UCSD - University of California - San Diego

01/30/2026 | Press release | Distributed by Public on 01/30/2026 16:05

Before Crisis Strikes — Smartwatch Tracks Triggers For Opioid Misuse

Published Date

January 30, 2026

Article Content

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
From left to right, Prof. Tauhidur Rahman, Ph.D. student Yunfei Luo and Prof. Eric Garland. Credit: HDSI

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 shape of 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".

UCSD - University of California - San Diego published this content on January 30, 2026, and is solely responsible for the information contained herein. Distributed via Public Technologies (PUBT), unedited and unaltered, on January 30, 2026 at 22:05 UTC. If you believe the information included in the content is inaccurate or outdated and requires editing or removal, please contact us at [email protected]