Core Processes in Brief Psychodynamic Psychotherapy

Advancing Effective Practice

Denise P. Charman

• Publisher: Routledge
• Year: 2003
• Print ISBN: 9780805840674
• eBook ISBN: 9781135639952
• Topic: Psychology
• Subtopic: Mental Health in Psychology
• Index: Psychology

Many students enter graduate programs with little or no experience of psychodynamic psychotherapy. Efforts to impart clinical skills have often been less than systematic and beginning psychotherapists have not always been encouraged to think about what they are doing and why they are doing it from a scientific standpoint.

Thoughtfully building on current debates over efficacy and effectiveness, this book outlines a promising approach to training in which the work of therapy is divided into tasks patterned after Luborsky's influential delineation of "curative factors"--significant developments in the course of the therapy that are crucial for effective change. Each task step for the therapist-cognitive, behavioral, affective, or a combination--is analyzed, taught separately, and then put in sequence with the other task steps. Curative factors have been extensively studied in recent years and the approach rests on a solid empirical base.

In a climate of increased accountability, clinicians must demonstrate that they are responding to providers' requests to conduct evidence-based practices. Core Processes in Brief Psychodynamic Psychotherapy will be an invaluable resource not only for students and trainees, but for established therapists who find themselves asked to justify their work.

PART I
INITIATING AND ASSESSING SUITABILITY FOR BRIEF PSYCHODYNAMIC PSYCHOTHERAPY

CHAPTER 1

Effective Psychotherapy and Effective Psychotherapists

Denise P. Charman
Victoria University

Psychotherapy is one of the most extensively researched treatments for mental disorders. When compared to no treatment or psychopharmacology, patient reports and comparative studies have demonstrated that psychotherapy is, in general, effective, efficient, and lasting (Asay & Lambert, 1999; Seligman, 1995). Researchers have also compared psychotherapy treatments hoping to find the most efficacious ones—a laudable undertaking. However, this undertaking has resulted in disenchantment on the part of many practitioners who feel keenly the gulf between research and practice (Talley, Strupp, & Butler, 1994). It is a paradox that efforts to demonstrate the beneficial effects of psychotherapy have led researchers to develop models and methodologies that are far removed from psychotherapy processes and practices.

The most stringent research is referred to as efficacy research, with the randomized controlled trial as the gold standard. Its methodology involves the comparison of manualized treatments administered by therapists who reach a criterion in competence and demonstrate satisfactory manual adherence. Treatments are delivered for a prescribed number of sessions to patients with clearly defined disorders who have been randomly allocated to treatment groups. Efficacy researchers have assumed that patients are not active in influencing outcomes, an assumption not well based (Tallman & Bohart, 1997) and now being discarded in drug trials (Bradley, 1997). Moreover, therapy in the field is not of fixed duration, is typically self correcting rather than manualized, involves active rather than passive patients, and focuses on functioning not symptoms (Seligman, 1998).

It is not just practitioners who feel disenchantment. Researchers also appear to be frustrated because despite intense efforts, efficacy research comparing psychotherapy treatments has demonstrated that they appear to be equivalent (Stiles, Shapiro, & Elliott, 1986). Some researchers find this incredible. They believe that if efficacy research methodologies were more finely tuned, the superiority of one treatment (perhaps the one to which they have an allegiance) over another would inevitably be demonstrated and their beliefs justified (e.g., King, 1998).

Unfortunately, the response of researchers to their disenchantment, a call for even greater methodological rigor, increases practitioner disenchantment. I have referred to this as a call to “polish the gold standard” (Charman, 2003). Treatments are to be even more precisely prescribed and monitored and definitions of disorder are to be more stringent (Lampropoulos, 2000). Disorders are defined by the Diagnostic and Statistical Manual of Mental Disorders (4th ed.) or International Classification of Diseases (American Psychiatric Association, 1995), even though diagnostic categories do not capture the differences in patient dynamics within the same disorder and are generally unhelpful to psychotherapists.

Given this situation of mutual disenchantment, it is important to note that efficacy research and “efficacy-for-disorder” research (Charman, in press) have produced a number of important outcomes. For example, they have defined clearly and demonstrated as efficacious a number of empirically supported treatments (Lampropoulos, 2000; Nathan & Gorman, 1998; Roth & Fonagy, 1997) and treatment components (Beutler & Harwood, 2001) that therapists can incorporate into their psychotherapy practices. An array of treatment manuals has been produced along with measures of manual adherence and treatment competence and purity (Luborsky & Barber, 1993), which are excellent devices for training purposes.

