(Editor's Note: Sidney J. Blatt was invited to prepare this article as part of this special section to exemplify the thorough empirical investigations behind the Psychodynamic Diagnostic Manual)
Results from empirical investigations of therapeutic process and outcome in intensive psychodynamic treatment of both outpatients and of seriously disturbed, treatment-resistant inpatients, as well as from investigations of brief outpatient treatment of major depression, demonstrate that therapeutic process variables and patient characteristics are major contributors to therapeutic outcome. These important process and patient dimensions are best conceptualized from a psychoanalytic perspective, and lead to several conclusions including the need for a psychodynamic diagnostic framework.
First, evaluation of therapeutic progress must go beyond assessment of symptom reduction as the primary measure of outcome and include assessments of decreased vulnerability to stress and the development of enhanced adaptive capacities (Blatt and Zuroff, 2005), including the development of more mature and adaptive representations (i.e., cognitive-affective interpersonal schemas) of self and others (e.g., Blatt, 1992; Blatt and Auerbach, 2001; Blatt and Ford, 1994; Blatt and Shahar, 2004; Blatt, Stayner et al., 1996; Fertuck et al., 2004; Vermote, 2005), especially changes in the content and structural organization of these representations (see also Mayman, 1967; Urist, 1977; Westen, 1991).
Second, extensive further analyses (see the summary in Blatt and Zuroff, 2005) of data from the NIMH-sponsored Treatment for Depression Collaborative Research Program (TDCRP; e.g., Elkin, 1994), a landmark randomized clinical trial comparing 16 weeks of two forms of brief, manual-directed psychotherapy (CBT and IPT) with medication (Imipramine) and a double-blind placebo, indicate that primary among the determinants of therapeutic outcome is the quality of the therapeutic relationship that patient and therapist establish very early in the treatment process (at the end of the second treatment session) and the patients' pretreatment personality characteristics (Blatt et al., 1995; Blatt, Zuroff et al., 1996; Zuroff et al., 2000), especially their pretreatment level of perfectionism. Specifically, the extent to which the patient perceived the therapist as concerned and empathic and lower pre-treatment levels of perfectionism both independently predicted significantly greater reduction of symptoms and of vulnerability to depression, as well as a significantly greater increase in adaptive capacities and stress resilience (Blatt and Zuroff, 2005). Pretreatment levels of perfectionism interfered with therapeutic progress, particularly in the latter half of the treatment, because these patients disengaged from the therapeutic process (Zuroff et al., 2000) and from their social relationships external to the treatment process (Shahar et al., 2004) as they approached the pre-established termination date (after 16 weeks of treatment).
Results from empirical investigations of therapeutic process and outcome in intensive psychodynamic treatment of both outpatients and of seriously disturbed, treatment-resistant inpatients, as well as from investigations of brief outpatient treatment of major depression, demonstrate that therapeutic process variables and patient characteristics are major contributors to therapeutic outcome.
The quality of the therapeutic relationship in these brief treatments, as assessed at the end of the second treatment session, however, moderated significantly the negative impact of pretreatment personality characteristics (i.e., perfectionism) on treatment outcome, especially at midlevels of perfectionism (Blatt, Zuroff et al., 1996). These findings, consistent with the recent emphasis on the importance of the therapeutic relationship in the treatment process (e.g., Krupnick et al., 1996; Norcross, 2002; Wampold, 2001), suggest that efforts to identify empirically based treatments (EBTs) require a much more complex view of the treatment process beyond evaluating the effects of particular treatments in reducing focal symptoms. Rather, these findings indicate that the investigation of the factors that contribute to therapeutic change in brief and long-term, intensive psychodynamic therapy requires a differentiated view of patients, therapists, and the treatment process in order to address complex questions like what kinds of treatment are most effective with what kinds of patients, in what kinds of ways, and through what types of therapeutic mechanisms (Blatt, Shahar and Zuroff, 2002).
Sidney J. Blatt, Ph.D., is professor of psychiatry and psychology, chief of the Psychology Section, Department of Psychiatry at Yale University, and on the faculty of the Western New England Institute of Psychoanalysis.
