Key Points

  • An NIH-funded AI tool reduced 30-day hospital readmissions among people with opioid use disorder by 47%, demonstrating that automated screening can close real treatment gaps in overwhelmed hospital settings.
  • The trial covered nearly 52,000 adult hospitalizations in Madison, Wisconsin, and estimated nearly $109,000 in healthcare savings over eight months.
  • A new $3.5 million NIDA grant at WashU Medicine will use mouse models to study how prenatal opioid exposure affects long-term brain development, comparing two standard treatment approaches. This is early preclinical research; no clinical conclusions should be drawn yet.
  • Medication-assisted treatment with buprenorphine or methadone during pregnancy is the established standard of care. Abruptly stopping opioids poses serious risks to both the pregnant person and the pregnancy.
  • A 2026 study found that medical students who screened positive for probable gambling disorder reported recent suicidal thoughts at more than double the rate of their peers, with odds roughly five times higher after adjusting for other factors.
  • Gambling disorder responds to therapies like cognitive behavioral therapy and often co-occurs with anxiety, depression, or a substance use disorder. Simple two-question screens can open the door to treating all of it together.

Addiction News Weekly Episode 1.6

In This Episode:

Episode Transcript

Welcome to Addiction News Weekly by Rehab.com, where we break down the biggest stories in addiction, recovery, and public health. This week the thread is detection, finding risk earlier, and in new ways, such as a new AI tool that spots opioid use disorder in hospitals, a new federal study on what opioid exposure does to a developing brain, and research linking gambling screening to lower suicide risk. Let’s get into it.

AI Screening Cuts Opioid Readmissions by 47% in Hospital Trial

First, we start with artificial intelligence in the hospital. In a clinical trial published in Nature Medicine and funded by the National Institutes of Health, researchers at the University of Wisconsin tested an AI tool that reads the notes and history already sitting in a patient’s electronic health record, looks for patterns linked to opioid use disorder, and prompts providers to consider an addiction medicine consult and to watch for withdrawal. Across nearly 52,000 adult hospitalizations in Madison, Wisconsin, the results were striking.1,2

Patients flagged by the AI tool had 47% lower odds of being readmitted within 30 days compared with those who got a standard provider-initiated referral. About 8% of the AI group came back within a month, against roughly 14% in the usual care group, and that held up after accounting for age, sex, race and ethnicity, insurance, and other conditions. Over eight months, the researchers estimated about $108,000 in healthcare savings, even after the cost of running the software.

NIDA director Dr. Nora Volkow noted that addiction care is easily overlooked in overwhelmed hospital settings, and that is exactly the gap this tries to close. A hospital stay is often a missed moment. People leave before an addiction specialist ever sees them, and the period right after discharge carries a high overdose risk.

The authors were candid about limits. Providers can get alert fatigue, and the tool still needs testing in other health settings.

Prenatal Opioid Exposure and the Developing Brain

Up next, the theme of catching things earlier runs through our next story, though this one is about a population we rarely talk about, babies.

We asked Rehab.com medical officer Dr. Sylvie Stacy to walk us through a newly funded federal study on opioid exposure before birth. The headline is that a researcher at WashU Medicine has received a $3.5 million grant from the National Institute on Drug Abuse to study how prenatal opioid exposure affects the developing brain, and to compare two treatment approaches head to head.3 Here’s Dr. Stacy for more.

Dr. Stacy: What This Research Means for Pregnant Patients

I’m glad that we’re talking about this grant because it deals with a population that really needs more attention. We focus so much on adults that neonates and infants are easily left out of a lot of conversations about substance use issues. I want to start out by pointing out that this research is early stage preclinical work using mouse models, so we’re looking at a study that’s just getting underway rather than a fresh set of clinical data in a patient population.

What makes this project really worth looking at is how it is set up. The research team is going to look at two different medical pathways that are very similar to what we already use in real-world care every day. One is maternal medication-assisted treatment where a pregnant patient is stabilized on a controlled medication like buprenorphine or methadone, and the other is treating the infant after delivery with carefully dosed medication to mitigate newborn withdrawal.

