Fentanyl is changing how doctors treat opioid use disorder
Traditional treatment approaches in an evolving illicit drug market are less effective, researchers say, highlighting the need for new clinical guidelines
HERSHEY, Pa. — For years, buprenorphine — one of the primary medications used to treat opioid use disorder — has been a critical bridge to recovery, helping to reduce illicit drug use and overdose deaths. But with the changing landscape of the illicit drug market, particularly the rise of the potent synthetic opioid fentanyl, health care providers have found that traditional treatment protocols aren’t as effective as they used to be.
A new national survey, led by researchers at the Penn State College of Medicine and the University of Pittsburgh, found that nearly three-quarters of clinicians encountered significant obstacles when starting buprenorphine treatment for patients using fentanyl. More than 67% have modified their treatment protocols, such as adjusting dosages.
But this is more than a technical hurdle for health care providers, the researchers said. It’s a major barrier for people seeking treatment for opioid use disorder who experience rapid withdrawal symptoms or prolonged symptoms as a result. The study, published in JAMA Open Network, highlights the complexity of treating opioid use disorder.
“It’s a public health crisis. Buprenorphine is a lifesaving option due to its safety profile and ease of access,” said lead author Sarah Kawasaki, associate professor of psychiatry and behavioral health and of medicine at Penn State College of Medicine. “We need more research and updated clinical guidelines for the fentanyl era.”
Treatment for opioid use disorder used to be predictable, Kawasaki said. Most people were using heroin, which could be verified on drug screenings. Clinicians could then give buprenorphine reliably at certain dosages and time intervals. Patients could access the medication at pharmacies and would do well.
Kawasaki said things started to change around 2017. More patients were testing positive for fentanyl. Unlike older opioids, fentanyl’s unique chemistry — specifically its ability to "hide" in the body's fat cells — makes the transition to buprenorphine more difficult. People experienced withdrawal symptoms for much longer than typical or had withdrawal symptoms so intense that they reported feeling allergic to the medication.
“Someone can get really sick or get tired of the whole process that they stop treatment. They might start using again and might overdose or die,” Kawasaki said.
Patients also began requesting methadone, another standard protocol for managing opioid use disorder. However, methadone is only available from an estimated 2,000 licensed facilities in the United States, Kawasaki said, compared to more than 70,000 pharmacies where patients can access buprenorphine.
To understand how doctors were responding to the challenges of starting patients on buprenorphine in the rapidly changing illicit drug market, the researchers surveyed 396 health care providers. The pool of participants was a nationally representative sample of physicians and advanced practice clinicians who initiated at least 10 patients with opioid use disorder onto buprenorphine during the prior year and at least one patient in the past 90 days.
The researchers found that:
- 72% of respondents reported experiencing challenges starting buprenorphine treatment among patients using fentanyl in the past year
- Nearly 62% of respondents reported one or more instances in which the patient experienced a severe and sudden onset of withdrawal symptoms
- 52.8% reported cases of prolonged withdrawal where symptoms lasted days instead of hours
- Those who worked in high-volume outpatient settings were more likely to report these challenges.
Approximately 67% reported modifying their standard protocols. In some cases, providers used doses that were much smaller than previously recommended while others used much higher doses. Some clinicians prescribed adjunct medications to help with withdrawal symptoms. Still others referred patients to inpatient treatment or to methadone because of the challenges of initiating buprenorphine treatment.
Despite these challenges, the researchers emphasized that buprenorphine remains a life-saving treatment for those with opioid use disorder and that many patients do not encounter problems when starting buprenorphine. According to Kawasaki, the study highlighted the need to develop evidence-based guidelines to successful initiate buprenorphine in light of more potent drugs.
“Buprenorphine still works. If you or a loved one needs help, don’t be afraid to reach out,” Kawasaki said.
Erin Winstanley, professor of medicine at the University of Pittsburgh, is senior author of the study. Other authors on the study from the University of Pittsburgh include Jane Liebschutz, professor of medicine; Cristina Murray-Krezan, associate professor of medicine; Galen Switzer, professor of medicine; Samantha Nash, clinical research coordinator; and Kwonho Jeong, biostatistician.
Funding from the National Institute on Drug Abuse supported this work.
