Gresham 2001.

Methods	RCT
Participants	Participants were elementary children in grades 1, 2 and 3 and their parents, from 2 urban areas in San Diego County (US). Number of participants: 1126 students in the intervention group and 851 students in the control group.
Interventions	Intervention: Think First For Kids programme. Children had 6 contacts, each lasting 35‐40 minutes, over a 6‐week period. There were 6 modules involving a range of video, a spinal cord speaker, hands on interactive teaching, maths, visual reinforcement and discussion. The intervention was delivered by teachers, district nurse, life skills educators as well as an external speaker/brain and spinal cord patient as well as input from parents in the form of parental support with a homework component. Control: unclear.
Outcomes	Self‐reported behaviour and safety skills and safety knowledge assessed using forced choice format questionnaires, 10 days following intervention.
Injury mechanisms	Brain and spinal cord injuries: violence and weapons safety; playground, recreation and sports safety; cycle safety; water safety; vehicle safety.
Notes	Intervention and control schools were matched on district, socioeconomic status, school‐defined reading scores and race/ethnic composition.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No information provided about the randomisation process.
Allocation concealment (selection bias)	Unclear risk	No information provided about the allocation process to determine if low or high risk, although children were matched on district, socioeconomic status, reading scores and ethnicity in the school.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not described. Participants were likely to know that they had received the intervention.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Insufficient information provided about the blinding process.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Post‐test results could not be matched for 20% of students, though the paper did not report whether these were control or intervention students. Intention‐to‐treat analysis not mentioned.
Selective reporting (reporting bias)	High risk	The authors did not separate out behaviour and knowledge outcomes and did not report the module scores.
Other bias	Low risk	Did not appear to be at risk of other bias.