Background: To promote healthy dietary and physical activity behaviour among primary school children, the city of Amsterdam structurally implements the school-based Jump-in intervention in over half of its primary schools. Previously shown to be effective in stimulating physical activity and outside recess play, our study is the first to evaluate Jump-in's effect on children's dietary behaviour. Evaluating the effectiveness and implementation process of an intervention in a real-life setting requests an alternative study design. Methods: we chose a mixed-methods, quasi-experimental Extended Selection Cohorts design to evaluate Jump-in's effectiveness and implementation process. Children and parents from the first ten primary schools that enrolled in the programme in 2016-2017 were invited to participate. The primary outcomes were children's dietary behaviour and behavioural determinants, assessed by child and parent questionnaires, and photographs of the food and drinks children brought to school. Process indicators, contextual factors and satisfaction with the programme were assessed by interviews with health promotion professionals, school principals, school project coordinators, and teachers; focus group discussions with parents and children; and document analysis. Discussion: Conducting research in a real-life setting is accompanied by methodological challenges. Using an Extended Selection Cohorts design provides a valuable alternative when a Randomized Controlled design is not feasible.
Keywords: Extended Selection Cohorts design; children; dietary behaviour; effect evaluation; mixed methods; process evaluation; school-based intervention; study protocol.