Epigenetic age vs biological age
"Biological age" is not one measurement. It is a category containing at least four quite different methods, which disagree with each other, cost wildly different amounts, and are useful for different purposes.
Epigenetic clocks
DNA methylation clocks measure chemical marks on DNA that change in patterned ways with age. Horvath's 2013 clock was the first widely used; Hannum, PhenoAge, GrimAge and DunedinPACE followed, each trained differently and each producing a different number from the same sample.
They are genuinely impressive research instruments. GrimAge in particular predicts mortality well, because it was trained on mortality rather than on chronological age — which is a meaningfully different design choice from the earlier clocks.
Their practical limitation is not accuracy. It is that the output is a number with no attached mechanism. A clock telling you that you are 47 does not tell you which system is responsible, which lever to pull, or whether anything you did last year worked.
Telomere length
Telomeres shorten with cell division and were the first popular biological age proxy. Consumer telomere testing is still sold.
The measurement is noisy, varies substantially between cell types and between assays, and individual-level change over a year is frequently smaller than the measurement error. It is worth understanding as biology and is poorly suited to tracking your own progress.
Composite biomarker scores
PhenoAge and Klemera-Doubal are built from ordinary blood markers — albumin, creatinine, glucose, CRP, white cell count and a few others — combined by a published formula. They predict mortality respectably and cost whatever a basic blood panel costs.
This is the family most consumer tools belong to, including this one. The inputs are things you can measure again next quarter, which is the property that makes a number actionable rather than merely interesting.
Which to use for what
If the question is "how fast am I ageing, in a way that would satisfy a researcher", an epigenetic clock is the better instrument and DunedinPACE is the most interesting of them, because it measures rate rather than state.
If the question is "what should I do differently this quarter", a biomarker panel answers it and a clock does not. The inputs are modifiable, the thresholds are published, and the retest interval is weeks rather than years.
They are not competitors so much as different tools. The mistake is paying several hundred dollars for a clock, receiving a number, and discovering there is no next step attached to it.
Why the numbers disagree
Two clocks run on the same blood sample routinely differ by several years, and a clock and a biomarker panel can differ by more. This is not a scandal; they are measuring different things and were trained on different outcomes.
It does mean a single number from any method should be treated as an estimate with a wide interval around it, and that comparing your result from one provider against a friend's from another tells you almost nothing.
Score yourself free — no account needed →Frequently asked questions
What is the difference between an epigenetic clock and this biomarker-based score?
"Biological age" is not one measurement. It is a category containing at least four quite different methods, which disagree with each other, cost wildly different amounts, and are useful for different purposes.
What does this guide cover?
This guide covers: Epigenetic clocks; Telomere length; Composite biomarker scores; Which to use for what; Why the numbers disagree.