Artificial intelligence could help early detection of mental health problems among hospital staff, according to research conducted by New York University’s Grossman School of Medicine.
WHY IT MATTERS
The study, led by researchers at NYU Langone Health, shows how AI could detect signs of mental stress in healthcare professionals. The research could hold great promise for the future of AI-based screening efforts in the mental health field.
The research, published this week in the Journal for medical internet research AIdelved into transcripts of virtual psychotherapy involving more than 800 healthcare professionals, including physicians, nurses, and emergency medical personnel.
For comparison, transcripts are included from 820 individuals who received psychotherapy during the first wave of COVID-19 in the United States but were not part of the healthcare workforce.
The findings, published online in the Journal of Medical Internet Research AI, revealed that among healthcare professionals, those who discussed topics such as their experiences in hospital wards, sleep deprivation or mood problems during therapy sessions were more likely to be diagnosed with anxiety and mood disorders. depression.
The research identified four distinct topics of conversation related to healthcare workers, including virus-related fears and experiences on hospital floors and ICU units.
Although the increased risk of anxiety and depression among health professionals discussing their experiences was relatively small at 3.6%, the model is expected to improve its ability to identify problems as more data becomes available.
According to lead author Matteo Malgaroli, research assistant professor in the Department of Psychiatry at NYU Langone Health, this represents a significant advance in mental health screening for healthcare professionals.
THE BIG TREND
The use of AI could potentially have a role for a broader population, highlighting the effectiveness of AI technology in supporting mental health care, with natural language processing potentially evolving into a screening tool for detecting and monitoring anxiety and depression symptoms.
Burnout is growing among physician assistants and those in the medical profession at large, as they battle growing data volumes and a shrinking workforce.
According to a recent national survey of nurses, 100,000 nurses have left their jobs during the pandemic, and by 2027, nearly 900,000, or nearly a fifth of the 4.5 million registered nurses nationwide, plan to do the same, increasing the overall national healthcare system is at risk. if no action is taken.
In addition to using AI to detect stress symptoms, the technology can also be used to identify deficiencies in clinical workflow and create a strategy for improvements.
For example, Doximity’s AI-powered chatbot tool could reduce the time doctors spend on administrative burdens.
In July 2022, a health technology project in Hong Kong received $5 million in public funding to introduce an AI-based data-driven approach to mental health diagnosis and treatment.
Changi General Hospital has already deployed AI to detect curable hypertension, with its hypertension-focused laboratory with Shimadzu reducing the turnaround time of primary aldosteronism testing, which was previously done abroad.
Talkspace uses language regression to analyze anonymized behavioral health messages and alert healthcare providers to patients at risk of self-harm, illustrating AI’s potential to detect additional mental health issues on the patient’s side.
ON THE RECORD
“Our findings show that those who worked on the hospital floor during the most intense moment of the pandemic faced unique challenges in addition to their regular work-related stressors, putting them at high risk for serious mental health issues,” Malgaroli said in a statement.
“These results suggest that natural language processing may one day become an effective screening tool for detecting and monitoring anxiety and depression symptoms,” added senior author psychiatrist Naomi Simon, MD, a professor in the Department of Psychiatry at NYU Langone.
Nathan Eddy is a healthcare and technology freelancer based in Berlin.
Email the writer: nathaneddy@gmail.com
Twitter: @dropdeaded209