Patient personality characteristics and therapeutic integration: treating borderline personality and emotionally dysregulated-dysphoric personality features.

This study examines the relationship between patient personality characteristics and therapeutic integration. Within a sample of patients (N=93) receiving outpatient psychodynamically- oriented psychotherapy, we assessed patient Borderline and Emotionally Dysregulated personality features through the Shedler-Westen Assessment Procedure (SWAP-200), and therapeutic technique using the Comparative Psychotherapy Process Scale (CPPS) during an early treatment session. We examined personality dimensionally, psychotherapy interventions across different theoretical orientations, as well as psychotherapy integration. These analyses revealed an overlap between the Borderline Clinical Prototype and the Emotionally Dysregulated-Dysphoric Q-factor, with the former associated with higher use of integration and the latter associated with higher use of either psychodynamicinterpersonal or cognitive-behavioural interventions. Secondary analyses also indicated the greater presence of interventions oriented towards emotional exploration and to the didactic instruction of effective symptom coping techniques across both of these personality subtypes early in treatment. The key differences between these personality types, as well as the theoretical, empirical, and clinical implications of these findings are discussed.

Patient personality characteristics and therapeutic integration: treating borderline personality and emotionally dysregulated-dysphoric personality features

Patient personality features receive a great deal of attention within psychotherapy. Researchers and clinicians alike recognize the relevance of personality within interpersonal relationships (e.g., Hopwood, Wright, Ansell, & Pincus, 2013; Klimstra et al., 2013), intrapersonal perceptions (e.g., Luyckx, Teppers, Klimstra, & Rassart, 2014; Pulford & Sohal, 2006), and emotional expression (e.g., Izard, Libero, Putnam, & Haynes, 1993; Ng & Diener, 2009). As a result, the perceived influence of patient personality on psychotherapy process and outcome is not surprising. The relationship understood to exist between patient personality and psychotherapy is perhaps most readily evident through the extensive research investigating treatment efficacy for personality disorders (e.g., Dimaggio & Attina, 2012; Schut et al., 2005). These studies recognize that characterologically-entrenched challenges cannot be aided by a one-size fits all approach. This notion is further reflected in applied theory, with many theorists and clinicians espousing varied approaches to working with patients of differing personality constellations (e.g., McWilliams, 2011; Shapiro, 1965) during treatment. Furthermore, therapist emotional reactions and interactions to patients with various personality styles transcend theoretical orientation (Colli et al., 2014; Lingiardi et al., 2015). Thus, additional questions emerge regarding the possible interaction between patient personality and psychotherapy techniques given these findings.

Among personality clusters and disorders, borderline personality constellations appear to elicit increased negative emotional reactions from therapists (Liebman & Burnette, 2013). Contributing factors may include elevated clinical presentation (Gross et al., 2002; Paris, 2010; Zimmerman, Chelminski, & Young, 2008), increased clinical risk (Black, Blum, Pfohl, & Hale, 2004; Soloff, Lynch, Kelly, Malone, & Mann, 2000), and heightened treatment challenges (Biskin, 2015; Howe, 2013). Furthermore, these difficulties can be understood within the larger context of maladaptive interpersonal patterns, which typically characterize individuals carrying this diagnosis (American Psychiatric Association, 2013). Current approaches to treating borderline personality disorder (BPD) include dialectical- behavioural therapy (DBT), cognitive behavior therapy, and psychodynamic psychotherapy (Linehan, Dimeff, Koerner, & Miga, 2014; Liechsenring & Leibing, 2003; Reeves-Dudley, 2017). This foregoing research reflects the multitude of perspectives proposing important connections between patient personality characteristics, specifically among individuals diagnosed with BPD, and treatment interventions.

These perspectives not only demonstrate the importance of personality on the course of treatment, but they also highlight the growing attention placed on therapeutic technique. While there are many theoretical orientations to psychotherapy, several focus significantly on psychodynamic-interpersonal and cognitive-behavioural approaches to treatment. Although areas of difference and overlap exist between these two approaches, empirical evidence lends continued support for the respective efficacy of both treatment modalities (e.g., Storebo et al., 2020; Cristea et al., 2017; Finch et al., 2019; Keefe et al., 2019).

Given the prevalence of comparative research in both psychodynamic-interpersonal and cognitive-behavioural treatment modalities, the accompanying increase in integration of theory, practice, and research is understandable. Currently, psychotherapy integration appears to be in widespread use, with research demonstrating that a majority of therapists report regularly utilizing techniques outside of their identified orientation (Thoma & Cecero, 2009). Researchers have further proposed that thoughtful integration may better suit the specific needs of each individual patient (Ablon & Jones, 1998). Nevertheless, most experts agree that there is no single road to integration; rather, there are varied theories and frameworks from which to choose, with current research just beginning to explore the array of potential opportunities (Wachtel, 2010). Gold and Stricker (2001; Stricker & Gold, 1996) proposed one approach to integration characterized by the use of cognitive, behavioural, experiential, and other techniques in a relational, psychodynamic model. They theorized that this approach might facilitate a deeper experience of personal growth, while simultaneously providing didactic structure for developing effective problem solving strategies and for achieving behavior change. Researchers have further demonstrated that this approach to integration facilitates strong therapeutic alliance (Goldman, Hilsenroth, Owen, & Gold, 2013) examining the intersection of these domains. While previous researchers have investigated the relationship between therapist personality and associated technique (Scandell, Wlazelek, & Scandell, 1997), and patient personality and psychotherapy outcome (Mulder, 2011), the relationship between patient personality and therapeutic process warrants further exploration. Additional questions remain regarding the relationship between patient personality (as conceptualized within a dimensional frame), therapist technique (considering specific interventions, different theoretical frameworks) and integrative approaches.

The present paper seeks to address current gaps in understanding regarding the relationship between patient personality and therapeutic integration, focusing on borderline personality and emotionally dysregulated-dysphoric personality features. Specifically, this research seeks to expand understanding of the interaction between borderline and emotionally dysregulated features and specific psychotherapy interventions across different theoretical orientations, as well as psychotherapy integration.

