Until Dr. Shiri Sharvit, chief clinical officer at Eleos Health, began incorporating artificial intelligence into her work as a mental health clinician, her companion was a post-it note. However, these sticky notes were sometimes lost or spilled water.
then came Eleosis an AI solution that improves the workflow of behavioral clinicians, she said. Although she incorporates it into her work as a part-time clinician, she has struggled to get others to use it.
“When I tried to pass it on to my colleagues, and when I gave training on how to incorporate technology into clinical practice on behalf of the American Psychological Association, I felt that people still didn’t get it,” she said. Sharvit has been invited several times to speak at the American Psychological Association.
This prompted Sharvit and Katherine C. Kellogg, Professor of Management and Innovation at the MIT Sloan School of Management, to write a paper on how to implement AI into the workflow of mental health clinicians and the challenges and solutions in doing so. I wrote
Kellogg said they feel clinicians have historically been left out of the conversation about AI technology in mental health.
“I think there is a lot of research being done on the social impact of AI and how organizations can adopt it to improve quality, reduce costs and increase revenue,” said Kellogg. I’m here. “However, little research has been done on the benefits of AI technology for frontline healthcare providers, the challenges it poses in their day-to-day operations, and how healthcare leaders and technology developers can help address these challenges. has not been done.”
of Peer-reviewed paperswas published in Frontiers in Psychology on Sept. 6 and focuses on three types of AI solutions: automation technology, patient engagement technology, and clinical decision support technology. Automation technologies computerize structured or semi-structured tasks, such as screening and managing digital surveys. Patient engagement technologies are solutions such as conversational agents and chatbots that can provide coaching and scripted therapy. Clinical decision support technology uses machine learning algorithms to make predictions based on historical data. Examples of predictable conditions include depression, anxiety, stroke, and suicidal ideation.
However, researchers admit that the use of these AI technologies is not without its challenges. Difficulties in implementing AI include fear of being displaced, lack of evidence-based evaluation studies, and concerns about negative impacts on clinician-patient relationships.
“Currently, clinicians don’t have many use cases and role models to actually use AI,” Sharvit said. “We wanted them to feel a little more confident about exploring digital tools, and explained that digital tools are meant to enhance, improve, and complement practices. I can’t.”
There are also solutions to some of these challenges. For example, exposing clinicians to the possibility of failure, training them in computational thinking, and engaging them in the technology development process. If some aspect of the AI technology is inaccurate, developers should work with clinicians to determine how to improve.
But more research needs to be done, Kellogg said. She identifies leading organizations using AI for mental health to determine what the best practices are. However, she can’t name the organization just yet because she will start these studies soon. Clinicians should be at the center when it comes to developing AI solutions, she added, but collaboration with others in the field, including technology developers, executives and researchers, is needed.
“When we’re all working together, don’t forget these frontline clinicians,” Kellogg said. “They are the key to making these technologies work in the real world.”
For Sharvit, the difference between using AI and sticky notes in her work is something like “navigating a place with a map versus using Google Maps or Waze on your phone.” AI can help reduce stress during client sessions, improve note-taking, and recall conversations from previous sessions, she said.
“That’s a definite difference,” says Sharvit. “First of all, I have an AI buddy. I’m not the only one in the session.”
Photo: metamorworks, Getty Images
.
Comments
Post a Comment