Abstract
Although histories of cognitive behaviour therapy have begun to appear, their use with people with psychosis diagnoses has received relatively little attention. In this article, we elucidate the conditions of possibility for the emergence of cognitive behaviour therapy for psychosis (CBTp) in England between 1982 and 2002. We present an analysis of policy documents, research publications and books, participant observation, and interviews with a group of leading researchers and senior policy actors. Informed by Derksen and Beaulieu’s articulation of social technologies, we show how CBTp was developed and stabilised through the work of a variety of overlapping informal, academic, clinical, professional, and policy networks. The profession of clinical psychology played a key role in this development, successfully challenging the traditional ‘division of labour’ where psychologists focused on ‘neurosis’ and left ‘psychosis’ to psychiatry. Following Abbott's systems approach to professions, we identify a number of historical factors that created a jurisdictional vulnerability for psychiatry while strengthening the jurisdictional legitimacy of clinical psychology in providing psychological therapies to service users with psychosis diagnoses. The National Institute for Health and Clinical Excellence played a significant role in adjudicating jurisdictional legitimacy, and its 2002 schizophrenia guidelines, recommending the use of psychological therapies, marked a radical departure from the psychiatric consensus. Our analysis may be of wider interest in its focus on social technologies in a context of jurisdictional contestation. We discuss the implications of our study for the field of mental health and for the relationship between clinical psychology and psychiatry.
Introduction
In recent years, historians have begun to examine psychotherapies other than psychoanalysis, as evidenced by the ‘Psychotherapy in Historical Perspective’ special issue of History of the Human Sciences, which included articles on a range of approaches, including ‘third wave’ cognitive and behavioural therapies (e.g. Drage, 2018; Jansson, 2018). Indeed, there is now an emerging literature about cognitive and behavioural therapies in the UK (e.g. Marks, 2012, 2015) and in the US (e.g. Rosner, 2014, 2018; Stark, 2017), countries where critics have described their dominance in mental health policy as hegemonic (e.g. Pilgrim, 2011).
To date, the history of cognitive behaviour therapy (CBT) has focused largely on its use with people with common mental health diagnoses like anxiety and depression (Marks, 2012, 2015; Rosner, 2014, 2018). Historians and social scientists examining contemporary developments in CBT have been particularly interested in the Increasing Access to Psychological Therapies (IAPT) programme (Marks, 2015; Pickersgill, 2019; Pilgrim, 2011). Established within the National Health Service (NHS) in England in 2007, it has been one of the mechanisms by which CBT has become a dominant form of psychotherapy in England. It is also an example of the influence that the evidence-based practice movement has had on mental health policy (Pilgrim, 2011). For example, three years before the IAPT programme was initiated, the National Institute for Clinical Excellence (NICE) published treatment guidelines on anxiety and on depression recommending the use of CBT. 1 The guidelines helped to justify the programme (Marks, 2015). In turn, argues Pickersgill (2019), IAPT helped to legitimise NICE's guideline-setting role.
This article focuses on the development, in the UK, of cognitive and behavioural therapies for people with psychosis diagnoses (CBTp). 2 This is a somewhat neglected aspect of the history of CBT, one that has been only briefly discussed in previous work (e.g. Marks, 2012, 2015; Pilgrim and Rogers, 2009). Indeed, a focus on the development of CBTp also sheds light on the evolution of NICE's clinical guidelines, since both behavioural family intervention (BFI) and CBTp were recommended in its first treatment guideline. This guideline concerned schizophrenia (National Institute for Clinical Excellence, 2002), and it preceded its anxiety and depression guidelines by two years. The development of CBTp and the 2002 NICE guidelines on schizophrenia are important for a number of reasons. Firstly, international observers credit British researchers with pioneering the development of these therapies (Bellack, Buchanan, and Gold, 2001; Hollon, 2010; Wooding et al., 2010). Secondly, the history of CBT in the UK is closely intertwined with the history of the profession of clinical psychology (Marks, 2015), and, as we will see, clinical psychologists played a major role in the development of CBTp. Thirdly, the 2002 guidelines helped establish the standards for NICE's later mental health guidelines. Finally, and perhaps most importantly, the recommendation of psychological therapies was a radical departure from the psychiatric consensus. As exemplified by books like The Biological Basis of Schizophrenia (Hemmings and Hemmings, 2012[1978]), the received view within biological psychiatry was that schizophrenia was a coherent diagnostic category and an immutable disorder of primarily biological aetiology requiring pharmaceutical intervention. Just two decades before the 2002 guidelines were published, psychological therapies had been eclipsed by drug treatments and had come to be regarded as ineffective, potentially harmful, and blaming of families (Harrington, 2019). It is, therefore, surprising that cognitive and behavioural therapies for people with schizophrenia diagnoses were the first to be recommended by NICE, since psychological therapies for anxiety and depression were much less contested. Although this article focuses predominantly on CBTp, our analysis locates it historically in relation to the BFI research programme. To some extent BFI was a condition of possibility for the emergence of CBTp in the UK: three of the four research centres were involved with both traditions; CBTp studies drew on the conceptual and methodological heritage of BFI; and the success of BFI trials afforded psychological therapies for those with psychosis diagnoses enough credibility for researchers to secure funding for further developments.
Since schizophrenia has long been seen as psychiatry's ‘heartland’ and ‘paradigm condition’ (Goodwin and Geddes, 2007: 189), the development of these therapies posed a number of challenges for the professions of clinical psychology and psychiatry. Within the profession of clinical psychology, it prompted debates about medicalisation (a recurring theme in the article). However, NICE's validation of psychological therapies posed a potential challenge to biomedical dominance in psychiatry. An editorial by two psychologists in the British Journal of Psychiatry emphasised the way in which NICE's recommendations marked a departure from the received view:
How times have changed! It was not long ago that talking to people about their psychotic beliefs was deemed impossible or harmful. Yet cognitive–behavioural therapy (CBT) for psychosis is now recommended by the National Institute for Clinical Excellence … to ‘reduce psychotic symptoms, increase insight and promote medication adherence’. (Birchwood and Trower, 2006: 107)
The history of CBTp is therefore a significant lacuna in the history of psychological therapies. In this article we aim to elucidate the conditions of possibility for the emergence of these therapies in the UK between 1982 (when the first UK BFI trial was published) and 2002 (when NICE's schizophrenia guidelines were published). In tracing its emergence, we also identify a number of historical factors that generated a jurisdictional vulnerability for biological psychiatry while strengthening the jurisdictional legitimacy of clinical psychology in the provision of psychological therapies to people with psychosis diagnoses.
Throughout, we refer to the UK as a whole, since, although the four clinical psychology research centres we discuss were based in England, at least two researchers worked for a time in Wales, while professional bodies like the British Psychological Society are nationally organised. In addition, the NHS encompasses the whole of the United Kingdom, but in Wales, Northern Ireland, and Scotland, it is organised and funded by the devolved governments. We do not wish to give the impression that developments occurred only in England. Practitioners, of course, were spread across the UK, while a key conference took place in Edinburgh. NICE guidelines apply to both England and Wales, and UK government ministers select the ‘conditions’ for which NICE produces guidance. Northern Ireland essentially adopts NICE guidelines after confirming they are applicable locally, while in Scotland, the equivalent guidance is produced by the Scottish Intercollegiate Guidelines Network (SIGN). Indeed, SIGN became the first guideline-setting body in the UK to recommend CBT and BFI in its Psychosocial Interventions in the Management of Schizophrenia (Scottish Intercollegiate Guidelines Network, 1998).
