Perceived Causes of Personal versus Witnessed Overdoses among People who Inject Opioids.

Background: Studies show that people who inject drugs (PWID) underestimate their overdose risk. We sought to explore this phenomenon by comparing how PWID perceive causes of personal overdoses compared to witnessed overdoses. Methods: We analyzed 40 interviews from participants enrolled in a randomized-controlled behavioral intervention to reduce overdose among at-risk PWID in San Francisco from 2014 to 2016. Subjects were current illicit opioid injectors with opioid use disorder, had received take-home naloxone, and had overdosed within five years. Interviews were audio-recorded and transcribed verbatim. Using thematic content analysis, three analysts coded the interviews and measured interrater reliability. The analysts developed a codebook of a priori and inductively generated codes, and applied it to all interviews. Coding discrepancies were discussed. Results: We used two theoretical frameworks - actor observer bias (AOB) and intragroup stigma - to analyze participants' descriptions of personal and witnessed overdoses. AOB suggests individuals may assign responsibility of their actions to external factors, while assigning responsibility for others' actions to internal mechanisms. Intragroup stigma describes the process whereby people perpetuate stigma within their own group. Related to these concepts, two principal themes were used to describe personal overdose: (1) drug volatility and (2) ascribing blame to others, and witnessed overdoses: (1) greed and (2) inexperience/foolishness. Conclusion/Importance: The differences in perceived causes of personal versus witnessed overdose align with AOB and intragroup stigma. Understanding how these theories shape overdose experiences may improve behavioral interventions by introducing peer based supports and encouraging PWIDs to employ evidence-based safety precautions when using opioids.