The Theory-Practice Gap in Cognitive Behavioral Therapy: Reflections and a Modest Proposal to Bridge the Gap

BETH-00398; No of Pages 9; 4C: Available online at www.sciencedirect.com

Behavior Therapy xx (2013) xxx–xxx www.elsevier.com/locate/bt

The Theory-Practice Gap in Cognitive Behavioral Therapy: Reflections and a Modest Proposal to Bridge the Gap

Keith Dobson Shadi Beshai University of Calgary

adolescents in the United States show serious This paper examines the issues related to the gap between psychological distress and are in need of treatment theory and practice in the area of cognitive-behavioral (Kessler, Chiu, Demler, & Walters, 2005; Kessler et therapy. The article begins with a review of the evidence for al., 2009; Lopez-Duran, 2011). These rates appear to such a gap, and having demonstrated that the gap exists, be similar around the globe. Fortunately, there are provides a discussion of some of the factors that are likely over 400 psychological treatments currently in important in its genesis and maintenance. The article then practice (Corsini & Wedding, 2011). Even more focuses on potential strategies to reduce the theory-practice fortunately, there is a growing trend to empirically gap that go beyond the common recommendation for both identify efficacious, effective, and efficient psychother- efficacy and effectiveness research. In particular, we provide apies. This trend has been referred to as the emphasis recommendations for protocol planning and design, train- on empirically supported treatment or the develop- ing and competency maintenance, dissemination research, ment of evidence-based practice (herein referred to as and implementation and policy change. We conclude with ESTs; Lambert, 2010). ESTs are treatments that have the proposition that theory and research should not only demonstrated efficacy in clinical research trials inform practice, but that practice should have a reciprocal (Kazdin, 2008), but many also incorporate sound benefit on theory and research. procedures for psychotherapy in general. ESTs are primarily the product of clinical theorists, are most often conducted in the clinical laboratory, Keywords: cognitive-behavior therapy; theory; practice; dissemination and are mostly conducted by clinical researchers. Clinical scientists have begun to notice a disquieting THE UNDERSTANDING OF PSYCHOPATHOLOGY has steadi- trend, however, which is that ESTs are seldom put ly increased with the scientific and technological into use by practitioners in the community and health advances made in recent decades. Both psychological care at large (Kazdin, 2008; McHugh & Barlow, assessment tools as well as the classification systems 2012). Despite the fact that a number of interven- used to diagnose and predict the course of mental tions have been found to be efficacious for a variety illness have dramatically improved over the last 30 of major psychological disorders (e.g., depression, years (Hyman, 2011; Meyer et al., 2001). These anxiety, bulimia nervosa, etc.), surveys of practi- recent advances shed new light on the prevalence tioners indicate that these treatments are rarely and perniciousness of psychological disorders. For utilized to treat patients who suffer from these instance, 30% of adults and 19% of children and conditions. Low utilization of ESTs in community mental health clinics has also been noted for eating disorders (Haas & Clopton, 2003), substance-use Address correspondence to Keith Dobson, Ph.D., Department of disorders (Santa Ana et al., 2008), and mood Psychology, University of Calgary, Calgary, Alberta Canada T2N disorders (Kessler, Merikangas, & Wang, 2007). 1N4; e-mail: [email protected]. For instance, a study by Ehlers, Gene-Cos, and Perrin 0005-7894/xx/xxx-xxx/$1.00/0 © 2013 Association for Behavioral and Cognitive Therapies. Published by (2009) of family physicians in the United Kingdom Elsevier Ltd. All rights reserved. revealed that only 11% of patients treated for

Please cite this article as: Keith Dobson, Shadi Beshai, The Theory-Practice Gap in Cognitive Behavioral Therapy: Reflections and a Modest Proposal to Bridge the Gap, Behavior Therapy (2013), http://dx.doi.org/10.1016/j.beth.2013.03.002 2 dobson & beshai posttraumatic stress disorder (PTSD) were provided Many practitioners have questioned the relevance of with evidence-based treatment for the disorder. efficacy trials, and claim that the conditions studied Further, only 6.9% of bulimia nervosa patients in in such trials are often less severe than what is the U.S. were provided with cognitive-behavioral typically observed in practice (Kazdin, 2008). therapy (CBT), despite its established efficacy for this Furthermore, some practitioners have indicated disorder (Chambless & Ollendick, 2001). Finally, that the samples typically allowed in clinical trials although clinical guidelines suggest the use of CBT include too many exclusion criteria and, as a result, with all patients who display signs of psychosis are too homogeneous and do not resemble the (NICE, 2009), only about 50% of schizophrenia diverse, heterogeneous clientele observed in routine sufferers are apparently provided with this treatment practice, who present with dual diagnoses and/or (Berry & Haddock, 2008). multiple problems. It should be noted, however, that It appears that the lack of utilization for ESTs in it is also possible to question how diverse clinical routine practice is not the only characteristic of this samples truly are, and recommend the use of treat- gap. Emerging evidence suggests that even when ment manuals for those patients who most closely practitioners attempt to implement ESTs, this approximate the presentation of cases treated in implementation is less than optimal. In other clinical trials (Stirman, Crits-Christoph & DeRubeis, words, treatment manuals are loosely followed, 2004). and thus the application of treatments bear only a Second, some practitioners have questioned the loose resemblance to the empirically tested protocols. practicality of using treatment manuals, as