A fillet of farmed salmon is shaped by its feed. The fat in the fish mirrors the fat in its diet, and as farms replaced fish oil with plant oils, the omega-3 fats EPA and DHA in fillets fell. In over 3,000 Scottish salmon, a 2015 portion had to be twice the size of a 2006 portion to give the same recommended amount, and Norway’s food safety committee put EPA, DPA and DHA in Norwegian farmed salmon at about half the 2006 level by 2014. Even so, farmed salmon still delivers more EPA plus DHA than most other fish, say the Scottish authors. The USDA entry on this page shows 1.67 g in a 3 oz raw serving, though fillets today vary with the feed.
Contaminants made farmed salmon controversial. In 2004 a study in Science analyzed over 2 metric tons of salmon and reported more organochlorine contaminants, a group that includes PCBs and dioxins, in farmed than in wild fish, with the highest levels in European farms. Later Norwegian data show them falling: dioxins and dioxin-like PCBs in Norwegian farmed salmon dropped between 1999 and 2011, and in 2014 the Norwegian Scientific Committee for Food and Environment put them at about 30% of the 2006 level. Mercury is low, and the FDA and EPA list salmon as a Best Choice.
The pink color is a feed effect too. Salmon take their color from astaxanthin, a pigment made by algae, and on farms it is added to the feed. US rules cap it at 80 mg per kilogram of feed and require the fish to be labeled, and a lab study found wild fillets held about 10 times more of it. Antibiotic use also depends on where the fish was farmed, as covered below. Wild fish have a page of their own: wild Atlantic salmon.
Health benefits of farmed Atlantic salmon
None of the studies below reaches strong evidence, and none follows people eating farmed salmon for years. The best-supported results come from small feeding trials, mostly 4 to 18 weeks long, in which salmon raised blood omega-3 and in some trials lowered triglycerides, and in which salmon raised on rapeseed oil still raised the omega-3 index. A trial of 205 children found no effect on overall IQ, and a Cochrane review of omega-3 trials, mostly capsules, found little effect on deaths. Here is what 25 studies found, strongest evidence first, from feeding trials and cohorts to fillet analyses of omega-3 and contaminants, with a few animal studies at the end. Several Norwegian trials had authors employed by a feed maker or a salmon producer, and the cards say so.
Children who ate farmed Atlantic salmon three times a week for 16 weeks did not score higher on overall intelligence tests than children who ate meat meals
Randomized trial in 205 children aged 4 to 6. Blood EPA and DHA rose only in the salmon group, but the main test scale scores did not differ between groups. Two of eight raw subtest scores (symbol search and picture concepts) improved more with salmon, which the authors call a possible modest effect on some skills, not on global IQ.
Raising long-chain omega-3 intake made little or no difference to deaths from any cause and lowered triglycerides by about 15%
Cochrane review of 86 randomized trials (162,796 people) lasting at least 12 months. Most trials gave EPA and DHA as capsules, not fish, and the authors say there is little evidence on eating fish itself. Heart disease events fell slightly (risk ratio 0.91), with low-certainty evidence.
Two portions a week of farmed salmon raised on rapeseed oil raised the omega-3 index as much as salmon raised on fish oil in healthy adults
Randomized trial in 51 healthy adults over 18 weeks, compared with no extra salmon. Fillets held 2.1 g of EPA plus DHA per 100 g on the fish-oil feed and 0.9 g on the rapeseed-oil feed. The rapeseed-oil group also had higher vitamin D status and lower triglycerides than the control group, while heart rate was lower only in the fish-oil group at 9 weeks.
In a Norwegian trial, salmon fed fish oil lowered triglycerides and two inflammation markers in heart patients, while salmon fed rapeseed oil did not
Double-blind trial in 60 people with coronary heart disease who ate about 700 g a week for 6 weeks of salmon fed fish oil, rapeseed oil or half of each. Blood fatty acids mirrored the fillets and the feeds. It compared feeds with each other, not salmon with no salmon.
Show 21 more studies
Salmon fed more vegetable oil carried fewer dioxins, PCBs and flame retardants, while mercury and lead were similar in all fillets
Preliminary follow-up of 58 heart patients from the Norwegian trial above, who ate 700 g a week of salmon for 6 weeks. Contaminant levels in fillets were highest with fish-oil feed and lowest with rapeseed-oil feed. Patients’ plasma dioxin and PCB levels fell over the 6 weeks, and plasma flame retardants (PBDEs) rose only in the fish-oil group.
