This site is intended for health professionals only

Food is medicine: eat your way out of depression

Patrick Holford
BSc DipION FBant
Institute for Optimum Nutrition

Despite all the advances in modern medicine, one in 10 women is diagnosed with breast cancer, one in 14 men is diagnosed with prostate cancer, one in seven people is clinically depressed, one in six people develops diabetes, one in six people dies prematurely from heart attack or stroke, one in four people over 80 has Alzheimer's, one in three people is obese, and one in two is overweight by age 50. They cost the NHS a small fortune, estimated to be in excess of £15 billion a year.(1) Each one of these conditions is diet-related, yet most sufferers are prescribed drugs or medical interventions, not diets or nutritional supplements. Why?
The reason given is usually "there's not enough evidence" or "people don't want to change their diet, they'd rather take drugs" or "there isn't the resource or time to advise people about their nutrition". Are these valid reasons or just excuses? Good medicine means three things: it works, it's safe and it's cost-effective and practical (see Table 1).


Let's take depression as a case in point and examine the evidence. The most common treatment is selective serotonin reuptake inhibitor (SSRI) antidepressants. A recent report on all treatments for depression from NICE says, "There is little important, clinical difference between antidepressants and placebo for mild depression."(2) For mild depression it doesn't recommend antidepressants, favouring exercise and "guided self-help", a kind of journalling and counselling. Unfortunately nutrition has not yet made it onto their agenda.
For moderate-to-severe depression three major reviews show a significant but not spectacular improvement comparing antidepressants with placebo.(3) For example, one such review found that 58% of people taking an antidepressant improved, compared with 45% of those taking placebo: not much difference.(4)
Another found that the difference in Hamilton Rating Scale (HRS) scores, the most widely used psychological test, was between those taking SSRIs and placebo was only 1.7 points - in other words, answering two out of the 17 questions differently (for example, that you'll sleep better and have gained weight). Most meta-analyses of all the published studies find that these "state-of-the-art" SSRI antidepressants lower HRS scores by 10-20%.(5)
What about nutritional approaches? The top three are omega-3 fats, the amino acid 5-HTP and folic acid.

Mood-boosting fish oils
Let's look at omega-3 oils. In this case, there have been five double-blind placebo-controlled trials to date, four of which show significant improvement, although not all used the Hamilton Rating Scale (HRS).(6) The first trial, published in the Archives of General Psychiatry, gave 40 depressed patients either omega-3 supplements or placebo and found a highly significant improvement.(7) The next, published in the American Journal of Psychiatry, tested the effects of giving 20 people suffering from severe depression, who were already on antidepressants but still depressed, a highly concentrated form of omega-3 fat, called ethyl-EPA, versus a placebo. By the third week the depressed patients were showing major improvement in their mood, while those on placebo were not.(8) Of those that were measured using the HRS, the improvement was around 50%. That's twice as effective as SSRIs. Something as simple as recommending a person eat oily fish such as salmon, mackerel, herring or kippers, or take a fish oil supplement, could make a big difference.

How the brain makes serotonin
Serotonin is made in the body and brain from tryptophan, an essential amino acid. Tryptophan is slightly changed into 5-hydroxy tryptophan (5-HTP), which is also found in some foods, most notably meat, fish, beans and eggs. Low levels of serotonin in the brain, or tryptophan in the diet are associated with depression. For example, in studies in which people were given food deficient in tryptophan they became rapidly depressed in a matter of hours.(9) Both tryptophan and 5-HTP supplements have proven effective antidepressants in clinical trials. But just how good are they, compared with antidepressants?
In the case of 5-HTP, which has proven more effective than tryptophan, there have been 27 studies, involving 990 people to date, most of which proved effective.(10) Of these, 11 were double-blind placebo-controlled, six of which measured depression using the HRS. Let's look at these six, since we know that SSRIs average a 15% improvement compared with placebos (see Box 1).