Materials and methods

Participants

All participants in this programmatic study (N=93) were patients admitted to psychodynamic psychotherapy treatment at a university-based community outpatient clinic (for greater methodological details please see Hilsenroth, 2007). Cases were assigned to treatment practitioners and clinicians in an ecologically valid manner (i.e., based on clinician availability, etc.), regardless of disorder or comorbidity. In this sample of 93 individuals, 65 patients were female, and 28 were male. The mean age was 30.1 years (SD=11.6). Table 1 displays the demographic information as well as the distribution of patients’ primary Axis I and II diagnoses for the entire sample in accordance with the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV; American Psychiatric Association, 1994). Independent clinical ratings of DSM diagnoses for depression, anxiety and Axis II disorders for this sample have demonstrated excellent interrater reliability (Cersosimo & Hilsenroth, 2021; Hilsenroth et al., 2004; Hilsenroth, 2007, Katz et al., 2019; Slavin-Mulford, Hilsenroth, Weinberger, & Gold, 2011; Stein, Pinsker-Aspen, & Hilsenroth, 2007). This sample consisted of primarily mood-disordered patients with relational problems manifested in either Axis II personality disorders or subclinical traits/features of Axis II personality disorders (Cluster A, N=8; Cluster B, N=38; Cluster C, N=28). After a description of programmatic research examining the process and outcome of psychotherapy was provided to the subjects, written informed consent was obtained.

Table 1.

Demographic information of sample (N=93) variable.

Gender
Female	65	(70%)
Male	28	(30%)
Mean age (SD)	30.1	(11.6)
Marital status divorced	14	(15%)
Married	21	(23%)
Single	57	(61%)
Widowed	1	(1%)
Primary axis I diagnosis adjustment disorder	12	(13%)
Anxiety disorder	11	(12%)
Eating disorder	3	(3%)
Impulse disorder	1	(1%)
Mood disorder	50	(54%)
Substance-related disorder	1	(1%)
V code relational problems	15	(16%)
Axis II diagnosis	52	(56%)
Axis II trait/features	22	(24%)
Pretreatment psychiatric severity mean BSI-GSI (t-score; SD)	66.5	(7.9)
Mean GAF (SD)	60.3	(5.9)

BSI-GSI, brief symptom inventory-global severity index; GAF, global assessment of functioning.

Therapists

Clinicians in the study were 29 advanced doctoral students (14 men and 15 women) enrolled in an American Psychological Association-approved clinical Ph.D. program. Each clinician received a minimum of 3.5 hours of supervision per week (1.5 hours of individual supervision and 2 hours of group supervision) on the therapeutic model of assessment (TMA; Finn & Tonsager, 1997; Hilsenroth, 2007), clinical interventions, organization of collaborative feedback, psychodynamic therapy, and review of videotaped case material. Individual and group supervisions focused heavily on the review of the videotaped case material and technical interventions. Each participating clinician saw an average of three patients. All clinicians were trained in psychodynamic psychotherapy through guidelines delineated by Book (1998), Luborsky (1984), McCullough et al. (2003) and Wachtel (1993), and through readings on psychological assessment, psychodynamic theory and psychodynamic psychotherapy (for a more detailed description of this training process, see Hilsenroth, Kivlighan & Slavin-Mulford, 2015).

Therapeutic model/treatment

Patients first received a psychological evaluation from the TMA (Finn & Tonsager, 1997; Hilsenroth, 2007), which utilized a multi-method assessment and consisted of four steps, including three meetings between the patient and the clinician totalling approximately 4.5 hours, and one patient appointment to complete a battery of self-report measures. The three meetings included: i) a semistructured diagnostic interview (Westen & Muderrisoglu, 2003; Westen & Muderrisoglu, 2006); ii) an interview follow- up; and iii) a collaborative feedback session. The clinician and the patient also worked together to develop treatment goals and negotiate explicit treatment frames (i.e., scheduling session times, the frequency of treatment sessions, missed sessions and payment plan). In all cases, the clinician who carried out the psychological assessment was also the clinician who conducted the formal psychotherapy sessions.

Individual psychotherapy consisted of once or twice weekly sessions organized, aided, and informed (but not prescribed) by a short-term psychodynamic psychotherapy treatment model that included (Blagys & Hilsenroth, 2000): i) focus on affect and the expression of emotion; ii) exploration of attempts to avoid topics or engage in activities that may hinder the progress of therapy; iii) identification of patterns in actions, thoughts, feelings, experiences and relationships; iv) emphasis on past experiences; v) focus on interpersonal experiences; vi) emphasis on the therapeutic relationship; and vii) exploration of wishes, dreams or fantasies. In addition to these areas of treatment focus, relational patterns, case presentations, and symptoms were conceptualized in the context of cyclical patterns (Book, 1998; Leichsenring & Salzer, 2014; Luborsky, 1984; McCullough et al., 2003; Wachtel, 1993). Safran and Muran’s (2000) model of intervention was used for identifying and repairing treatment ruptures as they occurred in the therapeutic relationship. Treatment was open-ended in length rather than of a fixed duration (number of sessions, M=26, SD=22). Whenever a termination date was set, this became a frequent area of clinical focus, as issues related to the termination are often linked to key interpersonal, affective and thought patterns prominent in that patient’s treatment.

Treatment goals were first explored during the assessment feedback session, and a formal treatment plan was reviewed with each patient early in the course of psychotherapy that was subsequently reviewed at regular intervals for changes, additions or deletions. Reassessment of patient functioning on a standard battery of outcome measures, as well as process ratings, were completed by patients and therapists immediately after selected sessions prior to these review points. Patients were informed both verbally and in writing that their therapist would not have access to their responses on any psychotherapy process measure. Additionally, all sessions were videotaped (not just the sessions in which reassessment ratings were completed). Independent technique ratings for this study were collected for the third session of treatment (post-TMA assessment). We used this early treatment session because it is a standard process assessment point in our programmatic study of psychodynamic psychotherapy.

Measures

The Shedler-Westen Assessment Procedure. The therapists used the SWAP-200 (Shedler & Westen, 1998, 2004a, 2004b; Westen & Shedler, 1999a, 1999b) to describe their patients after completing the therapeutic assessment and the first two therapy sessions (approximately 5-6 contact hours). The SWAP-200 is a clinically derived, empirically based diagnostic measure that has been shown to have excellent retest reliability as well as good interrater, discriminant, and convergent validities with a range of external criteria (Diener & Hilsenroth, 2004; Shedler & Westen, 2004b; Smith, Hilsenroth, & Bornstein, 2009; Westen & Muderrisoglu, 2003). To complete the SWAP-200, the rater arranges a set of 200 personality descriptions into eight different categories ranging from 0 (irrelevant or inapplicable to the patient) to 7 (highly descriptive of the patient). The Q-Sort has a fixed distribution that mitigates measurement error. Correlation coefficients are then calculated to assess the match between the characteristics of a particular patient and the empirically derived, aggregate descriptions. These descriptions are presented in two diagnostic scales: Clinical Prototypes which aggregate clinicians’ SWAP descriptions of hypothetical patients, and Q-factors which derive from Q-analysis of clinicians’ description of actual patients who met diagnostic criteria for a DSM-IV PD. Several studies support the reliability and validity of the SWAP-200 in the diagnosis of PDs (Shedler & Westen, 1998; Westen & Shedler, 1999a, 1999b). Norms for the SWAP-200 were established from an outpatient sample of therapy patients diagnosed with a DSM-IV personality disorder (Westen & Shedler, 1999a). That is, the average patient diagnosed with a personality disorder would accordingly have a SWAP 200 t-score of 50, and a standard deviation of 10. Therefore, both sub-clinical and normative t-scores would be below T=50.

