Fig. 1.

Adjusted odds ratios* for the association between total urinary arsenic (TUA1) in quartiles and total anti-HAV seropositive prevalence by self-reported HAV immunization: (a) received ⩾2 doses, (b) received <2 doses, (c) 0 dose and (d) ‘don't know’. [* Odds ratios from a logistic regression model that included the interaction between arsenic exposure by quartiles (TUA1) and hepatitis A immunization adjusted for log-transformed creatinine, age, sex, race, family income/poverty ratio, country of birth, body mass index and survey year.]