Changing what you eat can ease depressive symptoms a little, mostly in people who are not clinically depressed, and the evidence that it treats clinical depression rests on one small trial. A review of 16 randomised trials found dietary interventions reduced depressive symptoms (g = 0.275), though 15 of the 16 studies were in people without clinical depression. [firth-2019] In the SMILES trial, 32.3% of adults with moderate to severe depression who had 12 weeks of dietary support were in remission, against 8.0% of a befriending control, [jacka-2017] while a large prevention trial found a daily supplement did nothing. [bot-2019] This guide covers all three, the limits of each, and what to do with them. Our overview of depression covers the wider picture.
Can changing your diet reduce symptoms of depression?
A little, on average, in trials that mostly enrolled people without clinical depression. Firth and colleagues pooled 16 randomised trials with outcome data for 45,826 participants, and 15 of the 16 studies examined samples with nonclinical depression. Dietary interventions significantly reduced depressive symptoms (Hedges’ g = 0.275, 95% confidence interval 0.10 to 0.45), including in high-quality trials (g = 0.321) and when compared with active controls (g = 0.174, 0.01 to 0.34). [firth-2019]
On the usual rule of thumb, a g of 0.2 is small and 0.5 is medium, so this is a small effect. The same review found no effect on anxiety (g = 0.100, 95% confidence interval −0.04 to 0.24), and studies with female samples saw larger benefits for both depression and anxiety. The authors concluded that dietary interventions hold promise, and that research is still needed on which components of a diet do the work. [firth-2019]
| Study | Who took part | What was tested | What it found |
|---|---|---|---|
| Firth 2019, review of 16 trials | 45,826 people, mostly without clinical depression | Dietary interventions against control conditions | Depressive symptoms fell (g = 0.275); anxiety did not change (g = 0.100) |
| SMILES 2017 | 67 adults with moderate to severe depression | Seven dietician sessions on a modified Mediterranean diet against befriending, 12 weeks | Remission in 32.3% against 8.0% |
| MooDFOOD 2019 | 1,025 overweight adults at raised risk, not currently depressed | Daily supplement or 21 food-related therapy sessions, one year | No reduction in new episodes of major depression |
What did the SMILES trial find in people with depression?
It found more remission with dietary support than with befriending, in a small trial. SMILES was a 12-week, single-blind randomised trial of an adjunctive dietary intervention in moderate to severe depression: seven individual sessions with a clinical dietician against a social support protocol of the same length. The primary outcome was the Montgomery–Åsberg Depression Rating Scale (MADRS) at 12 weeks. [jacka-2017]
Of 67 people enrolled (33 in the diet group, 34 in the control group), the diet group improved significantly more on the MADRS (Cohen’s d = −1.16). Remission, a MADRS score under 10, was reached by 32.3% (10 people) of the diet group and 8.0% (2 people) of the control group. The number needed to treat for one extra remission was 4.1 (95% confidence interval 2.3 to 27.8). [jacka-2017]
Left view: remission at 12 weeks in Jacka et al. (2017), SMILES, 10 of 31 completers on diet support and 2 of 25 on social support. Right view: new episodes of major depression over one year in Bot et al. (2019), MooDFOOD, by whether people took the supplement or placebo and whether they had food-related therapy.
The diet was a modified Mediterranean diet based on the Australian and Greek dietary guidelines. It focused on 12 food groups, beginning with whole grains (5 to 8 servings a day), vegetables (6 a day), fruit (3 a day), legumes (3 to 4 a week) and low-fat unsweetened dairy (2 to 3 a day). Participants were chosen because their diet was poor, with low fibre, lean protein, fruit and vegetables and high sweets, processed meats and salty snacks, and 55 of the 67 were already using medication, psychotherapy or both. [jacka-2017]
Does a Mediterranean-style diet prevent depression?
