Beck Never Lived in Birmingham

This article appeared in Clinical Psychology, 2004, issue 34, pp 4-10

Beck never lived in Birmingham:

Why CBT may be a less useful

treatment for psychological

distress than is often supposed

Paul Moloney and Paul Kelly, Solihull & Birmingham Mental Health

NHS Trust

This article suggests that CBT has a long way

to go before addressing the difficulties of

typical NHS patients attending clinical

psychology services.

In recent years, a growing number of central

government agencies and therapeutic psychologists

have argued that Cognitive Behaviour

Therapy (CBT) should be the therapy of

choice for mental health practitioners – and particularly for those working in the time-limited

and pressured circumstances of primary care services.

This position is supported by claims that

CBT may be the most effective of all therapies,

that it is well founded in clinical outcome research

and in the findings of cognitive science and, finally,

that CBT offers a collaborative and inevitably empowering way of helping people in distress (e.g.

Beck, 1995; Fenell, 1997). In this paper we wish to

question all of the above claims. We suggest that

the current popularity of CBT may at least equally

reflect the needs and values of the mental

health professions, and of those political and

social institutions that help to shape their aims

and activities.

The paper will begin with a brief discussion of

the nature of CBT as practised within the NHS,

and will then move on to a critical examination

of the approach from three perspectives. These

will include, first, a body of psychological research

that may cast doubt on the conceptual

basis of the CBT approach; second, that section of

the psychotherapy outcome research literature

that bears on the effectiveness of CBT; and third,

the mental health epidemiological literature, which

may point to the social origins of much of the

distress encountered by CBT therapists working

in public services, and to the consequent need

for therapeutic approaches that emphasise social

action over the exploration of putative internal

psychological spaces.

The scope and nature of CBT

In the UK, CBT has enjoyed a strong historical

association with the profession of clinical psychology, particularly in NHS outpatient settings

(Clegg, 1998; Pilgrim & Treacher, 1992). After a

recent period of therapeutic eclecticism within

clinical and counselling psychology, CBT may be

returning as the preferred approach for most

therapeutic psychologists working in the public

health services, as supported by recent government

planning for mental health care (Department

of Health, 2001; NHS Executive, 1996).

In terms of both theory and application, the

term ‘Cognitive Behaviour Therapy’ encompasses

a group of perhaps 15 approaches, in which the

link between practice and theoretical foundations

may vary widely (Chadwick et al., 1996; Boyle,

2002). The origins of CBT are likewise multistranded, and include an (arguably incompatible) mixture of applied learning theory, the more accessible aspects of the psychodynamic tradition, selected elements of laboratory-based cognitive psychology and pragmatic clinical experience (Beck & Weishaar, 1989; Hawton et al., 1989; Willis & Sanders, 1997).

In general, CBT practitioners tend to share a

view of the human condition as a product of the

four interacting elements of cognition, physiology,

behaviour and emotion. Perhaps for most CBT

therapists, the treatment of distress will entail a

process of enabling the client to change ‘unhelpful’

beliefs and behaviours by means of a series of

structured exercises. This process is based upon

the key assumptions that the client’s more accessible thoughts – and the deeper patterns or

‘schemas’ that are said to underlie them – will

mediate their reactions to events, and that these

thoughts can be readily examined and then modified so as to yield therapeutic change (Davidson, 2000; Padesky, 1994; Trower & Casey, 1989).

Conceptual difficulties with CBT

Perhaps one of the strongest arguments in favour

of CBT is that the therapy is seen as being derived

from a scientifically valid body of knowledge

based within the discipline of cognitive science.

However, this claim may be challenged on a

number of different levels.

As already noted, CBT is based on the idea that

cognitive processes are fundamental in the origin

and amelioration of personal distress (Beck, 1995;

Beutler & Guest, 1989). Yet, as some reviewers

have recognised, evidence for this idea is equivocal

at best (Bracken, 2002; Cromby & Standen,

1998; Godsi, 1998; Hughes, 1997). For example,

although there are indications that depressed

people may say negative things more frequently

(or quickly) than non-depressed individuals, this

may actually reveal little about any causal relationship between cognition and emotion. Such a

process could easily reflect the effect of aversive

environments having primed many individuals to

more readily access pessimistic beliefs about

themselves and their world. Likewise, the finding

that successful completion of laboratory

tasks by depressed individuals can elevate their

mood may offer little support for the claim that

this has been achieved by the direct falsification

of the person’s underlying negative beliefs (e.g.

