Digital therapeutics (DTx) uses digital technologies to deliver high-fidelity evidence-based treatments at scale to improve health outcomes and reduce the burden on health care professionals [1]. One of the primary benefits of DTx is its ability to overcome geographical and time-related barriers to health care accessibility. Children and young people have not been left behind in the advancement of digital technology in health care. As smartphones are becoming more available to children and families, DTx may improve and maintain the health of children and young people.

Providing health care for children and young people, a socially vulnerable population, requires a specific skill set owing to their distinct characteristics and needs [2]. In children and young people, the communication skills necessary for making medical decisions and understanding the information related to medical treatment may vary depending on their verbal language development level [3]. In addition, children and young people’s emotional (eg, anxiety or depression) and physical (eg, pain or a range of medications) conditions may affect their level of commitment to the treatment [3]. Finally, the children and young people’s cultural and religious background may influence how much they value medical treatment and how they interact with health care professionals [3]. Therefore, when conducting clinical practice or research on children and young people, various factors must be carefully considered.

Children and young people are also known for being early adopters of digital technologies, including digital health technologies. Hence, the advancement of digital technologies dedicated to children and young people offers considerable benefits and is frequently discussed in the literature. However, as children and young people are susceptible to manipulation, especially via digital devices and tools, digital platforms have terms of service that are specifically adapted to children and young people [4-6]. For example, the Google Play Store necessitates an in-app reminder within all social media apps to stay safe on the web and be aware of the real-world risks of web-based activities [4]. The Apple App Store requires parental permission to protect children’s privacy and prevent them from engaging in commerce or following external links to websites, social networks, or other apps without their parent or guardian’s knowledge [5,6]. Each of the Google Play Store and Apple App Store app review guidelines provides specific details regarding their requirements. Therefore, when developing digital interventions for children and young people, a holistic and careful approach based on the distinct guidelines of the digital platform is necessary.

Although there is growing interest in digital interventions for children and young people, we have not found reviews that provide an overview of the characteristics of digital interventions for children and young people, as well as the potential challenges to be considered when developing them. Thus, we performed a scoping review that summarizes the current characteristics of the digital interventions dedicated to children and young people, such as their medical specialties, target users, and features, and the potential challenges associated with their development, as described in peer-reviewed literature.

Our study aimed to provide a broader perspective that summarizes the characteristics of digital interventions and potential challenges for children and young people, as published in the peer-reviewed literature. The research questions were as follows:

This scoping review used the framework of Arksey and O’Malley [7]. This scoping review is reported based on the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) guidelines [8]. The completed PRISMA-ScR checklist is presented in Multimedia Appendix 1.

We performed a search in 5 databases (PubMed, Scopus, Embase, MEDLINE, and CINAHL) and Google Scholar to identify studies that examined at least 1 digital intervention targeting children and young people with specific medical problems or their caregivers. The analyses were limited to clinical trials published in English between January 1, 2018, and August 19, 2022. The complete search strategy is outlined in Multimedia Appendix 2.

The inclusion and exclusion criteria were applied to screen titles, abstracts, and full-text papers for study selection (Multimedia Appendix 3). When there was a disagreement over study selection, consensus was reached through discussion.

The selected studies were analyzed to identify the characteristics and possible complications of digital interventions for children and young people. On the basis of data items, relevant information was extracted and charted into a specific form. The form included the following data items: authors, year of publication, title, aim, population, target disease, user, study design, duration of intervention, intervention protocol, and results.

To address the research questions, charted data were used as supplementary sources and full-text papers were read multiple times. Results were independently collected from each article, and cross-validation was performed. Two categories, descriptive characteristics and potential challenges, were used to group the research results. Reporting of the results about the identified characteristics and potential challenges of digital intervention for children and young people was done using both narrative and numerical descriptions.

The following 4 review phases were conducted based on the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) flowchart (Figure 1): identification, screening, eligibility assessment, and final synthesis. Our search query initially retrieved 3775 articles. A total of 652 articles were identified during the automatic duplicate removal process, and 132 articles were manually identified as duplicates and removed. Of the remaining 2991 articles, 2723 were removed based on title and abstract screening and the inclusion and exclusion criteria (Multimedia Appendix 3). The remaining 268 articles were assessed for eligibility using full-text screening, and 19 articles were eligible; 15 articles identified on Google Scholar were included in this review. Finally, 34 articles were included in the analysis.

