Children living in low socio-economic status (SES) communities are more likely to have unhealthy behaviours that include physical inactivity and unhealthy dietary habits, compared to those living in high-SES communities [1
]. These unhealthy behaviours can lead to health problems such as overweight/obesity, type 2 diabetes, cardiovascular diseases, and mental health problems, which can develop even at a young age [3
]. Promoting healthy behaviours at an early age may help to improve children’s health, as well as their educational achievements; both may lead to improved health in later life [4
]. Schools can play an important role in this, since they reach all children and form a strong social network of teachers and children who can influence one another, and since a significant proportion of a child’s day is spent at school [5
]. Moreover, school and home are both part of a child’s mesosystem: changes in the school may also influence the home environment, which could enhance the effects of school health promotion [6
Despite the school’s potential to help improve children’s health, school health promotion in the Netherlands at the beginning of the 21st century was marked by relatively low priority and lack of coordination and fragmentation, and was often supply-driven [8
]. Traditionally, the Dutch primary school system (for children aged 4–12 years) has been mainly driven by educational requirements determined at the national level by the Ministry of Education, Culture and Science, which focus on maths, language, reading, and world orientation [9
]. Due to these educational requirements, the interest in, and priority given to, the implementation of health promoting (HP) changes at schools is limited, as these changes do not directly address the educational requirements [8
]. As a result, HP changes are only coincidentally implemented and often lack systematic coherence and sustainability, as the changes are not embedded in the school system. Whereas educational requirements have been defined at the national level, the Ministry of Health, Welfare and Sports has delegated the responsibility for health promotion to the local authorities, leading to further fragmentation. Local health organizations have to compete for the attention of schools to make them realize the importance of health promotion and its implementation. This leads to a supply-driven approach, which can irritate schools due to the abundance of initiatives offered to them that do not really match their needs [11
]. On top of all these issues, the school itself is a complex system, characterized by a large number of interacting institutional elements [12
]. This means that it depends on the specific school context how suitable a change towards health promotion is [13
]. It also means that each implemented change will have different effects at each school; there is always an interaction between the change and the school context [13
This complexity of the school system, together with the contextual differences between them, the fragmentation of school health promotion, and the worrying increase in unhealthy behaviours among school children, have induced the local educational board ‘Movare’, the Regional Public Health Services (RPHS), and Maastricht University (UM), all situated in the southern part of the province of Limburg, to take action. Movare’s primary schools are in the former mining area of Parkstad, in which eight municipalities around the city of Heerlen collaborate (211 square kilometres; 250,000 inhabitants). This region is characterized by a low SES, compared to the mean for the Netherlands [16
]. Health and educational issues, such as a high prevalence of overweight/obesity and high school dropout rates compared to the national average, are a persistent problem in this region, and continue to exist from generation to generation [18
]. Therefore, the three organizations (Movare, RPHS, UM) developed the ‘Healthy Primary School of the Future’ (HPSF): a Dutch initiative which aims to sustainably integrate health and well-being within the school system [18
]. HPSF intends to go beyond traditional temporary and superficial top-down solutions and to establish a co-creation movement in schools towards systematic incorporation of health and well-being. This incorporation ideally leads to sustained changes that become embedded in the DNA of the school. In other words, HPSF aims to add-in
, instead of adding-on
, health and well-being to the school system. The initiative builds upon the principles of the health-promoting school (HPS) framework, which aims to create a healthy school environment using a school system approach. HPS focuses not only on classroom-based health education, but also on changes in school policy and the schools’ physical and social environment [22
], using bottom-up involvement of pupils, parents, teachers, and staff.
Co-creation processes within the school system are challenging for researchers to study in a scientifically sound manner. Traditionally, many action researchers in school interventions have followed the cycle of needs assessment, development, implementation, monitoring, and evaluation of change [23
]. However, the limitation of following these steps is that they suggest a logical, causal process, which is difficult to identify in a complex school system initiative [25
], where changes are interacting with each other and with other contextual aspects of the school [15
]. Therefore, we felt the need to find a way to adapt our research to this complexity. As this study is not the first to deal with initiatives in complex systems, we have been inspired by the existing literature, from which various considerations and insights were gathered [12
]. These insights into systems thinking have led us to adapt the principles of action research into a contextual action-oriented research approach (CARA). The purpose of CARA is to contribute as researchers to the process of a complex intervention initiative, and to conduct a thorough evaluation of the process and its final outcomes that addresses the importance of the implementation context. Basic properties of CARA are its specific focus on contextual differences, and the use of monitoring and feedback to both support and evaluate the process of change. The approach centers around four key questions: (1) What is the pre-existing context of each school?; (2) How does the process of change in each school evolve and which factors affect this process?; (3) How can research contribute to the process of change?;
and (4) Do children’s health and health behaviours improve as a result of the HP changes?
