Behaviour Research and Therapy

Transcription

1 Behaviour Research and Therapy 47 (2009) Contents lists available at ScienceDirect Behaviour Research and Therapy journal homepage: Three preparatory studies for promoting implementation of outpatient schema therapy for borderline personality disorder in general mental health care Marjon Nadort a, *, Richard van Dyck a, Johannes H. Smit a, Josephine Giesen-Bloo b, Merijn Eikelenboom a, Michel Wensing c, Philip Spinhoven d, Carmen Dirksen e, Jeroen Bleecke b, Bianca van Milligen a, Michiel van Vreeswijk f, Arnoud Arntz b a GGZ ingeest, Department of Psychiatry and EMGO Institute, VU University Medical Center Amsterdam, A.J. Ernststraat 887, 1081 HL Amsterdam, The Netherlands b Maastricht University, Department of Clinical Psychological Science, The Netherlands c Radboud University Nijmegen Medical Centre, Scientific Institute for Quality of Healthcare, The Netherlands d Leiden University, Department of Psychology, The Netherlands e Department KEMTA Academic Hospital Maastricht, The Netherlands f G-kracht, private mental health care outdoor clinic, Delft, The Netherlands abstract Keywords: Borderline personality disorder Outpatient therapy Schema therapy Pilot- implementation Process evaluation Objective: Three studies were conducted to prepare for the implementation of Schema Therapy (ST) for Borderline Personality Disorder (BPD) in general mental healthcare settings. Two were surveys to detect promoting and hindering factors, one was a preliminary test of a training program in ST. Methods: In 2004, a diagnostic analysis of factors promoting and hindering implementation of a new treatment for BPD was conducted among both managers (n ¼ 23) and therapists (n ¼ 49) of 29 Dutch mental healthcare institutes through a written survey (Study 1). Next, a training program, including a set of DVDs displaying the major therapeutic techniques, was developed and tested among eight therapists. The training program was evaluated by the participants. After the training, three independent raters evaluated therapists adherence and competence, viewing videos of the therapists completing structured role-plays (Study 2). In 2008, a second written survey was conducted in 22 mental health institutes to study factors for future nationwide implementation of ST (Study 3). Results: Both surveys indicated that the situation in most institutes was favorable for implementing a new effective treatment, as participants were not satisfied with the existing treatments, had suitable professional backgrounds, worked in settings with (B)PD-oriented care programs, and expressed a need for change. The surveys yielded clear results for promoting or hindering successful implementation of ST. Promoting factors included scientific evidence for the effectiveness of the treatment, structural changes in the patient s personality, rapidly noticeable effects for the patient, low drop-out rates and a favorable cost-effectiveness. Possible barriers included implementation mandated unilaterally by management, choosing ST based on financial or organizational needs, extending implementation over a lengthy period of time and providing telephone support by therapists beyond office hours. The eight-day training program received very positive ratings. After the training, therapists were rated as sufficiently adherent and competent applying ST to treat BPD patients, with peer supervision and supervision recommended as a supplement to the training. Conclusion: This study showed that the situation in 2005 was advantageous to start implementation of ST. Evaluation of the training and the achieved competence scores of trainees concluded that the training program was a good basis for training therapists in ST. Outcome of the survey in 2008 demonstrated that there was a clear interest for implementation of ST for BPD patients in the future. Ó 2009 Elsevier Ltd. All rights reserved. Background Implementation * Corresponding author. address: m.nadort@ggzingeest.nl (M. Nadort). For successful implementation, crucial elements and principles recur through most publications examining models and theories /$ see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi: /j.brat

2 M. Nadort et al. / Behaviour Research and Therapy 47 (2009) used in the field of planning the implementation of a new treatment (Grol & Grimshaw, 2003; Grol & Wensing, 2006; Weinmann, Koesters, & Becker, 2007). While the scientific basis of these principles is still limited, they provide a framework for setting up an implementation plan. According to Grol and Wensing (2006), these elements are: A systematic approach to, and good planning of, implementation activities. A diagnostic analysis of the target group and settings should take place before the start of the implementation. An analysis of the care as usual and deviation from the proposed behaviour is necessary and the choice of implementation activities should link with the results of the diagnostic analysis. Usually, a single method of measure is insufficient and it is better to look for a cost-effective mix of methods tailored to the identified obstacles and incentives to change. The target group should be involved in the development and adaptation of the innovation, as well as in planning the implementation. Focusing on organizational and practical issues is important. Continuous evaluation of both the implementation process and its results is necessary as is feedback to the target group. Implementation should become an integral part of the existing structures and long-term effects should be the aim. According to Grol and Wensing s principles, the present authors conducted three studies to prepare the implementation of schema therapy for borderline personality disorder in general mental healthcare settings in the Netherlands. Borderline personality disorder and schema therapy Borderline Personality Disorder (BPD) is a severe and chronic psychiatric disorder. It is marked by chronic instability in multiple areas (emotional dysregulation, self-harm, impulsivity, and identity disturbance). The life-time prevalence of BPD