The percentage of digital interventions targeting mental disorders was higher than that targeting physical diseases (26/34, 76% vs 8/34, 24%; Table 1). The Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition criteria were used to differentiate between mental and physical disorders. If the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition criteria were satisfied, it was classified as a mental disorder. Regarding the target users, 85% (29/34) of interventions targeted children and young people, 6% (2/34) targeted their caregivers, and 9% (3/34) targeted both groups.

Most of the digital interventions were web based (17/34, 50%) or mobile app based (13/34, 38%). A total of 4 studies combined 2 different modalities to deliver the intervention (eg, web-based + eye tracker [9] and mobile app–based + Google Glass; Alphabet Inc [10]). The details are listed in Table 2.

More than one-third of the studies (13/34, 38%) applied cognitive behavioral theory as the theory of digital interventions. A total of 5 studies were based on behavior change theory, and 4 studies were based on cognitive theory. Three studies were based on applied behavior analysis. Guideline-based intervention was used in 3 studies. One study applied 3 different theories, namely social learning theory, family systems theory, and cognitive behavioral theory, to inform their intervention [11]. The psychodynamic theory was used in 1 study [12]. The details are listed in Table 3.

Regarding the active intervention duration, which refers to users actively receiving the intervention components, digital interventions ranged from a single session to 28 weeks (average 8.8, SD 6.49 weeks) for mental disorders and from 4 to 24 weeks (average 11, SD 6.05 weeks) for physical disorders. Digital interventions ranged from a single session to 24 weeks (average 8.1, SD 5.09 weeks) for children and young people, 11 to 28 weeks (average 19.5, SD 12.02 weeks) for caregivers, and 8 to 24 weeks (average 14, SD 8.72 weeks) when targeting both children and young people and their caregivers. The details are listed in Table 4.

By adapting existing frameworks to the design of digital interventions [13], the intervention components were classified into the following 5 categories: guidance, task and activity, reminder and monitoring (RM), supportive feedback (SF), and reward system (RS; Figure 2).

Guidance components provided users with educational information and were included in 22 studies. Of these, 19 provided users with education, for example, cognitive behavioral therapy, behavioral interventions, social story interventions, parent training programs, coping skills, or physical activity education [11,12,14-30]. Two studies provided users with tailored educational information, including personalized health and wellness information for someone with Down syndrome [31] and nutritional information with individualized meal plans [32]. One study used a robot to guide children in engaging in the repetitive motions required for motor learning [33]. One study provided users with contact information for mental health support services [25].

Task and activity components involved engagement or participation within the intervention [13] and were included in 17 studies. Four offered cognitive training [34-37], 1 offered a virtual reality game [33], and 2 offered emotion-recognition activities [9,10]. Seven studies provided users with homework exercises to help them practice learning [11,12,14,18,21,26,38]. To enhance emotional regulation skills, some interventions used several activities for self-management of overwhelming emotions [39] or reading and writing tasks for depression, anxiety, and well-being [40]. Furthermore, other studies included phonological awareness activities [38], videoconferencing with a therapist [30], peer-to-peer live streaming [15], or dance sessions [27].

In the RM component, users’ progress or status was monitored. Reminders were sent via SMS text messages [15,35], emails [11,31,34,41], or a digital intervention platform [11,12,14,34,37]. In total, 13 studies used self-monitoring, including daily check-ins [11], emotional self-monitoring [25], activity and blood glucose monitoring [27], symptom monitoring [29], bolus and nutrient content calculating [42], web-based assessment [43], assessment of depressive symptoms and sleep diary [20], daily mood queries [22], mood history [25], mood activity diary [18], ketosis monitoring [32], current health data recording [32], and scores in the virtual reality game [33]. The other 6 studies relied on monitoring features to ensure user compliance [9,15,16,21,34,37]. A list of the overall summary of the intervention components used in each study on digital interventions for children and young people or caregivers is provided in Multimedia Appendix 4 [9-12,14-43].

