مطالعه تجربی از واکنش ضدانتقال نسبت به بیماران مبتلا به اختلال شخصیت

Abstract Objective The study examined to what extent patients with cluster A + B personality disorders (PDs) evoked other countertransference reactions among psychotherapists compared with patients with cluster C PDs as well as the relationship between the different countertransference reactions and outcome. Methods A total of 11 therapists at the Department for Personality Psychiatry, Ullevaal University Hospital, Oslo, Norway, filled out the Feeling Word Checklist-58 (FWC-58), 2 weeks after admission and 2 weeks before discharge, for 71 patients admitted to the day treatment program. The patients were diagnosed with the Structured Clinical Interview for DSM-IV Axis II Disorders (SCID-II). Results The study revealed that patients with cluster A + B PDs evoked more negative and less positive countertransference reactions than those with cluster C PDs. The psychotherapists varied significantly more in their reported countertransference reactions toward patients with cluster A + B PDs than toward those with cluster C PDs. Patients who dropped out of treatment evoked significantly more negative countertransference reactions after 2 weeks than patients who completed the treatment. In addition, the study revealed strong correlations between countertransference feelings and change during the treatment. Conclusions This empirical study confirms clinical narratives on specified relationships between countertransference reactions, different PDs, and treatment course.

Introduction Countertransference reactions play an important role in psychotherapy. Although there are numerous definitions of countertransference, all emphasize the emotions evoked in therapists when they are working with patients. Freud introduced the concept and viewed it as therapists' unconscious reactions toward patients, triggered by the therapists' own neurotic conflicts [1]. With this narrow view, countertransference reactions were considered a hindrance to treatment. However, a different view, including all the emotional reactions therapists have toward their patients, was introduced in the 1950s [2], and it is today the most frequently used definition of the concept [3]. With this “totalistic” view, many theorists have described the relevance and importance of countertransference reactions in the therapist-patient relationship [4], [5] and [6]. First, it may give therapists important information about patients and be helpful in understanding the problems patients are struggling with [7], [8], [9], [10] and [11]. Second, it may affect the outcome directly through the presence, or lack of, empathic attunement. This means that disengagement may strengthen a feeling of worthlessness in the patient, whereas interest, on the other hand, may increase patients' sense of self-cohesion [12]. The literature on countertransference reactions evoked by patients with personality disorders (PDs) is mainly anecdotal. Moreover, the main focus has, to a large extent, been limited to countertransference reactions evoked by patients with borderline PDs (BPD). It is claimed that patients with BPD elicit more troublesome and problematic countertransference reactions than other patients, and these, in particular, are feelings of anger, fear, and hatred to the therapists [13] and [14]. Furthermore, BPD patients are said to split their therapists in “all good” or “all bad” and evoke feelings in the therapists of being either omnipotent or helpless, exhausted, and rejected [15] and [16]. There are few empirical studies that have examined countertransference reactions toward patients with different PDs. Two studies have used clinical vignettes to describe patients with BPD and examine the extent to which these patients elicit different countertransference reactions than other groups of patients. Brody and Farber [17] found that vignettes of patients with BPD evoked more negative countertransference reactions than vignettes of patients with depression or schizophrenia. Comparing vignettes of patients with major depression and BPD, McIntyre and Schwartz [18] showed that vignettes of BPD evoked more extreme reactions of hostility and dominance. To our knowledge, only 3 studies have examined therapists' countertransference reactions with regard to different PDs in a clinical setting. Holmqvist [19] studied staff members' countertransference reactions toward severely disturbed psychiatric patients in 17 different treatment homes. Holmqvist found no significant differences in countertransference reactions toward patients with different axis I and axis II diagnoses. Moreover, Holmqvist found that the patient's self-image was more important than their diagnosis in influencing staff members' countertransference reactions. This is in line with a study by Colson et al [20] who found that patients' diagnosis influenced staff members' countertransference reactions to only a small degree. On the other hand, patient behavior, such as suicidality, and character pathology lead to specific countertransference reactions among the staff members. However, Betan et al [21] found that different clusters of PDs were associated with different countertransference reactions. Furthermore, the study revealed that countertransference patterns were systematically related to patients' personality pathology across therapeutic approaches, suggesting that clinicians, regardless of therapeutic orientation, can make diagnostic and therapeutic use of their own responses to the patient. Thus, there are divergent data on whether different PDs elicit different countertransference reactions among therapists, and further empirical studies are needed. In this study, we examined the therapists' countertransference reactions in a specialized day treatment program for patients with PDs. Our main aims were to examine whether patients with cluster A + B PDs (mainly BPD) would elicit different countertransference reactions than patients with cluster C PDs (mainly avoidant PD), and whether the staff varied more in their reported countertransference reactions toward patients diagnosed with cluster A + B PDs than toward patients diagnosed with cluster C PDs. We also wanted to explore whether patients who dropped out of treatment (noncompleters) evoked different countertransference reactions 2 weeks after the start of treatment than patients who completed treatment (completers), and explore the relationship between different countertransference feelings and improvement during treatment.

