Family-related obesity risk factors and dietary behaviours in high-risk populations : associations with child weight development

Abstract: Background Obesity rates in Swedish children are currently not increasing, however socioeconomic disparities are widening. Many children become obese as early as their preschool years. Hereditary and environmental family-related risk factors are the dominating determinants of child obesity, with parental obesity as the most important. Prevention is a high priority, and increased knowledge on risk factors specifically in high-risk populations is of vital importance for the development of efficient preventive interventions. Aims The primary aim was to analyse the impact of parental adiposity and parental educational level on child relative weight from infancy to adolescence, in high-risk populations. Further, to assess the associations between infant relative weight and early life factors, infant dietary intake, infant eating behaviours and parental food intake. Also, to validate a questionnaire measuring obesity-related eating behaviours in Swedish preschool children. Finally, the aim was to validate the dietary intake in infants and parents using a biomarker for fat intake. Material and methods Four child-parent populations were studied: 231 obese children followed longitudinally, from the Swedish National Childhood Obesity Centre in Stockholm (Study I); 197 (Study II) and 193 (Study IV) one-year old infants in high- and low-risk families (determined by parental weight status), participating in Early STOPP (Stockholm Obesity Prevention Project) and recruited from child healthcare in Stockholm County; 174 children 1-6 years old recruited from kindergartens in Stockholm (Study III). Study I: Associations between severity of obesity at age 7 and 15, age at onset of obesity and parental BMI were analysed in obese children, using data from BORIS, the Swedish quality registry for childhood obesity. Study II: Infant relative weight at 3, 6 and 12 months and rapid weight gain 0-6 months were analysed in relation to parental adiposity, parental educational level and early life factors. Study III: A factorial validation of the Swedish translation of an eating behaviour questionnaire (CEBQ) was performed on children 1-6 years old, and associations with child age, gender, relative weight and parental weight status were explored. Study IV: Infant dietary intake, infant eating behaviours and parental food intake were compared between high- and low-risk families. Associations between child and parental diet quality as well as between child weight, dietary intake and eating behaviours were assessed. The reported intake of fat from selected foods was correlated with the fatty acid composition in plasma. Results Study I: Maternal BMI was associated with degree of obesity at age 7, and both parents’ BMI was associated with degree of obesity at age 15. The relationships were stronger in adolescence. High parental educational level was associated with a lower degree of obesity in adolescence. Parental BMI was not associated with obesity onset. Study II: Child relative weight during the first year was associated with parental education but not with parental adiposity. Birth weight was the most important predictor of growth during the first year. Study III: The factor analysis of CEBQ revealed a seven-factor solution with good reliability. Several eating behaviour factors varied by child age. Study IV: Dietary intake and eating behaviours at age one did not differ between high- and low risk infants, but parents in high-risk families had a higher intake of obesogenic foods. Infant and paternal intake of vegetables and fish were related, but no parent-child associations were found for obesogenic foods. Relative weight at age one was associated with obesity-related eating behaviours but not with dietary intake. Overall, fat from selected foods correlated with the corresponding fatty acids in plasma, indicating satisfactory validity of reported dietary habits. Conclusions Parental adiposity and educational level are both important independent determinants of child weight development and degree of obesity, and can be used to identify children at high risk of obesity. However the timing of their influence differs. Parental adiposity does not affect the relative weight of infants, nor the obesity onset, but the impact of both maternal and paternal BMI becomes stronger with child age and affects the degree of obesity later in childhood and in adolescence. Low parental educational level is associated with higher infant weight as early as their first year, and a protective effect of high educational level may be long-term. Infant relative weight is also independently associated with birth weight and eating behaviours, but not with dietary intake, which is not influenced by parental adiposity or parental obesogenic food intake as early as at age one. This thesis indicates that preventive efforts targeting clearly identified high-risk families should be initiated as early as in infancy.