IntelliCare Study: Artificial Intelligence in a Mobile (AIM) Intervention for Depression

This study will evaluate and compare a smartphone intervention for depression and anxiety that uses machine learning to tailor treatment for participants with the same intervention without the machine learning component. The intervention, referred to as IntelliCare, delivers participant-specific treatment material and motivational messaging via a mobile phone to help individuals with depression and/or anxiety. Information and data received from the participant will inform the tailored treatment approach through machine learning. The purpose of this study is to compare different versions of the main IntelliCare Hub App (the centralized program delivery system) and participant user experience whether with the support of a coach or used independently. The randomized clinical trial (RCT) aims to obtain information on the feasibility and effectiveness of IntelliCare in improving symptoms of depression and anxiety.

Major depressive disorder (MDD) is a common mental disorder, with up to 10.3% of the population experiencing the disorder in a given 12-month period. The relationship between depression and anxiety has been well documented. Depression is a significant predictor of future anxiety, and anxiety is a significant predictor of future onset of depression. Indeed, studies find that more than 50% of all individuals with MDD also have a current anxiety disorder. Although effective treatments have been developed over the years to address depression and anxiety, the lack of personalization and inability to adapt to patient needs or preferences contributes to poor treatment adherence and outcomes. An intelligent treatment system was developed that uses state of the art machine learning approaches within a mobile intervention application to treat MDD and anxiety. Machine learning, a branch of artificial intelligence, focuses on the development of algorithms that automatically improve and evolve based on collected data. The intervention, called IntelliCare, uses a mobile application to continuously collect patient data and adapt intervention content and motivational messaging to create a highly tailored and user-responsive treatment system. During the RCT study, 270 participants with Major Depression and/or Anxiety will use IntelliCare apps for up to 8 weeks and will be invited to complete online questionnaires at four follow-up time points: weeks 4 and 8 in the study program, and months 3 and 6 after the study program ends. All participants will first undergo initial assessments that will include a telephone interview and a series of online questionnaires about their mood. Eligible participants will receive up to 8 weeks of access to the IntelliCare system, which consists of apps with lessons and tools designed to teach skills for mood management. It is suggested that participants utilize the mobile phone tools every day. Participants will be randomly assigned to one of two versions of the IntelliCare Hub App, which is the central delivery system for the IntelliCare program, and participants will also be randomly assigned to get support from a human coach or use the IntelliCare program independently. Each week, participants assigned to work with a coach will receive a brief motivational intervention. This coach will also be available to participants via email throughout the 8 week study. Participants randomly assigned to use the IntelliCare program independently will receive an on-boarding guide to set up the IntelliCare Hub App on their phone, and participants will have access to our tech support team if technical difficulties arise while using the program. Data from the RCT study will be used to examine whether: the IntelliCare program is an effective intervention for reducing depressive and anxiety symptoms; the recommender system will produce greater program adherence and; support from a coach or independent use might affect user experience and adherence. Data collected will also be used to further develop and evaluate machine learning methods for future research and deployment efforts.