Omega-3: The Essential Anti-Inflammatory

Fish oil might be the most widely taken supplement that almost nobody ever measures. If you have been swallowing a capsule every morning without once seeing a number, you are in good company. What we do differently at Fishtown Medicine is treat omega-3 the way we treat your blood pressure or your ApoB: we measure it first, then dose toward a target. The standard guidelines call an Omega-3 Index of 8% or higher "low risk" and leave it at that. I would rather work with you toward 12 to 15%. That range buys you more heart resilience, more brain protection, and a calmer inflammatory baseline. It is also where the populations with the fewest cardiovascular and dementia events have historically lived.

What omega-3 is and what it does

Omega-3 fatty acids are healthy fats your body cannot build for itself, which means every bit of it arrives through what you eat or what you take. For adults, the 2 that matter are EPA (eicosapentaenoic acid) and DHA (docosahexaenoic acid), and you get them mainly from oily fish like salmon, sardines, and anchovies, or from algae oil. Once they are in you, they get built into your cell membranes, bring your triglycerides down, and lower inflammatory cytokines. They also hold up the structure of brain and heart tissue, and the signaling that tells inflammation when it is time to stop.

The blood test I care about most in this conversation is the Omega-3 Index. It measures how saturated your red blood cell membranes are with EPA and DHA. That tells us what your tissue levels have been across the past 120 days, rather than what you ate for dinner yesterday. Here is how I read your result:

• High-risk (under 4%): this is the range that gets my attention. Down here you carry a higher statistical risk of sudden cardiac events, and any chronic inflammation you have is running with very little pushing back against it.
• Intermediate (4 to 8%): most new patients walk in somewhere in this band. It is "standard" for how Americans eat, which is a low bar to clear, and it leaves you a lot of room to work with.
• Guideline-acknowledged "low risk" floor (8 to 12%): the conventional Harris/von Schacky target. Getting here is a meaningful improvement over the American baseline. It is also the point where sudden cardiac death risk drops off sharply, which is why the guidelines are comfortable stopping here.
• FTM target (12 to 15%): the saturation seen in populations with high marine intake, like Japanese coastal communities, whose cardiovascular and dementia event rates have historically been the lowest anywhere. It is where I aim, using a Rx-grade or third-party-tested formulation, and I bring up the AF caveat out loud every time rather than tucking it into a footnote.
• Above 15%: historically this was Inuit territory, where extreme marine intake was simply the diet. Climbing toward 18% buys you nothing measurable, the benefit curve has already flattened by then, so I do not push you up here without a specific clinical reason.

EPA and DHA are not interchangeable, and the ratio you pick should follow what you are trying to fix. EPA is the main driver for mood support and for bringing systemic inflammation down. DHA is the structural scaffolding of your brain and your retina. That makes it the priority when we are protecting cognition, supporting a pregnancy, or helping you come back from a concussion or a traumatic brain injury (TBI). Eye health comes with a wrinkle worth understanding: eating fish is linked with lower macular degeneration risk, while omega-3 supplements showed no benefit in the AREDS2 trial. For mood that looks inflammation-driven, I look for a 4:1 EPA to DHA ratio, or at least 60% EPA. For neuroprotection, I want you at 600 to 1,000 mg of DHA a day or better.

Who this is for (and who it isnt)

Omega-3 fits a wide range of adults. These are the situations where I bring it up early:

• Most Philly adults who don't eat oily fish 3 or more times per week. Cheesesteaks, hoagies, pretzels, and pizza hand you all the omega-6 fat you could ask for and almost no omega-3. That is most weeks for most of us here, and when it is, a supplement is usually the simplest correction.
• Heart and metabolic health patients. At 2,000 to 4,000 mg a day, omega-3 reliably brings your fasting triglycerides down by 20 to 30%. That is often further than you can get by changing how you eat alone.
• Brain and mood support. In randomized trials, higher-EPA formulations at 1,000 to 2,000 mg of EPA a day have shown modest benefit for depression, added on top of standard treatment.
• Joint and autoimmune conditions. At 2,000 to 4,000 mg a day, higher omega-3 intake is linked with less joint pain in rheumatoid arthritis and with modest improvement in psoriasis.
• APOE4 carriers. If you carry APOE4, the variant that raises your Alzheimer's risk, we go higher, 3,000 to 4,000 mg. I also push salmon roe or oily fish, because those give you omega-3 in the phospholipid form, which may get into the brain more efficiently.

