Training Addiction Counselors to Implement an Evidence-Based Intervention: Strategies for Increasing Organizational and Provider Acceptance

Abstract

One barrier to widespread public access to empirically supported treatments (ESTs) is the limited availability and high cost of professionals trained to deliver them. Our earlier work from two clinical trials demonstrated that front-line addiction counselors could be trained to deliver a manualized, group-based cognitive behavioral therapy (GCBT) for depression, a prototypic example of an EST, with a high level of adherence and competence. This follow-up article provides specific recommendations for the selection and initial training of counselors, and for the structure and process of their ongoing clinical supervision. Unique challenges in working with counselors unaccustomed to traditional clinical supervision are highlighted. The recommendations are based on comprehensive feedback derived from clinician notes taken throughout the clinical trials, a focus group with counselors conducted one year following implementation, and interviews with key organization executives and administrators.

Introduction

Although empirically supported treatments (ESTs) exist for a wide range of psychiatric disorders, efforts at effective dissemination in community clinical settings have been met with limited success (Addis, 2002; Shafran, et al., 2009). Little support has been provided to ensure quality delivery of ESTs in substance use disorder (SUD) settings, despite pressure from policymakers, funders, and local stakeholders (NIDA, 2004). At the most basic level, adequate guidelines to help organizations prepare clinical staff to implement ESTs are lacking (Glasner-Edwards & Rawson, 2010; Gotham, 2004; Marinelli-Casey, Domier, & Rawson, 2002; Sholomskas et al., 2005).

Such guidelines are especially crucial in public sector substance abuse treatment settings, where services are typically provided by addiction counselors who lack educational or clinical exposure to theories and principles underlying ESTs, and who may be highly skeptical of changing the treatment status quo (Campbell, Catlin, & Melchert, 2003; Horvatich, 2006; Mulvey, Hubbard, & Hayashi, 2003). Training addiction counselors to implement ESTs for co-occurring disorders, which affect large numbers of those diagnosed with SUDs (Conway, Compton, Stinson, & Grant, 2006; Substance Abuse and Mental Health Administration, 2009), may be particularly challenging because many of them lack substantive training in the recognition and treatment of such disorders (Kerwin, Walker-Smith, & Kirby, 2006; Nunes, Selzer, Levounis, & Davies, 2010). Appropriate treatment of co-occurring disorders is critical, as they are associated with a host of adverse outcomes, including greater severity of substance and affective symptoms, increased risk of suicide, poor treatment response, decreased remission rates, and diminished quality of life (Aharonovich, Liu, Nunes, & Hasin, 2002; Davis et al., 2006; Rush et al., 2005; Saatcioglu, Yapici, & Cakmak, 2008; Watkins, Paddock, Zhang, & Wells, 2006). Currently, only 4 to 13% of U.S. adults with co-occurring psychiatric and substance use disorder are able to access treatment for both (Substance Abuse and Mental Health Services Administration, 2009; Watkins, Burnam, Kung, & Paddock, 2001).

Increasing educational requirements for addiction counselors to the level of those for mental health counselors (i.e., requiring a master's degree) could increase the number of providers able to assess and effectively treat dually diagnosed clients (Amodeo, 2006). However, this option may not be realistic for public sector SUD programs because of the costs of hiring staff with advanced degrees to provide direct client services. A more realistic alternative is to train front-line addiction counselors to implement ESTs that are usually provided by mental health counselors with advanced degrees.

Such training is a considerable undertaking. It must address multiple factors, including the organizational climate in which new practices will be introduced, characteristics of clinicians to be trained, processes for preparing for and implementing treatment, and methods for evaluating these efforts (Addis, 2002; Damschroder & Hagedorn, 2011; Godley, Garner, Smith, Myers, & Godley, 2011; Gotham, 2006; Proctor et al., 2007; Rosenthal, 2002). To illustrate, Godley and colleagues' (2011) comprehensive model for the dissemination and implementation of a behaviorally based substance abuse intervention for adolescents includes 3.5 days of initial clinician training, knowledge tests on the manualized treatment, biweekly coaching calls to receive feedback on the implementation of the intervention, expert review of digital recordings of treatment and supervision sessions, provision of electronically based feedback, and post-certification fidelity monitoring.

Although such an undertaking presents considerable challenge, training addiction counselors to competently deliver ESTs for co-occurring disorders would provide advantages for both clients and provider systems. Clients could have treatment needs met by a single organization and group of providers, and understanding the interrelationship between their co-occurring disorders may be easier when these conditions are treated together. Provider systems would benefit from using a more comprehensive approach utilizing existing resources.

Cognitive behavioral therapy (CBT) would be a particularly relevant EST for addiction counselors to learn, given the strong empirical base supporting its effectiveness in treating both SUDs (Carroll et al., 2008; Hides et al., 2010) and depression (Butler, Chapman, Forman, & Beck, 2006), one of the most common co-occurring disorders. Promising findings on depression outcomes have been reported for clinicians without advanced degrees (paraprofessionals) who are trained to deliver CBT (Bright, Baker, & Neimeyer, 1999; Montgomery, Kunik, Wilson, Stanely, & Weiss, 2010; Thompson, Gallagher, Nies, & Epstein, 1983). A review of studies directly comparing the efficacy of CBT for anxiety or depression delivered by paraprofessionals with that delivered by professional therapists, found that client outcomes (e.g., symptom ratings) were generally commensurate (Montgomery, et al., 2010). Because group-based treatments are the predominant treatment modality in addiction settings (Sobell & Sobell, 2009), it is particularly important to explore the use of group-based CBT (GCBT) for the treatment of co-occurring disorders.

