Expanding access to harm reduction in rural communities through community-informed public health vending machines

Abstract

Background

The opioid epidemic continues to disproportionately impact rural communities across the United States, where structural barriers, including geographic isolation, limited public health infrastructure, and heightened stigma, restrict access to harm reduction services. Public Health Vending Machines (PHVMs) that distribute naloxone and other wellness supplies (e.g., hygiene, wound care, socks, glasses) offer a promising, low-barrier, anonymous method for increasing access to life-saving interventions. However, the implementation of PHVMs in rural areas remains limited, and few studies have examined how these tools can be effectively and sustainably integrated into such contexts.

Methods

We employed a community-based participatory research (CBPR) approach, guided by the structural indicators of community-based participatory action research (SI-CBPAR). A qualitative needs assessment was conducted in six rural counties in North Carolina. Individuals with lived experience of substance use were trained as interviewers to recruit and conduct semi-structured interviews with peers. A total of 60 interviews were completed between June and December 2024. Participants discussed access to naloxone, stigma, preferred PHVM locations and distribution models, and desired harm reduction and wellness supplies. Transcripts were coded using a priori codes, with coding validation through inter-rater reliability and team-based consensus.

Results

Participants described a range of community-level challenges and assets related to naloxone accessibility, leading to the first overall theme, current community context of naloxone accessibility, with subthemes highlighting the sources of naloxone, its perceived importance, and structural and social barriers to access. Participants also provided input on the implementation of PHVMs (the second theme), expressing preferences for 24/7 access, private locations to reduce stigma, and expanded content to include additional harm-reduction supplies. These findings underscore the need for community-informed strategies to improve equitable access to naloxone and related services.

Conclusion

This study demonstrates that PHVMs are viewed by community members as an acceptable and community-supported strategy for expanding access to harm reduction in rural areas. The findings provided critical insight into the social and contextual factors that shape community readiness for PHVM implementation. The CBPR approach ensured the relevance and cultural alignment of the findings, reinforcing the importance of engaging individuals with lived experience as partners in implementation science. Sustainable deployment of PHVMs in rural communities requires tailored strategies that address local stigma, logistical barriers, and community needs. The results support the development of rural-specific PHVM implementation toolkits to reduce overdose deaths and promote health equity.

Background

The opioid crisis continues to devastate communities across the United States, with rural populations facing disproportionately high rates of overdose and limited access to healthcare services [1]. These structural challenges, including geographic isolation, provider shortages, and heightened stigma associated with substance use, contribute to poor health outcomes in rural areas [2, 3]. The implementation of harm reduction initiatives in rural communities must consider the unique needs of the community. A variety of outreach and harm reduction methods, including mobile units, social networks, mail-order services, and harm reduction cafés, have been implemented by local health departments, faith-based groups, food banks, and grassroots organizations [4, 5]. However, additional innovative strategies are still needed in rural communities to expand access to life-saving interventions. One such strategy is the implementation of public health vending machines (PHVMs) that distribute naloxone, fentanyl test strips, and other wellness and harm reduction materials. PHVMs offer a low-barrier, anonymous, and 24/7-accessible method for individuals to obtain essential supplies, particularly in communities where traditional healthcare infrastructure may be sparse. In addition to traditional vending machines, communities have implemented wall-mounted and pole-mounted distribution boxes, as well as converted newspaper stands to dispense naloxone [6, 7].

Internationally, PHVMs have been used for decades to dispense sterile injection equipment and other harm reduction supplies (some may refer to these vending machines as Harm Reduction Vending Machines). Only recently have PHVMs been scaled up in the US, and very few currently operate in rural settings. As a result, much of the existing evidence on feasibility and implementation comes from work conducted outside the US. Russell et al., (2023) conducted a scoping review of 22 studies, all from non-US settings, and identified several best practices for vending machine implementation, including maintaining 24/7 accessibility, incorporating syringe disposal options, enabling routine data collection, and ensuring anonymity of use [8]. These recommendations, however, were not developed specifically for rural communities. A more recent systematic review of 45 articles published between 1994 and 2023, the majority from outside the US (n = 35), found that PHVMs can mitigate some of the stigma associated with substance use and improve reach to high-risk populations who experience difficulty connecting to traditional services [9]. Individual country studies reinforce these findings. In Norway, PHVM were reported as a successful way to provide harm reduction supplies to groups considered “hard to reach” [10]. Australian evaluations similarly highlight the importance of confidentiality and anonymity, particularly in smaller or rural communities [11]. In Germany, an evaluation of a statewide vending machine program found that reach was greater in rural areas than in urban areas, suggesting that PHVMs may be especially valuable where conventional harm reduction infrastructure is sparse [12]. PHVM implementation, naloxone access, and harm reduction practices are highly contingent on country-specific legal, regulatory, and funding environments, and even within the US, state-level variation in harm reduction policy makes cross-context comparisons complex. Taken together, this international literature points to the promise of PHVMs, while underscoring the need for context-specific evidence on how they can be designed and implemented within rural US communities.

A growing body of US evidence on PHVMs has recently emerged. In rural Appalachian Kentucky, Marschke et al. found that people who use drugs preferred vending machines that were discreet, easily accessible, and stocked with items that extended beyond traditional harm reduction supplies [13]. In the same region, Knudsen et al. reported that PHVMs were viewed as more acceptable and appropriate than traditional syringe service programs [14]. Studies have also documented willingness to use PHVMs among rural populations. On the Bois Forte Reservation, more than half of surveyed individuals who use drugs (56%) indicated that they would use a PHVM if one were available in their community [15]. Among young adults who misuse opioids, Wagner et al. found that participants were willing to use PHVMs, particularly when machines were located in convenient places and offered 24-h access [16]. It has been found that the people accessing PHVMs tend to be younger than those who use traditional harm reduction services [17]. There is also evidence within the US that when PHVMs are implemented, they are effective in reducing overdose fatalities. For instance, the Trac B Exchange program in Clark County, Nevada, observed immediate reductions in overdose fatalities following the introduction of naloxone dispensing PHVMs [18]. Similar outcomes have been reported in other urban and suburban settings where vending machines have been deployed across public spaces, medical facilities, and detention centers [19–25].

