Background
The decline in physical activity (PA) levels and the possible associated health problems of young people continue to be of growing concern internationally [
1]. The decline in PA levels is greatest during the secondary school years (aged 12–18 years), and previously published research consistently reports lower PA among adolescent girls than among adolescent boys [
1,
2]. Globally, an estimated 80% of adolescents (aged 13 – 15 years) are not meeting the PA recommendations of at least 60 minutes of moderate to vigorous PA daily [
1]. As a consequence increased attention has been directed towards the development and evaluation of PA interventions targeting young people, and in particular adolescent girls [
3‐
5]. Key settings for PA promotion among adolescents include schools, home, community and primary care; however, by far the most commonly targeted setting for this age group is schools [
3,
6].
Schools are considered a key health setting to promote PA since young people attend and spend a considerable amount of time in school [
7]. The convenience of targeting existing curricular opportunities like physical education (PE), in addition to extra-curricular opportunities such as sport programs makes schools an obvious setting in which to deliver programs designed to promote PA participation. Numerous PA promotion interventions have been conducted within school settings with varying success [
6]. A systematic review of interventions to promote PA within adolescent populations reported that the strongest evidence of success was associated with multicomponent school-based interventions that are combined with community opportunities for PA that address multiple levels of influence on behaviour (i.e. as outlined in socio-ecological frameworks) and include enjoyable PE experiences as a main component [
3]. Effective interventions are also reported to consist of both a PA and cognitive component [
6]. Peer support strategies are also considered promising; whereas family support has been reported by some as ineffective [
3]. Strategies considered most appropriate for public health agencies and partners to promote PA to adolescents include community-wide campaigns, increased access to places in which to be active, educational information or outreach programs, and enhanced PE programs integrated within a socioecological perspective [
8].
Most PA promotion interventions have been conducted in the United States and United Kingdom and have tended to focus on children (aged 6–12) rather than adolescents (aged 13 – 19) [
6]. They have also have tended to include both sexes [
6], despite adolescent girls being identified as a priority group for PA promotion and the need to design interventions to specifically address the needs of girls [
3]. Furthermore, few PA interventions for girls have been conducted in low-socioeconomic areas [
3], despite adolescents from low income communities having lower levels of PA [
9]. As a result, a school-community linked PA-promotion intervention program targeting adolescent girls living in low-SES Australian rural and regional communities was developed, implemented and evaluated. The aim of the program was to improve Health-Related Quality of Life (HRQoL), levels of PA, and a range of potential mediators of PA (e.g. self-efficacy, perceived sport competence). The outcomes of a pre- and post- evaluation of effectiveness of this program found that: 1) The intervention had a protective effect on the intervention group’s health-related quality of life (HRQoL), whereby the intervention group maintained their HRQoL, whilst a decrease was observed in the control group. 2) There were no statistically significant differences between groups for any of the PA measures including - mins of leisure-time moderate-vigorous physical activity (LTMVPA), MET-mins of LTMVPA, or in the proportion meeting PA guidelines. 3) Among 'completers’ — those who had participated in the both the in-school component and outside of school community component — the intervention had positive effects on intra-personal capacities (i.e. self-efficacy, self-management, perceived behavioural control, outcome expectancy-value) and inter-personal factors (i.e. support from family and friends) [
10]. Understanding the reach, adoption and implementation of this program is important to help understand why the program was successful in achieving some of the intended outcomes and why other elements were not achieved.
PA intervention studies commonly provide an evaluation of the effectiveness of the intervention, however, more recently there has been a greater emphasis placed upon the importance of evaluating the context in which interventions are implemented [
11]. Indeed, few studies have provided a process evaluation of PA interventions targeted at girls [
12,
13]. Process evaluation is useful to help understand why a program was successful in achieving its intended outcomes [
14]. Process evaluation of multicomponent interventions can be complex and mixed method research designs are required to understand the variety of contexts and settings [
13,
15]. The RE-AIM model (reach, effectiveness/efficacy, adoption, implementation and maintenance) provides a systematic framework by which researchers can evaluate the external validity of multicomponent interventions within real-world settings such as schools [
16], and has been used to evaluate school-based PA interventions, mostly in primary schools [
7,
17,
18] and recently in a secondary school [
12].
The aim of this research was to conduct a process evaluation to understand the reach, adoption and implementation of a school-community linked PA intervention for adolescent girls (aged 12 – 15 years) in low-socioeconomic regional and rural areas of Victoria, Australia. Effectiveness of the intervention has been published previously [
10] and maintenance was not evaluated due to the short duration of follow-up of this study. Reflection on the methods adopted in the process evaluation of this multicomponent intervention are also discussed to inform future process evaluation plans.
