π Today, 5.7 million people live with OUD in the U.S., and approximately 81,000 people die each year due to opioid overdose. A driving factor for these high death rates is that a majority of those affected by OUD experience relapse, putting them at a higher risk of overdose and death.
The opportunity to intervene before relapse has immense life-saving potential, and yet, there is currently no mechanism in routine care to detect risk of relapse in daily life.
β However, consumer technology has now advanced to a point where it can collect and analyze real-time sensor data from patients through LLM/AI algorithms. This provides a new opportunity to improve OUD care and support recovery by possibly enabling early detection and prevention of future opioid use and potential overdose.
The collaboration will work to define and collect physiological signals and behavior traits that can (1) predict relapse and (2) be measured by wearable technologies, including heart rate, insomnia, physical inactivity, and physiological stress.
π± Smartphones will also be used to capture mental health characteristics such as social isolation, patient-reported anxiety, and depression. These data points will be used to train a tool to prevent opioid relapse.
π Today, 5.7 million people live with OUD in the U.S., and approximately 81,000 people die each year due to opioid overdose. A driving factor for these high death rates is that a majority of those affected by OUD experience relapse, putting them at a higher risk of overdose and death.
The opportunity to intervene before relapse has immense life-saving potential, and yet, there is currently no mechanism in routine care to detect risk of relapse in daily life.
β However, consumer technology has now advanced to a point where it can collect and analyze real-time sensor data from patients through LLM/AI algorithms. This provides a new opportunity to improve OUD care and support recovery by possibly enabling early detection and prevention of future opioid use and potential overdose.
The collaboration will work to define and collect physiological signals and behavior traits that can (1) predict relapse and (2) be measured by wearable technologies, including heart rate, insomnia, physical inactivity, and physiological stress.
π± Smartphones will also be used to capture mental health characteristics such as social isolation, patient-reported anxiety, and depression. These data points will be used to train a tool to prevent opioid relapse.
π Official press release can be found here: https://www.prnewswire.com/news-releases/industry-and-scientific-leaders-announce-partnership-to-combat-the-opioid-epidemic-302440048.html?tc=eml_cleartime
π To learn more about this work, please visit the project web page: https://datacc.dimesociety.org/developing-a-digitally-derived-tool-to-prevent-relapse-in-opioid-use-disorder/