man- For example, a relatively recent study (Stobie, dated by the EST movement. For instance, some Taylor, Quigley, Ewing, & Salkovskis, 2007)found have claimed that treatment manuals are too rigid, that 40% of patients who suffered from obsessive- and thus do not account for individual differences compulsive disorder and who were allegedly offered in, or the preferences of, their clients (Borntrager, CBT had in actuality undergone a treatment that met Chorpita, Higa-McMillan, & Weisz, 2009). In an minimal criteria for CBT. Similarly, Kessler et al. early survey of this area, Addis and Krasnow (2007) found that only 20.9% of individuals who (2000) found that the major concern about treat- suffered from depression in a 12-month period were ment manuals was that they were rigid, inflexible, offered adequate psychological or pharmacological and did not allow for the integration of the art of intervention. psychotherapy into clinical practice. Difficulties in the transfer of laboratory produced Barlow, Levitt, and Bufka (1999) have reported and tested treatments to the community has been that a number of practitioners harbor misconceptions termed theory-practice gap. In describing this gap, regarding ESTs, which act as a major barrier to their Brownson, Colditz, and Proctor (2012) indicated adoption in routine practice. McHugh and Barlow that it signifies a discrepancy from “care that could (2012) more recently argued that these misconcep- be, where health care is informed by scientific tions largely stem from practitioners’ doubt of the knowledge, and the care that is in routine practice” applicability of the scientific approach to practice. (p. xi; italics in original). Thus, the theory-practice Such doubt is often centered in the belief that gap can simply be defined as the poor transport and psychotherapy is more “art” than science (McHugh uptake of knowledge and technology, as related to & Barlow). Similarly, Baker, McFall, and Shoham health care, from clinical science to clinical practice. (2009) have suggested that psychologists have yet to Acknowledging the magnitude of this problem, the assume a leadership role in using and disseminating Institute of Medicine (2001) described this gap as a empirically supported interventions, despite the “chasm.” impressive scientific record of CBT and other Although there is sometimes a willingness to blame psychosocial interventions. Baker et al. maintained clinicians for their apparent refusal to take up and that the relatively scarce use of ESTs among employ evidence-based therapies, in fact there are psychologists may stem from ambivalence among many factors that contribute to the theory-practice practitioners in regards to the role of science in gap in CBT. Expanding upon Shafran et al.’s(2009) practice, coupled with inadequacies in scientific evaluation, these factors can be grouped into three training, both of which conspire to lead to higher broad categories. The first two of these categories value upon clinical anecdotes than group-based concern practitioner beliefs regarding the relevance research outcomes. and practicality of ESTs. The remaining barrier, Third, some practitioners are either unaware of which acts to perpetuate the first two barriers, ESTs and/or are unable to effectively implement concerns the relatively poor training about and these interventions in their clinics or practice knowledge of dissemination of ESTs, which in turn settings (McHugh & Barlow, 2012). In this regard, stems from the underfunded nature of the field. the apparent gap between theory and practice may

Please cite this article as: Keith Dobson, Shadi Beshai, The Theory-Practice Gap in Cognitive Behavioral Therapy: Reflections and a Modest Proposal to Bridge the Gap, Behavior Therapy (2013), http://dx.doi.org/10.1016/j.beth.2013.03.002 gap 3 be more of a testament to the lack of resources Addressing the Barriers allocated to translational research than to the desire is cbt relevant in routine practice? to implement empirically supported interventions, per se. As Brownson and colleagues (2012) main- As noted, some CBT efficacy trials have been tained, only about 0.1% of total annual health criticized because the exclusion criteria in such expenditure is allocated toward dissemination and trials are so stringent that the target disorder and implementation (D&I) research. This poor knowl- patient samples eventually allowed into the trial do edge dissemination hinders practitioners from not resemble their counterparts in routine practice (Kazdin, 2008). In other words, clients in the “real receiving necessary training, delays the transition ” of practices from the laboratory to the clinic setting, world are more heterogeneous than those seen in and may even perpetuate common misconceptions research trials, and the disorders treated in clinical regarding ESTs, such as their inapplicability to settings are more severe, atypical, and often present clinical practice. with one or more comorbid condition. As such, Several other criticisms have been leveled against results of efficacy trials are sometimes typified as the EST movement. Most notably, opponents irrelevant in routine practice, and the application of criticize the movement for being biased toward ESTs becomes a futile pursuit. short-term therapies, such as CBT, to the exclusion Although it is true that most studies validating of other longer-term modalities. Also, opponents CBT have taken the form of efficacy trials, a number have criticized the outcome-oriented nature of the of effectiveness (conducted in real-world clinical movement, arguing that efficacy has narrowly been settings) trials have been carried out in recent years. defined as symptom diminution and behavioral Evidence from such effectiveness trials supports the change. Thus, concepts such as functioning, life use of CBT in clinical settings (McHugh & Barlow, satisfaction, and general optimism have been 2012). It appears that comparable outcomes are largely neglected (Chambless & Ollendick, 2001). achieved when CBT is administered to