Direct answer
Prescription omega-3 can lower high triglycerides; selected high-risk patients benefited from prescription EPA. Ordinary fish-oil products do not share that outcome evidence, and funding, formulation and atrial-fibrillation risks matter.
- Strongest evidence: modest triglyceride reduction (around 20–30% with 4 g/day prescription omega-3 in people with elevated triglycerides) and reduced cardiovascular events in secondary prevention at prescription doses — FDA has approved icosapent ethyl (EPA-only, 4 g) after the REDUCE-IT trial. (Primary source)
- For primary prevention of heart disease in the general population, effects are modest at best — a large Cochrane review of 79 trials (112,059 people) found little to no cardiovascular mortality benefit across varied interventions and populations; it was not restricted to OTC-dose supplements. (Primary source)
- Absorption depends on formulation, dose and meals; no universal fish-oil/krill hierarchy is established. A matched-dose four-week trial found similar plasma and red-cell EPA+DHA levels. Algal products can supply preformed DHA and sometimes EPA; flaxseed supplies ALA with limited conversion. (Matched-dose bioavailability trial)
- Weak evidence for depression as a standalone treatment (higher-EPA formulas 1–2 g/day showed some effect in RCTs but heterogeneity is high), dry eye, and cognitive decline; strong evidence against for slowing progression to advanced macular degeneration in the high-risk AREDS2 population. (Primary source)
- Bleeding warnings matter particularly with high-dose prescription EPA and antithrombotic drugs; a universal risk threshold at 3 g/day is not established, especially with anticoagulants — flag with your prescriber before combining with warfarin, DOACs, or antiplatelets.
- Evidence grade: Contested
Ingredient deep dive · Heart & Cardiovascular
Omega-3 fatty acids work best for lowering high triglycerides, but the evidence is mixed for preventing heart attacks or strokes with ordinary over-the-counter fish oil. EPA/DHA from fish, algal oil, or prescription products are the active long-chain forms; flaxseed ALA is essential but converts poorly to EPA/DHA, usually below 15% according to NIH ODS. People on anticoagulants, antiplatelets, with atrial fibrillation, fish allergy, liver disease, planned surgery, pregnancy, or high-dose supplement use should treat omega-3 as a medication-like ingredient and review it with a clinician.
Table of contents
- Evidence summary
- What omega-3 fatty acids are
- All forms and types
- EPA-only vs DHA-only vs EPA+DHA
- How omega-3 works
- What works and what does not
- Benefits by claim
- Risks and reported side effects
- Interactions and medication precautions
- Who should avoid omega-3 supplements
- Dosage and how to take
- FAQ
- Sources & funding notes
Evidence summary
| Claim | Evidence | Source country | Funding / conflicts traced | Strength |
|---|---|---|---|---|
| Prescription-strength EPA/DHA lowers triglycerides | AHA advisory and NIH ODS summarize triglyceride lowering; Cochrane interview reports about 15% triglyceride reduction and FDA labels approve 4 g/day prescription products for severe hypertriglyceridemia. | USA / UK | AHA and NIH are official clinical sources. The 2020 Cochrane review reports university, WHO and NIHR support, with no known author conflicts; this does not clear every included trial’s finances (original review). | Strong for triglyceride lowering |
| Ordinary fish-oil supplements prevent heart attack or stroke | Cochrane found little or no effect on all-cause mortality, overall cardiovascular events or stroke. Cardiovascular deaths were slightly lower in the pooled analysis (RR 0.92, 95% CI 0.86–0.99), but not in low-risk-of-bias trials (RR 0.95, 95% CI 0.88–1.03). VITAL’s primary composite was null; MI was lower as a secondary outcome. | UK / USA | The Cochrane review reports university/WHO/NIHR support and no known author conflicts. VITAL had NIH support plus donated products, placebos, packaging and assays. Included trials are not collectively financially independent. | Mixed |
| High-dose purified EPA reduces events in selected high-risk patients | REDUCE-IT showed icosapent ethyl 4 g/day reduced the primary composite cardiovascular outcome from 22.0% to 17.2% in statin-treated high-risk patients. | International | REDUCE-IT was sponsored by Amarin; several authors had Amarin roles, stock, consulting, or research funding. | Moderate, conflicted |
| High-dose mixed EPA+DHA prevents cardiovascular events | STRENGTH found no significant reduction in major cardiovascular events with 4 g/day EPA+DHA carboxylic acids and was stopped for futility. | International, 22 countries | STRENGTH was funded by AstraZeneca and multiple authors reported AstraZeneca ties. | Moderate, negative |
| Depression symptoms improve | Meta-analyses suggest a small EPA-rich depression-treatment signal; a separate review found little or no preventive effect on depression/anxiety symptoms. | China/Canada, UK, Iran | Liao 2019: NSFC grant 81761128030 and no conflicts declared. Deane’s prevention review: WHO research and meeting support, with no other relevant interests declared. Underlying trial financing remains incompletely verified. | Mixed |
| Cognitive decline or dementia improves | No meaningful overall cognitive benefit in the cited mild-to-moderate Alzheimer treatment trials or cognitively healthy older-adult trials; incident dementia was not measured in the prevention review. | International / UK | Treatment review: university grant and NIHR infrastructure support; no known author conflicts. Prevention review: one author led an included trial. Commercial support also appears among the underlying trials. | No demonstrated benefit; dementia prevention untested |
| Dry eye improves | Cochrane found long-chain omega-3 produced little to no symptom reduction versus placebo, while newer meta-analyses report improvements in symptoms and tear tests. | Australia/USA/International, Taiwan, Italy | Cochrane: NEI/NIHR support and an included-study author role with independent verification; 17/34 trials had commercial funding, materials or author ties. The 2023 review declared no external funding or conflicts, without clearing every included trial’s finances. | Mixed |
| Rheumatoid arthritis joint symptoms improve | A systematic review of 20 RCTs found improvements in several RA activity markers but noted low overall trial quality. | Greece | Funding and conflicts not disclosed in PubMed summary. | Weak to moderate |
| Pregnancy supplementation improves infant outcomes | Cochrane found reduced preterm birth and early preterm birth, while USDA systematic review rated neurodevelopmental evidence limited or inconsistent. | International / USA | Pregnancy review: public/institutional support, included-study investigator roles and Fonterra/Nestlé institutional honoraria; 11 trials reported industry funding. USDA funded its neurodevelopment review and its scientists participated throughout. | Mixed |
| Flaxseed/ALA replaces EPA+DHA | NIH ODS says ALA conversion to EPA and DHA is typically below 15%, so ALA is not a reliable direct substitute for EPA/DHA needs. | USA | NIH ODS official nutrient guidance; individual cited studies require their own financial assessment. | Doesn't as a substitute |
What omega-3 fatty acids are
Omega-3 fatty acids are polyunsaturated fats with a double bond three carbons from the methyl end of the molecule, and the main nutritionally relevant forms are alpha-linolenic acid (ALA), eicosapentaenoic acid (EPA), docosahexaenoic acid (DHA), and the less-discussed docosapentaenoic acid (DPA) (NIH ODS). ALA is an essential fatty acid found in plant foods such as flaxseed, chia, walnuts, soybean oil, and canola oil, while EPA and DHA are long-chain omega-3s found mainly in seafood, fish oil, cod liver oil, krill oil, and algal oil (NIH ODS).
The active distinction is simple: ALA is a precursor, EPA is more tied to triglycerides and inflammatory lipid mediators, DHA is a major structural fat in the brain and retina, and DPA sits between EPA and DHA but has far fewer direct human trials (NIH ODS, DPA review). NIH ODS reports that humans usually convert less than 15% of ALA into EPA and DHA, which is why flaxseed oil should not be marketed as equivalent to fish oil or algal EPA/DHA (NIH ODS).
