Omega-3: a small heart benefit with a heart rhythm cost

Fish oil is sold as protection for the heart and the joints. Across 15 randomized trials, omega-3 supplements cut major cardiovascular events by 5 percent and raised the risk of atrial fibrillation by 25 percent, and the EU did not authorize the claim that EPA and DHA help maintain healthy joints.

How much you need

• Raw farmed Atlantic salmon averages about 0.59 g DHA and 0.32 g EPA per 3.5 oz (100 g), nearly twice wild sockeye at 0.33 g and 0.21 g, with wide spread between samples. — USDA Foundation Foods samples, 8 per fish Evidence E source
• Ground flaxseed averages about 19.4 g of plant omega-3 ALA per 3.5 oz (100 g), but samples ranged from 5.6 to 23.5 g. — USDA Foundation Foods samples, 8 Evidence E source
• Laboratory data on canned fish show EPA and DHA generally fall with canning and storage, and part of the omega-3 migrates into the covering oil. — 20 food analysis studies of canned fish Evidence D source

How well your body absorbs it

• In a 12-week RCT in 62 adults, about 250 mg a day of EPA plus DHA from Calanus, fish or krill oil raised the Omega-3 Index by a similar 1.0 to 1.15 points. — 62 healthy volunteers, mean age 29.7, 12 weeks Evidence B source
• In a crossover RCT in women with obesity, 4 g a day of DHA from fish oil raised DHA-derived oxylipins 3.8-fold in 4 weeks, while 4 g of ALA from flaxseed oil changed none. — women aged 20-51 with BMI 30-51, 4 weeks per phase Evidence B source
• In a randomized trial in 60 low-income families in Bogota, cooking with soybean oil for one month raised children's blood ALA by 0.17 percentage points versus sunflower oil. — children aged 11-18 from 60 families, 1 month Evidence B source

Who is most likely to fall short

• In 9 RCTs of 5,710 pregnant women, 450 to 800 mg a day of DHA in late pregnancy raised birth weight by about 102 g and halved low birth weight, with no clear effect on preterm birth. — 9 RCTs, 5,710 pregnant women, latter half of pregnancy Evidence A source

What the evidence does and does not show

• Across 15 RCTs, omega-3 supplements cut major cardiovascular events by 5% and heart attacks by 10%, with no change in stroke or all-cause death. — 15 RCTs of omega-3 supplements Evidence A source
• A meta-analysis of randomized trials found omega-3 supplements lowered triglycerides by about 17 mg/dL and raised HDL by about 1.6 mg/dL. — 24 studies, 2,043 subjects; TG in 21 trials (n = 1491) Evidence A source
• In older adults with depression, a meta-analysis of RCTs found omega-3 gave only a small reduction in depression scores (SMD -0.14). — older adults, 31 RCTs of anti-inflammatory interventions Evidence A source
• In 12 studies of 759 people with peripheral artery disease, EPA or EPA plus DHA did not improve walking distance, ankle-brachial index or flow-mediated dilation versus placebo. — 759 patients with peripheral arterial disease, 12 studies Evidence A source
• Observational data link higher dietary omega-3 to 18% lower odds of macular degeneration, but pooled time-to-event data show no long-term risk reduction. — adults aged 50 or older, 18 observational studies Evidence C source
• The EU did not authorize the claim that omega-3 EPA and DHA help maintain healthy joints. — EU claims register, claim POL-HC-7039 Evidence E source

What too much does

• In a meta-analysis of 35 RCTs (114,592 people), EPA plus DHA above 1,500 mg a day raised the odds of atrial fibrillation by 43% in people at high cardiovascular risk, an absolute increase of 0.8%. — 35 RCTs, 114,592 participants aged 50 or older, at least 12 months of treatment Evidence A source
• In the same 15 RCTs, omega-3 raised atrial fibrillation risk by 25% but did not increase gastrointestinal problems, bleeding disorders or cancer overall. — 15 RCTs of omega-3 supplements Evidence A source
• EFSA found no safety concern for adults taking up to 5 g a day of EPA plus DHA combined from supplements, and set no formal upper limit. — adults, supplemental EPA plus DHA Evidence E source

Medicines it interacts with

• A systematic review of 65 RCTs on supplements taken with heart drugs found only low-strength evidence that omega-3 added to statins further lowers triglycerides, and harms data were insufficient. — 65 RCTs, 2 controlled trials, 1 observational study Evidence A source

What to do with this

If you eat fish, the type matters more than the label. USDA samples of raw farmed Atlantic salmon average 0.59 g of DHA and 0.32 g of EPA per 3.5 oz (100 g), close to twice wild sockeye at 0.33 g and 0.21 g, with wide spread between the eight samples of each. Laboratory analyses of canned fish show EPA and DHA generally fall with canning and storage, and part of it moves into the covering oil, with no human data behind that.

