BACKGROUND: Schools can play an important role in promoting health. However, many education policies and institutions are increasingly emphasising academic attainment targets, which appear to be diminishing the time available for health education lessons. Interventions that integrate both health and academic learning may present an ideal solution, simultaneously addressing health education and academic development. The theories of change underlying these interventions are therefore of interest, but are poorly studied. METHODS: A systematic review of evaluations of interventions that integrate academic and health education for reduced substance use and/or violence was carried out. As part of this, reports describing theory were assessed for quality and data extracted. Theoretical data were synthesised within and across individual interventions using reciprocal translation and meta-ethnographic line of argument synthesis to produce an overall theory of change for interventions that integrate health and academic education to prevent substance use and violence. RESULTS: Forty-eight reports provided theoretical descriptions of 18 interventions. An overarching theory that emerged was that eroding 'boundaries' at multiple and mutually reinforcing levels-by integrating academic and health education, by transforming relationships between teachers and students, by generalising learning from classrooms to the wider school environment and by ensuring consistent messages from schools and families-is intended to lead to the development of a community of engaged students oriented towards pro-social behaviour and away from substance use, violence and other risk behaviours. CONCLUSIONS: Eroding 'boundaries' between health and academic education, teachers and students, classrooms and the wider school and schools and families were seen to be the most critical to establishing new frameworks of family, classroom or school organisation that are conducive to promoting both academic and social-emotional outcomes. Whether such interventions are feasible to implement and effective in reducing risk behaviours will be examined in other reports arising from the review.