Implementing the Chronic Care Model for Opioid and Alcohol Use Disorders in Primary Care

Abstract

Background:

Effective treatments for opioid and alcohol use disorders (OAUD) are available, yet only a small percentage of those needing treatment receive it.

Objectives:

This paper describes a collaborative planning and development process used by researchers and community providers to apply the chronic care model to the delivery of treatment for OAUD in primary care. The goal was to develop and implement an intervention that would support the delivery of brief psychotherapy and medication-assisted treatment (MAT).

Methods:

We used focus groups and interviews to identify barriers and facilitators, and organized the results using the chronic care model. We then identified implementation strategies, the intended organizational changes, and the materials necessary to carry out each strategy, and pilot-tested the process.

Results and Conclusions:

We describe the methods and outcomes of the collaborative planning and development process, and discuss implications of the work for the integration of substance use treatment with primary care.

In 2013, more than 20 million individuals needed, but did not receive, treatment for a probable substance use disorder (SUD).¹ The consequences of untreated SUDs are great, and include increased risk of disease, injury, disability, and death, as well as billions of dollars in costs to the criminal justice, social welfare, and health care systems.^2–6 Substance use also complicates the management of other chronic disorders, such as hypertension and type II diabetes.^7–9 OAUD are of particular concern: almost one-third of emergency department admissions are alcohol related,¹⁰ an estimated 1.9 million people abuse or are dependent on prescription opioids, and 323,000 people are dependent on heroin.¹¹ Effective treatments for OAUD (i.e., psychotherapy, MAT) are available, yet only a small percentage of those needing treatment receive it.^10,12–18

Although addiction treatment delivered in specialty care settings plays an important role for individuals with severe SUD, limited availability, lack of insurance, cost, and the stigma associated with using specialty care means that specialty care alone is insufficient to address this large unmet need. In addition, many individuals do not recognize they have a SUD and thus never seek specialty treatment. For individuals who do seek care, treatment often addresses only the acute episode, with relapse and subsequent treatment delivered by a different provider, impeding continuity of care.¹⁹ Primary care is a logical setting in which to identify and provide treatment for less severe SUDs and for individuals with more severe disorders who are unwilling or unable to enter specialty treatment, because it is well-suited to longitudinal, comprehensive, and coordinated care, and many individuals with substance-related issues seek primary care because they are dealing with medical conditions related to or complicated by their substance use.^20–22 However, despite the feasibility and potential for primary care to increase access to treatment, few primary care settings offer treatment for OAUD.²³

Implementing evidence-based treatments for SUDs in primary care is considered a key strategy for increasing access and decreasing unmet need. The Patient Protection and Affordable Care Act and the Mental Health Parity and Addiction Equity Act decreased financial barriers to access and designated SUD treatment as an essential health benefit.^24,25 Additionally, the National Drug Control Strategy recommends that Federally Qualified Health Centers (FQHCs) substantially expand their delivery of SUD treatment services.²⁶ The chronic care model addresses the need to transform the way clinical care is delivered for chronic illnesses and has been applied to both medical and mental health conditions, but has not been applied to OAUD treatment in primary care,^27,28 despite the similarity of OAUD with other chronic illnesses.^27,29 Additionally, although robust evidence demonstrates the effectiveness of the model for improving both care processes and clinical outcomes,³⁰ there is little information about how to operationalize it for diverse chronic conditions such as OAUD.

Several studies have applied the principles of chronic care management to the treatment of SUDs. The AHEAD trial³¹ tested the impact of chronic care management delivered in a part-time addiction medicine clinic, and a recent trial for homeless women used care managers to refer patients from primary care to specialty SUD care³²; neither found that the chronic care management intervention impacted outcomes. However, neither of these trials implemented the chronic care model as originally conceptualized,²⁸as neither integrated OAUD treatment with the patient’s usual primary care, or addressed how the primary care system could be transformed to provide evidence-based treatment for OAUD delivered by a patient’s usual providers.

