Story
Fish Oil · EPA / DHA Supplement
Last updated
In one pass What actually does the work in fish oil is two long-chain omega-3s: EPA and DHA. Not this — Fish oil 1000 mg = 1 g EPA+DHA — 1000 mg is the total weight of the fish oil; the EPA and DHA in it are usually only 300-500 mg. Read the active content, not the headline number.
Educational content, not medical advice — consult a clinician.
Story path
Chapter 1
Choosing a form and checking freshness
What actually does the work in fish oil is two long-chain omega-3s: and . Whether a bottle is worth buying and can be trusted comes down mainly to two things: which chemical form it is made in (which decides how much you absorb), and how fresh it is (with time it oxidizes and spoils).
There are three common forms. The form that fish naturally carry (TG), and the re-esterified triglyceride form (rTG), which is purified and then put back on a glycerol backbone, absorb best. The ethyl-ester form (EE), a product of industrial distillation, is cheapest but absorbs a step worse, and worse still on an empty stomach. In krill oil, EPA and DHA are attached to phospholipids; makers say this absorbs well, but each capsule holds little and it costs the most. When choosing, prefer rTG or TG.
Freshness gets overlooked even more than form. In one survey of fish oils on sale in New Zealand (Albert 2015), most products exceeded the limits for oxidation markers. Over-oxidized fish oil has already lost part of its EPA and DHA, and it causes nausea and fishy burps; whether swallowing it harms the body is not settled. So choose a product with a freshness certification where you can, refrigerate it after opening, and finish it within 2–3 months. If it smells strongly fishy or makes you queasy, throw it out.
There are three common forms. The form that fish naturally carry (TG), and the re-esterified triglyceride form (rTG), which is purified and then put back on a glycerol backbone, absorb best. The ethyl-ester form (EE), a product of industrial distillation, is cheapest but absorbs a step worse, and worse still on an empty stomach. In krill oil, EPA and DHA are attached to phospholipids; makers say this absorbs well, but each capsule holds little and it costs the most. When choosing, prefer rTG or TG.
Freshness gets overlooked even more than form. In one survey of fish oils on sale in New Zealand (Albert 2015), most products exceeded the limits for oxidation markers. Over-oxidized fish oil has already lost part of its EPA and DHA, and it causes nausea and fishy burps; whether swallowing it harms the body is not settled. So choose a product with a freshness certification where you can, refrigerate it after opening, and finish it within 2–3 months. If it smells strongly fishy or makes you queasy, throw it out.
Mechanism · Absorption and oxidation, one molecule
Which form you buy and how fresh it is look like two unrelated questions, but both land on the same molecule. Walk through it once and it clicks.1 · Why ethyl esters (EE) absorb slowly
The gut cannot absorb fat as it is. Once oil reaches the small intestine, bile first breaks it into very fine droplets, and then lipase from the pancreas clips the fatty acids off their backbone one by one. Only the clipped fatty acids can cross the cells of the gut wall.
The natural form (TG / rTG) has a glycerol backbone. That is exactly the shape pancreatic lipase handles every day, so the clipping goes smoothly.The ethyl-ester form (EE) is a product of industrial distillation: the glycerol on the backbone has been swapped for ethanol. The same scissors cut this joint much more slowly, so the food mass has already moved downstream before the and are released.
So taking it on an empty stomach is even worse is not superstition. On an empty stomach, little bile is released and pancreatic lipase has not been woken up, so an already slow cut gets slower; taking it with a meal that contains fat is like calling that step to work. The idea behind rTG is to run the ethyl-ester step first to purify out contaminants, then put the fatty acids back on a glycerol backbone, so you get both purity and a backbone that is easy to clip.
2 · Why fish oil spoils on the shelf
Omega-3s are useful because of the double bonds along their chains (how those work is in the chapter What EPA and DHA each do), and they are fragile because of the same double bonds.
The carbon sitting between two double bonds holds its hydrogen especially loosely. Oxygen from the air pulls that hydrogen away, and the chain becomes a damaged molecule with an unpaired electron, which then turns and steals a hydrogen from the neighboring chain. One oxidized chain sets off the next. So oxidation is not a steady fade; once it starts, it collapses in a cascade.
The end products of this chain reaction are a set of small aldehydes. They are:
The rancid smell you noticeThe fishy taste that comes back up when you burpThe wave of nausea after a dose
So throwing it out when it smells strongly fishy is not fussiness about taste: what your nose is detecting is the oxidation products themselves. Refrigeration, darkness, brown bottles, and finishing the bottle soon after opening all slow down how fast this chain reaction gets started, because heat and light both ignite it.
DHA has even more double bonds than EPA, so high-DHA formulas are more fragile by nature. That is also why it comes in sealed softgels rather than an open syrup.
Numbers · How to read the milligrams on the label
The most common trap on a Fish Oil label is the phrase "1000 mg fish oil":1000 mg of fish oil is not 1000 mg of plus .A typical "1000 mg fish oil" capsule contains only 180 mg EPA plus 120 mg DHA, 300 mg of omega-3 in total.The American Heart Association (AHA) advises people who already have coronary heart disease to get about 1 g of EPA plus DHA a day; with these capsules that means 3–4 a day.Only concentrated products (60–80% omega-3) reach 700–800 mg per capsule.
