Cod is a very lean fish. Raw, a 3 oz (85 g) serving has 70 calories, 15.1 g of protein and 0.6 g of fat, so about 92% of its calories come from protein. That serving also covers 51% of the daily value of selenium and 32% of vitamin B12. These are raw numbers, and cooking drives off water.
The Atlantic cod genome holds a surprise. A 2011 sequencing study found that the fish has lost the genes for MHC II and CD4, key parts of the immune system in other jawed vertebrates, yet it is not unusually prone to disease. Its immune system seems to make up for the loss with an unusually large set of MHC I genes and a different mix of Toll-like receptors.
The low fat also means far less omega-3 than an oily fish has. Raw cod has 0.16 g of EPA plus DHA in 3 oz, against 1.22 g in the same amount of raw wild Atlantic salmon. Cod liver oil is a separate product. The oil is a supplement that carries vitamins A and D, while the fillet has only a little of either: 1% of the daily value of vitamin A and 5% of vitamin D per 100 g.
Health benefits of cod
None of the cod studies below reaches strong evidence. The best-supported results are that eating cod raised iodine status in pregnant women with mild-to-moderate deficiency, that hair mercury stayed low on 400 g of cod a week, and that a large trial of cod liver oil, a supplement, did not prevent respiratory infections. Cohorts of lean fish eaters show no clear heart benefit. Here is what 27 studies found, strongest evidence first, from iodine, mercury and blood fats to blood sugar, inflammation and the gut, including the trials that found nothing.
Lean fish and fish proteins in trials lowered blood triglycerides but did not change total cholesterol or lipoproteins
Meta-analysis of 24 studies, with 18 to 21 randomized trials (1,392 to 1,456 people, depending on the outcome) in the main comparison of lean fish with no fish. Parallel-group trials found larger triglyceride drops than crossover trials. The authors say better-designed, longer and larger trials are needed, and some trials used fish-derived protein rather than fish meals.
Eating 200 g of cod twice a week for 16 weeks in pregnancy raised iodine status in women with mild-to-moderate iodine deficiency
Randomized trial in 137 pregnant women in Bergen, Norway. Median urinary iodine reached 98 µg/L in the cod group against 73 µg/L in the control group, while maternal thyroid hormones did not differ. Infants of the cod group had a lower Bayley cognitive score at 11 months, which the authors say needs follow-up. The Norwegian Seafood Research Fund paid for the study and says it had no role in its design, analysis or reporting.
Eating 400 g of cod a week in pregnancy raised hair mercury only slightly and did not push more women over the US EPA reference dose
Randomized trial in 137 Norwegian pregnant women (the same trial as the iodine study above) who ate cod fillet for 16 weeks or kept their usual diet. Median hair mercury was 554 µg/kg in the cod group and 485 µg/kg in the control group, a difference that was not significant, and 8% of all women ended above the EPA reference dose (6 in the cod group, 4 in the control group).
Cod liver oil, a vitamin D supplement, taken as 5 mL a day for up to six months did not lower the risk of covid-19 or other respiratory infections
Supplement trial, not a study of eating cod. Quadruple-blind, placebo-controlled randomized trial in 34,601 adults in Norway, with 10 µg of vitamin D a day in the oil. At least one respiratory infection was reported by 22.94% of the cod liver oil group and 22.13% of the placebo group (relative risk 1.04). Orkla Health AS, the maker of the cod liver oil, paid for the trial and says it had no role in running it, analyzing the data or writing the paper, though it could comment on the manuscript.
Show 23 more studies
A diet with 60% of its protein from lean seafood for 4 weeks lowered fasting triglycerides by 0.31 mmol/L compared with a non-seafood diet
Randomized crossover trial in 20 healthy adults (7 men and 13 women). The lean seafood diet also prevented the rise in the ratio of total to HDL cholesterol seen after the non-seafood diet. The test meals used cod or lean beef. It was small, and the seafood was a group of lean fish, not cod alone.
4 weeks of a cod protein diet improved insulin sensitivity in insulin-resistant adults compared with a diet of other animal proteins
Randomized crossover trial in 19 insulin-resistant people. The comparison diet used lean beef, pork, veal, eggs, milk and milk products, and both diets matched in fiber and fats. The gain was statistically significant (P = 0.027), while a measure of beta-cell function only tended to improve (P = 0.055).
