Serum Cotinine-Defined Tobacco Exposure and Its Association With Stroke, Mortality, and Cognitive Outcomes: Evidence...

To examine the association between serum cotinine-an objective biomarker of recent tobacco smoke exposure from active smoking and secondhand smoke-and stroke, all-cause mortality, and cognitive outcomes among US adults using National Health and Nutrition Examination Survey (NHANES) 2003 to 2018 data.

We conducted a cross-sectional analysis of NHANES 2003 to 2018. Serum cotinine was categorized into quartiles and also evaluated per interquartile range (IQR) increase. All analyses accounted for the complex NHANES sampling design (weights, strata, and primary sampling units). Associations with stroke were examined using survey-weighted multivariable logistic regression. Mortality outcomes from NHANES-linked mortality files were evaluated using survey-weighted Cox proportional hazards models and Kaplan-Meier curves with log-rank tests; these analyses were interpreted as associational. Cognitive outcomes were dichotomized using test-specific 25th-percentile cutoffs and analyzed using survey-weighted logistic regression; ORs represent the odds of preserved cognition (≥25th percentile). Prespecified subgroup analyses (by sex, race/ethnicity, education, and Composite Dietary Antioxidant Index) were conducted, and multiplicative interaction terms were tested.

A total of 8285 participants were included, of whom 592 reported a physician diagnosis of stroke. Per IQR increase in serum cotinine, higher cotinine was associated with higher odds of stroke (OR 1.50, 95% CI 1.19-1.90). Among participants with stroke, higher cotinine was associated with higher all-cause mortality (per IQR increase: HR 1.30, 95% CI 1.08-1.56). Higher cotinine was also associated with lower odds of preserved cognition on Animal Fluency Test and CERAD (OR 0.67, 95% CI 0.62-0.73; OR 0.94, 95% CI 0.89-0.99, respectively). In exploratory subgroup analyses, associations appeared stronger in women and non-Hispanic White participants.

Higher serum cotinine levels were associated with stroke, all-cause mortality among individuals with stroke, and cognitive impairment in NHANES 2003 to 2018. These findings are observational and do not establish causality; prospective studies are warranted to clarify temporality and to evaluate the effects of tobacco exposure reduction or cessation interventions on stroke-related outcomes.