Text version of this infographic
| Omega-3 type | Main sources | Role | Practical interpretation |
|---|---|---|---|
| ALA | Flaxseed, chia, walnuts, soybean oil, canola oil | Essential plant omega-3 precursor | Useful nutrient, but conversion to EPA/DHA is usually below 15% according to NIH ODS. |
| EPA | Fish oil, krill oil, cod liver oil, algal EPA products, prescription icosapent ethyl | Triglyceride lowering and inflammatory lipid mediators | Most relevant for triglyceride lowering and EPA-rich depression research according to the depression meta-analysis. |
| DHA | Fatty fish, fish oil, cod liver oil, algal DHA | Brain and retinal structure | Important biologically, but DHA-dominant formulas have weaker evidence for depression than EPA-rich formulas according to the depression meta-analysis. |
| DPA | Seafood and some omega-3 oils in smaller amounts | Intermediate between EPA and DHA; potential specialized pro-resolving mediators | Promising but under-studied; one review found only two pure DPA human studies at the time of publication (DPA review). |
All forms and grades
| Form | Source | Bioavailability | Best use | India cost signal | Verdict |
|---|---|---|---|---|---|
| Fish oil | Fish oil supplies preformed EPA+DHA; check the exact formulation and amount per serving. | Formulation and meals can affect absorption. The matched-dose four-week trial found no statistically significant total plasma/RBC EPA+DHA difference; this is not proof of identical clinical outcomes. | Direct EPA+DHA nutrition; prescription products are a separate treatment category for medically indicated triglyceride lowering. | — | Common option for non-vegetarians; compare current cost per EPA+DHA dose. (Primary source) |
| Krill oil | Antarctic krill; EPA/DHA often in phospholipid form plus astaxanthin | Some small, mostly conflicted studies suggest higher short-term plasma incorporation, but a 4-week matched-dose trial found no significant plasma or red blood cell EPA+DHA difference versus fish oil. | An option if tolerated; compare the actual EPA+DHA dose, capsule size and price. | No demonstrated clinical advantage from these bioavailability trials. | Optional; compare verified EPA+DHA dose and current price. |
| Algal oil | Microalgae; vegan DHA and sometimes EPA+DHA | An exploratory reanalysis found intake-normalized plasma phospholipid response non-inferior to fish oil. Doses and EPA:DHA ratios differed; dsm-firmenich funded the work and employed or contracted authors. | Vegetarian or vegan preformed DHA/EPA; pregnancy use when advised. Fish allergy still requires ingredient and clinician review. | — | Best vegetarian EPA/DHA route. |
| Cod liver oil | Fish liver oil; EPA/DHA plus vitamins A and D | Provides omega-3, but its distinguishing feature is fat-soluble vitamins, not superior omega-3 evidence. | Only when vitamin A/D content is appropriate and total retinol exposure is monitored. | — | Use cautiously because excess vitamin A/D is the unique risk. |
| Flaxseed oil / ALA | Flaxseed oil capsules or food-grade flaxseed | ALA is absorbed, but conversion to EPA/DHA is limited; NIH ODS says conversion is usually below 15%. | Vegetarian ALA intake and dietary diversity, not correction of low EPA/DHA status. | — | Useful food, weak EPA/DHA substitute. |
| Prescription omega-3 | Regulated products such as omega-3-acid ethyl esters and icosapent ethyl | Standardized, studied at 4 g/day doses, and monitored like medication. | Severe hypertriglyceridemia and selected high-risk cardiovascular patients under medical supervision. | Availability and price vary by prescription brand and pharmacy; do not substitute OTC products for prescription therapy. | Clinical-use form for high triglycerides. |
Fish oil
Fish oil provides preformed EPA+DHA. Compare the combined amount per serving, rather than the total oil weight (NIH ODS). One current Nutrabay example lists 180 mg EPA plus 120 mg DHA in a 1000 mg capsule; it is a seller’s label, not a market-wide survey or proof of efficacy (Nutrabay fish oil listing).
Krill oil
The cited krill trials measured short-term blood incorporation, not clinical outcomes. The four-week trial reported no significant total plasma/RBC EPA+DHA difference; DSM funded it and employed several authors. The acute crossover study was supported by Olympic Seafood and SkatteFUNN, with a company employee among its authors (four-week fish/krill trial, acute krill trial). Compare actual EPA+DHA per serving and current cost before choosing a product.
Algal oil
Algal oil supplies preformed DHA and, in some products, EPA, making it a direct nutrient source for vegetarians and vegans (NIH ODS). A dsm-firmenich-funded exploratory reanalysis reported non-inferior intake-normalized plasma phospholipid response relative to fish oil. Its different doses/ratios and employee/contractor authors limit generalization to all products or clinical benefits (microalgal trial).
Cod liver oil
Cod liver oil is not simply “fish oil with extra benefits” because the extra vitamins A and D create a separate safety problem at repeated high intakes (NCCIH). NCCIH specifically warns that fish liver oils such as cod liver oil may contain vitamins A and D, which can be toxic in excessive amounts, so cod liver oil should not be stacked casually with multivitamins, vitamin D tablets, or retinol-containing products (NCCIH).
Flaxseed and ALA
Flaxseed is a source of essential ALA, which converts inefficiently to EPA/DHA. A product containing algal DHA and/or EPA provides those nutrients directly; it is not automatically necessary for every vegetarian (NIH ODS).
EPA/DHA versus DPA
EPA and DHA dominate clinical research, while DPA is biologically interesting but not ready for consumer-facing claims (DPA review). A DPA review described DPA as an “iceberg” omega-3 because it is often overlooked and noted that only two pure-DPA human studies and 11 animal studies had been conducted, which makes DPA an insufficient-evidence supplement target (DPA review).
Text version of this infographic
- Fish oil: direct EPA+DHA, a common option for non-vegetarians, and widely available. Compare actual EPA+DHA per serving and current prices ( Nutrabay fish oil listing ).
- Krill oil supplies EPA+DHA. The small cited trials studied blood incorporation and had manufacturer support; compare the label and current cost rather than assuming clinical superiority (fish oil versus krill trial).
- Algal oil: direct vegetarian DHA/EPA and the best route for vegans or vegetarians who want preformed long-chain omega-3 (microalgal oil bioavailability trial).
- Cod liver oil: direct EPA/DHA plus vitamins A and D, so the extra vitamins are a safety variable rather than a guaranteed advantage (NCCIH).
- Flaxseed oil: ALA-rich but not a reliable EPA/DHA substitute because conversion is usually below 15% (NIH ODS).
