Did the SMILES trial really show diet treats depression?

The SMILES trial, published in BMC Medicine in 2017, was a single-blind randomised controlled trial of 67 adults with moderate-to-severe depression on usual care. The intervention was a 12-week modified Mediterranean dietary pattern delivered by an accredited dietitian; the control was social support of equal contact time. The intervention group had a statistically and clinically meaningful improvement in MADRS depression scores, and approximately 32% achieved remission compared with about 8% in the control. It was a small trial and not all features replicated, but the direction was supported by subsequent trials in adolescents (AMMEND), young adults (HELFIMED), and an older-adult sample. Diet is now recognised as an adjunct, not a substitute, in depression care.

What diet pattern has the best evidence for mental health?

The Mediterranean dietary pattern — vegetables, legumes, fruit, whole grains, fish, olive oil as the primary added fat, nuts and seeds, fermented dairy in moderation, limited red meat, and limited processed food — has the strongest and most consistent evidence across observational studies and randomised trials. The DASH pattern and traditional Asian patterns share many features. Specific 'brain foods' marketed in isolation — blueberries, dark chocolate, turmeric — do not have the same trial-level support. The pattern matters more than any single food.

Should I take omega-3 supplements for depression?

Omega-3 supplementation — specifically EPA-dominant formulations (≥60% EPA) at doses around 1–2 g/day of combined EPA+DHA — has modest evidence for major depressive disorder, larger when used alongside antidepressant therapy. The 2019 Cochrane review noted small effect sizes and many heterogeneous trials. Australian practice typically considers omega-3 as an adjunct, not first-line monotherapy. Best discussed with your GP, especially if you are on blood-thinning medication.

Can ultra-processed food make depression worse?

Multiple large prospective cohorts — including the BMJ 2024 umbrella review of 45 meta-analyses — found that high consumption of ultra-processed food was associated with about a 22% higher risk of incident depression and worse mental-health outcomes overall. The association is consistent across countries and survives adjustment for confounders, but the evidence is observational, so causality is not proven. Mechanisms under investigation include effects on gut microbiota, glycaemic variability, low fibre, and additives. Australian dietary guidelines already advise limiting ultra-processed food on cardiovascular and metabolic grounds; the mental-health signal adds further reason.

Where does diet sit alongside therapy and medication?

For mild to moderate depression, the RACGP and RANZCP guidance recommend a stepped approach starting with psychoeducation, lifestyle (sleep, exercise, diet, alcohol moderation), and structured psychological therapy such as CBT. Medication is added when symptoms are moderate-to-severe, persistent despite the above, or when the diagnosis is bipolar depression or another condition needing pharmacological treatment. Diet is part of the lifestyle layer — meaningful, modifiable, and complementary, but not a stand-alone treatment for major depression.

Depression and anxiety — dietary management

Diet and mental health: what the SMILES trial and follow-ups actually showed

By Dr HB Lo, FACRRM Reviewed 15 May 2026 9 min read

Diet is one component of mental-health care in Australian general practice — not a replacement for psychological therapy or medication, but a modifiable factor with growing trial evidence.

The 2017 Melbourne SMILES trial showed that a 12-week Mediterranean-pattern dietary intervention produced meaningful improvement in moderate-to-severe depression on top of usual care (≈32% remission vs ≈8% in social-support control). Follow-up trials (HELFIMED, AMMEND) replicated a meaningful effect on depressive symptoms. The effect on anxiety is more modest.

RACGP and RANZCP guidance both treat diet as part of comprehensive depression care — an adjunct, not a substitute.

The conversation that has actually changed

For decades, diet was treated in mental-health care as a quality-of-life concern, not a clinical lever. That changed in 2017 when the SMILES trial — run from the Food and Mood Centre at Deakin University in Geelong — published the first randomised controlled trial to test whether dietary change improved moderate-to-severe depression on top of usual care. The trial was small (67 adults), the intervention was a modified Mediterranean pattern delivered by an accredited dietitian, and the result was unexpected by the field’s then-prevailing assumptions: about one in three participants in the dietary group reached remission at 12 weeks, compared with about one in twelve in the social-support control.

The trial was small enough that scepticism was reasonable. Follow-up research has answered some of that scepticism — the HELFIMED trial in older Australian adults, the AMMEND trial in young men, and several smaller adolescent and young-adult studies all show meaningful effects of Mediterranean-pattern eating on depression scores. The effects are not large in absolute terms; they are real and add to standard care.