Comparative Psychotherapy Process Scale. The Comparative Psychotherapy Process Scale (CPPS; Hilsenroth, Blagys, Ackerman, Bonge, & Blais, 2005) is a brief, empirically derived, descriptive measure designed to assess therapist techniques that represent characteristic features of psychodynamic-interpersonal (PI; defined broadly to include psychodynamic, psychodynamic-interpersonal and interpersonal therapies) and cognitive-behavioural (CB; defined broadly to include cognitive, cognitive–behavioural and behavioural therapies) treatments. The CPPS consists of 20 techniques (randomly ordered), rated on a 7-point Likert scale ranging from 0 (not at all characteristic), to 6 (extremely characteristic). Ten statements are characteristic of PI interventions, and ten statements are characteristic of CB interventions. The reliability and clinical validity of the CPPS has been well established (Author et al., 2005), and the data we utilize in the current study is derived from that report, follows procedures detailed there, and is rated by trained coders who have demonstrated the ability to rate these techniques at the good (ICC 1 [0.60-0.74]; Fleiss, 1973) to excellent (ICC 1 [>0.75]; Fleiss, 1973) range. CPPS data utilized in the analyses in the present study were based on mean CPPS scores averaged across judges. For a subset of 20 patients used in the current study, the ICC was in the excellent range (CPPS-PI=0.86, CPPS-CB=0.78). Additionally, for the current sample, the mean CPPS-PI scale score for the rated session (3rd) was 3.32 (SD=0.74), and the mean CPPS-CB scale score was 1.26 (SD=0.56). When comparing the two scales, the degree of difference was found to be significant (degrees of freedom [df]=92, t=–21.5, P<0.0001).

Results

The mean t-scores and standard deviations of SWAP- 200 Q-Factors and Clinical Prototypes were first calculated for descriptive purposes. Following this, bivariate correlations were conducted to analyse the relationship between SWAP-200 Emotionally Dysregulated-Dysphoric Q-Factor and Borderline Clinical Prototype with in-session use of therapeutic interventions. Therapist effects in therapeutic intervention-use were then controlled for all variables approaching (P≤0.10) and reaching (P≤0.05) significance. Ad hoc analyses were subsequently conducted to provide a richer understanding into the nature of the relationships identified. Given our N of 93, our statistical power for the bivariate analyses within this study had a 100% chance of detecting a large effect size (Pearson’s r>0.50) (Cohen, 1988). As previously discussed the present study will only explore data related to the borderline and the emotionally dysregulated-dysphoric personality SWAP-200 subtypes.

Preliminary analyses

For descriptive analyses, mean t-scores and standard deviations are presented for SWAP-200 Q-Factors and Clinical Prototypes (Table 2). While we present descriptive data for all SWAP variables the present paper focuses only on findings related to the Borderline Clinical Prototype and the Emotionally Dysregulated-Dysphoric QFactor. Bivariate Pearson correlations (two- tailed) were then conduced to assess the relationships between SWAP- 200 Borderline Clinical Prototype and Emotionally Dysregulated-Dysphoric Q-Factor, with CPPS subscales (psychodynamic-interpersonal techniques [CPPS-PI] and cognitive-behavioural techniques [CPPS-CB]). A significant relationship was identified between the Emotionally Dysregulated-Dysphoric Q-Factor and CPPS-CB (r=0.27; P=0.01), as well as a trend towards significance between this Q-Factor with CPPS-PI (r=0.18; P=0.09). The Borderline Clinical Prototype was not significantly related to either the CPPS-CB (r=0.13; P=0.22) or the CPPS-PI (r=0.15; P=0.16).

Table 2.

SWAP-200 descriptive T-Score data.

SWAP-200 clinical prototypes	Mean t-score	SD
Paranoid	44.3	7.4
Schizoid	49.8	6.7
Schizotypal	48.4	6.3
Antisocial	45.6	5.3
Borderline	48.8	8.5
Histrionic	47.2	7.5
Narcissistic	44.5	6.2
Avoidant	51.8	7.1
Dependent	53.7	7.3
Obsessive	50.3	7.3
High functioning	55.8	7.0
SWAP-200 Q-factors	Mean t-score	SD
Dysphoric	53.7	6.6
Antisocial-psychopathic	46.3	5.1
Schizoid	48.4	6.7
Paranoid	44.5	8.0
Obsessional	55.3	7.3
Histrionic	50.8	7.6
Narcissistic	46.0	8.8
Avoidant-dysphoric	52.7	7.0
High functioning neurotic-dysphoric	56.8	6.2
Emotionally dysregulated-dysphoric	46.4	7.4
Dependent-masochistic-dysphoric	53.8	8.5
Hostile-externalizing dysphoric	47.0	7.4
High functioning	55.8	7.0

N=93; SWAP-200, Shedler-Westen assessment procedure.

MLM analyses

MLM analyses were employed to account for therapist effects in treatment outcome. This was conducted using Hierarchical Linear Modeling 7 software (Raudenbush, Bryk, Cheong, & Congdon, 2005). This was a particularly important aspect of our analyses, as different clinicians vary in their treatment delivery approach. In other words, patients who are treated by the same therapist may be more likely to receive similar types of interventions than patients treated by other therapists. As a result, by controlling for these effects in multilevel models, we ensured that the findings would not be confounded by the interdependency of the data (i.e., multiple patients being treated by the same therapist). Therapist effects (ICCs) were calculated for each dependent variable in the study by dividing the level-two variance by the total variance (level-one plus level-two variance). These ICC values identify the proportion of total variance in the variable that is attributable to therapists. The following ICC values were observed: CPPSPI= 0.38; CPPS-CB=0.07; PI x CB=0.17.

