According to Beck, the cognitive approach to psychotherapy “is best viewed as the application of the cognitive model of a particular disorder with the use of a variety of techniques designed to modify dysfunctional beliefs and faulty information processing characteristic of each disorder”. At its current stage of development, CBT is considered one of the best validated psychotherapies available (Bennett-Levy et al. 2004, Salkovskis 1996), indeed, surveys indicate that CBT is fast becoming the majority orientation of practising psychologists (Gaudiano 2008). The fundamental premise of CBT is that emotional experience, behaviour and physiological sensations are influenced by individuals’ cognitive processes. Each of these realms of human experience is synergistically co-determining, with cognition mediating their reciprocal interrelationship (Sanders and Wills 2005); Clark & Steer in Salkovskis, 1996). CBT asserts that the appraisal or meaning assigned by an individual to these processes that will determine their impact on psychological well-being (Beck 1976). Essentially, there may be several alternative ways of looking at experiences (Salkovskis 1996), that individuals actively construct their realities (Clark & Steer in Salkovskis, 1996). The crucial role of cognitions in predisposing and maintaining psychological difficulties is emphasised, cognitive techniques are therefore viewed as a key element of the intervention process (Clark & Steer in Salkovskis, 1996). The CBT model proposes that psychopathology is the product of faulty information processing that manifests itself in distorted and dysfunctional thinking which directly leads to negative emotions and maladaptive behaviours. Therapy aims to help clients identify and modify these distorted patterns of thinking (Bennett-Levy et al. 2004) by empowering them to choose alternative ways of interpreting and reacting, drawing from the fullest possible range of alternatives available. Psychopathology is conceptualised as occurring on a continuum, symptoms being viewed as extreme variants of relatively common human experiences. For example, in anxiety disorders, cognitive models specify circumstances under which otherwise normal cognitive processes become “stuck” resulting in pathological levels of anxiety (Salkovskis 1996). Furthermore, CBT postulates the existence of cognitive styles and biases characteristic of particular psychological disorders known as the cognitive content specificity hypothesis.
In practice, CBT is a collaborative, time limited, structured, educational, empirical and active form of therapy utilising a variety of methods, some specific to CBT, others adopted from other therapeutic approaches. Not only has CBT developed rapidly since Beck’s seminal work on depression, it has also changed greatly over the last 15 years with the emergence of CBTs 3rd wave approaches (Mansell 2008). Therapies falling under the CBT umbrella now include problem-solving therapy, DBT, ACT, rational-emotive behaviour therapy, cognitive processing therapy, MBCT and cognitive behavioural analysis of psychotherapy (Gaudiano 2008). Initially, this paper will explore the defining characteristics of conventional, mainstream CBT providing a brief overview of therapy’s structured format, its emphasis on collaborative therapeutic relationships and its cognitive and behavioural intervention techniques. The paper will endeavour to highlight throughout the defining characteristics of CBT, namely, its empirical, educational, parsimonious and adaptable approach to treatment using examples from a variety of clinical disorders. Examination of the main tenets of CBT will be followed by discussion of limitations associated with the approach. CBT has been criticised on multiple fronts, both from within CBT and from alternative therapeutic perspectives (Andersson and Asmundson 2008). A number of CBTs critiques will be examined including CBTs coherence as a theoretical framework, the role of the therapeutic relationship, its applicability across various populations and its current status within the broader socio-political context of mental health care provision. Limitations will be evaluated given recent developments within the discipline, primarily the emergence of the complementary therapeutic approaches heretofore mentioned. This paper will examine whether recent developments have sufficiently addressed CBTs limitations and will conclude with a discussion of possible areas of future research and development within the discipline.
“CT is not a technique driven treatment, it should be driven by individual case conceptualisations that are based on a specific cognitive model” (Wells 1997). The CBT assessment is a valuable opportunity to engage and socialise the client in the CBT approach, in which the therapist encourages the client to view therapy as a no lose experiment. In contrast to other therapeutic approaches, CBT assessments primarily focus on a detailed description of the current presenting problem (frequency, circumstances/situational aspects and associated distress levels), analysis of symptom, cognitive and behavioural details, historical and aetiological factors (primarily predisposing or triggering factors) and social/environmental influences. The majority of assessment is devoted to the identification and objectification of current presenting problems. Symptom presentation may be analysed in terms of its antecedents, associated appraisals and beliefs and emotional or behavioural consequences (Wells 1997). In panic disorder, assessment would focus on the nature of catastrophic misinterpretations, a detailed description of the primary feared symptoms and the nature of safety and avoidance behaviours (Wells 1997). During assessment and indeed throughout the course of therapy, symptom and mood measures are used to establish base line rates and to assist in monitoring progress over time.
