Bond 2004.

Methods	Study name: The Gatehouse Project Study design: cluster RCT Intervention arm(s): whole‐school intervention Comparator arm (s): N/S Sample size calculation performed: yes Subgroups prespecified: none Subgroup analyses: NA Start date: 1997 Duration of follow‐up: 4 years Number of follow‐ups: 2 Follow‐up time points: end of school years 1997, 1998, 1999, 2001 (end of school year, 1 year, 2 years, 4 years) ICC (if reported): 0.01 to 0.06
Participants	Number of schools randomised: 26 Number of participants randomised (total and by arm): 12 (intervention), 14 (control) Age (range or mean (SD)) or grade at the start: 13 to 14 years Gender: 52% female (intervention), 54% female (control) Ethnicity: non‐Australian born: 16% intervention, 9% control. Student with first language other than English: 24% intervention, 22% control SES: N/S Inclusion criteria: N/S Exclusion criteria: N/S
Interventions	Randomisation before or after baseline survey: before Duration of the intervention (excluding follow‐up): 24 months Description of the intervention: The Gatehouse Project was a school‐based primary prevention programme, which included both institutional and individual focused components to promote the emotional and behavioural well‐being of young people in secondary schools. Components of the intervention included establishment and support of an adolescent health team, identification of risk and protective factors in the social and learning environment from student surveys, and use of these data for identification of effective strategies to address these issues. A curriculum component was also implemented. Brief description of the theoretical model: health promoting schools framework, ecological approach Description of the comparator: N/S
Outcomes	Primary outcomes: depression, alcohol use, tobacco use, marijuana use, bullying Secondary outcomes: poor availability of attachments, arguments with 3 or more people
Setting	Country: Australia; State: Victoria Setting: secondary schools Focus: universal
Process measures	Process data reported: yes Method (qualitative or quantitative): both Description: field notes, informant interviews, school audit
Statistics	Sample size: 1335 intervention, 1342 control Unit of randomisation: school Unit of analysis: school Method to promote equivalence between groups: stratification Statistical models: ordinal logit models Baseline differences adjustment: N/S Repeated measures methods in analysis: no
Notes	Equity: baseline information on gender, ethnicity, parental separation, non‐Australian born Funding: grants from the Queen’s Trust for Young Australians, Victorian Health Promotion Foundation, National Health and Medical Research Council and Department of Human Services, Victoria, Murdoch Children’s Research Institute, Sydney Myer Fund, Catholic Education Office Randomisation method, e.g. block, stratification, computer: stratification by school administration Clustering accounted for in sample size calculation (if relevant): yes Cluster randomisation methods to account for clustering in analysis: yes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No details provided
Allocation concealment (selection bias)	Unclear risk	No details provided
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Not blinded but intervention was integrated with usual curriculum and administrative processes
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	No details provided
Incomplete outcome data (attrition bias) All outcomes	Low risk	Low levels of attrition and non‐differential rates between study arms
Selective reporting (reporting bias)	Unclear risk	No protocol available
Other bias	Unclear risk	Proportion responding to baseline survey differed between study arms (81% from intervention schools, 68% from comparison schools).