See discussions, stats, and author profiles for this publication at: esearch on psychotherapy integration: Buildingon the past, looking to the futureARTICLE in PSYCHOTHERAPY RESEARCH · MARCH 2015Impact Factor: 1.75 · DOI: S11471675 AUTHORS, INCLUDING:Louis CastonguayCatherine F. EubanksPennsylvania State UniversityYeshiva University181 PUBLICATIONS 2,774 CITATIONS26 PUBLICATIONS 246 CITATIONSSEE PROFILESEE PROFILEJohn Christopher MuranWolfgang LutzAdelphi UniversityUniversität Trier142 PUBLICATIONS 1,557 CITATIONS172 PUBLICATIONS 1,499 CITATIONSSEE PROFILESEE PROFILEAvailable from: John Christopher MuranRetrieved on: 20 August 2015

This article was downloaded by: [Society for Psychotherapy Research ]On: 24 March 2015, At: 12:48Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UKPsychotherapy ResearchPublication details, including instructions for authors and subscription esearch on psychotherapy integration: Building onthe past, looking to the futureabcdLouis G. Castonguay , Catherine F. Eubanks , Marvin R. Goldfried , J. Christopher Murancef& Wolfgang LutzaDepartment of Psychology, Penn State University, State College, PA, USAbDepartment of Psychology, Yeshiva University, New York, NY, USAcDepartment of Psychiatry, Mount Sinai Beth Israel, New York, NY, USAdDepartment of Psychology, Stony Brook University, Stony Brook, NY, USAeDerner Institute of Advanced Psychological Studies, Adelphi University, Garden City, NY,USAClick for updatesfDepartment of Psychology, Universität Trier, Trier, GermanyPublished online: 24 Mar 2015.To cite this article: Louis G. Castonguay, Catherine F. Eubanks, Marvin R. Goldfried, J. Christopher Muran & Wolfgang Lutz(2015): Research on psychotherapy integration: Building on the past, looking to the future, Psychotherapy Research, DOI:10.1080/10503307.2015.1014010To link to this article: ASE SCROLL DOWN FOR ARTICLETaylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of theContent. Any opinions and views expressed in this publication are the opinions and views of the authors, andare not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon andshould be independently verified with primary sources of information. Taylor and Francis shall not be liable forany losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use ofthe Content.This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at ns

Psychotherapy Research, 0METHOD PAPERResearch on psychotherapy integration: Building on the past, lookingto the futureLOUIS G. CASTONGUAY1, CATHERINE F. EUBANKS2,3, MARVIN R. GOLDFRIED4,J. CHRISTOPHER MURAN3,5, & WOLFGANG LUTZ6Downloaded by [Society for Psychotherapy Research ] at 12:48 24 March 20151Department of Psychology, Penn State University, State College, PA, USA; 2Department of Psychology, Yeshiva University,New York, NY, USA; 3Department of Psychiatry, Mount Sinai Beth Israel, New York, NY, USA; 4Department ofPsychology, Stony Brook University, Stony Brook, NY, USA; 5Derner Institute of Advanced Psychological Studies, AdelphiUniversity, Garden City, NY, USA & 6Department of Psychology, Universität Trier, Trier, Germany(Received 12 August 2014; revised 14 January 2015; accepted 27 January 2015)AbstractIntegration has become an important and influential movement within psychotherapy practice, reflected by the fact thatmany treatment providers now identify as integrative. However, integration has not had as great an influence onpsychotherapy research. The goal of this paper is to highlight the growing body of research on psychotherapy integration,and to identify future directions for research that may strengthen the integration movement as well as the field ofpsychotherapy as a whole. We first summarize the past 25 years of research on integration, with a focus on four approachesto integration: theoretical integration, technical eclectic, common factors, and assimilative integration. Next, we identifydirections of research within these four areas that could strengthen and support integrative practice. We then propose waysin which the perspective of integrationists could contribute to psychotherapy research in the critical areas of harmful effects,therapist effects, practice-oriented research, and training. We end this paper by suggesting that a greater collaborationbetween integrationists and psychotherapy researchers will help to create a unified landscape of knowledge and action thatwill benefit all participants and advance the field.Keywords: psychotherapy integration; psychotherapy research; theoretical orientation; trainingIntegration, as a theme or movement in psychotherapy, is old but vibrant. We have seen, over the last100 years, a rich conceptual and clinical literaturedelineating points of convergence and complementarity between different therapeutic approaches, aswell as describing how they can be combined orintegrated with