heteroge- Moreover, opponents of ESTs in general, and CBT neous clientele with varied clinical presentations in specific, have noted the relative lack of attention (Farrell, Schlup, & Boschen, 2010; Houghton, given to therapeutic relationship factors in treat- Saxon, Bradburn, Ricketts, & Hardy, 2009; Stewart ments. Such an assertion, however, ignores the fact & Chambless, 2007). A meta-analysis of 56 effec- that treatments that rely heavily on the therapeutic tiveness studies that examined CBT for adult anxiety relationship have also been written into treatment revealed that CBT produced large pre-post treatment manuals (e.g., Gibbons et al., 2012). Further, a effects, which strongly suggests that CBT is generally number of CBT manuals maintain that relationship effective in remediating adult anxiety (Stewart & factors are probably necessary, although it is Chambless, 2009). Of key relevance to the current unlikely that such factors are sufficient for thera- discussion, the effect sizes from effectiveness trials are peutic change (Dobson & Dobson, 2009). roughly comparable to those obtained in efficacy Unfortunately, it appears that populations that studies, which gives further credence to the applica- might benefit most from ESTs are the same tion of ESTs in real-world settings. populations that have no access to, or are rarely offered such treatments. For instance, there is is cbt practical? evidence that members of ethnic and racial minority The American Psychological Association Division groups tend to make less use of mental health services 12 Task Force on Promotion and Dissemination of than the majority population and terminate therapy Psychological Procedures (Chambless et al., 1998) sooner than members of majority groups (Wang et has developed a list of criteria to evaluate the level al., 2005; Ward, 2007). These disturbing facts exist of scientific support for a given intervention. One despite the fact that ethnically and racially diverse criterion for the highest level of support (i.e., populations experience greater economic and func- well-established treatments) stipulates that a treat- tional burden from emotional and behavioral ment must utilize a treatment manual, and thus disorders in comparison to individuals of majority have a standardized form of administration. From a groups (Huang, 2002). A study by Preciado (2012) dissemination perspective, the need for a treatment found an absence of cultural elements in randomized manual is understandable: Any practitioner who control trials of CBT, and some clinical trials had wants to implement an efficacious treatment must loosely defined “cultural adaptations” of CBT. comprehend what it is (and is not) in order to know Thus, in some studies the delivery of otherwise whether or not the intervention is being delivered as unchanged CBT protocols, when provided by intended. Unfortunately, it appears that practi- therapists of the target culture, were characterized tioners may be uneasy regarding the idea of as adaptations. manualized care (Kendall, 1998). For example,

Addis and Krasnow (2000) discovered a range of dialogue between researchers and practitioners. One opinions and concerns toward therapy manuals in a fairly direct implication of such research is that national survey of U.S. psychologists. They also proponents of evidence-based psychotherapy (cf. demonstrated that beliefs about the inflexibility of Goodheart, Kazdin, & Sternberg, 2006)needtodo manuals was related to more negative overall a considerably better job to translate research opinions about treatment manuals and their de- evidence into useful and practical models, which ployment in clinical practice. can be utilized in clinical practice by the modal Ollendick, King, and Chorpita (2006) summa- practitioner. For example, it may be that therapists rized this uneasiness, expressing concern that understand the conceptual framework related to clinicians may believe that treatment manuals are certain models of care, but do not know how to associated with inflexibility and may harm or even translate these models into practice. Alternatively, eliminate such issues as creativity, innovation, or they may know specific interventions, but not the what are sometimes seen as the “artistic” aspects of conceptual model that underpins such interventions, the process of psychotherapy. A number of pro- and therefore cannot flexibly use these models in ponents of the EST movement have responded to practice and fail to see their relevance. To our the concern about inflexible treatment manuals. For knowledge, no research to date addressed these instance, Kendall (1998) has argued that treatment potential competing models related to the theory- manuals can be used in a flexible, adaptive manner. practice gap. He maintained that good manuals allow for innovation and creativity within the theoretical why address the gap? and conceptual bounds of therapy, without sacrific- Some practitioners and indeed some researchers ing treatment fidelity. Kazdin (2000) argued that “it acknowledge the presence of a gap between theory is a convenient straw man to argue that manuals are and practice, but they are reluctant to bridge the gap, too rigid for clinical use” (p. 136). From his as they argue that the demands and dictates of perspective, some critics of manuals have perhaps research and practice are fundamentally irreconcil- used perceived rigidity as a justification to continue able. Proponents of this view maintain that research, to practice consistent with their preferred models. by definition, is highly structured, planned, and He has also asserted there are no good alternatives “clean,” while practice is often unstructured, unpre- to the use of manuals in the delivery of therapy, as dictable, and “messy” (Goodheart et al., 2006). As they provide the primary way to demonstrate that such, these disciplines may deal with nonoverlapping evidence-based interventions are actually being subject matters, and therefore it is not important to deployed as they were written and intended. Some