In the ENERGY-project - in line with how European Commission-funded projects are organized - the work to go through the different planning steps is conducted in eleven work packages (WPs) (see Figure ).
Division of work and timeline of the ENERGY-project.
Within all WPs the same state-of-the-art methodology will be used: all WPs have a comparable structure according to which the timely finalization of all deliverables will be accomplished. We will conduct systematic literature reviews; secondary data analyses of existing data sets; focus group research; a cross-European school-based survey; and a school-based-preliminary trial to evaluate the newly developed intervention scheme. Finally, the results will be disseminated among key stakeholders, including researchers, policy makers and the general population.
WP 1 comprises the coordination task to promote and ensure integrated and timely progress of the ENERGY-project. This WP is responsible for the overall running and implementation of the project, will carry out the administrative tasks, and is responsible for the financial and organizational management of the project. This WP is also responsible for setting up the structure for communication through regular meetings and a website.
WP 2 identifies the most important EBRB among school children aged 10-12 years to contribute to step 2 of the planning model depicted in Figure by reviewing the literature and conducting secondary data analyses of existing datasets.
WP 3 reviews the literature and conducts secondary data analyses to identify the most important personal, family environmental and school environmental correlates, predictors and determinants of the EBRB identified in WP 2 in order to contribute to step 3 of the planning model. Because most of the evidence on potential determinants of EBRB to date is based on cross-sectional data, WP2 and 3 will primarily focus on reviewing and re-analyzing longitudinal studies.
will identify successful intervention schemes, mediating pathways of intervention strategies, and important factors that moderate the effect of intervention strategies. These explorations are primarily based on secondary analyses of existing successful schemes promoting EBRB among schoolchildren and adolescents. WP 4 specifically addresses parental involvement in school-based schemes, by exploring what factors may promote parental involvement and what the impact of parental involvement on EBRB change in their children is. Earlier studies have indicated that parental factors and involvement is of key importance for school-aged children's EBRB and for effectiveness of primarily school-based interventions [20
], but determinants of parental involvement in interventions have hardly been studied before. Therefore, this WP includes using qualitative methods, which are especially useful for exploring new research issues, as well as quantitative analyses of existing data. WP 5 explores what successful strategies or successful factors of ongoing intervention schemes consist of by means of a literature review and secondary data analyses. The review and secondary analyses will especially focus on mediation and moderation analyses to gain insight in the underlying processes of the intervention, in what determinants need to be addressed in the new intervention, and to explore what strategies work best according to, for example, gender, socioeconomic position or ethnicity. WP 6 explores economic strategies/incentives and whether these strategies can be effective and useful in school-based studies. Apart from a literature review, WP 6 makes an inventory of ongoing initiatives regarding economic incentives and further explores what are the possibilities to implement such strategies in the school setting by means of a survey among school staff and/or school board members. Economic strategies or incentives have hardly been used nor studied in school-based intervention schemes to date. In addition, WP 6 updates recent insights in the importance of marketing to children as a driving force behind the obesity epidemic especially related to family and school environments.
WP 7 comprises a cross-European survey on personal, family environmental and school environmental correlates of the different EBRB identified in WP 2. Europe has a relatively wide variety in social and physical environmental contexts, dietary practices, physical activity levels as well as rates of overweight and obesity, and the cross-European survey of WP 7 therefore offers the right framework for further exploration of potential determinants of EBRB and geographical differences in such determinants. The survey is conducted in seven countries (i.e. Belgium, Greece, Hungary, The Netherlands, Norway, Slovenia, Spain), located in different regions of Europe. After the ENERGY-project was approved, an eighth country - Switzerland - has joined the cross sectional survey WP of ENERGY with its own funding. This survey accumulates data on prevalence of overweight - based on objective measures of height, weight and waist-circumference -, on the most important EBRB identified in WP2, and on the most important correlates of these EBRB identified in WP 3, as outlined and detailed in Figure . This survey results in up-to-date data on EBRB and their personal and environmental correlates in different parts of Europe - including countries for which such data is as yet scarce or unavailable. It will provide intervention entry points for modification of specific EBRB and their determinants for WP 8.
WP 8 will carefully design the intervention using the IM protocol by using, integrating and translating information from WPs 2 -7 into specific change objectives, selection of effective strategies and development of intervention materials and protocols.
WP 9 is devoted to the pilot implementation and preliminary evaluation of the newly developed intervention in five different countries, again located in different regions of Europe: Norway, Germany, Belgium, Greece, and Hungary. The intervention will be implemented at five schools in each country and five other schools will serve as controls. The evaluation study will focus especially on process variables and intermediary outcomes.
WP 10 will use information from WP 9 to prepare the final intervention package, and an implementation, adoption, dissemination, and impact monitoring plan. WP 10 is also responsible for the further exploitation of the ENERGY-intervention scheme.
