Statewide implementation of MOUD in jails: aligning barriers and strategies in the quest for acceptability and feasibility

Abstract

In 2019, the Maryland legislature passed a bill mandating that county jails provide medications for opioid use disorder (MOUD) to incarcerated persons. While vast scholarship points out common barriers to MOUD implementation in carceral settings, and some also depict facilitators of implementation, largely missing from these evaluative studies is a clear understanding of the process of identifying barriers and identifying ways to overcome these barriers (i.e., strategies). To address this scholarly gap, we conducted in-depth interviews with jail and community health leaders in Maryland counties in 2022 and follow-up interviews one year later, after the January 2023 compliance deadline. These leaders were participating in a Learning Collaborative as part of a larger comparative effectiveness trial comparing extended-release buprenorphine versus extended-release naltrexone. In interviews, participants focused their discussion on nine main implementation barriers and associated strategies, including: 1) staffing shortages, 2) financial barriers, 3) staff buy-in, 4) concerns of medication diversion, 5) space, 6) client buy-in, 7) intra- and extra- organizational relationships, 8) peer support, and 9) leadership. Our analysis suggests that while some barriers to MOUD implementation were mitigated through targeted strategies, others remained persistent or worsened. Notably, it appears that some important and salient challenges were ignored (or at least, not acted on) and sometimes leaders came up with strategies that were targeting non-existent (or not important and salient) challenges. Specifically, we find that while barriers including staff buy-in, client buy-in, space concerns, and fears of medication diversion lessened as strategies including staff education, client education, space accommodations, and diversion protocols increased, others, including staffing shortages, financial constraints, strained intra- and extra-organizational relationships, and leadership challenges, were particularly persistent and even intensified over time. Further, comparing barrier and strategy frequencies revealed critical mismatches, where the most frequently cited barriers (e.g., staffing and medication diversion) were met with disproportionately fewer strategies while lesser cited barriers (e.g., client buy-in) were frequently innovated on. This work identifies insights that may inform carceral-based MOUD efforts more broadly, especially as more states consider legislation mandating carceral-based MOUD.

Introduction

When an organization wants to solve a problem, they often turn one of two directions: 1) they create a new policy, practice, program, or reform or 2) they find an existing policy, practice, program, or reform (or accept one forced upon them) to adopt/adapt and implement. Once the reform is created, chosen, or accepted, organizational members generally craft an implementation plan and begin the process of training, coaching, and implementing the reform. However, as decades of implementation science suggest, change is difficult and organizations often face both anticipated and unanticipated barriers to implementation efforts (McNulty & Ferlie, 2004; Rosenberg & Mosca, 2011). Sometimes these barriers are known, obvious, and/or manifest and other times, they are unknown, hidden, and/or latent (Hoover & Harder, 2015; Nafei, 2016; Piderit, 2000). Addressing and consistently monitoring these barriers with innovative responses is crucial for the success or failure of implementation efforts and their associated outcomes (Barach, 2016).

Historically, evaluation research of reform implementation within criminal legal organizations, such as jails, focuses most heavily on the success or failure of the outcomes of an implementation as the primary, or often sole, measure. In doing so, the nuance of the contextualized processes at work during implementation are often overlooked or disregarded. Implementation Science (IS)—a framework that originated to more rigorously and deeply study reform within the health services field—offers some promise for criminal legal organizations seeking to understand implementation processes in addition to outcomes (Taxman, 2025).

One criminal legal area ripe with implementation efforts and studies are carceral settings such as jails and prisons. As jails and prisons struggle with substance use disorders, illicit substance use, and overdose (Dalbir et al., 2024), many are implementing medications for opioid use disorder (MOUD) such as buprenorphine, methadone, and naltrexone. A growing and robust body of evidence supports the effectiveness of MOUD in jails and prisons (Cates & Brown, 2023; Gordon et al., 2014; Moore et al., 2019; Springer, 2024), but many jails and prisons still prefer drug-free interventions (Friedmann et al., 2012; Zaller et al., 2013). Moreover, recent national surveys indicate less than half (43.8%) of jails offer some form of MOUD to some individuals, and only 12.8% offer these medications to anyone with an OUD (Balawajder et al., 2024). In fact, research and evaluation of MOUD implementation within prisons and jails often shows various challenges stemming from issues related to acceptability (Is it tolerable?) and feasibility (Is it suitable? Is it do-able?) of MOUD in carceral institutions. While many different barriers are cited, common among these are: limited resources (Krawczyk, et al., 2022; Staton et al., 2024); space limitations (McNeeley et al., 2024); stigma and buy-in (Brinkley-Rubenstein et al., 2019; Staton et al., 2024); diversion (Bandara et al., 2021); and opposition from leadership (Pivovarova et al., 2022). Yet, while the current literature identifies common barriers to MOUD implementation in carceral settings, and some also depict facilitators of implementation such as training, resources, and staff support (Ferguson et al., 2019), largely missing from these evaluative studies is a clear understanding of the process of identifying barriers and identifying ways to overcome these barriers (i.e., strategies) to improve both acceptability and feasibility of MOUD within jails or prisons.

To address this scholarly gap, this paper considers acceptability and feasibility processes during MOUD implementation in 10 county jails and 10 corresponding community treatment settings in Maryland, a state which recently mandated MOUD treatment in county jails. Specifically, this paper considers both staff identified and understood barriers and if/when strategies were employed to overcome these barriers. Additionally, the paper considers if/how specific identified barriers were matched with specific identified strategies to improve implementation processes. To do this, we interviewed Maryland jail and community health leaders as part of a larger comparative effectiveness trial comparing extended-release buprenorphine versus extended-release naltrexone. Leaders were participating in a Learning Collaborative, where the aim of the sessions was to understand factors related to the acceptability of MOUD in jail settings and the feasibility of providing medication continuity of care from jail to community treatment providers. Understanding the perspectives of leadership during implementation processes is crucial, as these individuals play a key role in setting priorities, shaping organizational culture, allocating resources, and informing front-line staff’s discretionary decisions regarding how policies are enacted at the ground level (Lipsky, 1980).

