To analyse the association between family structure and adiposity in children.

Cross-sectional and longitudinal analysis of the IDEFICS (Identification and prevention of dietary- and lifestyle-induced health effects in children and infants) study cohort.

Primary schools and kindergartens.

Children (n 12 350; aged 7·9 (sd 1·8) years) for the cross-sectional analysis and children (n 5236; at baseline: normal weight, aged 5·9 (sd 1·8) years) for the longitudinal study underwent anthropometry. Family structure was analysed as (i) number and type of cohabiting adults and (ii) number of siblings.

In the cross-sectional analysis, after controlling for covariates, children living with grandparents had significantly higher BMI Z-score than those living with both parents (0·63; 95 % CI 0·33, 0·92 v. 0·19; 95 % CI 0·17, 0·22; P < 0·01); in addition, the higher the number of siblings, the lower the BMI Z-score (only child = 0·31; 95 % CI 0·24, 0·38; 1 sibling = 0·19; 95 % CI 0·16, 0·23; 2 siblings = 0·15; 95 % CI 0·09, 0·20; >2 siblings = 0·07, 95 % CI 0·04, 0·19; P < 0·001). Over the 2-year follow-up, differences in weight gain were observed across family-structure categories. Further, the risk of incidence of overweight/obesity was significantly lower the higher the number of siblings living in the household (v. only child: 1 sibling = 0·74, 95 % CI 0·57, 0·96; 2 siblings = 0·63, 95 % CI 0·45, 0·88; >2 siblings = 0·40, 95 % CI 0·21, 0·77), independently of confounders.

The study suggests that an independent association between family structure and childhood obesity exists.