Discussion
The implementation of the program aimed to make positive changes to rural and regional-living adolescent girls’ health-related quality of life (HRQoL), PA levels and mediators of PA. The program was implemented in a real world setting, with PE teachers engaging in professional development opportunities that sought to introduce them to the principles underpinning the design of lesson plans and facilitate links to community-based sport and recreation providers. There were a number of significant positive results in terms of program effectiveness [
10], which were influenced by implementation factors related to the capacity building framework adopted in this study including satisfaction with professional development opportunities, including the training, resource manual and opportunities to work with coaches and instructors during PE classes. Furthermore, a number of positive perceptions from students and teachers were reported in relation to the program design which supported student learning and engagement. In particular, students reported increased awareness of PA opportunities outside of school, the development of new sport specific and PA skills and enjoyment of the program. Students also talked about having an understanding of when certain strategies would be used in game situations demonstrating the development of higher order thinking skills.
However, notwithstanding some positive outcomes, the degree of success of the intervention was limited by a range of factors. This study provides critical discussion around the barriers to implementation of a school-community linked program to inform practitioners and future intervention-based research in schools.
The socio-ecological model was a useful framework to consider and address the wide range of factors that influence PA behaviour and as such the program was underpinned by several elements reported to be most effective in promoting PA among girls [
3,
6]. These elements included: a multicomponent intervention with school and community opportunities for PA, addressing multiple levels of influence on behaviour, and applying student-centred pedagogical approaches and self-management strategies to engage students in a traditional team sport (football), a lifestyle sport (tennis) and a range of lifestyle physical activities (leisure centre activities) both in PE and in the community. The intervention, however, did not include a focus on family support and instead focused on inclusive peer strategies. Whilst family support has been reported as ineffective by some [
3], others have reported a positive impact of family support on sports club membership of adolescent girls [
45]. In particular, family support may be important in a regional context, especially where public transport is limited, to ensure young people can access PA opportunities outside of school [
45]. This study found that students experienced barriers related to travel including getting to and from PA opportunities. Strategies to support young people and their families may be required, particularly in low socioeconomic communities where access to familial resources such as a parent’s ability to make themselves available to transport children to/from opportunities to engage in PA are hampered [
24].
The likelihood of such a program achieving the intended health outcomes, such as changes in HRQoL, PA or mediators of PA is highly dependent on its implementation. Schools are undoubtedly a complex and challenging setting in which to implement change and are often constrained by barriers such as organizational factors, policy constraints, lack of resources and a crowded curriculum [
12]. More specifically, difficulties associated with implementing interventions in schools have included the: lack of teacher adoption, lack of program readiness, absence of program champions (advocates), inadequate or unsustainable funding, inadequate infrastructure, poor integration of the program within existing structures and programs, limited teacher training, insufficient program support materials and staff turnover [
31]. In this study, implementation was sometimes negatively influenced by a lack of individual and organisational readiness to adopt program principles. This was despite the apparent positive response to the individual-focused strategies implemented under the capacity building framework.
Individual and organisational readiness to implement the program varied. For instance, self-management strategies were not well implemented by teachers because they tended to implement cognitive components of the intervention in a selective manner. Furthermore, there were deeply embedded ideologies based upon performance discourses that reinforce the acquisition of skills through demonstration, explanation and practice teaching approaches [
44], which was in contrast to the principle of Game Sense, which seeks to refocus attention from skill performance to the tactical and strategic (cognitive) dimensions of a game. The Game Sense model was purposefully adopted for the sports of football and tennis to reduce the focus on skill performance in PE settings as raised by adolescent girls in the ethnographic fieldwork phase of the study [
22]. It was also selected because Game Sense aligned with recent shifts in both sports coaching and the teaching of games [
46].
Teaching pedagogies such as Game Sense often incite resistance from practitioners because they challenge the traditional and enduring belief that fundamental movement skills are prerequisite to game play [
44]. The implementation of Game Sense is often fraught with challenges including the change in decision making from the teacher to the student, variations in the interpretation of Games Sense [
47], lack of exposure to effective game-based professional development opportunities, and performance discourses embedded within PE and youth sport programs [
48]. Game Sense approaches have reported opportunities for student decision making, social interaction and cognitive understanding and/or high order thinking [
49]. Further, higher levels of enjoyment, particularly among girls has been reported in approaches that draw on Game Sense principles [
50]. Whilst capacity building strategies to empower teachers and coaches to implement a Game Sense approach, as well as other elements of the intervention like self-management strategies, were provided further work is required. It appeared that teachers and coaches would select aspects of the lessons that resonated with their own theories of practice. This may have occurred as their professional development (in relation to the program) was not necessarily ongoing or sustained, rather it was limited to a one day professional development session supplemented with collaborative learning opportunities with the introduction of coaches and instructors into the PE setting. It has been suggested that professional development for PE teachers has greater impact when it is ongoing and teachers place high value on learning collaboratively with and from each other in professional learning communities or networks [
51]. Therefore, ongoing sustained opportunities to engage in critical reflection throughout implementation of the intervention and/or provision of digital resources that support implementation (e.g. case studies and video clips that describe some of the challenges of using a Game Sense approach) is important to further build individual capacities in future school-focused and pedagogical-based PA interventions. Particularly, to realise the potential of Game Sense and broader game-based approaches to enhance girls participation in PA [
42].
