Leveraging the Power of Nondisruptive Technologies to Optimize Mental Health Treatment: Case Study

Effects of an Artificial Intelligence Platform for Behavioral Interventions on Depression and Anxiety Symptoms: Randomized Clinical Trial

BACKGROUND

The need for scalable delivery of mental health care services that are efficient and effective is now a major public health priority. Artificial intelligence (AI) tools have the potential to improve behavioral health care services by helping clinicians collect objective data on patients' progress, streamline their workflow, and automate administrative tasks.

OBJECTIVE

The aim of this study was to determine the feasibility, acceptability, and preliminary efficacy of an AI platform for behavioral health in facilitating better clinical outcomes for patients receiving outpatient therapy.

METHODS

The study was conducted at a community-based clinic in the United States. Participants were 47 adults referred for outpatient, individual cognitive behavioral therapy for a main diagnosis of a depressive or anxiety disorder. The platform provided by Eleos Health was compared to treatment-as-usual (TAU) approach during the first 2 months of therapy. This AI platform summarizes and transcribes the therapy session, provides feedback to therapists on the use of evidence-based practices, and integrates these data with routine standardized questionnaires completed by patients. The information is also used to draft the session's progress note. Patients were randomized to receive either therapy provided with the support of an AI platform developed by Eleos Health or TAU at the same clinic. Data analysis was carried out based on intention-to-treat approach from December 2022 to January 2023. The primary outcomes included the feasibility and acceptability of the AI platform. Secondary outcomes included changes in depression (Patient Health Questionnaire-9) and anxiety (Generalized Anxiety Disorder-7) scores as well as treatment attendance, satisfaction, and perceived helpfulness.

RESULTS

A total of 72 patients were approached, of whom 47 (67%) agreed to participate. Participants were adults (34/47, 72% women and 13/47, 28% men; mean age 30.64, SD 11.02 years), 23 randomized to the AI platform group, and 24 to TAU. Participants in the AI group attended, on average, 67% (mean 5.24, SD 2.31) more sessions compared to those in TAU (mean 3.14, SD 1.99). Depression and anxiety symptoms were reduced by 34% and 29% in the AI platform group versus 20% and 8% for TAU, respectively, with large effect sizes for the therapy delivered with the support of the AI platform. No group difference was found in 2-month treatment satisfaction and perceived helpfulness. Further, therapists using the AI platform submitted their progress notes, on average, 55 hours earlier than therapists in the TAU group (t=-0.73; P<.001).

CONCLUSIONS

In this randomized controlled trial, therapy provided with the support of Eleos Health demonstrated superior depression and anxiety outcomes as well as patient retention, compared with TAU. These findings suggest that complementing the mental health services provided in community-based clinics with an AI platform specializing in behavioral treatment was more effective in reducing key symptoms than standard therapy.

TRIAL REGISTRATION

ClinicalTrials.gov NCT05745103; https://classic.clinicaltrials.gov/ct2/show/NCT05745103.

Machine Learning Model to Predict Assignment of Therapy Homework in Behavioral Treatments: Algorithm Development and Validation

BACKGROUND

Therapeutic homework is a core element of cognitive and behavioral interventions, and greater homework compliance predicts improved treatment outcomes. To date, research in this area has relied mostly on therapists' and clients' self-reports or studies carried out in academic settings, and there is little knowledge on how homework is used as a treatment intervention in routine clinical care.

OBJECTIVE

This study tested whether a machine learning (ML) model using natural language processing could identify homework assignments in behavioral health sessions. By leveraging this technology, we sought to develop a more objective and accurate method for detecting the presence of homework in therapy sessions.

METHODS

We analyzed 34,497 audio-recorded treatment sessions provided in 8 behavioral health care programs via an artificial intelligence (AI) platform designed for therapy provided by Eleos Health. Therapist and client utterances were captured and analyzed via the AI platform. Experts reviewed the homework assigned in 100 sessions to create classifications. Next, we sampled 4000 sessions and labeled therapist-client microdialogues that suggested homework to train an unsupervised sentence embedding model. This model was trained on 2.83 million therapist-client microdialogues.

RESULTS

An analysis of 100 random sessions found that homework was assigned in 61% (n=61) of sessions, and in 34% (n=21) of these cases, more than one homework assignment was provided. Homework addressed practicing skills (n=34, 37%), taking action (n=26, 28.5%), journaling (n=17, 19%), and learning new skills (n=14, 15%). Our classifier reached a 72% F₁-score, outperforming state-of-the-art ML models. The therapists reviewing the microdialogues agreed in 90% (n=90) of cases on whether or not homework was assigned.