in the general population is about 2%. In psychiatric outpatient settings, 10% of the patients suffer from BPD. The medical and other societal costs for BPD are substantial (Ten Have, Lorsheyd, Van Bijl, & Osterthun 1995; van Asselt, Dirksen, Arntz, & Severens, 2007). In a randomised controlled trial (RCT) comparing Schema Therapy (ST) and Transference Focused Psychotherapy (TFP) (Giesen-Bloo et al., 2006), ST was found to be a superior treatment to TFP. ST had a recovery rate of 45.5% and a reliable change rate of 65.9%. The dropout rate for ST was significantly lower than for TFP. The positive results regarding the effectiveness of ST were substantiated by an economic evaluation showing that ST is also more cost-effective than TFP (Van Asselt et al., 2008). Because these results were promising, an implementation study for ST was conducted. The article on the implementation study conducted between December 2005 and February 2009 can be found separately in this special issue on implementation (Nadort et al., 2009). This implementation study was prepared using three smaller studies: two surveys to help identify barriers and factors promoting successful implementation of ST in regular mental healthcare settings in the Netherlands, and one small test of an ST-training program for therapists developed to use in an implementation study of ST in the Netherlands. The results of these studies follow. Part of the project that tested the efficacy and cost-effectiveness of ST versus TFP for BPD included a set of dissemination and implementation activities. The first two studies presented in this paper were part of these activities, while the third was part of a follow-up project studying actual implementation of ST in regular practice. Study I: 2004 survey Introduction To prepare implementation activities, potential factors hindering and promoting implementation of the treatment are identified. Before data of the Giesen-Bloo et al. (2006) trial were analysed, preparations were made to implement the most effective treatment of the RCT. Preparations included conducting a survey among some of the most important stakeholders for implementation: therapists and managers of mental health institutes in the Netherlands. The survey s goals were to assess whether the situation in these institutes was ripe for implementation of a highly specialized psychological treatment for BPD, and to identify the major factors promoting and hindering implementation, according to these stakeholders. The survey was based on the theories by Grol and Wensing (2006), which stress that different levels should be addressed to detect obstacles and to identify supportive factors, such as the individual care provider (competence, attitude, motivation for change and personal characteristics), the social setting and network (opinions of colleagues, opinion leaders and professional network), and the management level (structural conditions, interests and resources). Method To prepare for implementation of the most (cost-) effective treatment of the RCT (Giesen-Bloo et al., 2006), in 2004 a survey was conducted in Dutch mental healthcare institutes (Arntz, Dirksen, & Bleecke, 2005). The goal of this survey was to identify factors promoting or interfering with successful implementation of a new psychotherapeutic treatment of Borderline Personality Disorder (BPD). Based on Grol and Wensing s recommendations (2001), both managers and therapists of Dutch mental healthcare institutes were sampled. Successful implementation entails managers supporting the introduction of a new treatment, and therapists effectively executing the new treatment. Using Grol s framework, first trial interviews with both managers and therapists were conducted, addressing issues related to individual, social and contextual levels. On the basis of these pilots, a survey was developed that would be mailed to the sample. According to the levels listed by Grol and Wensing, several items were written and included in the survey. Participants and procedures Conducted in 2004, data were collected from 29 mental health institutes, distributed to nine Dutch provinces. The institutes were randomly selected from Dutch mental healthcare institutes that provided outpatient treatment for BPD patients. Before actual sampling of the institutes, a list of Dutch mental healthcare institutes was created. Next, sampling was stratified over region (North, East, Mid, West and South Netherlands). In each of the sampled institutes, one or more managers and one or more therapists involved with the treatment of BPD patients were mailed surveys, and asked to complete and return them. The number of respondents was 72. They were distributed as follows: 49 therapists (28 psychotherapists, 11 psychiatrists, 9 psychiatric nurses and 1 assistant physician); 23 managers (6 managerial psychiatrists, 6 treatment program heads, 6 policy managers and 5 heads of treatment teams). Analysis For the survey data, descriptive statistics were calculated. Results of the 2004 survey In the mental health institutes, cognitive therapy appeared to be the most prevalent psychotherapeutic treatment (72%). Among therapists and managers, 67.3% of the therapists said they provided cognitive treatment, and 82.6% of the managers reported cognitive