EPA-only vs DHA-only vs EPA+DHA
EPA alone, DHA alone and combined EPA+DHA have different trial records. REDUCE-IT and JELIS reported benefits with selected prescription EPA regimens; the contemporary EPA+DHA trials cited below were null on their primary composite outcomes. These are indirect comparisons across different populations, doses, comparators and designs; they do not isolate DHA as the cause of a difference. Both positive EPA trials had manufacturer support, while the null mixed trials also included public/charitable grants, commercial contributions or direct industry funding.
| Dimension | EPA-only | DHA-only | EPA+DHA combined |
|---|---|---|---|
| Best cardiovascular-events evidence | REDUCE-IT and JELIS — both positive, both industry-funded (REDUCE-IT; JELIS) | No DHA-only hard cardiovascular-outcome trial is included in the cited comparison. | STRENGTH, VITAL, ASCEND, OMEMI — all null on the primary composite endpoint (STRENGTH; VITAL; ASCEND; OMEMI) |
| Triglyceride lowering | ~18–21% at 3.8–4 g/day (Innes & Calder 2018) | ~13% at 2.7 g/day, slightly greater than EPA at matched dose | ~19% at 4 g/day in STRENGTH (STRENGTH) |
| LDL cholesterol | Small increases in some head-to-head biomarker trials | Larger increases than EPA in some trials; response depends on population and dose | LDL may rise with mixed prescription products; monitor when treating severe hypertriglyceridemia (FDA LOVAZA label). |
| Depression (adjunctive) | Pooled signal in EPA-rich formulations (≥60% EPA); exploratory subgroup, not guaranteed response (Sublette 2011). | No convincing signal in the small DHA-dominant evidence base; DHA antagonism remains an unconfirmed hypothesis. | No pooled benefit in the <60% EPA subgroup of that review; individual trial results and clinical settings vary. |
| Dry eye | The cited dry-eye evidence does not establish a large EPA-only trial. | The cited dry-eye evidence does not establish a large DHA-only trial. | DREAM’s primary symptom result was null, with NEI grants and manufacturer product donations. Reviews combine heterogeneous trials and comparators (DREAM). |
| Atrial fibrillation risk | Elevated at high dose (REDUCE-IT 3.1% vs 2.1%); the pooled EPA estimate was versus control, not a direct comparison of all three formulations (Khan 2021 meta-analysis) | No dedicated DHA-only AF safety comparison established (Khan 2021 meta-analysis) | AF signal in high-dose mixed trials (STRENGTH 2.2% vs 1.3%); dose meta-analysis does not establish every EPA:DHA ratio (AF dose meta). |
| Funding independence of key trials | Both positive trials industry-funded (Amarin, Mochida) | MIDAS had manufacturer funding; the cited null Alzheimer's and DOMInO trials had public support plus manufacturer contributions | Null findings span public/charitable support with commercial contributions (VITAL, ASCEND), the cited OMEMI trial, and AstraZeneca-funded STRENGTH |
| Overall evidence grade | Moderate (CV events); Moderate (depression) | Weak to Insufficient | Contested (CV events); Weak (dry eye, depression) |
EPA-only
EPA-only products deliver eicosapentaenoic acid in near-total purity, usually as an ethyl ester. The best-studied form is icosapent ethyl (Vascepa in the US, Epadel in Japan), a prescription-grade EPA whose US label approves it alongside maximally tolerated statin therapy for adults with triglycerides ≥150 mg/dL and established cardiovascular disease, or diabetes plus at least two additional cardiovascular risk factors (FDA Vascepa label, March 2026). Over-the-counter “high-EPA” fish oil concentrates exist but do not automatically share the formulation or clinical-outcome evidence of the prescription product.
Selected prescription EPA cardiovascular-outcome trials. In REDUCE-IT (n=8,179 statin-treated patients with triglycerides 135–499 mg/dL, icosapent ethyl 4 g/day vs mineral-oil placebo), the primary composite endpoint fell from 22.0% to 17.2% (HR 0.75, 95% CI 0.68–0.83, p<0.001), with cardiovascular death alone reduced from 5.2% to 4.3% (HR 0.80, p=0.03) (REDUCE-IT). JELIS (n=18,645 Japanese patients, EPA 1.8 g/day added to a statin) cut major coronary events from 3.5% to 2.8% — a 19% relative risk reduction (p=0.011) (JELIS, Lancet 2007). A 2021 meta-analysis of 38 RCTs (149,051 participants) reported stronger EPA-only subgroup estimates for MACE (RR 0.78 vs 0.99) and non-fatal MI (RR 0.72 vs 0.92). The MACE interaction was significant (p=0.000005), but the cardiovascular-mortality interaction was not (p=0.19; subgroup RRs 0.82 vs 0.94). These comparisons are between different trials (Khan et al. 2021, EClinicalMedicine).
Triglycerides. EPA lowers triglycerides roughly 18–21% at 3.8–4 g/day, though DHA lowers them slightly more at matched doses (Innes & Calder 2018). Depression. A meta-analysis of 15 trials involving 916 participants found omega-3’s antidepressant effect concentrated in supplements with ≥60% EPA content (SMD 0.558, p=0.001), versus a null effect for EPA<60% formulations — and the authors found no evidence DHA alone is effective (Sublette et al. 2011, J Clin Psychiatry, partly NIMH-funded; this funding statement does not establish that every included trial was independent).
Where EPA-only wins and loses. In the large trials discussed here, positive cardiovascular outcomes came from selected EPA-only trials, and it carries the strongest depression signal. It is inferior to DHA for raising HDL, lowering blood pressure and heart rate in some head-to-head trials; inflammatory-marker comparisons were inconclusive overall (Innes & Calder 2018). Prescription-dose EPA carries clinically important atrial-fibrillation and bleeding signals.
Safety. REDUCE-IT reported atrial fibrillation/flutter hospitalization in 3.1% vs 2.1% placebo (p=0.004), and a pooled analysis found EPA monotherapy increased AF (RR 1.35) and total bleeding (RR 1.49) versus control; these estimates are not direct EPA-versus-EPA+DHA comparisons (REDUCE-IT; Khan 2021). Bleeding with anticoagulants and antiplatelets requires monitoring.
Typical dose. Cardiovascular risk reduction: 4 g/day (REDUCE-IT) or 1.8 g/day (JELIS, added to statin). Triglycerides: 3.8–4 g/day. Depression adjunct: the exploratory range was 200–2,200 mg/day EPA in excess of DHA, with EPA ≥60% of EPA+DHA. (Primary source)
DHA-only
DHA-only supplements deliver docosahexaenoic acid in isolation, most commonly as algal oil, the vegetarian/vegan source widely used in prenatal and infant-nutrition products because it supplies DHA without a fish ingredient. DHA-only ethyl esters and triglyceride forms are also sold for maternal/infant nutrition and eye-health products. (Primary source)
Cardiovascular risk factors (not hard events). The cited comparison does not include a DHA-only hard cardiovascular-outcome trial. In head-to-head comparator trials, at 2.7 g/day in one trial, DHA lowered triglycerides by 13% versus 12% for EPA and raised HDL by 8%. Blood-pressure and heart-rate reductions came from different trials at approximately 3.7–3.8 g/day. Inflammation findings varied by population and study; the review did not establish universal DHA superiority (Innes & Calder 2018).
Cognition and infant development. The NIA-funded DHA Alzheimer’s trial found no benefit on its primary cognitive outcome (ADAS p=0.41). Martek supplied study drugs, assays and some MRI funding; manufacturer employees were co-authors ( JAMA 2010 ). NHMRC-funded DOMInO randomized 2,399 pregnant women to 800 mg/day DHA or control and found no significant overall child cognitive or language score benefit at 18 months. Girls had lower language and adaptive-behavior scores in secondary analyses without adjustment for multiple comparisons; these do not establish a general developmental harm. Efamol donated capsules and some authors disclosed nutrition-industry advisory roles ( DOMInO, JAMA 2010 ). MIDAS reported a 24-week learning/memory signal in older adults without dementia; Martek funded it and employed most authors (original MIDAS report). These trials cannot be described collectively as financially independent.
Depression. The small DHA-dominant evidence base did not show a convincing signal in Sublette’s review. The proposed EPA-minus-DHA and 1:1 antagonism explanations were exploratory and require prospective confirmation (Sublette 2011).
Where DHA-only evidence differs. Some small head-to-head trials favored DHA for HDL, blood pressure or heart rate; inflammatory-marker differences were inconsistent (Innes and Calder 2018). These biomarker trials cannot establish DHA’s inferiority for hard cardiovascular events. The cited Alzheimer and DOMInO trials were null on their overall cognitive/language endpoints; their populations and interventions differ (DHA Alzheimer trial, DOMInO).
