Picture the canned fish aisle. A can of tuna you know by heart, a salmon fillet in the chiller, and a flat tin of sardines your family treats with suspicion. All three are called healthy. They are not the same.
For omega-3, salmon and sardines are well ahead of tuna. A 4 oz (113 g) serving of raw farmed salmon holds about 2.22 g of EPA plus DHA, canned sardines about 1.11 g, and raw skipjack tuna 0.29 g (our arithmetic on USDA values [1][2]). For mercury, sardines and salmon sit at the bottom of the FDA’s table, while tuna runs from low to high depending on the kind [3].
So the short answer is salmon or sardines for omega-3, and tuna as a lean protein you buy with one eye on the label. Below: the numbers side by side, what trials say about the heart, and the weekly amounts the FDA and EPA suggest.
Which fish has the most omega-3?
Fish earn their reputation from two omega-3 fats, EPA and DHA. The bars show both for 4 oz (113 g), the serving size the FDA and EPA use in their advice [4]. The values are the USDA numbers on our fish pages, scaled to that weight [1][2].
Drawn to scale. Raw fish, except sardines, which are canned in oil and drained, with bones. Our arithmetic: grams per 100 g times 1.13. Data: USDA FoodData Central via NutriScrolls.
Atlantic mackerel leads with 2.60 g, then farmed salmon at 2.22 g and herring at 1.78 g. Wild Atlantic salmon has 1.62 g and the canned sardines 1.11 g. Raw skipjack tuna trails at 0.29 g, about a ninth of the mackerel.
Farmed beating wild surprises people, so check per calorie. Wild salmon is leaner, at 142 against 208 kcal per 100 g, and gives 1.01 g of EPA plus DHA per 100 kcal against 0.95 g for farmed. Count grams of fish and farmed wins. Count calories and the two are close.
Farmed salmon has one caution: its omega-3 follows its feed. In more than 3,000 Scottish farmed salmon, EPA and DHA fell between 2006 and 2015 as farms replaced fish oil with plant ingredients, so a portion had to double to meet recommended intakes, though it still beat most other fish (Sprague et al., 2016). Read the USDA figure as a snapshot.
Tuna needs a footnote. Ours is raw skipjack, the species in most light cans [1]. Canned tuna is not on the site, so we have no canned omega-3 figure and will not guess one. Here are the three fish of the title side by side (sardines canned, the others raw).
| Per servingPer 100 g | Farmed Atlantic Salmonper 3 oz (85 g) | Sardinesper 1 can (92 g) | Tunaper 3 oz (85 g) |
|---|---|---|---|
| CaloriesEnergy | 177 kcal740 kJ / 177 kcal | 191 kcal801 kJ / 191 kcal | 88 kcal366 kJ / 88 kcal |
| Protein | 17.4 g (35%) (35%) | 22.7 g (45%) (45%) (higher) | 18.7 g (37%) (37%) |
| EPA + DHA | 1.67 g (higher) | 0.9 g | 0.22 g |
| Vitamin D | 9.4 µg (47%) (187%) (higher) | 4.4 µg (22%) (88%) | No data |
| Vitamin B12 | 2.7 µg (114%) (110%) | 8.2 µg (343%) (329%) (higher) | 1.6 µg (67%) (65%) |
| Calcium | 7.65 mg (<1%) (<1%) | 351 mg (27%) (44%) (higher) | 24.7 mg (2%) (3%) |
| Selenium | 20.4 µg (37%) (37%) | 48.5 µg (88%) (88%) (higher) | 31 µg (56%) (56%) |
| Saturated fat | 2.59 g (13%) (13%) (higher) | 1.41 g (7%) (7%) | 0.28 g (1%) (1%) |
| CaloriesEnergy | 208 kcal870 kJ / 208 kcal | 208 kcal870 kJ / 208 kcal | 103 kcal431 kJ / 103 kcal |
| Protein | 20.4 g (41%) (41%) | 24.6 g (49%) (49%) (higher) | 22 g (44%) (44%) |
| EPA + DHA | 1.97 g (higher) | 0.98 g | 0.26 g |
| Vitamin D | 11 µg (55%) (220%) (higher) | 4.8 µg (24%) (96%) | No data |
| Vitamin B12 | 3.2 µg (135%) (129%) | 8.9 µg (373%) (358%) (higher) | 1.9 µg (79%) (76%) |
| Calcium | 9 mg (<1%) (1%) | 382 mg (29%) (48%) (higher) | 29 mg (2%) (4%) |
| Selenium | 24 µg (44%) (44%) | 52.7 µg (96%) (96%) (higher) | 36.5 µg (66%) (66%) |
| Saturated fat | 3.05 g (15%) (15%) (higher) | 1.53 g (8%) (8%) | 0.33 g (2%) (2%) |
● = higher amount. Source: USDA FoodData Central (FDC 175167, 175139, 175156).
