PhenoAge blood-test age

Phenotypic age (PhenoAge) — biological age from nine routine blood markers. Diagnostics & early detection. Reviewed Oct 2026.

A well-validated mortality risk score you can compute from a normal blood panel for free. It predicts; no trial shows that driving the number down changes what happens to you.

Stage

Available now, as a risk score

Rung: Available now. Approved or established. Usable today, with a clinician.

Evidence grade

large validated cohorts, no trial

Randomised human data on surrogate markers, or consistent large cohorts. How grades work.

What it is

A biological-age estimate calculated from chronological age plus nine markers that are already on an ordinary blood panel: albumin, creatinine, glucose, C-reactive protein, lymphocyte percentage, mean cell volume, red cell distribution width, alkaline phosphatase and white cell count. Developed by Morgan Levine and colleagues. It outputs an age in years, which can be older or younger than your real age. The regression coefficients are published in full in the 2018 paper, so the score can be computed from an existing blood result at no cost; it needs no proprietary test.

The evidence

Liu et al., PLoS Medicine 2018 (PMID 30596641) validated PhenoAge in NHANES IV against all-cause and cause-specific mortality, outperforming chronological age and doing so consistently across subpopulations. In a 2024 NHANES analysis (Precis Clin Med, PMID 39309670) each extra year of PhenoAge carried a 4% higher risk of all-cause and of cardiovascular death (HR 1.04, 95% CI 1.04–1.05) independent of chronological age, and people scoring 80–100 on the AHA's Life's Essential 8 had a PhenoAge 3.30 years younger. The underlying methylation version (Levine, Aging 2018, PMID 29676998) is the DNAm PhenoAge clock. The gap, stated plainly: every one of these is observational. PhenoAge predicts risk well; no randomised trial has tested whether managing care to lower someone's PhenoAge lowers their mortality. Mediation percentages (36% of the Life's Essential 8 effect on all-cause death) are statistical decompositions, not demonstrated causal levers. Single markers also mislead individually — creatinine tracks muscle mass and creatine intake, not only kidney function.

What you can do today

This is available today. Whether it suits you depends on your health, your other medicines and your risks, so discuss it with a clinician who knows your history before starting or changing anything.

Sources

Primary sourcepubmed.ncbi.nlm.nih.gov

Change history

9 Oct 2026

Phenotypic age (PhenoAge) added to Diagnostics at grade B, and lithocholic acid added to Aging at grade D.

PhenoAge is the rare case where the cheap measure is the better validated one: nine routine blood markers, validated against mortality in NHANES, free to compute. It still only predicts — no trial has tested whether lowering the number helps. Lithocholic acid goes in at the bottom of the ladder: two Nature papers in animals, no human data, and the molecule doubles as a standard rodent model of cholestatic liver injury, which the popular write-ups leave out.