This paper describes the process and outcome of a collaborative planning and development process used by researchers and community providers to apply the chronic care model to the delivery of treatment for OAUD in primary care. The goal of the process was to develop and implement a feasible and sustainable clinical intervention that would support the delivery of a six-session, motivational interviewing-based, brief psychotherapy treatment (BT) delivered by licensed social workers and MAT for OAUD with either long-acting injectable naltrexone or buprenorphine/naloxone within routine primary care. The intervention development work was undertaken within the framework of the chronic care model to ensure that the intervention would support longitudinal and proactive treatment, and would include the elements known to be important for the effective delivery of chronic illness care.^28,30 In doing so, our study was unique and distinct from previous trials^31,32 because we integrated the intervention within participants’ regular source of primary care in an effort to decrease barriers to access and increase the use of evidence-based treatments within primary care settings. We focused on OAUDs because both are prevalent in primary care settings, have evidence-based treatments, are major contributors to morbidity and mortality, have standards for appropriate care, and can be treated with medications approved by the U.S. Food and Drug Administration, making them feasible and appropriate for treatment within primary care.^33,34 The clinical intervention is being tested in a randomized controlled trial, currently underway.³⁵ The study was approved by the RAND Human Subjects Protection Committee.

PARTNERSHIP AND SETTING

The partnership for the collaborative planning process included researchers, staff, and providers at a multisite FQHC in Eos Angeles. The FQHC provides comprehensive medical care, and licensed social workers integrated into the FQHC provide behavioral health services to individuals with depression and anxiety. Before the collaboration, the clinic did not screen patients for SUDs and did not provide any OAUD treatment. The clinic serves more than 22,000 low-income families and individuals each year; patients are ethnically and racially diverse, with 58% of Hispanic origin (primarily Mexican and Central American), 26% Caucasian, 11% African American, and 4% Asian. A contracting agreement was developed between the research organization and the FQHC that specified roles, responsibilities, and tasks to be accomplished. Although the entire FQHC was considered a member of the partnership, the Chief Executive Officer of the FQHC authorized the medical director and director of behavioral health to be the key provider participants in the collaborative planning process, which took place during 2-hour, bi-weekly implementation team meetings over an 18-month period, and then during 90-minute monthly implementation team meetings with email communication between meetings. In addition to the medical director and director of behavioral health, the meetings were attended by a primary care physician with health services training who was a member of the FQHC board of directors and who coordinated the internal medicine resident training program at the clinic, four researchers (psychiatrist and health services researcher, PhD social welfare researcher, PhD psychologist, and a survey director), a note taker, and other staff and providers from the FQHC depending on the topic being discussed. All individuals who were involved in the delivery of OAUD care were employees of the clinic; no new staff was hired for the study.

APPLICATION OF THE CHRONIC CARE MODEL TO THE DELIVERY OF OAUD CARE

The chronic care model is a framework that guides health care delivery systems—particularly primary care practices—to translate general ideas for quality improvement into specific applications.²⁸ The model posits that, to improve health outcomes, the health care system needs to consider six elements of care, which have been shown in previous studies to improve outcomes. Throughout the first 18 months of the collaborative planning process, the partnership considered how these six elements could be applied to the treatment of OAUDs within the context of a FQHC. Each element was framed as a question, which guided the work of the partnership (Table 1).

Table 1.

Application of the Chronic Care Model to Delivering Care for OAUD

Element of the Chronic Care Model	Application to delivering care for OAUD
Delivery system redesign	Is the delivery system designed to ensure the delivery of evidence-based care for OAUD?
Self-management support	Are patients encouraged to become informed and competent self-managers of their OAUD?
Decision support	Is there expert consultation available to help clinicians adhere to evidence-based OAUD treatment practices?
Clinical information systems	Do clinical information systems help providers plan and deliver care for both individuals and populations of patients with OAUD and to monitor performance?
Health care organization	Is the leadership supportive, and are resources provided to support the delivery of OAUD care?
Community resources	Are patients linked to community OAUD recovery resources?

OAUD, opioid and/or alcohol use disorder.

COLLABORATIVE PLANNING AND DEVELOPMENT PROCESS

The goal of the collaborative planning process was to develop and implement a feasible and sustainable intervention that would support the delivery of both BT and MAT as clinically appropriate and chosen by the patient. The implementation team used focus groups and interviews conducted by the researchers to solicit input on barriers and facilitators to implementing OAUD treatment from all staff members, key administrators, and clinic board members. Protocols were developed by two of the researchers with input from the clinical partners and then reviewed by the entire team. Focus groups and interviews were audiotaped and transcribed and reviewed by the researchers to identify key themes, which were then shared with and analyzed by the larger implementation team.