Before examining these conditions of possibility and jurisdictional contestation, we first discuss theoretical influences on our analysis and our methodological approach, outline key developments in British clinical psychology in the 1970s and 1980s, and provide a brief summary of key publications between 1982 and 2002.
Theoretical and methodological approach
The history of mental health is a field of contestation where theoretical traditions, professions, and therapeutic approaches compete and where there may be significant variation between countries because of their differing socio-political contexts. In her examination of US psychiatry, Harrington (2019) has suggested that each successive theoretical movement within the discipline raised expectations with over-optimistic predictions but then failed to meet them on their own terms. Thus, she argues, biological psychiatry became dominant because psychodynamic approaches had over-promised and under-delivered. Biological psychiatry then went on to make the same mistake, failing to make the breakthrough discoveries in aetiology, diagnostic classification, and pharmacology it had promised as it re-emerged in the US in the late 1970s.
Harrington focuses primarily on the waxing and waning of theoretical traditions within the discipline of psychiatry, but, of course, psychiatry is only one of the professions operating in the mental health field. Following Abbott (1988), we view professions from a systems perspective where they compete with one another to establish the legitimacy to claim jurisdiction over particular kinds of human problems. Abbott has proposed that jurisdictional settlements are reached between professions but that these are unstable and social changes can create jurisdictional vulnerabilities for some professions while, at the same time, creating opportunities for other professions to make competing claims for legitimacy. The historical competition between those professions focused on personal problems was one of the three case studies discussed by Abbott (1988). Despite this, Scull (2011) is one of the few mental health historians to have taken up his ideas. Scull (2011) drew on the framework of jurisdictional contestation to examine the relationship between psychiatry, clinical psychology, and psychoanalysis in the US between 1940 and 2010. Scull argued that US clinical psychology had successfully competed with psychiatry and psychoanalysis by moving out of the large asylums and into the provision of outpatient psychotherapy for mild and moderate problems. Here, psychologists were freed from medical hierarchies, securing their autonomy when they gained access to third-party payments from health insurance companies. Scull saw CBT as facilitating US clinical psychology's expansion into out-patient psychotherapy. This therapy was attractive to insurance companies, he argued, because, in comparison with psychoanalysis, it was time-limited while its legitimacy was bolstered by research studies.
Scull (2011) did not examine the provision of psychological therapies for those with psychosis diagnoses, but his article highlights the importance of national differences in healthcare structures. Although, as we will see, British clinical psychologists also moved out of the large asylums in the late 20th century, they developed new services within the NHS, where the vast majority of clinical psychologists still work today (Hall, Pilgrim, and Turpin, 2015). In contrast, 41% of clinical psychologists in the US are in independent practice, with relatively few working in the public health sector (Norcross and Karpiak, 2012). In Europe, the definition and regulation of psychology and psychotherapy varies widely (Laireiter and Weise, 2019). Moreover, some psychotherapies may be allied with certain professions in some countries but not in others. Thus, in addition to theoretical traditions and professions, our account also attends to the specificities of the historical context in the UK.
A final conceptual resource on which we draw is Derksen and Beaulieu’s (2011) articulation of social technologies. Influenced by work in science and technology studies (STS), they define these as technologies developed from the social sciences that ‘consist entirely or predominantly of human action’, are situated in culturally specific ways, and ‘depend on social interaction for their constitution’ (ibid.: 705–6). Within the field of mental health, such social technologies might include psychiatric diagnostic categories, methods of assessment (e.g. diagnostic interview schedules and psychometric scales), psychological therapies and therapy manuals, and research methods. Derksen, Vikkelsø, and Beaulieu (2012: 139) call for an exploration of social technologies that pays ‘equal attention to the way so-called “technical” and “human” elements work together, and sometimes fail, to constitute particular effects’. From an STS perspective, social technologies develop through the activities of a variety of actors and formal and informal networks like those of clinical researchers, professional bodies, research funders, academic journals, advocacy organisations, the media, policymakers, and so on. Social technologies can act as social agents: they offer new ways of viewing the world and new ways of acting on it. For example, in an analysis informed by Actor-Network Theory, Manning (2000, 2002) has shown how the construct of dangerous and severe personality disorder was constituted in the UK through a complex interweaving of policy networks, legislation, and clinical research. Similarly informed by STS, Pickersgill (2013) has examined how, in the UK, psychological therapies developed in such a way that personality disorder was transformed from being seen as untreatable to being treatable, as the aims of clinicians and policymakers came to be closely aligned. Such a perspective invites new questions for historical research and scholarship. How do such social technologies become stabilised and institutionally legitimised over time? In what ways do these technologies act as social agents? In what ways do they open up new territories for their deployment and for the expansion of professions allied with those technologies?
We draw on an analysis of policy documents, CBTp research publications and books, participant observation, and interviews with a group of clinical psychologists who are leading researchers and senior policy actors. 3 We approached for interview key individuals who were actively engaged in psychosis research from the 1980s and involved in the establishment of each of the four major English clinical psychology CBTp research centres in Liverpool (Professor Richard Bentall), Birmingham (Professor Max Birchwood), London (Professor Elizabeth Kuipers), and Manchester (Professor Nick Tarrier). All those approached agreed to be interviewed. In addition, we approached a ‘second-generation’ researcher who had worked in two of the research centres: Professor Gillian Haddock began psychosis research at the start of the 1990s with Richard Bentall as one of her supervisors before moving to the Manchester centre. A sixth interview was with Professor Steve Pilling, co-director of the National Collaborating Centre for Mental Health, which, until 2016, was responsible for producing NICE guidelines. He is also director of the Centre for Outcomes Research and Effectiveness at University College London. Steve Pilling served as facilitator of the 2002 NICE Guideline Development Group (GDG). Elizabeth Kuipers chaired the GDG for the 2009 edition of the guidelines. She also went on to chair the GDG which produced the 2014 edition of the guidelines, and on which Max Birchwood also sat. Interviews were conducted by the second author between September and November 2013. Given the difficulty in anonymising transcripts where participants are discussing their own work, the interviewees kindly gave written consent to be interviewed ‘on the record’. The interview study was approved by the university's ethics committee.
We will first outline the national context and then summarise BFI and CBTp research between 1982 and 2002. In the remainder of the article, we seek to contribute to the emerging scholarship on cognitive and behavioural therapies, illustrating the value of examining their development attentive to the context of healthcare systems, debates about professional jurisdiction, and how the wider policy environment shapes responses to particular kinds of mental health problem. Our study shows how jurisdictional vulnerabilities can occur in relation to psychiatry's foundational categories, not just those at its borders. In documenting how a group of cognitive behavioural clinical psychologists established jurisdictional legitimacy by challenging the traditional ‘division of labour’ between psychologists and psychiatrists (Goldie, 1977; Pilgrim and Treacher, 1992), it is potentially of wider interest in two respects. Firstly, it identifies some of the processes that enable one professional group to expand into a therapeutic terrain previously dominated by another profession. Secondly, it shows how jurisdictional legitimacy can be adjudicated by guideline-setting bodies like NICE.
British clinical psychology, cognitive and behavioural therapies, and the changing fortunes of psychological therapies for schizophrenia
British clinical psychology's development was largely coterminous with that of the NHS, founded in 1948. In the 1950s and 1960s, psychologists provided psychometric assistance to aid psychiatric diagnosis. Subsequently they began to offer therapeutic interventions, initially behaviour therapy and, then later, introducing cognitive therapy, resulting in what has come to be known as CBT (Marks, 2015). According to Pilgrim and Treacher (1992), it was a pragmatic, eclectic, technocratic, and meritocratic profession, which had ‘imbibed’ the values of the welfare State. Today, the NHS funds UK clinical psychology training programmes and, as noted earlier, employs most British clinical psychologists (Hall, Pilgrim, and Turpin, 2015).