the goal of providing more comprehensive views of psychopathology and more effectivepractice (Goldfried, Pachankis, & Bell, 2005). Psychotherapy integration could now justifiably bereferred to a “leitmotiv” or “zeitgeist” in the field ofpsychotherapy (see Castonguay & Goldfried, 1994).Reflecting this influential status in the literature, alarge number of mental health providers across manycountries define themselves as integrative or eclectic(e.g., Caspar, 2008; Kazantzis & Deane, 1998;Muller, 2008; Norcross, Karpiak, & Santoro, 2005;Yin, Huang, & Fu, 2009). In contrast with some ofthe “pure” forms of therapy, however, empiricalefforts within the psychotherapy integration movement have lagged behind conceptual and clinicalcontributions. Perhaps reflecting the divide betweenscience and practice, there is a vast discrepancybetween the influence that an integrative perspectiveappears to have in day-to-day clinical work and theimpact it has on academic research. Nevertheless, webelieve that research that has been done under theumbrella of psychotherapy integration can providehelpful clinical guidelines. We also believe that anintegrative perspective is likely to be relevant ininvestigating some of the most complex and intriguing questions facing the field.Correspondence concerning this article should be addressed to Louis G. Castonguay, Department of Psychology, Penn State University, 354Moore Building, University Park, State College, PA 16802, USA. Email: [email protected] 2015 Society for Psychotherapy Research

Downloaded by [Society for Psychotherapy Research ] at 12:48 24 March 20152L. Castonguay et al.As part of a special series devoted to the 25thanniversary of Psychotherapy Research, the first goal ofthis paper is to briefly describe empirical investigations that have been conducted over the last quartercentury on issues related to psychotherapy integration. Not intended to be comprehensive (andundoubtedly reflecting the cognitive-behavioral leanings of many of the authors of this paper), oursynopsis will highlight how some of the findingssuggest new understandings of the therapeutic process, as well as innovative ways to expand and/orimprove the clinical repertoire—whether it is acrossorientations or within particular approaches. Oursecond goal is to delineate lines of research thatcould provide further support to integrative practiceand, reciprocally, to identify research directions thatcould benefit from the integration movement. Wealso hope to show that paving this two-way streetmay foster greater collaboration between researchersand clinicians.A Synopsis of the Past 25 Years of Research inPsychotherapy IntegrationBroadly defined, psychotherapy integration refers toa movement of conceptual and clinical rapprochement “which is not only an effort to integrate diversemodels and techniques but also an attempt to betterunderstand and improve psychotherapy by considering the perspective of different approaches” (Castonguay & Goldfried, 1994, p. 160). Arguably, themost prominent contributions in psychotherapyintegration can be parceled into four major trends(see Norcross, 2005): theoretical integration, technical eclectic, common factors, and assimilativeintegration. Theoretical integration involves theintegration of the theories and techniques of two ormore psychotherapies into a new conceptualizationof change or treatment approach. Technical eclectic,by contrast, entails the use of techniques fromdifferent approaches without attempting to create anew conceptual model that integrates the diversetheories that underlie them. The common factorsapproach focuses on the components and principlesthat are shared across orientations and highlights thetherapeutic impact of these common elements overaspects of treatments that are purportedly unique.Finally, assimilative integration, in contrast to theother three forms of integration, involves remaininganchored in a primary theoretical orientation whilethoughtfully integrating techniques and principlesfrom other orientations. Below, we describe thesetrends further and highlight a few prominent examples of research programs conducted within each ofthese frameworks.Theoretical IntegrationA large number of models have been developed withthe goal of either integrating or transcending constructs originally associated with divergent orientations. Unfortunately, relatively few of them havegenerated substantial research. Among the exceptionsare two theoretical models about the process ofchange. The earliest of these is the trans-theoreticalmodel, which portrays the development of psychotherapy across five stages of change (pre-contemplation, contemplation, preparation, action, andmaintenance), each of them characterized by processes of change (e.g., awareness raising, contingencymanagement) that are likely to be optimized byinterventions