Effective, Ineffective and Iatrogenic Treatment

This brief article explores some of the factors that are thought to account for successful outcomes in psychotherapy. It also looks at aspects of the therapeutic relationship that militate against positive outcome. In the main, the paper is written from an attachment/relational psychoanalytic perspective.

In a meta-analysis of outcome research, Roth and Fonagy (1996) review the relationship between outcome and therapist training, experience and technique, with specific reference to psychodynamic approaches. Findings indicate that certain dynamic techniques appear to facilitate the development of a therapeutic alliance or attachment relationship which, in itself, significantly influences outcome. However, the authors point out that psychodynamic therapy may require a fuller professional training than other modes of therapy. The evidence suggests that with increased training and experience “therapists may develop a capacity for matching their patient’s progress and to time interventions appropriately, engage their patient, select appropriate goals, and develop a productive collaborative relationship” (p. 355).

Roth and Fonagy (1996) cite evidence showing that experience and expertise “becomes a more important predictor of outcome with patients who are more severely disturbed” (ibid), emphasising again that “training and experience have a role in the therapist’s ability to repair inevitable ruptures in the therapeutic relationship, with consequent gains in outcome” (p. 356). The authors conclude by stating that “the literature clearly indicates that whereas good knowledge of the key components of a technique lays an important foundation for effective practice, expert practitioners are those who are able to use technical recommendations flexibly, and deviate and go beyond them at times when the clinical situation seems to require this” (p.357).

In looking at the therapeutic process from a developmental perspective, Hurry (1998) stresses that underlying models of insecure attachment inevitably arise in the therapeutic situation, thereby allowing for their recognition within the analytic relationship. Aspects of the transference will, therefore, emerge and coalesce around actual aspects of the setting, as well as in response to the analytic work itself. Hurry (1998) argues that concomitant to these processes the developmental relationship will need to be grounded within reasonable limits as this helps the patient to confront the reality of separateness and separation.

Hurry (1998) suggests that developmental work is rooted in the individual personalities of the patient and analyst and in their spontaneous interaction. Citing research by Kantrowitz (1995), she points to the importance of the fit or match between the patient and the analyst. Indeed, Kantrowitz (1995), in her long term follow-up study, found that patient-analyst match was the only factor relevant to successful outcome. Hurry (1998) argues that such matching is most important with regard to the spontaneous developmental aspects of the therapeutic process.

In expanding on this issue, Hurry (1998) notes Tahka’s (1993) recognition of the analyst’s legitimate experience of parent-like feelings of pleasure and pain as part of a stage-appropriate developmental relationship with the patient. However, Tahka (1993) acknowledges that there is a heightened countertransferential risk in developmental work of the analyst using the patient variously as a source of narcissistic gratification, to relieve guilt, to overcome feelings of helplessness or to gratify his or her own infantile needs. Hurry (1998) emphasises that these risks demand ongoing self-reflective monitoring of the countertransference.

In this context, Racker (1991) argues that the capacity for adequate countertransference experience depends primarily and decisively on the degree of the analyst’s own integration, and the degree to which the analyst is able to divide his or her ego into an irrational part that experiences and a rational part that observes the irrational part. In cases where the analyst’s attitude towards the patient is influenced by his or her own neurotic countertransference, the patient will be faced once again (and now within the analysis itself) with a reality that coincides in part with his or her neurotic inner reality.

In developing attachment theory, Bowlby (1988) dispensed with the theory of developmental stages, focusing, instead, on the security provided by the bond between mother and child. However, he emphasized that attachment dynamics are not outgrown with childhood but characterise intimate relationships in adulthood throughout the life-cycle. In outlining the clinical application of attachment theory in adult psychotherapy, Bowlby (1988) gives a strong indication as to why attachment theory did not incorporate a stage/phase theory. He notes that a therapist who adopts attachment theory would ask what his or her patient’s experience in childhood may have been to take account of why, for example, the patient is afraid to express his or her feelings openly and is distrustful of the therapist. Bowlby contrasts this mode of responding to that of a therapist who applies a stage or phase psychoanalytic theory. He argues that “. . . such a therapist might regard his patient’s reactions [to a separation] as being rather childish, even infantile, and as an indication that the patient was fixated in an oral or a symbiotic phase. What the therapist then might say, and especially the way he might say it, could well be experienced by the patient as lacking in respect for his (the patient’s) current feelings of attachment, distress, or anger. Here again there would be danger that the therapist might appear to be responding in a cold unsympathetic way and all too like one or other of the patient’s parents. Were that so the exchange would be anti-therapeutic” (p. 154).