3 940 M. Nadort et al. / Behaviour Research and Therapy 47 (2009) therapy was provided in their departments. The other forms of psychotherapy, including client centered psychotherapy, psychodynamic psychotherapy, and behavioral therapy, were each provided in 20 40% of the departments. Approximately 20% of the therapists applied at least one of the non-cognitive therapies. In 77.8% of the institutes a program for the treatment of BPD was used, and most of these institutes had a specific BPD program (55.6% of the total). In 22.2% of the institutes, a program for the treatment of Personality Disorders (PD) was used. In 19.4% of the institutes, no program was employed to treat BPD patients. On average, 9.75 therapists were involved in the treatment of BPD patients. In two thirds of the departments, more than seven therapists were involved in treating patients with BPD. For the respondents, the most important sources of information about new treatments were: 1.Written specialized resources (69.4%); 2. Conferences (56.9%); 3. Contacts with colleagues at the institute (50.0%); 4. Contacts with specialist colleagues outside the institute (44.4%); 5. Courses (41.7%).None of the responders, therapists or managers reported to be unwilling to change their existing treatment possibilities for BPD. Of the therapists, 24.5% reported to be interested in new treatments for BPD, but were not yet willing to change the treatments they delivered. 75.5% of the therapists claimed to be willing to change the treatments they delivered for BPD, in the short or long-term. For managers, 39.1% were interested in new treatments for BPD, but were not yet willing to change the treatments their departments delivered. 60.9% of managers were interested in new treatment possibilities and would be willing to implement them when a new, suitable treatment was available. Both managers and therapists rated the present treatment satisfaction for BPD patients with a median satisfaction score of 4.9 on a 0-10 scale. More than 50% of both groups rated the present treatment for BPD as unsatisfactory. For therapists, 77.6% rated the satisfaction of treatments they used for BPD less than 7 on a 0-10 scale, with only 22.4% moderately to highly-satisfied. For managers, 69.6% rated their present treatment satisfaction for BPD less than 7 on the 0-10 scale, with 30.4% moderately to highly-satisfied. Factors that may promote or hinder a successful implementation of a new psychotherapeutic treatment were described in 25 propositions. The following factors were rated most often as moderately or highly supportive (percentage is proportion of respondents rating the proposition as moderately or highly supportive):table 1 The following factors were reported to be highest to hinder or interfere with implementation (moderately or highly):table 2 Factors that could be considered as either important or unimportant with respect to implementation of a new psychotherapy were described in 25 propositions. The following factors were rated as moderately to highly important by the indicated percentage of respondents. Table 3 Table 1 Promoting factors for implementation. 93.0% Scientific foundation of the new treatment 87.3% The therapists that have to execute the new treatment have an important say in the choice of the new treatment. 87.2% An important opinion leader in the organization supports the new treatment. 84.5% The new treatment is supported by BPD patient organizations. 83.1% The possibility to first try out the new treatment before it is decided to deliver it. 74.6% The implementation is executed under the direction of a project group of the own institute. 74.3% The patient s personality is structurally changed by the new treatment. 72.8% There are weekly peer supervisions for the new treatment. Table 2 Hindering factors for implementation. 80.2% The choice for the new treatment is based on financial or organizational arguments, not on the basis of its effectiveness. 69.0% The implementation process takes more than a year. 64.8% The choice for the new treatment is imposed by the management. Discussion Considering the answers of the respondents, it could be concluded that in 2004 the environment was favorable to start implementation of ST for treatment of BPD in regular mental healthcare settings. Many cognitive therapists were working in these institutes, with an excellent background for further training in ST and they had the support of their managers. In general, enough cognitive therapists existed in mental health institutes to create peer supervision groups, a necessity for optimal implementation of ST. Since most of the institutes already had treatment programs for BPD (either specifically or as part of a broader defined PD treatment program), existing programs would facilitate the implementation of ST. With present treatments of BPD not particularly satisfying, and a clear interest in new and better treatments noted, a second study to prepare the implementation of ST for BPD was created to develop and test a training program for therapists. The training was supplemented with a set of DVDs displaying the major therapeutic techniques. Study 2: training program and audiovisual training material Introduction The next implementation steps consisted of several elements: production of a set of DVD s showing the basic techniques and methods of ST (Nadort, 2005), the development of a training program and creation of a website 1 ( (2005)). Study 2 tested the structured training in ST for BPD developed to disseminate ST among therapists and to implement ST in mental healthcare institutes. Method DVDs with ST techniques demonstrations In 2004 a six-hour DVD set of schema therapy techniques was produced, consisting of 34 segments averaging 10 min in length (Nadort, 2005). Key elements were chosen from Young, Klosko and Weishaar s work (2003), and from the protocol tested in the RCT (Arntz & Van Genderen, 2009; Van Genderen & Arntz, 2005). In addition to the specific techniques, the therapeutic approach was of primary importance in the sample segments. On the DVDs, professional actors played the borderline patients, and actual therapists who had participated in the main ST BPD outcome study played the therapists. Produced in Dutch, the DVDs were later subtitled in English. The DVD segments include: introduction, making contact and explanation of rationale of therapy, schema 1 The website offers information to the lay public about ST and offers a list of ST therapists that are member of the Dutch Registry of Schema Therapists. The website offers a link to order the DVD box, and provides the manual for the use of the DVDs. In addition, all (homework) forms, questionnaires and other forms that can be used in ST are available on the website for downloading. All of the continuing education links in the field of schema therapy can be found on the website, as well as links to recent research literature.

