AI Screening Tool Effective in Identifying Opioid Use Disorder and Referring Patients to Treatment
- Marcus Rucker
- 2 days ago
- 2 min read
A recent study published in Nature Medicine demonstrated that an artificial intelligence (AI)-driven screening tool was just as effective as a healthcare provider-only approach in identifying hospitalized adults at risk for opioid use disorder (OUD). The research found that the AI screening significantly reduced hospital readmissions and assisted in the referral of patients to inpatient addiction specialists. Supported by the National Institutes of Health (NIH), the study found that patients screened using the AI tool had 47% lower odds of being readmitted within 30 days compared to those who received traditional provider-initiated consultations. This automation preserved the quality of care while saving an estimated $109,000 in healthcare costs during the eight-month trial period at the University of Wisconsin Hospital.
“AI holds promise in medical settings, but many AI-based screening models have remained in the development phase, without integration into real-world settings,” said Majid Afshar, lead author and principal investigator of the study and associate professor of medicine at the UW School of Medicine and Public Health. “Our study represents one of the first demonstrations of an AI screening tool embedded into addiction medicine and hospital workflows, highlighting the pragmatism and real-world promise of this approach.”
The clinical trial screened 51,760 adults for OUD, with 34% screened with the AI tool. Conducted by researchers at the University of Wisconsin School of Medicine and Public Health–Madison, researchers compared traditional physician-led addiction consultations with the use of an AI-powered screening tool for OUD. A total of 727 addiction medicine consultations were conducted during the study period. Only 8% of patients in the AI group were readmitted within 30 days, compared to 14% in the provider-only consultation groupThe AI model, integrated into electronic health records, analyzed real-time data, including clinical notes and patient histories, to prompt addiction medicine consults and monitor withdrawal symptoms.
The article was co-authored by a multidisciplinary team of researchers and clinicians from the University of Wisconsin–Madison, including Dr. Majid Afshar, Dr. Felice Resnik, Dr. Cara Joyce, Madeline Oguss, Dr. Dmitriy Dligach, Dr. Elizabeth S. Burnside, Dr. Anne Gravel Sullivan, Dr. Matthew M. Churpek, Dr. Brian W. Patterson, Dr. Elizabeth Salisbury-Afshar, Dr. Frank J. Liao, Cherodeep Goswami, Dr. Randy Brown, and Dr. Marlon P. Mundt. Together, the team brought expertise across hospital medicine, informatics, addiction treatment, and health services research to evaluate a pragmatic, AI-powered intervention to improve care for patients with OUD.
Reference
Afshar, M., et al. (2025). Clinical implementation of AI-based screening for risk for opioid use disorder in hospitalized adults. Nature Medicine. DOI: 10.1038/s41591-025-03603-z