We then conducted MLM analyses of CPPS-PI, CPPSCB, and their interaction (PI x CB) in predicting the Borderline Clinical Prototype and the Emotionally Dysregulated-Dysphoric Q-Factor. When accounting for therapist effects and examining the relationship between patient personality characteristics and therapist technique, we conducted separate, one-level fixed intercept MLMs (e.g., patients nested within therapists). The level-one models were as follows:

where β0j is the intercept for client i treated by therapist j, β1j is the overall in-session use of PI for client i treated by therapist j, β2j is the overall in-session use of CB for client i treated by therapist j, β3j is the overall in-session use of integration (PI x CB) for client i treated by therapist j, and rij is the level 1 error. Of note, PI, CB, and PI x CB were entered into the model as uncentered, as they were already centred prior to the creation of the interaction effect. In instances where relationships approached (P≤0.10) and reached (P≤0.05) significance between the examined SWAP-200 Clinical Prototype and Q-Factor, we presented graphs to help visually clarify the different variables in relation to one another.

Borderline clinical prototype

When examining the relationship between the Borderline Clinical Prototype with therapist technique (PI, CB, and PI x CB), a significant positive relationship was identified between the Clinical Prototype and PI x CB (b=1.83, SE=0.83, t=2.18, P=0.03). In other words, even after controlling for therapist effects, patients with more prototypic borderline personality characteristics tended to receive significantly more integrated techniques in their early treatment than others. These findings can be found in Table 3. As seen in Figure 1, in the treatment of patients with higher elevations in the Borderline Clinical Prototype, therapists were significantly more likely to utilize both higher psychodynamic-interpersonal and cognitive- behavioural interventions, through integration.

Table 3.

MLM analysis of CPPS-PI, CPPS-CB, and their interaction in predicting SWAP-200 borderline clinical prototype.

Variable	Coefficient	SE	t-value	P-value
Intercept	48.89	0.90	54.09	<0.001
CPPS-PI	0.79	0.73	1.08	0.29
CPPS-CB	1.19	0.86	1.38	0.17
PI x CB	1.83	0.83	2.18	0.03*

N=93; SWAP-200, Shedler-Westen Assessment Procedure; CPPS-PI, Comparative Psychotherapy Process Scale - Psychodynamic-Interpersonal Subscale; CPPS-CB, Comparative Psychotherapy Process Scale - Cognitive-Behavioral Subscale; PI x CB, Comparative Psychotherapy Process Scale - Integrated CPPS Psychodynamic-Interpersonal and Cognitive-Behavioral Subscales. *P≤0.05; **P≤0.01; ***P≤0.001.

Figure 1.

Interaction between PI and CB techniques with borderline clinical prototype scores.

Emotionally dysregulated-dysphoric q-factor

When examining the relationship between the Emotionally Dysregulated-Dysphoric Q- Factor with therapist technique (PI, CB, PI x CB), a positive relationship trending towards significance was identified between the Q-Factor and CPPS-PI (b=1.08, SE=0.65, t=1.66, P=0.10), and a positive relationship reaching significance was identified between the Q-Factor and CPPSCB (b=2.03, SE=0.74, t=2.75, P=0.01). This finding reflects a trend in which patients with emotionally dysregulated- dysphoric personality characteristics tended to receive more psychodynamic-interpersonal interventions in early treatment than others. Additionally, even after controlling for therapist effects, these patients tended to receive significantly more cognitive-behavioural techniques in their early treatment. As emotionally dysregulated- dysphoric characteristics increased in patients, therapists were more likely to have higher use of either psychodynamic-interpersonal interventions or cognitivebehavioural interventions. These findings can be found in Table 4.

Table 4.

MLM Analysis of CPPS-PI, CPPS-CB, and their interaction in predicting SWAP- 200 emotionally dysregulated-dysphoric Q-Factor.

Variable	Coefficient	SE	t-value	P-value
Intercept	46.38	0.79	59.02	<0.001
CPPS-PI	1.08	0.65	1.66	0.10
CPPS-CB	2.03	0.74	2.75	0.01**
PI x CB	0.85	0.65	1.31	0.20

N=93; SWAP-200, Shedler-Westen Assessment Procedure; CPPS-PI, Comparative Psychotherapy Process Scale - Psychodynamic-Interpersonal Subscale; CPPS-CB, Comparative Psychotherapy Process Scale - Cognitive-Behavioral Subscale; PIxCB, Comparative Psychotherapy Process Scale - Integrated CPPS Psychodynamic-Interpersonal and Cognitive-Behavioral Subscales. *P≤0.05; **P≤0.01; ***P≤0.001.

Post hoc analyses

Given the above findings, we believed that it would be prudent to add post hoc analyses to understand the individual techniques (session 3) that may be in play during the session related to the observed relationships. As a result, we conducted Bivariate Pearson correlations (twotailed) to assess the relationship between the individual CPPS items composing the CPPS-PI and CPPS-CB subscales with the SWAP-200 Borderline Clinical Prototype and Emotionally Dysregulated-Dypshoric Q-Factor. Significant findings for specific interventions from the CPPSPI and CPPS-CB subscales with the two SWAP-200 personality scales can be found in Table 5.

Table 5.

Comparison of CPPS technique features associated with borderline clinical prototype (BC) and emotionally dysregulated- dysphoric Q-factor (EQ).

BC	EQ	Item # and Content (Scale)
Yes	Yes	1. The therapist encourages the exploration of feelings regarded by the patient as uncomfortable (PI)
r=0.22	r=0.27
P=0.03*	P=0.01**
Yes	Yes	7. The therapist focuses discussion on the relationship between the therapist and patient (PI)
r=0.22	r=0.19
P=0.04*	P=0.07
Yes	Yes	8. The therapist encourages the patient to experience and express feelings in the session (PI)
r=0.19	r=0.26
P=0.06	P=0.01**
Yes	Yes	18. The therapist teaches the patient specific techniques for coping with symptoms (CB)
r=0.20	r=0.23
P=0.05*	P=0.03*
Yes	Yes	20. The therapist interacts with the patient in a teacher-like (didactic) manner (CB)
r=0.23	r=0.34
P=0.02*	P<0.001**
Yes	No	10. The therapist addresses the patient’s avoidance of important topics and shifts in mood (PI)
r=0.18
P=0.09
No	Yes	3. The therapist actively initiates the topics of discussion and therapeutic activities (CB)
	r=0.17
	P=0.09
	Yes	16. The therapist allows the patient to initiate the discussion of significant issues, events, and experiences (PI)
	r= –0.18
	P=0.09

CPPS, Comparative Psychotherapy Process Scale; PI, psychodynamic-interpersonal; CB, cognitive-behavioural; items included when P≤0.10.

