Barriers and Facilitators to Establishing Partnerships for Substance Use Disorder Care Transitions Between Safety-Net Hospitals and Community-Based Organizations

Abstract

Background

The effectiveness of hospital-based transitional opioid programs (TOPs), which aim to connect patients with substance use disorders (SUD) to ongoing treatment in the community following initiation of medication for opioid use disorder (MOUD) treatment in the hospital, hinges on successful patient transitions. These transitions are enabled by strong partnerships between hospitals and community-based organizations (CBOs). However, no prior study has specifically examined barriers and facilitators to establishing SUD care transition partnerships between hospitals and CBOs.

Objective

To identify barriers and facilitators to developing partnerships between hospitals and CBOs to facilitate care transitions for patients with SUDs.

Design

Qualitative study using semi structured interviews conducted between November 2022-August 2023.

Participants

Staff and providers from hospitals affiliated with four safety-net health systems (n=21), and leaders and staff from the CBOs with which they had established partnerships (n=5).

Approach

Interview questions focused on barriers and facilitators to implementing TOPs, developing partnerships with CBOs, and successfully transitioning SUD patients from hospital settings to CBOs.

Key Results

We identified four key barriers to establishing transition partnerships: policy and philosophical differences between organizations, ineffective communication, limited trust, and a lack of connectivity between data systems. We also identified three facilitators to partnership development: strategies focused on building partnership quality, strategic staffing, and organizing partnership processes.

Conclusions

Our findings demonstrate that while multiple barriers to developing hospital-CBO partnerships exist, stakeholders can adopt implementation strategies that mitigate these challenges such as using mediators, cross-hiring, and focusing on mutually beneficial services, even within resource-limited safety-net settings. Policymakers and health system leaders who wish to optimize TOPs in their facilities should focus on adopting implementation strategies to support transition partnerships such as inadequate data collection and sharing systems.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11606-024-08883-8.

INTRODUCTION

The national substance use disorder (SUD) epidemic continues to deepen, with over 109,000 deaths due to drug overdose reported in 2022, most of which involved opioids.¹ Amidst this crisis, the rapid increase in emergency department (ED) visits and hospitalizations for patients with SUDs, such as opioid use disorder (OUD), represents a failure to engage patients in treatment before the most serious complications develop. It is estimated that one of every nine hospitalized patients in the United States (US) has an SUD,² and ED visits among patients with SUD as a primary diagnosis increased from 74.4 per 10,000 population in 2018–2019 to 103.8 in 2020–2021.³

Importantly, the burden of untreated SUD is not shared equally in the US, with profound racial/ethnic inequities in access to treatment.⁴ Within safety-net hospitals, which disproportionately serve vulnerable and uninsured patients and have fewer resources available to devote to addressing SUD, admissions related to SUD have risen sharply.^5,6 For example, a recent study from a large safety-net health system found that from 2008 to 2020, hospital admissions related to any SUD increased by 42%, accounting for over one-third of total hospital admissions in 2020.⁷ Similar increases in SUD-related admissions have been observed at non-safety hospitals, with one study finding that ED admissions due to substance use at U.S hospitals increased by 44% between 2006 and 2014.⁸

In this context, interventions that increase access to evidence-based SUD treatment are critical, and hospitals are a key juncture to engage patients with SUD in care.⁹ In response to this crisis, a number of hospital-based approaches to addressing SUDs have emerged in recent years. One successful model is transitional opioid programs (TOPs), which aim to transition patients to ongoing treatment in the community following initiation of medication for opioid use disorder (MOUD) treatment in the hospital.⁹ Several models of TOPs exist nationwide, varying with regards to staffing,^10,11 types of SUD services offered,¹¹ location (ED or inpatient),¹² and funding source.⁹ Overall, evaluations of these programs demonstrate their effectiveness at improving rates of treatment receipt amongst SUD patients.^9–11 Importantly, disparities exist in the availability of TOPs; these programs are more likely to be available in academic health centers and are less frequently offered within safety-net hospital settings.¹² Strategies to increase access to TOPs within safety-net hospitals are essential to improving equity in addiction treatment.

