top of page

The NICE guidelines are misleading, unscientific, and potentially impede good psychological care and help

(A pre-print of an article that appeared in Psychodynamic Practice: Individuals, Groups and Organisations Volume 15, Issue 1, 2009)

Phil Mollon


NICE guidelines have been given the authority to determine what psychological therapies can be provided within the UK National Health Service. This also has implications for private practice. The guidelines are based on a medical model and consider psychotherapy as analogous to a drug. Psychological thinking is discouraged by this approach. A large amount of psychotherapy research evidence is ignored by NICE, particularly the persistent finding that differences in effectiveness between therapies are minimal and elusive. Differences between the skills of therapists are a more significant factor. Practice-based evidence, and learning from the patient’s feedback (both conscious and unconscious) may be a better approach. For most forms of psychological distress, none of the main psychological therapies studied in randomised controlled trials can be considered clinically effective, even though they facilitate some degree of statistically significant change.

Keywords: NICE, CBT, psychotherapy research, medical model, practice-based evidence

About 125 years ago, Freud began to explore the dynamic unconscious – an unruly, vast, and awesome realm of passion, pain, and rage, lurking in the depths of the psyche. His investigations spawned many different forms of psychotherapy (including cognitive therapy[1]), each seeming to have their own merits and spheres of relevance. Today that unconscious is to be put back in its psychic box, and the diversity of therapies that have been Freud’s legacy are to be culled down to a very small number that meet with government approval.

The National Institute for Clinical Excellence (NICE) – originally established to assess the cost-effectiveness and value of drugs for use within the National Health Service – has been given authority to determine what forms of psychological therapy should be provided within the NHS[2]. Inevitably they also have implications for private practice and other non-NHS settings, since clients do read them and may complain if their treatment does not conform to that which is officially recommended as ‘evidence-based’. Increasingly these NICE documents are used not as ‘guidelines’ but as prescriptive and prohibitive instructions. They are harnessed to the government’s Improving Access to Psychological Therapies (IAPT) initiative[3], resulting in a huge amount of resources being provided to develop cognitive behaviour therapy services.

Many influential clinical psychologists within the British Psychological Society, committed to research along CBT lines, have been closely involved in developing these guidelines (some of which are co-published by the BPS). Indeed, it has often seemed difficult to question or challenge NICE in clinical psychology circles – and more moderate or inclusive views have tended to be sidelined. Psychotherapists and counsellors from other organisations have seemed intimidated by the claims of NICE for scientific authority. Others may simply have ignored them, perhaps not recognising their potential impact. Given what is at stake, in terms of resources and control over the marketplace of psychotherapy, it is surely time to recognise the flimsy garments of this particular emperor.

My perspective is that of a psychoanalyst and clinical psychologist, working within the NHS – and my view is that the competitive branding and marketing of psychological therapies is detrimental to working in a broadly-based way with the multiple factors involved in psychological distress and recovery[4].

The appeal of NICE

First we should consider the arguments often proposed in favour of NICE. From lectures by, and conversations with, the proponents of NICE, I have compiled the following common claims and assumptions:

• The various NICE guidelines for mental health problems have placed psychological therapies in central positions within the programmes of care recommended by the government.

• In so doing, the science and professions of psychology and psychotherapy are given due recognition in public perception and government policy.

• The guidelines facilitate the emergence of an evidence-based approach, and thus provide a scientific underpinning for psychological therapy.

• Those approaches that are not sufficiently evidence-based will tend to fade, leaving the more empirically-grounded methods as the basis of mainstream expected standards of care.

• The guidelines helpfully summarise a large amount of data, packaging it into clear and practical recommendations for use in the clinic.

• Through implementation of NICE, mental health care in the UK will become coherently based around scientifically supported forms of psychological therapy – this will represent a vast step forward in psychotherapeutic provision and a triumph for the science of psychology and the decades of work by clinical psychologists and other empiricists in establishing effective forms of therapy.

• Moreover, the establishment of empirically grounded protocols for effective forms of CBT means that these can be taught to many more therapists (from diverse backgrounds), greatly enhancing the availability of good therapy.

All this can sound very benign and rational – a triumph of decades of empirical work by the psychological professions, resulting in the government’s commitment to providing ‘state of the art’ and ‘evidence-based’ psychological treatments for all who need them, putting these on a comparable scientific basis to other components of modern health care. Unproven and unscientific approaches will be discarded, thus protecting the public and saving precious resources. Best of all, clinicians need no longer be troubled by uncertainty; they can consult NICE for the scientifically derived answer as to how to proceed with their client.

