Category: The Evidence
Fish oil reliably lowers triglycerides, that part is settled. Whether it protects your heart depends on dose, on EPA versus EPA plus DHA, and on who you are. We walk through the four trials that disagree.
Category: Research Reviews | Reading time: ~12 min | Level: Intermediate
Between 2018 and 2020, four large randomised trials set out to answer more or less the same question: does fish oil protect the heart? All four were well designed, well funded, and published in major journals. They did not agree. One found no meaningful benefit in generally healthy adults. Another found no benefit in adults with diabetes. A third, using a different, purified form of the supplement at a higher dose, found a large reduction in cardiovascular events. A fourth, at a similarly high dose but a different formula, found no benefit at all, and was stopped early after turning up more cases of an irregular heart rhythm in the treated group than the placebo group.
That is not a story about bad science. It is a story about a supplement that behaves differently depending on three things that most bottles on a shelf never mention: the dose, whether the product is EPA alone or a combined EPA and DHA formula, and who is taking it. Fish oil is not one intervention with one answer. It is closer to four separate experiments that happened to use the same broad category of ingredient, and reading only one of them, which is what most marketing does, tells you a fraction of the truth.
There is a part of this story that is genuinely settled. Fish oil lowers triglycerides, a type of blood fat distinct from cholesterol, and it does so in a clear, dose-dependent way that has been confirmed across dozens of trials. That is real, useful, well-established evidence. Whether fish oil also protects your heart from a heart attack or stroke is a separate question, and it is the one this review spends most of its time on, trial by trial, dose by dose, because the honest answer depends entirely on which of those four studies looks most like your own situation.
The short answer: fish oil's triglyceride-lowering effect is strong and dose-dependent, confirmed by a Cochrane pooling of 86 trials and almost 163,000 people showing roughly a 15% reduction, though the doses that achieve this, 2 to 4g a day, sit well above a typical wellness capsule [1]. Fish oil does not lower LDL cholesterol, and high doses can raise it slightly, correcting a common assumption [1]. Cardiovascular protection is moderate and population-specific: the four landmark outcome trials disagree, with two finding no meaningful benefit in generally healthy or diabetic adults, one finding a large benefit using a purified, EPA-only prescription drug in high-risk patients, and one finding no benefit and more atrial fibrillation at a high combined-formula dose [2]. A separate meta-analysis of 149,051 participants found EPA-only supplementation outperformed combined EPA and DHA [3], while another covering 81,210 patients found a dose-graded increase in atrial fibrillation risk [4]. The benefit, where real, concentrates in people with existing heart disease, high triglycerides, or low baseline fish intake.
Fish oil supplies two long-chain omega-3 fatty acids, EPA and DHA, that the body uses only in small amounts from the plant-derived omega-3 ALA, found in flaxseed and walnuts, since human conversion of ALA into EPA and DHA is inefficient. EPA and DHA are incorporated into cell membranes throughout the body, including in the heart and blood vessels, where they influence inflammation, blood clotting, and triglyceride metabolism. The best-documented downstream effect of this is a reduction in triglyceride production by the liver, which is the mechanism behind fish oil's clearest, most reproducible clinical effect [1]. Everything else, the cardiovascular outcome question in particular, is a longer, less direct chain of effects that has proven much harder to confirm consistently in large human trials, which is precisely why the outcome trials tell such a mixed story.
Start with the part of the fish oil story that does not require hedging. A Cochrane review pooling 86 randomised controlled trials and 162,796 participants found that omega-3 supplementation lowers triglycerides by roughly 15% on average, with the effect scaling up at higher doses, and the American Heart Association recognises a prescription dose of 4g a day as an established triglyceride-lowering treatment [1]. This is the single most consistent finding in the entire omega-3 literature, replicated across a huge combined sample, and it is not seriously disputed.