A critical factor in the effectiveness of TOPs is the successful transition to ongoing treatment in the community, which is enabled by the development of partnerships between hospitals and community-based substance use treatment settings.¹¹ Studies have found that the percentage of patients who engage in outpatient SUD treatment after involvement with TOPs are less than 50%,^13,14 in part due to social determinant of health-related barriers, such as limited transportation and housing, as well as time lags between hospital discharge and open spots in outpatient treatment programs.^11,13 This signals that even among established TOPs, there are opportunities to improve outcomes through better partnerships that enable patient advocacy and enhanced care coordination to overcome social determinant of health (SDOH) related challenges to transitions in care.¹¹ Previous research has described the barriers and facilitators to forging partnerships between health systems and community-based organizations (CBOs), which include poor communication, lack of data, and privacy concerns as barriers,^15,16 and having designated points of contact, clear requirements, and mutual trust as facilitators to these relationships.^15,17 Similarly, while prior research studies have discussed elements of TOPs that contribute to their success, such as pharmacotherapy initiation and peer navigation,^11,18,19 the development of partnerships between TOPs and external organizations remains understudied, and no study has specifically examined barriers and facilitators to establishing care transition partnerships between hospitals and community-based substance use treatment organizations.

To address these gaps, our study uses qualitative interviews with leadership and clinical and non-clinical staff at four safety-net health systems and their CBO partners to describe barriers and facilitators to developing and sustaining partnerships between these care settings. Care coordination and care transitions are key to combating the opioid epidemic and, for many individuals with OUD, hospitals are a major entry point into the healthcare system. It is important to understand how partnerships that enable hospitals to transition SUD patients to ongoing treatment in the community can be feasibly implemented within these settings.

METHODS

Study Settings and Participants

The qualitative data utilized in this study were collected as part of a larger study focused on how safety-net hospitals in the US implement TOPs. Interviewees included staff in administrative, clinical, and non-clinical support roles at hospitals associated with four safety-net healthcare systems: 1) City Health (n=4) based in Illinois; 2) Lake Health (n=6) in Ohio; 3) Metro Health (n=6) based in New York; 4) Mountain Health (n=6) based in Ohio. We assigned health systems pseudonyms to mask the identity of participants and institutions. Participants also included leaders and staff from CBOs that both support and provide treatment to patients with SUDs, who were referred to the study team by their health system partners (n=5 across 4 CBOs). Specifically, we asked health systems to identify partners that support TOPs in the community setting, which varied from private addiction treatment organizations (n=2) to public health departments (n=1) and county mental health and addiction services boards (n=2) that help coordinate treatment and harm reduction services in the community, including naloxone distribution and syringe services. We selected participating health systems based on their organizational characteristics (e.g. hospital size) as well as the characteristics of the surrounding community (e.g. urban vs. rural location), to ensure a diversity of viewpoints was represented in this study. Hospital participants were selected based on conversations with leaders from each health system, and each health system provided consent for participation in the study in addition to the individual interviewees.

Data Collection and Measures

Data collection entailed semi-structured interviews that occurred from November 2022 until August 2023. The initial interview guide included 15 questions that fell into five categories: general background, opioid use in the community, TOPs, facilitators to implementing TOPs, and barriers to implementing TOPs. Within our interviews, partnerships were defined as referral networks or relationships that organizations have with partners in the community, and successful partnerships were those that result in consistent and effective transitions from hospitals to community-based organizations. When appropriate, we used interview probes that asked about the history and motivation behind established partnerships, short term vs. longer term engagement with partners, and strategies organizations used to maintain partnerships. As per established procedures for semistructured interviews, we asked follow-up questions to gain further detail on participants' answers, tailoring these questions to their individual context.²⁰ Appendix Table 1 contains the full interview guide.

At least two study team members conducted each interview, with one team member asking questions and the second taking notes. Interviews ranged from 45-60 minutes in length and were conducted via Zoom. Interviews were audio recorded, professionally transcribed, and de-identified. Following each interview with health system participants, we followed up via email to ask for the names and contact information of partners at CBOs to be interviewed for this study. Participants were compensated with $50 Amazon gift cards and approval for this study was obtained from the Ohio University Institutional Review Board.

