Abstract: “Unclean” and “Unkempt”: How Opioid Stigma Mark Messages Harm Attitudes and Behavioral Intentions

◆ Victoria Ledford, Auburn University


With rising opioid overdose death rates in the United States (NIH 2023), creating more humanizing portrayals of people with Opioid Use Disorders (OUDs) has far-reaching implications. For example, stigma labels like “addict” can harm people’s emotions, attitudes, interpersonal support for people with OUDs, and support for helpful public health policies like rehabilitation and administration of Narcan, an overdose response drug (Ashford et al., 2018; Goodyear et al., 2018; Kelly & Westerhoff, 2010; Ledford et al., 2022). Though much opioid stigma research has examined stigmatizing labels (e.g., “addict”), little attention has been paid to the visual portrayals of people with OUDs through news images or imagery-laden language. These portrayals often situate people with OUDs as criminals (Scott, 2021) or use language like “clean” to describe those in recovery, implying that people currently using OUDs are “dirty” (McGinty et al., 2019). Such representations embody what Smith (2007) calls stigma marks – linguistic or visual cues that use disgust-evoking imagery to create distance between the stigmatized and non-stigmatized groups.

The current study subsequently uses and extends Smith’s (2007) mark theorization in the model of stigma communication (MSC) to test the effects of opioid stigma mark messages on a series of stigma-related mediators and outcomes. A sample of Amazon Mechanical Turk workers (N = 371) participated in the online between subjects 2 (stigma mark message: high, low) x 3 (opioid type: prescription, heroin, fentanyl) experiment. Participants were randomly assigned to one of six experimental conditions that included the same hypothetical news article about the opioid epidemic and described a person with an OUD, Sam. In the high mark stigma conditions, the article described Sam as wearing “unclean clothes” with an “unkempt appearance” and as someone who “does not seem put together.” The low stigma mark condition described Sam wearing “clean clothes” with a “polished appearance” and as someone who “seems put together.” After reading the article, participants answered key study questions and were debriefed.

PROCESS was used to examine five dual mediation models with mark messages influencing each of the five key study outcomes through both negative affect and stigma beliefs. Mark messages exerted significant indirect effects on all five study outcomes: more negative attitudes toward people with OUDs (R2 = .44, p < .001), more desired social distance from people with OUDs (R2 = .38, p < .001), more desired behavioral regulation of people with OUDs (R2 = .53, p < .001), less support for opioid related public health policies (R2 = .13, p < .001), and more intent to share the stigma message with others (R2 = .08, p < .001). However, dual mediation was only present for attitudes and social distance; stigma beliefs solely explained the relationship between mark messages and the other three outcomes of behavioral regulation, policy support, and stigma sharing intentions. Results of this study provide both theoretical and practical insights by extending the existing mediation pathways of the current MSC (Smith et al., 2019) and illuminating the harms that linguistic opioid stigma marks (i.e., “unclean”) can create.