What is worth correcting is a common assumption that rides alongside it: that fish oil also lowers cholesterol generally. It does not lower LDL cholesterol, the marker most routine blood tests report as "bad cholesterol," and at higher doses omega-3 supplementation can nudge LDL up slightly [1]. Triglycerides and LDL cholesterol are different blood fats with different biology, and fish oil's real, strong effect on one should not be mistaken for an effect on the other. This is also why the American Heart Association does not recommend routine omega-3 supplementation for people who are not already at elevated cardiovascular risk: a real effect on one blood marker is not the same as a documented, broad protective benefit for everyone [1].
This is the heart of the honest story, and it deserves to be walked through trial by trial rather than summarised away. VITAL enrolled 25,871 generally healthy adults on 1g a day and found no significant reduction in combined major cardiovascular events, although total heart attacks fell by around 28%, a secondary finding worth noting but not the trial's primary result [2]. ASCEND enrolled 15,480 adults with diabetes, also at 1g a day, and found no significant reduction in serious vascular events [2]. Both trials used relatively modest, over-the-counter-comparable doses in populations without established heart disease, which is close to how most people actually take fish oil, and both came back essentially negative on their main outcome.
REDUCE-IT told a different story. It enrolled 8,179 higher-risk patients, most with existing cardiovascular disease or diabetes plus other risk factors, and used a much higher dose, 4g a day, of a purified, EPA-only prescription drug called icosapent ethyl. Cardiovascular events fell by around 25%, a genuinely large effect [2]. But REDUCE-IT carries a real, widely discussed limitation: its placebo was mineral oil, and some researchers argue mineral oil may have mildly worsened cholesterol markers in the comparison group, which could inflate how large the apparent benefit looks. This is a recognised, unresolved debate in cardiology, not a footnote to dismiss the trial, but a genuine reason to hold the 25% figure with some caution [2].
STRENGTH then tested a similarly high dose, 4g a day, but of combined EPA and DHA rather than EPA alone, in 13,078 high-risk patients. It found no cardiovascular benefit at all, was stopped early for futility, and recorded more cases of atrial fibrillation in the treated group than in the comparison group [2]. Put side by side, REDUCE-IT and STRENGTH used comparable high doses in comparable high-risk populations and reached opposite conclusions, and the most-discussed explanation for the gap is the formula difference: EPA alone versus EPA combined with DHA.
That explanation gets independent support from a separate line of evidence. A 2021 systematic review and meta-analysis of 38 randomised controlled trials covering 149,051 participants found that omega-3 supplementation overall reduced cardiovascular mortality, with a relative risk of 0.93, and non-fatal heart attacks, and that EPA-only supplementation outperformed combined EPA and DHA formulas specifically [3]. Four large trials disagreeing is unsettling if you want a single clean answer. Four large trials disagreeing in a pattern that a separate, much larger pooled analysis can partly explain is closer to how real cardiovascular research usually looks: messy, but not directionless.
If you have established heart disease, very high triglycerides, or other high cardiovascular risk factors, the case for a properly dosed, medically supervised omega-3 regimen, and specifically the question of whether an EPA-only prescription option is right for you, is worth a real conversation with a doctor, because that is the population in which the strongest trial evidence, REDUCE-IT and the EPA-focused meta-analysis, actually applies [2][3]. If you are a generally healthy adult without those risk factors, the honest picture from VITAL and ASCEND, and from the American Heart Association's own position, is that routine fish oil supplementation has not been shown to meaningfully lower your cardiovascular event risk, whatever a bottle's front label implies [1][2].
If you are taking fish oil mainly for triglycerides, know that the doses shown to move that number meaningfully, 2 to 4g of combined EPA and DHA a day, are well above what a standard low-dose capsule provides, and this is a conversation for a healthcare professional rather than a self-directed high-dose trial. And if you are taking a low-dose daily capsule mostly as a general wellness habit, a reasonable, low-risk one, be honest with yourself that it is unlikely to be delivering either the strong triglyceride effect or the population-specific cardiovascular benefit documented in the trials above, since both cluster at doses well beyond what that capsule contains. Fish oil for everyday focus or cognition, incidentally, is the weakest of the common claims: the NIH Office of Dietary Supplements fact sheet describes generally null results for cognitive outcomes in healthy adults, so that particular marketing claim should be weighted accordingly [1].
