Family and Lifestyle Factors Mediate the Relationship between Socioeconomic Status and Fat Mass in Children and Adolescents

摘要

Socioeconomic status (SES) is strongly associated with childhood overweight. The underlying mechanism and the role of family and lifestyle factors as potential mediators of this relationship remain, however, unclear. Cross-sectional data of 4,772 girls and boys aged 5–16 years from the Kiel Obesity Prevention Study were considered in mediation analyses. Fat mass (FM) was assessed by bioelectrical impedance analysis and converted into a percent FM SD score (FM%-SDS). SES was defined by the parental educational level, classified as low, middle, or high. Characteristics of family and lifestyle factors were obtained via validated questionnaires and considered as mediators. In 3 different age groups, the product-of-coefficients method was used to examine age-specific mediator effects on the relationship between SES and FM%-SDS (c = total effects) and their ratio to total effects, adjusted for age, sex, puberty, and nationality. The prevalence of overweight was 6.9%. In all age groups, SES was inversely associated with FM%-SDS as follows: 5–7 years, c₁ = −0.11 (95% CI −0.19 to −0.03); 9–11 years, c₂ = −0.21 (95% CI −0.27 to −0.14); and 13–16 years, c₃ = −0.23 (95% CI −0.28 to −0.17). The relationship between SES and FM%-SDS was fully (5–7 and 9–11 years) and partly (13–16 years) mediated by similar and age-specific mediators, including parental BMI, parental smoking habits, media consumption, physical activity, and shared meals. Overall, these variables resulted in a total mediating effect of 77.8% (5–7 years), 82.4% (9–11 years), and 70.6% (13–16 years). Consistent for both sexes, the relationship between SES and FM%-SDS was therefore mediated by parental weight status, risk-related behavior within families, and children's and adolescents' lifestyle factors. Strategies for obesity prevention, which are predominantly targeted at socially disadvantaged groups, should therefore address the family environment and lifestyle factors.