Organizational Capacity and Readiness to Provide Medication for Individuals with Co-Occurring Alcohol Use Disorders in Public Mental Health Settings

Abstract

Purpose

Alcohol use disorders (AUD) in individuals with mental illness are largely untreated. The purpose of this study was to identify gaps in organizational capacity and readiness to provide medications for AUD in outpatient public mental health clinics.

Methods

We selected a purposive sample of eight publicly funded outpatient mental health clinics operated by the Los Angeles County Department of Mental Health; clinics were chosen to maximize heterogeneity. Guided by theories of organizational capacity and readiness and research on the adoption of pharmacotherapy for AUD in primary and specialty care treatment settings, we conducted semi-structured interviews and focus groups with administrators, providers and staff, and a qualitative analysis of the results.

Results

Respondents described significant organizational capacity and behavioral readiness constraints to providing medication treatment for AUD. Both groups articulated a perception that mental health clinics were not designed to provide co-occurring AUD treatment because of large caseloads, staffing configurations, and time constraints that did not support the delivery of appropriate treatment, and a lack of protocols and workflow procedures.

Conclusions

We documented organizational capacity and readiness constraints which impede the delivery of medication treatment for AUD in a large mental health system. While some constraints have straightforward solutions, others require structural changes to the way care is delivered, and state-level funding and policy changes.

Co-occurring alcohol and other substance-use disorders (COD) are prevalent and can have devastating consequences for people with serious mental illness (SMI). Among the 8.2 million adults with a SMI, nearly 40%, or 3.2 million adults, have a current substance use disorder (Substance Abuse and Mental Health Services Administration, 2019a), with alcohol use being the most commonly reported COD (Kamal, 2017; Substance Abuse and Mental Health Services Administration, 2017). Alcohol use disorders (AUD) among people with SMI increase morbidity and mortality (Dickey, Dembling, Azeni, & Normand, 2004; Maynard, Cox, Hall, Krupski, & Stark, 2004; Yoon, Chen, Yi, & Moss, 2011) and are associated with higher healthcare and social costs (Chang et al., 2015; Clark, Samnaliev, & McGovern, 2009; Ding et al., 2011; Heslin, Elixhauser, & Steiner, 2006), homelessness, and incarceration (Hawthorne et al., 2012; McNiel, Binder, & Robinson, 2005; Rosenberg et al., 2001). They are also associated with worse treatment outcomes (Lacro, Dunn, Dolder, Leckband, & Jeste, 2002).

Despite the availability of effective treatments, substance use in the SMI population largely goes untreated (Harris & Edlund, 2005; Substance Abuse and Mental Health Services Administration, 2019a; Watkins, Burnam, Kung, & Paddock, 2001). Estimates are that fewer than 13.6% of adults with COD receive any substance use treatment (Substance Abuse and Mental Health Services Administration, 2019a) – treatment that could significantly improve outcomes. National surveys indicate that when individuals with COD access behavioral health care, it is primarily through the mental health care system. In 2018, 55.9% of adults with COD accessed mental health care only, while 2.7% accessed substance use disorder treatment only. Fewer than 10.9% received treatment for both their mental health and substance use disorders (Substance Abuse and Mental Health Services Administration, 2019a).

There are several reasons why individuals with SMI do not receive substance use treatment. Mental health and substance use treatment in the United States have historically taken place under separate systems of care, posing barriers for the present-day integration of services (Burnam & Watkins, 2006; Timko, Dixon, & Moos, 2005). In the 1970s, rather than integrate the delivery of substance use treatment into the mental health care system, the federal government created a separately funded and administered substance use treatment system. While this increased the availability of substance use disorder treatment, it formalized the separation of substance use and mental health treatment. The resulting differences in Medicaid reimbursement, quality assurance standards, licensing requirements, funding sources, and staff training requirements have made it difficult to provide integrated care and for individuals to access both types of care within a single system. Mental health and substance use disorder treatment also have different philosophical underpinnings. Traditional substance use treatment is based on an abstinence model (Aletraris, Edmond, Paino, Fields, & Roman, 2016) that emphasizes group treatment and counselling, in contrast to the central role of pharmacotherapy treatment in mental health. Although there have been recent efforts to integrate care treatment services are still siloed (McGovern, Lambert-Harris, Gotham, Claus, & Xie, 2014; Padwa, Guerrero, Braslow, & Fenwick, 2015; Spivak et al., 2020).

