Conduct, anxiety and depressive disorders account for over 75% of the adolescent mental health burden globally. The current protocol will test a low-intensity problem-solving intervention to improve mental health symptoms and recovery rates for school-going adolescents in India. The protocol also tests the effects of a classroom-based sensitization intervention on the demand for counselling services in an embedded recruitment trial. Methods We will conduct a two-arm individually randomized

more »

... olled trial in six Government-run secondary schools in New Delhi. The targeted sample is 240 adolescents in grades 9-12 with persistent, elevated mental health symptoms and associated impact. Participants will receive either a brief problemsolving intervention delivered over 3 weeks by lay counsellors (intervention), or enhanced usual care comprised of problem-solving booklets (control). Self-reported adolescent mental health symptoms and idiographic problems will be assessed at 6 weeks (co-primary outcomes) and again at 12 weeks post-randomization. In addition, adolescent-reported impact of mental health difficulties, perceived stress, mental wellbeing and clinical remission, as well as parent-reported adolescent mental health symptoms and impact scores, will be assessed at 6 and 12 weeks post-randomization. Parallel process evaluation, including estimations of the costs of delivering the interventions, will be conducted. An embedded recruitment trial will apply a stepped-wedge, cluster (class)-randomized controlled design in 70 classes across the six schools. This will evaluate the added impact of a classroom-based sensitization intervention over school-level recruitment sensitization activities on the primary outcome of referral rate (number of referred students as a proportion of the total sampling frame in each arm). Secondary outcomes will be the proportion of referrals eligible to participate in the host trial, proportion of self-generated referrals, and severity and pattern of symptoms across the arms. Power calculations were undertaken separately for each trial. A detailed statistical analysis plan will be developed separately for each trial prior to unbinding. Discussion