Karla D. Wagner

The disclosure of one’s HIV-positive status is a personal decision that can play an important role in engagement in HIV treatment and care. Social network approaches situate individuals within broader webs of social connections that influence health and wellbeing and thus hold the potential to elucidate key social factors shaping the process of disclosure. As part of a larger clinical study examining how substance use and mental health shape HIV outcomes, we recruited 61 individuals newly diagnosed with HIV at two clinics in Western Kenya to participate in an egocentric social networks study. We conducted a survey that generated a visual network map to guide a subsequent qualitative interview about experiences being diagnosed and living with HIV. We thematically analyzed the qualitative data and visual network maps to examine the social contexts and patterns of HIV disclosure, with the goal of identifying supportive contexts of disclosure. The mean age was 36.7 years (range: 20–62); women were significantly younger than men and more likely to self-report a mental health issue. Typically, participants disclosed their HIV-positive status to a small number of close, trusted alters in their network, including intimate partners, siblings and other family members, and friends. The need for mental health support in the wake of a new diagnosis, especially among women, encouraged disclosure. Across nearly all stories, stigma was a powerful deterrent to sharing one’s status, and multiple people noted prior disclosure to others not named in their current networks because their relationships were negatively impacted. Our study shows how the intersection of social relationships, mental health, and stigma is critical in understanding decision-making processes around disclosure. Selective disclosure typically improved participants’ wellbeing while stigma precluded disclosure and exacerbated distress. Our study offers suggestions for social network interventions to support people living with HIV/AIDS.

We explored the prevalence and correlates of intention to use a harm reduction vending machine (HRVM) among people who use drugs (PWUD) in Nevada ( n = 219) from 2022 to 2023. Participants responded to a cross-sectional survey assessing HRVM awareness, intent to use, desired inventory, perceived barriers and facilitators of use, drug use patterns, and sociodemographics. Descriptive statistics, bivariate analyses, and Poisson regression models were conducted. Observations were excluded from the analysis if the demographic variables were missing ( n = 6) or if participants reported having used an HRVM before ( n = 23), resulting in an analytic sample of n = 190. Results indicated that 30% of participants were aware of HRVMs and 69% expressed intent to use one. Intention to use a HRVM was less common among people who identified as Hispanic and more common among people whose most common source of drug use supplies was a syringe service program. Perceived barriers included not knowing the HRVM locations, lack of transportation, and surveillance concerns. Conversely, perceived facilitators included access to needed supplies, free supplies, and 24/7 access. While awareness of HRVMs was low, future intent to use one was high, suggesting that HRVMs are an acceptable form of service delivery among PWUD.

Introduction: Emergency department (ED)-based interventions for patients with substance use concerns have been implemented across the USA to reduce drug-related ED utilization and improve other outcomes. To date, findings about their effectiveness have been mixed. Methods: We conducted a nonrandomized, quasi-experimental evaluation of an ED-based intervention implemented by Certified Peer Recovery Support Specialists (CPRSS) in 3 hospitals in Nevada, USA. We compared patients who received the intervention to those who did not on 1 primary (drug/alcohol poisoning/overdose ED visits) and 4 secondary outcomes (number of drug/alcohol-related ED visits, number of nondrug/alcohol-related ED visits, number of drug/alcohol-related inpatient hospital admissions, number of nondrug/alcohol-related inpatient hospital admissions) 9 months post-intervention. De-identified patient-level data were obtained from the Nevada Department of Health and Human Services. Analyses included within- and between-group comparisons of the number of visits in each outcome category. Results: A total of 4643 patients were included in this analysis (631 intervention and 4012 comparison patients). We observed significant differences in the demographic and ED utilization profiles of the 2 groups. Within-group pre–post comparisons showed small but statistically significant increases in the mean number of drug/alcohol-related ED visits in both groups, and a statistically significant but small increase in the mean number of drug/alcohol poisoning/overdose and nondrug-related ED visits in the comparison group, but not the intervention group. Overall, the change in mean number of visits was small and similar across both groups. Conclusion: We did not observe a clinically significant effect of the intervention on the outcomes, consistent with an emerging literature showing mixed results of ED-based CPRSS interventions. However, we identified several opportunities for further study, including a focus on multiple substances and patient-centered outcomes. OSF pre-registration: https://doi.org/10.17605/OSF.IO/YHEC5

Co-use of stimulants and opioids is often deliberate. However, the possibility remains that some people are unintentionally consuming fentanyl. To advance understanding of overdose risk, we examined the rate of concordance between self-reported fentanyl use and corresponding urine toxicology screen results. Between August 2022-August 2023, 411 participants (adults who reported any non-medical drug use in the past three months) in Nevada and New Mexico completed a cross-sectional survey, of whom 64% (n = 270; the analytical sample) also completed a urine toxicology screen, which detects fentanyl use in the past three days. Positive predictive value, negative predictive value, sensitivity, and specificity were calculated using self-reported past three-day fentanyl use (yes/no) and urine toxicology screen results for the presence of fentanyl (positive/negative). Of the 270 participants who provided a urine sample, 268 are included in the descriptive statistics (two with inconclusive urine toxicology screen results were excluded). Of the 268 participants, 146 (54.5%) had a fentanyl-positive urine toxicology screen result, 122 (45.5%) had a fentanyl-negative urine toxicology screen result, 137 (51.1%) reported past three-day fentanyl use, and 130 (48.5%) reported no past three-day fentanyl use. Only 6.9% of those with a fentanyl-positive urine toxicology screen did not report recent fentanyl use. The sensitivity of self-reported fentanyl use was 93%, specificity was 97%, positive predictive value was 97%, and negative predictive value was 92%. The rate of unanticipated exposure to fentanyl (that is, positive urine screen and negative self-report) in this sample was low, at 6.9%. This runs counter to the national narrative that there is widespread unknown contamination of fentanyl in the drug supply. Future research is needed to further explore how people who use multiple substances interpret their overdose risk and what harm reduction methods they employ.

Background: People who inject drugs (PWID) experience high rates of drug overdose death with the risk of mortality increasing after each non-fatal event. Racial differences exist in drug overdose rates, with higher rates among Black people who use drugs. Psychological factors may predict drug overdose. Methods: Cross-sectional data from a survey administered to PWID in Baltimore, MD enrolled in a social network-based intervention were analyzed. Linear regression methods with generalized estimating equations were used to analyze data from indexes and network members to assess for psychological factors significantly associated with self-reported number of lifetime drug overdoses. Factors associated with number of overdoses were assessed separately by race. Results: Among 111 PWID enrolled between January 2018 and January 2019, 25.2% were female, 65.7% were Black, 98.2% reported use of substances in addition to opioids, and the mean age was 49.0 ± 8.3 years. Seventy-five individuals (67.6%) had a history of any overdose with a mean of 5.0 ± 9.7 lifetime overdoses reported. Reports of feeling fearful (β = 9.74, P = 0.001) or feeling lonely all of the time (β = 5.62, P = 0.033) were independently associated with number of drug overdoses. In analyses disaggregated by race, only the most severe degree of fearfulness or loneliness was associated with overdose among Black participants, whereas among White participants, any degree of fearfulness or loneliness was associated with overdose. Conclusions: In this study of PWID loneliness and fearfulness were significantly related to the number of reported overdose events. These factors could be targeted in future interventions.