Feasibility, acceptability and potential efficacy of a virtual physical activity program in primary and secondary schools in New South Wales, Australia: A quasi‐experimental study

Abstract Issue Addressed Child and youth participation in physical activity (PA) is fundamental for healthy development and obesity prevention. Government policy requires schools to offer 150 minutes of PA each week, however compliance is low. Race around Australia (RAA) is a New South Wales (NSW) Department of Education, virtual PA program aimed at assisting schools in meeting the PA guidelines. Methods A pre‐ and post‐intervention, quasi‐experimental study was conducted using a mixed‐methods approach comprising teacher interviews, a student questionnaire and a 1.6 kilometre (km) timed run. Data were collected from April to September 2021 among students and teachers in Grades 5 to 8, from 10 schools in NSW, Australia. Results The analytical sample included data from 918 students and 17 teachers. The RAA program was deemed feasible and acceptable in primary schools, whereas there were several systemic and intrapersonal barriers to implementation success for secondary schools. In primary schools, RAA increased PA opportunities and the 1.6 km timed runs revealed a statistically significant treatment by time effect in favour of the intervention group for cardiorespiratory fitness (−36.91 seconds, 95% CI [−63.14, −10.68], P = .006). Conclusions RAA has demonstrated feasibility and potential efficacy in improving cardiorespiratory fitness. We recommend that program refinement be made to deliver an intervention that addresses the unique barriers of the secondary school setting through a multi‐level ecological approach. So What? Despite evident benefits, implementation of PA initiatives in the school setting reveals many challenges. Stronger consideration of the Health Promotion with Schools Framework is evidently needed.


| INTRODUCTION
In Australia, almost one-quarter (24.9%) of children and young people aged 5 to 17 years are overweight or obese, 1 whilst in the state of New South Wales (NSW), 19.3% of children are affected. 2 Sufficient levels of PA are key for obesity prevention. 3 Appropriate PA levels in children and young people are beneficial for the development of cardiorespiratory and muscular fitness, and healthy weight [4][5][6] ; are positively associated with motor skill development, bone health and reduced psychological stress; enhance wellbeing, cognitive performance, and classroom behaviour and are predictive of overall health. 4,6 These associations are strengthened by participation in moderate-to vigorous-intensity PA (MVPA). 6 Insufficient PA levels in children and young people are a major risk factor for obesity, chronic disease and health problems into adulthood. 4,7 An increasing evidence base recognises youth sedentary behaviour as an independent risk factor for poor physical, social and mental health. 8 Specifically, higher levels of sedentary behaviour have been positively associated with cardio-metabolic risk factors, unfavourable body composition, low fitness, poor behavioural conduct and low self-esteem. 8,9 Schools have been identified as a key setting for increasing PA levels and decreasing sedentary levels, 2,3,10,11 given children and young people spend approximately 35 hours a week attending school. 2 Schools present an opportunity to reach a range of students, regardless of ethnicity, gender or socio-economic status. 10,11 The NSW Department of Education (DoE) Sport and PA Policy outlines the requirement that schools offer students opportunities to participate in a minimum of 150 minutes of planned, moderate PA with some vigorous PA each week. 12 Data indicate that compliance with the policy is low, with only 66% of primary school students and 38% to 45% of Grade 8 to 10 secondary school students receiving at least 120 minutes of physical education each week. 13 Classroom movement integration aims to increase PA and decrease sedentary time beyond school physical education lessons, by integrating movement into scheduled classroom time. 11 Systematic reviews have demonstrated positive outcomes in relation to participation in PA, [14][15][16][17] enjoyment in PA 16 and school-related outcomes such as classroom behaviour 14,15,17 and academic performance. 16 The literature predominantly focuses on the primary school setting [14][15][16][17] with few studies reporting on secondary schools, and few studies reporting cardiorespiratory fitness outcomes of active-break interventions. Evidence suggests that cardiorespiratory fitness is a proxy indicator for PA outcomes in children and youth, 18 and could lead to a more accurate representation of intervention success than focusing solely on PA markers. 19 Further research is needed to understand the outcomes associated with classroom-based movement integration programs. 16,17,20,21 Additionally, research specific to the adoption and maintenance phases of implementation is required to identify facilitators for implementing school-based PA interventions in real-world conditions, at scale. 21,22 1.1 | The race around Australia program As part of their mandate, the DoE School Sport Unit are required to provide advice and assist in developing supporting materials to facilitate the implementation of the Sport and PA Policy in NSW schools. 12 Race around Australia (RAA) is a School Sport Unit PA initiative that is part of a suite of interventions that make up the NSW Premier's Sporting Challenge. 23 RAA was developed by a Grade 10 secondary school teacher, it was piloted across 46 NSW DoE schools in 2020 and in Term 2 2021, was formally offered to all 2220 NSW DoE schools. Once registered, students complete short bursts of PA during class time, utilising the RAA workouts, convert these activities to kilometres (kms) and enter this distance on the RAA online platform ( Figure S1). The distance is then displayed on a map, tracking students, classes and schools as they race a set course around the coast of Australia over the course of a term. As students progress around the map ( Figure S2), they unlock stage-specific curriculum blogs. 24

