TL;DR
Biological age is an estimate of how much physiological deterioration has occurred in a person’s body relative to others of the same chronological age. Unlike chronological age, it is modifiable.
- Five main measurement approaches exist. Epigenetic clocks, telomere length, blood biomarker panels, glycan analysis, and composite multi-omic scores. They do not measure the same thing.
- Epigenetic clocks have gone through three generations. DunedinPACE (third-generation, pace-of-aging) is the current gold standard for tracking intervention response.
- Blood biomarker panels offer the fastest feedback, in weeks. Epigenetic clocks shift in 6 to 12 months. Telomere length is too noisy for individual clinical decisions.
- Biological age responds to intervention. Caloric restriction, exercise, sleep optimization, and inflammation management have documented effects on epigenetic age.
- Inside a clinical workflow, the question is not “what is my biological age” but “which measurement gives the right signal for the intervention being tracked.”
The Horvath Paradox and the Illusion of Accuracy
In 2013, Steve Horvath at UCLA published an epigenetic clock that estimated a person’s age from DNA methylation patterns with a correlation of 0.96 to chronological age (Horvath, Genome Biology 2013). That accuracy looks impressive. It is also, paradoxically, why first-generation clocks are not the most useful tool inside a longevity clinic.
A clock trained to predict the number on a passport cannot reliably distinguish someone aging fast from someone aging slow if their methylation happens to match their age. The biological age field has moved through three generations of clocks since Horvath’s original paper. Five distinct measurement approaches exist in current practice. Each measures something different.
This piece covers what biological age actually is, the five measurement methods, what each method is good for clinically, and how the underlying biology responds when patients change their behavior.
Biological age vs chronological age
Chronological age is time elapsed since birth. Biological age is the physiological state of the body relative to others of the same chronological age. The two diverge in most adults.
Two 45-year-olds can have biological ages varying by a decade or more. The clinical relevance is direct. Biological age predicts disease incidence and mortality more reliably than chronological age alone. Jylhävä, Pedersen, and Hägg’s 2017 EBioMedicine review remains the foundational reference for this distinction.
The concept itself is older than the measurement tools. Alex Comfort proposed a “test-battery to measure ageing-rate in man” in The Lancet in 1969. The idea was sound. It took 44 years for the technology to make it operational, with Horvath’s 2013 epigenetic clock as the first scalable implementation.
One distinction is worth establishing early. Biological age can be measured as a static state (a single number representing accumulated aging) or as a rate (how fast aging is happening right now). These are different questions with different answers. The static-state framing dominated the first decade of biological age testing. The rate framing is increasingly where serious clinical work is moving.
The five main ways biological age is measured
Five measurement approaches dominate the current literature. They do not measure the same thing. The table summarizes the practical differences. Detail follows.
| Method | What it measures | Accuracy / Reliability | Actionability | Time to detect change |
| Epigenetic clocks (DNA methylation) | CpG site methylation patterns | High (varies by generation) | Moderate | 6 to 12 months |
| Telomere length | Length of chromosome end caps | Low to moderate (noisy) | Low | 12+ months |
| Blood biomarker panels | Organ function, inflammation, metabolic markers | Moderate to high | High | 4 to 12 weeks |
| Glycan analysis (IgG glycosylation) | Antibody sugar structures | Moderate | Moderate | 3 to 6 months |
| Composite / multi-omic scores | Multiple data streams combined | Varies by algorithm | High | Depends on inputs |
Biological Age Test Methods and Feedback Speed
Epigenetic clocks read DNA methylation patterns at specific genomic sites and translate them into an age estimate. The methodology is covered in detail in the next section.
Telomere length is the oldest commercial biological age test. Telomeres are protective caps on chromosome ends that shorten with each cell division (Blackburn et al. 2015 Science). The correlation with chronological age is roughly 0.51 to 0.55, considerably weaker than methylation-based correlations. Telomere length varies heavily between individuals for genetic reasons and shows poor test-retest reliability in commercial assays. It remains useful for population research. It is too noisy for single-point individual decisions inside a clinical workflow.
