Quantile regression analyses of associated factors for body mass index in Korean adolescents

Abstract

OBJECTIVES: This study examined the influence of home and school environments, and individual health-risk behaviours on body weight outcomes in Korean adolescents.

STUDY DESIGN: This was a cross-sectional observational study.

METHODS: Quantile regression models to explore heterogeneity in the association of specific factors with body mass index (BMI) over the entire conditional BMI distribution was used. A nationally representative web-based survey for youths was used.

RESULTS: Paternal education level of college or more education was associated with lower BMI for girls, whereas college or more education of mothers was associated with higher BMI for boys; for both, the magnitude of association became larger at the upper quantiles of the conditional BMI distribution. Girls with good family economic status were more likely to have higher BMIs than those with average family economic status, particularly at the upper quantile of the conditional BMI distribution. Attending a co-ed school was associated with lower BMI for both genders with a larger association at the upper quantiles. Substantial screen time for TV watching, video games, or internet surfing was associated with a higher BMI with a larger association at the upper quantiles for both girls and boys. Dental prevention was negatively associated with BMI, whereas suicide consideration was positively associated with BMIs of both genders with a larger association at a higher quantile.

CONCLUSIONS: These findings suggest that interventions aimed at behavioural changes and positive parental roles are needed to effectively address high adolescent BMI.