Assessment is an important element of the formulation or cognitive case conceptualisation process. Case conceptualisation is envisaged as the process whereby client and therapist reach a shared understanding of the origins, development and maintenance of the clients’ problems. The formulation is an overview, essentially a working hypothesis, which is open to testing and verification throughout therapy and provides a tentative intervention plan. Conceptualisation is an active and continuing process throughout therapy, thus, the formulation is open to addition and revision as therapy proceeds (Sanders and Wills 2005). As the client becomes increasingly socialised in the CBT approach, SBs and cognitive biases may be incorporated into the initial formulation to aid clarity, understanding and planning. Crucially, the evolving formulation educates the client in the interaction between thoughts, emotions, behaviour and physiology, illustrating how the elements interact, trigger and maintain their problems. The assessment and formulation process allow the therapist to draw upon their knowledge of cognitive models of psychological difficulties, tailoring these models and associated protocols to the individuals case conceptualisation, thereby creating a bridge between theory and practice. At its simplest level, the formulation may be a diagrammatic representation of the clients’ problems, in which therapist and client plot the sequence of events in symptom presentation, often using vicious cycles (Wells 1997). The formulation represents the first stage in creating a cognitive set for the processing of alternative explanation of the clients problems (Wells 1997). However, there is a scarcity of evidence linking quality or style of formulation and therapeutic outcome (Sanders and Wills 2005); its value lies, from a practitioner viewpoint, in representing a roadmap of how therapy might proceed and possible intervention avenues.
Structure and Format
The brief nature of CBT necessitates a highly structured and goal directed format to be adopted from the outset. Through the negotiation of agendas, goals, a problem list, summaries and completion of self-guided activities (homework) (Padesky and Greenberger 1995), CBT maintains a focused, results oriented approach. At the beginning of each therapy session, an agenda is collaboratively devised and agreed upon, each session typically including a review of the clients’ mood, feedback, client and therapist agenda items, review and agreement of homework tasks (Sanders and Wills 2005). In early therapy sessions, the generation of a problem list (a practice inherited from behaviourism) highlights the main areas in which the client wishes to see improvement and guides target areas (Sanders and Wills 2005). Goal setting, which typically occurs at the end of the assessment, allows client and therapist to clearly define and operationalise the gains the client wishes to make by the end of therapy (Sanders and Wills 2005).
A central component of CBT practice is self guided activity or homework. CBTs educational and self-help ethos seeks to empower the individual to ultimately become their own therapist, the completion of homework tasks outside session reinforce this learning experience. Homework can be enormously varied, from reading self help material to conducting behavioural experiments. Ideally, homework is devised collaboratively with a clear rationale for treatment gain. As therapy proceeds and the client becomes increasingly socialised to CBT, the client can adopt an increasingly independent role in devising homework with decreasing levels of therapist input. CBTs brief duration is greatly aided by the completion of homework outside session, enabling therapy to occur continuously in the clients own time. Successful homework completion has been associated with improved therapeutic outcomes, with clients who consistently complete homework between sessions improving more than patients who do not (Burns & Auerbach in Salkovskis, 1996). Treatment including homework produced improved outcomes than treatment consisting entirely of in-session work (Kazantzis et al. 2003)
Collaborative Therapeutic Alliance
CBTs core competencies (Roth & Pilling, 2007) affirm the importance of a collaborative therapeutic relationship within clinical practice. Competency as a CBT practitioner requires not only knowledge of cognitive models of psychopathology but also the ability to apply its theory in a structured manner. In practice, therapists must be able to formulate a useful case conceptualisation and apply empirically based clinical interventions within a collaborative therapeutic relationship (Padesky and Greenberger 1995). The collaborative element of the therapeutic relationship necessitates the adoption of a team work approach. This is underscored during therapy through the elicitation of client feedback, joint agreement of goals and frequent summaries by therapist and client to check understanding and learning. Collaboration is fostered by curiosity and inquisitiveness on the therapists’ behalf, both parties working in parallel towards achieving therapy goals.