Despite these promising outcomes, the implementation of PHVMs in rural areas remains limited. Rural residents often express a strong desire for increased access to naloxone and other harm reduction supplies. Yet, barriers such as stigma, fear of law enforcement, and lack of tailored implementation strategies persist [2]. Furthermore, few studies have examined how PHVMs can be sustainably implemented and maintained in rural communities with unique cultural and logistical needs.

North Carolina (NC) is a prime example of challenges impacting rural communities. With 80 out of its 100 counties classified as rural, the state has experienced a stark increase in drug-related overdose deaths, particularly in rural areas [26]. The age-adjusted rate of drug overdose deaths in NC for 2023 was 33.7 per 100,000 residents [27]. In 2022 alone, the state’s age-adjusted drug overdose mortality rate nearly doubled the national average [28]. These overdose deaths carry significant economic and social consequences, with the combined cost of opioid use disorder and fatal overdoses reaching over $39 billion [29]. Public health initiatives in NC have prioritized expanding access to naloxone, including through the implementation of PHVMs. Similar to other states, PHVMs have been primarily implemented in urban or suburban detention centers; thus, rural counties still lack widespread access to this critical infrastructure.

We sought to address this gap in naloxone access by conducting a community-engaged project to implement PHVMs in six rural NC counties: Carteret, Jackson, Surry, Stanly, Cabarrus, and Swain. To address the pressing need for PHVM implementation that considers the specific needs, assets, and lived experiences of people in rural communities, we utilized a community-based participatory approach for this project. Our partnership consisted of representatives from the North Carolina Harm Reduction Coalition (NCHRC) and Community Impact North Carolina (CINC), individuals with lived experience of substance use who resided or worked in the six rural counties, and academic researchers. The local community members who have personal experience with substance use were trained by the academic researchers to recruit and interview peers with lived experience of substance use about their preferred locations for PHVMs, desired supplies to be dispensed in the PHVMs, and perceived barriers and facilitators to the utilization of PHVMs. By centering the voices of people with lived experience as both researchers and study participants, the partnership was able to consider the cultural, social, and structural determinants of health that shape substance use in rural areas to inform the development of tailored implementation strategies to enhance harm reduction service delivery and reduce overdose deaths in underserved rural communities.

Methods

Study design

Community-based participatory approach

This study utilized a community-based participatory research (CBPR) approach to identify, recruit, and train individuals with lived experience to serve as interviewers for a qualitative needs assessment focused on the optimal placement of PHVMs stocked with naloxone. Community engagement was central to every phase of this work, and local harm reduction partners played a crucial role in shaping the recruitment and training process.

The CBPR approach is a collaborative methodology that actively involves community members, researchers, and other stakeholders throughout the research process. CBPR emphasizes the co-creation of knowledge and acknowledges the unique strengths each partner contributes to the research endeavor. It offers a framework for the equitable involvement of all participants, fostering mutual respect, shared decision-making, and the integration of diverse perspectives [30, 31]. CBPR is an essential strategy for engaging communities in research that directly impacts their lives. To guide our approach, we applied the Structural Indicators for Community-Based Participatory Action Research (SI-CBAR) framework [32]. The framework includes six key categories that are important to CBPR. They include 1) study initiation and funding, 2) coalition engagement, 3) research implementation, 4) community capacity development, 5) dissemination and attribution, and 6) power sharing. Each category has a series of indicators to help focus on the structural aspects of the project [32]. This study focused on the development of meaningful partnerships with people who use drugs in our target communities. These relationships were foundational to producing research that supports desired implementation outcomes of increasing naloxone and other harm reduction supplies via a PHVM [33].

People with lived experience

This study engaged individuals who have lived experience throughout the entire research process. Prior to the grant being written, a partnership was formed that included people with lived experience to discuss the mechanisms for improving naloxone distribution in rural counties. During this time, ideas were shared and finalized concerning improving naloxone access via the implementation of PHVMs. Our community partners with lived experience provided letters of support for the grant application, and funding was included in the budget for these partners (SI-CBAR steps 1 & 4). Upon receiving the grant award, the study design and materials (e.g., interviewer training, recruitment, data collection) were developed in collaboration with individuals who have lived experience. One aspect of this was reviewing the materials for stigmatizing language and usability of the materials (SI-CBAR steps 2 & 3). A key component of our data collection method was to hire and train people with lived experience (interviewers) to recruit and interview our study participants who also had lived experience of substance use (SI-CBAR step 3). The inclusion of peer interviewers from the same communities as participants was not merely a methodological choice, but a recognition of their unique subject knowledge. The early and ongoing collaboration with local partners played a crucial role in ensuring that interview questions reflected the realities of participants’ lives. Our team is currently applying the results of this study to inform the implementation of the PHVM in each community (SI-CBAR step 3). Dissemination of the results is a collaborative effort among all the study team members (SI-CBAR step 5). Throughout the entire project, the partnership worked collaboratively with each community to determine the best way to implement and sustain a PHVM (SI-CBAR step 6) [32].