The development of effective pharmacotherapies for AUDs that can be used in individuals with COD has made understanding and addressing barriers to integration particularly timely, as stand-alone psychosocial treatments for COD are resource-intensive, difficult to implement and sustain, and only modestly effective (Drake, O'Neal, & Wallach, 2008; Horsfall, Cleary, Hunt, & Walter, 2009). There are currently four medications for the treatment of alcohol use disorders that have been approved by the U.S. Food and Drug Administration: oral naltrexone, acamprosate, extended-release injectable naltrexone and disulfiram.

Medication treatment for AUD (MAUD) has demonstrated effectiveness and cost savings for individuals with AUD or co-occurring AUD (Center for Medicaid and CHIP Services, 2014; Schackman, Leff, Polsky, Moore, & Fiellin, 2012), yet MAUD has not been widely adopted, in either primary care or specialty treatment settings (Harris, DeVries, & Dimidjian, 2004; Jonas et al., 2014; Roman, Abraham, & Knudsen, 2011; Substance Abuse and Mental Health Services Administration, 2019b). Barriers to adoption include the 12-step approach as the primary treatment model, resource constraints and lack of access to prescribers in specialty substance use treatment settings. MAUD may not be appropriate for all individuals with a co-occurring mental health condition but based on the American Society of Addiction Medicine Standards of Care, it should be available to those patients for whom it is indicated and acceptable (American Society of Addiction Medicine, 2014). It is unknown what factors influence the availability of MAUD in the public mental health systems.

Organizational theory provides a framework for understanding implementation of new healthcare practices (Holt, Helfrich, Hall, & Weiner, 2010; Meyer, Davis, & Mays, 2012; Weiner, Amick, & Lee, 2008), such as the adoption of MAUD in public mental health care settings. Organizational capacity refers to the resources of an organization to implement a change in practice; whereas, organizational readiness refers to the willingness and ability of an organization’s members to institute the change (Weiner et al., 2008). Although the concepts of capacity and readiness overlap, previous research has treated them as separate constructs. Meyer et. al. propose that organizational capacity can be evaluated in terms of seven constructs, which provide a theoretical framework for evaluating the potential for change: fiscal and economic resources, workforce, physical infrastructure, inter-organizational relationships, data and information technology, policies and procedures, and organizational culture and leadership (Meyer et al., 2012). Organizational readiness theories are relevant to the workforce and organizational culture and leadership constructs as they help explicate whether an organization’s workforce will be able to implement a practice change. However, while readiness can be seen as a component of capacity, we treat the two as separate although related constructs because of the extensive work done to develop and test theories of readiness, which has been done outside of the literature on organizational capacity (Weiner et al., 2008).

Previous research suggests that capacity and readiness are both essential precursors to successful implementation of changes to care in public health systems (Brown, LaFond, & Macintyre, 2001; Brownson, Fielding, & Green, 2018; Holt et al., 2010; Meyer et al., 2012). In mental health and substance use treatment programs, concepts related to organizational capacity and readiness are associated with providers’ willingness to adopt evidence-based practices and implementation (Aarons & Sawitzky, 2006; Bonham, Sommerfeld, Willging, & Aarons, 2014; Jacobson et al., 2020; Saldana, Chapman, Henggeler, & Rowland, 2007; Thomas, Wallack, Lee, McCarty, & Swift, 2003) and prior work has also found organizational readiness to influence the perceived appropriateness of integrating SUD treatment in primary care settings (Ober et al., 2017) and the likelihood of SUD programs coordinating care with external organizations (Guerrero, Aarons, & Palinkas, 2014). To date, however, there has been little analysis of organizational capacity and readiness in the context of MAUD adoption in mental health clinics.

Because of the large role of the mental health system in treating individuals with COD, we applied theories of organizational capacity and readiness to guide our exploration of factors related to MAUD adoption in public outpatient mental health settings. The purpose of the study was to identify the gaps in organizational capacity and readiness to provide MAUD in these settings in order to develop a toolkit that could be used by public mental health systems to address these constraints.