Results The study revealed only one significant difference in countertransference reactions, between patients with cluster A + B PDs and patients with cluster C PDs 2 weeks after the start of treatment. The staff reported they felt more confident on the FWC-58 toward cluster C PDs than toward cluster A + B PDs (P = .029). However, several significant differences between the 2 groups were found 2 weeks before discharge. As displayed in Table 2, the psychotherapists reported feeling more confident and less rejected, on guard, overwhelmed, and inadequate toward patients with cluster C PDs than toward patients with cluster A + B PDs. Table 2. Differences between staff members' countertransference reactions toward cluster A + B and cluster C PDs 2 weeks before discharge FWC subscales Cluster n Mean SD t Significance levela 95% CI of the difference Important A + B 20 1.52 0.26 −0.87 .390 −0.23 to 0.09 C 10 1.59 0.17 Confident A + B 20 1.73 0.18 −2.24 .034 −0.24 to −0.01 C 10 1.86 0.12 Rejected A + B 20 0.19 0.15 2.88 .008 0.03 to 0.19 C 10 0.09 0.05 Bored A + B 20 0.32 0.29 0.50 .620 −0.15 to 0.24 C 10 0.28 0.22 On guard A + B 20 0.34 0.17 2.76 .010 0.04 to 0.24 C 10 0.21 0.10 Overwhelmed A + B 20 0.41 0.21 2.28 .030 0.01 to 0.26 C 10 0.27 0.12 Inadequate A + B 20 0.58 0.30 3.01 .005 0.08 to 0.39 C 10 0.35 0.11 CI indicates confidence interval. a Independent t test (2 tailed). Table options Table 3 displays the significant differences of the variances in the psychotherapists' countertransference reactions toward the same groups. The staff members disagreed significantly more on the cluster A + B PDs than on the cluster C PDs on being rejected and feeling inadequate. This means that the variance in the reported countertransference reactions from the staff varied more within cluster A+ B patients than within patients with cluster C PDs. The differences in variance between the 2 groups are also displayed in Table 2, which shows that the SDs for all 7 subscales are larger for patients with cluster A + B PDs. The probability for this finding given equal variance is 1/27 (P < .01). Table 3. Levene's test for equality of variances between cluster A + B and cluster C FWC-58 subscales F Significance Important 2.328 .138 Confident 2.392 .133 Rejected 8.537 .007 Bored 0.655 .425 On guard 3.385 .076 Overwhelmed 4.025 .055 Inadequate 5.909 .022 Table options Of the 71 patients included in the study, a total of 11 patients dropped out during treatment. Table 4 displays differences between the staff members' countertransference reactions 2 weeks after the start of treatment toward patients who completed the treatment (completers) and patients who dropped out during the treatment period (noncompleters). As displayed in Table 4, the psychotherapists reported feeling more overwhelmed and inadequate toward noncompleters than toward completers 2 weeks after the start of treatment. Table 4. Differences between completers of the treatment program and noncompleters 2 weeks after the start of treatment FWC-58 subscales Completers/noncompleters n Mean SD t Significance levela 95% CI of the difference Important C 60 1.24 0.26 1.31 .206 −0.06 to 0.24 NC 11 1.15 0.20 Confident C 60 1.59 0.19 1.33 .207 −0.06 to 0.25 NC 11 1.50 0.22 Rejected C 60 0.17 0.17 −0.49 .629 −0.14 to 0.09 NC 11 0.20 0.16 Bored C 60 0.28 0.23 −0.73 .475 −0.19 to 0.09 NC 11 0.33 0.20 On guard C 60 0.32 0.22 −1.20 .247 −0.16 to 0.04 NC 11 0.38 0.13 Overwhelmed C 60 0.28 0.16 −2.27 .042 −0.31 to 0.01 NC 11 0.44 0.22 Inadequate C 60 0.33 0.23 −2.34 .034 −0.33 to 0.01 NC 11 0.50 0.22 C indicate completers; NC, noncompleters. a Independent t test (2 tailed). Table options To correlate the degree of improvement on the GAF and the therapists' countertransference reactions at discharge, only patients with cluster A + B PDs or cluster C PDs were included (n = 30). As revealed in Table 5, the reported countertransference reactions were strongly correlated with the degree of patient improvement. Overall, improvement correlated positively with important and confident and negatively with rejected, bored, on guard, overwhelmed, and inadequate among the staff members. Table 5. Correlations between countertransference reactions and improvement on GAF (n = 30) FWC subscales Improvement on GAF-S Improvement on GAF-F Important 0.32 0.44⁎ Confident 0.54⁎⁎ 0.43⁎ Rejected −0.44⁎ −0.24 Bored −0.32 −0.38⁎ On guard −0.71⁎⁎ −0.48⁎⁎ Overwhelmed −0.51⁎⁎ −0.38⁎ Inadequate −0.69⁎⁎ −0.53⁎⁎ ⁎ P < 0.05 (2 tailed). ⁎⁎ P < 0.01 (2 tailed).