Literature review

Implementation science

Implementation Science (IS) is the “scientific study of methods to promote the systematic uptake of research findings and other evidence-based practice into routine practice and hence, to improve the quality and the effectiveness of health services” (Eccles & Mittman, 2006, 1). As such, IS is the movement of a reform from inception, to adoption, to use, to routine use, and finally to sustained use. Yet, while easily defined, implementing reform is often fraught with difficulties. An IS-based strategy for understanding (and evaluating) change efforts affords the opportunity to peer into the contextual abyss to see what is getting in the way of the reform (barriers) and what is helping it stick (facilitators).

While numerous and complex, IS frameworks consider five primary origins of potential barriers and facilitators including: 1) the characteristics of the reform, 2) the change processes, 3) the inner setting factors (such as the organization/agency itself), 4) the outer setting factors (such as stakeholders), and 5) the individuals involved in the reform/change process (Taxman & Belenko, 2011). Within these groupings the behaviors, perspectives, views, and actions of the organizational actors are critically important. For an implementation to succeed at any level organizational actors must see, or begin to see, the reform as acceptable or suitable and feasible or realistic within the organization. Without acceptability and feasibility, most reforms will slow or fail.

MOUD Implementation in carceral settings

Jails face unique barriers to MOUD implementation, and numerous studies depict implementation problems that derailed, halted, or prevented MOUD implementation in these spaces. For example, in a recent study of seven rural Colorado jails, McNeely and colleagues (2024) interviewed jail staff regarding MOUD implementation. While jail nurses and collaborative partnerships facilitated MOUD-implementation in some jails, other jails struggled with numerous implementation barriers. These included physical space limitations, staffing and training challenges, inadequate funding, and perceptions of diversion in addition to other contextual issues such as insufficient post-release service connections. In another example, Krawczyk and colleagues (2022) interviewed staff in 11 New Jersey county jails and found two of several common barriers to MOUD implementation included limited resources and short jail stays that staff perceived as preventing appropriate MOUD treatment. However, they also found several MOUD facilitators including starting MOUD quickly upon jail intake and building community partnerships to facilitate post-release linkage to services. While critically important for understanding MOUD implementation, the distinct lack of connection between perceived barriers and facilitators creates a knowledge gap that prevents IS from fully explaining reform challenges and hinders organizations from learning how to specifically address barriers with innovative facilitators to improve change success.

Gap in the literature and study significance

While research on MOUD implementation in jails has offered insight into barriers and facilitators of program success, missing from these studies is an understanding of whether strategies are deliberately and strategically matched to barriers they aim to resolve. Understanding this is both theoretically and practically important, and this study aims to address this gap. Additionally, compared to prisons, jails face unique challenges to MOUD implementation (Krawczyk et al., 2022; McNeely et al., 2024). This study aims to expand the existing literature by not only examining barriers and strategies of implementation success, but also whether these barriers and strategies strategically align.

Method

Background & study sites

The data for this study comes from interviews with jail and community health leaders that were conducted during a larger comparative effectiveness trial comparing extended-release buprenorphine versus extended-release naltrexone (Gordon et al., 2021) in six county jails in Maryland. As a part of that study, a learning collaborative (LC) was conducted in parallel with the randomized controlled trial partnering with some of the same county jails, and interviewees participated in that LC.

In 2019, the Maryland legislature passed a bill (HB0116) mandating that all 24 county jails provide MOUD to incarcerated persons. Following this legislation, Maryland joined a small cohort of states that have mandated jail-based MOUD (Culli, 2025). Following the legislation, jail staff must screen for, evaluate, and treat opioid use disorder among people in custody. Facilities must provide three forms of MOUD: methadone, buprenorphine, and extended-release naltrexone. Jails were mandated to come under compliance with the law on a rolling basis with all jails being in compliance by January of 2023.1 After the passage of the legislation, jail leaders and staff and community health workers became front-line MOUD providers, learning about MOUD in real time and developing plans to implement the mandate.

Maryland’s MOUD mandate has not been systematically evaluated. While studies might be in progress, only one has been published. Recently, Kramer and colleagues (2025) interviewed 24 recently incarcerated individuals and reported that, from the perspective of incarcerated people, treatment access has improved post-legislation. Though, as the authors underscore, gaps in care still exist, such as timely treatment access and stigma from staff. As such, there is a pressing need to understand Maryland’s MOUD implementation process and identify insights that may inform carceral-based MOUD efforts more broadly.

Data and sample

The jail and community health leaders that we interviewed were participating in the larger project’s LC. As participants in the LCs, each interviewee voluntarily agreed to participate in an interview. Jail leaders were typically either wardens or medical directors, and community health leaders were typically either program directors or managers. We interviewed leaders both pre- (2022) and post- (2023) LC sessions. Leaders worked within Maryland jails (n = 7 pre-LC and n = 9 post-LC) and Maryland community treatment centers (n = 10 pre-LC and n = 9 post-LC).

Learning collaborative

The aim of the LC sessions was to understand factors related to the acceptability of MOUD in jail settings and the feasibility of providing medication continuity of care from jail to community treatment providers. From May 2022 to December 2023, there were four LC sessions. The first was in-person and the rest were held virtually. The in-person kick-off session in May 2022 included four breakout meetings: 1) jail health staff and medication considerations; 2) Maryland House bill 116 jail data reporting requirements; 3) continuity of care for individuals leaving incarceration on MOUDs; and 4) Ask the doc: A conversation with an addiction medicine specialist. The second virtual session in April 2023 included three breakout meetings: 1) buprenorphine telemedicine models; 2) regulatory and licensing of methadone OTPs; and 3) a patient’s journey through addiction and incarceration. The third virtual session in August 2023 included three breakout meetings: 1) treating co-occurring opioid use disorders and other mental illnesses; 2) federal regulations for controlled substances and medication assisted treatment programs; and 3) staff stigma towards people with addictions. The final virtual session in December 2023 included three breakout meetings: 1) expanding peer recovery coaches as staff; 2) how to increase acceptance of MOUDs; and 3) reducing MOUD diversion.