Methods

Study Context

The study was conducted in partnership with the Los Angeles County Department of Mental Health (LAC DMH) which provides care to over 250,000 individuals yearly and is the largest county-operated mental health organization in the United States. The study was motivated in part because in late 2017 the LAC DMH changed its policy to permit naltrexone long-acting injection as a pharmacy benefit, and a pharmacy bulletin was sent to all providers letting them know of its availability as a treatment for co-occurring AUD (County of Los Angeles, 2017).

Compliance with Ethical Standards

All study procedures were approved by both the research organization’s and LAC DMH’s institutional review boards. No study investigators reported any potential conflicts of interest. All study subjects provided documented oral consent.

Study Design and Sampling Strategy

We selected a purposive sample of eight of the 25 publicly funded outpatient mental health clinics directly operated by the LAC DMH. The county is divided into eight service planning areas and covers nearly 5,000 square miles. One clinic was selected from each service area; clinics were chosen to maximize heterogeneity and included large, medium, semi-rural, suburban and urban sites; and served clients with a broad range of races and ethnicities. Six of the clinics are considered “large” (>2500 active clients); one of these large clinics was rural, and one was suburban. Two clinics were medium-sized (1000-2500 active clients); both of these were suburban. One clinic had a large Asian population, and one had a primarily African-American/Latino population, and one was primarily Latino. Among all LAC DMH clients, Latinos are the largest ethnic group at 48.8%, followed by Whites at 26.7%, Asian Pacific Islanders (API) at 14.0%, African Americans at 8.1%, and Native Americans at 0.19%. Total population with two or more races was 2.2% (Sherin, 2018). In 2018, of the 24,267 adults initiating mental health care at one of the eight participating clinics, 1,575 had a documented co-occurring AUD diagnosis of which 254 received a prescription for AUD pharmacotherapy, indicating the potential for substantial unmet need (Watkins, 2019).

To understand gaps in capacity and readiness, we conducted semi-structured interviews with key administrators and focus groups with medical providers and other non-medical staff. Broadly, our intent was that the interviews would focus on obtaining factual information about the organizations capacity to deliver MAUD and focus groups to solicit perceptions of this capacity. Because we were eliciting primarily factual information, we felt that administrator interviews would be a better vehicle to elicit such information than group discussions. The purpose of the focus groups was to solicit perceptions about the organization’s readiness to deliver MAUD care and potential ways clinics could improve delivery. We used focus groups for this purpose so that we could feasibly gather information from staff with a diverse range or roles, responsibilities, expertise and experience in these settings. We also wanted staff to be able to discuss the topic with one another so that we had a better understanding of the range in perceptions and experiences across each participating clinic.

Data Collection Procedures

Semi-structured Interviews with Key Administrators.

The program director of each clinic was asked to provide the name and contact information of 6-8 key administrators, including the supervising psychiatrist, supervising psychologist and head of triage. We contacted them by email and scheduled a telephone interview at a time convenient for the interviewee. Two trained interviewers (CC, MM) conducted telephone interviews between early June through August 2018. Interviews were audio recorded but not transcribed. Detailed field notes were collected, and the recordings were referred to for clarification. The sample size is consistent with recommendations for achieving thematic saturation in formative qualitative research with health care providers (Sandelowski, 1995). We did not provide incentives because answering questions about the capacity for treatment services is within the interviewee’s scope of work and the clinics considered this to be part of quality improvement.

Focus Groups with Providers.

At a staff meeting at each of the eight clinics, we described the study and asked providers to sign up to participate in a focus group—one for medical providers (i.e., psychiatrists, nurse practitioners who can prescribe medication) and one for non-medical providers (i.e., clinicians and other direct service staff who do not prescribe medication, such as social workers). Program Directors were then asked to identify a day and time to conduct one-hour discussion groups. If an insufficient number of people signed up, we asked the program direct to provide email addresses of others who might want to participate and we invited them to the group. Before each discussion group began, participants gave verbal consent for participation. Each group was attended by a facilitator and notetaker; groups were audio recorded and recordings were transcribed. We did not provide financial incentives but did provide refreshments at each group.