Some common target doses
Most people: if you eat fatty fish at least twice a week, you usually do not need a supplement.Existing cardiovascular disease, or raised (TG, a line on the blood-lipid report): EPA plus DHA 1–2 g/day, following your doctor's advice.The prescription dose used in REDUCE-IT: icosapent ethyl (a highly purified EPA ethyl ester), 4 g/day, EPA only. The trial enrolled people already on a statin whose triglycerides were still raised and who had cardiovascular disease, or diabetes plus other risk factors.Pregnancy: at least 200 mg of DHA a day is the amount most often recommended (the fetal brain and retina need a lot of DHA).Upper limit: the safety reference usually quoted from the US FDA is 3 g a day of EPA and DHA combined. It is not a formal Tolerable Upper Intake Level (), but the higher the dose, the more care bleeding deserves.
Fish oil and bleeding
VITAL, at 1 g a day, saw no increase in bleeding.In REDUCE-IT, at 4 g a day, serious bleeding was slightly more common (2.7% vs 2.1%), a difference that did not reach statistical significance.Stopping 7 days before surgery is a cautious approach. If you take warfarin or an antiplatelet drug, tell your doctor before adding fish oil; it is usually a minor issue.
How the three forms differ
Natural triglyceride (TG) and re-esterified triglyceride (rTG): TG is the form fish naturally carry. rTG is first purified as EE and then put back on a triglyceride backbone, which combines purity with absorption.Ethyl ester (EE): the glycerol on the triglyceride is replaced with ethanol, a common product of industrial distillation and purification. In the Dyerberg 2010 trial (72 volunteers, about 3.3 g of EPA plus DHA a day for 2 weeks), with natural fish oil set at 100%, bioavailability was 124% for rTG and only 73% for EE; the free-fatty-acid form was 91%, not significantly different from natural fish oil. EE does even worse on an empty stomach. Most cheap fish oils, and the icosapent ethyl used in REDUCE-IT, are EE.Phospholipid-bound (krill oil): EPA and DHA are attached to phospholipids. Makers often say it absorbs best, but there are not enough head-to-head comparisons; each capsule holds little EPA plus DHA, it costs the most, and the red antioxidant pigment astaxanthin comes along as a by-product.
The TOTOX value: how freshness is measured
TOTOX = 2 × peroxide value (PV) + anisidine value (AV). The limit commonly used in the industry is TOTOX < 26 (GOED and the EU), and below 10 is often treated as excellent.Shelf surveys are not encouraging. Of the 32 fish oils Albert 2015 bought in New Zealand, 83% exceeded the peroxide limit, 50% exceeded the TOTOX limit, and only 8% met every limit; most also contained less EPA plus DHA than the label claimed.Look for IFOS or GOED certification, keep it refrigerated, and finish it within 2–3 months of opening. A strong fishy smell or nausea means it has oxidized: throw it out right away.
dyerberg-2010-ee-vs-tg
Chapter 2
What EPA and DHA each do
and are both omega-3s in fish oil. Chemically they are almost the same chain, but the body uses them for somewhat different jobs.
DHA is mainly building material. It is built into cell membranes and keeps that layer soft. The retina and the brain's gray matter are unusually rich in DHA: anywhere proteins must change shape quickly in the membrane to pass a signal along needs a soft membrane.
EPA is more of a messenger. The raw material the body uses to make inflammatory signals (arachidonic acid, an omega-6) looks a lot like EPA, so EPA can squeeze into the same enzymes' seats and fewer pro-inflammatory signals get made. The body can also turn it into a class of molecules thought to help wind inflammation down.
So which one to choose is not about which is better; it is about what you want it to do. For lowering , EPA and DHA both help; only the cardiovascular trial results for a prescription pure-EPA drug rest on EPA alone. For pregnancy and brain development, the amount of DHA is what matters.
DHA is mainly building material. It is built into cell membranes and keeps that layer soft. The retina and the brain's gray matter are unusually rich in DHA: anywhere proteins must change shape quickly in the membrane to pass a signal along needs a soft membrane.
EPA is more of a messenger. The raw material the body uses to make inflammatory signals (arachidonic acid, an omega-6) looks a lot like EPA, so EPA can squeeze into the same enzymes' seats and fewer pro-inflammatory signals get made. The body can also turn it into a class of molecules thought to help wind inflammation down.
So which one to choose is not about which is better; it is about what you want it to do. For lowering , EPA and DHA both help; only the cardiovascular trial results for a prescription pure-EPA drug rest on EPA alone. For pregnancy and brain development, the amount of DHA is what matters.
Mechanism · Where each chain ends up
and differ chemically by only 2 carbons and 1 double bond, but they end up in different places in the body.EPA (20:5, eicosapentaenoic acid): leaning toward messenger
It is the raw material for one family of inflammation-resolving molecules (the E-series resolvins).It competes with arachidonic acid (AA, an omega-6) for two enzyme systems, COX and LOX, so fewer pro-inflammatory prostaglandins get made.The high-dose pure-EPA REDUCE-IT trial (NEJM 2019): in high-risk cardiovascular patients already on a statin whose were still raised, 4 g of icosapent ethyl a day lowered major adverse cardiovascular events () by 25% in relative terms.
DHA (22:6, docosahexaenoic acid): leaning toward membrane building material
In the rod outer segment (the part of the retina that handles dim-light vision), about half of the membrane fatty acids are DHA.In the brain's gray matter, DHA makes up about 15–20% of the fatty acids.It is a key lipid for keeping membranes fluid and the receptors in them working.It is also the raw material for another set of resolving molecules (the D-series resolvins and the protectins).Fetal brain development in pregnancy and visual development in infants depend mainly on DHA.
Can they convert into each other?
EPA → DHA: partly (it has to go through the full → EPA → DPA → DHA chain).DHA → EPA: partly (retroconversion).Neither direction is efficient. If you want one of them, take that one directly.
