What is it about?
Cardiovascular-kidney-metabolic (CKM) syndrome is a condition where heart, kidney, and metabolic problems overlap. It progresses very differently from person to person, so it is hard to know who is at highest risk. Existing biological age measures were not designed for CKM syndrome. We used machine learning and data from 6,896 US adults in NHANES (2005–2018) to build a new CKM-specific aging index, called CKMAI. It combines 18 routine blood tests, physical measures, lifestyle factors, and social factors. CKMAI predicted death from any cause, death from cardiovascular disease, and high-risk CKM status better than three existing indices. It also showed that depression partly explains the link between accelerated aging and worse outcomes. We validated CKMAI across different time periods and in a small hospital cohort in China. We developed a free online calculator and opensource R package so doctors and researchers can use CKMAI easily. CKMAI may help identify people who need earlier or more intensive care, but more validation in diverse populations is needed.
Featured Image
Photo by Madison Agardi on Unsplash
Why is it important?
CKM syndrome is a major and growing health burden, but risk tools often treat patients as a single group. CKMAI is the first aging index tailored to the CKM pathway and built with machine learning. It uses only routinely available clinical, nutritional, lifestyle, and social data—no costly omics tests. It outperformed universal biological age indices for mortality and high-risk CKM status, showed a super-additive interaction with PhenoAge, identified six distinct aging-metabolic phenotypes, and quantified depression as a partial mediator. By providing a free online calculator and open-source R package, CKMAI is immediately implementable in primary care and resource-limited settings. It supports the American Heart Association’s call for practical CKM risk stratification and precision prevention.
Perspectives
As authors, we were motivated by the many patients with similar CKM stages who follow very different trajectories. We wanted to capture the biological wear and tear along the CKM continuum, not just chronological age or single risk factors. Seeing CKMAI outperform established indices was encouraging, but the most surprising findings were the synergy with PhenoAge and the measurable role of depression. We hope this work encourages clinicians to consider mental health and social factors alongside metabolic and kidney markers, and to use CKMAI as a conversation tool with patients. We also hope researchers will test whether CKMAI-guided care improves outcomes in prospective studies and randomized trials.
Dong Wang
The First Affiliated Hospital of Anhui University of Chinese Medicine
Read the Original
This page is a summary of: A machine learning-derived aging index for risk stratification and mortality prediction in cardiovascular-kidney-metabolic syndrome: A retrospective cohort study, PLoS Medicine, October 2026, PLOS,
DOI: 10.1371/journal.pmed.1005078.
You can read the full text:
Contributors
The following have contributed to this page
