Opioid use disorder (OUD) is one of the most pressing public health problems of our time, with staggering morbidity, social impact, and economic costs. Prescription opioids play a critical role in the opioid crisis as they increase exposure and availability in the general population, making them an attractive target for much needed prevention and risk mitigation strategies. Opioid exposure, including legitimate prescription use, leads to a variety of physiologic adaptations (e.g. dependence) that may be leveraged to understand and identify risk of misuse. Mobile health (mHealth) tools, including wearable sensors have great potential in this space, but have been underutilized. Of specific interest are digital biomarkers, or end-user generated physiologic or behavioral measurements that correlate with events of interest, health, or pathology. Preliminary data support the concept that wearable sensors can detect digital biomarkers of opioid use and may provide clues regarding individual physiologic adaptations to opioid use over time. This dissertation follows a path though the exploration and refinement of these digital biomarkers of opioid use in various clinical use cases. Longitudinal data from individuals treated with opioids for acute pain will be explored through various machine learning models to detect opioid use and to explore patient and treatment factors that impact model performance. Next, a signal processing approach will be undertaken to explore the effects of opioid agonism in a different population of individuals- those presenting with opioid toxicity and precipitated withdrawal. Both approaches will be combined to further refine the digital biomarker capabilities, this time with a focus on the difference between opioid naive and chronic users. And finally, usability, facilitators and barriers to use of a sensor-based monitoring system for opioids will be evaluated through a qualitative lens. Taken together, theses data support the development of a smart technology, driven by empirically derived algorithms which can be used to monitor opioid use, support safe prescribing practices, and reduce OUD and death.
| Identifer | oai:union.ndltd.org:umassmed.edu/oai:escholarship.umassmed.edu:gsbs_diss-2182 |
| Date | 09 March 2022 |
| Creators | Carreiro, Stephanie |
| Publisher | eScholarship@UMassChan |
| Source Sets | University of Massachusetts Medical School |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | Morningside Graduate School of Biomedical Sciences Dissertations and Theses |
| Rights | Copyright is held by the author, with all rights reserved., select |
Page generated in 0.0018 seconds