Two fifty-year-olds walk into a clinic for a routine blood panel. One has the physiology of a typical forty-three-year-old; the other, of someone pushing sixty. Their passports say the same thing. Their blood tells a very different story — and a growing body of evidence suggests that the blood is the better predictor of what comes next.
This is the core insight behind the concept of biological age: that the rate at which your body is ageing matters more than the number of birthdays you have had. Over the past decade, a new generation of ageing clocks has turned that insight from an abstract idea into something measurable — and the results, replicated across several large independent cohorts, are reshaping how longevity science thinks about health, risk, and what is modifiable.
Evidence rating: STRONG for biological age clocks predicting mortality and morbidity; EMERGING for whether lifestyle changes can meaningfully lower a biological age score. The honest gap: no large randomised trial has yet proved that lowering a biological age score extends human lifespan.
From birthday candles to biology
Your chronological age is simple arithmetic — today's date minus the day you were born. It tells you nothing about how well your cardiovascular system, kidneys, liver, immune cells, or metabolic machinery are functioning relative to the average person your age. Two people born on the same day can be decades apart biologically, and the divergence widens with time.
This has been intuited by clinicians for as long as medicine has existed ("she looks good for her age"), but it remained subjective until researchers began building composite models that could quantify the gap. The first breakthrough came in 2013, when Steve Horvath published a multi-tissue DNA methylation clock trained on over 8,000 samples spanning 51 tissue types [1]. By measuring chemical tags on DNA at 353 specific sites, the clock could estimate a person's age with striking accuracy across tissue types. That 2013 paper established the clock itself; the link to hard outcomes came two years later, when an analysis pooling four independent cohorts showed that the gap between methylation age and chronological age in blood (known as "age acceleration") was associated with all-cause mortality, and remained so after adjustment for chronological age and the other covariates measured in those cohorts [9].
Second-generation clocks: prediction, not just correlation
Horvath's original clock was trained to match chronological age as closely as possible. It was a remarkable proof of concept, but the clocks that followed asked a sharper question: instead of matching passport age, what if you trained a clock directly on the things that matter — disease, disability, and death?
PhenoAge (Levine et al., 2018) did exactly this. Rather than training on chronological age alone, it was built from nine routine clinical blood biomarkers — albumin, creatinine, glucose, C-reactive protein, lymphocyte percentage, mean cell volume, red cell distribution width, alkaline phosphatase, and white blood cell count — that together captured mortality risk better than any single marker [2]. The resulting "phenotypic age" strongly predicted all-cause and cause-specific mortality, as well as healthspan and physical functioning. Crucially, PhenoAge could be computed from a standard blood panel — no expensive methylation array required — which puts a second-generation clock within reach of routine blood work.
A follow-up study in over 11,000 adults from the nationally representative US NHANES cohort confirmed the finding: each year of phenotypic age acceleration (appearing older than expected given your chronological age) was significantly associated with higher all-cause and cause-specific mortality, even among people who reported no existing diseases and had a normal BMI [3]. In other words, biological age captured risk that traditional clinical assessment missed.
GrimAge (Lu et al., 2019) pushed further, incorporating DNA-methylation-based surrogates of seven plasma proteins (including PAI-1 and growth differentiation factor 15) plus a methylation estimate of smoking pack-years [4]. The result was a clock that, in the authors' validation data, predicted time-to-death with extraordinary statistical strength (Cox regression p = 2.0 × 10⁻⁷⁵) and outperformed earlier clocks on outcomes including coronary heart disease and comorbidity count.
DunedinPACE (Belsky et al., 2022) took a different approach entirely. Instead of estimating how old you are biologically, it estimates how fast you are ageing right now [5]. Trained on the Dunedin Study birth cohort — tracking within-individual decline across 19 organ-system indicators over two decades — DunedinPACE distilled a twenty-year trajectory of multi-organ decline into a single DNA methylation blood test. A score of 1.0 means you are ageing at the expected rate; above 1.0 means faster; below, slower. In independent validation, DunedinPACE predicted morbidity, disability, and mortality, and showed markedly better test–retest reliability than earlier methylation clocks — which matters if you intend to measure the same person more than once [5].
