Mental Health Issue
Conditions
Keywords
ai, llm, cbt, cognitive behavioural therapy, cognitive behavioral therapy, artificial intelligence, large language model, app, cbt-i, homework, limbic, worksheets, workbook, cbt intervention
Brief summary
The goal of this clinical trial is to compare two different ways of delivering cognitive behavioral therapy (CBT) materials to people with symptoms of anxiety and depression. The main questions it aims to answer are: 1. Do people using the app experience greater reductions in anxiety and depression symptoms compared to those using the workbook? 2. Does the personalization offered by the app increase engagement with CBT materials compared to the workbook? Participants will be given access to either the Limbic Care app or an equivalent PDF workbook and will be asked to engage with the materials for several days each week over a period of 6 weeks. Each week, participants will complete a survey about their anxiety and depression symptoms and report any new health issues and their experiences with the materials.
Detailed description
Motivation Homework is a crucial component of cognitive-behavioral therapy (CBT) as it allows patients to actively implement therapeutic techniques and skills outside of sessions, promoting long-term mental health improvements. However, patient engagement with CBT homework can be challenging due to a lack of motivation, forgetfulness, or a lack of personalization, which can hinder the overall effectiveness of the therapy and retention in a therapy program. Design This randomized controlled trial (RCT) aims to evaluate the efficacy, acceptability, and safety of an AI-enabled smartphone application (Limbic Care) compared to a standard digital workbook (PDF format) in delivering cognitive behavioral therapy (CBT) materials to individuals with clinically significant symptoms of anxiety and/or depression. The study will involve 500 participants, who will be randomly allocated in a 3:2 ratio to either the intervention group (Limbic Care app) or the control group (PDF workbook). A large cohort (\ 2,000 participants) will initially be screened for inclusion and exclusion criteria. Eligible participants will be invited to participate in a baseline survey (capped at 550 to account for attrition and reach a target of 500 participants for the final data collection point). This survey will measure multiple validated symptom scales (e.g., for anxiety, depression, obsessive-compulsive disorder, health anxiety disorder) and gather subjective ratings on variables such as familiarity with CBT, previous experience with mental health apps, preferences for apps vs. PDF workbooks, medication status, and more. After the baseline survey, participants will be randomly allocated to one of the two experimental arms and invited to download the app (intervention group) or to access the PDF workbook via a weblink (control group). Over a 6-week period, participants will be asked to engage with their assigned materials (app or workbook) for a specified number of days each week. Weekly surveys will measure changes in anxiety (GAD-7) and depression (PHQ-9) symptoms, as well as any new adverse health events. Additionally, engagement metrics such as time spent using the app or viewing workbook pages will be automatically recorded. Outcome Measures The primary outcomes are changes in GAD-7 and PHQ-9 scores from baseline to week 6, measured weekly. We will also use engagement (operationalised as the time spent with the materials, as well as the number of exercises interacted with) as a primary outcome, with the aim of investigating whether this is a mediating factor for any effect of material format (app vs workbook) on symptom reduction. The secondary outcome is the incidence of adverse health events, to estimate and compare the safety of each material format. Additional outcome measures include changes in the Work and Social Adjustment Scale (WSAS), self-reported satisfaction and acceptability, perceived utility and effectiveness, motivation to engage with the materials, and changes in sleep quality (measured by the Mini Sleep Questionnaire, MSQ). Data Analysis Data will be analyzed using appropriate statistical methods to compare the intervention and control groups. Primary and secondary outcomes will be analyzed using repeated measures ANOVA or mixed-effects models, adjusting for baseline scores and potential confounders. Subgroup analyses will be conducted to explore the effects of demographic variables on treatment outcomes. Ethics and Safety The study has been approved by the UCL research ethics committee and will be conducted in accordance with the Declaration of Helsinki. Participants' safety and confidentiality will be prioritized throughout the study. Adverse events will be closely monitored, and participants will have the option to withdraw from the study at any time without penalty.