Interviewer recruitment and training

In collaboration with our community partners, NCHRC and CINC, the academic team conducted virtual town hall meetings with substance use prevention providers from the selected counties. These meetings served to introduce the project and initiate discussions about potential community members who could be recruited as interviewers. Follow-up one-on-one meetings between the academic team and local providers or community delegates were arranged and facilitated by NCHRC and CINC to identify suitable candidates to be interviewers.

Community partners were asked to assist in recruitment by identifying local organizations that work with individuals with lived experience. These organizations included peer recovery groups, faith-based initiatives, re-entry programs, health departments, and post-overdose response teams. The community partners were also asked to facilitate “warm handoffs” via introductory emails. Eligible candidates to be interviewers were required to be 18 or older, reside in one of the six selected counties, have lived experience with substance use (e.g., opioids, fentanyl, cocaine, or psychostimulants), and possess prior experience working or volunteering in harm reduction. References were requested to confirm experience and suitability for the role.

Interested candidates submitted an application that detailed their lived experiences, prior involvement in research (either as participants or collaborators), community engagement, and harm reduction efforts. The academic team, in consultation with local partners, conducted individual interviews with each candidate to evaluate their alignment with project goals and readiness to serve as interviewers. Those selected were hired as consultants and received $100 per interview (10 interviews per county).

After being hired, the interviewers took part in a 2.5-h virtual training developed by the academic team and community partners. The training included a customized module adapted from the Collaborative Institutional Training Initiative (CITI) to improve accessibility and relevance for community members. The adapted CITI training was developed in collaboration with the institution's IRB, individuals with lived experience, and the academic team. It covered all core topics from standard CITI training, including the history and ethical principles of research, human subjects’ protections, informed consent, investigator responsibilities, federal regulations, assessing risk, privacy and confidentiality, vulnerable populations, conflicts of interest, and unanticipated problems. Topics for the training covered included the adapted CITI training, research ethics, informed consent, conducting semi-structured interviews, data collection and analysis, safety protocols, and dissemination of findings. Interviewers received training in participant recruitment using a variety of outreach materials, including flyers, recruitment scripts, and pass-along cards, which were sent to them in advance via the US Postal Service. Interviewers also engaged in guided practice using the interview guide, including probing techniques, note-taking strategies, and approaches for adapting questions to participants’ comfort and communication styles. Training additionally covered how to safely conclude an interview if they sensed discomfort or felt unsafe.

During initial planning and community engagement, the research team asked community members and interviewers not to share their preferences for PHVM locations or machine types with other team members or participants until all interviews were completed. This approach was used to minimize bias in both data collection and analysis, and to avoid influencing community partners involved in implementation. After each interview, the interviewer met with the Principal Investigator for a brief debriefing session. These meetings provided an opportunity to reflect on the interview experience, troubleshoot recording issues, review field notes, plan subsequent recruitment, and refine probing questions as needed.

A comprehensive description of the protocol for identifying, hiring, and training community interviewers, created in partnership with our community partners, is available for review [34]. The study was deemed expedited by the East Carolina University Institutional Review Board (UMCIRB 23–001800).

Recruitment

Study recruitment was achieved through word-of-mouth, advertisements by harm reduction organizations across NC, health departments, community health workers, and snowball sampling among interview participants. To be eligible to participate in the interview, the participant was required to be over the age of 18, reside in the county where the interviews were being held, and have used opioids, fentanyl, cocaine, and/or psychostimulants in the last two years. Interview participants’ payment options were made in collaboration with the community partners, the academic team, and the interviewers. Two factors contributed to the decision to use gift cards as payment. The first being that the funders for the grant had specific requirements concerning the use of payment, e.g., participants had to sign a waiver that they would not use the funds for firearms, tobacco, or alcohol products. The second was that cash payments required extensive paperwork and delays due to the institutions’ procedures. Due to this, the team made the collective decision to use gift cards as participant incentives. The team worked collaboratively to decide on the type of gift card; the gift card type was based on each county's decision. Each participant received a $30 (USD) gift card for their participation. Interview anonymity was encouraged for participant protection; thus, an alias was used for each participant during the interview. Interviews were conducted in a private setting, such as parks, private clinic rooms, or other pre-identified locations, including parking lots or recovery homes.

Data collection

Data collection occurred between June and December 2024. The team of five trained interviewers with lived experience of substance use conducted face-to-face interviews lasting up to 1 h with 60 people (10 people in each county). At the time of the study, none of the six participating counties had a PHVM in place. As such, participant responses reflect anticipated perceptions and potential barriers rather than direct experience with PHVMs.

To mitigate social desirability bias, we used peer interviewers with lived experience, private interview locations, and emphasized confidentiality and anonymity prior to each interview. Questions were open-ended, nonjudgmental, and framed to normalize a range of experiences. Interviewers were trained to avoid leading prompts and to reassure participants that both positive and negative feedback about PHVMs was valuable.

Informed consent procedures

Prior to each interview, participants were provided with detailed information to support informed consent. The interviewer reviewed the purpose of the study, the reason the individual was being invited to participate, and potential reasons they might choose not to take part. Participants were informed of their right to decline participation and were provided with alternative options if they chose not to continue. The interviewer explained the estimated duration of the interview and what participation would entail, including responding to questions related to PHVMs and naloxone. Participants were also informed about the length of time their interview transcript would be retained, what to expect during the interview, and how and when they would receive compensation.

Confidentiality was discussed in detail, including the assignment or selection of an alias, to protect participant identity. The interviewer described the measures that they and the academic team would take to safeguard participants' data and explained their right to withdraw from the study at any point without penalty. Contact information for the academic team was provided in case participants had questions or concerns. The informed consent process was conducted privately between the participant and the interviewer and was not audio-recorded.

