This scoping review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) checklist to identify and map key characteristics, facilitators, barriers, and related themes in AYAs’ use of DMHIs (Multimedia Appendix 1) [35]. The review protocol was registered on PROSPERO (CRD42023479880).

The following major electronic databases were searched as of October 31, 2023: PubMed, Web of Science, PsycINFO, and CNKI. Although ScienceDirect was initially listed in our registered protocol, it was excluded from the final search due to significant content overlap with other databases. However, we ensured comprehensive coverage by searching additional relevant databases. Given that DMHIs are emerging technologies, no start date was set for the literature search to capture a broader range of publications. To supplement the search strategy, we manually retrieved bibliographies from relevant studies and included gray literature sources (OpenGrey and APA PsycExtra). The searches were restricted to English and Chinese. Further details are provided in Multimedia Appendix 2.

This research included studies examining facilitators and barriers to DMHIs among AYAs with disorders or symptoms of depression, anxiety, and stress. Reviews, recommendations, comments, newspapers, letters, conference abstracts, and research from other stakeholders’ perspectives were used to enhance the understanding of the topic but were excluded from the analysis. The following exclusion criteria were applied: (1) studies involving participants with serious illnesses (eg, heart failure or trauma), as these conditions could impose greater limitations on their thoughts and behaviors; (2) studies with participants younger than 10 or older than 26 years; (3) studies on digital health services unrelated to mental health; (4) studies assessing only the effectiveness of DMHIs, participants’ attitudes, or willingness—without discussing influencing factors; and (5) studies with unavailable full texts due to access restrictions or incomplete records were excluded to ensure transparency and replicability of the review.

The literature screening process was conducted using EndNote 20 (Clarivate Plc), and Microsoft Excel 2020 was used to record the extracted data. Two independent researchers (YW and YH) reviewed the literature and extracted data. First, YW identified and removed duplicate records, which were then manually checked for accuracy by YH. Second, the researchers independently screened studies based on titles and abstracts, followed by a full-text review to determine eligibility. Third, data extraction was conducted independently by both researchers. Any discrepancies at each stage were discussed and resolved by 3 researchers (YW, YH, and SZ). The extracted data from each included study comprised 4 dimensions: (1) metadata and study context (eg, authors, year, study design, and sample size); (2) characteristics of the study population (eg, country or region, race, sample type, and basic demographics); (3) characteristics of DMHIs (eg, type, delivery mode, number of sessions, and inclusion of self-help or self-directed tools); and (4) facilitators and barriers influencing AYAs with CMDs in accessing DMHIs.

In line with the WHO’s 3-level digital health outcomes assessment framework [36], which includes the health system, provider, and client perspectives, this review adopted a thematic analysis framework comprising 3 levels: (1) external level, (2) intervention level, and (3) individual level.

Thematic analysis in this study followed a 6-phase process (Textbox 1) [37].

After identifying themes and subthemes, a relative frequency meta-analysis was conducted to assess the occurrence of each facilitator and barrier across different delivery modes: completely nonportable devices, portable devices, single-platform, and multiple-platform interventions. Studies that did not specify the delivery model were excluded. The analysis used the relative frequency of occurrence (RFO) with 95% CIs, performed using the metaprop function in R (R Foundation) [38].

The quality of the included literature was assessed using the Mixed Methods Appraisal Tool (MMAT) [39]. This study applied the 2018 version, which was developed through a literature review of critical appraisal tools, user interviews, and electronic Delphi consultations with international experts. The MMAT evaluates 5 research categories: (1) qualitative studies, (2) randomized controlled trials, (3) nonrandomized studies, (4) quantitative descriptive studies, and (5) mixed methods studies. In addition to 2 general screening questions applicable to all study types, 5 specific criteria were established for each research category—qualitative, quantitative randomized controlled trials, quantitative nonrandomized studies, quantitative descriptive studies, and mixed methods studies [39].

A total of 6063 records were retrieved from electronic databases and gray literature sources. After removing duplicates, 2498 records remained. Title and abstract screening identified 131 studies for full-text review, of which 27 met the inclusion and exclusion criteria and were included in data extraction, synthesis, and analysis (Figure 1; also see Multimedia Appendix 1).

Table 1 presents the characteristics of the included studies. Among the 27 studies, 7 (26%) were quantitative, 6 (22%) were qualitative, and 14 (52%) applied mixed methods. Geographically, 4 (15%) studies were conducted in England, 5 (19%) in the United States, 3 (11%) in New Zealand, 7 (26%) in Australia, 2 (7%) in Ireland, and 2 (7%) in China. Additionally, 1 (4%) study was conducted in each of Canada, Sweden, and South Africa, while 1 (4%) spanned both Chile and Colombia. Regarding participant recruitment channels, 5 (19%) studies recruited participants from health institutions, 4 (15%) from the community, 6 (22%) from schools, and 2 (7%) from youth centers and a survey, respectively. Additionally, 10 (37%) studies recruited participants from multiple sources, including health institutions, schools, communities, teams, and existing research studies. The DMHIs examined in the included studies can be broadly categorized into web/internet-based, computer-based, app-based, game-based, and other formats. Specifically, 13 (48%) studies delivered DMHIs via web/internet-based programs, 6 (22%) were app-based, and 2 (7%) were game-based. One study (4%) utilized a text messaging tool, while 1 (4%) implemented a chatbot. Several interventions employed hybrid delivery methods, such as text messaging combined with web-based programs (1/27, 4%), computer-based combined with web-based programs (2/27, 7%), and web-based combined with app-based interventions (1/27, 4%). Regarding conflicts-of-interest disclosures, 19 studies included a conflicts-of-interest component, with 7 studies [5,8,10,16,17,40,41] explicitly declaring potential or actual conflicts of interest.