Though they were initially able to receive clinical referrals only via psychiatrists, the Trethowan report (Department of Health and Social Security, 1977) granted psychologists the professional autonomy they had long been seeking (Goldie, 1977; Pilgrim and Treacher, 1992). Once they were able to receive referrals directly from general practitioners, the demand for psychologists began to increase. This, together with other structural and policy changes in the NHS, led to a rapid expansion of the workforce. Figures from Turner et al. (2015) show that the number of NHS clinical psychologists in 2000 (5316) was nearly 30 times what it had been in 1960 (179), whereas the number of consultant psychiatrists increased only fourfold over the same period (679 in 1960 and 3057 in 2000).
In the 1970s, the jurisdictional settlement between psychiatry and clinical psychology was formulated by Goldie (1977) as a ‘division of labour’, where psychologists focused on ‘neurosis’ and left ‘psychosis’ to psychiatry (see also Pilgrim and Treacher, 1992). An implicit assumption here was that neurosis was understandable using normal psychological principles and thus amenable to psychological therapies. In contrast, psychosis was not, since symptoms were seen, following Jaspers (1997[1959]), as inherently un-understandable. This division of labour was reinforced by two developments. Firstly, by the mid 1980s, the efficacy of psychodynamic psychotherapy for schizophrenia, pioneered in the US in the post-War period, had been dealt a serious blow. Three major studies reported poor outcomes, including high rates both of suicide and of patients dropping out of treatment (Gunderson et al., 1984; McGlashan, 1984; Stone, 1986). Drake and Sederer (1986: 313) concluded that ‘patients with chronic schizophrenia’ were ‘highly vulnerable to negative treatment effects’ and that intensive treatments like individual psychotherapy were ‘overstimulating and intrusive’. At this point, cognitive and behavioural therapies were not in a position to pose a challenge. Indeed, behaviour therapy research into schizophrenia declined between 1963 and 1988 (Scotti, McMorrow, and Trawitzki, 1993), and the diagnostic category was referred to as ‘behavior therapy's forgotten child’ by Bellack (1986: 199). Pioneer of cognitive therapy, A. T. Beck, was initially pessimistic about its application to psychosis (Beck, 1976).
Secondly, as de-institutionalisation policies proceeded, British psychologists expanded beyond psychiatric hospitals, although a small group of largely behaviourally oriented psychologists continued to work in the field of psychiatric rehabilitation (e.g. Watts and Bennett 1983). As psychologists expanded into primary care and other community settings, they provided a broader range of therapies (Pilgrim and Treacher, 1992) for a broader range of problems (Bowden et al., 2015). This expanding therapeutic repertoire included cognitive and behavioural therapies (e.g. Hawton et al., 1989; Marks, 2015), which drew on well-established conceptual traditions in the wider discipline of psychology. American psychiatrist A. T. Beck's attempts to persuade his psychiatric colleagues of the merits of his cognitive therapy met with little success (Rosner, 2018), but British and American clinical psychologists took these ideas up much more readily. Beck's cognitive therapy was widely disseminated in the UK, and he made extended visits to the country in the 1970s and 1980s (Hollon, 2010). His 400-page Cognitive Therapy of Depression (Beck et al., 1979) provided a detailed guide on how to use the approach (Rosner, 2018). 4 Its epistemological assumptions meant that it lent itself to the production of manualised therapy protocols thus enabling comparative outcome studies to be conducted. Research trials began to demonstrate its efficacy (Hollon, 2010; Rosner, 2018).
In the next section, we give a brief overview of the research conducted between 1982 and 2002 – summarised in Table 1 – before identifying the conditions of possibility for the emergence of CBTp and examining aspects that strengthened the jurisdictional legitimacy of clinical psychology.
Selected timeline of key studies and books in the development of behavioural family intervention (BFI) and cognitive behavioural therapy (CBT).
An overview of research into cognitive and behavioural therapies for psychosis, 1982–2002
Behavioural family intervention
After the linking of schizophrenia relapse rates with family interaction by Brown, Birley, and Wing (1972) at the Institute of Psychiatry's Social Psychiatry Unit in London, Vaughn and Leff (1976) reported that relapse was associated with high levels of ‘expressed emotion’ in families (EE; criticism, hostility, and ‘emotional over-involvement’). Building on this, a team including Elizabeth Kuipers and led by psychiatrist Julian Leff developed a psychosocial intervention comprising a number of elements, including psychoeducation for relatives and family therapy sessions, which appeared to reduce EE (Leff et al., 1982). The London group's work was followed by trials of similar psychoeducational behavioural interventions with families by Max Birchwood and a psychologist colleague in Birmingham (Smith and Birchwood, 1987) and a multidisciplinary group led by Nick Tarrier in Manchester (Tarrier et al., 1988). In the 1990s, the London and Manchester groups both published guides to their respective approaches (Barraclough and Tarrier, 1997; Kuipers, Leff, and Lam, 1992).
Cognitive behaviour therapy for psychosis
Case studies using a number of different behavioural and cognitive techniques with individuals rather than families were published by British clinical psychologists in the late 1980s and early 1990s in Leeds (Fowler and Morley, 1989), in Wales (Chadwick and Lowe, 1990), and in Liverpool by Gillian Haddock and Richard Bentall with clinical psychologist Peter Slade (Haddock, Bentall, and Slade, 1993). In addition, two British psychiatrists began to apply Beck's cognitive therapy (e.g. Beck et al., 1979) with people with psychosis diagnoses in a case series (Kingdon and Turkington, 1991b). These collections of techniques began to be stabilised as ‘cognitive behaviour therapy’ in the early 1990s.
Kingdon and Turkington (1991a, 1991b) undertook some small-scale studies, and in 1994, they published the first CBTp textbook (Kingdon and Turkington, 1994). This was followed by guides published by psychologists in London (Fowler, Garety, and Kuipers, 1995) and in Birmingham (Chadwick, Birchwood, and Trower, 1996), while Liverpool psychologists published an edited collection (Haddock and Slade, 1996). Between 1996 and 1998, three randomised controlled trials (RCTs) of CBTp were conducted, in Birmingham (Drury et al., 1996), London and East Anglia (Kuipers et al., 1997), and Manchester (Tarrier et al., 1998). The latter study was conducted by psychologists alone, whereas the first two were co-authored with psychiatrists. Turkington and Kingdon (2000) followed with their own small RCT and an international collaboration with American colleagues (Sensky et al., 2000).
Conditions of possibility for the emergence of CBTp and the establishment of clinical psychology's jurisdictional legitimacy
A conservative and cautious approach
Given the scepticism following the failed US psychodynamic therapy trials of the mid 1980s, the small group of clinical psychologists evaluating psychological therapies initially adopted a conservative approach. Kuipers said that, at this time there was ‘pessimism’ and ‘nobody was interested in doing any treatments’ because ‘people were very scared that you’d make people worse'. As a result, they sought to establish these therapies as feasible, safe, and efficacious. The psychologists largely worked within a broadly psychiatric conceptual framework, using diagnostic categories and medical terminology (e.g. ‘symptom’, ‘treatment’, ‘relapse’, etc.) when reporting trial results. Moreover, since participants in the psychological therapy arms of comparative outcome trials were generally taking medication, these therapies could be seen as an adjunct to – rather than an alternative to – drug treatment.
