Potential and Pitfalls of Mobile Mental Health Apps in Traditional Treatment: An Umbrella Review

While the rapid growth of mobile mental health applications has offered an avenue of support unbridled by physical distance, time, and cost, the digitalization of traditional interventions has also triggered doubts surrounding their effectiveness and safety. Given the need for a more comprehensive and up-to-date understanding of mobile mental health apps in traditional treatment, this umbrella review provides a holistic summary of their key potential and pitfalls. A total of 36 reviews published between 2014 and 2022—including systematic reviews, meta-analyses, scoping reviews, and literature reviews—were identified from the Cochrane library, Medline (via PubMed Central), and Scopus databases. The majority of results supported the key potential of apps in helping to (1) provide timely support, (2) ease the costs of mental healthcare, (3) combat stigma in help-seeking, and (4) enhance therapeutic outcomes. Our results also identified common themes of apps’ pitfalls (i.e., challenges faced by app users), including (1) user engagement issues, (2) safety issues in emergencies, (3) privacy and confidentiality breaches, and (4) the utilization of non-evidence-based approaches. We synthesize the potential and pitfalls of mental health apps provided by the reviews and outline critical avenues for future research.


Introduction
Mobile mental health applications (i.e., apps) are virtual, smartphone-delivered platforms which provide self-directed or remotely facilitated mental health services in the areas of communication, self-monitoring, diagnosis, and treatment [1][2][3]. In order to circumvent user barriers associated with traditional treatment methods-including issues of poor availability, accessibility, and acceptability-these apps offer timely, cost-effective, and discreet channels for users to manage their mental health [3][4][5][6]. Specifically, help-seekers can overcome constraints of traditional clinical settings, such as long waitlists, restricted clinic hours, and living in regions with poor access to mental healthcare [3][4][5]7]. Instead of waiting an average of 14.5 days to consult a clinician [8], relevant information and interventions may be accessed in a timely manner and users may utilize apps for on-demand venting of thoughts and emotions [9,10]. Rather than seeking mental health support in face-to-face settings that require individuals to identify themselves, individuals may access support via apps anonymously and remotely, thus evading negative social evaluation [3][4][5]7].
Critically, as a reflection of the growing demand for mental healthcare [11], mental health apps have undoubtedly seen a rapid increase in their development and adoption. Between 2016 and 2018, they have grown threefold in number [5], offering help-seekers over 10,000 mental health apps to choose from [12]. Further, in a survey of 320 outpatient help-seekers from four clinics in the United States, 70% indicated interest in using apps to facilitate self-monitoring and management of mental health difficulties [13]. Considering their prominence and growing demand, therefore, it is important to inquire into how mental health apps may be utilized in conjunction with traditional interventions. 2 of 27 While an emerging body of research has investigated the utilization of mobile mental health apps in traditional treatment, however, findings have been scant and somewhat polarized. For instance, Torous et al. [14] focused on examination of challenges generated by mental health smartphone apps, while Eisenstadt et al. [15] concentrated on possibilities created by apps. On one hand, several studies have revealed the utility of mental health apps in supplementing different stages of traditional intervention, such as by providing education about treatment techniques prior to enrolment, facilitating symptoms-monitoring during the treatment process, and ensuring continued access to interventions after the treatment period [4,7,16]. On the other hand, a growing body of research has highlighted risks associated with app usage, such as the lack of safeguards around the privacy of users' information as well as utilization of non-evidence-based approaches [3,5,6]. Given this equivocality, there is a need for a more comprehensive view of the current mobile mental health apps landscape, to guide interested researchers toward a holistic understanding of apps as an adjunct to traditional treatment. As there is an increasing volume of reviews looking into the present mobile mental health landscape, we have chosen to conduct an umbrella review in the hope of presenting a big picture of the evidence base, as well as to discuss congruous or inconsistent findings. An umbrella review is a synthesis of systematic reviews, offering readers opportunities to look at a broad scope of factors investigated by scholars and consider whether consensus in the field has been met. Thus far, past investigations have provided an insightful outline of the current mobile mental health landscape, yet there is a relative lack of umbrella reviews that examined existing overviews. We aim to compile evidence from existing reviews to offer a higher level of summary.

