Philip Richardson  consultant clinical psychologist/ evidence-based lead.

BMJ Editor- Dr G. Andrews in his article on RCTs in psychiatry makes the misleading and unreferenced statement: 'long term psychodynamic psychotherapy..... has not been shown to be superior to talking to a mature and kindly advisor' (1).  The application of RCT methodology to long-term psychotherapeutic approaches is notoriously difficult, and we know of no trial that has done this using non-expert therapists as a control condition.  An evidence-based review of psychotherapy services in England (2) does not support Andrews' views on dynamic psychotherapy, and warns against conceptual slippage from 'scarcity of evidence' to 'evidence against'.  Most psychotherapy RCTs study short term treatments (12-20 sessions) and usually show no, or marginal differences in efficacy between cognitive and dynamic treatments.  Effect sizes are generally small albeit clinically significant; follow-ups are short.  With these brief interventions any coherent, skilfully-applied therapy model will apparently have moderate, though not necessarily enduring, effect.  Reviews habitually exclude all but studies of atypical, highly homogeneous patient populations.

The small amount of psychoanalytic psychotherapy available in the NHS is mostly used to treat tertiary-referral patients with moderate to severe, persisting disturbance, a diagnostically heterogeneous group of patients who have frequently failed to respond to short term, cognitive treatments.  Andrews' pessimistic views of dynamic psychotherapy are out-of date, as a number of excellent RCTs are in progress or recently published showing the advantage of psychodynamic psychotherapy over other approaches for severe disorders.  Bateman and Fonagy (3) for example demonstrate the value of a day hospital-based psychoanalytic approach to severely personality disordered patients.  Sandahl et al (4) demonstrate the superiority of psychodynamic group psychotherapy over CBT in reducing alcohol intake in alcohol-dependent patients, as judged by abstinence at 15 months follow-up.  These studies and others show change not just in well-being and in interpersonal relationships, but as demonstrated by 'hard' measures of reduction in utilisation of health and social services.

In practice CBT may have low patient acceptability.  The London Depression Intervention Study (5) set out to compare three treatment approaches for severe depression: systemic/dynamic couple therapy, drug therapy and cognitive therapy.  However most patients allocated to the cognitive therapy group found the treatment unacceptable and dropped out.  The final comparison favoured couple over drug therapy overall, even at 2-year follow-up.  These results challenge the idea that an evidence-based approach to the treatment of depression can rely exclusively on drug treatments and CBT.

References:

1. Andrews G. Randomised controlled trials in psychiatry: important but poorly accepted. BMJ 1999;319: 562-564.

2. Parry G, Richardson A. NHS psychotherapy services in England: review of strategic policy. London: Department of Health, 1996.

3. Bateman A, Fonagy P. (in press).  The effectiveness of partial hospitalisation in the treatment of borderline personality disorder- a randomised controlled trial.  Am J Psychiatry.

4. Sandahl C. et al.  Time-limited group therapy for moderately alcohol dependent patients: a randomised controlled trial.  Psychotherapy Research 1998; 8: 361-378.

5. Leff J, Vearnals S, Brewin C, et al. (in press)  The London Intervention Trial: an RCT of antidepressants versus couple therapy in the treatment and maintenance of depressed people with a partner: Clinical outcome and cost.  Brit J Psychiatry.

Jane Milton  consultant psychiatrist

Philip Richardson  consultant clinical psychologist/ evidence-based lead.

Robert Hale  consultant psychiatrist/ director, Portman Clinic

The Tavistock and Portman NHS Trust

120 Belsize Lane, London NW3 5BA

    Back to top