Haddock: a low-mercury lean fish studied mostly as cod

Almost no human research names haddock. The science sits under lean fish and under cod, its close relative, and this page borrows it on those terms. The strongest result is a meta-analysis of 18 to 21 randomized trials with 1,392 to 1,456 participants, in which lean fish lowered blood triglycerides compared with no fish and made no significant difference to total cholesterol or lipoproteins. The abstract gives no size for the triglyceride drop, and its lean fish category covers cod, pollock and fish proteins rather than haddock. What is measured on haddock itself is composition and mercury. Raw haddock has 0.45 g fat and 25.9 mcg selenium per 3.5 oz (100 g) across 8 USDA samples. Fifty FDA samples averaged 0.055 ppm mercury, in testing that ended in 2009. FDA and EPA put haddock on their Best Choices list, at 2 to 3 servings a week in pregnancy and breastfeeding.

The numbers, per 3.5 oz (100 g)

As people eat it, per 3.5 oz (100 g)

How much is actually in it

• Raw haddock has 25.9 ug selenium and 0.45 g fat per 3.5 oz (100 g) in USDA samples, with sodium ranging widely from 76 to 452 mg. — USDA Foundation Foods samples of raw haddock (n = 8) Evidence C source
  Selenium 25.9 ug (23.0-33.5, 8 samples); fat 0.45 g (0.33-0.55); protein 16.3 g (12.9-18.4); B12 1.83 ug (1.5-2.1, 3 samples); sodium 213 mg (76-452)
• Smoking raises haddock sodium about sixfold, from 68 mg to 420 mg per 3.5 oz (100 g) raw flesh in the UK table. — CoFID food composition records Evidence C source
  Raw 68 mg Na, smoked raw 420 mg; steamed 84 mg; selenium raw 34 ug, steamed 51 ug per 3.5 oz (100 g) (water loss)

What a real portion looks like

• A regular baked haddock fillet in US survey data weighs 4.9 oz (140 g) and gives about 28 g of protein. — USDA FNDDS portion weights Evidence C source
  1 small/regular fillet 140 g = 28.4 g protein; 1 large fillet 210 g = 42.5 g; protein 20.26 g/100 g

What your body absorbs

• In challenge studies, cod allergens reached the blood 10 minutes after eating, and cod digested at weaker stomach acid set off reactions at 10 to 30 times lower doses in allergic patients. — healthy volunteers and patients with diagnosed codfish allergy Evidence C source
  Allergen absorption from 10 min, peak 1-2 h; pH 3.0 digest triggered reactions at 10-fold to 30-fold lower cumulative doses

What cooking and storage change

• By the USDA retention table for lean finfish baked without drippings, about 90% of vitamin B12, B6, thiamin and folate survive cooking. — USDA R6 group Finfish and Shellfish Products Evidence C source
  Retention: B12, B6, thiamin, folate 90%; riboflavin, niacin 95% (method 2301)

What it does to your health

• A meta-analysis of 18 to 21 randomized trials with about 1,400 people found lean fish lowered blood triglycerides versus no fish, with no significant difference in total cholesterol or lipoproteins. — 1392 to 1456 participants in crossover and parallel RCTs (n = 1392-1456) Evidence A source
  Triacylglycerol-lowering vs no fish; no significant difference in total cholesterol or lipoproteins; size not in abstract
• Pooled cohort data link the highest lean fish intake to a 19% lower stroke risk than the lowest intake. — five prospective cohort studies, 5 comparisons for lean fish (n = 5 studies) Evidence C source
  RR 0.81 (95% CI 0.67-0.99) for lean fish; fatty fish RR 0.88 (0.74-1.04)
• In 83,304 Norwegian women followed about 21 years, swapping 20 g a day of processed meat for lean fish was linked to 8% lower all-cause mortality among high processed meat eaters. — 83 304 women in the Norwegian Women and Cancer Study, median 21.0 years (n = 83304) Evidence C source
  All-cause HR 0.92 (0.89-0.96), CVD HR 0.82 (0.74-0.90); no association in women eating 30 g/d or less of processed meat
• In a 4-week randomized crossover trial in 20 healthy adults, a lean seafood diet lowered post-meal C-peptide and the TG/HDL ratio without changing glucose or insulin. — 20 healthy adults (7 men and 13 women), 4-week diet periods (n = 20) Evidence B source
  Postprandial C-peptide P = 0.04, lactate P = 0.012, TG/HDL ratio P = 0.002 vs non-seafood diet; insulin and glucose unchanged
• In a randomized trial in pregnant women, 7.1 oz (200 g) of cod twice a week for 16 weeks raised median urinary iodine from 73 to 98 ug/L. — pregnant women in Bergen, Norway, gestational week 20-36; 61 per group analyzed (n = 137) Evidence B source
  UIC median 98 (IQR 64-145) vs 73 (52-120) ug/L, p = 0.028; infant Bayley-III cognitive score lower in cod group (p = 0.045)