Fundamentally, despite the use of a range of capacity building strategies, the implementation of the program required organizational changes in order to maximise individual changes at the student level. Weiner [
52] conceptually theorized organizational readiness as “a shared psychological state in which organizational members feel committed to implementing an organizational change and confident in their collective abilities to do so” (p.1). Further Weiner [
52] states,
Organizational readiness for change varies as a function of how much organizational members value the change…When organizational readiness for change is high, organizational members are more likely to initiate change, exert greater effort, exhibit greater persistence, and display more cooperative behavior. The result is more effective implementation.
Organizational readiness for change has been identified as an under-recognised area in health promotion practice [
53] and there is a need for greater recognition of the time it takes to facilitate organizational change, particularly within schools, to support the implementation of health promoting programs [
54]. However, organizational readiness and capacity for change has not been subjected to extensive theoretical development and empirical study [
52]. Considering the varying understandings of, and commitment to, the intention of a Game Sense approach and the cognitive component within this study; along with a perception that PE was marginalized within the wider school curriculum, understanding individual and organizational readiness may be especially important to determine the shared commitment and ability of program deliverers to implement the program. As such, future interventions and research studies should seek to assess and develop individual and organizational readiness within the study design to better engage schools and sport and recreation providers in the provision of school and community linked PA opportunities for adolescent girls.
Single-sex classes were a contentious issue in this study. Since the mid 1980’s there has been a shift in Australia to the implementation of co-educational classes for PE; although there is continuing controversy on co-educational and single-sex PE [
55,
56]. As reported in previous research, girls often need to feel emotionally safe to fully participate as boys often dominate play in game-situations [
57] and trivialise girls’ interests, concerns and physical behaviours [
58]. Students construct meaning from their experiences in PE lessons by drawing upon their thinking about many factors including their bodies, gender and societal norms. It has long been argued that the dominant masculine practice of traditional team sports in the PE curriculum and has tended to privilege boys and marginalised girls [
59]. Further, the role of the teacher in the construction and reproduction of, and resistance to dominant gender power relations has been of interest to educational researchers for some time [
59,
60]. Teachers expect girls to be tentative and reluctant to reaffirm their femininity; whereas, boys are assumed to be more compliant, because for a boy to display resistance in PE would question his masculinity [
59]. Various strategies were suggested within the lesson plans to emphasise inclusive interactions among students regardless of how the PE classes were organised (single sex or co-educational). In this study, schools tended to implement single sex classes to deliver the program, which was received positively for a number of students. However, it appeared that efficient skills in managing group dynamics were important. The impact teachers, coaches and instructors can make on students’ social interactions within the PE, sport and recreation setting remains an area warranting further study.
Some limitations to this study need to be acknowledged. Direct observation of program implementation may have strengthened the assessment of process evaluation; although this too is not without measurement issues. Direct observation was not practical for this large study due to the high cost associated per observation [
61]. In addition, the presence of observers may affect the implementation behaviour of program deliverers and/or student behaviour [
61]. Instead, data triangulation was implemented which involved teachers completing a short feedback form after each class to capture immediate thoughts and reflections and later teachers participated in a focus group to provide an overall reflection on the program. The response rate for the feedback forms was moderate (60%) and attendance by teachers at focus groups was high (>85%), which was useful to generate meaningful discussion about their experiences. Upon reflection of the evaluation plan, data triangulation was a useful method as the feedback forms indicated a high level of program implementation, although focus groups revealed that the understanding of, and commitment to, the intention of a Game Sense approach varied and subsequently influenced the consistency of program implementation. In future evaluation plans, an online reporting system, with all registered program deliverers sent email reminders may improve recording of immediate feedback.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
WP and AT were co-principal investigators on this study. They led the design of the study and obtained funding. WP was the chief investigator and oversaw all aspects; whilst AT led the development of the school component. John Smyth was also a co-principal investigator on the study and led the ethnographic fieldwork. MC was the project manager and was responsible for data collection, implementation, and analysis. AT and AM designed the intervention materials and provided professional development to schools. AM also assisted with qualitative data collection and critically commented on program implementation and qualitative analysis. JH oversaw statistical design, analysis and reporting; RE and AM provided expertise in sport and recreation and physical education respectively. All authors contributed to the design of the study and critically commented on the manuscript, and read and approved the final manuscript.