Biological Age Tests Compared
The major biological-age tests — epigenetic clocks (GrimAge, PhenoAge, DunedinPACE, Horvath, Hannum), glycomics (GlycanAge), telomere length, and commercial composites — compared on one axis: how well each is validated for predicting an aging outcome, not how well it is marketed.
Built and medically reviewed by Charles Kamen, MD, board-certified neurologist ·
No biological-age test is FDA-approved, and none is clinically validated to change a medical decision for an individual. The best-validated epigenetic clocks — GrimAge, PhenoAge, and DunedinPACE— predict mortality, disease, or pace of aging in independent cohorts, which is real and useful science. But that validation is population-level: they predict risk across groups of people, not whether an intervention is working in you. Of the methods compared here, 3 reach our top (outcome-validated) tier; the rest are age-correlation validated or exploratory. A single home-test number is not a diagnosis, a treatment target, or proof that anything is reversing.
Why this page exists: people search “best biological age test” and “TruDiagnostic vs Elysium vs GlycanAge” and get marketing copy. This is the honest version — a physician-built comparison of the underlying methods, graded by validation, so you can tell a well-validated research instrument from a proprietary score with thin outcome data. It pairs with our longevity biomarker ranges (the routine labs that actually drive decisions), our longevity drug evidence index (what the molecules are graded for), and our hallmarks of aging map (why these markers matter).
A word on the tiers here: clocks are not drugs, so we do not use our drug A–D rubric. Instead each method is graded on a 3-tier validation scale— outcome-validated (predicts mortality or pace of aging in cohorts), age-correlation validated (tracks chronological age, weaker outcome prediction), or exploratory/composite/proprietary (thinner aging-specific outcome validation). The tier reflects how well a method is validated as a research instrument, never its accuracy for you personally.
What a biological-age test can and cannot tell you
They predict risk across populations — not whether a therapy is working in you.
A result a few years off chronological age is within normal measurement noise.
No biological-age test is FDA-approved or a validated medical decision tool.
Different tests measure different biology — methylation, glycans, telomeres — and disagree.
Population validation is real science; individual clinical use is not yet established.
A test is one optional input — never the reason to start or stop a therapy.
The validation tiers
Each method is graded by how well it is validated for predicting an aging-relevant outcome— not by how well it is sold.
Shown in multiple independent human cohorts to predict an aging-relevant OUTCOME — all-cause mortality, disease incidence, or rate of aging — beyond chronological age. The strongest validation any biological-age method has today. Still population-level: they predict risk across groups, not whether an intervention is working in you.
Strongly correlates with chronological age and serves as the foundational reference clock. Used heavily in research, but mortality/disease prediction is weaker than the outcome-validated clocks. First-generation.
Measures a correlate of aging (e.g., immune-glycan patterns, telomere length) or combines measures into a proprietary score. Aging-specific outcome validation is weaker, thinner, or not independently reproduced. Useful as a research or screening signal, not a validated clinical tool.
Scope note: these tiers grade research/cohort validation— how robustly a method predicts an aging outcome across human populations. They do not grade clinical accuracy for an individual, diagnostic validity, or FDA status (no method is FDA-approved). A Tier 1 clock is the most validated research instrument available; that is a different question from whether its number should guide your care.
