Teun Remmers

36 | Chapter 3 be more constrained by external factors (e.g. homework) in their engagement in PA, irrespective of their PA enjoyment. Third, one study showed that an elevated Body Mass Index (BMI) may lead to weight-criticism and lower PA levels as a result of avoidant coping (18). In this way, BMI may attenuate the association between PA enjoyment and PA behavior as heavier children may experience limitations to perform PA by their overweight. Finally, several conceptual frameworks postulate that the relationship between the social environmental factors and behavior can also be moderated by personality traits (11,19). A personality trait that might be of importance in this case is the degree of automaticity as this moderates a person’s reaction to the environment, and thus plays a considerable role in the behavior of adults and children (20,21). In addition, high tendencies towards automaticity/spontaneity (i.e. impulsivity ) have already been established as a key moderating factor of influences on child snacking behavior (22,23). PA in children is often characterized as spontaneous (24). Therefore, impulsivity may attenuate the association between PA enjoyment and PA behavior, as in impulsive children PA may occur rather automatically without involvement of cognition. We addressed the following research question: what is the association between PA enjoyment and objectively measured PA behavior and are 1) gender, 2) age, 3) BMI, and 4) impulsivity of the child moderators in this relationship? Methods Study design and participants This study was embedded in the KOALA Birth Cohort Study. The cohort was recruited among healthy pregnant women in the general population via obstetric practices (conventional recruitment group, n=2343); and partly in healthy pregnant women recruited among alternative lifestyle channels, such as organic food shops and anthroposophic doctors’ practices (alternative recruitment group, n=491) (25,26). From the total KOALA-cohort, 584 children were eligible and selected by the availability of previous accelerometer measurements between ages 4-5 years (27) and/or 6-7 years at the occasion of home visits for lung function testing (26). At child’s age of 8-9 years old, we measured PA behavior by accelerometry, and PA enjoyment by questionnaire. For these measurements, we excluded children with asthma diagnosis ( n =33), growth hormone medication use ( n =1), loss-to-follow up ( n =11), and residing ≥ 30 kilometers from the nearest study location ( n =69). The flow of participants is presented in Figure 1. Finally, 171 participants provided valid accelerometry and data PA enjoyment at 8-9 years old. All parents gave written informed consent. The present study was approved by the Medical Ethics Committee of Maastricht University Medical Center+.