Individual-, family-, and contextual-level variables do not explain the protective effect of parental nativity status on changes in 3-15-year-old children's BMI.

Individual-, family-, and contextual-level factors can simultaneously and interactively affect a child's body mass index (BMI). We examine parental nativity as a key determinant of changes in children's BMI over time. Prior research on this topic has been inconclusive. A longitudinal sample of households with children residing in four low-income, high minority New Jersey cities provided data on demographics, socioeconomic status, anthropometric measures, as well as dietary and physical activity behaviors for one randomly selected child. The baseline interview for two separate cohorts took place in 2009/10 and 2014-15, with a follow-up interview 2-5 years later. The outcome variable, change in BMI z-score was divided into three categories (decrease in BMI z-score; no meaningful change; increase in BMI z-score) and analyzed using ordinal logistic regressions. About 28% of the children in the sample had at least one foreign-born parent. For the two major racial/ethnic groups, i.e., Hispanics and non-Hispanic blacks, having a foreign-born parent was associated with a favorable change in BMI-children of foreign-born parents were more likely to experience a decrease BMI z-score between baseline and follow-up. Multivariate analyses reveal that the initial association between parental nativity and children's BMI change (OR = 0.20; p < 0.001) persists after controlling for an extensive set of covariates, such as child dietary and physical activity behaviors, family-level variables, census tract characteristics, and measures of food environment (OR = 0.17; p < 0.001). Through a series of sensitivity analyses, we verified that our results are consistent across different model specifications. In our sample, having a foreign-born parent was a protective factor for children's BMI change that operates through different pathways than might be anticipated.