Omega-3 and biological aging: what actually changed in DO-HEALTH

You read that omega-3 may make you almost four months younger, and the temptation arrives quickly: should I buy a bottle today? The short answer is no. The finding is real and worth discussing, but it describes a small change in certain blood markers, not whole-body rejuvenation or a prescription for everyone.

The DO-HEALTH Bio-Age analysis studied omega-3, vitamin D and exercise in older adults. Its greatest value is not an anti-aging capsule. It is a clear example of how easily a molecular signal can be mistaken for a clinical benefit. Let us separate the two.

Clinical and editorial review: 30 August 2026. This content is educational and does not replace individual medical care.

The answer in one minute

Question	What we know
Who took part?	777 Swiss adults aged 70 or older who were relatively healthy and active, drawn from the DO-HEALTH trial.
What did they receive?	Vitamin D3 at 2,000 IU daily, omega-3 at 1 g daily and a home exercise programme, alone or in combination, for three years.
What was measured?	Four blood DNA methylation measures: PhenoAge, GrimAge, GrimAge2 and DunedinPACE.
What changed?	Omega-3 showed a favourable signal in PhenoAge, GrimAge2 and DunedinPACE. All three treatments together added an effect in PhenoAge.
By how much?	A small effect, equivalent to about 2.9 to 3.8 months over the three-year follow-up.
What was not shown?	The analysis did not prove that people lived longer, avoided dementia or had fewer heart attacks because of that shift.

What the study actually did

DO-HEALTH was a randomised trial involving 2,157 European adults aged 70 or older. The biological-age paper, published in 2025, was a later analysis of 777 Swiss participants who had blood samples available at baseline and after three years.

The design combined three interventions: 2,000 IU of vitamin D3 daily, 1 g of omega-3 daily and 30 minutes of home exercise three times per week. The team then compared four methylation clocks. Omega-3 produced a favourable signal in three, while the full combination added an effect in PhenoAge.

Two details stop this becoming a sweeping conclusion. First, it was a post hoc analysis, so this was not the main clinical question for which the trial was originally built. Second, the subgroup was fairly healthy and active. We cannot assume the result will be the same in a frail adult, a 50-year-old or someone who already eats plenty of oily fish.

What an epigenetic clock is, without the jargon

An epigenetic clock examines chemical patterns on DNA, mainly methylation, and summarises them as a number related to age or pace of aging. It does not count wrinkles, look directly at the brain or predict exactly how long one person will live.

Each clock also answers a slightly different question. PhenoAge and GrimAge were built to relate to health and risk; DunedinPACE tries to estimate how quickly the body is changing. Two clocks can therefore respond differently to the same intervention. Our guide to epigenetic clocks and biological age explains what these tests measure and where their limits begin.

In this study, the samples came from blood and only two time points were compared, baseline and year three. The result can be useful for research, but it does not show that muscle, arteries and brain became younger by the same amount.

The result that keeps this grounded

The original DO-HEALTH trial provides the most important counterweight. Among its 2,157 participants, none of the three interventions, alone or combined, produced statistically significant differences in the six primary clinical outcomes after three years: blood pressure, physical performance, cognitive function, non-vertebral fractures and infections.

That does not cancel the epigenetic analysis. It reminds us that a biomarker and a lived health outcome are not the same thing. A signal may appear before a measurable clinical effect, may be too small to change it or may never translate into a meaningful benefit.

DunedinPACE, for example, has been associated in other cohorts with morbidity, disability and mortality. That association makes it interesting, but it does not turn an individual result into a diagnosis. We still lack universal clinical thresholds showing that a particular shift should trigger the purchase, withdrawal or dose increase of a supplement.

Omega-3, vitamin D and exercise tell different stories

Omega-3: an interesting signal, not an automatic indication

Omega-3 produced the most consistent clock signal. Even so, the study did not compare dietary strategies or select people because they were deficient. It cannot tell us whether someone who already gets enough EPA and DHA has the same room to respond.

The useful question is not "does omega-3 work?" but "what problem am I trying to solve?". Eating little fish, treating high triglycerides under medical supervision and chasing a biological-age score are different situations. Dose, product and the balance between benefit and risk change too.

Vitamin D: correct a need, do not chase a high number

Vitamin D alone did not consistently change the four clocks. The Endocrine Society guideline suggests against routine supplementation above recommended intakes and routine 25-OH vitamin D testing in healthy adults aged 50 to 74 without an established indication. The USPSTF concludes that evidence is insufficient to assess the balance between benefits and harms of screening asymptomatic adults.

That does not mean testing is never useful. It may change care when there are compatible symptoms, osteoporosis, malabsorption, selected conditions, medication or risk factors. The aim is to identify and treat a real need, not turn the highest possible result into a longevity medal.

Exercise: even when the clock stays still, the body can gain

The home programme did not produce a clear independent clock effect. It would be a mistake to conclude that exercise failed. Physical activity improves outcomes that matter directly, including strength, mobility, balance, functional capacity and cardiometabolic health. The WHO emphasises that any amount is better than none and that muscle strengthening benefits every age group.