Safety, limits and interactions

• FDA testing of 50 Atlantic haddock samples found a mean mercury level of 0.055 ppm, with a maximum of 0.197 ppm. — FDA monitoring samples 1991-2009 (n = 50) Evidence E source
  Mean 0.055 ppm, median 0.049, SD 0.033, max 0.197

What people believe that the data does not support

• The EU authorizes the claim that selenium supports normal immune function, but no claim exists for haddock or cod themselves. — EU nutrition and health claims register Evidence E source
  Authorized for selenium; haddock and lean fish: no entries; cod: 3 claims, all non-authorized

Who this works differently for

• In a randomized trial in 133 pregnant women, cod twice a week did not change infant general development at 3 to 11 months but lowered socio-emotional problem scores. — 133 pregnant women (19 weeks of gestation or less), 16 weeks (n = 133) Evidence B source
  ASQ-2 total no difference (P = 0.633); ASQ:SE 20.9 vs 26.1 at 3 mo and 20.5 vs 26.8 at 6 mo, P = 0.020 (lower is better)
• In a randomized trial in 137 pregnant women, 14 oz (400 g) of cod a week for 16 weeks did not significantly raise hair mercury (median 554 vs 485 ug/kg). — Pregnant women (n = 137), gestational week 20-36, Norway (n = 137) Evidence B source
  Hair mercury 554 (392-805) vs 485 (341-740) ug/kg, p = 0.186; exceeding US EPA reference dose: 6 vs 4 women
• Among 108 Singapore children with fish allergy, cod triggered 31.5% of reactions, yet 75.9% tolerated at least one other fish species. — 108 children with IgE-mediated fish allergy, tertiary hospital, 2015-2020 (n = 108) Evidence C source
  Cod 31.5% of reactions; anaphylaxis 6.5%; 25.9% outgrew the allergy at median 60 months
• FDA and EPA list haddock among Best Choices and advise 2 to 3 servings a week of these fish in pregnancy and breastfeeding. — people who are or might become pregnant, breastfeeding, and children Evidence E source
  Best Choices: 2 to 3 servings a week in pregnancy and breastfeeding; 2 servings a week for children

The bottom line

Pooled data from five cohort studies link the highest lean fish intake to a 19 percent lower stroke risk than the lowest (RR 0.81, 95% CI 0.67 to 0.99), an association from few studies that do not separate species. In 83,304 Norwegian women followed for a median of 21 years, replacing 20 g a day of processed meat with lean fish was linked to 8 percent lower all-cause mortality (HR 0.92, 95% CI 0.89 to 0.96). That held only among women eating more than 1.1 oz (30 g) of processed meat a day. In a 4-week crossover trial in 20 healthy adults, a lean seafood diet lowered post-meal C-peptide and the TG/HDL ratio without changing glucose or insulin. It was an exploratory secondary outcome, and the abstract gives no sizes. A regular baked fillet in US survey data weighs 4.9 oz (140 g) and gives about 28 g of protein. Sodium is the number that moves. USDA raw samples ranged from 76 to 452 mg per 3.5 oz (100 g), a spread the dataset does not explain. In the UK table smoking raises sodium about sixfold, from 68 to 420 mg per 3.5 oz (100 g). Baking lean fish keeps about 90 percent of vitamin B12, B6, thiamin and folate by the USDA retention table, a group factor dated 1975. Selenium has an authorized EU claim for normal immune function, and no claim exists for haddock or cod. Fish allergy is the first caution. Among 108 Singapore children with fish allergy, cod triggered 31.5 percent of reactions and 6.5 percent had anaphylaxis, yet 75.9 percent tolerated at least one other species, most often salmon or tuna. Haddock was not tested there, so the study says nothing about whether a child allergic to cod can eat haddock. In challenge studies, cod allergens reached the blood 10 minutes after eating, and cod digested at weaker stomach acid set off reactions at 10 to 30 times lower doses in allergic patients. Pregnancy data come from cod trials in Norway. Eating 14 oz (400 g) of cod a week for 16 weeks did not significantly raise hair mercury (median 554 against 485 mcg/kg, p = 0.186) and raised median urinary iodine from 73 to 98 mcg/L. Infants of mothers in the cod group scored lower on the Bayley-III cognitive composite (p = 0.045), a finding the authors flagged for follow-up. In 133 women, infant general development at 3 to 11 months did not differ, and parent-reported socio-emotional problem scores were lower in the cod group.