Beck, 1995). A more reasonable interpretation of

this work may be that the person’s basic belief

repertoire includes positive dimensions that are

activated by positive experiences, and that a given

individual’s prevailing negativity may be far more

an outcome of the kinds of negative experiences

that they have recently been undergoing (Erwin,

1996; Fancher, 1995).

However, we would argue that perhaps the key

flaw within most models of CBT lies in the poor

fit between the concepts of mind offered by many

writers in the field (e.g. Hawton et al., 1989;

Padesky, 1994) and those offered within other

branches of psychology.

For example, the wider perspective of both

historical and cross-cultural research indicates

large variations in the way in which human beings

have customarily understood the causes of their

thoughts and actions and behaviours, in turn suggesting that currently accepted Western views

on the nature of the self need have no special

claim to validity (Crook, 1980; Gray, 2002; Sue &

Sue, 1990). In the context of clinical problems,

there is some evidence that the self-abnegating

language commonly found among depressed

Westerners may be much less common among

other cultural groups, including South East Asians

and Africans (Marsella, 1981; Littlewood &

Lipsedge, 1997). One reason for this difference

may be that, for depressed non-Westerners, the

explanatory repertoires offered by their cultures

are less likely to promote those expressions of

guilt and responsibility that derive from Western

Christianity and the Protestant work ethic (Chan,

1990; Sue & Sue, 1990). Conversely, when seeking

to account for the likely origins of their own experiences of ill health, contemporary Westerners

may be inclined to discount the effects of those

social and material adversities with which they

may be struggling (Blaxter, 2003; 1997; Cornwell,

1984).

Indeed, the practice and theory of CBT seems to

be premised upon the quintessentially Western

idea of our being able to scrutinise and then

modify our own thinking, although it is sometimes

unclear whether this Cartesian notion of

an internal observer should be regarded as a

metaphor or a reality (see Baars, 1997). Nevertheless, a wide variety of research suggests that the process of introspecting into the causes of

our thoughts, feelings and actions may often be

inaccurate and misleading, and to an extent that

seldom seems to be recognised within the cognitive

and behavioural therapies (e.g. Beck, 1995;

Willis & Sanders, 1997).

In the field of neuropsychology, for example,

the well-known experiments conducted with

people whose brains have been surgically divided

have shown that the explanations that they give

for the causes of their feelings and actions can be

blatantly in error from the standpoint of an external

observer, yet completely compelling for the

individuals concerned (Gazzaniga, 1993; McKay,

1980). These results agree with a number of similar

findings in other branches of clinical neurology,

which suggest that even for physically normal

individuals, there is no necessary link between beliefs about the sources of subjective experiences

and of conduct on the one hand, and their demonstrable neurological, bodily and environmental underpinnings on the other (Claxton, 1999; 1996; Dennett, 1991; Parfitt, 1987). Indeed, neuroscientists increasingly view the brain as a set of parallel systems without any central controller or ‘Cartesian theatre’ in which thoughts may be viewed and then manipulated in the way that many CBT writers seem to imply (Blackmore, 2001; Damassio, 1994; Dennett, 1991; Norrentranders, 1998).

The results of several decades of social psychological research seem to complement this

picture. Here, researchers have consistently shown

that our degree of insight into the likely reasons for

our thoughts and behaviours can be surprisingly

limited, and that we may instead habitually rely

upon a priori (and often erroneous) causal theories

in order to explain ourselves to ourselves

and to one another (Caldini 1994; Nisbett &

Wilson, 1980; Wegner, 2002; Wilson, 2002).

Overall then, the notions of consciousness, introspection and deliberation that are central to

the theory and practice of CBT seem to match

poorly with current knowledge of the mind. This

incongruence may underline both how the therapeutic task of finding a cognition that putatively causes other thoughts or feelings may be far from straightforward, and also how the sensations of viewing and controlling our thoughts

within CBT may have little connection with the

webs of social and material influence that may in

reality shape much, and perhaps all, of our experience and conduct (Smail, 2001; Smillensky,

2000; Wegner, 2002).

CBT and the psychotherapeutic

outcome research literature: A critique

Despite the above difficulties with the theoretical

bases of CBT, there seems to be a growing consensus within the mental health field that the

effectiveness of this approach is well supported

by clinical research (Department of Health, 2001;

Roth & Fonagy, 1996). This research appears

to provide valid evidence, based largely upon

randomised control trials (RCTs), of the effectiveness of CBT in the treatment of many forms of distress. However, a thorough review of the literature reveals that this claim may lack firm support, for a number of reasons.

