The British Journal of Psychiatry
Psychodynamic psychiatry’s green shoots
Jeremy Holmes
  • Declaration of interest

    J.H. was Chair of the Faculty of Psychotherapy at the Royal College of Psychiatrists 1998–2002.


Psychodynamic psychiatry makes a significant educational, scientific and therapeutic contribution to contemporary psychiatry. Recent developments in gene–environment interaction, neuropsychoanalysis and the accumulating evidence base for psychoanalytic therapies and their implications for practice are reviewed.

Within the broad church that is psychiatry, psychodynamic approaches have until recently played a significant role. Understanding the developmental dynamics and defences which underlie symptoms, sensitivity to transference and countertransference as manifest in the therapeutic setting, and skilfulness in handling the psychodynamics of the doctor–patient relationship, even when the focus is primarily psychopharmacological, are basic aspects of psychiatric expertise integral to training and practice.

Despite this history, and being viewed by service users as an essential component of mental healthcare, psychodynamic psychiatry has become something of an endangered species. Can it survive as a significant force within psychiatry or is it doomed at best to ‘optimal marginalisation’,1 at worst a backwater of esoteric nostalgia? The focus of psychodynamic psychiatry is a developmental and interpersonal perspective on the origins and maintenance of psychological illness, together with specialist knowledge of psychoanalytically informed treatments able to alleviate it. The purpose of this editorial is to review recent developments, suggesting a muted revival of psychodynamic psychiatry’s contribution, in basic science, outcome studies and clinical practice.

Developmental psychopathology and the science of intimacy

A developmental perspective is indispensible for understanding psychiatric illness. The standard ‘diathesis–stress model’ views disorder as resulting from the exposure of genetically susceptible individuals to environmental trauma. A psychodynamic viewpoint combines with recent research to gloss this in two distinct ways. First, environmental susceptibility can be health-enhancing as well as illness-promoting. To take a recent finding in dopamine D4 receptor (DRD4) polymorphism: given maternal insensitivity, the 7-repeat allele is a vulnerability factor for externalising disorders in toddlers, but with sensitive mothers the 7-repeat allele confers less deviant behaviour. Significantly, from a psychodynamic psychiatry perspective, such infants respond better to ‘enhanced parenting’, namely psychotherapeutic interventions. Belsky and colleagues suggest that these and comparable gene–environment findings are best seen in terms of ‘plasticity’ rather than ‘vulnerability’ genes,2 which confer reactivity to environmental context, with potential for negative health effects in adversity, but positive effects under conditions of sensitivity and support, the latter including the kinds of health-promoting relationships inherent in psychodynamic therapies. Second, implicit in this model is the need for measures of the relational environment equal in sophistication to those of modern genomics. Psychodynamic psychiatry provides just such a ‘science of intimacy’. An example is the Adult Attachment Interview,3 which, in contrast to crude pen-and-paper checklists, taps into a person’s perception of their relational environment, both current and developmental, and can be used to track changes in the inner world in the course of psychotherapy.


A second growth point for psychodynamic psychiatry is in the new-minted niche of neuropsychoanalysis, which has captured the imagination of leading figures in both psychoanalysis and neuroscience.4 Psychoanalysts in search for scientific credibility can now visualise physical correlates of their black-box postulates, whereas neuroscientists learn from psychoanalysis how meanings can emerge from brain biology. There are numerous areas of mutual interest, including: the distinction between the declarative memory system and procedural memories and the ways in which the latter may encode early trauma; parallels between the conscious/unconscious dichotomy and the interplay between cortical and subcortical structures, especially the amygdala; ‘mirror neurons’ as the basis for empathy; the role of the right brain in processing the sensory aspects of memory and the relevance of this to post-traumatic stress disorder; neuroplasticity and the ameliorative impact of psychotherapy; functional magnetic resonance imaging-revealed changes in response to therapeutic intervention. No less a figure than Nobel prize-winning Eric Kandel endorses the need for an approach which reconciles the ‘rigorous empirical framework of molecular biology yet incorporates the humanistic concepts of psychoanalysis’.5

General evidence for psychodynamic psychiatry’s efficacy

An earlier cri de cœur about the imminent demise of psychodynamic psychiatry called for major research effort to establish its efficacy and relevance.6 More recently, a number of meta-analyses of dynamic psychotherapy have appeared in high-impact journals, with somewhat contradictory results. Leichsenring &Rabung7 found that long-term psychodynamic psychotherapies (LTPPs) produced large within-group effect sizes (average 0.8–1.2) comparable with those achieved by other psychotherapy modalities; that gains tended to accumulate even after therapy has finished, in contrast to non-psychotherapeutic treatments; and that a dose–effect pattern was present, with longer therapies producing greater and more sustained improvement. Two key mutative factors are a secure, sensitive and interactive working alliance; and facilitating experiencing of previously avoided painful feelings.8 Although expensive, psychodynamic psychiatry is able in some circumstances to ‘pay for itself’,9 thanks to offset costs of other expenses (medication, hospital stays, welfare payments, etc). Smit et al,10 by contrast, found the evidence for LTPP limited and conflicting, suggesting that positive findings may to an extent be an artefact of inadequate controls. More research is needed, but on balance the evidence does seem to favour psychodynamic psychiatry.