Search Strategy and Selection Criteria
We included reviews of mental health apps that reported on: (1) the effectiveness and pitfalls of mobile mental health intervention program(s); and are (2) quantitative or qualitative reviews, rather than individual studies, aimed at reducing subclinical or clinical mental health symptoms. Eligible reviews, up to 31 May 2022, were identified from the Cochrane library, Medline (via PubMed Central), and Scopus databases by two co-authors (J.K., G.T.), using the following search terms: "mental health app *" OR "e%mental health" OR "mobile%based psychotherapy intervention *" OR "app%based mental health intervention *" OR "smartphone%based mental health intervention *" OR "digital mental health" OR "digital app * for mental health" OR "technology in psychotherapy" OR "mental health smartphone app *") AND ("review*" OR "synthesis" OR "meta-analysis" OR "meta-analytic").

Quality Assessment
We conducted a methodological quality assessment, using the JBI critical appraisal tool for systematic reviews [17], to evaluate the systematic reviews and meta-analyses included in our umbrella review. This critical appraisal tool comprises eleven items which are rated as "yes", "no", "unclear", or "not applicable". These include methodological evaluations of each review's inclusion criteria, search strategy, data synthesis, and strategies to minimize biases in data extraction and study appraisal. For each appraisal item, J.K. and G.T. conducted their evaluations independently and any disagreements were resolved through discussion after independent review. Assessments with at least five "yes" responses were included. In sum, the score (i.e., number of "yes" ratings) of the eligible reviews ranged from a moderate score of five or six (n = 4) to a high score of seven and above (n = 10). Our quality assessment identified that items four (i.e., "were the sources and resources used to search for studies adequate?") and six (i.e., "was critical appraisal conducted by two or more reviewers independently?") had the lowest proportion of "yes" ratings. This highlighted that (1) ensuring a comprehensive search strategy including grey literature; and (2) minimizing bias in critical appraisals are common methodological issues in systematic reviews and meta-analyses. Nevertheless, all fourteen eligible reviews for assessment had at least five "yes" ratings and, therefore, none were excluded from our umbrella review (see Table 1 for critical appraisal results).

Data Extraction
In line with Aromataris et al.'s [31] data extraction protocols for umbrella reviews, the following information was extracted from included reviews: (a) review details (author, year of publication, type of review, review objectives including interventions and outcomes assessed, total sample size, participant demographics, country), (b) search details (number of databases/sources searched, date range of included studies, number of studies included), and (c) analysis details (method of analysis, key findings). The extracted characteristics of included reviews are summarized in Table 2.     The majority of programs were effective or partially effective in producing beneficial changes in the main psychological outcome variables.  Evidence that e-mental healthcare can be a viable option for care delivery but that specific accessibility and acceptability factors must be considered.

Data Synthesis
Due to the heterogeneity of the included reviews in the study aims, mental health interventions, and outcome variables investigated across the included reviews, it was unfeasible to synthesize our results statistically. Instead, we narratively synthesized evidence from various systematic reviews, meta-analyses, scoping reviews, and literature reviews based on the primary findings of each review.

Results
The main search string returned 103 unique articles (see Figure 1 for PRISMA diagram [51]); and two additional articles were identified via Google Scholar. Thereafter, the article review proceeded in two phases. First, two co-authors reviewed the title and abstract for all 105 articles to determine initial eligibility based on our aforesaid selection criteria, and 48 articles were removed at this phase as they were protocols for literature reviews or articles that did not constitute quantitative or qualitative reviews (e.g., individual studies). In the second phase of article review, the remaining 57 articles were reviewed in full by two co-authors. At this stage, we excluded 13 articles which examined web-based mental health interventions and 5 articles which focused on the implementation of mobile mental health services (e.g., role of therapeutic alliances or gamification elements) [52,53]. In addition, three other reviews were excluded because they focused on outcomes other than mental health (i.e., academic performance [54]), the assessment of mobile mental health services [42], and the types of e-mental health systems and their degree of technological advancement [55]. As a result, a total of 21 articles were excluded at this stage, and 36 articles were included in the final review.

Data Synthesis
Due to the heterogeneity of the included reviews in the study aims, mental health interventions, and outcome variables investigated across the included reviews, it was unfeasible to synthesize our results statistically. Instead, we narratively synthesized evidence from various systematic reviews, meta-analyses, scoping reviews, and literature reviews based on the primary findings of each review.