In contrast to alternative therapeutic approaches such as Psychodynamic, the therapeutic relationship is not conceived in of itself, to be the primary vehicle of clinical change in CBT. Beck, 1976, argued that although the qualities of a good therapeutic relationship are necessary, they are not sufficient as therapies active ingredient. Once the basic elements of the therapeutic relationship are established, CBT interventions may proceed (Burns & Auerbach in Salkovskis, 1996). The importance of the therapeutic relationship within CBT has been reasserted over recent years. Demonstration that non-specific therapy factors, that is, the therapeutic relationship have an additive contribution to clinical improvement independent of CBT techniques has emphasised the importance of the alliance within CBT. These research findings illustrate that client subjective appraisal of the quality of the therapeutic relationship and therapist empathy are crucial (Squier 1990). Clients of practitioners who where rated as warmest and most empathic improved significantly more than patients of therapists with the lowest empathy ratings. Therefore, the quality of the therapeutic relationship, even in a technical form of therapy like CBT, has a significant impact on clinical improvement (Sanders and Wills 2005). As with other forms of therapy, the fundamental characteristics of a good therapeutic relationship such as accurate listening, empathy and congruence are vital for effective CBT practice (Sanders and Wills 2005).
Guided discovery, the primary learning process within CBT, teaches clients to question their own thoughts and beliefs (Padesky and Greenberger 1995). Therapists guide discovery verbally through questioning and experientially through BEs. Through this process, clients evaluate alternative perspectives or information which may be outside their immediate conscious awareness but when considered informs and impacts their existing thoughts, predictions and beliefs. Through the practice of collaborative empiricism, client and therapist adopt an experimental approach (Wells 1997), treating thoughts as hypotheses open to testing and verification (Padesky and Greenberger 1995).
CBT encompasses interventions targeting cognitive content, from NATs to schema level and cognitive processes known as metacognition. Using Socratic dialogue, the therapist guides the client to uncover new information, integrating and generalizing these new insights into their thinking. The use of synthesising questions allows the therapist and client to extrapolate from the concrete to the abstract, applying information gleaned in the current discussion to previous ideas, thus facilitating generalizability and the construction of new beliefs. Socratic dialogue has been conceptualised as a means of “loosening” NATs, priming clients to consider alternative possibilities, increasing their range of attention, memory and highlighting cognitive biases in action.
Thought records, a content focused intervention, enables the examination of evidence for and against a negative automatic thought, the identification of alternative perspectives and associated belief levels. The identification of the most salient or affect inducing thought (hot thought) focuses client and therapist on emotively charged and potentially debilitating or maintaining thoughts. Belief ratings assist in the distancing process, emphasising that thoughts are not true or false in a dichotomous sense. Moreover, thought records allow client and therapist to identify predictions which can be subjected to testing (Wells 1997). Whilst working with DAs, it may be more appropriate to emphasise the construction of new, more adaptive alternatives rather than attempting to disprove old rules using evidence for/against or examining pro’s/con’s (Padesky and Greenberger 1995). Cognitive techniques targeting the core belief and schema levels such as positive data logs and retrospective thought records enable the generation of new helpful, alternative beliefs over a prolonged time period or the re-evaluation of past experiences using present day focus.
Traditionally, CBT has tended to focus upon language content; metacognitive perspectives (thinking about thinking or the meaning of thinking) assert the importance of maladaptive cognitive processes in the development, maintenance and treatment of psychological problems. Advocates suggest that challenging of negative thoughts or beliefs targets the output or end product of dysfunctional metacognitive processes. Metacognitive awareness therefore highlights process level maintaining factors, for example, through the use of frequency counts; the client may record the number of intrusive thoughts or duration of ruminative thinking and associated impact on mood (Sanders and Wills 2005).
