Although school has been an important intervention venue for obesity prevention, the role of school-level factors in obesity development or prevention has not been well-documented. This study aimed to systematically examine the current evidence on school-level factors associated with obesity outcomes in longitudinal studies. The literature search was performed in PubMed, EMBASE, CINHAL, and PsycINFO. Peer-reviewed articles using longitudinal study designs and published in English from 1991 to 2018 were eligible. Twelve articles met eligibility criteria for final systematic review. Nine studies reported significant long-term associations between school-level factors and obesity outcomes. Higher parental education, longer minutes of recess, meeting recommended recess and physical education time, higher socio-economic status, suburban compared with rural area, higher parental involvement in school, and healthful school food environment were significantly associated with lower rates of obesity or obesity trajectory. However, due to the small number of studies and heterogeneity of measures and variables used in their analytic models, the overall level of evidence from this review suggests the importance of further, systematic study. Empirically rigorous research is needed to identify additional aspects of the school context and environment that may contribute to the risk of obesity throughout the life course. KEYWORDS childhood obesity, longitudinal study, schools, systematic review 1 | INTRODUCTIONThe school setting has been recognized as an important intervention venue for obesity prevention because children and adolescents spend a significant amount of time in schools and share the surrounding environment that might influence their energy balance-related behaviours, including diet and physical activity. 1-4 Overall, schoolbased obesity prevention programmes have been found to be most effective when they are theory-based, behaviourally focused, longterm (1-4 years), include multicomponents such as parent involvement, 3-6 and when the school environment and policy modifications are incorporated within the intervention programmes. 3,4,7,8 Systematic reviews and meta-analyses of school-based obesity prevention programmes have indicated promising results on dietary and/or physical activity behaviour changes, yet limited success on adiposity reduction has been reported, with pooled effect sizes ranging from 0.072 to 0.17. 1,3,5,9 Possible reasons for the limited success rates