Blood biomarker panels measure functional outputs of aging. Inflammation markers (hsCRP, IL-6), metabolic markers (HbA1c, fasting insulin), atherogenic markers (ApoB, Lp(a)), liver and kidney markers (GGT, cystatin C), and immune markers (lymphocyte percentage). Levine and colleagues built PhenoAge in 2018 from nine such markers plus a chronological-age input. The advantage of blood-based composite scores is feedback speed. Changes show up in weeks rather than months.
Glycan analysis examines patterns of sugar attachment to immunoglobulin G antibodies. The patterns shift with age and inflammation (Gudelj et al. 2018 Cell Immunology). Glycan-based scores capture immune and inflammatory aging primarily. They are useful as a complement to blood biomarkers rather than as a primary aging signal.
Systems Age and Integration in Clinical Workflow
Composite and multi-omic scores combine multiple data streams. The 2025 Sehgal and Levine paper in Nature Aging introduced Systems Age, a single methylation test producing 11 system-specific aging scores. It is the most clinically actionable of the multi-omic approaches available now.
Across longevity clinic practice, the methods that hold up clinically are the ones that respond on timescales operators actually use. Weeks to quarters for blood biomarker panels. Two to four quarters for epigenetic measures. Telomere testing tends to be an artifact of consumer marketing rather than part of clinical workflow.
The opportunity is not just more biomarkers. It is better integration, systems that connect biology, data, and clinical reasoning in ways that actually extending health span.
Samir Mitra, Founder and CEO of Reya.ai. LinkedIn, 2024.
Three generations of epigenetic clocks
First-generation clocks
They (Horvath 2013, Hannum 2013) were trained to predict chronological age from DNA methylation patterns. Horvath’s clock uses 353 CpG sites across 51 tissue types and achieves a 0.96 correlation with chronological age. Hannum’s contemporaneous clock works on blood samples. The very accuracy that made first-generation clocks impressive is also their clinical limitation. A clock trained to predict the number on a passport carries limited information about how fast biology is actually decaying.
Second-generation biological age clocks
They were trained on health outcomes rather than chronological age. PhenoAge (Levine et al. 2018, Aging) was trained on a phenotypic age estimate derived from nine clinical biomarkers and mortality risk. GrimAge (Lu et al. 2019, Aging) was trained directly on mortality, incorporating methylation surrogates for plasma proteins including PAI-1, adrenomedullin, and lifetime smoking pack-years.
GrimAge acceleration independently predicts cardiovascular disease, cancer, and all-cause mortality after adjusting for known risk factors. These clocks measure something clinically meaningful, not just an age estimate.
Third-generation clocks
They measure rate rather than state. DunedinPACE (Belsky et al. 2022, eLife) was trained on the rate of decline in 19 biomarkers tracked from age 3 to 51 in the Dunedin longitudinal cohort. The result is a pace-of-aging score. 1.0 means aging one year per calendar year. 0.85 means 15 percent slower than average. 1.2 means 20 percent faster.
Why the Clinical Question Must Drive Clock Selection
Test-retest reliability (ICC) is above 0.90, the highest of any aging clock. The Waziry CALERIE caloric restriction trial documented an 11 percent pace-of-aging reduction over two years, the first randomized-trial evidence of clock responsiveness to a longevity intervention.
The clinical implication is direct. A clinic running DunedinPACE quarterly can detect intervention response within a single quarter. A clinic running first-generation Horvath has effectively no short-term signal. The choice of clock determines whether the test is useful for the question being asked.
From operator experience, the clock selection decision is often made for the wrong reasons. Cost, brand recognition, or whatever the testing vendor packages by default. The right starting point is the clinical question. The clock is the second decision, not the first.
“Aging biology is inherently multi system and probabilistic. AI’s maturity in the form of technology today is predictive/probabilistic (not yet handling causaulity).”
Samir Mitra, Founder and CEO of Reya.ai. LinkedIn, 2024.
Why pace of aging matters more than a static number
A static biological age estimate tells you the accumulated state. A pace-of-aging measurement tells you current trajectory. These are different clinical questions.
A 40-year-old with a biological age of 35 might still be aging at 1.15 years per calendar year right now. The static number is encouraging. The pace might be the actual problem. The reverse is also possible. Biological age 45, pace 0.85. The cumulative state is concerning but the current trajectory is good.