Barriers were organized according to how they mapped onto the model, and team members discussed how each barrier could be overcome. Team members then identified implementation strategies to address barriers, the intended organizational change, and the materials necessary to carry out each strategy. We defined an implementation strategy as a method designed to enhance the adoption of a particular clinical practice; strategies were identified through the collective experience of the team and by reviewing the literature on strategies for implementing organizational change.^36–42

In parallel with the process used to identify and address implementation barriers, the team developed current and future state workflow diagrams and draft care protocols to allow the organization to visualize how the continuum of care for OAUD could be integrated into clinic practices. Where choices emerged about how to operationalize a specific process (e.g., should screening be conducted by the medical assistants or self-administered by the patients?), the team conducted plan–do–study–act cycles with small groups of clinicians or staff to determine which choice would be most compatible with existing clinic workflow and culture, and then discussed the results (e.g., number of patients who completed a screening, quality of the screening, staff preferences) and the relative merits and drawbacks of each. Plan–do–study–act cycles are a structured approach to improving a process by engaging the end-user in planning a test of change (plan), carrying out the test (do), observing the result of the test (study), and refining the process (act).^38,41,43,44The draft set of processes were pilot tested and revised over an 18-month period to further ensure they worked, were compatible with existing clinic processes, and were acceptable to clinic staff. The pilot test revealed lower than expected rates of identification of patients with OAUD, difficulties engaging homeless or developmentally disabled clients, information gaps in the registry, inconsistencies in the way patients were referred to the care coordinators (the warm handoff), and the need for procedures to identify and address clients at risk of not following up after meeting with the care coordinator, or clients with co-occurring mental health disorders. As a result, we revised the screening and warm handoff procedures, revised the registry, and expanded the role of the care coordinator to include additional motivational interviewing–based strategies during the warm handoff. Patients identified with co-occurring mental health disorders during the initial assessment with the behavioral health provider were given the choice of addressing their mental health condition first, or concurrently with their substance use. The final processes were incorporated into three manuals,^45–47 which contain detailed workflows and were reviewed by subject matter experts from outside the organization to ensure they were consistent with the best available science.

OUTCOMES OF THE COLLABORATIVE PLANNING PROCESS

Tables 2 through 6 show the implementation barriers identified for five of the six chronic care model elements, along with the corresponding implementation strategies, and desired change. The sixth element, health care organization, is not included because respondents did not identify barriers in this domain. Several of the implementation strategies (e.g., organize clinician–researcher implementation team meetings and conduct a local needs assessment) were part of the collaborative planning process, and, although not a result of the process, are included for completeness. We addressed all barriers with the goal of developing a new set of processes that would be compatible with the clinic workflow and culture and increasing provider and organizational readiness to deliver MAT and BT through training and modeling by the identified clinic champions.

Table 2.

Chronic Care Model Component: Delivery System Redesign

Implementation Barriers	Implementation Strategies	Desired Changes
1. No consistent way to identify patients with OAUD. 2. No work flow process to identify and link patients to treatment; no accountable person to provide linkage. 3. Time constraints on visit length. 4. Providers lose clinical time if patient does not show up for intake appointment.	Obtain leadership buy-in for procedural changes (barriers 1, 2, and 3):  Clinic leadership and executive board endorsed program (“part of mission”).  Medical assistant training in new OAUD screening protocol.  Clinic leadership authorized extended time for both medical and therapy intake visits.  Leadership authorized the creation of a CC position and revised professional roles. Organize and hold clinician/researcher implementation team meetings (barriers 1, 2, 3, and 4):  Monthly meetings held to develop and implement screening protocol.  Monthly meetings held to develop a written care protocol with detailed specifications of who does what to whom, when and how. Revise professional roles (barrier 2):  Protocol included a new position, the CC, who used a registry to track and follow patients identified with OAUDs, and who functioned as a link between the BHP and primary care providers.  Researcher/clinician implementation team considered how CC, therapist and medical provider would work together, and what tasks could be shifted away from medical provider. Conduct cyclical small tests of change (barriers 1 and 2):  Care protocol was iteratively tested and refined during a pilot phase before formal testing, using the plan-do-study-act method.	Medical assistants screen all patients at every visit using a validated screener. Formalized care protocol specifies who does what to whom, when and how. Care protocol provides examples of scripted interactions with patients and allows for multiple ways patients can access treatment. For example, patients can access MAT directly or by being referred by a BHP. Visit lengths adjusted. BHP conducts initial assessment and discussion of treatment options; only patients with likely dependence are referred for MAT. CC position created. Bi-weekly clinical status assessment entered into EHR for medical providers to view.