Data Analysis

To create a codebook, 5 study team members (ZL, CF, BF, AL, JC) independently coded a transcript and discussed similarities and differences in the resulting codes. We agreed on an initial codebook and developed descriptions of how these codes should be used in the coding process. Following the creation of a codebook, two study team members (ZL and CF) coded each interview transcript to ensure codes are applied consistently. Additional codes were created as needed, and reviewed and confirmed by the study team. ZL then extracted segments of interview texts that had been open coded with the codes ‘Barrier’ and ‘Facilitator’ and performed axial coding to identify segments describing barriers and facilitators to developing partnerships with more details and context. In the open coding process, we coded data segments as a barrier if it described challenges to forging partnerships with external organizations, and a facilitator if it described strategies to enable the development of partnerships between organizations. After conducting axial coding, in associations between and within codes are examined,²¹ ZL then grouped the codes into larger, more abstract categories; this entailed asking “whether the emerging themes suggest concepts that may help us describe and explain the phenomena we are observing”.²²

RESULTS

Participants described several barriers to developing SUD transition partnerships with external organizations, which fell into four domains: 1) policy and philosophical differences in treatment of substance use; 2) communication; 3) limited trust and 4) lack of connectivity between systems. Participants also described facilitators that enabled them to overcome barriers and develop and sustain external partnerships, which fell into three domains: 1) building partnership quality; 2) Strategic hires and staffing; and 3) organizing partnership processes. The specific barriers discussed within each domain are presented below, followed by a description of the facilitators, with selected quotes included in each section.²³ We selected quotes that were able to capture the essence of the themes that emerged from our analysis, as well as to illustrate the context in which our findings were based. Figure 1 displays these results and Table 1 contains additional excerpts from study participants organized by barrier/facilitator category. Overall, the barriers and facilitators described by hospital participants and CBO participants were consistent, and Table 2 indicates how many times specific barrier/facilitator categories were discussed by hospital participants vs. CBO participants.

Figure 1.

Barriers and facilitators to developing transitions partnerships between hospitals and CBOSs.

Table 1.