So
Lowering triglycerides: both EPA and DHA work. Reproducing REDUCE-IT's cardiovascular result takes that prescription pure EPA, prescribed by a doctor.Pregnancy, infants, or a focus on brain development: DHA cannot be skipped.Balanced supplements (EPA:DHA roughly 1.5:1 to 2:1) suit general use.The purest EPA belongs to REDUCE-IT-type indications, not everyday wellness.
Mechanism · A soft membrane, and a fight for a seat
is building material and is a messenger: behind each of those sentences is a complete physical process. Walk through them and you can reason out other situations yourself.DHA · A chain with six bends holds up a soft membrane
Saturated-fat chains are straight. Lined up side by side, they pack as neatly and tightly as a bundle of chopsticks. Each extra double bond on a chain adds a bend that will not straighten out.
DHA bends six times. Once it is inserted into a cell membrane, that patch can no longer pack neatly or squeeze tight, so the membrane stays soft and able to change shape.
Why the retina and gray matter in particular need this soft membrane:
The rod outer segment is a stack of extremely thin membrane discs. The instant light arrives, the light-catching protein in the membrane must change shape at once to pass the signal on. If the membrane were stiff, that change would have to push against resistance and would come a beat late. The soft membrane DHA holds open gives that movement room.Synaptic membranes in gray matter work the same way: receptors and ion channels are all embedded in the membrane, and they work by changing shape. How soft or stiff the membrane is decides how much effort those proteins need to change shape.
So DHA is not a vague "good for the eyes, good for the brain" claim. It is the floor under the fast shape-changing proteins in those two organs. A soft floor is what makes fast movement possible.
EPA · Grabbing a seat in an enzyme's pocket
First, where do inflammatory signals come from?
Normally, arachidonic acid (AA, a long-chain omega-6 fatty acid) does not float in the blood; it hangs on the phospholipids of cell membranes. When a stimulus arrives, an enzyme clips it off the membrane and hands it to two enzyme systems, COX and LOX, which turn it into prostaglandins and leukotrienes. The redness, swelling, heat, and pain you feel are that batch of molecules at work.
EPA's chain length and shape are close enough to AA's that it can sit in the same enzyme pocket. Then two things happen at once:
While EPA occupies the pocket, that enzyme molecule cannot get to AA. The same stimulus produces fewer pro-inflammatory molecules.Even when EPA does get processed, its products are milder.
Joining the two lines · The membrane is both building material and an ammunition store
Notice the small detail above: AA is clipped off the membrane. EPA and DHA hang on the membrane in the same way.
That gives the key link: at the moment a stimulus arrives, whether the enzyme clips off AA or EPA depends on what your membrane is built from. Whether the omega-3 you eat actually gets swapped into your membranes decides not only how soft they are, but also which raw material is at hand when a signal is sent.
This also explains something often treated as a mystery: fish oil has to be taken for several months before you can talk about an effect. Swapping membranes is slow. Cells load new fatty acids into phospholipids a little at a time, and one meal does not change the stock.
A more active layer than "less inflammation" (still debated)
One theory holds that inflammation does not simply burn itself out. Instead the body sends a separate set of wind-down signals: calling in the cells that clear up, eating the debris, and returning blood vessels to normal. It is a program that has to be switched on. EPA and DHA are the raw materials for these wind-down molecules (resolvins and protectins).
On this theory, running short of raw material does not make inflammation burn hotter; it means inflammation cannot be wound down. It fails to end when it should and drags on as long-term low-grade inflammation. To be clear, these wind-down molecules are often present in human blood at very low levels and are hard to measure, and how much they do in the human body is not settled. For now this layer is a plausible mechanism that has not been confirmed.
Even so, keeping less inflammation and inflammation that can end apart is useful, because the two failures look completely different. Once you tell them apart, you can see why the word anti-inflammatory is so often misused in supplement ads.
Evidence · Why the big trials reversed the story
The history of randomized trials of fish oil is one of the most dramatic reversals in nutrition medicine.2002–2010 (the optimistic years): early trials such as GISSI-Prevenzione (patients after a heart attack) and JELIS (Japanese patients with high cholesterol, given pure on top of a statin) found fewer cardiovascular events in people taking fish oil.
2018–2020 (recalibration): several larger, more rigorous randomized trials found no benefit in the general population.
The VITAL trial (NEJM 2019, N = 25,871, median follow-up 5.3 years)
Who: middle-aged and older adults in the US with no cardiovascular disease and no cancer, taking 1 g of EPA plus a day or a placebo.Primary endpoints (major cardiovascular events, invasive cancer): no significant difference.A secondary endpoint: 28% fewer heart attacks overall ( 0.72, 95% 0.59–0.90). A secondary endpoint is a lead, not a conclusion.Heart attacks fell by 77% in the Black subgroup. The number is large, but it comes from a subgroup analysis and needs another trial to replicate it.
The ASCEND trial (NEJM 2018, N = 15,480, mean follow-up 7.4 years)
Who: adults with diabetes but no cardiovascular disease yet, taking 1 g of omega-3 a day or an Olive Oil placebo.Serious vascular events: no significant difference ( 0.97, P=0.55).
The STRENGTH trial (JAMA 2020, N = 13,078)
Who: statin-treated patients at high cardiovascular risk with raised and low high-density lipoprotein () cholesterol, the so-called good cholesterol, taking 4 g a day of a carboxylic-acid form of EPA plus DHA, compared with corn oil.An interim analysis found a low chance of benefit and the trial was stopped early. Major cardiovascular events were almost identical in the two groups, and atrial fibrillation was about 70% more common.This stands in sharp contrast to the positive result REDUCE-IT got with 4 g of pure EPA.
