May 13, 2025

An artificial intelligence (AI)-based screening tool was more effective than traditional in-person screening and addiction consultation at preventing hospital readmissions within 30 days for hospitalized adults with opioid use disorder (OUD), a study has found. The AI-screening tool was as effective as a screening conducted by a clinical professional at identifying those with OUD and initiating consultation with an addiction specialist. People screened by the AI tool were 47% less likely to be readmitted during the first 30 days of discharge. About 8% of hospitalized adults who received the AI screening were readmitted to the hospital, compared to 14% who received traditional screening by a clinical professional.

The AI screening tool was tested from March to October 2023 at the University Hospital in Madison, Wisconsin. The outcomes of hospital-wide AI screening were compared to ad-hoc in-person screenings and consultations for OUD conducted between March 2021 and October 2022. The AI screening tool reminded the hospital staff of the consumer’s need for an addiction specialist consultation for OUD. During the post-intervention period, the AI screener served as the intervention, identifying patients at high risk for OUD and providing a recommendation for consultation with the addiction medicine service, along with prompting the initiation of the Clinical Opiate Withdrawal Scale and order set. Over the three study years, 51,760 adult hospitalizations resulted in an OUD screening; 66% were screened without the AI screener, and 34% took place after the AI screener was deployed hospital-wide.

The AI-based method was just as effective as a health provider-only approach in initiating addiction specialist consultations and recommending monitoring of opioid withdrawal. A total of 1.51% of hospitalized adults received an addiction medicine consultation when health care professionals used the AI screening tool, compared to 1.35% without the assistance of the AI tool.

The AI screener avoided a net cost of $6,801 per readmission for the consumer, health care insurer, and/or the hospital. This amounted to an estimated $108,800 in health care savings for the eight-month study period in which the AI screener was used, even after accounting for the costs of maintaining the AI software. The average price of a 30-day hospital readmission is currently estimated at $16,300.

These findings were reported in Clinical Implementation Of AI-Based Screening For Risk For Opioid Use Disorder In Hospitalized Adults by Majid Afshar, Felice Resnik, Cara Joyce, Madeline Oguss, and colleagues. The goal was to evaluate whether an AI-driven OUD screener embedded in the EHR was non-inferior to usual care in identifying consumers for addiction medicine consultations, aiming to provide a similarly effective but more scalable alternative to human-led ad hoc consultations.

The full text of Clinical Implementation Of AI-Based Screening For Risk For Opioid Use Disorder In Hospitalized Adults was published April 3, 2025, by Nature Medicine. A free abstract is available online at https://www.nature.com/articles/s41591-025-03603-z (accessed May 1, 2025).

More information about the study methodology was reported at https://clinicaltrials.gov/study/NCT05745480 (accessed May 1, 2025).

For more information, contact: Majid Afshar, M.D., Associate Professor, Pulmonary and Critical Care Medicine, Department of Medicine, University of Wisconsin–Madison, 610 Walnut Street, Warf Office Building, Suite 500, Madison, Wisconsin 53726-2336; Email: majid.afshar@wisc.edu; Website: https://www.medicine.wisc.edu/directory/afshar_majid