Two portions of farmed salmon a week in pregnancy raised EPA by 80% and DHA by 90% in breast milk five days after birth
Randomized trial in 123 women who rarely ate oily fish, from week 20 of pregnancy until delivery, compared with their usual diet. The salmon gave 3.45 g of long-chain omega-3 fats a week. Breast milk secretory IgA, an immune factor, was lower in the salmon group, and the authors say its meaning is unclear.
In a Norwegian trial, 750 g a week of fatty fish (farmed salmon) for 8 weeks improved blood sugar 2 hours after a meal in overweight adults, but lean fish (cod) did not
Randomized trial in 68 overweight or obese adults, with a no-fish control group. Fasting glucose did not change, and fatty fish raised the omega-3 content of white blood cell membranes. Two authors work for a fish feed maker and a salmon producer.
In a Norwegian trial, 750 g a week of farmed salmon for 4 weeks lowered triglycerides and raised HDL cholesterol in healthy adults compared with chicken
Randomized trial in 38 young, healthy, normal-weight adults. Cod, the lean fish, did not do this, and glucose regulation did not differ between groups. Two authors work for a fish feed maker and a salmon producer.
In a Norwegian trial in autumn, five salmon dinners a week slowed but did not stop the fall in vitamin D in adults with overweight
Randomized trial at 60° north latitude in 63 adults who ate 750 g a week of salmon or cod for 8 weeks, or kept their usual diet. Serum vitamin D3 fell in all groups, though less with salmon. Other vitamins in the blood did not change. Two authors work for a fish feed maker and a salmon producer.
In a Norwegian trial, 750 g a week of salmon for 8 weeks changed the gut bacteria profile but not blood lipids in overweight adults
Randomized trial in 76 adults with overweight or obesity (gut bacteria analyzed in 33), comparing salmon, cod or no fish. Fecal output and serum bile acids did not change. The paper reports funding from the Bergen Medical Research Foundation; two authors work for a fish feed maker and a salmon producer, which the paper says were not involved in on-site data collection.
Eating farmed Atlantic salmon twice a week for 4 weeks lowered triglycerides and raised HDL cholesterol in overweight adults
Crossover trial in 19 adults who ate 90, 180 or 270 g of salmon twice a week for 4 weeks each, with a 4 to 8 week break between. Large LDL particles rose with the dose, a shift the authors link to lower heart disease risk. It was a small study.
200 g of smoked farmed salmon a week for 8 weeks did not differ from controls in total cholesterol, HDL or triglycerides in women with overweight, though LDL fell by 8.2 mg/dL against a 9.5 mg/dL rise in controls
Randomized study of 38 pair-matched women aged 18 to 30. Among women with a high waist-to-height ratio, triglycerides rose 50.0 mg/dL in the salmon group, so the authors call for more research.
Three fish meals a week, built on farmed Atlantic salmon and albacore tuna, raised the omega-3 index from 5.63% to 7.33% in women with overweight over 8 weeks
Secondary analysis of a randomized trial in 42 women. The comparison group ate shrimp and imitation crab, and its index stayed flat (5.90% to 5.80%). Body fat and starting omega-3 status explained 83% of the response. The meals mixed salmon with tuna, so this is not salmon alone. Funded by USDA, the National Dairy Council and Campbell Soup Co.
Eating the most fatty fish was linked to an 8% lower rate of coronary heart disease, while lean fish showed no such link
Meta-analysis of 19 reports from prospective studies. The highest fatty fish intake also went with 17% fewer coronary deaths and 3% lower total mortality. This covers fatty fish in general, not farmed salmon, and cohorts cannot prove cause.
People who ate the most fish had a 6% lower risk of dying from any cause than those who ate the least
Meta-analysis of 12 prospective cohorts (672,389 people, 57,641 deaths). Compared with never eating fish, 60 g a day went with a 12% lower risk. It covers fish in general, not farmed salmon, and cohorts cannot prove cause.