try to reconcile these various efforts. It could even be authors have gone further still and argued that argued that research trials are conducted as “proof of some practitioners may actually prefer the use of concept” efforts, to show that a particular interven- treatment manuals, given their organized and tion can work under ideal circumstances, rather than instructional nature (e.g., LeCroy, 2008). Other whether it might work in the clinic setting. authors have advocated a broader view of dissem- Our perspective on this divide is that such ination, arguing that traditional education and arguments stem from the poor scientific training supervision methods are only some of the possible and general cynicism regarding the scientific enter- tools to promote the adoption of evidence-based prise that is held by a large number of practicing therapies in practice (Addis & Waltz, 2002). Some psychologists on the one hand, and a potential of the ideas related to dissemination are discussed arrogance and disrespect for the vagaries of clinical below. practice by some research psychologists on the other. As Baker and colleagues (2009) argue, by do clinicians know about cbt and adopting this spirit, psychologists have thus far other ests? failed to create a vital and robust applied science. A study by Boisvert and Faust (2006) revealed that As a result, psychologists are losing battles to other many clinicians had little knowledge about outcome health-care providers, especially medical personnel, research. Moreover, many clinicians believed that despite the lustrous empirical track-record of CBT research findings were more negative than what they in comparison to pharmacotherapy and other are in reality. This lack of knowledge was not related medical interventions. to therapist experience, theoretical orientation, or It has been argued that the broad implementation perceived knowledge of the literature. Boisvert and of ESTs is important for both practical and ethical Faust suggested that practitioners’ poor familiarity reasons. At a practical level, knowledge about with the outcome literature serves to exacerbate the ESTs allows program managers to recruit appropri- gap of theory-practice, and functions to sever the ately trained clinicians. At the level of the clinician,

Please cite this article as: Keith Dobson, Shadi Beshai, The Theory-Practice Gap in Cognitive Behavioral Therapy: Reflections and a Modest Proposal to Bridge the Gap, Behavior Therapy (2013), http://dx.doi.org/10.1016/j.beth.2013.03.002 gap 5 knowledge about ESTs allows decisions about which chief question: “How do we get our treatments to interventions to prioritize for identifiable disorders. the many people in need of service?” (p. 202). Thus, this knowledge base permits efficiency in Given the urgency of the matter, a number of making administrative and clinical decisions. From researchers have offered suggestions on how to an ethical perspective, knowledge about ESTs also mend the gap between theory and practice. The helps therapists to provide the highest level of care, recommendations made in the literature are related which is an ethical imperative. For example, if a to the planning and design of effectiveness trials, clinician is aware of the literature base, and knows training or maintenance of practitioner competen- that CBT is recommended as a first line of treatment cies after the conclusion of effectiveness trials, or to for moderate levels of major depression, it is arguably dissemination and policy change. unethical to use a nonsupported treatment, as such course of action will likely result in a longer course of protocol planning and design treatment and with less certain outcomes. In other Shafran and colleagues (2009) have made a number words, clinicians who are scientifically trained are of excellent recommendations regarding the devel- more likely to conduct a cost-benefit analysis of using opment and evaluation of ESTs in clinical practice. certain treatment protocols for particular clients. For example, they suggested that one way to With all other things being equal, and if such an address concerns of the relevance of ESTs in routine analysis is done correctly and impartially, the practice is to have treatment developers explicitly clinician will more likely resort to an EST since the describe how their trials address issues of comor- validated nature of ESTs (even if not for the specific bidity and disorder severity. It was also recom- configuration of problems and moderating factors mended that practitioners be provided with the that a specific patient presents with) makes them the appropriate supervision and training over the preferred choice for a number of conditions. As an course of effectiveness trials, as to ensure proper analogy, it falls below the required standard of care training and protocol fidelity. In addition, Shafran and is likely a cause of legal action if a surgeon et al. argued that clinicians should be encouraged to performed an intervention that is not evidence based. use outcome measures at set intervals throughout Why would or should the standard of care be any less the protocol, and that training in the use of such in the area of mental health? measures should be easily accessible and affordable Given all of the above, we believe that the to clinicians. Shafran et al. also noted that most correction of cynicism about ESTs should start trials fail to analyze therapist effects as a moderator early in the graduate training of mental health of outcome. As such, they recommend that trials therapists. Put differently, a solid scientific education consider and analyzes the effects of therapist coupled with quality training regarding the flexible expertise and level of training on treatment outcome. implementation of most ESTs is required to help If such effects are to be routinely used as part of combat the divide between science and practice. effectiveness trials, quality measures of therapist Further, we echo Kazdin’s(2008)sentiments in that competence and skill level need to