BHP, behavioral health provider; CC, care coordinator; EHR, electronic health record; MAT, medication-assisted treatment; OAUD, opioid and/or alcohol use disorder.

Table 6.

Chronic Care Model Component: Linkages to Community Resources

Implementation Barriers	Implementation Strategies	Desired Changes
1. No systematic way to provide inpatient detoxification services. 2. Linkage with community resources used infrequently.	Created community linkages (barrier 2):  Visited local specialty substance use treatment provider to build working relationships and develop referral processes for clinic patients. Obtain formal commitments (barrier 1):  Obtained commitment from provider of inpatient detoxification services to expedite access to services.  Developed mutually agreed-upon protocol for patient referral. Promote network weaving (barriers 1 and 2):  Lists of available community resources, including AA/NA and specialty providers developed and provided to patients by CC.  Patients referred to a more intensive level of care if provider and patient together determine that this level of care is insufficient.	Patients able to access inpatient detoxification services without long waiting periods. Patients able to access a higher level of care if needed.

AA/NA, Alcoholics Anonymous/Narcotics Anonymous; CC, care coordinator.

The end product of the collaborative planning process was a clinical intervention, Collaborative Care (CC)^45,46 (Table 7). Together, the different components of the intervention are intended to support the delivery of treatment for OAUD in primary care. For each aspect of the continuum of care (screening and identification, referral to behavioral health, longitudinal assessment, treatment planning, and initiation of either BT or MAT or both), a care protocol in the CC manual⁴⁷ specifies what action should be done, who is to carry it out, how it is to be accomplished, when it should occur, and the resources needed. For example, after the medical assistant screens the patient for risky OAUD use, the results of the screening are given to the primary care provider, who assesses the patient, delivers a brief intervention, and refers the patient to a care coordinator. Because most primary care patients identified by screening are not seeking care for their substance use, we assumed that they would be ambivalent about starting treatment. Thus, motivational interviewing techniques and multiple opportunities for patient engagement were incorporated into the protocols. The CC intervention included a new position, that of the care coordinator, whose role was to meet with patients after their medical appointment, assess motivation, and encourage treatment initiation and adherence, as well as to follow patients longitudinally whether or not they elected to start treatment. The care coordinators used a patient registry to do population-based management on all patients identified and to track outcomes for those who initiated treatment, by assessing patient outcomes on a monthly basis. The care coordinator role was shared between two individuals with BA-level education, who underwent 2 days of training on motivational interviewing techniques and participated in weekly supervision and case management meetings with the behavioral health providers and a clinical psychologist, who led these meetings.

Table 7.

Components of Collaborative Care

Care protocol for continuum of care

BT and MAT for primary care treatment protocols

Care coordinator

Specialized patient registry

Provider training materials

Decision support system for providers

Provider alerts in patient management system

Culturally adapted patient education materials

Linkages with community resources

BT, brief treatment; MAT, medication-assisted treatment.

DISCUSSION

Policymakers and medical providers are increasingly recognizing the importance of primary care in increasing access to treatment for SUDs and reducing unmet need. There has been relatively little guidance, however, on how primary care practices can implement the continuum of care, given the historic separation between the substance use treatment system and general medical care, and the difficulties inherent in creating the organizational changes needed to implement any new practice.^40,48 In this article, we describe a process used within a research-provider partnership to develop and implement a clinical intervention based on the chronic care model which supports the delivery of effective treatments for OAUD in a multisite FQHC. The intervention was designed to address the barriers to implementing the continuum of care for OAUD in primary care. The chronic care model was used as an organizing framework because of the similarities OAUDs share with other chronic illnesses.²⁷ Although the resources needed to develop and implement the CC intervention may not be available to some community clinics, clinics interested in adopting the intervention can build on the published manuals^45–47 and experience of the current study to assess what resources and strategies would be needed to implement a similar program.