Disenchantment with efficacy studies has led to the development of a variant approach referred to as effectiveness research, which focuses on the feasibility and clinical utility of psychotherapy treatments in the real world (Seligman, 1995). Support for effectiveness research can be seen in the advocacy for the Public Health Model by the National Institute for Mental Health (NIMH) (Niederehe, Street, & Lebowitz, 1999). This advocacy is heartening to practitioners, although they may still feel that even this model is more of the same, especially because there are moves to find a gold standard effectiveness methodology (e.g., Marginson et al., 2000).

Effectiveness research is packaged in an individualized form as the “patient- focused approach” advocated by Howard and his colleagues (e.g., Howard, Moras, Brill, Martinovich, & Lutz, 1996). This approach provides practitioners with a method to monitor the progress of their own patients by collecting data from each patient, which are mapped as an individualized patient recovery curve. Any patient curve can then be statistically compared with normed curves to determine if clinically meaningful change has occurred. It is essentially a case-based quality assurance system (Beutler, 2001) that addresses the question, “Is this patient’s condition responding to the treatment that is being applied?” (Howard et al., 1996, p. 1060). The approach can signal when a patient is not improving sufficiently, enabling the therapist to adjust treatment. However, if practitioners find that they are less effective than desired, the patient-focused approach does not, as yet anyway, provide insight as to how they might adjust their treatment, as it does not monitor within-session events and is not related to psychotherapy theory.

TREATMENT RESEARCH IDENTIFIES EFFICACIOUS AND EFFECTIVE THERAPISTS

The relation between treatment research and real-world psychotherapy is not as dichotomized as the previous section implies. In the course of efficacy studies, researchers have identified efficacious therapists. Their findings prompt a revision of the research question away from which therapy is more efficacious to which therapist is. For example, researchers in the NIMH’s Treatment of Depression Collaborative Research Program (TDRCP) generated therapeutic efficacy scores for all TDRCP therapists and reanalyzed their data (Blatt, Sanislow, Zuroff, & Pilkonis, 1996). The most efficacious therapists had patients improve significantly more, had less variability in patient outcomes, and had stronger alliances as perceived by patients. Alliance is considered a common factor across therapies. Efficacious therapists were not significantly different from less efficacious therapists on demographics or level of clinical experience.

Significantly, the two most efficacious therapists (from the placebo and imipramine treatments plus clinical management; Blatt et al., 1996) were providing treatments consistent with their preferred mode of working. This is what happens in the real world of psychotherapy: Psychotherapists opt to train in therapies of their own choosing and generally train for effectiveness in their preferred field. This finding highlights an important shortcoming in research methodologies. Therapists in research studies are not randomly allocated to treatment groups (only patients) and are often recruited to deliver a particular treatment via a process of self-selection, treatment preferences, and predisposing personal characteristics. The predilections of the therapist toward the therapy has been referred to as “therapy allegiance” by Wampold (2001), who suggested that this is another common factor across therapies.

Peer Identification of Effective Therapists

Recruitment processes for research therapists have often obliged peers or supervisors to nominate “good” therapists. Yet, it has been a consistent finding that when such good therapists were compared, their patient outcomes have shown great variability (e.g., Goldfried, Raue, & Castonguay, 1998; Masterton, Tolpin, & Sifneos, 1991). However, despite the apparent invalidity of their ratings of effectiveness, peer and supervisor ratings continue to be valued. Therefore, it is important to understand the bases for these ratings.

A pilot study consisting of a telephone survey (n = 15) of psychotherapists listed in the business telephone book sheds some light. Eighty percent of therapist respondents were psychodynamic in orientation and had at least six years experience. Respondents were asked, “In your opinion, what makes for a good psychotherapist?” The adjectives that respondents gave were categorized into dimensions, which were then ordered according to the number of respondents providing adjectives related to that dimension. The three most salient dimensions were, in order, personal qualities, interpersonal qualities, and training (equal to interpersonal qualities).

