Sevinc 2011.

Study characteristics
Methods	Study design: cluster‐RCT Intervention period: 8 months Follow‐up period (post‐intervention): nil Differences in baseline characteristics: reported Reliable outcomes: reported Protection against contamination: NR Unit of allocation: group (2 schools in each group) Unit of analysis: individual
Participants	N (controls baseline) = 2926 N (controls follow‐up) = 2654 N (intervention 1 baseline) = 1932 N (intervention 1 follow‐up) = 1897 N (intervention 2 baseline) = 1989 N (intervention 2 follow‐up) = 1815 Setting (and number by study group): schools, intervention 1 (N = 2), intervention 2 (N = 2), control (N = 2). Each group comprised of one low‐SES and one high‐SES school Recruitment: all schools involved in a half‐day education system; randomly sampled Geographic region: Denizli, Turkey Percentage of eligible population enrolled: 98.9% participants Mean age: (intervention + control): 7‐13 years Sex: intervention 1: 50.3% female; intervention 2: 49.8% female; control: 49.1% female
Interventions	Aim: to determine the efficiency of applying both PA and healthy nutrition programmes and only a healthy nutrition programme for preventing obesity in primary school students (aged 7‐13) in Denizli, to determine the relationship of this efficiency with the possible variables, and to construct an obesity control programme aimed at the students. Associated study name: Get into motion for health Intervention description Intervention group 1: PA combined with healthy nutrition education programme. Initial weekly PE classes were 2 h in total and increased to 3 h on different days of the week. During these sessions standard PA and sport programmes, specific to the age range of the children were applied. Does not state who delivered the PA programme. Intervention group 2: healthy nutrition education programme only. Education on the importance of healthy nutrition and the methods of preventing obesity were given to the students, their parents, and the teachers. Boxed milk was also distributed to the students for them to drink during meal time. Moreover, to supply healthy eating options for the students in the school canteens, water, freshly squeezed fruit juice, buttermilk, milk and seasonal fruits were sold. Who delivered/training: personnel of the Health Training Division of the City Health Administration/teachers Diet and PA combination intervention vs dietary intervention only vs control PA intervention vs control (health education)
Outcomes	Outcome measures Primary outcome: BMI Secondary outcomes: NR Process evaluation: NR
Implementation‐related factors	Theoretical basis: NR Resources for intervention implementation: reported Who delivered the intervention: reported PROGRESS categories assessed at baseline: gender; parent: SES (income) PROGRESS categories analysed at outcome: gender; parent: SES (income) Outcomes relating to harms/unintended effects: NR Intervention included strategies to address diversity or disadvantage: targeted both low‐ and high‐SES schools Economic evaluation: NR
Notes	Funding: NR In one of the schools in intervention group 1, some of the equipment required for the PA programme could not be obtained, and an insufficiency in directing the school canteens to supply healthy food instead of fast food and carbonated drinks might be counted among the limitations of the study.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	From the low‐ and high‐value SES regions, 3 schools each (a total of 6 schools) were selected by using a simple random sampling method. These schools were randomly divided into 3 groups consisting of 1 school from the low‐ and 1 school from the high‐SES level. Of these groups, 2 were again randomly selected as intervention groups and the remaining 1 as the control group
Allocation concealment (selection bias)	Unclear risk	NR
Blinding (performance bias and detection bias) All outcomes	Unclear risk	NR
Incomplete outcome data (attrition bias) All outcomes	Low risk	94% retention of those ‘reached’
Selective reporting (reporting bias)	Unclear risk	Protocol or trial register not found
Other bias	Low risk	No additional threats to validity
Other bias‐ timing of recruitment of clusters	Low risk	Figures suggest recruitment happened prior to randomisation