For an older adult, preserving the ability to rise from a chair or climb stairs is usually more valuable than reducing a molecular score by a fraction. The practical goal is to prevent sarcopenia and loss of independence, using a progression suited to the person's real starting point.

Turning the study into a sensible decision

Situation	A reasonable first step
I eat little fish and want to improve my diet	Review the whole dietary pattern first and choose an option that is sustainable.
I have high triglycerides	Assess cardiovascular risk and treatment with a clinician; do not copy the trial dose.
I suspect vitamin D deficiency	Confirm whether there is a clinical reason to test and treat instead of supplementing blindly.
I have lost strength or balance	Prioritise progressive exercise, enough protein and functional assessment.
I want to repeat my epigenetic clock	First define what decision the result would change and allow an interpretable interval.

When a measurement changes no decision, it often adds more anxiety than information. This guide to longevity biomarkers worth tracking can help set priorities before expanding a testing panel.

Safety: one gram is not the same as four

The daily gram used in DO-HEALTH is not equivalent to 4 g daily of certain omega-3 ethyl ester medicines prescribed for hypertriglyceridaemia. The UK medicines regulator warns of a dose-dependent increase in atrial fibrillation risk among patients with cardiovascular disease or risk factors, with the highest signal at 4 g daily.

That does not mean every capsule causes an arrhythmia. It does mean dose should not be escalated without guidance, products should not be combined without checking their composition, and prescribed treatment should not be stopped after reading an article. Palpitations, dizziness or shortness of breath warrant medical advice.

With vitamin D, more is not better either. Persistently excessive doses can raise calcium too far and harm bones, kidneys and the heart. A general upper limit does not replace personal advice, because some conditions and treatments require different doses or greater caution.

Conclusion: interesting, small and still no shortcut

DO-HEALTH leaves us with good news, even if it is less dramatic than the headline: a simple intervention produced a small signal in several epigenetic clocks. It also leaves a better lesson: moving a biomarker is not the same as living longer or feeling better.

If the study encourages you to eat better, investigate a real need and train consistently, excellent. If it pushes you to buy three products and chase biological age every month, it is worth pausing. Useful longevity is visible in sustainable decisions, safety and function, not an isolated number.

Frequently asked questions about omega-3 and biological aging

Does omega-3 slow biological aging?

In the DO-HEALTH Bio-Age analysis, 1 g daily for three years produced a small favourable shift in three of four DNA methylation clocks. It is a biomarker signal, not proof of whole-body rejuvenation or a universal recommendation.

How many months did the epigenetic clock change?

The estimated effects were equivalent to about 2.9 to 3.8 months over three years. That figure describes a statistical difference in blood, not extra months of life.

Does that mean taking omega-3 helps you live longer?

This study cannot tell us. The analysis was not designed to prove longer survival, fewer heart attacks or less dementia. Diet, clinical indication, risks and personal goals matter when making a decision.

Did vitamin D alone move the clocks?

It did not show a consistent effect by itself across the four clocks. The combination of vitamin D, omega-3 and exercise added a favourable signal in PhenoAge, but not in every marker.

Was the exercise in the study intense?

No. It was a simple 30-minute home programme three times per week. Its lack of an independent clock effect does not reduce the known benefits of exercise for strength, mobility, balance and general health.

Should I take 1 g of omega-3 each day?

Not simply because that was the trial dose. The decision depends on diet, triglycerides, cardiovascular history, medication and product type. High prescription doses belong to a different clinical setting.

Do I need a vitamin D test before supplementing?

Routine screening is not advised for every healthy person. Testing can be useful when symptoms, conditions, treatments or risk factors suggest deficiency and when the result would change the decision.

How often should an epigenetic clock be repeated?

There is no universal clinical interval. If it is used, keep the same laboratory and allow enough time for a change to be interpretable. Repeating it every few weeks is more likely to capture noise than progress.

Sources

1. Bischoff-Ferrari HA, et al. DO-HEALTH Bio-Age analysis of vitamin D, omega-3, exercise and DNA methylation clocks. Nature Aging, 2025. Europe PMC.
2. Bischoff-Ferrari HA, et al. Main clinical results from the DO-HEALTH trial. JAMA, 2020. Europe PMC.
3. Belsky DW, et al. Development and validation of DunedinPACE. eLife, 2022. Europe PMC.
4. Endocrine Society. Clinical practice guideline on vitamin D for disease prevention, 2024. Recommendations.
5. US Preventive Services Task Force. Screening for vitamin D deficiency in adults. Recommendation.
6. Medicines and Healthcare products Regulatory Agency. Omega-3 ethyl ester medicines and atrial fibrillation risk. Safety update.
7. NHS. Vitamin D, doses and risks of excess. Clinical information.
8. World Health Organization. Physical activity and benefits for older adults. Fact sheet.

If you take medication, have atrial fibrillation, kidney disease, a calcium disorder or a medical indication for vitamin D or omega-3, seek advice before changing a dose or stopping treatment.