Finally, WP 11 files and administrates all data collected in the different WPs and oversees and guards quality-controlled data management and storage.
Different qualitative and quantitative methods are used in the WPs in ENERGY, including systematic reviews, focus group interviews, secondary analyses of existing data sets, as well as original survey and evaluation research.
Reviews and secondary data analyses
The systematic reviews within WPs 2-6 are conducted according to a standard protocol. Basically, after a search strategy is developed and run, relevant papers are identified by subsequent selections of article titles, abstracts, and full papers, by two independent reviewers. Relevant data from the included studies is extracted and their methodological quality is scored, again independently by two reviewers. If possible and appropriate, the data of the retrieved studies are combined to conduct a formal meta-analysis.
Within WPs 2-5 analyses of existing datasets are conducted. Relevant datasets owned or managed by the partners in ENERGY will be used as well as additional potentially relevant datasets identified through the literature reviews. The secondary data analyses focus on mediation and moderation analyses of longitudinal data.
Although a number of school-based EBRB interventions have shown some effects, only very few evaluation studies have looked into how these effects came about, i.e. what the mediators of intervention effects were. Within the ENERGY-project the secondary data analyses will exactly do that: assess the mediators of intervention success or failure in order to gain insight in the underlying reasons why interventions may be effective or not. For these mediation analyses we will apply the procedure suggested by MacKinnon [21
]. Furthermore, we will test moderators of intervention effects by including possible moderator X intervention interaction terms in the analyses to explore if interventions have differential effects according to, for example, gender, socio-economic position or ethnicity.
The study has been approved by the Medical Ethics Committee of Durham University.
To explore what reasons parents have to be involved or not in their children's school-based health promotion interventions, a minimum of four focus group interviews will be conducted in each of four countries (Belgium, Norway, Spain, Hungary). Separate focus groups will be organized for one random group of parents and three separate groups with parents who (1) often participate in parent activities of school-based interventions; (2) parents with low socio-economic positions; and (3) parents with a low interest in EBRB. The focus groups will be conducted according to a standard protocol in all four countries. The interviews will be sound-recorded, and transcribed and submitted to content analyses.
The cross-European survey in WP7 will take place in schools in seven countries in the spring of 2010. We aim to collect data among approximately 7000 children to further study EBRB and their potential determinants according to region, ethnicity, gender and socio-economic status. The 1000 children per country will be recruited in a school-based manner. First three provinces in each country will be randomly selected. Within each province three municipalities with > 20,000 inhabitants will then be randomly selected. Lists of schools in these municipalities are obtained and schools will be randomly selected to be invited to participate. All children in the right school-classes for the study will be included in each school. WP 7 will administer survey questionnaires among schoolchildren, among their parents and school staff, and conduct audits of the school environment, and anthropometric measurements of the participating children. Children's height, weight and waist circumferences will be measured by trained staff, and accelerometers will be used to objectively assess physical activity in a sub-sample of about 600 children in three of the participating countries (Figure ). All measurements will be performed according to written protocols and thus standardized between countries and schools. The research assistants responsible for the fieldwork receive a joint training.
The survey instruments used within WP7 will be developed based on the input of WPs 2-6, applying previously validated instruments where possible and appropriate. The survey questionnaires will be pilot-tested, evaluated for test-re-test reliability and internal consistency, and the pre-tested version will be translated to the languages of the other participating countries, and then back-translated for quality control.
The school-environment audit instrument will be adopted from the ENDORSE study [2
Recruitment of schools, children and parents is conducted according to a standardized stepwise approach ensuring comparable samples in all countries, aiming for inclusion of lower and higher socio-economic position groups.
All data collected, i.e. completed surveys and audit instruments, reports on measured height, weight and waist circumference, and accelerometer records, will be sent to the coordinating center for data entry and data cleaning.
The evaluation study in the five countries will focus on comprehensibility, readability, relevance, credibility, and attractiveness of the intervention materials, next to preliminary evaluation of effects on EBRB and potential behavioral determinants. The effectiveness of the intervention is evaluated in a pre-test-post-test design, including an intervention and a control condition. This preliminary evaluation is conducted in a school-randomized controlled design. The intervention will be implemented in five schools, with two classes per school, in each of countries, and the results are compared to a five control schools in each country. Schools will be randomly allocated to intervention or control.
We aim for inclusion of approximately 2500 children in the preliminary evaluation study. Power calculations based on previous studies of school-based interventions indicate that this sample size is sufficient to detect relevant changes in EBRB, i.e. a 5% difference in change between intervention and control group, and their determinants [22
Evaluation will be based on surveys among the participating children and their parents before and after the intervention has been implemented and on questionnaires completed by the school boards and the teachers in the intervention schools. The survey questionnaire and measurement protocols will be similar to those used in WP 7, but enriched with process evaluation questions.