To address the need for ESTs that target co-occurring disorders, we developed and tested a manualized GCBT for depression for use in addiction treatment settings. We developed two treatments, one focused specifically on depression— BRIGHT (Broadening Recovery by Improving Goals, Habits, and Thoughts; Hepner, Miranda, et al., 2011) — and one that integrated depression and substance use treatment (BRIGHT-2; Hepner, Muñoz, et al., 2011; Osilla, Hepner, Muñoz, Woo, & Watkins, 2009). We found that addiction counselors with little or no prior mental health experience could be effectively trained to deliver both BRIGHT (Hepner, Hunter, Paddock, Zhou, & Watkins, 2011) and BRIGHT-2 (Hunter et al, in press). Independent ratings of audiotaped sessions showed that the average adherence rate was 94% for BRIGHT and 95% for BRIGHT-2 across all coded sessions, indicating that counselors maintained high adherence to the treatment. Independent raters also assessed CBT competency using a modification of the Cognitive Therapy Adherence and Competence Scale (Barber et al., 2003). A score of 4.0 or higher on this 7 point scale indicates competent CBT delivery, and the average competence score across all coded sessions was 4.1 for both BRIGHT and BRIGHT-2. In the BRIGHT study, the treatment provided by the addiction counselors resulted in significant clinical improvements over conventional treatment. Specifically, BRIGHT clients reported significantly fewer depressive symptoms, as assessed by the Beck Depression Inventory-II (BDI-II; Beck, Steer, & Brown, 1996) compared to usual care clients at three-and six-month follow-ups. At the 6 month follow-up, BRIGHT clients also reported significantly fewer drinking days and fewer days of problem substance use compared to clients receiving usual care (Watkins, et al., 2011). The results from BRIGHT-2 indicated that the treatment reduced both depressive symptoms and substance use at 3 and 6-months. However, the results were not statistically different from usual care; this may be partially due to the small sample size (n = 73; Hunter et al., in press).

This article describes our experience in training addiction counselors to deliver the BRIGHT interventions. It includes information regarding the selection, training, and supervision of counselors and provides recommendations for future EST dissemination. Our goal is to provide an implementation model which details the changes made in the traditional mental health professional training and supervision model to tailor it for addiction counselors. We discuss some of the unique challenges that may arise in such trainings and specific strategies to address these challenges.

Method

We briefly describe the interventions, treatment manuals, overall training plan, trainee and trainer/supervisor characteristics, and methods used to document challenges we faced during counselor selection, training, and supervision.

The BRIGHT Interventions

We implemented our manualized training and supervision model in two clinical trials, BRIGHT and BRIGHT-2, which were implemented over 2 ½ and 1 ½ year periods, respectively. Therapy groups were co-led by pairs of addiction counselors. The two treatments were adapted from an existing CBT manual (Muñoz, Ippen, Rao, Le, & Dwyer, 2000) to create a 16 session depression treatment (Hepner, Miranda, et al., 2011) and an 18-session integrated depression and substance abuse treatment (Hepner, Muñoz , et al., 2011). The group leader manuals and group member workbooks are identical, except that the former include instructions and sample scripts for introducing exercises. The group leader manuals focus on instructing counselors to apply CBT skills within the context of specific exercises, rather than providing extensive information on CBT theory or higher-level case conceptualization. We believe that these higher-order concepts can be most usefully woven into weekly clinical supervision over time, after counselors have gained confidence in the application of basic techniques. Both trials received approval from the RAND Human Subjects Protection Committee.

BRIGHT Training Programs

Addiction counselors selected for the BRIGHT trials first underwent an interactive two-day in-person training session that included information on depression and basic CBT concepts. The basic CBT model was described, highlighting the interrelationships between thoughts, mood, and behavior (using the CBT circle) and the way depression symptoms and substance abuse could be understood and modified using this approach. Time was spent reviewing basic counseling skills, beginning CBT competencies, and group management issues. These skills were further developed during ongoing, weekly two-hour clinical supervision that included audiotaped CBT session reviews. These procedures are detailed below.

Trainer/supervisors

Training was provided by five clinical psychologists (4 women, 1 man) who had extensive CBT experience, including graduate and post-graduate training and supervised clinical experience delivering CBT, as well as private practice CBT work. Three of the trainers taught graduate-level CBT coursework in university-based clinical training programs. One trainer was the author of the treatment protocol from which the BRIGHT manuals were adapted. Responsibility for ongoing clinical supervision was shared by four of the trainers, all of whom had prior clinical supervision experience and/or were significantly involved in the development of the treatment manuals.

Counselors

Eight addiction counselors (6 women, 2 men) were trained to deliver one or both of the interventions. Two counselors dropped out of the study because of personal circumstances and a new job opportunity, respectively. The counselors were employed by one of the largest publicly funded substance abuse treatment providers in a large metropolitan area. Most had between two and four years of experience with SUD treatment, and one had 31 years. Only one of the counselors had an advanced degree (master's degree) at the start of training. Seven counselors had completed a state SUD certification program, and five were in recovery from drug or alcohol use.

Training Challenges and Solutions

We used three methods to obtain information about the implementation of training and supervision. First, we reviewed detailed notes that two of the training clinicians made throughout the implementation period. Second, we conducted a focus group with the participating addiction counselors one year after the initial training (N = 4). The focus group asked the counselors about their experience with training and supervision, perceptions of clients' experience, support within the organization and general satisfaction. Third, interviews with key executives and administrators provided information on the organizational context in which the training and supervision were implemented. The focus group and executive/administrator interviews were facilitated by project team members who were not involved in the training and supervision component in order to reduce reporting bias.

