Implementation of strategies to increase adolescents’ access to fruit and vegetables at school: process evaluation findings from the Boost study

BACKGROUND Access to fruit and vegetables (FV) is associated with adolescents' FV consumption. However, little is known about implementation of strategies to increase access to FV at schools. We examined the implementation of two environmental components designed to increase access to FV at Danish schools. METHODS We used data from 20 intervention schools involved in the school-based multicomponent Boost trial targeting 13-year-olds' FV consumption. The environmental components at school included daily provision of free FV and promotion of a pleasant eating environment. Questionnaire data was collected by the end of the nine-month intervention period among 1,121 pupils (95%), from all school principals (n = 20) and half way through the intervention period and by the end of the intervention among 114 teachers (44%). The implementation of the components was examined descriptively using the following process evaluation measures; fidelity, dose delivered, dose received and reach. Schools with stable high implementation levels over time were characterised by context, intervention appreciation and implementation of other components. RESULTS For all process evaluation measures, the level of implementation varied by schools, classes and over time. Dose received: 45% of pupils (school range: 13-72%, class range: 7-77%) ate the provided FV daily; 68% of pupils (school range: 40-93%, class range: 24-100%) reported that time was allocated to eating FV in class. Reach: The intake of FV provided did not differ by SEP nor gender, but more girls and low SEP pupils enjoyed eating FV together. Dose delivered: The proportion of teachers offering FV at a daily basis decreased over time, while the proportion of teachers cutting up FV increased over time. Schools in which high proportions of teachers offered FV daily throughout the intervention period were characterized by being: small; having a low proportion of low SEP pupils; having a school food policy; high teacher- and pupil intervention appreciation; having fewer teachers who cut up FV; and having high implementation of educational components. CONCLUSIONS The appliance of different approaches and levels of analyses to describe data provided comprehension and knowledge of the implementation process. This knowledge is crucial for the interpretation of intervention effect. TRIAL REGISTRATION Current Controlled Trials ISRCTN11666034.