BACKGROUND: Young people's engagement in gambling can be linked to gambling-related harm. There is a lack of independently funded and evidence-based school-based interventions that seek to prevent and reduce the harms associated with gambling.

OBJECTIVES: To conduct a pilot cluster randomised controlled trial of a gambling prevention intervention (Preventing Gambling Related Harm in Adolescents) among young people aged 13-15 years to determine the utility of conducting a phase III randomised controlled trial assessing effectiveness and cost-effectiveness.

DESIGN AND METHODS: Two-arm, pilot cluster randomised controlled trial with an embedded process evaluation, health economic scoping study and social network analysis.

SETTING: Six state schools across Scotland: four intervention and two control.

PARTICIPANTS: Students aged 13-15 years (baseline: intervention  = 762 students, control  = 352 students. Follow-up: intervention  = 598 students, control  = 295 students).

INTERVENTION: Preventing Gambling Related Harm in Adolescents is a peer-led, social network intervention to prevent adolescent gambling and reduce gambling-related harm. Students, nominated by peers in their year group, attended a 2-day training workshop outside of school, which was delivered by trained youth workers. After completing the 2-day training workshop, students were then assigned the role of 'peer supporters'. Peer supporters then attended three in-school follow-up sessions, with the same youth workers, to refresh the learning they had received during the workshop. After the workshop and during the follow-up sessions, peer supporters were encouraged to have conversations about their learning on the topic of gambling and gambling-related harm with their friends and family. Social network maps were created by peer supporters to: (1) help them identify people to speak to and (2) record to who they spoke to about gambling and gambling-related harm.

MAIN OUTCOME MEASURES: Progression to a full-scale Phase III cluster randomised control trial, using pre-set progression criteria. Progression criteria were: (1) successful recruitment of six schools; (2) five schools remain in the pilot study; (3) the intervention being delivered with 80% fidelity to the manual; (4) the process evaluation indicates the intervention is acceptable to students and staff and (5) 70% of students complete the student questionnaire at baseline and follow-up.

RESULTS: All five progression criteria were met. All schools were recruited and retained in the study and there were low levels of missing data on outcomes. The process evaluation indicated that Preventing Gambling Related Harm in Adolescents was acceptable to all stakeholders and was delivered with fidelity to the delivery manual. Some minor refinements to the intervention and trial methods would aid student perception of the real-world effects of gambling. Indicative costs to deliver two cycles of Preventing Gambling Related Harm in Adolescents in one school were £8313.00, with a mean cost per student of £28.08.

LIMITATIONS: A limitation is student self-reported gambling behaviour. This relied on their interpretation of what gambling is, which varied.

CONCLUSIONS: Delivery of the Preventing Gambling Related Harm in Adolescents intervention was both feasible and acceptable, indicating that the study could progress to a full-scale Phase III randomised controlled trial, with minor amendments to the intervention and trial processes.

FUTURE WORK: Preventing Gambling Related Harm in Adolescents should progress to a full-scale randomised controlled trial.