The Cocktail Report (sound really smart around your friends):
A comprehensive October 2025 study evaluated the technical and biological reliability of 18 epigenetic clocks (tools that estimate biological age from chemical modifications on DNA). Nearly all performed well for technical reproducibility. But biological reliability, measured across repeated samples taken hours apart, before and after meals, under stress, and across days, was markedly lower for most clocks.
The key distinction: technical reliability means the same blood sample tested twice gives the same result. Biological reliability means a person's real-world reading stays stable enough to be meaningful. Most clocks fail the second test.
Only PCGrimAge achieved a biological reliability score (ICC above 0.75) considered "good" for clinical use. Most popular consumer clocks, including first- and second-generation clocks trained to predict chronological age, showed only moderate or poor biological reliability.
A separate Yale School of Medicine study found that some common epigenetic clocks can give readings that differ by up to nine years on two tests from the same blood sample. Adding a mathematical technique called principal component analysis reduced that gap to under one year.
DunedinPACE, one of the more reliable newer-generation clocks trained on mortality and pace of aging rather than chronological age, showed better stability and detected lifestyle-driven changes in as little as 12 months in published clinical trials.
If you have spent money on a biological age test, here is what the science actually says: the test almost certainly worked as advertised, measuring the chemical modifications on your DNA accurately. But the number it gave you may have been meaningfully different from the number you would have gotten a week later.
That is a measurement problem researchers are still solving, not a marketing scandal. A 2025 study from the TranslAGE platform evaluated 18 epigenetic clocks and found that biological reliability, how stable a reading is across real-world conditions, was the critical unresolved gap.
Your measured biological age can shift based on what you ate before the test, whether you were under stress, or the time of day the sample was drawn. Clocks trained on chronological age were the most susceptible to this noise, while clocks trained on mortality risk and pace of aging, such as GrimAge and DunedinPACE, were more stable and meaningful.
For you personally, this matters because many consumer tests use older-generation clocks. If your result says you are biologically five years older than your calendar age, there is a real possibility that the reading reflects measurement noise rather than signal.
To be candid, this does not mean epigenetic clocks are useless. In population research, they have consistently identified what accelerates or slows biological aging, including smoking, poor sleep, chronic stress, and low physical activity.
The problem is translating a population-level tool to individual health decisions, which requires a level of precision most current clocks have not yet achieved. The researchers' own practical guidance: use a newer-generation clock with documented biological reliability, test under consistent conditions, and wait at least three months between tests.
Why Should You Care?
Most biological age test marketing implies a single result tells you something definitive about your health trajectory, but the precision required for individual clinical decisions is still being built. Knowing which clocks to trust, and how to interpret results, is the difference between useful data and expensive noise.
Sources:
Higgins-Chen AT, et al. Biological versus Technical Reliability of Epigenetic Clocks and Implications for Clinical Translation. Aging Cell. 2025 Oct 14. PMID: 41279914. https://pubmed.ncbi.nlm.nih.gov/41279914/
Higgins-Chen AT, et al. A computational solution for bolstering reliability of epigenetic clocks. Nature Aging. 2022. https://medicine.yale.edu/news-article/a-computational-solution-for-bolstering-reliability-of-epigenetic-clocks-implications-for-clinical-trials-and-longitudinal-tracking/
Belsky DW, et al. DunedinPACE, a DNA methylation biomarker of the pace of aging. eLife. 2022. https://pmc.ncbi.nlm.nih.gov/articles/PMC8853656/
US News / The Conversation. Can Biological Age Tests Help Patients Get a Handle on Their Health? May 4, 2026. https://www.usnews.com/news/health-news/articles/2026-05-04/can-biological-age-tests-help-patients-get-a-handle-on-their-health