SF features enabled users to receive personalized feedback on their progress. One study provided SF to users via SMS text messages [43], 5 via telephone calls [19-21,26,35], 3 via emails [14,26,42], 3 via a digital intervention platform [12,29,31], and 1 via a human coach [33]. Some studies included chat sessions [18,23] and videoconferences with trained therapists [30]. Three studies offered >1 method of assistance delivery [12,24,27].

RS components are mechanisms that provide reinforcements to enhance the user’s behavioral change. Different types of rewards were used in 6 studies, including badge and RSs [11], reward loops [34], tokens to unlock minigames [9], incentives [35], unlocking the game Balloon Blast as a reward [39], and credits [43].

Our findings support the growing interest in digital mental health as a field of digital health research. The mental health of children and young people was targeted for digital interventions 3 times more frequently than physical health. In addition, a substantial number of digital interventions directly targeted children and young people. Digital health approaches are expected to provide major benefits to children and young people [44]. However, as children and young people are particularly susceptible to manipulation, targeted digital interventions require a careful risk-benefit analysis based on ethics, regulatory frameworks, monitoring mechanisms, and clinical decision-making.

We found that digital interventions for children and young people are more likely to be delivered via computers than via smartphones. This may be explained by the expansion in smartphone ownership by age; in 2019, more than half of all children and young people aged ≥11 years owned smartphones, whereas less than half of all children and young people aged ≤10 years owned smartphones [45]. On the basis of this expansion, the age of the target user is an essential factor to consider when developing digital interventions.

The duration of digital interventions for children and young people was more likely to vary depending on the target user rather than the target disease. The average duration of digital interventions targeting caregivers was more than twice that of the digital interventions targeting children and young people. Furthermore, the duration of digital interventions targeting children and young people showed the highest variability (from a single session to 24 weeks). The developmental level of children and young people may influence the duration of digital interventions [3]. Therefore, before implementing digital interventions for children and young people, their communication capabilities based on developmental stage must be considered.

In our study, the digital intervention components primarily consisted of guidance and RM. This finding supports the importance of providing children and young people with health education guidance, which can lead to positive well-being, academic success, and healthy adult outcomes [46]. In addition, as engagement in the intervention significantly impacts clinical outcomes [47], RM may be an essential component of digital interventions. Surprisingly, RS was the least used component in children and young people digital interventions. RS may lead to addictive behaviors when misused, similar to smartphones or computer games [48]. Thus, when selecting intervention components for susceptible populations such as children and young people, it is critical to define the component and how it will be applied in the digital intervention.

Most digital interventions required double consent procedures (children and young people and their caregivers) for real-world implementation. This process is frequently used to protect socially susceptible populations during health interventions. However, under certain circumstances, waiving consent from children and young people or their caregivers may be acceptable, depending on the age and characteristics of the children and young people. In the United States, if children and young people are <7 years, it is possible to waive their consent (as opposed to legal consent) and obtain consent only from their caregivers [49]. In the United Kingdom, children and young people aged <16 years mandate double consent procedures unless they prove maturity based on the Gillick ruling [50]. Although there is no universal consensus on when children should be considered capable of making medical decisions, the possibility that altered decision-making patterns may occur with the onset of adolescence, which occurs at approximately 12 years of age, should be considered [48].

On the basis of our results, consent waivers to caregivers may occasionally be permitted depending on the characteristics of the children and young people. For example, lesbian, gay, bisexual, transgender, queer children and young people are more likely than their peers to encounter stigma in health care settings and rejection from their caregivers [17,51]. This may cause them to be separated and isolated in health care settings, resulting in health disparities. Therefore, logistical barriers may need to be lowered for specific children and young people populations when deploying digital interventions.

