Omega-3 and fish oil: uses, evidence, dosing and safety

Fish oil occupies an unusual position. It has the single strongest result of anything covered on this site, rated at the highest level of certainty researchers use. It also has a harm signal that scales with dose, which the bottles do not mention.

Both facts come from the same body of evidence, and this page covers thirteen further areas where it has been tested.

What omega-3 fats are

Omega-3 fatty acids are a type of polyunsaturated fat, and the reason they matter is stated in the same reference: your body does not make omega-3 fatty acids on its own.

Three of them are worth naming. EPA, or eicosapentaenoic acid, and DHA, or docosahexaenoic acid, are the long-chain forms found in oily fish and in most capsules. ALA, or alpha-linolenic acid, is the plant form in flaxseed, walnuts and rapeseed oil, and the body converts only a small fraction of it into the other two.

We need these fats to build brain cells and for other important functions, which is why the deficiency case is not in dispute even where the supplement case is.

How omega-3 works

Omega-3 fats are built into cell membranes, where they change how flexible the membrane is and which signaling molecules the cell can make from it.

That second part drives most of the claims. The body converts omega-3 and omega-6 fats into competing families of signaling compounds, one broadly calming and the other broadly inflammatory, and they compete for the same enzymes. Shift the ratio and you shift the balance of signals.

It is a good mechanism, and this page is largely the story of how imperfectly it translates.

What the evidence says about omega-3 and triglycerides

Start here, because it is the strongest result and the least advertised.

The 2020 Cochrane review found that increasing LCn3 reduced triglycerides by ˜15% in a dose-dependent way (high-certainty evidence), where LCn3 is long-chain omega-3, meaning EPA and DHA.

Two things make this stand out. High-certainty evidence is rare anywhere in nutrition and almost unheard of for a supplement. And dose-dependent means more produced more, which is the pattern a real effect makes and a spurious one usually does not.

Triglycerides are a blood fat measured alongside cholesterol, and high levels are treated in their own right. This is the one thing fish oil unambiguously does.

What the evidence says about omega-3 and heart disease

The claim on the front of the box, and the answer is a much quieter version of it.

Cochrane’s assessment describes itself as the most extensive systematic assessment of effects of omega-3 fats on cardiovascular health to date. Its finding is carefully worded: moderate- and low-certainty evidence suggests that increasing LCn3 slightly reduces risk of coronary heart disease mortality and events.

Slightly, at moderate to low certainty. That is a real benefit and a small one, and it sits a long way below the high-certainty triglyceride result on the same page.

The plant form did something similar and separate: increasing ALA slightly reduces risk of cardiovascular events and arrhythmia.

Everything else was flat. Omega-3 had little or no effect on serious adverse events, adiposity, lipids and blood pressure.

What the evidence says about omega-3 and heart rhythm

This section exists because the finding is genuinely important and almost never mentioned alongside the benefit.

A 2026 meta-analysis found that high-dose EPA/DHA treatment is associated with an increased risk of AF in patients at high cardiovascular disease risk. AF is atrial fibrillation, an irregular heart rhythm that itself raises the risk of stroke.

The size is modest in absolute terms, at an absolute risk difference of 0.8%, meaning roughly eight extra cases per thousand people treated.

The dose dependence is the actionable part. Low-dose EPA/DHA does not appear to increase AF risk, even in high-risk populations.

So the picture is not “fish oil is dangerous”. It is that the high-dose products, taken by exactly the people with heart disease who are most likely to be sold them, carry a rhythm risk that the modest benefit has to be weighed against. That is a conversation with a cardiologist, not a shelf decision.

What the evidence says about omega-3 and depression

The mental health claim is large, and the best-quality evidence behind it is in children, where it is weak.

A 2024 Cochrane review reached an unusually plain verdict: the effect of omega-3 PUFA supplementation, PUFA meaning polyunsaturated fatty acid, in improving self-reported depression symptoms or achieving depression remission in children and adolescents is very uncertain.

Safety was not the issue. While no harms are evident, more data are needed to confirm potential risks.

What the reviewers call for is telling about the state of the field: larger sample sizes, varied dosages, eicosapentaenoic acid/docosahexaenoic acid ratios, treatment durations, and safety profiles all remain unsettled after decades of trials.