Imagery based cognitive techniques are used when the clients thoughts and emotions present in visual rather than verbal form or when imagery is their preferred form of expression (Sanders & Wills, 2005). For example, in social phobia, the client’s fear of social rejection and perceived lack of coping skills may come to mind as an image of themselves blushing and being publicly humiliated. Therapeutic interventions focus upon restructuring or altering the form or content of images. The client can be encouraged to re-imagine an image, altering its appearance or creating a new ending “finishing out” (Wells 1997). In PTSD, in which intrusive memories and images maintain the current threat cycle, imagery techniques are used to contextualise the trauma image, embedding it within its historical context, thereby disabling its contribution to current threat perception.
Behavioural intervention techniques such as activity scheduling and BEs have been described as “the most powerful means of cognitive change” (Wells 1997). The effectiveness of behavioural activation (BA), graded increases in activity levels and exercise has been demonstrated, particularly in depression (Westbrook et al. 2007). BA promotes improvements in mood and feelings of self-efficacy through focus upon increasing opportunities for enjoyment, mastery and pleasure. Activity schedules enable a baseline level to be established, plan future activities, increase problem solving and demonstrate to the client their current activities and achievements (Beck in Salkovkis, 1996). BA is typically the first intervention target in depression, progressive increases in activity levels and associated improvements in mood allow therapy to proceed to more cognitively focused tasks which may have been initially impossible.
BEs are experiential exercises which enable clients to test hypotheses through action, reality check their predictions and generate new perspectives thus facilitating emotionally grounded change (Bennett-Levy et al. 2004). BEs are highly adaptable in design and implementation, in which the client may be an active participant or observer, can occur independently or with therapist support either in session or in vivo. BEs are most successful, that is, produce maximum cognitive change if outcomes that support the various hypotheses are clearly operationalised, results are reviewed in detail and their impact on existing beliefs openly discussed (Bennett-Levy et al. 2004). The potency of BEs as learning experiences means that a limited number of well designed BEs can not only target NATs but also begin to undermine dysfunctional assumptions and beliefs.
Informed by the experiential learning cycle and adult learning theory, BE design is envisaged as containing four key elements: planning, experiencing, observing and reflecting. Each stage represents an opportunity to question predictions, generate new ideas and test them in everyday life. Theoretical considerations of the mechanisms of change underlying BEs invoke multilevel information processing models such as Teasdale’s ICS model (Teasdale in Salkovskis, 1996). ICS outlines two qualitatively different information processing systems, a propositional, logical, rational, verbal, affect-free processing system and an implicational, non-linguistic, rapid, experiential, emotionally salient learning system. According to the ICS model, the value of BEs is their ability to create “enactive procedures” that activate different schematic models. Essentially, BEs enable enactment of different ways of being rather than solely providing evidence which when rationally considered leads to belief change. In contrast to the cognitive techniques previously discussed, BEs are though to operate on both the propositional and implicational levels enabling a felt sense of their results. It is assumed that verbal cognitive techniques exclusively operate on the propositional level unless affect is stimulated. Research conducted with trainee practitioners has demonstrated that BEs produce significantly greater cognitive and behavioural change than thoughts records (Bennett-Levy 2003), although evidence for their specific effectiveness is sparse (Bennett-Levy et al. 2004).
CBTs oft quoted limitations have been described as “myths” by those within mainstream CBT (Westbrook et al. 2007), critiques have originated however both within and external to the discipline (Andersson and Asmundson 2008). The following discussion will examine some of the main themes of this somewhat polarized debate, contextualising views within the reality of present day evidence-based, cost/efficiency focused, stepped care service delivery. It will argue that recent developments within CBT have done much to counteract its limitations and that CBTs future course will be defined by its ability to integrate increasingly fragmented perspectives into its ever evolving framework.