Looking more closely at NICE guidelines, and some of the unsound assumptions behind them, this positive perspective seems less certain. The fundamental problem becomes apparent when the purpose of NICE guidelines is considered. NICE purports to offer clinical guidelines concerning: “The appropriate treatment and care of people with specific diseases and conditions within the NHS”[5] - thus begging or sidestepping the question of whether common psychological problems are indeed ‘specific diseases’ – analogous to physical disease.

The medical model

The medical model of ‘mental illness’ was well expressed by Benjamin Rush, founder of American psychiatry, who claimed there is no essential difference between mental and bodily diseases:

“I infer madness to be primarily seated in the blood-vessels, from the remedies which most speedily and certainly cure it, being exactly the same as those which cure fever or disease in the blood vessels from other causes and in other parts of the body” [quoted in Binger 1966]

Whilst some conditions do give a relatively strong impression of having some biological basis, the tradition of overextending the category of ‘disease’ has a long history. For example, Henry Maudsley, father of English psychiatry, wrote in the 19 Century of a nasty condition called ‘masturbatory insanity’:

“The habit of self-abuse gives rise to a particular and disagreeable form of insanity, characterised by intense self-feeling and conceit, extreme perversion of feeling and corresponding derangement of thought, in earlier stages, and later by failure of intelligence, nocturnal hallucinations, and suicidal and homicidal propensities”

[quoted in Hare 1962]

Thomas Szasz, one of the most trenchant critics of the medical model of psychological distress, drew attention to the psychiatric view of homosexuality just a few decades ago:

“Our secular society dreads homosexuality in the same way as the theological societies of our ancestors dreaded heresy. The quality and extent of this aversion is revealed by the fact that homosexuality is considered both a crime and a disease … By medical definition, every homosexual act is the symptom of a medical ‘disease’” [page 272] and “I reject both the assumption that homosexuality is a medical illness and that accepted methods of psychiatric examination are a species of medical examination”.

[Szasz, 1970, p279]

Contact with real clients presenting in mental health settings soon reveals the complexity of factors behind psychological distress. For example, a single woman suffers with anxiety and depression. She is living in a council flat, suffering persecution by drug dealing neighbours who play loud music through the night, combining this with verbal abuse and threats of violence. In addition she has experienced harassment by drug users outside her flat and was mugged by ‘hoodies’. Does she have a medical ‘disease’, for which NICE can indicate the appropriate pharmacological or psychological treatment? Or is it that she is reacting to stressful events with understandable depression and anxiety, and associated neurobiological concomitants.

The NICE guidelines would recommend her ‘disease’ be treated with either a Selective Serotonin Reuptake Inhibitor (SSRI antidepressant) or with Cognitive Behaviour Therapy. These alternative treatments are presented as if within a similar category of phenomena. Thus the guideline on anxiety states:

“Any of the following interventions should be offered

Psychological therapy [CBT]; pharmacological therapy [SSRI], or if an SSRI is unsuitable or there is no improvement, imipramine or clomipramine may be considered; self-help bibliotherapy [based on CBT].” [abbreviated text]

Note that the notion of ‘disease’ and its quasi-medical treatment with either medication or CBT forecloses a meaningful exploration of the client’s world. No doubt some CBT practitioners would protest that this is not the case and that enquiry into the client’s circumstances and experiences and inner mental world are all crucial to their work. However, the conceptualisation of the client’s problems as ‘disease’ inherently annihilates meaning and individuality by homogenising emotional distress. Moreover, it implicitly reduces psychological therapy to a standardised (manualised) product resembling a drug. If it were not doing this, it would not be possible to make such prescriptive dictates. Instead, the guidelines would say something like:

“Consider the client’s life circumstances, relationships, sources of stress, predisposing history of trauma or loss, core beliefs, patterns of automatic thought, psychodynamic conflicts, physical health and nutrition – try to formulate what has contributed to the client’s state of anxiety - then consider what form, level, and combination, of intervention is most appropriate. These might include social work support, advice on finding work, discussion of parenting issues, exploration of family dynamics, lifestyle counselling, trauma-focused therapy for past experiences, CBT to look at patterns of thought and behaviour, and the provision of a safe place for emotional exploration and reflection.”