By tracking brain development, learning, emotion, and behavior over time using neuroimaging, the goal is to see how these strategies support the developing brain. One really important thing to emphasize is what the evidence already shows us. For a pregnant person with an opioid use disorder, stopping opioids abruptly is very dangerous, both for the parent and for the pregnancy.

It triggers acute withdrawal that threatens the fetus, which is why the absolute standard of care is stable, long-term medication for the duration of the pregnancy. It’s not detox. I’ve seen this play out in my own practice where patients sometimes face a pretty big sense of societal guilt, and they want to stop using opioids cold turkey.

I always explain that stabilization using medication like buprenorphine is really the safest choice for their baby and for them as new moms. This study is not questioning that established standard. It’s really just trying to refine it and refine our understanding of how these strategies work.

Because this is animal research, nobody should draw hard conclusions about specific medications or alter our existing treatment plans that we use in clinic based on this. For someone who is pregnant and struggling with opioid use disorder, it’s so important that they connect with an addiction medicine specialist or an OB-GYN who understands substance use and is willing to work with the mother in getting her substance use disorder adequately treated during pregnancy. That really is a key first step, and then the next step after that is managing the newborn after delivery and making sure that any withdrawal signs and symptoms are adequately addressed.

Looking forward, as we collect more data on prenatal exposures like this, one big challenge will be making sure that our healthcare infrastructure can actually deliver this specialized care to pregnant patients without the threat of stigma or criminalization that sometimes prevents them from even seeking care in the first place.

Gambling Disorder Linked to Elevated Suicide Risk in Medical Students

Our third story widens the idea of screening from drugs and alcohol to gambling. A 2026 study published in Scientific Reports looked at 775 medical students at a French university and found that those who screened positive for a probable gambling disorder reported recent suicidal thoughts at more than double the rate of their peers, roughly 35% versus 15%.4

After adjusting for other factors, the odds were about five times higher. Two things stand out for treatment. First, gambling disorder and significant anxiety were each linked to that risk independently, even after accounting for depression and substance use, which tells researchers that suicide prevention should not focus on depression alone.

Second, gambling disorder is a treatable behavioral addiction. It runs on many of the same reward and compulsion pathways as substance use. It responds to therapies like cognitive behavioral therapy, and it often travels with anxiety, depression, or a substance use disorder.

So screening opens the door to treating all of it together. There are simple two-question screens that can start the conversation. The practical message is that catching a hidden problem early can prevent a financial, relational, or mental health crisis before it happens.

Conclusion

Put it all together and the bottom line is simple. Whether it’s an AI tool in the hospital, a screening question in the clinic, or a research grant studying the youngest patients of all, the gains come from finding risk earlier, and they only count if treatment is there to meet people when you do. If you or someone you love is looking for treatment, Rehab.com lists thousands of verified centers across the country, and free, confidential support is available at any time through the SAMHSA National Helpline at 1-800-662-4357.

For more, visit Rehab.com. We will be back next week, and thank you for listening.

You Might Also Like:

Sources in This Episode

  1. Afshar M, Resnik F, Joyce C, et al. Clinical implementation of AI-based screening for risk for opioid use disorder in hospitalized adults. Nat Med. 2025;31:1863-1872. doi:10.1038/s41591-025-03603-z. Published April 3, 2025. Accessed July 7, 2026. https://www.nature.com/articles/s41591-025-03603-z
  2. National Institutes of Health. AI screening for opioid use disorder associated with fewer hospital readmissions. Published May 7, 2025. Accessed July 7, 2026. https://www.nih.gov/news-events/news-releases/ai-screening-opioid-use-disorder-associated-fewer-hospital-readmissions
  3. Washington University in St. Louis. NIH grant supports research on brain development after opioid exposure in the womb. The Source. Published June 2026. Accessed July 7, 2026. https://source.washu.edu/2026/06/nih-grant-supports-research-on-brain-development-after-opioid-exposure-in-the-womb/
  4. Luquiens A, Bourgier C, Fabbro-Peray P. Gambling disorder as a risk factor for suicidal ideation in medical students. Sci Rep. 2026;16:12294. doi:10.1038/s41598-026-37805-3. Published March 5, 2026. Accessed July 7, 2026. https://www.nature.com/articles/s41598-026-37805-3
Dr. Sylvie Stacy
By Dr. Sylvie Stacy
Published on July 7, 2026