Harrington (2019) notes how some of the psychodynamic and family therapies in the 1950s and 1960s had appeared to blame families. In contrast, commenting on her BFI work with families, Kuipers said that ‘the thing that seemed to work most was … to be very, very positive with them and try and build on their strengths … not worrying about cause. Lots of stuff will tell you … you caused it, which is not true’.
Such pragmatic strategies enabled these researchers to avoid being drawn into heated debates about causality that might have led to these therapies being dismissed by psychiatrists or relatives. However, critics like Johnstone (1993) argued that BFI was a medical approach because it viewed schizophrenia as an illness with a biological basis requiring medication and that, while family communication was seen as having a role in relapse it was not seen as an aetiological factor.
In addition to the adoption of a conservative and cautious approach, there were other aspects fostering the legitimacy of CBT, which we will discuss in turn.
The alignment of clinical psychology's professional identity with important cultural values
Abbott (1988) argues that cultural values are a major source of professional legitimacy. For example, ‘academic knowledge legitimizes professional work by clarifying its foundations and tracing them to major cultural values’ (ibid.: 54). These include ‘rationality, logic, and science’ (ibid.), as well as what he refers to as individual, social, political, and economic values. However, clinical psychology's professional identity was aligned not only with science but also with emerging values like humanistic approaches to psychiatric care. Behaviour therapy had exemplified modernist assumptions of technicism, rationality, amorality, and humanism (Woolfolk and Richardson, 2018), while Beck's cognitive therapy championed empiricism, objectivity, and experimentalism (Rosner, 2018). These values are enshrined in the ‘scientist-practitioner’ model of clinical psychology practice, which balances a dual commitment to the scientific method and to psychotherapeutic practice (Pilgrim and Treacher, 1992). The foreword to Haddock and Slade’s (1996) Cognitive-Behavioural Interventions With Psychotic Disorders identified four principles underlying this work. The third and fourth principles emphasised scientific values: the importance of research in developing and evaluating new treatments. Similarly, Tarrier described those involved in early work as ‘people who were interested … in the clinical work, but were also willing to explore and evaluate it from a research perspective’.
However, Abbott (1988) also notes that shifts in cultural values may create both jurisdictional vulnerabilities and competing claims. Such a shift occurred in the 1970s, signalled by the UK mental health charity MIND's increasing criticism of psychiatric care and its advocacy for patients' rights (Toms, 2020). In the 1980s, consumerism began to play a role in mental health policy (Pilgrim and Ramon, 2009). Rogers and Pilgrim’s (1993) survey of 516 British mental health service users illustrates the influence of these changes. Although they do not report their participants' diagnoses, it is clear that many are likely to have been given a psychosis diagnosis: 70% had received antipsychotic medication and 65% had been admitted to hospital for more than three months at a time, and over half had been admitted more than four times. Sixty percent of the sample had received counselling or psychotherapy, and it was rated as helpful by more people than drug treatment and ECT (received by 48% of the sample). One survey respondent noted how they had valued their therapist's ‘understanding, validation and acceptance of experience from my perspective’, while another had benefitted from ‘an explanation that I could understand of psychotic episodes’ (ibid.: 624). Haddock and Slade’s (1996: xi–xii) first and second principles – ‘listening and attempting to make sense of’ and ‘acknowledging the meaning and role of’ psychotic symptoms – implied that CBTp was aligned with these emerging values.
Rogers and Pilgrim (1993) point out that implicit in their respondents’ positive comments about psychotherapy were ‘unfavourable messages about the value of traditional psychiatry’ (ibid.: 624). This was another way in which early researchers were aligned, since a similar contrast was found throughout the interviews. Bentall reported that many patients had said to him ‘nobody's ever asked me about my life’. Their clinical experience as practitioners led them to question many traditional psychiatric assumptions. As Haddock put it:
Working with people made me realise … that asking people about their experiences was exactly what they wanted. They wanted to talk about the fact that they'd started to hear these really horrible voices or they wanted to talk about how they felt when they felt really paranoid and it was a sort of a bit of a revelation in a way that you know, although the … taught stuff from a lot of … health professionals' … training said you know, this is the way you treat it, you treat it with drugs and you look after people and you might be able to do some behavioural management and token economy stuff but that's the limit of what you can do. And I think what was happening in the sort of eighties and nineties was starting to demonstrate actually there is a heck of a lot more that can be done and maybe some of those things that you know are traditionally accepted are actually not that valid really.
Criticisms of standard psychiatric care by service users and advocacy organisations thus generated jurisdictional vulnerability for the psychiatric profession, while the demand for humanistic care and access to talking treatments helped strengthen the cultural legitimacy of clinical psychology and psychological therapies. Of course, clinical psychologists were not the only professionals listening to service users and offering psychotherapies, but, we would argue, it was the profession's blending of humanistic and scientific values – summarised in Richards’ (1983: 179) description of the profession's ‘scientific humanism’ – that helped to strengthen its jurisdictional legitimacy. 5
Developing experimental and translational research programmes
What became known as CBTp was co-produced through an assemblage of researchers, funding bodies, journals, research centres, and both experimental and translational research adapting therapy protocols. Kuipers observed:
Psychology was a bit of a cohort effect, a rising profession which was showing success, in … treatments for anxiety and depression, where medications had not been so successful. It was a new profession, so people didn't feel particularly boundaried, I think, that they couldn't do things.
Thus, the successful use of CBT by psychologists as they had moved from hospitals to the community, facilitated its use with people with psychosis diagnoses. The psychologists working in this area shared ideas with each other through a range of informal networks of friendships and more formal academic networks: reading publications; attending conferences; and giving or inviting others to give talks, including people from other research centres.
Bentall described the early history of CBTp as representing ‘a kind of quirky idea to dabbling stage, followed by a stage where people started to do … small scale RCTs’. Funding for pilot studies came, according to Birchwood, from ‘local schemes of one sort or another’. By the early to mid 1990s, the clinical psychology research centres in Birmingham, Liverpool, London, and Manchester had become established, producing a series of publications (see Table 1) and attracting postgraduate researchers. Once RCTs demonstrating the efficacy of BFI had been published, these centres began to focus on how they might use cognitive therapy with individuals rather than families, partly for pragmatic reasons. According to Tarrier, therapists ‘found it difficult to get families involved’, but it was ‘a lot easier practically to get a patient involved’ in individual therapy. Moreover, for those, like Kuipers, working in inner cities, ‘perhaps 20% of people have families, so … if there isn't [a] family you need to be doing something’.
Psychological therapy researchers adopted two key research strategies: innovation via single-case methods; and translation, adapting CBT protocols for anxiety and depression. Single-case research designs involved evaluating the impact of an intervention by measuring behavioural changes (Shapiro, 1966), and this was an important experimental paradigm within behaviour therapy (Derksen, 2000). Such cases acted as laboratories of experimentation, providing an ideal opportunity for practitioners and postgraduate students to innovate, since they required little, if any, research funding. At this point, the notion of CBT as a coherent entity was not yet stabilised and an eclectic range of theories of hallucinations and delusions were investigated. Earlier case studies had failed to develop traction, but in the late 1980s and early 1990s, the cadence increased as a second generation of researchers published studies completed during their clinical training or research PhDs (e.g. Chadwick and Lowe, 1990; Fowler and Morley, 1989; Haddock, Bentall, and Slade, 1993), with several reporting positive results. These studies ‘created a climate of, well this might just work’ (Bentall). Working at the time in North Wales, psychologist Paul Chadwick used cognitive techniques to modify delusional beliefs, drawing on Beck's ideas (e.g. Hole, Rush, and Beck, 1979). For example, the therapist and client could collaboratively design ways of empirically testing the beliefs (Chadwick and Lowe, 1990).
