Efficacious psychosocial treatments for mental disorders are available. However, these interventions are rarely routine in settings where most people receive services (1). Lack of community-based implementation of evidence-based treatments may be due in part to challenges faced by those seeking to implement treatments, such as selecting staff to train, buffering implementation efforts to withstand the negative impact of high staff turnover (2), maintaining treatment model fidelity, and maintaining administrative support (3).
Experts suggest that successful implementation of evidence-based treatment requires the commitment of consistent, long-term financial resources (4), sufficient personnel for training and implementation (5), and support of organizational leaders (6). Despite the integral role of mental health agency leaders in the implementation process, relatively few studies have examined their perspectives. One exception documented mental health administrators' concerns about evidence-based treatment and found that administrators were worried about high startup costs; the impact of the new treatment on the larger system, necessitating large-scale buy-in, infrastructure development, and development of new organizational relationships; limited training resources; and the necessity of ongoing training because of personnel turnover (7).
Studies that have examined the role of organization leaders have not conducted personal interviews (5). To enhance our understanding of mental health administrators' perspectives on opportunities and challenges in implementing an evidence-based treatment, we conducted semistructured interviews with administrators of community mental health agencies. These agencies were implementing dialectical behavior therapy as part of a multicounty effort to improve the quality of behavioral health care provided to publicly insured individuals in eastern Pennsylvania.
Dialectical behavior therapy is a well-supported approach for treating borderline personality disorder, and it has demonstrated efficacy in improving retention in treatment and global and social adjustment in a range of populations, including adolescents with bipolar disorder (8) and individuals with substance use disorders (9). Community mental health providers are increasingly adopting dialectical behavior therapy (3), which involves an intensive training commitment.
County-level mental health administrators in four central and eastern Pennsylvania counties partnered with a large nonprofit managed behavioral health organization that is responsible for managing Medicaid behavioral health care in those counties. This partnership supported training in dialectical behavior therapy for ten provider organizations.
Administrators from all organizations participating in the implementation of dialectical behavior therapy were contacted by e-mail, provided with information about the study, and asked to participate. Thirteen administrators from nine of the ten organizations participated in the study, which was approved by the University of Pittsburgh Institutional Review Board.
Semistructured interview questions were developed from a systematic review of the literature, and they were revised according to suggestions from stakeholders—that is, experts in dialectical behavior therapy, administrators of a managed behavioral health organization, and county mental health administrators. Interview questions included the following topics: perceived progress of the implementation initiative, opinions about dialectical behavior therapy, ideas about the training protocol, and agency adjustments to accommodate treatment or training. Examples of the questions asked are "What are your impressions of dialectical behavior therapy as a treatment model?" and "What are your thoughts about the training protocol?" Interviews were conducted via phone in early 2007, before the start of training, and took approximately 45 minutes to complete.
Field notes from the interviews were reviewed by research team members to explore general topics, and a primary coder read detailed interview field notes to identify recurring themes to develop a codebook. Two coders independently coded field notes for the major domains of inquiry using qualitative data analysis software (Atlas ti), and they subsequently identified and discussed disagreements in coding until reaching consensus. In cases where the two coders could not reach a consensus, the research team discussed the item until consensus was reached. Provisional categories were finalized after a process of constant comparison in which each statement was checked against similar data and against a more inclusive category that described statements in a class. Subsequently, the research team discussed the content of each domain and refined the coding scheme by expanding, collapsing, or eliminating codes until there was consensus. Additional coding was done based on the range of responses within each domain (10).
Four themes emerged in the interviews with community mental health agency administrators regarding their organization's implementation of dialectical behavior therapy. These include opinions about dialectical behavior therapy and its fit with existing practices, resource concerns (for example, reimbursement issues, time commitment, and staff training), staff selection for training and staff turnover, and ongoing client referrals. Below we describe themes in detail and provide exemplars from the interviews.
In regard to the first theme, opinions about dialectical behavior therapy and its fit with existing practices, agency administrators were generally positive about dialectical behavior therapy, believing it has good face validity. Quotes from administrators included, "I think it is reasonable." "It is worthwhile. It's frustrating to work with borderline personality disorder—dialectical behavior therapy is practical and makes common sense." Another administrator mentioned that "The treatment was humane, and [we] liked what we heard about it. It is very worthwhile and valued."
Many administrators felt that dialectical behavior therapy would complement established services for consumers, regardless of whether they had borderline personality disorder. Administrators described their current treatment model for consumers with borderline personality disorder as a combined approach: "[Clinicians use] cognitive-behavioral therapy and [their own] theoretical orientation, [which may involve] dialectical behavior therapy and cognitive-behavioral therapy, but not one treatment philosophy."
