Associations Between Urinary and Blood Heavy Metal Exposure and Heart Failure Risk in Elderly Adults: Insights From an Interpretable Machine Learning Model Based on NHANES (2003-2020)

Abstract

This study examines the link between heavy metal exposure and heart failure risk in individuals aged 50 and older, utilizing machine learning models and data from the NHANES dataset. Five models were evaluated, with Gradient Boosting Decision Trees (GBDT) selected for its accuracy, interpretability, and ability to capture nonlinear relationships. The GBDT model achieved an accuracy of 0.78, sensitivity of 0.93, and an AUC of 0.92. Analysis revealed that higher levels of urinary iodine, blood cadmium, urinary cobalt, urinary tungsten, and urinary arsenic acid were significantly associated with increased heart failure risk. Synergistic effects with age and BMI further amplified these risks. Interpretability tools such as SHAP, per-muted Feature Importance, ICE, and PDP were used to enhance model transparency and understanding. These findings underscore the importance of ongoing research into the mechanisms linking heavy metals and heart failure and the need for monitoring and regulatory measures to protect vulnerable populations.