Why REDUCE-IT and STRENGTH disagree
Same dose (4 g), both in high-risk cardiovascular patients, yet one positive and one negative.Two main explanations, neither settled:Different ingredients: REDUCE-IT used EPA only and raised blood EPA higher; STRENGTH used EPA plus DHA.Different comparators: REDUCE-IT's placebo was mineral oil, which some argue worsened lipids and inflammatory markers in the control group and so widened the gap; STRENGTH's comparator was corn oil.For the use of lowering cardiovascular events, what the US FDA approved is icosapent ethyl, not ordinary fish oil.
Where the consensus stands
Healthy people taking fish oil to prevent heart disease: the large trials mostly saw no benefit.Very high triglycerides (> 500 mg/dL): prescription-dose omega-3 is one of the first-line options for lowering them.Very high cardiovascular risk plus raised triglycerides: a doctor decides whether to use icosapent ethyl (Vascepa).Eating fatty fish (salmon, sardines, mackerel) at least twice a week: this is the advice of the US Dietary Guidelines Advisory Committee (DGAC) and the AHA. It rests mainly on observational studies in which people who eat more fish have less heart disease. That is an association, not a trial, and it asks a different question from the supplement trials (the chapter Eating fish or taking capsules explains why).
nicholls-2020-strengthascend-2018
Chapter 3
Eating fish or taking capsules
The omega-3 in one salmon meal is roughly equal to a small handful of Fish Oil capsules. That conversion already tells you something: to match eating fish with capsules, you would have to swallow quite a few.
But the bigger difference is not the amount. When you eat fish, you are not eating omega-3 alone. The same piece of fish carries good-quality protein, vitamin D, iodine, selenium, and vitamin B12. And a meal with fish is a meal without something else, often red or processed meat.
Nutrition science calls this the food-matrix effect: what does the work may be the whole food, and the meal it replaced, rather than one molecule pulled out on its own. In supplement trials the capsule is added on top of the usual diet, and nothing is displaced. So studies of eating fish (mostly observational) and trials of capsules are not answering the same question.
At the table it is simple: eat fatty fish two or three times a week, which is also what dietary guidelines advise, and most people do not need a supplement. It is people who do not eat fish who need another plan.
But the bigger difference is not the amount. When you eat fish, you are not eating omega-3 alone. The same piece of fish carries good-quality protein, vitamin D, iodine, selenium, and vitamin B12. And a meal with fish is a meal without something else, often red or processed meat.
Nutrition science calls this the food-matrix effect: what does the work may be the whole food, and the meal it replaced, rather than one molecule pulled out on its own. In supplement trials the capsule is added on top of the usual diet, and nothing is displaced. So studies of eating fish (mostly observational) and trials of capsules are not answering the same question.
At the table it is simple: eat fatty fish two or three times a week, which is also what dietary guidelines advise, and most people do not need a supplement. It is people who do not eat fish who need another plan.
Numbers · How many capsules one fish meal equals
Eat fish or take fish oil? Start with the numbers.Omega-3 in common fatty fish ( plus , per 100 g cooked)
Salmon (Atlantic, farmed): 1.8–2.5 gSardines (canned in oil): 1.4–1.5 gMackerel: 1.0–2.7 g (varies a lot by season)Cod: 0.2–0.3 g (a lean fish, so low)Tuna (canned): 0.2–0.5 g (water-packed cans lose some)Trout: 0.8–1.5 g
For comparison, one standard Fish Oil capsule has about 0.3 g of omega-3. Two 100 g salmon meals a week come to about 20 capsules a week, or 3 a day.
But eating fish brings far more than omega-3
Good-quality protein (about 20–25 g per 100 g)Vitamin D (fatty fish is one of the few natural sources)Iodine (sea fish)Selenium (sea fish and shellfish)Vitamin B12Fullness, and it replaces red and processed meat
In observational studies, people who eat more fatty fish have a lower risk of heart disease, while randomized trials of omega-3 capsules in the general population were mostly negative. That cannot be read as the evidence for fish is stronger. The first is an association, and people who eat a lot of fish tend to be healthier in other ways too; and the two kinds of study are not asking the same question (the food-matrix effect below).
In practice
Fatty fish 2–3 times a week (≥ 150 g each time): most people get enough omega-3 from food and do not need a supplement.No fish, or vegetarian: consider algal oil for DHA, 200–500 mg/day.Pregnancy: because of mercury, choose low-mercury fish (salmon, sardines, trout) and avoid large predatory fish such as shark, swordfish, and bigeye tuna.Liver disease, or taking an anticoagulant: talk to your doctor first.
Mechanism · What separates a fish meal from a capsule
Food-matrix effect is often treated as a pleasant-sounding excuse. It actually points to something specific: three things that each make sense on their own.1 · Fish delivers omega-3 in the setting where it is easiest to absorb
The chapter Choosing a form and checking freshness explains that fat has to be broken up by bile and clipped by pancreatic lipase before it can cross the gut wall. How efficient that step is depends heavily on the setting.
The and in fish flesh already sit in phospholipids and , and they reach the small intestine with a whole meal containing fat and protein. Plenty of bile is released and the lipase is awake: the best conditions for that clipping step.Swallowing an ethyl-ester capsule on an empty stomach is the worst condition for the same step.
The same number of milligrams on a label does not mean the same amount reaches your blood.
2 · Fish packs the omega-3 together with what protects it
What polyunsaturated fat fears most is oxidation (the chain reaction that collapses in a cascade). Sea fish are also one of the main sources of selenium.
