Omega-3 (Fish Oil) for Anxiety: What the Evidence Shows

Fish oil is sold as a gentle, natural way to take the edge off anxiety. The evidence is more specific — and more interesting — than the label.

The honest version: omega-3 does appear to reduce anxiety, but almost entirely under two conditions. The dose has to be high — around 2 grams a day — and the signal is in people with a diagnosed condition, not in healthy people looking to feel calmer. In the general population, the best evidence says omega-3 does not prevent anxiety at all. Two large bodies of research that look like they disagree actually don’t. Here is how they reconcile.

Does omega-3 actually reduce anxiety?

The strongest “yes” comes from a 2018 meta-analysis in JAMA Network Open (Su et al.). Pooling 19 clinical trials across 2,240 participants from 11 countries, it found that people taking omega-3 had a significant reduction in anxiety symptoms versus controls (Hedges g, 0.374; 95% CI, 0.081–0.666; P = .01).

A Hedges g of 0.37 is a small-to-moderate effect — real, but not dramatic. And the average masks where the effect actually lives. That’s the part the supplement pitch skips.

Why does the dose matter so much?

Because below roughly 2 grams a day, the effect largely disappears.

In the JAMA analysis, trials using at least 2,000 mg/day showed a significant benefit (Hedges g, 0.213; P = .02), while lower-dose trials did not reach significance. A separate 2024 dose-response meta-analysis in BMC Psychiatry (Bafkar et al.) — 23 trials, 2,189 participants — sharpened the point: the greatest improvement landed at 2 g/day (SMD, −0.93), and supplementation below 2 g/day did not affect anxiety symptoms at all.

For context, a standard fish-oil softgel delivers only 200–300 mg of combined EPA and DHA. Hitting 2 grams of active omega-3 means several concentrated capsules a day — far more than most people casually take. If you’re dosing “one a day,” you’re almost certainly in the range the trials found does nothing.

Does it work for everyone — or only if you’re already anxious?

Only if there’s a clinical condition present. This is the cleanest split in the data.

The JAMA meta-analysis separated participants with a specific clinical diagnosis from those without one. The result was stark: in people with a condition, omega-3 clearly helped (Hedges g, 0.512); in people without any specific clinical condition, the effect was essentially zero (Hedges g, −0.008; 95% CI, −0.266 to 0.250; P = .95).

That last number is the honest answer to “will fish oil calm me down if I’m generally stressed but healthy?” On this evidence: no.

So why do some sources say omega-3 does nothing?

Because they’re answering a different question — prevention in healthy people — and getting a defensible null.

A WHO-funded systematic review (Deane et al., 2019, British Journal of Psychiatry, summarized by the UK’s NIHR) pooled 32 randomized trials and 46,467 adults. Its conclusion: increasing long-chain omega-3 intake has little or no effect on preventing anxiety or depression in people who don’t already have them. For anxiety specifically, the effect in people without anxiety at baseline was SMD 0.15 — and it slightly favored the lower-omega-3 group.

There’s no real contradiction. The JAMA and BMC meta-analyses tested treatment — high doses, in people with symptoms or a diagnosis. The WHO review tested prevention — in healthy people. Both can be true: omega-3 may modestly help a clinical anxiety problem at 2 g/day, and do nothing to keep a healthy person from developing one. The mistake is buying fish oil for the second job on the strength of the first.

Does the type of omega-3 (EPA vs DHA) matter?

This is where anxiety breaks from the depression playbook. In depression research, EPA-rich formulas (≥60% EPA) tend to outperform. For anxiety, the JAMA data pointed the other way: the benefit showed up in formulas with less than 60% EPA (Hedges g, 0.485), not in the high-EPA ones (Hedges g, 0.092; P = .35). The authors flagged this as genuinely puzzling and likely a different underlying mechanism. Translation: don’t assume the “high-EPA for mood” advice transfers to anxiety. The composition question isn’t settled.

The takeaway

Omega-3 is cheap, safe (the 2024 review found no increase in adverse events), and worth considering as an adjunct — but only if you respect the two conditions the data keeps repeating: a real dose (about 2 g/day of active EPA + DHA) and a real target (clinical anxiety, not everyday stress). For healthy people hoping to head anxiety off, the evidence says save your money.

If you want a supplement with clearer trial data for stress specifically, ashwagandha has stronger cortisol-lowering evidence — with its own liver-safety caveat worth reading first. And magnesium sits in a similar “modest, mostly if you’re low” bucket to omega-3. For the system all of these plug into — how a nervous system actually winds down — see our anxiety regulation and sleep restoration work.