Sources

1. hdk-daily-r7-13 · food composition data
   Fish, haddock, raw | Selenium, Se 25.9 UG range 23.0-33.5 (median 24.6), samples 8 | Total lipid (fat) 0.45 G range 0.33-0.55 | Protein 16.3 G range 12.9-18.4 | Vitamin B-12 1.83 UG range 1.5-2.1, samples 3 | Sodium, Na 213.0 MG range 76.0-452.0 (median 126.0), samples 8
   USDA FoodData Central, Foundation Foods, release 2026-04-30: Fish, haddock, raw (FDC 333374) · checked 2026-10-01
2. hdk-daily-r7-14 · food composition data
   16-375 Haddock, flesh only, raw | Sodium (mg) 68 | Selenium (µg) 34 | Protein (g) 17.8; 16-492 Haddock, flesh only, smoked, raw | Sodium (mg) 420; 16-377 Haddock, flesh only, steamed | Sodium (mg) 84 | Selenium (µg) 51
   CoFID 2021, Public Health England (OGL v3.0), codes 16-375, 16-492, 16-377 · checked 2026-10-01
3. hdk-daily-r7-16 · food composition data
   Fish, haddock, baked or broiled | Protein 20.26 G per 100 g | 1 small/regular fillet 140 g 28.4 G | 1 large fillet 210 g 42.5 G
   USDA FNDDS 2021-2023, code 26117120 Fish, haddock, baked or broiled (fdc_id 2706255) · checked 2026-10-01
4. hdk-daily-r7-09 · controlled challenge study
   Ingestion experiments in subjects without allergy revealed absorption of biologically active fish allergens only 10 minutes after ingestion with maximal serum levels after 1 to 2 hours. ... Fish digested at pH 3.0 revealed comparable reactivity patterns as undigested extracts. Moreover, these test materials triggered reactions at 10-fold to 30-fold lower cumulated challenge doses in patients with allergy.
   J Allergy Clin Immunol, 2007 · checked 2026-10-01
5. hdk-daily-r7-15 · food composition data
   2301 FINFISH,<5%FAT,BAKED,WO/DRIPPINGS | vitamin_b12_ug 90% | vitamin_b6_mg 90% | thiamin_mg 90% | folate_total_ug 90% | riboflavin_mg 95% | niacin_mg 95%
   USDA Table of Nutrient Retention Factors, Release 6 (2007), code 2301 FINFISH,<5%FAT,BAKED,WO/DRIPPINGS · checked 2026-10-01
6. hdk-daily-r7-01 · meta-analysis of RCTs
   Meta-analysis of data from 18 to 21 eligible crossover and parallel-design RCTs with a total of 1392 to 1456 participants found triacylglycerol-lowering effects for lean fish compared with no fish consumption. Lean fish intake showed no significant differences related to total cholesterol or lipoprotein levels.
   Nutr Rev, 2021 · checked 2026-10-01
7. hdk-daily-r7-02 · meta-analysis of cohorts
   We identified five prospective studies, including 7 comparisons for fatty fish intake and 5 comparisons for lean fish intake. Compared with the highest category of intake with lowest category, the summary relative risk was 0.88 [95% confidence interval (CI), 0.74-1.04] for fatty fish intake and 0.81 (95% CI, 0.67-0.99) for lean fish intake.
   Lipids Health Dis, 2018 · checked 2026-10-01
8. hdk-daily-r7-03 · prospective cohort
   Of 83 304 women in the Norwegian Women and Cancer Study (NOWAC) study, 9420 died during a median of 21·0 years of follow-up. ... Among women with higher processed meat intake (> 30 g/d), replacing 20 g/d with lean fish was associated with lower all-cause (HR 0·92, 95 % CI 0·89, 0·96), cancer (HR 0·92, 95 % CI 0·88, 0·97) and CVD mortality (HR 0·82, 95 % CI 0·74, 0·90)
   Br J Nutr, 2024 · checked 2026-10-01
9. hdk-daily-r7-04 · randomized crossover trial