Regarding the mental health symptoms and therapeutic approaches covered by the DMHIs in the included studies, 22 out of 27 (81%) aimed to improve depressive symptoms. Additionally, 14 studies addressed other mental health challenges, including anxiety, stress, worry, anger, and suicidal ideation. Notably, 3 studies [12,32,42] specifically targeted anxiety disorders. Among the 20 studies that explicitly stated a theoretical foundation for intervention development, 14 incorporated cognitive behavioral therapy, with 3 [8,15,30] utilizing computerized cognitive behavioral therapy. Furthermore, 2 studies [17,43] were based on peer support, 1 on behavioral activation [44], 1 on behavior change and resiliency [40], 1 on positive psychology [7], and 1 on mindfulness [45].

The quality assessment (Table 2) indicated that qualitative studies and quantitative randomized controlled trials were generally of high quality. However, all quantitative descriptive studies exhibited varying degrees of sample representativeness.

Based on the generated themes, the predominant facilitators and barriers varied across different delivery modes, as shown in Figures 3 and 4. For completely nonportable devices, quality and effect emerged as the most relevant facilitator (RFO=53%, 95% CI 0.24-0.81) and barrier (RFO=42%, 95% CI 0.01-0.91) to DMHI use. By contrast, for portable devices, the primary facilitators included high quality and effect (RFO=21%, 95% CI 0.12-0.31), design harmony (RFO=20%, 95% CI 0.11-0.28), and content engagement (RFO=13%, 95% CI 0.04-0.23), while the dominant barrier was low quality and effect (RFO=15%, 95% CI 0.02-0.35). For single-platform DMHIs, usage was primarily associated with 2 key facilitators: high quality and effect (RFO=32%, 95% CI 0.13-0.50) and perceived benefits (RFO=18%, 95% CI 0.07-0.31). The predominant barrier was low quality and effect (RFO=30%, 95% CI 0.05-0.60). Regarding DMHIs available across multiple platforms, the most influential facilitators were design harmony (RFO=25%, 95% CI 0.12-0.38), high quality and effect (RFO=23%, 95% CI 0.10-0.37), and content engagement (RFO=10%, 95% CI 0.01-0.26). The primary barriers were perceived risks (RFO=17%, 95% CI 0-0.54) and low quality and effect (RFO=13%, 95% CI 0-0.36). Further details are provided in Multimedia Appendix 5.

This review systematically identified facilitators and barriers to AYAs’ access to DMHIs by analyzing quantitative, qualitative, and mixed methods literature using a 6-step thematic analysis approach. Following WHO guidelines, a 3-level theoretical framework was used to categorize these influential factors. Fourteen facilitators were identified, including integration with schools and other resources, social norms, and strategic marketing. Conversely, 13 barriers were recognized, such as integration with schools, content gaps, and design limitations.

The DMHIs examined in our review primarily targeted depression and were largely based on cognitive behavioral therapy. However, their development and application vary widely. Some studies have shown that DMHIs incorporating mindfulness training and peer support can effectively alleviate psychological challenges, including anxiety, stress, and resilience [43,45]. These findings underscore the potential of diverse DMHI approaches in mental health support and contribute to the growing digital landscape of psychological interventions.

During the RFO analysis, we focused specifically on digital modes of intervention, excluding studies that did not specify their delivery modes. Research supports the notion that different platform types may lead to variations in functionality, usage, and performance [55]. In our study, the theme of “quality and effect” emerged as the predominant facilitator and barrier across all DMHI delivery modes. This finding suggests that the perceived quality and effectiveness of DMHIs were central concerns for both participants and researchers. High-quality interventions that demonstrate positive outcomes are more likely to enhance user engagement, support sustained use, and encourage recommendations to others [51]. Conversely, interventions that fail to meet user expectations or inadequately address their needs may result in disengagement or abandonment [7].

When examining the subthemes, the most common and consistent points across the 3 levels were summarized. Participants showed greater willingness to engage when DMHIs were integrated with other resources [46,47], possibly due to increased accessibility and a reduced risk of judgment or embarrassment [56]. Previous research and professionals particularly valued the benefits of combining DMHIs with traditional medical resources and incorporating them into daily life [54,57,58]. The use of multimedia, also referred to as “aesthetics” or “visual assets,” was widely accepted and recommended for its ability to attract attention, generate interest, facilitate understanding, and enhance usability and satisfaction [12,59,60]. Perceived helpfulness or usefulness aligns with the Technology Acceptance Model, as it influences participants’ attitudes toward use, thereby shaping behavioral intentions and actual usage behaviors [61]. Negative emotional experiences led to reduced engagement among AYAs [15], with certain emotions, such as frustration and nervousness, contributing to physical and mental stress [62]. Privacy, security, and credibility concerns remained common barriers in DMHIs, similar to traditional psychotherapy. Despite technological advances and digitization, these remote services pose unique and heightened privacy risks to clients [54,63].

