Omega-3 in Scottish farmed salmon 2006-2015

Overview

Sprague M et al.. Sci Rep 2016 · PMID 26899924 · DOI 10.1038/srep21892

Omega-3 in farmed Scottish salmon roughly halved from 2006 to 2015 (2.74 to 1.36 g per 100 g), but it still had more omega-3 per 100 g than the wild salmon tested (0.76 g) and most other foods.

Does the study support the specific claim? Partly · Methodological quality: high

Original publication

Plain-language summary

Scientific in-depth review

1. Full reference

Authors

Sprague M et al.

Journal

Sci Rep

Year

2016

Identifier

PMID 26899924 DOI 10.1038/srep21892

Verification

verified – PubMed efetch 2026-10-25

Retracted

no

Corrections

none

Extent checked

full text

Conflicts of interest and funding

No funding or competing-interests statement in the available full text; samples supplied by salmon producers and feed manufacturers; analyses by the University of Stirling Nutrition Analytical Service.

2. Research question

How did EPA/DHA levels in farmed salmon change as feeds shifted to terrestrial ingredients?

3. Study design and population

Study design

Measurement / residue study – Retrospective analysis of fatty-acid profiles of commercial farmed Scottish salmon analysed by a university analytical service (2006-2015), plus a one-off comparison with retail fish and meat samples

Population

Farmed Scottish Atlantic salmon (standard quality cuts supplied year-round by producers and feed manufacturers); comparators: 21 wild (mainly Pacific) salmon from supermarkets and retail mackerel, cod, tuna, sardines, beef, chicken, lamb, pork (2014/2015)

Country

UK (Scotland)

Period

2006-2015

Sample size

>3,000 farmed salmon profiles (yearly n 2006-2015: 106, 174, 247, 81, 85, 393, 212, 523, 546, 687); wild salmon n=21

Inclusion and exclusion criteria

Included samples deemed representative of commercial Scottish production; excluded trial fish, supermarket fish and fish of unknown farm location

4. Methods and endpoints

Exposure / intervention

Year/feed composition

Comparator

2006 levels

Endpoints

Fatty-acid composition (% of total fatty acids), EPA+DHA g/100 g wet weight, lipid content, portions needed to reach 3.5 g EPA+DHA/week (ISSFAL) with 130 g portions

Statistical methods

One-way ANOVA with Tukey post hoc tests (P<0.05), transformations where needed; means +/- SD

5. Actual results

Results

Terrestrial fatty acids roughly doubled from 2010 to 2015 (18:1n-9 15% to ~30%, 18:2n-6 5% to ~10%, 18:3n-3 2% to ~5%) while EPA and DHA fell by about half. EPA+DHA in flesh fell from 2.74 g (2006) to 1.36 g per 100 g ww (2015); as % of lipid ~24% (2006), 21.7% (2010), 13.0% (2015). Lipid content stable (~12-13%). Wild salmon 0.76 g EPA+DHA/100 g ww vs farmed 2.75 (2006), 2.21 (2010) and 1.36 g (2015). One 130 g portion met the 3.5 g/week ISSFAL level in 2006; two portions needed in 2015; one 2015 portion still provides 1.8 g EPA+DHA. Farmed salmon delivered more EPA+DHA than most fish and all terrestrial livestock tested. The n-3:n-6 ratio of salmon flesh by year is not reported in the main text (only in a supplementary table not available).

Absolute risks

not applicable (composition study)

Authors' conclusion

Nutritional value (EPA+DHA) of farmed Scottish salmon has declined with more terrestrial feed ingredients, requiring double portions vs 2006 to meet ISSFAL intake, but farmed salmon still delivers more EPA+DHA than most other fish and all terrestrial livestock; microalgae and GM oilseeds may be future sources.

6. Strengths

• Large 10-year dataset
• Absolute (g/100 g) as well as relative (% fatty acids) values reported
• Comparison with wild salmon and other foods

7. Weaknesses

• Convenience samples submitted by industry; representativeness not verified
• Wild salmon n=21, mostly Pacific species from supermarkets (not wild Atlantic)
• Comparison foods sampled once (2014/2015)
• No funding/COI statement; producers supplied samples
• Data only to 2015; feed formulations confidential
• No health outcomes

8. Risk of bias

Orienting tool

No standard tool fits (composition/measurement study); structured validity judgement

Type of assessment

structured assessment based on the full text by a single reviewer (no formal double review)

Full formal assessment

no

Overall risk of bias

some concerns

Rationale

Single-reviewer structured judgement based on the full text; not formal (needs two independent reviewers). Changed from low to some_concerns because of industry-supplied convenience samples and missing funding/COI statement.

Full-text review (25 October 2026): Added absolute EPA+DHA values (2.74 to 1.36 g/100 g; wild 0.76 g), yearly n, % lipid values, portion calculation, sample sourcing from producers/feed companies and exclusions, missing funding/COI statement; noted ratio not reported in main text; bias overall changed from low to some_concerns.

9. Assessment of the statistical interpretation

Decline in EPA+DHA content is large (about 50%) and consistent; comparisons with wild salmon rely on 21 samples. The paper shows farmed salmon remains richer in EPA+DHA per 100 g than wild Pacific salmon despite lower percentages of lipid.

10. Relevance to the specific claim

C-FSH-04 (supports ratio/EPA decline), C-FSH-05 (contradicts loss of benefit), C-FSH-11 (feed shift).

11. Independent contrary findings

12. Transferability to Germany and Europe

High (salmon sold in EU).

13. Overall methodological judgement

Shows a real decline in omega-3 content of Scottish farmed salmon with plant-based feeds (supports the direction of C-FSH-04/C-FSH-11), but also that per portion farmed salmon still provided more EPA+DHA than the wild salmon tested and than other common foods, contradicting 'the benefit disappears' (C-FSH-05). It does not report the 10 vs 3-4 omega-3:omega-6 ratio stated in the claim.

14. Open questions

Trends after 2015.

15. Linked fact checks

• The omega-3 to omega-6 ratio is about 10 in wild salmon but only 3-4 in farmed salmon. – Studies with mixed or unclear results
• Because of the lower omega-3:omega-6 ratio, farmed salmon is converted from an anti-inflammatory into a pro-inflammatory meal, and the expected nutritional benefit disappears. – Studies contradicting or limiting the claim
• Farming carnivorous fish species depletes the wild fish stocks used as feed rather than relieving pressure on the oceans. – Studies contradicting or limiting the claim
• All salmon is farmed and salmon should not be eaten; sea bass, sea bream and similar fish should be eaten instead. – Studies contradicting or limiting the claim