The effect of peer education based on adolescent health education on the resilience of children and adolescents: A cluster randomized controlled trial

Considering that this study was a cluster RCT, the sample size should be increased through design effect (DE) to ensure the efficiency of the test [41, 42]. When the average cluster size (m) is 40 and the intra-cluster correlation coefficient (ICC) is 0.05, the calculated DE is 2.95. The DE was multiplied by the sample size without clustering effect (n), resulting in a sample size of 811 students in each group and a cluster number of 20 in each group. Study design, participants, and processes A cluster RCT was conducted using 4 schools (2 primary schools, 2 middle schools) in the Qijiang district of Chongqing. The inclusion criteria were four comparable schools in geo- graphical distribution, educational philosophy and cultural level. Stratified by primary and junior high schools, 4 schools were randomly divided into the intervention group with peer education intervention and the control group without any intervention (blank control group). Finally, the intervention group and the control group each included a primary school and a middle school. All participants in the intervention group received peer education intervention, and all participants in the control group received no intervention. The survey was expected to take a year and a half. Considering that students in grade 6 and grade 9 will be lost to follow up due to entering a different school, they were not included in this study. The inclusion criteria were students in grades 4–5 in primary school and grades 7–8 in middle school who were able

to complete the questionnaire independently. A total of 1,678 students in selected schools were eligible to participate in this study, but 8 individuals were excluded due to intellectual disabili- ties and being illiterate (exclusion criteria). Therefore, 758 and 912 students were included in the intervention group and control group, respectively. At baseline in December 2017, with the help of school teachers, the researchers led 758 stu- dents (19 classes) in the intervention group and 912 students (24 classes) in the control group to complete the questionnaire, which mainly included the participants’ general demographic characteristics (gender, family economic status, parenting style, etc.) and psychological resil- ience related questions. Most participants completed the questionnaire within 40 minutes, with no obvious logical errors or missing items. In April 2,018, peer educator training was con- ducted by experts in the health of children and adolescents. The last follow-up was conducted in May 2019, one year after the intervention, 732 students (26 missings) in the intervention group and 881 students (31 missings) in the control group were investigated with the same questionnaire as the baseline. The reason for missing data was changing schools and home- sickness. Finally, the data analysis included 1,613 students, as shown in Fig 1. The study was approved by the Biomedical Ethics Committee of Peking University (IRB 00001052–13,034) and the ethical committee of Chongqing Medical University, and written informed consent

was obtained from students and their parents before investigation in the research. The trial was pre-registered in ClinicalTrials.gov before recruitment started (identifier NCT02343588). Peer education involving adolescent health education In the intervention group schools, negotiating with head teachers in every class, 4 excellent, responsible and well-communicated students, including 2 boys and 2 girls, were selected as peer educators who served on the class committee, actively participated in extracurricular activities and had activity organization experiences. The training session for peer education activities was divided into primary school group and middle school group. The training teacher of each group was composed of 1 professor from Chongqing Medical University and 3 graduate students from the research group. The training forms were intensive teaching, group discussion and knowledge competition. The training content was divided into four parts: ado- lescent physical health knowledge, adolescent mental health knowledge, healthy behavior and PLOS ONE The effect of peer education based on adolescent health education on resilience of children and adolescents PLOS ONE | https://doi.org/10.1371/journal.pone.0263012 February 2, 2022 4 / 18