Following the implementation period, the clinician training and supervision notes were reviewed by the clinicians who provided weekly supervision to extract key themes, which were then organized into the following categories: (a) pre-implementation issues; (b) process issues in the supervision of addiction counselors; (c) development of non-specific counselor competencies during supervision; and (d) development of CBT-specific competencies during supervision. Responses in the focus group were organized by the interview questions and summarized to determine whether there was general agreement or disagreement. The focus group results were then integrated into the domains specified by the clinician training and supervision notes.

Results

Challenges in Preparing to Implement Counselor-Led GCBT

Table 1 summarizes challenges faced prior to implementation. These challenges related to the organizational context in which the treatment was delivered, the selection of counselors, and initial training efforts.

Table 1. Preparing to Implement Addiction Counselor-Led Group CBT.

Domain	Challenges	Solutions
Organizational	• Staff skepticism of ESTs	• Obtain consistent, visible support from program administration for EST
Context	• Staff concern about altering usual treatment	implementation.
	approaches	• Highlight benefits to organization, staff, and clients in communications
	• Implementation of a new treatment in an	about EST implementation.
	overburdened system	• Ensure release time and work support for counselors to be trained.
		•Collaboratively develop administrative procedures that are minimally disruptive to competing staff demands
CBT Counselor	• Low counselor buy-in to EST due to concerns	• Select counselors with high interest in learning about CBT and
Selection	about:	depression.
	◦ Required time commitment	• Link learning CBT to enhancement of professional development.
	◦ Unfamiliar nature of the intervention
Initial Counselor	• Counselors may be overwhelmed by amount of	• Provide in-person training balancing didactic material and interactive
Training	material to be learned.	exercises.
		• Frame training as building on pre-existing counselor competencies.
		• Focus on counselor ability to understand and explain basic CBT model.
	• Counselors may have philosophical differences	• Acknowledge differences in treatment approaches, but highlight
	about what constitutes an effective treatment	compatibility between CBT and counselors' existing treatment
	approach.	philosophy.
		• Use CBT skills to address counselor concerns.

Organizational environment

Organizational factors are critical yet frequently overlooked elements in the literature on implementing ESTs in substance use treatment settings (Damschroder, et al., 2011). The addiction counselors who became GCBT leaders initially expressed significant concerns about the time and effort required to learn and lead the interventions. In addition, before the interventions were implemented, the organization did not have a consistent policy supporting the use of ESTs. Some training in empirically supported approaches (e.g., motivational interviewing) and the use of treatment manuals were available to counselors, but were not mandatory. Although administrators expressed openness to ESTs, they cited high counselor caseloads as a limiting factor in providing standardized training across their treatment sites.

Therefore, first securing organizational support prior to training and treatment implementation was key to increasing staff acceptance of GCBT. At monthly meetings we presented information to executives and other administrators (e.g., program directors) about the potential value of GCBT (e.g., improved client outcomes, enhancement of staff members' professional skills, long-term organizational cost and time savings). We also addressed questions and concerns, beginning 6 months before the interventions were implemented and throughout the study period. Administrative support for GCBT was then communicated to the entire counseling staff by (1) providing release time to attend mandatory meetings in which the study interventions were reviewed, (2) supporting the use of feedback and incentives for staff assisting in screening clients for GCBT, (3) providing additional release time for counselors who became GCBT group leaders, (4) having key administrators present at staff meetings regarding the study, and (5) featuring the study in the staff newsletter. To limit the burden of implementing a new treatment in an already overtaxed system, we provided ongoing technical assistance and were open to modifying administrative procedures (e.g., documentation, transportation scheduling) based on feedback from and preferences of staff.

Counselor selection

A primary challenge in selecting addiction counselors for GCBT training was overcoming their wariness of the large time commitment required (e.g., two-day initial training, leading the two-hour group twice weekly, weekly two-hour group supervision) and differences between the content and structure of a GCBT intervention and that of the groups addiction counselors typically lead (which are generally substance-abuse-focused and process-oriented). To surmount this challenge, we interviewed prospective GCBT leaders to find individuals who demonstrated flexibility and openness to alternative therapeutic approaches, and a willingness to learn a manualized treatment, work with a co-facilitator, and receive the type of feedback associated with intensive clinical supervision. We used the interviews to learn about counselors' general treatment philosophies, as well as their previous experience with and knowledge of depressed clients and CBT (which we found was typically limited). We also used the interviews to build upon counselors' expressed desire to enhance their clinical knowledge with new skills/tools. A year after the training, the selected counselors reported that improving their clinical skills to better address client needs was the primary reason for deciding to participate in the training. They also indicated that they thought they had been selected because they demonstrated more willingness to learn new material and dedication to addressing mental health issues than their peers.

Initial training

The counselors selected for GCBT training had little experience using a manualized intervention, and they were worried about their ability to adequately grasp the manual material prior to leading the groups. We therefore facilitated an engaging two-day training session that allowed counselors to actively practice new skills while enabling their trainers to gauge the counselors' initial grasp of the concepts presented. We opted for an intensive initial training rather than briefer trainings spread over time, for both practical and team-building reasons. Cohesiveness and rapport were built by having counselors who would be co-leading groups complete the intensive training together and having the training given by individuals who would be counselors' ongoing clinical supervisors. Framing the training as building upon skills counselors already possessed helped to reduce their anxiety about the amount of information to be learned. Importantly, the training provided a platform on which the trainers could model behaviors and shape expectations of what would later occur in weekly supervision, including collaborative discussion, detailed review of treatment elements, and provision of constructive feedback on counselor performance.