Interviews

As participants in the LCs, each interviewee voluntarily agreed to participate in in-person, telephone, or web-based interviews both pre- and post-LC. Before each interview, the researcher read an informed consent script to participants, and all interviewees provided consent to participate in the interview. The baseline interview guide asked participants about their organizational roles and experiences implementing MOUD, including attitudes toward different medications, screening and treatment processes, barriers and facilitators to compliance with Maryland’s legislative mandate, and coordination with community providers. The follow-up interview guide focused on changes over time, specifically examining whether barriers had shifted, what facilitators had emerged, and what strategies organizations were using to improve implementation and strengthen partnerships. Each interview took about an hour. One interviewer conducted all the interviews, and interviews were audio recorded and later transcribed into Microsoft Word documents. All study protocols were reviewed and approved by WCG IRB.2

Coding and analysis

After data collection concluded, the lead author linked all transcripts to the qualitative data management program Atlas.ti. Before beginning to code, the lead author developed a codebook informed by the interview questions. These codes were deductive and focused on five topic areas: 1) thoughts and attitudes regarding MOUD; 2) facility substance use disorder service availability; 3) implementation acceptability and feasibility (including barriers and strategies); 4) MOUD medication comparisons (methadone, buprenorphine, extended-release naltrexone); and 5) system linkage in the community. As she coded the interviews, she then developed and applied inductive codes (e.g., Implementation-Barrier-Cost) as needed to capture emergent themes. This hybrid deductive-inductive coding approach allows for the researcher to examine a priori themes derived from the literature and interview guide while remaining open to new themes that emerged from the data (Corbin & Strauss, 1990).

After coding, the lead author met with the other authors to discuss emergent themes and preliminary findings. The lead author then moved into the analysis stage, where she focused on themes of acceptability and feasibility as this was the most frequently discussed topic by interviewees and most directly aligned with the study’s central research question regarding MOUD implementation. She used a phronetic iterative approach (Tracy, 2019), moving iteratively between the data, emerged codes, and implementation science literature. Specifically, her analysis was informed by implementation science literature emphasizing acceptability and feasibility as key determinants of implementation success, as well as scholarship identifying barriers and facilitators arising from the reform itself, implementation processes, organizational contexts, external environments, and the individuals involved in implementation efforts (Taxman & Belenko, 2011). These concepts offered a framework to explore how participants understood and navigated implementation challenges. Using analytic memos and constant comparison, the lead author refined categories and interpreted how barriers and strategies reflected broader implementation processes. This involved synthesizing and organizing codes as needed. For example, first-cycle codes such as “Implementation-Barrier-Cost, “Implementation-Reforms-More Funding” and “Mandate-State Provided No Funds” were initially grouped into the broader category of “Financial Barriers”, which captured how financial limitations shaped MOUD feasibility across sites. During subsequent rounds of analysis, these data were further interpreted as reflecting broader concerns about the feasibility of implementation and perceived disconnects between policy mandates and local implementation realities. Similar comparisons were conducted across categories to identify how participants understood barriers, the strategies used to address them, and the organizational conditions that shaped implementation outcomes. This coding cycle resulted in nine notable barriers and related strategies, including 1) staffing shortages, 2) financial barriers, 3) staff buy-in, 4) concerns of medication diversion, 5) space, 6) client buy-in, 7) intra- and extra- organizational relationships, 8) peer support, and 9) leadership.

After completing analysis, the lead author then explored two additional questions. First, to examine whether (and how) perspectives on barriers and strategies shifted during the implementation process, the lead author compared mentions of barriers and strategies during baseline interviews compared to follow-up interviews. Frequencies reflected the number of interviews in which a barrier or strategy appeared, not the total number of times it was mentioned. If a participant referenced the same barrier or strategy multiple times within a single interview, it was counted only once. This approach aligns with prior qualitative research incorporating code frequencies (e.g., Hochwald et al., 2023; Sandelowski, 2001; Seale & Kelly, 2004). Second, to examine if there is any mismatch between the frequency at which a barrier was mentioned compared to the frequency at which a related strategy was mentioned, barriers and innovations were grouped into low (0–19%), moderate (20–49%), and high (50% +) frequency categories, and each barrier-strategy combination was placed into a matrix accordingly. This allowed for the identification of mismatches, cases where a barrier was frequently mentioned but the corresponding strategy was not (or vice versa). This process was used for both baseline and follow-up interviews to make comparisons over time.

Findings

Jail and community health leaders identified a range of factors that shaped the implementation of jail-based MOUD programs. To address the central research question—What are leader’s perceived barriers to MOUD implementation and how do they address these barriers? —we organize our findings into three sections. First, we outline the key barriers identified by participants, along with the strategies leaders developed to address them. Next, to identify whether barriers and strategies change as implementation progresses, we examined whether (and how) perspectives on barriers and strategies shift from baseline to follow-up interviews. Finally, to shed light on areas where innovation may be lacking despite widespread concern, we examine whether there is any mismatch between the frequency at which a barrier was mentioned compared to the frequency at which a related strategy was mentioned.

Barriers to implementation and organizational responses

Leaders focused their discussion on nine main implementation barriers, including: 1) staffing shortages, 2) financial barriers, 3) staff buy-in, 4) concerns of medication diversion (e.g., trading, selling, or unauthorized possession of MOUD medications), 5) space, 6) client buy-in, 7) intra- and extra- organizational relationships, 8) peer support, and 9) leadership. Below, we explain each of these barriers as well as the innovative strategies that leaders used to address the barriers and come under compliance with the mandate.