Interview and Focus Group Protocols

The development of the semi-structured interview and focus group protocols was informed by research on MAUD adoption in primary care and specialty treatment settings (Gueorguieva et al., 2010; Ober et al., 2017; Storholm et al., 2017; Substance Abuse and Mental Health Services Administration, 2009; Urada, Teruya, Gelberg, & Rawson, 2014; van Boekel, Brouwers, van Weeghel, & Garretsen, 2014; West et al., 1999) as well as by the concepts of organizational capacity and readiness. Table 1 shows seven organizational capacity constructs identified by Meyer et al. (2012) and how the construct was applied to MAUD; the table also provides examples of questions asked during administrative interviews.

Table 1.

Organizational Capacity & Readiness Constructs, Dimensions, and Example Questions

Construct (Meyer et al., 2012) 1	MAUD-Relevant Dimension	Interview Example Questions
Fiscal and Economic	Funding mechanism Financial resources to support delivery of co-occurring AUD care and MAUD	How do you bill for AUD medication management visits? What are the billing mechanisms for specific pharmacotherapies? (I) Can you bill for behavioral services for co-occurring AUD? (I)
Workforce Capacity and Readiness	Availability of providers with specific expertise (addiction psychiatry certification/ substance use counsellors) Co-occurring AUD services provided Staff knowledge/expertise Fit of delivering MAUD care within current practices; appropriateness and acceptability Motivation Perceived benefit to patients and providers Existence of MAUD “Champions”	How knowledgeable are the clinicians and staff in this clinic/do you feel about MAUD? (I) How many prescribers are certified in addiction medicine and/or addiction psychiatry? (I) What is the availability of substance use counselors to conduct behavioral counselling? (I)
Physical Infrastructure	Refrigerator space (for medication), pharmacy access, waiting room space Appointment time structure/flexibility	What is the process for initiating a patient on long-acting injectable naltrexone? Acamprosate? (I) What is your capacity for point of care urine drug testing? (I)
Inter-Organizational Relationships	Relationships with specialty substance use providers and inpatient detoxification facilities Access to expert MAUD consultation	What is your relationship with specialty substance use treatment providers and inpatient detoxification facilities? (I) What is your referral process for accessing specialty substance use treatment? (I)
Data and Information Technology (IT) Resources	Capacity of electronic information systems to record and track co-occurring diagnoses and treatment information; configuration flexibility to accommodate outcomes monitoring Data quality, reporting capability	How useful is the electronic health record in the management of co-occurring AUD? (I)
Policies and Procedures2	Availability of policies and procedures covering the population and treatments provided	Are there co-occurring AUD and MAUD treatment protocols and policies? What are they? (I)
Organizational Culture & Leadership	Treatment of co-occurring AUD and provision of MAUD as consistent with organizational mission Climate Leadership buy-in Training resources	How does treating clients with co-occurring AUD and providing MAUD fit within organization’s mission and goals? (I) What resources are available for training and consultation? (I)

Note:

^(I)Refers to question asked in the interview.

Focus group questions were similar to those asked of administrators but were mainly open-ended. All questions aimed to elicit information about the organization’s and providers’ capacity and readiness to deliver care, including MAUD, to clients with co-occurring alcohol use disorders (e.g., “What are your thoughts about providing care for clients with co-occurring alcohol use disorders at this clinic”, “Tell me your thoughts about how providing care for clients with co-occurring alcohol use disorders fits into your current work practices?”, “How well do you think this clinic could provide care for co-occurring alcohol use disorders?”, “What resources or supports do you think are needed?”, “What specific challenges or barriers do you anticipate implementing co-occurring alcohol use disorder care at this clinic?”). We used the Weiner et al. (2008) concept of organizational readiness to further explicate the dimension of workforce and leadership readiness, assessing both psychological and behavioral readiness (e.g., “How confident are you about your ability to provide care for co-occurring alcohol use disorders at this clinic?”, “Do you think you and your colleagues are adequately prepared to successfully implement it?”, “What do you think about leadership support for care for co-occurring alcohol use disorders at DMH?”).

Data Analysis

Interview Data.