According to Salkovskis, 2002, theory and empirical evaluation are central to developments within CBT; there is a continuous reciprocal relationship between science and practice. In recent years however, the necessity of cognitive intervention techniques, one of CBTs defining characteristics which differentiates it from other therapeutic approaches has been questioned (Jacobson et al. 1996). The necessity of logico-rational strategies to directly challenge and modify maladaptive thoughts has been questioned (Longmore and Worrell 2007);(Ilardi and Craighead 1994). A number of empirical anomalies within CBTs outcome literature have inspired this debate (Hayes et al. 2004). Component analyses investigating the necessary and effective elements of CBT have in some cases failed to demonstrate added value of cognitive intervention techniques. In a study of major depression, comparison of treatment conditions comprising of BA, BA with challenging of automatic thoughts and CBT resulted in equivalent performance across conditions at conclusion of therapy, 6 month follow up and in relapse rates at 2 years (Jacobson et al. 1996). The ambiguity has further intensified due to CBTs delay in researching its mechanisms of change, which according to (Burns and Spangler 2001) often fail to conform to cognitive model predictions. Additionally, measuring changes in the cognitive mediators of a disorder (thoughts and beliefs implicated by a cognitive model in disorder maintenance) do not appear to precede symptomatic improvement, thereby challenging cognitions assumed mediating role in therapeutic change. The course of this ongoing debate would appear to be directly in contrast with Salkovskis’ description of CBTs empirical focus above. The clarification of the essential and effective components of CBT practice across ranges of severity and CBT formats appears to be required (Waddington,)
This equivocation is further exacerbated by characteristics instilled by CBTs integrative development. CBTs foundations in BT and CT have resulted in the development of a diverse and at times contradictory nomenclature (Mansell 2008). Its roots in clinical observations rather than empirical research, led to a disconnect with cognitive science and neuroscience (Gaudiano 2008). Furthermore, CBT has been criticised for its inability to define itself and its lack of coherence as a theoretical framework (Mansell, 2008). Although there is general agreement on the key characteristics of CBT, outlined earlier in this paper, there is not an accepted definition of the essential elements that comprise CBT. The recent proliferation of competing theoretical frameworks such as ICS, SPAARS, S-REF & RFT has propelled CBT further from its foundations and towards increasing complexity and fragmentation.
A reductionistic, mechanistic model of distress
CBTs opponents have argued its approach is mechanistic, overly rationalistic (Greenberg and Safran 1987), fails to address the entire individual and has limited utility for people with long term or complex problems (Gaudiano, 2010). Critics cite its focus on “the here and now”, its lack of attention to developmental history, interpersonal relationships and CBTs technique driven focus governed by rigid protocols (Sanders and Wills, 2005). CBTs response has been two fold, further research and refinement of its cognitive models and intervention enhancements tackling enduring and complex problems utilising Schema Focused Therapy (SFT). For example, Beck’s initial schematic model of depression was augmented to include the concepts of modes and charges to account for findings of diathesis-stress, the relationship between cognition and personality and the phenomena of sensitization and remission (Beck in Salkovkis, 1996). This pattern can be observed in many cognitive models in which research findings have been integrated to provide a more comprehensive account, for example, the finding that self-directed attention is a critical mediator in social phobia.
Schemas are unlikely to become a primary therapeutic target in conventional, short term CBT unless necessitated by client material or pose a significant risk factor for relapse. SFT developed from studies of CT non-responders and relapsers, so called treatment failures. These individuals were characterised as having more rigid cognitive structures, a history of chronic, lifelong psychological difficulties and deeply engrained maladaptive belief systems. These schemas were formed in troubled or abusive childhoods, resulting in the child’s development of maladaptive coping or survival styles. Developed for the treatment of Borderline Personality Disorder (BPD), SFT builds upon its CBT foundations assimilating elements from attachment, psychodynamic (particularly object relations) and emotion focused perspectives, incorporating a range of therapeutic techniques from these approaches (Kellogg and Young 2006). A primary mechanism of change within SFT is limited reparenting, therapist and client dialogues enable the nurturing of the client as an abandoned child thereby challenging harsh and punitive adult relationships from early childhood. The therapeutic relationship is the antithesis of traumatic childhood relationships, an arena for clients to identify and test beliefs about relationships, practice alternative new behaviours and learn new ways of relating (Sanders and Wills, 2005). Patients are guided to generalise what they have learned in the therapeutic relationship to relationships outside of therapy.
Confrontational, adversarial and dehumanizing
Critics have asserted that CBT frames the client as a passive recipient of technical interventions (Strong et al. 2008), the CBT practitioner as controlling, medicalising, concerned with employing techniques and seeking evidence (Mansell, 2008; Sanders & Wills, 2005). The use of technical terminology such as ‘administered’ and ‘implemented’ within CBT does little to assuage these concerns. Conversely, CBT advocates maintain that it is not an assembly of techniques applied mechanistically (Salkovkis, 1996). Research has demonstrated that CBT patients rate their therapists higher on various relationship variables (interpersonal skills, accurate empathy and support) than psychodynamic therapists, their level of active listening was found to be equivalent to insight-oriented therapists (Keijsers et al. 2000).
