Family structure in relation to body mass index and metabolic score in European children and adolescents

BACKGROUND

Living in single parent and blended families or as an only child-compared to living in two-parent biological families or with siblings, respectively-is associated with a higher body mass index (BMI) in cross-sectional studies. However, longitudinal research addressing the children's BMI in this context is scarce. Further, little is known about the association between family structure and metabolic health.

OBJECTIVES

This study aimed at investigating the association between both aspects of family structure with BMI and a metabolic score (MetS).

METHODS

Cross-sectional data from 7804 children participating in the European multi-center I.Family study (2013/2014) and longitudinal data from 5621 children who also participated previously in the IDEFICS study (2007-2010) were used. Family structure was assessed by a detailed interview. BMI z-score and the MetS were based on measured anthropometry, blood pressure, high-density lipoprotein, blood glucose, and triglycerides. Linear regressions were performed to model associations between family structure with BMI and MetS.

RESULTS

Children from single-parent families had higher BMI z-scores in the cross-sectional (β = 0.09, 95% confidence interval [CI]: 0.001 to 0.18) and longitudinal analyses compared to those from two-parent families. Cross-sectionally, the number of siblings was associated with lower BMI z-scores (β = -0.07, 95% CI: -0.10 to -0.03) and lower MetS (β = -0.14, 95% CI: -0.26 to -0.01). Longitudinally, only children between baseline and follow-up had higher BMI z-scores at follow-up (β = 0.07, 95% CI: 0.01 to 0.14) compared to stable siblings.

CONCLUSION

Obesity prevention measures should focus on single-parent households and families with an only child.

Is there stress eating in children and adolescents? A first validation of the Dutch version of the Salzburg Stress Eating Scale

The recently developed Salzburg Stress Eating Scale (SSES) is the first to specifically address the relationship of stress experiences and coping failure on the one hand and increases or decreases in food intake on the other hand. The SSES demonstrated good psychometric properties in English and German speaking adult samples. However, it has not been applied to younger age groups, which may differ in their expression and reporting of stress related eating behavior. Therefore, the current study evaluated the psychometric properties and correlates of the SSES in a Dutch speaking community sample of adolescents (study 1; N = 244, 9-18 years, adjusted BMI range = 71-189%, M = 103%) and in an inpatient sample of adolescents with obesity (study 2; N = 100, 9-16 years, adjusted BMI range = 151-310%, M = 194.5%). Factor analysis confirmed the original one factor structure in both studies. Internal consistency was 0.85 in study 1 and 0.90 in study 2 (Cronbach's alpha). In both the obesity and community sample, SSES scores were positively correlated with other eating-related measures (e.g., emotional eating, external eating), and uncorrelated with non-eating-related wellbeing measures (e.g., perceived stress, depressive symptoms). Unexpectedly, however, no association with adjusted BMI was found in the community sample, obesity sample, or combined sample. The scores of the obesity sample were also compared to the community sample in an age and sex-matched subsample of study 1. Against our hypothesis, the obesity sample did not report higher SSES scores than the community sample. The good internal consistency and the evidence for convergent and discriminant validity indicate that the Dutch version of the SSES seems useful in examining this clinically important eating behavior in adolescents. However, validity with regard to stress eating and BMI remains to be further explored.