The use of digital tools in the field of mental health, including applications assisted by artificial intelligence (AI) algorithms, was the subject of a free seminar at the IECS headquarters with the participation of three renowned international speakers. The activity took place within the framework of COLMA (CO-morbid Long-term disease Management in the Americas: blended intervention), an international project focused on the management of comorbid chronic diseases (hypertension, diabetes and depression) in vulnerable populations in Peru and Argentina, as stated in the introductory remarks by the Dr. Vilma Irazola, director of the Department of Research in Chronic Diseases at IECS, who is the principal investigator of that initiative in Argentina.
Dr. David Mohr, Professor of Preventive Medicine and Director of the Center for Behavioral Intervention Technologies at Northwestern University in Chicago, presented his experiences applying digital mental health tools to patients with depression in a rural South Carolina community and to Northwestern Medicine affiliates who were offered this option while on the waiting list for psychotherapy. Preliminary results show that the approach is effective and cost-effective for the healthcare system.
“Digital mental health tools are effective in the real world and can be implemented and distributed at scale. But integration within the healthcare system is difficult and costly, requiring stronger collaboration between academia, industry, providers, and funders,” he concluded.
Dr. Pepijn van de Ven, Professor of Artificial Intelligence and Machine Learning and Director of the Data-Driven Computer Engineering research group at the University of Limerick, Ireland, spoke about various emerging applications of AI and machine learning in the field of mental health, including identifying patients at higher risk of depression, shortening or personalizing mental health questionnaires, supporting clinical decisions, and health education.
However, when it comes to using AI for treatment, complementing or even replacing the functions of human therapists, Dr. van de Ven contrasted the potential advantages (e.g., improved screening, connecting with vulnerable populations, and affordable scaling) with the negative aspects: catastrophic advice, trust in chatbots, inequity due to biases (algorithms trained on populations that do not include minorities), and cost of access.
“The discomfort with AI is not unjustified,” acknowledged Dr. van de Ven. “But it’s crucial to understand that not all AI models are generative and that the risk depends on how the model is used. And there are many applications that carry very little risk.”
Finally, Dr. Marcia Scazufca, a researcher at the Hospital de Clínicas of the Faculty of Medicine of the University of São Paulo, in Brazil, presented results of two experiences of psychosocial interventions in older adults with depression, one of which (VIVA VIDA) consisted of a self-guided digital program that was deployed through 48 educational and behavioral activation messages via WhatsApp over 6 weeks.
The program was tested in a randomized clinical trial in more than 600 participants, with good operational performance and “promising results in terms of technology adherence and engagement, with weekly message viewing rates of 80-90% and short-term effectiveness [measured by improvements in the Patient Health Questionnaire (PHQ-9)],” said Dr. Scazufca.