Using Machine Learning to Optimize User Engagement and Clinical Response to Digital Mental Health Interventions

Using Machine Learning to Optimize User Engagement and Clinical Response to Digital Mental Health Interventions

Using Machine Learning to Optimize User Engagement and Clinical Response to Digital Mental Health Interventions

Digital mental health interventions are a cost-effective and efficient approach to expanding the accessibility and impact of psychological treatments; however, little guidance exists for selecting the most effective program for a given individual. In the proposed study, decision rules will develop for selecting the digital program that is most likely to be the optimal intervention for each user. These treatment recommendations can be implemented in the context of large healthcare delivery systems to improve the delivery of digital mental health interventions at scale. The overarching aim of the current study is to better understand for whom and how leading digital interventions work in a large healthcare setting. The study builds on the existing literature and follows expert recommendations by using machine learning (ML) methods to develop precision treatment rules (PTRs) for three leading digital interventions for emotional disorders (e.g., anxiety, depression, and related mental health disorders). Specifically, ML methods will be used to develop PTRs to optimize clinical outcomes and associated intervention engagement. This study will leverage a unique partnership between Boston University (BU), SilverCloud Health (SC)--a leading provider of digital mental health care--and Kaiser Permanente (KP)--one of America's leading health care providers. A clinical trial (RCT) will be conducted to evaluate the relative effectiveness of three distinct empirically supported digital mental health interventions (from SC's existing library of programs) in a sample recruited from KP primary care and other clinical settings. Data from this trial will be used to develop theoretically and empirically informed, reliable selection algorithms for managing treatment delivery decisions. Algorithms will be validated in a separate "holdout" dataset by examining whether allocation to predicted optimal treatment is associated with superior outcomes compared to allocation to a non-optimal treatment. The role of user engagement will be determined, and other mechanisms in treatment outcome.

Kaiser Permanente (KP) members referred to SilverCloud Health (SC) through established operating procedures will be eligible to participate. After referral, potential participants will be informed of the study through the SC digital health platform. Interested participants will be provided information about the study through a digital version of an approved consent form. After signing consent, participants will complete baseline assessment questionnaires through Qualtrics, an online data collection platform. After completing the baseline assessment, eligible participants will be randomly assigned to one of three existing digital mental health interventions hosted by SC: (1) Unified Protocol for Transdiagnostic Treatment of Emotional Disorders (UP), (2) Space from Depression (SFD), and (3) Space for Resilience (SFR). A SC program supporter will also be established at that time. This differs from standard operating procedures at SC in that participants are generally able to select their preferred program from SC's offerings. There is nothing that would preclude participants from seeking a different SC treatment following study participation. Potential participants will have up to two weeks to enroll in the study, complete the baseline assessment, and begin a treatment program. If the participant does not initiate treatment within two weeks, they will be withdrawn from the study though they may continue with clinical services at SC. Following randomization, study staff at BU will send participants Qualtrics questionnaires via email throughout the treatment phase of the study, at 4-, 8-, and 12-weeks after treatment is initiated. The participant completes these questionnaires online. Regular asynchronous written support will be provided by SC supporters, consistent with their standard procedures.

Digital, transdiagnostic, emotion-focused CBT intervention that consists of five "core" modules or components that have been shown to target temperamental characteristics (i.e., neuroticism) and resulting emotion dysregulation that are believed to underlie all anxiety, depressive, and emotional disorders. The core components of the program are psychoeducation, mindfulness, cognitive flexibility, behavioral strategies to counter emotion-driven behaviors, interoceptive, and emotion exposures.

Digital CBT program designed to minimize the impact of depressive symptoms. This program emphasizes the use of cognitive behavioral strategies as well as mindfulness through a series of seven structured modules. The core components of the program include psychoeducation around the relationship between thoughts, feelings, and behaviors; cognitive behavioral practices aimed at restructuring negative beliefs; behavioral strategies to improve self-esteem; and mindfulness techniques that focus attention on the present moment.

Digital program based on positive psychology principles and designed to promote resilience and well-being through a series of seven modules. The core components of the program include psychoeducation, values exploration, building relationships, promoting self-esteem and self-efficacy, and building gratitude and optimism.

This is a cognitive behavioral treatment (CBT) for emotional disorders. This transdiagnostic intervention consists of eight modules and can be effectively applied to various disorders and problems.

Digital CBT program designed to minimize the impact of depression symptoms. Emphasizes CBT strategies and mindfulness through a series of seven structured modules.

This program is built from positive psychology principles and is designed to promote resilience and well-being through seven modules.

World Health Organization-Five (WHO-5) is a 5-item self-report questionnaire of psychological well being. The WHO-5 has demonstrated good validity and utility as an outcome measure in clinical trials including several digital intervention trials. Respondents rate each statement in relation to the past two weeks on a scale ranging from 0= At no time to 5= All the time (Ware, 1995).

Generalized Anxiety Disorder-7 (GAD-7) is a reliable and valid self-report measure of symptoms of generalized anxiety that is widely used in clinical research. This measures consists of 7 questions in a likert scale format, and respondents rate each question on a scale ranging 0=Not at all or 3= Nearly every day, indicating how often they experience anxiety-related problems (Spitzer, Kroenke, Williams, Lowe, 2006).

Patient Health Questionnaire 9 (PHQ-9) is a reliable and valid self-report measure of depression symptoms that is widely used in clinical research. Consisting of 9 questions, respondents rate each question on a scale ranging 0= Not at all or 3= Nearly every day, indicating how often they experience depression-related problems (Pfizer, 1999).

Inclusion Criteria: English-speaking adults Ages 18 or older Have a device that can connect to the internet.