In the body, selenium is built into a class of selenoproteins, and one kind (glutathione peroxidase) has the job of taking apart lipid peroxides that have already formed, which cuts that chain reaction short. Following the mechanism, fish sells the flammable goods and the fire extinguisher in one package; this has not been tested on its own in people. Capsules usually contain only the flammable goods, so manufacturers have to add antioxidants, seal them, and keep them from light.
The same piece of fish also carries vitamin D. It and omega-3 are both fat-soluble and cross the gut wall in the same bile-salt micelles, so the fat in that meal carries both across at once.
3 · The fish meal displaces what you would otherwise have eaten
This is the easiest point to miss, and the one that best explains why the evidence seems to conflict.
In observational studies, people who eat fish are using it to replace another meal, often red or processed meat. Part of the association comes from what the fish brought, part from what that meal did not contain, and part from the lifestyles of people who eat fish.In supplement trials, people who take capsules are adding one on top of their usual diet, and nothing is displaced.
So the two kinds of study are not asking the same question. One asks what happens if you swap a meal; the other asks what happens if you add a little more omega-3. Their answers differ not because they contradict each other, but because the designs differ.
So you cannot set the association with eating fish in observational studies against the results of fish-oil trials and declare one side's evidence stronger. A supplement can copy only the one molecule. It cannot copy the absorption setting or the selenium and vitamin D that come along, and it certainly cannot copy the processed meat that got displaced. That is also why dietary guidelines recommend eating fish rather than starting with capsules.
Safety · Mercury, pollutants, and fish in pregnancy
The risks that really matter when you eat fish:① Methylmercury
It concentrates up the food chain: large, long-lived predatory fish carry the most.High mercury: shark, swordfish, marlin, bigeye tuna, tilefish from the Gulf of Mexico, king mackerel, orange roughy.Low mercury: salmon, sardines, mackerel (Atlantic mackerel, not king mackerel), trout, shrimp, scallops, tilapia.Weighing and against mercury: the pooled analysis by Mozaffarian and Rimm 2006 concluded that for most adults the benefits outweigh the risks; pregnant women, children under 5, and women of childbearing age need to choose fish with particular care.
② Persistent organic pollutants (POPs, such as PCBs and dioxins)
Often higher in farmed fish (feed concentrates them).Modern certified fish oil is distilled and purified, which removes most PCBs and mercury.For products labeled for purity, IFOS 5-star certification means something.
③ Microplastics and PFAS (perfluorinated compounds)
Microplastics and PFAS have been detected in most fish (they are also in water and salt, not only fish).Their health effects are still being studied, but they are not a reason to give up fish.
④ Fish-oil oxidation (covered in the chapter Choosing a form and checking freshness): in practice a more common supplement-quality problem than mercury or PCBs.
Fish in pregnancy and breastfeeding (the joint advice of the US Food and Drug Administration, FDA, and the Environmental Protection Agency; here EPA is the agency, not the fatty acid)
2–3 servings a week (about 8–12 oz, or 230–340 g) of low-mercury fish.Avoid: shark, swordfish, marlin, orange roughy, tilefish from the Gulf of Mexico, bigeye tuna, king mackerel.Limit: yellowfin tuna (1 serving a week) and canned white (albacore) tuna.Neither agency recommends eating no fish at all in pregnancy, because DHA matters too much for the fetus.
In short: for the average adult, 2–3 meals of low-mercury fatty fish a week, with no supplement on top, is the most common advice today. It rests on observational studies of eating fish plus dietary guidelines; and because large Fish Oil trials in the general population were mostly negative, the order fish first, then think about capsules makes even more sense.
fda-epa-fish-advice-2021
Chapter 4
Can flaxseed oil replace fish oil?
Flaxseed, chia seeds, and walnuts do contain omega-3, but a different one: (alpha-linolenic acid). It is shorter than the two in fish and has fewer double bonds.
The catch is that to turn ALA into usable and , the body has to lengthen the chain itself and then open new double bonds along it. That conversion line is slow and congested: only a small fraction makes it all the way through, and even less reaches DHA.
So flaxseed oil cannot replace fish oil. It is not a matter of eating more of it; the process itself does not get through.
For people who do not eat fish, the real route is algal oil. Fish do not make much omega-3 themselves; they build it up by eating algae (through the food chain). Algal oil skips the fish and takes ready-made DHA and EPA straight from the source.
The catch is that to turn ALA into usable and , the body has to lengthen the chain itself and then open new double bonds along it. That conversion line is slow and congested: only a small fraction makes it all the way through, and even less reaches DHA.
So flaxseed oil cannot replace fish oil. It is not a matter of eating more of it; the process itself does not get through.
For people who do not eat fish, the real route is algal oil. Fish do not make much omega-3 themselves; they build it up by eating algae (through the food chain). Algal oil skips the fish and takes ready-made DHA and EPA straight from the source.
Mechanism · Where the conversion line jams
Turning into and then into is an assembly line with several steps. It is slow, and the slowness sits in two very specific places. Walk through them and you can explain for yourself why no amount of extra flaxseed oil makes up the DHA.Bottleneck 1 · The machine that opens double bonds is shared with omega-6
Opening a new double bond on a fatty-acid chain is done by a class of enzymes called desaturases. The first machine on this line is Δ6-desaturase (its gene is called FADS2).
The trouble is that it is also the first step on the omega-6 line. The linoleic acid in your meals (from the most common cooking oils: soybean, corn, and sunflower) is turned into arachidonic acid on the same machine.
And a modern diet usually contains far more linoleic acid than ALA. The two raw materials queue at the same window, the one with the numbers takes most of it, and ALA can only shuffle forward slowly. This is one reason that simply eating more flaxseed oil raises blood EPA only modestly: the raw material is not what runs short; the window is.