Stern’s theory of the development of self would seem to most complement attachment theory. Stern (1985) proposes four senses of self, suggesting that each emergent sense of self defines a new domain, that is, a sphere of influence or activity of social relatedness. He views the initial formation of a domain as constituting a sensitive period of development, arguing that each domain will hold predominance during that time. However, he points out that once all domains of relatedness are available to the infant no one domain will necessarily claim preponderance during any particular age period.  

Stern (1985) suggests that the infant’s newly developed senses of self provide the arena within which clinical issues are played out. Such issues include attachment, autonomy, orality, symbiosis, mastery, control, trust, dependence, independence, and separation. However, he argues that clinical issues are not merely age- or phase-specific but life-course issues. He therefore contends that though these issues may take different forms during different developmental epochs they are continually being worked on, negotiated and reorganized in dyadic interaction throughout the life-cycle.

Feltham (2007) points out that therapists work from the necessary injunction “do no harm”. However, he questions whether this might slide into a defensive mind-set of “take no risks”. He suggests that the therapist who plays everything by the book, employing a set of rigid prescriptions to maintain an ethical frame, a la Langs (1988), may foreclose the space to develop a humane, intersubjective relationship, leaving the patient with a sense of having been abused by a too rigidly correct therapist. The “take no risks” approach may, therefore, end up breaching the “do no harm” injunction – in effect, the treatment becomes iatrogenic, that is actively harmful.

Marrone (1998) in addressing similar issues to those raised by Bowlby (1988) and Feltham (2007), argues that psychoanalytic treatment may not only be ineffective but also iatrogenic. He characterizes an iatrogenic relationship as one in which the patient becomes trapped in long term work with an analyst who adopts a cold, detached, dismissive style, and who employs persecutory and stereotypical techniques as a means of defending against countertransferential feelings of vulnerability and weakness in him or herself. Marrone (1998) details a range of persecutory techniques. These include frequent and intrusive interpretations and questions, derogation, invalidation of subjective experience and false neutrality.

Marrone (1998) suggests that the analyst who interacts with the patient in these ways is, in effect, identifying with the aggressor (A. Freud, 1993) that is, identifying with parental qualities and characteristics as a way of defending against anxiety and guilt. He contends that, in so doing, the analyst is re-enacting in the transference-countertransference matrix unresolved aspects from past relationships with his or her own parents.

Similarly, Aron and Hirsch (1992), in proposing that money is the last taboo in psychoanalysis, suggest that the analyst is faced with repeated personal and professional dilemmas concerning money matters. Specifically, the authors argue that the analyst may prolong an unproductive treatment for reasons of greed or economic dependency upon the patient. For the same reasons, the analyst may “fail to confront, challenge, or raise certain issues for fear that the patient may leave” (p. 244).

Aron and Hirsch (1992) go on to suggest that the analyst, not wishing to have weaknesses exposed, may attempt to hide his or her personal relationship to money. The patient, on becoming aware of this vulnerability, may be drawn into helping the analyst perpetuate self-deception about this issue. The authors argue that this form of protection or collusion may result in the patient remaining “sick” or, if better, staying in treatment for longer than is necessary. Aron and Hirsch (1992) suggest that in such a situation, the patient is, in effect, acting as a therapist to his or her analyst. They see similarities between this dynamic and the way in which children help their troubled parents by reversing roles and becoming a “parental child” (Bowlby, 1988). As with Bowlby (1988), the authors view this kind of interaction as a primary cause of psychopathology in children. Aron and Hirsch (1992) therefore conclude that, like a parent, the analyst must be prepared to sacrifice loss of dependence upon the patient. They argue that, unless these issues are openly addressed in the transference-countertransference relationship, the analyst may become enmeshed in the interactional world of the patient and thus interminably repeat early and current unresolved internalized relational patterns. In so doing, the analyst fulfils the patient’s expectations, becoming a familiar toxic transference object rather than a new developmental object (Hurry, 1998).