The fish trade strengths. Per 100 g, sardines have 8.9 µg of vitamin B12 against 3.2 µg in farmed salmon, and 382 mg of calcium against 9 mg, while salmon has 11 µg of vitamin D against 4.8 µg. Shop for omega-3 alone and you land on salmon, mackerel or herring. Shop for the rest too and sardines hold their own.
How much mercury is in each fish?
The FDA publishes the average mercury it found in each fish it tested, in a table called Mercury Levels in Commercial Fish and Shellfish (1990-2012) [3]. The bars use its mean in parts per million (ppm), colored by the list each fish falls on in the FDA and EPA advice [4].
Drawn to scale. Means from the FDA table (page current as of February 2022); the Best, Good and Avoid lists are from the FDA and EPA advice revised in October 2021. Data: FDA.
Tuna is where the table gets interesting. Canned light tuna averages 0.126 ppm over 545 samples. Canned albacore averages 0.350 ppm over 451 samples, almost three times as much. Fresh or frozen bigeye averages 0.689 ppm, which puts it on the Choices to Avoid list. All three are sold as tuna. The FDA also lists fresh or frozen skipjack at 0.144 ppm, but from only 3 samples.
A study of 168 cans bought in New Jersey agrees: white-style tuna averaged 0.407 ppm and light-style 0.118 ppm. White tuna is albacore, a relatively larger species than the skipjack in light cans, and draining the can made no difference (Burger and Gochfeld, 2004).
The low end is just as clear. Sardines average 0.013 ppm and fresh or frozen salmon 0.022 ppm, so albacore carries about 27 times the mercury of the same weight of sardines (our arithmetic). The salmon row covers 94 samples and does not split farmed from wild. The Atlantic mackerel value comes from a 1978 report, and the word mackerel also covers king mackerel, at 0.73 ppm a fish to avoid.
A can that says only tuna tells you little. Light is the low-mercury end of the tuna aisle, and albacore or white sits in the middle.
Which fish gives the most omega-3 for the least mercury?
Put the two tables on one chart and a pattern shows. Up is more omega-3 in a 4 oz serving, right is more mercury, so the best corner is the top left.
Omega-3 from USDA samples and mercury from separate FDA samples, so each dot is a pairing of two averages. Salmon has one FDA mercury value, shared by farmed and wild. Tuna is raw skipjack, with mercury from 3 FDA samples. Canned albacore (0.350 ppm) would sit off the chart to the right, but our site has no omega-3 figure for it. Data: USDA FoodData Central via NutriScrolls; FDA.
Mackerel, salmon and herring float high, and sardines are lowest on the mercury axis. Skipjack tuna sits alone at the lower right with the least omega-3 and the most mercury of the six. Salmon, sardines and herring all give more omega-3 per serving than tuna for less mercury.
How much omega-3 is enough? One review of risks and benefits found that 250 mg a day of EPA plus DHA appears sufficient for primary prevention (Mozaffarian and Rimm, 2006). That is 1.75 g a week. One 4 oz serving of mackerel, farmed salmon or herring covers it, and wild salmon falls just short at 1.62 g. Sardines take about two cans, and raw skipjack would take about six servings (our arithmetic).
A weekly fillet, or sardines on toast twice, gets you there without leaning on tuna.
Does eating fish protect your heart?
Fish and heart health is one of the most studied links in nutrition, and the answer is more careful than the headlines. Pooling cohorts and trials, the JAMA review found that 1 to 2 servings a week went with 36% lower risk of coronary death (95% CI 20% to 50%) and 17% lower total mortality (Mozaffarian and Rimm, 2006). Trials of capsules are less rosy.
