Relationship of cotinine-verified and self-reported smoking status with metabolic syndrome in 116,094 Korean adults

Abstract

No study has reported the relationship between cotinine-verified and self-reported smoking status with metabolic syndrome (MetS). This study was performed to evaluate the relationship between urinary cotinine-verified and self-reported smoking status with MetS and determine the effects of unobserved smokers on MetS in Korean adults. A total of 116,094 individuals (66,875 men and 49,219 women) with mean age of 36.7±6.8years included in Kangbuk Samsung Health Study and Kangbuk Samsung Cohort Study between 2011 and 2013 who had urinary cotinine measurements were enrolled. Cotinine-verified current smoking was defined as urinary cotinine level of above 50ng/mL. Unobserved smoking was defined as urinary cotinine level of above 50ng/mL in self-reported never smokers. The overall prevalence rates of cotinine-verified current smokers and MetS were 22.9% and 10.5%, respectively. The misclassification rate to cotinine-verified current smokers among self-reported never smokers was 1.7%. A multivariate logistic regression model adjusted for variables with univariate relationship (model 1) showed that cotinine-verified current smokers significantly increased the odds ratio for MetS compared with cotinine-verified never smokers (odds ratio [95% confidence interval], 1.30 [1.23, 1.37]). Log-transformed cotinine levels were also associated with MetS (1.04 [1.03, 1.05]). However, the association was not significant in the previously mentioned model including the traditional 5 components of MetS (model 2). Unobserved smokers significantly increased the ORs for MetS in both model 1 (1.43 [1.23, 1.67]) and model 2 (1.57 [1.06, 2.33]). This study shows that unobserved smoking and cotinine-verified current smoking are associated with MetS but urinary cotinine could be 1 conditional factor that interacts with traditional MetS components.