Associations between countertransference reactions towards patients with borderline personality disorder and therapist experience levels and mentalization ability

Abstract Objective This exploratory study locates countertransference as a pan-theoretical concept, comprising of thoughts, feelings, and behaviors expressed or experienced by therapists toward their patients. It aims to understand the patterns of countertransference experienced in working with borderline personality disorder. Associations between countertransference reactions and therapist-related variables of experience and mentalization ability are also examined. Method Psychotherapists (n = 117) completed the Therapist Response Questionnaire to assess patterns of countertransference experienced with a representative patient diagnosed with borderline personality disorder. They also completed a measure of mentalization ability that examined self-related mentalization, other-related mentalization, and motivation to mentalize. Results The profile of responses across eight countertransference dimensions is discussed, with the most strongly endorsed reactions being positive/satisfying, parental/protective, and helpless/inadequate. More experienced therapists reported less negative countertransference reactions in select dimensions. Therapists’ self-reported ability to reflect on and understand their own mental states was negatively correlated with a range of difficult countertransference experiences. There were few associations between their ability to make sense of others’ mental states, the motivation to mentalize, and the strength of their countertransference reactions. Conclusion The implications for countertransference management as well as therapist training and development are highlighted.


Introduction
The increasing focus on the relational aspects of the therapeutic encounter 1 has drawn attention to countertransferential reactions that may well be inevitable in the process of therapeutic work. 2 Countertransference, originally embedded in the psychoanalytic tradition as unconscious, conflict-based responses to the client's transference, 3 has undergone conceptual shifts and expansions. More recently, countertransference is seen as a pan-theoretical construct, relevant across theoretical orientations. 4,5 The contemporary expanded view of countertransference includes therapists' conscious and unconscious reactions to the client (sensory, affective, cognitive and behavioral).
These reactions could also be "induced" among psychotherapists as a response to the presenting concerns, personality characteristics, or interpersonal styles of clients, beyond the psychotherapists' own internal conflicts. 6 Theoretical literature and anecdotal accounts from clinicians suggest that working with certain client groups, e.g. borderline personality and antisocial personality, 7,8 evokes strong and distinct patterns of countertransference reactions. Borderline personality disorder is classified as a Cluster B personality disorder, along with antisocial, histrionic, and narcissistic personality disorders; all of which are marked by frequent dramatic, emotional, or erratic behaviour. 9 Individuals with BPD can be challenging; with frequent manifestations of emotion dysregulation, the use of primitive defenses like splitting, relational disruptions, and difficulties in maintaining boundaries in the therapeutic interaction. 10 Intense and difficult emotional reactions; anxiety, guilt, rage, helplessness, worthlessness, rescue fantasies, and even terror, have been reported in work with patients with borderline personality disorder, 11 who "often evoke a sense of walking on eggshells" among therapists. 10  Three research questions guided this study. The Therapist Response Questionnaire (TRQ) 18 This 79-item self-report measure assesses a wide spectrum of thoughts, feelings, and behaviors expressed or experienced by therapists toward their patients. These range from relatively specific feelings (e.g., "I feel bored in sessions with him/her") to complex constructs, such as projective identification (e.g., "More than with most patients, I feel like I've been pulled into things that I didn't realize until after the session was over").
Each item is individually rated using a five-point Likert scale (1 = not true at all, 3 = somehow true, and 5 = fairly true). The items are written in a way that they can be responded to by therapists of any theoretical orientation. 34 The factor structure of the TRQ comprises eight  35 In the present study, Cronbach's alpha for internal consistency ranged from acceptable (0.6 < α < 0.7) to good (0.7 < α < 0.9) and excellent (α > 0.9). 36 The estimates were: overwhelmed/ In the present study, each participant was asked to select their most recently-seen patient with borderline personality disorder and respond to the TRQ.
The Mentalization Scale (MentS) 37 This 28 item measure assesses mentalizing capacity, and has three sub-scales; self-related mentalization

Data analysis
Descriptive statistics (mean, SD, and frequencies) and inferential statistics were used to analyze data.
Bivariate correlations were computed to examine associations between countertransference dimensions (TRQ) and therapist experience levels and mentalization (MentS). T tests were used to assess differences in countertransference dimensions between therapists who received supervision and those who did not.

Results
The pattern and intensity of therapist countertransferential responses in their work with a representative patient with borderline personality disorder are reported in Table 1.
The results ( Table 2)  Additionally, therapists with more years of experience were less likely to give their patient special status or be over-concerned about them and over-involved with them (r = -0.25, p = 0.006). Therapists who had seen more clients with BPD over the course of their careers also reported less negative countertransference in similar domains and less disconnection or feelings of being distracted, withdrawn, annoyed, or bored in sessions (r = -0.23, p = 0.015).
The magnitudes of correlations were interpreted based on guidelines recommended by Gignac and Szodorai 38 as small (r = 0.10-1.9), medium (r = 2.0-2.9), or large (r ≥ 0.30). Most associations between therapist experience levels and countertransference reactions ranged between small and medium effects.
In an additional analysis (      to experience a sense of therapeutic optimism and a view of the patient as "special". 10 However, emotional responses that are "too positive", 5 or behaviors that reflect therapists' needs to protect and "rescue" their patients, 13 have also been recognized as problematic.
Positively-valenced feelings and behaviors can be associated with the risk of therapist over-involvement with the client and a loss of objectivity. Breivik et al. 39 speculated that therapists may report more positive feelings either due to defensive processes or lack of awareness of their negative feelings towards patients with personality disorders. This study has limitations arising from a possible sample selection bias; therapists who chose to respond to this survey may not form a representative group.
The use of self-report to assess complex processes such as countertransference and mentalization has potential shortcomings. There