   This was a secondary outcome and explorative randomized controlled trial with a crossover design in 20 healthy adults (7 men and 13 women) ... The diets did not alter serum insulin and glucose concentrations. However, relative to the nonseafood diet period, the lean seafood diet period reduced postprandial C-peptide (P = 0.04) and lactate (P = 0.012) concentrations and fasting and postprandial TG/HDL-cholesterol ratios (P = 0.002).
   J Nutr, 2016 · checked 2026-10-01
10. hdk-daily-r7-05 · randomized controlled trial
    pregnant women in Bergen, Norway, recruited through Haukeland University Hospital, were randomly assigned (1:1) to an intervention of 200 g of cod twice a week for 16 weeks (gestational week 20-36) or to continue with their standard diet (control group). ... Postintervention UIC was higher in the intervention group (n = 61) [median (interquartile range, IQR) 98 (64-145) μg/L], compared with control (n = 61) [median (IQR) 73 (52-120) μg/L] (p = 0.028), also after adjusting for baseline UIC (p = 0.048). Infants of mothers in the intervention group had a lower cognitive composite score on the Bayley-III compared with the control group (p = 0.045).
    Thyroid, 2021 · checked 2026-10-01
11. hdk-daily-r7-10 · agency data
    HADDOCK (Atlantic) | 0.055 | 0.049 | 0.033 | ND | 0.197 | 50 | FDA 1991-2009
    FDA, Mercury Levels in Commercial Fish and Shellfish (1990-2012) · checked 2026-10-01
12. hdk-daily-r7-12 · regulatory decision
    POL-HC-6463 Selenium Selenium contributes to the normal function of the immune system Authorised
    EU Register on nutrition and health claims, claim POL-HC-6463, snapshot 2026-09-21 · checked 2026-10-01
13. hdk-daily-r7-06 · randomized controlled trial
    In an RCT, 133 pregnant women (≤19 weeks of gestation) were randomly assigned to receive 200 g cod fillet twice weekly (intervention) or to continue with their habitual diet (control) for 16 wk. ... We found no difference between infants in the intervention and control groups on total ASQ-2 scores (P = 0.633), but a difference on the ASQ:SE scores in favor of the intervention group (20.9 and 20.5 in the intervention group and 26.1 and 26.8 in the control group for 3 and 6 mo, respectively; P = 0.020).
    J Nutr, 2021 · checked 2026-10-01
14. hdk-daily-r7-07 · randomized controlled trial
    Pregnant women (n = 137) were enrolled in the RCT "Mommy's Food" (2016-2017), which was designed to increase iodine status. Participants were randomly assigned to intervention (400 g of cod fillets per week) or control (continued habitual diet) groups for 16 weeks (gestational week 20-36). ... 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 2026-10-01
15. hdk-daily-r7-08 · retrospective cohort
    The diagnosis of fish allergy was made in 108 patients based on a convincing history of IgE-mediated allergic reaction and a positive skin prick test. ... The most common fish causing reactions were threadfin (48.1%), salmon (33.3%) and cod (31.5%). ... Anaphylaxis occurred in 6.5%. ... Most can tolerate another species of fish (75.9%), most commonly salmon (37.0%), tuna (24.1%) and cod (22.2%).
    Asian Pac J Allergy Immunol, 2025 · checked 2026-10-01
16. hdk-daily-r7-11 · agency guidance
    Eat 2 to 3 servings a week from the "Best Choices" list
    FDA/EPA, Advice about Eating Fish, 2024 · checked 2026-10-01