Facilitators and Barriers Associated With Implementation of Evidence-Based Psychotherapy in Community Settings

IPSRT takes a two-pronged approach to ameliorating current mood symptoms and preventing manic and depressive relapse. First, IPSRT focuses on regularizing patients’ social rhythms (daily routines) and titrating patients’ level of activity. Second, using strategies drawn from interpersonal psychotherapy (IPT) for unipolar disorders, IPSRT addresses the resolution of current interpersonal and social role problems. Outpatients who receive IPSRT have better outcomes than comparison groups (1,9).

We recruited eight clinicians and four administrators or supervisors who had participated in IPSRT training between 2006 and 2009. They were employed by four moderate to large nonprofit community mental health center outpatient clinics located in urban or suburban communities in different parts of the country. We obtained a list of IPSRT training participants from the clinics and invited randomly selected individuals to complete a brief semistructured phone interview. After obtaining informed consent, we conducted interviews with the participants, beginning with open-ended questions about IPSRT implementation and followed by probes and questions designed to explore facilitators of and barriers to IPSRT implementation. The interviewer took detailed field notes, and the interviews were audio recorded to facilitate clarification and verification of field notes. Study procedures were approved by the University of Pittsburgh Institutional Review Board.

Consistent with template coding (10), preliminary codes were developed on the basis of common themes in the implementation literature and were reviewed and refined during coding by research team members. Recurring themes were identified and categorized by an initial coder and were independently reviewed and confirmed by a second coder. Themes were aggregated into larger themes by the research team, and disagreements were resolved by consensus.

The interviews revealed substantial differences in the extent to which the clinics successfully implemented IPSRT following the training. Participants from two clinics (hereafter referred to as successful implementers) described clinicians’ ongoing use of many IPSRT components, whereas participants from the other two clinics (hereafter referred to as unsuccessful implementers) reported very limited and quickly diminishing use of IPSRT subsequent to the training. Across both types of clinics, the following five major IPSRT implementation themes emerged: pretraining familiarity with IPSRT, administrative support for implementation, IPSRT model fit with usual practice and clinic culture, implementation team and plan, and supervision and consultation. Below, we provide exemplars of these themes and discuss each one in greater detail, using examples from specific clinics when appropriate to contrast clinics that were successful versus unsuccessful implementers.

Pretraining awareness of IPSRT was a common theme. Participants at successful clinics frequently described having some familiarity with IPSRT before the training, whether through workshops, readings, or conferences. One clinician “had an ‘informal training’ with [her supervisor] three months before the formal training,” and another “had started doing readings and going to meetings about IPSRT in the 1990s and started using it in 2001.”

In contrast, lack of familiarity with IPSRT was often mentioned by participants from unsuccessful implementers. One administrator related, “Other than the psychiatrists, few [clinicians] were [familiar] with IPSRT before the training, and there was resistance. . . . They did not want to incorporate the approach.” One administrator at another unsuccessful implementer contrasted implementation of IPSRT with the clinic’s successful implementation of a different intervention “There was prior familiarity with the [successfully implemented intervention)],” he said. “There was a foundation to build upon.”

Participants often stressed that administrative support was an integral component of successful implementation. Participants from successful implementers commonly discussed the importance of having a clinic champion—someone at the clinic who “represented” IPSRT and whom clinicians could approach with questions or concerns. A clinician related, “I can’t imagine [successfully implementing IPSRT] without all of the administration on board,” and an administrator observed, “It helps to have someone in administrative power that is sold on the treatment and wants to make it happen.” A supervisor said her clinic’s “head doctor was very interested in IPSRT and . . . pushed for it,” and another supervisor noted, “Clinicians would stop her in the hallway to talk about [IPSRT] . . . because they knew she was helping to facilitate the use of IPSRT.” Administrators from successful implementers also supported IPSRT when training disrupted clinicians’ normal work schedule. One clinician recounted, “[I] lost hours with patients, took days off for the training, and was not expected to make up days missed.” Such comments were absent during interviews with participants from unsuccessful implementers.