At Penn State, researchers are solving real problems that impact the health, safety and quality of life of people across the commonwealth, the nation and around the world.
For decades, federal support for research has fueled innovation that makes our country safer, our industries more competitive and our economy stronger. Recent federal funding cuts threaten this progress.
Learn more about the implications of federal funding cuts to our future at Research or Regress.
Journal
JAMA Network Open
Method of Research
Survey
Subject of Research
People
Article Title
Barriers to Buprenorphine Initiation in Patients Using Fentanyl
Article Publication Date
5-Jan-2026
Before crisis strikes — smartwatch tracks triggers for opioid misuse
A University of California San Diego study is working on a potentially life-saving measure that may be as simple as strapping on a smartwatch.
Opioid overdoses continue to take a devastating toll across the United States. According to the U.S. Centers for Disease Control and Prevention (CDC), in 2023, the nation recorded roughly 105,000 drug overdose deaths overall, with nearly 80,000 deaths involving opioids. Worldwide, opioids are also responsible for the majority of drug-related deaths. A University of California San Diego study is working on a potentially life-saving measure that may be as simple as strapping on a smartwatch.
Researchers have long known that people living with chronic pain and long-term opioid prescriptions can experience downward spirals of elevated stress, pain flare-ups and craving — shifts that may raise the risk of opioid misuse and addiction. The problem is that clinicians usually only see snapshots of how someone is doing: a clinic visit, a questionnaire, a check-in every few weeks. That can miss critical “in-between” moments when risk spikes.
The UC San Diego team’s study proposed a different approach: enabling a common smartwatch to continuously track subtle changes in heart rhythm, then apply machine learning to estimate when someone may be slipping into a high-risk state — facilitating earlier and potentially life-saving support. The study was led by Professor Tauhidur Rahman and Ph.D. student Yunfei Luo at the Halıcıoğlu Data Science Institute (HDSI), part of the University of California San Diego’s School of Computing, Information and Data Sciences (SCIDS) and Eric Garland, PhD, professor of psychiatry at UC San Diego School of Medicine and endowed professor at Stanford Institute for Empathy and Compassion
The team built a system that uses a wearable device to collect inter-beat interval data, the tiny timing differences between heartbeats. From these signals, the system estimates heart rate variability (HRV), a measure that often shifts when the body is under strain. In simple terms, HRV provides a window into how the nervous system is responding to stress.
The system tracks risk-related states such as stress, pain and craving, then looks for patterns that occur more often in people at higher risk of opioid misuse compared with those taking medication as prescribed.
The idea: a “smoke alarm” for risk — without constant check-ins.
The study at-a-glance
- Participants and Data: 10,140 hours of wearable data from 51 adults with chronic pain on long-term opioid therapy;
- Device: a commercially available Garmin Vivosmart 4 smartwatch;
- Setting: daily life outside the clinic over an 8-week period;
- Comparison groups: participants were categorized using Current Opioid Misuse Measure (COMM), a standard questionnaire to help clinicians identify whether a patient who is taking prescription opioids for chronic pain may be showing signs of misuse; and
- Key outputs:
- Predicted stress/pain/craving levels over time
- A final “misuse risk” classification based on patterns in those trajectories, plus clinical record text
Stress, pain, craving: hard-to-quantify risk factors
Luo described the approach: “We built a system that uses a wearable device to collect inter-beat interval data, the tiny timing differences between heartbeats. From these signals, the system estimates heart rate variability (HRV), a measure that often shifts when the body is under strain. In simple terms, HRV provides a window into how the nervous system is responding to stress.”
The heart rate variability was mapped to opioid misuse risk in two steps:
Step 1: Personalized prediction of stress, pain, and craving
The lead clinical scientist involved in the study, UC San Diego Health’s Eric Garland, indicated that every monitor must be individually tailored. “One major challenge is that HRV is deeply personal,” Garland said. “What looks like ‘high craving’ for one person may be normal for another. To account for that, the team trained personalized models,not a one-size-fits-all predictor.”
Luo added that the team used a learning-to-branch technique to dynamically identify clusters of participants with similar characteristics. “This makes the model more data-efficient and enables personalized predictions of stress, pain and craving,” he said.
