Enhancing Engagement With Digital Mental Health Care

University of Washington

Status: Completed

Conditions: Engagement, Patient

Treatments: Behavioral: Study 1, Generic Resources; Behavioral: Study 1, Generic Response; Behavioral: Study 2, DIY tool without AI; Behavioral: Study 1, Next Steps Survey; Behavioral: Study 1, Tailored Resources by Demographics; Behavioral: Study 1, Tailored Response; Behavioral: Study 1, Demographics Survey; Behavioral: Study 2, DIY tool with AI; Behavioral: Study 1, Tailored Resources by Desired Resources; Behavioral: Study 2, Control

Study type: Interventional

Funder types: Other; Industry; NIH

Identifiers

NCT04507360

1R01MH125179-01 (U.S. NIH Grant/Contract)

STUDY00010958

Details and patient eligibility

About

This proposal is a partnership between Mental Health America (MHA), a nonprofit mental health advocacy and resource organization, Talkspace (TS), a for-profit, online digital psychotherapy organization, and the University of Washington's Schools of Medicine and Computer Science Engineering (UW). The purpose of this partnership is to create a digital mental health research platform leveraging MHA and TS's marketing platforms and consumer base to describe the characteristics of optimal engagement with digital mental health treatment, and to identify effective, personalized methods to enhance motivation to engage in digital mental health treatment in order to improve mental health outcomes. These aims will be met by identifying and following at least 100,000 MHA and TS consumers over the next 4 years, apply machine learning approaches to characterizing client engagement subtypes, and apply micro-randomized trials to study the effectiveness of motivational enhancement strategies and response to digital mental health treatment.

Full description

Digital mental health (DMH) is the use of technology to improve population well-being through rapid disease detection, outcome measurement, and care. Although several randomized clinical trials have demonstrated that digital mental health tools are highly effective, most consumers do not sustain their use of these tools. The field currently lacks an understanding of DMH tool engagement, how engagement is associated with well-being, and what practices are effective at sustaining engagement. In this partnership between Mental Health America (MHA), Talkspace (TS), and the University of Washington (UW), the investigators propose a naturalistic and experimental, theory-driven program of research, with the aim of understanding 1) how consumer engagement in self-help and clinician assisted DMH varies and what engagement patterns exist, 2) the association between patterns of engagement and important consumer outcomes, and 3) the effectiveness of personalized strategies for optimal engagement with DMH treatment.

This study will prospectively follow a large, naturalistic sample of MHA and TS consumers, and will apply machine learning, user-centered design strategies, and micro randomized and sequential multiple assignment randomized trials (SMART) to address these aims. As is usual practice for both platforms, consumers will complete online mental health screening and assessment, and the investigators will be able to classify participants by disease status and symptom severity. The sample that the investigators will be working with will not be limited by diagnosis or co-morbidities. Participants will enter the MHA and TS platforms prospectively over 4 years. For aim 1, participant data will be analyzed statistically to reveal differences in engagement and dropout across groups based on demographics, symptoms and platform activity. For aim 2, the investigators will use supervised machine learning techniques to identify subtypes based on consumer demographics, engagement patterns with DMH, reasons for disengagement, success of existing MHA and TS engagement strategies, and satisfaction with the DMH tools, that are predictive of future engagement patterns. Finally, based on the outcomes from aim 2, in aim 3a the investigators will conduct focus groups applying user centered design strategies to identify and co-build potentially effective engagement strategies for particular client subtypes. The investigators will then conduct a series of micro-randomized and SMART trials to determine which theory-driven engagement strategies, co-designed with users, have the greatest fit with subtypes developed under aim 2. The investigators will test the effectiveness of these strategies to 1) prevent disengagement from those who are more likely to have poor outcomes after disengagement, 2) improve movement from motivation to volition and, 3) enhance optimal dose of DMH engagement and consequently improve mental health outcomes. These data will be analyzed using longitudinal mixed effects models with effect coding to estimate the effectiveness of each strategy on client engagement behavior and mental health outcomes.