There are also a few situations where we talk it through first, or start you somewhere else:

• You take a blood thinner (Eliquis, Xarelto, warfarin, daily aspirin), or you have a bleeding disorder. Omega-3 thins the blood modestly on its own, so the dose and the monitoring get coordinated rather than stacked.
• You have a history of atrial fibrillation. At 4 grams or more per day, omega-3 has been linked with a small increase in AFib risk, and your cardiologist belongs in that decision with us.
• You have surgery scheduled in the next 7 to 14 days. Most surgeons want fish oil stopped before an elective procedure to lower bleeding risk, so we plan around the date.

How we evaluate it: safety, then effectiveness, then cost

Every supplement we recommend goes through the same 3 gates, in that order, and we go deeper on the whole process in how we choose supplements.

• Safety first. I want third-party testing on the bottle, either IFOS (International Fish Oil Standards) certification or USP verification. High-quality oils that carry it are filtered down to undetectable levels of mercury, PCBs, and dioxins, while cheaper oils, particularly store-brand ethyl esters, do not always get there. If a bottle smells strongly fishy, the oil has gone rancid and it belongs in the trash.
• Effectiveness second. Your one-a-day multivitamin probably carries 300 mg of fish oil, which is far too small a dose to move a biomarker. Read the combined EPA plus DHA per serving rather than the total fish oil figure printed on the front. We also prefer the re-esterified triglyceride (rTG) form, Nordic Naturals for example, over the cheaper ethyl ester (EE) form, which absorbs poorly and is harder on your stomach.
• Cost last. A 90-day supply of a high-quality rTG or algae omega-3 usually costs $30 to $60, and a direct-to-consumer Omega-3 Index test typically runs $50 to $100. Once we are down to the pure, well-absorbed options, we take the best value.

How to dose it, and when

Read the label for combined EPA plus DHA, because a softgel sold as "1,000 mg fish oil" may carry only 300 mg of omega-3s.

• Foundational. I usually start you at 2,000 mg of total EPA and DHA daily, which is typically 2 large softgels of rTG oil, and we build from there.
• Therapeutic. If your triglycerides are high, or we are working on an autoimmune condition, we may titrate up to 4,000 mg daily with close supervision.
• APOE4 protocol. 3,000 to 4,000 mg daily, weighted toward phospholipid-form sources like salmon roe and oily fish.
• Take it with fat. Omega-3s are fat-soluble. If you take them on an empty stomach with black coffee, you lose most of the dose. Take them with the largest meal of your day instead; avocado, olive oil, eggs, and salmon all do the job.
• Consistency beats timing. These fats saturate your cell membranes over weeks, so a missed dose is no crisis. What fills the reservoir is taking it steadily.

Triglycerides and joint stiffness can change within 4 to 8 weeks. Moving your Omega-3 Index from a low value up into the 12 to 15% target usually takes 3 to 4 months of daily dosing, because red blood cells turn over slowly. We retest at month 3 to confirm you got there.

Cerebrovascular and stroke evidence

Stroke is a more nuanced story than sudden cardiac death, and it only makes sense once you separate observational biomarker data from RCT supplementation data.

• Observational, biomarker-based: A pooled analysis of 29 prospective cohorts covered 183,291 participants and 10,561 strokes. Compared with the lowest quintile of EPA, the highest quintile was associated with a 17% lower total stroke risk (HR 0.83) and an 18% lower ischemic stroke risk (HR 0.82). DHA moved the same direction, a little more weakly.
• RCT, supplementation-based: A Cochrane review of 31 trials found that omega-3 supplementation made little or no difference to stroke risk (RR 1.02).
• The exception: EPA monotherapy. In REDUCE-IT, prescription icosapent ethyl at 4 g/day cut non-fatal stroke by 29% (RR 0.71) in patients who already had cardiovascular disease or diabetes and were already taking a statin. The EPA+DHA combination products have not reproduced it.

