Accepted for/Published in: Journal of Medical Internet Research
Date Submitted: Nov 14, 2024
Date Accepted: Aug 14, 2025
Identifying Stigma Phenotypes in Social Media Narratives of Substance Use: An Observational Study
ABSTRACT
Background:
Individuals with substance use problems experience stigma in different contexts. Identifying characteristic situations in which stigma occurs or manifests – stigma phenotypes – can serve as important leverage points for future intervention.
Objective:
This paper aims to (1) identify stigma phenotypes expressed in social media narratives relating to substance use stigma and (2) explore the similarities and differences between the stigma phenotypes from a social ecological perspective.
Methods:
We collected Reddit posts pertaining to three substances – alcohol, cannabis, and opioids. We performed feature engineering using a combination of content analysis, machine learning (ML), and keyword-based methods to predict variables at different levels of the Social Ecological Framework. Leveraging these features, we applied the Fuzzy C-Means (FCM) clustering algorithm on the subset of posts containing stigma to extract stigma phenotypes, where a phenotype is defined by four main dimensions: (1) the stigma mechanism(s) present, e.g. internalized stigma, anticipated stigma, and enacted stigma; (2) the substance(s) used, e.g. alcohol, cannabis, and opioids; (3) the settings involved, e.g. work, school, home, etc.; (4) the actors involved, e.g. family, friends, partners, etc. Finally, we use Kruskal-Wallis and Dunn’s post-hoc tests to examine the differences between stigma phenotypes with respect to specific ecological factors.
Results:
We derived 7 stigma phenotypes from stigma-related posts by 8627 authors. The phenotypes can be categorized into 4 groups: internalized stigma only, anticipated stigma, enacted stigma only, and mixed stigma phenotypes. Narratives in internalized stigma phenotypes focus on the self, with minimal reference to settings and actors. One phenotype focused on anticipated stigma and was characterized by high proportion of opioid use mentions (707/1217, 58.09%) and references to the healthcare setting (647/1217, 53.16%). Posts associated with the enacted stigma only phenotypes included substantial representation of settings and actors. Narratives in the mixed stigma phenotypes often involve more than one stigma mechanism, setting, and actor, with home and family being the most salient factors. The phenotypes differed from one another with respect to social ecological factors including loneliness and social isolation, use of treatment services, presence of healthcare providers, community and support groups, society, and legalization.
Conclusions:
These findings provide valuable insights that help inform the design and development of interventions targeted at different stigma phenotypic groups from a social ecological perspective.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.