The comparison
| Test / method | Category | Validation | What it measures | Validation evidence | Commercial availability |
|---|---|---|---|---|---|
| GrimAgeBuilt on methylation surrogates of known mortality biomarkers (Lu 2019), which is partly why it predicts death better than age-mirroring clocks. | Epigenetic clock | Tier 1 | DNA methylation surrogate for mortality-related biomarkers (incl. smoking pack-years, GDF-15, inflammatory proteins) + chronological age adjustment | Strongest mortality predictor among published clocks; GrimAge residuals predict time-to-death, cancer, and cardiovascular disease in independent cohorts. | Reported by several consumer epigenetic tests (e.g., TruDiagnostic, Elysium) as one component. |
| PhenoAge (Levine)Trained against a phenotypic-age score built from routine blood markers tied to mortality (Levine 2018). | Epigenetic clock | Tier 1 | DNA methylation version of a clinical biomarker composite (albumin, creatinine, glucose, CRP, lymphocyte %, MCV, RDW, alkaline phosphatase, chronological age) | Predicts all-cause mortality, healthspan, and age-related disease incidence beyond chronological age in cohort studies. | Widely reported under consumer epigenetic tests. |
| DunedinPACEBuilt from longitudinal within-person change, which is why it is the leading candidate for tracking whether something shifts pace of aging (Belsky 2022). | Epigenetic clock | Tier 1 | Rate of aging (pace), not age level — derived from 19 biomarkers tracked longitudinally in the Dunedin birth cohort | Validated to detect 5-year change in pace of aging; predicts later decline. Designed to be responsive to intervention, unlike most clocks. | Core metric for TruDiagnostic and increasingly licensed into other tests. |
| Horvath pan-tissue clockThe original multi-tissue clock and the methodological foundation of the field (Horvath 2013). | Epigenetic clock | Tier 2 | DNA methylation at 353 CpG sites, estimable across most tissues (skin+blood, brain, buccal, etc.) | Strongly tracks chronological age across tissues; predicts mortality modestly but less than second-generation clocks. | Underlies many early consumer tests (e.g., MyDNAge). |
| Hannum clockA co-first blood-based clock published the same year (Hannum 2013). | Epigenetic clock | Tier 2 | DNA methylation at 71 CpG sites, developed on whole blood | Correlates strongly with chronological age in blood; same first-generation era as Horvath, weaker mortality prediction than GrimAge/PhenoAge. | Used in research and some lab panels; less prominent in DTC products. |
| GlycanAgeMeasures immune-glycan aging, a real biological correlate — but a different signal than DNA-methylation age and not independently validated to predict mortality as strongly. | Glycomics | Tier 3 | IgG N-glycosylation patterns in blood (glycome), which shift with age and track inflammation | IgG glycans correlate with age and inflammatory states; aging-specific outcome (mortality/pace) validation is thinner than epigenetic clocks and largely cohort-specific. | Sold direct-to-consumer as GlycanAge. |
| Telomere lengthA genuine hallmark of aging (see our hallmarks map), but population-level telomere length is a noisy individual marker. | Telomere | Tier 3 | Average telomere length (typically from blood leukocytes) | Short leukocyte telomeres associate with mortality and age-related disease, but individual predictive value is weak and measurement variability is high. Aging-relevant but not a precise biological-age clock. | Sold by several labs and DTC tests. |
| Proprietary composites (e.g., Elysium Index)Composite scores can add signal but make external validation harder — the score and its components change with product iteration. | Composite | Tier 3 | Branded index combining multiple epigenetic and/or biochemical signals into a single score | Validation varies by product and is often partly proprietary; independent head-to-head mortality/pace validation against GrimAge/PhenoAge/DunedinPACE is limited. | Sold direct-to-consumer (e.g., Elysium’s biological-age index). |
How to read this table:“Validation evidence” describes what each method has been shown to predict in published cohort studies. GrimAge, PhenoAge, and DunedinPACE are the best-validated because they predict aging outcomes beyond chronological age (Marioni and others showed DNA-methylation age predicts all-cause mortality). GlycanAge and telomere length capture real biology but with weaker aging-outcome validation, which is why they sit in Tier 3. Commercial examples (TruDiagnostic, Elysium, GlycanAge, MyDNAge) illustrate how a method reaches consumers; inclusion is not a recommendation.
Scope & compliance: this comparison grades research/cohort validation only, never clinical accuracy, diagnostic value, or safety for any individual. It contains no treatment claims, no protocols, and no patient outcomes. No biological-age test is FDA-approved; results vary and should never drive a medical decision without a licensed physician. Nothing here is medical advice.
First-generation vs second-generation clocks
The field splits into two waves. First-generation clocks (Horvath pan-tissue, Hannum, both 2013) were trained to mirror chronological age— they are excellent at estimating how old your methylation looks. That makes them foundational, but it also caps their outcome value: a clock built to match calendar age is not optimized to predict who ages fastest or dies earliest.