Step 2: Estimating misuse risk by studying the shapeof daily patterns
Rahman said the team looked beyond stress, craving or pain at any single moment and instead focused on how these states evolve over time. “Using nonlinear dynamical analysis, we examined whether a person’s daily patterns were more rigid and predictable or more flexible and variable,” he explained. “People at higher risk of opioid misuse showed more repetitive trajectories and tended to get stuck in high stress, pain or craving — what appears in our analysis as lower entropy, or reduced flexibility over time. In contrast, those taking opioids as prescribed showed more fluctuation and rebound, reflected as higher entropy.”
Adding clinical context for more accurate prediction
To improve accuracy, the system also uses information already found in medical records, such as demographics, prescription history, symptoms and related conditions. Instead of relying on a large cloud-based chatbot, the researchers used smaller, clinically trained language models to convert these records into compact numerical summaries that the prediction model can use. Combining smartwatch signals with clinical context improved performance. This approach could help clinicians detect risk shifts between visits, trigger timely check-ins, reduce the burden of constant self-reporting, and better target prevention for chronic pain patients.
What’s next
The team points toward exploring how this kind of monitoring might support “just-in-time interventions” — help delivered at the moment it’s most needed.
Rahman, study supervisor and director of the Mobile Sensing and Ubiquitous Computing (MOSAIC) Laboratory, is hopeful that mobile and wearable sensors and AI/machine learning may be a key to reversing an increasingly deadly trend. “As overdose deaths remain high nationally, the long-term hope is that tools like this could help clinicians move from periodic snapshots to continuous, patient-friendly monitoring — and intervene earlier, before risk becomes tragedy.”
This study was published in Nature Mental Health.
A full U.S. utility patent application (US2025/016369) was also filed for this technology, titled “System and Method for Personalized Closed-Loop Opioid Addiction Management with Mobile and Wearable Sensing of Administrations, Affective States and Misuse Risk Scores”.
Journal
Nature Mental Health
Powerful AI can help diagnose substance use disorder
System could help clinicians get treatment faster to people who need help
image:
University of Cincinnati Professor Hans Breiter and his research partners developed a powerful new AI that can help diagnose substance use disorder.
view moreCredit: Andrew Higley
Diagnosing substance-use disorder can be difficult because of patient denial related to the stigma attached to addiction.
But a new study by the University of Cincinnati uses a novel artificial intelligence to predict substance use defining behaviors with up to 83% accuracy and 84% accuracy to predict the severity of the addiction. Researchers say this could allow clinicians to provide treatment faster to patients who need it.
The clinical standard for psychiatry defines substance use disorder by four categories of destructive behaviors related to impaired control, physical dependence, social impairments and risky use irrespective of the substance being used. Successful prediction of these can help researchers understand the general processes defining addiction.
The study is one of the first of its kind to use a computational cognition framework with artificial intelligence to assess how human judgment can be used to predict substance use disorder defining behaviors, identify the substances used and determine the severity of the addiction.
“This is a new type of AI that can predict mental illness and commonly co-occurring conditions like addiction. It’s a low-cost first step for triage and assessment,” UC College of Engineering and Applied Science Professor Hans Breiter said.
The study was published in the Nature journal Mental Health Research.
Previously, Breiter and his team demonstrated that their novel AI was effective at predicting other health issues such as patient anxiety and willingness to get vaccinations.
Breiter worked with longtime collaborator and UC Senior Research Associate Sumra Bari, the paper’s lead author, to apply their novel AI system to substance use disorder.
The study examined 3,476 participants ages 18 to 70 who provided written, informed consent and answered questionnaires that were then used as the target of AI-based prediction.
Respondents also rated the degree to which they liked or disliked 48 pictures with mildly emotional stimuli. The picture rating data were used to quantify mathematical features of people’s judgments, including variables commonly related to behavioral economics. These variables along with a small set of demographics were then used with artificial intelligence algorithms to predict substance use disorder-defining behaviors and identify both the substances being used and the severity of the disorder.
“Anyone with a smartphone or computer can do the picture rating task. It’s low cost, scalable and resilient to manipulation,” Bari said.