Safety. The clinically relevant caveat is LDL elevation: DHA raised LDL by ~7% (p<0.0001) in head-to-head trials versus ~2% for EPA, a significantly greater effect (Innes & Calder 2018) — relevant for anyone treating high cholesterol. The cited evidence does not establish that DHA-only supplementation carries lower AF risk than EPA: STRENGTH tested a combination, not isolated DHA. (Khan 2021 meta-analysis)
Typical dose. Cardiometabolic risk-factor trials: 2.7–3.8 g/day. Cited cognition trials: 800 mg/day in DOMInO to 2 g/day in Alzheimer's disease. Maternal/prenatal: 800 mg/day (DOMInO).
EPA+DHA combined
This is the most common and most-studied form: fish oil, cod liver oil, krill oil, and prescription products like omega-3-acid ethyl esters (Lovaza) and omega-3 carboxylic acids (Epanova). It contains both EPA and DHA in ratios that vary substantially by product; read the EPA and DHA amounts separately.
Primary cardiovascular events in the cited contemporary EPA+DHA trials. The contemporary EPA+DHA cardiovascular-outcomes trials cited here did not significantly reduce their primary composite endpoint. STRENGTH (n=13,078, omega-3 carboxylic acid 4 g/day vs corn oil, AstraZeneca-funded) reported 12.0% vs 12.2% (HR 0.99, p=0.84) and was stopped for futility (STRENGTH, JAMA 2020). VITAL (n=25,871, omega-3 1 g/day, NIH-supported with manufacturer-donated study products) was null on its primary composite (HR 0.92, p=0.24) while its whole-cohort secondary MI outcome was lower (HR 0.72; 95% CI 0.59–0.90), without adjustment for multiple comparisons (VITAL, NEJM 2019). ASCEND (n=15,480 people with diabetes, 1 g/day vs olive oil; BHF support plus manufacturer products and packaging funding) was null (rate ratio 0.97, p=0.55) (ASCEND). OMEMI (n=1,027 elderly post-MI, 1.8 g/day, with study products sponsored by Orkla Health) was null (HR 1.08, p=0.60) (OMEMI). The 2020 Cochrane review found little or no effect on all-cause mortality or overall cardiovascular events. Its pooled cardiovascular-death estimate was slightly lower (RR 0.92, 95% CI 0.86–0.99), while the low-risk-of-bias analysis was not significant (RR 0.95, 95% CI 0.88–1.03) (original Cochrane review).
Triglycerides. Prescription-level EPA+DHA lowers triglycerides; STRENGTH reported a 19% reduction from baseline versus −0.9% with corn oil (p<0.001) — even though the cardiovascular-outcome was null (STRENGTH). Depression. The <60% EPA subgroup was null in Sublette’s review; the pooled signal favored EPA-rich formulations, but several high-EPA trials were negative and the ratio analysis was exploratory (Sublette 2011).
Dry eye. The NIH-funded DREAM trial (n=535, EPA 2,000mg + DHA 1,000mg/day vs olive oil; NIH-funded with capsules donated by Access Business Group and fish oil by Epax) found no significant difference in OSDI symptom score (−13.9 vs −12.5, p=0.21) or any sign-based outcome (DREAM, NEJM 2018). A more heterogeneous meta-analysis of 19 RCTs was positive, but the large DREAM trial was negative (dry-eye meta 2023).
Where EPA+DHA wins and loses. The cited mixed EPA+DHA trials provide a large evidence base. Selected EPA-only trials reported benefit where the cited EPA+DHA cardiovascular trials did not; this is an indirect comparison, with a negative dry-eye result in DREAM, which had public funding and product donations. The null primary results occur across different funding models. VITAL had public support with product donations; ASCEND had BHF support with manufacturer products and packaging funding. Those contributions need to remain disclosed (original ASCEND trial). Cross-trial results do not prove EPA+DHA is inferior to EPA alone in every clinical setting.
Safety. STRENGTH found new-onset atrial fibrillation in 2.2% vs 1.3% (HR 1.69, p<0.001) despite the null efficacy result, and GI adverse events in 24.7% vs 14.7%. A dose-response meta-analysis found a pooled AF signal that was stronger above 1 g/day (HR 1.49) than at ≤1 g/day (HR 1.12). Signals occurred with EPA-only and mixed products, but every EPA:DHA ratio was not tested (AF dose meta). These separate trials do not establish a direct formulation-specific ranking of bleeding risk.
Typical dose. Cardiovascular megatrials: 1–4 g/day (all null on primary composite). Primary prevention (VITAL, ASCEND): ~1 g/day. Dry eye (DREAM): 3,000 mg/day (null).
How omega-3 works
Omega-3s become part of cell membranes and are used to make signaling molecules involved in cardiovascular, pulmonary, immune, and endocrine function (NIH ODS). EPA and DHA can reduce hepatic triglyceride production, alter very-low-density lipoprotein metabolism, and compete with omega-6 fatty acids in pathways that produce inflammatory and platelet-active lipid mediators (NIH ODS, MedlinePlus).
What the body does with it
ALA can be elongated and desaturated into EPA and DHA, but the conversion is limited and competes with omega-6 fatty acid metabolism (NIH ODS). EPA, DPA, and DHA can also generate specialized lipid mediators involved in resolving inflammation, but mechanistic plausibility does not equal clinical proof for every inflammatory condition (DPA review).
Why form and delivery matter
Form matters because “omega-3” can mean ALA, EPA, DHA, DPA, fish oil, krill oil, cod liver oil, algal oil, or a prescription ethyl ester product, and those are not clinically interchangeable (NIH ODS). Delivery matters because high-dose prescription trials used 4 g/day standardized products, while many retail capsules deliver far less EPA/DHA per capsule and have not been tested for the same endpoints (FDA LOVAZA label, FDA VASCEPA label).