Interview data were organized by interview guide question in an excel document. Each column was a question, which as per the interview guide, were ordered by organizational capacity domain. Each row corresponded to a single interview. The two researchers who had conducted interviews summarized the organizational capacity limitations to MAUD care articulated in each interview and clinical site (CC, MM), and then met with a third researcher (KW) to develop an overall narrative synthesis of each identified capacity issue. In the process of developing this narrative synthesis, themes often corresponded directly to the domains, but identified problems occasionally overlapped across domains. In these cases, the researchers reexamined respondent comments throughout the interview to decide how best to group the problems and organize resulting themes. Inconsistencies were discussed and resolved. We used a table to summarize results from the interviews.

Focus Groups.

A researcher (SBH) and notetaker, who facilitated more than 90% of the focus groups, reviewed several transcripts and agreed that the majority of responses aligned with the seven Meyer et al. (2012) constructs. Next, an excel document was used to organize the transcript content by these constructs. First, two transcripts were reviewed and coded independently by two members of the research team. The team reviewed each other’s codes to assess consensus. Any inconsistencies were discussed and resolved. Next, the two coders split up the remaining transcripts and coded using similar strategy. Next, the transcripts were reviewed to select illustrative quotes for each capacity construct where participants from more than four sites and discussion groups mentioned and were consistent with the constructs identified from the administrator interviews. Because we had quotes from the focus groups, we present these results in a narrative format.

Results

Administrator Interviews

We received 53 names of administrators from the program managers across the eight sites; 39 agreed to participate (3-8 interviewees per site, overall participation rate 74%). Interviewees held the role of Mental Health Clinical Supervisors (n=27), seven of whom were also licensed social workers. Interviewees also included Supervising Psychiatrists (n=6), Program Heads/Managers (n=3), Mental Health Nurse Counselors (n=2) and Administrative Supervisor (n=1).

Table 2 shows a narrative synthesis of the organizational capacity constraints identified by the administrators, organized by organizational capacity dimension. The number of clinics from which at least one administrator identified the problem are noted in parentheses. Problems related to workforce and human resources, and the lack of standardized AUD workflows and treatment guidelines were identified by administrators at eight out of eight clinics.

Table 2.

Organizational Capacity & Readiness Themes: Administrator Interviews (N=39)

Organizational Capacity/Readiness Construct	Theme (Number of clinics)
Fiscal and Economic	Billing issues. Medi-Cal does not cover substance use treatment and will deny reimbursement for care unless progress note clearly documents how substance use is affecting mental health and how treatment of the co-occurring AUD improves the mental health condition (5)
Workforce	Lack of knowledge/expertise/self-efficacy. Lack of knowledge about co-occurring AUD treatment, including pharmacotherapy. Clinics do not have access to addiction medicine expertise or have a provider champion (8) Poor fit with current practice – inadequate staffing/time. Mental health clinics are not designed to provide AUD care. Caseloads are high and can’t provide all the necessary treatment components (7) Poor fit with current practice – lack of appropriate staffing/expertise. Staff are apprehensive about treating something that isn’t primarily mental health; primary care providers should provide this treatment (5)
Physical Infrastructure	Internal resource constraints. Difficulty obtaining urine drug screens (4)
Interorganizational Relationships	External resource constraints. Scarcity of inpatient beds for residential treatment or withdrawal management (6)
Information Technology	Communication barriers. No easy way for different providers to see the results of AUD screening and assessment, diagnosis, and treatment provided (3) Electronic medical record limitations. Difficult to record co-occurring AUD diagnosis or add to the problem list (2) Electronic medical record limitations. No way to identify and track the population of patients with co-occurring AUD (2)
Policies and Procedures3	Lack of MAUD workflow protocol. No standardized MAUD workflow or treatment guidelines (8); where protocols existed such as for screening, administrators were confident they were being followed.
Organizational Culture and Leadership	Lack of knowledge. Lack of awareness of both internal and external specialty substance use disorder treatment availability and resources, or awareness of resources is siloed (7)

Focus Groups

Fifty-two medical providers, including psychiatrists, nurse practitioners, and registered nurses participated in one of the eight focus groups (mean per site was 6.5 with range of 3-9) and 72 nonmedical providers, including psychiatric social workers, clinical psychologists, case managers, and community workers participated in one of eight focus groups for a mean number of nine staff per site with range of 6-12. The capacity challenges identified by staff in the focus groups echoed challenges identified in the administrator interviews.