The guidelines do not present anything remotely resembling such broad enquiry.

The problems with psychiatric diagnoses and the medical model have been well-known since Szasz first wrote about them in the 1960s, and the psycho-bio-social model of mental health, proposed by Adolf Meyer (1950), is well-established in psychiatric training – but NICE blithely ignores this whole issue. NICE guidelines lie entirely within a medical model. It seems extraordinary that psychologists have colluded with this. We look in vain in the guidelines for psychological content – perspectives or hypotheses regarding the psycho-bio-social processes that give rise to mental health conditions. Instead, the repeated comparisons with drugs imply a notion of a disease entity that can be corrected by the right combination of pharmacological or psychological ingredients. A pharmacological or psychological pill is prescribed.

Our ‘patients’ do not have single ‘diagnoses’.

NICE guidelines could only be written by people who are not immersed in real clinical work. Any clinician who has worked extensively with real clients knows that each one is unique. Whilst some broad generalisations can be made, the underlying structures of cognition, trauma, neurobiology, formative experiences, developmental history etc. are very different from one client to another.

The guidelines are based on diagnoses – specific disease entities. However, most psychotherapists or counsellors are uneasy about using psychiatric diagnoses, feeling that these do not do justice to the complexity and individuality of the person. Moreover, it is known that many patients acquire multiple differing diagnoses from different psychiatrists. A high proportion of clients seen in secondary care could attract a ‘diagnosis’ of ‘personality disorder’ – although the qualities giving rise to this might not be immediately apparent. Whatever the merits of this particular diagnostic concept, it does mean that people to whom it can be applied have problems that are more complex than can be accommodated within the NICE guidelines for ‘specific diseases’.

The more pervasive and deeper disturbance might not be immediately apparent. Many clients seen in secondary care present with anxiety or depression, but gradually show characteristics broadly captured by the concept of Borderline Personality Disorder. A clinician following NICE guidelines for anxiety and depression would offer CBT. This may not be appropriate for the more complex personality problems. Jeffrey Young, developer of the Schema Therapy variant of cognitive therapy, has stated that standard CBT is actually contraindicated in work with Borderline Personality Disorder for at least the first year [Cambridge workshop 5.2.08]. He argued that people with BPD tend to experience standard CBT techniques as patronising and expressing a lack of understanding of their particular problems. 

A medical model assumes that specific diseases are somewhat independent. However, ‘patients’ seen within mental health services commonly present with several conditions: e.g. depression, anxiety; panic disorder, phobias, OCD, sexual problems, personality disorder – even schizophrenia. Statistically, this co-morbidity would be highly unlikely if the conditions were separate disease entities (Ross 1995; 2008). It seems more likely that the superficially different clinical presentations are essentially just different expressions of, and defences against, the same underlying mental pain – an argument originally advanced by the Scottish psychoanalyst, Ronald Fairbairn back in 1952.

This clustering of mental health problems has been demonstrated empirically. Studies find that co-morbidity of axis 1 disorders with other Axis 1 or Axis II disorders is up to 90% (Westen et al. 2004). Yet most trials of psychological therapies select patients with only one specific disorder. Eddy et al. (2004) found that in studies on psychological treatments for OCD an average of 62% were excluded because of co-morbidity. Westen et al. (2004) found that in trials generally, 40-70% of patients are excluded. Thus the patients studied in the formal trials that inform NICE are quite different from those seen in real life settings. This can lead to significant distortions of fit when the data are forced onto actual clients.

Jeremy Holmes comments as follows:

“Cognitive behaviour therapy works well in university based clinical trials with subjects recruited from advertisements, but the evidence about how effective it can be in the real world of clinical practice is less secure. In the London depression trial, for example, couple therapy performed better than antidepressants for treating severe depression in patients living with partners, but cognitive behaviour therapy came nowhere, having been discontinued early in the trial because of poor compliance from a particularly problematic (but clinically typical) group of patients.”

[Holmes 2002]

NICE distorts the expectations of both public and referring doctors

The lack of psychological content in NICE, combined with the pervasive implied view of CBT as a sort of verbal drug, encourages a misleading and unthinking ‘prescription’ of ‘CBT’ – as those of us who work within the NHS find evidenced both in referral letters and in requests from ‘patients’. Some colleagues tell me of their careful psychological work being criticised by those with little direct knowledge of psychological therapy because it does not conform sufficiently to their assumptions regarding CBT.

