Cardiology · Nephrology · Endocrinology · 3 h ago
CKM-specific aging index predicts mortality in retrospective NHANES cohort
A retrospective study of 6,896 NHANES adults developed an aging index for cardiovascular-kidney-metabolic risk assessment that outperformed three comparator indices. Mortality discrimination was high, but validation in geographically independent cohorts with complete mortality follow-up remains lacking.
- Retrospective development cohort included 6,896 NHANES adults.
- Five-year all-cause mortality AUC was 0.907.
- CKMAI outperformed three comparator indices across reported outcomes.
- Independent geographic validation of mortality prediction remains lacking.
Researchers developed a cardiovascular-kidney-metabolic aging index (CKMAI) in a retrospective cohort of 6,896 adults from NHANES 2005–2018, weighted to represent the US adult population. A machine-learning framework evaluated more than 100 candidate survival models and selected a simple Cox model. Outcomes were all-cause mortality, cardiovascular mortality, and high-risk CKM status.
The selected model achieved a mean C-index of 0.893. At five years, CKMAI’s area under the curve was 0.907 (95% CI 0.884–0.927) for all-cause mortality and 0.937 (95% CI 0.892–0.960) for cardiovascular mortality. For identifying high-risk CKM status, the AUC was 0.790 (95% CI 0.765–0.816). CKMAI outperformed PhenoAge, KDM, and the cardiometabolic index across the reported outcomes; reclassification and decision-curve analyses also favored CKMAI.
The index combines laboratory measures, nutritional assessments, lifestyle factors, and social determinants of health, suggesting potential utility for risk stratification. Temporal validation within NHANES and preliminary validation in a Chinese hospital cohort of 261 participants supported identification of high-risk CKM status. However, independent geographic validation with complete mortality follow-up is lacking, and this observational study does not establish that using the index improves clinical outcomes. The authors reported public and academic funding and no competing interests.
Is this summary clinically accurate?
Help fellow clinicians: your rating sends inaccurate summaries straight to our editors.
Sign in to rate this summary →Source
PLOS Medicine: A machine learning-derived aging index for risk stratification and mortality prediction in cardiovascular-kidney-metabolic syndrome: A retrospective cohort study ↗This is an automated AI-condensed summary that has not yet been reviewed by an editor. Always consult the full item at the original source.