4 M. Nadort et al. / Behaviour Research and Therapy 47 (2009) Table 3 Important factors for implementation. 98.6% The new treatment is a clearly identifiable part of the total package of treatments. 95.8% The new treatment has a clearly defined crisis management policy. 95.8% Peer supervision and supervision are important parts of the learning phase of the new treatment. 94.4% The employer financially stimulates participation in training in the new treatment. 94.3% Immediate colleagues are positive about the new treatment. 91.5% The new treatment fits in the existing structure of and collaboration between treatment programs. 91.5% The government admits the new therapy in the treatment package. 88.7% The new treatment has a promptly noticeable effect for the patient. 85.9% BPD patients have difficulties to fulfill the minimal conditions of the treatment. 85.9% The new treatment incorporates a specialized crisis facility. 84.9% The new treatment is evidence-based. 80.3% The new treatment has a relatively low drop-out rate. 78.9% The new treatment has a favorable cost-effectiveness. 78.9% The management as a whole supports the choice for the new treatment. 78.8% Institutes increasingly work with treatment programs for BPD. 74.7% The barrier for participating in the treatment is low for the patients. 74.6% The characteristics of the patients of the scientific study are representative for the patients treated by the therapists. questionnaires, case conceptualization, explanation of the mode model, limited reparenting, detached protector mode (pros and cons, imagery work, mode dialogue), diagnostic imagery, imagery rescripting by the therapist, imagery rescripting by a helper, imagery rescripting by the Healthy adult, Punitive parent mode (multiple chair techniques), empathic confrontation, Angry child mode, Angry protector mode, angry side of Healthy adult mode, schema diary and mode diary, flashcard, downward arrow technique, historical role-play, motivating new behaviour, imagery rescripting future, breaking through dysfunctional partner choices and learning to make a healthy partner choice. All segments were used in the training program, and were evaluated by both instructors and participants. Training program: method and components A training course was developed to train psychotherapists, psychologists and psychiatrists in schema therapy techniques for the treatment of borderline patients. The training addressed the theoretical model, treatment frame, different phases, and the use of strategies and techniques (Arntz & Van Genderen, 2009; Van Genderen & Arntz, 2005; Young, Klosko, & Weishaar, 2003; Young & Klosko, 1999; Young, Klosko, & Weishaar, 2005). Central to ST is the assumption of five schema modes specific to BPD. Schema modes are sets of schemas expressed in pervasive patterns of thinking, feeling and behaving (Lobbestael, Arntz, & Sieswerda, 2005, Lobbestael, Vreeswijk, & Van, Arntz, 2008). Change is achieved through a range of behavioral, cognitive and experiential techniques that focus on (1) the therapeutic relationship (2) daily life beyond therapy and (3) past (traumatic) experiences. Recovery in ST is achieved when dysfunctional schemas no longer control or rule the patient s life. The training program consisted of the following components: intake and assessment phase, case conceptualization, schema questionnaire completion and results discussion, therapeutic relationship, empathic confrontation, setting limits, cognitive techniques, behavioral techniques, experiential techniques, and the mode model. A great deal of attention was devoted to address various modes of the borderline patient, such as the Abandoned child, the Angry child, the Detached protector, the Punitive parent, and the Healthy adult mode (Lobbestael et al., 2005, 2008). The training program also emphasized how to set-up treatment and phase-in various techniques. Specific themes and techniques were covered on each training day, starting with a short introduction by the instructors on a given topic, a DVD demonstration, and dyadic role-playing in sub-groups. Two instructors then offered feedback. Finally, a plenary discussion of the techniques that had been practiced occurred. Therapists were required to actively participate. The course was organized within the norms of the Dutch Association of Behavioral and Cognitive Therapy and the Dutch Society of Psychiatry, and was accredited by these associations. Accreditations by both associations ensured that the training course could also be used for future implementation. Participants and procedures of the training program The training course (group format, 50 h, eight days) was delivered in 4 2 days from January 2005 to April Five practitioners of a regular mental health service and three practitioners of an Out-patient Forensic Psychiatric Service participated. They attended all sessions. Two trainers with expertise in ST and the treatment of borderline patients trained the participants, who were psychotherapists, psychologists and psychiatrists. The profile of participants was: 1 male, 7 female; mean age 34 years-old (range 25 55); 5 licensed psychotherapists; all licensed in CBT. Evaluation of the training program Each training day was evaluated separately, documented in a written report. The mean evaluation per day on a scale from 0 (bad) to 10 (very good) was calculated. Further, a final evaluation over the complete course was also conducted on a scale from 0 (bad) to 5 (good). To examine the participants competence as ST therapist after the program, all participants were asked to participate in three 10- minute role-playing situations with actors (script-based). During the first role-play, an actress played a patient s role in the Abandoned child mode, while the therapists demonstrated limited reparenting techniques. For the second role-play, an actress portrayed the Punitive parent mode, and therapists recognized the mode and demonstrated mode-specific techniques to address it. In a third role-play, the actor dramatized the Detached protector mode and therapists demonstrated how to break through the detached mode. All role-plays were videoed, and segments were coded and rated by three independent evaluators: two therapists well versed in schema therapy, and one BPDpatient who had been successfully treated with schema therapy. The evaluators were trained before completing the ratings. The Young Therapy Adherence and Competence Scale (Young, Arntz, & Giesen-Bloo, 2006) was used to rate the video segments. The purpose of this rating scale is to determine if therapists are following the procedures and practices of ST, and to assess their competence as an ST therapist. While there may be some overlap, adherence differs from competence. Therapy adherence addresses the question: is the therapist providing the therapy according to the protocol. Therapy competency addresses the issue: how skillfully is the therapist providing the therapy. As for the rating, since only 10-minute video segments were evaluated, not all of the items on the Therapy Adherence Scale could be demonstrated in such a short amount of time. Therefore, items that were selected included the general therapeutic style, limited reparenting, therapeutic relationship, psycho-education about schemas and modes, and connecting daily situations with schemas and modes. The items were rated on a scale (0 2) with skills ranging from (0) not present to (2) strongly present. An independent research assistant entered all of the data. There were no missing values.