Bottleneck 2 · The last step to DHA takes a detour
The body has no machine that can open one more double bond directly on a chain that is already that long. To make DHA, it has to detour:
First lengthen the chain past the targetOpen the double bond on this longer versionThen send the whole chain to another workshop in the cell (the peroxisome)There, clip off two carbons to get DHA, and ship it back
One extra loop, plus a trip between workshops. So this last step loses the most. Converting to EPA is already hard and converting to DHA is harder. These are not two separate pieces of bad news; they are one line that narrows the further along you go.
Women convert a little more, and that is the same line
The Burdge 2006 review reports that women convert a higher share of ALA into EPA and DHA than men, possibly because of a regulating effect of estrogen. Looking at pregnancy, that makes sense: the fetal brain and retina are exactly where DHA is used most, and the supply can come only from the mother. But the body widening the narrow road when it is needed most is only a plausible explanation; it has not been directly shown.
So what to do in practice
Keep eating flaxseed, walnuts, and chia. ALA itself is an essential fatty acid, and these foods bring much more than ALA.But if your goal is to raise blood DHA, do not push harder on the conversion line; that is fighting a channel that narrows by design. Supply the finished product directly: algal oil.The DHA in algal oil and in fish oil is biochemically the same molecule. The only difference is which link of the food chain it comes from: microalgae are where it is made, and fish are just a warehouse in the middle.
Numbers · Conversion rates, sources, the ω-6 debate
For vegetarians, plant omega-3 can stand in for fish only partly; this is one of the topics nutrition most often oversimplifies.Sources of (alpha-linolenic acid, 18:3 ω-3)
Flaxseed oil: 53–57% ALAChia seeds: 18%Walnuts: 9%Hemp seeds: 22%Soybean oil and canola oil: 7–10%
The key point: the body converts only a small share of ALA into and
ALA → EPA: 5–8% (slightly higher in women, possibly linked to estrogen)EPA → DHA: 0.5–5%Overall ALA → DHA: 0.1–4%
These are estimates from different tracer studies. They vary widely, so read them only as orders of magnitude.
This means someone who eats no fish and relies only on flaxseed oil will have clearly lower EPA and DHA in the blood, but that is not the same as complete deficiency, since ALA itself is also an essential fatty acid. Long-term vegetarians have 20–50% less EPA and DHA in their blood than fish eaters, yet their cardiovascular event rates are not correspondingly higher, possibly because their overall diet is healthier (this is what has been observed; the reason has not been proven).
Algal oil is the real source of DHA and EPA for vegetarians
DHA and EPA extracted from microalgae are biochemically the same as those in fish oil (fish get their omega-3 by eating algae).A typical product: 200–500 mg of DHA per capsule.It costs more (about 3–5 times the price of fish oil).Vegetarians who are pregnant or breastfeeding are often advised to take it.
The omega-6 to omega-3 ratio
The modern Western diet: about 15–20 : 1.Estimates for our evolutionary past: about 1–4 : 1."Too much omega-6 is the root of inflammation" is a popular claim, but the evidence is weak. Randomized trials that replaced saturated fat with omega-6 (mainly linoleic acid) mostly saw low-density lipoprotein () cholesterol, the so-called "bad" cholesterol, fall. Results on cardiovascular events were not fully consistent, and the American Heart Association's 2017 advisory (Sacks 2017) comes down overall in favor of this swap.So the point is not cutting omega-6 but bringing the absolute amount of omega-3 up.There is no need to worry about omega-6; focus directly on your total omega-3 intake.
Evidence · What ALA does on its own
Even without being converted to and , itself is an essential fatty acid: the body cannot make it. Its roles:1. A source of energy: like other fatty acids, it can be burned by beta-oxidation.
2. A small share goes into membrane phospholipids, adding to membrane fluidity.
3. The heart (separately from EPA and DHA): the American Heart Association's 2017 advisory on dietary fats (Sacks 2017) favors replacing saturated fat with unsaturated fat. Looking at ALA alone, pooled cohort studies show that people who eat more ALA have about 10% lower all-cause mortality. That is an observed association and cannot show that ALA caused it. Some suggest part of the effect works through fewer heart-rhythm disturbances, but that step has not been proven.
4. Slightly better inflammation markers (limited evidence): diets high in ALA lower C-reactive protein (, a common blood marker of inflammation) and a little.
Practical advice (vegetarian, or no fish)
1–2 tablespoons of flaxseed a day (ground; whole seeds pass through the gut almost intact): about 2–3 g of ALAA handful of walnuts a day (28 g): about 2.5 g of ALA1–2 tablespoons of chia seeds a day: about 5 g of ALAAlgal DHA on top, 200–500 mg/day (especially in pregnancy, while breastfeeding, and for children under 5)Do not cook with flaxseed oil: polyunsaturated fats oxidize very easily when heated
Claims that do not hold up
Flaxseed oil equals fish oil: no; conversion is too low.A vegan diet needs no supplements at all: vitamin B12, DHA, and often iron and zinc all deserve attention.Omega-3 and omega-6 must be tuned to 1:1 to be healthy: no evidence supports such an extreme ratio.
naghshi-2021-ala-mortality
Chapter 5
Which conditions it helps
Does fish oil work? is the wrong question. The right one is: what do you want it to treat?
The same bottle, used for different things, has very different strength of evidence behind it. The two firmest uses are not wellness at all: bringing very high down with prescription doses, and giving a prescription pure- drug to specific high-risk cardiovascular patients. Taking in pregnancy is common advice, based mainly on the mechanism (the fetus needs a lot of DHA) and on expert consensus.