be designed and clinicians should be encouraged to adopt research utilized. In addition, it was suggested that mecha- roles more often than the current tradition allows. nisms of action, an often illusive and neglected part of This activity would help the integration of science effectiveness research, should be routinely examined into regular practice, and ultimately bridge the gap in through dismantling protocols. perspectives. Similarly, as argued below, we believe Finally, Shafran and colleagues (2009) recom- that scientists would benefit from the experience of mended that future research should abandon the “the trenches” of routine practice, and this exposure one-size-fits-all model. Thus, there should be an would likely help program developers better address enhanced focus on matching clients to treatment practitioners’ concerns. type and intensity in an effort to reduce costs and allocate resources appropriately. In turn, adminis- Bridging the Gap tering less intensive, self-help, or web-based inter- Increased access to and training for optimal delivery ventions for clients who only require such services of ESTs is now viewed by the funders of health care could allow for increased benefit to those who systems, and those who deliver health care as an require face-to-face contact or more intensive urgent mandate. The widespread recognition about interventions, as the resources would be more the economic, social, and personal costs of mental available for such purposes. Significant efforts are disorders, as well as the demand for evidence-based needed to examine to what extent treatments can be treatments, has never been greater. Kazdin (2008) tailored to meet the needs of individual clients, aptly argued that internal debates in the field while retaining their optimal levels of efficacy. regarding the relevance of ESTs distract from the There may well be a trade-off between fidelity to the

Please cite this article as: Keith Dobson, Shadi Beshai, The Theory-Practice Gap in Cognitive Behavioral Therapy: Reflections and a Modest Proposal to Bridge the Gap, Behavior Therapy (2013), http://dx.doi.org/10.1016/j.beth.2013.03.002 6 dobson & beshai original treatment model and adaptation to the more, some studies (e.g., Beidas & Kendall, 2010) specific treatment model and exigencies and partic- suggest that didactic training alone is not sufficient ularities of a given case, and this interaction likely to change a practitioner’s behaviors. Rather, on- has implications for a variety of issues, such as going consultation and case supervision appear to acceptability of the treatment to the client, the be critical elements of competency enhancement perceived “fit” of the treatment to the client, the and maintenance (Rakovshik & McManus). therapist-client relationship, and outcome. More- Armed with this evolving literature base, treatment over, for a more complete understanding of the developers should pay particular attention to their psychotherapy process, the movement towards training models and procedure to ensure appropri- evidence-based practice can likely be integrated ate implementation of their protocols. We argue into the literature on empirically supported rela- that research needs to be devoted to which methods tionship factors. of competency training are most effective for Finally, developers should create manuals that psychotherapy practice. For example, most training are widely accessible at a reasonable cost. For programs precede practical training by didactic example, open access web sites could be created for knowledge. It may be, however, more effective to the dissemination and evaluation of ESTs. Finally, intersperse education and training, increasing the as Chorpita, Daleiden and Weisz (2005) have difficulty and complexity of cases as the trainee recommended, treatment manuals should allow develops and shows competency of basic building for an idiographic element, wherein the protocol blocks. Such an approach has been used for some is systematically adapted to fit the characteristics of time in the development of competent surgery and individual clients. medical skills (Leung, 2002), and this training model might perhaps be of use in evidence-based training and competency maintenance psychotherapy. Yet another model is the case-based The training of mental health providers remains one model of training, as has been used in fields such as of the most important pillars of effective CBT nursing (Thomas, O’Connor, Albert, Boutain, & dissemination and implementation. The difficulty Brandt, 2001). This model uses actual cases, with thus far has partly stemmed from the utilization of all of their complexities, to train clinicians to one-size-fits-all training models, which tend to be conceptualize and plan interventions. It has been focused around theoretical approaches to therapy, argued by Malcolm Gladwell in his book Outliers rather than to specific protocols for specific clinical that it requires 10,000 hours of experience to problems. Therapy training also tends towards attain world-class competence at a given activity didactic, course and workshop training, which can (Gladwell, 2008). If this assertion is valid, then be a necessary first step in training, but does not training programs should never expect emerging substitute for hands-on experiential practice with clinicians to be competent, but they should rather expert supervision. Also, given the diffused nature of expect that competency will evolve over time and as training in ESTs, a number of details are neglected in a result of ongoing experience. training. For instance, McHugh and Barlow (2012) argued that competency training (or “procedural dissemination, implementation, and learning for the application of knowledge to a clinical policy change encounter”; p. 44), as opposed to didactic training The EST movement has for the most part relied on (efforts designed to increase knowledge about the passive and, what McHugh and Barlow (2012) intervention), are understudied and consequently not called, diffused efforts of