Abstract
Both low family socioeconomic status (SES) and low neighborhood SES have been associated with higher levels of childhood obesity. However, little is known about how these two factors operate together. The purpose of this study was to determine if the association between neighborhood SES and obesity varies across household SES. We used the first-grade round of the Early Childhood Longitudinal Study, Kindergarten Class of 2011 (ECLS-K:2011). Household SES was defined based on income, education, and occupation. Neighborhood SES was defined by the percent of households living in poverty in the child's school district. Log-binomial regression models estimated the association between neighborhood SES and obesity and tested whether this association varied by household SES. We found the association between neighborhood SES and obesity varied significantly by household SES (p-interaction = 0.002). For children in the lowest tertile of neighborhood SES, prevalence of obesity was not statistically significantly different comparing children with low, middle or high household SES (Predicted probability (PP)(lowest) 0.20 (95% CI: 0.17, 0.23), PP(middle) 0.21 (95%CI: 0.18, 0.24), PP(highest) 0.16 (95%CI: 0.12, 0.20)). Conversely, within the highest and the middle tertiles of neighborhood SES, children with high household SES have significantly lower prevalence of obesity compared to children with the lowest household SES (PP: 0.09 (95%CI: 0.07, 0.11) vs 0.19 (0.16, 0.21) and (PP: 0.07 (95%CI: 0.05, 0.09) vs 0.17 (0.13, 0.21) for highest vs lowest household SES in middle and high neighborhood SES, respectively). Hence, low-SES in either variable is enough to be associated with increased prevalence of obesity.