What the evidence says about omega-3 and anxiety

Separate from depression, and interesting mainly for the dose threshold it identifies.

A dose-response meta-analysis found improvement in anxiety symptoms with the greatest improvement at 2 g/d, and that supplementation in a dose lower than 2 g/d did not affect anxiety symptoms.

That threshold matters commercially. Many supermarket capsules contain a few hundred milligrams of combined EPA and DHA, well under the level at which anything was observed here.

The certainty is the catch, and the authors state it: evidence of very low certainty. Safety was reassuring, with no increase in adverse events at moderate certainty.

What the evidence says about omega-3 and memory

Neither the yes nor the no that gets reported.

A 2025 review in people with mild cognitive impairment found omega-3 affected some domains of cognitive function, such as FSIQ, information processing, and digit span/working memory/attention, where FSIQ is full-scale IQ.

And it found no effect it could distinguish from chance for some domains, such as picture completion, picture arrangement, or block design.

Partial movement across a test battery, in people who already have measurable cognitive impairment. That is a reason for more research. It is not evidence that fish oil protects a healthy brain from dementia, which no section of this evidence base supports.

What the evidence says about omega-3 and diabetic nerve pain

A Cochrane review, and a clean statement of not knowing.

There are inadequate data to draw conclusions about the effects of omega-3 PUFA supplementation on peripheral nerve impairment in adults with diabetes mellitus, and there may be little to no benefit for symptoms or quality of life compared with placebo.

Nerve pain in diabetes is common, poorly treated and miserable, which makes it a natural target for supplement marketing. On the evidence, there is nothing here to buy.

What the evidence says about omega-3 and fatty liver

A marker improved, the scans did not, and the reviewers flag a safety point worth noting.

A 2025 meta-analysis found omega-3 supplementation may improve GGT levels and hepatic steatosis assessed by US. GGT, or gamma-glutamyl transferase, is a liver enzyme; hepatic steatosis means fat in the liver; US is ultrasound.

But the better imaging did not agree. Magnetic resonance methods, hepatic stiffness and biopsy showed no such improvement, and ultrasound is the least precise of the three.

The safety line is the one to carry. The omega-3 group was more likely to experience overall AEs compared with the control group, AEs being adverse events, and the reviewers call for careful monitoring and long-term safety data.

What the evidence says about omega-3 in cancer care

The strongest positive results on this page are in hospital patients, not in shoppers.

A 2025 review in gastrointestinal cancers found omega-3 supplementation was associated with a 24% reduction in the risk of overall complications, and reduced the length of hospitalization by roughly two and a half days.

A separate review in gastric cancer surgery reached the same territory, finding it associated with reduction of the occurrence of infectious complications and possible improvement in immune function, at moderate certainty for the infection outcome.

This is nutritional support around major surgery, given under medical supervision, in people who are often malnourished. It is a genuine clinical use and it is not a reason for a well person to take capsules.

What the evidence says about omega-3 and weight

Two different situations, opposite in what they are trying to achieve.

In cancer cachexia, the severe muscle and weight loss of advanced illness, omega-3 may be a promising agent to increase body weight, with the clearest effect in patients aged 67 or over and those starting under 60 kg.

In children and adolescents with obesity, a 2025 meta-analysis found benefit on BMI, HOMA-IR, and TG levels. Those are body mass index, a measure of insulin resistance, and TG meaning triglycerides. But there was no favorable effect on weight, BMI z-score, TC, LDL-C, HDL-C, FBS and insulin.

Note what that list contains. Weight itself did not move, and neither did the cholesterol measures. In a page about a supplement sold for heart health, that is worth reading twice.

What the evidence says about omega-3 and inflammation

The mechanism section, quantified, and it comes with a practical detail about which product to buy.

A 2026 review found that EPA+DHA doses of 1-3 g/day were associated with the most consistent reductions in CRP, TNF-α, and IL-6. All three are standard blood markers of inflammation, IL standing for interleukin.

It also found the ratio between the two long-chain fats mattered, identifying the EPA:DHA blood ratio as the strongest predictor of the downstream change.