Farmed Scottish salmon from 2015 needed double the portion size of 2006 salmon to supply the recommended EPA plus DHA, after feeds shifted to plant oils
Fatty acid analysis of over 3,000 Scottish Atlantic salmon farmed from 2006 to 2015, with no people tested. Plant fatty acids rose as EPA and DHA fell. The authors note that this salmon still delivers more EPA plus DHA than most other fish and all land livestock.
Farmed salmon carried significantly more organochlorine contaminants than wild salmon, and European-raised fish had the most
Chemical analysis of over 2 metric tons of farmed and wild salmon from around the world, with no people tested. The authors judged that eating farmed Atlantic salmon may pose risks that detract from the benefits of fish. The paper drew published replies in Science, and later data below show lower levels.
Dioxins, dioxin-like PCBs, mercury, arsenic and DDT fell in Norwegian farmed salmon between 1999 and 2011
Analysis of more than 2300 fillet samples collected by the Norwegian Food Safety Authority. Excluding other foods, 2011 dioxin and PCB levels allowed up to 1.3 kg of salmon a week before reaching the then tolerable weekly intake. The paper drew a published comment and a reply.
Wild Atlantic salmon had higher levels of dioxins, PCBs and mercury than farmed salmon in one comparison, but farmed fish had less DHA
Analysis of wild and farmed Atlantic salmon for contaminants and fatty acids, with no people tested. Farmed fillets had more total fat and more omega-6, while EPA did not differ. All levels were well below EU limits, and a 200 g portion a week gave almost twice the EPA plus DHA that EFSA considers adequate (1.75 g a week).
UK farmed salmon products still gave more EPA and DHA per portion than wild salmon because they hold far more fat
Analysis of salmon products from major UK retailers. Wild salmon had a higher share of long-chain omega-3 fats (20 to 40% against 9 to 26%) but less total fat (1 to 6% against 7 to 17%). Farmed products with more vegetable-oil fatty acids had a lower EPA and DHA share, and products varied widely.
A benefit-risk calculation found the recommended EPA plus DHA could not be reached from farmed or wild salmon alone without exceeding an acceptable cancer-risk level
Risk model built on contaminant data for farmed Atlantic and wild Pacific salmon, with no people tested. The benefit-risk ratio was greater for wild Pacific salmon as a group, though some farmed subgroups matched it. The authors advise women of child-bearing age and young children to choose the least contaminated fish. It predates the later fall in contaminant levels.
In one laboratory analysis, farmed salmon held about 25% of the vitamin D of wild salmon
Vitamin D3 measured in several fish species, and the effect of baking and frying. Content varied widely even within a species, and the authors say food tables may be out of date. A 2014 Norwegian assessment found vitamin D in farmed salmon appeared unchanged since 2006.
Wild salmon flesh held about 10 times more astaxanthin than farmed salmon, the pigment that colors the fillet
Laboratory digestion model and intestinal cell cultures, with no people tested. Cooking cut astaxanthin by 48 to 57% in wild and 35 to 47% in farmed fillets. Of the astaxanthin in uncooked fillets, 43% reached the digestive micelles for wild salmon against 12% for farmed.
Mice fed farmed salmon raised on soybean oil developed more insulin resistance and liver fat
Mouse trial using fillets from salmon fed fish oil or with 80% of it replaced by rapeseed, olive or soybean oil. Vegetable oils cut PCB and DDT levels by more than 50%, but the soybean-oil fish, high in linoleic acid, worsened insulin resistance, while rapeseed oil improved glucose tolerance. A feed company produced the salmon, and an author is funded by it.
Rats on a high-fat diet with crude oil from farmed salmon developed insulin resistance, belly fat and fatty liver, while refined oil did not
A 28-day rat study plus fat cell experiments that tested the pollutant mixtures found in the crude oil. It tested fish oil, not fillets, so it cannot show what a portion of salmon does in people.
Who should be careful with farmed Atlantic salmon
Most people can eat farmed salmon freely. Take a little care if one of these applies:
- Pregnancy and smoked salmon. The FDA and EPA put salmon on the Best Choices list for mercury and advise 2 to 3 servings of 4 oz a week in pregnancy and breastfeeding. Refrigerated smoked fish such as lox is on the CDC list of foods to avoid in pregnancy, at 65 and older and with a weak immune system, because of Listeria.