However, several administrators were concerned about the fit of dialectical behavior therapy with the existing clinic structure and population. As one said, "There is an expectation [with dialectical behavior therapy] that clients [in crisis] could call clinicians at home. This is different than our current model … we see "on-call" as going backwards ten years. That's what they used to do, and it's ineffective." Another expressed the view that dialectical behavior therapy "would be ideal for academics, but with fee-for-service and Medicaid populations, I'm unsure if it will work." Another was concerned about "the mantra of fidelity." This administrator said, "As for fidelity—one shoe won't fit all. Dialectical behavior therapy is good to work with for some, but not all, circumstances. I will encourage staff to use good judgment. I don't want to take fidelity too much to heart."
The second common theme was resource concerns. Participants were concerned about the availability of sufficient resources, including personnel, to support implementation. They were concerned about sufficient funding to adequately support adoption and implementation in the current reimbursement model, because clinician training (for example, reading, attending training, and completing homework) would be not be reimbursable and would keep clinicians from direct service provision. One administrator related that "a concern [is] the big commitment, and that's why some other providers … chose not to participate in the training." Another administrator observed, "[The] biggest drawbacks at this time being the time-intensive model, readings, and caseload; balancing is tough." Several were concerned about losing money, illustrated when one administrator said, "There is an expectation of fidelity to the model that doesn't fit with payment structure. Training, weekly meetings, data collection, a lot is not billable … it all adds up."
The third common theme was staff selection for training and staff turnover. Administrators commonly described carefully selecting clinicians for dialectical behavior therapy training groups, weighing factors such as whether they volunteered, clinician seniority, credentials, familiarity with dialectical behavior therapy, staffing needs, and diversity. This selection criterion was described by an administrator seeking to "balance [the staff who receive training by] race, gender, and [clinical] discipline." Many were concerned about staff turnover, and they described selecting senior staff more likely to remain with the agency. One participant said, "Everyone selected for training started with the program and is committed to the organization. All of them have ownership and value the input in the agency." Another administrator mentioned, "First we went to people who were stable with the agency for a while and excited to participate. These were people who had been with the agency for five years or more, were mature, clinically licensed, and grounded in the company—it was clear that they had a relationship with the company." Other administrators "selected those for training who were open to it the most, interested … excited" or "selected [clinicians] on various criteria, such as having a master's degree, doing therapy, leading groups, and seeing lots of clients."
Some administrators expressed concern that training more experienced clinicians would make them unavailable for supervision and mentorship. Concerned with retention of experienced clinicians, some hesitated to train clinicians because it might result in their leaving the agency. One participant said, "We have labor problems. There is a high demand and low supply of master's-level clinicians. Young people are interested in attending the training to advance their careers but will then move on after they've received it, which doesn't help the agency."
Turnover of trained clinicians would also compromise implementation; one administrator discussed a previous initiative where "implementation was going well, but this was lost due to high turnover." Several administrators described the importance of ongoing training to accommodate staff turnover, exemplified by the administrator who said, "There also is no provision for training new people once the training is over. We don't like that [the trainers are] out after that instead of providing training on an ongoing basis."
Ongoing client referrals for dialectical behavior therapy was the fourth common theme. Agency administrators wondered whether a sufficient number of clients who were appropriate for dialectical behavior therapy would be seen. As one wondered, "Will we receive more referrals of this type? We see some of these patients now but would need to see more after training [to make participation in the training worthwhile]." Many administrators believed that securing an adequate referral stream was an issue that could not be adequately addressed at the clinic level. Instead, it required the support of larger systems—for example, county mental health administrators and the managed behavioral health organization. Without larger system support, many agency administrators felt the initiative would fail. This idea was reflected when one agency director observed, "To make this [dialectical behavior therapy implementation] work, it is necessary to have buy-in and understanding of top administration."
In our analysis of interviews with mental health administrators who were implementing dialectical behavior therapy, we found that administrators anticipated a range of challenges in successfully implementing and sustaining dialectical behavior therapy. The challenges mentioned were most often related to resources, both staff and financial. Despite their concerns, administrators and their organizations were devoting substantial time, energy, and resources to training in dialectical behavior therapy.
Researchers have identified individual characteristics associated with improved training results. However, administrators' comments suggest that these research-based predictors of who will benefit most from training were not a factor in decisions on who would be trained. Instead, administrators gave careful consideration to the agency's need to minimize disruptions caused by clinician absence and to ensure that staff most committed to the agency were trained so that they could train and supervise additional staff. Researchers must better articulate and share with administrators information about who would be most likely to acquire and implement skills. Constructing training scenarios that minimize the short-term risks that administrators associate with training—for example, high up-front costs—would also likely increase adoption of innovative treatments, as would research demonstrating the positive impact of training on clinicians' job satisfaction, job tenure, and successful implementation of evidence-based treatments.