of diverse schools of therapy (Prochaska, 1979; Prochaska & DiClemente, 1984,2005). A substantial amount of research has beenconducted based on this approach, especially onhealth behaviors and the treatment of substanceabuse. Findings suggest, for example, that the client’spretreatment stage of change reliably predicts psychotherapy outcome (Norcross, Krebs, & Prochaska,2011). Stiles’ assimilative model also defines theprocess of therapeutic change along several phasesthat progress from the warding off to the assimilationof internal voices that represent the client’s problematic experiences. Stiles and colleagues have built abody of studies that provide promising preliminaryempirical support for this model (see Stiles, 2011).In addition to these two theoretical approaches,newer integrative models have begun to garnerempirical support. Guided by dynamic or chaostheories of change, Adele Hayes has developed amodel for the treatment of depression that involves aseries of interventions (associated with diverse orientations) aimed at fostering three phases of change:stabilization of the client’s life functioning, destabilization of the client’s emotional and cognitive experience, and re-stabilization of the client’s view of selfand behavior. Built in part on intriguing processfindings (such as the link between emotional processing and exploration of the past and outcome incognitive therapy [CT]), both the process and theoutcome of this integrative treatment have begun toreceive empirical support (Grosse Holtforth et al.,2011; Hayes, Beevers, Feldman, Laurenceau, & Perlman, 2005; Hayes, Feldman, & Goldfried, 2007).Tying together constructs of psychodynamic theory(attachment), CT (schema), and basic (social psychology) literatures, Constantino and Westra (2012)have offered an expectancy-based model to explain aspecific mechanism or process of change, i.e., corrective experiences in therapy. Their approach alsoinvolves a sequence of distinct phases that are fosteredby a variety of interventions, such as the validation of

Downloaded by [Society for Psychotherapy Research ] at 12:48 24 March 2015Psychotherapy Researchinterpersonal expectations early in treatment, theprovision of challenging (corrective) feedback midtreatment, and confirmation late in therapy of revisedexpectations of self and others. Evidence supportingthis model includes the finding that in cognitivebehavioral therapy (CBT), early confirmation bythe therapist of the client’s view of self is relatedto the alliance (Constantino, Arnow, Blasey, &Agras, 2005).Interestingly, a number of similarities can be foundacross these models, such as a focus on a period ofacceptance or stabilization, followed by challenge oraction, and finally solidification. In addition, all ofthese theories offer guidelines that could help clinicians adjust their interventions based on the specificneeds of clients and/or phases of intervention.Because the interventions are not restricted to oneorientation, these models can also help therapists toexpand their repertoire of techniques to meet specifictreatment goals or objectives.It should be noted that broad theoretical models ofpersonality, psychopathology, and psychotherapyhave been developed by some of the pillars of theintegration movement, such as Paul Wachtel (1977)and Barry Wolfe (2005). To a large extent, however,these models have not yet been the direct source ofsystematic research programs, with the noteworthyexception of the work of Klaus Grawe (see Caspar &Grosse Holtforth, 2010). Grawe proposed a conceptual framework aimed at explaining human functioning from a complex interaction of motivation,cognition, and learning factors. Based on constructsderived from several theoretical traditions (includingbasic sciences such as psychopathology, cognitiveand developmental psychology), this model emphasizes the determining role of needs, approach andavoidance motives, core schemas, and past experiences. Interventions are aimed at realizing fourtherapeutic factors: clarification, problem activation,resource activation, and mastery. Corrective experiences, both emotional and cognitive, play an important role in the process of change. This model hasserved as the basis of a large number of studiesconducted by Grawe and colleagues, most notablyFranz Caspar (2007), related to assessment, caseformulation, process, and outcome. Some of thefindings, for example, have shown that the fosteringof different types of corrective experiences are linkedto improvement in different phases of therapy(Grosse Holtforth, Grawe, & Castonguay, 2006).Furthermore, a randomized controlled trial (RCT)has found that an integrative CBT protocol guidedby Grawe’s consistency theory (General Psychotherapy; Grawe, 2004) led to better outcome thantraditional CBT for highly symptomatic clients(Grosse Holtforth