Eating 750 g of cod a week for 8 weeks lowered serum neopterin, a marker of immune activation, while the same amount of salmon did not
Randomized trial in 63 adults with overweight or obesity: cod (22 people), salmon (22) or no fish (19). Neopterin fell by a median of 2.65 nmol/L on cod, and blood selenium rose only after cod. The Bergen Medical Research Foundation paid for the research and says it had no role in the design or reporting. Lerøy Seafood Group supplied the fish free of charge and employs one author, and says it was not involved in collecting data on site.
The same cod protein diet lowered hsCRP, a blood marker of inflammation, in insulin-resistant adults
Same randomized crossover trial as above, in 19 overweight or obese men and women aged 40 to 65, with 4 weeks on each diet. hsCRP fell on cod (P = 0.021) and tended to rise on the mixed-protein diet. Interleukin-6, TNF-alpha and adiponectin did not differ between diets, and total cholesterol and LDL fell only on the mixed-protein diet.
In the same trial, the lean seafood diet lowered after-meal C-peptide and lactate, although insulin and glucose did not change
Secondary, explorative analysis of the 20-person crossover trial above. The ratio of triglycerides to HDL cholesterol was also lower on the seafood diet. Serum insulin and glucose were not changed, and the authors say the findings may matter for long-term diabetes risk, which this trial could not show.
Eating 750 g of cod or salmon a week for 8 weeks changed the gut bacteria profile but not blood lipids or fecal fat
Randomized trial in 76 adults with overweight or obesity, with gut bacteria analyzed in 33 of them. The cod group looked more like the salmon group than like the no-fish group. Fecal output of fat, cholesterol and bile acids and serum cholesterol, triglycerides and bile acids did not change. The Bergen Medical Research Foundation paid for the research and says it had no role in the design or reporting. Lerøy Seafood Group supplied the fish free of charge, and two authors work for Skretting and Lerøy Seafood Group.
Women who ate 200 g of cod twice a week in pregnancy had infants with fewer socio-emotional problems at 3 and 6 months, with no difference in general development
Secondary analysis of the randomized trial in 133 pregnant women, with mothers filling in questionnaires when infants were 3, 6 and 11 months old. Total scores on the general development questionnaire did not differ (P = 0.633). Socio-emotional scores were 20.9 and 20.5 in the cod group and 26.1 and 26.8 in the control group at 3 and 6 months (P = 0.020). The main trial found a lower cognitive score at 11 months in the cod group.
Meals with cod or veal burned the same amount of energy after eating and gave the same appetite and later food intake
Crossover trial in 23 overweight men and women who ate four test meals: cod or veal with mashed potatoes or wholegrain pasta. Diet-induced thermogenesis, appetite sensations and energy eaten at a buffet 3.5 hours later did not differ. Glucose and insulin peaks were higher after the mashed potato meals, whichever protein was eaten.
Non-heme iron from a cod protein meal was absorbed 2.7 times better than from a fava bean protein meal in healthy women
Single-blinded isotope trial in healthy Swedish women aged 18 to 45, each serving as her own control. Adjusted to a 40% reference dose, absorption was 9.2% for the cod meal and 4.2% for the fava bean meal, and 21.7% for a beef meal tested in a second trial. The Swedish Research Council FORMAS and Region Västra Götaland funded the work.
Eating 750 g of cod a week for 8 weeks raised TMAO in blood and urine more than salmon did
Randomized trial in 62 adults with overweight or obesity (cod, salmon or no fish). The authors propose TMAO as a marker of cod intake and 1-methylhistidine as a marker of salmon intake. The trial aimed to find intake markers, not to test health effects.
Tablets with 2.5 g a day of cod protein for 8 weeks lowered after-meal glucose within the cod group, but no more than herring or salmon protein did
Supplement trial, not a study of eating cod. Randomized, double-blind trial in 93 overweight adults who got 2.5 g of protein a day from herring, salmon, cod or milk as tablets, with 77 analyzed. In the cod group, glucose at 90 and 120 minutes fell, and salmon protein did not affect any glucose measure. Milk protein tablets lowered the glucose area under the curve by 7%.
4 g a day of cod protein hydrolysate for 8 weeks did not change blood sugar, insulin or blood lipids in adults with metabolic syndrome
Supplement trial, not a study of eating cod. Randomized, double-blind trial with 15 people on cod protein hydrolysate and 15 on placebo. Fasting triglycerides, total cholesterol and LDL fell in both groups, with no difference between them. Funders included the Norwegian Council of Research and the company Firmenich Bjørge Biomarin AS.