Interview guide

The development of the semi-structured interview guide was a collaborative process that centered on the voices and expertise of individuals with lived experience of substance use. These collaborators contributed to the wording, sequencing, and relevance of the questions to ensure the interview guide was respectful, non-stigmatizing, and grounded in community realities. The goal was to elicit meaningful feedback on naloxone access and the acceptability of different naloxone distribution models.

Before beginning each interview, participants were asked to provide verbal consent by responding to the prompt: “(Alias), do you agree to participate in this study?” If a participant declined, the interview was respectfully concluded, and the recording was stopped. For those who agreed, we began by collecting brief demographic information, including age, gender, and race.

The interview then transitioned to a series of open-ended questions designed to understand how community members currently access naloxone, their experiences with accessing naloxone through pharmacies, and the broader challenges associated with obtaining naloxone in their community. Follow-up probes were used to gather more details on issues such as stigma, cost, and preferences for accessing naloxone.

To explore potential implementation strategies for improving community-wide naloxone access, participants were shown images of four proposed distribution models: (1) a public health vending machine, (2) a wall-mounted vending machine, (3) a free-standing outdoor distribution stand, and (4) a pole-mounted distribution box. Each option was described, and participants were asked to provide feedback on the perceived fit and feasibility of each option for their community. They were then asked to select their preferred model and explain their reasoning.

Additional questions focused on identifying ideal locations for naloxone placement, suggestions for health-related items to accompany naloxone in the distribution units, and ideas for community outreach and promotion. Each interview concluded with an opportunity for participants to share any additional thoughts.

Analysis

We conducted a reflexive thematic analysis following the six phases outlined by Braun and Clarke: (1) familiarizing ourselves with the data, (2) generating initial codes, (3) constructing themes, (4) reviewing potential themes, (5) defining and naming themes, and (6) producing the final report [35]. Interviews were audio recorded, transcribed, deidentified as necessary, and checked for accuracy by the Principal Investigator, Project Manager, and a Research Assistant. We used a hybrid inductive/deductive approach, beginning with a set of a priori codes informed by our research aims while allowing additional codes and themes to emerge organically from the data. The five a priori codes were (1) current naloxone accessibility in the community, (2) PHVM location, (3) items in the PHVM, (4) preferred type of PHVM (i.e., traditional vending machine, wall or pole mounted, converted newspaper stand), and (5) stigma.

Two researchers, independently coded all transcripts. An initial coding conference was held to identify and manage discrepancies. A qualitative coding framework was then developed to guide the analysis of participant responses related to naloxone access and PHVM implementation. Consistent with CBPR principles and the SI-CBPAR process, we actively involved the interviewers and community partners. As a team, prior to data collection, we decided that the initial coding would be conducted by the academic team, with themes and codes subsequently reviewed and confirmed by the community partners and interviewers. We met individually with each interviewer to review the preliminary coding structure and discuss quotes within each code category. Interviewers contributed critical context and interpretation, offering insight into cultural nuances, community dynamics, and the meaning behind participants’ language that may have otherwise been overlooked. Because our interviewers were residents of the counties where they conducted the interviews, they were able to assist with analyzing community needs, local nicknames, and references to people and places within their communities.

Codes were subsequently clustered into higher-order categories, from which overarching themes and subthemes were derived. Themes reflected shared patterns of meaning relevant to the research questions, while preserving the specificity and lived experience conveyed in participant accounts. This analytic process resulted in two primary themes (1) Current Community Context of Naloxone Accessibility and (2) Implementation of PHVMs. The theme Current Community Context of Naloxone Accessibility encompassed three subthemes: 1) Sources of Naloxone; 2) Importance of Naloxone; 3) Barriers to Accessing Naloxone. The theme Implementation of PHVMs reflected perspectives on integrating community-based naloxone distribution through vending mechanisms. It included three subthemes: 1) Location of PHVMs; 2) Type of Public Health Machine; 3) PHVM Items. (See Table 1.)

Table 1.

Exemplar quotes on naloxone access and PHVM implementation

Themes	Subthemes	Define	Exemplar quote
Current community context of naloxone accessibility	Sources of naloxone	Describing where and how people currently obtain naloxone in their communities	“People phone friends, you call, you know, hysterically until you can find somebody that maybe has some (naloxone). So, you don't have to do the last resort of calling the ambulance.”
	Importance of naloxone	Reflecting community members' understanding of its life-saving potential	“Most people who are dealing with those around them who are using aren't familiar, with naloxone, obviously people who use are familiar with it, but those are the ones that it's critical to have loved ones and family and friends also be informed.”
	Barriers to accessing naloxone	Structural barriers (availability, transportation, cost)	“More or less (retracted town name) to where it's so far to get to any kind of emergency services. I mean, you got all the way to (retracted town name) and there's nothing, but a couple convenience stores. I mean, if you need to save a life, you need to be able to get to it quickly.”
		Social barriers (personal discomfort, unawareness, stigma)	“Nobody likes to go anywhere and basically admit that they're using drugs…but you know, you're still in your mind, you don't want to go put yourself out there.”
Implementation of PHVMs	Location of PHVMs	Preferences for 24/7 access	“So pretty much having like a little vending machine, kind of like Coca-Cola or soda machine to where someone can walk up, put money in and they come out. I think that would help out because then you wouldn't have to worry about times available to get it being closed to get it. You're able to get it 24/7 when needed.”
		Machines located in private versus public spaces	“Private. Definitely a private environment. It's like a stigma thing. Like it doesn't matter if we're an addict or not an addict. If we go ask them for some (naloxone), then they're automatically going to say that's an addict.”
	Type of PHVM	Perceptions of the different distribution models	“The vending machine. Yeah, it would probably be best; because honestly, with the polices around here; if you get seen at the needle exchange or anything like that, you're automatically put into a, you know, oh well he's a drug user.”
	PHVM Items	Suggestions for additional health-promoting items to be included	“Condoms, syringes, alcohol pads, sanitary products, napkins, Band-Aids even. I mean, like, any kind of thing for cuts. First aid equipment stuff. Test strips.”