| Present evaluation
The previous pilot evaluation of RAA across 46 schools, reported positive results for the program through the analysis of self-reflection and feedback from staff and students, an evaluation of teacher and student engagement in the online platform, semi-structured interviews and website analytics. However, the evaluation did not objectively measure any health indicators associated with the intervention. 24 Consequently, our evaluation aimed to investigate the potential efficacy of RAA by including an assessment of cardiorespiratory fitness. A secondary aim was to examine the feasibility and acceptability of the upscaled version of RAA across primary and secondary school settings, by exploring barriers and facilitators to implementation.

| Study design
A pre and post intervention, quasi-experimental study was conducted using a between-subject design and mixed methods approach. 25 Ethical approval was obtained from the University of Wollongong Human Research Ethics Committee (Reference 2021/ETH00065) and the NSW State Education Research Applications Process. Informed written consent was obtained from all school principals and individual teachers involved in the focus groups. An opt-out approach was utilised for student recruitment, with study information distributed to parent/caregivers via each school's preferred forum. Opt-out consent was deemed the most appropriate approach for students in this study due to the low risk associated with the research and the need to recruit a representative sample. This study aligns with all opt-out consent requirements as outlined in Section 2.3 within the National Statement on Ethical Conduct in Human Research. 26 Additionally, student assent was sought prior to data collection.

| Participants
All DoE primary and secondary schools within the Illawarra Shoalhaven and South Western Sydney Local Health Districts were invited to participate in the evaluation. We aimed to recruit three control and three intervention schools within each health district, restricting data collection to children in Grades 5 to 8. By selecting these age groups, analyses would capture the transition from primary to secondary school and highlight any differences between the two settings. Additionally, the tools utilised in this study have been validated for these age groups. 18 Intervention schools were recruited from schools that had registered an interest in implementing RAA; conversely control schools were recruited from those who had not registered interest in RAA.
We aimed to match intervention and control schools across a range of variables, including school size, geographic location, Indices for Socio-Economic Disadvantage, and the Indices of Community Socio-Educational Advantage, to ensure compatibility.
All schools who participated in the evaluation received $1000 to be spent on sporting equipment, as an acknowledgement of their efforts, presented at the conclusion of data collection.

| Intervention
Intervention schools registered in the RAA challenge and, with support from the School Sport Unit, implemented the program autonomously for 9 weeks. Control schools continued with their regular activities throughout the term.

| Theoretical framework
Two key frameworks commonly utilised in children's health promotion, formed the theoretical basis of the analysis. 'Health Promotion with Schools: A Policy for The Health System' 27 is a NSW Health policy framework developed on the foundations of the World Health Organization's 'Global Standard for Health Promoting Schools' and acknowledges partnerships between health and education sectors as key to effective health promotion. 27 The framework emphasises consideration of the curriculum, community and school environment as essential for successful implementation of health programs in schools. 27 Programs that target PA determinants at the various levels of the ecological model, 28 hold the greatest potential for preventing and

| Outcome measures
The quantitative and qualitative tools utilised to assess the outcomes are detailed in Table 1. (1 = Strongly Disagree to 5 = Strongly Agree), to rate their experiences and feelings about participating in RAA. In most schools, surveys were administered to students by their teachers. Student cardiorespiratory fitness was measured at baseline and follow up via a 1.6 km fitness test conducted within each school by the research team. All teaching staff who participated in the delivery of RAA were invited to attend school level focus groups to evaluate the feasibility and acceptability of the initiative. Due to COVID-19 lockdowns, these were conducted via videoconference using Zoom (Zoom Video Communications Inc.). A semistructured interview guide was developed, asking open-ended questions, to ensure that key barriers or facilitators unknown to the interviewer were captured. 30 The interviews were recorded and later transcribed using transcription software (Otter.ai). Due to an imbalance in pre and post data collection numbers for the fitness measures, data were imputed utilising the Baseline Observation Carried Forward method 31 and analysed separately by school type. Linear mixed models were run to determine changes in the outcome measures between time points and treatment groups, with school name included as a random effect in the model to account for clustering. Differences between treatment types were considered statistically significant at P < .05. Effect sizes (Cohen's d) were calculated by dividing the difference between the means by the pooled standard deviation. 32 To date, there is no known software that produces effect size from a linear mixed model, so these were calculated using the unadjusted means.