Quotations from Interviews Corresponding to Barrier and Facilitator domains

Barrier/Facilitator Domain	Quotations
Barrier: Policy and Philosophical Differences in Treating Substance use	"We've got some really strong agencies in here that do that. The medically assisted part has been growing. It's an uphill battle in our communities. A lot of people just don't believe in it. So we've got a few, but they're increasing. Some of them work with us, not all of them work with us or are monitored by us for the behavioral health side." (Lake Health ID 5, CBO participant, leadership role) "It's got to be a community of providers because these are patients who have a lot going on in their life and they may not want to adhere to the ideas that you have for as a doctor for what you think a medical practice looks like. So, this whole thing is like, "Oh, we have a grace period of 20 minutes for you to check into your appointment." That's all out the window and you know that." (Metro Health ID 5, clinical role)
Barrier: Communication	"You have a system lead, you have system committees, people turn over, people may or may not show up, leaders may or may not cascade information. And that's everywhere, that's just everywhere." (Mountain Health ID6, hospital participant, clinical role) "I remember back in the early days, it was just trying to get people, when we said, "Hey, we're going to deploy a recovery coach to the emergency room." People were like, "What's a recovery coach?" And so we really had to define for people, what does a recovery coach do? What's their skillset? What's their background? What are the goals? What are we going to do?” (City Health ID1, hospital participant, administrative role) "It was challenging. I'll put it that way, in trying to get the facilities to work with you in some regards when it comes to, "Okay, I'm not giving you HIPAA. Can you just tell me if JK came?" Some people would, some people wouldn't. I was like, "I just want to be able to stop pestering somebody." Some would be agreeable, some weren't." (Lake Health ID4, hospital participant, clinical role) "Now if we discharge a patient from the hospital in the evening and they need to do an intake for a methadone program, the methadone program isn't open. They're not open on weekends. And I'm not really trying to pooh pooh them for their hours of operation, but let's just be realistic. It's not very accessible from a collaboration standpoint. It doesn't allow us to collaborate with them well when we are basically wide open and taking all comers and they are very constricted in their time of access and the services that they provide." (Metro Health ID 1, health system participant, clinical role) "Those services were also established in another county with another hospital, and when the hospital changed ownership, that was the first program to go, leaving an even bigger void in our community." (Lake Health ID7, CBO participant, leadership role)
Barrier: Limited trust	"And that's the problem of trust, well, at least on the treatment side, the problem of trusted providers is really hard in the age of Google where you can pay to be at the top of a search. And I don't think that's something that we've figured out" (Metro Health ID6, hospital participant, clinical role) "The question is are they offering the right services that our patients need? And I will just say the quality of care, I find it to be quite abysmal in a lot of our addiction treatment community. There's not been in the health systems, health hospital systems in particular, quality improvement, patient safety. Those are things we talk about every day." (Mountain Health ID2, hospital participant, clinical role)
Barrier: Lack of Connectivity Between Systems	"One of the things that we are hearing more and more and we struggle with is accessing data, to support the need or to support the ask that we make of whether it's our local governments, resources or our state or federal. We lack data to support what really is truly a successful program that works and putting those measures in place with our providers to support what we're doing." (Lake Health ID7, CBO participant, leadership role) “As a doctor in the hospital system, I'm not credentialed to practice medicine in that building. So I can't write prescriptions, I can't send lab orders. There's no connection between the medical care I provide here at the hospital and our program that is in that building." (Metro Health ID1, health system participant, clinical role) "It's one of the challenges we wish we could provide at all, so it'd be all on the same EMR and we could really see if someone showed up and what happened." (Mountain Health ID4, health system participant, clinical role)
Facilitator: Building Partnership Quality	"So it's being able to really identify what resources our health system has, whether it's data or clinical, and then what do our partners need and how can we support each other." (City Health ID1, hospital participant, administrative role) "The reality is that we can't do everything. It's the reality. And the things that we do, we try to do the best that we can do. But for some patients, for whatever reason, they're going to do better in a non-hospital setting. We don't claim to have a monopoly on successful programs, so it's important to have those relationships. And there are people that do things that we don't do." (Metro Health ID 2, hospital participant, clinical role) "I would say if somebody was starting up a new place, they need to give that person or those people the time to go visit the facility agencies. Show their face, get their information, say, "We're trying to start this up. What can we each do for each other?" (Lake Health ID 4, hospital participant, clinical role) “The community agencies partner and work very well together because since our resources are scarce, we want to make sure to avoid any duplication. So we've always had to work well together and collaborate on projects and things like that.” (Mountain Health ID 5, CBO participant, administrative role)
Facilitator: Strategic hires and staffing	"So for instance, in our emergency room, we do some challenges hiring, we reached out to one of our community treatment providers who has full services residential IOP that [organization name] doesn't offer. And so we actually use them to hire a couple of recovery coaches in our emergency room. And that partnership has been great because not only did it fill a staffing need that we had, but our patients also have really warm and easy handoff to this treatment provider and that treatment provider offers services that [organization name] doesn't." (City Health ID1, hospital participant, administrative role) "And then linkages. We have therapists within our clinic, but linkages to outpatient programming and meetings and all sorts of community things that people are interested in engaging. That was our focus in 2017, to start that up on. So we hired... recovery coaches were not a position we had in the system, that was new. So we hired recovery coaches, we hired a manager of behavioral health" (City Health ID2, hospital participant, clinical role) "I feel like that's a big thing because some of the people that work at the facilities that I have called, "Oh, yeah. I used to work with you. I didn't know you worked there," so it builds those bonds. I don't know how else to explain it, but you know what I mean? "Okay. Now I can make it easy on all of us. You know me, I know you." I think getting to know those people on that face-to-face makes a difference." (Lake Health ID 4, hospital participant, clinical role) "But again, I think the long term relationship and very, very low turnover in the key positions is what continues to allow that collaboration to occur." (Lake Health ID 1, hospital participant, clinical role)
Facilitator: Organizing Partnership Processes	"I think a lot of it has came down to just building really good relationships with partners and then staying close with them, being able to check in regularly, particularly about implementing new projects, how things are going, what's challenging, those sorts of things.” (City Health ID1, hospital participant, administrative role) “We tend to be that connection for all of them. So what we'll find is if agency A is not able to get into agency B with a client, they're like, "Man, we're really struggling," we'll get a phone call, it says, "Hey Jamie, can you help me?" And we play that mediator role. My executive director and I both are really... We have no problem leveraging those relationships if it's going to get people the services they need and the help they need." (Lake Health ID5, CBO participant, leadership role) "But we bring that data with us...And so when the hospital first says, well, we've got social workers, we're taking care of this. Well, they're really not okay. And we can really provide some data to show that some of these individuals are coming back several times, and the data is available." (City Health ID 5, CBO participant, leadership role)

Table 2.