Marrone (1998) argues that the therapist who adopts an attachment theory orientation is less likely to relate to the patient iatrogenically because attachment theory emphasises the importance of human qualities of warmth, support, sensitivity and responsiveness. Attachment and infancy research have demonstrated that these are the very qualities in the parent that engender a sense of security and self-agency in the child. Clearly, then, such personal qualities in the analyst are essential if a therapeutic alliance is to be established. However, as Symington (1996) argues, the therapist who emphasises attunement and empathy may avoid engaging with his or her own and the patient’s inner fears and psychic pain. In such instances, the therapist may enter into a defensive mutual collusion with his or her patient. In this context, Symington (1996) speaks about the therapist’s need of an inner act of courage to free him/herself and the patient from the grip of feared and menacing internalized objects.

Similarly, as Herman (1992) points out, a therapist may defend against countertransferential feelings of helplessness in the face of a patient’s trauma by adopting a stance of grandiose special-ness or omnipotence. This carries an attendant risk of extreme boundary violations, up to and including sexual intimacy. Such violations are frequently rationalized on the basis of the patient’s desperate need for rescue and the therapist’s extraordinary gifts as a rescuer.

The extent, therefore, to which qualities of empathy and attunement may consciously and consistently be maintained throughout a period of long term work and protect the patient from an iatrogenic relationship, would seem open to question. The therapist, in a similar way to the parent, may, at times, be less than “good enough” and thus is likely to make mistakes in expressing his or her unique subjectivity. Therefore, Hurry (1998), following Tahka (1993), and Marrone (1998), following Peterfreund (1983), are right to point to the importance of the therapist being able to recognize and appropriately acknowledge his or her own faults and failings as part of an ongoing self-reflective monitoring of the countertranference.

This reflective quality in the therapist again highlights the significance of the repair of inevitable ruptures to the therapeutic relationship that arise as a result of becoming drawn into ubiquitous mutual enactments. As the relational literature shows, non abusive enactments may lead to profound therapeutic change, enhancing a sense of felt security and the capacities for affect regulation and reflective functioning or mentalization. Crucially, the outcome depends on how the enactment is worked with in the therapeutic relationship – can it be acknowledged, contained and collaboratively explored, validating the patient’s subjective reality and, thereby, provide a shared, co-constructed emotional experience as the rupture to the working alliance is repaired, or will it remain dissociated, replicating past abusive relationships and re-traumatize the patient?

This dichotomy is succinctly summed up by Kernberg (1995) who, in discussing the erotic aspects of the transference and countertransference, notes that “There is probably no other area of psychoanalytic treatment in which the potentials for acting out and for growth experiences are so intimately condensed” (p. 114). In this context, Mann (1997) argues that the therapist’s erotic subjectivity needs to be openly recognised to lessen the risk of his or her sexual desires being enacted with the patient. A relational perspective holds that the adequate repair of non abusive ruptures constitutes therapeutic action, with the consequent gain of a constructive outcome, as noted by Fonagy (1998), by Roth and Fonagy (1996) and, in slightly different terms, by Benjamin (1992).
The potential for ineffective and iatrogenic treatment, particularly in long term therapy employing psychoanalytic techniques, emphasises the need for a comprehensive training therapy and the ability to work with unconscious interactional forms of communication - with transference-countertransference enactments (Davies and Frawley, 1994; Herman, 1992; de Zulueta, 2007), and cycles of projective and introjective identification (Bion, 1984). Indeed, within contemporary psychoanalysis it is widely accepted that the therapist’s capacity to monitor and reflect upon his or her countertransference may act as a safeguard against iatrogenia and the development of a persecutory, abusive and exploitative relationship about which Hurry (1998), Kernberg (1995), Mann (1997), Marrone (1998) and Tahka (1993) are so rightly concerned, as well as one characterized by unacknowledged economic dependency, as depicted by Aron and Hirsch (1992). Supervision of the therapist’s clinical work has an important part to play in this reflective process.

Whereas Freud (1913) viewed countertransference as consisting of the analyst’s unanalyzed unconscious conflict and, therefore, as an obstacle to the analytic process, Mitchell (1997), from a contemporary relational perspective, presupposes that countertransference is an inevitable and integral aspect of the therapeutic relationship. He argues that the countertransference consists not only of unresolved experiences and conflicts from the therapist’s own past or current life, but also of important information about the patient’s material which may provide clues about unconscious inner conflicts. Mitchell (1997) contends that the selective disclosure of the countertransference may help the patient and therapist connect on an emotional level and, thereby, open up and vitalize the patient’s subjective experience, providing him or her with a sense of being valued and understood. Handled insensitively, however, disclosure may close down and deaden the patient’s experience. Clearly, then, the decision to disclose requires a responsible, sensitive and judicious approach by the therapist that takes full account of the need to protect the patient’s personal integrity and boundaries. A delicate balance needs to be achieved, therefore, whereby the therapist is neither excessively emotionally detached from the patient nor excessively intimate and intrusive (Mitchell, 1997).