| Qualitative analysis
Data were analysed following the guidelines for thematic analysis outlined by Braun and Clarke. 33 Following familiarisation with the data, each transcript was analysed using an inductive, open coding process, whereby meaningful quotes or key examples from teachers, were assigned a code. These codes were then grouped together to develop themes. Review of the themes by the research team helped attenuate individual bias from the analysis and add credibility to the findings.

| FINDINGS AND DISCUSSION
The findings are presented together with a discussion of relevant literature as suggested by Blignault and Ritchie. 34   Interviews with teachers revealed that one of the intervention secondary schools had not implemented the program as intended, hence the decision was made by the research team to exclude this school's data from the quantitative analysis. Therefore, of the 901 secondary students enrolled, 63.8% were included in the analytical sample for cardiorespiratory fitness.

| The sample
Demographic descriptors including gender, Indigenous status and country of birth were similar across intervention and control groups.
The mean age of primary school students in the intervention group was 10.6 years (SD = 0.68) and control group 10.3 years (SD = 1.55).
The mean age for secondary students was 11.9 years (SD = 0.47) in the intervention group and 12.0 years (SD = 0.46) in the control group. There was a statistically significant difference of 1.5 months between treatment groups in secondary schools, however, the implications for this current study were deemed not meaningful. There were no statistically significant differences in gender between intervention and control groups.
Six focus groups/interviews were conducted involving all intervention schools, with feedback from four primary and thirteen secondary school teachers and RAA Coordinators ( Figure 2). The qualitative analysis included the secondary school data that were excluded from the quantitative analysis, as their feedback was useful in understanding the barriers that contributed to the lack of program fidelity. responses to questions relating to enjoyment and future participation in RAA were more positively skewed when compared to secondary students (P < .01) with the mode and median responses of secondary school students 'neutral' for each question. Teacher interviews provided feedback on student enjoyment. One primary school teacher stated:

| The feasibility and acceptability of RAA
'… to just be able to do a fitness activity as a brain break was great… the kids really enjoyed it.' -Primary teacher

| Cross-curricular adoption
Generally, in primary schools, students belong to one class group and have one teacher with whom they spend most of the school day, hence implementation of RAA was flexible to that teacher's schedule.

| Confidence
Few secondary teachers reported using the activity resources, however those that did suggested they enhanced engagement, were useful for student leadership, and would be facilitating for non-healthoriented teachers if they were to adopt the program. Comparatively in their study of a movement integration intervention in secondary schools, Innerd and colleagues 40 discovered a lack of confidence amongst teachers in delivering activities that required physical skill and planning.
'It was great to have some stuff to show them (senior students) when they ran it (RAA)…. And I'm guessing that would be good for teachers who aren't physical education teachers as well.' -Secondary teacher One primary school teacher recommended creating videos to support the activity resources.
'I really liked the launch video where the two people actually did the activities and the kids could follow along with them… because sometimes that was a bit of a barrier, having to say the activity and then me myself, having to demonstrate' -Primary teacher

| Theme 3: Engagement with RAA program design elements
The RAA platform, in some cases, increased engagement with the program, by providing visuals and motivating students.
However, in upscaling RAA, the School Sport Unit developed a new online platform which was unable to keep up with the increased volume and demand of those utilising the website.
According to teachers, technical difficulties made interaction with the platform time consuming, limited the ability of students and teachers to experience the platform to its fullest potential and impacted student engagement. Teachers also suggested that a lack of consistent access to technology was a barrier to implementation in secondary schools.
'Technology in our school is scarce, at best. We have four computer rooms that are booked out continuously In a systematic review of school-based PA interventions in both primary and secondary schools, motivational interventions were deemed crucial for PA engagement, particularly for females. 44 'But once we got going, and we could see our class move along the map, they were very competitive and wanting to log their hours in… It was great, they loved seeing that visual.' -Primary teacher '…our school is a bit competitive, so it really appealed to the teachers and the students.' -Primary teacher Primary school teachers suggested that the number of milestones diminished the goal setting effect of the unlockable milestones.
'Because there were so many milestones… we didn't really

| Student leadership
A key aspect of successful implementation of school-based PA interventions is establishing a multi-component leadership committee, including students who may have the role of identifying enjoyable activities and promoting PA in the school. 45 In primary schools, the race element of RAA offered opportunities for students to adopt leadership roles, which was further motivating for students.
In one primary school, a teacher stated: 'I've got a student, in my class who became the leader… And he was the first one in our class to be like, "right, we need to do two runs before we start our fitness today so we can get those kms up"…he was really motivating.' -