Once coding was completed, the Principal Investigator coded a sample of interviews to examine inter-rater reliability. No discrepancies were identified. At which time, the coding scheme was finalized by the interviewers and academic team. A conference was held with the entire study team, which included our community partners and the interviewers, to confirm observations and to determine priority findings after coding was completed. The study team disseminated the results to each of the six counties via town hall meetings.

Results

Sixty interviews were conducted about the lived experiences of opioid users in six rural counties in North Carolina. The sample included 27 (45%) males, 31 (52%) females, and 2 (3%) who stated they preferred not to answer. The median age of the sample was 38.7 years. Participants’ race included White (n = 41; 68%), African American (n = 5; 8%), Native American/American Indian (n = 9; 15%), and declined to share (n = 6; 10%).

Current community context of naloxone accessibility

This code examined participants’ experiences and perceptions related to obtaining naloxone in rural communities. Participants identified various sources for obtaining naloxone. Interview responses highlighted several challenges, including identifying structural and social barriers.

Sources of naloxone

Participants identified several sources where naloxone could be obtained, including hospitals (particularly emergency departments), emergency medical services (EMS), pharmacies, churches, community centers (e.g., Restoration House, Grace Place), fire departments, recovery and treatment centers, syringe service programs, and public health offices.

“Different services like (retracted - health facility). I believe the hospital. I believe the police station. I've also heard rumors that the fire department was giving it out one time.”

Informal sources were also mentioned, such as friends, acquaintances actively using substances, and the individuals who were selling the substances.

“My drug dealers…when I would buy drugs, my people would give me something to have.”

Importance of naloxone

Another recurring theme was that everyone needed to carry naloxone. Participants often emphasized the importance of family, friends, and community members being prepared with naloxone.

“I mean everybody, I mean everybody, because even if somebody does not or doesn't have a past or present drug use, or they might have a friend or a family member and so it's important that anybody can get ahold of it.”

Barriers to accessing naloxone

Even when naloxone was technically available through formal sources, obstacles were often reported that made it effectively inaccessible.

Structural barriers: Participants reported structural barriers to naloxone access, including limited availability, transportation issues, and high costs.

Availability: During the interviews, participants consistently reported difficulties in obtaining naloxone. Across all counties, participants reported that naloxone was not always readily available in their communities. One participant plainly stated, “Truthfully, no,” when asked whether naloxone was accessible when they needed it. Another participant shared:

“Most of the time when I've seen that somebody needs it, it's always at 3:00 in the morning or something, or they're in the car going down the road, and their brother or sister or whoever, their girlfriend or husband just went in the bathroom at McDonald's while they were stopped and done something and then when they get back in the car and they're headed down the road, then they're overdosing; and there's nowhere to get it fast enough for it to help them anyway.”

Others echoed this sentiment, with one describing multiple instances where it was unavailable during emergencies:

“There’s been times where we’ve needed some, and it ain’t been there.”

Another participant emphasized the broader implications of limited access due to living in a rural area and not having 24/7 access:

“People overdose every day, multiple times a day.”

Participants noted logistical challenges, such as the unavailability of naloxone outside standard business hours. One participant explained,

“If it is after hours, then we wouldn’t be able to access it…feel like sometimes it could be late at night and someone wants to go get it, and they can't get it because, say, if you can get it from a pharmacy, the pharmacy is closed. Well, now you can't get it.

Transportation: Often, rural counties are spread out, and some of the areas where overdoses occur are not closely located to healthcare services. Participants expressed that having to travel significant distances impeded access:

“Having to walk across town is the most challenging thing, for a lot of us.” “In the most rural places, the most remote places of our county, you have to drive for like 20 or 30 minutes just to get to, you know, civilization.”

Cost: For many, financial barriers further complicated access to naloxone. Some participants cited the high cost of naloxone as a prohibitive factor:

“It’s expensive. I've always heard it's really expensive. Because uh, probably a lot of homeless people are probably the ones that overdose not only, but a lot. Like, you know, a large amount is probably homeless or somebody who can't, you know, somebody for some reason or another can't afford to be able to do stuff like that.”

Social barriers: Access was frequently constrained by social factors, including personal discomfort, knowledge gaps, and stigma.

Personal discomfort: Participants described a reluctance among some community members to engage with public services due to social anxiety or fear of exposure. As one individual explained,

“A lot of people out here don’t like to go out in public and don’t like to be around other people, so they’ll skip getting things that they need just to avoid going out around people.”

Others noted that lack of knowledge and awareness also contributed to the underutilization of naloxone:

“Everyone that uses ain’t always smart enough to get Narcan on their own. Lots of people think they don’t need it.”

Knowledge gaps: Several participants described gaps in their knowledge of naloxone, including uncertainty about what it is and where to obtain it. Multiple individuals in each county were unfamiliar with the term naloxone and referred to it by its brand name, Narcan. Participants also noted not knowing where naloxone was available in their community (e.g., pharmacies, public health departments). One participant reported:

“I feel like it's not widely known that you can get it. Obviously, I didn't even know that you could get it from a pharmacy. But it's not really known of places where you can get Narcan safely.”