However, it was two British psychiatrists – David Kingdon and Douglas Turkington – who attempted a more wholesale translational approach, applying the protocol for Beck's cognitive therapy for depression (Wooding et al., 2010). As Haddock observed:
[They] had been really influenced by Tim Beck and they were sort of looking at transferring just what had happened in depression into psychosis and that was where they were coming from, so a much more cognitive therapy model than I think where some of the other people were coming from.
Haddock described how CBTp became increasingly stabilised and instantiated through books published by the research centres: ‘I think that all coincided with, you know, becoming more popular, there were all the conferences, the books were available so there were sort of like treatment protocols available’. Such dissemination created momentum: ‘it was a very small group of us … we started publishing and then suddenly people got inspired by it … and then we just had students and people who came to visit us’ (Birchwood).
Research funding bodies like local NHS funders, the Department of Health, the Medical Research Council, and the Wellcome Trust were significant actors in the network. The epistemological assumptions of CBTp enabled its evaluation via RCTs though some of these researchers had misgivings about these designs (Birchwood and Trower, 2006) because it required the adoption of the ‘drug metaphor’. This assumes that a therapy's efficacy can be isolated from the role of the therapist and the therapeutic relationship (Stiles and Shapiro, 1989). Wampold and Imel (2015) refer to it as the medical model approach to psychotherapy research, and RCTs are associated with biomedicine (Pickersgill, 2019). For Birchwood, however, it was pragmatically necessary at the time to use RCTs ‘because there was such a lot of scepticism, to win your spurs it had to stand up to the kind of credibility of the drugs.… It tended to be a big mistake in retrospect, but to get the dosh you had to do it’.
By the end of the 1990s, a heterogeneous collection of ideas had stabilised as ‘CBTp’, the research centres had had a collectivising effect on the field, and research funders were supporting trials published in esteemed psychiatric journals. The cultural legitimacy of CBTp was increasing, and it began to attract international validation. In 1998, Beck asked British CBTp researchers to a conference in Philadelphia to meet North American colleagues as he had ‘become very interested in this work before we started publishing the trials’ (Kuipers), and this group has met annually since. Bentall described the meeting as a ‘critical event’, and Kuipers saw it as having a collectivising effect: ‘We all knew each other when Tim Beck asked us to go to America, but we hadn't quite met like that, so there was a bit more cohesion about it after that, which was quite helpful’. Following the meeting, Beck ‘became a prominent “champion” of CBT for schizophrenia in the US’ (Wooding et al., 2010: 107), which would facilitate the acceptance of these therapies internationally in the years after the 2002 NICE guideline.
Rendering the un-understandable understandable: Re-specifying psychosis in psychological terms
Using theories from the broader academic discipline of psychology, the psychologists engaged in this work attempted to systematically render intelligible what Jaspers (1997[1959]) had regarded as un-understandable symptoms. As Bentall puts it, they ‘were kind of ambitious about applying psychological ideas to phenomena which were thought to be un-psychological’. Social technologies draw from the social, rather than the life sciences (Derksen, Vikkelsø, and Beaulieu, 2012), and British clinical psychologists had an intellectual – as well as a professional – independence from psychiatry, since they could draw on their parent academic discipline's research methods as well as theories from a range of subfields (e.g. perception, developmental psychology, neuropsychology, cognitive psychology, social psychology, personality psychology, etc.).
Academic knowledge is a powerful vector for establishing the cultural legitimacy of a jurisdictional claim as it can ‘make connections … that may reveal underlying regularities that can ultimately reshape practical knowledge altogether’ (Abbott, 1988: 55). Research in experimental psychopathology enabled the development of cognitive models of delusions and hallucinations. For example, one line of research, inspired by survey evidence of hallucinations in non-clinical populations (Posey and Losch, 1983), examined whether environmental contexts of increased stress, perceptual ambiguity, and sensory deprivation could lead to a failure to discriminate the real from the imagined (Slade and Bentall, 1988). Chadwick and Birchwood (1994) argued for the importance of modifying voice hearers' beliefs about their voices as a way of reducing their distress leading them, according to Birchwood (personal communication), to question the notion of CBTp as a ‘quasi-neuroleptic’ (e.g. Birchwood and Trower, 2006). Garety and Freeman (1999) reviewed psychological theories of delusions, which conceptualised them as, variously, attempted explanations of anomalous experiences, ‘theory of mind’ deficits, probabilistic reasoning biases, or related to a defensive attributional style and the management of self-esteem.
Previously, within psychiatry, the meaning of psychosis had been sought in the traditions of phenomenology or psychoanalysis, but a new cognitive conceptual framework was developing, one that – apart from defensive attribution – owed little to these previous traditions. An important aspect of these theories was that they examined precipitating factors in the present (rather than the past) and within cognitive systems (rather than social relationships). Some psychologists argued that this avoided both the social context and the biographical meaning of experiences (e.g. Boyle, 1990, 2002; Johnstone, 1993), but it also meant that these approaches avoided the kinds of criticism that had been levelled at the psychodynamic and family-based therapies of the 1950s and 1960s (Harrington, 2019).
Whereas psychiatry draws on medical knowledge derived from clinical populations, psychology's disciplinary focus encompasses the ‘normal’ population. In addition, constructs could be developed and tested through the use of psychometric theory and statistical analysis. These two factors enabled Eysenck’s (1950) conceptualisation and measurement of personality in terms of dimensions rather than typological categories. His use of factor analysis led him to propose that ‘normality, neurosis and psychosis are not qualitatively different, but are positioned in a three-dimensional space charted by the dimensional theory’ (Derksen, 2000: 9). Thus, although it was a psychiatrist who proposed that delusions and hallucinations could be viewed as varying along a continuum from normality to pathology (Strauss, 1969), psychologists proved more able than psychiatrists to exploit such ideas in the 1980s. Psychologists were much more familiar with the concepts and methods for charting this three-dimensional space than psychiatrists, and they were more interested in understanding the movement from ‘normality’ to ‘psychosis’. Psychologists developed psychometric measures of delusions, defining them in psychological terms and measuring them along dimensions (e.g. Brett-Jones, Garety, and Hemsley, 1987). In contrast, the discipline of psychiatry has long resisted abandoning categories for dimensions. For example, the dimensional approach to diagnosis advocated by the drafters of DSM-5 had to be abandoned as they had ‘not been widely accepted’ (American Psychiatric Association, 2013: xliii). Psychometric theory also provided a basis to critique the reliability and validity of the heterogeneous diagnostic category of schizophrenia and to propose, instead, a focus on individual symptoms like delusions and hallucinations (Bentall, Jackson, and Pilgrim, 1988).
However, there were different perspectives within clinical psychology about key issues like medicalisation. 6 Both Richard Bentall and Mary Boyle criticised the diagnostic construct of schizophrenia and reductionist versions of biological psychiatry (Bentall, Jackson, and Pilgrim, 1988; Bentall, 1990; Boyle, 1990). While the Birmingham group accepted the validity of these criticisms (Chadwick, Birchwood, and Trower, 1996), the London group viewed them as overstated (Fowler, Garety, and Kuipers, 1995). However, even though the London group proposed a greater aetiological role for biopsychosocial vulnerabilities, their models (e.g. Garety et al., 2001) placed much more of an emphasis on psychological than on purported biological aetiological mechanisms. The finding of common ground among clinical psychology's different traditions would become increasingly important as the profession began to adopt a more assertive uni-disciplinary view of psychosis.
