Implementing a “Naloxone Hub With Many Spokes” Model in Kentucky: Partner Agency Perspectives on Implementation Strategies

Abstract

Background

Overdose education and naloxone distribution (OEND) is a vitally important evidence-based practice for addressing the ongoing opioid epidemic. During the HEALing Communities Study in Kentucky, OEND was dramatically scaled up in eight counties using a “hub with many spokes” model and multifaceted implementation strategies. This aim of this manuscript is to describe qualitative perspectives of partner organizations regarding the utility of the implementation strategies used to expand OEND.

Method

Twenty small-group and 24 individual qualitative interviews were conducted with staff from 44 agencies that implemented OEND through partnerships with the study team. Interviews were conducted 6–8 months after the study's Communities That HEAL intervention had ended, allowing participants the ability to reflect on their experiences. Inductive coding in NVivo 12 and thematic analysis were used to identify themes regarding agencies’ perspectives regarding the implementation strategies deployed during the study.

Results

In describing implementation strategies that supported their efforts to implement OEND, interview participants largely focused on strategies that provided resources, such as shipments of no-cost naloxone and overdose education tools. Flexibility in how the requisite overdose education was delivered allowed agencies to identify an educational approach that fit their workflow and addressed OEND recipients’ needs. Participants also viewed implementation facilitation and technical assistance provided by study staff as helpful in facilitating the implementation process.

Conclusions

These qualitative data highlight the importance of multifaceted implementation strategies in the process of scaling up EBPs in communities. Future efforts should continue to explore how implementation strategies can be optimized to meet the needs of diverse types of organizations seeking to implement OEND and other evidence-based practices that can mitigate the harms of the opioid epidemic.

Plain Language Summary

Why was this study done?

The opioid epidemic is a serious public health crisis. Teaching people how to recognize and reverse an opioid overdose using naloxone—a medication that reverses potentially fatal opioid effects—can save lives. In Kentucky, as part of the HEALing Communities Study, access to overdose education and naloxone distribution (OEND) was expanded in eight counties using a “hub with many spokes” model. This approach connected a central resource hub to community organizations to ensure wide distribution of overdose education and naloxone. This study describes how partner agencies viewed the strategies used to expand OEND.

What did the researchers do?

The research team interviewed staff from 44 agencies that helped deliver OEND. These interviews took place 6–8 months after the OEND expansion effort ended, giving those staff time to reflect on what worked and what could be improved. Researchers analyzed the interviews to identify common themes and insights about the implementation strategies.

What did the researchers find?

Agency staff appreciated several key strategies that supported their efforts:

Free naloxone and education materials made it easier for agencies to provide overdose education and naloxone to clients, their friends and family, and the community.

Flexibility in delivering OEND allowed agencies to tailor their approaches to fit their specific needs.

Support and guidance from the research team helped agencies navigate challenges and successfully implement OEND.

Why does this matter?

Previous studies have identified barriers to implementing OEND. This study shows that providing resources, offering flexible delivery methods, and giving agencies ongoing support are strategies for expanding OEND.

What should we take away?

Future efforts should focus on optimizing strategies that meet the needs of diverse organizations. Approaches that combine resource delivery, flexibility, and facilitation can help ensure success when implementing OEND.

Introduction

Deaths from drug overdoses continue to be a national crisis in the United States (US), such that about 42% of American adults know someone who has died from a drug overdose (Athey et al., 2024). In Kentucky, the overdose epidemic has largely been driven by illicit opioid use. From 2018 to 2021 in Kentucky, suspected opioid overdose encounters by emergency medical services increased by 44% (Rock et al., 2024), and opioid overdose deaths rose during the COVID-19 pandemic (Blair et al., 2023; Friedman & Akre, 2021; Slavova et al., 2021). Despite a 9.8% decrease in drug overdose deaths in Kentucky in 2023 (Kentucky Injury Prevention and Research Center, 2024), there is an ongoing need to scale up evidence-based practices (EBPs) to reduce opioid overdose deaths.

Overdose education and naloxone distribution (OEND) dramatically reduces the likelihood of death from opioid overdose (Clark et al., 2014; Holmes et al., 2022; Jones et al., 2022; McDonald & Strang, 2016; Razaghizad et al., 2021). When OEND has been scaled up at the community-level, rates of opioid overdose mortality have decreased (Naumann et al., 2019; Walley et al., 2013). Despite the likely benefits of scaling up OEND, access to naloxone remains challenging in many communities (Chatterjee et al., 2022; Kinnard et al., 2021; Meyerson et al., 2021; Nolen et al., 2022; Pollini et al., 2022; Spector et al., 2022).

