Press release

Thursday, April 3, 2025

Faced with the flu, health authorities call for “great vigilance” and vaccination
Read moreFaced with the flu, health authorities call for "great vigilance" and vaccination

An NIH-supported clinical trial shows that an AI tool is as effective as healthcare providers at generating referrals to addiction specialists.

An artificial intelligence (AI)-based screening tool, developed by a National Institutes of Health (NIH)-funded research team, successfully identified hospitalized adults at risk for opioid use disorder and recommended referrals to inpatient addiction specialists. The AI-based method was just as effective as a provider-only approach in initiating consultations with an addiction specialist and recommending follow-up for opioid withdrawal. Compared with patients who received provider-initiated consultations, patients who received AI screening were 47% less likely to be readmitted to the hospital within 30 days of their initial discharge. This reduction in readmissions translated into a total of nearly $109,000 in estimated healthcare savings over the study period.

The study, published in Nature Medicine, presents the results of a completed clinical trial, demonstrating the potential of AI to influence patient outcomes in real-world healthcare settings. The study suggests that investing in AI could be a promising strategy, particularly for healthcare systems seeking to increase access to addiction treatment while improving efficiency and reducing costs.

“Addiction care remains largely under-prioritized and can be easily overlooked, especially in overcrowded hospital settings where it can be difficult to integrate resource-intensive procedures such as screening,” said Nora D. Volkow, MD, director of the NIH’s National Institute on Drug Abuse (NIDA). “AI has the potential to strengthen the implementation of addiction treatment while optimizing hospital workflow and reducing healthcare costs.”

In a clinical trial, researchers from the University of Wisconsin School of Medicine and Public Health, Madison, compared physician-led addiction specialist visits to the performance of their AI screening tool, which had been developed and validated in previous work. The researchers first measured the effectiveness of provider-led visits at University Hospital in Madison, Wisconsin, between March and October 2021 and March and October 2022, during which healthcare providers conducted one-time addiction specialist visits for opioid use disorder. They then deployed the AI screening tool between March and October 2023 to assist healthcare providers and remind them, throughout the hospitalization, of the need to see an addiction specialist. From start to finish, the trial examined 51,760 adult hospitalizations, including 661 non-AI screening tool deployments and 341 hospital-wide deployments. A total of 727 addiction consultations were conducted during the study period.

The AI screening tool was designed to recognize patterns in data, similar to how our brains process visual information. It analyzed information from all available documents in electronic medical records, such as clinical notes and medical histories, in real time to identify characteristics and patterns associated with opioid use disorder. Once identified, the system sent an alert to healthcare professionals when the patient's medical record was opened, recommending that they order an addiction consultation and monitor and treat withdrawal symptoms.

The trial found that the AI-guided consultation was just as effective as the provider-initiated consultation, ensuring consistent quality while offering a more scalable and automated approach. Specifically, the study showed that 1.51 % of hospitalized adults received a substance use consultation when providers used the AI screening tool, compared to 1.35 % without the aid of the AI tool. Additionally, the AI screening tool was associated with a decrease in 30-day readmissions, with approximately 8 % of hospitalized adults in the AI screening group being readmitted to the hospital, compared to 14 % in the traditional provider-led group.

The reduction in 30-day readmissions was maintained after accounting for patient age, sex, race and ethnicity, insurance status, and comorbidities, as calculated via an odds ratio. Analyzing the results using the odds ratio, the researchers estimated a decrease of 16 readmissions using the AI screening tool. A subsequent cost-effectiveness analysis indicated a net cost of $6,801 per avoided readmission for the patient, health insurer, and/or hospital. This represents an estimated total of $108,800 in healthcare savings for the eight-month study period during which the AI screener was used, even after accounting for AI software maintenance costs. The average cost of a 30-day hospital readmission is currently estimated at 16,300 $.

“AI holds promise in the medical field, but many AI-based screening models have remained in the development phase, without real-world integration,” said Majid Afshar, MD, lead author of the study and associate professor at the University of Wisconsin-Madison. “Our study represents one of the first demonstrations of an AI screening tool integrated into addiction medicine and hospital workflows, highlighting the pragmatism and real-world potential of this approach.”

Although the AI screener demonstrated strong effectiveness, challenges remain, including potential alert fatigue among providers and the need for broader validation across different healthcare systems. The authors also note that although the different study periods—spanning several years—were seasonally matched, the evolving nature of the opioid crisis may have introduced residual bias. Future research will focus on optimizing the integration of the AI tool and assessing its long-term impact on patient outcomes.

The opioid crisis continues to strain healthcare systems across the United States, with emergency room admissions for substance use increasing by nearly 6 % between 2022 and 2023 to reach an estimated number of 7.6 million. Opioids are the second cause of these visits after alcohol, but screening for opioid use disorder in hospitals remains inconsistent. As a result, hospitalized patients with opioid use disorder often leave the hospital before seeing an addiction specialist, a factor linked to a tenfold increase overdose rates. AI technology has emerged as an innovative and scalable tool to potentially overcome these barriers and improve opportunities for early intervention and connection with medications for opioid use disorder , but further research is needed to understand how AI can be used effectively in healthcare settings.If you or someone you know is struggling or in crisis, help is available. Call or text

988 or chat on988lifeline.org . To learn how to get help for mental health, drug, or alcohol problems, visitFindSupport.gov. If you are ready to find a treatment center or provider, you can go directly toFindTreatment.govor call 800-662-HELP (4357).About the National Institute on Drug Abuse (NIDA):

NIDA is a component of the National Institutes of Health, part of the U.S. Department of Health and Human Services. NIDA supports most of the world's research on the health aspects of drug use and addiction. The Institute conducts a wide variety of programs to inform policy, improve practice, and advance the science of addiction. For more information about NIDA and its programs, visit www.nida.nih.gov .About the National Institutes of Health (NIH):

The NIH, the nation's medical research agency, comprises 27 institutes and centers and is a component of the U.S. Department of Health and Human Services. The NIH is the primary federal agency conducting and supporting basic, clinical, and translational medical research. It studies the causes, treatments, and cures for common and rare diseases. For more information about the NIH and its programs, visit www.nih.gov .NIH…Transforming Discovery into Health

®###