Individual and institutional collaboration and competition: The complex and dynamic relationship between psychiatry and clinical psychology
As Pilgrim and Rogers (2009: 958) have noted, contemporary mental health professions are engaged in ‘a multiplicity of alliances and disputes … organised around roles … and ideologies’, and there were both collaborative and autonomous ventures between 1982 and 2002. There is a strong tradition of collaboration between the two professions and, as noted earlier, a number of trials were conducted jointly by psychologists and psychiatrists. Moreover, other professions, like mental health nurses, were trained in BFI (Gournay, 2000). Multidisciplinary organisations, focused on particular therapeutic approaches, provided a forum for both collaborative work and dissemination to practitioners via talks, journals, and conferences. For example, the Association of Family Therapy was an important network for BFI, while several interviewees saw the British Association of Behavioural Psychotherapy (BABP) as playing a more important role at this time than the British Psychological Society (BPS). 7 Birchwood described the 1988 Third World Congress of Behaviour Therapy in Edinburgh, organised by the BABP, as a ‘watershed’, as ‘there was quite a bit of excitement there about schizophrenia’. Discussions of these ideas took place in other contexts too and were becoming increasingly popular.
Clinical psychologists also received more indirect assistance from those psychiatrists sitting on funding bodies who shared their psychosocial perspective. Due to the preponderance of psychiatrists on these committees, they were ‘very psychiatric bodies’ (Birchwood), but there was often a ‘toing and froing between those who had a more social view of the world and those who had more a biological view of the world’ (Tarrier). According to Tarrier, since mental health research bids were often competing with those from other medical specialties, ‘the decision the committee makes then will depend on the committee makeup’, and so psychological therapy bids ‘did need that lobby about people who are influential on these committees who thought it was a good thing’.
Psychologists and psychiatrists also collaborated on various reviews of mental health policy. This process was facilitated by the Department of Health's (DoH) employment of key members of each profession as policy advisors. For example, the 1999 National Service Framework on adult mental health (NSF; Department of Health, 1999) included members of each profession with a shared commitment to a more psychosocial perspective. Four psychologists (including Elizabeth Kuipers and also the DoH policy advisor, psychologist Glenys Parry) were involved in NSF working groups. Of the more numerous group of psychiatrists involved, two were also CBTp researchers, one of whom was David Kingdon (employed as a policy advisor by the DoH in the 1990s).
In addition, there were three explicit institutional collaborative ventures between the BPS and the Royal College of Psychiatrists (RCPsych) spurred by a shared interest in shaping the way in which evidence-based practice would be translated into healthcare policy. Firstly, in 1998 the two organisations set up a new jointly owned journal, Evidence-Based Mental Health, through BMJ Publishing (https://ebmh.bmj.com/pages/about/). This helped to stabilise and institutionalise the notion of evidence-based mental health research. Secondly, Pilling noted that a joint BPS and RCPsych working party was set up ‘at the request of the Department of Health’ to develop guidelines for schizophrenia on the basis of systematic reviews of trials. Policymakers were apparently interested in the efficacy and safety both of psychological interventions and of the atypical antipsychotic medications introduced in the 1990s. According to Pilling, both sets of reviews formed the basis for the 2002 NICE schizophrenia guidelines. Lastly, the two organisations made a successful bid to become one of the six guideline development units to be commissioned by NICE: the National Collaborating Centre for Mental Health (NCCMH) was established in 2001 under the co-directorship of Pilling (a psychologist) and Tim Kendall (a psychiatrist). 8
However, there is an equally strong tradition of a more ambivalent relationship between psychiatry and clinical psychology. This could range from more explicit competition and rivalry to uni-disciplinary work focused on developing specifically psychological models – some of the studies discussed in Table 1, for example, were conducted only by psychologists. Moreover, far more psychologists than psychiatrists practised cognitive behaviour therapy. Goldie’s (1977) analysis of interviews with British psychiatrists, clinical psychologists, and social workers identified strategies of compliance, eclecticism, and radical opposition in relation to traditional psychiatry. Although many psychologists sought to expand their role, Goldie observed that they largely avoided direct confrontation with psychiatrists lest this led to them losing professional autonomy, and so they worked within what he described as ‘realistic limits’ (ibid.: 154). He reported indications of disciplinary rivalry with one psychologist commenting that, apart from prescribing medication, much of the rest of the work of psychiatry ‘could be done by other members of staff, just as well, if not better’ (ibid.: 148). Interviewees in the present study varied in their views though many were critical of the routine psychiatric care of the time. Some couched disciplinary rivalry in humorous terms. For example, Tarrier described talking with a colleague about a BFI study by US psychiatrists ‘and we thought “Oh sod it, we could do this, and we could do it and probably better”. Arrogance of youth!’.
There were uni-disciplinary as well as inter-disciplinary interventions in the policy sphere. In the mid 1990s, Glenys Parry at the DoH, commissioned a review of the evidence on the effectiveness of psychological therapies by psychologists Tony Roth and Peter Fonagy (Parry, 2015). This review – subsequently published as Roth and Fonagy (1996) – included a chapter on schizophrenia that identified the potential utility of psychological therapies including CBTp and BFI. Another discipline-specific venture came, in 2000, from psychologists' professional body, the BPS, when it published Recent Advances in Understanding Mental Illness and Psychotic Experiences (British Psychological Society, 2000). The BPS report appeared to lay the foundations for a comprehensive re-specification of psychosis and schizophrenia in psychological terms. It focused on aetiology, explanatory models, and a range of forms of help and treatment, including CBT. It was a consensus statement, including not only CBTp researchers but also a psychodynamic psychologist and psychologists with more critical perspectives (e.g. Mary Boyle, Lucy Johnstone, and David Pilgrim). The report identified both areas of consensus between these different traditions and areas of ongoing debate. Although the early development of CBTp between 1982 and 2002 was shaped largely in academic, professional, and policy networks with little direct collaboration with the psychiatric survivor movement, this report represented a more active engagement. It quoted from service users (including psychologists who had received psychosis diagnoses) noting that many did not find a ‘medical model’ approach helpful. The report drew on not only a discourse of de-medicalisation but also a discourse promoting the profession of clinical psychology, implying, perhaps, that its approach was more aligned with service users’ preferences. The report could be seen as an indication that the profession was attempting to move beyond Goldie’s (1977) ‘realistic limits’ and ‘division of labour’, grounding these claims in a discourse of ‘scientific humanism’ (Richards, 1983). It was also a sign, perhaps, that the profession was constructing a new professional identity, one more clearly differentiated from psychiatry. This nascent identity may have been nurtured by two sets of policy developments in the 1990s, which had further increased psychiatry's jurisdictional vulnerability.