Within the CPPS-PI subscale, significant associations were identified between CPPS item #1 (‘The therapist encourages the exploration of feelings regarded by the patient as uncomfortable’) and both the Borderline Clinical Prototype (r=0.22; P=0.03) and the Emotionally Dysregulated-Dysphoric Q-Factor (r=0.27; P=0.01). Additionally, relationships both reaching and approaching significance were found between CPPS item #7 (‘The therapist focuses discussion on the relationship between the therapist and patient’) and both the Borderline Clinical Prototype (r=0.22; P=0.04) and the Emotionally Dysregulated-Dysphoric QFactor (r=0.19; P=0.07). Relationships respectively reaching and approaching significance were identified between CPPS item #8 (‘The therapist encourages the patient to experience and express feelings in the session’) and both the Borderline Clinical Prototype (r=0.19; P=0.06) and the Emotionally Dysregulated-Dysphoric Q-Factor (r=0.26; P=0.01). A trend towards significance was identified between CPPS item #10 (‘The therapist addresses the patient’s avoidance of important topics and shifts in mood’) and the Borderline Clinical Prototype (r=0.18; P=0.09). Lastly, a negative trend towards significance was identified between CPPS item #16 (‘The therapist allows the patient to initiate the discussion of significant issues, events, and experiences’) and the Emotionally Dysregulated-Dysphoric Q-Factor (r=-0.18; P=0.09).

Within the CPPS-CB subscale, a trend towards significance was identified between CPPS item #3 (‘The therapist actively initiates the topics of discussion and therapeutic activities’) and the Emotionally Dysregulated-Dysphoric Q-Factor (r=0.17; P=0.09). Additional relationships reaching significance were found between CPPS item #18 (‘The therapist teaches the patient specific techniques for coping with symptoms’) and both the Borderline Clinical Prototype (r=0.20; P=0.05) and the Emotionally Dysregulated-Dysphoric Q-Factor (r=0.23; P=0.03). Lastly, significant relationships were identified between CPPS item #20 [‘The therapist interacts with the patient in a teacher-like (didactic) manner] and both the Borderline Clinical Prototype (r=0.23; P=0.02) and the Emotionally Dysregulated-Dysphoric Q-Factor (r=0.34; P<0.001).

Discussion

The present study sought to address current gaps in understanding regarding the relationship between patient personality and therapy technique, with a specific focus on borderline and emotionally dysregulated personality features. Initial analyses within the present study provided descriptive information about the present sample. These analyses involved the calculation of mean t-scores and standard deviations for SWAP-200 Q-factors and Clinical Prototypes (Table 2). Clinical Prototype t-score means ranged from 44.3 to 55.8 (Borderline Clinical Prototype, t=48.8) while t-scores for SWAP-200 Q-factors ranged from 44.5 to 56.8 (Emotionally Dysregulated-Dysphoric Q-factor, t=46.4) that reflect clinical and subclinical ‘features’ of personality disorders on average, similar to the DSM-IV diagnoses observed the patients in this sample. Since norms for the SWAP-200 were established from an outpatient sample of therapy patients diagnosed with a DSM-IV personality disorder (Westen & Shedler, 1999a). The average patient diagnosed with a personality disorder would accordingly have a t-score of 50, with a standard deviation of 10. Considering this in relation to the present findings, we see that the present study’s sample is largely composed of patients with personality pathology of moderate severity and is thus consistent with the original SWAP-200 sample.

Additional analyses examined the relationships between SWAP-200 personality prototypes and clinical interventions, revealing relationships between the Borderline Clinical Prototype and Emotionally Dysregulated-Dysphoric Q-factor with therapeutic technique. Subsequent analyses demonstrated that these relationships remained significant even after accounting for therapist effects. When examining the relationship between the Borderline Clinical Prototype and therapist technique, while controlling for therapist effects, we found that therapists were significantly more likely to utilize the integration of more cognitive-behavioural interventions with more psychodynamic-interpersonal interventions. The same analysis with the Emotionally Dysregulated-Dysphoric Q-factor revealed that when working with patients with related personality features in psychodynamicallyoriented psychotherapy, therapists were more likely to have higher use of either psychodynamic-interpersonal interventions or cognitive-behavioural interventions. Given the high degree of overlap between the clinical features and presentation of the Borderline Clinical Prototype and the Emotionally Dysregulated-Dysphoric Q-factor, we thought it would be most informative to discuss and contextualize these findings together, while exploring their larger clinical implications.

As noted by Westen & Shedler (1999a), the initial development of the SWAP revealed differences between the DSM-IV (American Psychiatric Association, 1994) definition of BPD and their empirically derived portrayal of actual patients diagnosed with the personality disorder. The core features of the patients assessed were, as described by Westen & Shedler (1999a), ‘most distinguished by their intense, poorly modulated affect and, more generally, by their omnipresent dysphoria and desperate efforts to regulate it.’ As a result, the authors unsurprisingly noted the presence of overlap between the Emotionally Dysregulated-Dysphoric subfactor and related borderline features (Westen & Shedler, 1999b). Furthermore, additional research by Shedler & Westen (1998) demonstrated a strong correlation between the Emotionally Dysregulated Q-factor and the Borderline Clinical Prototype (r=0.61), as compared to a correlation not exceeding 0.06 with any of the other SWAP prototypes. To further understand the overlap between these two subtypes, we constructed a table comparing and contrasting the individual SWAP-200 items most strongly associated with both the Borderline Clinical Prototype and the Emotionally Dysregulated-Dysphoric Q-factor (Table 6).

Table 6.

Comparison of SWAP-200 items most strongly associated with borderline clinical prototype (BC) and emotionally dysregulated- dysphoric Q-factor (EQ).