Intentional implementation efforts can expand OEND in communities, as evidenced by our recent work as part of the HEALing (Helping to End Addiction Long-term^®) Communities Study in Kentucky (HCS-KY), which built partnerships with community coalitions and a wide range of local organizations (HEALing Communities Study Consortium, 2020). To expand OEND, the HCS-KY team developed a set of implementation strategies (Knudsen et al., 2023), drawing on the Expert Recommendations in Implementation Change (ERIC) compendium (Powell et al., 2012, 2015) to support partner organizations across implementation phases (Aarons et al., 2011). A centralized naloxone hub was tasked with naloxone dispensing, shipping, and expiration monitoring. Naloxone was provided at no cost to agency partners, along with a computer tablet(s) for submitting anonymous distribution data. The hub provided recipient-facing educational resources, including a short video and a brochure that addressed the state's educational requirements (Kentucky Revised Statutes 217.186); these resources were translated into multiple languages. In addition, implementation facilitators worked with partner organizations to implement the delivery of OEND. During initial meetings, implementation facilitators provided information, described options for overdose education that would meet state requirements, and collaborated with the agency to design an OEND workflow. At follow-up meetings, implementation facilitators shared data on the agency's distribution efforts, worked with agencies to reconcile their naloxone inventory, and collaborated with agencies to troubleshoot challenges and revise OEND workflows. These implementation strategies resulted in an adoption rate of 69% and considerable reach, with nearly 40,822 units of naloxone (i.e., a unit equals a two-dose package) distributed by partner agencies (Knudsen et al., 2023; Oyler et al., 2024). In the larger multistate analysis of naloxone distribution, HCS intervention communities experienced significantly greater increases in naloxone distribution than control communities (Freeman et al., 2025).

Although these quantitative measures of implementation outcomes provide some indication that the implementation strategies supported OEND scale-up, they provide little insight into how partner organizations perceived the implementation strategies deployed. Qualitative methods are useful for understanding perspectives on the value and utility of implementation strategies (Holtrop et al., 2018). Previous qualitative work has predominantly focused on the perspectives of OEND recipients (Abadie, 2023; Beharie et al., 2023; Bennett et al., 2020; Collins et al., 2024; French et al., 2023; Gray et al., 2023; Kahn et al., 2022; Kano et al., 2020; Kesich et al., 2023; Kline et al., 2020; Ko et al., 2021; Lowenstein et al., 2022; Natale et al., 2023; Schneider et al., 2023; Spadaro et al., 2023; Urmanche & Harocopos, 2023; Westafer et al., 2024), with fewer studies focused on organizational stakeholders involved in delivering OEND, such as pharmacists (Ijioma et al., 2021; Irwin et al., 2024; Rawal et al., 2023), first responders (Baumgart-McFarland et al., 2022; Carroll et al., 2023; Lloyd et al., 2023; Smiley-McDonald et al., 2022), harm reduction staff (Frost et al., 2022), and medical providers (Fraimow-Wong et al., 2024; Punches et al., 2020; Sasson et al., 2023). Prior studies often describe experiences with OEND in the absence of a larger structured effort to use implementation strategies to support scaling up this EBP. Relatively little is known about which implementation strategies are most salient and valuable to organizations working to expand access to OEND. Thus, the aim of this study is to describe organizational perceptions of implementation strategies used as part of HCS-KY's efforts to expand OEND.

Method

Study Context

This qualitative study was embedded within the HCS which tested the Communities That HEAL (CTH) Intervention (HEALing Communities Study Consortium, 2020), in which community coalitions in four states proceeded through a multi-phase process to prioritize OEND strategies and other evidence-based practices for implementation in their communities (Chandler et al., 2023; Sprague Martinez et al., 2020; Winhusen et al., 2020; Young et al., 2022). The CTH intervention was active from January 2020 through June 2022; in Kentucky's eight counties, the first partner organization began implementing OEND in April 2020, and implementation continued through June 2022. From July 2022 to December 2022, the hub provided final shipments to partner organizations, and implementation facilitators continued to provide support to agencies during the transition to sustainment through the state's naloxone program. All research procedures, including the design of the qualitative interviews reported in this manuscript, were approved by Advarra Inc., the HCS's single Institutional Review Board (Pro00038088).

