Health-record score predicts survival better than age alone
A health-record score improved 10-year survival classification for 1 in 13.3 patients compared with an age cutoff of 75 years.
Limit: Health benefits from score-guided care were not measured
Edited by Boris Djordjevic · AgingResearch.News
An observational cohort study analyzed 6,623,541 patients aged 30–90 in the US Veterans Health Administration. Women made up 6.2%; 25.2% had diabetes and 0.5% had HIV. Researchers used clinic visits from 2007–2017 and followed patients through 2021, with a median follow-up of 6.3 years.
The Veterans Ageing Cohort Study–Charlson Comorbidity Index combined age, laboratory results and other health information. Among patients with complete 10-year follow-up, a score cutoff of 42 correctly classified 79.0% as alive or dead, compared with 71.5% using an age cutoff of 75. Improvements were greater among patients with diabetes or HIV. The researchers defined “physiological age” by matching median score-based survival estimates to US population survival estimates at different ages.
Why it matters
Some guideline-recommended care takes 10 years before benefits outweigh harms, making remaining lifespan relevant to decisions as people age. The authors suggest that assessing physiological age could improve healthcare value.
Caveats
This observational study tested survival prediction, not whether score-guided care improved health. Its patients received US Veterans Health Administration care, and 6.2% were women.
The paper
Yale University
EBioMedicine, Jul 16, 2026, Open access