We did not feel it was crucial for counselors to master a great deal of CBT theory prior to leading groups. However, we believed it was essential for them to understand and be able to explain the basic CBT model (i.e., the interrelationships among thoughts, activities, and mood) using a large diagram (the CBT circle) that is prominently posted during BRIGHT group sessions. During the two-day training, counselors practiced explaining the CBT circle to the trainers and their fellow counselors, using depression symptom examples. We reinforced learning by explicitly identifying counselors' effective use of specific CBT techniques and encouraging them to apply those techniques on each other during the training process.

Throughout the initial training, differences between CBT and the treatment approaches typically used by counselors emerged. For example, we emphasized helping clients differentiate between lapses and relapses in their substance use, which was at odds with the abstinence-only approach endorsed by some counselors. We non-defensively discussed differences in perspectives, but also pointed out similarities between CBT and traditional treatment approaches. For example, we highlighted the similarity of the Serenity Prayer, popular in Alcoholics Anonymous, to CBT approaches that emphasize the individual's ability to choose how to think about and react to events, even if the events themselves cannot be changed.

We found it helpful to encourage counselors to use CBT skills to address their skepticism about the treatment. For example, some counselors believed a manualized treatment with structured sessions would be rejected by clients because it would preclude spontaneous discussion of and attention to client crises and would interfere with rapport building. We asked those counselors to withhold judgment about clients' rejection of the treatment until evidence could be gathered on actual (vs. presumed) reactions. We also collaboratively developed strategies for weaving client-raised concerns into session agendas to avoid derailing session goals. For example, counselors were taught how to encourage a client who came to a Thoughts Module session wishing to discuss an upcoming court date regarding custody of her children to identify negative automatic thoughts related to this event that could be focused on in the session. Further, we provided counselors with psychoeducation to address the common misperception that CBT focuses less on building a strong relationship with a client than other forms of therapy do, and we explicitly built in training on how to increase rapport with clients.

Supervising Addiction Counselors: Process Issues

Table 2 presents challenges in the process of supervising addiction counselors who are implementing a GCBT intervention.

Table 2. Supervising Addiction Counselors in Group CBT: Process Issues.

Challenges	Solutions
Counselors may be unaccustomed to	• Socialize counselors to the structure and process of supervision.
intensive clinical supervision.	• Make supervision sessions structured and focused on acquisition of practical skills.
	◦ Role playing
	◦ Concrete suggestions for successful group management
	• Model expected behaviors in supervision.
• Counselors may be hesitant to	• Keep ratio of positive reinforcement to constructive criticism high.
discuss their own clinical	• Encourage peer feedback.
performance.	• Highlight areas of counselor expertise.
	• Model openness to feedback by soliciting counselor opinions about supervision.
• Counselors may feel overwhelmed	• Provide consistent, individualized positive feedback on counselor growth.
or burned out by responsibilities.	• Provide tangible acknowledgment of appreciation.
	• Allow opportunities to “vent” frustrations about larger systems issues.
	• Discuss clinician self-care strategies.

Novelty of intensive clinical supervision

Unlike trainees in advanced degree mental health training programs, addiction counselors typically do not receive intensive clinical supervision that requires detailed discussion and review of therapy session content and individual client progress. We oriented counselors to these aspects of supervision and set expectations that they would critically evaluate their own performance in the groups and be open to receiving constructive criticism. During the weekly two-hour group supervision, listening to tape recorded portions of the group sessions was crucial. Counselors sometimes genuinely believed they were correctly presenting information to clients, but review of the tapes indicated otherwise. For example, counselors sometimes attempted to “correct” a client's subjective evaluation of the impact that a particular thought or activity had on the client's mood. Such instances allowed us to clarify the counselors' role in instructing clients on the use of different therapy tools (e.g., rating scales to evaluate mood) and to help them understand that the manner in which clients applied these tools would vary. Although we found audiotape review extremely valuable, some counselors had a somewhat different opinion. They reported at the one-year focus groups that the audiotaped sessions produced limited understanding of treatment delivery and suggested that videotaping would be more helpful in capturing the quality of the treatment sessions.

The supervision was more highly structured and didactic than that typically provided for supervisees at a master's or doctoral level. We routinely posed hypothetical situations and used role plays to evaluate counselors' ability to deal with different clinical scenarios (e.g., How would you deal with a client who says, “Doing the practice just makes me feel bad”?). This enabled us to evaluate specific counseling skills and model alternative ways of handling challenging group situations. Role plays also provided a vehicle for helping counselors adapt manual scripts to fit their own style while maintaining fidelity to session content. At the one-year follow-up focus group, the counselors agreed that role plays were the most valuable activity they engaged in during supervision.

We also provided specific recommendations to support counselors' success in learning the material, such as instructing them to set aside specific time each week to review upcoming session material and meet with their co-facilitator and to make notes after sessions of questions/difficulties that arose so that they could bring them to weekly supervision. Finally, we used supervision to model behaviors and attitudes that counselors would ideally display in the group, such as collaborative problem solving, timeliness, follow-up on action items, willingness to receive feedback (e.g., on what counselors felt was more and less helpful in supervision sessions), and open communication.

Counselor reluctance to discuss clinical performance

Counselors initially had difficulty discussing their performance during supervision. This seemed to stem from a combination of lack of experience in consciously trying to remember detailed information and reluctance to acknowledge difficulties encountered in delivering session material or working with clients. To encourage open discussion of counselor performance, we provided ample positive reinforcement and solicited feedback from other counselors prior to providing constructive criticism. The latter has obvious relevance for the development of peer supervisory skills and demonstrates respect for counselor opinions. We also consistently praised counselors for engaging in critical self-evaluation and requesting help in understanding and presenting the treatment material. We modeled willingness to receive feedback by asking counselors about positive and less helpful aspects of supervision and for their suggestions on how to improve the supervisory experience. Another factor that contributed to a positive, collaborative supervisory relationship was open acknowledgment of the counselors' experiential knowledge of addiction and recovery.