Comparison of Study Themes Discussed by Hospital and CBO Participants

Theme	# of Hospital Participants (n=21)	# of CBO participants (n=5)
Barrier: Policy and Philosophical Differences in Treating Substance use	7 (33%)	2 (40%)
Barrier: Communication	8 (38%)	1 (20%)
Barrier: Limited trust	4 (19%)	1 (20%)
Barrier: Lack of Connectivity Between Systems	3 (14%)	2 (40%)
Facilitator: Building Partnership Quality	9 (42%)	4 (80%)
Facilitator: Strategic hires and staffing	8 (38%)	3 (60%)
Facilitator: Organizing Partnership Processes	11 (52%)	3 (60%)

Barriers

Barrier: Policy and Philosophical Differences in Treating Substance Use

Several participants described how policy or philosophical differences in treating substance use between their own organization and an external organization created wariness toward forming partnerships and ultimately limited care continuity. These included policies surrounding the timing of urine toxicology testing, distributing naloxone, and the types of SUD treatment offered. As described by one participant:

"That's the thing about residential programs. Some of them will allow, some of them don't allow, a lot of them don't allow patients to carry naloxone. We always talk to our patients about Narcan, and understanding to make sure we bring it up when they're leaving the treatment program, because the treatment program has not let them pick it up the whole time while they were there." (City Health ID 2, Hospital participant, clinical role)

Other participants described how the policies surrounding MOUD dosing were more restrictive, or high barrier, at a partner organization. Another participant described hesitancy about linking patients to an organization that did not require counseling in addition to MOUD, which the participant perceived as potentially detrimental to patients. A common philosophical difference among participants included the perception that their own organization was more patient-centered than external organizations, or the belief that external organizations were more bottom-line oriented. For example, interviewees based at organizations that did not penalize patients with SUD for missed appointments expressed wariness around linking patients to treatment organizations that did not offer the same flexibility:

"I just think that at the core we are different organizations, different models of organizations at the very core and our decision making. When it comes down to it, I think we can be a little bit more patient-centered based on the fact that we are a public safety and a hospital. And when you have these nonprofits and these corporations, they may be a little bit more bottom line minded and maybe a little bit less flexible and accommodating than we are." (Metro Health ID 1, clinical role)

Different approaches to SUD care, in other words, limited successful partnerships between safety net hospitals and community-based organizations.

Barrier: Communication

Communication was a common barrier to enacting successful partnerships. Prior to beginning a partnership, communication barriers included a lack of awareness of partner activities or services offered, as well as poor understanding of specific staff roles, such as recovery coaches, which impacted the willingness of both parties to forge a partnership. As noted by a participant:

"I think that's something I've learned in this role over the years is that you can't do any of this in a vacuum, you really have to make sure that you're intentional with making sure that partners and people who share this space with you are aware of what you do and have opportunity to give input and work with you. And that's hard work, it's not easy." (City Health ID1, hospital participant, administrative role)

Confusion about roles was exacerbated by perceptions that certain types of organizations, such as MOUD treatment centers, are siloed, with limited engagement from other entities, and have different hours of operation. In particular, interviewees from hospitals noted a preference for partners with less standard operating hours, given that patients may benefit from a linkage at all hours of the day. Once a partnership was initiated, communication issues often stemmed from the person facilitating the communication; for example, if leadership at an organization was the main point of contact, there was a possibility that critical information may not cascade to frontline employees tasked with organizing the linkage process. Similarly, if the main point of contact left the organization, or if a change in hospital leadership led to a shift in priorities, there was a possibility that communication would falter and the partnership would cease to exist.