Since these studies differ in design, you can't just add up the scores to gain an average, but you don't need to be a mathematician to realise these results are a lot better than the average 15% improvement reported for antidepressants with a fraction of the side-effects. In play-off studies between 5-HTP and SSRI antidepressants, 5-HTP comes out slightly better. One double-blind trial at Basel University of Psychiatry gave 34 depressed volunteers either the SSRI fluvoxamine (Luvox) or 300mg of 5-HTP. At the end of the six weeks, both groups of patients had had a significant improvement in their depression. However, those taking 5-HTP had a slightly greater improvement, compared with those on the SSRI, in each of the four criteria assessed - depression, anxiety, insomnia and physical symptoms - as well as their own self-assessment, although this improvement was not statistically significant.(11)
B vitamins boost your mood
Our next food factor is B vitamins, and more specifically folic acid. It's known that people with either low blood levels of folic acid or high blood levels of homocysteine, which is an indicator of not getting enough B6, B12 or folic acid, are both more likely to be depressed and less likely to get a positive result from antidepressant drugs. One study compared the effects of giving an SSRI antidepressant, fluoxetine, with either a placebo or with folic acid. While 61% of patients improved on the drug plus placebo, 93% of patients improved on the drug plus folic acid.(12) But how does folic acid itself, a cheap vitamin with no side-effects, compare with antidepressants?
Three trials involving 247 people have set out to answer this question.(13) Two studies involving 151 people assessed the use of folic acid in addition to other treatment, and found that adding folic acid reduced HRS scores on average by a further 2.65 points. That's not as good as the results with 5-HTP or omega-3 fats, but as good as, if not better than, antidepressants. These studies also show that more patients treated with folate experienced a reduction in their Hamilton Rating score of greater than 50% after 10 weeks compared with those on antidepressants. Folate is abundant in green leafy vegetables, beans, nuts and seeds, although most people don't eat anything like the 800µg or more that works best in clinical trials, hence supplementation may work better. Recommending these foods or a daily folic acid supplement can be remarkably effective.

Nutritional medicine is safer than drugs
While folic acid, omega-3 fats and 5-HTP are virtually side-effect-free (very infrequently people get transient nausea on 5-HTP in large amounts), up to 25% of the people taking antidepressants experience significant side-effects, including nausea, vomiting, malaise, dizziness, and headaches or migraines. Fluoxetine, the original SSRI, prescribed to more than 38 million people worldwide, has 45 listed side-effects. However, there are two most serious side-effects - increased risk of suicide and serious withdrawal symptoms - that make SSRIs far from the magic bullet promised in the nineties.
Despite over a decade of denial from both the pharmaceutical industry and the Medicines Health Regulatory Authority, there's no doubt that the increased risk of suicide from taking SSRIs is real. The most comprehensive meta-analysis, a study of 702 trials involving 87,650 patients, recently published in the BMJ,(14) shows a doubling to tripling in the incidence of suicide of patients on SSRIs versus placebos. Since these are trials of depressed patients, you'd expect the opposite, so this really is a serious indictment of these drugs.
The true rate of death in Britain from all adverse drug reactions, excluding potential overdoses, may therefore be greater than 10,000 per year, concludes the author of a survey published in the BMJ in which two hospitals assessed the cause of admission in 18,820 patients.(15) Excluding any potential overdoses, they found that 1,225 (6.5%) of admissions were the result of adverse drug reactions, with 2.3% of these admissions resulting in death. That's a big price to pay when there are, in many cases, safer and highly effective alternatives.
Taking depression as our example, why aren't these highly effective and safe nutrients prescribed? Is it really to do with science, politics or money? After all, there is little profit to be made from a nonpatentable nutrient and virtually nothing taught in nursing or medical training on these approaches. Combining inexpensive and effective diet and supplements, together with other lifestyle factors, including exercise(16) and increased light exposure,(17) as well as counselling is likely to be much more effective, cost-effective and considerably safer than conventional antidepressant medication (see Box 2).


Patrick Holford is author of The Optimum Nutrition Bible and Optimum Nutrition for the Mind and publishes a bi-monthly newsletter exploring nutritional approaches for health problems.For more details see He is also director of the Brain Bio Centre, an outpatient treatment centre for mental health using an optimum
nutrition approach. For details see


  1. Fairweather-Tate S. The Royal Society 2003 Sept;358:1709-27.
  2. NICE. CG23 depression guideline. Available at URL:
  3. Geddes JR, Freemantle N, Mason J, et al. In: The Cochrane Library 2003;2. Oxford: Update Software.
  4. Hansen RA, et al. Ann Intern Med 2005;143(6):415-26.
  5. Khan A, et al. Arch Gen Psychiatry 2000;57:311-24.
  6. Peet M, Stokes R. Drugs 2005;65(8):1051-9.
  7. Stoll A, et al. Arch Gen Psychiatry 1999;56(5):407-12.
  8. Nemets B, et al. Am J Psychiatry 2002;159:477-9.
  9. Smith K, et al. Lancet 1997; 349:915-9.
  10. Turner E, et al. Pharmacol Ther In press 2005.
  11. Poldinger W, et al. Psychopathology 1991;24(2):53-81.
  12. Coppen A, Bailey J. Affective Disorders 2000;60:121-30.
  13. Taylor MJ, et al. In: The Cochrane Library 2003;2. Oxford: Update Software.
  14. Fergusson D, et al. BMJ 2005;330:653.
  15. Pirmohamed M, et al. BMJ 2004;329:15-9.
  16. Craft L, Perna F. J Clin Psychiatry 2004;6(3):104-11.
  17. Leppamaki S, et al. BMC Psychiatry 2004;4(1):22.