Stigma: Stigma surrounding opioid use remained a significant factor in one’s obtaining naloxone. Participants described feeling judged as one individual shared:

“I mean, there's a lot of people that judge other people and they don't know the situation or that they could have overdosed on an accident.”

A common trend was reported that people had experienced strong negative feelings from community members due to their use of substances.

“People have their own opinion of what that person may be because they use drugs. Just because they use drugs, doesn't mean they're a bad person because they use.”

Many of the participants reported that the judgment and feelings often lead to being treated poorly.

“People around are so used to being degraded and put down for the things we do.”

Participants expressed discomfort when requesting naloxone at pharmacies, fearing judgment or criminalization:

“You know, you got to ask for it. And they looking at you like you’re some kind of criminal or something.”

Stigma was also one of the key contributing factors for the location of the PHVM. This is discussed in the upcoming section.

Implementation of PHVMs

PHVM Location

Determining the best location involves considering several factors, including privacy, accessibility, and acceptance. Participants provided recommended locations that were both specific and general areas that would benefit the community. Insights on 24/7 access and preferences on whether the PHVM should be in a private or public space, and how stigma plays a role in that opinion.

Determining the location of the machine was an extensive process because each person interviewed described multiple locations, but did not always know the correct name for the place. Often, there were nicknames for certain county areas that might not be familiar to those outside the substance use culture. For example, in one county, multiple participants referred to “the island” as the ideal location. This location is the county pavilion. Other participants provided specific names of organizations within their community, such as what one participant shared: “I think the restoration house in (retracted name of county),” which is a social service organization in their community.

24/7 Access: Participants consistently suggested that access to PHVM in their communities should be available 24/7.

“I think it should be where anybody can access it any time. I think it should be available 24 hours a day, seven days a week. Okay. Period.”

Not having to interact with people when obtaining naloxone was mentioned by multiple participants. This is highlighted by this participant’s quote:

“A lot of people out here don't like to go out in public and don't like to be around other people, so they'll skip getting things that they need just to avoid going out around people.”

Private versus public spaces: For many participants, determining the most suitable location for PHVMs was significantly influenced by concerns about stigma. Participants often expressed the desire to access harm reduction supplies discreetly, without fear of judgment from others in their community.

“A coworker could see you purchasing that, you know, that could cost you your job because, I mean, my employer probably would not like knowing that I use drugs, it could cost my job. School teachers, you know, my kids, teachers could see that, you know, then, you know, teachers will talk about it. And then my kids will get treated differently at school. And there's a lot of stigma associated with that.”

Nearly all participants reported a preference for locations that provided a level of privacy or anonymity, such as areas that were less visible, open 24/7, or not directly linked to law enforcement or healthcare institutions.

“I feel like people would rather get it privately because they don't want to be known as someone that uses drugs. I mean, a lot of people are ashamed by it, but I feel like if it were public, a lot of people would still access it because it's something that we need. It's not want. It's something that we need.”

Some participants mentioned that placing PHVMs in or near commonly visited, yet neutral, community spaces, such as gas stations or laundromats, could help reduce the stigma associated with their use.

“I like public places, libraries, laundromats, convenience stores. Shopping centers like Walmart, Food Lion. Everyone goes grocery shopping. People go to Walmart. They visit those places. Fast food. Everyone eats fast food. So, fast food restaurants and a lot of times convenience stores are open 24 hours a day. Some of them are open all the time. So, in a real emergent situation, it would, you know, if it were, if it were placed in, in public places like that, then there would be full access with no stigma and not having to go explain to the pharmacist why you need it.”

Overall, participants emphasized the importance of the PHVM being as private and discreet as possible to avoid community stigma.

Type of public health machine

During the interviews, the participants were given descriptions and pictures of the four different machine options. The options included a traditional vending machine, a wall-mounted machine, a pole-mounted box, and a newspaper stand. Interviewers asked participants to select the option they believed would be most beneficial for their community. Participants shared their thoughts on each type of machine.

Overwhelmingly, most participants reported a preference for the full-size vending machine as it could hold more supplies.

“I say the vending machine, okay, because it holds more and there's probably more people that would need it. And depending on how, how much, like how, how long it takes them to restock it or whatever, um, people would probably go to it more.”

Some participants preferred the wall mount because it could still hold multiple items but also had the option of being installed in a more private location. As stated by one participant:

“I like that for a bathroom a lot better because you can just go in there. It'd be private, discreet. You wouldn't have to face any stigma. You know it'd be happening.”

Yet other participants felt that having the wall-mounted machine in a bathroom or other closed location was not the best idea, as stated here:

“It needs to be out in the open to further the purpose of awareness.”

Some appreciation was expressed for the design of the newspaper stand-style option, noting that the discreet appearance could help reduce stigma associated with accessing harm reduction supplies. However, concerns about safety and oversight, particularly regarding the potential for children to be drawn to the machine out of curiosity.

“It's really a sleek model, but the thing is that kids would be attracted to stuff like that, and you would really have to explain to the children what that is. If it's in a regular vending machine, they know all types of stuff come with vending machines.”

Additionally, concerns were expressed that the newspaper stand option was not secured properly and that all items could be removed.

“I think the newspaper stand thing is a good idea. You just have to find some way to limit how much is taken at once. You know the design is going to be the difficulty.”

A few participants went as far as to say that if the newspaper stand-style was not secured, it would be stolen.

“I don't like the one that's movable because if you put it outside, it would probably taken down. They take the whole damn thing. The whole thing.”

