Fish oil protects your heart

For an ordinary fish oil capsule, the effect on heart disease is small at most and may be zero. The largest synthesis available pooled 86 randomized trials covering 162,796 people and found little or no effect on death from any cause, or on cardiovascular events overall, rating both of those findings high certainty. The same review found a small reduction in coronary heart disease events and coronary heart disease deaths, rated low certainty, on the order of one coronary event avoided for every 167 people who increase their intake.

The trial closest to what most people actually do tested 1 gram a day in 25,871 US adults aged 50 and over who were not selected for heart problems. Over a median 5.3 years a major cardiovascular event occurred in 386 people taking omega-3 and 419 taking placebo, a difference that did not reach significance. Heart attacks specifically were fewer in the omega-3 group, 145 against 200. Stroke, cardiovascular death and death from any cause were unchanged. Larger benefits have been reported only at 4 grams a day in people already at high cardiovascular risk, and those trials disagree with one another, as described below.

A separate and much narrower question is high-dose omega-3, 4 grams a day, in people already at high cardiovascular risk. REDUCE-IT gave purified EPA to 8,179 statin-treated patients with raised triglycerides, and its primary composite endpoint occurred in 17.2% on treatment against 22.0% on comparator. STRENGTH gave EPA combined with DHA to 13,078 similar patients, the same kind of endpoint occurred in 12.0% against 12.2%, and the trial was halted early for futility. PISCES, published in 2026, gave a fish oil containing both EPA and DHA, or a corn oil placebo, to 1,228 adults on hemodialysis in Canada and Australia. Over about 3.5 years serious cardiovascular events occurred at roughly half the rate on fish oil, a hazard ratio of 0.57, with heart attacks, cardiac deaths and strokes all lower, and adverse events did not differ meaningfully between the groups. People on dialysis carry a very high cardiovascular risk and differ from the populations in the other trials, and this is a single trial.

Two documented differences separate REDUCE-IT and STRENGTH. One is the product: purified EPA in the first, EPA plus DHA in the second. The other is what they were compared against. REDUCE-IT used mineral oil, and a biomarker substudy of that trial found that over 12 months the mineral oil group's C-reactive protein rose 21.9%, interleukin-6 rose 16.2% and interleukin-1 beta rose 28.9%, while the treated group changed little. The authors of that substudy state that what this means for interpreting the trial's results is uncertain. STRENGTH used corn oil, chosen as an inert comparator, and found no association between achieved EPA or DHA blood levels and event rates. A 2024 pooled analysis of 15 large trials and 134,144 people found heart attacks about 11% lower and cardiovascular deaths about 8% lower across all of them, no significant reduction in either in the trials that gave EPA and DHA together, and larger reductions in the three trials of purified EPA alone. Two of those three were open-label trials in Japan with no placebo and the third was REDUCE-IT, so none of them compared purified EPA with an inert placebo. PISCES, which reported a large benefit from a product containing DHA, does not fit a pattern in which only purified EPA works. NICE's guideline on cardiovascular disease prevention advises against offering omega-3 fatty acid compounds to prevent cardiovascular disease, with an exception for icosapent ethyl, the prescription purified EPA used in REDUCE-IT, when it is used as described in a separate NICE appraisal.

There is a signal running the other way. Pooled across 8 cardiovascular outcome trials and 83,112 participants, atrial fibrillation occurred in 4.0% of people taking omega-3 against 3.3% on comparator, a risk ratio of 1.24, and the size of the association tracked the dose: 1.12 at one capsule a day or less, and 1.51 above that. A larger 2026 pooled analysis of 35 trials and 114,592 people aged 50 and over, which added trials not designed around the heart and unpublished counts from trial investigators, found the odds of atrial fibrillation 14% higher overall. Split by dose and by cardiovascular risk, the increase was statistically significant only in people at high cardiovascular risk taking more than 1.5 grams a day, where the odds were 43% higher and the absolute difference was 0.8 percentage points. At 1.5 grams a day or less the odds were 6 to 7% higher, in high-risk and low-risk groups alike, and neither difference was statistically significant. That analysis was supported in part by an omega-3 industry association, and its authors include investigators of several of the trials it pooled. In STRENGTH the rates were 2.2% against 1.3%, which is one additional case for every 114 people treated. Absolute event rates were low throughout. The US prescribing label for Vascepa, the brand of icosapent ethyl, lists atrial fibrillation or flutter requiring hospitalization and bleeding among its warnings, and states that bleeding was more frequent in people also taking antithrombotic drugs such as aspirin, clopidogrel or warfarin. It also states that it is not known whether people with fish or shellfish allergy are at increased risk of an allergic reaction.

Eating fish and taking capsules are not the same body of evidence. The Cochrane review reports little evidence on the effects of eating fish, because nearly all the trials tested supplements, while US dietary guidance separately recommends 8 ounces or more of seafood a week. The one effect that is clear and consistently quantified across this literature is on triglycerides, which fall by roughly 15% in a dose-dependent way. That is a blood measurement rather than a cardiovascular event, and whether lowering it changes outcomes is the question the trials above were built to answer.