Medication treatment for opioid use disorder (MOUD) is an effective evidence-based therapy for decreasing opioid-related adverse outcomes. Effective strategies for retaining persons on MOUD, an essential step to improving outcomes, are needed as roughly half of all persons initiating MOUD discontinue within a year. Data science may be valuable and promising for improving MOUD retention by using “big data” (e.g., electronic health record data, claims data mobile/sensor data, social media data) and specific machine learning techniques (e.g., predictive modeling, natural language processing, reinforcement learning) to individualize patient care. Maximizing the utility of data science to improve MOUD retention requires a three-pronged approach: (1) increasing funding for data science research for OUD, (2) integrating data from multiple sources including treatment for OUD and general medical care as well as data not specific to medical care (e.g., mobile, sensor, and social media data), and (3) applying multiple data science approaches with integrated big data to provide insights and optimize advances in the OUD and overall addiction fields.

Background: While medications for opioid use disorder (MOUD) effectively treat OUD during pregnancy and the postpartum period, poor treatment retention is common. Digital phenotyping, or passive sensing data captured from personal mobile devices, namely smartphones, provides an opportunity to understand behaviors, psychological states, and social influences contributing to perinatal MOUD non-retention. Given this novel area of investigation, we conducted a qualitative study to determine the acceptability of digital phenotyping among pregnant and parenting people with opioid use disorder (PPP-OUD).Methods: This study was guided by the Theoretical Framework of Acceptability (TFA). Within a clinical trial testing a behavioral health intervention for PPP-OUD, we used purposeful criterion sampling to recruit 11 participants who delivered a child in the past 12 months and received OUD treatment during pregnancy or the postpartum period. Data were collected through phone interviews using a structured interview guide based on four TFA constructs (affective attitude, burden, ethicality, self-efficacy). We used framework analysis to code, chart, and identify key patterns within the data.Results: Participants generally expressed positive attitudes about digital phenotyping and high self-efficacy and low anticipated burden to participate in studies that collect smartphone-based passive sensing data. Nonetheless, concerns were noted related to data privacy/security and sharing location information. Differences in participant assessments of burden were related to length of time required and level of remuneration to participate in a study. Interviewees voiced broad support for participating in a digital phenotyping study with known/trusted individuals but expressed concerns about third-party data sharing and government monitoring.Conclusion: Digital phenotyping methods were acceptable to PPP-OUD. Enhancements in acceptability include allowing participants to maintain control over which data are shared, limiting frequency of research contacts, aligning compensation with participant burden, and outlining data privacy/security protections on study materials.