Results
The main search string returned 103 unique articles (see Figure 1 for PRISMA diagram [51]); and two additional articles were identified via Google Scholar. Thereafter, the article review proceeded in two phases. First, two co-authors reviewed the title and abstract for all 105 articles to determine initial eligibility based on our aforesaid selection criteria, and 48 articles were removed at this phase as they were protocols for literature reviews or articles that did not constitute quantitative or qualitative reviews (e.g., individual studies). In the second phase of article review, the remaining 57 articles were reviewed in full by two co-authors. At this stage, we excluded 13 articles which examined web-based mental health interventions and 5 articles which focused on the implementation of mobile mental health services (e.g., role of therapeutic alliances or gamification elements) [52,53]. In addition, three other reviews were excluded because they focused on outcomes other than mental health (i.e., academic performance [54]), the assessment of mobile mental health services [42], and the types of e-mental health systems and their degree of technological advancement [55]. As a result, a total of 21 articles were excluded at this stage, and 36 articles were included in the final review.

Timely Support
In total, 16 out of 36 studies cited timely support as an advantage which mental health apps have provided, by transcending traditional help-seeking boundaries associated with waiting time and physical distance [9,12,15,25,29,[33][34][35][36]39,40,42,44,45,49,50]. Given that mental health apps provide in-the-moment support at the user's convenience, help-seekers can overcome constraints of traditional clinical settings, such as long waitlists, restricted clinic hours, and living in regions with poor access to mental healthcare [8]. For example, Struthers et al.'s [29] systematic review of 24 studies found that time-associated flexibility and level of control over treatment encourage the use of e-mental health services among youths, their parents, and their healthcare providers. Further, Chan and Honey's [9] integrative review identified that users perceive mobile mental health apps as being "easy to use", since app usage may be accessed on demand and can be easily integrated into the user's daily routines. Considering that delayed treatment contributes to more severe and enduring mental health difficulties [4,8,56], the timely nature of mobile mental healthcare is especially helpful in situations when an in-the-moment experience of relief is needed and traditional support might not be as helpful by the time it becomes available [33].

Cost-Effective
Further, as cited in 11 reviews, mental health apps afford users with the opportunity to access cost-effective treatment options according to their financial abilities [25,32,[34][35][36]42,44,46,[48][49][50]. For instance, Binhadyan et al.'s [35] literature review of 74 articles-which addressed e-mental health interventions for university students with ADHD-identified that the minimal (or no) fees for app-based interventions played a key role in enabling help-seekers to circumvent barriers to traditional mental healthcare. Echoing this, Oyebode et al.'s [46] thematic analyses of user reviews of 106 mental health apps found that the average price of 11 fee-based apps was USD 5.26, which is significantly lower than average psychotherapy fees ranging from USD 100 to USD 200 per session in the United States [57]. Hence, the lower cost of digital apps, compared to traditional psychotherapy, renders mental health apps a more accessible psychological tool for people of varying financial abilities.

Combat Stigma in Help-Seeking
Notably, eight studies noted that mental health apps provide the ability to access mental healthcare discreetly and thus circumvent the adverse stigma surrounding helpseeking [15,35,36,39,44,48,49]. For example, in Lal and Adair's [44] literature review of 115 articles about e-mental health interventions, it was highlighted that digital mental health interventions allow individuals who are uncomfortable with in-person treatment to receive help anonymously and bypass discomforts associated with identifying themselves and facing negative social evaluations. This may be especially helpful for people from collectivist cultures with prevalent "face" concerns, where conventional help-seeking has been found to be associated with poorer life satisfaction and lower positive affect [58]. Moreover, Wies et al.'s [50] review of 26 digital mental health treatments revealed that apps could serve as an initial point of contact and gradually facilitate transition to faceto-face interventions. In sum, mobile mental health apps potentially allow their users to overcome help-seeking barriers stemming from stigmatized attitudes toward conventional mental healthcare.

Enhance Therapeutic Outcomes
As highlighted in 25 studies, mobile mental health apps may also enhance therapeutic outcomes (see Table 3 for a summary of the target populations of the included reviews) including reducing symptoms of mood disorders [9,12,15,19,20,[22][23][24][25][27][28][29][30][35][36][37][38][40][41][42][46][47][48][49][50]. For instance, in Firth et al.'s [19] meta-analysis of 18 randomized controlled trials, smartphone interventions had a small-to-moderate effect in reducing depressive symptoms in an overall sample of 3, 414 adults from both clinical and nonclinical populations. Further, Petrovic and Gaggioli's [47] review of eight studies on mobile-based mental health tools showed that participants experienced reduced stress levels and improved coping skills after three weeks of app usage, suggesting that apps increase the likelihood of treatment success by providing opportunities to practice coping strategies in clients' natural environments. Harith et al.'s [41] umbrella review of seven studies also found significant evidence of the effectiveness of digital mental health interventions, including app-based programs, in alleviating depression, anxiety, stress, and eating disorder symptoms in university students. Table 3. Target population of included reviews.

