Abstract / Summary
Background: Digital health interventions (DHIs) offer scalable ways to reach youth. While DHI research often poses minimal risk, guardian permission is typically required for minors to participate. As a result, requiring guardian permission may unintentionally obstruct safe, high-quality research and delay its translation.
Objective: This review aims to explore practices related to guardian permission. This study systematically reviewed randomized controlled trials of DHIs with minors from the United States published between January 1, 2020, and June 17, 2026.
Methods: PubMed, Embase, PsycINFO, and ACM Digital Library were searched; Cochrane Library and Web of Science yielded only duplicates and were therefore excluded. Included studies had to be: conducted in the United States, randomized controlled trials evaluating DHIs for pediatric mental and/or behavioral health, written in English, and peer-reviewed. Studies were excluded if participation was decided at the school level, included caregivers or pregnant teens as participants, or had a mean sample age of more than 18 years. Two coders independently screened and extracted data, with a third, unique reviewer resolving discrepancies. Risk of bias was assessed using the revised Cochrane risk of bias tool for randomized trials (RoB 2). A narrative framework was used to identify themes. Inverse variance-weighted linear regressions and logistic regressions were conducted to compare studies with and without permission waivers.
Results: Of 7170 screened studies, 49 met inclusion criteria (21 used guardian permission waivers; 28 required permission). The most commonly reported rationale for waiver use was participant privacy. Logistic regressions showed that studies with waivers were more likely to assess gender beyond the binary of male-female (odds ratio [OR] 4.27, 95% CI 1.22-14.93; P=.02) and more likely to recruit online (OR 0.04, 95% CI 0.01-0.17; P<.001). Weighted study-level regression revealed significantly lower follow-up retention in studies with waivers (mean 77.03%, SD 12.32%) compared to those requiring guardian permission (mean 88.61%, SD 17.17%; P=.005). No significant differences emerged across age, race, ethnicity, assessment of suicidality, sensitive topic assessment, intervention focus, or treatment retention.
Conclusions: Though several frameworks exist for evaluating ethics around waiving guardian permission for minors in research, this is the first review to systematically evaluate research in DHI, which often varies in terms of its level of risk. Although small sample sizes and variable reporting practices limited statistical power, the lack of differences in sample characteristics or study features between studies that used waivers and those that did not suggests that institutional review boards may be differentially interpreting regulations and guidance on waiving guardian permission. These findings likely indicate the need for more evidence to guide risk and benefit evaluations-or more education about the existing evidence base, as well as the risks and potential benefits of pediatric DHI research.