5 942 M. Nadort et al. / Behaviour Research and Therapy 47 (2009) Analysis Results of the evaluation of the training program were analyzed by calculating the means and the standard deviations. The intra class correlation coefficients (ICC) of the therapy adherence and competence scores were analysed by means of reliability analyses. A two way mixed model (consistency) was used. Results of the training program The mean evaluation of the training program and the DVD series of therapy techniques per day was 8.77 (SD 0.56) (range 0 10; badgood), and the mean of the final evaluation was 5 (SD 0) (range 0 5; bad-good). Overall, the training program received a very good evaluation. The participants evaluated the instructors introduction as good, the examples as good and illustrative, and the exercises and instructors feedback as educational and useful. All participants evaluated the program as good and educational. Competence tasks Although the competence tasks only demonstrated brief therapy segments, they gave a good illustration of the therapists style and the therapeutic techniques. Analysis of the results showed that adherence to ST and competence, as for overall appropriateness of used methods and techniques in SFT, was sufficient (mean 1.18; ICC ¼ 0.88). The global competence/quality ST therapist rating was moderate-good (mean 1.39; ICC ¼ 0,84), with a cut off score of 1. One therapist had a score below 1. Discussion Based on the information presented, it can be concluded that the 50-h training program, supplemented by the DVD series of therapeutic techniques, was a good basis for the training of practitioners in schema therapy. But experiences from the RCT learned us that providing a training program is not sufficient for learning ST. Also following the guidelines of Grol and Wensing (2006) that usually, a single method of measure is insufficient and it is better to look for a cost-effective mix of methods tailored to the identified obstacles and incentives to change in addition, actual practice under supervision, supplemented by peer supervision, is necessary and strongly recommended (Van Vreeswijk, Broersen, & Nadort, 2008). Study III: 2008 survey Introduction In 2008, the Dutch government changed the organization and payment of mental healthcare dramatically, from a tax-based system to a commercial, insurance-based system. This shift necessitated a re-evaluation of potential barriers and factors supporting the implementation of ST through the creation of a new survey in Additional factors since the 2004 survey had also come to light. For instance, early experiences training ST therapists warned that telephone availability beyond office hours (which is also a component of dialectical behaviour therapy for BPD) was potentially problematic for the implementation of ST. Since this issue had not been addressed in the 2004 survey, it was added to the new survey. Based on the experiences during the implementation study, specific questions were included regarding the therapist s phone support beyond office hours, the organizational problems regarding travel time for therapists from different institutes, and problems with reorganization and cooperation among the different institutes. A shortened version of the questionnaire of Arntz and colleagues (Arntz, Dirksen, & Bleecke, 2005) was used. Additional issues that had to be addressed included recent changes in the Dutch Health Insurance System. Insurers became more critical of mental health treatment providers and costs. Moreover, insurers focused more on the demands of patients, who could easily switch to a competitor if they did not receive the care they desired. Insurers measured the quality of provider care through performance indicators, such as patient satisfaction and improvements in General Attainment Scores (GAS). Treatment providers who exceeded quality measurements received higher compensations and more patient referrals, entering into a preferred partnership contract with insurers. To understand the opinion of insurers regarding the efficacy and cost-effectiveness of schema therapy, a questionnaire was mailed to 15 large insurance companies. Insurers were asked four short questions: a) what kind of information do you need from a mental health institute to purchase specific mental healthcare, such as schema therapy b) what are important factors in your decisionmaking process when you can choose between different institutes for purchasing specific mental healthcare c) what kind of information do you need to assess the quality of the provided mental healthcare d) what are the conditions for a preferred partnership contract. Of the 15 companies, only 20% responded to the questionnaire, and most of the respondents were not familiar with schema therapy. All respondents preferred evidence-based or best practice care. Participants and procedures of the 2008 survey In 2008, a list of 31 Dutch mental healthcare institutes providing outpatient treatment for BPD patients was created encompassing 12 Dutch provinces. In each of the institutes, one or more managers and one or more therapists involved with the treatment of BPD patients were mailed the survey and asked to complete and return it. Of the mental healthcare institutes that were approached, 5 institutes refused to participate, so data were collected in the remaining 26 mental healthcare institutes agreeing to participate. Of the 26, 13 were already familiar with the Schema Therapy Implementation Study of Nadort and colleagues (see the paper regarding implementation in this special issue Nadort et al., 2009). Of these 13 institutes, 8 had already participated in the implementation study. For various reasons, five of the thirteen institutes had previously refused to participate in the implementation study. These five institutes were then asked to complete a survey to gather more information about their objections to participating. Thirteen new mental health institutes providing outpatient treatment for BPD patients were mailed surveys and asked to complete and return them.. The number of responding institutes was 22. Respondents were distributed as follows: 60.5% were therapists, 26.3% were financial managers, and 13.2% were directors of treatment teams. Analysis For the survey data, descriptive statistics were calculated. Results Mental health care institutes Most of the institutes provided a specialized program for BPD patients (71.1%) and were already using some form of ST for