Interviewees noted that there was at times also a disconnect when there was an overabundance of individuals attempting to develop a partnership, with different sides confused about the state of the partnership and partners’ activities. As described by a hospital participant:

"I feel like there's not a whole lot of collaboration. So I feel like there's too many chefs in the kitchen, if that makes sense. A lot of folks are trying to address the problem and how it impacts their specific area of care, but the left hand doesn't always know what the right hand's doing. So really I feel like that's the biggest disconnect." (Lake Health ID1, hospital participant, clinical role)

Barrier: Limited Trust

Limited trust also challenged the development of organizational partnerships. Some hospital participants perceived that external organizations offered lower quality services and did not believe that advertisements of services were a reliable source of information. Interviewees from hospital settings also noted that CBOs often preferred to speak with patients rather than care coordinators, which was based on the perception that staff would not ensure that patients complete the transition. As described by an interviewee:

"Because I think with agencies in our area, they don't trust that a patient will show up or follow through unless that patient calls themselves. So a lot of them are leery about working with me directly." (Lake Health ID2, hospital participant, non-clinical support role)

Once a patient successfully transitioned to a CBO, interviewees from hospitals discussed how CBO staff were often unwilling to discuss the patient’s status or progress. An interviewee from a CBO also noted that their organization preferred to place their own staff in hospital EDs to facilitate linkages, but that many hospitals were hesitant to share their space with staff from an external organization, which served as a barrier to partnerships.

Barrier: Lack of Connectivity Between Systems

An inability to share data or information systems was another common barrier discussed by interviewees. For example, hospital staff was unable to track whether patients were receiving ongoing care at external facilities, or continue to treat patients that had been initiated on treatment in the hospital. For example, in response to a question surrounding the difficulty establishing transitions partnership a participant noted:

"The data sharing, very much. Sharing that patient information. And then the follow-up, sharing afterwards. We've done it, we've sent them out and then nobody wants to share follow up information. Even just numbers, basic numbers." (Lake Health ID5, CBO participant, leadership role)

Participants also discussed how they lacked the measures and data needed to develop a successful transitions program, and that the measures tracked at one organization often differed from those collected and reported at external organizations; this created challenges related to assessing patients' needs or progress.

Facilitators

Facilitator: Building Partnership Quality

Facilitators that enabled interview participants to enhance the quality of their partnerships addressed barriers within the domains of ‘policy and philosophical differences in treating substance use,’ ‘communication,’ and ‘trust issues.’ To build trust and improve communications, participants discussed how establishing a strong and natural rapport enabled partners to easily address differences as they arise, as well as coordinate smoother patient transitions. As described by an interviewee:

“We actively went after other agencies to find, to build a rapport with them, so that we could get our patients in, we got inpatient, outpatient, but we needed to get them in quickly. And now, because we built such a good rapport, we'll call, "Hey this is X or this is Y from [organization name]." And they're like, "Oh yeah, well, wait, no. I don't think we have any beds. Well, let me check. We'll call you." And usually we can get people in within the day, the same day, or the next day. So that building that rapport with your community and the patients." (Mountain Health ID1, hospital participant, non clinical support role)

To overcome policy and philosophical differences, several interviewees described actively working to find mutual ground on disagreements, rather than expecting a partner to adopt a different approach or change policy. For example, for one organization, this entailed creating an agreement to allow their providers to continue titrating patients on MOUD via telemedicine at a partner facility with different dosing policies. Finding mutual ground also supported the quality of relationships, along with framing partnerships as opportunities to support partner programs and fill in service gaps, rather than supplant or hinder their activities. For one organization, this entailed asking partners early on what services they are missing, and demonstrating how their own organization can provide these services or enhance partners’ capacity in other ways. As noted by a participant:

"It's really challenging initially to be embedded in the hospital. If you don't have a relationship, you have to spend a lot of time talking to the hospital. And again, this is going back, about what the benefits are and how it could relieve some of the stress on their social workers." (City Health ID 5, CBO participant, leadership role)

Finally, to overcome partners’ initial wariness towards external organizations, and to foster trust and quality within the partnership, participants described the importance of demonstrating positive outcomes and high-quality work to potential partners, particularly during the early stage of partnership development. As described by an interviewee:

“So the more my patients actually showed up, I think they trusted me a little more." (Lake Health ID 2, hospital participant, clinical role)