Dose and target explain most of the gap. The trials that showed nothing used mixed EPA/DHA capsules at modest doses without measuring or targeting the Omega-3 Index. The trials and biomarker cohorts that showed a signal ran higher EPA fractions and / or sustained high tissue saturation. That is why we measure your O3I instead of guessing at it.

Both guideline bodies read that negative-trial evidence the way you would expect. The 2024 AHA/ASA Primary Prevention of Stroke Guideline assigned a Class 3 (no benefit) recommendation for omega-3 supplementation for stroke risk reduction. The 2026 ACC/AHA Dyslipidemia Guideline carries a separate Class 2b recommendation for icosapent ethyl (4 g/day). It applies to adults 50 and older with established ASCVD, or diabetes plus another risk factor, along with persistent TG 150-499 mg/dL and LDL under 100 on a maximally tolerated statin. We respect both positions, and our practice pattern still runs ahead of the formal criteria, because we are treating to an Omega-3 Index biomarker target rather than waiting for the trial-defined entry conditions.

How we use omega-3 in clinic (and why we move earlier than the Class 2b criteria)

Our clinical position is that omega-3 does reduce MACE and stroke risk - when product quality, dose, and biomarker targeting are right. Read the null trials closely. What they indict is under-dosed, poorly-characterized OTC fish oil given without measuring tissue levels, rather than omega-3 itself. Most of them used around 1 g/day of mixed EPA+DHA, the same exposure profile the biomarker cohorts call "low risk." That is the protocol those trials studied. It is not the protocol we use.

The supporting case for our position:

• REDUCE-IT (icosapent ethyl 4 g/day, pharmaceutical-grade EPA) showed a 25% MACE reduction, an NNT of 21 over 4.9 years, and a 28% reduction in stroke as a secondary outcome. Quality and dose were doing that work.
• The biomarker data (Omega-3 Index, EPA and DHA tissue levels) ties higher levels to lower MACE, stroke, and all-cause mortality across hundreds of thousands of participants, and it does so consistently. In the Framingham Offspring analysis, the top O3I quintile came in with 39% lower incident CVD.
• The 2024 stroke-prevention Class 3 position and the 2026 Class 2b dyslipidemia position both rest heavily on trials that used mixed EPA+DHA at 1 g/day or less without biomarker targeting. That is essentially the same "low risk" exposure profile again.

What we prescribe:

• Lovaza (omega-3 acid ethyl esters, EPA+DHA Rx) when insurance covers it. Its FDA approval is for triglycerides over 500 mg/dL. We use it for ASCVD and cerebrovascular risk reduction in selected patients, because the formulation is pharmaceutical-grade and we know what is in it.
• Icosapent ethyl (Vascepa) at 4 g/day if you meet the REDUCE-IT criteria: established ASCVD or diabetes plus a risk factor, already on a statin, LDL 41-100, TG 150-499, with insurance willing to cover it. Of everything on this page, that is the path with the clearest single trial standing behind it.
• Specific pure, third-party-tested EPA+DHA formulations at 2-3 g/day, dosed to bring your Omega-3 Index up into the 12-15% range. Generic OTC fish oil stays off the list. The dose, oxidation state, and EPA:DHA ratio vary from bottle to bottle, and that exposure is what produced the null outcomes data.
• We start earlier than the formal Class 2b criteria when you have elevated ApoB, residual ASCVD risk, a family history of early MI or stroke, or a measured low O3I, whatever your triglycerides happen to be doing. We are treating to a biomarker target rather than waiting for trial-defined entry criteria.
• We surface the AF risk directly. Both EPA-only and EPA+DHA raise AF risk in a dose-dependent way, particularly above about 1 g/day. In REDUCE-IT, hospitalization for AF ran 3.1% on IPE against 2.1% on placebo, an NNH of roughly 91, and among patients with prior AF the gap was much wider, 12.5% against 6.3%. So if you have a history of palpitations, prior AF, or a pacemaker, you and I have that risk-benefit conversation out loud before anything gets added.

For the broader stroke and MACE primary-prevention picture, see the Stroke Prevention guide.

Flaws, side effects, and interactions

No supplement is perfect, and being honest about the downsides is part of the job.

