Buckley 2010.

Methods	CBA
Participants	Children in year 9 of high school (95% were aged 13 to 14 years) from schools in an urban deprived area of Queensland, Australia. Number of participants: 360 students in the intervention group (97% of all eligible students) and 180 students (45% of all eligible students) in the control group.
Interventions	Intervention: SPIY programme. teacher training, a teacher's manual and student workbook for 8 lessons carried out in the school. Each lesson lasted 50 minutes, and included presentations of risk‐taking and injury scenarios, introduction to first aid and cognitive behavioural activities to prevent the risk‐taking behaviour, including protecting friends. Control: usual curriculum. The SPIY programme was made available after the study.
Outcomes	Self‐reported risk behaviour measured using the Australian Self‐Report Delinquency Scale, 2 weeks postintervention.
Injury mechanisms	Poisoning; road traffic accidents: cars, cycles, motorbike, pedestrian.
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Allocation to intervention/control (selection bias) (for non‐RCT and CBA studies)	High risk	Schools selected which group they wanted to be in (intervention vs control).
Blinding of participants and personnel (performance bias) All outcomes	High risk	Parents were sent information that an evaluation of an injury prevention programme was taking place.
Blinding of outcome assessment (detection bias) All outcomes	High risk	Self‐reported outcomes, high risk of allocations being detected.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Attrition was > 20% as the analyses were only based on children with complete before‐and‐after data.
Selective reporting (reporting bias)	Unclear risk	The authors only reported data for children with before‐and‐after data.
Other bias	High risk	The study did not take into account clustering effects. Only 45% of control group children were included (197 children) compared to 97% of the intervention group, indicating a differential selection bias.
Risk of bias due to confounding (for non‐RCTs and CBA studies)	Unclear risk	Although groups had similar sociodemographic factors, because schools self‐selected which groups they were in there could have been some residual confounding.