Another common theme at both successful and unsuccessful implementers was IPSRT’s fit with usual clinical practice. Many clinicians from all sites commented on the straightforwardness of IPSRT’s social rhythm component. As described by one clinician, “It just made sense to me and I ran with it.” However, participants also discussed how IPSRT’s interpersonal component was not as easily grasped or used. One clinician from a successful implementer described “glossing over the IPT part . . . and using the social rhythm part more.” Another clinician admitted, “The social rhythm part stood out for [me] more than the IPT part, and [I] remembered more of it [from the training].” An administrator from a successful implementer noted, “The IPT part of IPSRT did not take hold [among the clinicians], but people were using the social rhythm part.”

Participants also discussed challenges implementing components that did not naturally fit with current activities. For example, many described how IPSRT’s physiological rationale (stabilizing circadian rhythms) was a novel concept for both clinicians and patients and was therefore more difficult to teach to patients. Participants were also concerned that some treatment activities could potentially overwhelm patients who—by the very nature of bipolar disorder—might be fragile. One clinician from a successful implementer “loved the Social Rhythm Metric (a self-report form central to IPSRT) but found it a bit cumbersome. . . . The grids and mood rankings can be a little intimidating to patients.” Another clinician discussed difficulty in getting patients to understand and take steps to stabilize their circadian rhythms. “[Patients] would hear it and seem to get it,” the clinician reported, “but it’s hard to sustain. . . . There needs to be a better repertoire of materials to help patients understand the physiological aspect.”

Many participants also reported that their clinic valued the use of evidence-based interventions for mood disorders. One clinician from a successful implementer shared, “[Our clinic] was built around the goal of using EBPs (evidence-based practices) like IPSRT,” and another recalled that soon after being hired, she was “presented IPSRT as an EBP for bipolar [disorder] and told to read [the treatment manual], learn it, and try it.” An administrator at a successful implementer described the clinic as “a mood disorders clinic that focused on using best practices for mood disorders.”

The importance of having a plan for implementation and staff to support it was another common theme. Participants at unsuccessful implementers described relying primarily on training, with less attention given to developing a team approach and a logistical foundation to support implementation. As an administrator from an unsuccessful implementer candidly stated, “There was not enough groundwork before, during, and after training [about] how IPSRT was going to be incorporated into standard practice.” An administrator from another unsuccessful clinic lamented, “No one was interested in [an implementation team] after [the training],” adding that without such a team, “there was no institutional togetherness. . . . Clinicians did what they wanted to do.”

In contrast, at successful implementers, it was common for motivated, interested clinicians to meet before training to discuss the development of an implementation team and implementation plan. Some clinics hired only clinicians with prior experience in effective therapies. As one clinic administrator revealed, “We recruited clinicians specifically with training in EBPs for mood disorder.” One supervisor commented, “Clinicians came to us because they wanted to learn IPSRT and were excited about it.” Successful implementers also discussed the need for a clear implementation plan. A clinician at a successful implementer related, “We had a plan going in [to the training] and knew how we’d be learning [IPSRT] and using it with patients.”

Participants from successful implementers commonly discussed two additional implementation activities: marketing the treatment model to everyone in the clinic and having a referral plan in place before IPSRT training. For example, one clinician said, “IPSRT was marketed [in the clinic] as a new hope.” With respect to referrals, participants described a shared awareness and responsibility among all clinicians—regardless of discipline—for maintaining the flow of case referrals. As one clinician discussed, “Others not treating bipolar [disorder] or using IPSRT were very good about . . . referrals [for bipolar disorder] that could be passed along.” Another clinician related, “Cases came to [me] through internal referrals from the psychiatrist, and patients knew they were coming to [me] for IPSRT.” One administrator provided an overview, describing how “[his clinic] had a steady stream of referrals, and there was collaboration among social workers, psychologists, and psychiatrists, who would talk and hand off cases.”

All participants discussed the importance of adequate consultation after training. The participants who received adequate consultation found it extremely useful and those who did not wished it had been available. One clinician from a successful implementer related, “[Consultation] got us to stay true to the model and bring us back if we were going astray.” A supervisor asserted, “[The trainers] should follow up with [the trainees] and do consultation or some sort of follow-up.” An administrator from an unsuccessful implementer admitted, “The consultation piece wasn’t there, and that is where the initiative was lost.” Common among successful implementers were built-in support mechanisms and routine supervision and meetings. In contrast, a clinician from an unsuccessful implementer discussed wanting an “in-house, forum-style” venue focused on “checking in on how things are going, seeing if there are any challenges, and troubleshooting [the challenges].”