Plant omega-3 is a different fat. Ground flaxseed averages 19.4 g of ALA per 3.5 oz (100 g), with samples from 5.6 to 23.5 g. In a crossover trial in women with obesity, 4 g a day of DHA from fish oil raised DHA-derived oxylipins 3.8-fold in four weeks, while 4 g of ALA from flaxseed oil changed none. Those are blood markers with no health outcome measured.

If you take a capsule, the source of marine oil made little difference in one trial. In 62 healthy young adults, about 250 mg a day of EPA plus DHA from Calanus, fish or krill oil raised the Omega-3 Index by a similar 1.0 to 1.15 points over 12 weeks.

Dose is where the caution sits. In trials of people at high cardiovascular risk taking more than 1,500 mg a day of EPA plus DHA, the odds of atrial fibrillation rose by 43 percent. If you are at high cardiovascular risk, take a statin or a blood thinner, or are considering a high-dose capsule, that finding is a question for the person treating you. EFSA found no safety concern for adults taking up to 5 g a day of EPA plus DHA from supplements and set no formal upper limit. That opinion covers adults only.

Where to get it from food

USDA does not report omega-3 fats for individual foods, so this site cannot rank foods by it. The sections above name the foods the studies themselves looked at.

The bottom line

Omega-3 supplements do a little for the heart and lower triglycerides by about 17 mg/dL. Atrial fibrillation rose by a quarter across 15 trials, and in a pool of 35 trials the rise was significant only above 1,500 mg a day of EPA plus DHA, in people at high cardiovascular risk. Pregnancy is where the benefit is clearest, with higher birth weight in trials of 450 to 800 mg a day of DHA in late pregnancy. Joints, walking distance in peripheral artery disease and macular degeneration are weak or unsettled cases. We have no card on supplement safety in pregnancy, in children or with blood thinners, and that silence is a gap.

