Whenever people talk about their “real” biological age, the word GrimAge tends to come up. It is an epigenetic clock – an algorithm that uses chemical marks on DNA to estimate how fast you are ageing. Unlike older clocks, GrimAge was not built to estimate calendar age but to predict the risk of death. That is exactly what makes it interesting – and exactly why its result is easy to overinterpret.
How to read the numbers: Fact = a finding from research or biology
of the GrimAge clock
and for smoking
epigenetic clocks
What GrimAge is
GrimAge was introduced in 2019 by a team led by Ake T. Lu and Steve Horvath. It is based on DNA methylation – the attachment of small methyl groups to specific sites in DNA known as CpG sites. The methylation pattern changes with age and with lifestyle.
The authors first built methylation-based “surrogates” for seven blood plasma proteins linked to inflammation and metabolism – GDF-15, PAI-1, cystatin C, leptin, adrenomedullin, beta-2 microglobulin and TIMP-1 – and for smoking pack-years. They then combined these components with age and sex into a model that best predicted time to death. The result is expressed in years. GrimAge2, published in 2022, added surrogates for C-reactive protein (CRP) and glycated haemoglobin (HbA1c).
Three generations of epigenetic clocks
- First generation – the Horvath and Hannum clocks (2013) were trained on calendar age. They are less closely linked to disease risk.
- Second generation – PhenoAge (2018) and GrimAge (2019) were trained on markers of health and mortality.
- Third generation – DunedinPACE (2022) does not measure age but the pace of ageing: how many “biological years” accumulate per calendar year. A value of 1.0 corresponds to the average pace.
How GrimAge is measured
The test starts with a blood sample – from a vein or as drops from a finger prick dried on a card (dried blood spot). Saliva tests also exist, but GrimAge was developed on blood, and results from other tissues may not be comparable.
In the laboratory, DNA is isolated from blood cells and a methylation array reads the methylation status at hundreds of thousands of CpG sites. The algorithm then calculates two main numbers:
- GrimAge in years – the estimated epigenetic age.
- Age acceleration – the difference from your calendar age. A positive number means an “older” result, a negative one a “younger” result.
In Europe, commercial epigenetic tests typically cost in the range of a few hundred euros, and results arrive within a few weeks.
What GrimAge predicts
In large cohorts, a GrimAge higher than calendar age is associated with a higher risk of death from any cause, earlier onset of cardiovascular disease, and cancer. In comparative studies, GrimAge – together with DunedinPACE – is among the clocks that predict health risks best.
This is, however, a statistical association at group level. GrimAge is a risk marker, not a diagnosis. It does not replace measuring blood pressure, cholesterol or blood glucose, or regular check-ups.
A result several years “older” than your age is a signal to focus on the basics – not smoking, movement, sleep, body weight. It is not a reason to panic.
What influences GrimAge
Most of the evidence comes from observational studies, which show associations rather than necessarily causes.
- Smoking – the strongest known factor. It leaves a marked imprint on DNA methylation, and GrimAge includes it directly through the pack-years surrogate. In former smokers, some of these changes gradually ease.
- Obesity and visceral fat – repeatedly linked to higher GrimAge; components such as leptin and PAI-1 are closely tied to fat tissue.
- Physical activity and fitness – people who are more active and have better cardiorespiratory fitness tend to show lower GrimAge acceleration.
- Diet and alcohol – better diet quality is associated with more favourable values, higher alcohol intake with less favourable ones.
- Sleep – the findings are less consistent; associations have been found mainly with insomnia.
- Stress and living conditions – chronic stress and lower socioeconomic status are linked to faster ageing, so this is not only about personal choices.
What intervention trials have shown
Randomised trials are few and their effects are modest. In a secondary analysis of the CALERIE trial (two years of moderate calorie restriction), the pace of ageing measured by DunedinPACE slowed, but no clear effect was found on GrimAge or PhenoAge. In the DO-HEALTH trial (2025), three years of omega-3 fatty acids in people over 70 was associated with a small slowing of several clocks, including GrimAge2 – on the order of a few months. Evidence that supplements can “turn back” GrimAge by years is weak.
- Strong evidence Higher GrimAge acceleration is associated with higher mortality and cardiovascular disease in large cohorts; smoking is the strongest known factor that raises it.
- Moderate evidence Obesity, low physical activity and fitness, higher alcohol intake and lower socioeconomic status are linked to higher GrimAge (observational studies).
- Weak evidence Small slowing of clocks with interventions such as calorie restriction or omega-3 in older adults; associations with sleep and specific diets.
- No evidence “Rejuvenating” GrimAge by years with dietary supplements or commercial protocols.
Limits of measurement: what to watch out for
- Test-retest noise – even the same sample does not give exactly the same result twice; a difference of around a year can be purely technical.
- Differences between laboratories – different arrays, data processing and algorithm versions; results from two companies are not directly comparable.
- Repeat with the same provider – the same test, the same sample type, several months apart.
- A single result is only a snapshot – acute illness, infection or a shift in blood cell composition can move the result.
- No medical decisions based on GrimAge alone – do not change any treatment or medication because of this number.
What matters is the trend, not a single number. Treat a change as real only when it is repeated and consistent.
More accessible alternatives
PhenoAge can be estimated from a routine blood count and biochemistry panel (albumin, creatinine, glucose, CRP and others) – our site offers a free Biological Age calculator. GlycanAge measures sugar chains (glycans) attached to IgG antibodies, which are related to inflammation; the evidence base is still smaller. Functional markers such as VO2max and grip strength are strongly associated with longevity and can be improved with training.
If you decide to take a test
- Not every “biological age” test calculates GrimAge – check which clocks the laboratory uses (GrimAge, GrimAge2, DunedinPACE…) and from what sample.
- If GrimAge is what interests you, choose blood over saliva.
- Read the result as age acceleration, not as an absolute age.
- Measure change by repeating the test after 6 to 12 months, not after a few weeks.
- Prioritise proven steps: do not smoke, stay active, sleep well, keep a healthy waist circumference and limit alcohol.
- Red flag: offers that immediately sell expensive “rejuvenation” supplements alongside your result.
Conclusion
GrimAge is one of the best available epigenetic risk markers. At population level it predicts mortality and disease well. For an individual, however, it is an expensive number with natural noise. Treat it as motivation, not a diagnosis – GrimAge improves with the same unglamorous habits that help even without a test.