Fiscal and Economic Resources.

Staff from four sites articulated concerns about billing for substance use disorders in mental health settings. For example, “the biggest barrier is billing because we have to address the mental health, and substance abuse disorders are mental health impairments, but, you know, we have to go by this list of Medi-Cal approved— You know. Substance abuse is not one of them. That's like a secondary type of thing we would address” (Site 1, non-medical provider group). A staff member from a different clinic noted, “We're not a drug Medi-Cal program, so we can't bill specifically for just a substance use disorder. So, every time we treat a substance use disorder it's to minimize their psychiatric symptoms, not to minimize substance use disorder, although, obviously, they fall hand in hand” (Site 3, medical provider group).

Workforce.

Participants in all eight sites indicated that AUDs were prevalent among their patient population and overwhelmingly agreed that patients could benefit from AUD treatment. However, some staff were apprehensive about treating something that wasn’t considered primarily a mental health issue. They also noted that some staff had a lack of knowledge and expertise regarding how to treat COD. For example, a respondent mentioned, “I have a lot of clients who really struggle with alcohol use and other mental illness and I just sometimes don't feel prepared or I don't feel like I have the skills” (Site 7, non-medical provider group). Medical providers and non-medical providers thought that more trainings were needed; training topics that were mentioned included understanding alcohol dependency, including alcohol’s effect on the brain and behavior, effective interventions for alcohol use disorder, including medications, and protocols for medication use.

Despite seeing the benefit of integrating MAUD treatment, in different ways staff at every site questioned whether providing MAUD was a good fit with current practice. Lack of time was mentioned by several staff. For example, one respondent said, “It’s like, it requires a lot of time to do substance use. So it’s almost like an extra clinic being added onto this clinic. So, whether we do it, it just depends on how many people are added on to handle it” (Site 1, medical provider group). Lack of time was perceived as limiting the ability to provide quality care. For example, one clinician said, “So if you have a client like that and you're infrequently seeing them, let's say six to eight times a year, and they're struggling with an alcohol addiction, that’s really difficult because you're not really being authentic to any specific model of treatment” (Site 7, non-medical provider group). A medical provider responded that “It does take a great deal of time and dedication. Like I would say when I was working in [x location] I was actually running the MAT [medication assisted treatment] program there, and it took a great deal of time to track these folks, ensure that they’re safe, doing well, coming in for their visits. It was not an easy task, and it took a lot more time and a lot more support. So, it is definitely a concern that we would not do them justice as far as offering them the appropriate level of care that they need” (Site 1, medical providers).

Respondents also suggested that a staff configuration or composition with different expertise might be needed to address alcohol use disorders effectively. One respondent mentioned “I think most psychiatrists aren’t even really necessarily trained to determine the level of care that the substance patient may require. And so I think that in and of itself could be problematic” (Site 1, medical provider group). Other providers said, “I think it would be beneficial to have somebody come in who's got experience and training in substance use disorders, whether that be a psychiatrist who his main focus has been substance use or somebody that's a LCSW or MFT that that's been their focus” (Site 1, non-medical provider group). Respondents mentioned needing more nursing staff as well as psychiatrists with addiction expertise. For example, “Yeah, definitely more nursing staff. I mean, even just training the general staff up front, too. I mean, like what does an emergent situation look like where you need to contact the nurse?”(Site 1, medical provider group). Caseload size was mentioned as a barrier to providing adequate care, “Our caseloads are so high …how would we be able to do that? [It’s] impractical in that setting, it’s setting us up for failure” (Site 1, non-medical provider group). Staff mentioned they did not have enough staff with specialized knowledge, “we have one substance abuse counselor for 3,000 clients, and every time — sort of a management thing — but every time we ask for more substance abuse counselors it doesn't happen” (Site 3, medical provider group).

Physical Infrastructure.

Focus group participants at three sites mentioned physical space constraints that prevented them from providing co-occurring AUD care. More specifically, staff at one clinic mentioned that their site was not well designed to collect urine toxicology samples because the bathroom was in the waiting room. In another site, staff were concerned that there was no sink available to nurses to prepare and administer the medication. Participants from another clinic suggested the need to have more space overall, so that they would have more treatment rooms.