Data Collection

The four state research teams conducted semistructured qualitative interviews with a purposive sample of OEND partner organizations from January 2023 to March 2023, a period that was approximately 6–8 months after the CTH intervention ended. Internal site databases were used to select a purposive sample that included organizations located in rural and urban communities, and organizations in the three sectors prioritized by the larger study (health care, criminal legal system, and behavioral health, which included medication for opioid use disorder (MOUD) programs, counseling-based treatment, and social services). Purposive sampling was used because of resource constraints, as this data collection was added as a later amendment to the study protocol in order to better understand barriers and facilitators to implementation. In addition, a working group of implementation scientists from the four states determined that a saturation-based approach to sampling could not practically be conducted for this large-scale project that involved 34 communities and hundreds of partner agencies (McAlearney et al., 2023).

All interviews were conducted by 10 implementation facilitators on the HCS-KY team, most of whom had worked with partner agencies on OEND implementation and had served as qualitative interviewers for three previous rounds of interviews involving coalition members. Interviewers were trained on the interview guide, and weekly team meetings provided opportunities to clarify study procedures and troubleshoot any challenges. Interviewers contacted 123 potential participants by email or telephone to explain the purpose of these interviews and invite agency representatives to participate in a small group or individual interview. Interviews were conducted with 70 individuals working within 44 OEND partner organizations; of the 53 nonparticipants, 14 individuals refused to participate and 39 did not respond to repeated invitations. Of the 70 participants, 26 agency representatives participated in one-on-one interviews, and 44 participated in 20 small group interviews. The group interviews consisted of 15 sessions with two participants and five sessions with three participants. A script describing the study, the voluntary nature of participation, the need for audio-recording, efforts to protect confidentiality, and compensation was used to obtain verbal informed consent from all participants. After providing verbal informed consent, participants were asked open-ended questions about implementing OEND during video conference or by telephone interviews. All interviews were audio-recorded and professionally transcribed. Participating individuals received a $50 Amazon gift card, unless state, government, or employer policies did not permit compensation.

The working group of implementation scientists developed the interview guide during a series of meetings. The interview questions about implementation of OEND and other EBPs during HCS were informed by the RE-AIM/PRISM model (Feldstein & Glasgow, 2008; Glasgow et al., 2019) which had been adapted as part of the larger study (Knudsen et al., 2020). The interview guide for partner agencies that implemented OEND is presented in Supplemental File 1; the same guide was used for both individual interviews and those conducted with small groups.

Data Analysis

Individual and small group interviews were coded and analyzed using the same approach. Initial coding occurred as part of a cross-site, consensus-based deductive coding process (Hsieh & Shannon, 2005; McAlearney et al., 2023) that involved all four state research teams; this process was similar to other qualitative coding conducted as part of the parent study (McAlearney et al., 2023). Transcripts were coded using a jointly developed cross-site codebook based on the RE-AIM/PRISM model (Feldstein & Glasgow, 2008; Glasgow et al., 2019; Glasgow et al., 1999). Passages about OEND Adoption and OEND Implementation were further coded into subcodes of Internal Factors, External Factors, and HCS Factors (i.e., factors specific to the design of the parent study and CTH intervention), which encompassed barriers and facilitators to adoption and implementation (see Supplemental File 2). To ensure cross-site consistency in interpreting the codebook, pairs of individuals from each of the four sites coded the same transcripts until consensus was reached and the codebook was finalized. Next, coders within each site were trained and used a similar process until consensus was reached. In Kentucky, the coding team consisted of 8 individuals who conducted the interviews. This approach shared some of the hallmarks of coding reliability thematic analysis (Braun & Clarke, 2019) in its emphasis on identifying passages that fit within RE-AIM/PRISM domains, resulting in coded datasets to facilitate additional analysis.

For this manuscript, our approach aligned with codebook thematic analysis (Braun & Clarke, 2022) in that our codebook was intended to facilitate a team-based approach to theme development. Unlike codebook reliability approaches, we conceived of themes as the analytic outputs rather than inputs (Braun & Clarke, 2020). Passages from the Kentucky interviews within the OEND Adoption-HCS Factors and OEND Implementation-HCS Factors codes were extracted for additional inductive subcoding and analysis. First, the lead author (HK) reviewed all passages in these codes and then drafted initial subcode definitions with inclusion/exclusion criteria relevant to implementation strategies. A coding team of four individuals (SAH, SBH, MG, OD) worked independently to apply the subcodes to a common subset of passages. The team met to identify areas of consensus, to discuss disagreements in applying the codes, and to suggest potential new codes; these meetings also identified codes that required additional revision. We viewed this process as a way to work toward shared understanding and meaning within our team, which included an implementation scientist, three staff members who had worked with OEND-implementing agencies and thus had experiential knowledge, and a medical student with prior qualitative research experience. Supplemental File 2 presents the codebook for this analysis of HCS-KY implementation strategies.

