Fish oil safety: side effects, interactions and who should skip it

What the safety evidence covers, and what it doesn't

For most healthy adults taking a labeled amount, fish oil has a reassuring record. The largest direct look at tolerability is a 2018 meta-analysis of 21 trials and 24,460 people, which found no clear evidence of serious treatment-related adverse events from prescription-strength omega-3 [PMID 29482765]. It also concluded that these products are generally well tolerated but not free of side effects.

The caveat is the dose. Those trials used prescription products, well above what a typical over-the-counter softgel delivers. A lower dose could plausibly cause fewer or milder effects, but the review doesn't establish that directly.

Duration is a limit too. In a five-month trial of 115 healthy adults taking 300 to 1,800 mg a day of combined EPA and DHA, the omega-3 index rose in a dose-dependent way [PMID 24252845]. That trial tracked a blood marker, not long-term health outcomes, so it says little about years of continuous use.

For reference, the FDA advises against more than 2,000 mg a day of combined EPA and DHA from supplements, or 3,000 mg a day from all sources. Our fish oil topic page collects the rest of our coverage.

Common side effects and how people ease them

In the 2018 pooled analysis, fishy aftertaste and skin reactions (rash, itching, eczema) were more common than with placebo. Products combining EPA and DHA specifically caused more burping and nausea [PMID 29482765]. The NIH Office of Dietary Supplements also lists heartburn, stomach discomfort, diarrhea, headache and bad breath among the mild effects people report.

Diarrhea deserves one honest note: it appears on the NIH list, but the pooled trial did not isolate it as a separate outcome. We found no trial evidence tying fish oil to constipation.

The most common practical tip is to take it with a meal. There is a supporting rationale: in a small 1988 trial, a high-fat meal raised EPA absorption from the natural triglyceride form from 69% to 90%, and roughly tripled absorption of EPA and DHA from the ethyl ester form [PMID 2847723]. A separate single-dose trial without food found the ethyl ester form absorbed far less well than free fatty acids [PMID 3358766]. Better absorption does not prove fewer side effects, though, and we found no controlled trial testing meal timing against burping directly.

• Take it with food rather than on an empty stomach.
• Split a larger daily amount across two meals.
• If aftertaste or reflux persists, lower the dose or stop and ask a clinician rather than pushing through.

Blood thinners, aspirin and surgery

The best-documented interaction is with medications that affect clotting. A review of the clinical and basic science literature found fish oil has a weak blood-thinning effect on its own. The effect grows more noticeable alongside antiplatelet drugs and, to a lesser degree, warfarin or factor Xa inhibitors [PMID 29627068].

The same review found that fish oil alone was not linked to a meaningful increase in surgical bleeding risk. So the concern is mostly about combinations. Aspirin is itself an antiplatelet drug, so regular aspirin use with fish oil is worth mentioning to a doctor or pharmacist, especially if you also take a prescription anticoagulant. Our guide to fish oil and medications walks through the same checks.

Some surgeons ask patients to pause fish oil before a procedure. Evidence for a specific timeline is thin, so ask your surgeon or anesthesiologist rather than guessing.

Dose matters: heart rhythm and other signals

The more serious dose-related signal involves the heart. A 2021 meta-analysis of seven large cardiovascular trials (81,210 people) linked marine omega-3 supplementation to a higher risk of atrial fibrillation. The risk was greater in trials testing more than 1 gram a day than in those testing 1 gram or less [PMID 34612056].

A trial of a 4 g/day prescription EPA drug in 8,179 people already treated for cardiovascular disease or diabetes showed the same tension. It lowered a composite of cardiovascular events from 22.0% to 17.2%, but increased hospitalization for atrial fibrillation [PMID 30415628]. That trial involved a prescription product in a high-risk group, so it does not translate directly to a daily softgel.

A few smaller signals are worth knowing, with their limits stated:

• LDL cholesterol. A Cochrane review of 23 trials in type 2 diabetes (mean 3.5 g/day) found omega-3 lowered triglycerides but raised LDL slightly, with no effect on blood sugar [PMID 18254017].
• Prostate cancer. One observational study found higher prostate cancer risk in men with the highest blood levels of long-chain omega-3s [PMID 23843441]. It measured blood levels, not supplement use, and cannot show cause. A randomized trial of 1 g/day in 25,871 adults found no difference in invasive cancer diagnoses [PMID 30415637].
• H. pylori treatment. In an observational study of 4,014 people, higher dietary omega-3 and fish intake was associated with lower success of standard antibiotic treatment for H. pylori [PMID 28659302]. It is an association, not a reason to change anything on your own.

Who should check with a doctor first

A few situations call for a conversation before starting, or for skipping standard fish oil altogether.

Kidney and liver disease come up often. We found no trial evidence that label-dose fish oil harms people with normal kidney or liver function. For anyone with existing kidney or liver disease, the sensible default is to run any new supplement past a doctor. That is general caution, not a specific finding about fish oil. Our longer piece on who should not take fish oil covers these groups in more detail.

Pregnancy and breastfeeding

Pregnancy is a case where the evidence cuts both ways, which is why it belongs in a clinician conversation. A 2018 Cochrane review of 70 trials (19,927 pregnant women) found high-quality evidence that omega-3 supplementation lowered preterm birth (13.4% to 11.9%) and early preterm birth (4.6% to 2.7%). It also found a probable small increase in pregnancies continuing past term [PMID 30480773].

That is not a dosing instruction. It suggests that the right amount, product and timing are individual decisions best made with an obstetric provider. The same holds for breastfeeding, where we found no source in this review that gives a specific safe figure.

Warning signs to stop and get help

These are general safety cues rather than findings from the studies above, but they are sensible ones.

• Signs of an allergic reaction: hives, facial or throat swelling, or trouble breathing. Stop and seek emergency care.
• Unusual bleeding: frequent or hard-to-stop nosebleeds, easy bruising that is new for you, blood in urine, or black stools, particularly if you also take a blood thinner.
• New palpitations or an irregular heartbeat, especially at higher doses, given the atrial fibrillation data.
• Persistent digestive symptoms: ongoing diarrhea, severe reflux or nausea that does not settle with food or a lower dose.
• New rash or itching after starting, since skin reactions were reported more often than with placebo [PMID 29482765].

If any of these appear, stop the supplement and contact a clinician. Bring the product label so the actual dose of EPA and DHA is clear. For most people taking a modest amount with meals, the more likely outcome is a bit of burping and a fishy taste.