et al., 2011). Like the integrative3research described above, this study points to thebenefit of expanding our clinical focus and adjustingtreatment interventions to meet the needs of particular clients.Technical EclectismOne could argue that the earliest publications aboutintegration were mostly by and for academicians,such as French’s (1933) attempt to draw parallelsbetween Freud and Pavlov, Rosenzweig’s (1936)description of common factors, and Dollard andMiller’s (1950) seminal translation of psychodynamic concepts within learning principles. Onemight also argue that the first books and papersabout integration that were of direct clinical relevance were about eclectism—describing when andhow to use specific approaches, or a combination ofapproaches, for a particular client. Interestingly, thisprescriptive and individualized philosophy has beena point of convergence for clinicians and researchersfor a long time. In the late sixties, Gordon Paul(1967) made a recommendation for future researchthat became a mantra for many psychotherapyresearchers: identifying what treatment is the mosteffective for a particular client in a specific situation.A few years later, one of the earliest and mostinfluential surveys of clinical practice revealed thata majority of therapists identified as eclectic (Garfield & Kurtz, 1976). The pragmatic use and combination of theoretically different techniques is stillprominent in today’s clinical practice (Norcross,2005). Formal eclectic systems have also beenproposed; however, many of them have not generated substantial and sustained empirical research,including some of the most visible ones (e.g.,Lazarus, 2005). A noteworthy exception is the workof Larry Beutler and colleagues.Based on a review of both client characteristicsand processes of change underlying different formsof psychotherapy, Beutler (1979) set up a researchprogram to test specific propositions about treatmentmatching. Beutler’s systematic treatment selectionapproach has led to two prescriptive principles thathave been recognized as empirically supported(Norcross, 2011). The first is based on the client’slevel of reactance, or resistance toward efforts fromothers to control him or her. Research indicates thatclients with high levels of reactance will benefit morefrom treatment when therapists are less directive,whereas clients with low levels of reactance will showfurther improvement with therapists at the higherend of the directiveness spectrum. The secondprescriptive principle is based on the client’s copingstyle. Specifically, evidence suggests that clients whocope with stressful events by blaming themselves and

Downloaded by [Society for Psychotherapy Research ] at 12:48 24 March 20154L. Castonguay et al.ruminating (or internalizing) will be more responsiveto treatments that foster self-exploration. By contrast, interventions directly aimed at behavioralchange and symptom reduction appear to be moreeffective for clients who tend to blame others and actout (or externalize) in the face of difficult events.By focusing on client characteristics that are nottied to a particular disorder, these principles offertherapeutic guidelines that have broad implicationsin terms of client populations. Furthermore, theseprinciples are not restricted to specific types oftreatment and can therefore be relevant to cliniciansof different orientations. Perhaps more importantly,these principles do not impose drastic changes onthe practice of most clinicians. For example, directive cognitive therapists do not have to abandon theirpreferred set of cognitive-behavioral interventionswhen working with highly reactant or internalizingclients. Rather, these empirically based principlessimply suggest that cognitive therapists can adapt theway they use their preferred procedures to be moreattuned to their clients’ individual characteristics(see Castonguay, 2000; Goldfried & Castonguay, 1992).Common FactorsWhile eclectism may have been the first integrativetheme to directly guide clinicians, the delineation ofcommon factors—constructs and components oftherapy that cut across different theoretical orientations—may have been the earliest and most consistenttopic to generate researchers’ interest in the integration movement. Such interest was stimulated bymany factors, including reports that therapists ofdiverse allegiances did not restrict themselves tothe interventions prescribed by their respectiveapproaches, and many behaved in ways that weremore similar than dissimilar (e.g., Fiedler, 1950;Klein, Dittman, Parloff, & Gill, 1969); the boldhypothesis that some relationship variables mightnot only be necessary but sufficient for therapeuticchange to occur (Rogers, 1957); early findings suggesting that the outcome of theoretically divergentapproaches might be more equivalent than discrepant(e.g., Frank, 