The small group then worked to develop themes (Braun & Clarke, 2006, 2022, 2023; Kiger & Varpio, 2020). Each member reviewed all code reports independently to generate initial themes. During meetings, the initial themes were discussed to identify where shared meaning had arisen in the independently generated themes across coders. When a team member proposed a unique theme, the group discussed whether that theme made a substantive contribution to our shared understanding of agency perspectives on implementation strategies and whether the theme was distinct enough to stand on its own. The group discussed and refined the themes until consensus was reached regarding the final themes to include in this manuscript. Members proposed representative passages for each theme; a consensus-based process was used to select passages for inclusion in this manuscript. Consensus regarding themes and passages was reached after three team meetings. Supplemental File 3 contains our consolidated criteria for reporting qualitative research (COREQ) checklist (Tong et al., 2007).

Results

Characteristics of interviewees are presented in Table 1. In describing the factors that supported their efforts to implement OEND, partner organizations largely focused on implementation strategies that provided resources in terms of no-cost naloxone, overdose education tools, and support provided by HCS-KY staff, such as implementation facilitators and staff of the hub.

Table 1.

Characteristics of Interviewees (n = 70).

	% (n)
Sex
Male	15.7% (11)
Female	84.3% (59)
Ethnicity
Hispanic/Latino/a/e	7.1% (5)
Non-Hispanic	92.9% (65)
Race
Asian	1.4% (1)
Black	7.1% (5)
White	90.0% (63)
Missing	1.4% (1)
Age
18–34 years of age	24.3% (17)
35–49 years of age	42.9% (30)
50–64 years of age	28.6% (20)
65–74 years of age	4.3% (3)
Educational attainment
High school degree or equivalent	1.4% (1)
Some college, no degree	8.6% (6)
Associate degree	8.6% (6)
Bachelor's degree	28.6% (20)
Master's degree	48.6% (34)
Doctoral or professional (e.g., MD, DDS) degree	4.3% (3)
Agency type
Medication for opioid use disorder (MOUD) program	52.9% (37)
Non-MOUD behavioral health program	20.0% (14)
Health department	11.4% (8)
Primary care	5.7% (4)
Emergency medical services (EMS)	5.7% (4)
Criminal legal system organization (e.g., detention center, drug court)	4.3% (3)
Agency location
Urban	61.4% (43)
Rural	38.6% (27)

Note. Due to rounding, percentages may not sum to 100%. Some programs providing MOUD also provide non-MOUD services to treat other types of substance use disorder.

Naloxone Accessibility Was a Common Barrier Prior to the Study. Many participating agencies readily acknowledged previous challenges in affording to implement OEND or of keeping naloxone on hand in quantities that enabled growth of OEND efforts. Therefore, having readily available naloxone paid for and delivered by the HCS-KY team easily fit into the short-term and long-term goals of participating organizations. One study participant highlighted the successful expansion of OEND within the population served, attributing it to their consistent access to a supply (i.e., Narcan) through HCS-KY:

“We were now going to be able to … provide the Narcan to the people that we were going out to see or the community. And it wasn't just one place, it was multiple places that, ‘Oh, you're going to come to my door and you're going to give me Narcan?’ ‘Yes, and we'll come back, and we'll give you more if that's what you need.’ And it helped for us to also have those conversations about that de-stigmatizing of the Narcan and how beneficial it could be for individuals. So, it's definitely been an earmark for our program and being able to provide that to our community members that we go see.” (01072642, urban emergency medical services (EMS) organization)

In reviewing participant interviews, it was clear that more naloxone on hand meant more personal benefit and safety for the clients of participating agencies. Some study participants claimed that the naloxone supplied by HCS-KY was the greatest factor in successful implementation of the OEND model within their agency:

