Wrap It Up Rigorous Evaluation

Background Unintended pregnancy rates among teenagers are higher in rural counties than urban or suburban counties. Wrap It Up (WIU) is a sexual health education (SHE) "booster" program for older high school students attending school in rural communities that was designed to help address this issue. Though there is preliminary evidence of the program's effectiveness, it has not yet been rigorously evaluated. With this study, ETR will rigorously evaluate the effectiveness of WIU with a randomized controlled trial.

WIU is designed to equip 11th and 12th grade students with the knowledge and skills necessary to make informed sexual health decisions as they move from high school into adulthood, in order to prevent behavioral risk factors that contribute to unintended pregnancy and sexually transmitted infections (STIs). This program was created by Health Connected, a non-profit organization with over 30 years' experience in creating, evaluating, training on, and disseminating comprehensive sexual health education to young people and their families throughout California. Health Connected is ETR's implementation partner in this study.

Most students this age will have received prior SHE. WIU is a 4-to-5-hour "booster" intervention, delivered during late high school, intended to enhance the effectiveness of previous SHE. The WIU program will be administered to groups in school settings, such as science classes or student advisory sessions. We will use trained sexuality education professionals supported by Health Connected to deliver the intervention.

Research Questions

The principal purpose of WIU is to decrease behaviors that contribute to unintended pregnancy and STIs in students and increase behaviors that support prevention and wellbeing. The study's primary research question is:

What is the impact at 6-months post baseline of WIU relative to a "business as usual" comparison condition on the proportion of youth who self-report they either did not have vaginal or anal sex, or used condoms or effective birth control when they had vaginal or anal sex, in the past 90 days?

The study's secondary research questions are:

• What is the impact at 6-months post baseline of WIU relative to a "business as usual" comparison condition on the proportion of youth who self-report improved self-efficacy for self-advocating with healthcare providers, peers, and partners about their sexual health?
• What is the impact at 6-months post-baseline of WIU relative to a "business as usual" comparison condition on the proportion of youth who self-report improved self-efficacy to find and access sexual health services?
• What is the impact at 12-months post-baseline of WIU relative to a "business as usual" comparison condition on the proportion of youth who self-report they either did not have vaginal or anal sex, or used condoms or effective birth control when they had vaginal or anal sex, in the past 90 days?

Study Design This evaluation is a cluster randomized controlled trial (RCT) using a "school by year" approach. The investigators' implementation partners, Health Connected, will engage 16 California schools serving rural students to participate in the study for 2 years. Eight of the participating schools will be randomly assigned to the "intervention-first" group and eight to the "comparison-first" group. Each school will serve as its own control. Thus, half of the 16 participating schools will receive WIU in their first year. The other half will be in the comparison-first group and will receive the WIU intervention in their second year. During the comparison group year, schools will administer their normal SHE. Each school has an equal chance of being in either the intervention-first or comparison-first group.

Recruitment and randomization will occur during year 1 for the first cohort of 8 schools and during project year 2 for the second cohort of 8 schools; implementation will occur over years 2 and 3 for the first cohort of schools and over years 3 and 4 for the second cohort.

Upon recruitment and before randomization, schools will commit to a specific context (e.g., grade level, class setting) for implementing WIU and agree to use that same context for enrollment, baseline data collection, and study implementation across both years of the study regardless of which condition they are randomized at initial enrollment.

Investigators intend to enroll up to 1,600 students (roughly 50 students per school per year). Between 2024 and 2027, the study will survey participating students three times: at baseline (before implementing the WIU program), 6 months post-baseline, and 12 months post-baseline. Students who provide complete outcome data on the baseline survey will be considered part of the study sample that underwent random assignment and used to calculate attrition.

Data Collection The study team plans to administer the baseline survey in-person at the school sites on tablets provided by ETR. Audio-assistance will be available for student participants, if needed. Data will be collected approximately 1-4 weeks prior to implementation, with comparable times for schools in the comparison group. The project team will offer up to three make-up sessions for the baseline survey. In the baseline survey, the project team will assess demographic characteristics (e.g., sex, race, educational status); potential moderators (e.g., age, race/ethnicity, relationship status); sexual behaviors (e.g., abstinence, having sex with/without condom or contraception use, number of partners); and potential mediators (contraception knowledge, contraception attitudes, pregnancy attitudes) that will be used to establish baseline equivalence of the analytic sample. The timing of the implementation may vary across schools; for example, at some participating schools, baseline data collection may occur in the fall, while for others, it may occur in the spring. However, within a specific school, the timing will be consistent across the two years of study implementation.

Survey data will be collected in the same manner for both treatment and control groups by local data collectors hired and trained by ETR. If youth are still enrolled in the same school, the 6-month and 12-month post-program surveys may be collected onsite in the same manner as the baseline survey, or collected online. For youth who are no longer enrolled in the same school, they will receive an email or text message with a unique link so the survey can be accessed and completed on a device of their choosing. ETR will use its field-tested, digital tracking mechanism to best support locating youth for follow-up surveys.

Analysis Plan Investigators plan to recruit 16 schools across two years and randomize each to an initial experimental condition. For the second year of a school's implementation, the school will serve in the experimental condition opposite from its initial assignment, so that each school serves as its own control; this results in a total of 32 school-by-year clusters. Each school must enroll grade 11 and grade 12 students to be eligible for participation.

The study intends to enroll up to 1,600 students in grades 11-12 across the 16 recruited schools. We anticipate an analytic sample size of 1,408 students (88% response rate) at the 6-month follow-up time point, and an analytic sample size of 1,280 students (80% response rate) at the 12-month follow-up time point. The project does not anticipate any cluster-level attrition during the study.

The study estimates a minimum detectable impact of 8.25 percentage points, relative to a 20 percent prevalence rate (or equivalently, a minimum detectable effect size of 0.17) for the primary outcome of sexual abstinence/use of contraception, measured at the 6-month follow-up. This is based on the following assumptions: alpha = 0.05, power (Beta) or 0.80, an expectation that the covariates will explain 20 percent of individual level variation and 50 percent of cluster level variation in the model, and an intraclass correlation (ICC) of 0.01. The calculation's assumed randomized sample size is 1,600 students, which corresponds to approximately 50 students per recruited school in a given year. This model assumes an 88% response rate at the 6-month follow-up and an 80% response rate at the 12-month follow-up for the secondary outcomes, and these are based on response rates in a recent evaluation of a high school program conducted in 20 schools. The evaluation's power calculation was conducted cognizant of data on California high schoolers' sexual activity, which indicates that 20% of high school students in the state are sexually active. The study settled on this minimum detectable impact based on a meta-analysis by Chin et al. (2012) of similar group-based, comprehensive sexual risk-reduction programs for adolescents, in which the average effect size corresponded to a Cohen's d of 0.19.