1961); and the suggestion that many ofthe “unique” interventions of particular orientationsare idiosyncratic manifestations of more generalstrategies or principles of change, such as the increaseof positive expectations, establishment of a therapeutic relationship, provision of a new view of self,facilitation of corrective experiences, and the continued testing of change with day-to-day reality (Goldfried, 1980; Goldfried & Padawer, 1982).For a long time, many researchers (especially thoseassociated with the behavioral tradition) referred tocommon factors as “non-specific” variables. As such,they were defined as factors that are auxiliaries totheoretically driven techniques and whose nature andimpact are not yet understood (see Castonguay,1993). Today, common factors have not only beenrecognized as legitimate therapeutic processes, theyare by far the variables that have received the mostempirical attention in psychotherapy processresearch. The work of Bruce Wampold (e.g., Laska,Gurman, & Wampold, 2014; Wampold & Imel,2015) deserves special mention for both highlightingand building research support for the role of commonfactors in explaining therapeutic change acrossorientations.One common factor that has received considerableattention in the research literature is the workingalliance. In a recent review, Horvath, Del Re, Flückiger, and Symonds (2011) identified more than 200research reports on the working alliance alone (andonly for individual therapy with adults). The correlation between the alliance and outcome is robustacross different types of therapy, including CBT,and remains so even when moderators such as studydesign and researcher allegiance are included in theanalysis (Flückiger, Del Re, Wampold, Symonds, &Horvath, 2012). Clearly, the nature of the alliance canbe defined and measured. Moreover, its link withoutcome and other crucial aspects of therapy challenges the perception of it as a mere “auxiliary” ornoninstrumental factor. However, it must be notedthat there is controversy about whether the alliance isan important causal factor in producing change (e.g.,DeRubeis, Brotman, & Gibbons, 2005). Concernshave been raised that claims in favor of commonfactors like the alliance have been overstated, andevidence in favor of specific factors should not beoverlooked (e.g., Asnaani & Foa, 2014). Additionalresearch using rigorous methodologies, includingboth RCTs (see Crits-Christoph, Chambless, &Markell, 2014) and qualitative studies (e.g., Nilsson,Svensson, Sandell, & Clinton, 2007), is needed toclarify the relative importance of common versusunique factors.As reflected in the work of two recent Task Forces(Castonguay & Beutler, 2006a; Norcross 2002,2011) several other relationship variables that cutacross theoretical orientations have received empirical support, such as empathy and positive regard. Itshould also be noted, however, that common factorsare not restricted to relationship variables (Castonguay, 1993, 2006; Grencavage & Norcross, 1990;Lambert, 2013) and that a number of non-relationalfactors have also been the focus of research, such astherapist focus of interventions (Goldfried, Raue, &Castonguay, 1998), exposure (see Weinberger &Rasco, 2007), as well as procedures that foster the

Downloaded by [Society for Psychotherapy Research ] at 12:48 24 March 2015Psychotherapy Researchclient’s acquisition of a new perspective of self(Castonguay & Hill, 2007) and that facilitate corrective experiences (Castonguay & Hill, 2012). Asubstantial number of learning (e.g., feedback) andaction (e.g., modeling) factors have also been identified by Lambert (2013) as treatment commonalities. Along with supportive (mostly relationship)variables, these factors are presented as one explanation for the lack of marked differences betweenvarious forms of therapy. According to Lambert,most of the variables and constructs he identified“have been operationally defined and then correlatedwith outcome in research studies of therapy”(p. 199).It should also be noted that a number of commonfactors might be best defined as “faux unique”variables: variables that have been traditionally associated with one form of therapy but that may serve abeneficial role in other orientations. For example,although the deepening of emotional experience andthe exploration of the past have been found to be lessfrequent in CBT than in psychodynamic therapy, bothprocesses have been linked to improvement in CBT(Castonguay, 2011). Faux unique factors, however,are not only relevant to the practice of CBT. As foundin a recent investigation, the use of CBT interventionswas associated with an increase in insight, suggestingthat such interventions might facilitate treatment goalsemphasized in psychodynamic therapy (McAleavey &Castonguay, 2014). Also supporting the same idea aremuch older findings suggesting