dissemination and public well understood. Further, much like efforts to match exposure, to date. These initial efforts need to be therapies to clients based on client characteristics, supplanted by more focused, directed strategies. efforts to train practitioners should be tailored to First, such directed strategies must examine current their needs, skill level, and credentials. As McHugh models of dissemination and modify and synthesize and Barlow have argued, “the fundamental chal- them to better fit extant data (Schoenwald, McHugh lenges of training are when to train, what interven- & Barlow, 2012). Dissemination strategies, much tions to provide training in, and who should serve as like implementation efforts, range from ineffective to trainers” (p. 49). effective, and as such, research that identifies the Despite the relative infancy of the field of success of these strategies is welcome. dissemination research (Brownson et al., 2012), Barlow (2004) has proposed that nomenclature there is emergent evidence that longer durations of influences the speed at which information is widely training in general, and longer competency training accepted and disseminated. He recommended in particular, are associated with better outcomes that psychologists begin labeling ESTs such as for CBT (Rakovshik & McManus, 2010). Further- CBT as “psychological treatments” in order to help

Please cite this article as: Keith Dobson, Shadi Beshai, The Theory-Practice Gap in Cognitive Behavioral Therapy: Reflections and a Modest Proposal to Bridge the Gap, Behavior Therapy (2013), http://dx.doi.org/10.1016/j.beth.2013.03.002 gap 7 differentiate them from more generic psychothera- interprofessional and international in scope. It may py. He has argued that this relabeling would allow be possible that web-based technology can be for better integration of these treatments into the developed, perhaps sponsored by an international health care system. Barlow (2010) later argued that health organization, such as the World Health the dissemination of CBT and other supported Organization or a consortium of international treatments not only depends on their acceptance in research agencies, to form a repository of evidence- the mental health community, but also upon the based psychotherapies. These organizations could exclusion of harmful and unbeneficial therapies build on the excellent work of existing national from routine practice. As such, he suggested that groups, such as Britain’s National Institute for front-line researchers, in collaboration with practi- Health and Clinical Excellence (see http://www. tioners, should come to consensus about the nice.org.uk/), which, from our perspective, is the definitions of harmful treatments and treatment best example in the world to date of bridging effects (Lilienfeld, 2007). evidence and clinical practice—for a wide range of Finally, it has been suggested that researchers and health issues and diagnoses. protocol developers do not adequately share the results of their efficacy and effectiveness trials with The Practice-Theory Gap: How Should Practice key funders and stakeholders. Consequently, the Inform Theory? pillars of effective dissemination and implementation We want to close this article with a brief discussion of (development, training, contextualization, etc.) are one of the critical aspects of the theory-practice gap, not sufficiently bolstered. It is therefore recom- but one that is often neglected in discussions of the mended that researchers need to provide the results topic. The issue that we refer to is the critical issue of of their research with key stakeholders and policy the bridge between practice and science. In many makers. Figure 1 characterizes this hypothesized discussions of the science-practice or theory-practice relationship. One challenge for the field will be who, gap, there is a certain directionality to the discussion, or what agency, should take responsibility for such which is that theorists design the intervention, an effort. Although there do exist some notable clinical scientists develop and test the intervention, efforts to bring together psychotherapy research and then practitioners are expected to implement the results for the purposes of meta-analyses (Cuijpers, intervention. This idea, that interventions begin in Anderson, Donker, & van Straten, 2011; Cuijpers et the university or research context, are evaluated al., 2011), it is likely the case that no single research there, and then disseminated to the field or practice, group, professional organization, or even a nation- is highly normative in many areas of applied science, level system can take on the responsibility to accu- but fails to capitalize on the range and depth of mulate such data-bases, since the efforts must be knowledge of many practitioners. Also, although

FIGURE 1 The conceptual linkage among treatment development factors and dissemination research.

Please cite this article as: Keith Dobson, Shadi Beshai, The Theory-Practice Gap in Cognitive Behavioral Therapy: Reflections and a Modest Proposal to Bridge the Gap, Behavior Therapy (2013), http://dx.doi.org/10.1016/j.beth.2013.03.002 8 dobson & beshai there have been many instances where clinicians have Berry, K., & Haddock, G. (2008). The implementation of the been consulted during the development of cognitive- NICE guidelines for schizophrenia: barriers to the implementation of psychological interventions and recommenda- behavioral therapies (cf. Martell, Dimidjian & tions for the future. Psychology and Psychotherapy: Theory, Herman-Dunn, 2010), this expertise is not always Research and Practice, 81, 419–436. utilized. A modest proposal that we would make is Boisvert, C. M., & Faust, D. (2006). Practicing psychologists’ that one way to bridge this gap is for developers and knowledge of general psychotherapy research findings: – evaluators of interventions to always include an Implications for science practice relations. Professional Psychology: Research and Practice, 37, 708–716. expert panel of clinicians