Epigenetic Clocks and the Biological-Age Boom

Send a tube of blood or a swab of saliva to a lab, and within weeks a report can tell you that your body is “really” 47 even though your passport says 52. That is the promise of biological-age testing, a fast-growing corner of the longevity industry built on epigenetic clocks. These tests estimate biological age by reading chemical tags on your DNA rather than the DNA sequence itself. The science behind them is real and widely used in research, but the consumer numbers come with measurement noise and active scientific debate about how much a single result can tell you. The short answer to whether they “work”: the underlying biology is well established, yet the precise age figure on a one-time consumer report should be treated as an estimate, not a verdict.

What an epigenetic clock actually measures

Epigenetic clocks read DNA methylation, the addition of methyl groups to cytosine bases of the DNA. These tags switch genes on or off without changing the genetic code, and their pattern shifts in predictable ways as cells age. The first widely used clock, published by geneticist Steve Horvath in 2013, analyzes methylation at 353 selected sites and uses a statistical model to produce a single age estimate, according to a Nature Reviews Genetics overview of the field. Horvath’s was the first “pan-tissue” clock, meaning it could estimate age from blood, saliva, brain or other tissues, as the Max Planck Institute for Biology of Ageing explains.

The key point for readers: a clock does not directly measure damage or disease. It measures methylation patterns that correlate with chronological age and, in newer versions, with mortality risk. The output is a model’s best guess.

Horvath, GrimAge and DunedinPACE: three different questions

Not all clocks answer the same question, and conflating them causes much of the confusion in marketing material. A useful framing comes from longevity educators who compare the older clocks to an odometer and the newest to a speedometer.

First-generation clocks such as Horvath’s were trained to predict chronological age. Second-generation clocks, including GrimAge, were trained on health outcomes. GrimAge is built from methylation surrogates of seven plasma proteins linked to disease plus smoking pack-years, and it is among the strongest methylation predictors of mortality, according to a comparison of the major clocks. DunedinPACE, a third-generation tool, does something different: it estimates your current rate of aging. Per the DunedinPACE paper in eLife, a score of 1.0 means you are aging one biological year per calendar year, while 0.85 implies a slower pace.

That distinction matters when judging interventions. In the CALERIE randomized trial of caloric restriction, the calorie-restricted group showed roughly a 3% slowing of DunedinPACE, while GrimAge and PhenoAge did not change, according to the study published in Nature Aging. The same researchers note that in a separate cohort, a 3% slower DunedinPACE was associated with a 15% lower risk of death, as summarized by Columbia University’s Mailman School of Public Health.

The consumer-test boom and what it costs

What began as a research tool is now a retail product, riding the same wave of interest that surrounds projects like Bryan Johnson’s Blueprint protocol and the experimental longevity drugs such as rapamycin, metformin and senolytics.

TruDiagnostic, one of the larger players, sells its blood-based TruAge test for $499, or $249 on a subscription plan, and the company says its report includes a biological-age score, a DunedinPACE pace-of-aging measure, organ-system age estimates and dozens of additional biomarkers, according to a third-party review of the service. Elysium Health’s saliva-based Index retails around $499 with subscription pricing near $299, while Tally Health’s saliva test, TallyAge, is priced at $229 per kit, per the same review and a GlycanAge comparison of the two kits. Tally Health was co-founded with the involvement of Harvard geneticist David Sinclair, a detail buyers often cite, though these are company and reviewer descriptions rather than independent validation.

The reliability debate scientists want you to know about

Here is the caveat the glossy reports tend to underplay. Researchers distinguish between technical reliability, meaning whether the same blood sample gives the same answer twice, and biological reliability, meaning whether the same person gives a stable answer across repeated collections over hours or days. The two are not the same.

A 2024 analysis on bioRxiv, “When to Trust Epigenetic Clocks,” warns that retesting the same individual “can lead to fluctuations by several years owing to technical noise” and that epigenetic age “may fluctuate as much as 2 years during the course of a single day,” per the peer-reviewed preprint summary on PubMed Central. The authors caution that a single significant clock result “is not enough to indicate a reliable decrease in biological age.” A separate 2025 bioRxiv study found that across pooled data, most clocks reached only “moderate” to “good” reliability, with intraclass correlation coefficients of roughly 0.4 to 0.7 and no clock hitting the “excellent” range, according to the preprint on biological versus technical reliability.

The practical implication: short-term factors such as a recent meal, stress, poor sleep or even altitude can nudge a result, and a year-to-year change in your score may reflect noise rather than real biological change.

How to read your number sensibly

If you take a test, treat the figure as a population-based estimate with error bars, not a precise readout. Researchers favoring careful interpretation suggest looking for changes that show up across multiple newer-generation clocks rather than reacting to one moving number. Pace-of-aging measures like DunedinPACE respond to interventions faster than static age clocks, which is why trials increasingly rely on them.

These tests also do not replace conventional medicine. Standard checks of blood pressure, cholesterol, glucose, and the lifestyle basics of sleep, movement and diet remain the evidence-backed levers, and they cost far less than a methylation kit. Some clinics now bundle epigenetic testing into broader workups, as seen at members-only operations like Biograph, but bundling does not settle the reliability questions above.

This is journalism, not medical advice. Anyone considering a test, or making decisions based on one, should discuss the results with a qualified clinician.

FAQ

Are epigenetic clocks accurate? They are accurate in the sense that the underlying methylation science is well established and used widely in research. Individual consumer results, however, carry measurement noise; studies report that scores can shift by years from technical variation alone, so a single number is best read as an estimate.

Which clock is best for tracking lifestyle changes? Pace-of-aging measures such as DunedinPACE appear more responsive to interventions over weeks to months than older age-prediction clocks like Horvath’s or GrimAge, according to the CALERIE trial. No clock is a definitive verdict, and experts suggest watching for consistent signals across several clocks.

How much do biological-age tests cost? Published prices range from about $229 for Tally Health’s saliva kit to roughly $499 for blood-based TruDiagnostic or saliva-based Elysium Index, with lower rates on subscription plans, per third-party reviews.

Sources

1. Nature Reviews Genetics — DNA methylation-based biomarkers and the epigenetic clock theory of ageing
2. Max Planck Institute for Biology of Ageing — What is the epigenetic clock?
3. SuperAge — Epigenetic clocks explained: Horvath, GrimAge, DunedinPACE
4. eLife — DunedinPACE, a DNA methylation biomarker of the pace of aging
5. Nature Aging — Effect of long-term caloric restriction on DNA methylation measures of biological aging in the CALERIE trial
6. Columbia University Mailman School of Public Health — Calorie Restriction Slows Pace of Aging in Healthy Adults
7. KnowYourDNA — TruDiagnostic / TruAge review
8. GlycanAge — TruAge vs Elysium Index comparison
9. bioRxiv / PubMed Central — When to Trust Epigenetic Clocks: Avoiding False Positives in Aging Interventions
10. bioRxiv — Biological versus Technical Reliability of Epigenetic Clocks