de Villiers 2015.

Study characteristics
Methods	Trial name: HealthKick Study design: cluster‐RCT for implementation outcome Intervention duration: 3 years Length of follow‐up from baseline: 3 years Differences in baseline characteristics: not reported Unit of allocation: schools Unit of analysis: schools
Participants	School type: elementary schools Region: Cape Town, South Africa Demographic/socioeconomic characteristics: historically disadvantaged, low‐income communities from an urban area close to the city of Cape Town and from 2 rural areas outside of Cape Town, South Africa Inclusion/exclusion criteria Inclusion ‐ Whether the principal expressed the need for a health promotion programme to be implemented in the school. ‐ The presence of a shop or vendor selling food items at the school. ‐ Unhealthy diet and lack of PA among learners and teachers selected as a top health priority by the school principal. ‐ The view of the education district level managers of the potential of schools to effect changes, subjectively taking into account functionality (i.e. functional school‐based support team; school management team), ethos (co‐operation, will, inclination) and viability of school (e.g. results/performance of schools). ‐ Distance from the research office (< 105 minutes' drive). Exclusion ‐ School size (schools with < 50 grade 4 learners were excluded). Number of schools allocated: 16 Numbers by trial group n (controls baseline) = 8 n (controls follow‐up) = 8 n (interventions baseline) = 8 n (interventions follow‐up) = 8 Recruitment Schools:the HealthKick study comprised 16 eligible schools selected from the representative sample of 100 primary schools surveyed in 2 conveniently selected educations districts (1 urban and 1 rural) in the Western Cape Province of South Africa during the formative phase of the study. Students: not reported Recruitment rate: not clear
Interventions	Number of experimental conditions: 2 (1 intervention, 1 control) Policies, practices or programmes targeted by the intervention The HealthKick programme which was an adaptation of the National School Health Policy and Implementation Guidelines (since replaced by the Integrated School Health Policy (ISHP)). The specific objectives were to: ‐ Promote healthy eating habits. ‐ To develop an environment within the school and community that promotes and facilitates these objectives through an action planning process (APP). Implementation strategies EPOC:local opinion leaders ‐ To facilitate and drive the formal implementation of the APP, a champion (teacher) was identified at each school and they were encouraged to liaise with the project team whenever they required assistance. EPOC:educational materials ‐ An "educator’s manual" which contained an APP guide, a booklet for each action area containing guidelines for prioritising action as well as strategies to address identified priorities; the South African food‐based dietary guidelines; a poster listing the behaviour outcomes desired for the children; a poster for listing planned actions; and in 2011 a healthy lifestyle guide for teachers was included. ‐ A resource box with printed materials relating to a healthy lifestyle and its role in the school curriculum. ‐ A curriculum support manual integrating the HK goals with the existing Life Orientation curriculum, developed by an expert in a format familiar to educators. EPOC:educational outreach visits ‐ Optional intervention support was offered to the intervention schools in all 4 action areas during the 3 years of the intervention. The support took the form of structured activities by the research team to broaden the staff's knowledge and skills around actions to support a healthy lifestyle. Furthermore, the research team kept in regular contact with the schools who were encouraged to call for assistance/support from the research team at any time. EPOC:education meetings ‐ Implementation and nutrition training workshop. Theoretical underpinning: Social Ecological model Description of control: principals at schools in the control arm received a booklet with "tips" for healthy schools and a guide to resources that could be accessed to assist in creating a healthier school environment. No further engagement took place between the research team and these schools except for the annual learner and environmental survey (EPOC: educational material).
Outcomes	Outcome relating to the implementation of school policies, practices or programmes ‐ School with shops selling fruit salad ‐ Vegetable gardens at schools ‐ Schools having nutrition‐related policies Data collection method ‐ Data from the situational analysis were used as baseline information and an adapted version of the principal questionnaire and observational schedule used during the formative assessment which was completed annually at all 16 schools. ‐ Principals and school staff involved with the school nutrition programme, tuck shops and vegetable gardens were interviewed using semi‐structured questionnaires. The interviews were conducted by members of the research team. As above and project officers acted as both implementers and outcome assessors. Interviews were recorded and transcribed, and the data managed with ATLAS.ti Qualitative Data Analysis. Initial data analysis involved coding the focus group data as group interviews (i.e. similar responses coded only once per group). Validity of measures used: not reported/self‐report methods Outcome relating to cost: not reported Outcome relating to adverse consequences: not reported Outcome relating to child diet, PA or weight status: student level dietary intake Data collection method: 24‐hour recall Validity of measures used: not reported
Notes	Notes: PA was also a goal of the HealthKick Program; however, an implementation outcome was unavailable. Consequently, this trial is reported as a nutrition trial, with corresponding implementation strategies, and behavioural outcomes extracted. Research funding: World Diabetes Foundation. Conflicts of interest: authors reported no conflicts of interest.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Random sequence generated by drawing the names of schools typed on folded white paper of exactly the same shape and size from a container.
Allocation concealment (selection bias)	Unclear risk	Although the authors' state that the allocation sequence was decided on by the project co‐ordinator before the selection took place, there is no description of whether this was concealed.
Blinding of participants and personnel (performance bias) Implementation outcome	High risk	Outcome group: there was no mention that participants and personnel were blinded. Team members served as both implementers and researchers due to limited resources and, therefore, a high risk of performance bias.
Blinding of outcome assessment (detection bias) Implementation outcome	High risk	Outcome group: self‐reported records were kept of all activities/events planned by the schools and the numbers that were carried out. Observation was not undertaken by an independent observer blind to group allocation.
Incomplete outcome data (attrition bias) Implementation outcome	Low risk	Outcome group: no schools dropped out over the 3 years.
Selective reporting (reporting bias)	Low risk	There were no unreported process evaluation outcomes according to those planned in the published protocol.
Other bias	Low risk	Appeared free from other bias.
Overall risk of bias assessment	Low risk	Most domains were at low risk of bias.