Cheung 2018.

Study characteristics
Methods	Trial name: Power Up for 30 (PU30). Study design: non‐RCT Intervention duration: approximately 1 year (not explicitly stated) Length of follow‐up from baseline: baseline (2013–2014) and follow‐up (2015) Differences in baseline characteristics: PU30‐trained schools were more likely than untrained schools to be of high (> 75%) FRL but less likely to be of mid‐high FRL (50% < FRL ≤ 75%). Trained schools had a lower proportion of white students, had a higher proportion of Hispanic students, and were, on average, larger. A greater proportion of trained schools were in suburbs while more untrained schools were in rural areas. At baseline, PU30‐trained schools had 11 fewer minutes of recess, but 10 more minutes of in‐class PA breaks each week compared to untrained schools. Unit of allocation: school Unit of analysis Implementation outcomes: school Behavioural / health outcomes: PA frequency
Participants	School type: Georgia public elementary schools Region: Georgia USA Demographics/socioeconomic characteristics: PU30‐trained schools were more likely than untrained schools to be of high (> 75%) FRL but less likely to be of mid‐high FRL (50% < FRL ≤ 75%). Trained schools had a lower proportion of white students, had a higher proportion of Hispanic students, and were, on average, larger. A greater proportion of trained schools were in suburbs while more untrained schools were in rural areas. Inclusion/exclusion criteria Inclusion:must have a completed survey from ≥ 1 administrator, 1 PE teacher, and 3 grade level chairs by September 2014 Exclusion:not reported Number of schools allocated Schools:159 schools Students:not reported Numbers by trial group n (controls baseline) = 80 schools n (controls follow‐up) = 62 schools n (interventions baseline) = 79 schools n (interventions follow‐up) = 71 schools Recruitment Schools:159 schools Students:not reported Recruitment rate Schools:not reported Students:not reported
Interventions	Number of experimental conditions: 2 (1 intervention, 1 control) Policies, practices or programmes targeted by the intervention Training to increase adherence to policy, and PA opportunities offered (including frequency and duration). Implementation strategies EPOC: tailored interventions ‐ Telephone‐based technical support assisting with the engagement of school administrators and staff as well as selection of appropriate resources. EPOC: educational meetings ‐ Training workshop EPOC: educational materials ‐ Shared and modelled the use of low‐ and no‐cost resources including exercise DVDs, PowerPoint files, and an online resource guide containing links to web‐based PA videos, PA curricula, and integrated PA‐academic lessons; and monthly emailed newsletters. Theoretical underpinning ‐ Of the evidence‐based intervention/policy/practice or programme: not reported ‐ Of the implementation strategy: not reported Description of control 80 schools randomly selected to form the untrained comparison arm.
Outcomes	Outcome relating to the implementation of school policies, practices or programmes Mean (standard deviation) minutes of PA offered per week during PE, during recess, during lunch, after school and before school. Data collection method: adapted school PA survey. PE teachers provided data regarding PE, before‐school and after‐school PA opportunities, while grade‐level chairs provided data on recess and in‐class PA breaks. PE teacher and grade‐level chair surveys, which took 10–15 minutes to complete, included subsets of questions from the larger administrator survey which provided data not used in the current study. Validity of measures used: not reported Outcome relating to cost: not reported Outcome relating to adverse consequences: not reported Outcome relating to child diet, PA or weight status: not reported Data collection method: not reported Validity of measures used: not reported
Notes	Research funding: supported, in part, by Robert Wood Johnson Foundation (award number 71061), and support was given also from the Eunice Kennedy Shriver National Institute of Child Health and Human Development of the National Institutes of Health T32 Predoctoral Training Program in Reproductive, Perinatal, and Pediatric Epidemiology (award number T32HD052460). Conflicts of interest: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	Non‐randomised study. Randomly selected a pool of trained and untrained schools for study.
Allocation concealment (selection bias)	High risk	Non‐randomised intervention.
Blinding of participants and personnel (performance bias) Implementation outcome	High risk	Outcome group: nature of the trial precluded blinding.
Blinding of outcome assessment (detection bias) Implementation outcome	Unclear risk	Outcome group: all. Unclear/not stated whether data collectors/researchers were blinded.
Incomplete outcome data (attrition bias) Implementation outcome	Low risk	Outcome group: all. Reasonable level of follow‐up data.
Selective reporting (reporting bias)	Unclear risk	Unclear whether all planned outcomes were reported.
Other bias	Low risk	Did not appear to be at risk of contamination or other biases.
Recruitment to cluster	Unclear risk	Schools were included that provided baseline survey responses. Not enough information to make an assessment.
Baseline imbalance	Low risk	Baseline imbalances accounted for in analysis.
Loss of cluster	High risk	Loss of teachers to follow‐up indicated loss of clusters. 71/79 (90%) PU30 and 62/80 (78%) untrained schools provided follow‐up data between March 2015 and May 2015 from ≥ 1 PE teacher and ≥ 1 grade‐level chair.
Incorrect analysis	Low risk	Various methodology seems to account for baseline imbalances. Covariates included school baseline PA time, school‐level demographics and other school‐level characteristics. The unadjusted model assessed the impact of PU30 on PA opportunities alone using linear regression and generalised estimating equations, which accounted for clustering of multiple respondents per school and of schools within districts.
Compatibility with individually randomised RCTs	Unclear risk	Unclear, no statement regarding this.
Potential confounding	Low risk	Baseline differences accounted for in analysis.
Overall risk of bias assessment	Unclear risk	Most domains were at low or unclear risk of bias.