What works and what does not
| Claimed benefit | Verdict | Evidence grade | Key caveat |
|---|---|---|---|
| Lowering triglycerides | WORKS | Strong for triglyceride lowering | Best evidence is for prescription-strength EPA/DHA or EPA at 4 g/day under medical supervision, not low-dose “wellness” capsules (AHA advisory hub, FDA LOVAZA label). |
| Preventing heart attack or stroke in the general population | MIXED | Moderate | Cochrane finds little or no broad protection, while VITAL found fewer MIs in some analyses but not a significant reduction in the primary composite CVD endpoint (Cochrane 2020 interview, VITAL update). |
| Reducing cardiovascular events in selected high-risk patients | MIXED | Moderate | Purified EPA in REDUCE-IT was positive, but mixed EPA+DHA in STRENGTH was negative, and both major trials had industry funding (ACC REDUCE-IT summary, JAMA STRENGTH trial). |
| Depression | MIXED | Moderate | EPA-rich formulas show small treatment signals in depression, but prevention of depression/anxiety symptoms is not supported (depression meta-analysis, British Journal of Psychiatry prevention meta-analysis). |
| Anxiety | INSUFFICIENT EVIDENCE | Weak | A 2023 review found too few anxiety trials for reliable meta-analysis (anxiety/depression review). |
| Alzheimer treatment / cognitive tests in healthy older adults | NO DEMONSTRATED BENEFIT | Negative for measured cognition; dementia prevention untested | No meaningful overall cognitive benefit in the cited Alzheimer treatment trials or cognitive-test prevention trials. None of the latter measured incident dementia (Cochrane Alzheimer treatment, Cochrane cognition prevention). |
| Dry eye | MIXED | Weak to moderate | Cochrane was cautious, while newer meta-analyses report benefits; trial heterogeneity and comparator issues prevent a clean verdict (Cochrane dry eye review, 2023 dry eye meta-analysis). |
| Joint pain / rheumatoid arthritis | MIXED | Weak to moderate | Some RA markers improve, but trial quality is low and omega-3 is not a substitute for disease-modifying therapy (RA meta-analysis). |
| Infant development | MIXED | Moderate | Pregnancy omega-3 reduces early preterm birth in Cochrane, but neurodevelopment outcomes remain limited and inconsistent (Cochrane pregnancy review, USDA neurodevelopment review). |
| Inflammation in general | INSUFFICIENT EVIDENCE | Weak | Mechanisms are real, but “reduces inflammation” is too broad unless tied to a specific disease endpoint (NIH ODS, DPA review). |
Text version of this infographic
| Claim | Verdict | Why |
|---|---|---|
| Triglycerides | WORKS | AHA, NIH ODS, and FDA labels support prescription omega-3 for high triglycerides (AHA advisory hub, FDA LOVAZA label). |
| Heart attack/stroke prevention | MIXED | Cochrane found little or no broad benefit, while specific high-risk EPA trials reported benefit and EPA+DHA trials did not (Cochrane, JAMA STRENGTH trial). |
| Depression | MIXED | EPA-rich formulas show small benefits in depression, but prevention of depression or anxiety symptoms is not supported (depression meta-analysis, British Journal of Psychiatry prevention meta-analysis). |
| Alzheimer treatment / cognitive tests | NO DEMONSTRATED BENEFIT | No meaningful overall cognition benefit; incident dementia prevention remains untested in the cited review (treatment review, prevention review). |
| Dry eye | MIXED | Cochrane was cautious and newer meta-analyses were more positive (Cochrane dry eye review, 2023 dry eye meta-analysis). |
| Rheumatoid arthritis joint pain | MIXED | RA markers improved in a meta-analysis, but study quality was low (RA meta-analysis). |
| ALA replacing EPA/DHA | DOESN'T | ALA conversion to EPA/DHA is usually below 15% (NIH ODS). |
Benefits by claim
Triglycerides and severe hypertriglyceridemia
Omega-3’s cleanest benefit is lowering triglycerides, especially at prescription doses used for severe hypertriglyceridemia (AHA advisory hub). The FDA label for LOVAZA states that omega-3-acid ethyl esters are indicated as an adjunct to diet to reduce triglycerides in adults with severe triglycerides of at least 500 mg/dL and gives a daily dose of 4 g/day (FDA LOVAZA label).
The practical caveat is that triglyceride lowering is not the same as proven prevention of pancreatitis or cardiovascular events for every product, because the LOVAZA label states the effect on pancreatitis risk and cardiovascular mortality/morbidity has not been determined (FDA LOVAZA label).
Heart attack and stroke prevention
For healthy adults or people buying standard fish oil without a specific lipid target, the evidence does not support a simple “fish oil prevents heart attacks” message (Cochrane 2018 statement). The 2020 review included 86 trials and 162,796 participants. It found little or no effect on all-cause mortality, overall cardiovascular events or stroke, but a small pooled cardiovascular-death reduction (RR 0.92, 95% CI 0.86–0.99) and possible small coronary benefits. The cardiovascular-death result was not significant when restricted to low-risk-of-bias trials (RR 0.95, 95% CI 0.88–1.03) (original Cochrane review).
The high-risk prescription-EPA story is different but conflicted: REDUCE-IT reported fewer cardiovascular events with icosapent ethyl 4 g/day, while STRENGTH found no benefit with high-dose EPA+DHA carboxylic acids and showed more atrial fibrillation (ACC REDUCE-IT summary, JAMA STRENGTH trial). REDUCE-IT was sponsored by Amarin and had extensive author ties to Amarin, while STRENGTH was funded by AstraZeneca and had AstraZeneca author conflicts, so the best conclusion is not “industry studies are invalid” but “the benefit appears formulation-, population-, and comparator-specific” (REDUCE-IT ACS analysis, JAMA STRENGTH trial).
Depression and anxiety
Omega-3 is not an antidepressant replacement, but EPA-rich formulas may have a small adjunctive role in diagnosed depression (depression meta-analysis). A 26-trial meta-analysis found an overall small improvement in depressive symptoms and reported that EPA-pure or EPA-major formulations showed benefits, while DHA-major formulations did not show clinical benefits (depression meta-analysis).
The longer-duration prevention review found little or no benefit for preventing depression or anxiety symptoms (British Journal of Psychiatry prevention review). A 2023 adult meta-analysis reported an EPA-rich depression signal with substantial heterogeneity and only one eligible anxiety trial (published 2023 abstract). Its earlier author preprint discloses an Efamol Ltd grant, with no sponsor role reported and no author conflicts declared. The final paper’s full financial statement was not recovered, so that disclosure remains version-specific (author preprint).
Cognitive decline and dementia
The treatment review found no convincing overall benefit in mild-to-moderate Alzheimer disease on cognition, daily activities or dementia severity, and little evidence for other dementias (Cochrane treatment review). The prevention review found no cognitive-test benefit in healthy older adults, but none of its three trials measured new dementia cases; it cannot establish an effect on incident dementia (Cochrane prevention review).
Dry eye
Dry eye is genuinely mixed because a Cochrane review found long-chain omega-3 versus placebo showed little to no reduction in dry eye symptoms, while newer meta-analyses reported improvements in symptoms and tear-film measures (Cochrane dry eye review, 2023 dry eye meta-analysis). The most honest patient-facing verdict is “may help some people, not reliably enough to replace standard dry-eye care” because outcome measures, doses, and comparators vary across trials (Cochrane dry eye review).
Joint pain and rheumatoid arthritis
Omega-3 may modestly improve rheumatoid arthritis disease activity markers, but it should not be framed as a disease-modifying therapy (RA meta-analysis). A systematic review of 20 RCTs involving 1,252 patients found improvements in eight RA-related markers and reduced leukotriene B4, but the overall trial quality was described as low in the evidence brief (RA meta-analysis).
Infant development and pregnancy outcomes
Pregnancy evidence is split between obstetric outcomes and child neurodevelopment outcomes (Cochrane pregnancy review, USDA neurodevelopment review). Cochrane found omega-3 addition during pregnancy reduced preterm birth from 13.4% to 11.9% and early preterm birth from 4.6% to 2.7%, but also increased prolonged gestation beyond 42 weeks from 1.6% to 2.6% (Cochrane pregnancy review).
A USDA-funded systematic review found cognitive-development evidence was limited because 5 of 8 pregnancy supplementation RCTs reported some improvements but all 8 also reported at least one non-significant result, and evidence for language, social-emotional, motor, and visual outcomes was insufficient or inconsistent (USDA neurodevelopment review).
Inflammation
Omega-3s can influence inflammatory lipid mediator pathways, but “reduces inflammation” is too vague to be a reliable supplement claim without naming the disease and outcome (NIH ODS). DPA, EPA, and DHA can produce different specialized pro-resolving mediators, but DPA human supplementation evidence remains too sparse for consumer claims (DPA review).
