NIH grant to fund depression chatbot for Black patients

Researchers from George Mason University received a $70,906 grant from the National Institutes of Health to fund the development and validation of an antidepressant recommendation chatbot in Black and African American populations.

The chatbot will be built using an existing tool, created earlier this year by a George Mason University (GMU) team looking to match patients to antidepressants that fit their needs more effectively. The original tool has shown promise in making treatment recommendations for 16,775 general-population patient subgroups, and the new project will assess the appropriateness of the platform's recommendations for Black and African American cohorts.

The project -- which the research team believes is the first of its kind -- will use information from the National Institutes of Health (NIH) "All of Us" program database.

"Antidepressant medications are a first-line treatment for depression; however, a majority of depressed patients do not experience improvement with their first antidepressant. Additionally, minority populations, including Black and African Americans, are not well represented in antidepressant studies, contributing to reduced antidepressant effectiveness in these populations," explained Farrokh Alemi, Ph.D., project lead and professor of Health Informatics at GMU.

"There is a significant need to synthesize available evidence regarding antidepressant effectiveness and provide personalized treatment recommendations, and this project addresses a major gap in the management of Black and African Americans with depression," Alemi continued.

Alemi's team will develop a natural language processing tool called a Knowledge-enhanced Antidepressant Recommendation Dialogue System to engage in back-and-forth conversations with chatbot users. During the conversation, the AI will gather patient information and preferences, which will then be analyzed to identify potential treatments.

From there, the tool is designed to generate a list of recommended antidepressants, relevant studies about each medication and an explanation of its recommendations. The generative AI will share these details with the patient's clinician in a short point-of-care recommendation, with options to review the conversation history and supporting evidence.

Following development, the conversational AI will be tested in a group of Black and African American patients to assess its functionality and gather feedback on user preferences. The project also aims to train a doctoral or master's student from one of these groups to expand the healthcare AI workforce.

"Chatbots -- or patient-facing dialogue systems like the one we will create -- hold transformative potential for the healthcare sector and are increasingly prominent in psychiatric applications, predominantly through therapy-bot implementations," Alemi said. "Our chatbot will help improve the detailed, time-consuming, medical history intake process, and provide point-of-care summary and prescription recommendations to the patients' clinicians. The chatbot will make patients more comfortable because the natural language modality provides an intuitive, empathetic, stigma-free interface."

The funding is part of NIH's Artificial Intelligence/Machine Learning Consortium to Advance Health Equity and Researcher Diversity program, which encourages the increased participation of underrepresented communities and researchers in the development of AI and machine learning technologies.

Shania Kennedy has been covering news related to health IT and analytics since 2022.