Furthermore, 2 factors—DMHI integration with schools and personalization—can serve as both facilitators and barriers. On the one hand, integrating DMHIs into school settings may significantly enhance accessibility to psychological support by streamlining delivery to students and potentially addressing gaps in resources and teacher training [46]. Embedding mental health interventions within the school environment allows students to access support without requiring external referrals or additional logistical arrangements. On the other hand, integrating DMHIs with schools may make them less appealing to students and could evoke negative feelings [45,46]. Additionally, strict school regulations on smart device usage may limit students’ ability to engage freely with DMHIs [54], potentially reducing their effectiveness by restricting flexible and convenient access to these digital tools. Regarding personalization, participants frequently highlighted its advantages, such as greater flexibility, tailored monitoring experiences, feedback, and additional benefits [44,45]. However, some participants viewed personalization negatively, as it often required additional actions and steps, adding to their burden and inconvenience [15]. Balancing the positive and negative aspects of these 2 factors requires strategies that maximize benefits while minimizing drawbacks. Suggested approaches include offering opt-in participation and flexible access options, allowing students to decide whether to engage and choose between on-campus or off-campus access to school-integrated DMHIs to enhance their sense of control and reduce stress. Additionally, making personalization optional by preconfiguring basic settings can improve ease of use, enabling users to engage with personalized features only to the extent they require.

Although many randomized controlled trials indicated that DMHIs were satisfying [8,10], they were not considered a substitute for traditional mental health services but rather an augmentation, potentially expanding accessibility and enhancing engagement [56,64,65]. From users’ perspectives, the most intuitive feedback was their preference for face-to-face communication, along with concerns and dissatisfaction regarding privacy issues and the mechanical, fixed responses of DMHIs [5,14,66]. However, this does not imply that users are inclined to reject DMHIs; rather, they support DMHIs as a supplementary form of care alongside traditional mental health services [65]. From a scientific research and development perspective, a hybrid model integrating DMHIs with traditional mental health services can maximize the advantages of both, such as addressing low retention rates in DMHIs and assisting mental health professionals in providing follow-up care [67,68]. Consequently, the notion that DMHIs can serve as a “digital glue” to enhance user engagement in mental health services is more widely accepted, as it facilitates a seamless transition between digital and nondigital services [56]. Therefore, supporting hybrid digital and traditional mental health services should be the central focus for the future development and implementation of DMHIs.

An effective design process is crucial for the success of DMHIs in the mental health field, with a particular emphasis on human-centered or user-centered design [69,70]. Developers and designers must adopt this approach to refine and enhance DMHIs, ensuring they align with users’ needs. Future research should focus on identifying the key components among the myriad and complex facilitators and barriers, sharpening the design process, and exploring which types of DMHIs are best suited for different mental health conditions.

Additionally, basic user characteristics may serve as unique influencing factors. Current findings suggest that women tend to spend more time using mental health apps [71], whereas men show greater interest in simulation game-based DMHIs [72]. Limited research has explored differences in DMHI usage based on race and ethnicity, though some evidence indicates that people of color may face barriers to accessing culturally responsive care in DMHIs [3]. Further systematic exploration and synthesis are needed to identify specific cultural factors that facilitate or hinder DMHI uptake among AYAs.

This scoping review summarized the facilitators and barriers to DMHIs for AYAs with depression, anxiety, and stress, categorizing them in a structured manner. By synthesizing the literature, the review offers insights for future intervention service designers and therapists, supporting the translation of DMHIs from research to practice—an essential step for advancing mental and public health. However, our review has some limitations. First, due to our inclusion and exclusion criteria, the quality of the included studies varies, with some not meeting high-quality standards. Second, the participants’ characteristics and age ranges in the reviewed studies were constrained by our exclusion criteria. Factors that facilitate or hinder adults or older adults with CMDs from engaging with DMHIs may similarly affect AYAs. Third, we excluded studies published in languages other than English and Chinese, which may have led to the omission of culturally specific factors. Another potential limitation is the search strategy. While we used keyword searches across multiple databases, this approach may have been less comprehensive than using built-in search structures (eg, MeSH terms in PubMed), potentially resulting in the omission of relevant studies. Finally, we acknowledge that the frequency of an individual factor does not necessarily indicate its significance, underscoring the need for follow-up studies to clarify its importance.

This scoping review systematically searched, screened, and synthesized the literature on facilitators and barriers to DMHIs for AYAs with depression, anxiety, and stress. Through thematic synthesis, we identified a series of themes and subthemes at the external, intervention, and individual levels, highlighting key factors influencing the use and adoption of DMHIs. By consolidating these factors, this review provides insights that can inform the design and implementation of more effective DMHIs for AYAs.