on the development and Borntrager, C., Chorpita, B. F., Higa-McMillan, C., & Weisz, J. R. evaluation team. This panel should be empowered to (2009). Provider attitudes toward evidence-based practices: make broad comments on the scope or nature of the Are the concerns with the evidence or with the manuals? intervention, and to even modify specific elements Psychiatric Services, 60(5), 1–5. of the treatment, including the treatment manual. Our Brownson, R. C., Colditz, G. A., & Proctor, E. K. (Eds.). (2012). Dissemination and implementation research in suspicion is that this incorporation of clinicians health: Translating science to practice. New York, NY: would go some distance to ensure that the resulting Oxford University Press. interventions are more applicable to the clinical Chambless, D. L., Baker, M., Baucom, D. H., Beutler, L. E., world than is sometimes currently the case, more Calhoun, K. S., Crits-Christoph, P., . . . Woody, S. (1998). acceptable to the working clinician, and more easily Update on empirically validated therapies, II. Clinical Psychology, 51(1), 3–16. implemented in the clinical setting. Although we can Chambless, D. L., & Ollendick, T. H. (2001). Empirically well imagine that this process of direct consultation supported psychological interventions: Controversies and would require extra time and effort to develop and evidence. Annual Review of Psychology, 52, 685e716. test any given intervention, we suspect that the http://dx.doi.org/10.1146/annurev.psych.52.1.685 potential benefits far outweigh the initial costs. Chorpita, B., Daleiden, E., & Weisz, J. R. (2005). Modularity in the design and application of therapeutic interventions. Overall, our belief is that the field of CBT has Applied and Preventative Psychology, 11, 141–156. evolved tremendously since the early days of clinical Corsini, R. J., & Wedding, D. (Eds.). (2011). Current research. As a field, we now can point with justifiable psychotherapies (9th ed.). Florence, KY: CENGAGE Learning. pride at the number and range of interventions that Cuijpers, P., Andersson, G., Donker, T., & van Straten A. exist and meet criteria for being empirically support- (2011). Psychological treatment of depression: Results of a “ ” series of meta-analyses. Nordic Journal of Psychiatry; Early ed. Our suspicion is that the next generation of Online, 1–11. research must be focused on implementation, dis- Cuijpers, P., Donker, T., Johansson, R., Mohr, D. C., van semination, and effectiveness work, and that the Straten A., & Andersson, G. (2011). Self-guided psycholog- emerging field of dissemination science (Brownson et ical treatment for depressive symptoms: A meta-analysis. al., 2012) will have much to provide the field as it PLoS ONE 6(6), e21274. http://dx.doi.org/10.1371/ journal.pone.0021274. moves in this direction. This is indeed an exciting Dobson, K. S., & Dobson, D. J. G. (2009). Evidence-based time to address the gaps among science, theory, and practice of cognitive-behavioral therapy. New York, NY: practice. The Guilford Press. Ehlers, A., Gene-Cos, N., & Perrin, S. (2009). Low recognition of post-traumatic stress disorder in primary care. London References Journal of Primary Care, 1,36–42. Addis, M. E., & Krasnow, A. D. (2000). A national survey of Farrell, L. J., Schlup, B., & Boschen, M. J. (2010). Cognitive- practicing psychologists' attitudes toward psychotherapy behavioral treatments of child-hood obsessive-compulsive treatment manuals. Journal of Consulting and Clinical disorder in community-based clinical practice: Clinical Psychology, 68, 331–339. significance and benchmarking against efficacy. Behavior Addis, M. E., & Waltz, J. (2002). Implicit and untested Research and Therapy, 48, 409–417. assumptions about the role of psychotherapy treatment Gibbons, M. B. C., Thompson, S. M., Scott, K., Schauble, L. A., manuals in evidence-based mental health practice: Commen- Mooney, T., Thompson, D., . . . Crits-Christoph, P. (2012). tary. Clinical Psychology: Science and Practice, 9,421–424. Supportive-expressive dynamic psychotherapy in the com- Baker, T. B., McFall, R. M., & Shoham, V. (2009). Current status munity mental health system: A pilot effectiveness trial for and future prospects of clinical psychology. Psychological the treatment of depression. Psychotherapy, 49, 303–316. Science in the Public Interest, 9,67–103. http://dx.doi.org/ Gladwell, M. (2008). Outliers. New York, NY: Little, Brown. 10.1111/j.1539-6053.2009.01036.x. Goodheart, C. D., Kazdin, A. E., & Sternberg, R. J. (Eds.). (2006). Barlow, D. H. (2004). Psychological treatments. American Evidence-based psychotherapy: Where practice and research Psychologist, 59, 869–878. meet. Washington, DC: American Psychological Association. Barlow, D. H. (2010). Negative effects from psychological Haas, H. L., & Clopton, J. R. (2003). Comparing clinical and treatments: A perspective. American Psychologist, 65,13–20. research treatments for eating disorders. International Barlow, D. H., Levitt, J. T., & Bufka, L. F. (1999). The Journal of Eating Disorders, 33, 412–420. dissemination of empirically supported treatments: A view Houghton, S., Saxon, D., Bradburn, M., Ricketts, T., & Hardy, to the future. Behavior Research and Therapy, 37, 147–162. G. (2009). The effectiveness of routinely delivered cognitive Beidas, R. S., & Kendall, P. C. (2010). Training therapists in behavioral therapy for obsessive-compulsive disorder: A evidence-based practice: A critical review of studies from a benchmarking study. British Journal of Clinical Psychology, systems-contextual perspective. Clinical Psychology, 17,1–30. 49, 473–489.