Staffing

Staffing shortages among security and medical personnel were the most significant barrier cited by jail and community health leaders. This was expected, given widespread understaffing in both corrections and addiction medicine (Heffernan & Li, 2024; McNeely et al., 2021). The added demands of the unfunded state mandate left staff at all levels feeling overwhelmed. Many security staff saw implementation of the mandate as another task added to an already full plate; as one jail leader stated, “It’s not our only job… we still have a jail to run.” Leaders feared shortages would hinder implementation, and post-interviews confirmed this: there weren’t enough officers to escort patients, enough medical staff to provide MOUD, or enough supervision to prevent diversion. As one leader reflected, “I wish we were fully staffed, but that’s not reality. We do the best we can with what we can.” Participants often described staffing shortages as a broader challenge of organizational capacity rather than simply a personnel problem. Leaders emphasized that implementing MOUD required additional labor, coordination, and oversight at a time when many facilities were already struggling to meet their basic operational needs. As a result, MOUD implementation was frequently viewed as competing with other institutional responsibilities for limited staff time and attention. These tensions left leaders balancing the demands of a new healthcare initiative against the ongoing need to maintain safety, security, and daily jail operations. To address staffing shortages, leaders mandated overtime and created hiring initiatives. Many institutions also created MOUD coordinator positions to facilitate the program. However, the success of these strategies was mixed: while hiring initiatives brought in some new staff, facilities continued to struggle with chronic understaffing, and existing employees reported feeling burned out from overtime.

Financial barriers

Financial constraints were pressing for jail and community health leaders. The state mandate did not include funding, leaving jails to cover (often through grants) costs estimated between $250,000 and $500,000 annually. Some struggled to pay staff overtime or afford more expensive medications like extended-release naltrexone. In the words of one jail leader, “once you have the money you can do almost anything”, and leaders felt stuck with the resources available to them. In many instances, they were also frustrated that the state mandated them to provide this care without providing the funding to do so. For example, one jail leader told us, “Yeah, they need to pay for it like they said they would when [state legislation] was passed and not just through grants.” and another jail leader remarked, “If it’s a law and they’re telling us we have to do it then shouldn’t they give us money…”. Another jail leader explained their thought process in more detail, saying,

I think the state made this law and they didn’t bring the right people to the floor. They didn’t bring the local detention centers or the local council, or the local government to the floor… what [one county’s] budget is for the detention center it’s three times bigger than what [another county’s] budget is… So you’re thinking that everybody has the same budget. And then you’re implementing grants that are so cumbersome a lot of places don’t even have grant writers… And so you didn’t help us [and] now we’re just trying to piece it together.

As their responses show, some jail leaders viewed the state’s unfunded MOUD mandate as disconnected, inequitable, and burdensome. Rather than empowering local facilities, the legislation left them scrambling to improvise under resource constraints and administrative obstacles. For many participants, the challenge was not simply the cost of implementation itself, but the perception that state policymakers underestimated the realities of implementation at the local level. Leaders described feeling excluded from policy discussions and frustrated by what they viewed as one-size-fits-all solutions that failed to account for substantial differences in jail resources, staffing, and infrastructure. As a result, financial barriers often reflected broader concerns about inadequate implementation support and a disconnect between those creating policy and those responsible for carrying it out. To address these financial concerns, organizations reallocated their budgets to account for rising costs in some areas and put a lot of effort into going after grants, in some instances hiring grant writers to increase their odds of securing funding. However, grants themselves were a challenge for leaders when they did not have (or lost) grant writers or did not get the funding they anticipated.

Staff buy-in

Security and medical staff buy-in was also noted as a significant barrier to implementation. Leaders shared that many jail staff viewed the program as an added burden or as “bringing those filthy drugs into their clean facility,” expressing skepticism about its value and concerns over diversion. Some jail staff resented the additional responsibility of observing dosing, often doing so inattentively, which increased the risk of diversion and reinforced their negative perceptions. A common sentiment was that MOUD was “just trading one drug for another.” Others questioned the program's effectiveness, pointing to the frequent re-incarceration of individuals on treatment. These views were often rooted in broader community stigma and “cold turkey” mindsets, which several leaders acknowledged were mirrored inside the facility. These concerns reflect broader tensions about the role of treatment within carceral settings and competing beliefs about addiction, recovery, and public safety. Rather than simply lacking information about MOUD, many staff questioned whether providing MOUD was compatible with the mission of the jail or whether individuals with OUD deserve access to such services while incarcerated. As a result, staff buy-in was closely tied to perceptions of the program’s legitimacy, requiring leaders to address not only knowledge gaps but also deeply rooted assumptions about substance use treatment and correctional practice.

To address this barrier, jail and community health leaders implemented ongoing education efforts aimed at dispelling myths, building confidence, and fostering shared ownership over the program. Staff training was both formal and informal, often starting with supervisors who then educated line staff, ensuring staff had opportunities to ask questions and understand the program’s purpose. In most cases, the more formal training was offered by medical staff at the facility. Jail leaders emphasized a relational approach, encouraging peer learning and consistent communication between custody and health staff. Education was strategically framed to align with existing routines and workloads, reducing resistance by assuring staff they were not “doing more work”, especially in an environment already ripe with staffing and resource shortages. Moreover, the training content, which covered medication types, diversion prevention, and specific dosing protocols, which was paired with ongoing reassurance that this effort would “save people’s lives,” helped shift the culture from skepticism to endorsement.