This restrictive ‘prescription’ is encouraged by NICE guidelines, which demand that:

“CBT should be delivered only by suitably trained and supervised people who can demonstrate that they adhere closely to empirically grounded treatment protocols” [Anxiety guideline, p 16]

The recurrent emphasis upon manualised procedures and ‘empirically grounded protocols’ means that clinical judgement, innovation, and adaptation to the individual client, are all discouraged. Instead of the creative use of observation, intuition, and inspiration, the practitioner must look to NICE and the protocols used in research trials.

This is a misuse of research. The requirements of a research design and those of clinical practice are different. Researchers have to develop a fixed protocol in order to standardise their experimental condition so that it can be compared with another standardised condition. Westen et al comment:

“The reality is that researchers generally solidify treatment packages (manuals) so early and on the basis of so little hard data on alternative strategies, even within the same general approach, that clinicians have to accept on faith that the treatment as packaged is superior to the myriad variants one could devise or improvise with a given patient”. [Westen et al. 2004]

When research design is inappropriately generalised to the clinic, the therapist is invited to ask: Given this diagnosis, what does NICE tell me is the appropriate treatment? Such a question follows from the implicit medical model. More relevant questions might be:

‘Why is this person presenting with this particular symptom/problem at this point? What is the person telling me of their current and past experiences and their patterns of thought, belief, and perception? What is the current context? What is the triggering event? What earlier developmental contexts are resonating with the current circumstances? Are their particular biological factors to consider? What is going on in this person’s therapy at this point – what might be blocking progress? What is the client’s view of therapy? What is his or her implicit or explicit model of how therapy should work?’

NICE can be a distraction from the task of figuring out what is really going on in the person’s mind and life experience.

The rich and diverse discourse of over a century of psychotherapy is foreclosed. Instead we have the repetitive drone of ‘CBT CBT CBT’. Moreover, the form of CBT prescribed is often a far cry from the cognitive therapy originally developed by the psychoanalyst, Aaron Beck. When presented with videos of Beck at work with clients[6], CBT practitioners are sometimes astonished and wonder ‘where is the CBT?’. What we observe is a genial and subtle psychotherapist, gently enquiring about the client’s experiences, thoughts, and history – and generally behaving in the facilitative manner found with many experienced and effective therapists. At a recent annual conference of the American Psychological Association, Beck emphasised the role of empathy, saying that some of his past students who lacked empathy had been unable to become effective therapists, although they could be effective researchers!

None of these subtleties bother the writers of NICE. Instead we find a relentless prescriptive certainty – often distinctly cavalier in its disregard for clinical difficulty and complexity. For example, the guideline on Obsessive Compulsive Disorder repeatedly advocates Exposure and Response Prevention [ERP] – which really means little more than an instruction to get dirty and not wash your hands until your anxiety subsides. But Jeffrey Schwartz, from the UCLA OCD research unit, notes that “controversy swirls around exposure and response prevention therapy” and he quotes a colleague who worked with Edna Foa, a noted proponent of CBT, stating that 25% of their patients refused to undergo a single session of ERP (Schwartz 2002 p.59-60).

Even Paul Salkovskis, a leading exponent of cognitive therapy, cautions against simple behavioural exposure for people with OCD, arguing instead for cognitive strategies and a treatment approach that combines discussion techniques and behavioural experiments, referring to a study by Van Oppen (1995) showing that “Cognitive treatment without the incorporation of exposure has, surprisingly, been shown to be at least as effective as behavioural treatment” (Salkovskis & Wahl 2004 p75). Moreover, the approach called Danger Ideation Reduction Therapy [DIRT] has been found, in preliminary studies, to be superior to ERP, in terms of reduction in symptomatology, cognitive change, and fewer dropouts (Jones & Krochmalik 2007). Thus, even within the CBT paradigm, NICE appears unduly restrictive, prescriptive, and simplistic.

NICE impedes innovation and thought

NICE guidelines, in their current form, are inherently conservative – recommending the research findings of the past. Inevitably NICE will lag behind innovative clinical practice in the field. Clinicians are constantly sharing observations, thoughts, testing hypotheses, and so forth, presenting detailed clinical accounts of their work, in seminars and conferences. We should certainly expect that the best clinical practice would go beyond the meagre recommendations of NICE.