The picture ranking task might seem simple, she said. But it evaluates an individual’s unique profile of preferences among 1.3 trillion possibilities, creating a surprisingly powerful tool.
The system utilizes concepts familiar in the world of economics such as aversion to losses, aversion to risk and desire for insurance against bad outcomes It quantifies a set of variables that describe human judgments.
The system was able to identify the type of substance used (stimulants, opioids, or cannabis) with up to 82% accuracy and the severity of the addiction with up to 84% accuracy. A statistical evaluation of the judgment data revealed that participants with higher substance use disorder severity were more risk-seeking, less resilient to losses, had more approach behavior and had less variance in preferences, informing a behavioral profile of individuals with substance use disorder.
By predicting substance use disorder behaviors directly, this approach could enable assessment across a broader spectrum of addictions, potentially including behavioral addictions such as excessive social media use, gaming or food consumption, Bari said.
Journal
npj Mental Health Research
Method of Research
Experimental study
Subject of Research
People
Article Title
Predicting substance use behaviors with machine learning using small sets of judgment and contextual variables
Penn Nursing study identifies key predictors for chronic opioid use following surgery
University of Pennsylvania School of Nursing
PHILADELPHIA (February 5, 2026) — For many Americans, a routine surgical procedure serves as their first introduction to opioid pain medication. While most stop using these drugs as they heal, a considerable number of "opioid-naïve" patients transition into New Persistent Opioid Use (NPOU)—continuing use long after the typical recovery period.
A new systematic review and meta-analysis led by Penn Nursing researchers, published in Pain Medicine, has identified the specific patient-related risk factors that most accurately predict which individuals are at the highest risk for this dangerous transition. The study, which synthesized data from 27 high-quality studies, found that four primary factors significantly increase the odds of a patient becoming a long-term opioid user following surgery: Medicaid enrollment, preoperative benzodiazepine use, mood disorders, and anxiety.
“Identifying who is at risk before the first incision is made is a critical step in combatting the opioid crisis,” said lead author and doctoral student Yoonjae Lee, DNP, APRN. “Our findings provide a roadmap for clinicians to implement targeted interventions, ensuring that high-risk patients receive enhanced monitoring and alternative pain management strategies.”
The Risk Factor Breakdown
Through a rigorous meta-analysis, the research team derived "pooled odds ratios," which quantify how much each factor is associated with the odds of persistent use:
- Medicaid Enrollment & Preoperative Benzodiazepines: These were the strongest predictors, with patients in these categories having 77% higher odds of developing NPOU (Odds ratio: 1.77).
- Mood Disorders: Patients with a history of depression or other mood disorders faced 24% higher odds compared to those without.
- Anxiety: Patients with pre-existing anxiety disorders had 17% greater odds of persistent use.
A Call for Holistic Preoperative Screening
NPOU is defined as continued opioid use beyond three months post-surgery and has been linked to increased morbidity, higher mortality rates, and long-term complications.
The study’s findings emphasize that "opioid-naïve" status, meaning the patient has not used opioids recently, is not a standalone guarantee of safety. By highlighting that social determinants (such as insurance type) and psychological factors (such as anxiety) are just as influential as the surgery itself, the researchers advocate for a more comprehensive approach to preoperative screening. Integrating these data-driven insights into clinical practice can help prevent the unintended consequences of surgical pain management and improve long-term outcomes for patients nationwide. Co-authors for this study include Rosemary C. Polomano; Heath D. Schmidt, PhD; Jungwon Min, PhD; and Peggy A. Compton, PhD; all of Penn Nursing.
# # #
The University of Pennsylvania School of Nursing (Penn Nursing) is a global leader in education and the top NIH-funded nursing research institution in the U.S. Ranked the #1 nursing school in the country by QS University for a decade, Penn Nursing consistently earns top honors from U.S. News & World Report for its BSN and graduate programs. By integrating innovation in research, education, and practice, Penn Nursing prepares nurse scientists and leaders to meet the health needs of a global society. Follow Penn Nursing: Facebook | LinkedIn | YouTube | Instagram.
Journal
Pain Medicine
Article Title
Patient-related risk factors for new persistent opioid use after surgery among opioid-naïve individuals in the United States: a systematic review and meta-analysis