On the basis of our findings, 2 types of potential risks are associated with digital interventions for children and young people: adverse events [10,12,14,18,23,28,34,35] and data privacy [10,17,26-28,34,37]. Generally, monitoring and responding to adverse events ensures safety while also advancing our understanding of how users interact with digital interventions. Therefore, it is critical for developers to proactively detect potential safety issues and for regulators to record and process adverse events. For example, the US Food and Drug Administration enforces MedWatch, a web-based reporting system that allows users and health professionals to voluntarily disclose adverse events that occur while using a Food and Drug Administration–regulated product [52]. Currently, the primary risks of digital interventions are privacy, confidentiality, and security of user data and information. Privacy vulnerabilities include unauthorized third-party access to confidential patient-related information, cybercrimes, and accidental data leakage, which could render digital interventions less acceptable to children and young people [31]. To ensure data security and information privacy, digital interventions for children and young people should use a secure platform with deidentified data under different sharable conditions [9-11,14,15,17,22,26-28,34,37].

Regarding the engagement and clinical effectiveness of digital interventions for children and young people, the involvement of their families and caregivers must be considered. The level of engagement of children and young people in digital interventions and clinical trials is frequently influenced by the preferences or barriers of their caregivers [24,30,32,35]. This might be more prevalent when the target disease has a strong family history [32]. This correlates with a recent systematic review that highlighted caregiver involvement as a critical component in treatments and studies of children and young people [53]. Although digital interventions are often likely to increase children and young people’s autonomy and empowerment in their treatment [30], children and young people are susceptible populations that may rely on the decisions of their caregivers, which may differ based on race and ethnicity, sex, or other social determinants of health.

Our findings reveal several societal issues that must be considered when developing and deploying digital interventions for children and young people, including their diversity and accessibility. The diversity of populations and settings in clinical trials is closely linked to the generalizability of outcomes. Although this is a critical point for scalability and implementation of digital interventions, most studies had a limited ethnicity or sex distribution in their recruitment [11,17,18,23,27,31] and settings [19,26]. To overcome these barriers, clinical trials should be conducted iteratively based on the DTx Real World Evidence Framework [54], highlighting partnership development as a co-design process and using diverse research methods. Similarly, when implementing interventions, language differences should be considered [22].

Although ease of access is one of the primary benefits of digital interventions, considerable structural barriers exist, such as socioeconomic inequality; this is called the digital divide. Many studies report that accessing digital interventions in rural areas is difficult, making it problematic for children and young people to undertake relevant interventions [9,15,17,29,40]. With adequate access, overcoming issues with digital literacy and facing severe disadvantages of novel health care solutions may be easier. To address these issues, policy makers may partner with educational institutions to integrate these technologies into the school setting. School-based health practice has the potential to address unmet children and young people health needs by increasing access to advanced health care for disadvantaged and marginalized populations.

This study had several limitations. First, some studies on digital interventions for children and young people may have been omitted, despite our detailed methodology. However, by combining the keywords used to search the 5 databases, we retrieved additional studies from Google Scholar that were absent from these databases. Second, no quality assessment of the included papers was performed, given the scope of this study. This should be considered in future research to provide an understanding of the quality of current research. Finally, only articles written in English were included in this study.

Researchers may consider involving caregivers in future research on digital intervention for children and young people, which may influence the engagement of children and young people in interventions as well as the prognosis of a disease that affects children and young people. Researchers may also consider setting up an appropriate follow-up safety plan with children and young people and their caregivers after they have left or completed the research with digital interventions as their responsibility to manage safety and deal with any emergencies for children and young people does not end when the research is completed. More research is needed to establish the standard of appropriate secure platform and data-sharing conditions when implementing digital intervention for children and young people by carefully examining and understanding the theory and features of digital interventions that researchers are going to conduct. In addition, future studies may include a clear breakdown of the development and implementation efforts of digital interventions for children and young people because these details are necessary for assessing their effectiveness and accessibility, which will be very useful to other researchers.

We identified potential challenges based on the distinctive features of children and young people and digital-based interventions published in peer-reviewed journals. We provided suggestions regarding ethical, interpersonal, and societal aspects to consider when developing and deploying such interventions for children and young people (Figure 3). To advance work in this dynamic area, we recommend thoroughly understanding the characteristics of children and young people at every developmental stage and considering their caregivers throughout the intervention period. To produce rigorous real-world evidence, a rigorous partnership for a co-design process that meaningfully involves children and young people and their caregivers in designing and implementing digital health interventions is recommended.