BC	EQ	Item # and Content
Yes	Yes	12. Emotions tend to spiral out of control, leading to extremes of anxiety, sadness, rage, etc.
Yes	Yes	16. Tends to be angry or hostile (whether consciously or unconsciously).
Yes	Yes	77. Tends to be overly needy or dependent; requires excessive reassurance or approval.
Yes	Yes	90. Tends to feel empty or bored.
Yes	Yes	109. Tends to engage in self-mutilating behavior (e.g., self-cutting, self-burning, etc.).
Yes	Yes	117. Is unable to soothe or comfort self when distressed; requires involvement of another person to help regulate affect.
Yes	Yes	142. Tends to make repeated suicidal threats or gestures, either as a ‘cry for help’ or as an effort to manipulate others.
Yes	Yes	157. Tends to become irrational when strong emotions are stirred up; may show a noticeable decline from customary level of functioning.
Yes	Yes	191. Emotions tend to change rapidly and unpredictably.
Yes	No	11. Tends to become attached quickly or intensely; develops feelings, expectations, etc. that are not warranted by the history or context of the relationship.
Yes	No	15. Lacks a stable image of who s/he is or would like to become (e.g., attitudes, values, goals, and feelings about self may be unstable and changing).
Yes	No	45. Tends to idealize certain others in unrealistic ways; sees them as ‘all good,’ to the exclusion of commonplace human defects.
Yes	No	79. Tends to see certain others as ‘all bad,’ and loses the capacity to perceive any positive qualities the person may have.
Yes	No	98. Tends to fear s/he will be rejected or abandoned by those who are emotionally significant.
Yes	No	103. Tends to react to criticism with feelings of rage or humiliation.
Yes	No	134. Tends to act impulsively, without regard for consequences.
Yes	No	153. Interpersonal relationships tend to be unstable, chaotic, and rapidly changing.
Yes	No	154. Tends to elicit extreme reactions or stir up strong feelings in others.
Yes	No	171. Appears to fear being alone; may go to great lengths to avoid being alone.
Yes	No	185. Tends to express intense and inappropriate anger, out of proportion to the situation at hand.
No	Yes	50. Tends to feel life has no meaning.
No	Yes	54. Tends to feel s/he is inadequate, inferior, or a failure.
No	Yes	56. Appears to find little or no pleasure, satisfaction, or enjoyment in life’s activities.
No	Yes	73. Tends to ‘catastrophize’; is prone to see problems as disastrous, unsolvable, etc.
No	Yes	81. Repeatedly re-experiences or re-lives a past traumatic event (e.g., has intrusive memories or recurring dreams of the event; is startled or terrified by present events that resemble or symbolize the past event).
No	Yes	127. Tends to feel misunderstood, mistreated, or victimized.
No	Yes	149. Tends to feel like an outcast or outsider; feels as if s/he does not truly belong.
No	Yes	168. Struggles with genuine wishes to kill him/herself.
No	Yes	189. Tends to feel unhappy, depressed, or despondent.
No	Yes	195. Tends to be preoccupied with death and dying.

Morey (1991) developed a conceptualization of borderline features that looked at four specific facets, or subtypes: i) Affective Instability; ii) Identity Problems; iii) Negative Relationships; and iv) Self-Harm. These classifications provide better insight into the areas of overlap between the Borderline Clinical Prototype and the Emotionally Dysregulated-Dysphoric Q-factor. In examining the SWAP-200 items compared in Table 6, a trend appears to emerge; the items associated with both personality types appear to relate to affective instability and selfharm, the items associated with just the Borderline Clinical Prototype appear to also relate to negative relationships, and the items associated with just the Emotionally Dysregulated-Dysphoric Q-factor appear to also relate to identity problems.

The shared presence of affective instability and selfharm in both the Borderline Clinical Prototype and the Emotionally Dysregulated-Dysphoric Q-factor is not surprising given an abundance of related research supporting this association. Similarly, emotional dysregulation has been described as central to conceptualizations of both the development of, and challenges associated with, BPD (Speranza, 2013). In fact, ‘the reduction of ineffective action tendencies linked with dysregulated emotions’ has been highlighted as a core process of successful change among BPD individuals through DBT (Lynch, Chapman, Rosenthal, Kuo, & Linehan, 2006).

In addition, there is an experiential avoidance model of deliberate self-harm that posits that related acts occur as a means of escaping or avoiding undesirable or unmanageable emotional experiences (Chapman, Gratz, & Brown, 2006). Likewise, utilizing a similar conceptual model, the Emotional Cascade Model (proposing a relationship between emotional dysregulation and certain BPD behaviours), similarly identified the emotional relief associated with these acts among individuals carrying this diagnosis (Selby & Joiner, 2009). While self-harm is not unique to BPD, it does appear to occur with greater frequency and severity within this population (Turner et al., 2015). Additionally, researchers investigating non-suicidal self- injury found that, within their adolescent sample, while not all participants met diagnostic criteria for BPD, all participants did meet criteria for different personality disorders with dysregulated traits (Ferrara, Terrinoni, & Williams, 2012). Clearly, there is a strong precedent for including self-harm within the conceptualizations of both borderline and emotional dysregulation personality features.

There is empirical support for the inclusion of interpersonal difficulties, or negative relationships, as a key feature of the Borderline Clinical Prototype. Previous research has found that individuals with borderline features display higher emotional reactivity to social rejection, and that this reactivity often results in lower levels of social support (Zielinski & Veilleux, 2014). Further, both theory and research have lent support to the notion of rejection sensitivity playing a crucial role within borderline symptomatology (Rosenbach & Renneberg, 2014).

Researchers have identified related interpersonal difficulties, noting the presence of simultaneous desire and fear of establishing relationships, with greater overall relationship dissatisfaction, among those with BPD when compared to individuals without BPD (Drapeau & Perry, 2004).

Conversely, the inclusion of identity problems as a feature of the Emotionally Dysregulated-Dysphoric Qfactor but not of the Borderline Clinical Prototype may seem surprising, there is preliminary empirical support for this classification. Although identity disturbance has been included as a feature of BPD from an early conceptualization by Kernberg (1975) to the recent DSM-5 (American Psychiatric Association, 2013), researchers recently identified emotion dysregulation as a significant predictor of identity disturbance (Neacsiu, Herr, Fang, Rodriguez, & Rosenthal, 2015). Of note, this relationship remained significant even after controlling for depression, anxiety, and most importantly, BPD. This study concluded that future research should consider identity disturbance in relation to emotion regulation on a transdiagnostic level, lending additional support to the associated SWAP-200 classification.

There are important clinical implications to these findings. Patients with borderline personality features and with emotionally dysregulated-dysphoric features share many core characteristics and appear to elicit similar interventions within treatment. When treating patients with both personality types, therapists tend to be more active in early sessions than average. Of note, this increased involvement occurs in the form of significantly greater use of integration with the Borderline Clinical Prototype population, whereas this involvement occurs in the form of significantly greater use of either psychodynamic-interpersonal or cognitive-behavioural interventions with the Emotionally Dysregulated-Dysphoric Q-factor population. These findings may be reflective of clinical necessity; individuals with borderline personality constellation interventions to fully address these varied aspects of functioning. Similarly, individuals with emotionally dysregulated- dysphoric personality constellations displaying affective instability, self-harm, and identity problems may also require higher therapeutic involvement, but with initial emphasis rooted in either cognitive-behavioural or psychodynamic-interpersonal interventions, perhaps depending on which of these related features is being emphasized first within the treatment.

These findings further clarify previous research investigating the treatment of both BPD and emotion dysregulation. Regarding the Borderline Clinical Prototype findings, there have been additional proposals of approaching the treatment of borderline pathology via an integration of cognitive therapy and psychodynamic therapy, through the thoughtful merging of prevailing cognitive (e.g., Beck & Freeman, 1990) and psychodynamic (e.g., Kernberg, Selzer, Koenigsberg, Carr, & Applebaum, 1989) models of BPD treatment (Louw & Straker, 2002).