Although the World Health Organization, the Centers for Medicare and Medicaid Services, the National Committee for Quality Assurance, and policy analysts have all endorsed the use of the chronic care model to improve the organization and delivery of care and to improve health outcomes, there is little experience to inform the operationalization and implementation of the model for OAUD in primary care settings. Our study is the first to use the chronic care model for OAUD treatment in primary care as originally conceptualized, and to describe the process used to operationalize the model. We found that although the model was a useful framework to guide the development of the clinical intervention and to ensure the intervention was consistent with the best available science, it needed to be operationalized before it could be implemented. The collaborative planning process was necessary to identify and address potential implementation barriers, develop the care protocols, ensure the intervention was aligned with the culture and processes of the organization, and ultimately to make the intervention feasible and acceptable to providers. Although the implementation barriers identified may not generalize to other primary care settings, they are consistent with the research literature on implementation barriers for other health care practices.⁴⁹

There are multiple hypothesized advantages to integrating OAUD treatment into primary care, although few studies have tested the effectiveness of integration. Longitudinal care—which in this case means that individuals are given multiple opportunities to engage and can increase or decrease treatment intensity depending on need and preferences—addresses the chronic nature of addiction²⁹ and changing levels of motivation and desire for treatment and recovery. Moreover, patients may experience less stigma when discussing substance use with a medical provider than when seeking specialty care^50,51 and be drawn to the convenience of treatment in non-specialty settings with their regular provider. Regular assessment of use may facilitate the early identification of harmful use when less intensive treatments can be successful, and reassessment at regular intervals may facilitate early identification of relapse and continuity of care. Patients may be more receptive to learning about the impact of substance use on their physical health if this information is provided in a factual and non-judgmental manner, and may be more receptive to an intervention that does not require them to identify as an addict.

Complex organizational change interventions typically are not described in enough detail for replication, and there has been a call for greater clarity about the functional components of interventions and the use of consistent terminology.⁵² Strengths of our report include use of terminology consistent with current implementation research methodology,⁴⁹ and the detailed description of the relationship between implementation barriers, implementation strategies, and desired organizational changes. Strengths of the CC clinical intervention include the focus on team-based, longitudinal care, and a population-based approach, which gives patients multiple opportunities and choices about engaging with care. Population-based management is important because most individuals with SUD do not perceive a need for treatment and could benefit from providers capitalizing on teachable moments, when readiness and motivation for treatment may be high. Bringing OAUD into the primary care setting can help to break down the wall that separates SUD treatment from other acute and chronic medical care. A collaborative planning process can be used to apply the chronic care model to the treatment of OAUD in primary care and to develop a feasible and acceptable clinical intervention. Data on the effectiveness of the intervention on outcomes are forthcoming.

Table 3.

Chronic Care Model Component: Patient Self-Management Support

Implementation Barriers	Implementation Strategies	Desired Changes
1. Patients do not know about treatment availability. 2. Patients are not ready to initiate treatment, or have other, more pressing problems. 3. Once treatment has begun, patients drop out before completing treatment. 4. Patients not prepared to partner in their treatment.	Develop and distribute educational materials (barrier 1):  Researchers and clinic staff partnered to create written educational materials in English and Spanish appropriate for clinic population; materials reviewed by patients and revised.  CCs provide patients with written educational materials at time of initial contact.  BHPs and medical providers educate patients about different treatment options (BT, MAT, detoxification, watchful waiting) at intake and, if needed, at subsequent visits. Encourage patients to initiate and engage with treatment (barriers 2 and 3):  CCs meet with patients and, using motivational interviewing techniques, attempt to engage patients with care.  Registry prompts CCs to call patients and offer multiple opportunities to engage with treatment.  CCs assess clinical status and patient goals bi-weekly using standardized questionnaire; data entered into registry along with urine toxicology results.  Registry creates an individualized patient report which provides visual feedback to patient on their goals and progress.	Educational materials made available to all patients identified with a OAUD. Posters placed throughout clinic to encourage patients to seek help for their substance use. Patients given multiple opportunities to access different types of treatment. Prompted by registry, CC made reminder phone calls. Individualized patient reports showing progress and goals given to patient. CC trained in motivational interviewing techniques during a 2-day training and used techniques to validate patients’ current readiness to change and encourage patients to try “one visit” to see if BT was a good fit.

