A school and community trial for improving 13-year-olds' fruit and vegetable consumption: the Boost study

1. Change in total fruit, vegetable and fruit & vegetable consumption in grams per day (in school + outside school) at 1st and 2nd follow-up versus baseline and between groups
1.1. The validated Pro Children 24-hour recall questionnaire (pupils' self-reported FV intake on the day prior to the data collection)
1.2. Pre-specified success indicator: We aimed at a 20% increase in FV intake at intervention schools compared to control schools at 1st follow-up based on the effect size obtained in the Pro Children study
2. Change in habitual intake of fruit, vegetables and fruit & vegetables
2.1 The validated Pro Children Food Frequency Questionnaire
2.2. Pre-specified success indicator: Significantly higher proportion of pupils eating fruit and vegetables daily at intervention schools at 1st and 2nd follow-up versus baseline and between groups

The Boost intervention is designed to create significant positive changes (at 1st versus baseline and between groups) in the following determinants of children's fruit and vegetable consumption identified from systematic literature reviews (Rasmussen 2006, Krølner 2011):
1. Environmental determinants:
1.1 Home availability/accessibility: quantity, variety, quality, convenience, visibility, appearance, parental facilitation (Pro Children and Boost items)
1.2 FV availability/accessibility in schools: quantity, variety, quality, convenience, visibility, appearance (Pro Children and Boost items)
1.3 FV availability in sport clubs and youth clubs: quantity, variety, quality, convenience, visibility, appearance (Pro Children and Boost items)

2. Social determinants:
2.1. Parental intake of FV (Pro Children item)
2.2. Parental knowledge of national fruit and vegetable recommendations (Pro Children item)
2.3. Parental attitude to the importance of children eating fruit and vegetables (Pro Children item)
2.4. Parental modelling (Pro Children item)
2.5. Parental barriers e.g. price, time issues (Adapted from CHAMP item)
2.6. Teachers' attitude to the importance of promoting children's FV intake in schools (Boost item)
Teacher modelling (Boost item)
2.7. Sports coaches' attitude to the importance of promoting children's FV intake in sports clubs (Boost item)
2.8. Sports coach modelling (Boost item)
2.9. Situational norms: number of meals, situations, occasions and settings perceived as appropriate for eating fruit and vegetables (Boost items)
2.10 Social peer norms for eating FV in school and leisure time activities (Boost items)

3. Personal determinants:
3.1. Children's knowledge of national fruit and vegetable recommendations (Pro Children item)
3.2. Children's awareness about whether they meet national fruit and vegetable recommendations (Pro Children item)
3.3. Children's taste preferences: number of fruit and vegetables liked (Pro Children item)
3.4. Short term outcome expectancies (Pro Children item)

4. Predefined explorative secondary outcomes:
4.1. Intended positive side-effects
4.2. A significant decrease in proportions of pupils at intervention schools who feel out of energy and have difficulties in concentrating during class in school at 1st follow-up versus baseline and between groups (measured by new Boost items)
4.3. A significant decrease in proportions of pupils at intervention schools
who eat unhealthy food almost daily at 1st follow-up versus baseline and between groups (measured by HBSC food frequency questionnaires)
4.4. Significant positive changes in pupil well-being at intervention schools at 1st follow-up versus baseline and between groups (measured by the HBSC questionnaire)
4.5. Second follow-up sustainability measure: At 2nd follow-up it will be measured whether participation in the Boost study has initiated some environmental changes and capacity building at the intervention schools (e.g. initiation of fruit and vegetable programmes, increased sale of FV in schools, increased focus on school food policies and limited access to unhealthy foods at school) by questionnaires to the school principals 2nd follow-up (based on Pro Children questionnaire and new Boost items).

5. Unintended adverse effects
We have aimed at designing an intervention which does not produce any adverse effects, but we will measure the following unintended adverse effects.
5.1. A significant increase in weight- and eating-related teasing and proportions of pupils being bullied at 1st follow-up versus baseline and between groups (measured by HBSC items and new Boost-items).
5.2 Assessment of whether the free fruit and vegetables in school replace fruit and vegetables eaten at other times of the day: A significant decrease in children eating FV at home (24 h recall questionnaire) and in children bringing FV to school from home (Pro Children items) at 1st follow-up versus baseline and between groups.

As unintended adverse effects are difficult to predict they will also be illuminated through process evaluation.