Counselor burnout

Over time, the counselors became increasingly invested in GCBT in response to seeing many clients significantly improve, in both depression and substance use outcomes. At the time of the one-year focus group, all the counselors reported interest in continuing to deliver the treatment. Nevertheless, the considerable demands of leading a group combined with their heavy standard caseloads resulted in notable counselor burnout at various times during the study. Counselor burnout can be defined as difficulty in appropriately performing clinical tasks due to personal discouragement, emotional/physical drain, and apathy toward system stress (Lee et al., 2007). Our counselors exhibited behaviors indicative of burnout, such as poorer outreach with absent group members, failure to consistently follow up on recommendations for upcoming sessions, and frequent expressions of stress regarding the time needed to prepare for and lead the group.

The following strategies were helpful for dealing with counselor burnout. Positive feedback was given regularly over time and included highly specific observations on ways in which counselors' skills progressed and improved (e.g., becoming better at helping clients understand the difference between thoughts and feelings). Acknowledging professional development was also key. To enhance the professional value counselors associated with learning GCBT, we awarded certificates of proficiency in leading the interventions, increased the complexity of concepts discussed in supervision over time (e.g., case-conceptualization skills), and included counselors in presentations to professional audiences. Provision of appropriate collegial emotional support included allowing counselors to occasionally express their frustration about other aspects of their jobs (e.g., caseloads, state audits). This also demonstrated that we understood the context in which counselors were learning and implementing GCBT. Discussing and encouraging counselor self-care (a topic not typically addressed in their work setting), including using CBT skills for this purpose, were also valuable in helping with burnout.

Supervising Addiction Counselors: Developing “Non-Specific” Counselor Competencies

Table 3 presents challenges faced in helping counselors develop or refine general clinical competencies considered crucial for treatment success.

Table 3. Supervising Addiction Counselors in Group CBT: “Non-specific” Counselor Competencies.

Challenges	Solutions
Counselors may use a confrontational	• Provide education about depression to correct misattributions of client behavior.
or overly disclosing approach with	• Review basic counseling skills for building rapport.
clients.	• Draw attention to the impact of a collaborative approach on client outcomes.
	• Address boundary issues in treatment.
• Counselors may have limited	• Routinely review results of depression assessments for each client.
experience in formally monitoring
client progress.
• Counselors may need to improve	• Address common challenges such as managing conflict between clients and responding to client
group management skills.	emotional displays.
• Counselors may need to improve	• Model co-leading skills in supervision.
co-leading skills.	• Explicitly check in on co-leader relationships.
• Counselors may place a low priority	• Provide a rationale for active outreach.
on active client follow-up.	• Provide a structured method for client follow-up.

Counselor approach to clients

The collaborative approach of GCBT was often directly at odds with the more confrontational stance addiction counselors were accustomed to using with clients (and that is common in substance abuse treatment settings). Counselors were particularly prone to using confrontation when they assumed features of clients' depression (e.g., apparent lack of engagement) reflected factors such as personality pathology or a lack of readiness or motivation to address depression or substance use problems. We found it helpful to continually review the ways depression could affect a client's approach to therapy, such as negative self-evaluations and hopelessness leading to beliefs that treatment and homework assignments would not help. Modeling a CBT approach, we encouraged counselors to catch and challenge themselves when making negative attributions about a client's behavior in a session (e.g., that the client was being manipulative). To further encourage a less confrontational approach, time was spent in supervision reviewing basic rapport-building techniques such as active listening (paraphrasing), reflection of feelings, summarizing, and monitoring nonverbal reactions.

This issue was complicated by the fact that many BRIGHT counselors were also in recovery and had probably directly experienced the confrontational approach themselves when receiving treatment. Thus, our message that a collaborative tone is more effective than confrontation may have created cognitive dissonance and resistance toward adopting the new approach. It was important to normalize the discomfort counselors initially felt in “trying on” a new way to lead a group. The most effective factor in convincing counselors that they did not need to cajole clients into changing or use a “tough love” approach was their direct observation of clients' behavior in the group. They observed that a more collaborative, problem-solving style of interaction that involved meeting clients where they were on a continuum of change typically led to greater participation in the group rather than a worsening of engagement problems. At the time of the one-year follow-up focus group, counselors reported that clients enjoyed GCBT, because “we don't use the rude approach of the ‘regular’ counselors.” They said that clients felt more comfortable sharing in GCBT groups than they did in the other addiction group treatment sessions they attended.

Another challenge was over-disclosure of counselors' beliefs or attitudes during group sessions, including discussion of their religious beliefs or personal life. Counselors also sometimes used humor excessively with group members. They tended to view these approaches as effective general treatment strategies that had been used without incident in other treatment contexts. Care was taken to respect counselors' previous experiences, but a clear rationale was presented for why such approaches were less helpful in the context of GCBT for depression (e.g., indiscriminate or poorly timed humor could negatively impact rapport with depressed clients because they might feel teased). We emphasized the need to be strategic in the use of self-disclosure and for counselors to think through how this could potentially facilitate group members' progress in treatment before using this strategy.

Limited experience using assessment measures

The counselors were not accustomed to using standardized methods of assessing client functioning and progress during treatment. Therefore, supervision always included a review of client depression levels, including Patient Health Questionnaire-9 (PHQ-9; Kroenke, Spitzer, & Williams, 2001; Spitzer, Kroenke, & Williams, 1999) scores, which were obtained at every other treatment session. We discussed graphs of PHQ-9 scores so that counselors could receive immediate visual feedback on changes or consistencies in a client' depression level across sessions. Counselors' motivation increased when they observed decreases in client depression over time. Discussion of client depression levels also reinforced the rationale for active outreach to clients who appeared less engaged or motivated.

