Developing a Wearable Sensor-Based Digital Biomarker of Opioid Dependence.
Stephanie Carreiro1, Pravitha Ramanand2, Washim Akram2
1From the Department of Emergency Medicine, Division of Medical Toxicology, University of Massachusetts Chan Medical School, Worcester, MA.
Anesthesia and Analgesia
|October 16, 2024
Summary
Wearable sensors can identify physiologic changes related to opioid dependence. Machine learning models accurately distinguish between opioid-naïve and chronic opioid users, aiding safer prescribing.
Area of Science:
- Digital health
- Pharmacovigilance
- Machine learning in medicine
Background:
- Opioid exposure causes physiological adaptations that can indicate risk for opioid-use disorder (OUD).
- Digital pharmacovigilance using noninvasive sensors offers a novel approach to monitor opioid use and promote safer prescribing.
- Identifying early signs of dependence is crucial for managing OUD risk.
Purpose of the Study:
- To identify wearable sensor-derived features associated with opioid dependence.
- To compare physiological adaptations in opioid-naïve individuals versus chronic opioid users.
- To develop a machine learning model for distinguishing between these groups.
Main Methods:
- A longitudinal observational study collected continuous physiological data from participants with acute pain receiving opioid analgesia.
- Data were gathered during hospitalization and for 7 days post-discharge, combined with electronic health record and self-reported opioid administration.
- Machine learning models were trained and validated using 30 sensor-derived features and 9 demographic/clinical features.
Main Results:
- Forty-one participants and 169 opioid administrations were analyzed.
- Four interpretable decision tree-based machine learning models were developed using 14 sensor and 5 clinical features.
- Participant-level model performance showed accuracy ranging from 70% to 90%, sensitivity from 67% to 100%, and specificity from 67% to 100%.
Conclusions:
- Wearable sensor data can generate digital biomarkers to predict opioid use status (naïve vs. chronic).
- The identified features may indicate opioid dependence.
- Further research should explore opioid dependence, withdrawal, and transition states in repetitive opioid exposure.
Related Concept Videos
01:25Analgesia and Pain Management
540
Pain is critical to various clinical pathologies, provoking an urgent need for effective management. Pain, whether acute or chronic, is a complex neurochemical process. Its alleviation depends on the type, with nonopioid analgesics effective for mild to moderate pain, such as musculoskeletal or inflammatory pain, while neuropathic pain responds best to anticonvulsants, tricyclic antidepressants, or serotonin/norepinephrine reuptake inhibitors. For severe acute or chronic pain, opioids may be...
540
01:15Drug Abuse and Addiction: Pharmacological Phenomena
447
Drug dependence, abuse, and addiction are complex phenomena that can precipitate various abnormal states. Physical dependence refers to a state of pharmacological adaptation to a drug. This adaptation often results in tolerance—a reduced response to the drug after repeated administrations. When the drug use is abruptly stopped, withdrawal symptoms occur due to the body's need to readjust from the pharmacologically induced imbalance. However, tolerance and withdrawal symptoms do not...
447