“I think the biggest factors that played probably the largest role is just the access of having it [naloxone] here was incredible. I think that being able to know that we're sending a patient home with it, and it wasn't going to get lost, oh, it went to your pharmacy, but if they weren't picking up any other prescriptions, it's maybe less likely that they'll take the time to go and maybe wait in line for 20, 20 min to get it picked up. And so that was one huge benefit was just having it hands-on that we know the patient was leaving with it that day and we didn't have to play any guessing games on whether they were able to pick it up.” (01074356, urban primary care organization)

Increased direct access to naloxone in conjunction with opioid education provided an opportunity for clientele to circumvent everyday logistical barriers to obtaining this life-saving medication, as one participant pointed out:

“And I do think that the ease of access with the education was a great tie in, because they do want to have the access to it. But again, sometimes it's just that extra step and having to go to a pharmacy, getting a prescription, going through that. And then sometimes they don't get the education piece. So, I think both of those things together were great.” (01072807, urban MOUD organization)

The Ease and Adaptability of OEND Tools, Resources, and Workflows Supported Contextually Tailored Implementation. Given the inherent differences in workflows between partner agencies, the OEND implementation process needed to be streamlined and adaptable. Such qualities allowed for faster adoption of the model and encouraged participants in scale-up efforts. Many agencies discussed the OEND process as easily adopted within the agency specifically due to HCS-KY's provision of tools and resources needed for successful implementation:

“I think that basically it really gave us the materials to take it to the next level with little to no front-end prep on our part, honestly. I really felt like everything was provided to us and what we needed and explained in a way that was very understanding, and we were able to take it the moment we had the resources and initiate this within our clinic. We didn't have any barriers to offering it that I'm aware of within the clinic. And so that was just something I think that the accessibility and the education that was provided to us beforehand really streamlined it and made it a smooth process delivering it to patients.” (01074356, urban primary care organization)

Adaptability was supported through having multiple overdose education options which allowed agencies to select a method that fit within their workflow. The HCS-KY team offered agencies the option to deliver overdose education through a 9-min video on tablets. Available in multiple languages, the video explained how to prevent an overdose, recognize an overdose, and administer naloxone. The use of a portable tablet coupled with the education offered in multiple languages provided a ready-made educational resource for overdose education. This flexible format allowed agencies to incorporate overdose education into busy schedules: individuals could watch it solo on tablets, online, or in groups.

“And so, I go on, ‘You can have them come in, they can watch the video. Here is a link that they can watch the video. Come in, tell us they've watched the video, complete the demographic sheet and we can send them out with Narcan.’ Of course, we ask them a couple questions, make sure that they have watched the video if they don't do it here, just to make sure. And so, it's been a really, really easy implementation and it's been a really positive experience.” (01062324, urban behavioral health organization)

While the video format was well-suited for some organizations, it proved less effective in fast-paced, busy, or remote settings. Some of the agencies implementing OEND programs encountered challenges in using video-based training. In these cases, the video failed to engage participants adequately, emphasizing the need for more interactive approaches. Live, in-person sessions were a more effective means of delivering overdose education and ensuring participant understanding, as noted by this participant:

“But when I originally started, I did attempt to use our tablet to play the video of how to administer Narcan. I felt like the video was too long. Initially, I was kind of losing people's interest when I would try to put that video on. And so, I started going to just strictly live sessions where I would show them how to administer it.” (01032216, rural EMS organization)

The research team also developed a brochure entitled, “The Opioid Overdose Recognition and Response Guide,” a step-by-step resource to help agency staff educate patients on the essential components of overdose response. With the support of these brochures, agency staff offered brief 2–5 min instruction, providing patients with the confidence and knowledge needed to carry and use naloxone effectively. One individual involved in an OEND program highlighted this guide as a key strategy in their successful overdose education efforts:

“Yeah, no, the brochure was fast and honestly a lot of them have seen it used, and it was just basically going over, ‘Okay, this is the real way you do it,’ and yeah.” (01092412, rural MOUD organization)

Along with this pamphlet, a County Resource Guide developed by the research team could be used to help patients locate additional naloxone distribution locations, OUD treatment, and local social services. This County Resource Guide was a key strategy in addressing the opioid overdose epidemic by connecting individuals with vital resources. By offering a comprehensive list of available services, the County Resource Guide made it easier for those struggling with substance use disorders to access the support they needed, ultimately improving their well-being.