The increasing alignment of policymakers' concerns with those of clinical researchers in the 1990s
The 1990s saw schizophrenia becoming a major policy concern in the wake of negative media coverage that directly questioned the quality of psychiatric care, deepening psychiatry's jurisdictional vulnerability. Both Manning (2000, 2002) and Pickersgill (2013) observe that policymaking does not occur in a vacuum and is affected not only by research but also by the media and advocacy groups. Meurk et al. (2015) argue that media reporting can influence the perceived importance of an issue and the kind of policy solutions seen as appropriate, particularly when there are ‘focusing’ events that open ‘policy windows’. In the 1970s and 1980s, British media discussion of de-institutionalisation policy had often focused on scandals involving the neglect of psychiatric patients, but it changed in character in the 1990s following a series of homicides committed by people with a diagnosis of schizophrenia (Rose, 1998). The killing of Jonathan Zito in 1992 by Christopher Clunis led to a public inquiry, and this was a key focusing event. As the health editor of the Independent newspaper concludes, ‘After the Zito killing, the nature of the debate about mental illness changed. The focus shifted from the care of the patients to the protection of the public’ (Laurance, 2003: xiii). A similar conclusion had been reached by studies of the media (Hallam, 2002; Rose 1998). The DoH instituted a policy requiring an official inquiry after every homicide associated with mental health services and the number of inquiries – and thus, media reports – rapidly increased, the cumulative effect of which was to ‘ratchet up concern’ (Laurance, 2003: 40). Cummins (2012) suggests that UK media reporting about community care in the 1990s shared many of the features of Cohen’s (1972) notion of a ‘moral panic’ and he argues that the framing of media reports can be affected by advocacy organisations. Groups like Schizophrenia: A National Emergency (SANE) and the Zito Trust – co-founded by Jonathan Zito's wife, Jayne – raised concerns in the media about the risk posed to the public by psychiatric patients because of poor quality community care. The policy response to this was ‘an increased managerialist culture with a focus on audit and risk’ (Cummins, 2012: 325). In addition to being a subject of negative media reporting, schizophrenia was also a significant cost to the NHS, with treatment estimated to be £2.6 billion per year with £652.2 million (over 5% of the total NHS budget) of that spent on in-patient care alone (Knapp, 1997). Health economist Martin Knapp recommended investment in ‘cost-effective community care arrangements, anti-psychotic drugs and psychological interventions’ (ibid.: 509).
In the 1990s, then, policymakers' concerns about the quality of standard psychiatric care aligned with the aim of clinical psychology researchers to improve the quality of care. However, the social technology of CBTp was also aligned with the goal of risk management, which had become a key aspect of the work of mental health professionals (Rose, 1996). Psychological research suggested that it was possible to identify the early warning signs of relapse (e.g. Birchwood et al., 1989), and this helped to open up a new terrain – the possibility that, using medication or CBTp, one could intervene at an early stage to prevent or to mitigate relapse (Birchwood, Spencer, and McGovern, 2000). Psychological therapies could, perhaps, have economic as well as scientific and humanistic value. But, at the same time as schizophrenia was increasingly becoming an object of concern for policymakers, the way in which treatments were to be evaluated was changing.
Capitalising on the growing influence of the evidence-based medicine movement
During the 1990s, healthcare decision-making became increasingly governed by the principles of evidence-based medicine (e.g. Eddy, 1990; Sackett et al., 1996) instantiated by the establishment of the Cochrane collaboration in the NHS (Chalmers, Dickersin, and Chalmers, 1992). Cognitive behavioural researchers could capitalise on these policy developments, since these values were consistent with their professional identity as scientist-practitioners. Since trials of BFI and CBTp had already been completed by the time NICE was established in 1999, these therapies had a head start on other approaches. As Woolfolk and Richardson (2018: 138) note, ‘At the dawn of the new era of evidence-based medicine, CBT arrived first, and was armed with most data’. The NSF (Department of Health, 1999) explicitly referenced Roth and Fonagy’s (1996) review and incorporated the Cochrane collaboration's hierarchy of evidence, which placed RCTs at the top. This validated the RCTs by Drury et al. (1996), Kuipers et al. (1997), and Tarrier et al. (1998), which the NSF explicitly referenced. It concluded that ‘for schizophrenia there is growing evidence of effectiveness for psychological therapies, including some cognitive approaches and anxiety management techniques. Psychological therapies with the families of those with schizophrenia, combined with medication, can prevent relapse, and reduce admission to hospital’ (Department of Health, 1999: 46). In 1998, Scotland's guideline-setting body produced Psychosocial Interventions in the Management of Schizophrenia (Scottish Intercollegiate Guidelines Network, 1998), recommending the use of CBT and BFI.
However, the treatment ecology was about to be significantly disturbed as the results of the previously mentioned BPS/RCPsych project on schizophrenia were published in a series of articles. Firstly, advocates of atypical antipsychotic medication were dealt a severe blow when the medication review concluded that there was ‘no clear evidence that atypical antipsychotics are more effective or are better tolerated than conventional antipsychotics’ (Geddes et al., 2000: 1371). Secondly, lingering concerns about psychological therapies were allayed. Pilling noted that, at the time the review was conducted ‘people were reasonably confident about the effectiveness of family interventions, but a little bit uncertain about some of the issues along with it, like for example, that question about harm’. However, when the review of BFI and CBTp was published (Pilling et al., 2002), the section on family interventions concluded that ‘there were no differences in suicide rates between family interventions and other treatments’ (ibid.: 774).
Evidence-based practice could, then, act as a ‘countervailing power’ (Light, 2010) on professions. Within the evidence-based movement, psychological and pharmaceutical interventions were evaluated by similar criteria. This meant that those psychological therapies that shared its epistemological assumptions could prosper, while overly optimistic claims about medication could be challenged. These were ideal conditions for the legitimisation of psychological therapies by NICE.
The 2002 NICE guideline on schizophrenia
The selection of the ‘conditions’ for which NICE guidelines are commissioned is ‘a political matter decided by ministers and the Department of Health’ (Kendall et al., 2004: 158). Thus the decision to focus NICE's first treatment guideline on schizophrenia was an indication of policymakers' concerns about it. The guideline was produced for NICE by the NCCMH (established by the BPS and RCPsych) and its GDG, which included two psychologists (Pilling as facilitator and the London group's Philippa Garety), four psychiatrists (including Kendall as chair), and a range of other stakeholders. As discussed earlier in relation to funding bodies, the constitution of these committees is a key issue, and Birchwood (who sat on the GDG for the 2014 edition of the guidelines) noted that the inclusion of ‘psychologists on the NICE committee is definitely very important’.
The final published guideline (National Institute for Clinical Excellence, 2002) reviewed a range of service designs and interventions including psychological therapies like CBT and BFI, providing a ‘clinical summary’, health economic evidence, clinical practice recommendations, and future research recommendations. It recommended the use of BFI and CBTp in a range of instances and also concluded that ‘providing CBT represents good “value for money”’ (ibid.: 136).
NICE's schizophrenia guideline had a significant impact. Pilling noted that ‘the methods that we developed for that review became the methods that we subsequently used, so it had influence on how we developed subsequent reviews for example for depression and anxiety disorders’. These guidelines were also important internationally in legitimising psychosocial interventions, especially CBTp (Gaebel et al., 2005). Pilling observed:
If you'd looked prior to the publication of that guideline … at the international guidelines on the treatment of schizophrenia, few of them actually mention psychological interventions at all. They were essentially a medication treatment guideline.… It helped people … think slightly differently about psychotic disorders and it prompted further research.
Discussion
We have attempted to identify the conditions of possibility that enabled the development and stabilisation of a heterogeneous collection of techniques as CBTp. This social technology provided a means for the profession of clinical psychology to expand into a terrain previously dominated by psychiatry. Psychiatry's jurisdictional vulnerabilities provided opportunities for this profession to establish jurisdictional legitimacy in the provision of therapeutic interventions for people with psychosis diagnoses. NICE acted as a countervailing power by providing a means of adjudicating jurisdictional contestation. A similar approach could usefully be adopted in relation to other topics both in mental health and in other fields where social technologies are utilised and where there is jurisdictional contestation.