BT, brief treatment; BHP, behavioral health provider; CC, Care Coordinator; MAT, medication-assisted treatment; OAUD, opioid and/or alcohol use disorder.

Table 4.

Chronic Care Model component: Decision Support

Implementation Barriers	Implementation Strategies	Desired Changes
1. Providers not psychologically prepared to deliver OAUD treatment. 2. Providers and support staff (nursing, medical assistants) not behaviorally prepared to implement OAUD treatment. 3. MAT and BT treatment protocols not specific for primary care.	Conduct a local needs assessment (barriers 1 and 2):  Researchers and clinic staff conducted a local needs assessment to assess provider readiness to deliver OAUD care. Adapt existing treatment protocols to meet local needs while maintaining fidelity (barriers 2 and 3):  Multidisciplinary research/clinician team with MAT and MI expertise adapted existing treatment manuals and developed treatment protocols for primary care setting.  Final treatment protocols were reviewed by at least two external experts to identify if recommendations were consistent with best practices. Identify and prepare champions (barriers 1 and 2):  Two champions (one medical and one behavioral health) came forward and began to use the two treatment protocols. Their experiences were shared with other clinicians, informally in conversations and formally in staff meetings. Provide training and technical assistance (barriers 1 and 2):  Brought in local addiction experts to provide training to all staff on the impact of substance use disorders on health of patients.  Facilitate providers obtaining needed DEA certification. Conduct cyclical small tests of change (barriers 1 and 3):  Champions and staff tried out different steps in the process and team made real-time changes to procedures based on user feedback. Purposely examine treatment protocols during pilot test for feasibility and acceptability (barriers 1, 2, and 3):  Protocols iteratively tried out and refined.  Feedback from all providers explicitly encouraged and incorporated into final protocols.  Care protocol and pilot test allowed providers to visualize how treatment could work in their system. Provide ongoing clinical supervision (barriers 1 and 2):  Local addiction experts provided ongoing supervision and consultation.	Medical providers trained in MAT and have DEA certification to prescribe buprenorphine. Provision of ongoing, OAUD-specific decision support for medical providers. Two-day training in BT for both the BHP and the CC, followed by weekly supervision/case management meetings attended by BHP and CC. Providers behaviorally and psychologically ready to integrate OAUD treatment, demonstrated by provider adoption of BT and MAT into routine practice. MAT and BT treatment protocols created for primary care.

BT, brier treatment; BHP, behavioral health provider; CC, Care Coordinator; MAT, medication-assisted treatment; MI, motivational interviewing; OAUD, opioid and/or alcohol use disorder.

Table 5.

Chronic Care Model Component: Clinical Information Systems

Implementation Barriers	Implementation Strategies	Desired Changes
1. No way to track treatment use and clinical outcomes for the population of individuals identified with OAUDs.	Develop and implement stand-alone registry for population-based management:  Researchers and clinical champions partnered to describe how the registry would be used and determined the registry could not be part of the EHR.  Researchers created registry.  CCs given access to registry and then they iteratively tested and provided feedback on ease of use and functionality.  Registry prompts CCs to make reminder phone calls to patients 2 days before each therapy appointment, assess outcomes on all patients attending BT using a validated clinical outcomes monitoring tool, and order monthly urine toxicology screens.  Registry was used by CCs to generate a list of all patients for population-based tracking during bi-weekly clinical supervision meetings attended by BHPs and CCs; Information from registry was relayed to medical providers by CCs through the EHR.  Registry prompts CCs to call out-of-treatment patients and offer patients multiple opportunities to engage with treatment. Revise the EHR system:  EHR alert placed on chart so that medical providers know to ask patient about substance use and desire for treatment at initial and subsequent visits.	Population-based tracking of all patients identified with OAUDs in bi-weekly meetings attended by CC and BHPs. Measurement-based care for all patients receiving treatment for OAUDs. Medical providers able to access information collected by CCs on patient’s clinical status.

BHP, behavioral health provider; CC, care coordinator; EHR, electronic health record; OAUD, opioid and/or alcohol use disorder.