Macgowan 1997.

Methods	Cluster‐RCT. Classes matched and randomised to intervention or control
Participants	802 middle school students enrolled, 440 were included in analysis (aged 11‐16 years, mean 12.6 years, SD 1.1; 193 males, 247 females) from a school in Miami, USA (62 students with learning difficulties and 300 students who did not complete minimum of 19/21 questions on survey were excluded)
Interventions	Intervention: discussion on recognising dating violence, characteristics of strong and weak relationships, and problem‐solving and communication skills. 5 sessions lasting 1 hour each over the course of 5 days. Delivered by facilitators who had received 3 hours of training by author of curriculum. Facilitators also had a checklist of each day's materials to be covered to hand. 241 students Control: wait list control. 199 students
Outcomes	Knowledge about dating violence, as measured by a questionnaire (developed by authors) based on the curriculum
Follow‐up	2 days post‐test
Notes	‐
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not stated
Allocation concealment (selection bias)	Unclear risk	Not stated
Incomplete outcome data (attrition bias) All outcomes	High risk	After excluding students with learning difficulties (n = 62) and those who did not complete minimum of 19/21 questions on survey (n = 300), only 440 of 802 eligible students included in analysis (45% attrition). This is high, especially for immediate follow‐up. Also may reduce generalisability of results
Selective reporting (reporting bias)	Low risk	Outcomes reported fully (means, numbers participating and summary statistics)
Blinding of participants and personnel (performance bias) All outcomes	High risk	Blinding not possible. Although facilitators were trained to minimise differences in delivery of intervention, no mention of how this was monitored
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Not stated