Two things follow. The dose that moves these markers is above what many capsules contain, and the EPA and DHA figures on the back of the pack are more informative than the fish oil total on the front.

These remain laboratory readings. No section above showed a symptom improving in step with them.

What the evidence says about omega-3 in pregnancy

One of the better-supported uses, and the outcome is a real one rather than a marker.

A 2026 review found omega-3 supplementation was linked to a reduced risk of preterm birth, along with modest improvements in maternal weight gain and infant birth size.

The reviewers support its inclusion in maternal nutrition programs, which is a stronger endorsement than anything else on this page receives.

One honest caveat from their own analysis: for some outcomes, significance was lost when the reviewers re-ran it excluding weaker studies. The preterm birth finding held; several of the size measurements were less robust.

What the evidence says about omega-3 and exercise

Modest, real, and carrying one result nobody expected.

A 2025 review of omega-3 added to exercise training found it may augment some effects of exercise training on body composition and cardiometabolic health, including a small gain in lower-body muscular strength.

It also increased LDL, the harmful cholesterol, which is an odd finding for a heart supplement and one the reviewers report without explaining.

Their summary is appropriately deflating: such effects appear to be modest, and for most other outcomes omega-3 added nothing to the training itself.

Myths about omega-3, and what the evidence says

“Fish oil prevents heart attacks.” Cochrane says it slightly reduces coronary events, at moderate to low certainty. Slightly is doing a lot of work in that sentence.

“It lowers cholesterol.” It lowers triglycerides, at high certainty. Cochrane found little or no effect on other lipids, and one exercise review found LDL going up.

“More is better.” The clearest dose-dependent finding on this page besides triglycerides is the atrial fibrillation risk, which appeared at high doses and not at low ones.

“It treats depression.” For children and adolescents, Cochrane rates the evidence very uncertain. For anxiety, the signal appears only at 2 grams a day and at very low certainty.

“It protects against dementia.” No evidence here supports that. The cognition work is in people who already have impairment and moved only some test domains.

“Plant omega-3 is the same thing.” ALA converts to EPA and DHA inefficiently. Cochrane examined it separately and found a different, smaller pattern of effects.

Dosing and forms of omega-3

The doses that produced results in these reviews are generally 1 to 3 grams a day of combined EPA and DHA, which is higher than many products supply.

The label problem is worth spelling out. A capsule advertising 1000 mg of fish oil may contain only 300 mg of EPA and DHA combined. The number that matters is on the back.

Forms differ in absorption. Triglyceride and re-esterified triglyceride forms generally absorb better than the ethyl ester form used in some cheaper products, though no trial on this page tested one against another for an outcome. Prescription-strength preparations exist for high triglycerides and are a different proposition from a supplement.

Oily fish twice a week is what most dietary guidelines actually recommend, and it carries none of the dose concerns below.

Safety, side effects and who should be careful

At ordinary doses the usual complaints are fishy aftertaste, reflux and loose stools, all of which improve by taking capsules with food or freezing them.

Three cautions carry more weight.

The atrial fibrillation finding above applies to high doses in people at high cardiovascular risk. The fatty liver review reported more adverse events overall in the supplemented group. And omega-3 affects clotting, which matters for anyone on blood thinners or facing surgery.

Anyone with existing heart disease taking high-dose fish oil should be doing so on medical advice rather than their own.

Interactions

The main one is with anticoagulant and antiplatelet medication, including warfarin, aspirin and the newer blood thinners, where the effect on clotting can add up.

Blood pressure medication is a lesser consideration, since omega-3 can lower blood pressure slightly, though Cochrane found little or no effect there.

Anyone on regular medication, and anyone scheduled for an operation, should tell their doctor or pharmacist they take it.

Bottom line on omega-3

Fish oil is not a myth and not a cure, and it is the rare supplement whose best evidence is specific enough to act on.

If your triglycerides are high, this is the one supplement with a high-certainty effect on them. If you have had a heart attack, the benefit is real and slight, and above a certain dose it comes with a rhythm risk worth discussing. If you are pregnant, the preterm birth evidence is among the better findings on this site.

For everything else, from depression to memory to weight, the evidence is thinner than the shelf implies. And for most people the recommendation that survives all of it is the least commercial one: eat the fish.