Not in the one large trial that tested a food-related therapy for prevention. MooDFOOD randomly assigned 1,025 overweight adults with elevated depressive symptoms and no depressive episode in the previous six months to a daily supplement or placebo, and to 21 sessions of food-related behavioural activation therapy, which emphasised a Mediterranean-style diet, or none. Over a year, 105 people (10%) developed major depression. The odds ratio was 0.93 for therapy (95% confidence interval 0.76 to 1.13) and 1.06 for supplements (0.87 to 1.29), so neither made a significant difference. [bot-2019]
The trial’s authors list reasons for caution. Fewer people developed depression than expected, which lowered the power to detect an effect, about a quarter were lost to follow-up, there was no active control for the therapy, and one year may have been too short. An analysis that accounted for how many sessions people attended found lower onset with the therapy (odds ratio 0.78, 0.64 to 0.95), which suggests a possible effect with enough sessions and needs testing on its own. [bot-2019]
The trial was run because prospective studies link higher-quality diets to lower future onset of depressive symptoms. [bot-2019] Those studies observe what people eat and cannot show that diet is the reason.
Do supplements help prevent depression?
Not in MooDFOOD, which is the largest trial of its kind reviewed here. The daily supplement contained omega-3 fatty acids (1,412 mg), selenium, folic acid and vitamin D with calcium, and 10.5% of the supplement group developed major depression against 9.9% on placebo (odds ratio 1.06, 95% confidence interval 0.87 to 1.29). [bot-2019]
On secondary measures the supplement group scored slightly worse than placebo for depressive and anxiety symptoms, and the authors concluded that the findings do not support using these supplements to prevent major depression. Participants were not chosen for a deficiency in the nutrients provided, so the trial does not test supplements in people who are deficient. [bot-2019]
What do these studies not show about diet and depression?
They do not show that diet alone treats depression, and they do not show that any single food or supplement does. The review pooled trials in mostly nonclinical samples, SMILES was small and could not blind participants, and MooDFOOD enrolled people who were not depressed at the start. [firth-2019] [jacka-2017] [bot-2019]
SMILES is the only trial here in people with clinical depression, and its authors say the small sample may have inflated the effect size. They also note that the inability to blind participants may have biased the results and caused different dropout rates: 31 of 33 in the diet group and 25 of 34 in the control group had complete data at 12 weeks. The trial was funded by Australia’s National Health and Medical Research Council, with food vouchers sponsored by a supermarket chain. [jacka-2017]
What can you do with this evidence?
Treat food as one supportive part of care and not a replacement for treatment. If you want to try the pattern SMILES tested, it was a modified Mediterranean diet built around whole grains, vegetables, fruit, legumes and dairy, with a dietician guiding it, and most participants were also using medication or therapy. [jacka-2017] A small first step, such as one more portion of vegetables a day, is easier to keep up than a rebuild.
Appetite changes are part of low mood for many people, so skipped meals or eating far more than usual are worth noticing in their own right. Our guide to exercise for depression and behavioural activation cover two other changes with trial evidence behind them.
How close is your eating to where the SMILES trial started?
Tick anything that is true for you most weeks. This is a prompt for reflection, not a test, and it produces no diagnosis.
0 of 5 ticked
Most of these match the pattern the SMILES participants started from. Pick one, such as a portion of vegetables with the evening meal, and keep it for two weeks before changing anything else. If your mood is low, tell a doctor as well.
A few of these are habits. Choose the one that would be easiest to change and try it this week. If your eating shifts with your mood, that is worth mentioning to a doctor.
Your eating does not look like the pattern the trial started from. Diet is only one part of how mood is managed, so if your mood is low, the depression screener is a next step.
When to seek help
Speak to a doctor if low mood or loss of interest has lasted for weeks, or if it is affecting work, relationships or daily life, whatever your diet looks like. Do not stop or change a prescribed medicine because of what you eat without speaking to the person who prescribed it.
If you feel unable to keep yourself safe, contact your local emergency services or a crisis helpline.
How MyFreud can help
A change in eating is easy to judge by feel and hard to judge by memory. Daily mood tracking in MyFreud shows whether your mood moves over the weeks after you change something, and the notepad is somewhere to write down what you changed and how the days went. It does not replace treatment.
Download MyFreud and start today: App Store or Google Play.