The way the guidelines are presented and used currently is acting as a damper on thoughtful clinical exploration. Whilst ostensibly promoting psychological therapy, they actually impede psychological thinking, replacing this with a simplistic medical model and prescriptions of standardised forms of therapy analogous to pharmaceutical products. Integration of new knowledge, such as the burgeoning influx of neuroscience (e.g. Hart 2008), is effectively blocked by NICE since it does not form part of the approved treatment protocols. To use a current cliché, the NICE guidelines create a ‘dumbing down’ effect on clinical practice.

Misrepresentation of evidence regarding psychotherapy

The deepest flaw in the NICE guidelines is their essential misalignment with vast swathes of psychotherapy research that show:

[1] evidence for differences in effectiveness between different forms of psychological therapy is elusive and minimal; [2] differences in outcome between therapists (of different skill and personal qualities) are greater than those between different therapies

[Lambert, 2007]

Duncan and Miller (2006) comment that the ‘DoDo Bird verdict’[1] is the most replicated finding in the psychological literature, embracing many different research designs, clinical presentations, and settings. Of course the developers of brands of psychological therapy may wish to promote their product and cite research demonstrating its efficacy. However, since much psychotherapy research is carried out by people who owe allegiance to the therapy being tested, this can influence results and interpretation of results. It has been estimated that up to 40% of results of trials may be due to these ‘allegiance effects’ of researcher bias (Duncan & Miller 2006).

Bias may also influence the selection of evidence. Duncan & Miller 2006 point out that whilst CBT enthusiasts may cite 15 comparisons showing an advantage for that approach in randomised controlled trials, there are 2985 comparisons that show no difference (Wampold 2001). One of the largest and most sophisticated outcome studies – the National Institute of Mental Health Collaborative Study of the treatment of depression (Elkin et al. 1989) compared 4 approaches: Beck’s Cognitive Therapy, Klerman & Wiessman’s Interpersonal Therapy, antidepressant medication, and placebo. There were no differences in overall effectiveness. The treatment model accounted for 0% of the variance in outcome, the therapeutic alliance accounted for 20% and qualities of the therapist accounted for 8%. The most effective therapists achieved outcomes twice as large as the least effective therapists. In another study, Wampold & Brown (2006) found that 46-69% of the variance of change was due to the therapist.

Because two forms of psychological therapy were found effective in the NIMH study, NICE recommends both these. It states in the Depression guideline that CBT is the psychological treatment of choice, but that Interpersonal Therapy can be offered if the patient expresses a preference for it. The deeper intriguing messages of the study – that the compared forms of therapy were found to be equally effective, and that qualities of the alliance and of the therapist were found to be of more importance than the purported therapy – are ignored. It is as if the compilers of NICE reasoned that it must be by lucky chance or good judgement that the researchers happened upon two forms of psychological therapy that are both effective. They do not appear to see the obvious implication that many other therapies might also have been shown to be equally effective if they had been included – that it is not really to do with the form of therapy used or the model associated with it. Of course, NICE cannot embrace this perspective because it is at odds with its task of advising on specific treatments for specific diseases – and the medical model that pervades NICE.

Wampold (2001) surveyed a vast amount of data on psychotherapy outcomes to conclude that the overall differential effects are as follows. A massive 87% of the variation in outcome is due to extra-therapeutic factors, such as characteristics of the client. Of the remaining 13% of change that is due to the treatment, 8% can be attributed to the therapeutic alliance (60% of the 13% treatment effects), 4% is accounted for by the quality of the therapist’s allegiance to, and enthusiasm for, the therapeutic model (30% of the therapeutic effects), and 1% is due to the model or technique itself (8% of the therapeutic effects).

In a large practice-based study – i.e. based on real life data rather than artificial randomised controlled trials - Stiles et al. (2006) surveyed 1309 patients who received cognitive-behavioural therapy (CBT), person-centred therapy (PCT) and psychodynamic therapy (PDT) at 58 NHS primary and secondary care sites during a three-year period. The patients completed the Clinical Outcomes in Routine Evaluation-Outcome Measure (CORE-OM) at the beginning and end of their treatment. The therapists indicated which treatment approaches were used on an End of Therapy Form. Outcomes of six groups were compared: three treated with CBT, PCT, or PDT only, and three treated with one of these plus one additional approach (e.g., integrative, supportive, art), designated CBT+1, PCT+1, or PDT+1, respectively. The results were that all six groups averaged marked improvement (pre-post effect size = 1.36). Treatment approach and degree of purity of the approach each accounted for statistically significant but comparatively tiny proportions of the variance in CORE-OM scores (respectively, 1% and 0.5% of pre-post change). Distributions of change scores were largely overlapping. No outcome differences were found between therapies in relation to anxiety and depression – conditions sometimes considered to show superiority for CBT[7]. Thus this very large study of routine clinical practice at 58 different sites adds further strength to the impression that therapeutic approach is not, in itself, a very important factor. Good psychotherapists, of diverse therapeutic allegiances, are helpful.