Raising omega-3 intake, mostly with capsules, made little or no difference to deaths from any cause across 86 trials
Meta-analysis of 86 randomized trials with 162,796 adults. All-cause mortality had a risk ratio of 0.97 (95% CI 0.93 to 1.01). Coronary heart disease events fell slightly (0.91, 95% CI 0.85 to 0.97, low certainty) and triglycerides by about 15%. Most trials used capsules, and the authors say there is little evidence on eating fish itself.
A daily marine omega-3 capsule did not lower major cardiovascular events in 25,871 adults at usual risk
Placebo-controlled trial in US men 50 and older and women 55 and older, followed a median 5.3 years. Major events occurred in 386 people on omega-3 and 419 on placebo (hazard ratio 0.92, 95% CI 0.80 to 1.06). Heart attacks, a secondary end point, were lower (0.72, 95% CI 0.59 to 0.90). Funded by the National Institutes of Health and others.
One capsule trial did win, REDUCE-IT, which tested a purified EPA drug at a high dose in 8,179 people with high triglycerides on statins, funded by Amarin Pharma. Cardiovascular events were 17.2% against 22.0% on placebo (Bhatt et al., 2019). That is a medicine in a specific group, and it says little about a can of fish.
Two servings of oily fish a week raised red blood cell EPA plus DHA as much as daily capsules
Randomized trial in healthy premenopausal women over 16 weeks. Both groups averaged 485 mg of EPA plus DHA a day, from two weekly servings of salmon and albacore tuna or from 1 to 2 capsules. Red blood cell EPA plus DHA rose from 4.0% to 6.2% in the fish group (11 women) and from 4.3% to 6.2% on capsules. Small, and it measured blood levels, not health outcomes.
Every extra 20 g of fish a day went with a 4% lower risk of death from cardiovascular disease in a pooled analysis of cohort studies
Meta-analysis of 14 cohort studies with 911,348 people and 75,451 deaths; 8 reported cardiovascular mortality (relative risk 0.96, 95% CI 0.94 to 0.98). All-cause mortality: 0.98 (95% CI 0.97 to 1.00). The link was significant only in Asian studies. In Western studies it looked U-shaped, lowest near 20 g a day.
Three points hold together. Capsules have not cut deaths in people at usual risk. Two oily fish servings a week raise blood omega-3 as well as capsules do. And fish eaters tend to live a little longer, though cohorts cannot prove the fish is why. Twenty grams a day is 140 g a week, a little more than one 4 oz serving (our arithmetic). With heart disease or high triglycerides, ask your doctor whether more fish or an omega-3 product suits you.
Is mercury a risk, and who should be careful?
For most adults, not at the levels in the fish above. A large study of two US cohorts looked for exactly this link.
Adults with the highest toenail mercury had no higher risk of cardiovascular disease than those with the lowest
Case-control study nested in two US cohorts of 51,529 men and 121,700 women, with 3,427 people who developed cardiovascular disease matched to controls. Top fifth against bottom fifth of mercury: relative risk 0.85 for total cardiovascular disease (95% CI 0.72 to 1.01), 0.85 for coronary heart disease and 0.84 for stroke, at the exposure levels seen in these adults. Funded by the National Institutes of Health.
Pregnancy is where mercury is taken seriously, because the developing brain is the sensitive target, and the research there shows both sides at once.
Children whose mothers ate no seafood in pregnancy had 1.48 times the odds of a low verbal IQ compared with mothers who ate over 340 g a week
Cohort of 11,875 pregnant women in the UK (ALSPAC), who reported seafood intake at 32 weeks; children were followed to age 8, with adjustment for 28 confounders (odds ratio 1.48, 95% CI 1.16 to 1.90). Weaker prosocial behavior, fine motor, communication and social scores followed the same pattern. The authors argue that advice to limit seafood could be harmful.
Each extra weekly fish serving went with 4.0 points higher infant memory scores, while each 1 ppm of hair mercury went with 7.5 points lower
Cohort of 135 mother-infant pairs in Project Viva (US); mothers ate 1.2 fish servings a week on average and had mean hair mercury of 0.55 ppm, and infants were tested at 6 months. Scores were highest when mothers ate more than 2 servings a week but kept hair mercury at or below 1.2 ppm. A small cohort.