To begin with, many RCTs involving CBT have

included inadequate control groups for comparison

purposes – usually individuals who remain

on a waiting list or receive an unconvincing form

of pseudotherapy, delivered with limited commitment by the researchers (Bolsover, 2002; Holmes 2002; Mair, 1992). Second, CBT has indeed sometimes compared well with other therapies in a number of trials involving selected research populations, which may often comprise middle class university students and therapeutic practitioners who are strongly convinced of the efficacy of the CBT approach (Dawes, 1994; Mair, 1992). The situation may be different when comparisons are made within inner city community-based clinical settings, characterised by hard-pressed clinicians serving populations that experience high levels of social and economic deprivation (cf. Hagan & Donnison, 1999; Richards, 1995). Here, comparisons of CBT with other psychological therapies have suggested that the former may offer little or no significant additional benefit in the treatment of such problems as alcohol and drug abuse, depression, chronic anxiety, and behavioural and emotional disturbance (Dawes, 1994; Dineen, 1999; Eisner, 2000; Epstein, 1995; Elkin et al, 1994; Hemmings, 2002; Leff et al., 2000; Sandell et al., 2000; Sanders & Tudor, 2001).

A third challenge to the evidence base for the

effectiveness of CBT consists in the large body of

comparative clinical outcome literature that has

accumulated over the last half century. This has

convincingly shown that, for a wide range of

clinical problems, such effectiveness as psychotherapy does have may bear scant relation to the therapist’s theoretical position, extent of professional training or alleged expertise (Dawes, 1994; House, 2003; King-Spooner, 1995; Spinelli, 2001). These startling conclusions seem to have passed almost unremarked in the professional training literature. Indeed, particularly in the field of CBT, it seems to emphasise the acquisition of ever more refined clinical skills (see Proctor, 2002).

Finally, it is perhaps worth noting that even

strong proponents of CBT recognise that this

treatment, like all other psychotherapies, will significantly help only about two-thirds of all recipients, even under ideal research trial conditions (Bergin & Garfield, 1994; Smith & Glass, 1977; Tarrier, 2002). Both in the professional literature and in discussions of clinical practice that we have witnessed, this commonplace observation rarely seems to prompt any searching discussion as to why this might be so or as to what alternative forms of help might be offered. Yet this issue is clearly an important one, when lack of clinical improvement can all too readily be attributed to a failure of motiv on the client’s part (Pilgrim, 1997; Smail, 2001; Willoughby, 2002).

An alternative view: social inequalities

as the fundamental determinants of

personal distress

In contrast to this individualised view of the

genesis of human difficulties, many writers and

researchers in the mental health field have highlighted the importance of toxic social influences in the origins of personal distress. There are abundant indications that the incidence and severity of a range of familiar psychological disturbances (including anxiety, depression, psychosis, substance abuse and self-harm) are linked to the cumulative effects of widespread social and economic inequalities (Bruce et al., 1992; Godsi, 1998; Mirowsky & Ross, 1989; Prilleltensky et al., 2000; Wilkinson, 2001). Critical psychologists have described these processes by reference to the operation of varieties of social power, which may work to profoundly (and negatively) shape the identity and self-efficacy of the least privileged individuals – and indeed of wide swathes of the population during times of economic and political upheaval (Bordieu, 1984; Smail, 1993; Stoppard, 2000; Wilkinson, 1996).

In the context of the practice of CBT, this

analysis suggests that, for the distressed person,

any attempt to modify their ‘negative’ thoughts

will have little capacity to beneficially change their

psychological state in the long term, beyond

that individual’s power to alter the landscape of

social and material influences in which they are

embedded (Franzblau & Moore, 2001; Smail,

2001; Wilkinson, 1996). However, these kinds of

observations have at best been embraced with

ambivalence by the mental health professions.

Perhaps because they have always had to operate

in political and institutional climates that have

favoured individualised and technical or treatmentoriented approaches to distress, of which CBT may be a paradigmatic example (Ferudi, 2003; Hansen et al., 2003).

Conclusion

As the title of this paper suggests, our overall

argument is that although aspects of CBT may be

helpful – particularly those parts that encourage

the client to confront the environmental causes

of their distress where this is possible – it may

none the less be the case that, overall, the theory

and practice of CBT can be seen as effective only

if viewed from the standpoint of those in positions

of socio-economic privilege. For the majority of

clients seen by psychologists in areas of relative

deprivation (such as many parts of Birmingham,

for example), the emphasis of CBT on alleviating

distress through challenging thoughts may be

profoundly misleading for the client and for the

therapist alike, and may indeed constitute a poor

reflection of their shared clinical experience. In