Binhadyan et al. [35] ADHD
Borghouts et al. [18] Eisenstadt et al. [15] Firth et al. [19] Garrido et al. [20] Harith et al. [41] Lattie et al. [22] Leech et al. [23] Nicholas et al. [26] Petrovic & Gaggioli [47] Six et al. [28] Thach [48] Thach [49] Anxiety/Depression/Stress/Well-being Chan & Honey [9] Henson et al. [12] Anxiety, Depression, Schizophrenia spectrum and psychotic disorders Larsen et al. [21] Suicide/Self-harm Simblett et al. [27] Post-traumatic stress disorder (PTSD) More specifically, mental health apps can amplify treatment outcomes by complementing different stages of traditional interventions in line with their specific purpose. For example, in Hwang et al.'s [42] scoping review, certain mental health apps (e.g., MoodPrism, mHealth)-which track and monitor users' emotional state and psychological stress-were found to reduce symptoms of depression, anxiety, and stress. Hence, by providing on-thego documentation of users' psychological well-being, these apps can tailor relevant goals for each user in real-time and supplement traditional treatment. In addition, Oyebode et al.'s [46] thematic analysis of user reviews of 104 mental health apps revealed positive themes such as "reminder and notification", "in-app support", "logging", "analytics and visualization", "assessment", and "data export"; which indicate the unique features of mental health apps that are valued by help-seekers. Together, this suggests that users could utilize mental health apps in conjunction with traditional treatment to enjoy higher therapeutic success as compared to only receiving the traditional face-to-face intervention alone. Nonetheless, common themes for the pitfalls of mental health apps have been identified as well.

User Engagement Challenges
Six reviews referred to high attrition rates and poor rates of sustained engagement prevalent among mental health apps [14,20,22,24,29,30]. For instance, Garrido et al.'s [20] review of 32 digital mental health interventions found that 39% of studies reported attrition rates of over 20%-levels indicative of potential attrition bias. Further, in Struthers et al.'s [29] review of 24 studies on acceptability of e-mental health for youths, the number of participants who completed the full intervention ranged widely across studies from 29.4% to 87.5%, with two studies suggesting the decreasing usage of e-mental health interventions over time. As theorized by Torous et al. [14], user engagement may be hindered by factors including unsatisfactory functionality of these apps and usability concerns (i.e., difficulties using apps).

Safety Issues in Case of Emergency
According to two reviews [14,21], mental health apps may also be poorly equipped to assist users through emergencies. For instance, Larsen et al. [21] reviewed publicly available apps which address suicide, and found that none of these apps abided with the best practice of providing visible crisis support information within the app. Similarly, in Torous et al.'s [14] clinical review of challenges surrounding user engagement, it was suggested that the vast majority of apps are limited in their ability to respond effectively during emergencies related to suicide or self-harm, or recognize anticipatory warning signs. In the event of a time-sensitive mental health emergency such as risk of suicide, therefore, help-seekers might not be able to access critical support needed through mental health apps).
First, eight reviews found that mental health app users were commonly concerned with their confidential information being shared with third parties or used for unauthorized purposes such as marketing [9,14,18,29,39,45,46,50]. In Wies et al.'s [50] scoping review of ethical challenges in digital mental health, it was shown that mental health app users' main concerns centered around the consequences of confidential information being leaked to third parties, which would implicate professional, personal, and social domains of their lives. In particular, two reviews identified inadequate passcode protection (i.e., to prevent external access to users' data) as a privacy-related weakness of mental health apps [32,49]. For instance, a thematic analysis of user reviews of 106 mobile mental health apps revealed that mental health app users were dissatisfied with the lack of passcode protection (e.g., a unique PIN) to prevent external access to sensitive information [32].
A second concern was the lack of clear privacy policies which explain the protection of users' information, as highlighted by five reviews [21,26,32,46,49]. More specifically, only 22% of apps targeted at bipolar disorder and 29% of apps targeted at suicide or deliberate self-harm provided a clear privacy policy which informs users on how their data are used [21,26]. Moreover, Wies et al. [50] reported that there is insufficient clarity about the adequacy of consent obtained through digital mental health apps, in particular regarding the type of data processing or intervention that the user is consenting to. Taken together, therefore, the use of mental health apps is often accompanied by risks of being identified as a help-seeker or the leakage of personal information to third parties, thus endangering users' privacy and impeding trust and engagement with these apps.