The pole-mounted option was the least favorable out of each item. The main reason participants expressed concern about this mechanism was the limited capacity to hold items (only 6 items). They questioned the practicality of such limited-capacity machines in meeting the diverse needs of individuals who rely on harm reduction resources in rural areas that may not have access elsewhere. One participant noted that a machine with so few items could quickly run out of essential supplies, especially in high-need areas, reducing its overall effectiveness.

“How fast would it run out? In a day and a half. A day you'd have to constantly refill it. Unless you had 100 of them all throughout the city.”

PHVM Items

Participants were asked to identify items other than naloxone that they believed were needed in their community. Items included other harm reduction products (e.g., sterile needles, testing strips), hygiene items, over-the-counter medications, resource pamphlets, and non-perishable groceries. During the interviews, when asked about different items, many participants referenced specific items because they believed it would benefit the unhoused population.

“Maybe some wipes or something that people a lot of people don't have access to a shower. A lot of the addicts are homeless, so they don't have access to things like that. I don't.”

Participants also mentioned that having pamphlets or resources posted on or inside the machines would be beneficial, such as information on treatment and recovery options, job openings, housing, and other social service support.

“Informational pamphlets…like rehab.” “Any kind of resources.

Housing application. Anything job. Recent job or opening job.”.

Many participants emphasized the importance of stocking the PHVM with a variety of supplies that support overall health and well-being. Suggested items included syringes, condoms, and basic first aid supplies.

“Condoms, syringes, alcohol pads, sanitary products, napkins, Band-Aids, even. I mean, like, any kind of thing for cuts. First aid equipment stuff. Oh, maybe like the test strips. Like, the xylazine test strips.”

Discussion

This study effectively employed a CBPR approach to evaluate access to naloxone and the implementation considerations of PHVMs in six rural counties in North Carolina. Sixty qualitative interviews, 10 per county, were conducted with individuals who have experienced substance use, yielding rich, contextually grounded accounts of how overdose risk, service availability, stigma, and geography intersect in rural communities. These findings speak directly to rural public health infrastructure, highlighting how current systems often leave naloxone and other harm reduction supplies technically available yet functionally out of reach. In the sections that follow, we discuss how participants envisioned PHVMs as a concrete, rural-specific strategy to address these gaps, focusing on structural and social barriers to naloxone access, design and placement preferences for PHVMs, and implications for building sustainable, community-led harm reduction infrastructure in rural settings.

Rural public health impact

This study demonstrates how rural residents with lived experience of substance use envision public health vending machines (PHVMs) as a concrete solution to long-standing gaps in naloxone access. Thus, providing a mechanism to improve health outcomes, thereby contributing to an emerging body of research advocating for scalable, community-rooted public health infrastructure in rural areas [36–39]. Participants described a landscape in which naloxone is technically available through hospitals, EMS, pharmacies, and public health agencies, yet often functionally out of reach due to distance, limited hours, cost, and stigma. Studies have repeatedly called for flexible services that create meaningful opportunities for engagement with harm reduction practices [40], and our findings underscore the urgent need for low-barrier, community-informed interventions in rural settings [41]. Participants described significant, multilayered challenges to obtaining naloxone and meeting broader harm reduction needs, even when resources were technically available. In this context, participants saw PHVMs not as a generic one-size-fits-all harm reduction intervention, but rather as a rural-specific, locally tailored tool that could have the capacity to meet people where they are.

Our findings highlight several specific ways PHVMs may address structural barriers in rural settings. Existing literature demonstrates the efficacy of PHVMs in urban and suburban areas, where they have been associated with reductions in overdose deaths and increased community engagement with harm reduction services [24, 42–45]. However, rural adaptation remains limited despite documented disparities in overdose mortality rates and healthcare infrastructure between rural and urban communities [1–3]. Participants repeatedly emphasized the mismatch between overdose risk and service hours, describing overdoses occurring “at 3:00 in the morning” or in remote locations where “there’s nowhere to get it fast enough.” In this context, 24/7 access was not a preference but a necessity for meaningful overdose response and is recognized as a best practice [8]. Locating PHVMs in neutral, commonly visited spaces such as gas stations, laundromats, or grocery stores was seen as a strategy to reduce travel time and incorporate naloxone within everyday routines. These accounts suggest that, in rural communities where health care facilities are sparse and far apart, PHVMs can serve as decentralized harm reduction care sites.

These dynamics, rooted in stigma, created a tension between privacy and visibility that directly shaped preferences for PHVM location and machine type. Many participants favored discreet placement (e.g., in or near but not inside highly surveilled spaces, or in neutral community locations) to minimize the risk of being observed. At the same time, some participants argued that PHVMs “need to be out in the open to further the purpose of awareness.” For many, this “awareness” referred to making the machine, and by extension, naloxone and other supplies, easy to find for people at risk of overdose and for those who could respond. For others, the statement also reflected a broader desire to signal that overdose and harm reduction are visible, legitimate public health concerns in their community. Together, these perspectives suggest that rural PHVM implementation must balance two goals: ensuring that machines are visible enough to be discovered and normalized, while preserving sufficient privacy to protect individuals from stigma. Design strategies must explicitly account for stigma, as prior studies have identified it as a central barrier to seeking substance use care [46, 47], and our participants similarly described avoiding pharmacies and public settings for fear of judgment. Placing machines in neutral yet busy locations, providing clear signage, and offering multiple machines across a county without requiring face-to-face requests for supplies may help reconcile these competing needs.