Enhancing group management skills

Effective group management skills, such as integrating new members into the group, monitoring and intervening in violations of group rules, helping clients manage strong emotions, and keeping group members informed in advance of planned counselor absences or changes in group leadership, were an ongoing focus of supervision. For example, we found that counselors often let conflicts between group members escalate without intervening because they believed that this helped clients fully express their feelings toward one another and work through them. Some counselors were uncomfortable with actively responding to clients' expressions of strong, negative feelings (e.g., openly acknowledging clients' crying), which may have resulted from inexperience in dealing with very depressed clients. Such challenges required teaching counselors that active intervention (e.g., helping clients label feelings and learn effective ways to express feelings toward others) was a more effective strategy for helping clients manage their feelings than passive approaches.

Enhancing co-leading skills

The BRIGHT groups were co-led by teams of two addiction counselors; however, most of these counselors had limited co-leading experience. In supervision, we modeled co-leading skills such as direct communication when talking with counselors and (in the case of jointly led supervision) with each other. We also addressed the importance of setting aside time for counselors to discuss the upcoming group, the division of labor between counselors, and evaluation of the co-leading experience. The latter included frank discussions of feelings of resentment, discomfort, or disagreement that periodically arose between the counselors about ways in which the group was being led.

Prioritizing client follow-up

In implementing the BRIGHT interventions, we emphasized the importance of promptly reaching out to clients who showed up late or missed sessions. Counselors initially disagreed with this and viewed it as “babying” clients. However, after we provided education about the effects of depression on clients' ability to consistently engage in treatment, counselor follow-up improved. Nonetheless, some difficulties related to client follow-up continued to occur. For example, counselors sometimes waited until the last minute to contact missing group members prior to the next group session. We therefore found it helpful to explicitly instruct counselors to make “to do” lists of items to complete (including client contacts) prior to the next supervision session, and we followed up on these action items to maintain counselor accountability. We also regularly reviewed guidelines for following up on instances of suicidal ideation expressed by clients either verbally in a session or on the PHQ-9.

Supervising Addiction Counselors: Developing CBT-Specific Counselor Competencies

Table 4 presents challenges faced in helping counselors develop CBT-specific competencies.

Table 4. Supervising Addiction Counselors in Group CBT: Counselor CBT Competencies.

Challenges	Solutions
• Counselors may have limited	• Use repeated role-plays to practice explaining the CBT model to clients.
preexisting knowledge of the CBT	• Teach counselors to use behavioral principles in group sessions.
model.	• Over time, extend understanding of CBT by individualizing treatment.
• Counselors may have misconceptions	• Introduce concepts of collaborative dialogue and guided discovery.
of CBT as a class and the counselor's	• Model and role play Socratic questioning and other specific techniques.
role as lecturer.
• Counselors may have difficulty	• Provide a rationale for structured treatment
following a structured, manualized	• Frame initial reliance on manual scripts as an aid to delivering treatment as intended.
intervention.	• Discuss strategies for time management of session material.
• Counselors may lack experience in	• Review counselor follow-up of client homework at each supervision session.
assigning homework to clients.

Limited prior knowledge of CBT

Most counselors had only a cursory understanding of CBT prior to being trained. The two-day training exposed them to basic CBT theory, but throughout supervision we still routinely reviewed the counselors' ability to explain the CBT circle and to correct client errors in explaining it by listening to audiotaped portions of group sessions. To solidify their understanding of basic behavioral principles, we modeled the use of learning principles and encouraged counselors to use them. We reviewed concepts such as reinforcement, repetition, planned ignoring as an alternative to verbal punishment for unwanted behavior, and giving clients opportunities to summarize and process material in different ways (e.g., “What did you learn from this exercise?”).

As counselors' comfort in leading the basic BRIGHT group exercises increased, we focused on ways they could personalize aspects of the treatment for individual group members. Examples included identifying and focusing on each client's key cognitions and behaviors (e.g., making notes after the sessions and referring to them in subsequent sessions), tailoring exercises to individual clients in terms of their level of comprehension (e.g., spending extra time gathering examples of automatic thoughts with clients who are struggling with this skill), integrating information learned from practice review into the session, and using data from PHQ-9 assessments to guide approaches to specific exercises (e.g., highlighting the importance of sleep hygiene techniques for clients complaining of disrupted sleep). As supervision progressed, we also introduced basic case-conceptualization skills, including the relationship between core and intermediate beliefs and automatic thoughts and using client historical information to develop hypotheses about core beliefs and reinforcement experiences that were relevant to presenting problems.

Misperception of CBT as a “class”

Initially, counselors were more likely to give direct advice rather than working in a more collaborative way with group members. We actively discouraged counselors from lecturing at clients or even referring to the group as a “class” (which could promote a less interactive style), and used multiple modeling examples and role plays in supervision to teach techniques that would facilitate greater client participation and interactive dialogue (e.g., Socratic questioning, downward-arrow technique). It was useful to explain that when counselors solicit answers from clients rather than suggest them (e.g., allowing clients to draw connections between thoughts, emotions, and behaviors for themselves), client learning is facilitated, client self-efficacy is enhanced, and client options are emphasized. The counselors reported at the one-year focus group that they thought the clients liked GCBT in part because it provided more opportunities to express themselves than the other addiction treatment groups they attended did.