6 weeks of 2.5 g a day of cod protein hydrolysate did not ease irritable bowel symptoms more than placebo
Supplement trial, not a study of eating cod. Double-blind trial with 13 people on cod protein hydrolysate and 15 on placebo. Symptom scores fell in both groups (from 298 to 236 on cod, from 295 to 202 on placebo), and end scores did not differ (P = 0.395). Funders included the Norwegian Council of Research and the company Firmenich Bjørge Biomarin AS.
8.1 g a day of cod residual protein for 8 weeks lowered blood ketone bodies and α-hydroxybutyrate in lean adults, without changing fasting glucose or insulin
Supplement trial, not a study of eating cod fillet. Randomized, double-blind trial in 50 lean adults, 40 of whom finished, using capsules of protein from cod residuals. TMAO rose in plasma and urine. The Research Council of Norway, K. Halstensen AS and a regional fund paid for it, one author is a shareholder in K. Halstensen AS, and another was employed there.
6 g a day of protein from cod residuals for 8 weeks lowered after-meal free fatty acids in overweight adults in a pilot trial
Supplement trial, not a study of eating cod fillet. Double-blind pilot trial in 42 overweight or obese adults split into three groups. Only the presscake meal lowered after-meal free fatty acids, while the version with stickwater did not. Fasting insulin rose within the presscake group, and that change did not differ from the other groups. The Research Council of Norway, K. Halstensen AS and others funded it, and one author is a shareholder in K. Halstensen AS.
Cod-liver oil for 4 weeks did not change the main blood sugar measures in women with gestational diabetes, though later measures looked better
Supplement trial, not a study of eating cod. Double-blind trial in 550 women with gestational diabetes, using cod-liver oil or a mineral oil placebo. After 12 to 16 weeks, secondary outcomes such as glucose, lipids, hs-CRP and HOMA-IR were lower in the oil group, but the primary outcomes had not differed. The NHS advises against cod liver oil in pregnancy because of its vitamin A. A Jilin science fund paid for the study.
People who ate the most lean fish did not have a lower risk of heart disease or death than people who ate the least, in pooled cohort studies
Meta-analysis of 19 prospective reports. Lean fish studies covered 937,362 person-years, 21,636 coronary heart disease cases and 16,831 deaths, and cod is a lean fish. Fatty fish was linked to lower coronary heart disease (RR 0.92) and total death (RR 0.97), but lean fish was not. Cohorts show links, not cause.
Norwegian women who ate 75 to 100 g of lean fish a day had a lower risk of type 2 diabetes (risk ratios 0.71 and 0.67)
Cohort of 33,740 women free of diabetes at the start, with lean fish, fatty fish, fish products, total fish and cod liver oil supplements analyzed separately. Only lean fish was linked to lower risk. The authors say lean fish eaters may have a protective lifestyle that the study could not fully account for.
A pooled analysis of cohorts found no link between lean fish and type 2 diabetes (RR 1.03)
Systematic review and meta-analysis of prospective cohort studies. Fatty fish looked protective (RR 0.89), but the link vanished when one study was removed (RR 0.93). Fish products and fried fish showed no link either. It conflicts with the Norwegian cohort above, and the authors call for more long-term studies.
Eating at least 175 g of fish a week went with a lower risk of death in people with vascular disease, but not in the general population
Pooled analysis of 4 cohorts with 191,558 people, studying fish in general, not cod. In 3 cohorts of patients with vascular disease, the hazard ratio for total death was 0.82, and for major cardiovascular events 0.84. In the PURE general-population cohort, 350 g a week or more was not linked to either outcome. Fish rich in omega-3 fats drove the benefit, while other fish were neutral in the one cohort that recorded types.
Each extra 100 g of fish a day went with an 8% lower risk of dying from any cause in pooled cohorts
Umbrella review of 34 meta-analyses covering 298 prospective cohort studies of fish in general, not cod alone. Moderate-quality evidence linked each 100 g a day to lower risk of coronary heart disease (0.88), stroke (0.86), heart failure (0.80) and depression (0.88). Most cancers showed no link.
Rats fed cod protein did not develop the muscle insulin resistance that casein or soy protein allowed on a high-fat diet
Animal study with rats on a high-fat diet using casein, cod or soy protein, plus chow-fed controls. Cod protein kept insulin-driven muscle glucose uptake close to that of chow-fed rats, without less weight gain or fat. Muscle cells exposed to cod-derived amino acids also took up more glucose with insulin. It was not a human study.