Utilization of Non-Evidence-Based Approaches
Lastly, limited empirical and theoretical evidence has been found for both (1) the efficacy of mental health apps and (2) the basis of therapeutic interventions used in mental health apps.
First, 10 reviews found limited evidence for the effectiveness of mental health apps in reducing symptoms of psychological distress (e.g., depression, anxiety, stress) and improving socioemotional competency [14,23,24,36,38,40,42,45,47,50]. For example, Drissi et al.'s [38] systematic review of studies examining e-mental health interventions developed for healthcare workers found that only two studies (27%) included empirical evaluations of the reported interventions, and the empirical evaluations were based on a limited number of participants. Similarly, Gould et al.'s [40] review of mental-health-related apps created by the Veteran Affairs or the Department of Defense showed a pressing lack of evidence for the effectiveness of these apps, with the exception of two apps (PTSD Coach, Virtual Hope Box). Further, in studies examining the efficacy of app interventions, there has been a lack of empirical support for their long-term effectiveness. In Carter et al.'s [36] review of 37 digital mental health intervention studies, for instance, 23 studies (62%) reported results from less than 6-months follow-up. In addition, in Leech et al.'s [23] systematic review of mental health apps for adolescents and young adults, all 11 randomized controlled trials examined the immediate or short-term effects of app interventions, except four studies which incorporated 6-week to 6-month follow-up assessments. Together, this suggests that the long-term benefits of mental health app usage have not been established by empirical evidence, and help-seekers should not rely entirely on these platforms for mental health treatment.
Second, apart from the efficacy of mobile mental health apps, four reviews cited an insufficient theoretical and empirical basis for therapeutic techniques employed by mental health apps [14,42,47,50]. For example, Petrovic and Gaggioli [47] conducted a scoping review of digital mental health tools catered to informal caregivers in Europe, and found that only a small portion of their 16 reviewed papers defined a clear therapeutic rationale behind the interventions used, such as adopting principles of cognitive behavioral therapy or stress inoculation training. In addition, Hwang et al. [42] conducted a scoping review of 14 studies about mental-health-related apps for adults over 18 years of age, and identified two studies that did not provide theoretical evidence for their intervention methods, involving a breathing exercise app and mood-monitoring app. Seeing as such unsupported practices could unintentionally pose serious risks to the well-being of helpseekers in dangerous situations, it is crucial that clinicians and researchers remain astute as to the scientific evidence informing app-based mental healthcare.

Key Findings
In sum, mobile mental health apps can potentially circumvent barriers of traditional mental healthcare to provide timely, cost-effective, and discreet support which facilitates various stages of treatment and improves outcomes. On the other hand, it is imperative that app users (clinicians and help-seekers) are mindful of the pitfalls surrounding apps usage: these involve engagement challenges, safety issues, confidentiality breaches, and a lack of evidence-based practices (see Figure 2 for an overview).
In sum, mobile mental health apps can potentially circumvent barriers of traditional mental healthcare to provide timely, cost-effective, and discreet support which facilitates various stages of treatment and improves outcomes. On the other hand, it is imperative that app users (clinicians and help-seekers) are mindful of the pitfalls surrounding apps usage: these involve engagement challenges, safety issues, confidentiality breaches, and a lack of evidence-based practices (see Figure 2 for an overview).

Strengths and Limitations
Our review has several limitations that should be noted. First, given that our umbrella review provided a higher-level synthesis of a wide range of previous reviews, this introduced significant heterogeneity-regarding review methodologies (e.g., systematic reviews, narrative reviews, thematic analyses), primary focus of the reviews (e.g., efficacy, user engagement, ethical challenges), sample demographics (e.g., adolescents, caregivers, young adults), and outcome measures used (e.g., posttraumatic stress symptoms, depression symptoms, emotion regulation)-hence introducing difficulties with interpretation of common potential and pitfalls of mobile mental health apps. Nonetheless, this crossreview heterogeneity reinforces the need for the present umbrella review which identifies converging themes of mental health apps' advantages and downfalls despite varying aims and measures. Second, since we included only published peer-reviewed reviews written in English, unpublished work and reviews in other language mediums were not included in our search strategy; hence, this may have influenced the findings of this review. Further, as we searched three key databases, our search strategy may have excluded relevant reviews from other databases such as PsycINFO and EMBASE. Our inclusion criteria for reviews may have resulted in overlap of primary studies between reviews. Finally, due to the rapidly advancing nature of digital mental health interventions, it is possible that some of the mobile mental health apps assessed may now be outdated. In spite of these limitations, strengths of the present umbrella review include its strict adherence to methodology