Difficulty following a structured treatment

We spent considerable time helping counselors adjust to the structured nature of GCBT. We discussed why session agendas were used (e.g., they provide a “road map” that gives focus and direction to a session), how to pace session activities (e.g., having counselors take turns “watching the clock” during the session to keep on track), what counselors could do if they ran short on time, and how to incorporate client concerns into existing agenda items. We also strongly encouraged counselors to rely on scripts provided in their BRIGHT group leader manuals when initially introducing exercises/discussions in the group. We normalized reliance on the scripts as an expected part of the learning process and as important in delivering the treatment with fidelity. This helped prevent counselors from viewing the use of the scripts as an indication of lack of expertise. Over time, role plays during supervision provided a vehicle for counselors to adapt the scripts to fit their own style while maintaining fidelity to session content. At the one-year follow-up focus group, the counselors said that they enjoyed the structure of the manuals because it kept them on track. They also reported that the manuals were empowering to clients because they provided “tools” to take away from the treatment session and gave clients an additional chance to express themselves (through the writing exercises in the workbooks). However, some counselors expressed the concern that certain group sessions included too much information to present and others described having had the mistaken belief early on in the group work that they were supposed read the manuals verbatim.

Assigning client homework

Like many beginning CBT therapists, counselors often neglected review of client homework assignments during the early rounds of leading GCBT. We modeled the importance of this aspect of CBT through our regular discussion in supervision of group members' homework assignments. Iin group referred to homework as “practice” given the negative connotations the former term could have for some group members. We helped modify counselors' perceptions of homework review as “correcting client mistakes” in order to decrease direct advice-giving to clients on ways to increase adherence. Homework review was instead framed as an opportunity to reinforce client learning of key session concepts and also as a forum for utilizing CBT principles (e.g., group brainstorming and active, collaborative problem solving) to overcome obstacles.

Discussion

Our previously published findings demonstrated that addiction counselors with limited educational and clinical experience in ESTs could be trained to deliver GCBT for depression, a prototypic EST, with a high degree of treatment fidelity (Hepner, et al., 2011; Hunter et al., in press). In addition, clients receiving BRIGHT demonstrated significantly fewer depressive symptoms at 3 and 6 month follow-ups and significantly less problem substance use at a 6 month follow-up than clients receiving usual care (Watkins, et al., 2011). Our work is consistent with other models of EST dissemination that address multiple factors influencing treatment implementation. For example, our observations parallel Stirman and colleagues (2010) Assess and adapt, Convey basics, Consult, Evaluate, Study outcomes, Sustain (ACCESS) model in recommending interactive, experiential training, ongoing supervision/consultation, and adaptation of treatment approaches to address the unique attitudes, needs, and skills of counseling staff. Similarly, our study confirms the Consolidated Framework for Implementation Research (CFIR) model (Damschroder, et al., 2011) in noting that factors related to the intervention itself, the setting, implementation processes, and the individuals involved will influence the success of dissemination efforts.

Although our focus was on co-occurring disorder treatment, we utilized the kind of intensive initial training and ongoing clinical supervision recommended in the limited literature on training addiction counselors to use ESTs for substance use problems. For example, Morgenstern, Morgan, McCrady, Keller, and Carroll (2001) found that addiction counselors could competently deliver a CBT coping skills program after undergoing 35 hours of didactic training, clinical case training with three to four clients, and two-hour weekly supervision. Sholomskas and colleagues (2005) found that independent ratings of adherence and skill in implementing a manualized CBT drug abuse intervention were higher for clinicians who participated in a three-day didactic seminar plus clinical case supervision than for those who read the treatment manual only or who read the manual and participated in a web-based interactive training course. Guided by these findings, we utilized a two-day didactic and experiential training session followed by intensive two-hour weekly clinical supervision that incorporated audiotaped session review.

However, our dissemination model differs in important ways from previously published models. It is the first, to our knowledge, to provide specific recommendations for training addiction counselors to use CBT to treat co-occurring disorders, a common yet frequently overlooked problem in substance use treatment. The extra demands placed on counselors to learn a new treatment approach (EST) and apply it to clinical problems they have less experience treating (e.g., depression) led us to expand upon additional considerations and strategies in our model. For example, the counselor selection process is a central element. We believe that enthusiasm for learning new treatments, willingness to work outside one's “comfort zone,” and openness to feedback are critical characteristics to seek in counselors, since they will likely increase buy-in to ESTs, which in turn will increase the likelihood that the ESTs will be implemented as intended (Stirman et al., 2010).

Our model also differs from others in detailing the ways in which traditional clinical supervision can be modified to address the needs of addiction counselors. We include explicit socialization of counselors to the process and expectations of clinical supervision (e.g., the ability to discuss sessions in detail) and implementation of strategies for encouraging critical self-evaluation (e.g., providing high levels of positive reinforcement and modeling openness to self-evaluation and feedback). Another difference from previously published models is our recommendation to focus on both the development of EST-specific competencies and basic counseling skills (e.g., group management skills, development of co-leading relationships, monitoring client progress, active client outreach) during the training and supervision period. We found significant variability in counselors' mastery of basic counseling skills prior to training, and we considered it important to strengthen this skill set in order to provide a more solid foundation of clinical competencies upon which to build EST-relevant skills.

A final important aspect of our model is the attention paid to understanding the role that larger systems issues, such as counselor burnout, play in the implementation process. Addiction counselors working in public sector treatment settings carry heavy caseloads, treat challenging clients, frequently work in understaffed facilities, and have large paperwork burdens (Carise, Love, Zur, McLellan, & Kemp, 2009). Burnout is common and turnover rates are high (McLellan, Carise, & Kleber, 2003). Thus, the implementation of a new intervention that is challenging and time-consuming to learn cannot take place in a vacuum where the heightened stress that counselors face in fitting new responsibilities and demands into already busy schedules is not acknowledged. We therefore highlight establishing strong, visible administrative support prior to intervention implementation. Our model also encourages periodic discussion and problem solving in supervision of burnout-related issues to help counselors consistently deliver quality care over time. A large-scale survey of addiction counselors has found that quality clinical supervision appears to play a role in reducing emotional exhaustion and burnout (Knudsen, Ducharme, & Roman, 2008).

