A new study published in Nature Medicine has found that an artificial intelligence (AI) screening tool can be as effective as healthcare providers in identifying hospitalized adults at risk for opioid use disorder and recommending referrals to addiction specialists. The AI-based method successfully reduced hospital readmissions by 47% compared to patients who received provider-initiated consultations.
Developed by a National Institutes of Health (NIH)-funded research team, the AI screening tool analyzed electronic health records in real-time to identify features and patterns associated with opioid use disorder. Upon identification, the system issued an alert to providers recommending addiction medicine consultation and monitoring of withdrawal symptoms.
The study, which screened 51,760 adult hospitalizations over an eight-month period, found that patients who received AI-prompted consultations had a lower rate of 30-day readmissions compared to those who received provider-initiated consultations. The AI screener was also associated with estimated healthcare savings of nearly $109,000 during the study period.
The researchers suggest that investing in AI may be a promising strategy for healthcare systems seeking to increase access to addiction treatment while improving efficiencies and saving costs. However, challenges remain, including potential alert fatigue among providers and the need for broader validation across different healthcare systems.
Source: https://www.nih.gov/news-events/news-releases/ai-screening-opioid-use-disorder-associated-fewer-hospital-readmissions