In this study of IPSRT implementation in outpatient clinics, we identified five major themes related to effective implementation, each illustrated with concrete examples provided by participants of strategies that facilitated successful implementation. Ours is one of the first studies to examine the implementation of an evidence-based psychotherapy in community mental health center outpatient clinics. Many of the themes we identified are consistent with those identified in studies of other interventions and settings (5,11–13). The description of specific strategies for success is an important contribution for those seeking to turn concepts into concrete activities that support implementation.

The participants discussed the importance of both management support of implementation and the intervention’s fit with the clinical setting, themes commonly mentioned in the implementation literature (6,11). The importance of a good fit between the intervention and the culture of a clinical setting for successful implementation has been reported for other interventions (6,14), as has the need for ongoing supervision and consultation for recent trainees as they begin using and seeking to master the intervention (15). The participants’ comments were consistent with these findings and also provided some concrete examples of successful and unsuccessful approaches. A number of comments, however, were about the importance of pretraining familiarity with the intervention. Many participants felt that pretraining familiarity was important for later successful implementation, a factor not as commonly identified in previous studies of implementation success. We were unable to examine the extent to which pretraining familiarity played a causal role in successful implementation or reflected clinic characteristics that were not discussed but that were associated with more successful implementation. Nevertheless, attention to pretraining familiarity may be useful for future implementation efforts.

Notably, however, the discussion of fit focused on intervention components, with the more concrete and behavioral social rhythm components being implemented more commonly than the interpersonal therapy component. Other studies also suggest that even when interventions are implemented successfully, some components are more likely than others to be implemented (14). For example, the social rhythm component was adopted more successfully than the interpersonal therapy component. Several factors may have facilitated greater use of the social rhythm components, including the handouts and other documents about this component that were made available to clinicians. In addition, the social rhythm component is less complex than the interpersonal therapy components, so clinicians have an easier time trying aspects of the social rhythm component with patients and assessing the results (6). As we seek ways to better implement effective psychotherapies in outpatient clinics, this finding suggests the need to consider differential levels of implementation support for the components of therapy. It also indicates the need to devote greater attention to identifying the components of existing evidence-based psychotherapies that are the most effective.

“Train and pray” has long been recognized as an approach unlikely to result in effective implementation (12,13). A common theme of our results was the importance of clinics having a planned implementation approach and an identified team to carry it out. Notably, several recommendations were related to making staff aware of the intervention and encouraging referrals for the intervention, illuminating the importance of providing sufficient patient flow for clinicians providing the intervention, especially immediately after training, when clinicians are working to master new techniques.

This study must be considered within the context of its limitations. We focused on a relatively small number of community mental health center outpatient clinics and a single therapy. Interviewing several individuals from each clinic allowed us to identify general issues within the clinics. However, we were unable to assess variation in implementation plan or experience within clinics, nor do we know how involving more clinics would change our findings. All participating clinics sought IPSRT training, suggesting substantial interest in IPSRT, and we do not know to what extent our findings would generalize to other psychotherapies, to other outpatient settings providing psychotherapy, or to clinics in states that mandate use of evidence-based practices.

Despite these limitations, our study suggests a road map for community mental health center outpatient clinics seeking to successfully implement an effective psychotherapy, such as IPSRT. When key factors are present, these outcome-enhancing treatments can be both implemented and sustained, even in community mental health center clinics with relatively limited resources.

The National Institute of Mental Health of the National Institutes of Health (NIH) provided support (award R34MH091319) for this study. The authors are indebted to Gina Boyd, M.L.I.S., for research assistance and assistance with the preparation of the manuscript and to the community mental health clinicians, clinical supervisors, and administrators who shared with us their IPSRT implementation experiences. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.

Dr. Frank is a consultant to Servier International and Lundbeck. She receives royalties from Guilford Press and the American Psychological Association Press. The other authors report no competing interests.