Does phenotypic age from routine blood tests predict mortality?
Answered November 2, 202610 sources
EvidenceAStrongGood randomized trials in people that agree, with no serious weaknesses. New studies are unlikely to change the answer.
Yes. People whose phenotypic age is older than their actual age die sooner, and in large studies it predicted death somewhat better than age alone.
- Who it's for
- Adults wanting a risk marker; not shown to guide treatment for one person.
- What to expect
- Each extra year of acceleration goes with roughly 4% to 24% higher death risk.
- Watch out for
- It shows group-level links, so it can't forecast one person's lifespan.
Each dot is a risk ratio for death with a line for its 95% range; 1 means no difference in risk.
- Cancer survivors, all deaths2.07 (1.69 to 2.54)Clear worsening
- Cancer survivors, cancer deaths2.15 (1.52 to 3.04)Clear worsening
- Cancer survivors, non-cancer deaths2.06 (1.66 to 2.57)Clear worsening
- US adults, all deaths5.52 (4.26 to 7.15)Clear worsening
| Result | Estimate | 95% range | Reading |
|---|---|---|---|
| Cancer survivors, all deaths | 2.07 | 1.69 to 2.54 | Clear worsening |
| Cancer survivors, cancer deaths | 2.15 | 1.52 to 3.04 | Clear worsening |
| Cancer survivors, non-cancer deaths | 2.06 | 1.66 to 2.57 | Clear worsening |
| US adults, all deaths | 5.52 | 4.26 to 7.15 | Clear worsening |
Sources 3 BMC Cancer, 2025; 9 Cardiovasc Diabetol Endocrinol Rep, 2026
Figure 3. Relationship between Phenotypic Age, chronological age, and PhenoAgeAccel.
Figure 3 from A new aging measure captures morbidity and mortality risk across diverse…, PLoS Med 2018. Licence: CC0 1.0
Figure 7. Receiver operating characteristic curves for 10-year mortality.
Figure 7 from A new aging measure captures morbidity and mortality risk across diverse…, PLoS Med 2018. Licence: CC0 1.0
What the studies show
- A study that followed 11,432 US adults for 12.6 years found phenotypic age was linked to death from any cause, after accounting for age and sex. The link held in seemingly healthy people and in adults 85 and older2.
- Two Chinese cohorts (83,571 and 21,229 people) found each standard-deviation rise in phenotypic age acceleration went with a 24% higher risk of death. Prediction was good, with scores of 0.81 and 0.87 on a scale where 1.0 is perfect8.
- A UK Biobank study of 296,931 people over 12.8 years found an improved score built from 17 blood markers beat phenotypic age slightly (0.749 against 0.743). Both beat age alone (0.706)4. Phenotypic age adds only a modest gain over age.
- In 2,643 US cancer survivors followed a median 9.2 years, accelerated phenotypic age was linked to about double the risk of death. Each one-unit rise went with 4% higher risk3.
- In 609 people with multivessel heart disease followed a median 33.5 months, each 10-year rise in phenotypic age was linked to 51% higher mortality risk10.
- In 332,012 UK adults over 13.6 years, those with the fastest aging had a 54% higher risk of death than those with the slowest6.
These figures compare groups. The gain or loss for one person depends on how high their risk was to begin with.
Key studies
| Study | Who | What it found |
|---|---|---|
| Study following people, 20182 | 11,432 US adults | Phenotypic age linked to death after adjusting for age |
| Study following people, 20244 | 296,931 UK adults | Predicted death better than age alone |
| Study following people, 20258 | 83,571 Chinese adults | 24% higher death risk per step of acceleration |
| Study following people, 20253 | 2,643 cancer survivors | About double the death risk with acceleration |
What we don't know
- Whether using the score in clinics changes decisions or improves health. Studies show prediction, not benefit from acting on it.
- Whether lowering phenotypic age extends life. One trial of calorie restriction suggested slower biological aging, but it did not measure lifespan1.
Try it yourself: How old do your labs look?
Ask a follow-up
Written by AI from the studies listed, without review by a clinician, and graded with the GRADE method. It can be wrong, so check the studies it cites. Education, not medical advice.