Similar variations of this notion have been proposed, all with the shared root of integration as an ideal treatment approach to BPD (e.g., Links, 2015; Merced, 2015; Paris, 2015). Additionally, a case study with inpatients diagnosed with BPD demonstrated the necessity of a more flexible, and at times more structured, treatment approach with this population (Goodman, Anderson, & Diener, 2014). However, although proposed, there is limited research on the implementation of integration with this population. Past research has demonstrated a positive relationship between the use of cognitive-behavioural techniques within psychodynamic psychotherapy and certain aspects of therapeutic alliance (Goldman, Hilsenroth, Owen, & Gold, 2013). Our findings further advance this body of research by demonstrating the tendency for patients with borderline features in a psychodynamicallyoriented treatment to receive significantly more integrated interventions in their early treatment than other patients.

In understanding the larger context of our Emotionally Dysregulated-Dysphoric Q-factor findings, there is more limited prior research available. This may be in part due to the more diffuse diagnostic nature of emotional dysregulation as compared to BPD, evidenced even through the absence of the former in the DSM-5 (American Psychiatric Association, 2013) and the presence of the latter. Of note, this has been a point of ongoing criticism, with researchers discussing the importance of including aspects of identity functioning (a characteristic that we associated with emotional dysregulation but not borderline features) within DSM conceptualizations of personality (Bender, Morey, & Skodol, 2011; Morey et al., 2011; Skodol et al., 2011). Furthermore, as previously discussed, the two constellations overlap in a number of important ways, and as such, treatment conceptualizations and approaches invariably share commonalities. Nevertheless, there are important points of divergence between these two constructs, and as such, emotional dysregulation warrants further research so that a more comprehensive understanding about effective treatment approaches can be obtained. Our findings lay an early framework through which this personality type can be understood, indicating that patients with emotional dysregulation personality features treated in psychodynamically- oriented psychotherapy tend to receive significantly more interventions of either psychodynamic-interpersonal orientation or cognitive-behavioural orientation within their early treatment.

Post hoc analyses were conducted to provide further insight into the specific types of related interventions utilized within the treatment of these patients. These analyses demonstrated elevated use of the specific psychodynamic-interpersonal interventions across both the Borderline Clinical Prototype and the Emotionally Dysregulated-Dysphoric Q-factor oriented towards affect, both within and outside of the therapeutic relationship. Also, within the CPPS-PI scale trends towards significance also emerged between the Borderline Clinical Prototype and the therapist addressing patient’s avoidance of important topics and shifts in mood, which may be reflective of both the affective instability and the negative relationships that characterize related features. Among the cognitive-behavioural interventions examined, these post hoc analyses demonstrated elevated use of interventions oriented towards building more effective coping skills during times of distress. Also within the CPPS-CB scale, a trend towards significance also emerged between the Emotionally Dysregulated-Dysphoric Q-factor and a more active therapist involvement with this population.

There is limited research investigating the specific interventions that are used most frequently in the treatment of these populations. Although much previous research has investigated the implementation of manualized treatment with diagnostically similar patients (particularly those diagnosed with BPD), less is known about treatment occurring in a naturalistic setting. Notably, Goodman, Edwards, and Chung (2015) completed an empirical clinical case study that looked at the psychotherapeutic processes and outcomes of five inpatients with BPD. This research is especially relevant to the current study, as these patients were engaged in psychodynamic psychotherapy. While clinicians were trained in a manualized psychodynamic therapy with a focus on transference interpretations, there were no adherence checks, enabling clinicians to operate more flexibly, similar to the present study. Of note, while the researchers used a different psychotherapy process scale (Psychotherapy Process Q-Set; PQS; Jones, 2000), they found that the most characteristic psychotherapy process items within the sample included the following: ‘Patient brings up significant issues and material’ (similar to CPPS-PI item #16, which approached significance for the Emotionally Dysregulated-Dysphoric Q-factor); ‘Therapist communicates with patient in a clear, coherent style’ (similar to CPPS-CB item #20, which reached significance for both variables); and ‘Patient experiences discomforting or painful affect’ (similar to CPPS-PI item #1, which reached significance for both variables). Additionally, among the interventions most strongly associated with overall decreases in patient distress levels were the following: ‘Therapist emphasizes patient’s feelings in order to help him/her experience them more deeply’ (similar to CPPS-PI items #1 and #8, significant across variables); and ‘Therapist actively exerts control over interaction’ (similar to CPPS-CB item #3, which approached significance for the Emotionally Dysregulated-Dysphoric Q-factor). Although there are important differences between both the setting (inpatient) and treatment (transference focused) across settings, these significant areas of overlap are important to consider within the treatment of patients with borderline and emotional dysregulation personality features.

In further attempting to contextualize these findings, we considered the wider application of the Emotionally Dysregulated Dysphoric Q-factor, beyond features shared with the Borderline Clinical Prototype. In doing so, overlap with eating disorders, an area with growing research devoted to psychotherapy process, emerged as an important consideration. Past research has consistently demonstrated a relationship between emotion dysregulation and disordered eating (e.g., Burns, Fischer, Jackson, & Harding, 2012; Lavender & Anderson, 2010; Lavender et al., 2014; Merwin et al., 2013; Racine & Wildes, 2013). Other research investigating the possible association between personality and eating disorders has found especially prevalent links between BPD and emotion dysregulation with bulimia nervosa and combined anorexia nervosa-bulimia nervosa (Westen, Thompson-Brenner, & Peart, 2006). Additional research has gone so far as to link selfinjury and disordered eating through emotion dysregulation, conceptualizing both as a bodily expression of internal emotion dysregulation (Muehlenkamp, Peat, Claes, & Smits, 2012). As a result, past research has led to the recommendation that clinicians address emotion dysregulation within the wider treatment for eating disorders and disordered eating (Lavender et al., 2014). Of particular note, this link between emotion dysregulation and eating disorders has been replicated with samples using the same SWAP-200 diagnostic criteria as that in the current study (Thompson-Brenner, Eddy, Satir, Boisseau, & Westen, 2008; Thompson-Brenner & Westen, 2005a; Westen & Harnden-Fischer, 2001). Given this strong connection, research investigating therapeutic process within an eating disordered population may relate to our current findings with the Emotionally Dysregulated Dysphoric Q-factor.