Fish oil is well tolerated at low, everyday doses, with the most common side effects being mild, fishy aftertaste, burping, or occasional stomach upset. The safety picture changes at the higher, prescription-level doses used in the outcome trials above. A meta-analysis of 7 trials covering 81,210 patients found marine omega-3 supplementation was associated with a 25% relative increase in atrial fibrillation risk, and this risk was dose-graded, rising further above 1g a day [4]. STRENGTH independently recorded more atrial fibrillation cases in its high-dose treatment arm [2]. High-dose fish oil can also have a mild blood-thinning effect, which is a genuine consideration for anyone already on anticoagulant or antiplatelet medication, or scheduled for surgery. None of these signals are strongly established at the low doses found in a typical daily wellness capsule, but anyone with existing atrial fibrillation, a bleeding disorder, or on blood-thinning medication should treat a high-dose fish oil regimen as a decision to make with a doctor, not a supermarket purchase to make alone.
Pregnant, breastfeeding, or on medication? Check with a healthcare professional first.
Fish oil is a good test of whether a supplement story can survive contact with its own best evidence, and this one mostly does, once the dose and the population are put back into the sentence. The triglyceride effect is real and strong. The cardiovascular protection claim is real too, but only for some people, at doses most capsules do not contain, and the four biggest trials on the subject genuinely disagree with each other. We would rather say that plainly than round it up to "supports heart health" and leave the nuance out. Wherever omega-3 appears in the Remedy Library, the grade reflects this split: strong for triglycerides, moderate and population-specific for cardiovascular outcomes, with the dose and formula caveats attached rather than smoothed over.
If you want to see how your own risk profile lines up with which of these four trials actually applies to you, Remy can walk through the dose and population differences in plain language, and the Remedy Library keeps the same honest, split grading in one place rather than a single reassuring headline.
Four major trials, four different results, is not a failure of the science. It is what happens when researchers finally ask a specific enough question of a supplement that had spent years being sold as a single, universal answer. The specific questions, EPA or EPA plus DHA, what dose, which patients, are exactly the ones a bottle on a shelf was never built to answer, and they turn out to matter more than whether you take fish oil at all.
1. NIH Office of Dietary Supplements. Omega-3 Fatty Acids: Fact Sheet for Health Professionals. 2023. Institutional synthesis citing a Cochrane pooling of 86 RCTs (162,796 participants) showing a dose-dependent triglyceride reduction of roughly 15%, an AHA-recognised 4g/day prescription dose for triglyceride lowering, the absence of an LDL-lowering effect (with a possible slight LDL increase at high doses), the AHA position against routine supplementation in low-risk adults, and generally null results for cognition in healthy adults. 2. NIH Office of Dietary Supplements. Omega-3 Fatty Acids Fact Sheet: VITAL, ASCEND, REDUCE-IT and STRENGTH trial summaries. 2023. Institutional synthesis of the four landmark outcome trials: VITAL (n=25,871, primary prevention, 1g/day, no significant reduction in combined major cardiovascular events, roughly 28% reduction in total myocardial infarction); ASCEND (n=15,480, diabetes, 1g/day, no significant reduction in serious vascular events); REDUCE-IT (n=8,179, high risk, EPA-only icosapent ethyl 4g/day, roughly 25% reduction in cardiovascular events, mineral oil placebo debated as a confounder); STRENGTH (n=13,078, EPA+DHA 4g/day, no cardiovascular benefit, stopped early, increased atrial fibrillation). 3. Khan SU, Lone AN, Khan MS, et al. Effect of omega-3 fatty acids on cardiovascular outcomes: a systematic review and meta-analysis. EClinicalMedicine. 2021. Pooled 38 RCTs (149,051 participants); found reduced cardiovascular mortality (RR 0.93) and non-fatal myocardial infarction, with EPA-only supplementation outperforming combined EPA and DHA formulas. 4. Gencer B, Djousse L, Al-Ramady OT, et al. Effect of long-term marine omega-3 fatty acid supplementation on the risk of atrial fibrillation: a meta-analysis of randomized controlled trials. Circulation. 2021. Pooled 7 RCTs (81,210 patients); found marine omega-3 supplementation associated with increased atrial fibrillation risk (HR 1.25), with risk rising further above 1g/day.