“Having access to this allowed us to do far more than just provide people naloxone. We got to dismantle beliefs about substance use and misuse. We got to give people access to treatment resources, treatment information. Good God, how many questions we answered with residential treatment, mental healthcare, physical healthcare, just all the things. And [Quick Response Team lead]'s right, that resource guide is a gem, and we still use it here.” (01022769, rural behavioral health organization)

The County Resource Guide listed many agencies affiliated with HCS-KY, providing locations where individuals could access no-cost naloxone and other needed services.

To further support overdose education and naloxone distribution, HCS-KY supplied agencies with marketing materials that encouraged carrying naloxone and raised awareness that OEND was available at the agency. These materials not only helped reduce the stigma around substance use disorders but also gave agencies an effective tool to promote this vital service within their communities.

“We were able to utilize some of HEAL's marketing and, I guess, promotional materials that we were able to co-brand. Then, we were able to add a QR code on some of these materials that would drive people to the website that they would go to in order to scan to get to the mailed naloxone.” (01131207, urban health department)

HCS-KY Staff Supported Implementation. HCS-KY staff worked with the partnering agencies to raise awareness and discuss the availability of naloxone, along with providing technical assistance regarding OEND throughout the study period. Each partner was assigned an implementation facilitator who worked with the agency throughout the intervention period. In addition, HCS-KY staff at the hub worked to dispense and ship naloxone based on the needs of partner agencies. If needed, implementation facilitators or hub staff provided technical assistance via video conference and even traveled to partner agencies to provide technical assistance or deliver naloxone. HCS-KY also funded peer recovery coaches and care navigators who were deployed into some partner agencies, who could also provide OEND to clients. These staff deployed within agencies helped to promote awareness and availability of naloxone as one participant pointed out:

“And it's like the [recovery coach] promoting the availability of Narcan were greater than those who may say something negative against it.” (01061185, urban health department)

Due to the HCS-KY staff supporting the agency with implementing OEND coupled with other implementation resources, partnering agencies had the resources to make the OEND process easily adaptable to the agency's needs as one participant described:

“It was easy. You provided us the tablet, you provided us the video, you provided us instructions for when we forgot, you even came out one time and updated our tablet for us. It was easy. It was just easy to do. There was no massive hoops, too. It was so easy.” (01064357, urban primary care organization)

In addition to the initial implementation support provided to the partnering agency, HCS-KY staff remained in constant communication with the partnering agency about the OEND efforts, which included emails, phone calls, and video conferencing meetings via Zoom as one participant mentioned:

“Interviewer: So, what do you think were the biggest factors that helped your organization implement this model of OEND during the HEALing Communities Study?

Interviewee: Well, I think probably the communication with you guys. I know our staff was meeting with you guys quite often, and I think just the communication back and forth and then of course actually delivering it to us and getting it in our hands. I think all of that really helped get it going.” (01022768, urban MOUD organization)

Lastly, one participant shared the positive impact of the technical assistance support HCS-KY staff provided when needed to ensure a smooth implementation of the OEND process:

“Our [HCS-KY] team was absolutely phenomenal in everything. There was no complaints that I'm aware of in that regard. They were really attentive to everything we needed when we reached out. When we had technical difficulties, we were able to get that implemented because that was a thing, too. Working our systems on our tablets, knowing how to be educated on that, the training for that and making sure that all of our staff was trained in that and how to utilize that system, giving us that time anytime we needed it, sending out those dates. ‘What can we do? How can we do it to make things more beneficial for you all? Or how can we get this implemented quickly?’ And I don't think we had any issues whatsoever. If they were any, it was just on our part, us just learning. But for the most part, everything I think went really smoothly with that implementation process.” (01072642, urban EMS organization)

Required Data Reporting Was Simple for Some Agencies but Challenging for Others. HCS-KY required their partners implementing OEND to collect and report data back to the study, which helped to measure reach for the study's outcomes but also was used as an “audit and feedback” strategy in which implementation facilitators would share summaries of the agency's naloxone distribution during follow-up meetings. Many participants described the data reporting process for OEND implementation as simple and straightforward. However, some participants did share their encounters with the data reporting process as being challenging:

“Interviewer: So, you would say you probably haven't had many difficulties in the process at all?

Interviewee: Oh, absolutely not. The only thing we had a little bit of difficulty with is the REDCap system, and that was putting things in over and over and over again.” (01062324, urban behavioral health organization)

The minor challenges shared by participants included data reporting requirements, such as scanning each QR code before distributing naloxone, which were difficult or time-consuming, connectivity issues with the tablet provided by HCS-KY, and navigating the REDCap data reporting system, especially when implementing OEND in a group setting.

