The short version
- Pooling 17 prospective cohorts with 42,466 people and 15,720 deaths, all-cause mortality was 15 to 18% lower in the highest versus lowest quintile of circulating long-chain omega-3 fatty acids.
- In 2,500 Framingham Offspring participants, those with an omega-3 index above 6.8% had a 34% lower risk of death from any cause than those below 4.2%, while total cholesterol was not significantly related in the same models.
- VITAL randomised 25,871 adults to 1g of marine omega-3s daily for a median 5.3 years and found a hazard ratio of 0.92 for major cardiovascular events and 1.03 for invasive cancer, neither significant.
- A meta-analysis of 10 trials in 77,917 high-risk individuals found no significant association of omega-3 supplements with coronary heart disease death or major vascular events, overall or in any subgroup.
- The trials enrolled people regardless of baseline omega-3 status, so a benefit confined to those in the lowest part of the distribution would not be detectable in their design.
Most longevity biomarkers fail on the first question: is the thing you are measuring actually related to dying? The omega-3 index passes that test comfortably. It then walks straight into the second question, which is whether changing it changes anything, and that is where the story gets uncomfortable.
What the index is
Harris and von Schacky proposed it in 2004: the percentage of EPA plus DHA in your red blood cell membranes. Red cells turn over slowly, so unlike a plasma reading this reflects months of intake rather than what you ate yesterday. In their original review of primary and secondary prevention data, an index at or above 8% was associated with the greatest protection against coronary heart disease death, and 4% or below with the least.
That is a well-designed biomarker: stable, cheap, physiologically meaningful, and modifiable.
The association held up at scale
The Fatty Acids and Outcomes Research Consortium pooled 17 prospective cohorts, 42,466 individuals and 15,720 deaths over a median 16 years. After multivariable adjustment, risk of death from all causes was 15 to 18% lower in the highest versus lowest quintile of circulating long-chain omega-3s. The same pattern appeared for cardiovascular, cancer and other-cause death. Notably, no association was seen for alpha-linolenic acid, the plant-derived 18-carbon omega-3 — the signal was specific to the marine fatty acids.
The Framingham Offspring cohort gives the sharper version: 2,500 participants with a mean age of 66, 350 deaths over a median 7.3 years, models adjusted for 18 variables. Those in the highest index quintile (above 6.8%) had a 34% lower risk of death from any cause and 39% lower risk of incident cardiovascular disease than those in the lowest (below 4.2%). In the same models, total cholesterol was not significantly related to those outcomes and the omega-3 index was.
One disclosure the paper itself makes and we will repeat: the lead author holds stock in the laboratory that sells omega-3 index testing. That does not make the data wrong. It is exactly the kind of thing you should know before you order the test.
Then the trials
VITAL randomised 25,871 US adults to 1g per day of marine omega-3s or placebo for a median 5.3 years. Major cardiovascular events: hazard ratio 0.92, confidence interval 0.80 to 1.06. Invasive cancer: 1.03. Death from any cause: 1.02. No excess bleeding, and no benefit on the primary endpoints.
The Omega-3 Treatment Trialists' Collaboration pooled 10 trials and 77,917 high-risk individuals, assessing 6,273 coronary events. Coronary heart disease death: rate ratio 0.93. Major vascular events: 0.97. No significant association overall, and none in any subgroup — including people with existing coronary disease, diabetes, high lipids or statin use. Their conclusion states it plainly: no support for current recommendations to use these supplements in people with a history of coronary heart disease.
How both things can be true
Three explanations, and they are not exclusive.
The first is confounding. People with a high omega-3 index eat fish, and people who eat fish differ from people who do not in a hundred ways a statistical model handles imperfectly.
The second is reverse causation of a subtle kind — chronic illness alters fatty acid metabolism, so a low index can be a consequence of poor health as much as a cause.
The third is the one the field finds most interesting: the trials did not select on the biomarker. VITAL enrolled a general older population regardless of baseline omega-3 status, then gave everyone the same 1g. If the benefit lives in moving people out of the bottom of the distribution, a trial that mostly recruits people already in the middle will find nothing, and will keep finding nothing however many times you run it.
That is exactly the argument now being made about the vitamin D trials, where the genetic evidence points to benefit concentrated in deficiency and the trials enrolled the replete. The same structural mistake, twice, in two different nutrients.