Effectiveness in specific disorders

Two other challenges to the claims for LTPP efficacy are that patients studied are diagnostically heterogeneous, and it is therefore difficult to determine the specific indications for psychodynamic therapies, and the paucity of head-to-head studies comparing psychodynamic treatment with briefer, possibly cheaper therapies. Responding to the second point is a research challenge for the future. For the first, the role of psychodynamic psychiatry in a number of specific disorders is beginning to emerge. Anxiety disorders are typically the preserve of cognitive–behavioural therapy (CBT), so it is noteworthy that Busch and colleagues have developed an evidence-based psychodynamic therapy for anxiety, concentrating particularly on the role of unconscious unexpressed anger, demonstrating good outcomes in 21 sessions compared with controls.11 Replication is clearly needed. Similarly, there is now a number of tailored short-term therapies for depression, for which a recent meta-analysis found within-group effect sizes of 0.69, and pre-/ post-therapy 1.34, both maintained at 1-year follow-up.12 Compared with CBT there was a small (0.30) immediate advantage to CBT but no differences at 3 months’ and 1-year follow-up.

The complexity posed by people with borderline personality disorder represents a major problem for psychiatric services. Two manualised modified psychodynamic therapies, mentalisation-based therapy13 and transference-focused therapy have demonstrated significant improvements for patients with borderline personality disorder compared with treatment-as-usual controls. Mentalisation-based therapy sees individuals with borderline personality disorder as having difficulties with ‘mentalising’, that is ‘reading’ one’s own and others’ thoughts and feelings, which leads to recurrent interpersonal conflict. The therapy encourages reflection on everyday affective crises, including those with the therapist, rather than offering ‘deep’ interpretations. Without the capacity to mentalise the latter are often incomprehensible or may precipitate feelings of shame and humiliation. Mentalisation-based therapy is typically delivered by specially trained non-medical mental health workers. However, this group of patients, for whom suicide and self-harm are commonplace, is arguably best helped by teams in which psychodynamic psychiatry-trained psychiatrists play a major role, as diagnosticians, supervisors and buck-stoppers.

The history of psychoanalytic approaches to schizophrenia illustrates in microcosm the rise, fall and tentative rebirth of psychodynamic psychiatry. Despite Freud’s misgivings, psychoanalytic treatment for schizophrenia was widely practised in the 1950s and 1960s, especially in the USA. The picture changed as effective drug treatments for psychosis became available and it was shown that many patients dubbed ‘schizophrenic’ in the USA in fact had borderline personality disorder. Eventually, McGlashan, a psychodynamic psychiatrist, published a much-publicised follow-up study of patients with schizophrenia showing that psychoanalytic therapy could lead to deterioration and was contraindicated.14 At this stage, it seemed that psychodynamic therapy for psychosis was obsolete. However, the Scandinavian ‘needs-adapted’ approach to schizophrenia now provides a model in which the uniqueness of each patient is recognised, medication kept to a minimum, the family dynamics around psychosis charted, and a long-term one-to-one relationship with a key-worker seen as vital to improvement.15 The needs-adapted approach is not strictly speaking psychodynamic, but contains psychodynamic psychiatry ingredients in attempting to help patients understand the nature and meaning of their symptoms, rather than simply seeing them merely as manifestations of a biologically based ‘disease’.

Psychodynamic psychiatry as an integrative discipline

Psychodynamic psychiatry’s integrative viewpoint distinguishes it from its parent discipline of psychoanalysis in a number of ways. First, although effective psychiatric work needs space and time for self-scrutiny, today only a minority of psychiatrists undertake personal analysis. Staff sensitivity or Balint groups, mindfulness training and personal coaching are alternative routes to the capacity for self-reflection needed if countertransferential iatrogenesis is to be avoided. Second, rather than espousing specific psychoanalytic ideologies – Kleinian, Relational, Kohutian, etc. – psychodynamic psychiatry fosters treatments which are tailored to particular disorders, drawing on the best and most effective psychoanalytic approaches. Third, the psychodynamic psychiatrist will be conversant in and respectful of other psychological therapies – CBT, interpersonal therapy, family therapy – understanding their indications and different roles. Fourth, the psychodynamic psychiatrist needs to be skilled both in prescribing medication and delivering psychotherapy, ever alert to the ways in which managing medication may enact a transference relationship rather than a medical need. The role of the psychodynamic psychiatrist, conversant with psychodynamic psychiatry’s increasingly scientific basis, is to develop advanced expertise in these areas, thereby strengthening the increasingly challenged culture of local mental health services.


  • Jeremy Holmes is Visiting Professor at the University of Exeter. His latest book, Exploring in Security: Towards an Attachment-Informed Psychodynamic Psychotherapy, was awarded the 2010 Goethe Award.

  • Received February 19, 2012.
  • Revision received February 19, 2012.
  • Accepted February 23, 2012.


View Abstract