The Australian guidance — including the RANZCP Mood Disorders guideline — now treats lifestyle factors, including diet, as one component of comprehensive depression care, alongside psychological therapy and (where indicated) medication.

A. Core clinical — what the evidence actually says

The strongest signals come from three places: randomised trials of dietary intervention, large prospective cohorts on dietary pattern and incident depression, and umbrella reviews of ultra-processed food and mental health.

Randomised trials. The trials that matter most are SMILES (Australian adults with moderate-to-severe depression), HELFIMED (Australian adults aged ≥35), AMMEND (young Australian men), and several international replications. Meta-analyses of these and similar trials consistently show a modest but reproducible reduction in depression symptom scores with Mediterranean-pattern eating. The standardised mean difference is in the small-to-moderate range — meaningful but not transformative. The effect sits alongside, rather than substituting for, therapy and medication.

Cohort studies. Long-term cohorts including the SUN cohort in Spain and the Whitehall II cohort in the UK consistently associate higher adherence to a Mediterranean dietary pattern with a 20–35% lower risk of incident depression over follow-up of 4–12 years. The size of the association is similar across cohorts and survives adjustment for many confounders, though residual confounding from socioeconomic status and overall lifestyle pattern cannot be fully excluded.

Ultra-processed food. The 2024 BMJ umbrella review pooled 45 meta-analyses across 32 health outcomes. For mental health, the strongest associations were with depression (≈22% higher risk in highest vs lowest UPF consumption) and worse overall mental wellbeing. The evidence quality was moderate — the mechanisms most plausibly involve gut microbiota, low fibre, glycaemic variability, additives, and displacement of whole foods, but the evidence base is observational.

Specific nutrients.

Omega-3 fatty acids. The 2021 Cochrane review found a small effect in major depressive disorder, larger with EPA-dominant formulations and when used alongside antidepressants. Evidence quality is moderate; effect size is small to moderate.
Vitamin D. The major randomised trials (VITAL-DEP, D-Health) did not show a meaningful effect of vitamin D supplementation on depression in adults who were not deficient. Where vitamin D is genuinely low and the clinical picture warrants treatment, replacement is reasonable, but routine supplementation for mood in adults with normal levels is not supported.
Folate and B12. Replacement in deficient adults is supported. L-methylfolate has been studied as an adjunct for treatment-resistant depression with mixed results; it is not routine.
Probiotics. A small effect on depressive symptoms has been reported in some meta-analyses; trials are heterogeneous and not yet of a quality to recommend specific strains routinely.
Magnesium, zinc, selenium. Limited evidence; replacement in deficient adults is supported, but routine supplementation is not.

B. Evidence appraisal and controversies

The “treats depression” overstatement. The SMILES trial result has sometimes been described publicly as “diet treats depression” — a framing the original authors have explicitly pushed back against. The trial showed that on top of usual care, a Mediterranean-pattern dietary intervention reduced depression scores meaningfully more than social support of equal contact time. That is an adjunct effect, not a stand-alone treatment effect. For people with major depressive disorder, dietary change does not replace trial-supported psychological therapy or pharmacotherapy where these are indicated.

Generalisability of SMILES. The trial enrolled 67 Australian adults, 95% of whom were already on antidepressant therapy or in psychological treatment at baseline. The intervention was delivered by an accredited dietitian over multiple sessions — not a brochure. Generalising the effect size to people not receiving any other care, or to people receiving only a leaflet about diet, is not supported by the trial.

The microbiome story. Gut microbiota composition is associated with mental-health outcomes in observational data and in animal models. Human trials of specific probiotic or prebiotic interventions have shown modest signals but no clear winner. The 2024 evidence does not yet support routine probiotic prescribing for depression or anxiety. The microbiome story is real and likely mechanistically important; specific clinical recommendations beyond a high-fibre whole-food dietary pattern are premature.

“Inflammation is the cause of depression”. Inflammation is one of several biological correlates of depression in subgroups, particularly in treatment-resistant illness or where there is co-existing metabolic disease. It is not the cause of all depression. Anti-inflammatory dietary patterns make sense on cardiovascular and metabolic grounds and align with the Mediterranean-pattern evidence above. They are not a stand-alone explanation for mood disorders.

Elimination diets, “leaky gut”, and individualised IgG-based food panels. The Australasian Society of Clinical Immunology and Allergy advises against IgG-based food-intolerance panels in clinical decision-making, including for mood symptoms. Coeliac disease should be ruled out where there is a clinical suspicion (chronic gastrointestinal symptoms, family history, iron deficiency anaemia); it does occasionally present with mood symptoms and improves with treatment. Beyond that, restrictive elimination diets for mental-health symptoms are not supported and can cause harm (nutrient deficiency, eating-disorder triggering, social isolation).