Second-generation clocks (PhenoAge 2018, GrimAge 2019) were trained instead to predict aging outcomes— phenotypic age, mortality biomarkers, time-to-death. That is why they predict mortality and disease better than the clocks that came before them. DunedinPACE (2022) took a different route: built from longitudinal within-person change in the Dunedin cohort, it measures paceof aging — how fast you are aging, not how old you look — which makes it the leading candidate for detecting whether an intervention shifts the rate of aging.
The practical takeaway: if a biological-age test is going to be informative, it should report at least one second-generation clock (GrimAge or PhenoAge) and ideally a pace measure (DunedinPACE). A test reporting only a first-generation clock, a glycan score, or a telomere length tells you less about aging outcomes — useful context, weaker signal.
How a physician uses these tests
A biological-age test is optional, not essential. The labs that actually drive most longevity decisions are routine and cheap — fasting insulin, HbA1c, lipids, hs-CRP, hormone panels, CBC — the ones on our longevity biomarker rangespage. A methylation clock adds a different, research-grade layer on top of those, and for some motivated patients it is a worthwhile signal to track over time — with repeat testing, consistent methodology, and realistic expectations about noise.
What we do notdo: treat a single high biological-age result as an emergency, or a single drop as proof a therapy is working. If a test is used, it is one input into a clinical picture built from your labs, history, and goals — never the basis for a stack of interventions. Care starts with an $88 evaluation. For the full service overview, see physician-led longevity medicine.
Biological-age testing in Las Vegas
The science does not change by ZIP code — but the sales pressure does. Las Vegas has a dense market of concierge and med-spa longevity programs that sell biological-age testing as a lead-in to expensive “age-reversal” protocols, often before the foundational labs and lifestyle basics are in place. The honest read is the opposite of that funnel: a well-validated clock is a research-grade optional signal, not a reason to start a stack.
Whether you visit our southeast Las Vegas clinic at Eastern Avenue and the 215 from Henderson, Summerlin, or Green Valley — or meet Dr. Kamen by secure telehealth anywhere in Nevada — the conversation is the same: what each test measures, what it can and cannot tell you, and whether it is worth the cost for your goals. Visit our Las Vegas clinic to start.
Selected Research
- Horvath S. DNA methylation age of human tissues and cell types. Genome Biol. 2013;14(10):R115.
- Hannum G, et al. Genome-wide methylation profiles reveal quantitative views of human aging rates. Mol Cell. 2013;49(2):359-367.
- Levine ME, et al. An epigenetic biomarker of aging for lifespan and healthspan. Aging (Albany NY). 2018;10(4):573-591.
- Lu AT, et al. DNA methylation GrimAge strongly predicts lifespan and healthspan. Aging (Albany NY). 2019;11(2):303-327.
- Belsky DW, et al. DunedinPACE, a DNA methylation biomarker to measure the pace of aging. eLife. 2022;11:e73420.
- Marioni RE, et al. DNA methylation age of blood predicts all-cause mortality in later life. Genome Biol. 2015;16:25.
Anchors for the clock/method notes only — first- and second-generation epigenetic clock development and cohort mortality/pace validation. Methodological reference, not a treatment or diagnostic claim. Nothing here is medical advice.
Biological Age Tests FAQ
What is the best biological age test?
There is no single "best" biological age test, because it depends what you are measuring. Among published methods, the best-validated for predicting aging outcomes are the second-generation epigenetic clocks — GrimAge (mortality biomarker surrogates), PhenoAge (phenotypic age), and DunedinPACE (pace of aging). They have been shown to predict mortality, disease, and rate of aging in independent cohorts. But "best validated" is not the same as "clinically useful for you": no biological-age test is FDA-approved, and none is validated to guide a medical decision or confirm an intervention is working in an individual.
Are epigenetic age tests accurate?
Epigenetic clocks are among the most reproducible biological-age methods — they correlate strongly with chronological age and, for the second-generation clocks, with mortality and pace of aging. But "accuracy" has two limits worth knowing. First, a clock gives a statistical estimate with a confidence band, not an exact number — a result a few years off chronological age is within typical noise. Second, a clock can be accurate at predicting risk across a population while being uninformative for any one person. Accuracy as a research instrument is high; accuracy as a personal decision tool is unproven.