What this means in practice
A representative study of older US adults (from the Health and Retirement Study) tested PhenoAge, GrimAge, and DunedinPACE head-to-head in a population-based sample [6]. All three "second and third generation" clocks were consistently and significantly associated with cognitive dysfunction, functional limitations, chronic conditions assessed two years later, and four-year mortality — confirming that these are not laboratory curiosities. They capture something real about the body's ageing trajectory.
The practical implication is straightforward: two people of the same chronological age can sit at very different points on the morbidity curve, and a composite biological age measure — built from multiple organ-system signals, not a single molecule — can reveal the gap. This is why the field has moved from single-biomarker thinking ("my cholesterol is fine") to composite, multi-system scoring.
Can you change your biological age?
This is the question that drives most consumer interest, and the honest answer is: the early evidence is encouraging, but it is early.
A small pilot randomised controlled trial (n = 43 healthy men aged 50–72) tested an eight-week diet and lifestyle programme — including dietary guidance, sleep, exercise, relaxation, probiotics, and phytonutrients — against a no-intervention control [7]. The treatment group showed an average 3.23-year decrease in Horvath DNAm age compared with controls (p = 0.018). The trial was small, used the first-generation Horvath clock rather than the newer clocks with stronger mortality prediction, and has not yet been replicated at scale. It is a signal, not a conclusion.
Longitudinal observational data from the Melbourne Collaborative Cohort Study (n = 1,041, followed over ~11 years) added another piece: improvements in diet quality (measured by the Alternative Healthy Eating Index) were associated with slower epigenetic ageing by GrimAge, PhenoAge, and DunedinPACE, while weight gain was associated with faster ageing [8]. Stable weight was linked to the lowest epigenetic ageing at follow-up.
The honest gap: no large, adequately powered randomised trial has yet demonstrated that lowering a biological age score extends human lifespan or prevents specific diseases. Association is not proof of cause. The clocks are validated as predictors of outcomes in large cohorts, but the question of whether deliberately moving the score translates into the outcomes the score predicts remains under active investigation. This is the difference between a validated risk marker and a proven intervention target — an important distinction that the consumer longevity market does not always make clearly.
Measurement noise: why a single reading is not the whole story
Biological age clocks — whether blood-based or epigenetic — carry measurement noise that is worth understanding.
A blood-based biological age score depends on clinical biomarkers that can shift day to day. An acute infection will spike your C-reactive protein and white blood cell count, temporarily inflating your biological age. A dehydrating run the morning before your blood draw could shift creatinine. This is not a flaw in the clock — it is reflecting real physiological state — but it means a single snapshot should not be over-interpreted.
Epigenetic clocks face their own technical variability: batch effects between laboratory runs, differences in DNA extraction, and the inherent noise of methylation array measurement. DunedinPACE was specifically designed with test-retest reliability in mind, restricting its model to methylation probes with high reliability [5], but no clock is noise-free.
The practical takeaway is that biological age is most informative as a trend over time, not as a single number on a single day. A baseline reading gives you a starting point; a repeat reading months later, under similar conditions, tells you the direction of travel. It is the trajectory — not the decimal — that carries the signal.
Composite, not single-molecule
One of the most important lessons from the development of these clocks is that ageing is a multi-system process. PhenoAge draws on nine organ-system markers. GrimAge incorporates seven protein surrogates plus smoking exposure. DunedinPACE tracks decline across 19 indicators of organ integrity.
