Principal Results and Comparisons With Previous Work
The aim of this review was to systematically examine the literature on mobile apps for mental health in children and young people. Our review identified 24 papers describing 15 apps or prototypes, two of which were available to download from Google Play or iTunes [40,42,48]. We identified only two small RCTs [44-47], one of which was a laboratory-based experimental study [47], and both failed to demonstrate a significant effect on their intended outcomes (depression or body image). Therefore, we conclude that currently there is no evidence to support the effectiveness of apps designed for adolescents with mental health problems. Because we did not identify any study participants younger than 9 years, there is no evidence to support the effectiveness of apps designed for children with mental health problems either.
Our conclusion is consistent with previous reviews and highlights that the evidence base has barely increased over the past 4 years [18,23-25]. The lack of empirical studies contrasts starkly with the commercial development of mobile apps. From October 2013 to June 2016, the number of apps available to download from the app store doubled to 2 million [49], of which 1.98% (39,600) were classed as medical. Given the significant increase in the availability of mobile apps, the lack of evidence to support their safety or effectiveness with vulnerable populations is concerning.
Although the evidence base is currently lacking, this does not rule out the fact that well-designed, adequately tested, evidenced-based mobile apps could be effective. The evidence base for the clinical effectiveness of mobile apps in adult mental health is slowly emerging [18,50-52]. Our review suggests that the comparative literature for children and adolescents is significantly lagging, a trend also noted within the literature on other forms of eHealth, such as computerized CBT [6,7].
In terms of acceptability, it has been suggested that apps and eHealth in general are particularly suited for adolescents who are familiar with and regular users of technology [23]. We identified 12 small pilot feasibility trials [26-37] that suggested, in the short term, adolescents may be favorably disposed to this form of delivery. Acceptability was generally positive with ratings of ease of use, satisfaction, and usability rated average to high [26,29,32,33,35,37]. The privacy and discretion afforded by well-designed apps were of importance to young people [28,29,39]. However, many participants were healthy, nonreferred adolescents and less is known about whether those with mental health problems would have similar views. It is important to also note that although adolescents may have positive attitudes toward mHealth, it does not necessarily mean they would prefer it over a face-to-face intervention [24]. It is also important to consider whether the affinity that people have for their mobile phones and the trust and expectations placed in them positively influence clinical outcomes and user satisfaction [53]. This “digital placebo effect” may account for why some people continue to download and use mobile apps for mental health even though the evidence base is largely absent [53]. Nonetheless, our findings support previous conclusions and suggest that apps may provide an acceptable way of supporting mental health interventions for some adolescents [18,54].
Therapist perspectives on mobile apps were mixed, with concerns relating to patient security, increased responsibility and workloads, and the need to set clear boundaries between sessions [28]. These are different concerns to those surrounding the use of other forms of eHealth, such as computerized CBT, in which clinicians were concerned about the effectiveness of computerized CBT with more severe mental health problems and the lack of a therapeutic relationship [55]. This may reflect differences in the purpose of these interventions (ie, computerized CBT being a therapeutic intervention compared to apps that are an adjunct to therapy). Interestingly, therapists who used the app Mobile Mood Diary in clinical practice reported benefits such as facilitation of client engagement [28]. Lack of technical confidence was the most common barrier to implementation. This lack of technical confidence may be addressed by improving the user-friendliness of the app, either by codesigning apps with therapists or providing training for therapists.
App usage, where reported, was moderate and adherence ranged from 65% to 83%, which is comparable to those seen in Internet interventions for depression and anxiety [56]. There was a suggestion that self-monitoring of mood via apps promoted higher adherence compared to paper self-monitoring [33,35]. Information on longer-term usage is scarce, but the included studies suggest app usage begins high and declines over time [33,35]. This “law of attrition” [57] is also a common challenge for computerized CBT and eHealth interventions [57]. As with these other technology-based interventions, using mHealth apps with support from a therapist offers one strategy for increasing longer-term engagement [24,58]. Indeed, the SmartCAT app used in conjunction with face-to-face support demonstrated an 83% completion rate [33], similar to completion rates demonstrated in face-to-face CBT (84%) and guided Internet CBT (81%) [59]. Making mHealth apps inherently more engaging by design is another strategy for increasing longer-term engagement. One promising proposition is the use of serious gaming, gamification principles, telepresence, and persuasive technology in eHealth (and by extension mHealth) design [58,60]. The evidence base for the benefits of these principles as applied to mHealth and eHealth is currently in its infancy, however, and is a burgeoning area of research [60].
This review highlights several methodological concerns about the quality of the research evidence for mental health mobile apps, especially those for adolescents. Sample sizes tend to be small and reporting of demographic data such as gender and age inadequate, particularly in pilot feasibility studies. Few participants have an identified mental health problem and, as such, little is known about the acceptability and use of apps with clinical groups. As far as can be determined, the youngest participant in these studies was 9-years-old, meaning there is no research evidence for the use of mobile apps in children younger than this age. Where reported, symptoms tended to be mild to moderate in severity and, as such, the appropriateness of mobile apps for complex or more severe problems is unknown. Studies tend to be short in duration and there is sparse information on whether positive gains from using mobile apps are maintained. Finally, none of the apps in this review have been evaluated using a suitable RCT comparing a mobile app to an adequate control group. Future research should address these methodological concerns. Given the beneficial role that parent participation and engagement can have in adolescent mental health treatment [61], future research may also want to consider the role of parents/guardians in supporting adolescents using apps for mental health.
Our review has focused on the academic literature and of the apps identified, two of which were available to download. This contrasts starkly with the large number available from commercial sites and raises questions about the safety, quality, and efficacy of those that are available [11,22,62]. Content analyses [21,43] of six apps for children and adolescents available to download highlighted that none have been subject to any research evaluation. The authors also noted that the apps did not reflect best practice guidelines and lacked privacy policies [21,43]. Some of these apps claim to address a worrying number and type of complex problems, including “child abuse,” “daughter’s abusive relationship,” and “teen suicide, depression, and stress.” Ineffective or detrimental apps are a significant concern and incur costs to patient safety and care [11]. Therefore, our review adds to calls for better regulatory oversight to ensure app quality and safety [11,18,22,62].
