The goal of this pilot study is to determine the feasibility of a randomized trial of the YES-CAN! (Youth Engaged Strategies to Change Adolescent Norms) program to prevent nicotine vaping among adolescents. The program integrates the following evidence-based strategies: youth-adult collaboration; youth-developed narrative videos to convey health messages; peer leaders as change agents; and sustained implementation to change the normative environment. Two middle/high school communities will receive the YES-CAN! intervention. In each school community, a trained teacher will deliver a credit-earning middle or high school class to 25-30 middle or high school students, who will produce 6-8 short videos intended to increase refusal skills; promote stress management and positive coping; change social norms; prevent vaping initiation; and promote vaping cessation among current users. Videos will use a narrative approach and integrate known determinants of vaping. Middle or high school students will collaborate with the teacher and researchers to develop discussion guides and skills-building activities based on best practices for substance use prevention. In 6-8 sessions, middle or high school students will deliver their videos to all students in the associated middle school. A text messaging or other media component will reinforce and boost the effectiveness of the classroom sessions.

Aims for this pilot/feasibility study are:

1. To determine the feasibility and acceptability of implementing the YES-CAN! program
2. To determine the feasibility and acceptability of the research protocols that will be used in a future efficacy trial

This is a pilot/feasibility study of a school-based intervention in which middle and/or high school students develop and deliver a video-narrative-based intervention to middle school students in their school community. To provide evidence of feasibility for the purposes of planning for a large cluster-randomized trial, the investigators will implement the intervention and data collection protocols in two school communities within which students will be followed longitudinally. These school communities are middle schools and/or middle and high school pairs, where each middle school is one of the main schools that feeds into the related high school. The intervention will be delivered over one school year. Baseline and outcome measurements will be conducted via online survey within these schools at baseline, immediately after intervention delivery, and 1-year after baseline.

Within the middle and/or high schools, 25-30 students will participate in the video production and delivery. The school will be encouraged to invite a diverse range of students varying in academic success, gender, socio-economic status, and race/ethnicity to enroll in the credit-earning class. In the fall semester, enrolled students will receive a curriculum related to vaping, health risk communication, and video production. The fall semester will culminate in the completion of 6-8 short narrative videos that are ready for dissemination to middle schools. In the spring semester, students will develop skills and a structure for delivering a vaping prevention program to middle school students. Small groups of participating students will deliver the program in middle school classrooms. Students will be responsible for contacting teachers and scheduling sessions, which will occur mid-February through April, for a total of 6-8 sessions. The middle school sessions will address: health risks of vaping; resistance/refusal skills; healthy ways to cope with stress and anxiety; social skills; and other factors known to be related to nicotine vaping and other forms of substance use in youth. Students will also develop and implement a text message or other type of media campaign for middle school students to boost the effectiveness of the classroom sessions.

Analyses will be conducted using IBM SPSS Statistics 28, SAS Version 9.4 and Mplus Version 8.8. Analyses will be separate for middle school program recipients and the middle/high school program developers. Analyses will address the following domains:

Response rate: An aim is to understand survey response rates and missing data and to develop procedures to address these in the larger trial. Inferential statistics will be used to estimate parameters for key comparisons that assess response rates, student retention/missing data patterns, and random sampling relative to school-level characteristics. The investigators plan to stratify by sex and grade when selecting students for survey participation. This will provide a precise denominator for understanding response rates by these demographics. Response rates will first be assessed descriptively as a proportion of total sampled students within each school, summarized according to age, grade, sex, and race/ethnicity, and compared to overall school-level demographics using t-tests and chi-square or Fisher's exact analyses.

Retention: Given the longitudinal design and student mobility, attrition is anticipated. Students will be invited to complete follow-up surveys even if they missed earlier surveys to allow us to examine missing data patterns. To examine bias generated by attrition over time, the investigators will use independent samples t-tests, chi-square analyses, and regression modeling to compare key baseline variables such as demographic characteristics, vaping behaviors, knowledge, attitudes, perceived norms, and academic aspirations between those who complete 1 and 2 follow-up surveys vs. those lost to follow-up. The investigators will examine attrition in relation to baseline characteristics by regressing baseline constructs (e.g., vaping status, attitudes, norms) on retention status. This regression approach will allow for the testing of interaction terms (e.g., interaction terms of school, grade, and sex with retention status) to determine whether attrition differs by subgroups. Survey response rates will also be compared between middle and high school students and assessed descriptively in terms of time to completion and number of reminders needed. This will provide a way to gauge whether different methods are needed between types of schools, as well as the perceived value of incentives across schools.

