Among 10 sociodemographic and lifestyle variables, smoking is strongly associated with biomarkers of acrylamide exposure in a representative sample of the U.S. Population

Abstract

Hemoglobin adducts of acrylamide (HbAA) and glycidamide (HbGA) have been measured as biomarkers of acrylamide exposure and metabolism in a nationally representative sample of the U.S. population in the NHANES 2003-2004. We assessed the association of sociodemographic (age, sex, race-ethnicity, education, and income) and lifestyle (smoking, alcohol consumption, BMI, physical activity, and dietary supplement use) variables with these biomarkers in U.S. adults (aged ≥ 20 y). We used bivariate and multiple regression models and assessed the magnitude of an estimated change in biomarker concentration with change in a covariable for 2 biomarkers of acrylamide exposure. Smoking was strongly and significantly correlated with HbAA and HbGA concentrations (rs = 0.51 and 0.42, respectively), with biomarker concentrations being 126 and 101% higher in smokers compared with nonsmokers after adjusting for sociodemographic and lifestyle covariates. Age was moderately and significantly correlated with both biomarkers (rs = -0.21 and -0.22, respectively). BMI (rs = -0.11) and alcohol consumption (rs = 0.13) were weakly yet significantly correlated with HbAA concentrations only. The estimated percentage change in biomarker concentration was ≤ 20% for all variables other than smoking after adjusting for sociodemographic and lifestyle covariates. Using multiple regression models, the sociodemographic variables explained 9 and 7% whereas the sociodemographic and lifestyle variables together explained 46 and 25% of the variability in HbAA and HbGA, respectively, showing the importance of considering and adequately controlling for these variables in future studies. Our findings will be useful in the design and analysis of future studies that assess and evaluate exposure to acrylamide and its metabolism to glycidamide.

Figure 1

Estimated change in hemoglobin adduct of acrylamide and glycidamide with change in sociodemographic and lifestyle variables for adults ≥20 y, NHANES 2003–2004 Model 1 represents simple linear regression, model 2 represents multiple linear regression with sociodemographic factors, and model 3 represents multiple linear regression with sociodemographic and lifestyle factors; changes derived from a linear regression model while holding any other variables in the model constant; asterisk accompanying the percent change indicates significance (P <0.05). NHB, non-Hispanic black; NHW, non-Hispanic white; HS, high school; PIR, poverty income ratio.