Food and Behaviour Research

Donate Log In

Randomised double-blind placebo-controlled trial of fish oil in the treatment of depression

Silvers KM, Woolley CC, Hamilton FC, Watts PM, Watson RA. (2005) Prostaglandins Leukot Essent Fatty Acids.  72(3) 211-8. 

Web URL: Licensed users of PLEFA can view the full text of this article here


Converging evidence suggests that omega-3 polyunsaturated fatty acids have aetiological importance in depression. To determine the effect of adding fish oil to existing therapy in participants who were being treated for depression in a community setting,

77 participants were randomly assigned to receive 8 g of either fish or olive oil per day in addition to their existing therapy. Fifty-nine (77%) participants completed 12 weeks of treatment. Dietary, biochemical and lifestyle factors were measured throughout the study. Mood was assessed using the Short Form Hamilton Depression Rating Scale (HDRS-SF) and the Beck Depression Inventory II. Sample size calculations were based on the HDRS-SF. Intention-to-treat and per protocol analyses were carried out using residual maximum likelihood.

There was no evidence that fish oil improved mood when compared to the placebo oil, despite an increase in circulating omega-3 polyunsaturated fatty acids. However, mood improved significantly in both groups within the first 2 weeks of the study (P<0.001) and this improvement was sustained throughout.

In conclusion, fish oil was no more effective than the control as an add-on therapy for depression in this setting.


Increasing evidence suggests that an increased dietary intake of omega-3 fatty acids might help to prevent or reduce depression and other mood disorders. (See Logan (2004). for an excellent introductory review). This evidence includes results from several controlled trials, although the most promising results have come from studies using either EPA-enriched fish oils or pure ethyl EPA (see Freeman et al 2006)

This study involved community mental health patients suffering from a depressive episode, who were treated for 12 weeks with either high-DHA fish oil or an olive oil placebo in a randomised controlled trial. Results showed no advantages for the fish oil over placebo, but both groups of patients showed significant improvements within 2 weeks that were sustained for the duration of the study.

At least three overlapping factors may be relevant in the interpretation of these findings

  • the high total dose of oils used
  • the very low ratio of EPA to DHA provided by this particular fish oil, and
  • the strong placebo effect that was found.

The treatment involved very high doses of the supplemented oils, at 8 x 1000mg capsules daily, providing 3g/day of EPA+DHA combined. More is not always better. The only dose-ranging study of omega-3 treatment in depression published to date (See Peet & Horrobin 2002) found 1g/day of pure ethyl EPA more effective in reducing depressive symptoms than either 2g or 4g (which were no better than placebo)

The treatment in this study provided 2400mg DHA and 600mg EPA each day, an unusually low ratio of EPA:DHA at 0.25 (the usual ratio in ordinary fish oils is around 1.5, i.e. they provide 50% more EPA than DHA, not 75% less)

Successful controlled trials of omega-3 treatment in depression have used either pure ethyl EPA or much higher ratios of EPA:DHA than did this study (and a higher total EPA content). The only controlled trial of pure DHA for depression to date found no benefits over placebo ( Marangell et al 2003). This study found a similar lack of benefit from a high-DHA fish oil.

The placebo was olive oil, providing just over 5g of the monounsaturated oleic acid. Again, this is a high dose, and as oleic acid has some psychoactive properties (including calming and sleep-inducing effects), it is already acknowledged that this is not an ideal placebo.

Both treatment groups showed significant and sustained improvements in depressive symptoms, suggesting that strong 'placebo effects' were operating in this study. It is possible that there might have been some effects from both the olive oil and the fish oil, or from general expectations, but depression often improves simply with the passage of time.