RESUMEN
BACKGROUND: Early environmental influences have been linked to child weight status, however further understanding of associations in diverse populations is needed. METHODS: A cross-sectional analysis of household and family factors associated with overweight was conducted on a culturally diverse, urban dwelling sample of 820 first through third graders (mean age 7.6 ± 1.0 years) residing in three eastern Massachusetts cities. Overweight was defined as BMI > 85th percentile, based on measured height and weight, and the CDC growth reference. Multivariate logistic regression was used to identify demographic, behavioral, and social environmental variables significantly related to weight status. Independent variables included race-ethnicity, age, sex, servings of sugar-sweetened beverages/week, hours of screen time/week, parent overweight, (calculated from self-reported weight/height), parent education, household food restriction rules regarding snacking and/or kitchen access, frequency of having dinner as a family (reported as "a lot" vs. "sometimes/rarely/never") and child vitamin/mineral supplement use. Selected interactions were explored based on prior studies. RESULTS: Prevalence of overweight was 35.5% in girls and 40.8% in boys. In the final, adjusted model, compared to white children, the odds of overweight were higher in children of Hispanic race-ethnicity (odds ratio (OR) = 2.4, 95% CI = 1.4 - 4.1). In the same adjusted model, compared to children with no household food restriction rules, the odds of overweight were 2.6 (95% CI = 1.3-5.1) times higher and 3.5 (95% CI = 1.9-6.4) times higher for children having one rule or two rules, respectively. Parent report of frequent family dinner and child vitamin use were protective, with a halving of risk for overweight for each behavior (OR = 0.47, 95% CI = 0.31-0.71 and OR = 0.54, 95% CI = 0.37-0.78, respectively). CONCLUSIONS: In the presence of other factors, frequent family dinner and vitamin use were associated with lower risk of overweight and household food restriction rules with higher risk. Although such relationships have previously been reported, this investigation is among the first to demonstrate these associations in a low-income, racially-diverse early elementary school population, and suggest potential targets of opportunity within the family context that could reduce child overweight risk in a subgroup of children at elevated risk of obesity.