In further investigations of processes that contribute to therapeutic change, we have been especially interested in the role of patient personality characteristics, informed by psychodynamic formulations. In particular, we found that distinction between anaclitic and introjective forms of psychopathology (e.g., Blatt and Shichman, 1983) have enabled investigators to systematically introduce distinctions between types of patients into the investigation of therapeutic process and outcome (e.g., Blatt and Felsen, 1993; Cronbach, 1953). The validity of the anaclitic-introjective diagnostic distinction has been demonstrated by extensive recent research on depression (see summary in Blatt, 2004) and personality disorders (e.g., Levy et al., 1995; Morse, Robins and Gittes-Fox, 2002; Ouimette et al., 1994). In contrast to increasing criticism of the DSM in terms of its lack of a cohesive unifying theory, the forced demarcation between normal and pathological through arbitrarily defined threshold values, excessive concerns with manifest symptoms and signs, a high degree of overlap or “comorbidity” among presumed distinct disorders, failure to consider possible relations among various disorders and their links with variations in normal personality development as well as to considerations of etiology and therapeutic intervention (Blatt and Levy, 1998), various forms of psychopathology in the anaclitic and introjective configurations are viewed as interrelated forms of maladaptation that occur in response to serious disruptions of the normal integrative, dialectic development of interpersonal relatedness and self-definition (Blatt, in press; Blatt and Blass, 1996).
Psychopathology emerges as individuals, at different developmental levels, become preoccupied with a distorted one-sided effort to establish and maintain satisfying interpersonal relations at the expense of the development of self-definition or of a sense of self-definition at the expense of interpersonal relatedness. The anaclitic-introjective distinction derives from psychodynamic considerations including differences in instinctual focus (libidinal versus aggressive), types of defensive organization (avoidant versus counteractive), and predominant character style (e.g., emphasis on an object versus a self orientation, and on affects versus cognition). In these formulations, continuity is maintained among normal psychological development, variations in normal character or personality organization, and different forms of psychological disturbance. Even further, continuity is maintained within the two primary clusters or configurations of disorders so pathways of potential regression and progression as well as the nature of therapeutic change can be more fully understood and appreciated.
The psychodynamic distinction between anaclitic and introjective forms of psychopathology, between patients preoccupied with issues of interpersonal relatedness or with issues of self-definition (e.g., issues of self-worth and control), has enabled investigators to identify two groups of patients that are differentially responsive, possibly through different mechanisms, to different types of therapy. Though, as noted above, introjective patients (those with elevated scores on perfectionism) did relatively poorly in brief treatments for depression, introjective patients were particularly responsive to long-term intensive psychodynamic treatment (Blatt, 1992; Blatt and Ford, 1994; Blatt and Shahar, 2004; Fertuck et al., 2004; Vermote, 2005). These research studies also indicated that anaclitic patients were responsive primarily to supportive interpersonal or relational dimensions while introjective patients were responsive primarily to the interpretative or explorative aspects of the treatment process. Also, anaclitic and introjective patients appear to change in different ways, in aspects most relevant to their personality organization. Anaclitic patients change primarily in the quality of their interpersonal relationships and in their representation of self and other, while introjective patients change primarily in the frequency and intensity of clinical symptoms and in their cognitive efficacy (Blatt and Ford, 1994).
In summary, our findings about the importance of patient characteristics, such as the anaclitic and introjective personality configurations, and their interaction with aspects of the treatment process, underscore the importance of establishing a systematic psychodynamic diagnostic framework for conducting clinically relevant evaluations and for the investigation of aspects of the treatment process. A systematic psychodynamic diagnostic framework informed by current research, such as the anaclitic-introjective distinction, has the potential to contribute to a new generation of research that will further clarify, refine, and hopefully improve our therapeutic efforts.
[Editor's Note: The reference list for this article as well as a more comprehensive bibliography can be obtained by contacting the author via email at or by visiting his Web page at ]