Intended for healthcare professionals


The future of psychotherapy in the NHS

BMJ 2004; 329 doi: (Published 29 July 2004) Cite this as: BMJ 2004;329:245
  1. Sandy Goldbeck-Wood, associate editor (sgoldbeck-wood{at},
  2. Peter Fonagy, Freud Memorial professor of psychoanalysis (p.fonagy{at}
  1. BMJ
  2. Psychoanalysis Unit, University College London, London WC1E 6BT

    More evidence based services are taking shape to meet growing demand

    The demand for psychological therapies in Britain has never been greater,1 yet their claim on scientific legitimacy and therefore on public resources has never been under greater scrutiny.2w1-w4 At a meeting held by the UK Council for Psychotherapy in November 2003 to address the future of psychotherapy in the NHS, the clearest messages were conflicting ones—that although the taxpaying public demands increased access to psychological therapies and the government espouses both patient choice and user centred services,1 the evidence on the efficacy and cost effectiveness of the many different psychotherapies is patchy. Randomised trials cover only a limited number of treatments, and many treatments remain unevaluated in relation to many conditions.3 Exclusion rates of 40-70% of presenting patients limit their generalisability to the treatment seeking population,4 and a dearth of long term data, data on quality of life, non experimental evidence, user perspectives, and evidence of the generalisability to NHS practice of studies carried out in other settings hampers rational purchasing decisions. Little is known about equity of access to therapy across socioeconomic or ethnic groups, and with neither a career structure nor a pay scale of its own, psychotherapy is not even formally recognised as an independent profession. The result is a lottery for patients and piecemeal and ad hoc arrangements for recruitment and supervision of staff.

    One of the key obstacles has been a conflict of cultures. The narrative based hermeneutic culture of post-Freudian psychotherapy and the empirical culture that dominates medicine have proved reluctant bedfellows. Theoretical differences are reinforced by longstanding political antagonisms and resentments. Analytic therapists in particular have been defensive and suspicious in the face of the evidence culture as though the very idea of objective scrutiny represented a hostile intrusion into a quasi sacred private world.w5 Cognitive behaviour therapy has done much better at embracing the need for evidence and the requirements of purchasers. As a result its strong research and dissemination strategy has found favour with service providers and research funding bodies lost in a jungle of conflicting claims and vested interests, regardless of whether it is the most effective for any given indication.5

    Real difficulties exist in providing meaningful evidence on psychotherapies.w6 Although research is legitimately and necessarily a public activity, much of what is most important in psychotherapy is legitimately and necessarily private.w7 As with surgical research, factors related to the individual practitioner and patient are probably at least as important a part of the “active ingredient” as the modality of therapy.6w8 Individual, patient related factors such as coping style may also be influential. A meta-analysis of 16 trials showed that self reflective and introspective individuals seem to benefit more from insight oriented therapies whereas impulsive and aggressive patients respond better to symptom focused procedures.6 In short, some of the influences that randomised trials would normally seek to exclude seem actually to be part of the active ingredient of psychotherapy. It seems that a good therapist can achieve results with the right patient almost irrespective of his or her declared brand of therapy.6 To complicate matters, therapists themselves disagree over fundamentals such as relevant outcome measures, and what constitutes a “therapeutic dose.” Analytic therapists favour a long term view, in which short term worsening of presenting symptoms may be seen to play a necessary part and success is measured less in terms of symptom relief than of self understanding. All these represent real barriers to the pursuit of evidence but barriers which are being surmounted.

    Evidence from randomised controlled trials now exists for the efficacy of psychotherapy in depression, panic, generalised anxiety disorder, eating disorder, and personality disorder. Cognitive behaviour therapy, family therapy, interpersonal therapy, cognitive analytic therapy, and some psychodynamic therapy have variously been shown to reduce life impairing mood states and behaviours substantially.3 But with notable exceptions, such as the use of cognitive behaviour therapy for panic disorders, most trials still leave patients with residual symptoms, and most participants in trials relapse or seek further treatment within a year or two of “successful treatment.”3 Most patients presenting at clinics are polysymptomatic, and improvement is inversely proportional to the chronicity and complexity of their presenting condition. Duration of treatment is also highly correlated with comorbidity and chronicity, and we know that theoretically based therapies with a coherent theoretical mechanism of action deliver better results than interventions, such as simple interpersonal support, which lack a coherent theoretical model.7

    But though the overall picture is diffuse, differences between individual therapies in relation to specific disorders are now emerging. The efficacy of cognitive behaviour therapy in conditions such as panic disorder, specific phobias, and obsessive compulsive disorder is well established, although critics argue that effect sizes have been overestimated because of inappropriate controls.8w9 Psychodynamic therapy has been shown to be effective in severe personality disorder,9 which responds poorly to cognitive behaviour therapy.w10-w12 Family therapy is effective in anorexia nervosa,10 and structured brief psychological therapies in depression11 are generally effective at least in the short term, although with no clear evidence of advantage to any particular approach.7 Again few trials have extended follow up, and those that do show a clear tendency for patients to relapse.4w13

    Unlike medicine and allied professions, psychotherapy has established itself primarily outside the state sector. From Freud's first analyses of the Viennese bourgeoisie to modern Britain, private provision still exceeds state provision Britain has no legal definition of psychotherapy, so the British market is richly varied in type of therapy, training, price, and quality. Regulation with bodies such as the United Kingdom Council for Psychotherapy, and the British Association for Counselling and Psychotherapy, is still purely voluntary. But if evidence based psychotherapy is to be more than a private luxury for the wealthy, and if the NHS is to meet growing demand for mental health services, the internecine warfare between the cultures of medicine and psychotherapy must stop. The NHS needs to recognise psychotherapists as professionals with proper career and pay structures, rather than treating them as petitioners at the gate. Psychotherapists need to speak coherently and convincingly as a profession in language that others can understand. And the joint pursuit of rigorous, theory based, patient centred, therapist led research into talking therapies must go on. Only then will patients receive the accessible, affordable, and effective talking treatments they need and want.


    • Embedded Image Additional references w1-w13 are on

    • Competing interests SG-W has received training in psychodynamic and humanistic therapies. She is not currently practising as a therapist.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    View Abstract