In high-fat-fed rats, cod protein prevented a drop of about 50% in whole-body and muscle glucose disposal that casein and soy protein did not
Animal study over 4 weeks. Cod protein also prevented a 60% fall in a muscle insulin signal (PI 3-kinase activity) and improved GLUT4 movement to part of the muscle cell. The authors call cod protein a natural insulin-sensitizing agent in rats, which has not been shown in people.
Who should be careful with cod
Most people can eat cod freely. Take a little care if one of these applies:
- Fish allergy. Fish is one of the major food allergens, and US law makes labels name the species, such as cod, instead of only “fish”. If you are allergic to fish, ask your allergist before eating cod or products made with it.
- Pregnancy and cod liver oil. The oil is a supplement with vitamin A, and the NHS says not to take cod liver oil or vitamin A supplements in pregnancy, because too much vitamin A can harm a baby’s development. Cod fillet is a different food, and the FDA and EPA list it as a Best Choice.
- Mercury. This is a small concern. For pregnancy and breastfeeding the FDA and EPA advise 2 to 3 servings of 4 oz a week from the Best Choices list, which includes cod. In a trial, 400 g a week for 16 weeks kept hair mercury low in most women.
How to choose, store and eat cod
- Baked or pan-fried: in a lab test these methods did not change the total iodine in a fillet slice, and no cooking method changed mercury much.
- Boiled: boiling lowered the iodine in the fillet, matching the iodine found in the boiling water, so a soup or stew keeps it.
- Frozen: thawing frozen fillets cost 12% of the iodine on average, lost to the thaw water.
Buy cod that is refrigerated or properly iced, and refrigerate or freeze it right away if you will not cook it soon. Cook it to 145°F (63°C), or until it is opaque and flakes with a fork. Keep raw seafood apart from ready-to-eat foods, in sealed containers or bags.
Sources
- Advice about eating fish. US Food and Drug Administration and Environmental Protection Agency. 2024.
- Food allergies. US Food and Drug Administration.
- Pregnancy vitamins and supplements. NHS.
- Cooking (Food Safety for Moms-to-Be). US Food and Drug Administration. 2018.
- Meat, poultry and seafood (Food Safety for Moms-to-Be). US Food and Drug Administration. 2018.
- Dahl L, Duinker A, Næss S, et al. Iodine and mercury content in raw, boiled, pan-fried, and oven-baked Atlantic cod (Gadus morhua). Foods. 2020.
- Næss S, Kjellevold M, Dahl L, et al. Effects of seafood consumption on mercury exposure in Norwegian pregnant women: a randomized controlled trial. Environ Int. 2020.
History of cod
Cod fed people long before refrigeration. Stockfish, cod dried in cold air and wind, is Norway’s longest-running export. The 13th-century Egil’s Saga mentions stockfish shipped from mid-Norway to Britain in the year 875, and the dried fish fed Viking voyages and Iceland for centuries. Ancient DNA from fish bones shows that Viking Age cod (800 to 1066 CE) found at Haithabu, Germany, came from the Northeast Arctic population that still supports the Lofoten fisheries of Norway.
Across the Atlantic, northern cod fishing off Newfoundland dates to the 16th century. The catch peaked at 810,000 tons in 1968, about three times the largest yearly catch before the super-trawlers arrived. About 8 million tons were caught between 1647 and 1750, and factory trawlers took the same amount in 15 years.
In 1992 the Canadian fisheries minister, John Crosbie, declared a moratorium on northern cod, first for two years. By 1993 six cod populations had collapsed, and the closure was extended indefinitely. Northern cod spawning biomass had fallen by 99%, and more than 35,000 fishers and plant workers from more than 400 coastal communities lost their work, in what is described as the largest industrial closure in Canadian history.
Recovery has been slow. A 2022 study of cod off southern Newfoundland found that the stock had not recovered after 30 years of far less fishing, and showed evidence of Allee effects, in which a population below a threshold struggles to rebuild. In 2024 Canada allowed commercial cod fishing off the north and east coasts of Newfoundland and Labrador for the first time since 1992, with a total allowable catch of 18,000 tons.
Overfishing also changed what the fish carry. A study using more than 30 years of data found that methylmercury in Atlantic cod rose by up to 23% between the 1970s and 2000s because of diet shifts that began with overfishing. Iceland’s extensions of its fishing limits led to three Cod Wars with Britain, in 1958 to 1961, 1972 to 1973 and 1975 to 1976, in which the Royal Navy escorted British trawlers.