the impact of operantconditioning on the practice of Rogerian therapy(Truax, 1966).While common factors have clearly sparked theinterest of researchers, they can also have meaningfulclinical relevance. For example, research on commonfactors can and should serve as evidence-based guidelines for the training of graduate students, irrespectiveof the theoretical orientation of the training program(Beck et al., 2014; Boswell & Castonguay, 2007).Furthermore, such research provides clinicians of allexperience levels with a list of interventions that theycan use in establishing clinical rapport, facilitatingclient engagement in prescribed interventions, as wellas identifying and resolving alliance ruptures ordifficulties during treatment. An excellent exampleof the clinical application of common factors researchis Swift and Greenberg’s (2015) investigation ofpremature termination in psychotherapy. The authorsdelineate evidence-based strategies to reduce dropoutthat draw on common factors such as strengtheninghope and enhancing motivation.While several common factors, such as the allianceand empathy, should be viewed as basic (albeit notsimple to enact and optimize) tools for all forms ofpsychotherapy, several faux unique variables can5help clinicians to expand their repertoire of interventions. And like the principles related to theclient’s reactance level and coping style, the integration of faux unique variables does not necessarilyrequire drastic changes on the part of practicingtherapists. CBT therapists do not have to cease theirattempts to impart coping skills because they havebegun to pay more systematic attention to developmental issues, and psychodynamic therapists do nothave to deny the importance of insight just becausethey are looking to other theoretical orientations fortools to foster it.Assimilative IntegrationAlthough the psychotherapy integration movementhas been influential, we also believe that the fourprimary traditions of psychotherapy—CBT, humanistic/experiential, psychodynamic, and systemic—willsurvive and continue to grow. One might say that thisprediction is simply reflecting an epistemologicaldestiny, as these four paradigms can be viewed as acontemporary manifestation of longstanding ways ofaccumulating and using knowledge, either by relyingon observation and logical thinking, by focusing onphenomenological experience, by interpreting orconstructing the reality with which humans are confronted, or by appreciating the complex relationalsystems that guide human development across thelifespan. We would also argue that one way for thesefour major approaches to grow is to assimilate, in acohesive way, concepts and techniques of otherorientation. This assimilative trend of integration isthe youngest of the four trends of integration, but it isone that could, in our opinion, be of strong appeal toresearchers and clinicians alike.1For researchers whose theoretically driven researchprograms are associated with one of the majororientations, an assimilative perspective offers a window for bold and innovative expansions. Rather thanengaging in a frequently illusory attempt to reinventthe wheel, assimilative integrationists can devote theircreativity and insightfulness to refining a solid tradition of thinking and practice. It is difficult to imaginea better way to accumulate knowledge and thusadvance science. Similarly, for clinicians, an assimilative approach allows for an expansion of clinicalrepertoire without shaking the foundations of theirmost typical ways of practicing. As emphasizedelsewhere (Castonguay, 2011, 2013), it may well bethat one fruitful way to improve the effectiveness ofpsychotherapy is to build upon our conceptual,empirical, and clinical foundations while openingourselves to potential contributions of researchersand practitioners working in other communities ofknowledge seekers.

Downloaded by [Society for Psychotherapy Research ] at 12:48 24 March 20156L. Castonguay et al.A number of research programs can be identifiedunder the theme of assimilative integration. Two ofthem are directly related to Jeremy Safran’s expansion of the CT concept of schema (Safran, 1990a,1990b; Safran & Segal, 1990). By emphasizing theinterpersonal, developmental, motivational, emotional, and conflictual aspects of this concept thathave traditionally received less attention in CBTthan in other orientations (see Blagys & Hilsenroth,2000; Jones & Pulos, 1993), Safran has provided aconceptual basis to widen the scope of the cognitiveinterventions. Among the interventions described bySafran to change core schem

Research on psychotherapy integration: Building . Adelphi University, Garden City, NY, USA f Department of Psychology, Universität Trier, Trier, Germany Published online: 24 Mar 2015. . appears to have in day-to-day clinical work and the impact it ha