The purpose of aim 3b is to identify effective engagement strategies tailored to client needs and demographics to increase MHA website engagement, and to better understand how self-help mental health resources can help people overcome negative thinking and support healthier thought processes. The investigators will compare effective engagement strategies tailored to subtypes developed under aim 2 to study the mediated impact of engagement strategies on consumer mental health outcomes. The study team will determine if engagement strategies targeted to consumer engagement subtype will enhance engagement and in turn result in improved clinical outcomes. These will be compared to generic strategies that are not subtype targeted.

All aim 3b activities will occur with MHA, broken down into two parts: (Study 1) a sequential multiple assignment randomized trial (SMART) and (Study 2) a Do-It-Yourself (DIY) tool longitudinal randomized control trial (RCT). Study 1 will use a SMART to examine methods to optimize engagement with MHA's website, and Study 2 will recruit participants for a longitudinal month-long study where they are randomly assigned to a control group, the use of a DIY tool without Artificial Intelligence (AI), or the use of a DIY tool with AI to examine the efficacy of using a digital tool to improve mental health functioning. An AI tool that uses machine learning/Natural Language Processing (NLP)/AI methods was developed to personalize and tailor an intervention to improve engagement and completion outcomes. The study focuses on a specific, popular DIY tool that teaches cognitive restructuring. Pilot work showed that (1) engagement and completion rates on DIY tools can be low, and (2) a pilot AI tool had significantly higher engagement and completion rates. These differences may arise due to AI support, User Interface/User Experience/design differences, other factors, or a combination thereof. Additionally, the efficacy of the digital tool to improve mental health functioning is unknown. Study 2 will recruit participants who will be randomly assigned to one of three groups for a longitudinal month-long study: thrice-weekly DIY tool use with AI, without AI, or a control group.

Enrollment: 78,390 patients

Sex: All

Ages: 14+ years old

Volunteers: Accepts Healthy Volunteers

Inclusion criteria

• Phase 3b Study 1 (Sequential Multiple Assignment Randomized Trial; SMART): Users of the Mental Health America (MHA) website, engaging from Internet Protocol (IP) addresses in the United States, who have chosen to start the PHQ-9 depression screener in English. Must be those who can read English.
• Phase 3b Study 2 (Do-It-Yourself; DIY): Patient Health Questionnaire (PHQ)-9 or Generalized Anxiety Disorder (GAD)-7 score of 10 or greater, users of MHA website, 18 years of age or older.

Exclusion criteria

• Phase 3b Study 1 (SMART): None
• Phase 3b Study 2 (DIY): Younger than 18 years old, Non-English or Non-Spanish speaking, PHQ-9 less than 10, outside of US, have more than a little familiarity with the concept of cognitive reframing.

Trial design

Primary purpose: Health Services Research

Allocation: Randomized

Interventional model: Sequential Assignment

Masking: Triple Blind

---

78,390 participants in 11 patient groups

Study 1, Demographic Survey + Generic Response + Generic Resources

Experimental group

Description:

This arm included participants who completed the Demographics Survey (stage 1 randomization). The Generic Response + Generic Resources condition (stage 2 randomization) will feature the response-as-usual on Mental Health America (MHA)'s website and the four generic resource pages on MHA's website.

Treatment:

Behavioral: Study 1, Demographics Survey

Behavioral: Study 1, Generic Resources

Behavioral: Study 1, Generic Response

Study 1, Demographic Survey + Generic Response + Tailored Resources by Demographics

Experimental group

Description:

This arm included participants who completed the Demographics Survey (stage 1 randomization). The Generic Response + Tailored Resources by Demographics condition (stage 2 randomization) will feature the response-as-usual on Mental Health America (MHA)'s website and resources tailored to two demographics. People who endorse being Lesbian, Gay, Bisexual, Transgender, or Queer (LGBTQ) will receive 4 resources associated with LGBTQ issues. People who endorse being 8-17, 18-24, 25-44, or 45+ years of age will receive 4 resources used by people in those age groups. If someone enters both age and LGBTQ status, they will be provided with 2 resources tailored to age and 2 resources tailored to LGBTQ status, randomly chosen.

