| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ALL ISSUES | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Educational Abstract |
Key Words: Research Abstracts • Case Formulation • Core Conflictual Relationship Theme
 |
INTRODUCTION |
|---|
 Top INTRODUCTION ABSTRACTSCONCLUSIONREFERENCES |
|---|
Formulation has been considered a core clinical skill and can aid the practitioner in multiple ways. A good formulation helps organize complex and sometimes contradicting information about patients. It facilitates understanding of the patient and consequently may increase empathy. Ideally, a formulation guides treatment, increases the consistency of one's interventions, and serves as a marker for change. It is best considered a "work in progress" that is altered as new information emerges and as the patient responds to interventions based on a preliminary formulation.
In the past several years, although primarily in the 1980s, more than 15 structured formulation methods were developed, several of which have been tested for reliability and validity. Characteristics these methods share are that 1) the formulation process is broken down into separate components; 2) they are based on clinician judgment rather than patient self-report; 3) inferences are relatively low-level; and 4) the methods include both cognitive and cyclical dynamic aspects.
The first and most researched of these methods is Lester Luborsky's Core Conflictual Relationship Theme (CCRT).3 It focuses on identifying a patient's central relationship pattern by examining narratives, out of which are extracted a wish of the patient, an imagined or actual response from an other, and a response of the self. Examples of research findings based on the CCRT are that 1) interventions based on a patient's CCRT predict psychotherapy outcome; 2) gaining mastery of the CCRT is associated with successful outcome in therapy; and 3) CCRTs derived from dreams and from narratives elicited in structured interviews are similar to those derived from psychotherapy transcripts.
In this set of research abstracts, I summarize and comment on recent research on psychotherapy case formulation. Of the six studies reviewed, three investigate the CCRT. One investigates another approach called the Plan Formulation Method.4 The remaining two examine patients' own formulations and a cognitive-behavioral case formulation method.
|
ABSTRACTS |
|---|
|
TopINTRODUCTIONABSTRACTSCONCLUSIONREFERENCES |
|---|
Summary: Using Luborsky's Core Conflictual Relationship Theme (CCRT) method, the authors identified the most pervasive central relationship patterns among a group of 20 male Vietnam-combat veterans being treated for combat-related posttraumatic stress disorder (PTSD). The most pervasive patterns that emerged were a wish to be "close and accepted," a "rejecting and opposing" response from others, and a reaction of self to others of being "disappointed and depressed" and tending to "oppose and hurt others." Relationship themes were identified by asking the veterans to tell 10 stories about themselves in relationships. These narratives were then transcribed and coded by the first author, using a CCRT coding method based on standard categories for wishes, responses of others, and responses of the self. A senior psychologist coded CCRTs for 5 of the 20 patients, producing an overall agreement rate of 88.3%.
Comment: This study found similar relationship themes among PTSD patients as have previously been found among depressed patients, a group of patients with mixed diagnoses, and as we shall see below, a group of patients with borderline personality disorder diagnoses. Interestingly, the authors point out that none of their 20 subjects fit the overall pattern for the sample. This leaves open the question of whether studies following similar methodologies are identifying psychic structures that exist within individuals, or patterns that characterize a sample, but no individual within the sample. The methodology also raises the question of whether a genuine relationship pattern has been identified in which a person's wish leads to a response from others and a subsequent response from the self. In order to make that determination, data analysis would have to occur at the level of the individual patient—prior to being aggregated across patients as has been done in this and in similar studies.
Chance SE, Bakeman R, Kaslow NJ, Farber E, Burge-Callaway K: Core conflictual relationship themes in patients diagnosed with borderline personality disorder who attempted, or who did not attempt, suicide. Psychotherapy Research 2000; 10:337–355
Summary: Using the CCRT, the authors explored three hypotheses in a sample of 22 psychiatric inpatients reliably meeting diagnostic criteria for borderline personality disorder, 11 of whom were suicide attempters, the other 11 of whom were not. First, they predicted that suicidal patients would be more likely than nonattempters to describe a wish to be taken care of, a rejecting response from other, and a response of anger by the self. Second, they predicted that the same relationship pattern would co-occur at a greater frequency in their entire sample as compared with base rates. Third, they predicted that suicide attempters would experience others as interfering with their wishes, would respond negatively, and would have difficulty expressing these feelings directly, as compared with the nonattempters. None of these hypotheses received significant support. The two groups appeared much more similar than different. The most frequent CCRT components were a wish to be loved, a rejecting response from others, and a disappointed response of the self. The authors also conducted exploratory sequential analyses to determine patterns within relationship episodes. They found that a wish to be loved was linked to a response of rejection by others in 70 of 340 relationship episodes, and was present in 80% or more of the subjects. They found that experiencing others as rejecting and responding with disappointment characterized 104 of the relationship episodes and over 90% of the subjects. Finally, they found that experiencing others as "liking me" was associated with a response of respecting in 18 episodes, and in 4 of 6 attempters and all 7 nonattempters.