The integration of a compassion based focus within CBT has further enriched and reinforced the importance of the therapeutic alliance. Compassion focused therapy (CFT) developed from observations that people with high levels of shame and self-criticism find it difficult being warm, compassionate and kind to themselves (Gilbert 2009), that they often use a harsh, bullying manner when attempting to change their thoughts and behaviours. People who are unable to self-sooth find it difficult being reassured or remaining calm when considering alternative thoughts or engaging in new behaviours. In CFT, the role of the therapist to help the client experience safety in their interactions, to feel safe with what is explored in therapy and to ultimately replace self-criticism with self-kindness. In compassionate mind training, the client learns the skills and attributes of compassion through modelling of the therapists compassionate abilities. CFT focuses upon the client experiencing alternative thoughts as kind, supportive and helpful. Throughout therapy, the client is taught to use warmth, compassion and gentleness as their foundation from which to move into more challenging activities.
Limited applicability to certain populations: the psychologically minded
Critical evaluations of CBT have highlighted its potential weakness and limited evidence-base for children, older adults (Kazantzis et al. 2003) and people with intellectual disabilities (ID). Critics argue whether CBT models and techniques can be applied to these disparate populations or whether CBT is best suited for the population in which it was developed. The extant literature has highlighted particular characteristics that engender suitability for CBT including, the ability to access thoughts, recognise, differentiate and label emotions, link events and emotions, understand the mediating role of cognitions and assume responsibility for change (James et al, 2001). Within ID, research is required on potential barriers to treatment that may or may not arise from capacity and motivational factors. CBT being an essentially linguistic method presents unique challenges within ID, cognitive techniques being more difficult to apply as verbal ability decreases (Willner and Hatton 2006). Furthermore, there has been a paucity of research within mainstream CBT regarding the applicability and validity of CBT models to ID (Willner and Hatton 2006).
NICE guidelines for children indicate CBT, in either group or individual format for depression (NICE, 2005) and moderate to severe ADHD (NICE, 2006). Research investigating child focused CBT interventions have been almost exclusively derived from adult treatment protocols, with insufficient attention devoted to their applicability to children (Cartwright-Hatton and Murray 2008) or developmentally appropriate intervention techniques (Doherr et al. 2005); (O’Connor and Creswell 2008). For example, it has not yet been demonstrated, that a child’s developmental stage predicts treatment outcome. The role of family in child-focused CBT is receiving increased attention. Research investigating parental cognitions as triggering or maintaining factors in childhood problems have resulted in their incorporation within child-focused models. A recent study demonstrated that socially phobic mothers encourage their infants to interact less with a friendly stranger than mothers with GAD, which is predictive of the extent to which the child subsequently shows anxiety in the presence of a stranger (de Rosnay et al. 2006). Parental anxiety has also been found to be a significant predictor of treatment failure of individual child treatment. Preliminary evidence suggests that treating one family member whether it is parent or child can have secondary effects on other family members’ difficulties.
A panacea for psychological distress
Clinical significance analysis reveals that one third to a half of clients achieve recovery following CBT (or any other form of psychological therapy) (Westbrook et al, 2007). Therefore, it is clear that CBT and other therapeutic approaches cannot be conceived as a panacea for psychological problems. For a number of disorders, specifically tailored variants of CBT are recommended, often in parallel with family centred approaches, for example, anorexia and bulimia nervosa (NICE, 2004). CBTs impressive evidence base for depression and anxiety disorder treatment has encouraged the creation of the Increasing Access to Psychological Therapies (IAPT) programme in the UK (Ghosh 2009). IAPT focuses upon rapid throughput of patients aiming to reduce waiting list times and move 50% of people treated towards recovery (Rachman and Wilson 2008). IAPT draws heavily upon the NICE depression guidelines and Layard’s work on happiness (Layard, 2006) which outlined the social and economic costs of depression and psychological problems.
Mental health research has been dominated by symptom based