Interorganizational Relationships.

Staff from four sites mentioned the challenges of working with outside entities to provide detoxification services because of the scarcity of those services. This challenge was associated with attitudes that taking care of patients with AUDs was time consuming. For example, “because that is a lot of our time is spent calling detox and detox trying, and then okay, we've got one, but it's all the way out in [x location]. And so now we've got to get a client from [x location] to [x location] today, right now” (Site 5, non-medical provider group). Respondents from five clinics mentioned that the scarcity of alcohol treatment resources also led to long wait times for patients, “it is just more availability is what's needed, you know, a lot of the — for Medi-Cal have waiting lists. And you have a client that may want to go into detox, but, "Okay, well, you have to wait 30 days" (Site 1, non-medical provider group).

Information Technology.

Information technology also was mentioned as a capacity constraint. Respondents from two sites mentioned that there was no formal way to track the diagnosis and treatment plan around AUD, even if it was entered into the system. For example, “I mean, you can put the diagnosis in under the diagnosis list, and it's there. I don't know about tracking. Right?” (Site 3, medical provider group) and “So, it's there and it goes in the diagnosis, and then people track it in their notes, but there is no form or —There's no treatment plan for it” (Site 3, medical provider group).

Policies and Procedures.

Echoing the theme identified in the administrative interviews, respondents in the focus groups from five sites also indicated the need for treatment and workflow protocols. “Yeah, it’s more the person doesn’t know what to do … it’s more like that, like people feel uncomfortable” (Site 5, medical provider group). A respondent from another clinic stated, “I think it’s just putting some of the things that we’re talking about into action, like how do we implement it now? Like, the things that we’re talking about, it would be nice if we had something very specific, like okay, this happened, like you said, you give this medication, this happens, this is what you do. These are the things to look out for, very easy, short and concrete”(Site 4, medical provider group).

Organizational Culture and Leadership.

Staff at five sites suggested it would be helpful if leadership would communicate more clearly and consistently about COD care, and requested consistent information and additional support for providing care. One respondent stated, “[if] everybody gets the same information…It’s not going to be misinterpreted or interpreted differently from one doctor or one nurse to another” (Site 5, medical provider group). Another respondent stated, “Well, they always say it is a priority, but they don’t offer the resources, because they keep saying oh, we’ll have this person do this project, but it’s like three years and there’s no follow through” (Site 4, medical provider group).

Discussion

This study examined administrator and provider perceptions of organizational capacity and readiness constraints to providing MAUD to individuals with co-occurring AUD receiving mental health treatment in a purposive sample of publicly funded mental health clinics in Los Angeles County. Results from the administrator interviews and the focus groups with providers were similar. While administrators and providers generally were willing to provide care for co-occurring AUD, including MAUD, and felt like it was worthwhile, they described significant capacity and readiness constraints. Notably, both groups articulated a perception that mental health clinics were not designed to provide co-occurring AUD treatment and MAUD because of large caseloads, staffing configurations and time constraints that did not support the delivery of MAUD, and a lack of protocols and procedures to set standards indicating a lack of organizational capacity and behavioral readiness constraints. Other key barriers identified were a lack of knowledge surrounding billing Medicaid, documentation processes that did not support communication and team-based care, and inconsistent leadership support. These issues were perceived as substantial barriers to adopting MAUD treatment.

Despite different philosophical underpinnings to the treatment of mental health and substance use disorders and different views as to the role of pharmacotherapy treatment (Aletraris et al., 2016; Uebelacker, Bailey, Herman, Anderson, & Stein, 2016), it is noteworthy that both administrators and providers could see the benefits of integration. Delivering MAUD was perceived as both acceptable and worthwhile, and most prescribing providers expressed a willingness to treat, indicating psychological readiness. While respondents noted that many staff lacked the necessary expertise, suggesting a need to address behavioral readiness, they were open to receiving more training. Where protocols and procedures existed, such as for screening and assessment, administrators felt confident that these protocols were being followed. This suggests that at least in one large urban county, the United States is making progress towards realizing a “no wrong door” approach to treatment access (Centers for Medicare & Medicaid Services, undated; National Institute on Drug Abuse, 2004). This is especially notable given that in Los Angeles County mental health and substance use treatment continue to be administered under different systems of care, unlike most other states and counties which have integrated behavioral health care at the administrative level. It also suggests that philosophical separation between mental health and substance use treatment approaches is lessening, which is an important and encouraging sign.

