Identifying Patterns of Tobacco Use and Associated Cardiovascular Disease Risk Through Machine Learning Analysis of Urine Biomarkers.

ABSTRACT:

Background

Tobacco use remains a leading cause of disability-adjusted life years lost in the United States. Cardiovascular harm varies by tobacco product type and usage patterns, yet reliable methods for assessing exposure and harm across different products, especially novel tobacco products, are limited.

Objectives

The authors aimed to identify distinct biomarker exposure patterns associated with different tobacco products using cluster analysis and validate this approach through longitudinal analysis of cardiovascular disease risk.

Methods

Using the Population Assessment of Tobacco and Health data set, we performed cluster analysis and geometric mean modeling of tobacco-related biomarkers, followed by a longitudinal retrospective cohort study with Cox proportional h

SUBMITTER: Siegel NA 

PROVIDER: S-EPMC11904550 | biostudies-literature | 2025 Feb

REPOSITORIES: biostudies-literature