Experts have also discussed clinicians' need for a "learning period" in which they take fewer clients and have their work time offset to enable them to prepare for consumer sessions and observe other clinicians implementing treatment (5,11), all of which is difficult from an administrative standpoint because of lost productivity and revenue. Although precise numbers may be unavailable, rough estimates from treatment researchers regarding resources needed to implement a new intervention would allow administrators to make more informed choices and would permit administrators to better plan for implementing and sustaining the treatment.
Another consistent theme was the challenge of workforce instability coupled with fragile financial infrastructures. In the current fee-for-service environment, a typical response is to point to inadequate resources (for example, reimbursement rates) and reliance on client volume. The impact of a labor-intensive cost structure with substantial overhead greatly complicates efforts to adopt new and improved practices. Our findings support the view that implementing and sustaining effective new treatments will require open dialogue with multiple stakeholders about financial best practices in order to fairly examine the ratios of overhead-to-direct care costs. The problem is complex, and adjusting rates and payment structure alone, without an improved financial management practice at the provider level, is unlikely to successfully support necessary changes to improve care.
Administrators also emphasized that evidence-based treatments are implemented in the context of an already complex, ongoing, clinical enterprise. Change at multiple levels including that of the practitioner, team, organization, and larger system is necessary to effect large-scale change in treatment delivery. Evidence-based treatments need to be perceived by community-based mental health professionals as easy and compatible with existing services (12). Similarly, the intervention's fit with an organization's current service delivery structure, mission, interests, and resources influences its ease of implementation (13). Brief administrator surveys to assess organizational readiness before training may help identify possible system-level challenges to implementation as well as their solutions. Organizational interventions may also help better prepare for a successful implementation of evidence-based treatments.
Administrators expressed concern that the treatment, while "ideal," might not be realistic, being too time and cost intensive. Some have suggested a potential solution as implementing "active ingredients" or components of the treatment (14). However, implementing components of dialectical behavior therapy, rather than the full model, does not result in the same treatment benefit (15). Another alternative may be to train clinicians in evidence-based practices (for example, evidence-informed assessment, clinical decision making, and treatment practices) rather than specific evidence-based treatments. Such an approach has the potential of improving care for a broader base of consumers, as well as addressing administrators' concerns about not having a sufficient number of clients who would benefit from any one approach.
Administrators were interviewed regarding the planned implementation of dialectical behavior therapy. Therefore, we are not sure whether their concerns are warranted or reflect another construct, such as motivation, leadership abilities, or limitations of the implementation plan. Once implementation is completed, a follow-up assessment will be necessary. We also don't know whether findings would generalize to implementation of other interventions. Organizations were in rural and suburban areas; organizations in such areas may face different constraints than urban organizations regarding available clinicians, integrated services, and rates of treatment participation. It is unclear how such factors might influence study findings.
Agency administrators can promote and enhance the organization's readiness for change, and are essential for successful implementation of a treatment. Ultimately, agency administrators make decisions about whether and which treatment will be implemented, which clinicians will be trained, and how the established system will change to accommodate the treatment. Consequently, attending to agency administrators' concerns and needs before inception of implementation can help facilitate processes and attenuate or prevent obstacles that may arise.
Community mental health centers are challenging organizations to manage. They typically have high staff turnover, operate on tight budgets, treat difficult populations, and must be responsive to changes in their system—for example, county systems and funding shifts from payers (3). Agency administrators are often highly skilled and understand the unique nuances of sustaining a community mental health agency. Their valuable opinions about the context and constraints provide a perspective that treatment developers and academic researchers cannot. Ultimately, merging the perspectives of all involved—administrators, treatment developers, researchers, and other stakeholders—will help facilitate the implementation of evidence-based treatment.
This research was supported, in part, by a Community-Academic Partnership Grant funded by the University of Pittsburgh, Department of Psychiatry, and by Community Care Behavioral Health, as well as by a National Institute of Mental Health Career Development Award to Dr. Herschell (K23 MH074716). The authors thank the agency administrators for their insightful comments and participation in this project; Mary C. Hennigh, Michelle Hovis, Nancy McDonald, Beth Pickering, Sarah Reynolds, and Barb Schroeder for their feedback on initial drafts of the agency administrator interview; Susan Essock for her feedback on initial drafts of the manuscript; and Amanda Costello.
The authors report no competing interests.