It depends who you are and how much you take. Fish oil reliably lowers triglycerides, which is well established. Protection against heart attacks and cardiovascular death is more mixed: two large trials in generally healthy or diabetic adults found no meaningful benefit, while a trial in high-risk patients using a purified EPA drug found a large reduction in events. The honest summary is moderate, population-specific evidence, strongest in people who already have heart disease or high triglycerides.
Not LDL cholesterol, the type usually meant by "cholesterol" in a routine blood test. Omega-3s do not lower LDL, and at high doses can raise it slightly. What fish oil reliably lowers is triglycerides, a different type of blood fat. Confusing the two is one of the most common misunderstandings about fish oil supplements.
A large meta-analysis of 38 trials covering over 149,000 participants found that EPA taken alone was associated with a bigger reduction in cardiovascular mortality and non-fatal heart attacks than combined EPA and DHA formulas. This is one reason a purified, high-dose EPA prescription drug produced the strongest results among the four landmark outcome trials, while a high-dose combined EPA and DHA trial found no benefit at all.
There is a real, dose-graded signal. A meta-analysis of 7 trials covering over 81,000 patients found marine omega-3 supplementation was associated with a 25% relative increase in atrial fibrillation risk, with the risk rising further above 1g a day. This signal appears mainly at the higher, prescription-level doses used in cardiovascular trials, not necessarily at typical low-dose wellness capsules, but it is a real consideration, especially for anyone with existing heart rhythm issues.
The triglyceride-lowering effect is dose-dependent, and the doses shown to meaningfully move triglycerides in trials and prescription use sit around 2 to 4g of combined EPA and DHA a day, well above what a typical over-the-counter capsule provides in a single dose. Anyone considering fish oil specifically for high triglycerides should discuss a proper dose with a healthcare professional rather than assuming a standard capsule is doing the job.
This is the weakest of the common fish oil claims. According to the synthesis from the NIH Office of Dietary Supplements, omega-3 supplementation has generally shown null results for cognition and focus in healthy adults. The DHA-and-brain-development story in infants and the omega-3-and-mood research area are separate questions with their own, distinct evidence bases, but "fish oil for everyday focus" in a healthy adult is not well supported.
Match the evidence to your own risk profile rather than looking for one universal answer. If you have high triglycerides or existing heart disease, the case for a proper, medically supervised dose is genuinely stronger. If you are a generally healthy adult without those risk factors, the American Heart Association itself does not recommend routine omega-3 supplementation, and a low-dose daily capsule is closer to a reasonable, low-risk habit than a proven protective measure.
Two differences likely matter: REDUCE-IT used a purified, EPA-only prescription drug, while STRENGTH used a combined EPA and DHA formula, and the EPA-outperforms-EPA-plus-DHA pattern shows up elsewhere in the literature too. REDUCE-IT also compared against a mineral oil placebo, which some researchers argue may have mildly worsened the comparison group's cholesterol markers, inflating the apparent benefit. Both explanations are debated, which is exactly why one striking trial should not be read as the final word.
Probably not much, honestly. Most of the meaningful effects in this review, both the triglyceride benefit and the atrial fibrillation risk, cluster around doses of 2 to 4g a day, well above the roughly 300mg to 1g typically found in an everyday capsule. A low daily dose is unlikely to cause the atrial fibrillation signal, but by the same logic it is unlikely to deliver much of the studied cardiovascular benefit either.