Discussion

As part of efforts to scale up OEND in communities, a bundle of implementation strategies was deployed, and this study focused on the perspectives of organizations that partnered with HCS-KY regarding these strategies. Overall, participating organizations viewed the strategies as acceptable and feasible, contributing to their ability to deliver OEND.

One dominant theme in these interviews was that the provision of no-cost naloxone by HCS-KY was critically important to OEND implementation. The purchase of naloxone was supported by grant funds as well as a financial contribution from the Kentucky Opioid Response Effort (KORE), which was established through the Substance Abuse and Mental Health Services Administration's State Targeted Response and State Opioid Response grants and is housed within the state's Department for Behavioral Health, Developmental and Intellectual Disabilities. To some degree, this finding about the importance of no-cost naloxone is not surprising given that prior research has pointed to funding challenges coupled with the cost of naloxone and, at times, supply chain issues as barriers to implementation in community distribution programs (Frost et al., 2022; Hincapie et al., 2021; Winstanley et al., 2016). Qualitative work by Rawal et al. (2023) found that the cost was a barrier to pharmacy-based naloxone dispensing by standing order, and similarly, qualitative interviews focused on first responder-based OEND programs also noted that funding was a key facilitator of implementation (Carroll et al., 2023). Nearly 88% of all partner organizations were interested in sustaining OEND by joining KORE's program that also offers no-cost naloxone through its federal funding (Knudsen et al., 2023), pointing to the longer-term viability of this approach.

The unit cost of naloxone has decreased substantially over time, including when it was only available by prescription and more recently with its transition to an over-the-counter product (Marley et al., 2024). Barriers remain for accessing naloxone, including its cost and availability in community pharmacies. In a national study of people prescribed opioids for chronic pain, a group at risk of overdose, only 54% reported being willing to pay for naloxone (Huang et al., 2024). Although many insurance plans provide some coverage for naloxone, recent work has shown that cost-sharing in commercial insurance and Medicare reduces the odds that individuals will fill a naloxone prescription (Chua et al., 2024). Pharmacy stocking is also an ongoing issue. In a recent study of North Carolina pharmacies, fewer than 60% had naloxone in-stock and available for dispensing without prescription, and the mean out-of-pocket price was nearly $63, which exceeded the manufacturer's suggested retail price of $45 (Marley et al., 2024).

Developing educational materials is an implementation strategy included in the Expert Recommendations in Implementation Change (ERIC) compendium (Powell et al., 2015), and this strategy served at least two purposes. Because naloxone was available only by prescription at the time, provision of overdose education was necessary to comply with state regulations which mandated that overdose education needed to cover seven content areas. However, overdose education was important for reasons beyond regulatory compliance, as other studies have noted the importance of educational resources for training recipients of naloxone (Drainoni et al., 2016; Fraimow-Wong et al., 2024; Punches et al., 2020; Sasson et al., 2023) as well as those who deliver OEND (Ijioma et al., 2021; Rawal et al., 2023).

Our approach to resources for overdose education evolved in response to early feedback from partner organizations that reliance on a pre-existing 20-min interactive web-based training developed by another entity was not feasible for many agencies. Concerns about the time required for OEND had been documented in prior qualitative work (Ijioma et al., 2021; Rawal et al., 2023; Sasson et al., 2023). This early feedback from partner organizations points to the importance of organizational compatibility as a facilitator of innovation adoption and complexity of innovations as a barrier (Damschroder et al., 2022; Rogers, 2003). In response, our team developed a nine-minute video, available in multiple languages, that was installed on tablets and available on YouTube (https://www.youtube.com/watch?v=lZMleZybx_Q). Some partner organizations valued this video-based option, but others still perceived that its length was a barrier to use, particularly in fast-paced environments or in settings where many individuals had previously been trained in OEND (e.g., syringe service program participants). Adding review of an OEND pamphlet helped to address this concern to some extent, but nonetheless, attention to how overdose education needs may vary between settings and allowing for flexibility in how overdose education is delivered is an important consideration when planning for implementation of this EBP.

In addition to overdose education resources, partner organizations were provided communication materials upon request to raise awareness about the availability of OEND. This strategy aligns with the ‘increase demand’ implementation strategy in the ERIC compendium (Powell et al., 2015). These communication materials included customized information about how to request OEND while typically emphasizing the brief message of ‘Save a life. Get naloxone.’ Other qualitative research has reported the need for raising awareness about OEND availability (Fraimow-Wong et al., 2024; Gray et al., 2023; Natale et al., 2023) and the value of this type of message, which frames naloxone around a shared sense of responsibility to reduce opioid fatalities (Bennett et al., 2020). Not all partner organizations requested these materials, but among those that did, there was a general sense that the communication materials were helpful.

