The study aims to evaluate whether use of a mobile mental health app improves outcomes of depression and anxiety for students.

There is a mental health crisis on college campuses with a growing demand for mental health services and increasing wait times. At the same time, there has been a dramatic surge in the development of mobile mental health apps. Thus, there is an opportunity to leverage these tools to improve mental health services. Our recent pilot study demonstrated the acceptability of using a mental health app to augment college mental health services and the feasibility of a fully digital approach to recruit and randomize students. The findings serve as a proof of concept for a full trial.

Specific Aims:

This study aims to evaluate the effectiveness of a mental health app to improve symptoms of depression and anxiety for students compared to treatment as usual.

In our pilot study, the investigators limited recruitment to students waiting for an initial appointment at Harvard University Health Services (HUHS) Counseling and Mental Health Services (CAMHS). In this trial, the investigators will broaden enrollment to include students seeking care at CAMHS as well as students who are not currently enrolled in treatment but who could benefit from additional mental health support.

In addition, in the full trial, the investigators will explore mediators of change, including perceived stress and isolation, through which a Cognitive Behavioral Therapy (CBT)-based Chatbot app ("Wysa") may contribute to improvement in depression and anxiety among students.

This will also involve exploring possible moderators of the effect such as students' demographic characteristics, concurrent or previous mental health treatment, and degree of usage of the mobile app.

The mobile mental health app will not be offered as a replacement for treatment at HUHS CAMHS but as a way of offering initial support and a potential gateway to seeking further treatment. In addition, the app is intended to promote mental wellness as using the app will help give students skills that may prevent them from needing further mental health treatment. For those with active symptoms, it is meant to increase students' interest in engaging with mental health services.

This study will make a significant and novel contribution to research in college mental health regarding the effectiveness of mobile mental health applications in the college population. Additionally, it will help provide understanding of the mediators of this effect and whether an innovative tool such as a CBT chatbot app can be used to promote mental wellness, prevent worsening mental health symptoms and increase uptake with other mental health related services for students who are not engaged in traditional treatment.

The investigators propose to address the following aims:

Aim 1: Evaluate the use of a CBT-based mobile Chatbot app ("Wysa") to improve outcomes of depression and anxiety for students compared to treatment as usual as measured by improvement in PHQ-9 and GAD-7 scores.

Hypothesis 1: Participants randomly assigned to the intervention will be more likely to report improved anxiety and depression outcomes, assessed after the intervention, compared to the participants randomly assigned to the control group (Cohen d = 0.4).

Aim 2: Assess the potential mediators of change for a CBT-based Chatbot app ("Wysa") to improve depression and anxiety.

Hypothesis 2: Perceived stress and social isolation will be identified as potential mediators of the relationship between the intervention and depression and anxiety.

Background and Significance

Untreated or under-treated depression and anxiety in college-age students has tremendous costs for the individual, the larger community and society at large, with the fatal outcome of suicide leading to devastating consequences for college communities. The frequency that students report having seriously considered suicide has increased by almost 40% in the last five years and the demand for counseling services has also grown five times faster than average student enrollment. Technological innovations such as mobile apps hold promise to improve access to care, allow students to seek care in a way that is potentially more comfortable for them, and to alleviate demand on over-burdened service centers. However, there is limited evidence on the effectiveness, feasibility of use and mechanisms by which these mobile platforms may improve mental health outcomes. Consequently, there is a pressing need to further research the use of mobile apps in this population to improve care and inform novel treatment approaches.

Considering the ubiquity of mobile phone usage in this population-98% of young adults aged 18-29 own a smartphone -there is significant interest in the use of mobile phone apps as an alternate care delivery platform. Additionally, the number of mobile mental health apps is exponentially growing with currently thousands of mental health apps available. Mobile apps have the potential to reach people outside of the traditional care model and allow people to use these tools when they need them, and as often as they like without having to wait for a mental health professional. Additionally, mobile health applications offer advantages such as continuous availability and immediate support, anonymity, customizable content and self-guided treatment approaches. These tools have also been seen as ways to increase service capacity and efficiency, reduce other barriers to face-to-face help seeking such as stigma and shame, and may be particularly well suited to young people who may be more accepting of technological approaches.

However, the evidence base for the efficacy of digital technologies for the mental health of college students remains unclear and research is significantly lagging behind app development. A systematic review found that even though many studies found that digital mental health interventions were effective, about half of the studies failed to present any usability or acceptability outcomes. They found that further research on user experience and engagement would be vital to the success of digital mental health interventions on college campuses. Furthermore, another study found that the largest problem for top-ranked apps for depression was attrition; the value of an AI-based chatbot lies in its ability to respond real-time to a user's inputs. Lastly, a scoping systematic review focused specifically on mental digital health platforms for university students found that students worldwide were prepared to engage with digital mental health interventions but placed a high emphasis on a need for reliable devices handling their sensitive data.