Respondents in the telephone survey described personal qualities using words and phrases such as “mindfulness,” “not having an agenda,” “concern for others,” “intelligent,” “not a rigid personality structure,” “sense of self,” “intuitive,” “self-aware,” “thoughtful,” “knows own issues,” “able to take care of self,” “heart is able to be open,” “patience,” “creative,” and “able to separate.” These descriptive words convey a sense of self-relatedness. Therapist self-relatedness is the “interactive self-experience” with “self-awareness, self-control, self-esteem, and so forth, being manifested clinically in varying degrees of openness versus defensiveness” (Orlinsky, 1994, p. 105). No respondent used descriptors relating to suffering and survival, although earlier in human history and in some indigenous communities today suffering and survival have been significant life experience qualifications for shamans (Ellenberger, 1983). Indeed, psychotherapists, especially those who choose a psychodynamic orientation have been found likely to come from dysfunctional families (Halgin & Murphy, 1991).

Interpersonal qualities were indicated by respondents’ use of words and phrases such as “listening,” “responding,” “having empathy,” “an accepting presence,” “being authentic,” “genuine,” “in tune,” “trust in the inherent qualities of the other,” and “available to the patient.” Interpersonal aspects “reflect joint contributions to the global quality and atmosphere of the emergent dyadic/group process … characterized in particular by varying levels of therapeutic teamwork” (Orlinsky, 1994, p. 105). The therapist respondents in the telephone survey did not use descriptive words available within their theoretical (psychodynamic) framework such as containment nor did they take into account the need to tailor technique according to the type of patient. For example, for a patient with paranoia, warmth, caring, and empathy may exacerbate the paranoia (Mohr, 1995).

Thus, the telephone survey confirmed that peers rely on assessment of personal and interpersonal qualities to assess effectiveness. The results were remarkably consistent in this small sample; therefore, a larger number of respondents may not alter the findings. However, on scrutiny, these nominated qualities are idealized and limited, for example by not being based on their personal experience as survivors or as having the ability to modify technique based on patient characteristics. Therefore, to the extent that peer ratings of therapist effectiveness are based on global judgments of personal and interpersonal qualities, it is little wonder that many studies have had mixed results in relating peer ratings to patient outcomes. Indeed, effectiveness as assessed by ratings by supervisors and peers have been found to be not consistent among each other (Najavits & Strupp, 1994), not reach significance (Luborsky, McLellan, Diguer, Woody, & Seligman, 1997), and not to be reliable (Lambert & Okiishi, 1997).

The ways in which personal and interpersonal qualities are related or unrelated to effectiveness have been addressed in a limited number of empirical studies. One of the earliest studies was conducted by Ricks (1974). It was a follow-up study of distressed male adolescents who had been seen by either Therapist A or Therapist B. When the adult clinical status of the patients was examined, both therapists were found to have been effective with less distressed patients. However, with very distressed patients Ricks found differential effectiveness related to personal and interpersonal qualities of the therapists. Therapist A, the “supershrink,” had invested more time, made use of resources outside of therapy, was firm and direct with parents of the boys, encouraged moves to autonomy, implemented problem solving, was more consistent, and had strong alliances (Najavits & Strupp, 1994). That is, the supershrink had effective interpersonal qualities. On the other hand, Therapist B, the “subshrink,” had invested less time, withdrew from the boys, was frightened of pathology, seemed to become depressed with difficult cases, and was hopeless about outcomes (Najavits & Strupp, 1994). In short, the subshrink had poor personal qualities. It appears that high and low therapist effectiveness may have different predictors, with poor therapist personal qualities associated with poor patient outcomes and good therapist interpersonal qualities with good patient outcomes.

The most effective therapists in the Najavits and Strupp (1994) study were distinguished from least effective therapists in that they had positive qualities of warmth, affirmation, understanding, helping, and protecting. The most effective therapists had few actively negative qualities and were more self-critical. Less effective therapists in the Najavits and Strupp study were more actively hostile (Najavits & Strupp, 1994), had hostile introjects, or were high on scales of dissaffiliativeness (Hilliard, Henry, & Strupp, 2000). That is, less effective therapists had controlling and hostile views of themselves, which were related to engaging in more patient interactions that were more hostile, were less prone to grant friendly autonomy, and gave mixed messages that were simultaneously affiliative and hostile (Henry, Schacht, & Strupp, 1990; Strupp, 1993).

