Background: With rapid extensions to education in many emerging market economies, secondary schools have the potential to be an important platform for health promotion and prevention. A ‘health promoting school’ approach has become an increasingly popular framework internationally with which to address the health needs of school communities. A growing evidence base indicates that, if applied successfully, a health promoting school framework can lead to improvements in both health and educational outcomes. Methods: A cluster randomized controlled trial was conducted to assess the effectiveness and cost-effectiveness of a whole-school health promotion intervention (Mitra, meaning a friend) in Bihar, India. Two intervention delivery models using a lay school counsellor (the SEHER Mitra (SM) arm) or a teacher (Teacher as SM (TSM) arm), were compared against the standard Adolescent Health Education Program, in 74 government-run secondary schools in Bihar, India. All grade IX students were assessed at the start and end of the academic year (i.e. June 2015-March 2016; 8 months apart). The primary outcome was school climate, (the perceived ethos or atmosphere of the school) measured with the Beyond Blue School Climate Questionnaire (BBSCQ). Secondary outcomes included self-reported bullying, violence, depressive symptoms, attitudes towards gender equity, and knowledge of reproductive and sexual health. A qualitative study was nested in the trial to evaluate the reasons why the difference in the delivery agents may have yielded different results for the two arms when compared with the control. For this study, data were collected through one on one interviews and focus group discussions with key stakeholders. Qualitative data were analysed thematically using Framework Analysis. Findings: The baseline survey was conducted in July 2015, and included 13,035 participants (SM: 4524; TSM: 4046; control: 4465; 52.5% boys). The endpoint survey included 14,414 participants (SM: 5316; TSM: 4475; control: 4623; 52.9% boys). School climate scores were similar by arm at baseline, but schools receiving the SM-delivered intervention had significantly larger gains in school climate scores at endpoint (mean BBSCQ=24.13) compared with those receiving the TSM-delivered intervention (mean BBSCQ= 17.16; adjusted mean difference (aMD)=7.91, 95%CI:6.34, 9.47; effect size (ES)=1.98 95%CI:1.93, 2.03) or the control intervention (mean BBSCQ=17.75; aMD=7.44, 95%CI:5.88, 8.99; ES=1.86 95%CI:1.81,1.91). School climate scores were similar in the TSM and control arms at the study endpoint (aMD=-0.47, 95%CI:-2.03, 1.08; ES=-0.12 95%CI:-0.17,-0.07). Schools with the SM-delivered intervention showed significant improvements in all secondary outcomes compared with both the TSM and control arms. From the qualitative sub-study, a number of fundamental implementation factors were identified as not being sufficiently well developed to facilitate the effective implementation of the SEHER in the TSM arm relative to SM arm. These included: a lack of a shared understanding of the SEHER amongst all key stakeholders; reluctance of principals to be the leader of the programme implementation in schools; poorly developed forms of collaboration within school; and the lack of a properly functioning School Health Promotion Committee; and overburdened TSMs with academic and non-academic responsibilities. Conclusions: The multicomponent whole-school health promotion intervention had major beneficial effects on school climate and related outcomes when it was delivered by lay school counsellors, but no consistent effects when delivered by teachers compared with the standard Government program.