No single molecule captures the complexity of how a body ages, and the history of the field is littered with candidate markers that looked decisive in one cohort and faded under wider scrutiny. That is precisely why the clocks above were built as composites in the first place: each was trained on many organ-system signals at once, and each was validated against hard outcomes rather than against a single molecule's plausibility. A composite, multi-system score is far harder to fool — or to be fooled by — than any one trending molecule.
How Omniwo helps you measure this
Omniwo's BioAge test computes a blood-based biological age using a PhenoAge-family algorithm built from nine core clinical biomarkers — the same class of multi-system composite that the research above validates. It is a finger-prick, at-home test: no clinic visit, no expensive saliva methylation kit, no waiting weeks for a lab in another country.
What you get is a composite score — not a single-marker snapshot — alongside the individual biomarker values that drive it. That means you can see which systems are contributing to an older or younger biological age: is it your inflammatory marker (hs-CRP)? Your glycaemic control (HbA1c)? Your kidney function? The score tells you where to look; the trend over re-tests tells you whether the changes you are making are registering.
The honest framing: a BioAge test measures your current physiological state relative to population norms. It does not diagnose any disease, it does not tell you how long you will live, and it does not replace clinical advice. What it does is give you a baseline — and a reason to re-test. If the research holds, the direction you move your biological age over months and years may be one of the most meaningful health signals you can track.
Omniwo's tests and content are for wellness and educational insight. They are not a medical device, do not diagnose, treat, cure or prevent any disease, and do not replace advice from a qualified healthcare professional.
Start with a baseline: explore the BioAge test in the shop →
Sources
-
Horvath S. DNA methylation age of human tissues and cell types. Genome Biology. 2013;14(10):R115. doi:10.1186/gb-2013-14-10-r115 (PMID: 24138928)
-
Levine ME, Lu AT, Quach A, et al. An epigenetic biomarker of aging for lifespan and healthspan. Aging. 2018;10(4):573–591. doi:10.18632/aging.101414 (PMID: 29676998)
-
Liu Z, Kuo PL, Horvath S, Crimmins E, Ferrucci L, Levine M. A new aging measure captures morbidity and mortality risk across diverse subpopulations from NHANES IV: A cohort study. PLoS Medicine. 2018;15(12):e1002718. doi:10.1371/journal.pmed.1002718 (PMID: 30596641)
-
Lu AT, Quach A, Wilson JG, et al. DNA methylation GrimAge strongly predicts lifespan and healthspan. Aging. 2019;11(2):303–327. doi:10.18632/aging.101684 (PMID: 30669119)
-
Belsky DW, Caspi A, Corcoran DL, et al. DunedinPACE, a DNA methylation biomarker of the pace of aging. eLife. 2022;11:e73420. doi:10.7554/eLife.73420 (PMID: 35029144)
-
Faul JD, Kim JK, Levine ME, Thyagarajan B, Weir DR, Crimmins EM. Epigenetic-based age acceleration in a representative sample of older Americans: Associations with aging-related morbidity and mortality. Proceedings of the National Academy of Sciences. 2023;120(9):e2215840120. doi:10.1073/pnas.2215840120 (PMID: 36802439)
-
Fitzgerald KN, Hodges R, Hanes D, et al. Potential reversal of epigenetic age using a diet and lifestyle intervention: a pilot randomized clinical trial. Aging. 2021;13(7):9419–9432. doi:10.18632/aging.202913 (PMID: 33844651)
-
Li DL, Hodge AM, Cribb L, Southey MC, Giles GG, Milne RL, Dugué PA. Body Size, Diet Quality, and Epigenetic Aging: Cross-Sectional and Longitudinal Analyses. The Journals of Gerontology: Series A. 2024;79(4). doi:10.1093/gerona/glae026 (PMID: 38267386)
-
Marioni RE, Shah S, McRae AF, et al. DNA methylation age of blood predicts all-cause mortality in later life. Genome Biology. 2015;16:25. doi:10.1186/s13059-015-0584-6 (PMID: 25633388)
This article is educational and not medical advice. See our medical disclaimer.