HCS-KY staff, especially the implementation facilitators who deployed the strategy of external facilitation (Powell et al., 2015), were also viewed favorably by partner organizations, who perceived that HCS-KY staff worked collaboratively with them to overcome implementation barriers. In the context of OEND, a recent randomized trial deployed a bundle of implementation strategies in syringe service programs that relied heavily on facilitation, finding that the facilitation-based approach significantly increased naloxone reach relative to a no-facilitation condition (Lambdin et al., 2024). As has been noted by others in the field of implementation science, facilitation is vitally important for organizational change initiatives (Kitson & Harvey, 2015; Moussa et al., 2019) and has been identified as an evidence-based implementation strategy (Kirchner et al., 2014; Ritchie et al., 2024; Wang et al., 2018).

It is important to note several limitations of this study. First, these interviews were only conducted with staff from partner organizations that successfully implemented OEND in a single state. It is unknown whether the HCS-KY implementation strategies would have overcome the barriers that may have impeded the subset of agencies who declined to partner or whether perspectives on the implementation strategies would be similar in other states. However, the varying types of organizations involved may suggest that the findings would have applicability for others seeking to expand OEND. Second, agencies that did not participate in the interviews may have had negative perceptions regarding their partnership with HCS-KY or the strategies used during the study. Finally, most interviewers had provided facilitation to agency staff during the implementation phase of the study, and that prior relationship may have resulted in some staff being less candid about their experiences.

Conclusions

These qualitative data highlight the importance of multifaceted implementation strategies in the process of scaling up EBPs in communities. Partner organizations noted readily accessible naloxone, supportive overdose education and other accompanying resources, and continued implementation support as noteworthy practices during this study. Overall, the strategies developed by HCS-KY were generally viewed positively by partner organizations, and OEND was greatly expanded during the study, with growth in the rate of naloxone distribution in Kentucky's intervention communities being three times greater than waitlist communities (Freeman et al., 2025). Future efforts should continue to explore how implementation strategies can be optimized to meet the needs of diverse types of organizations seeking to implement OEND.

Supplemental Material

sj-docx-1-irp-10.1177_26334895261438500 - Supplemental material for Implementing a “Naloxone Hub With Many Spokes” Model in Kentucky: Partner Agency Perspectives on Implementation Strategies

Supplemental material, sj-docx-1-irp-10.1177_26334895261438500 for Implementing a “Naloxone Hub With Many Spokes” Model in Kentucky: Partner Agency Perspectives on Implementation Strategies by Hannah K. Knudsen, Michael Goetz, Shaquita Andrews-Higgins, Sandra Back-Haddix, Olivia A. Davis, Patricia R. Freeman, Douglas R. Oyler and Sharon L. Walsh in Implementation Research and Practice

Supplemental Material

sj-docx-2-irp-10.1177_26334895261438500 - Supplemental material for Implementing a “Naloxone Hub With Many Spokes” Model in Kentucky: Partner Agency Perspectives on Implementation Strategies

Supplemental material, sj-docx-2-irp-10.1177_26334895261438500 for Implementing a “Naloxone Hub With Many Spokes” Model in Kentucky: Partner Agency Perspectives on Implementation Strategies by Hannah K. Knudsen, Michael Goetz, Shaquita Andrews-Higgins, Sandra Back-Haddix, Olivia A. Davis, Patricia R. Freeman, Douglas R. Oyler and Sharon L. Walsh in Implementation Research and Practice

Supplemental Material

sj-docx-3-irp-10.1177_26334895261438500 - Supplemental material for Implementing a “Naloxone Hub With Many Spokes” Model in Kentucky: Partner Agency Perspectives on Implementation Strategies

Supplemental material, sj-docx-3-irp-10.1177_26334895261438500 for Implementing a “Naloxone Hub With Many Spokes” Model in Kentucky: Partner Agency Perspectives on Implementation Strategies by Hannah K. Knudsen, Michael Goetz, Shaquita Andrews-Higgins, Sandra Back-Haddix, Olivia A. Davis, Patricia R. Freeman, Douglas R. Oyler and Sharon L. Walsh in Implementation Research and Practice