One tier down are rheumatoid arthritis, as an add-on treatment, and EPA-dominant formulas in depression: there is a signal, but it is not firm.
Further down are the claims that show up most in ads. Cancer prevention, protecting the brain, and protecting the eyes were all tested directly in large trials and showed no effect; making you smarter or slowing aging have no trials behind them at all.
So the real place of fish oil is an add-on treatment for specific patients, plus a way to restock for people who do not eat fish, not a wellness product everyone should take.
The same bottle, used for different things, has very different strength of evidence behind it. The two firmest uses are not wellness at all: bringing very high down with prescription doses, and giving a prescription pure- drug to specific high-risk cardiovascular patients. Taking in pregnancy is common advice, based mainly on the mechanism (the fetus needs a lot of DHA) and on expert consensus.
One tier down are rheumatoid arthritis, as an add-on treatment, and EPA-dominant formulas in depression: there is a signal, but it is not firm.
Further down are the claims that show up most in ads. Cancer prevention, protecting the brain, and protecting the eyes were all tested directly in large trials and showed no effect; making you smarter or slowing aging have no trials behind them at all.
So the real place of fish oil is an add-on treatment for specific patients, plus a way to restock for people who do not eat fish, not a wellness product everyone should take.
Mechanism · How high-dose fish oil pulls triglycerides down
Lowering is the firmest use of fish oil. The process happens in the liver and is worth walking through; once you have, you will also see why this job needs a dose far above the everyday wellness amount.First, where blood triglycerides come from
The triglycerides measured in a fasting blood test are mostly not the oil you just ate. They are what the liver packaged and shipped out itself.
Liver cells assemble fatty acids into triglycerides, load them into a cargo hold, and wrap it in a shell protein called , making a transport particle called VLDL (very-low-density lipoprotein) that is released into the blood. The more the liver ships, the higher blood triglycerides go. So to lower them, this assembly line has to slow down.
Once and reach liver cells, they slow the line in three places at once (these three come from mechanistic studies)
The raw material is awkward to use. The enzymes that assemble triglycerides do not handle these two bent long chains well, so less can be loaded into the cargo hold than with ordinary fatty acids.The switch is flipped toward burning. Liver cells have a switch (PPARα) that decides a fatty acid's fate: package it and ship it, or send it into the mitochondria to be burned. Omega-3 flips it toward burning, so another batch of raw material waiting to be packaged disappears.A half-full truck does not leave. If there is not enough cargo, the apoB shell protein is broken down on the spot; the truck is dismantled before it leaves the factory.
The three add up to one result: the liver sends fewer VLDL particles into the blood. Blood triglycerides come down with them.
This also explains why the dose has to be so high
Notice that what has to change is not the speed of one reaction but the running rate of a whole organ's assembly line. For that, the omega-3 concentration inside liver cells has to genuinely turn over. To put it the way the chapter What EPA and DHA each do does: the membranes and the stock have to be swapped first.
The everyday wellness amount cannot do that, so the dose used to lower triglycerides is several times higher, and it has to be taken for some time before blood lipids show a change. This is not a manufacturer upselling; it is what the mechanism requires.
Which also explains why a pure-EPA product is a drug, not a wellness product
The dose it uses is already in the range a doctor needs to watch. A dose high enough to change how an organ behaves is also high enough to have other consequences: high-dose omega-3 lengthens bleeding time, and large trials have also seen more atrial fibrillation (the chapter What EPA and DHA each do compares the trials).
The dose that can change physiology and the dose that brings side effects often arrive together. That is not only about fish oil: whenever you see a supplement whose treatment dose is far above its wellness dose, you can reason through it this way first.
Evidence · Indication by indication
Beyond heart disease, fish oil ( plus ) has other evidence-based uses, but the strength of the evidence varies widely. The groups below are sorted by type of evidence, not by letter grade.Firmer evidence (several randomized trials, or large trials of a prescription drug)
Very high (TG > 500 mg/dL): prescription doses of EPA plus DHA, 2–4 g/day, lower triglycerides by 20–50%.Icosapent ethyl (Vascepa) 4 g plus a statin: in high-risk cardiovascular patients already on a statin whose triglycerides were still raised, major cardiovascular events fell by 25% in relative terms (REDUCE-IT).
Mainly advice (expert consensus plus mechanism)
More than 200 mg of DHA a day in pregnancy: the fetal brain and retina need a lot of DHA, and this is common prenatal nutrition advice.
A signal, but not firm
Rheumatoid arthritis (RA): 3-4 months of supplementation reduced morning stiffness (SMD -0.43) and the use of painkillers (SMD -0.40), pooling 17 randomized trials (the Goldberg 2007 ). SMD, the standardized mean difference, puts results from different scales onto one ruler. Note: that 3-4 in the paper is the number of months of treatment, not g/day; the paper gives no dose figure.Add-on treatment for major depression: in the Liao 2019 meta-analysis, the benefit appeared in the tier where EPA is ≥ 60% of the formula and the EPA dose is ≤ 1 g/day; DHA-dominant formulas showed none. Note that the dose line is an upper bound, not a lower bound: taking more is not the lever here.ADHD in children: the Bloch 2011 meta-analysis found a small effect (SMD about 0.31); it is not a first-line treatment.Recovery after exercise: limited evidence that it reduces muscle soreness.Allergy and asthma: supplementing in pregnancy and early infancy may slightly lower a child's allergy risk.
