Association of arsenic exposure with smoking, alcohol, and caffeine consumption was investigated. Data from National Health and Nutrition Examination Survey for the years 2005–2010 were used for this investigation. Urinary levels of total arsenic (UAS) and dimethylarsonic acid (UDMA) were evaluated for children aged 6–12 years and adolescents and adults aged ≥12 years. Urinary levels of arsenobetaine (UAB) were evaluated for adolescents and adults only. Regression models were fitted for log transformed values of UAB, UAS, and UDMA. For the models for children, however, gender, race/ethnicity, SES, and fish/shell fish consumption during the last 30 days were the only independent variables that were included in the models. Nonsmokers were found to have higher levels of UAS and UDMA than smokers. Elevated levels of UAB, UAS, and UDMA were associated with higher amounts of daily alcohol consumption. The associations were in the opposite direction for daily caffeine consumption. Females were found to have statistically significantly lower adjusted levels of UDMA than males for those aged ≥12 years. Irrespective of age, those with unclassified race/ethnicity had the highest levels of UAB, UAS, and UDMA and non-Hispanic whites had the lowest levels. Adolescents had the higher levels of UAB, UAS, and UDMA than adults. Higher SES was associated with higher levels of UAB, UAS, and UDMA among adolescents and adults. Irrespective of age, fish consumption was associated with higher levels of UAB, UAS, and UDMA.