Comment: Although the hypotheses were not supported, these authors replicated previously reported findings showing the most frequent wishes and responses, as described above. However, these authors took a deeper look at the entire range of relationship episodes and may be the first to explore the CCRT with sequential analysis at the level of the individual patient, which is the level at which Luborsky conceptualizes the CCRT. More research applying sequential analysis to CCRT data is needed.
Wilczek A, Weinryb RM, Barber JP, Gustavsson JP, Asberg M: The core conflictual relationship theme (CCRT) and psychopathology in patients selected for dynamic psychotherapy. Psychotherapy Research 2000; 10:100–113
Summary: The authors explored and failed to find support for four hypotheses related to the CCRT method for identifying central relationship patterns, using a naturalistic sample of 55 Swedish patients, 80% of whom were women. CCRTs were extracted from narratives told in a structured interview procedure. Good reliability was found across most measures, including CCRT components. The authors expected 1) CCRTs to differ systematically across DSM-III-R categories; 2) more pervasive CCRTs (as measured by the experience of similar wishes and responses across multiple relationships) to correlate positively with psychopathology; 3) negative responses from others and of the self to correlate positively with psychopathology; and 4) specified negative responses of the self to correlate with specific symptom reports. None of these hypotheses was supported with the exception of partial support for hypothesis 3: patients who felt a diminished sense of belonging and of being needed tended to expect others to respond negatively to them. The authors found what is becoming a familiar modal central relationship pattern: a wish to be close to others and to be accepted, a response by others of rejection, and a response of the self of depression. These were the most common wishes and responses across the sample, which included a significant number of individuals who did not meet criteria for any DSM-III-R diagnosis.
Comment: This is one of the few negative findings published on the CCRT method, but it raises significant concerns about the discriminant validity of the method, particularly when one considers the commonly held clinical observation that relationship patterns vary across patients with different personality organizations and diagnoses. The modal CCRT appears to be a general one across patients with differing Axis I and Axis II diagnoses. The authors question whether the standard CCRT categories are sufficiently varied and whether they might be improved if each wish and response were also rated for level of personality development. The authors note that a similar wish (e.g., to be close) would likely be expressed in significantly different ways in a neurotically organized person as compared with a personality-disordered or psychotic person. Differences of this type might also be detected if the structured interview process were altered such that patients were asked to relate stories about themselves "at your best" in relationships and "at your worst." The criterion of frequency of wishes and responses used for identifying the central relationship pattern might also be reconsidered. The frequency with which a wish or response is expressed in a set of narratives may not reflect its psychological significance for the individual in question.
Caspar F, Pessier J, Stuart J, Safran JD, Samstag LW, Guirguis M: One step further in assessing how interpretations influence the process of psychotherapy. Psychotherapy Research 2000; 10:309–320
Summary: The authors reanalyzed data from a 1986 study showing that therapist interventions compatible with a structured case formulation predicted deeper levels of patient experiencing and better outcome. The current study explored whether separating process and content aspects of an intervention adds to the understanding of how case-formulation-based interventions affect process and outcome in therapy. On the basis of psychotherapy transcripts, they found marginally acceptable levels of reliability in distinguishing process from content aspects, with intraclass correlation coefficients ranging from 0.48 to 0.64. Using analyses of three cases with varying outcomes, the authors showed that distinguishing between process and outcome aspects of interventions may increase the amount of explained variance in ratings of patients' depth of experiencing. When the compatibility of the intervention with the formulation differed between its process and content aspects, the aspect of compatibility that correlated with depth of experiencing varied across cases. That is, in one case, the content aspect appeared more important, whereas in another case, the process aspect was the more important. Case assessments were based on the Plan Formulation Method developed by Curtis and Silberschatz.
Comment: This is a creative study suggesting that how one makes an intervention can be as important as what one says. The "how" may need to be compatible with the formulation, just as the "what" may need to be. One hopes that future studies will demonstrate greater levels of reliability and will explore the process–content distinction in formulation-related interventions with a larger sample size.
