There is little empirical research on patients at the severe level of personality organization who generally need more clarifications of roles, tasks, boundaries, and supportive interventions, and often create increased problems with countertransference. The purpose of this study was to investigate the relationship between clinicians' ratings of their new patients' level of personality organization traits (i.e.: Defensiveness, identity integration, object relations, and reality testing) and degree of expected countertransference and treatment interventions.
We surveyed 509 Clinicians about their reactions to a recently seen patient.
We found that the more severe the level of personality organization traits: 1. The more expected need for ongoing clarification of roles and tasks throughout the course of treatment; 2. The more clinicians' expectation of their patients being confused about boundaries; 3. The more expected negative countertransference; 4. The more expected need for supportive interventions.
These findings point out the importance for clinicians to accurately assess and understand their patients' level of personality organization in order to intervene appropriately regardless of theoretical orientation.
Personality organization, Clarification, Boundary, Countertransference, Supportive therapy
The level of patients' personality organization (neurotic, borderline or psychotic) may have a significant effect on the practitioners' degree of emotional reaction (i.e., countertransference) and use of clarifications of roles, tasks, boundaries and supportive techniques. Despite the diagnostic parsimony and treatment implications of personality organization (PO) as a category in the taxonomy of mental disturbances, it is not found in the Diagnostic and Statistic Manual  or the International Classification of Diseases . Personality organization is only addressed in the Psychodynamic Diagnostic Manual -1  and PDM-2 .
Although there has been much theoretical writings about patients at the borderline and psychotic levels of P.O. and their need for clarifications of roles, tasks, and boundaries, the increased problems with countertransference and need for more supportive interventions [5-10], there is little empirical research for this assumption.
Kernberg  identified four levels of personality organization. At the most pathological level, psychotic personality organization, reality testing is not intact, there is often delusional thinking and the person uses psychotic distortion and denial. The second level is borderline personality organization (BPO); (not identical to but incorporating borderline personality disorder) characterized by variable reality testing, primitive defense mechanisms, such as projective identification and splitting, identify diffusion and poor object relations. The third level is neurotic personality organization, characterized by intact reality testing, neurotic defense mechanisms such as repression, a well-integrated self and generally good object relations. The fourth level is the mature personality organization, characterized by good reality testing, mature defense mechanisms (such as humor, anticipation and sublimation), and well-integrated self and other representations.
There is a growing body of literature that suggests that PO is related to differences in treatment approach and outcome. For example, Koelen and colleagues  conducted a systematic review of 18 studies concerning the association between PO and psychotherapy response. The authors found that higher initial levels of PO are moderately to strongly associated with better treatment outcome.
Moreover, some studies indicate that level of PO may interact with the type of intervention (i.e., interpretive versus supportive ). Patients at the lower end of the PO continuum struggle with poor reality testing and require more use of clarifications than higher functioning individuals. They often benefit from the therapists' clarifications of roles, tasks, boundaries throughout treatment [6,8-10].
Further supporting this notion, the use of clarifications has been found to be helpful with borderline patients [14,15]. Horowitz  noted that traditional techniques in psychodynamic psychotherapy involve clarifications of reality leading to interpretations of self-defeating patterns. However, lower functioning patients may benefit from more frequent non-threatening clarifications. Horowitz stated that by clarification, the therapist is advancing toward interpretation, and the interpretations might compare and contrast dysfunctional beliefs of the patient with more realistic views.
Boundary issues are also a common problem with patients at the psychotic to borderline range of personality organization [5,6,8-10,16,17]. Auerbach and Blatt  wrote that the schizophrenic patient has extreme boundary problems as if a core self or basic self-other boundaries were lacking altogether. Additionally, in a qualitative study of therapist experiences with boundary challenges in clinical work  some therapists recalled experiencing boundary crossings from clients who were psychologically decompensating, which is consistent with lower levels of P.O. While the borderline level patients often have boundary problems, they have better boundary articulation than the schizophrenic patients.
Patients at the lower end of PO, regardless of the theoretical orientation of their treatment, generally require more supportive interventions. Supportive interventions include strengthening the therapeutic alliance, environmental interventions, education, advice and suggestion, encouragement and praise, limit setting and prohibitions, undermining maladaptive defenses while strengthening adaptive defenses, and emphasis on strengths and talents . Barber, et al.  wrote that supportive techniques are often used in different forms of psychotherapy or counseling, and Rockland  proposed that supportive interventions can be used with any form of treatment.
