Every zero-result search is logged — it is how we decide what to curate next.
Also: EPA · DHA · eicosapentaenoic acid · docosahexaenoic acid · marine n-3 fatty acids · omega 3
Four separate measurements, never averaged. Independent evidence counts only studies with no sponsor attached. Methodology version v2.0-2026-10 · how we grade evidence
Claimed to reduce cardiovascular events and mortality, lower triglycerides, reduce inflammation and improve body composition, mood, joint pain and cognition. Triglyceride lowering is consistent and dose-dependent across trials and is the basis of the prescription indications. Cardiovascular event reduction from supplement doses is not supported: VITAL (1 g/day, n=25,871), ASCEND (1 g/day, n=15,480, mean follow-up 7.4 years in people with diabetes) and STRENGTH (4 g/day EPA+DHA, n=13,078, stopped early for futility) were all null.
REDUCE-IT is not a supplement trial. It tested icosapent ethyl, a prescription highly purified EPA ethyl ester at 4 g/day, against a mineral-oil comparator, and was funded solely by its manufacturer. Reporting it as evidence for fish-oil capsules is, by this report’s account, the single most common error in the category.
There is a quantified harm signal running the other way. A meta-analysis of seven cardiovascular-outcome trials and 81,210 participants found incident atrial fibrillation HR 1.25 (95% CI 1.07–1.46), dose-dependently, and REDUCE-IT itself recorded more hospitalisation for atrial fibrillation or flutter (3.1% vs 2.1%) and more serious bleeding (2.7% vs 2.1%). STRENGTH recorded gastrointestinal adverse events in 24.7% versus 14.7% on corn oil. VITAL and ASCEND reported no excess bleeding or serious adverse events at 1 g/day. Net: the category moved from "protective" to "null with a harm signal".
At a glance
Incorporation into membrane phospholipids alters fluidity and signalling; EPA and DHA are substrates for 3-series eicosanoids and for the specialised pro-resolving mediators (resolvins, protectins); and both suppress hepatic VLDL-triglyceride synthesis and secretion. The hepatic triglyceride mechanism is confirmed in humans, consistently and dose-dependently. The anti-atherosclerotic extension of it is not: three large independent trials found no event reduction at supplement doses, so the inflammation-resolution account remains a hypothesis about mechanism rather than an explanation of a demonstrated clinical effect.
Dated determinations, each flagged for whether it is final and whether it covers safety only. A safety review is not an efficacy finding, and most reporting in this category conflates the two.
Specific GRAS notice numbers were not verified in this pass and are therefore not recorded. GRAS is a safety determination and says nothing about efficacy. Note separately that icosapent ethyl is a prescription drug, not a supplement — the REDUCE-IT result belongs to the drug, not to this regulatory status.
Announced 19 June 2019 for conventional foods and dietary supplements. These are *qualified* claims, which is the whole point: each must carry the statement that "FDA has concluded that the evidence is inconsistent and inconclusive", must declare the grams of EPA and DHA per serving, and requires a minimum of 0.8 g combined EPA+DHA per serving. A qualified claim is a regulator saying the evidence is weak while permitting the words anyway.
The existence of authorised EPA/DHA claims is recorded, but the exact wording and conditions were not verified in this pass and are deliberately not reproduced here.
Educational information only
Educational information only. Not medical advice, not a diagnosis, and not a substitute for your clinician. Never start, stop, or change a prescription medication based on anything you read here. We do not sell, prescribe, dispense or recommend any medication, peptide or supplement, and we do not link to anyone who does. If you are having a medical emergency, call 911.
Published by Nu Link, 10111 Clodine Rd, Richmond, TX 77407.