CHAMPAIGN, Ill. — New work led by an Urbana-Champaign University of Illinois researcher harnesses the power of generative artificial intelligence, using it in tandem with metrics-based models of care and access to care in a simulated case study, creating a new framework that promotes personalized mental health treatment, addresses common barriers to access, and improves outcomes for diverse people.
Social work teacher Cortney VanHook led the research, in which he and his co-authors used generative AI to simulate the mental health journey of a fictional client named “Marcus Johnson,” a composite of a young, middle-class Black man with depressive symptoms who navigates the health care system in Atlanta, Georgia.
In response to the researcher’s requests, the AI platform created a detailed case study and treatment plan for the fictional client. Based on the personal details the researchers used in the prompt, the AI platform examined the simulated client’s protective factors, such as their supportive family members and their potential barriers to care. – including gender-specific cultural and family expectations – as well as her concerns about receiving culturally competent treatment due to the shortage of Black male providers in her employer-sponsored health plan’s network.
Real-world simulations allow practitioners to understand individuals’ pathways to mental health care, common access issues and demographic disparities, VanHook said.
Additionally, using a simulated client alleviates concerns about violating patient confidentiality laws, allowing practitioners, trainees and students to explore and refine potential interventions in a low-risk environment, thereby promoting more equitable, responsive and effective mental health systems, he said.
Graphic by Cortney VanHook and Daniel Abusuampeh
“What’s unique about this work is that it’s practical and evidence-based,” VanHook said. “It goes from just theory to the actual use of AI in mental health care. I see this framework applying to educating students about populations they may not be familiar with but will come into contact with in the field, as well as its use by supervisors in the field when they are training their students or by clinicians on how to understand and better support clients who come into their settings.”
VanHook and co-authors Daniel Abusuampeh from the University of Pittsburgh and Jordan Pollard from the University of Cincinnati prompted the AI platform to apply three evidence-based theoretical frameworks to create its simulated case study and treatment plan for the virtual client.
AI software was asked to use Andersen’s behavioral model – a theory about the factors that determine individuals’ use of health services – examine personal, cultural, and systemic factors that supported or hindered the client’s use of mental health services. Additionally, the proposed treatment plans incorporated theory on the five components of access to assess availability, accessibility, accommodation, affordability, and acceptability of care for the client; as well as measurement-based care, a clinical approach that applies standardized and reliable measurements for ongoing monitoring of the client’s symptoms and functioning.
The team used metrics-based care to refine the treatment approaches recommended by AI. To ensure that AI-generated simulation reflects real-world clinical practice, VanHook and Pollard – who are both licensed mental health professionals – reviewed the proposed treatment plan for clinical accuracy and compared the record to published research findings.
As all three authors identify as Black men, they confirmed the material’s cultural sensitivity and conceptualization of the barriers that Black men often face in the American mental health system.
“Each population – regardless of race, age, gender, nationality and ethnicity – has a unique mental health care journey, and AI contains a lot of information to understand different populations and how they interact with the mental health field. AI has the ability to account for complex barriers as well as facilitate population-wide mental health care,” VanHook said.
The authors acknowledged that content generated by AI technology will be limited by the data and models contained in the platform’s training set, which may not reflect the diversity, unpredictability, or emotional nuances of clinical encounters. Likewise, despite the evidence-based frameworks that were applied in the project, VanHook said these do not address all of the systemic and structural barriers faced by Black men nor capture all of the social, cultural or individual factors that influence clients’ care.
Nonetheless, the team maintained in the paper published in the journal Frontiers in Health Services that generative AI holds great promise for improving access, cultural competence, and client outcomes in mental health care when integrated with evidence-based models.
“AI is a train that’s already rolling, and it’s picking up speed. So the question is: How can we use this amazing tool to improve mental health care for many populations? I hope it will be used in the field, as a teaching tool, and within higher-level management and administration when it comes to mental health services,” VanHook said.
In August, Illinois Governor JB Pritzker signed a new law: The Well-being and Monitoring of Psychological Resources Act – which limits the use of AI in mental health care “to administrative and additional support services” provided by licensed behavioral health professionals. The new policy follows reports of young Americans committing suicide after interactions with AI chatbots.
“Using AI in the manner used in our study is consistent with new state law if it is used in the education and clinical supervision process,” VanHook said. “The described process based on the measure may blur the lines, so I urge caution against its use beyond the purposes of education and clinical supervision until we receive more guidance from the state.”