Association between arsenic exposure and behavior among first-graders from Torreón, Mexico

Abstract

Previous studies suggest adverse effects of arsenic exposure on children's cognitive function. In this study, we examined the potential association between arsenic exposure and children's behavior. Five hundred and twenty-six children, 6-7 years old, living near a metal foundry in Torreón, Mexico, participated in the study. Arsenic exposure was measured as total urinary arsenic (UAs) and arsenic metabolites-monomethylarsonic acid (MMA) and dimethylarsinic acid (DMA) concentrations. Children's behavior was assessed by Conners Behavior Rating Scales for parents and teachers. The median (interquartile range) concentrations of UAs, MMA and DMA were 55.2 (39.7), 6.7 (5.9) and 39.3 (28.5) μg/L, respectively. The mean behavior scores from parent and teacher ratings were within the clinically normal range (T<65). The relationship between behavior and urinary arsenic was modeled in linear and logistic regression models, with UAs, MMA and DMA tested in separate models and entered as quartiles. No significant association was found between any measure of urinary arsenic and parent ratings of behavior. However, higher UAs was modestly associated with higher scores on the Oppositional, Cognitive Problems and ADHD sub-scales of the teacher ratings; a dose-response relationship was not established between UAs quartiles and behavior. Higher urinary DMA was associated with higher ratings on the Oppositional, Cognitive Problems and ADHD Index by teachers. The associations between UAs and behavior became statistically non-significant after adjustment for the Peabody Picture Vocabulary Test scores, suggesting that the harmful effects of arsenic on behavior may be secondary to arsenic-induced cognitive deficits. These data suggest a potential adverse association between arsenic and children's behavior and indicate a need to further study the effects of arsenic and arsenic metabolites on neurobehavioral outcomes in children.