- Fish allergy. Fish is one of nine major food allergens in the US, and labels must name the species. Reactions can be serious and children usually do not outgrow it, so talk to an allergist before eating salmon if you have a fish allergy.
- Eating it several times a week. Norwegian analysis of 2011 samples allowed up to 1.3 kg of farmed salmon a week within the dioxin and PCB limit, excluding other foods, but EFSA cut that limit from 14 to 2 pg per kg of body weight a week in 2018. Norway’s 2022 assessment found 2 to 3 fish meals a week would take almost all age groups past the new limits, yet judged the benefits larger than the risks.
How to choose, store and eat farmed Atlantic salmon
- Raw or cooked: the numbers on this page are for raw fish, and the FDA advises cooking most seafood to 145°F (63°C). Cooking cut astaxanthin by 35 to 47% in farmed fillets in a lab test. Norway’s food safety authority says farmed Norwegian salmon can be eaten raw without freezing, because its heat-treated feed leaves parasites unlikely.
- Color and label: astaxanthin in the feed gives the flesh its pink to orange-red color, and US rules say the presence of the color additive in the fish must be declared.
- Origin and antibiotics: use differs by country. Norwegian vets prescribed 548 kg of antibacterials for all farmed fish in 2023, while Chilean salmon farms used 382.5 metric tons in 2016, about 0.53 kg per ton of harvested salmon.
Choose fillets that smell fresh and mild with firm flesh that springs back when pressed, keep them at 40°F (4°C) or below, and use them within two days. Norway’s VKM judged residues of veterinary medicines, pesticides and other feed contaminants in Norwegian farmed fish to be very low, often not detectable, and not a food safety concern.
Sources
- Advice about eating fish. FDA and EPA. Content current as of 5 March 2024.
- Mercury levels in commercial fish and shellfish (1990-2012). FDA.
- Listeria prevention. CDC.
- Food allergies. FDA.
- Sharp MF, Lopata AL. Fish allergy: in review. Clin Rev Allergy Immunol. 2014.
- Benefit and risk assessment of fish in the Norwegian diet: an update of the report from 2006 based on new knowledge. VKM report 2014:15. 2014.
- Benefit and risk assessment of fish in the Norwegian diet. VKM report 2022:17. 2022.
- EFSA Panel on Contaminants in the Food Chain. Risk for animal and human health related to the presence of dioxins and dioxin-like PCBs in feed and food. EFSA J. 2018.
- 21 CFR 73.35: Astaxanthin. US Code of Federal Regulations.
- Farmed Atlantic salmon and rainbow trout are safe for sushi and sashimi. Norwegian Food Safety Authority (Mattilsynet).
- Selecting and serving fresh and frozen seafood safely. FDA.
- Miranda CD, Godoy FA, Lee MR. Current status of the use of antibiotics and the antimicrobial resistance in the Chilean salmon farms. Front Microbiol. 2018.
- NORM/NORM-VET 2023: usage of antimicrobial agents and occurrence of antimicrobial resistance in Norway. Norwegian Veterinary Institute and Norwegian Institute of Public Health. 2024.
History of farmed Atlantic salmon
Atlantic salmon, Salmo salar, lives in rivers and seas around the North Atlantic. People were already fertilizing its eggs by hand in Germany in 1763, and by the late 1800s hatcheries in Europe, North America and Japan raised young salmon to restock rivers, not to sell as food.
Modern salmon farming with floating sea cages began in Norway in the late 1960s. Norwegian industry history credits the brothers Ove and Sivert Grøntvedt on the island of Hitra, who put salmon in a floating net pen on 28 May 1970 and harvested the first generation in 1971. A Norwegian trade magazine argues they were not first and points to the brothers Olav and Karstein Vik at Sykkulven, who it says put salmon and trout in floating cages in 1959.
Farms then spread to North America in the early 1970s, to Chile and New Zealand in the late 1970s and to Australia in 1985. In 1996 world farmed salmon production passed the wild salmon harvest, and by 2022 farmed salmon and trout came to about 4.2 million metric tons against about 1.4 million wild.
Norway also worked to sell the fish abroad. A delegation went to Tokyo in 1985 to persuade Japan that raw Norwegian salmon belonged in sushi. According to the Norwegian Seafood Council, an industry body, Norway exported 2 tonnes of salmon to Japan in 1980 and more than 45,000 tonnes a year two decades later.