The tiny amount of outcome variance that is due to the treatment model means that the NICE endeavour is analogous to an attempt to establish an evaluation of restaurants on the basis of what brand of salt is used in the cooking!

In many cases, outcome data supporting particular therapies are simply ignored. For example, at the Savoy Psychological Therapies Conference 30 Nov/1 Dec 2007 [London], Elliot, Friere, & Cooper presented a paper on Empirical Support for Person Centred Psychotherapy. They stated:

“For reasons that are not clear, the large empirical literature that supports the Person-Centred practice is generally not known or reflected in mental health policy…”, referring to 140 outcome studies, showing large amounts of pre-post change, which were maintained at follow-up. Person-centred therapy was shown to be statistically and clinically equivalent to other therapies, including CBT. The strongest support was shown for its use in depression, PTSD, and couples’ problems. At the same conference, Milrod presented an rct of psychoanalytic psychotherapy for panic disorder, demonstrating its effectiveness (published as Milrod et al. 2007) – a finding that is, of course, quite at odds with NICE recommendations.

But none of these therapies are as effective as we need them to be

Perhaps the most grotesquely misleading feature of NICE is its implication that we have developed clinically effective psychological treatments that actually cure people. When reading accounts of rcts, it is easy to slip into assuming that statistically significant change means that people actually resolved their psychological difficulties. Whilst for some people this fortunate result may have occurred, it is not the case for most participants in psychotherapy trials. The sobering truth is stated by Westen et al (2004), as follows:

“… the existing data support a more nuanced and, we believe, empirically balanced view of treatment efficacy than implied by widespread use of terms such as empirically supported, empirically validated, or treatment of choice …

With the exception of CBT for panic, the majority of patients receiving treatments for all the disorders we reviewed did not recover. They remained symptomatic even if they showed substantial reductions in their symptoms or fell below diagnostic thresholds for caseness; or they relapsed at some point within 1 to 2 years after receiving ESTs conducted by clinicians who were expert in delivery of the treatment, well supervised, and highly committed to the success of their treatment of choice” [p615]

Commenting on the huge NIMH Collaborative Research Program, comparing CBT, IPT, medication, and placebo, they conclude:

“Despite a promising initial response, by 18 months posttreatment, the outcome of brief psychotherapy was indistinguishable from a well-constructed placebo” [p599]

Similarly, Hollon et al. (2002) conclude:

“Despite real progress over the past 50 years, many depressed patients still do not respond fully to treatment. Only about half of all patients responded to any given intervention, and only about a third eventually meet the criteria for remission. Moreover, most patients will not stay well once they get better unless they receive ongoing treatment.” [p70]

For the most part, our mainstream psychological therapies, including CBT, are not clinically effective, in the sense of reliably and predictably eliminating all the manifestations of psychological dysfunction.

Is there an alternative to NICE?

If we put aside the dubious methodologies and recommendations of NICE, are we left with any basis on which to select and improve psychological therapy? Fortunately the answer is yes. Part of what has been so exasperating about NICE is that the guidelines could have been helpful and broadly acceptable if it were not for their hubris and false certainty. A summary and review of relevant evidence in relation to psychological treatment of various conditions, along with some tentative conclusions and recommendations as to what approaches and issues might helpfully be considered, would obviously be of value and widely welcomed. Such a document would facilitate a thoughtful and enquiring approach to clinical work. As they stand, the NICE guidelines impede thought. We can be grateful that there are indeed more useful alternative surveys of research, of real relevance to the clinician (e.g. Cooper 2008).

