Identifying Patterns of Tobacco Use and Associated Cardiovascular Disease Risk Through Machine Learning Analysis of Urine Biomarkers
Main finding · Composite outcome of heart failure, myocardial infarction, or strokeResult — go to the resultHazard ratio 2.24 95% CI 1.17–4.30Authors' conclusion
Our categorization of exposure through cluster analysis provides a potential tool for evaluating the use of emerging tobacco products and establishing a connection between novel exposures and cardiovascular risk. This approach may contribute to the validation of a valuable tool for assessing the risk associated with the use of different tobacco products.
Study context
- Population
- Individuals in the Population Assessment of Tobacco and Health data set
- Participants
- 6,463 individuals
- Exposure
- Associations between clusters and a primary composite outcome of heart failure, myocardial infarction, or stroke
- Outcome
- composite outcome of:
- Heart failure
- Myocardial infarction
- Stroke
Methodological detailscluster analysis, geometric mean modeling, Cox proportional hazard modeling
- Cluster analysis
- Geometric mean modeling
- Cox proportional hazard modeling
Main finding
As reported in the abstract
Hazard ratio
2.24
95% CI 1.17–4.30
Cluster 2 vs cluster 1
HR 2.24 (95% CI 1.17–4.30)
Observational data: associations, not effects.
Further findings
- All clusters showed elevated biomarkers of oxidative stress and inflammation compared to cluster 1, with clusters 2 and 3 showing the highest levels
Original publication
Identifying Patterns of Tobacco Use and Associated Cardiovascular Disease Risk Through Machine Learning Analysis of Urine Biomarkers. JACC Adv. Published 22 February 2025. doi:10.
Summarised from the published abstract; every number is checked against it