(4) Unveiling Predictors and Impact: Harm Reduction Vending Machines in Clark County, NV

Background & Introduction: In 2017, the Southern Nevada Health District (SNHD) partnered with Impact Exchange (Trac-B Exchange), a non-profit, to introduce Public Health Vending Machines (PHVMs) in Clark County, Nevada. These machines provided essential disease prevention supplies, including sterile syringes, offering an alternative to those who couldn't access traditional syringe service programs (SSPs). Their recognition in the United States, led by Nevada, has been a recent development. The PHVMs' primary mission in Clark County was to combat the spread of HIV and viral hepatitis among people who inject drugs, later expanding to include the distribution of naloxone. Recent analysis by Allen et al. found that naloxone dispensation at PHVMs in Clark County was associated with an immediate reduction in opioid-involved overdose fatalities. This evaluation assesses changes since the initial PHVM placement in 2017, focusing on the period from August 1, 2021, to May 31, 2023. It offers an updated overview of PHVM participant demographics, usage patterns, accessed items, factors influencing utilization, and includes a survey analysis of drug-related behaviors among recent PHVM enrollees from January 1, 2021, to July 31, 2023.

Methods: Three de-identified datasets were used in the current study: (1) enrollment data from the Trac-B Exchange; (2) vending machine transaction data exported from the VendNovation website; and (3) data from an optional harm reduction survey administered between January 1st, 2021 to July 31st, 2023.
Following enrollment in Trac-B Exchange, participants are issued a unique card number. A participant identifier was created to link cards to individual participants using administrative notes, and data were examined to identify and remove duplicate enrollments. Enrollments without an issued card number (pending application approval) were also removed. Vending machine transaction data were linked to the enrollment data by card number for analyses.
A negative binomial regression model was fit to the transaction data to examine for factors associated with use (overall and by item type). The model adjusted for age, sex, race, ethnicity, and homelessness status. To examine for factors impacting high use (>= 95th percentile of transactions) and low use (<= 25th percentile of transactions), logistic regression was used adjusting for the same factors included in the negative binomial regression. Significance was set at p<0.05. All analyses were conducted in R (version 4.3.1).

Results: A total of 606 enrollments were included in the transaction analysis, representing 606 unique participants with at least 1 vending machine transaction between 8/1/21 – 5/31/23. A total of 13,174 transactions were available for analysis. Most users were under 55 years old (88%), male (59%), White (84%), non-Hispanic (86%), and not experiencing homelessness (88%). The mean number of transactions per client was 21.7 (SD 34.5; Range: 1-348). Compared to individuals 18 to 34 years, those 45 to 54 years old (IRR 1.67 [1.26, 2.22], p<0.001) and 55 years and older (IRR 1.45 [1.04, 2.04], p=0.031) had a significantly higher rate of transactions. Compared to the housed, the homeless displayed a lower rate of transactions (IRR 0.43 [0.31, 0.59], p<0.001). Compared to age 18 to 34 years, ages 35 to 44 years (OR 2.08 [1.25, 3.47], p=0.005) and 55 years and older (OR 1.94 [1.05, 3.55], p=0.031) associated with increased risk for high utilization status. Compared to the housed, homelessness associated with reduced risk for high utilization status (OR 0.34 [0.15, 0.70], p=0.006). No significant differences in rate of transactions or risk for high utilization status were observed by sex, race, or ethnicity. Analysis of responses to the optional harm reduction survey conducted at participant intake from January 1st, 2021, to July 31st, 2023, revealed: Methamphetamine (64%) and heroin (47%) were the two most reported drugs being injected. About 64% reported injecting drugs between 2-4 times a day. About 18% of participants reported sharing needles and 30% reported sharing injection equipment. About 59% requested naloxone.

Conclusion & Discussion: Older individuals are more likely to make any and frequent use of PHVMs, while those experiencing homelessness are less likely to make any or frequent use of PHVMs. These insights indicate a need for increased outreach to young adults and the unhoused to identify barriers impacting use, hypothesized to include poor awareness of this resource, transportation, and stigma. Additionally, enhancing the incentives associated with the harm reduction survey is paramount to gathering more comprehensive data on participant behaviors and beliefs. In conjoint with previous research, it is imperative for stakeholders to place heightened emphasis on implementing PHVMs as a proactive measure to curb opioid-involved overdose fatalities. Addressing this concern calls for concerted efforts to centralize and standardize data collection methods. Such improvements will serve to enhance the precision for future evaluations.

References: Allen, S. T., O’Rourke, A., Johnson, Jessica. A., Cheatom, C., Zhang, Y., Delise, B., Watkins, K., Reich, K., Reich, R., & Lockett, C. (2022). Evaluating the impact of naloxone dispensation at public health vending machines in Clark County, Nevada. Annals of Medicine, 54(1), 2680–2688. https://doi.org/10.1080/07853890.2022.2121418
Centers for Disease Control and Prevention. (2019). Life-Saving Naloxone from Pharmacies. https://www.cdc.gov/stopoverdose/naloxone/index.html
Otiashvili, D., Kirtadze, I., Mgebrishvili, T., Beselia, A., Tabatadze, M., Vardanashvili, I., & Ober, A. J. (2022). Implementation and evaluation of a syringe vending machine trial in Tbilisi, Georgia. International Journal of Drug Policy, 103, 103649. https://doi.org/10.1016/J.DRUGPO.2022.103649
Russell, E., Johnson, J., Kosinski, Z., Kaplan, C., Barnes, N., Allen, S., & Haroz, E. (2023). A scoping review of implementation considerations for harm reduction vending machines. Harm Reduction Journal, 20, 33. https://doi.org/10.1186/s12954-023-00765-2