Zuber I: Patients' own problem formulations and recommendations for psychotherapy. Journal of Psychotherapy Integration 2000; 10:403–414
Summary: Using a Swedish sample of 159 patients with at least moderate difficulties in global functioning, the author explored and found support for three hypotheses related to patients' characterizations of their problems and to treatment recommendations made by therapists. First, formal DSM-IV diagnosis was not weighted heavily by clinicians when recommending a treatment modality. Second, treatment recommendations could be predicted by the types of problems reported by the patients. Specifically, patient self-formulations that focused on symptoms were related to recommendations for cognitive or behavioral therapy. On the other hand, dynamic psychotherapy was recommended for patients who formulated their problems in terms of relationship difficulties. A third hypothesis was that the more verbally expressive patients were, the more likely they were to be recommended for insight-oriented psychotherapy. An additional finding was that problem formulations were related to DSM-IV Axis I diagnoses, but not Axis II diagnoses. Patients whose self-formulations focused on symptoms were more likely to be diagnosed with anxiety disorders, whereas those focusing on relationship problems were more likely to be diagnosed with an affective disorder. Also, patients reporting relationship problems were rated as functioning better than those reporting symptom formulations, as measured by the therapists' Global Assessment of Functioning ratings.
Comment: This study is more about therapists than about patients. It shows that the therapists in this sample were sensitive to the patients' problems when recommending modes of therapy; further, the therapists assumed that cognitive and behavioral treatments would be more effective for symptom amelioration than dynamic or expressive therapies, and that the opposite would be the case for relationship problems. As Zuber points out, the body of psychotherapy outcome research partially supports these differential recommendations, particularly with regard to the responsiveness of anxiety disorders to cognitive and behavioral treatments. The author also makes the excellent point that therapists can use these results to guide their focus in treatment. That is, carefully listening to patients' conceptualizations of their problems can help therapists decide what relative focus to give to symptoms and relationship problems.
Persons JB, Bertagnolli A: Inter-rater reliability of cognitive-behavioral case formulations of depression: a replication. Cognitive Therapy and Research 1999; 23:271–283
Summary: The authors tested how well a group of 47 mental health professionals could identify components of two cognitive-behavioral case formulations derived by a team of experts in the method. After viewing initial psychotherapy interviews and reading transcripts of those interviews, participants were asked to list the patients' problems and schemas. The therapist/subjects were participants in a day-long workshop on cognitive-behavioral case formulation and had completed 2½ hours of training prior to constructing their formulations for the study. To aid them in developing schemas of the patients, the study authors provided a multiple-choice questionnaire listing 15 adjectives describing the patient's view of self, other, and the world (e.g., self: defective, passive; others: unsupportive, helpful, dominating; world: bad, unpredictable, benevolent). The therapists were asked to rate the patients' degree of "belief" in the adjective on a scale from 0 to 10. Results showed that the clinicians identified 67% of the patients' overt problems. Interrater reliability coefficients averaged over sets of five judges averaged 0.72, which is considered good. Single judges showed poor reliability, producing interrater reliability coefficients averaging 0.37.
Comment: This is one of a handful of studies evaluating the reliability of a cognitive-behavioral case formulation method. The results are comparable to those found for dynamically based methods. The design is similar to those found in early research on dynamic case formulation methods, in that the basic test is how well clinicians can approximate a formulation constructed by an expert or team of experts in the method in question. This test has a lower threshold for reliability than would a test investigating whether individuals or teams working independently can construct a similar formulation based on the same clinical material.
|
CONCLUSION |
|---|
|
TopINTRODUCTIONABSTRACTSCONCLUSIONREFERENCES |
|---|
The individual practitioner who does not have the benefit of a group to aid in formulation still benefits from this case formulation research. First, using structured and systematic methods that include a fixed set of components should raise reliability above what it would be otherwise. Using these methods also helps organize clinical information, and increases the comprehensiveness and clarity of the formulation. Second, a focus on low-level inferences should increase reliability and serve to keep interventions focused more closely on what the patients says in therapy. Finally, some argue that it may not be necessary to have an accurate and comprehensive formulation initially. Instead, it may be more important to have an explicit working hypothesis that can be tested with the patient, then altered and improved as therapy proceeds.
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TopINTRODUCTIONABSTRACTSCONCLUSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ALL ISSUES | SEARCH | TABLE OF CONTENTS |