Pace of aging is the clinically actionable metric. It responds to intervention within months. Static biological age takes six to twelve months to shift detectably.
Even small elevations in pace matter at the population level. Belsky 2022 documented that fast agers (DunedinPACE above 1.0 by one standard deviation) had 56 percent higher seven-year mortality and 54 percent higher seven-year chronic disease risk than slow agers.
Inside a longevity clinic workflow, pace of aging is the metric that determines whether quarterly interventions are working. Static biological age is the patient-facing story. The pace of aging is the clinician’s tracking signal. The two have different jobs.
What about telomere length and blood biomarkers
Telomere length deserves a brief honest treatment. Telomeres shorten with each cell division (Blackburn et al. 2015 Science). The correlation with chronological age sits around 0.51 to 0.55, considerably weaker than methylation-based correlations near 0.96. Inter-individual variability is high and largely genetically determined. Test-retest reliability in commercial assays is poor. Telomere testing has its place in population research. As a single-point clinical decision tool for an individual member, it carries too much noise to anchor protocol decisions.
Blood biomarker panels are the operational foundation of clinical longevity practice. The standard panel runs across inflammation (hsCRP, IL-6), metabolism (HbA1c, fasting insulin, HOMA-IR), atherogenic markers (ApoB, Lp(a)), liver and kidney function (GGT, cystatin C, eGFR), immune function (lymphocyte percentage), and hormones (testosterone, estradiol, DHEA-S, thyroid panel). Composite algorithms such as PhenoAge translate the panel into a single biological age estimate. The advantage is feedback speed. Biomarker changes register in four to twelve weeks.
The limitation is coverage. Blood biomarkers capture metabolic, cardiovascular, and inflammatory aging well. They do not directly capture brain aging or musculoskeletal aging. Functional measures (VO2 max, grip strength, gait speed) complement the blood layer.
Glycan analysis captures immune and inflammatory aging through IgG glycosylation patterns. It is faster than epigenetic clocks. Best used as a complement to blood biomarkers, not as the primary signal.
In Reya’s customer base, the blood biomarker layer is the fastest-feedback signal a clinic can use for tracking lifestyle interventions. That is why it tends to be the foundation of clinical longevity protocols, with epigenetic testing layered on annually.
The Systems Age development and where measurement is going
The 2025 Sehgal and Levine paper in Nature Aging introduced Systems Age, a single blood methylation test that produces 11 system-specific aging scores. Heart, Lung, Kidney, Liver, Brain, Immune, Inflammatory, Blood, Musculoskeletal, Hormone, and Metabolic.
The framework matters because aging does not happen uniformly. Two 50-year-olds with the same global biological age can have very different patterns of system-specific aging. One might have a 40-year-old metabolic system and a 60-year-old cardiovascular system. The other might be the reverse. They should be treated differently. A single number obscures the pattern that should drive the protocol.
Sehgal and Levine’s Systems Age outperformed existing global clocks for predicting system-specific diseases. It is the first multi-system clock that translates into actionable clinical differentiation.
Proteomic clocks (Argentieri 2024 Nature Medicine, covering more than 4,000 proteins in body fluids) and metabolic clocks (Deelen 2019 Nature Communications, 14 markers) are advancing in parallel. The direction of the field is clear. Biological age measurement is becoming less about a single number and more about a multi-system multi-omic profile.
“Immune shifts, metabolic changes, and epigenetic drift interact over time. Yet healthcare often treats them in isolation.”
Samir Mitra, Founder and CEO of Reya.ai. LinkedIn, 2024.
What changes biological age
Biological age responds to specific interventions. The evidence base is concrete enough now to name the studies.
Caloric restriction: It has the strongest randomized-trial evidence. The Waziry CALERIE trial documented an 11 percent pace-of-aging reduction (DunedinPACE) over two years of moderate caloric restriction in non-obese adults.
Exercise: It produces measurable epigenetic age reductions in multiple intervention trials. Sustained aerobic and resistance training combined seem to produce the largest effects, though direct head-to-head comparison data is limited.
