Credit score: Pixabay/CC0 Public Area
A synthetic intelligence (AI)-driven screening software efficiently recognized hospitalized adults in danger for opioid use dysfunction and beneficial referral to inpatient dependancy specialists.
The AI-based methodology was simply as efficient as a well being provider-only method in initiating dependancy specialist consultations and recommending monitoring of opioid withdrawal. In comparison with sufferers who acquired provider-initiated consultations, sufferers with AI screening had 47% decrease odds of being readmitted to the hospital inside 30 days after their preliminary discharge.
This discount in readmissions translated to a complete of almost $109,000 in estimated well being care financial savings in the course of the research interval.
The research, revealed in Nature Drugs, studies the outcomes of a accomplished medical trial, demonstrating AI’s potential to have an effect on affected person outcomes in real-world well being care settings. The research suggests funding in AI could also be a promising technique particularly for well being care methods searching for to extend entry to dependancy remedy whereas enhancing efficiencies and saving prices.
“Addiction care remains heavily underprioritized and can be easily overlooked, especially in overwhelmed hospital settings where it can be challenging to incorporate resource-intensive procedures such as screening,” mentioned Nora D. Volkow, M.D., director of NIH’s Nationwide Institute on Drug Abuse (NIDA). “AI has the potential to strengthen the implementation of addiction treatment while optimizing hospital workflow and reducing health care costs.”
In a medical trial, researchers on the College of Wisconsin College of Drugs and Public Well being, Madison, in contrast physician-led dependancy specialist consultations to the efficiency of their AI screening software, which had been developed and validated in prior work.
Researchers first measured the effectiveness of provider-led consultations at College Hospital in Madison, Wisconsin, between March to October 2021 and March to October 2022, whereby well being care suppliers carried out advert hoc dependancy specialist consultations for opioid use dysfunction. They then carried out the AI screening software between March to October 2023 to help the well being care suppliers and remind them all through hospitalization of a affected person’s want for an dependancy specialist’s care.
From begin to end, the trial screened 51,760 grownup hospitalizations, with 66% occurring with out deploying the AI screener and 34% with the AI screener deployed hospital-wide. A complete of 727 dependancy drugs consultations had been accomplished in the course of the research interval.
The AI screener was constructed to acknowledge patterns in information, like how our brains course of visible info. It analyzed info inside all of the documentation out there within the digital well being information in actual time, similar to medical notes and medical historical past, to establish options and patterns related to opioid use dysfunction.
Upon identification, the system issued an alert to suppliers once they opened the affected person’s medical chart with a suggestion to order dependancy drugs session and to observe and deal with withdrawal signs.
The trial discovered that AI-prompted session was simply as efficient as provider-initiated session, guaranteeing no lower in high quality whereas providing a extra scalable and automatic method. Particularly, the research confirmed that 1.51% of hospitalized adults acquired an dependancy drugs session when well being care professionals used the AI screening software, in comparison with 1.35% with out the help of the AI software.
Moreover, the AI screener was related to fewer 30-day readmissions, with roughly 8% of hospitalized adults within the AI screening group being readmitted to hospital, in comparison with 14% within the conventional provider-led group.
The discount in 30-day readmissions nonetheless held after accounting for sufferers’ age, intercourse, race and ethnicity, insurance coverage standing, and comorbidities, as calculated by way of an odds ratio. When analyzing the outcomes utilizing the percentages ratio, the researchers estimated a lower of 16 readmissions by using the AI screener.
A subsequent cost-effectiveness evaluation indicated a web value of $6,801 per readmission averted for the affected person, well being care insurer, and/or the hospital. This amounted to an estimated complete of $108,800 in well being care financial savings for the eight-month research interval wherein the AI screener was used, even after accounting for the prices of sustaining the AI software program. The typical value of a 30-day hospital readmission is at present estimated at $16,300.
“AI holds promise in medical settings, but many AI-based screening models have remained in the development phase, without integration into real-world settings,” mentioned Majid Afshar, M.D., lead creator of the research and affiliate professor on the College of Wisconsin-Madison. “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.”
Whereas the AI screener confirmed robust effectiveness, challenges stay, together with potential alert fatigue amongst suppliers and the necessity for broader validation throughout totally different well being care methods. The authors additionally be aware that whereas the assorted research intervals—spanning a number of years—had been seasonally matched, the evolving nature of the opioid disaster could have launched residual biases. Future analysis will give attention to optimizing the AI software’s integration and assessing its longer-term influence on affected person outcomes.
The opioid disaster continues to pressure well being care methods within the U.S., with emergency division admissions for substance use rising by almost 6% between 2022 to 2023 to an estimated 7.6 million. Opioids are the second main trigger of those visits after alcohol, however screening for opioid use dysfunction in hospitals stays inconsistent. Because of this, hospitalized sufferers with opioid use dysfunction continuously depart the hospital earlier than seeing an dependancy specialist, an element linked to a tenfold improve in overdose charges.
AI know-how has emerged as a novel, scalable software to probably overcome these boundaries and enhance alternatives for early intervention and linkage to medicines for opioid use dysfunction, however extra analysis is required to grasp how AI can be utilized successfully in well being care settings.
Extra info:
Majid Afshar et al, Scientific implementation of AI-based screening for threat for opioid use dysfunction in hospitalized adults, Nature Drugs (2025). DOI: 10.1038/s41591-025-03603-z. www.nature.com/articles/s41591-025-03603-z
Offered by
Nationwide Institutes of Well being
Quotation:
AI screening for opioid use dysfunction related to fewer hospital readmissions (2025, April 4)
retrieved 4 April 2025
from https://medicalxpress.com/information/2025-04-ai-screening-opioid-disorder-hospital.html
This doc is topic to copyright. Other than any honest dealing for the aim of personal research or analysis, no
half could also be reproduced with out the written permission. The content material is offered for info functions solely.