6 M. Nadort et al. / Behaviour Research and Therapy 47 (2009) Table 4 Promoting factors for implementation. Percentage ST gets positive attention in the media 89.5% The patient s personality is structurally changed by ST. 89.4% ST fits in with evidence-based working. 89.4% ST is evidence-based. 86.9% Immediate colleagues are positive about ST. 84.2% ST is a clearly identifiable part of the total package of 84.2% treatments. The employer financially stimulates participation in ST training 81.6% ST has a promptly noticeable effect for the patient. 81.5% ST training is provided by an external expert. 78.9% The ministry of health admits ST in the treatment package. 78.9% ST has a favorable cost-effectiveness. 76.3% ST fits in with the existing structure of and collaboration 76.3% between treatment programs. Peer supervision and supervision are important parts of the 73.7% learning phase of ST ST has a flexible treatment protocol 73.7% ST is supported by BPD patient organizations. 73.7% An important opinion leader in the organization supports ST. 71.1% borderline patients (84.2%). Most patients were treated with outpatient therapy (92.1%). Managers rated their average satisfaction of current therapy programs for BPD patients 5.1 (SD 1.9) on a scale from 1 to 10, (1 ¼ minimal satisfaction, 10 ¼ maximum satisfaction), and the therapists satisfaction score was 5.5 (SD 2.1). 39.5% of the respondents worked in institutes where no evidencebased treatments were investigated. 15.8% of the respondents reported that there were evidence-based treatments that were investigated, but that they were not personally involved. 42.1% of the respondents were personally involved in BPD treatment investigations. The respondents received information about new therapies from written specialized reports (78.9%), conferences (76.3%), and workshops (52.6%). Factors that may promote or hinder successful implementation of schema therapy were described in 25 propositions. The following factors were rated most often as moderately or highly promoting (percentage is proportion of respondents rating the proposition as moderately or highly promoting):tables 4 and 5 Discussion Based on the respondents answers, the situation in 2008 seemed encouraging for future implementation of ST. Most of the institutes already had treatment programs for BPD (either as such or as part of broader defined PD treatment programs), but present treatments of BPD were not viewed as particularly effective, and a clear interest in new and better treatments was expressed. 84% of the respondents were already using some form of ST. It was unclear from the results whether the programs were structured, and the Table 5 Hindering factors for implementation. Percentage Travel time necessary for peer supervision and supervision 68.4% Telephone support by the therapist outside office hours 57.9% The choice for ST is based on financial or organizational arguments, 50% not on the basis of its effectiveness. Therapy sessions of 45 min twice a week 50% Patients have to pay a personal contribution for sessions 39.5% Treatment period takes 3 years 39.5% ST requires weekly peer supervision. 36.8% Changes on organizational level 34.3% The implementation is region coordinated 29% ST requires monthly supervision 26.3% ST requires a training program of 8 fulltime days 23.7% ST requires specialized care providers 23.7% high percentage of respondents already using a form of ST may be partially explained by the fact that 8 institutes already participated in the implementation study of Nadort and colleagues. General discussion The three preparatory studies followed the guidelines of Grol and Wensing (2006), using a good planning and systematic approach, and both surveys offered a diagnostic analysis of the target group and settings, an analysis of the care as usual, and deviation from the proposed behaviour. The situation in 2004 seemed favorable to start implementation of ST as treatment of BPD in the regular mental health service. Many cognitive therapists were working in these institutes, with an excellent background preparing them for further training in ST. Generally, enough cognitive therapists were already working in mental health institutes to create peer supervision groups, a key component for an optimal implementation of ST. Most of the institutes already had treatment programs for BPD (either as a specialized program, or as part of broader defined PD treatment programs), and their existence would facilitate the implementation of ST. Present treatments of BPD were not particularly satisfying, and a clear interest in new and better treatments was desired. Most of the respondents indicated that their most important sources of (initial) information were specialized reports (i.e., articles, books) and conferences, relatively easy outlets to relay information on ST s (cost-) effectiveness. Grol and Wensing recommend to make a choice of implementation activities that are linked with results of the diagnostic analysis and to use a cost-effective mix of methods that are tailored to the identified obstacles and incentives to change. It is important that the target group is involved in the development, adaptation and planning of the implementation, and to focus on organizational and practical