Facilitator: Strategic Hires and Staffing

Interviewees discussed facilitators relating to hiring decisions and staff that enabled them to overcome barriers related to trust and communication. Facilitators to improve communication included hiring people specifically to develop partnerships and facilitate linkages, which could be a time-consuming task, and identifying the appropriate point person at a partner organization with whom to initiate a partnership, ideally a staff member in a position with low turnover. As described by an interview participant:

“You can know who the director is. But if you want to get Mary or Johnny into the program, it's good to know who the impact person is to expedite the process rather than just looking at a name on a directory, which would be my name, but I may not be the best person to call. You might need the person down the hall. So it's about maintaining those relationships that makes the hand-off more friendly and quicker." (Metro Health ID 2, hospital participant, clinical role)

Facilitators to enhance trust included cross-hiring staff from potential partner organizations to build familiarity, and leveraging networks and professional reputations to remove doubts surrounding the trustworthiness or work quality of external organizations.

Facilitator: Organizing Partnership Processes

Facilitators that helped participants organize partnerships with external organizations, as well as implement processes to organize partnership activities, addressed barriers within all four domains. In beginning a partnership, these facilitators included having a mediator bring organizations together and vouch for the trustworthiness of both parties, and establishing partnerships in the context of regional professional organizations or conferences. As noted by an interviewee:

"And the events that we go to, usually there are many, many tables of other facilities, that they get up and they say, "Okay, this is what we offer." And X and I sit there in the background saying, "That would be a great place." So we go to that table, we talk to those people, we get their cards, we get their information." (Mountain Health ID1, hospital participant, non clinical support role)

Interviewees also discussed the importance of visiting partner organizations in-person prior to engaging in the care transition processes to promote familiarity, as well as to ensure the quality of external organizations. Participants from one CBO described using data on their previous successes to demonstrate to partners how they can help them improve their patient engagement rates, and overcome initial mistrust. In maintaining a partnership, participants noted that having standing meetings with partner organizations ensured that communication did not depend on a singular staff member, as well as promoted quality by allowing partners to resolve operational challenges on a regular basis.

Finally, to address barriers related to interoperable systems, participants from two organizations discussed developing initiatives to support data transfer between organizations; one health system had successfully linked their electronic medical record with an external organization, while the second health system was in the process of getting a data sharing agreement in place. Other facilitators included having formal, business contracts with partner organizations in place, to ensure that partners work closely together in developing care plans for patients.

DISCUSSION

Despite the recent emphasis on developing hospital-based TOPs with the goal of improving patient access to ongoing SUD treatment, the process of establishing partnerships between hospitals and CBOs is poorly understood and rarely documented. Understanding barriers and facilitators to creating these partnerships across sectors and developing recommendations of what works could improve patient transition processes for hospitals both with and without established TOPs. The barriers identified in this study - ranging from divergent policies and philosophies regarding SUD treatment, poor or inadequate communication between partner organizations, distrust, and an inability to share data or information systems - highlight the complexities inherent in bridging partnerships across differing entities. Our results also demonstrate that to overcome these barriers, it is critical for partnerships to be based in collaborative work. Purposeful actions must be taken to ensure that interactions are organized and outcomes-driven, with the appropriate stakeholders from each organization engaged in partnership development. These findings align with existing literature on healthcare and CBO partnerships,^24,25 and apply it specifically in the context of SUD care transition.

The findings of this study are also consistent with national policy efforts to enhance cross-sector collaboration to address the SUD and opioid crisis. For example, in 2018, the National Academy of Medicine (NAM) developed the Action Collaborative on Combating Substance Use and Opioid Crisis initiative to bring together key stakeholders across multiple sectors to collaboratively develop solutions to decrease overdose death rates and increase access to SUD treatment.²⁶ As part of this initiative, NAM convened a working group which identified four overarching guidelines to improve the SUD treatment ecosystem: Capacity, Competency, Consistency, and Compensation. Capacity refers to whether an organization or system is appropriately sized and resourced to address the needs of the community they serve. Competency pertains to the training, preparation, and assessment of individuals employed within the treatment framework. Consistency refers to the quality of care delivered by health systems, and Compensation describes whether the treatment system's financial structure incentivizes adherence to best practices through reimbursement.²⁶ These principles resonate with the facilitators identified in our study, emphasizing the importance of building quality partnerships, strategic hires, and organizing partnership processes.