Primary teacher
In primary and secondary schools, teachers suggested that if they were to implement RAA in the future, they would emphasise a student-led component, to lighten the burden of implementation on teachers and to further engage students. After speaking about the difficulties faced in engaging teachers, one secondary school pondered:  [15][16][17] Participating primary school teachers, on average, reported facilitating a RAA workout 3 to 5 times a week, which is typical of teacher-led movement integration interventions. 20 Comparatively, in secondary schools, only Personal Development Health and Physical Education teachers adopted the program, implementing it during practical lessons and at times, as active breaks in theory lessons. However, exposure to RAA and therefore the impact of the intervention was low due to an overall lack of cross-curricular adoption.
One primary school teacher specifically highlighted the increased PA participation amongst girls.
'…our girls were more eager to participate as it went on. So those participation levels for girls were really good.' 'And even some students that were a little bit reluctant towards PA, became involved. So I thought that part was great.' -Primary teacher Teachers also suggested that it increased the students' selfawareness of PA levels and encouraged students to complete more PA outside of school.
'I just noticed that some of my students were becoming more aware of how much PA that they were doing. And then we would have a discussion and there were certainly some that were trying to complete more, even on the weekends…' -Primary teacher In secondary schools, low exposure seemed to impact the benefits perceived by teachers.
'It was good, it did increase involvement, but not to the extent that it was extra beneficial for most of the kids.' -

Secondary RAA Coordinator
The feedback from both primary and secondary school teachers demonstrated the potential of RAA to increase PA levels when implemented as intended.

| Cardiorespiratory fitness
Changes in primary school students' fitness outcomes, from baseline to follow up, are depicted in Table 3. The mean difference between the intervention and control groups was statistically significant (À36.91 seconds [À63.14, À10.68], P = .006) and indicated a small intervention effect for primary school students (d = 0.33).
Few studies have explored this outcome previously. A study of a 10-month 'active lesson' intervention in students 9 to 10 years of age, showed improvements in aerobic fitness for the least fit students, however, no overall effects were found. 46 Further, in a study of an 8-week curriculum-based PA intervention in 9-to 11-yearolds, no substantial group by time effects were found for aerobic fitness. 40 3.10 | Primary school teachers' accounts of an overall change in attitude toward pa, and the use of workout resources beyond raa, reveal the potential for cardiorespiratory fitness outcomes to be sustained beyond the 9-week intervention period Changes in fitness outcomes from baseline to follow up for secondary school students are also depicted in At the interpersonal level of the ecological model, support and encouragement amongst peers, teachers and class groups, through student leadership and within school competition, were facilitators for engagement and implementation. These themes were less prominent in secondary schools; however, student leadership was utilised in some instances and teachers suggested that strengthening this aspect could be an effective approach going forward.
At the intrapersonal level, primary school teachers and students demonstrated greater enjoyment in the intervention than secondary schools, due to both the feasibility of the intervention and the attitude of teachers and students. Teachers' skills, attitudes and experiences with classroom PA, influenced their adoption of the intervention. Lower levels of motivation amongst secondary school students when compared to primary school students impacted the ability of secondary school teachers to engage students.
A multi-level ecological approach is needed to ensure the intervention is feasible and sustainable in both school settings across the various domains of the ecological model.

| LIMITATIONS
Nonrandomisation could be a limitation of this study. In comparison with RCTs, in nonrandomised trials, there is a greater chance that the differences between treatment groups will influence observed outcomes in ways unrelated to intervention exposure, as often, nonrandomised trials rely on enthusiastic participants to volunteer allocation to the intervention group. 48 Poor response rates for the student surveys highlighted the need for assistance from researchers with completion. Further, due to high numbers of missing data, the Baseline Observed Carry Forward imputation method was utilised, whereby all students with missing follow up data were assigned their baseline value, and all follow up data from students that did not have baseline data, were removed from analysis. This method is labelled as conservative, and therefore can strongly underestimate the effects of an intervention. 31 Ideal methods for treating missing data are highly debated. 31 Lastly, this evaluation failed to measure a dose-response effect for the intervention, and therefore it is difficult to make meaningful comparisons between exposure and outcome. Choosing an appropriate measure for PA in youth is complex. 49 The study demonstrated the need to pilot tools and implement the selected tools with researcher supervision. We recommend using both cardiorespiratory fitness testing as a proxy-indicator for PA, and accelerometers to provide evidence of the effect of the intervention on in-and out-ofschool physical activity levels.

| CONCLUSION
The current research was the first known to directly compare the fea-

CONFLICT OF INTEREST
All authors declare that they have no conflicts of interest.

DATA AVAILABILITY STATEMENT
Quantitative data are available from the corresponding author on reasonable request. Data for this study is not openly available as participants did not provide informed consent for data sharing.

ETHICS STATEMENT
Ethics approvals were sought and gained from: the University of Wollongong Human Research Ethics Committee (Reference 2021/ ETH00065) and the NSW State Education Research Applications Process (SERAP).

PATIENT CONSENT
Through opt out consent, all participants have consented for this research to be published.