Strengths and Limitations
Our review has several limitations that should be noted. First, given that our umbrella review provided a higher-level synthesis of a wide range of previous reviews, this introduced significant heterogeneity-regarding review methodologies (e.g., systematic reviews, narrative reviews, thematic analyses), primary focus of the reviews (e.g., efficacy, user engagement, ethical challenges), sample demographics (e.g., adolescents, caregivers, young adults), and outcome measures used (e.g., posttraumatic stress symptoms, depression symptoms, emotion regulation)-hence introducing difficulties with interpretation of common potential and pitfalls of mobile mental health apps. Nonetheless, this cross-review heterogeneity reinforces the need for the present umbrella review which identifies converging themes of mental health apps' advantages and downfalls despite varying aims and measures. Second, since we included only published peer-reviewed reviews written in English, unpublished work and reviews in other language mediums were not included in our search strategy; hence, this may have influenced the findings of this review. Further, as we searched three key databases, our search strategy may have excluded relevant reviews from other databases such as PsycINFO and EMBASE. Our inclusion criteria for reviews may have resulted in overlap of primary studies between reviews. Finally, due to the rapidly advancing nature of digital mental health interventions, it is possible that some of the mobile mental health apps assessed may now be outdated. In spite of these limitations, strengths of the present umbrella review include its strict adherence to methodology protocols for umbrella reviews (e.g., utilization of JBI critical appraisal checklist), holistic synthesis of evidence for both potential and pitfalls of mobile mental health applications, and inclusion of a broad evidence base including systematic reviews, meta-analyses, scoping reviews, and literature reviews.

Future Research Directions
To support the continued examination of app usage as an adjunct to traditional treatment, future research could inquire into three key areas.

App Functions
First, in terms of app functions, further research should examine the efficacy of mental health apps in supporting individuals with differing degrees of symptom severity. Considering that mental health apps are commonly designed and utilized to manage and relieve mild symptomatology [59], there is currently a lack of understanding regarding how these approaches may be applied to more severe symptoms. As app effectiveness may vary across the mild, moderate, and severe ranges of mental health conditions, future investigations could probe into how people with different levels of symptom severity (e.g., depression severity) respond to symptom relief provided by mental health apps.

App Regulation
Second, regarding app regulation, there is a need for further research to develop overarching evaluation guidelines for mental health apps. Due to the present lack of such guidelines, standardized criteria for "approved-for-use" apps remain unclear to both app developers and clinicians alike [5,60]. Hence, future studies should examine key elements for the regulation of mental health apps, such as the presence of evidenced-based approaches, existing randomized controlled trials conducted to assess app efficacy, as well as visibility of emergency services contacts. In so doing, both app developers and mental health professionals may achieve a shared understanding of the key elements guiding evaluation and regulation of mental health apps.

Individual Differences in Apps Usage
Finally, with regard to individual differences in apps usage, future research should look into the role of individual differences-including demographic factors and individual needs and preferences-in modulating the effectiveness of mental health apps [61,62]. Research has suggested that trait-like demographic and usage factors, including socioeconomic background, individual motivations underlying digital technology use, perceptions of usefulness, and smartphone use preferences could potentially influence access to and wellbeing outcomes of digital technology, including mental health apps [5,62,63]. Given that the potential of mental health apps has primarily been examined in adolescents and young adults (see [23] for a review), however, there is currently a lack of understanding about the role of these individual differences, such as demographics (e.g., socioeconomic status, age) and other usage factors, in shaping engagement with and effectiveness of mental health apps. Therefore, subsequent research should inspect how these individual difference factors influence apps effectiveness.

Conclusions
In sum, this umbrella review provided a comprehensive synthesis of existing quantitative and qualitative evidence regarding the potential and pitfalls of mobile mental health apps as an adjunct to traditional psychotherapy. Further, we offer three key areas for future research, concerning app functionality, app regulation, and individual differences in app usage. Our review highlights that mobile mental health apps' unique potential, such as providing timely support, being cost-effective, combating stigma surrounding help-seeking, and enhancing treatment outcomes, could be tapped into to supplement mental health interventions, although associated risks (i.e., user engagement challenges, safety issues, confidentiality breaches, and non-evidence-based approaches) need to be understood and managed. Specifically, one viable risk management strategy would be adhering to the American Psychiatric Association's hierarchal framework that emphasizes clinicians' responsibilities to examine stages of the framework with clients, discuss queries, and support shared decision-making on app usage [61].