Huang, L. (2002). Reflecting on cultural competence: A need secondary care. NICE Clinical guideline 82. London: for renewed urgency. Focal Point, 16,4–7. National Institute for Health and Clinical Excellence. Hyman, S. E. (2011). Diagnosis of mental disorder in light of Ollendick, T. H., King, N. J., & Chorpita, B. (2006). modern genetics. In D. A. Regier, W. E. Narrow, E. A. Kuhl, & Empirically supported treatments for children and adoles- D. J. Kupfer (Eds.). The conceptual evolution of DSM-5 cents. In P. C. Kendall (Ed.), Child and adolescent therapy: (pp. 3–17). Arlington, VA: American Psychiatric Association. Cognitive- behavioral procedures. New York, NY: Guilford Institute of Medicine. (2001). Crossing the quality chasm: A new Press. health system for the 21st Century. Washington, DC: Author. Preciado, J. (2012). Culture in Ibero-America: A neglected issue Kazdin, A. E. (2000). Psychotherapy for children and adolescents: in behavioral and cognitive randomized control trial Directions for research and practice. New York, NY: Oxford interventions. International Journal of Clinical and Health University Press. Psychology, 12, 489–501. Kazdin, A. E. (2008). Evidence-based treatment and practice: new Rakovshik, S. G., & McManus, F. (2010). ESTablishing opportunities to bridge clinical research and practice, enhance evidence-based training in cognitive-behavioral therapy: A the knowledge base, and improve patient care. The American review of current empirical findings and theoretical guid- Psychologist, 63,146–159. http://dx.doi.org/10.1037/ ance. Clinical Psychology Review, 30, 496–516. 0003-066X.63.3.146 Santa Ana, E. J., Martino, S., Ball, S. A., Nich, C. Frankforter, Kendall, P. C. (1998). Directing misperceptions: Researching T. L., & Carroll, K. M. (2008). What is usual about the issue facing manual-based treatments. Clinical Psychology: “treatment-as-usual”? Data from two multisite effectiveness Science and Practice, 58,729–740. trials. Journal of Substance Abuse Treatment, 35, 369–379. Kessler, R. C., Avenevoli, S., Green, J., Gruber, M. J., Guyer, Schoenwald, S. K., McHugh, R. K., & Barlow, D. H. (2012). M., He, Y., . . . Merikangas, K. R. (2009). National The science of dissemination and implementation. In R. K. comorbidity survey replication adolescent supplement McHugh & D. H. Barlow (Eds.), Dissemination and im- (NCS-A): III. Concordance of DSM-IV/CIDI diagnosis with plementation of evidence-based psychological interventions clinical reassessment. Journal of the American Academy of (pp. 16–42). New York, NY: Oxford University Press. Child and Adolescent Psychiatry, 48,386–399. Shafran, R., Clark, D. M., Fairburn, C. G., Arntz, A., Barlow, Kessler, R. C., Chiu, W. T., Demler, O., & Walters, E. E. D. H., Ehlers, A., . . . Wilson, G. T. (2009). Mind the gap: (2005). Prevalence, severity and comorbidity of 12 month Improving the dissemination of CBT. Behaviour Research DSM-IV disorders in the national comorbidity survey and Therapy, 47, 902–909. replication. Archives of General Psychiatry, 62, 617–627. Stewart, R. E., & Chambless, D. L. (2007). Does psychotherapy Kessler, R. C., Merikangas, K. R., & Wang, P. S. (2007). research inform treatment decisions in private practice? Journal Prevalence, comorbidity, and service utilization for mood of Clinical Psychology, 63,267–281. http://dx.doi.org/ disorders in the United States at the beginning of the twenty-first 10.1002/jclp.20347. century. Annual Review of Clinical Psychology, 3,137–158. Stewart, R. E., & Chambless, D. L. (2009). Cognitive-behavioural http://dx.doi.org/10.1146/annurev.clinpsy.3.022806.091444 therapy for adult anxiety disorders in clinical practice: A Lambert, M. J. (2010). Using outcome data to improve effects meta-analysis of effectiveness studies. Journal of Consulting of psychotherapy: Some illustrations. In M. J. Lambert (Ed.), and Clinical Psychology, 77(4), 595–606. Prevention of treatment failure: The use of measuring, Stirman, S. W., Crits-Christoph, P., & DeRubeis, R. J. (2004). monitoring, and feedback in clinical practice (pp. 203–242). Achieving successful dissemination of empirically supported Washington, DC: American Psychological Association. psychotherapies: A synthesis of dissemination theory. LeCroy, C. W. (Ed.). (2008). Handbook of evidence-based Clinical Psychology: Science and Practice, 11, 343–359. treatment manuals for children and adolescents (92nd ed.). Stobie, B., Taylor, T., Quigley, A., Ewing, S., & Salkovskis, P. M. New York: Oxford University Press. (2007). “Contents may vary”: A pilot study of treatment Leung, W-C. (2002). Competency based medical training: A histories of OCD patients. Behavioural and Cognitive review. British Medical Journal, 325(7366), 693–696. Psychotherapy, 35,273–282. http://dx.doi.org/10.1017/ Lilienfeld, S. O. (2007). Psychological treatments that cause S135246580700358X harm. Perspectives on Psychological Science, 2,53–70. Thomas, M. D., O’Connor, F. W., Albert, M. L., Boutain, D., & Lopez-Duran, N. (2011, October 18). Fifty percent (50%) of Brandt, P. A.. (2001). Case-based teaching and learning teens have experienced a psychiatric condition by their 18th experiences. Issues in Mental Health Nursing, 22, 517–531. birthday. Child Psychology Research. Retrieved September Wang, P. S., Lane, M., Olfson, M., Pincus, H. A., Wells, K. B., 21, 2012, from http://www.child-psych.org. & Kessler, R. C. (2005). Twelve-month use of mental health Martell, C. R., Dimidjian, S., & Herman-Dunn, R. (2010). services in the United States: results from the National Behavioral activation for depression: A clinician’s guide. Comorbidity Survey Replication. Archives of General New York, NY: Guilford Press. Psychiatry, 62(6), 629–640. Meyer, G. J., Finn, S. E., Eyde, L. D., Kay, G. G., Moreland, K. L., Ward, E. C. (2007). Examining differential treatment effects for Dies, R. R., . . . Reed, G. M. (2001). Psychological testing and depression in racial and ethnic minority women: a qualita- psychological assessment: A review of evidence and issues. tive systematic review. Journal of the National Medical American Psychologist, 56,128–165. Association, 99, 265–274. McHugh, R. K., & Barlow, D. H. (Eds.). (2012). Dissemination and implementation of evidence-based psychological interventions New York, NY: Oxford University Press. RECEIVED: October 4, 2012 NICE. (2009). Schizophrenia: Core interventions in the treat- ACCEPTED: March 1, 2013 ment and management of schizophrenia in primary and Available online xxxx