issues (2006, p 66 70). Factors rated by many respondents as highly important or supportive were incorporated. The surveys indicated that information about ST should include several components. First, ST is evidence-based. Second, ST leads to a structural change in the patient s personality, not merely (temporary) symptomatic improvements. Third, ST generally leads to prompt reductions in the patients problems. Considering the average rapid decrease of BPD manifestations and other problems in the first six months of ST treatment, this seemed easy to support. Fourth, ST is suitable for a broad range of patients, with low drop-out rates. Given the nonstringent in- and exclusion criteria and the finding that baseline severity at the start of treatment did not predict recovery (but instead predicted more relative improvement), even more severe BPD patients benefited from ST treatment. Finally, the vast majority of the BPD patients in the RCT were able to adhere to the treatment protocol. The results of the RCT (Giesen-Bloo et al., 2006) and the study by Van Asselt et al. (2008) on the cost-effectiveness of ST support all five points. The survey s results concluded several important suggestions for the implementation study of Nadort et al. (2009). First, participants clearly wanted to have an important part in choosing the treatment, especially the therapists. Management mandating therapists to practice a particular treatment without the therapists say was not advised. Creating a positive and supportive environment between management, opinion leaders, and colleagues plays an important role for successful treatment. Additional recommendations included forming a project group in the institute, starting peer supervision groups, creating an appropriate crisis facility and policy, and allowing trained therapists to experiment with the new treatment before the implementation becomes finalized. Supervision by an external expert was also advised. Despite the time needed to employ a new treatment, the objective is to create

7 944 M. Nadort et al. / Behaviour Research and Therapy 47 (2009) a specialized treatment integrated within the institute s total program. In 2005, preparations for the actual implementation study that began in 2006 were made. According to the survey s outcome, the target group was involved in the development and testing of the training program and in the implementation process. Elements that had been stressed as important by the respondents of the survey were incorporated into the implementation package. Information about the (cost-) effectiveness of ST was provided. Therapists and managers of healthcare centers were invited to participate in the implementation study. Therapists were offered an 8-day training program, with supervision provided on location by an external expert. Another part of the implementation package included the formation of peer supervision groups and appropriate crisis facilities and policies. According to Grol and Wensing (2006), evaluation of both the implementation process and its results is necessary, as well as feedback to the target group. Implementation should become an integral part of the existing structures and long-term effects should be the aim (p 66 70). These elements were also integrated and the therapists were informed about the results of the implementation study on a regular basis. One of the supportive factors mentioned in the survey is the positive media attention for ST. Following publication of the main outcome study (2006), publicity was generated by informing insurance companies, the ministry of Public Health, patient organizations, the press, and the Dutch health counsel about the positive results of ST. The goal was to expand the availability of ST as a treatment of BPD. Finally, there are some important factors beyond mental healthcare institutes. In the first survey it was emphasized that ST needed to be recognized in the Dutch reimbursement system. Recently, ST has been recommended as one of the evidencebased treatments in the Dutch Guidelines on Personality Disorders (2008), and insurance companies reimburse for treatment. In the second survey, some obstacles for implementing ST were noted, including telephone availability of the therapist beyond office hours. The impact of telephone support was investigated in the implementation study of Nadort et al. (2009), and the results of the treatment with and without telephone support can be found separately in this special issue on Implementation. Some limitations existed in this study, however. The response rate of the insurance companies was very low (20%), probably due to the questionnaire being ed to potential respondents. Ideally, insurers would have received the questionnaire via mail, followed by a personal interview. Nevertheless, the 20% of insurers who responded did not have a broad knowledge of schema therapy, which is disappointing. For nationwide implementation it is recommended to offer insurance companies detailed information about the (cost-) effectiveness of schema therapy based upon the studies of Giesen-Bloo et al. (2006), Van Asselt, Dirksen, Arntz, and Severens (2007, 2008) and the preliminary results of the study completed by Nadort et al. (2009). The competence task of therapists was another limitation of this study. Therapists adherence and competence to the model were assessed directly after the training, based upon three short video segments. While showing a good sample of treatment, they were not actual sessions, so the entire Therapy Adherence and Competence Scale could not be applied, possibly influencing the outcome. Partly based on what we learned in our 2004 survey, we used a broad dissemination and implementation strategy that produced some clear successes. For instance, more than 1000 Dutch therapists were trained in ST. Many copies of the protocol