Participants’ strong preference for full-size vending machines over smaller or less secure options was similarly pragmatic and context-specific. Larger machines were perceived as better suited to rural areas where restocking may be infrequent, and demand for a variety of supplies is high. Participants also valued that a full-size machine could hold multiple types of items, such as naloxone, hygiene products, safer-use supplies, socks, and resource information on social service and treatment options, so that if someone were seen using the machine, observers would not know which specific item was being accessed, potentially reducing stigma. Concerns that smaller or unsecured machines would quickly empty, be tampered with, or even be stolen underscore the importance of capacity, durability, and tamper resistance when implementing PHVMs outside densely resourced urban settings. That participants spontaneously weighed issues such as capacity, safety, child access (e.g., worries that children might be drawn to a small, attractive box), and stigma illustrates how lived experience can inform practical design decisions that might otherwise be overlooked.

Cost emerged as another critical barrier that PHVMs alone cannot solve. Participants perceived naloxone as expensive, particularly for people experiencing housing instability or financial hardship. Although our study did not test specific pricing models, participants’ accounts support the implementation of PHVMs as low- or no-cost distribution points. In response, our team worked with community partners to develop sustainability plans that would allow naloxone to be offered free of charge through PHVMs, supported initially by grant-funded bulk purchasing. Future research should examine how different funding and stocking models influence uptake and equity in rural PHVM programs.

Finally, this work underscores the value of involving rural residents with lived experience in every stage of PHVM implementation. Consistent with calls for substance use care and harm reduction interventions to be developed through open, exploratory processes rather than predetermined models of health engagement [33, 41], peer interviewers and community partners contributed detailed local knowledge, such as nicknames for locations, informal supply networks, and nuanced understandings of stigma, that directly shaped where machines should be placed, which machine types would be acceptable, and what items should be stocked. Rather than relying on “grass tops” leaders and other decision-makers to design harm reduction interventions, this study demonstrates how community-driven insights from people with lived experience can be translated into specific, actionable design features for PHVMs in rural settings. Taken together, these findings suggest that PHVMs, when co-designed with rural communities, have the potential to transform naloxone and harm reduction access from sporadic, institution-bound services into an embedded, everyday part of rural public health infrastructure. By aligning machine placement, visibility, privacy, and contents with the lived realities of rural residents, PHVMs can help close critical gaps in overdose prevention and advance health equity in communities that have long been underserved.

Limitations

This study utilized a CBPR approach. While this approach supported meaningful collaboration with community stakeholders, it also imposed certain constraints on the types of data that could be collected and the depth of information that participants were willing or able to share.

Despite the efforts to mitigate social desirability bias, some participants may have overstated support for the PHVMs, minimized experiences of stigma, or reported more inaccurate naloxone access. Participants may have felt they needed to align with perceived expectations. As such, findings may be a combination of genuine experience and participants’ efforts to present themselves in a manner they perceived as desirable. The reliance on self-reported data may have introduced recall bias or social desirability bias. Moreover, although efforts were made to ensure a diverse range of participant perspectives, some voices, particularly those of individuals experiencing the most severe marginalization, may not have been fully captured [48].

The rural setting of the study and the sensitive nature of substance use conversations constrain the transferability of these findings beyond the specific counties included. Consistent with the aims of qualitative inquiry, the results are context-bound and intended to offer in-depth insight rather than broad generalizability.

Conclusion

This study illustrates how rural residents with lived experience of substance use conceptualize PHVMs as a practical response to persistent gaps in naloxone and harm reduction access. Participants described PHVMs as a rural-specific, community-tailored strategy that could provide 24/7, low-barrier access to supplies in locations that fit the rhythms and realities of their daily lives. Their perspectives highlight that the success of PHVMs will depend not only on installing machines but on careful decisions about placement, visibility, privacy, and cost in communities where health care infrastructure is limited, and stigma remains pervasive.

By engaging peer interviewers and community partners throughout the research process, this work demonstrates how CBPR can generate actionable guidance on PHVM design, including preferred machine types, stocking priorities, and location strategies that address both structural and social barriers. These findings offer a concrete direction for rural communities considering PHVMs as part of their overdose prevention strategy. Future research should examine how different funding models, stocking approaches, and implementation strategies influence the uptake, equity, and sustainability of PHVMs in rural communities.

Abbreviations

CBPR

Community-based participatory research

CINC

Community Impact North Carolina

CITI

Collaborative institutional training initiative

NCHRC

North Carolina harm reduction coalition

NC

North Carolina

PHVM

Public health vending machines

SI-CBAR

Structural indicators for community-based participatory action research

Author contributions

Conceptualization (LAE, KLE); Data Curation (LH, CD, LAE); Formal Analysis (LAE, LH, CD); Funding acquisition (LAE, KLE); Investigation (LAE, KLE, LH, CD, MB, SB, VF, WO, AW), Methodology (LAE, KLE, LH, CD, LAE, JC, AB); Project Administration (LAE, KLE); Writing Original Draft (LAE, KLE, CD, LH), Writing review and edit (LAE, KLE, LH, CD, MB, SB, VF, WO, AW, JC, AB).

Funding

This project is supported by the Health Resources and Services Administration (HRSA) of the U.S. Department of Health and Human Services (HHS) as part of an award totaling $300,000 with no percentage financed with non-governmental sources. The contents are those of the author(s) and do not necessarily represent the official views of, nor an endorsement, by HRSA, HHS, or the U.S. Government.

Data availability

Availability of data and materials can be requested from the PI, and the request will be reviewed.

Declarations

Ethics approval and consent to participate

The study was deemed expedited by the East Carolina University Institutional Review Board (UMCIRB 23–001800).

Consent for publication

NA.

Competing interests

The authors declare no competing interests.