Project Summary/Abstract Stigma associated with substance use disorders (SUDs) can have severe negative consequences for physical and mental health; affect employment, housing, and social relationships; result in social alienation; and serve as a barrier to obtaining treatment. Existing stigma reduction interventions have largely focused on structural stigma (e.g. education of medical students or professional groups that work with persons with SUDs), but stigma can occur at multiple levels and in different contexts. In particular, there is a need for more stigma reduction interventions targeting those affected by self-stigma. We propose to address this gap by characterizing stigma in different substance use contexts through narratives collected from two different sources, social media and survey data. The long-term objective of this proposal is to derive insights concerning stigma phenotypes – characterized by setting (e.g., work, home, school), actors (persons involved in the stigma-related experience), and situations in which stigma reduction interventions are needed, as well as intervention mechanisms that are appropriate for their target populations. The significance of this proposal is that it focuses on stigma as it directly affects the persons who experience stigma – an area in which stigma reduction interventions are lacking. We focus on stigma relating to three substances: alcohol, cannabis, and opioids. Our three aims are: 1) we will integrate natural language processing and qualitative research methods to identify stigma processes, their contextual factors, and barriers and facilitators of stigma reduction in a social media corpus; 2) we will administer a survey to elicit narratives from persons with SUDs about how they have dealt with stigma in different contexts; and 3) we will triangulate the findings from Aims 1 and 2 to identify settings, target populations, and intervention mechanisms for future stigma reduction interventions. Our multidisciplinary research team possesses expertise in multiple areas relating to this proposal, including NLP, qualitative research, and substance use, and are ideally positioned to conduct the research.