Dietary patterns and mental health outcomes: A Narrative Review of Original Research Evidence

This narrative review examined original human research on dietary patterns and mental health outcomes. PubMed was the primary database, with Google Scholar used as a supplementary source. Searches covered publications from 1 January 2017 to 14 September 2026. Study selection was guided by explicit eligibility criteria. Data were extracted using a structured framework, appraised with design-specific tools, and synthesised narratively. Eligible studies evaluated Mediterranean or Paleolithic dietary patterns, ultra-processed food consumption, or broader dietary patterns in relation to depression, anxiety, stress, psychological distress, or mental health-related quality of life. Thirty-three principal analyses were included: eight randomised controlled trials, 19 prospective, longitudinal, or target-trial-emulation analyses, and six analytical cross-sectional studies. Observational evidence most consistently linked higher ultra-processed food consumption with poorer depressive outcomes. Mediterranean-style and other healthier dietary patterns were associated with more favourable outcomes in several trials and prospective studies, but null and inconsistent findings were also identified. Evidence for Paleolithic dietary patterns was limited to cross-sectional studies. Overall, dietary quality was associated with mental health outcomes, but methodological heterogeneity, risk of bias, and residual confounding limit causal interpretation. Further well-designed randomised and longitudinal studies are required.