book were sold. The majority of post-graduate training programmes in Psychotherapy and Clinical Psychology incorporated ST into their programmes. Also a Dutch Registry was set-up with three levels of ST therapists (junior, senior and supervisor) that can be found on a website that was created to serve the public and to promote exchange among therapists ( (2005)). From 2007 until 2009 the website had more than 28,000 unique visitors. Additionally, schema therapy received a lot of positive media attention. However, we did not assess to what degree ST was actually implemented in regular practice, nor did we assess how effectively such applications of ST for BPD were. Our implementation study (Nadort et al., 2009) demonstrated that the effectiveness of an implemented ST treatment can be surprisingly high, but this was assessed in the context of a scientific study, which may not fully represent what therapists and institutes actually do in their regular practice when applying ST. Further studies are needed to assess this issue. Finally, we want to emphasize the use of different elements in the learning process of ST. Because the training program and series of therapy techniques were all rated positively during the implementation study, we think these elements provide a sufficient basis to learn ST. In addition, as discussed in the second survey, to achieve successful implementation we strongly advise peer supervision and supervision as important elements in the learning phase of ST. It is recommended for future implementation studies to plan these type of preparatory studies in order to achieve successful implementation. Funding/Support This research was funded by Grant application Doelmatigheid/ deelprogramma Implementatie van ZonMw, aanvraagnummer , Role of the sponsor The sponsor played no role in the data collection and analysis, manuscript preparation, or authorization for publication. Acknowledgements Acknowledgements: We wish to acknowledge Hannie van Genderen en Remco van der Wijngaart for their contributions to the training program. We would like to express our gratitude to Travis Atkinson for his invaluable help in correcting the manuscript. References Arntz, A., Dirksen, C., & Bleecke, J. (2005). Promoting and interfering factors related to implementation of schema focused therapy for borderline personality disorder in Dutch mental health care institutes. Eindrapportage college voor Zorgverzekeraars. Arntz, A., & Van Genderen, H. (2009). Schema therapy for borderline personality disorder. Chichester: Wiley. Multidisciplinaire Richtlijn Persoonlijkheidsstoornissen. (2008). (Dutch Guidelines on Personality Disorders). Utrecht: Trimbos Instituut. Giesen-Bloo, J., Van Dyck, R., Spinhoven, P., VanTilburg, W., Dirksen, C., Van Asselt, T., et al. (2006). Outpatient psychotherapy for borderline personality disorder: a Randomized trial of schema-focused therapy vs. Transferencefocused psychotherapy. Archives of General Psychiatry, 63, Grol, R., & Grimshaw, J. (2003). From best evidence to best practice; effective implementation of change in patients care. Lancet, 362, Grol, R., & Wensing, M. (2001). Implementatie; effectieve verandering in de patiëntenzorg. Maarssen: Elsevier. Grol, R., & Wensing, M. (2006). Implementatie: effectieve verbetering van de patiëntenzorg. Maarssen: Elsevier gezondheidszorg. Lobbestael, J., Arntz, A., & Sieswerda, S. (2005). Schema modes and childhood abuse in borderline and antisocial patients. Journal of Behaviour Therapy and Experimental Psychology, 36, Lobbestael, J., Vreeswijk, M. F., & vanarntz, A. (2008). An empirical test of mode conceptualisations in personality disorders. Behaviour Research and Therapy, 46,

8 M. Nadort et al. / Behaviour Research and Therapy 47 (2009) Nadort, M. (2005). Schematherapie voor de Borderline Persoonlijkheidsstoornis. Therapietechnieken. DVD-box. Nadort, M., Arntz, A., Smit, J. H., Giesen-Bloo, J., Eikelenboom, M., Spinhoven, P., et al. (2009). Implementation of outpatient schema therapy for borderline personality disorder with versus without crisis support of the therapist outside office hours: a Randomized trial. Behaviour Research and Therapy, 47(11), Ten Have, M. L., Lorsheyd, J. J. G., van Bijl, R., & Osterthun, P. (1995). Jaarboek Geestelijke Gezondheidszorg 1995/1996 (Annual report mental health care 1995/ 1996). Utrecht: De Tijdstroom. Van Asselt, A. D. I., Dirksen, C. D., Arntz, A., Giesen-Bloo, J. H., Van Dyck, R., Spinhoven, P., et al. (2008). Outpatient psychotherapy for borderline personality disorder: cost effectiveness of schema-focused therapy versus transference focused psychotherapy. British Journal of Psychiatry, 192, Van Asselt, A. D. I., Dirksen, C. D., Arntz, A., & Severens, J. L. (2007). The cost of borderline personality disorder: societal cost of illness in BPD-patients. European Psychiatry, 22, Van Genderen, H., & Arntz, A. (2005). Schemagerichte cognitieve therapie bij borderline persoonlijkheidsstoornis. Amsterdam: Nieuwezijds. Van Vreeswijk, M., Broersen, J., & Nadort, M. (2008). Handboek Schematherapie, theorie, praktijk en onderzoek. Houten/Diegem: Bohn Stafleu Van Loghum Website for schematherapy: website voor schematherapie: (2005) Weinmann, S., Koesters, M., & Becker, T. (2007). Effects of implementation of psychiatric guidelines on provider performance and patient outcome: systematic review. Acta Psychiatrica Scandinavica, 115, Young, J., Arntz, A., & Giesen-Bloo, J. (2006). Therapy adherence and competence scale. http// Accessed Young, J. E., & Klosko, J. S. (1999). Leven in je leven. Leer de valkuilen in je leven kennen. Lisse: Swets & Zeitlinger. Young, J. E., Klosko, J., & Weishaar, M. E. (2003). Schema therapy: A practitioner s guide. New York: Guilford. Young, J. E., Klosko, J. S., & Weishaar, M. E. (2005). Schemagerichte therapie; handboek voor therapeuten. Houten: Bohn Stafleu van Loghum.