Medication diversion

Another concern that contributes to a lack of buy-in among staff are fears of medication diversion among incarcerated people, including the trading, selling, or unauthorized possession of MOUD medications. Leaders reported that staff were anxious about how some clients might “cheek” or “cup” medications to pass it to others and described repeated efforts to adjust procedures and tighten supervision in response. These issues contribute to staff resistance, as some feel the program enables manipulation or adds to their workload, making diversion a key barrier to implementation. More broadly, concerns about diversion reflect tensions between treatment goals and custodial responsibilities. Leaders described diversion as a threat not only to the integrity of the MOUD program but also to institutional safety and control, raising concerns about contraband, manipulation, and liability. As a result, successful implementation required leaders to reassure staff that treatment could be delivered without compromising security, often through the development of new monitoring practices and accountability measures. These fears were addressed by introducing new protocols, as described by one jail leader below,

They sit there 15 minutes… now we’re using the black barber capes, it’s over them so they can’t spit it in their pocket, none of that stuff. The gym’s floor is nice because it’s a painted floor, so any kind of substance shows up on the floor pretty good so that works out well. So, it’s just trying to stay one step ahead of them as they try to sneak it back to the housing unit and sell it.

In addition to adopting these protocols, some facilities also hired drug-sniffing dogs that can detect diverted medication. Maryland has a history of using dogs to detect illicit substances (e.g., alcohol, medications) but the extent to which these efforts improve detection and reduce contraband remains contested (Watson, 2016).

Space

Jail leaders explained that their facilities were not designed for the necessary medication and observation areas. As one put it, “The medication is the easy part; it’s the programming and having space… space is a challenge.” Many jails were small and did not have the extra space to accommodate a new MOUD program. One jail leader emphasized that in order to fully accommodate the program, they would have to drastically restructure their medical units, which would be quite expensive,

We don’t have the funding to… we don’t even have the space to actually assess bringing people down to medical and try to do what the medical needs. Once we get an infirmary, we add space, add cells where people can go in if they need to be seen at a certain time… Right now we don’t have that… ten million dollars would do more than fix up the back of the jail, would do more than help expand and open up spaces…but until I get ten million…

This response suggests that even if staff wanted to provide treatment, the (current) physical infrastructure simply could not accommodate it. The leader emphasizes that meaningful implementation would require significant funding to address special concerns, underscoring how space constraints made MOUD delivery feel unattainable to many and compounding the stress of the unfunded mandate that many leaders already felt. More broadly, participants described space constraints as reflecting a mismatch between the physical design of jails and the demands of contemporary healthcare delivery. Many facilities were originally constructed to maximize security and housing capacity rather than support ongoing treatment services, leaving leaders to adapt existing environments to accommodate a new program. As a result, implementation often depended on leaders’ ability to creatively repurpose existing infrastructure, highlighting how the built environment can shape the feasibility of policy implementation. To address space concerns, institutions opted to use telehealth as frequently as possible. They also repurposed existing spaces such as classrooms, multipurpose rooms, and medical wings to make it work, such as one jail leader who installed benches in each housing area so that medications could be passed out on the blocks.

Client buy-in

Client buy-in was also seen as a barrier to MOUD implementation. Leaders believed that many incarcerated individuals are skeptical of the treatment because they are fearful about how participation might be perceived by judges, family, or peers. One community health leader shared that their patients often share sentiments like, “I’m afraid of what my family’s going to think if I’m on prescribed dope instead of street dope,” and “if I take [MOUD] then I’m not really clean.” This skepticism can be intensified by misinformation or stigma around MOUD in recovery communities, a lack of trust in jail-based services, or negative past experiences. Jail leaders observed low uptake despite program availability: “We just haven’t had very many takers on it,” and “people aren’t lining up at the door asking for a [long-acting injectable buprenorphine] injection.” Others described individuals initially expressing interest but backing out later, often due to release timing, changing motivations, or misunderstandings. One jail leader shared, “by the time we get ready to medicate them either they change their mind, or they get released and it sort of passed.” These concerns suggest that client resistance was often rooted in broader issues of trust, stigma, and treatment legitimacy rather than a simple unwillingness to participate. Leaders described how individuals weighed the potential benefits of MOUD against fears of judgment from family members, peers, recovery communities, and even themselves. In this way, implementation depended not only on making treatment available but also on helping clients view MOUD as a legitimate and acceptable pathway to recovery. Leaders therefore faced the challenge of overcoming social barriers that extended well beyond the walls of the jail.

Jail and community health leaders emphasized client education as a critical strategy to promote agency, motivation, and long-term recovery. Education revolved around explaining how MOUD works, what clients could expect during treatment, and how it could support their recovery journey. Community health staff framed education as a way to empower clients with knowledge and compassion, aiming to interrupt cycles of relapse and incarceration without blame. Jail staff took a more logistical approach, focusing on setting clear expectations around program procedures, medication types, and treatment timelines.

Relationships

Intra- and extra-organizational relationships were seen as a barrier to MOUD implementation due to poor coordination and communication between jails and community providers. Community health leaders believe that reentry planning often lacks consistency, leaving community organizations unaware of when individuals are being released or what care they received while incarcerated. As one community health leader shared, jail personnel responsible for re-entry services sometimes did not initiate contact with them: “To date we have never received one call or e-mail from a person who’s supposed to be doing re-entry services.” This lack of communication disrupts continuity of care, as people are frequently discharged without medications, follow-up plans, or basic reentry support like housing, phones, or benefits. And, even when partnerships exist, timing and logistics can break down. A jail leader noted the difficulty in ensuring access to methadone for a person released on a weekend: “We were scrambling to try to figure out how he was going to get his dose for Saturday and Sunday.” As one community health leader summarized, “If we’re going to really, really try to get treatment services in the jail, that connection to care to me is the biggest missing link.” These challenges suggest that successful MOUD implementation depends not only on the actions of individual organizations but also on the strength of coordination across organizations. Leaders frequently describe implementation breakdowns occurring at points of transition, particularly when responsibility shifted between jails and community providers. As a result, continuity of care was often shaped by the quality of communication, trust, and information sharing between organizations. For many participants, MOUD implementation was ultimately a cross-system endeavor, requiring leaders to bridge gaps between institutions that historically operated independently of one another.