Sugar. Reducing added sugar is reasonable on cardiovascular and metabolic grounds and is consistent with the Australian Dietary Guidelines. The specific causal link between sugar intake and clinical depression is not as well established as the broader UPF association — most of the signal in cohort data is captured by overall dietary pattern rather than sugar specifically.

The honest picture: dietary pattern matters. A Mediterranean-style or similar whole-food pattern is supported by the evidence as a meaningful adjunct in mental-health care. Specific superfoods, supplements, and elimination diets are markedly more aggressively marketed than the evidence supports.

C. Australian operations — how this fits into the visit

A focused conversation about diet and mental health in Australian general practice usually happens during one of:

A Mental Health Treatment Plan (item 2715, 60 minutes, or 2717, 90 minutes) — the structured plan that opens access to ten subsidised psychology sessions per calendar year via Better Access. Diet, sleep, exercise, alcohol, and substance use are part of the lifestyle section.
A long consultation (item 36) or very long consultation (item 44) as part of an unscheduled review for low mood, fatigue, or anxiety.
A GP Management Plan (item 721) with Team Care Arrangement (item 723) when there are co-existing chronic conditions — diabetes, hypertension, obesity — that overlap with mental-health care. This combination opens up to five subsidised allied-health visits per calendar year, including with an accredited practising dietitian via the Chronic Disease Management items.

(MBS / PBS items verified 2026-05-15 via WebSearch — workspace egress to mbsonline.gov.au + pbs.gov.au still blocked; spot-check confirms current.)

For someone presenting with low mood, the standard Australian general practice approach is:

Screen for severity with PHQ-9 (depression) and GAD-7 (anxiety).
Rule out reversible medical causes that present as fatigue or low mood — thyroid disease, B12 deficiency, anaemia, sleep apnoea, alcohol, prescription medication side effects.
Stage of care. For mild to moderate depression: psychoeducation, lifestyle (sleep, exercise, diet, alcohol moderation), structured psychological therapy as the foundation. For moderate-to-severe: add pharmacotherapy alongside the above.
Refer to a dietitian if the eating pattern is far from a Mediterranean or DASH framework, if there is co-existing diabetes or cardiovascular disease, or if there is an eating-disorder concern. Dietitians Australia maintains a public directory.

For severe mental illness, treatment-resistant depression, suspected bipolar disorder, perinatal mental-health concerns, or significant suicidality, general practice moves to specialist referral (psychiatrist, perinatal mental-health team, or hospital-based service) and the dietary conversation is part of, not central to, the plan.

D. Integrative considerations

A practical, AU-evidence-grounded order of operations for diet in mental-health care:

Step	Evidence	What it looks like
1. Reduce ultra-processed food, particularly UPF that displaces vegetables, fruit, legumes, whole grains, and oily fish	Moderate (BMJ 2024 umbrella review)	Plain whole foods on the plate; cooking from raw ingredients more often
2. Build toward a Mediterranean-pattern week	Strong (SMILES, HELFIMED, AMMEND)	Vegetables and legumes at most meals; fish 2–3 times/week; olive oil as primary added fat; nuts and seeds; limited red meat
3. Fibre target	Strong (general health)	25–30g/day from vegetables, legumes, whole grains, fruit, nuts
4. Alcohol moderation within NHMRC 2020 guidelines	Strong	≤10 standard drinks/week, ≤4 in any one day; less is better
5. Caffeine moderation, especially after midday in anxiety	Moderate	Under 400 mg/day; none after early afternoon if sleep is affected
6. Sleep, movement, sunlight	Strong	7–9 hours sleep, 150+ minutes/week moderate aerobic activity, daily morning daylight
7. Omega-3 from food first; supplementation as an adjunct in major depression after discussion	Moderate (Cochrane)	Oily fish 2–3 times/week; supplementation only after a clinical conversation
8. Correct nutrient deficiencies (iron, B12, vitamin D) when present	Strong	Repletion based on bloods, not on routine supplementation
9. Probiotics, “gut healing” protocols	Weak	Not routine; reasonable as an adjunct if a specific gut disorder is being managed
10. Elimination diets without a specific clinical indication	Not supported	Avoid; risk of harm > benefit

A few practical anchors for someone wanting to make this concrete in an Australian context:

The CSIRO Total Wellbeing Diet, the Mediterranean meal plans from Heart Foundation, and the NHMRC Australian Dietary Guidelines all describe practical AU-relevant patterns.
An accredited practising dietitian (APD) via Dietitians Australia provides individualised guidance and is partially subsidised under chronic-disease and mental-health care plans.
For people with disordered eating concerns, the Butterfly Foundation helpline and InsideOut Institute provide AU resources.