Particularly for health systems with limited resources, such as safety-net hospitals,²⁷ developing effective partnerships may be an important strategy to enhance their capacity, competency, consistency, and compensation related to delivering SUD treatment. For example, multiple participants in this study described how framing partnerships as opportunities to fill in service gaps and provide mutual support for partner organizations helped them overcome barriers related to lack of trust and policy/philosophical differences; partnerships formed for this purpose can also be viewed as enhancing the capacity and consistency of partner organizations. Similarly, cross-hiring staff from partner organizations, as well as hiring staff specifically to facilitate transition partnerships, can support competency between organizations, in addition to fostering trust and communication. Finally, developing formal, business partnerships between organizations can help ensure that goals, standards, and incentives related to patient transitions are aligned between partner organizations.

Other challenges highlighted in our study, such as system connectivity, resonates with the broader issue of interoperability in healthcare IT systems. Antiquated data systems have been a persistent challenge for safety-net organizations; for example, studies have found that in the years following Medicaid expansion, safety-net hospitals implemented EHRs at a slower rate compared with their non-safety-net counterparts.²⁸ Similarly, a lack of available resources continues to hinder the implementation of new programs and measures, as well as the adoption of innovations, within safety-net hospitals.²⁹ As such given the myriad of challenges facing safety-net health systems, the implementation of measures and data-sharing initiatives to support care transitions for SUD patients may not be a priority in many safety-net settings. Yet, hospital and community partnerships have been found to be successful in decreasing hospital readmissions,^30,31 lowering costs, and enhancing quality of care.³² Given that data sharing and transparency have been cited as key precursors to the development of partnerships between hospitals and external organizations,³³ safety-net settings attempting to demonstrate improvements in publicly-reported metrics should consider strategies focused on enhancing data collection and sharing with CBOs, particularly given that US hospitals are responsible for billions in non-reimbursed SUD-related medical costs each year.^34,35

This study has a number of limitations. First, we used a qualitative study design which allows us to thoroughly examine barriers and facilitators to partnerships among study participants, but limits the generalizability of our results to other settings. Furthermore, while the sites included in our study were diverse in regards to geographical location, size, and development of TOPs, future studies are necessary to assess whether the barriers and facilitators identified persist within a greater range of organizations. We also only interviewed staff and providers within these organizations and do not include the perspective of patients, who are the primary targets of transition partnerships. Finally, the CBO participants in this study were suggested by their health system/hospital partners, and their inclusion is subject to selection bias and may not represent the full spectrum of institutions where patients can receive ongoing SUD treatment. Additionally, while the hospitals suggested CBOs to contact, they did not always make direct connections, and one hospital did not suggest any partner organizations after multiple prompts; as such, the study team faced challenges establishing contact with several CBOs, and their viewpoint is underrepresented in this analysis which may limit the generalizability of our findings. However, it is notable that each barrier and facilitator in our study was discussed by participants from both CBOs and health systems, which is a strength of our analysis.

Conclusions

By focusing on a relatively unaddressed aspect of SUD patient transitions– partnerships between hospitals and CBOs–this study is an important contribution to the literature on hospital-based TOPs. In particular, we demonstrate that while multiple barriers to developing hospital-CBO partnerships exist, stakeholders can adopt strategies that mitigate these challenges, such as cross-hiring between organizations and focusing on mutually beneficial services, even within resource-limited safety-net settings. Policymakers and other health system leaders who wish to optimize TOPs in their facilities should focus on developing implementation strategies to barriers unaddressed by existing facilitators, such as inadequate data collection and sharing systems. Future studies should encompass the perspectives of patients and a wider range of CBOs, to better understand their support for patients transitioning from acute care settings such as hospitals. Future research must also assess whether adopting implementation strategies, informed by these determinants, is associated with improved rates of successful transitions for SUD patients between facilities.

Supplementary Information

Below is the link to the electronic supplementary material.

FUNDING

This study was supported by the National Institutes of Health, National Institute on Drug Abuse (grant R34DA055228).

DECLARATIONS:

Conflict of Interests:

The authors have no conflicts of interest to report.