Egan 2018.

Study characteristics
Methods	Trial name: PACES Study design: non‐RCT with pre–post control group Intervention duration: 3 consecutive academic semesters (about 12 months) Length of follow‐up from baseline: ‐ Baseline: data collected in autumn 2014 ‐ First intervention semester: spring 2015. 1‐on‐1, semi‐structured interview with each teacher participating in the intervention (9 teachers) end of spring 2015 Differences in baseline characteristics: not reported for baseline information Unit of allocation: schools Unit of analysis ‐ Implementation outcomes: school ‐ Behavioural/health outcomes: none
Participants	School type: elementary schools Region: 4 local elementary schools in 2 school districts from a greater metropolitan area in 1 south‐eastern state. Demographics/socioeconomic characteristics ‐ Schools A and B were magnet schools (companion campuses, i.e. 2 separate school campuses that share administrators and curricula focus) from 1 school district and schools C and D were public schools from a different school district. ‐ Participating teachers (10 female, 2 male) aged 23–54 (mean 33.2, SD 10.3 years) and self‐identified as 11 non‐Hispanic white teachers and 1 African American teacher. The teachers' years of teaching experience ranged from 1 to 33 years (mean 10.2, SD 11.0 years). Inclusion/exclusion criteria Inclusion Participants were selected from 4 local elementary schools in 2 school districts from a greater metropolitan area in 1 southeastern state. School selection was based upon location (close proximity to the researchers' university), access (receptive to participating in research) and stated priorities (3 of the schools identified school health as a priority in their strategic plan). Exclusion For consistency teachers that taught outside grades 1–3 were excluded, as grade 3 was the highest grade in school B. Number of schools allocated Schools:4 (1 control, 3 intervention) Students:161 (within 4 schools) Teachers: 12 Numbers by trial group n (controls baseline): 1 school n (controls follow‐up): 1 school n (interventions baseline): 3 schools n (interventions follow‐up): 3 schools, 9 teachers Recruitment Schools:school selection was based upon location (close proximity to the researchers' university), access (receptive to participating in research) and stated priorities (3 schools identified school health as a priority in their strategic plan). The first 3 schools that accepted the invitation were assigned to participate (referred to from this point onward as schools A, B and C) to receive the PACES intervention. The fourth school (school D) agreed to participate as a waitlist control. Students:all 181 students in the participating teachers’ classrooms were eligible to participate, and 161 participated (48.45% female). Recruitment rate Schools:not reported Students:89%
Interventions	Number of experimental conditions: 3 experimental conditions: 1. 3 experimental conditions + 1 waitlist control; 2. school A (treatment level 1): low implementer; and 3. school B and C (treatment levels 2/3): high implementer. Policies, practices or programmes targeted by the intervention: PA MI Implementation strategies EPOC: tailored interventions ‐ A member of the research team meeting with each teacher individually to share baseline PA and MI results, identify current MI strengths and areas for improvement, collaboratively set personalised MI goals, and consider suitable resources, including those posted on the community of practice (schools B and C). EPOC: educational outreach visits ‐ A member of the research team meeting with each teacher individually to share baseline PA and MI results, identify current MI strengths and areas for improvement, collaboratively set personalised MI goals, and consider suitable resources, including those posted on the community of practice (schools B and C). ‐ Additionally, preservice classroom teachers delivering physically active lessons and movement breaks in intervention classrooms (with the teacher present) ≥ 3 times (on data collection days) during PACES programme implementation (schools C). EPOC: educational meetings ‐ A member of the research team orienting each teacher to a virtual professional learning community, Move for Thought (moveforthought.ning.com). Also, the website included MI materials, videos, links and a blog for members to ask questions, share ideas and connect with fellow members (Schools A, B and C). EPOC: audit and feedback ‐ Strengths and areas for improvement, collaboratively set personalised MI goals (school B and C only). Theoretical underpinning ‐ Of the evidence‐based intervention/policy/practice or programme: not reported ‐ Of the implementation strategy: 3 partnership approaches (communities of practice, community‐based participatory research and service learning) were employed based on the partnership model of Webster and colleagues (Webster 2015). Description of control: waitlist control
Outcomes	Outcome relating to the implementation of school policies, practices or programmes SOSMART video recordings, and implementation score based off recordings. Data collection method: coding of video recordings by research assistants using System for Observing Student Movement in Academic Routines and Transitions (SOSMART). The implementation score was calculated for each classroom at baseline and outcome to examine the effect of PACES on the extent of MI during the first year of the 2‐year intervention. The implementation score was calculated using a 2‐step process. First, the percentage of scans for 7 of the SOSMART variables was calculated at both baseline and outcome separately to determine the percent of time each variable was observed. Second, the percentage of time each variable was observed for baseline and for outcome for each teacher to create a baseline and outcome implementation score was totalled. After creating the implementation score at baseline and outcome for each teacher, the mean implementation score (baseline and outcome) for each group (treatment level 1, treatment level 2, treatment level 3, and control) was calculated. Validity of measures used ‐ The authors provided the reference of Russ and colleagues for validity and reliability (Russ 2016). Intraobserver reliability resulted in 97.5% agreement and exceeded 80% on all variables. Additionally, construct validity was supported for 8/11 MI variables. 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: this research did not receive any specific grant from funding agencies in the public, commercial or not‐for‐profit sectors. Conflicts of interest: authors declared there was no conflict of interest.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	Non‐randomised study.
Allocation concealment (selection bias)	High risk	No blinding of allocation.
Blinding of participants and personnel (performance bias) Implementation outcome	High risk	Outcome group: all. The nature of the intervention precluded blinding. Community‐based participatory research involved a member of the research team meeting with each teacher individually to share baseline PA and MI results, identify current MI strengths and areas for improvement, collaboratively set personalised MI goals, and consider suitable resources, including those posted on the community of practice.
Blinding of outcome assessment (detection bias) Implementation outcome	High risk	Outcome group: objective data captured by trained data collectors. Teachers would have been aware of the outcome of interest.
Incomplete outcome data (attrition bias) Implementation outcome	Low risk	Outcome group: SOSMART implementation outcomes appear to be reported for all included schools.
Selective reporting (reporting bias)	Unclear risk	No protocol available for this study.
Other bias	Low risk	Did not appear to be at risk of contamination or other biases.
Potential confounding	Unclear risk	Unclear if confounding had occurred. Schools A and B were magnet schools (companion campuses, i.e. 2 separate school campuses that share administrators and curricula focus) from 1 school district.
Overall risk of bias assessment	Unclear risk	Most domains were at high and low risk of bias.