Keeping the above in mind, recent research by Colli and colleagues (2016) is especially relevant to the present study. These researchers utilized naturalistic data from both psychodynamic and cognitive-behavioural clinicians treating patients for eating disorders and analysed this data through both the SWAP-200 and a specialized version of the CPPS for eating disorders (Comparative Psychotherapy Process Scale-Bulimia Nervosa; CPPS-BN; Hilsenroth et al., 2005; Thompson-Brenner & Westen, 2005a). Of note, they found that the highest-rated interventions used by psychodynamically-oriented clinicians included: ‘Preferred that the patient, rather than the therapist, initiate the discussion of significant issues, events, and experiences’ (CPPS-PI item #16, which approached significance for the Emotionally Dysregulated-Dysphoric Q-factor); ‘Encouraged the patient to experience and express feelings in the session’ (CPPS-PI item #8, which approached significance for the Borderline Clinical Prototype and reached significance for the Emotionally Dysregulated-Dysphoric Q-factor); ‘Used the therapeutic relationship to offer the patient a different model for relationships than she had previously experienced’ (CPPS-PI item #7, which reached significance for the Borderline Clinical Prototype and approached significance for the Emotionally Dysregulated-Dysphoric Q-factor); and ‘Encouraged the awareness and exploration of feelings the patient found uncomfortable or unacceptable’ (CPPS-PI item #1, which reached significance for both variables). The researchers also investigated the intervention use of cognitive-behavioural therapists. Of additional note, the authors also found that the dysregulated/impulsive eating disorder personality subtype, measured by the SWAP-200 and resembling the Emotionally Dysregulated-Dysphoric Q-factor, significantly impacted the type of psychodynamic interventions received by patients.

Additionally, Thompson-Brenner & Westen (2005b) conducted a naturalistic study, investigating therapeutic interventions utilized within the treatment of bulimia nervosa. Similar to the present study and that of Colli and colleagues (2016), their research also utilized the SWAP- 200 and the CPPS-BN. These findings are especially relevant to the present study given their focus on specific differences in the treatment of patients with BPD comorbidity, with significantly higher usage found among certain interventions, including the following: ‘Addressed the patient’s avoidance of important topics and shifts in mood’ (CPPS-PI item #10, which approached significance for the Borderline Clinical Prototype); ‘Helped the patient regulate intense emotions (e.g., anger, fear, etc.)’ (similar to CPPS-PI item #1 and CPPS-CB item #18, which reached significance across variables); and ‘Taught the patient specific techniques for coping with her symptoms (again similar to CPPS-CB item #18, which reached significance across variables)’. These findings are important given the additional finding of dysregulation playing a significant role in negatively impacting treatment outcome within this population.

Research investigating the specific techniques used within the naturalistic treatment of patients with borderline and emotionally dysregulated-dysphoric personality features is relatively limited, despite the important clinical implications of related findings. While the exact nature of these differences needs to be further explored across different settings, the current findings provide important early impressions about these trends in treatment. Specific to the current study, the treatment of both borderline and emotionally dysregulated-dysphoric personality features appears to be characterized by greater use of interventions oriented towards emotional exploration and to the didactic instruction of effective symptom coping techniques. In working with patients with borderline and emotionally dysregulated-dysphoric personality features, therapists should also consider that these individuals might benefit from greater therapeutic involvement. There is some preliminary research indicating positive gains, both in terms of alliance and outcome, related to the use of these interventions with these two groups of patients. However, further research is needed to better clarify the nature of both orientation and specific interventions on the Borderline Clinical Prototype and the Emotionally Dysregulated-Dysphoric Q-factor.

Related to the limitations of this study and the need for continuing examination, this study utilized an outpatient sample with mild to moderate psychopathology, and it is difficult to say whether these findings would be replicated with an inpatient sample, or with more severe patients. We also utilized graduate trainee therapists who received regular videotaped supervision throughout the course of treatment with their patients. As a result, there is a need for replication of these findings with more experienced therapists. The present study also examined therapy conducted in a psychodynamically-oriented approach. Future research should attempt to extend findings in treatments utilizing other primary treatment modalities, such as cognitive-behavioural therapy. Additionally, in utilizing a naturalistic approach to treatment, the current study implemented greater flexibility in both patient selection and application of treatment protocols. These factors may have contributed to aspects of the treatment process to a minor degree. While this is important to consider, we feel that the benefits of generalizability afforded by this treatment approach outweigh the costs of this potential impact.

Conclusions

Numerous findings related to the treatment of patients with borderline and emotionally dysregulated personality features emerged from the present research, and these findings all have direct implications for clinical practice. These findings are especially important to consider given their utilization of prevailing approaches to psychotherapy, namely psychodynamic-interpersonal, cognitive-behavioural, and integrated techniques. We see that, when working with these patients, psychodynamically oriented clinicians tend to utilize significantly more integrated interventions (combining both psychodynamic-interpersonal and cognitivebehavioural interventions) early in treatment. Relatedly, in treating patients with personality features consistent with the SWAP-200 Emotionally Dysregulated-Dysphoric Qfactor, therapists tend to utilize either significantly more psychodynamic-interpersonal techniques or significantly more cognitive-behavioural techniques within early treatment. There are important similarities and differences between both these personality types and the treatment process trends that we found. In sum, a better understanding of patients’ personality characteristics appears to be useful in improving the impact of specific therapeutic interventions. And when these personality characteristics seem to fall outside the purview of the DSM-5, other models such as the Psychodynamic Diagnostic Manual (PDM-2; Lingiardi & McWilliams, 2017) diagnostic framework may better aid in planning clinically sophisticated interventions examining psychotherapy process and outcome (Hilsenroth, Katz & Tanzilli, 2018).

Across both personality types, we see a trend of overall increased therapeutic involvement, which is not surprising given the greater symptom severity that frequently characterizes related personality features. Post hoc analyses shed additional light on the nature of the interventions used most frequently across both populations. These include, across both the Borderline Clinical Prototype and the Emotionally Dysregulated-Dysphoric Q-factor, emphases on patients’ emotional experiences, teaching patients techniques for symptom management, and interacting with patients in a didactic manner. While the current study did not look at outcome, it is important to note that the overall study sample saw significant therapeutic gains over the course of treatment across a range of different outcomes and diagnostic groups (Hilsenroth, 2007; Hilsaenroth et al., 2007; Hilsenroth et al., 2018; Katz et al., 2019; Kuutmann & Hilsenroth, 2012; Levy, Hilsenroth, & Owen, 2015; Mullin, Hilsenroth, Gold, & Farber, 2017; Pitman, Slavin-Mulford, & Hilsenroth, 2014). As a result, we cannot reliably conclude that the intervention trends seen yielded therapeutic gains, although there is reason to believe that they are reflective of a successful attunement to the treatment needs of these patients. In treating patients with personality types resembling the Borderline Clinical Prototype and the Emotionally Dysregulated-Dysphoric Q-factor, clinicians should consider these findings when approaching treatment.