The fish itself is odd. Its genome, sequenced in 2016, shows that the ancestor of all salmonids copied its entire genome about 80 million years ago. Feeds changed too: fish oil made up 24% of Norwegian salmon feed in 1990 and 11% in 2013, according to the 2016 Scottish study cited above.
Sources
- Salmon farming. Wikipedia.
- Celebrating 50 years of modern aquaculture. Norwegian Seafood Council.
- Myten om Hitra. iLaks (in Norwegian).
- Norwegian salmon continues to lead global sushi category 40 years on from its introduction to Japan. Norwegian Seafood Council press release, 16 June 2025.
- Sprague M, Dick JR, Tocher DR. Impact of sustainable feeds on omega-3 long-chain fatty acid levels in farmed Atlantic salmon, 2006-2015. Sci Rep. 2016.
- Lien S, Koop BF, Sandve SR, et al. The Atlantic salmon genome provides insights into rediploidization. Nature. 2016.
- Astaxanthin. Wikipedia.
Frequently asked questions about farmed Atlantic salmon
Is farmed salmon good for you?
For most people, yes. A 3 oz (85 g) raw serving has 177 calories, 17.4 g of protein, 114% of the daily value of vitamin B12 and 47% of vitamin D, per the USDA entry on this page. In trials farmed salmon raised blood omega-3 levels, while a Cochrane review of omega-3 trials, mostly capsules, found little effect on deaths. Cohorts link fatty fish to an 8% lower rate of coronary heart disease, but they cannot prove cause.
Is farmed salmon healthier than wild salmon?
Neither is clearly healthier. Farmed fillets hold more fat (208 against 142 calories per 100 g on this site), and a UK analysis found they still gave more EPA plus DHA per portion. Wild fish had more vitamin D and astaxanthin in lab tests, and contaminant studies point both ways: Hites found more in farmed fish in 2004, while a later comparison found more in wild fish. See wild Atlantic salmon for its numbers.
Is farmed salmon still high in omega-3?
Yes, but lower than it used to be. The USDA entry on this page shows 2.51 g of total omega-3 and 1.97 g of EPA plus DHA per 100 g raw. Scottish fillets lost much of their EPA and DHA from 2006 to 2015, and Norway reported about half the 2006 level by 2014. Two portions a week of salmon with only 0.9 g of EPA plus DHA per 100 g still raised the omega-3 index in a trial of 51 adults.
Is farmed salmon safe to eat?
Yes for most people, according to official risk assessments. Norway’s VKM found in 2014 that the benefits clearly outweigh the negligible risk from current contaminant levels, and in 2022 that benefits outweigh risks at 2 to 3 fish meals a week. Dioxins and dioxin-like PCBs in Norwegian farmed salmon fell to about 30% of 2006 levels. EFSA lowered the tolerable weekly intake for dioxins and PCBs in 2018, so eating a lot every week can still pass it.
Is farmed salmon dyed?
The pink color comes from astaxanthin added to the feed, a pigment wild salmon also take in from what they eat. US rules cap it at 80 mg per kilogram of feed and require the fish to be labeled. In a lab comparison, wild fillets held about 10 times more astaxanthin than farmed ones.
Does farmed salmon have mercury?
Only a little. FDA tests found a mean of 0.022 ppm mercury in fresh or frozen salmon, though the table does not split farmed from wild, and the FDA and EPA list salmon as a Best Choice. They advise 2 to 3 servings of 4 oz a week in pregnancy and breastfeeding. Norway found mercury in farmed Atlantic salmon had fallen to about 50% of its 2006 level by 2014.
Does farmed salmon contain antibiotics?
Farmed fish can be treated with antibiotics, but residues in Norwegian farmed fish were very low and often not detectable, according to VKM in 2014. Use differs by country: Norwegian vets prescribed 548 kg for all farmed fish in 2023, while Chile used 382.5 metric tons in 2016. Ask where your fish was farmed if this matters to you.
How much farmed salmon should you eat a week?
Official advice covers fish in general, about 2 portions a week. The Dietary Guidelines for Americans advise at least 8 oz of seafood a week, and Norway advises 300 to 450 g of fish with at least 200 g fatty fish. No trial here tested an ideal amount of farmed salmon, and trials used from 2 portions a week up to 750 g a week.