Miller 1999.
| Methods | RCT. Students randomly allocated to treatment or control using random number generator | |
| Participants | 90 students enrolled, 41 were included in analysis (aged 10‐14 years, mean 12.8 years, SD 1.0, majority aged 13 years; 12 males, 29 females) from grades 6‐9 at public middle schools in Miami Dade County, USA. Students came from 2 schools representing diverse populations | |
| Interventions | Intervention: teen dating violence intervention and prevention (Teen VIP). Group discussed domestic violence and community resources available, and received psycho‐educational sessions exploring the ideas of violence, equality, self esteem, dating violence and anger management. The intervention was delivered over 5 days (duration unclear). 20 students Control: group discussed domestic violence and community resources available but using a person‐centred approach. 21 students The study targeted students with a history of exposure to domestic violence or involvement in an abusive relationship. Delivered by National Council of Jewish Women |
|
| Outcomes | Healthy functioning, equalitarian/non‐controlling/non‐abusive dating relationships as measured by Behaviour Assessment System for Children, Justification of Verbal and Coercive Tactics Scale, Adolescent Coping Orientation for Problem Experiences and Attitudes Toward Relationships and Achievement Scale | |
| Follow‐up | 1 week post‐test | |
| Notes | ‐ | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Random number generator used to allocate students |
| Allocation concealment (selection bias) | Unclear risk | Not stated |
| Incomplete outcome data (attrition bias) All outcomes | High risk | High attrition rate: 55% (50/91). This was due to students not completing the intervention programme (minimum 50% of 5‐week intervention) or not completing the questionnaires. It is unclear whether attrition was evenly distributed within the 2 arms |
| Selective reporting (reporting bias) | Low risk | All outcomes reported fully (number of participants, means and SDs) |
| Blinding of participants and personnel (performance bias) All outcomes | High risk | Blinding not possible. No mention of whether counsellors delivering the intervention received guidance or monitoring |
| Blinding of outcome assessment (detection bias) All outcomes | Unclear risk | Not stated |