Researchers at the University of Wisconsin-Madison have developed an innovative AI tool designed to identify hospital patients at risk of opioid use disorder, which is critical as the opioid crisis continues to strain U.S. healthcare systems.

Emergency department admissions for substance use have surged by nearly 6% from 2022 to 2023, totaling an estimated 7.6 million admissions, underscoring the urgent need for effective interventions.

The AI screener analyzes electronic health records in real-time to detect patterns associated with opioid use disorder, prompting alerts for consultations and monitoring, which has resulted in a slight increase in addiction consultations from 1.35% to 1.51% of hospitalized adults.

In a trial involving 51,760 hospitalizations, the AI tool facilitated 727 addiction medicine consultations, demonstrating that AI-prompted consultations were as effective as those initiated by healthcare providers.

Results indicated that 8% of patients in the AI group were readmitted within 30 days, compared to 14% in the traditional group, suggesting a significant reduction in readmissions.

The AI screening tool has been linked to a decrease in hospital readmissions, suggesting it effectively identifies patients who could benefit from specialty addiction care.

Dr. Majid Afshar, the lead author of the study published in Nature Medicine, asserts that the AI method is as effective as traditional health provider-only approaches, enhancing efficiency by reducing human error.

This tool not only recommends referrals to addiction specialists but also operates with effectiveness comparable to traditional methods, highlighting its potential in improving patient outcomes.

Despite its effectiveness, challenges such as provider alert fatigue and the need for broader validation across healthcare systems remain, which could hinder widespread adoption.

Dr. Afshar emphasizes that investing in AI could improve access to addiction treatment and save costs for healthcare systems, making it a promising avenue for future healthcare strategies.

A cost-effectiveness analysis revealed a net cost of $6,801 per readmission avoided, contributing to significant savings despite the ongoing costs associated with maintaining the AI software.

Future research will focus on optimizing the integration of AI tools in healthcare settings and evaluating their long-term impact on patient outcomes in addiction care.