in analyses restricted to high-income countries, pediatric-only populations, and soft drink consumers, although differences between income and age subgroups were not statistically significant.

Significance

This systematic review and meta-analysis concludes that nonalcoholic beverage consumption was associated with depression and anxiety, while evidence for stress was less stable. The major strength of this meta-analysis is the inclusion of studies from diverse global regions, which the authors said improves the representativeness of the evidence across different populations.

Associations differed substantially by beverage type. Several sweetened or carbonated beverage categories were positively associated with depression, while bubble tea and carbonated beverages were positively associated with anxiety. Coffee was inversely associated with depression but was not significantly associated with anxiety. The anti-inflammatory and antioxidant properties of various bioactive compounds present in coffee may partly explain its inverse association with depression.

Among various study populations, the strongest association estimates between nonalcoholic beverage intake and mental health issues were observed among adolescents. Maturation of the prefrontal cortex continues throughout this developmental phase, which may heighten adolescents' sensitivity to neurobehavioral effects of beverage components.

Studies conducted in lower-middle-income countries also reported higher estimates for depression, highlighting the complex influence of socioeconomic factors, such as nutritional transition and accessibility to mental health resources.

For anxiety, effect estimates were higher in East Asia and the Pacific region, suggesting the potential influence of region-specific factors. For stress, the association estimate was higher among females than males, although the sex-specific findings were not statistically significant, making any sex difference uncertain.

The authors proposed that glycemic fluctuations and insulin resistance associated with sugar-sweetened beverage consumption may impair neuroplasticity by affecting brain-derived neurotrophic factor (BDNF) levels and activating inflammatory pathways. These mechanisms were not directly tested in the meta-analysis but may partly explain the observed associations.

Furthermore, the authors proposed that high-sugar diets may disrupt the hypothalamic–pituitary–adrenal (HPA) axis, altering normal cortisol rhythms and potentially contributing to dysregulation of the stress response system.

However, the studies included in the meta-analysis exhibited methodological variations and high statistical heterogeneity across the three mental health outcomes, which may contribute to inconsistencies in the strength of the associations. Compared to clinically diagnosed depression, depressive symptoms showed stronger associations, which may be due to the variations in depression definitions across studies.

Overall, the findings of this meta-analysis may help inform future mental health-related dietary research and public health strategies. However, more prospective and interventional studies are needed before developing specific dietary recommendations for nonalcoholic beverage consumption.

The majority of studies included in this meta-analysis are cross-sectional, which restricts the certainty of the causal inference. The depression association was not statistically significant in the cohort subgroup; only one cohort report contributed to the anxiety analysis, and the stress analysis consisted entirely of cross-sectional studies. Residual confounding by socioeconomic status, smoking, overall diet, physical activity, sleep, and other lifestyle factors also cannot be ruled out. Furthermore, these studies relied on self-reported information on beverage intake, which may introduce recall bias. Future studies should therefore consider quantitative dietary assessment methods and more detailed exposure analysis to explain the underlying mechanisms.