Although our work addresses the notable gap in the EST dissemination literature on specific guidelines for effectively teaching clinical skills to front-line treatment providers, there may be limitations to the generalizability of our conclusions. Our work was conducted with a single multi-site organization and a relatively small group of counselors in a single state. It is also important to note that the results are based on the clinical supervision team observations, and feedback from a counselor focus group and interviews and meetings throughout study with addiction treatment supervisors and administrators. Given the qualitative nature of the data, the results are subject to bias. For example, although we received both positive and negative feedback about the training and supervision during the focus group with the trained counselors, the group setting may have introduced demand characteristics. In addition, our model has limitations. Most notably, the costs associated with the intensive initial training and ongoing clinical supervision recommended here may be readily supported when discretionary grant funds are available but difficult to continue after such funding ends. It may also become increasingly difficult to implement an EST with the loss of trained staff and the high turnover rates.

We recognize that some organizations may need to make adaptations of some of our recommended training practices. For example, in lieu of the initial intensive training we describe, a shorter introductory training could be supplemented with a complete role play of selected group sessions (with supervisors or peers playing group members) before counselors lead an actual group session. Also, once counselors are proficient in leading the interventions, they could become involved in the training process (which provides excellent peer modeling for new group leaders and additional professional development opportunities for advanced group leaders). Finally, if a counselor is unable to attend an intensive pre-group training, it would be advisable to have him or her observe several group sessions from behind a one-way mirror, with sufficient preparation from a supervisor (e.g., advising the counselor to pay attention to what counselors, rather than clients, are doing; to follow the manual and anticipate what counselors in session will do; etc.). Future research should explore alternative training modalities that may further facilitate EST dissemination by making implementation more cost-effective and flexible. Online training is an effective and appealing way to present new treatment techniques to addiction counselors (Larson, et al., 2009), and it can be effectively combined with more traditional teaching methods (e.g., use of written materials, face-to-face training) in a “blended learning” strategy (Cucciare, Weingardt, & Villafranca, 2008). Sholomskas and colleagues (2005) found that clinicians who participated in either web-based or traditional seminar-based training demonstrated steadier or more improved EST-related skills over time than clinicians whose preparation consisted solely of reading a treatment manual.

We further recognize that the intensive clinical supervision we provided may not be feasible for many organizations. Although ongoing supervision is likely to be critical, the proper amount necessary to establish and maintain competence remains an empirical question. It would therefore be fruitful to explore less costly supervision models such as peer supervision.

Another potential limitation of our model is that it focuses on the training and dissemination of a manualized treatment. Some have suggested that a more realistic goal for disseminating ESTs in settings where well-developed systems of addiction care already exist is to train clinicians in a set of specific evidence-based skills (Glasner-Edwards & Rawson, 2010). An additional issue concerns ongoing monitoring of treatment fidelity. Routinely implementing a system for assessing counselor adherence and competence in community settings may be challenging, as this is usually done in the context of research studies. If a system for assessing adherence/competence is adopted, counselors must be socialized to the rationale and value of such evaluation, and it must be framed as a tool to assist their skill development.

Important challenges and unresolved issues pertaining to the training and supervision of addiction counselors to implement CBT for co-occurring depression remain that are beyond the scope of this article. Prominent among these is the scope of practice for addiction counselors. Certain training, certification, and regulations of addiction counselors may prohibit their engaging in treatment of co-occurring mental health disorders. For example, the Maryland Department of Health and Mental Hygiene Board of Professional Counselors (n.d.) specifically indicates that alcohol and drug counselors “may not diagnose and treat mental and emotional disorders.” Given the emerging evidence that addiction counselors with adequate training and supervision can competently deliver depression treatment, there may soon be a firmer foundation on which to advocate for expanding the role of addiction counselors to treat a broader range of mental health problems. On the basis of our experiences and the research literature to date, we would advocate for this role expansion only in the context of a highly structured and professionally supervised treatment program such as the one described here or possibly through a specialized certification process.

The development of a program of specialized certification to ensure a standard of competence that would be recognized by government agencies (and insurance companies) would be helpful for promoting ongoing dissemination of ESTs such as CBT. Such certification could be offered by a nationally based professional organization, board or association, such as the Association for Behavioral and Cognitive Therapies. Certification, especially if linked to job opportunities for counselors, would increase the attractiveness of learning ESTs, and it might offer the additional benefit of teaching more effective skills to address client issues within the counselors' traditional scope of practice.

Conclusions

It is our hope that the guidelines and model we have provided for training and supervision of addiction counselors in GCBT will facilitate wider dissemination and implementation of ESTs in SUD treatment settings, including those directed at the treatment of co-occurring disorders. Our recommendations for intensive training and ongoing supervision are based on two clinical trials in which training and supervision were provided over several years. We recognize that the variability in financial and staff resources in public sector SUD treatment settings necessitates some flexibility and modification in the implementation of these recommendations. Further research on implementation efforts and on the unresolved issues outlined here can assist in identifying the requirements for training addiction counselors to deliver ESTs.

Glossary

AOD

alcohol and other drug

BRIGHT

Building Recovery by Improving Goals, Habits, and Thoughts

CBT

cognitive behavioral therapy

EST

empirically supported treatment

GCBT

group cognitive behavioral therapy

NSDUH

National Survey on Drug Use and Health

PHQ-9

patient health questionnaire

SUD

substance use disorder