This study seeks to advance the initial findings from our pilot study and address current gaps in the literature. By selecting an interactive, AI-chatbot app, our study also aims to explore the role of an innovative approach to providing mental health resources and improving mental wellness for students.

Research Design

This is a randomized controlled trial conducted fully digitally to evaluate the effectiveness of a mobile chatbot app to alleviate symptoms of depression and anxiety in students.

Methods

Recruitment and Screening This trial will include a broad recruitment strategy with a focus on mental health promotion. Students who are both engaged in treatment at HUHS and those who are not engaged in treatment will be allowed to participate.

The recruitment strategy is detailed below:

HUHS/CAMHS:

Students visiting the HUHS website or the CAMHS website will be able to view information about the study and a link which will direct them to the iWellness Study website in order to enroll.

Students who seek care at HUHS CAMHS for symptoms of depression and anxiety will be offered the option to participate in the study when registering for an initial assessment at HUHS CAMHS. Students will be directed to the study website to view information about the study and a link to complete an enrollment eligibility survey.

Other recruitment efforts:

• Harvard graduate students who have completed a mental wellness survey as part of the Graduate Student Mental Health Initiative led by Dr. Paul Barreira would be invited to participate in this study after completing the survey by clicking on the link to enroll. The graduate wellness survey is accessible and distributed to graduate students in departments who agree to participate.
• Students will be able to obtain information about the study through campus outreach efforts for mental health promotion including the college and graduate/professional schools.
• Students will also be able to obtain information through relevant campus organizations and campus resource centers. Students would be directed to study information on the HUHS website.

Sample size: By targeting a broader population for recruitment our goal will be to recruit approximately 250-300 students in the study over approximately a 3-month enrollment period.

This is estimated based on data from our recruitment sources including students making new appointments at HUHS CAMHS, visiting the HUHS website, and Harvard graduate students who complete a mental health screening who would all be invited to participate in the study.

This sample size would allow us to obtain 0.80 power to detect significant improvement (Cohen d = 0.4) and significance 0.05.

Baseline Assessment and Group Assignment

Once a participant is recruited, they will go through the same enrollment process as in the pilot study in which they will be directed to the study enrollment website where interested participants will complete a survey to determine eligibility including information about basic demographics and mobile device ownership. Those meeting basic eligibility requirements will also complete screening using the Patient Health Questionnaire - 9 (PHQ-9) and the Generalized Anxiety Disorder 7-item scale (GAD-7) to determine final eligibility. If eligible, participants will agree to participate via an electronic consent form.

In order to test the possible mediator of change, participants will also be asked to fill out a baseline assessment of perceived stress using the Perceived Stress Scale and isolation using the Hughes Loneliness Scale.

Participants will then be randomly assigned to either the Wysa intervention app or the Harvard College Mobile app and given instructions on how to download their assigned app. Participants from both groups will continue to use their assigned app for 8 weeks after enrollment. Participants will receive a secure Qualtrics survey via email at 2-weeks, 4-weeks and 8-weeks. The endpoint of the study will be 8-weeks after enrollment. At the 8-week endpoint, participants will again receive a secure survey including the PHQ-9, GAD-7, Perceived Stress Scale, and the Hughes Loneliness Scale and questions about their satisfaction with the app and their overall care.

Inclusion Criteria:

• Participants will be required to be at 18 years or older
• Own a smartphone with Wi-Fi or 3G/4G capabilities
• Current student at Harvard University
• A PHQ-9 score of 5 or greater (indicating mild symptoms of depression or greater), or a score of 2 or greater on PHQ item 10 (indicating that they felt disabled in their life because of their mood) or a GAD-7 score of 8 or greater (indicating mild symptoms of anxiety or greater) will be required for enrollment.

Exclusion Criteria:

• Those who score less than 5 on PHQ-9 or less than 8 on the GAD-7 during study screening will be excluded from the study.
• Anyone who indicates thoughts of harm to themselves or others (PHQ-9 item 9 score > 0) during initial screening will be excluded from participation in the study.
• Participants with a PHQ-9 Q9 suicide item score > 0 will receive an automated notification that they are not eligible to participate in the study and that they should call the Harvard Counseling and Mental Health Services 24/7 line. Study staff will also inform Chief of HUHS CAMHS to reach out to the participant.

Participants will be randomly assigned to one of two apps and will be blinded to which is the intervention app and which is the control.