Exposure and Engagement Drive Impact: Results From a Large-Scale Trial of a Digital Family Planning and Reproductive...

We examined how exposure and engagement with CyberRwanda, a digital family planning and reproductive health intervention for Rwandan adolescents, influenced the intervention's impact.

A 3-arm cluster randomized hybrid type 2 effectiveness-implementation trial was conducted among 6,078 students in 60 secondary schools from 2021 to 2023. Schools were randomized 1:1:1 to CyberRwanda self-service (self-guided tablet access to digital platform), facilitated (self-guided and group tablet access plus in-person activities), or control. Intention-to-treat analyses demonstrated no impact of CyberRwanda on the primary outcomes of childbearing, contraceptive use, and HIV testing; however, several secondary outcomes relating to knowledge, attitudes, self-efficacy, and behavior were significantly improved in one or both CyberRwanda arms. In the present analysis, we estimate dose-response, per-protocol, and complier average causal effects to determine CyberRwanda's impact on effectiveness outcomes among participants with varying levels of exposure (any: ≥1 month, low: <12 months, and high: ≥12 months) and engagement (high frequency users: used CyberRwanda many times vs. low frequency users: did not use CyberRwanda or used only once or a few times). We also assessed individual, implementation, and school-level characteristics associated with high exposure and engagement to understand the factors that influence use.

Overall, 91.6% of participants were exposed to CyberRwanda. Of these, 61.1% were exposed for ≥12 months and 31.7% were high frequency users. Negligible effects were detected among participants with <12 months of exposure; however, the ≥12-month exposure subgroup had significant improvements on every secondary outcome that was improved in the full sample but with stronger effect sizes, including contraceptive use among sexually active participants (60.0% CyberRwanda vs. 49.4% control, prevalence difference: 0.12, 95% confidence interval: 0.05, 0.20). The complier average causal effect analysis revealed that contraceptive use among sexually active participants increased by 40 percentage-points for high frequency users with ≥12 months of CyberRwanda exposure versus control. No effects were observed on the primary outcomes. High exposure and high engagement were more common among males and younger participants; high frequency users were also more likely not to report tablet-related challenges.

Exposure and engagement were the primary drivers of CyberRwanda's impact, underscoring the importance of prioritizing access to and engagement with digital platforms to maximize effectiveness.