Aron, L. and Hirsch, I. (1992). Money Matters in Psychoanalysis. In Relational Perspectives in Psychoanalysis, eds. N. J. Skolnick and S. C. Warshaw. Hillsdale, N J: The Analytic Press.
Benjamin, J. (1995). Recovery and Destruction: An Outline of Intersubjectivity. In Like Subjects, Love Objects: essays on recognition and sexual difference. New Haven: Yale University Press.
Bion, W. (1984). Elements of Psycho-Analysis. London: Routledge.
Bowlby, J. (1988). A Secure Base: Clinical Applications of Attachment Theory. Bristol: Arrowsmith.
Davies, J.M. & Frawley M.G. (1994). Treating the Adult Survivor of Childhood Sexual Abuse. New York: Basic Books.
de Zulueta, F. (2007). The Perversion of the Professional Caring Relationship. In Attachment: New Directions in Psychotherapy and Relational Psychoanalysis, Vol. 1, No. 3, vii – xii.
Feltham, C. (2007). Ethical Agonising. In Therapy Today, Vol. 18, No. 7, 4-6.
Fonagy, P. (1998). Moments of Change in Psychoanalytic Theory: Discussion of a New Theory of Psychic Change. In Infant Mental Health Journal, 19 (3): 346-353.
Freud, S (1913). The Disposition to Obsessional Neurosis, in vol. 11 of The Penguin Freud Library. London: Penguin.
Freud, A. (1993). The Ego And The Mechanisms Of Defence. London: Karnac Books. Herman, J.L. (1992). Trauma and Recovery. New York: Basic Books.
Hurry, A. (1998). Psychoanalysis and Developmental Therapy. In A. Hurry ed. Psychoanalysis and Developmental Therapy. London: Karnac Books.
Kantrowitz, J. (1995). Outcome research in psychoanalysis: review and recognitions. In T. Shapiro and R. Emde (eds), Research in Psychoanalysis: Process, Development, Outcome (pp. 313-328). Madison, CT: International Universities Press.
Kernberg. O.F. (1995). Love relations: Normality and pathology. New Haven, CT & London: Yale University Press.
Langs R. (1988). A primer of psychotherapy. New York: Gardner Press.
Mann, D. (1997). The Psychotherapist’s Erotic Subjectivity. In Psychotherapy: An Erotic Relationship. London: Routledge.
Marrone, M. (1998). Attachment and Interaction. London: Jessica Kingsley Publishers.
Mitchell, S. A. (1997). Influence and Autonomy in Psychoanalysis. Hillsdale, NJ: The Analytic Press.
Peterfreund, E. (1983). The Process of Psychoanalytic Therapy. New York: The Analytic Press.
Racker, H. (1991). Transference and Countertransference. London: Karnac Books.
Roth, A. and Fonagy, P. (1996). What works for whom: Limitations and implications of the research literature. New York: Guilford Press.
Stern, D. N. (1985). The Interpersonal World of the Infant: A View from Psychoanalysis and Developmental Psychology. New York: Basic Books.
Symington, N. (1996). Mental pain and moral courage. In The Making of a Psychotherapist, chapter 5, 50-60. London: Karnac Books.
Tahka, V. (1993). Mind and Its Treatment. Madison, CT: International Universities Press.

The views expressed in this article are those of the author. All articles published on Counselling Directory are reviewed by our editorial team.

Share this article with a friend
Twickenham TW2 & London WC1X
Written by Paul Renn
Twickenham TW2 & London WC1X

I am a relational psychoanalytic psychotherapist and author in private practice in central and southwest London (see below in 'Extra Information' for details of my practice addresses). I am accredited by the United Kingdom Council for Psychotherapy (UKCP). I work with individuals, offer couples coun...


Find the right counsellor or therapist for you

All therapists are verified professionals

All therapists are verified professionals