Sleep optimization: It matters more than most members and many clinicians recognize. DNA methylation maintenance happens during restorative deep sleep. Chronic short sleep accelerates epigenetic aging in measurable ways within days.
Inflammation reduction: It translates directly to biological age improvement on second-generation clocks. GrimAge and PhenoAge incorporate inflammatory surrogates, so lowering hsCRP through diet, omega-3 intake, and exercise produces measurable biological age effects.
Stress management: It has documented effects. Chronic cortisol elevation accelerates epigenetic aging. Meditation, social connection, and time in nature each show measurable epigenetic signatures in intervention studies.
Accelerators: Smoking is the strongest single lifestyle accelerator. It is dose-dependent on every epigenetic clock (Klopack et al. 2022 Clinical Epigenetics). Chronic heavy alcohol consumption shows similar patterns.
One anchor RCT: Fitzgerald and colleagues in 2021 (Aging) documented a 3.23-year decrease in Horvath methylation age over eight weeks in healthy adult males following a methylation-supportive diet and lifestyle program.
The biology responds. The implementation gap is the actual clinical problem.
“The key issue in longevity medicine, is that you must get people to change their behavior, and so you need a system that is going to basically enable behavior change.”
Samir Mitra, Founder and CEO of Reya.ai. Longevity.Technology interview, February 2025.
What this means inside a longevity clinic workflow
The choice of the biological age test is not an academic question. It determines whether the result is actionable inside a clinical workflow.
A practical framework looks like this. Blood biomarker panel quarterly as the fast-feedback foundation. DunedinPACE annually, or every three to six months for aggressive protocols, as the pace-of-aging tracker. GrimAge2 is annually used as the mortality-prediction layer. Systems Age was once commercially available as the multi-system profile. Functional measures (VO2 max, grip strength) are independent longevity predictors that complement lab-based testing.
The harder question
How does a clinic actually coordinate biological age data, blood biomarkers, wearables, functional measures, and behavior-change protocols across hundreds of members without burning out the clinician or overwhelming the member with data?
That is the workflow design Reya was built around. Coordinating biological age data alongside biomarkers, wearables, and behavior-change protocols so the data turns into outcomes.
In Reya’s customer base, the operator pain point is not choosing a test. It is integrating results across multiple modalities over time without the workflow falling apart.
From measurement to outcomes
Biological age is a real, measurable, clinically meaningful concept. Not wellness marketing.
The maturity of the field has moved past science. The methodology question is increasingly answered. The harder question is how to use the answers consistently across years and patients. The clinic that solves that question is the clinic that earns the membership.
Three of the five measurement approaches are clinically robust today. Multi-system measurement is the direction of the next two to three years. The workflow architecture around the data is where the real clinical work happens.
Frequently Asked Questions
Chronological age is the number of years since birth. Biological age estimates how old the body is functioning at the cellular and physiological level, based on biomarkers like DNA methylation patterns, inflammation, and metabolic health. Two people with the same chronological age can have biological ages that differ by a decade or more depending on lifestyle, genetics, and disease burden.
Five main approaches dominate current practice. Epigenetic clocks based on DNA methylation patterns (Horvath, PhenoAge, GrimAge, DunedinPACE). Telomere length testing. Blood biomarker composite scores (PhenoAge-blood). Glycan analysis of IgG antibody patterns. And multi-omic composite scores like Systems Age that combine multiple data streams. They do not measure the same thing.
Some biological age markers do shift downward with sustained lifestyle change, particularly methylation-based clocks responding to caloric restriction, exercise, sleep optimization, and inflammation reduction. The Waziry CALERIE trial documented an 11 percent pace-of-aging reduction over two years. Marketing claims of dramatic age reversal over weeks tend to outpace the underlying evidence.
Blood biomarker composite scores can be retested quarterly because they respond to lifestyle change in weeks. DunedinPACE can be retested every three to six months for aggressive protocols, annually for most practice. First-generation clocks like Horvath need at least 6 to 12 months between tests to detect change above noise. Telomere length is too variable to track usefully at any short interval.
