Optimising a multi-strategy implementation intervention to improve the delivery of a school physical activity policy at scale: findings from a randomised noninferiority trial

Abstract

Background: To maximise their potential health benefits, school-based physical activity policies need to be implemented at scale. This paper describes the third in a sequence of trials that sought to optimise an effective strategy (PACE) to assist schools' implementation of a physical activity policy. Specifically, it aimed to determine the probability that a multi-strategy intervention adapted to reduce in-person contact (Adapted PACE) was "as good as" the original intervention (PACE) in increasing the weekly minutes of structured physical activity implemented by classroom teachers.

Methods: A noninferiority cluster randomised controlled trial was undertaken with 48 primary schools in New South Wales, Australia. Schools were randomised to receive PACE or a model with adaptations made to the delivery modes (Adapted PACE). Teachers' scheduled minutes of weekly physical activity was assessed at baseline (Oct 2018-Feb 2019) and 12-month follow-up (Oct-Dec 2019). The noninferiority margin was set at - 16.4 minutes based on previous data and decision panel consensus. A linear mixed model analysed within a Bayesian framework was used to explore noninferiority between the two PACE models. A cost minimisation analysis was conducted from the health service provider perspective, using the Australian dollar (AUD).

Results: The posterior estimate for the between group difference at follow-up was - 2.3 minutes (95% credible interval = - 18.02, 14.45 minutes). There was an estimated 96% probability of Adapted PACE being considered noninferior (only 4% of the posterior samples crossed the noninferiority margin of - 16.4 minutes). That is, the minutes of physical activity implemented by teachers at Adapted PACE schools was not meaningfully less than the minutes of physical activity implemented by teachers at PACE schools. The mean total cost was AUD$25,375 (95% uncertainty interval = $21,499, $29,106) for PACE and AUD$16,421 (95% uncertainty interval = $13,974, $19,656) for Adapted PACE; an estimated reduction of AUD$373 (95% uncertainty interval = $173, $560) per school.

Conclusions: It is highly probable that Adapted PACE is noninferior to the original model. It is a cost-efficient alternative also likely to be a more suitable approach to supporting large scale implementation of school physical activity policies.

Trial registration: Retrospectively registered with the Australian New Zealand Clinical Trials Registry (ACTRN12619001229167).

Keywords: adaptations; children; implementation; noninferiority; optimisation; physical activity; policy; scale-up; school.

Fig. 1

Flowchart of school enrolment and data collection throughout the study

Fig. 2

Distribution of the posterior estimated differences in teacher's total scheduled minutes of physical activity between groups (uninformative prior)