Psychometric and distributional evaluation of measures: Baseline demographics and behavioral/attitudinal variables will be summarized using standard descriptive statistics (mean, SD, proportions). Outcome variables are dichotomous, continuous, and ordinal. The investigators will examine the distributional properties of observed data to inform analytic approaches for the larger trial. For psychometric analyses, the investigators will use a split half approach to randomly create EFA and CFA subsamples. Measurement models will be estimated separately for middle school recipient and middle/high school program developer samples, though the investigators will combine the two schools at each level to create the EFA and CFA subsamples (i.e., the middle school and high school EFA/CFA subsamples will include a random selection from each school). The multi-item scales have been derived from prior research, but use of both EFA and CFA will ensure their appropriate use in our local context. Similar constructs (e.g., descriptive norms, injunctive norms) are expected to be related but distinct, though this will be tested empirically using EFA and verified using CFA. EFA modeling will be carried out using maximum likelihood estimation and promax rotation, with final models chosen based on fit (RMSEA), simple structure, and magnitude of factor loadings. The resulting measurement structure from EFA modeling will be estimated in the independent random sample using CFA and interpreted based on model fit (model chi square, RMSEA, and comparative fit index), magnitude and significance of factor loadings, correlations among factors, and internal consistency reliability (Cronbach's alpha).

Subgroup Comparisons: Formal subgroup comparisons (e.g., sex, grade) are proposed to test differences in baseline variables (e.g., vaping use and related attitudes, perceived norms, resistance skills) to better understand how to address unique determinants of vaping among student subgroups in the videos.

School: Analyses will be carried out to compare the two middle schools that receive the program and the two middle/high schools that develop the program on all baseline demographic, behavioral, and attitudinal variables. This is necessary to understand between-school differences for planning a larger trial. Independent samples t-tests and chi-square models, as well as models adjusted for grade and sex using a regression-based framework, may be used. In the larger future efficacy trial, inclusion of additional schools will allow for modeling of school-level fixed and random effects (i.e., three-level model of time nested within students nested within schools).

Sex and Grade: Following the same analytic approach used for school differences, the investigators will test baseline psychosocial variables by sex and grade. The initial approach will be to use bivariate models (t-tests, chi-square) to test sex and grade effects separately, though the investigators will extend this approach to regression modeling of adjusted effects (e.g., sex differences adjusted for grade) as well as potential interaction effects (e.g., sex differences depend on grade).Finally, among middle/high students, the investigators will compare student characteristics measured at baseline between the group of middle/high school students enrolled in the intervention class vs. the broader sample of those students not in the class, both within school and across schools. Student characteristics include sex, age, race/ethnicity, vaping measures (knowledge, attitudes, perceived norms, and behaviors), and positive youth development measures.

Inclusion Criteria:

• Children enrolled in middle school grades 6-8 and high school grades 9-12 (age approximately 10-13 and 14-17); adults (over age 18) involved in educational system

Exclusion Criteria:

• Not able to read and write in English or Spanish

Survey data will be fully de-identified, preserved and shared. Scientific data will be made available in CSV format that can be processed in a variety of statistical software packages. To facilitate data use, the study will use standard processing and documentation protocols adopted by the Inter-university Consortium for Political and Social Research (ICPSR) for data formats and dictionaries as well as for variable names, descriptions, and labels. Variable descriptions include a brief explanation of the questionnaire item content or of the constructed measure. Value labels tie individual numeric response codes to descriptive responses from the questionnaire. Public use study data and associated documentation will be made available to the research community free of charge through the Data Sharing for Demographic Research (DSDR) data repository hosted at ICPSR. Datasets in DSDR will be findable and identifiable through a study digital object identifier (DOI) minted by ICPSR.

Final submission and release of the study data will occur no later than time of an associated publication or end of the performance period, whichever comes first. Study data deposited in DSDR will be available to the research community in perpetuity. All deidentified study data will be made available as public use data to the research community via DSDR.

Users of the public use data must register with ICPSR and agree to the Terms of Use, which are designed to protect study participants by limiting data use to scientific research and aggregate statistical reporting, prohibiting attempts to identify study participants, and requiring immediate reporting of any disclosure of study participant identity. Data users also agree not to share or redistribute any data downloads.