Risks and reported side effects
| Side effect or risk | Frequency / dose relationship | Mechanism or reason | What to do |
|---|---|---|---|
| Fishy aftertaste, unpleasant taste, taste perversion | Common; LOVAZA label lists taste perversion as >3% and greater than placebo. | Oil reflux and flavor compounds. | Follow the product’s food instructions; ask the prescriber about persistent reflux or taste effects (FDA LOVAZA label). |
| Burping / eructation | Common; LOVAZA label lists eructation >3% and greater than placebo. | Reflux of oil from stomach. | Take with food or split dose; stop if severe (FDA LOVAZA label). |
| Dyspepsia, heartburn, stomach discomfort, nausea, vomiting, diarrhea, constipation | Common to occasional; STRENGTH reported gastrointestinal adverse events in 24.7% with omega-3 carboxylic acids versus 14.7% with corn oil. | Fat load and GI irritation. | Discuss persistent symptoms with the prescriber before changing a prescription dose (STRENGTH, MedlinePlus). |
| Bad breath, bad-smelling sweat, headache | NCCIH lists these possible effects as usually mild; it does not give their frequency. | Reported symptoms; exact mechanisms are not established by the cited overview. | Switch brand/form or stop if bothersome (NCCIH). |
| Musculoskeletal pain, peripheral edema, constipation, gout, atrial fibrillation with icosapent ethyl | VASCEPA label lists these as common in the cardiovascular outcomes trial at incidence at least 3% and at least 1% more frequent than placebo. | Observed adverse-event pattern in outcomes trial. | Use prescription EPA only under clinician monitoring (FDA VASCEPA label). |
| Atrial fibrillation or flutter | Dose-related signal at high doses; REDUCE-IT reported AF/flutter 5.3% versus 3.9%, and STRENGTH reported new-onset AF 2.2% versus 1.3%. | Uncertain electrophysiologic effect; higher risk in people with prior AF/flutter. | Avoid self-directed high-dose omega-3 if you have AF/flutter; monitor closely if prescribed (ACC REDUCE-IT summary, JAMA STRENGTH trial). |
| Bleeding tendency / prolonged bleeding time | The prescription icosapent ethyl trial and VASCEPA label report more bleeding, especially with concomitant antithrombotics. This does not quantify the risk for every OTC dose or formulation. | Reduced platelet aggregation and prolonged bleeding time. | Monitor with anticoagulants, antiplatelets, bleeding disorders, and surgery (FDA VASCEPA label, NIH ODS). |
| Fish or shellfish allergic reaction | Potentially serious; increased risk in people with fish/shellfish allergy is unknown in the prescription label. | Hypersensitivity can occur; the label does not establish contaminants as the cause. | Seek medical guidance; known hypersensitivity to the product or its ingredients is a contraindication. An algal product also requires ingredient review (FDA VASCEPA label). |
| LDL cholesterol increase | Relevant to EPA+DHA prescription products; LOVAZA label says LDL may increase. | DHA-containing products may raise LDL in some hypertriglyceridemia patients. | Monitor LDL periodically when treating high triglycerides (FDA LOVAZA label). |
| Liver enzyme concerns in hepatic impairment | Not a routine issue for healthy users; LOVAZA label recommends ALT/AST monitoring in hepatic impairment. | Medication-level omega-3 processing and underlying liver disease. | Clinician monitoring if liver disease is present (FDA LOVAZA label). |
| Vitamin A or D excess from cod liver oil | Dose-related and stack-related; risk increases when combined with multivitamins or vitamin D/retinol products. | Cod liver oil contains fat-soluble vitamins that accumulate. | Check vitamin A/D amounts and avoid stacking unless prescribed (NCCIH). |
| High-dose immune suppression concern | The Institute of Medicine noted that 900 mg/day EPA plus 600 mg/day DHA or more for several weeks might reduce immune function, while EFSA and FDA assessments cited by NIH ODS found up to about 5 g/day EPA+DHA appears safe for bleeding and immune endpoints. | Suppression of inflammatory responses. | Avoid chronic high-dose self-supplementation, especially with immunosuppression or active infection risk (NIH ODS). |
Interactions and medication precautions
| Interacts with | Type | Severity | Mechanism | Action |
|---|---|---|---|---|
| Warfarin | Medication | Monitor | Omega-3 may prolong bleeding time and may theoretically raise INR, although a retrospective warfarin study found fish/krill oil did not significantly affect time in therapeutic range or bleeding. | Do not start, stop, or change high-dose omega-3 without INR/clinician monitoring (NIH ODS, warfarin retrospective study). |
| DOACs: apixaban, rivaroxaban, dabigatran, edoxaban | Medication | Monitor / caution | The VASCEPA label advises bleeding monitoring when prescription icosapent ethyl is used with anticoagulants; it does not provide individual DOAC interaction estimates. | Review prescription icosapent ethyl and any proposed supplement with the prescriber (FDA VASCEPA label). |
| Heparin, low-molecular-weight heparins, fondaparinux | Medication | Monitor / caution | The VASCEPA label advises monitoring prescription icosapent ethyl with anticoagulants; drug-specific estimates for these agents are not provided. | Discuss the full regimen and procedural plans with the treating clinician (FDA VASCEPA label). |
| Antiplatelets: aspirin, clopidogrel, prasugrel, ticagrelor | Medication | Monitor | VASCEPA reports greater bleeding incidence with antithrombotics. LOVAZA describes possible bleeding-time prolongation without establishing clinically significant prolongation in its cited trials. | Monitor bruising, nosebleeds, GI bleeding signs, and perioperative plans (FDA VASCEPA label, FDA LOVAZA label). |
| NSAIDs: ibuprofen, naproxen, diclofenac, high-dose aspirin | Medication | Use caution | The cited omega-3 fact sheet does not establish a specific NSAID interaction. A pharmacist should assess the full medication list. | Review frequent NSAID use with the clinician, especially alongside antithrombotics (MedlinePlus medication review). |
| SSRIs/SNRIs: sertraline, fluoxetine, escitalopram, venlafaxine, duloxetine | Medication | Use caution | The cited fact sheet does not establish an omega-3-specific SSRI/SNRI interaction or a clinical effect size. | Discuss the complete medicine/supplement list with the prescriber; these sources do not quantify an added SSRI/SNRI bleeding risk (MedlinePlus). |
| Blood pressure medicines: ACE inhibitors, ARBs, beta blockers, calcium-channel blockers, diuretics | Medication | Monitor if symptomatic | The cited fact sheet does not establish an interaction with each named blood-pressure drug class. | Keep prescribed BP treatment and review high-dose supplementation with the clinician (MedlinePlus). |
| Diabetes medicines: insulin, sulfonylureas, metformin, GLP-1 drugs, SGLT2 inhibitors | Medication / condition | Monitor | NIH ODS cites EFSA safety assessment stating up to about 5 g/day EPA+DHA has not been shown to affect glucose homeostasis, but diabetes is a condition requiring medication review. | Monitor lipids and glucose as usual; do not use omega-3 as diabetes treatment (NIH ODS, MedlinePlus). |
| Statins, especially atorvastatin | Medication | Generally compatible | VASCEPA label reports 4 g/day did not significantly change atorvastatin exposure in a drug-interaction study. | Commonly combined in high-risk patients under medical care; monitor lipids and adverse effects (FDA VASCEPA label). |
| Omeprazole | Medication | No meaningful interaction shown for icosapent ethyl | VASCEPA 4 g/day did not significantly change omeprazole exposure in a drug-interaction study. | Follow the product and prescriber instructions; this study does not establish a timing-separation benefit (FDA VASCEPA label). |
| Rosiglitazone | Medication | No meaningful interaction shown for icosapent ethyl | VASCEPA 4 g/day did not significantly change rosiglitazone exposure. | Follow diabetes prescriber guidance (FDA VASCEPA label). |
| Orlistat | Medication | Possible reduced absorption | An eighteen-rat study found reduced fatty-acid incorporation with dietary orlistat; a corresponding human effect was not established. | Discuss nutritional goals with the prescriber. The cited study does not show that separating doses restores absorption (rat EPA/DHA study). |
