“Omega-3” is not one intervention—and the trial results do not transfer

The bottom line

VITAL's one-gram daily EPA+DHA product did not significantly reduce its primary major-cardiovascular-event endpoint: 386 versus 419 events, hazard ratio 0.92. OMEMI's 1.8-gram EPA+DHA product was also neutral after myocardial infarction: 108 of 505 participants (21.4%) versus 102 of 509 (20.0%), hazard ratio 1.08.

REDUCE-IT was different. Four grams daily of prescription icosapent ethyl reduced its primary composite in statin-treated high-risk adults with elevated triglycerides: 705 of 4,089 (17.2%) versus 901 of 4,090 (22.0%), hazard ratio 0.75—an absolute difference of 4.8 percentage points. That result cannot be reassigned to dietary fish or an over-the-counter EPA+DHA capsule.

Why this matters

“Omega-3,” “fish oil” and “EPA” are often treated as synonyms. They are not. Trials can differ in EPA versus DHA content, purity, amount, comparator, background statin use, triglyceride levels, prior disease and primary endpoint. A positive result for one prescription product in one selected population does not validate every product sold under a broad nutrient label.

What researchers did

VITAL randomized 25,871 generally healthy US adults in primary prevention to one gram daily of marine EPA+DHA or placebo. OMEMI randomized adults aged 70–82 who had recently experienced myocardial infarction to 1.8 grams daily of EPA+DHA or corn-oil placebo. REDUCE-IT randomized statin-treated adults with triglycerides of 135–499 mg/dL and established cardiovascular disease or diabetes plus risk factors to four grams daily of prescription icosapent ethyl or mineral-oil placebo.

Those are different clinical questions. VITAL asked about broad primary prevention; OMEMI asked about older post-heart-attack patients; REDUCE-IT asked whether a prescription EPA-only product added benefit in selected high-risk statin users.

What they found

VITAL was neutral for its primary cardiovascular composite. Myocardial infarction was lower—145 versus 200, hazard ratio 0.72—but it was a secondary component without multiplicity adjustment. Stroke (148 versus 142) and cardiovascular mortality (142 versus 148) were neutral. (VITAL)

OMEMI's primary composite was neutral. New atrial fibrillation occurred in 28 participants (7.2%) versus 15 (4.0%), hazard ratio 1.84 with p=0.06: an imprecise safety signal, not a definitive result. Major bleeding was similar, 54 versus 56. The trial recruited fewer than its planned 1,400 participants, and its final composite included heart-failure hospitalization absent from the original primary-outcome description. (OMEMI)

REDUCE-IT reduced its primary and key secondary composites; the key secondary endpoint was 11.2% versus 14.8%, hazard ratio 0.74. All-cause mortality was not significantly lower. Atrial-fibrillation or flutter hospitalization was 3.1% versus 2.1%, and serious bleeding was 2.7% versus 2.1%. (REDUCE-IT)

How strong is the evidence?

All three studies were randomized, but each directly supports only its own population, product, comparator and endpoint. VITAL's large primary-prevention sample strongly supports its neutral primary result. OMEMI's smaller-than-planned sample and endpoint history reduce precision. REDUCE-IT provides strong evidence for prescription icosapent ethyl in its selected high-risk population.

REDUCE-IT used mineral oil as placebo. Mineral oil may have worsened some biomarkers, which could inflate part of the estimated treatment effect. That is a real limitation; it does not prove that comparator harm explains the entire result. The most defensible conclusion keeps both facts visible.

What this evidence does not show

The trials do not show that eating more fish, taking any retail fish-oil product, combining EPA and DHA, or using prescription icosapent ethyl produces the same effect. They do not recommend a product, brand or dose for an individual.

None tested biological-aging rate, healthspan or lifespan extension. A lower cardiovascular composite in a high-risk population is a disease-outcome result—not proof of a general anti-aging effect.

Safety and conflicts

Atrial fibrillation and bleeding must remain visible in interpretation. OMEMI's atrial-fibrillation finding was imprecise; REDUCE-IT and the current prescription label report atrial-fibrillation/flutter and bleeding concerns, especially relevant alongside antithrombotic medicines. This reporting is not a personal safety assessment.

VITAL received public funding and donated products. OMEMI was foundation-funded and Orkla supplied product and placebo. REDUCE-IT was funded by Amarin, the prescription product's manufacturer, and investigators disclosed industry relationships. Product quality and composition can vary outside a prescription trial.

What happens next

Future evidence should clarify which observed REDUCE-IT effects are specific to purified EPA, its high-risk population and its comparator. Independent replication with well-characterized comparators, consistent endpoint reporting and transparent product testing would strengthen translation. Readers should expect every new claim to name the exact product, population and outcome.