Does that justify supplementing?
Not on its own. "The trial design might have hidden a benefit" is a hypothesis, and a hypothesis is not a result. The honest state of it is that a low omega-3 index reliably marks higher risk, and that giving unselected people a gram of fish oil has repeatedly failed to lower it.
What the third explanation does justify is measuring before you buy. If your index comes back at 3%, you are in the part of the distribution the association is strongest in and food or a supplement is a defensible response. If it comes back at 8%, the trials say your gram of fish oil is not doing what you hoped.
What we would actually tell you
Eat oily fish twice a week if you like it. That is the version of this with the cohort data attached and it comes with protein and a meal.
If you want the test, know that it costs roughly what two months of the supplement costs, and that it turns an open-ended purchase into a decision with a number behind it. That is a better use of money than a subscription bought on a mortality statistic from an observational study.
And keep the scale honest. The omega-3 index gap in Framingham was 34% between extreme quintiles of an observational cohort. Being in the least fit fifth for VO2 max carried a fourfold hazard in a cohort of three-quarters of a million people. Biological age tests, by contrast, cannot even agree with each other. Not every marker deserves the same share of your attention.
Good questions
Should I take fish oil to live longer?
The randomised evidence does not support it. VITAL gave 25,871 adults 1g daily for over five years with no reduction in cardiovascular events, cancer or death, and a pooled analysis of 77,917 high-risk people found nothing either. The blood level predicts mortality; the capsule has not been shown to change the outcome.
What is a good omega-3 index score?
The original proposal put the greatest cardioprotection at 8% or above and the least at 4% or below, and the Framingham cohort split its quintiles at 6.8% and 4.2%. Those cut-points come from observational data, so treat them as a map of the risk distribution rather than a target with a proven benefit to hitting it.
Is the omega-3 index test worth paying for?
It is worth it if the result would change what you do. A low reading puts you in the part of the distribution where the association is strongest and makes eating more fish a defensible response. A high reading tells you a supplement is unlikely to help. Be aware the researcher who proposed the index holds stock in a testing laboratory.
Why did the fish oil trials fail if the blood level matters so much?
Three reasons, probably overlapping. Fish eaters differ from non-fish eaters in ways models cannot fully adjust for. Chronic illness itself lowers omega-3 levels. And the trials enrolled people regardless of baseline status, so any benefit confined to the deficient would be diluted away. That last one is a hypothesis, not an excuse.
Is plant-based omega-3 as good as fish oil?
Not for this marker. In the pooled analysis of 17 cohorts, no association with mortality was seen for alpha-linolenic acid, the 18-carbon plant omega-3, while the marine 20 to 22-carbon fatty acids showed the 15 to 18% difference. Conversion of the plant form to EPA and DHA in humans is limited. Algal oil is the vegetarian route to the marine forms.
Do I need to take fish oil if I already eat fish?
Probably not. Two servings of oily fish a week is the exposure the cohort data was built on, and it comes with protein and a meal rather than a capsule. If you want certainty, measure your index once and decide from the number. Adding a supplement on top of an already adequate intake is the situation the null trials describe best.
Sources
- Harris WS, Von Schacky C. The Omega-3 Index: a new risk factor for death from coronary heart disease? Prev Med, 2004. View study
- Harris WS, Tintle NL, Imamura F, et al. Blood n-3 fatty acid levels and total and cause-specific mortality from 17 prospective studies. Nat Commun, 2021. View study
- Harris WS, Tintle NL, Etherton MR, Vasan RS. Erythrocyte long-chain omega-3 fatty acid levels are inversely associated with mortality and with incident cardiovascular disease: The Framingham Heart Study. J Clin Lipidol, 2018. View study
- Manson JE, Cook NR, Lee IM, et al. Marine n-3 Fatty Acids and Prevention of Cardiovascular Disease and Cancer. N Engl J Med, 2019. View study
- Aung T, Halsey J, Kromhout D, et al. Associations of Omega-3 Fatty Acid Supplement Use With Cardiovascular Disease Risks: Meta-analysis of 10 Trials Involving 77 917 Individuals. JAMA Cardiol, 2018. View study
These statements have not been evaluated by the Food and Drug Administration. This product is not intended to diagnose, treat, cure, or prevent any disease.