Abbott’s (1988) approach suggests that a range of cultural values can help professions establish their jurisdictional legitimacy. Studies of the IAPT programme have focused on the importance of scientific and economic values (e.g. Pickersgill, 2019). However, although economic issues were not unimportant in the institutional legitimisation of CBTp, they were, perhaps, less important than in the development of IAPT. BFI and CBTp gained scientific legitimacy through researchers' use of diagnostic categories and methods like RCTs at the moment that evidence-based practice became an important countervailing power in healthcare. However, it appears that scientific values were not the only route to legitimacy. Psychiatry's jurisdictional vulnerability was not only due to biological psychiatry's failure to make the scientific advances it had promised (Harrington, 2019). It was also generated by the increasing criticisms of the quality of psychiatric care for people with a schizophrenia diagnosis made by service users, advocacy groups, the media, policymakers, and the mental health professions (including some within psychiatry itself). Clinical psychologists and psychosocially oriented psychiatrists engaged with policymakers as the political salience of schizophrenia changed in the 1990s. Also important were how the values of British clinical psychology aligned with broader cultural shifts towards consumerism and humanistic approaches to psychiatric care. As Abbott (1988) suggests, jurisdictional claims may gain legitimacy in response to shifts in cultural values, and these included not only science and rationality but also what Abbott describes as individual, social, economic, and political values. This broader conception of jurisdictional legitimacy may be of use both in the history of the psychotherapies and in other fields.
Our analysis has been attentive to national factors like the composition of the mental health workforce, professional cultures and histories, health policy, the media, and guideline-setting bodies like NICE. It is difficult, in the absence of comparative research, to fully answer the question of why CBTp developed in the UK rather than, say, in the US. However, our analysis suggests that the mutually influential relationship between the NHS, the professions within it, and currents in mental health policy were significant factors. In an article published after our interview with him, Nick Tarrier argued that ‘the NHS provided the infrastructure, opportunity and ideology for innovation’, which ‘provided fertile ground for CBTp to develop’ (Tarrier, 2014: 256). Structural factors like the arrangement of healthcare provision and the relative size of professions are also important, and both Abbott (1998) and Harrington (2019) have observed that a profession cannot maintain a jurisdictional claim if it does not grow in size to meet the demand it generates. As we noted earlier, it would appear that, as it has expanded into new therapeutic territories, making new jurisdictional claims and generating increased demands for its services, the British clinical psychology workforce has managed to avoid this problem and has expanded rapidly (Turner et al., 2015). Future research could examine how the trajectories of psychological therapies in different countries are influenced by national factors like their healthcare structures.
The challenge to the traditional division of labour between psychologists and psychiatrists raises the question of what new jurisdictional settlement was reached between the two professions by 2002. On the one hand, clinical psychology was expanding rapidly into a new area offering a psychological therapy that addressed, to some degree, the demand for more humanistic and patient-centred care. Moreover, psychologists' use of continuum and dimensional models emphasised the potential intelligibility of unusual beliefs and experiences, contrary to the received psychiatric view that they were un-understandable and that there was a sharp discontinuity between the ‘normal’ and the ‘abnormal’. There was also some engagement with the ideas of the growing mental health service user movement as evidenced by the BPS report (British Psychological Society, 2000). However, on the other hand, critics argued that CBTp could be assimilated too easily into a medical approach because of its individualistic focus and the use of RCTs, medical language (e.g. symptoms, treatment, relapse etc), and diagnostic constructs (as illustrated in the naming of the therapy: initially CBT for schizophrenia and then, subsequently, CBT for psychosis). Critics could view BFI and CBTp simply as an adjunct to standard psychiatric treatment for schizophrenia. However, for some, as we have seen, the adoption of a theoretically and methodologically conservative approach was a pragmatic strategy. Moreover, the research centres adopted different positions in debates about the value of diagnostic categories like schizophrenia and the role of biological aetiological factors and drug treatment. Thus the jurisdictional settlement between psychiatry and clinical psychology is, perhaps, still a work in progress.
As Pickersgill (2019) has observed, in practice, binary evaluations of medicalisation may not be appropriate, and we would argue that CBT has both medicalising and de-medicalising aspects. Future work might, as Halfmann (2012) has suggested, more usefully conceptualise medicalisation along multiple dimensions (i.e. discourses, practices, and identities and actors) and across multiple levels (micro, meso, and macro). Of course, such evaluations are even more complicated by the fact that CBT now comprises a heterogeneous array of approaches (Marks, 2012). As Woolfolk and Richardson (2018: 141) note, it is now a church broad enough to include ‘both Beck and the Buddha’.
A limitation of the present work is that interviewees were describing, in 2013, events that had occurred sometimes 20–30 years before and oral histories may be affected by the vagaries of memory. Moreover, the interviews were ‘on the record’, and this may have influenced the expressed views of the interviewees. However, we have sought to place these interviews in the context of broader scholarship. As an ‘elite’ sample, our interviewees were able to give a nuanced ‘insider’ perspective on the development of cognitive and behavioural therapies. It may well be that this has obscured the role of other influences. Our account should, therefore, be seen as provisional, and it is to be hoped that further research and scholarship might shed more light on the development and legitimisation of CBTp. Researchers could build on this work by interviewing a broader range of clinical psychologists and other professional disciplines from more geographically varied locations.
Future work could examine developments over the 20 years since the 2002 NICE guidelines. How has the jurisdictional contestation between psychiatry and clinical psychology, prompted by the development of the social technology of CBTp, progressed, and what new jurisdictional settlement has been reached? Psychiatry's jurisdictional vulnerability in relation to schizophrenia remains and, in some respects, has deepened. Harrington (2019) notes that debates about diagnostic classification and about the value of biological psychiatry are ongoing, and she reports signs that the pharmaceutical industry is withdrawing from the development of antipsychotic medication. Moreover, questions remain about the efficacy and acceptability of medication (Pilgrim and Rogers, 2009). CBTp has become increasingly legitimised, and NICE's recommendations were strengthened in the two subsequent revisions of the schizophrenia and psychosis guidelines such that the 2014 edition recommended that CBT be offered ‘to all people with psychosis or schizophrenia’ (National Institute for Health and Care Excellence, 2014: 242; emphasis added). However, this has not occurred without some debate (McKenna and Kingdon, 2014). In addition, clinical psychology has been increasingly challenged: the 2014 revision of the British Psychological Society (2000) report was criticised for failing to adequately address issues of race and culture (Kalathil and Faulkner, 2015), and a revised version was produced in response (Cooke, 2017). However, it is important not to exaggerate psychiatry's jurisdictional vulnerability. Currently, on average only 26% of NHS patients have been offered CBT, while 97.6% are prescribed medication (Royal College of Psychiatrists, 2018). Over a decade ago, Pilgrim and Rogers (2009: 959) concluded that, despite the challenges facing it, psychiatry was ‘surviving well’. This appears to hold true today and the question of how British psychiatry has navigated its jurisdictional vulnerabilities merits further study.
Footnotes
Declaration of conflicting interests
The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: David Harper was an ‘additional contributor’ to the British Psychological Society (2000) report and a ‘contributor’ to Cooke (2017). He has co-authored publications with Mary Boyle, Anne Cooke, Lucy Johnstone and David Pilgrim and Richard Bentall supervised his clinical psychology master’s thesis in 1990-1.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