Tested in large trials, with no effect found
Dry eye: the 2018 DREAM trial found no significant benefit (some earlier small studies had been optimistic).Age-related macular degeneration: in the AREDS2 trial, adding omega-3 gave no extra benefit.Cognitive function and Alzheimer's disease: most randomized trials were negative. Yurko-Mauro 2010 saw a little improvement with DHA alone, but the ADCS DHA trial (Quinn 2010, JAMA) was negative in people who already had Alzheimer's disease.Cancer prevention: VITAL was negative.Cardiovascular prevention in people with diabetes: ASCEND was negative.
No trials in support
Making you smarter, slowing aging: marketing language with no randomized trial behind it.
So the real place of fish oil
An add-on treatment for specific patients (very high triglycerides, rheumatoid arthritis, certain types of major depression)A source of omega-3 for people who do not eat fishDHA in pregnancyNot an all-purpose wellness supplement for healthy people
goldberg-2007-omega3-joint-pain-metaliao-2019-omega3-depression-metaquinn-2010-adcs-dhaareds2-2013-jamaascend-2018
In practice · Four questions before you buy
If you have decided to take fish oil, answering these four questions first helps you avoid most substandard products.1. What exactly are you trying to solve? (the indication)
Heart health without a specific diagnosis: eat fish first rather than taking a supplement; if you do buy one, 1 g/day of plus is enough. > 500: 2–4 g/day (a prescription dose, decided by a doctor).Pregnancy: DHA-dominant, 200–500 mg/day.Rheumatoid arthritis or other arthritis: take it for 3–4 months (that is the treatment length, not g/day; Goldberg 2007 gives no daily dose).Add-on treatment for depression: an EPA-dominant formula (≥ 60% EPA), with an EPA dose ≤ 1 g/day.If you cannot answer, do not buy.
2. How much EPA plus DHA is actually in the bottle?
Read the nutrition panel, not the big print on the front.1000 mg of fish oil is not 1000 mg of omega-3 (see the chapter Choosing a form and checking freshness).Target: at least 500 mg of EPA plus DHA per capsule before calling it concentrated.
3. Which form is it (TG, rTG, EE)?
rTG or TG beats EE (in Dyerberg 2010, EE's bioavailability was only 73% of natural fish oil's).If you buy EE, take it with a meal that contains fat.Vascepa (icosapent ethyl) is an EE, but it is EPA only, 4 g a day, and needs a prescription: it is a treatment, not a wellness product.
4. Oxidation (TOTOX) and certification
IFOS 5-star or GOED certification.TOTOX < 10 is excellent.A fishy smell, nausea, or fishy burps: it has already oxidized, so return it.Refrigeration and brown glass beat clear plastic.Finish it within 2–3 months of opening.
If you see any of these three, put it down
It says deep-sea fish oil but does not list how many milligrams of EPA and DHA: the label is fudging.100 capsules at an absurdly low price: most likely EE, with high oxidation.It promises beauty, anti-aging, or making you smarter: that is outside the evidence.
liao-2019-omega3-depression-metagoldberg-2007-omega3-joint-pain-metaquinn-2010-adcs-dhadyerberg-2010-ee-vs-tg
References · 7
- National Institutes of Health, Office of Dietary Supplements. (2023). Omega-3 Fatty Acids — Fact Sheet for Health Professionals. ods.od.nih.gov/factsheets/Omega3FattyAcids-HealthProfessional
- Albert, B. B., Derraik, J. G., Cameron-Smith, D., Hofman, P. L., Tumanov, S., Villas-Boas, S. G., Garg, M. L., & Cutfield, W. S. (2015). Fish oil supplements in New Zealand are highly oxidised and do not meet label content of n-3 PUFA. Scientific Reports, 5, 7928. 10.1038/srep07928
- Bhatt, D. L., et al. (2019). Cardiovascular risk reduction with icosapent ethyl for hypertriglyceridemia (REDUCE-IT). NEJM, 380(1), 11–22. 8,179 statin-treated patients with CVD or diabetes plus risk factors, triglycerides 135-499 mg/dL; icosapent ethyl 4 g/day vs placebo, median 4.9 years. Primary end point 17.2% vs 22.0%, HR 0.75 (0.68-0.83); key secondary HR 0.74; CV death 4.3% vs 5.2%. Atrial fibrillation or flutter hospitalisation 3.1% vs 2.1%; serious bleeding 2.7% vs 2.1% (P = 0.06). Funded by Amarin (abstract, PMID 30415628). 10.1056/NEJMoa1812792
- Manson, J. E., et al. (2019). Marine n-3 fatty acids and prevention of cardiovascular disease and cancer (VITAL). NEJM, 380(1), 23–32. 10.1056/NEJMoa1811403
- 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. 10.1016/j.jacc.2011.06.063
- Sacks, F. M., Lichtenstein, A. H., Wu, J. H. Y., Appel, L. J., Creager, M. A., Kris-Etherton, P. M., et al. (2017). Dietary fats and cardiovascular disease: A presidential advisory from the American Heart Association. Circulation, 136(3), e1-e23. Coconut oil is ~82% saturated; the AHA advises against its use because it raises LDL-C with no known offsetting favorable effect, and recommends replacing saturated with unsaturated fat. 10.1161/CIR.0000000000000510
- Burdge, G. C. (2006). Metabolism of α-linolenic acid in humans. Prostaglandins, Leukotrienes and Essential Fatty Acids, 75(3), 161–168. Conversion of dietary ALA is limited; the 2002 young-women / young-men tracer papers are different studies (EPA ~21% / DHA ~9% in women; little DHA enrichment in men) and are not this record. 10.1016/j.plefa.2006.05.013