Thus, the therapist’s (not just the patient’s) capacity to relate is very important for patient outcome (Hilliard et al., 2000; Luborksy, Barber, & Crits-Christoph, 1999). Hostility in the therapist denies patients the opportunity to attain the hopefulness that therapy will be helpful (Snyder, Michael, & Cheavers, 1999) and achieve remoralization (Howard, Orlinsky, & Lueger, 1994). The two maxims of “do no harm” and “instill hope” have been given vital new tighter meanings based on research findings.

Lafferty, Beutler, and Crago (1989) found that the least effective psychotherapists (of 30 trainees) had less empathy and valued comfort, stimulation, and intellectual goals more than effective therapists did. These personal qualities, however, can make for success as a student and as an academic. There lies a conundrum in the identification of effective therapists by academic psychotherapists.

The previously mentioned studies support the call by Stein and Lambert (1995) for graduate programs to emphasize personal qualities of trainees in selection. This is especially important because Hilliard et al. (2000) found that extensive training in psychodynamic psychotherapy was not adequate in correcting the impact of therapists’ own interpersonal histories on their psychotherapeutic work. Such findings confirm Strupp’s (1976) early call for psychodynamic psychotherapy research and psychotherapy training research to be interdependent. There are research opportunities here on the relation between selection, effectiveness, and personal and interpersonal qualities, especially with the development of circumplex models, for example Structural Analysis of Sequential Behavior (Benjamin, 1996), Millon Personality Model (Muran, Samstag, Jilton, Batchelder, & Winston, 1997), and Inventory of Interpersonal Problems (Horowitz, Rosenberg, & Bartholomew, 1993). An alternative to good selection process is to mandate personal psychotherapy as a component of training. However, this alternative does not seem ethical at this time because positive psychotherapy outcomes cannot be guaranteed (Mohr, 1995).

Effectiveness Defined as Patient Outcomes

Because peer and supervisor reports are not reliable and valid assessments of therapist effectiveness, effectiveness needs to be defined in alternative ways. It is likely that these alternatives will be based on work samples (Luborsky et al., 1997) and criteria for assessing work samples based on patient outcomes (e.g., Lueger et al., 2001).

Interestingly, changes in patient status as an outcome of psychotherapy are not usually measured by changes in diagnosis. Patient outcome is a complex, multidimensional construct that has been operationally defined in multifarious ways. Examples include measures of symptomology, interpersonal functioning, social role performance, and specific treatment targets such as cognitions, behavior, affect, or physiological arousal (Docherty & Streeter, 1993). Patient outcomes assessed at the end of therapy with multiple measures have consistently revealed inconsistent results across those measures even in the same study, with some measures showing improvement and others not. This creates a confusing picture (Elkin, 1994; Hilliard et al., 2000).

This confusion may be due to the lack of an apparent theoretical rationale for the selection of outcome measures. Howard and his colleagues (Howard, Moras, Brill, Martinovich, & Lutz, 1996; Howard, Orlinsky, & Lueger, 1994) in their phase model have provided a theoretical rationale. This model has not yet been widely adopted. However, it does indicate that the nature of any expected beneficial outcome would be related to the phase of treatment that the patient has completed. Sequential improvement in outcomes would be expected, first in subjective well-being such as fulfilled expectations or increased hope (remoralization), then symptom reduction (remediation), and finally recovery of life functioning (rehabilitation; Orlinsky, 1994). This model implies that phase-related outcome measures should be assessed throughout treatment and clinically meaningful change monitored according to the expected phase-related changes. It is unknown if this order of recovery always applies. Sometimes the patient may show improvement in functioning before subjective well-being.

The confusing picture is also related to outcome measures varying according to vantage point: that is, whether it is the patient, observer, or therapist doing the rating. However, from alliance studies patient self-reports appear to be the best predictors of patient outcomes (Horvath & Greenberg, 1994). After all, it is the patient who needs to experience the outcome. Confusion abounds for other reasons, which are related to the timing of the measurement of therapy outcomes. End-of-therapy assessment may indicate that some therap...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• List of Contributors
• Preface
• Part I Initiating and Assessing Suitability for Brief Psychodynamic Psychotherapy
• Part II Determining and Maintaining a Focus
• Part III Keeping an Eye on the Relationship
• Part IV Ending and Evaluating
• Part V Special Issues
• Appendix