Sources

1. om3r3-14 · Position of an expert body · dataset
   Fish, salmon, Atlantic, farm raised, raw: PUFA 22:6 n-3 (DHA) 0.585 g, range 0.488-0.739, 8 samples; PUFA 20:5 n-3 (EPA) 0.318 g, range 0.228-0.43, 8 samples. Fish, salmon, sockeye, wild caught, raw: DHA 0.326 g, range 0.214-0.48; EPA 0.211 g, range 0.081-0.286, 8 samples (value from the source's table, not a sentence)
   USDA FoodData Central, Foundation Foods, release 2026-04-30 (Fish, salmon, Atlantic, farm raised, raw; Fish, salmon, sockeye, wild caught, raw) · checked 2026-10-02 · independently re-checked
2. om3r3-15 · Position of an expert body · dataset
   Flaxseed, ground: PUFA 18:3 n-3 c,c,c (ALA) 19.42 g per 100 g, range 5.559-23.48, median 21.3, 8 samples (value from the source's table, not a sentence)
   USDA FoodData Central, Foundation Foods, release 2026-04-30 (Flaxseed, ground; PUFA 18:3 n-3) · checked 2026-10-02 · independently re-checked
3. om3r3-12 · Mechanistic or laboratory data · systematic review
   Five studies showed that n-3 FAs migrate from fish to the filling medium ... EPA and DHA content generally decreased due to canning and storage, with retention varying by fish species, filling medium, and sterilization method.
   Mar Drugs, 2026 · checked 2026-10-02 · independently re-checked
4. om3r3-09 · Randomized controlled trial(s) · RCT
   The post-interventional (mean ± SD) O3I increase was comparable between the three groups (CO: 1.09 ± 0.55%; FO: 1.0 ± 0.53%; KO: 1.15 ± 0.65%, all p < 0.001).
   Lipids, 2023 · checked 2026-10-02 · independently re-checked
5. om3r3-10 · Randomized controlled trial(s) · RCT
   there were higher plasma levels of nearly all oxylipins derived from DHA (3.8-fold overall; P < 0.001) and EPA (2.7-fold overall; P < 0.05) after 28 days of fish-oil supplementation, while there were no changes to oxylipins after flaxseed-oil supplementation.
   J Nutr, 2021 · checked 2026-10-02 · independently re-checked
6. om3r3-11 · Randomized controlled trial(s) · RCT
   Soyabean oil significantly increased ALA concentrations by 0.05 percentage points of total serum fatty acids whereas sunflower oil decreased them by 0.12 percentage points (soyabean v. sunflower oil effect=0.17; 95% CI 0.11, 0.24).
   Public Health Nutr, 2015 · checked 2026-10-02 · independently re-checked
7. om3r3-05 · Meta-analysis or systematic review · meta-analysis of RCTs
   nine RCTs involving 5710 subjects were included. This meta-analysis showed that DHA supplementation (450-800 mg/day) was associated with a significantly higher birth weight of infants (Inverse variance [IV]: 101.71 [57.36-146.06] at 95% CI, P = .00001, I2 = 0%), and fewer low birth weights (LBWs) (Mantel-Haenszel [M-H]: 0.53 [0.33-0.86] at 95% CI, P = .01, I2 = 72%), with lesser but statistically insignificant pre-term births (PTB)
   Nutr Res, 2024 · checked 2026-10-02 · independently re-checked
8. om3r3-02 · Meta-analysis or systematic review · meta-analysis of RCTs
   The incidence of major cardiovascular events (RR 0.95, 95%CI 0.91 to 0.99, P = 0.026), myocardial infarction (RR 0.90, 95%CI 0.83 to 0.98; P = 0.021), and cardiovascular death (RR 0.94, 95%CI 0.88 to 0.99; P = 0.028) was reduced in the omega-3 fatty acid group compared with the control group.
   Cardiovasc Drugs Ther, 2024 · checked 2026-10-02 · independently re-checked
9. om3r3-04 · Meta-analysis or systematic review · meta-analysis of RCTs
   There were 24 studies with 2043 subjects showed significant effects: (1)TG decreased by 16.95 mg/dl (21 trials, n = 1491; 95 % CI: 23.25, -10.66), (2)HDL increased by 1.55 mg/dl (22 trials, n = 1914; 95 % CI: 0.69, 2.42)
   Nutr Metab Cardiovasc Dis, 2026 · checked 2026-10-02 · independently re-checked
10. om3r3-06 · Meta-analysis or systematic review · meta-analysis of RCTs
    Sub-group analyses supported the use of omega-3 fatty acids (SMD = -0.14, 95% CI = -0.27 to -0.02, p = 0.03)
    Transl Psychiatry, 2025 · checked 2026-10-02 · independently re-checked
11. om3r3-07 · Meta-analysis or systematic review · meta-analysis
    Of 1067 citations, 12 studies (n = 759 patients) met the predefined inclusion criteria. Supplementation of EPA, or EPA + DHA did not alter the primary outcome measures of pain-free walking distance, maximal walking distance; ankle brachial index, or flow mediated vasodilation versus placebo.
    Nutr Metab Cardiovasc Dis, 2026 · checked 2026-10-02 · independently re-checked
12. om3r3-08 · Observational data · meta-analysis of observational studies
    Higher ω-3 intake was associated with reduced odds of AMD overall (odds ratio [OR] 0.82, 95% confidence interval [CI] 0.70-0.96; p = 0.01) ... Time-to-event pooling showed no significant long-term risk reduction (hazard ratio [HR] 0.98, 95% CI 0.84-1.15; p = 0.83).
    J Nutr, 2026 · checked 2026-10-02 · independently re-checked
13. om3r3-16 · Position of an expert body · regulatory decision
    POL-HC-7039 EPA and DHA Omega-3 fatty acids Omega-3 EPA and DHA help maintain healthy joints Non-authorised (value from the source's table, not a sentence)
    EU Register on nutrition and health claims, claim POL-HC-7039, snapshot 2026-09-21 · checked 2026-10-02 · independently re-checked
14. om3r3-01 · Meta-analysis or systematic review · meta-analysis of RCTs
    A total of 35 randomized controlled trials (37 data sets; n=114 592) were included in this meta-analysis. Only studies including patients at high-risk for cardiovascular disease who were treated with high-doses of EPA/DHA (>1500 mg/d) showed a statistically significant increase in AF risk with a pooled odds ratio (OR) of 1.43 (95% CI, 1.14-1.79) and an absolute risk difference of 0.8% (0.40%-1.1%).
    Circ Arrhythm Electrophysiol, 2026 · checked 2026-10-02 · independently re-checked
15. om3r3-03 · Meta-analysis or systematic review · meta-analysis of RCTs
    An increased risk of atrial fibrillation (RR 1.25, 95%CI 1.10 to 1.41; P = 0.000) was observed in patients in the omega-3 fatty acid group. ... For safety endpoints, there were no statistically significant differences between the two groups in gastrointestinal problems, bleeding-related disorders, and cancer.
    Cardiovasc Drugs Ther, 2024 · checked 2026-10-02 · independently re-checked
16. om3r6-17 · Position of an expert body · agency opinion
    supplemental intakes of EPA and DHA combined at doses up to 5g a day do not raise safety concerns for adults
    EFSA, Scientific Opinion on the Tolerable Upper Intake Level of EPA, DHA and DPA, press release 27 July 2012 · checked 2026-10-07 · independently re-checked
17. om3r3-13 · Meta-analysis or systematic review · systematic review
    Evidence was obtained from 65 randomized clinical trials, 2 controlled clinical trials and 1 observational study. With only a few small studies available per supplement, evidence was insufficient for all predefined gradable clinical efficacy and harms outcomes ... Incremental benefits were noted for triglyceridemia with omega-3 fatty acid added to statins
    Syst Rev, 2012 · checked 2026-10-02 · independently re-checked