Cod: 34 thousand people, one winter, no effect

Cod liver oil is the oldest supplement in the northern European medicine cabinet, and it has now had the trial it deserved. Norway randomised 34601 adults to 5 mL a day or placebo through a covid winter, quadruple blinded, with four endpoints named in advance. None of them improved: not infection, not serious illness, not other respiratory infections, not symptoms. The dose delivered 10 micrograms of vitamin D, deliberately low, so higher doses remain untested. Norwegians also eat a lot of fish already, which makes any extra harder to detect. It is still the cleanest answer this question has ever had.

The numbers, per 100 g

How much is actually in it

• Eating 750 grams of cod a week lowered an immune activation marker and raised blood selenium, where salmon did neither. — 63 participants with overweight or obesity, 22 on cod, 22 on salmon and 19 control (n = 63) Evidence B source
  Serum total neopterin fell in the cod group against control (median change -2.65 nmol/L, p = 0.018) and not in the salmon group (0.00, p = 0.59); estimated intake of selenium, iron, magnesium and zinc was similar across groups, yet serum selenium rose only after cod (p = 0.017)

What cooking and storage change

• A third of cod carries the parasite that causes anisakiasis, against one salmon in twenty. — 83 studies included from 519 identified, with data from 13 countries including Spain, Germany, Chile, Denmark, Norway and Iceland (n = 83 studies) Evidence C source
  Anisakidae occurred widely with high variability between fish species and fishing seas; mean prevalence was 33% in cod (Gadus morhua) and 5% in Atlantic salmon; fish caught in the Northeast Atlantic showed the highest levels

What people believe that the data does not support

• In a trial of 34 thousand Norwegians through a covid winter, daily cod liver oil prevented nothing. — 34601 adults aged 18 to 75 not already taking daily vitamin D supplements, 17278 on cod liver oil and 17323 on placebo (n = 34601) Evidence B source
  Supplementation was not associated with a reduced risk of any co-primary endpoint: positive SARS-CoV-2 test, serious covid-19 defined as dyspnoea, hospital admission or death, negative test result, or self-reported symptoms; participants took the supplement for a median of 164 days
• All three European health claims for cod were refused, including one about keeping skin young. — — Evidence E source
  Three register entries name cod and all three carry the status Non-authorised; the example is cod liver oil standardised in vitamin A claiming to maintain the youth capital of the skin and contribute to the integrity of skin tissues

Who this works differently for

• Swapping other animal protein for cod improved insulin sensitivity when nothing else in the diet changed. — 19 insulin-resistant men and women (n = 19) Evidence B source
  Insulin sensitivity improved significantly on the cod diet (p = 0.027) with a strong tendency towards a better disposition index (p = 0.055); an interaction with baseline insulin sensitivity was seen on the beta-cell function index (p = 0.022)

What is still unknown

• Bottling the same cod protein as a four-gram supplement produced no effect at all. — 30 adults with metabolic syndrome, 15 per arm (n = 30) Evidence B source
  No difference between hydrolysate and placebo in insulin, glucose or GLP-1; fasting triglycerides, total cholesterol and LDL fell within both groups from baseline with no difference between them; body composition unchanged

The bottom line

Cod as food is more interesting than cod as oil. In a crossover trial where two diets were matched on fibre and every class of fatty acid and differed only in protein source, insulin sensitivity measured by clamp improved on the cod diet. Extract that protein and bottle 4 grams of it as a hydrolysate and nothing happens. In another trial, 750 grams of cod a week raised blood selenium while the same estimated intake from salmon did not, which points at absorption rather than content. One safety note for anyone eating it raw or home-cured: about a third of cod carries Anisakis larvae against 5 percent of Atlantic salmon, and freezing and thorough cooking are what kill them.

Sources

1. cod-dr-04 · randomised clinical trial
   Increased serum concentration of selenium was observed only after cod intake when compared to the Control group (P = 0.017).
   Eur J Nutr, 2021 · checked 2026-09-22
2. cod-dr-05 · systematic review and meta-analysis of prevalence studies
   Cod (Gadus morhua) and Atlantic salmon (Salmo salar), the most consumed fish species in Belgium, have a mean prevalence of 33% and 5% respectively.
   Int J Food Microbiol, 2020 · checked 2026-09-22
3. cod-dr-01 · randomised placebo-controlled trial
   Supplementation with cod liver oil was not associated with a reduced risk of any of the co-primary endpoints.
   BMJ, 2022 · checked 2026-09-22
4. cod-dr-06 · regulatory register entry
   POL-HC-7271 Huile de foie de morue : Cod liver oil standardized in vitamin A, EPE and HA Maintain the youth capital of the skin. Contribute to the integrity of the skin tissues Non-authorised
   EU Register on nutrition and health claims, заява POL-HC-7271, знімок 2026-09-21 · checked 2026-09-22
5. cod-dr-02 · randomised crossover trial
   There was a significant improvement in insulin sensitivity (P = 0.027) and a strong tendency for a better disposition index (beta-cell function x M/I) (P = 0.055) in subjects consuming the cod protein diet compared with those consuming the BPVEM diet.
   Diabetes Care, 2007 · checked 2026-09-22
6. cod-dr-03 · randomised double-blind trial
   In conclusion, supplementing with a low dose of CPH in subjects with MetS for 8 weeks had no effect on fasting or postprandial levels of insulin, glucose or GLP-1, lipid profile or body composition.
   Nutrients, 2020 · checked 2026-09-22