Physical activity and obesity prevention in early childhood

Abstract

Background: Obesity is a worldwide epidemic with adverse health outcomes and finding effective interventions to prevent this disease is crucial. Physical activity (PA) has beneficial health effects and is often included in obesity prevention strategies. However, it has not been established whether it is possible to influence PA during early childhood or if PA is correlated to weight status. Knowledge on patterns and correlates to PA during early childhood is still scarce, especially using longitudinal data.

Aim: The aim of this thesis was to investigate patterns and correlates to child accelerometer measured PA during early childhood including child sex, weight status, motor skill, obesity risk, SES, and parental accelerometer-measured PA. Further, the aim was to investigate if a long-term, multicomponent obesity prevention project could affect child weight status and child PA.

Material and Method: This thesis consists of three studies, all sub-studies of the Early Stockholm Obesity Prevention Project (Early STOPP). Early STOPP was a clustered randomized control obesity prevention trial with a long-term, low-intense, family-based design. The intervention targeted children at high risk of obesity based on parental BMI and in total, 238 children were recruited. Of these, 181 had high obesity risk and were randomized to intervention (n=66) and control (n=115) in addition 57 children with low obesity risk were recruited as a reference group. Data on child and parental accelerometer measured PA was collected yearly from age 2-6 years using an Actigraph GT3X. Weight status and other potential correlates were collected simultaneously and at baseline (age 1). Study I was a cross-sectional study investigating patterns and correlates to child PA at three years of age (n=57). Study II had a prospective design studying patterns of PA over time as well as investigating potential correlates to child PA during early childhood. Study III was a clustered randomized control trial evaluating the effects of a long-term obesity prevention intervention on the main outcome weight status as well as secondary behavior outcome, including PA.

Results: Children were from three-years of age more physically active during weekdays than weekend days and the level of PA varied across the day (Study I and II). On average, child PA increased with 11% per year from age 2 to age 6. No significant differences in PA patterns between boys and girls nor high and low obesity risk were found. The least active children (based on tertiles) did not have higher levels of PA during weekdays than weekend days, which was observed in the middle and mostly active children. (Study I and II) Over time, maternal PA, was correlated to child PA as was time in preschool and season of year. Child sex, weight status, motor skill, obesity risk, SES, and paternal PA was not correlated to PA cross-sectionally nor over time. (Study I and II) Children in the intervention group gained less weight per year compared to children in the control-group but none of the other measured outcomes for weight status differed significantly. There were no differences between groups in total PA nor in weekday PA or weekend PA. (Study III)

Conclusion: Based on results on longitudinal patterns and correlates to child PA, the least active children could benefit from targeted PA interventions in the preschool setting. Also, maternal PA seem to be of importance for child PA and might possibly be of interest for future research. Early STOPP, a long term, low-intensive, family-based, multicomponent obesity prevention intervention, was not successful in its goal to reduce the development of obesity during early childhood. Nor was it able to affect the secondary outcome, PA. Results from this thesis indicates that obesity during early childhood is not affected by PA.