Developing and sustaining strong intra- and extra-organizational relationships was a central strategy used by jail and community health leaders to implement MOUD legislation effectively. Leaders emphasized the importance of teamwork, daily communication, and collaboration across roles to ensure smooth operations and sustained change. Regular interdisciplinary meetings brought together security, medical, mental health, peer support, classification, and external providers to coordinate care and better understand each other's roles. Community-based meetings were also used to share available resources and improve communication among local agencies. Beyond formal meetings, ongoing collaboration supported discharge planning, continuity of care, and warm handoffs. Community health staff maintained a frequent presence in the jail, and facility leaders fostered partnerships through joint facility tours and open communication. These coordinated efforts helped streamline service delivery and reflected a shared commitment to recovery, reinforcing the idea that successful MOUD implementation depends on collective, cross-system engagement.

Peer support

Although peer support (i.e., hiring staff with lived experience) was not identified as a barrier to MOUD implementation, it was frequently cited as a key strategy, likely because it was mandated by the legislation. Jail and community health leaders quickly recognized the value of hiring staff with lived experience, highlighting peers' unique ability to build trust and rapport with clients in ways traditional staff often could not. Peers, some of whom had used MOUD themselves, could speak authentically about treatment and recovery, making them powerful messengers and motivators. Leaders described peer support as an essential part of implementation, praising their contributions to daily operations, client engagement, and care coordination. Despite challenges with hiring—especially due to background check restrictions—peers were seen as irreplaceable assets. As one community health leader put it, “the mandate to include peer specialists was such a gift”. Though not a grassroots strategy, peer support became a vital tool that enhanced program credibility and helped both facilities and clients more fully embrace treatment. More broadly, leaders viewed peer support as a mechanism for building trust and legitimacy within the treatment process, making treatment information more relatable and fostering engagement among individuals who might otherwise remain skeptical of services. In this way, peer support served as a tool for strengthening relationships and promoting confidence in the program.

Leadership

Leadership changes pose a barrier to MOUD implementation, particularly when new jail administrators are unfamiliar with existing or effective models of care. Transitions in leadership often lead to disruptions in programming. Leadership also influences the direction and scope of MOUD programs. A community health leader described their experience with a former jail leader, saying “She really didn’t deal with the Health Department… she was a strong advocate of Vivitrol [extended-release naltrexone]… Methadone really never took off.” After her departure, a new administrator welcomed broader engagement, opening the door for methadone provision and partnerships with external treatment providers. This shift illustrates how leadership buy-in and orientation toward MOUD can either restrict or expand treatment options and cross-agency collaboration.

Strong, supportive leadership was central to translating Maryland’s MOUD mandate from policy to practice. Effective leaders helped navigate organizational change, align staff, and maintain momentum, setting the tone for how seriously the mandate would be taken. In some facilities, leadership shifts transformed hostile environments into collaborative ones, with staff becoming more open and participatory. One community health leader described a new team as “sharp,” “ready to go,” and fully “supportive,” contrasting it with previous leadership that blamed outside providers for institutional problems. Leadership styles varied across sites, from authoritative, top-down approaches that emphasized compliance, to more collaborative, hands-on leadership that engaged deeply with the community. Regardless of style, strong leadership consistently emerged as a key driver in successful MOUD implementation. These findings suggest that leadership influences implementation not only through administrative decision-making but also through the values, priorities, and relationships leaders cultivate within their organizations. Participants described how leaders shaped staff attitudes toward treatment, determined the extent of collaboration with external partners, and influenced which treatment options were viewed as legitimate or feasible. As a result, implementation was often contingent upon leadership vision and commitment.

Shifts in perspective over time

To identify whether challenges and strategies change as implementation progresses, we examine whether (and how) perspectives on barriers and strategies shift from baseline (n = 17) to follow-up (n = 16) interviews. Below, Fig. 1 displays how perceptions of barriers and strategies changed over time.

Fig. 1.

Shifts in perspective of barriers and innovations between baseline and follow-up. Notes: Baseline n = 17, Follow-up n = 16

Examining changes in perceptions of barriers over time suggests that concerns about staffing, financial barriers, interorganizational relationships, and leadership increased from baseline to follow-up interviews. In most cases, concerns rose steadily (e.g., 76% of leaders were concerned about staffing in baseline interviews and 88% were concerned about staffing in follow-up interviews) but one area, relationships, saw a large jump from 12% in baseline to 50% in follow-up interviews. Moreover, in some cases, barriers were slightly alleviated over time but were still concerning. For example, one jail leader told us, “We’re still understaffed. I don’t foresee that unfortunately going away any time soon. We’re better staffed than we were five months ago I would say but not by much.” In other cases, barriers not only persisted but intensified as MOUD implementation progressed. For instance, one community health leader said, “We haven’t made a whole lot of growth and change we’re still struggling maybe even more so to find any sort of folks that we can hire to do this work.” This trend may indicate that the strategies designed to address these barriers were either insufficiently resourced or inconsistently applied.

In contrast, declining mentions of staff buy-in, medication diversion, space limitations, and client buy-in suggest that strategies targeting these barriers may have achieved partial or perceived success. For example, one jail leader said, “Once we started talking about it and training staff everybody was a little bit more receptive” and another remarked, “Honestly I am very impressed with all of our staff because any time change is introduced into corrections we’re sometimes hard to accept change.” Notably, peer support was not identified as a barrier at either time point.

Interestingly, implementation efforts expanded across all targeted barrier areas over time, regardless of whether concerns intensified or dissipated. Some areas had smaller increases, like financial barriers (65% BL vs. 81% FU) and staff buy-in (53% BL vs. 63% FU). Concerns about financial barriers rose at the same time that financial-related strategies increased, while strategies regarding staff buy-in increased even though concerns about staff buy-in decreased. The same is true for strategies regarding diversion, space, and client buy-in barriers. Other areas had larger increases in strategies, like staffing (18% BL vs. 50% FU), relationship building (18% BL vs. 63% FU), and peer support (6% BL vs. 38% FU). Both staffing and relationship strategy trends mirror barrier trends, while peer support strategies increased even though there was no uptick in concerns about peer support.