• Fish burps and reflux. Usually this means the oil is oxidized (rancid), it is in ethyl ester form, or you took it on an empty stomach. Switching to a rTG brand, freezing the capsules, and taking them with your largest meal settles it most of the time, and enteric-coated softgels, which dissolve in the small intestine, are another option. If the bottle smells strongly fishy, throw it out.
• Mild blood thinning. Omega-3s have a mild anti-platelet effect. Highly saturated cell membranes are good for blood flow generally, but your clotting time does lengthen a little, so we plan a washout of 7 to 14 days before scheduled surgery.
• Blood thinners. Taking omega-3 alongside prescription blood thinners (Eliquis, Xarelto, warfarin, daily aspirin) can raise your bleeding risk. We often keep the dose modest, around 1 gram per day, and watch for unusual bruising.
• Atrial fibrillation. High-dose omega-3, around 4 grams per day, has been linked with a small bump in atrial fibrillation risk. REDUCE-IT and STRENGTH both show the rhythm risk rising slightly while major cardiac events generally fall. For metabolically healthy patients the cognitive and longevity benefits usually outweigh that, and if AFib is part of your history we coordinate with your cardiologist. If you feel a thumping or a skip in your chest, tell me and we adjust right away.
• LDL cholesterol. Some patients see a small rise in LDL when they start high-dose omega-3, particularly on DHA-heavy products. I confirm with an ApoB test, which counts the harmful particles. If ApoB is rising along with LDL, we adjust the formulation or the dose.
• Statins. Omega-3 does not block statins. The combination has its strongest data in people with stubbornly high triglycerides or known plaque on imaging.

What we recommend, and what we dont

• We look for: the re-esterified triglyceride (rTG) form, IFOS or USP certified for purity and freshness, with a verified combined EPA plus DHA dose printed on the label. If you eat plant-based, or fish does not agree with you, algae oil gives you bio-identical EPA and DHA from the same algae the fish eat to build their own omega-3.
• Worth considering: krill oil, which delivers omega-3 in the phospholipid form and absorbs efficiently, if cost is less of a concern. Prescription Vascepa (icosapent ethyl, 4 grams per day) is purified EPA only, and it carries strong randomized trial data for reducing major cardiac events in high-risk patients on a statin who have high triglycerides. We choose between them based on your risk profile, your insurance, and your lab values.
• We avoid: one-a-day multivitamin fish oil at 300 mg, which is too small to move a biomarker, and ethyl ester bulk oils with no third-party testing. We also avoid ALA from flax or walnuts as a substitute, since you convert under 5% of it to EPA and DHA and usually fall short. And we do not go above 4,000 mg per day without a clear rationale and a way to monitor.

Guidance from the Clinic

"I often see patients taking fish oil for years, yet their Omega-3 Index is still sitting at 4%. Usually, it is a dosing or absorption issue. It is not enough to swallow the capsule. We need to verify that the EPA and DHA are getting into your cell membranes. Read for combined EPA plus DHA, pick the rTG form, take it with a full meal, and recheck your index at 3 months. That is most of the game."

Dr. Ash

If you'd like us to source it for you:

A note on cost: any discount we negotiate on professional-grade supplements passes straight through to you, with no markup. Here is how we choose and source supplements.

Scientific References

1. Harris, W. S., et al. (2017). The Omega-3 Index and relative risk for coronary heart disease mortality: Estimation from 10 cohort studies. Atherosclerosis, 262, 51-54.
2. Bhatt, D. L., et al. (2019). Cardiovascular Risk Reduction with Icosapent Ethyl for Hypertriglyceridemia (REDUCE-IT). New England Journal of Medicine, 380, 11-22.
3. Mozaffarian, D., & Wu, J. H. Y. (2011). Omega-3 Fatty Acids and Cardiovascular Disease: Effects on Risk Factors, Molecular Pathways, and Clinical Events. Journal of the American College of Cardiology, 58(20), 2047-2067.
4. Yokoyama, M., et al. (2007). Effects of eicosapentaenoic acid on major coronary events in hypercholesterolaemic patients (JELIS). Lancet, 369(9567), 1090-1098.