| Cyclosporine and transplant immunosuppression | Medication / condition | Specialist only | Older cyclosporine studies evaluated fish oil in transplant or vascular-toxicity settings, and high-dose omega-3 has immune-function theoretical concerns. | Do not self-supplement after transplant without the transplant team (cyclosporine renal transplant trial, NIH ODS). |
| Other supplements discussed in bleeding reviews: ginkgo, garlic, ginger, ginseng, high-dose vitamin E, turmeric/curcumin | Supplement | Review combinations | Evidence varies by supplement and antithrombotic use. The 2022 review found no clinical bleeding association for fish oil or ginseng. | Disclose all products before procedures or while taking antithrombotics. These reviews do not prove that each combination adds bleeding risk (2022 bleeding review, 2012 review). |
| Vitamin A, vitamin D, retinoids, multivitamins | Supplement / medication | Cod liver oil caution | Cod liver oil can contain vitamins A and D, creating additive fat-soluble vitamin exposure. | Avoid cod liver oil stacks unless total vitamin A/D is calculated (NCCIH). |
| Omega-6-rich diet | Dietary interaction | Diet context | Omega-6 and omega-3 pathways compete for desaturation enzymes. | NIH ODS says an optimal omega-6:omega-3 ratio is not defined; adequate EPA/DHA intake is generally more useful than simply lowering linoleic acid (NIH ODS). |
| Atrial fibrillation or flutter | Medical condition | Use caution / avoid high-dose self-use | Prescription labels and major trials show increased AF/flutter risk, especially with prior AF/flutter. | Discuss with cardiologist before any high-dose omega-3 (FDA VASCEPA label, JAMA STRENGTH trial). |
| Liver disease, thyroid disease, pancreatic disease | Medical condition | Medical supervision | MedlinePlus specifically tells patients to inform clinicians about liver, thyroid, pancreatic disease, diabetes, or atrial fibrillation before prescription omega-3 use. | Use clinician-guided dosing and monitoring (MedlinePlus). |
| Fish/shellfish allergy | Medical condition | Avoid fish-derived forms unless cleared | Fish-derived prescription products warn that allergic-reaction risk is uncertain in fish/shellfish allergy. | Consult an allergist or clinician and review the exact ingredients, including any algal product; algal origin alone does not establish allergy safety (FDA VASCEPA label). |
| Surgery, dental extraction, childbirth bleeding risk | Clinical situation | Tell clinician | Bleeding-time concerns matter most when combined with procedures or anticoagulants. | Disclose omega-3 use before procedures; follow surgeon/obstetrician instructions (FDA LOVAZA label). |
Text version of this infographic
| Risk level | Medication, supplement, or condition | Why it matters |
|---|---|---|
| High caution | Warfarin, DOACs, heparins, aspirin, clopidogrel, prasugrel, ticagrelor | The VASCEPA label advises bleeding monitoring for prescription icosapent ethyl with anticoagulants or antiplatelets. It does not quantify risk for every OTC dose/form or each named drug (FDA VASCEPA label). |
| High caution | Atrial fibrillation or flutter | High-dose trials and prescription labels show increased AF/flutter risk (JAMA STRENGTH trial). |
| High caution | Surgery, dental extraction, bleeding disorders | Bleeding-time effects matter most around procedures and in people with baseline bleeding risk (FDA LOVAZA label). |
| Caution | Other medicines and supplements, especially with antithrombotic treatment | Check the full list with a clinician. Evidence is heterogeneous; the cited review found no clinical bleeding association for fish oil or ginseng and does not prove additive effects for every combination (bleeding review). |
| Medication review | Blood-pressure and diabetes medicines | Tell the prescriber about all medicines and relevant conditions; this guidance does not establish a specific interaction with every BP or diabetes drug (MedlinePlus). |
| Usually compatible under care | Statins and omeprazole | VASCEPA interaction studies did not show significant exposure changes for atorvastatin or omeprazole (FDA VASCEPA label). |
| Avoid stacking | Cod liver oil plus vitamin A, vitamin D, retinoids, or multivitamins | Fish liver oils can contain vitamins A and D, which can be toxic in excess (NCCIH). |
Who should avoid omega-3 supplements
Discuss other medicines and relevant conditions with the clinician before prescription treatment or high-dose supplementation: antithrombotic use, frequent NSAIDs, bleeding disorders, planned procedures, atrial fibrillation/flutter, fish or shellfish allergy, liver/pancreatic/thyroid disease, pregnancy or breastfeeding, and transplant treatment. These are topics for review, not a list of universal contraindications. Known hypersensitivity to the prescribed product or its ingredients is a contraindication (MedlinePlus, FDA VASCEPA label). Fish liver oils may add substantial vitamin A/D; review total intake alongside other vitamin products or retinoid medicines rather than assuming every combination is unsafe (NCCIH).
Dosage and how to take
| Goal | Evidence-matched form | Typical evidence dose | Practical guidance |
|---|---|---|---|
| General omega-3 nutrition | Food first: oily fish for non-vegetarians; algal DHA/EPA for vegetarians; flax/chia/walnuts for ALA | No single official EPA/DHA RDA is set in the NIH ODS fact sheet, but ALA adequate intakes are listed and seafood is emphasized as a source. | Do not confuse ALA with EPA/DHA; check total EPA+DHA on labels (NIH ODS). |
| High triglycerides | Prescription omega-3-acid ethyl esters or icosapent ethyl | 4 g/day in FDA labels. | Use only under medical supervision; monitor LDL, bleeding risk, and AF risk (FDA LOVAZA label, FDA VASCEPA label). |
| High-risk cardiovascular patients with elevated triglycerides | Icosapent ethyl, not generic fish oil | 4 g/day in REDUCE-IT and FDA label. | This is a prescription decision because REDUCE-IT was positive but industry-funded and AF/bleeding monitoring is needed (ACC REDUCE-IT summary, FDA VASCEPA label). |
| Depression adjunct | EPA-major formula | Meta-analyses suggest EPA-rich products, often around 1 g/day EPA range, have the strongest signal. | Use only as adjunctive care, not a replacement for therapy or prescribed antidepressants (depression meta-analysis, anxiety/depression review). |
| Pregnancy | Food-based low-mercury seafood or clinician-approved DHA/EPA, including algal DHA for vegetarians | Cochrane analyzed varied pregnancy regimens and found lower preterm birth but more prolonged gestation. | Pregnancy dosing should be clinician-guided; avoid cod liver oil unless vitamin A/D exposure is reviewed (Cochrane pregnancy review, NCCIH). |
| Dry eye or joint pain | EPA+DHA products | No universally reliable dose because trials vary. | Consider only as a time-limited adjunct if standard care is in place and bleeding/AF risks are low (Cochrane dry eye review, RA meta-analysis). |
Follow the exact product instructions: VASCEPA is taken with food, while MedlinePlus states LOVAZA can be taken with or without food. Meals and formulation can affect absorption, so there is no single direction for every omega-3 product. Swallow prescription capsules whole and discuss reflux or dose changes with the prescriber (MedlinePlus, current VASCEPA label).
Text version of this infographic
- General nutrition: use food first, and if supplementing, choose a product that states actual EPA+DHA rather than only “fish oil mg” (NIH ODS).
- High triglycerides: evidence-matched dosing is prescription-level 4 g/day with medical monitoring, not casual OTC dosing (FDA LOVAZA label).
- Depression adjunct: EPA-rich formulas show the strongest signal, but they are adjuncts rather than replacements for mental-health treatment (depression meta-analysis).
- Pregnancy: use clinician-guided DHA/EPA and avoid uncalculated cod liver oil because fish liver oils can contain vitamins A and D (NCCIH).
Related research
For deeper, evidence-graded context on the conditions omega-3 is most often used for, see Pure City Research's condition guides:
- Heart disease: prevention and management guide
- Type 2 diabetes: prevention and management guide
- Stress, anxiety & depression: prevention and management guide
Funding and conflict disclosures
Research funding at a glance
12 disclosure entries. The counts below summarize independence tiers explicitly assigned in this article. They count disclosures, not studies, funding amounts or evidence quality.
