Separating age and physiology can preserve death-risk ranking
The combined score reached 0.843 on a measure of death-risk ranking, exceeding the age-and-sex score by 0.016 in 6,739 validation participants.
Research Square
An observational modelling study analysed 19,829 National Health and Nutrition Examination Survey participants with complete biological measurements. Researchers developed separate scores for age and sex and for body measurements adjusted for age and sex. They developed the model in 13,090 participants from 1999–2010 and tested its fixed scoring rules in 6,739 participants from 2011–2016.
The body-measurement score retained information about death risk and had a low correlation with age. Combining the two scores improved how well the model ranked people by death risk compared with age and sex alone. Comparison with another mortality score, calculated without one of its usual measurements, provided no evidence of a difference in risk ranking. The combined score’s predicted mortality closely matched observed mortality at five and eight years.
Why it matters
Chronological age contributes to many mortality scores, making it harder to distinguish age-related risk from differences in physiology. Separating these contributions could help aging researchers interpret what a mortality score represents.
Caveats
The analysis included only participants with complete measurements, did not account for the survey’s sampling weights, and used later data from the same survey for validation. This preprint has not been peer-reviewed, and validation in an independent population remains necessary.
The paper
Allen GS
Research Square · 6 Oct 2026 · Preprint, not peer-reviewed