There is a large and accumulating amount of data giving some insights into the qualities characterising the most effective therapists. Along with the capacities for conveying warmth and empathy, well-known to researchers for many years, the best therapists tend to be flexible and to seek and take account of feedback from the client (Duncan & Miller 2008). Psychoanalytic therapists can, in addition, draw upon the facility of listening for unconscious feedback from the client – data that is plentifully available once one knows how to hear it. (e.g. Casement, 1985; Langs, 1978; 1982). The capacity to decode unconscious communication is a psychodynamic skill of far-reaching importance.

One of the implicit lessons of psychotherapy research is that since our models and associated techniques account for very little of the outcome variance, it is best to loosen our attachment to these. The tribal insignia of ‘brand name’ therapies should be worn lightly – perhaps even ironically – for these divisions are surely matters of foolish narcissism. No model or theory can remain pure once it is sphere of application is extended. It must evolve and change. Bion (1970) puts it as follows:

“As soon as the deductive system proves its value as an exploratory tool, the self-consistency, which appeared to exist when the domain in which the system applied was limited, is imperilled because readjustments become necessary to fit the theorem to its extended domain.” [p24]

This has always been the case with psychological and psychoanalytic theories of the mind, and their associated techniques. They begin as an attempt to account for, and treat, a particular condition of mind – and are then applied to other conditions, resulting in an ever-increasing conceptual and technical fuzziness, often conveniently overlooked until some major shift of paradigm is proposed. It thus becomes very difficult to know quite what is meant by terms such as ‘CBT’ or ‘Psychoanalytic Psychotherapy’, since these are continually evolving and vary greatly amongst practitioners. Attempts to ‘manualise’ or issue decrees that practitioners should “demonstrate that they adhere closely to empirically grounded treatment protocols” (NICE anxiety guideline) are efforts to fix and freeze the art of psychotherapy.

A few years ago, Robert Langs (1982), influenced by Bion, proposed a distinction between ‘truth therapy’ and ‘lie therapy’ – the latter consisting of attempts to promote comforting clichés in place of disturbing emotional truth. Along these lines, we might consider the lies inherent in the NICE guidelines:

  • that psychological distress falls into specific disease patterns;
  • that we have adequate and clinically effective psychological therapies for these ‘diseases’;
  • that diagnosis can be used as a sound basis for selecting a correct psychological ‘treatment’;
  • that some forms of therapy have been shown consistently to be superior than other forms;
  • that CBT should be regarded as the treatment of choice for most ‘patients’;
  • that psychological ‘diseases’ such as anxiety and depression are ‘caused’ by faulty thinking.

Bion (1970), in his discussion of ‘Lies and the Thinker’, pointed out that lies require effort of thought and persuasion for their manufacture, whereas truth simply exists. It is there to be apprehended – but can be obscured by our sense of already knowing, whether by means of our training or through consulting NICE guidelines. Bion comments:

“The more his interpretations can be judged as showing how necessary his knowledge, his experience, his character are to the thought as formulated, the more reason there is to suppose that the interpretation is psychoanalytically worthless.” (1970, p105)

Let us turn from the false knowing offered by NICE and, with honesty and relief, reclaim our ignorance!


Atkins, D.C. & Christiensen, A. 2001. Is professional training worth the bother? A review of the impact of psychotherapy training on client outcome. Australian Psychologist. 36 122-130

Beck, A. T. 1976. Cognitive Therapy and the Emotional Disorders. Republished by Penguin, 1991. London.

Binger, C. 1966. Revolutionary Doctor. New York. Norton

Bion, W.R. 1970. Attention and Interpretation. In Seven Servants. New York. Aronson.

Casement, P. 1985. Learning from the Patient. London. Routledge.

Cooper, M. 2008. Essential Research Findings in Counselling and Psychotherapy: The Facts are Friendly. London. Sage.

Duncan, B., & Miller, S. 2006. Treatment manuals do not improve outcomes. In Norcross, J. Levant, R., & Beutler, L. [Eds.] Evidence-based practices in mental health. Washington, D.C. APA press.

Duncan, B. & Miller, S. 2008. Lessons of the Supershrinks.

Eddy, K.T., Dutra, L., & Westen, D. 2004. A multidimensional meta-analysis of psychotherapy and pharmacotherapy for obsessive-compulsive disorder. Unpublished manuscript. Emory University. Atlanta, GA. (cited in Westen et al. 2004)

Elkin, I et al. 1989. The NIMH TDCRP: General effectiveness of treatments. Archives of General Psychiatry. 46. 971-982

Fairbairn, W.R.D. 1952. Psychoanalytic Studies of the Personality. London. Tavistock.