Alignment between barriers and strategies

To examine whether there are any areas where innovation may be lacking despite widespread concern (or the opposite), we examine if there is any mismatch between the frequency at which a barrier was mentioned compared to the frequency at which a related strategy was mentioned. Below, Table 1 displays the alignment between barriers and strategies. To identify mismatch, we group barriers into three categories: infrequently mentioned barriers (where 0 to 19% of leaders mentioned the barrier), somewhat mentioned barriers (where 20 to 49% of leaders mentioned the barrier), and frequently mentioned barriers (where over 50% of leaders mentioned the barrier). We grouped strategies into similar categories: infrequently mentioned innovation (where 0 to 19% of leaders identify strategies to address the associated barrier), somewhat mentioned innovation (where 20 to 49% of leaders identify strategies to address the associated barrier), and frequently mentioned innovation (where over 50% of leaders identify strategies to address the associated barrier).

Table 1.

Alignment between barriers and innovations

Baseline
	Infrequently Mentioned Innovation (0–19%)	Somewhat Mentioned Innovation (20–49%)	Frequently Mentioned Innovation (50% +)
Infrequently Mentioned Barrier (0–19%)	Relationships (B:12%, I:18%)* Peer Support (B:0%, I:6%) Leadership (B:0%, I:12%)
Somewhat Mentioned Barrier (20–49%)	Diversion (B:47%, I:12%)** Space (B:35%, I:18%)	Client Buy-In (B:29%, I:29%)
Frequently Mentioned Barrier (50% +)	Staffing (B:76%, I:18%)		Financial Barriers (B:53%, I:65%) Staff Buy-In (B:53%, I:53%)

Follow-up
	Infrequently Mentioned Innovation (0–19%)	Somewhat Mentioned Innovation (20–49%)	Frequently Mentioned Innovation (50% +)
Infrequently Mentioned Barrier (0–19%)	Leadership (B:13%, I:19%)	Space (B:19%, I:25%) Client Buy-In (B:6%, I:44%) Peer Support (B:0%, I:38%)
Somewhat Mentioned Barrier (20–49%)		Diversion (B:31%, I:25%)	Staff Buy-In (B:38%, I:63%)
Frequently Mentioned Barrier (50% +)		Staffing (B:88%, I:50%)	Relationships (B:50%, I:63%) Financial Barriers (B:69%, I:81%)

* B denotes the percentage of participants who mentioned the barrier, while I denotes the percentage of participants who mentioned innovations related to the barrier. ** Italicized text denotes a mismatch between the frequency at which a barrier was mentioned compared to the frequency at which a related innovation was mentioned

At baseline, three barrier-strategy combinations are mismatched. Medication diversion was mentioned by 47% of leaders as a barrier to implementation (a somewhat mentioned barrier) but only 12% of leaders had developed strategies to address this (an infrequently mentioned strategy). Similarly, space was described as a barrier to implementation by 35% of leaders (a somewhat mentioned barrier) but only 18% of leaders tried to address, this issue (an infrequently mentioned strategy). Finally, staffing was viewed by 76% of leaders as a barrier (a frequently mentioned barrier) but only 18% of staff discussed strategies to address, this issue (an infrequently mentioned strategy). These mismatched barrier-strategy combinations highlight a critical implementation gap, where notable—sometimes higher-priority—barriers are not being addressed with adequate resources. Although leaders did not explicitly discuss why certain barriers received less attention, these decisions may reflect broader structural constraints, such as limited funding and staffing shortages. These constraints may help explain why some barriers persist throughout the implementation process.

At follow-up, new patterns emerge. Now, five barrier-strategy combinations are mismatched, though predominantly in the direction of more innovation than concern. Space issues were mentioned by 19% of leaders as a barrier to implementation (an infrequently mentioned barrier) but a larger percentage of leaders (25%) developed strategies to address this (a somewhat mentioned strategy). Client buy-in (6% saw as barrier versus 44% innovated) and peer support (0% saw as barrier versus 38% innovated) follow a similar pattern. Staffing remains a frequently mentioned barrier (88% mention it), though advances to a somewhat mentioned strategy (50% of leaders mention a staffing-related strategy). Finally, staff buy-in was noted by 38% of leaders as a barrier (a somewhat mentioned barrier) while 63% of leaders reported innovating in this area (a frequently mentioned strategy). These patterns reflect the findings of the previous section, that strategies increased over the implementation process.

Discussion

Through interviews with jail and community health leaders working in Maryland following the passage of state legislation mandating jail-based MOUD, analysis for this paper suggests that leaders and organizations faced numerous and evolving barriers to implementing jail-based MOUD programs. While some barriers, including staff buy-in, client buy-in, space concerns, and fears of medication diversion, lessened as strategies like staff education, client education, space accommodations, and diversion protocols increased, others, including staffing shortages, financial barriers, strained intra- and extra-organizational relationships, and leadership challenges, were particularly persistent and even intensified over time. Further, comparing barrier and strategy frequencies revealed critical mismatches, where the most frequently cited barriers (e.g., staffing and medication diversion) were met with disproportionately fewer strategies while lesser cited barriers (e.g., client buy-in) were frequently innovated on. One possible explanation for this trend is that staffing is largely a budgetary concern and as this was an unfunded mandate, it is unsurprising that there were fewer strategies. Medication diversion can also be a downstream effect of staffing. From our analysis, it appears that some important/salient challenges were ignored (or at least, not acted on) and sometimes leaders came up with strategies that were targeting non-existent (or not important/salient) challenges such as client buy-in. Our findings underscore that sustained jail-based MOUD implementation requires not only identifying barriers but strategically aligning resources and efforts to address them in meaningful and enduring ways.