Our work has organizational and policy implications. Some of the capacity and readiness constraints have relatively straightforward solutions. Clinics can provide training and make expert consultation available to increase behavioral readiness. Obtaining a regular supply of urine drug testing equipment to address physical infrastructure barriers is also relatively straightforward if funding is available. Developing protocols and procedures, while initially time-consuming, ultimately should simplify and standardize care. Likewise, formalizing relationships with specialty substance use provider to address intra-organizational capacity challenges for individuals who need a higher level of care such as inpatient detoxification services, should make accessing this care less time-consuming.

Other organizational capacity and readiness constraints, however, are less straightforward and will require structural changes to the way care is delivered in mental health clinics and state-level funding and policy changes. Changing staffing configurations, visit constraints and caseloads will require addressing the fiscal/economic and workforce construct in the Meyer et al. (2012) framework. Respondents felt that without additional staff, it would not be possible to deliver high quality treatment for co-occurring AUD, which they defined as including both MAUD and psychosocial support. Adding substance use counselors to the workforce in public mental health settings, may address the need for more intensive psychosocial recovery support. Not surprisingly, the requirement for specific documentation practices in order to bill Medicaid (Medi-Cal in California) for co-occurring AUD treatment, including MAUD, is a significant fiscal organizational constraint to integrating care. Changing these requirements will involve state level policy changes.

Our work has implications for the theory behind organizational change. We were able to easily identify, using empirical data, support for the Meyer framework, as a comprehensive way to assess the organizational capacity for a specific change. While the Weiner concept of organizational readiness overlaps and expands upon the workforce and leadership dimensions of the Meyer framework, it does not address important structural constraints such as fiscal and economic resources, policies and procedures and information technology support, which respondents identified as important barriers to adoption. Future work should integrate the two theories, as each contributes to our understanding of organizational change. Our work also supports the use of broader implementation frameworks, like the EPIS (Exploration, Preparation, Implementation, Sustainment) (Aarons, Hurlburt, & Horwitz, 2011) which posits that both inner and outer setting characteristics are important drivers for the adoption of new practices in public service sectors.

Our study has several limitations. The data are descriptive and can be used to generate rather than test hypotheses. We selected a purposive sample of clinics – one clinic from each service planning area within the county -- limiting generalizability of findings. Within each clinic, administrators and providers volunteered to participate, possibly leading to self-selection bias. While this approach could have led to a biased, more positive perspective about MAUD delivery, as shown in the results, administrators and providers still reported reservations about the current capacity to provide MAUD care in their settings. The data come from a single mental health system in a single state. Although Los Angeles County is large and we collected data from urban, suburban and semi-rural clinics with extensive ethnic diversity, our results may not be representative of other mental health clinics.

Conclusion

Evidence-based treatment recommendations promote the integration of care for co-occurring AUD (Iqbal, Levin, & Levin, 2019; Rundell, Oza, Greco, & Cruzado, 2019; Watkins, Hunter, Burnam, Pincus, & Nicholson, 2005). The availability of effective medications for the treatment for AUD brings further opportunities for integrated COD care delivery in outpatient mental health settings. However, we found that the current public mental health system is not adequately structured to provide pharmacotherapy treatment for AUD. More specifically, we documented organizational capacity and readiness constraints in one large mental health system that impedes the effective delivery of MAUD treatment, including fiscal, workforce, physical infrastructure, interorganizational relationship, information technology, policies and procedures, and organizational culture and leadership issues. External factors, such as regulatory changes, as well as inter- and intra-organizational changes will be needed to further realize the goal of effective co-occurring AUD treatment delivery in public mental health settings. Next steps for this work include the development of a “toolkit”, or resources that public mental health systems can use to promote the adoption of MAUD treatment in mental health systems.