Fish: the omega-3 and the mercury arrive together

Advice about fish usually tells you one half of the story. A trial in Norwegian preschoolers measured 86 biomarkers after 16 weeks of oily fish three times a week, and the largest single change was hair mercury, ahead of the omega-3 fatty acids everyone was looking for. In pregnant women, 400 grams a week of cod, deliberately one of the lowest-mercury fish, did not raise hair mercury measurably, and 8 percent of the whole study population was already above the United States reference dose regardless of which group they were in. So the species matters, the baseline matters, and a page that mentions only the good half is not being straight with you.

The numbers, per 100 g

What your body absorbs

• Europe allows fish to be sold as something that helps you absorb iron from the rest of the meal. — — Evidence E source
  Of 20 register entries naming fish, one is Authorised and 19 Non-authorised; the authorised wording is that meat or fish contributes to the improvement of iron absorption when eaten with other foods containing iron

What it does to your health

• People eating the most fish had about a third lower odds of osteoporosis, in studies that cannot show cause. — 14 studies with 26145 participants (n = 26145) Evidence C source
  Highest against lowest fish intake gave odds ratio 0.62 (95% CI 0.49-0.79, I2 79.4%); among perimenopausal women 0.54 (0.32-0.90); case-control studies 0.28 (0.16-0.50) against cross-sectional 0.70 (0.55-0.87); lumbar or femoral sites 0.59 and calcaneus 0.73
• In people on dialysis, fish oil went with far fewer heart events, on evidence the authors call unconfirmed. — 12 studies in dialysis-dependent chronic kidney disease (n = 12 studies) Evidence C source
  Fish oil supplementation lowered cardiovascular events by 44% (HR 0.56, 95% CI 0.46-0.68) and myocardial infarction by 48% (0.52, 0.34-0.78); higher baseline omega-3 levels went with 31% lower all-cause mortality (0.69, 0.54-0.88); higher dietary omega-3 intake showed only a non-significant trend (0.92, 0.79-1.08)

Safety, limits and interactions

• Four hundred grams of cod a week through pregnancy did not measurably raise mercury in hair. — 137 pregnant women in Norway (n = 137) Evidence B source
  Post-intervention median hair mercury 554 micrograms/kg in the cod group against 485 in controls (p = 0.186); adjusting for baseline, a difference appeared only among women starting below 534 micrograms/kg; 8% of the whole study population exceeded the US EPA hair reference dose of 1000 micrograms/kg, 6 in the intervention group and 4 in the control
• When preschoolers ate oily fish three times a week, hair mercury rose more than any of the other eighty-five markers measured. — 232 children aged 4 to 6 in Norway (n = 232) Evidence B source
  Standardised effects for red cell EPA 0.61, DHA 0.43, total n-3 0.41, n-3 to n-6 ratio 0.48, urinary iodine 0.32 and hair mercury 0.83; arachidonic acid -0.54 and total n-6 -0.31; the authors name n-3 fatty acids, urinary iodine, hair mercury and plasma 1-methylhistidine as the strongest effects of the 86 tested

What is still unknown

• Brain scans of people who eat more fish show less small-vessel damage, and the studies that found nothing rarely get quoted. — 24 studies, predominantly in older adults (n = 24 studies) Evidence C source
  Higher fish consumption was associated with less severe white matter hyperintensities and fewer cerebral microbleeds, with preserved hippocampal, temporal lobe and periventricular white matter volume and cortical thickness in the precuneus, parietal and cingulate grey matter; some analyses found no association and isolated findings suggested possible adverse associations that were not replicated
• Official advice on fish in pregnancy exists and mostly fails to reach the people giving the advice. — biomonitoring studies, Spanish guidelines and surveyed healthcare professionals Evidence C source
  Consistent evidence of mercury exposure shaped by fish consumption patterns; guidelines showed substantial heterogeneity, frequent omission of official recommendations and use of outdated or incomplete information; surveyed professionals showed limited awareness of official guidance and inconsistent communication to pregnant women

The bottom line

Fish eaters have less osteoporosis, less small-vessel damage on brain scans and lower death rates, all in observational data where fish eating travels with income, education and exercise. The review of brain scans is unusually honest about it, flagging the studies that found nothing and the unreplicated results that pointed the wrong way. What nobody has done is weigh omega-3 against mercury in one number for one species, which is the question people actually have. Europe does allow one claim: meat or fish improves iron absorption from the rest of the meal, which sits awkwardly beside a 2025 trial where beef was no better than a plant-based burger at that job. Both are in our base, and neither settles it.

Sources

1. fsh-dr-07 · regulatory register entry
   POL-HC-6417 Meat or fish Meat or fish contributes to the improvement of iron absorption when eaten with other foods containing iron Authorised
   EU Register on nutrition and health claims, заява POL-HC-6417, знімок 2026-09-21 · checked 2026-09-22
2. fsh-dr-03 · meta-analysis of observational studies
   The pooled OR for osteoporosis comparing the highest to lowest fish consumption levels was 0.62 (95% CI, 0.49-0.79; I² = 79.4).
   Nutr Res, 2026 · checked 2026-09-22
3. fsh-dr-05 · systematic review and meta-analysis
   In hemodialysis-dependent CKD, fish oil supplementation lowered cardiovascular events by 44% (HR 0.56; 95% CI 0.46-0.68) and myocardial infarction by 48% (HR 0.52; 95% CI 0.34-0.78).
   Future Cardiol, 2026 · checked 2026-09-22
4. fsh-dr-01 · randomised controlled trial
   Post-intervention, the intervention group had median (inter-quartile range) THHg concentrations of 554 (392-805) µg/kg, and the control group 485 (341-740) µg/kg (p = 0.186).
   Environ Int, 2020 · checked 2026-09-22
5. fsh-dr-02 · randomised controlled trial
   Of the 86 biomarkers, the strongest effect of fatty fish intake was on n-3 PUFAs, UIC, hair mercury, and plasma 1-MH.
   J Nutr, 2021 · checked 2026-09-22
6. fsh-dr-04 · systematic review of observational studies
   Some analyses found no association and isolated findings suggested possible adverse associations that were not consistently replicated.
   Aging Clin Exp Res, 2026 · checked 2026-09-22
7. fsh-dr-06 · scoping review with guideline analysis and cross-sectional survey
   Guideline analysis revealed substantial heterogeneity, frequent omissions of official recommendations, and the use of outdated or incomplete information.
   Environ Int, 2026 · checked 2026-09-22