The aim of this article is to elaborate on the way in which we are dealing with the complexity of the school system and the HPSF initiative by using CARA and how we are able to contribute to the initiative and at the same time conduct a thorough evaluation. HPSF has a broad focus on different aspects of health and well-being. The present study focuses on two key aspects, i.e., healthy nutrition and physical activity (PA).
The present paper has introduced the contextual action-oriented research approach that we developed to deal with the complexity of both the school system and the HPSF initiative. The paper has shown how we aim to contribute to the initiative and at the same time conduct a thorough evaluation. Some methodological, practical, and/or integrity limitations and strengths of the approach need to be discussed. First, fully assessing and understanding all aspects of each context, the process of change, and the implementation of each change, is impossible due to limitations of time, resources, and participant burden [13
]. Therefore, CARA researchers have to make difficult selection choices about which data to collect and in how much detail. Luckily, we do not have to start from scratch. We build on previous work published in the international literature on, e.g., relevant concepts in the process of change in complex (school) systems, which was essential for this decision process. We included methods that are appropriate to obtain rich information, that are feasible for the researchers and the schools, and that can be translated into rapid and real-time feedback for the schools.
Second, we believe that the data collection instruments in CARA need to fit the context to be able to get meaningful data. Therefore, existing instruments might not be available or have to be adapted to the context. In the current study, we have adopted the strategy of preferring adjustment of existing and tested instruments over the development of new instruments. By using the principle of data triangulation in our analysis, we combine the accuracy of the quantitative questionnaires with the in-depth insights that interviews and observations afford.
Third, CARA includes not only an evaluation of the process of change in the schools, but also an effect study to investigate the evidence for the behavioural and health effects of these changes among the children. Since randomization is neither desirable nor feasible here [26
], a quasi-experimental study design is used. To investigate for whom and under which circumstances the changes cause the greatest effect, the school-specific effects are combined with relevant process and contextual factors at each school. Although we do recognize the importance of assessing the implementation fidelity, the focus in our study is not on the fidelity of intervention components, but on identifying when and how adaptation take place, and which factors prove to be crucial for sustained changes. This specific focus of interest is based on the notion that even small changes may produce large effects in a specific context (i.e., ‘tipping’ point) [12
]. Better (i.e., high fidelity) implementation of a change does thereby not necessarily mean greater effect [13
]. Another aspect to consider is that CARA researchers are not external observers, but actively participating partners in the initiative. By including an effect study with a quasi-experimental design in CARA as well, we believe we combine the best of two worlds: the advantages of a researcher involved in the process of change and still being able to study the effects objectively.
Finally, CARA also implies time-consuming research, as a thorough insight into the school context is necessary. This insight requires a relationship of trust between the researchers, the schools, and all other partners involved, which takes time to build. In this relationship, nobody should be afraid to say what really bothers them, which should yield data that reflect the real situation. Moreover, CARA requires flexible time planning of the researchers; they need to be able to react quickly to what happens in each school, to be able to give relevant support, and to analyse the data quickly to be able to give rapid and real-time feedback. At the same time, the feedback process needs to take place in a careful manner, as both the initiative and the research can benefit from an open discussion of the real situation of those involved without losing the trust of the informants. To maintain a relationship of trust, we aim to offer honest but discrete feedback. When results cannot be fed back anonymously, we aim to ask permission of the informants before communicating the feedback to others. CARA can also be time-consuming for the schools. Researchers need to consider this and ensure that studying the initiative is feasible for them and the schools. Due to our critical selection of mixed methods, we believe we have found a feasible way to support and evaluate the initiative in the schools. By means of the feedback provided and the focal points included to further improve HPSF, we aim to offer added value to the schools which we hope outweighs their time investment.
In addition to these methodological, practical, and integrity aspects, there is another important aspect to consider: whereas the current paper mainly focuses on the school complexity, the school is only one of a child’s microsystems. Changes in children’s home setting and neighbourhood, their other microsystems, also interact with the impact of changes at school [7
]. Thus, the complexity goes beyond the focus of the research described in this paper.