Both studies point the same way: fish helps, mercury hurts, so choose fish with the first and little of the second. The FDA and EPA turn that into a chart [4]. A serving is 4 oz, about the palm of your hand. In pregnancy and breastfeeding, eat 2 to 3 servings a week from the Best Choices list, or 1 serving from the Good Choices list. Children eat 2 servings a week from Best Choices, in portions that grow with age. The Dietary Guidelines for Americans suggest at least 8 oz of seafood a week for adults, and 8 to 12 oz in pregnancy and breastfeeding.
Drawn to scale. Ounces a week from the FDA and EPA advice (revised October 2021, page current as of March 2024). Child totals are our arithmetic: two servings times the serving size for the age. Data: FDA.
Of the fish in this post, salmon, sardines, herring, Atlantic mackerel and canned light tuna are Best Choices. Albacore, fresh or canned, and yellowfin are Good Choices. Bigeye tuna and king mackerel are the two to skip. If you are pregnant, planning a pregnancy or feeding a young child, ask your doctor or midwife how to apply this to you.
Are sardines worth it for the bones?
Sardines have one trick the others lack. Canned in oil and eaten with their soft bones, a can (92 g) holds 351 mg of calcium, 27% of the daily value, and tops our calcium ranking per serving [1].
Drawn to scale. Each food in its own serving size, as on its page. Percent of the US daily value (1,300 mg) is our arithmetic. Data: USDA FoodData Central via NutriScrolls.
Calcium from small soft-boned fish was absorbed about as well as calcium from milk
Absorption test in 19 healthy adults aged 21 to 28. Each meal had about 397 mg of calcium from small Bengali fish or 377 mg from skimmed milk. Absorption was 23.8% from the fish and 21.8% from the milk (p = 0.52). These were small whole fish, not canned sardines.
Nobody has measured calcium absorption from canned sardines themselves, so read the match with milk as a good sign, not a promise. If you avoid dairy, sardines with their bones are still a real option.
A simple way to choose
- Most omega-3 for one meal: Atlantic mackerel, farmed salmon or herring, one 4 oz serving a week covers 1.75 g.
- Best all-round tin: sardines with bones: 1.11 g of EPA plus DHA per 4 oz, 351 mg of calcium a can and the lowest mercury on the FDA table.
- Tuna: the label matters more than the fish name. Pick canned light, a Best Choice at 0.126 ppm, and keep albacore to one serving a week in pregnancy or breastfeeding.
- Pregnant, breastfeeding or feeding young children: stay on the Best Choices list and ask your doctor before changing anything.
Frequently asked questions
Is canned tuna safe to eat every week?
Canned light tuna is a Best Choice on the FDA and EPA list, with a mean of 0.126 ppm mercury across 545 samples. In pregnancy and breastfeeding that list allows 2 to 3 servings of 4 oz a week. Albacore averages 0.350 ppm and counts as a Good Choice, one serving a week. Check the label for light or albacore, and ask your doctor if you are pregnant.
Which has more omega-3, salmon or sardines?
Salmon does per weight. A 4 oz serving of raw farmed salmon has about 2.22 g of EPA plus DHA, against 1.11 g for canned sardines. One can of sardines (92 g) has 0.9 g, and a 3 oz serving of farmed salmon 1.67 g on our food pages. Sardines add 351 mg of calcium a can, which salmon does not.
Is farmed or wild salmon better?
Neither is clearly better. Per 100 g, farmed salmon has 1.97 g of EPA plus DHA and 208 kcal, and wild has 1.44 g and 142 kcal. The FDA gives one mercury figure for fresh or frozen salmon, 0.022 ppm, and does not split farmed from wild. Farmed fillets vary with their feed, so the USDA value is a snapshot.
Can you eat the bones in canned sardines?
Yes, the bones in canned sardines are soft and edible, and they hold much of the calcium. The 351 mg per can on our page is for sardines drained with their bones. In a 19-person test, calcium from small soft-boned fish was absorbed about as well as from milk, but those were not canned sardines.
How much fish should children eat?
The FDA and EPA advise 2 servings a week from the Best Choices list for children. A serving is about 1 oz at ages 1 to 3, 2 oz at 4 to 7, 3 oz at 8 to 10 and 4 oz at 11. Salmon, sardines, herring and canned light tuna are on that list. Ask your pediatrician about your own child.
