BackgroundObesity in childhood is well known to transit adult obesity and to increase risk in metabolic diseases and cancer. Although population based intervention is effective to prevent and manage child obesity, there is few studies target family. We conducted this study to determine the association between the prevalence of obesity in elementary school-aged children and mostly eating meal as a family.MethodsWe cross-sectionally analyzed data for 3,355 children at primary school, aged 5-11 years, from the Korea National Health and Nutrition Examination Survey 2007-2011. We assessed the child obesity (BMI≥95 percentile or ≥25 kg/m2) with eating meals as a family. Analyses were adjusted for the child’s class, maternal obesity, maternal education, parental status, household income, optimal or excess of daily caloric intake. The study population was divided into 4 groups according to the response about mostly eating meals as a family (breakfast/dinner) (YY: both breakfast and dinner group, NY: dinner only group, YN: breakfast only group, NN: neither breakfast nor dinner group)ResultsMultinomial logistic regression analyses were used to test the association between obesity and family meal group. After adjusting for covariates, the odds of obesity associated with family meal groups of NY, YN, NN (compared with YY) were NY: 1.30(95% CI: 0.86-1.96), YN: 2.81(95% CI:1.36-5.80), NN:3.95(95% CI:1.87-8.36)), respectively.Conclusion The primary school children group with meals as a family had a higher prevalence of obesity than those without meals as a family, especially dinner. Meals as a family may be promising targets for obesity prevention efforts in primary school children.