BackgroundIn contrast to curriculum-based health education interventions in schools, the school environment approach promotes health by modifying schools' physical/social environment. This systematic review reports on the health effects of the school environment and processes by which these might occur. It includes theories, intervention outcome and process evaluations, quantitative studies and qualitative studies.Research questionsResearch question (RQ)1: What theories are used to inform school environment interventions or explain school-level health influences? What testable hypotheses are suggested? RQ2: What are the effects on student health/inequalities of school environment interventions addressing organisation/management; teaching/pastoral care/discipline; and the physical environment? What are the costs? RQ3: How feasible/acceptable and context dependent are such interventions? RQ4: What are the effects on student health/inequalities of school-level measures of organisation/management; teaching/pastoral care/discipline; and the physical environment? RQ5: Through what processes might such influences occur?Data sourcesA total of 16 databases were searched between 30 July 2010 and 23 September 2010 to identify relevant studies, including the British Educational Index, the Cumulative Index to Nursing and Allied Health Literature, the Health Management Information Consortium, EMBASE, MEDLINE and PsycINFO. In addition, references of included studies were checked and authors contacted.Review methodsIn stage 1, we mapped references concerning how the school environment affects health and consulted stakeholders to identify stage 2 priorities. In stage 2, we undertook five reviews corresponding to our RQs.ResultsStage 1: A total of 82,775 references were retrieved and 1144 were descriptively mapped. Stage 2: A total of 24 theories were identified (RQ1). The human functioning and school organisation, social capital and social development theories were judged most useful. Ten outcome evaluations were included (RQ2). Four US randomised controlled trials (RCTs) and one UK quasi-experimental study examined interventions building school community/relationships. Studies reported benefits for some, but not all outcomes (e.g. aggression, conflict resolution, emotional health). Two US RCTs assessed interventions empowering students to contribute to modifying food/physical activity environments, reporting benefits for physical activity but not for diet. Three UK quasi-experimental evaluations examined playground improvements, reporting mixed findings, with benefits being greater for younger children and longer break times. Six process evaluations (RQ3) reported positively. One study suggested that implementation was facilitated when this built on existing ethos and when senior staff were supportive. We reviewed 42 multilevel studies, confining narrative synthesis to 10 that appropriately adjusted for confounders. Four UK/US reports suggested that schools with higher value-added attainment/attendance had lower rates of substance use and fighting. Three reports from different countries examined school policies on smoking/alcohol, with mixed results. One US study found that schools with more unobservable/unsupervised places reported increased substance use. Another US study reported that school size, age structure and staffing ratio did not correlate with student drinking. Twenty-one qualitative reports from different countries (RQ5) suggested that disengagement, lack of safety and lack of participation in decisions may predispose students to engage in health risks.LimitationsWe found no evidence regarding health inequalities or cost, and could not undertake meta-analysis.ConclusionsThere is non-definitive evidence for the feasibility and effectiveness of school environment interventions involving community/relationship building, empowering student participation in modifying schools' food/physical activity environments, and playground improvements. Multilevel studies suggest that schools that add value educationally may promote student health. Qualitative studies suggest pathways underlying these effects. This evidence lends broad support to theories of social development, social capital and human functioning and school organisation. Further trials to examine the effects of school environment modifications on student health are recommended.FundingThe National Institute for Health Research Public Health Research programme.