Cod liver oil has its own history. In the 1800s it became a popular bottled medicine, after doctors in Germany and other countries noticed that children given it did not develop rickets. In 1922 the American researcher Elmer McCollum showed that the rickets cure in the oil was a separate factor from vitamin A, which he named vitamin D, and by the 1930s, with vitamin D found and the anti-rickets effect of cod liver oil understood, rickets could be treated and eradicated in the United States.
Sources
- Star B, Nederbragt AJ, Jentoft S, et al. The genome sequence of Atlantic cod reveals a unique immune system. Nature. 2011.
- Star B, Boessenkool S, Gondek AT, et al. Ancient DNA reveals the Arctic origin of Viking Age cod from Haithabu, Germany. Proc Natl Acad Sci U S A. 2017.
- Perälä T, Hutchings JA, Kuparinen A. Allee effects and the Allee-effect zone in northwest Atlantic cod. Biol Lett. 2022.
- Schartup AT, Thackray CP, Qureshi A, et al. Climate change and overfishing increase neurotoxicant in marine predators. Nature. 2019.
- Rajakumar K. Vitamin D, cod-liver oil, sunlight, and rickets: a historical perspective. Pediatrics. 2003.
- Atlantic cod. Wikipedia.
- Collapse of the Atlantic northwest cod fishery. Wikipedia.
- Stockfish. Wikipedia.
- Cod Wars. Wikipedia.
- Cod liver oil. Wikipedia.
Frequently asked questions about cod
Is cod good for you?
For most people, yes. Raw cod has 70 calories, 15.1 g of protein and 0.6 g of fat in 3 oz (85 g), and covers 51% of the daily value of selenium. In trials, eating cod raised iodine status in pregnant women with mild-to-moderate deficiency, and lean fish and fish proteins lowered triglycerides. None of this reaches strong evidence, and pooled cohorts found no lower heart disease risk for lean fish.
Is cod high in mercury?
No. The FDA and EPA list cod as a Best Choice, their lowest-mercury group. In a trial of 137 pregnant women, 400 g of cod a week for 16 weeks left median hair mercury at 554 µg/kg, against 485 µg/kg in the control group, a difference that was not significant. Mercury in Atlantic cod did rise by up to 23% between the 1970s and 2000s, according to a study of diet shifts from overfishing.
Can you eat cod every day?
No study here tested cod every day. Trials used 750 g a week for 8 weeks, or 400 g a week for 16 weeks in pregnancy. The FDA and EPA advise 2 to 3 servings of 4 oz a week from the Best Choices list, which includes cod, for pregnancy and breastfeeding. Vary your fish, and ask your doctor if you have a health condition.
Does cod have omega-3?
A little, but far less than oily fish. Raw cod has 0.16 g of EPA plus DHA in 3 oz (85 g), against 1.22 g in the same serving of raw wild Atlantic salmon. In a pooled analysis of cohorts, fatty fish but not lean fish went with a lower risk of heart disease (RR 0.92) and death (RR 0.97).
Can you eat cod while pregnant?
Yes, if it is fully cooked. The FDA and EPA list cod as a Best Choice, and in a trial, 200 g twice a week for 16 weeks raised iodine status in pregnant women with mild-to-moderate deficiency. That trial also found a lower infant cognitive score at 11 months in the cod group, which the authors say needs follow-up. Cook cod to 145°F (63°C), and skip cod liver oil, which the NHS says not to take in pregnancy because of vitamin A.
Is cod liver oil good for you?
The evidence is thin. In 34,601 adults, 5 mL a day for up to six months did not lower the risk of covid-19 or other respiratory infections (22.94% against 22.13% reporting an infection). Cod liver oil supplements showed no effect on type 2 diabetes risk in a cohort of Norwegian women. It contains vitamin A, and the NHS says not to take it in pregnancy.
Is cod high in protein?
Yes. Raw cod has 17.8 g of protein per 100 g, or 15.1 g in 3 oz (85 g), which is 30% of the daily value. About 92% of its calories come from protein, and it scores 100 out of 100 for essential amino acids. These are raw weights, and cooking drives off water.
Is cod good for weight loss?
No study here tested weight loss. Raw cod has 82 calories and 0.7 g of fat per 100 g, so it is a low-calorie protein food. In 23 overweight adults, meals with cod or veal burned the same energy after eating and gave the same appetite and later food intake.