Obesity's Role in Stroke Incidence Among Smokers: An Assessment of Confounding, Interaction, and Mediation Effects

To examine the association between body mass index (BMI) category and history of stroke among current smokers, while assessing potential confounding, effect modification by age, and mediation by cardiovascular comorbidities.

Data were analyzed from 53,939 current adult smokers participating in the 2021 Behavioral Risk Factor Surveillance System (BRFSS). Survey-weighted logistic regression estimated the association between BMI categories (underweight, normal, overweight, obese) and self-reported stroke history. The study assessed multicollinearity and adjusted for sociodemographic and health variables. Interaction terms were tested to evaluate effect modification by age, and exploratory decomposition analysis was used to assess mediation by conditions such as hypertension and diabetes.

Stroke prevalence was highest among underweight smokers (6.3%). In adjusted models, the association between high BMI and stroke was attenuated or null. A significant interaction between BMI and age was identified; while younger underweight smokers showed lower stroke odds, older obese smokers exhibited odds ratios near or below 1.0 (such as ages 60-64: OR 0.70; 95% CI: 0.57-0.87), consistent with an "obesity paradox." Mediation analysis indicated that traditional cardiovascular risk factors did not mediate the relationship in underweight or overweight groups, with only partial mediation by hypertension and myocardial infarction observed in obese smokers.

Among smokers, BMI is not consistently associated with increased stroke risk, and its impact is strongly modified by age. The findings suggest that the potent vascular toxicity of smoking may overwhelm or alter typical weight-related risk pathways, reinforcing smoking cessation as the primary preventive strategy.