Consult this article’s source and funding notes for named funders, countries, relationships and exceptions where available. Institutional backing, researcher interests and trial sponsorship are separate questions. Public funding alone does not establish independence; commercial ties alone do not prove a claim false. This overview is not a new financial audit.
| Source | Country | Funding source | Conflict status | How we used it |
|---|---|---|---|---|
| NIH ODS omega-3 fact sheet | USA | US government; no commercial funding disclosed on page. | Official government evidence summary; individual studies have separate funding and conflict disclosures. | Background, forms, ALA conversion, safety, interactions. |
| NCCIH omega-3 overview | USA | US government. | Official government consumer guidance; underlying clinical studies require separate disclosure assessment. | Consumer safety, side effects, cod liver oil caution. |
| Cochrane cardiovascular review summary | UK / international evidence | 2020 original review: University of East Anglia internal support, WHO funding for the extension and NIHR Cochrane infrastructure support (full review). | Authors declare none known; included-trial commercial support is not ruled out. | Primary weight for broad cardiovascular prevention. |
| VITAL update | USA | NIH grants; Pharmavite and Pronova/BASF agents, placebos and packaging; Quest no-cost assays. NIH participated in design/conduct/interpretation; investigators retained final decisions. | Public funding with commercial in-kind contributions; separate ancillary-trial financial details are not fully cleared. | Primary-prevention composite and secondary MI outcome. |
| REDUCE-IT ACS analysis | France/USA/Canada; parent trial international | Amarin Pharma sponsored REDUCE-IT. | Conflicted; several authors had Amarin employment, stock, consulting, or research funding. | Used cautiously for prescription EPA in selected high-risk patients. |
| STRENGTH trial | International, 22 countries | AstraZeneca AB. | Conflicted; multiple author AstraZeneca ties. | Counterweight to REDUCE-IT and high-dose EPA+DHA safety. |
| Depression meta-analysis | China / Canada | National Natural Science Foundation of China, grant 81761128030 | Authors declared no conflict of interest. | EPA-rich depression signal. |
| British Journal of Psychiatry prevention meta-analysis | UK | WHO NUGAG funded the research via UEA (grant 2017/695622-0) and some meeting attendance; commissioning design and GRADE roles disclosed. | No other relevant financial relationships declared. Underlying trial funding remains separately assessed. | Depression/anxiety prevention claim. |
| Cochrane dementia treatment review | International | Roux university grant and NIHR Cochrane infrastructure support (original review). | No known author interests declared; included trials had commercial support/roles. | Mild-to-moderate Alzheimer treatment; not incident dementia prevention. |
| Cochrane dry eye review | Australia/USA/international | NEI methodological/salary support and NIHR infrastructure support; 17/34 included trials had industry funding, materials or relevant author ties. | One author was senior author on two included studies; independent verification used. | Dry-eye caution. |
| Cochrane pregnancy review | International | SAHMRI/Women’s & Children’s Hospital plus DFID, MRC, NHMRC and Danish National Research Foundation support; 11 trials reported industry funding. | Included-study investigator roles; Fonterra/Nestlé institutional honoraria disclosed. Other reviewers handled those studies. Unreported trial support remains unknown. | Pregnancy outcomes. |
| Microalgal oil bioavailability trial | Switzerland | Internal dsm-firmenich funding. | Conflicted; multiple authors were employees or contractors. | Used only for bioavailability, not broad benefit claims. |
Which omega-3 form is best: fish oil, krill oil, algal oil, cod liver oil, or flaxseed?
Fish and algal oils supply direct EPA/DHA; compare the dose, ingredients and current price. Cod liver oil adds vitamin A/D exposure, krill bioavailability trials do not prove superior clinical outcomes, and flaxseed supplies ALA with limited conversion (NIH ODS, fish/krill trial).
Can vegetarians get enough omega-3 from flaxseed alone?
Vegetarians can get ALA from flaxseed, chia, walnuts, and oils, but flaxseed alone is not a dependable source of EPA/DHA because conversion is inefficient. If the goal is direct DHA or EPA, algal oil is the more evidence-matched vegetarian option (NIH ODS, microalgal oil bioavailability trial).
Does omega-3 help depression or anxiety?
Omega-3 may modestly help depression symptoms when formulas are EPA-rich, but it does not have strong evidence for preventing depression or anxiety symptoms. A 26-trial meta-analysis found benefits mainly for EPA-pure or EPA-major formulations, while a prevention meta-analysis found long-chain omega-3 probably has little or no effect on depression or anxiety symptoms (depression meta-analysis, British Journal of Psychiatry prevention meta-analysis).
Does omega-3 thin the blood?
Omega-3 can reduce platelet aggregation and prolong bleeding time, especially at high doses, but most research does not show major bleeding problems at typical doses. The FDA labels still advise monitoring when omega-3 products are used with anticoagulants or antiplatelets, and people on warfarin, DOACs, aspirin, or clopidogrel should not start high-dose omega-3 without medical advice (NIH ODS, FDA VASCEPA label).
Is cod liver oil better than fish oil?
Cod liver oil is not automatically better because it adds vitamins A and D, which can be useful only when needed and risky when stacked with other supplements. NCCIH warns that fish liver oils may contain vitamins A and D that can be toxic in excessive amounts (NCCIH).
How much omega-3 should I take daily?
The right dose depends on the goal: food-first intake for general nutrition, clinician-guided 4 g/day prescription omega-3 for severe high triglycerides, EPA-rich adjunct formulas only when treating depression under care, and pregnancy dosing only with obstetric guidance. FDA labels for LOVAZA and VASCEPA use 4 g/day for prescription indications, but that dose should not be copied with over-the-counter capsules without monitoring (FDA LOVAZA label, FDA VASCEPA label).
Frequently asked questions
Is fish oil actually good for your heart?
Prescription omega-3 at 4 g/day lowers elevated triglycerides; AHA describes about 20–30% reduction in the 200–499 mg/dL range (AHA advisory summary). The 2020 Cochrane review included 86 trials and 162,796 participants, finding little or no effect on all-cause mortality or overall cardiovascular events, alongside a small pooled cardiovascular-death reduction (RR 0.92, 95% CI 0.86–0.99) and possible small coronary benefits. The cardiovascular-death estimate was not significant in low-risk-of-bias trials (RR 0.95, 95% CI 0.88–1.03) (Cochrane review). REDUCE-IT found fewer events with prescription icosapent ethyl in selected statin-treated high-risk patients; this does not establish an OTC fish-oil benefit (REDUCE-IT).
Fish oil vs krill vs algal — which is best?
Formulation and meals affect absorption; the evidence does not establish one universal ranking. A matched-dose four-week trial found similar plasma and red-cell EPA+DHA levels for fish-oil ethyl esters, triglycerides and krill oil. Algal products provide DHA and sometimes EPA; check the label. Flaxseed provides ALA, which converts inefficiently (NIH ODS omega-3). (Matched-dose bioavailability trial)
Can omega-3 cause bleeding?
The cited 2022 review did not find a clinical bleeding association for ordinary fish-oil supplements. Prescription EPA evidence is different: the current VASCEPA label reports more bleeding, especially with antithrombotics. There is no universal 3 g/day safety cutoff; review the dose, formulation and other medicines with the prescriber (bleeding review, current VASCEPA label).
Have a question — or want us to cover something?
Ask about anything on this page, or request the next deep dive: an ingredient, a supplement, or a health concern. We use published research, evidence syntheses, and regulatory guidance, with clear source links.
One daily research roundup
Get the topics, key findings and links from our new articles in one email. At most one digest a day, only when there is something new.