Hare, E.H. 1962. Masturbatory insanity: The history of an idea. Journal of Mental Science. 108.1-25

Hart, S. 2008. Brain, Attachment, Personality. An Introduction to Neuro-Affective Development. London. Karnac.

Hollon, S.D., Thase, M.E., & Markowitz, J.C. 2002. Treatment and prevention of depression. Psychological Science in the Public Interest. 3 39-77

Holmes, J. 2002. All you need is CBT. British Medical Journal 324 288-94

Jones, M. & Krochmalik, A. 2007. Obsessive-compulsive washing. In Menzies & de Silva [eds.] Obsessive Compulsive Disorder: Theory, Research & Treatment. Chichester.Wiley.

Lambert, M. 2007. Presidential address. What we have learned from a decade of research aimed at improving psychotherapy outcome in routine care? Psychotherapy Research. Journal of the Society for Psychotherapy Research. 17 [1] 1-14

Langs, R. 1979. The Listening Process. New York. Aronson.

Langs, R. 1982. Psychotherapy: A Basic Text. New York. Aronson.

Meyer, A. 1950. Collected Papers of Adolf Meyer. Baltimore. John Hopkins University Press.

Milrod et al. 2007. A randomized controlled clinical trial of psychoanalytic psychotherapy for panic disorder. American Journal of Psychiatry 164. 1-8

Rosenzweig, S. 1936. Some implicit common factors in diverse methods in psychotherapy. Journal of Orthopsychiatry. 6. 412-415.

Ross, C. 1995. Pseudoscience in Biological Psychiatry. New York. Wiley.

Ross, C. Presentation at the Cardinal Clinic trauma conference 12-13 June 2008. Windsor, UK.

Salkovskis, P & Wahl, K. 2004: Treating Obsessional Problems Using Cognitive Behavioural Therapy. In Reinecke, M.A. & Clark, D.A.[Eds.]Cognitive Therapy Across the Life Span:. Cambridge University Press

Schwartz, J. 2002. The Mind and the Brain. New York. Harper Collins

Szasz, T.S. 1970. The Manufacture of Madness. London Paladin,.

Stiles, W.B., Barkham, M., Twigg, E., Mellor-Clark, J., & Cooper, M. 2006. Effectiveness of cognitive behavioural, person-centred, and psychodynamic therapies as practiced in UK National Health Service settings. Psychological Medicine. 36. 555-566.

Wampold, B.E. 1997. A meta-analysis of outcome studies comparing bona fide psychotherapies. Empirically, ‘All must have prizes’. Psychological Bulletin. 122(3) 203-215

Wampold, B.E. 2001. The Great Psychotherapy Debate. New York. Lawrence Erlbaum.

Wampold, B.E., & Brown, J. 2006. Estimating variability in outcomes attributable to therapists. A naturalistic study of outcomes in managed care. Journal of Consulting and Clinical Psychology. 73(5) 914-925

Westen, D., Novotny, C.M., & Thompson-Brenner, H. 2004. The empirical status of empirically supported psychotherapies. APA Psychological Bulletin. 130 [4] 631-663

[1] ‘All have won and all must have prizes’ from Lewis Carrol’s Alice in Wonderland.

[1] In his 1976 book, Cognitive Therapy and the Emotional Disorders, Aaron Beck describes how he developed cognitive therapy out of his practice as a psychoanalyst. He gives a detailed account of his discovery of ‘automatic thoughts’ as a more hidden form of free-association. [2] At the time of writing, NICE have published 14 guidelines in relation to mental health, with more in preparation: [3] [4] One reviewer of this paper prior to publication commented that a particular therapy that might soon be endorsed by NICE had not been mentioned. My concern is not to advocate or defend therapy X against therapy Y, but with drawing attention to the flawed reasoning, unjustified conclusions, and inappropriate certainty that pervades NICE. [5] This somewhat hidden statement is found on the ‘What we do’ section of the NICE website , where it explains their three ‘centres of excellence’. The Centre for Clinical Practice produces the guidelines: “The Centre for Clinical Practice develops clinical guidelines. These are recommendations, based on the best available evidence, on the appropriate treatment and care of people with specific diseases and conditions”. [6] Available from [7] This additional analysis of the data was reported to me as a personal communication from Professor Mick Cooper.

bottom of page