Interventions
DEVICELimbic Care
Limbic Care is an AI-enabled smartphone application that delivers cognitive behavioral therapy (CBT) interventions and psychoeducation to users via an interactive, conversational interface using AI. Participants can choose from three courses (sleep, worry, and low mood) that sets the type of content they will see in the app. For each course, different interventions and psychoeducation material will appear in the to do list section of the app on a fixed schedule. The app also allows users to engage in open-ended conversation with the AI, which serves as an non-interventional, empathetic listener.
DEVICEDigital Workbook
The digital workbook in PDF format will be provided to participants via the Internet. Participants can choose from three workbooks that each focus on a different topic: sleep, worry, and low mood. Each workbook contains psychoeducation material and CBT intervention worksheets, consisting of text, images, and response boxes to write in (if printed out).
Sponsors
Limbic Limited
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
TREATMENT
Masking
NONE
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
Yes
Inclusion criteria
* Has a personal smartphone with Internet access with an accepted operating system: iOS (any) or Android (10 or later) * Has a GAD-7 score \>= 8 OR a PHQ-9 score \>= 10 (indicating anxiety and/or depression symptoms above a clinical threshold, according to NHS Talking Therapies guidelines) * US resident * Fluent in English
Exclusion criteria
* Currently undergoing mental health therapy * Changed psychotropic medication in the past 8 weeks (this includes starting/stopping medication or increasing/decreasing dosage) * Self-identify as being at risk of harming themselves or others * Have used the Limbic app before * Use recreational drugs more frequently than once per week * Consume \>= 10 alcohol units per week
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Change in GAD-7 (Anxiety) | Baseline to Week 6, and weekly (from Baseline to Week 6) | The GAD-7 (Generalized Anxiety Disorder-7) is a 7-item self-report questionnaire used to screen for and measure the severity of generalized anxiety disorder (GAD). |
| Engagement Duration | Summed across Week 1 to Week 6 | Time spent in the Limbic Care application (intervention) or viewing workbook pages (control). |
| Therapy Completion | Summed across Week 1 to Week 6 | Proportion of therapy content viewed (exercises in-app, or PDF pages viewed) |
| Change in PHQ-9 (Depression) | Baseline to Week 6, and weekly (from Baseline to Week 6) | The PHQ-9 (Patient Health Questionnaire-9) is a 9-item self-report screening and diagnostic tool for depression. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Incidence of Adverse Health Events | Summed across Week 1 to Week 6 | Self-reported incidence of any adverse health events (physical or mental) experienced in the past week. |
Other
| Measure | Time frame | Description |
|---|---|---|
| Change in WSAS (Wellbeing) | Baseline to Week 6, and weekly (from Baseline to Week 6) | The Work and Social Adjustment Scale (WSAS) is a 5-item self-report measure used to assess the degree of functional impairment attributable to an identified problem or disorder (for this study, participants were asked to consider their problems more generally). |
| Change in MSQ (Sleep) | Baseline to Week 6, and weekly (from Baseline to Week 6) | The Mini Sleep Questionnaire (MSQ) is a 10-question self-report instrument designed to screen for excessive daytime sleepiness (EDS) and sleep disorders. |
| Motivation | Weekly, from Week 1 to Week 6 | Self-report ratings on Likert scales 1 (worst) to 5 (best) for how motivation (App Preference, Likelihood of Trying Therapy, and Trust in Wellbeing Apps that Use AI) |
| Utility and Effectiveness | Weekly, from Week 1 to Week 6 | Self-report ratings on Likert scales from 1 (worst) to 5 (best) for usability and effectiveness (Effectiveness, Future Use, Learning, Understandability, Usefulness, Accessibility, and Ease of Use) |
| Satisfaction and Acceptability | Weekly, from Week 1 to Week 6 | Self-report ratings on Likert scales from 1 (worst) to 5 (best) for satisfaction and acceptability (Achievement, Motivation, Personalisation, Satisfaction, Structuredness). |
Countries
United Kingdom
Outcome results
None listed