Regression models to estimate the total concentration of polycyclic aromatic hydrocarbon metabolites in urine

Polycyclic aromatic hydrocarbons (PAH) are potentially mutagenic and carcinogenic and as such their exposure is of serious concern. Regression models to estimate the total concentration of polycyclic aromatic hydrocarbon metabolites (∑PAH) in urine were developed. Using data from National Health and Nutrition Examination Survey for the years 2003-2008 and 2011-2012, regression models were developed to estimate ∑PAH in urine. The performance of the fitted models were computed by comparing percent predicted values located between 0 and 5%, 5-10%, 15-20% and >20% of the observed values. While data for 2003-2008 were used to fit the regression models, the data for 2011-2012 were solely used to evaluate the performance of the fitted models. R(2) of the fitted model was 94.7%. About 46% of the predicted values for the 2003-2008 data and over 48% of the predicted values for the 2011-2012 data were found to be between 0 and 10% of the observed values. In order to use these models, in addition to age, gender, and smoking status, and optionally race/ethnicity, the knowledge of the levels of only 1-hydroxynaphthalene, 2-hydroxynaphthalene, and 9-hydroxyfluorene is necessary. Models that optionally require the knowledge of the levels of urine creatinine were also developed.