Blood biomarker composite scores respond in 4 to 12 weeks. DunedinPACE responds within months. Static epigenetic clocks like Horvath and PhenoAge shift over 6 to 12 months. Telomere length is too noisy to track meaningful short-term change in individuals. Interventions that change inflammation and metabolic markers show up fastest because second-generation clocks incorporate those signals directly.
Telomere length has a correlation of roughly 0.51 to 0.55 with chronological age, considerably weaker than methylation-based correlations near 0.96. Inter-individual variability is high and largely genetic. Test-retest reliability in commercial assays is poor. Telomere length has its place in population research. It carries too much noise to anchor individual clinical decisions.
References
1. Horvath, S. (2013). DNA methylation age of human tissues and cell types. Genome Biology 14(10), R115. https://doi.org/10.1186/gb-2013-14-10-r115
2. Jylhävä, J., Pedersen, N. L. & Hägg, S. (2017). Biological age predictors. EBioMedicine 21, 29–36. https://doi.org/10.1016/j.ebiom.2017.03.046
3. Comfort, A. (1969). Test-battery to measure ageing-rate in man. The Lancet 294(7635), 1411–1415. https://doi.org/10.1016/S0140-6736(69)90950-7
4. Blackburn, E. H., Epel, E. S. & Lin, J. (2015). Human telomere biology: A contributory and interactive factor in aging, disease risks, and protection. Science 350(6265), 1193–1198. https://doi.org/10.1126/science.aab3389
5. Levine, M. E., Lu, A. T., Quach, A., et al. (2018). An epigenetic biomarker of aging for lifespan and healthspan. Aging 10(4), 573–591. https://doi.org/10.18632/aging.101414
6. Gudelj, I., Lauc, G. & Pezer, M. (2018). Immunoglobulin G glycosylation in aging and diseases. Cellular Immunology 333, 65–79. https://doi.org/10.1016/j.cellimm.2018.07.009
7. Sehgal, R., Markov, Y., Qin, C., et al. (2025). Systems Age: A single blood methylation test to quantify aging heterogeneity across 11 physiological systems. Nature Aging 5, 1880–1896. https://doi.org/10.1038/s43587-025-00958-3
8. Hannum, G., Guinney, J., Zhao, L., et al. (2013). Genome-wide methylation profiles reveal quantitative views of human aging rates. Molecular Cell 49(2), 359–367. https://doi.org/10.1016/j.molcel.2012.10.016
9. Lu, A. T., Quach, A., Wilson, J. G., et al. (2019). DNA methylation GrimAge strongly predicts lifespan and healthspan. Aging 11(2), 303–327. https://doi.org/10.18632/aging.101684
10. Belsky, D. W., Caspi, A., Corcoran, D. L., et al. (2022). DunedinPACE, a DNA methylation biomarker of the pace of aging. eLife 11, e73420. https://doi.org/10.7554/eLife.73420
11. Waziry, R., Ryan, C. P., Corcoran, D. L., et al. (2023). Effect of long-term caloric restriction on DNA methylation measures of biological aging in healthy adults from the CALERIE trial. Nature Aging 3(3), 248–257. https://doi.org/10.1038/s43587-022-00357-y
12. Argentieri, M. A., Xiao, S., Bennett, D., et al. (2024). Proteomic aging clock predicts mortality and risk of common age-related diseases in diverse populations. Nature Medicine 30, 2450–2460. https://doi.org/10.1038/s41591-024-03164-7
13. Deelen, J., Kettunen, J., Fischer, K., et al. (2019). A metabolic profile of all-cause mortality risk identified in an observational study of 44,168 individuals. Nature Communications 10, 3346. https://doi.org/10.1038/s41467-019-11311-9
14. Klopack, E. T., Carroll, J. E., Cole, S. W., Seeman, T. E. & Crimmins, E. M. (2022). Lifetime exposure to smoking, epigenetic aging, and morbidity and mortality in older adults. Clinical Epigenetics 14, 72. https://doi.org/10.1186/s13148-022-01286-8
15. Fitzgerald, K. N., Hodges, R., Hanes, D., et al. (2021). Potential reversal of epigenetic age using a diet and lifestyle intervention: A pilot randomized clinical trial. Aging 13(7), 9419–9432. https://doi.org/10.18632/aging.202913