Sources
- NutriScrolls. Food pages for farmed Atlantic salmon, wild Atlantic salmon, sardines, tuna, mackerel and herring, and the comparison and calcium rankings. nutriscrolls.com/food/seafood/. Accessed 24 Oct 2026.
- U.S. Department of Agriculture, Agricultural Research Service. FoodData Central, entries used on NutriScrolls: salmon, Atlantic, farmed, raw (FDC 175167); salmon, Atlantic, wild, raw (FDC 173686); sardine, Atlantic, canned in oil, drained solids with bone (FDC 175139); tuna, skipjack, raw (FDC 175156); mackerel, Atlantic, raw (FDC 175119); herring, Atlantic, raw (FDC 175116); milk, 2% (FDC 172205). fdc.nal.usda.gov. Accessed 24 Oct 2026.
- U.S. Food and Drug Administration. Mercury Levels in Commercial Fish and Shellfish (1990-2012). Content current as of 25 Feb 2022. fda.gov. Accessed 24 Oct 2026.
- U.S. Food and Drug Administration and U.S. Environmental Protection Agency. Advice about Eating Fish (revised October 2021; content current as of 5 Mar 2024). fda.gov. Accessed 24 Oct 2026.
- Mozaffarian D, Rimm EB. Fish intake, contaminants, and human health: evaluating the risks and the benefits. JAMA. 2006;296(15):1885-1899. doi:10.1001/jama.296.15.1885
- Abdelhamid AS, Brown TJ, Brainard JS, et al. Omega-3 fatty acids for the primary and secondary prevention of cardiovascular disease. Cochrane Database Syst Rev. 2020;3(3):CD003177. doi:10.1002/14651858.CD003177.pub5
- Manson JE, Cook NR, Lee IM, et al. Marine n-3 fatty acids and prevention of cardiovascular disease and cancer. N Engl J Med. 2019;380(1):23-32. doi:10.1056/NEJMoa1811403
- Bhatt DL, Steg PG, Miller M, et al. Cardiovascular risk reduction with icosapent ethyl for hypertriglyceridemia. N Engl J Med. 2019;380(1):11-22. doi:10.1056/NEJMoa1812792
- Harris WS, Pottala JV, Sands SA, Jones PG. Comparison of the effects of fish and fish-oil capsules on the n-3 fatty acid content of blood cells and plasma phospholipids. Am J Clin Nutr. 2007;86(6):1621-1625. doi:10.1093/ajcn/86.5.1621
- Jayedi A, Shab-Bidar S, Eimeri S, Djafarian K. Fish consumption and risk of all-cause and cardiovascular mortality: a dose-response meta-analysis of prospective observational studies. Public Health Nutr. 2018;21(7):1297-1306. doi:10.1017/S1368980017003834
- Hibbeln JR, Davis JM, Steer C, et al. Maternal seafood consumption in pregnancy and neurodevelopmental outcomes in childhood (ALSPAC study): an observational cohort study. Lancet. 2007;369(9561):578-585. doi:10.1016/S0140-6736(07)60277-3
- Oken E, Wright RO, Kleinman KP, et al. Maternal fish consumption, hair mercury, and infant cognition in a U.S. cohort. Environ Health Perspect. 2005;113(10):1376-1380. doi:10.1289/ehp.8041
- Mozaffarian D, Shi P, Morris JS, et al. Mercury exposure and risk of cardiovascular disease in two U.S. cohorts. N Engl J Med. 2011;364(12):1116-1125. doi:10.1056/NEJMoa1006876
- 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;6:21892. doi:10.1038/srep21892
- Burger J, Gochfeld M. Mercury in canned tuna: white versus light and temporal variation. Environ Res. 2004;96(3):239-249. doi:10.1016/j.envres.2003.12.001
- Hansen M, Thilsted SH, Sandström B, et al. Calcium absorption from small soft-boned fish. J Trace Elem Med Biol. 1998;12(3):148-154. doi:10.1016/S0946-672X(98)80003-5
Not medical advice
This article is general information, not a diagnosis or treatment. Talk to a doctor or registered dietitian before changing your diet, especially if you are pregnant, take medication or have a health condition. Read our medical disclaimer.
