The evidence for the impact of policy on physical activity outcomes within the school setting: A systematic review

Background: Despite the well-established health benefits of physical activity (PA) for young people (aged 4-19 years), most do not meet PA guidelines. Policies that support PA in schools may be promising, but their impact on PA behavior is poorly understood. The aim of this systematic review was to ascertain the level and type of evidence reported in the international scientific literature for policies within the school setting that contribute directly or indirectly to increasing PA. Methods: This systematic review is compliant with Preferred Reporting Items for Systematic Review and Meta-Analysis guidelines. Six data-bases were searched using key concepts of policy, school, evaluation, and PA. Following title and abstract screening of 2323 studies, 25 progressed to data synthesis. Methodological quality was assessed using standardized tools, and the strength of the evidence of policy impact was described based on pre-determined codes: positive, negative, inconclusive, or untested statistically. Results: Evidence emerged for 9 policy areas that had a direct or indirect effect on PA within the school setting. These were whole school PA policy, physical education, sport/extracurricular PA, classroom-based PA, active breaks/recess, physical environment, shared use agreements, active school transport, and surveillance. The bulk of the evidence was significantly positive (54%), 27% was inconclusive, 9% was significantly negative, and 11% was untested (due to rounding, some numbers add to 99% or 101%). Frequency of evidence was highest in the primary setting (41%), 34% in the secondary setting, and 24% in primary/secondary combined school settings. By policy area, frequency of evidence was highest for sport/extracurricular PA (35%), 17% for physical education, and 12% for whole school PA policy, with evidence for shared use agreements between schools and local communities rarely reported (2%). Comparing relative strength of evidence, the evidence for shared use agreements, though sparse, was 100% positive, while 60% of the evidence for whole school PA policy, 59% of the evidence for sport/extracurricular PA, 57% of the evidence for physical education, 50% of the evidence for PA in classroom, and 50% of the evidence for active breaks/recess were positive. Conclusion: The current evidence base supports the effectiveness of PA policy actions within the school setting but cautions against a "one-sizefits-all" approach and emphasizes the need to examine policy implementation to maximize translation into practice. Greater clarity regarding terminology, measurement, and methods for evaluation of policy interventions is needed.