Global Co-Occurrence of COPD and Peripheral Artery Disease in Older Adults: A Machine Learning Analysis of Spatial Patterns and Population-Attributable Risks.

BACKGROUND

Chronic obstructive pulmonary disease (COPD) and peripheral artery disease (PAD) significantly contribute to morbidity and mortality among older adults. The geographic co-occurrence may indicate shared exposures; however, global population-level evidence remains scarce.

METHODS

Utilizing Global Burden of Disease (GBD) 2021 estimates, the analysis focused on COPD and PAD incidence in adults aged ≥ 70 years across 202 countries and territories. Disease-specific incidence quartiles characterized spatial co-occurrence. Random forest models with SHAP (Shapley Additive Explanations) ranked exposure contributions, while negative binomial regression with population offsets estimated associations and population-attributable fractions (PAFs). A quartile-based composite risk index summarized cumulative exposure.

RESULTS

The global median incidence was 1150.16 per 100,000 for COPD and 695.4 per 100,000 for PAD. High-incidence regions exhibited partial geographic overlap. Smoking, high LDL cholesterol, and kidney dysfunction correlated with the incidence of both diseases, whereas several additional exposures were disease-specific. Elevated LDL cholesterol had the largest positive association with COPD incidence (RR 1.149), whereas high sugar-sweetened beverage intake exhibited the largest positive association among PAD-specific exposures (RR 1.059). Model-derived cumulative PAFs were 29.7% for COPD and 23.7% for PAD.

CONCLUSION

COPD and PAD demonstrate distinct yet overlapping global spatial patterns among older adults, sharing several population-level exposure correlates. Due to the cross-sectional and ecological nature of this study, the observed associations and PAFs should not be interpreted as individual-level causal effects. These findings may inform regional public health planning and generate hypotheses for prospective validation of integrated cardiopulmonary risk assessment.