A multicomponent school-based intervention in disadvantaged secondary schools to reduce the decline in physical activity associated with adolescence: the Physical Activity 4 Everyone randomized controlled trial

Research Doctorate - Doctor of Philosophy (PhD)Background and Aims: Physical inactivity has been described as a primary cause of most chronic conditions, as important as both tobacco and obesity as a major modifiable risk factor for chronic diseases. The economic burden of physical inactivity globally is INT $53.8 billion. Despite this, as few as 20% of adolescents globally meet current physical activity recommendations, with socio-economically disadvantaged adolescents less likely to be physically active. Given evidence suggests physical activity levels throughout adolescence track into adulthood, effective interventions targeting socio-economically disadvantaged adolescents are warranted. Comprehensive school-based physical activity interventions have the potential to impact on physical activity levels, yet few such interventions have targeted socio-economically disadvantaged adolescents. As a result, development of cost-effective school-based physical activity interventions targeting socio-economically disadvantaged adolescents is a public health priority. Methods: The primary aim of this thesis was to evaluate a 24-month, school-based physical activity intervention in a trial targeting a cohort of Grade 7 students attending schools located in socio-economically disadvantaged communities (<i>Physical Activity 4 Everyone (PA4E1)</i>). The <i>PA4E1</i> intervention was evaluated using a cluster randomised controlled trial (RCT) involving 1100 adolescents (Grade 7, mean age 12.0 years at baseline) from five intervention and five control schools located in the Hunter, Central Coast and Mid North Coast regions of New South Wales, Australia. The two year multicomponent intervention was guided by socio-ecological theory and the Health Promoting Schools Framework, incorporating seven physical activity strategies and six implementation support strategies. The three physical activity strategies implemented across the curriculum were teaching strategies to increase physical activity in physical education lessons, student physical activity plans and enhanced school sport programs; the two school environment strategies were recess/lunchtime activities and school physical activity policy; and two broader school environment strategies were linking schools with community physical activity providers and linking with parents. Six additional strategies supported school implementation of the physical activity intervention strategies including an in-school physical activity consultant, leadership and executive support, teacher training, resources, prompts and intervention implementation performance feedback. The primary outcome was mean duration of moderate-to-vigorous physical activity (MVPA) minutes per day assessed using Actigraph (GT3X) accelerometers at baseline, and 12- and 24-months post randomisation. Additional physical activity outcome measures included: mean minutes per day of vigorous and moderate activity, counts per minute, % wear time spent in MVPA, vigorous and moderate activity, in-school and out-of-school physical activity. Secondary outcome measures were weight, body mass index (BMI), and BMI Z-score. In addition, a cost effectiveness evaluation was undertaken whereby intervention costs and incremental cost effectiveness ratios were calculated for both physical activity and adiposity. Physical activity and weight status data were analysed using repeated measures linear mixed models with models developed for the baseline to 12-month period, as well as baseline to 24-month period. Results: Parental consent was provided for 1233 of the 1468 Grade 7 students from participating schools. At baseline, 1150 students wore an accelerometer (mean age 12.0 years, 54% female), with 965 providing at least three days of valid wear data (83% of accelerometer wearers, 78% of those with consent). At 24-month follow-up, 985 students wore an accelerometer (mean age 14.0 years, 57% female), with 441 of these (45%) providing valid wear data. At both 12- and 24-month follow-up there was a significant group-by-time effect in favour of the intervention group for MVPA. At 12-month mid-intervention follow-up, students in the intervention group participated in 3.85 minutes (95% CI= 0.79, 6.91) more MVPA per day than students in the control group. At 24-month follow up students in the intervention group participated in 7.02 minutes (95% CI= 2.68, 11.36) more MVPA per day (p = ≤0.01) than students in the control group. The mean duration of daily MVPA increased by 4.39 minutes for intervention group students and decreased by 2.63 minutes for control group students. The intervention group students participated in 2.53 minutes more vigorous physical activity (p=0.03, 95% CI= 0.27- 4.79) and 4.5 minutes more moderate physical activity (p≤0.01, 95% CI= 1.98, 7.03) than the control group students at 24-months post randomisation. At 12-month mid intervention, there was a significant group-by-time effect for weight (mean difference=-0.90kg) and BMI (-0.28kg/m2) in favour of the intervention group. At 24-months, there were statistically significant group-by-time effects for weight (mean difference= -0.78 kg, 95% CI= -1.40; -0.16, p=0.03) and BMI (mean difference= -0.28, 95% CI= -0.50,-0.06, p=0.01) in favour of the intervention group. The intervention cost was AUD $329,952 over 24-months. The incremental cost effectiveness ratio per additional minute of MVPA per day was AUD$56 ($35 - $147) and AUD$563 ($282 - $3,942) per 10% reduction in BMI z-score. Conclusion: The <i>PA4E1</i> trial showed the intervention was effective in not only reducing the decline in physical activity among adolescents attending schools located in socio-economically disadvantaged areas, but in increasing physical activity in comparison to a decrease in the control group. In addition, the intervention had a significant positive effect on adiposity and BMI. The findings suggest that implementation of the intervention by socio-economically disadvantaged secondary schools has the potential to reverse the decline in physical activity in this population group at a relatively small marginal cost. Further understanding of the mechanisms for implementation of the program at scale is required to contribute towards achieving health gains at a population level. The results of the trial suggest an opportunity for the dissemination of the evidenced based program to a larger number of schools. Measuring the sustainability of the intervention, inclusive of effect on both student level outcomes and school practice implementation level outcomes is suggested