Examining the associations among adults' physical activity, serum cotinine, and the risk of metabolic syndrome.

INTRODUCTION/

OBJECTIVE

This study aims to assess the associations among tobacco exposure, confirmed through serum cotinine (SC) levels, physical activity (PA), and the likelihood of developing metabolic syndrome (MetS).

METHODS

This study involved 8,423 adults from the National Health and Nutrition Examination Survey (NHANES) spanning years 2007 to 2018. Logistic regression analyses were conducted to estimate the associations among SC, PA and the risk of MetS with its components. We also investigated the sex differences on these outcome variables via analysis of variance (ANOVA) and Chi-square test.

RESULTS

The covariate-adjusted logistic regression analysis results suggest that SC concentration (unit: ng/mL) was significantly and positively associated with MetS (OR: 6.41, 95%CI: 5.81-7.08), abdominal obesity (OR: 1.92, 95%CI: 1.75-2.11), hypertension (OR: 2.70, 95%CI: 2.46-2.96), hyperglycemia (OR: 12.50, 95%CI: 10.63-14.71), hypertriglyceridemia (OR: 2.10, 95%CI: 1.90-2.33), and low serum high-density lipoprotein cholesterol (HDL-C) (OR: 1.61, 95%CI: 1.46-1.77). PA energy consumption (unit: MET*h/wk) was significantly and inversely associated with MetS (OR: 0.56, 95%CI: 0.51-0.61), abdominal obesity (OR: 0.71, 95%CI: 0.65-0.78), hypertension (OR: 0.55, 95%CI: 0.50-0.60), hyperglycemia (OR: 0.59, 95%CI: 0.52-0.67), and hypertriglyceridemia (OR: 0.84, 95%CI: 0.76-0.93). For males, an increase of approximately 6 MET*h/wk in PA energy consumption was negatively associated with the higher odds of MetS linked with a 1 ng/mL higher SC level. Similarly, for females, the corresponding estimate was 7 MET*h/wk.

CONCLUSIONS

This study indicates a positive relationship between SC concentration and MetS risk, whereas higher PA energy expenditure was inversely associated with MetS risk. Moreover, the findings indicate a potential compensatory association between PA and SC levels in relation to MetS risk. These estimates are based on statistical modeling and should be viewed as illustrative approximations rather than definitive causal or prescriptive thresholds. Future longitudinal and interventional studies are warranted to confirm these relationships.