CONCLUSIONS

The findings of this study demonstrate the potential of ML and natural language processing to improve the detection of therapeutic homework assignments in behavioral health sessions. Our findings highlight the importance of accurately capturing homework in real-world settings and the potential for AI to support therapists in providing evidence-based care and increasing fidelity with science-backed interventions. By identifying areas where AI can facilitate homework assignments and tracking, such as reminding therapists to prescribe homework and reducing the charting associated with homework, we can ultimately improve the overall quality of behavioral health care. Additionally, our approach can be extended to investigate the impact of homework assignments on therapeutic outcomes, providing insights into the effectiveness of specific types of homework.

Self-Management of Subclinical Common Mental Health Disorders (Anxiety, Depression and Sleep Disorders) Using Wearable Devices

RATIONALE

Common mental health disorders (CMD) (anxiety, depression, and sleep disorders) are among the leading causes of disease burden globally. The economic burden associated with such disorders is estimated at $2.4 trillion as of 2010 and is expected to reach $16 trillion by 2030. The UK has observed a 21-fold increase in the economic burden associated with CMD over the past decade. The recent COVID-19 pandemic was a catalyst for adopting technologies for mental health support and services, thereby increasing the reception of personal health data and wearables. Wearables hold considerable promise to empower users concerning the management of subclinical common mental health disorders. However, there are significant challenges to adopting wearables as a tool for the self-management of the symptoms of common mental health disorders.

AIMS

This review aims to evaluate the potential utility of wearables for the self-management of sub-clinical anxiety and depressive mental health disorders. Furthermore, we seek to understand the potential of wearables to reduce the burden on the healthcare system.

METHODOLOGY

a systematic review of research papers was conducted, focusing on wearable devices for the self-management of CMD released between 2018-2022, focusing primarily on mental health management using technology.

RESULTS

We screened 445 papers and analysed the reports from 12 wearable devices concerning their device type, year, biometrics used, and machine learning algorithm deployed. Electrodermal activity (EDA/GSR/SC/Skin Temperature), physical activity, and heart rate (HR) are the most common biometrics with nine, six and six reference counts, respectively. Additionally, while smartwatches have greater penetration and integration within the marketplace, fitness trackers have the most significant public value benefit of £513.9 M, likely due to greater retention.

A Comparison Between Clinical Guidelines and Real-World Treatment Data in Examining the Use of Session Summaries: Retrospective Study

BACKGROUND

Although behavioral interventions have been found to be efficacious and effective in randomized clinical trials for most mental illnesses, the quality and efficacy of mental health care delivery remains inadequate in real-world settings, partly owing to suboptimal treatment fidelity. This "therapist drift" is an ongoing issue that ultimately reduces the effectiveness of treatments; however, until recently, there have been limited opportunities to assess adherence beyond large randomized controlled trials.

OBJECTIVE

This study explored therapists' use of a standard component that is pertinent across most behavioral treatments-prompting clients to summarize their treatment session as a means for consolidating and augmenting their understanding of the session and the treatment plan.

METHODS

The data set for this study comprised 17,607 behavioral treatment sessions administered by 322 therapists to 3519 patients in 37 behavioral health care programs across the United States. Sessions were captured by a therapy-specific artificial intelligence (AI) platform, and an automatic speech recognition system transcribed the treatment meeting and separated the data to the therapist and client utterances. A search for possible session summary prompts was then conducted, with 2 psychologists validating the text that emerged.

RESULTS

We found that despite clinical recommendations, only 54 (0.30%) sessions included a summary. Exploratory analyses indicated that session summaries mostly addressed relationships (n=27), work (n=20), change (n=6), and alcohol (n=5). Sessions with meeting summaries were also characterized by greater therapist interventions and included greater use of validation, complex reflections, and proactive problem-solving techniques.

CONCLUSIONS

To the best of our knowledge, this is the first study to assess a large, diverse data set of real-world treatment practices. Our findings provide evidence that fidelity with the core components of empirically designed psychological interventions is a challenge in real-world settings. The results of this study can inform the development of machine learning and AI algorithms and offer nuanced, timely feedback to providers, thereby improving the delivery of evidence-based practices and quality of mental health care services and facilitating better clinical outcomes in real-world settings.