What is the difference between PhenoAge, GrimAge, and DunedinPACE?
They measure different things. PhenoAge (Levine 2018) is a DNA-methylation version of a clinical biomarker score built from routine blood tests tied to mortality. GrimAge (Lu 2019) uses methylation surrogates of mortality-linked proteins and smoking, which is why it is the strongest published predictor of time-to-death. DunedinPACE (Belsky 2022) measures pace of aging — how fast you are aging, not how old your methylation looks — and is designed to be responsive to change, making it the leading candidate for tracking whether an intervention shifts the rate of aging.
Can a biological age test tell me if a treatment is working?
Not reliably, and no manufacturer is permitted to claim it can. The one clock engineered to be responsive to change is DunedinPACE (it detects 5-year pace-of-aging differences), and it is the most plausible candidate for tracking change over time — but using a single home test result to judge whether an anti-aging intervention is "working" for you is not clinically validated. A meaningful read requires repeat testing, consistent methodology, and clinical context, and even then it is a research-grade signal rather than a proven outcome measure.
Is GlycanAge as good as an epigenetic clock?
They measure a different biological signal. GlycanAge tracks IgG glycosylation (the glycome), which genuinely shifts with age and inflammation — so it captures an immune-inflammation dimension epigenetic clocks do not. But its aging-specific outcome validation (mortality, pace of aging) is thinner and more cohort-specific than the best-validated epigenetic clocks, which is why we place glycomics in the exploratory/composite tier. It can be a useful complementary inflammation signal; it is not a stronger validated aging predictor than GrimAge, PhenoAge, or DunedinPACE.
Does telomere length tell you your biological age?
Only loosely. Short leukocyte telomeres are associated with mortality and age-related disease, and telomere attrition is a genuine hallmark of aging (see our hallmarks map). But as an individual biological-age marker, telomere length has weak predictive value and high measurement variability, which is why it sits in the exploratory tier rather than with the validated clocks. Telomere testing is better understood as one piece of an aging picture than as a precise biological-age number.
Are biological age tests FDA-approved?
No. As of 2026, no biological-age or epigenetic-age test is FDA-approved to diagnose, treat, or guide care for any condition. They are sold as laboratory-developed tests or consumer wellness products. The validation cited for the stronger clocks is research-grade — cohort studies showing they predict mortality or pace of aging — not the clinical-trial evidence the FDA would require to authorize them as medical decision tools.
Why are biological-age tests marketed so heavily in longevity clinics?
Because they are compelling, quantifiable, and commercially effective — a single "biological age" number is a powerful hook. The strongest tests (GrimAge, PhenoAge, DunedinPACE) are genuinely well-validated research instruments, so the marketing is not baseless. The problem is the leap from "validated at the population level" to "this number should drive your treatment," which the evidence does not support. A thoughtful physician uses these tests as one input into a clinical picture — alongside your labs, history, and goals — never as the basis for a stack of interventions.
Should I get a biological-age test in Las Vegas or by telehealth?
You can, with the right framing. Whether you visit our southeast Las Vegas clinic at Eastern Avenue and the 215 from Henderson, Summerlin, or Green Valley — or meet Dr. Kamen by secure telehealth anywhere in Nevada — the conversation about biological-age testing is the same: which method (if any) is worth the cost for your goals, what it can and cannot tell you, and how to read a result without over-interpreting a noisy number. We treat these tests as optional, research-grade inputs — never as the reason to start or stop a therapy.
Related reading: Longevity biomarker ranges · Longevity Drug Evidence-Grade Index · The 12 Hallmarks of Aging · Senolytics evidence grades
Physician-Led in Las Vegas
Read your biology, not just a number.
A board-certified neurologist will put any biological-age result in context — what it actually measures, what it can and cannot tell you, and what an honest plan looks like alongside your real labs and goals.
Ready to start your longevity care?
Book your $88 evaluation with Dr. Kamen today, online, by phone, or by text.
Book Your $88 EvaluationPrefer to talk? Call (702) 808-2650 or text us to book.
Physician-led peptide therapy in Las Vegas. Visit our clinic