When to seek help sooner, not later

Mental-health symptoms that warrant prompt care:

Thoughts of suicide or self-harm — call Lifeline 13 11 14 or present to an emergency department
Sudden severe change in mood, energy, or thinking
Inability to function at work or to look after dependents
Distressing intrusive thoughts or hallucinations
Concern from people close to you that something is significantly off

In acute distress or suicidal crisis, contacts are:

Lifeline: 13 11 14
Suicide Call Back Service: 1300 659 467
Beyond Blue: 1300 22 4636
Kids Helpline (5–25 years): 1800 55 1800

Mental-health crisis or risk to self warrants emergency department review — diet is a longer conversation that happens after safety.

What this article is and is not

This is general health information drawn from current Australian general practice guidelines, the RANZCP Mood Disorders guideline, the Australian Dietary Guidelines, and major randomised trials of dietary intervention in depression. It is not personal medical advice and does not create a doctor–patient relationship. Decisions about diagnosis, therapy, medication, and dietary referral are made with your own GP and treating clinicians, who know your history.

For Australian consumer-friendly summaries covering the same ground: HealthDirect — Depression, Better Health Channel, Beyond Blue, and the Black Dog Institute.

Sources cited

Frequently asked questions

Did the SMILES trial really show diet treats depression?

The SMILES trial, published in BMC Medicine in 2017, was a single-blind randomised controlled trial of 67 adults with moderate-to-severe depression on usual care. The intervention was a 12-week modified Mediterranean dietary pattern delivered by an accredited dietitian; the control was social support of equal contact time. The intervention group had a statistically and clinically meaningful improvement in MADRS depression scores, and approximately 32% achieved remission compared with about 8% in the control. It was a small trial and not all features replicated, but the direction was supported by subsequent trials in adolescents (AMMEND), young adults (HELFIMED), and an older-adult sample. Diet is now recognised as an adjunct, not a substitute, in depression care.
What diet pattern has the best evidence for mental health?

The Mediterranean dietary pattern — vegetables, legumes, fruit, whole grains, fish, olive oil as the primary added fat, nuts and seeds, fermented dairy in moderation, limited red meat, and limited processed food — has the strongest and most consistent evidence across observational studies and randomised trials. The DASH pattern and traditional Asian patterns share many features. Specific 'brain foods' marketed in isolation — blueberries, dark chocolate, turmeric — do not have the same trial-level support. The pattern matters more than any single food.
Should I take omega-3 supplements for depression?

Omega-3 supplementation — specifically EPA-dominant formulations (≥60% EPA) at doses around 1–2 g/day of combined EPA+DHA — has modest evidence for major depressive disorder, larger when used alongside antidepressant therapy. The 2019 Cochrane review noted small effect sizes and many heterogeneous trials. Australian practice typically considers omega-3 as an adjunct, not first-line monotherapy. Best discussed with your GP, especially if you are on blood-thinning medication.
Can ultra-processed food make depression worse?

Multiple large prospective cohorts — including the BMJ 2024 umbrella review of 45 meta-analyses — found that high consumption of ultra-processed food was associated with about a 22% higher risk of incident depression and worse mental-health outcomes overall. The association is consistent across countries and survives adjustment for confounders, but the evidence is observational, so causality is not proven. Mechanisms under investigation include effects on gut microbiota, glycaemic variability, low fibre, and additives. Australian dietary guidelines already advise limiting ultra-processed food on cardiovascular and metabolic grounds; the mental-health signal adds further reason.
Where does diet sit alongside therapy and medication?

For mild to moderate depression, the RACGP and RANZCP guidance recommend a stepped approach starting with psychoeducation, lifestyle (sleep, exercise, diet, alcohol moderation), and structured psychological therapy such as CBT. Medication is added when symptoms are moderate-to-severe, persistent despite the above, or when the diagnosis is bipolar depression or another condition needing pharmacological treatment. Diet is part of the lifestyle layer — meaningful, modifiable, and complementary, but not a stand-alone treatment for major depression.

Source quality

Sources grouped by evidence tier. AU primary tier first; international where AU is silent or lagging; named-author reconstruction where guidelines have not yet caught up. How tiers work.