Diet and depression: a free practice guide that also names what does not work
- A review with explicit practice recommendations on diet and depression, written for mental health clinicians by the Food & Mood Centre at Deakin University – the group that ran the SMILES trial – with co-authors from UNSW and the University of Sydney.
- Free and fully open access under a Creative Commons Attribution 4.0 licence on PubMed Central. Published online 4 December 2024, in the February 2025 issue of the Australian and New Zealand Journal of Psychiatry, the official journal of the Royal Australian and New Zealand College of Psychiatrists, 59(2):115-127.
- Covers what the evidence supports (Mediterranean-pattern diets: 33% lower risk of developing depression at highest versus lowest adherence; SMILES remission 32% versus 8%), what it does not (DASH: one RCT, no effect, insufficient evidence), and the microbiome-gut-brain mechanisms proposed to underlie both.
- Read or download the full text at PubMed Central, PMCID PMC11783990, DOI 10.1177/00048674241289010.
Ask a clinician about the gut-brain axis and you tend to get one of two answers: an eye-roll, or a probiotic recommendation. Neither is defensible, and the gap between them is precisely where the commercial "gut health" industry has set up shop. This paper is the shortest route I know from that impasse to something you can use in Monday's session – and it costs nothing.
The authors are not selling anything
The senior authors are Felice Jacka and Tetyana Rocks of the Food & Mood Centre at Deakin University in Geelong; the lead author is Heidi Staudacher, joined by Scott Teasdale (University of New South Wales), Caitlin Cowan (University of Sydney) and dietitian Rachelle Opie. This is the group behind SMILES, the trial that demonstrated dietary change could reduce depressive symptoms. The paper appeared in the Australian and New Zealand Journal of Psychiatry – the official journal of the Royal Australian and New Zealand College of Psychiatrists, published by SAGE – online on 4 December 2024, and in the February 2025 issue. The full text sits on PubMed Central under a CC BY 4.0 licence: free to read, download, quote and circulate to your team.
The mechanistic section rewards a second reading. It frames the microbiome-gut-brain axis as bidirectional communication running through the vagus nerve, the intestinal barrier and the immune system. It notes that the microbiome is altered in clinical depression, with consistent upregulation of taxa carrying inflammatory properties – the opportunistic pathogen Eggerthella among them – and downregulation of taxa with anti-inflammatory or short-chain-fatty-acid-producing capabilities. Those short-chain fatty acids arise from microbial fermentation of dietary fibre and modulate intestinal barrier and immune function. The gut microbiota also produce serotonin, gamma-aminobutyric acid, catecholamines and acetylcholine.
Note the authors' own verdict on how firmly all of this ties to mood: the microbiome is "probably involved". That single hedge, from the field's leading group, is worth more than a semester of confident webinars.
The value is in the negative space
The Mediterranean pattern carries the strongest signal. A meta-analysis of four longitudinal studies with roughly 10-year follow-up found a 33% risk reduction for developing depression at highest versus lowest adherence. SMILES (12 weeks, n=56 adults) achieved remission in 32% of the dietitian-delivered Mediterranean diet arm against 8% in the social support group. HELFIMED (6 months, n=152, incorporating nutrition education and fish oil supplementation) and AMMEND (12 weeks, n=72, young males) point in the same direction.
Then the paper does what a supplement-vendor course never will. DASH: a single RCT, no effect on depressive scores against a healthy-diet control, and the authors conclude there is insufficient evidence for its use in prevention or treatment. Prebiotics: a 3-month RCT in 106 people with obesity found 16 g/day of inulin alongside a high-fibre diet had limited effects on mood, with only an exploratory hint that responders had elevated faecal Coprococcus at baseline. Mediterranean interventions, the authors add, have had limited success in actually shifting the microbiome. And across 16 trials totalling nearly 46,000 individuals, healthy-eating interventions produced only a small effect on depressive symptoms – largest where the advice came from registered nutrition professionals.
Using it in the room
The practical section is deliberately modest, which is why it is usable. Dietary assessment starts with simple open questions about how many meals or snacks are eaten per day. Counselling is to be delivered sensitively, on the explicit understanding that many patients carry shame and guilt into any conversation touching on weight; the recommended stance is weight-neutral, aimed at health behaviour change rather than weight loss. Information goes out in accessible, bite-size format, calibrated to readiness to change. Goals are small, attainable and value-based, with an emphasis on forward planning.
The referral red flags are spelled out rather than implied: current or historical eating disorders, disordered eating, poorly managed comorbid medical conditions such as gastrointestinal disorders or type 2 diabetes, loss of appetite, significant weight change, and patients who are young, pregnant or elderly. The authors' framing is one I would endorse without reservation – mental health clinicians are well placed to provide dietary counselling and to use clinical judgement, but are encouraged to refer to a specialist dietitian where necessary. For a psychologist or psychiatrist who wants a defensible position on diet without drifting into a discipline they were never trained in, thirteen pages is a fair price.
A paper that says the Mediterranean pattern earns its place, the DASH diet does not yet, and the microbiome is probably involved is giving you something no commercial gut-health course ever will: the shape of its own uncertainty.
The authors are candid that much of the underlying evidence comes from general-population or physical-health samples rather than people with a diagnosed depressive disorder, relies on self-report symptom questionnaires, is short-term without long-term follow-up, and is biased toward high-income countries; the efficacy of dietary change as a sole intervention remains unconfirmed, since trials rarely restrict psychotropic medication. It is also a review with practice recommendations rather than a formally graded clinical guideline, and its referral pathway assumes access to a dietitian that many practices do not have.