Treatment:

Behavioral: Study 1, Demographics Survey

Behavioral: Study 1, Tailored Resources by Demographics

Behavioral: Study 1, Generic Response

Study 1, Next Steps Survey + Generic Response + Generic Resources

Experimental group

Description:

This arm included participants who completed the Next Steps Survey (stage 1 randomization). The Generic Response + Generic Resources condition (stage 2 randomization) will feature the response-as-usual on Mental Health America (MHA)'s website and the four generic resource pages on MHA's website.

Treatment:

Behavioral: Study 1, Next Steps Survey

Behavioral: Study 1, Generic Resources

Behavioral: Study 1, Generic Response

Study 1, Next Steps Survey + Generic Response + Tailored Resources by Desired Resources

Experimental group

Description:

This arm included participants who completed the Next Steps Survey (stage 1 randomization). The Generic Response + Tailored Resources by Desired Resources condition (stage 2 randomization) will feature the response-as-usual on Mental Health America (MHA)'s website and 4 resources tailored to a survey question that asks participants what they would like to do next on the website after screening is complete (e.g., "Learn more about depression", "Take another mental health test")

Treatment:

Behavioral: Study 1, Tailored Resources by Desired Resources

Behavioral: Study 1, Next Steps Survey

Behavioral: Study 1, Generic Response

Study 1, Next Steps Survey + Tailored Response + Generic Resources

Experimental group

Description:

This arm included participants who completed the Next Steps Survey (stage 1 randomization). The Tailored Response + Generic Resources condition (stage 2 randomization) will feature a response tailored to screening status (above or below criteria for depression) and expressed need for mental health support (e.g., "We're so glad to hear you're open to exploring how to improve your mental health. People who score with minimum or mild depression often notice that symptoms can get worse in the weeks after taking a Depression test.") They will also receive the 4 generic resources.

Treatment:

Behavioral: Study 1, Tailored Response

Behavioral: Study 1, Next Steps Survey

Behavioral: Study 1, Generic Resources

Study 1, Next Steps Survey + Tailored Response + Tailored Resources by Desired Resources

Experimental group

Description:

This arm included participants who completed the Next Steps Survey (stage 1 randomization). The Tailored Response + Tailored Resources by Desired Resources condition (stage 2 randomization) will feature a response tailored to screening status (above or below criteria for depression) and expressed need for mental health support, and will receive resources tailored to a survey question that asks participants what they would like to do next on the website after screening is complete (e.g., "Learn more about depression", "Take another mental health test").

Treatment:

Behavioral: Study 1, Tailored Resources by Desired Resources

Behavioral: Study 1, Tailored Response

Behavioral: Study 1, Next Steps Survey

Study 2, Control

Experimental group

Description:

Participants in the Do-It-Yourself (DIY) control group will receive psychoeducation materials in week 0. They will view content as usual (no DIY) and will receive surveys from week 1 to week 4 and follow-up surveys in week 5 and at the end of week 8.

Treatment:

Behavioral: Study 2, Control

Study 2, DIY tool without AI

Experimental group

Description:

Participants in the Do-It-Yourself (DIY) tool without Artificial Intelligence (AI) group will be instructed to use the DIY tool 3 times a week. They will receive surveys from week 1 to week 4 and follow-up surveys in week 5 and at the end of week 8.

Treatment:

Behavioral: Study 2, DIY tool without AI

Study 2, DIY tool with AI

Experimental group

Description:

Participants in the Do-It-Yourself (DIY) tool with Artificial Intelligence (AI) group will be instructed to use the DIY tool with AI 3 times a week. They will receive surveys from week 1 to week 4 and follow-up surveys in week 5 and at the end of week 8.

Treatment:

Behavioral: Study 2, DIY tool with AI

Study 1, Demographic Survey

Experimental group

Description:

This arm included participants who were randomized to the Demographics Survey (stage 1 randomization) and were not randomized to a stage 2 condition (i.e., did not view screening results page on which stage 2 randomization occurred).

Treatment:

Behavioral: Study 1, Demographics Survey

Study 1, Next Steps Survey

Experimental group

Description:

This arm included participants who were randomized to the Next Steps Survey (stage 1 randomization) and were not randomized to a stage 2 condition (i.e., did not view screening results page on which stage 2 randomization occurred).

Treatment:

Behavioral: Study 1, Next Steps Survey