Contributions to theory and research

Our findings contribute to both IS and the practice of carceral-based MOUD programs by deepening our understanding of how barriers are identified and sometimes misaligned with strategies during the implementation process. This research advances IS by demonstrating the dynamic and evolving nature of barriers and strategies in the context of mandated carceral reform, and specifically mandated MOUD care. Existing IS frameworks provide useful categorizations of where barriers may arise (e.g., individual, inner setting, outer setting), but they have been less attentive to how strategy efforts adapt—or fail to adapt—over time. Our approach, which captures leader perspectives at both baseline and follow-up, suggests that some barriers (e.g., staff buy-in, space limitations) become less salient as targeted strategies are employed, while others (e.g., staffing shortages, financial constraints) persist or intensify despite awareness of the barrier. This suggests that implementation success requires not only identifying barriers but regularly reassessing them and updating innovative strategies accordingly. Second, we identify a critical mismatch between barrier salience and innovative action, an insight that pushes IS beyond binary facilitator/barrier typologies. While much of the IS literature treats facilitators as assets and barriers as obstacles, our findings reveal the consequences of misalignment: salient challenges (like staffing) often received less strategic efforts, while less frequently cited issues (like client buy-in) received a disproportionate share of strategic focus. One possible explanation is that leaders focused on barriers they felt they could control, such as client buy-in, while neglecting more frequently cited challenges like staffing shortages, which they viewed as beyond their resources or ability to resolve. This mismatch reflects an under-theorized aspect of implementation—barrier-innovation match—and suggests the need to more closely examine how decision-makers interpret challenges and allocate resources/efforts in ways that may or may not be data-driven.

In doing so, this research fills a notable gap in the carceral-based MOUD literature. While previous research identifies both barriers and facilitators of MOUD implementation (e.g., Krawczyk et al., 2022; McNeeley et al., 2024), missing from these studies is whether strategies were deliberately and strategically matched to the barriers they aimed to resolve. Our findings suggest that such intentional alignment is relatively rare. In some cases, strategies were deployed without a clearly articulated connection to pressing challenges; in others, leaders acknowledged persistent issues but lacked the ability or drive to develop meaningful solutions. This disjuncture impedes progress toward both acceptability and feasibility, the two central IS criteria necessary for sustainable implementation.

Contributions to practice

Practically, these findings offer important insights for carceral and community health leadership, policymakers, and researchers. Most notably, sustained MOUD implementation cannot rest on initial reform mandates or static assessments of organizational readiness. Instead, it requires ongoing barrier identification, prioritization of resources based on salience and impact, and a willingness to revise or replace strategies that fail to address the most pressing problems. Furthermore, these findings underscore the importance of leadership, intra- and inter-organizational relationships, and adequate resources in building and maintaining MOUD programs in carceral settings. Addressing persistent challenges like staffing shortages and strained relationships with outside agencies will require changes that extend beyond what frontline staff or even jail or community health leaders can control. These issues demand broader policy solutions, such as sustained state-level investment in workforce development, competitive compensation to attract and retain staff, and formal mechanisms for interagency coordination. This reinforces the value of IS frameworks that account for outer setting factors, including state mandates, funding availability, and community partnerships.

Limitations and future research

This research is limited in a few ways. Interviewees might not have brought up all the barriers and strategies present in their organization, choosing to highlight what they perceived as most important or impactful. Relatedly, although we, similar to other qualitative researchers (e.g., Hochwald et al., 2023; Sandelowski, 2001; and Seale & Kelly, 2004), use frequencies to identify broad patterns in how barriers and strategies were discussed across baseline and follow-up interviews, frequency should not be interpreted as a direct indicator of importance, emotional salience, or depth of meaning. Additionally, because one researcher completed the coding, the findings may reflect some degree of subjective interpretation. However, findings and interpretations were discussed collaboratively among the research team to minimize bias. Moreover, this study focuses on the perspectives of jail and community health leaders and does not include insights from line staff or incarcerated individuals, both of whom play critical roles in the implementation process. Including these voices in future research would offer a more comprehensive understanding of how carceral-based MOUD implementation is enacted and experienced on the ground. Additionally, although numerous topical areas were identified during first-cycle coding, the present analysis focused on implementation acceptability and feasibility. While this allowed for a more in-depth examination of MOUD barriers and strategies, it may have limited attention to other relevant domains (e.g., attitudes toward MOUD or community linkage) that could be further explored in future research. Finally, the study timeline may not have been long enough to observe the full arc of implementation success or failure, particularly with regard to sustainability. Some strategies may have been too new to assess their long-term utility or alignment with evolving barriers, and future research could examine this phenomenon over a longer duration of time.

Conclusion

This study contributes new theoretical and empirical insights by tracing how jail and community health leaders in Maryland navigated, responded to, and sometimes misaligned their responses to barriers in implementing jail-based MOUD. Moreover, our findings challenge researchers and practitioners to think beyond isolated lists of barriers and facilitators and instead consider how alignment (or misalignment) between barriers and strategies impacts implementation success. Sustained, effective reform depends on strategies rooted in clear understanding of the problems at hand, grounded in contextual realities, and responsive to change over time.

Authors’ contributions

B.K.M conducted the data analysis and wrote the main manuscript text. D.R. assisted with data analysis and helped draft portions of the main manuscript text. T.B., F.V., and M.G. participated in project management/data collection and reviewed and edited the manuscript.

Funding

This research received funding from the National Institute on Drug Abuse (UG1DA050077) under the NIH HEAL JCOIN Initiative^SM.

Data availability

The data that support the findings of this study are not publicly available. De-identified data may be made available from the corresponding author upon reasonable request and with appropriate institutional approvals.

Declarations

Ethics approval and consent to participate

Before each interview, the researcher read an informed consent script to participants, and all interviewees consented to participating in the interview. The research was conducted in accordance with applicable ethical standards for research involving human participants and adhered to the principles outlined in the Belmont Report. All study protocols were reviewed and approved by WCG IRB. WCG IRB is a private, for-hire IRB agency. For more information, see https://www.wcgclinical.com/about/.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.