Kernberg  advocated that psychoanalytic institutes teach a broader range of psychoanalytic treatments that included the use of supportive interventions at a time when the use of supportive interventions were considered "un-analytic". However, a survey of 89 psychoanalysts (all members of the American Psychoanalytic Association or the International Psychoanalytical Association) about their own analysis, showed that the most beneficial analyses were associated with having a caring and emotionally engaged analyst who possessed positive relational and personality qualities, used supportive techniques in addition to classical techniques .
Luborsky  in his Principles of psychoanalytic psychotherapy: A manual for supportive-expressive treatment, viewed psychoanalytic treatment as a continuum between supportive and interpretive work based on the patients' level of functioning.
A number of studies have underscored the utility of supportive treatment for patients at more severe levels of PO. Abruzzi , for example, followed up on over 2,000 young people suffering from their first episode of acute psychosis who were treated with supportive psychotherapy. Over a 10-yr period, these people were seen and treated without chemicals, institutions, or restraints. Follow-up on 70-80% of the patients revealed results that compare favorably with drug and institutional approaches. Similarly, Rosenbaum and colleagues  compared psychodynamic psychotherapy for psychosis with standard treatment in patients with a first-episode schizophrenia spectrum disorder during the first two years of their study. They concluded that their study speaks in favor of including supportive psychodynamic psychotherapy in the treatment for patients with schizophrenic first-episode psychoses. Harder and colleagues  investigated the long-term outcome of supportive psychodynamic psychotherapy (SPP) in a prospective, longitudinal, comparative, multicenter investigation of successively referred patients diagnosed with first-episode schizophrenia spectrum disorder. Manualized SPP for up to 3 years as a supplement to standard treatment (ST) were compared to ST alone and followed up for 5 years (N = 269). They found that the combination of SPP + ST was significantly better than ST alone.
Jørgensen, et al.  found that both Mentalization based therapy (MBT) and supportive treatment are highly effective in treating borderline personality disorder (BPD) when conducted by a well-trained and experienced psychodynamic staff in a well-organized clinic. Newton-Howes and Wood  examined whether CBT reduces psychopathology in patients with schizophrenia more effectively than the use of non-cognitive psychotherapies. They did a systematic review and meta-analysis of the literature. They found that theoretically based CBT therapies might not outperform more accessible and simpler forms of supportive therapy.
Patients at the lower end of personality organization are often difficult and provocative. Supporting this notion, Gordon, et al.  and Spektor, Luu and Gordon  found that clinicians expected the greatest degree of countertransference reactions to patients at the lower levels of PO. Brody and Farber  surveyed 336 therapists and found that borderline patients evoked the greatest degree of anger and irritation and the least degree of liking, empathy, and nurturance. The authors also found that schizophrenic patients received the most complex mix of countertransference feelings, along with the highest perceived need for referral.
It is important to understand and address countertransference reactions with difficult patients, as studies have connected failure to manage countertransference reactions with ethical violations and treatment success. For example, Gutheil  reviewed types of boundary violations in working with difficult patients that led to civil lawsuits, complaints to licensing boards, and reports to professional societies and found that mismanaging countertransference was frequently associated with clinical transgressions. How a clinician deals with countertransference is not just an issue of risk-management; it can also affect the treatment process and outcome. To that effect, Rossberg, Karterud, Pedersen and Friis  found that symptom change was positively correlated with positive countertransference feelings and negatively correlated with negative countertransference (CT) feelings.
In ameta-analysis by Hayes, Gelso & Hummel , they noted that "Four conceptions of CT have emerged as the most prominent over the years: the classical, the totalistic, the complementary, and the relational" (p.88) and "In the professional literature and in everyday dialogue, these four conceptions ... are all used, sometimes interchangeably and in contradictory fashion. It is often unclear which of the four, or which combination, is intended at any given time..." (p.89). (For the purposes of this study, we use an over simplified definition of "countertransference reaction" as commonly used by ethics boards, licensing boards and malpractice courts. We will further clarify this in the procedure section). The authors found that managing countertransference successfully is related to better therapy outcomes.
Gordon and Stoffey  took 7 mental functions from the Psychodynamic Diagnostic Manual (Identity, Object Relations, Affect Tolerance, Affect Regulation, Super Ego Integration, Reality Testing, and Ego Resilience) and found with a step wise regression that: Affect Regulation (or level of defensive functioning) (β = 0.35, t(93) = 6.01, p < 0.001), Reality testing (β = 0.32, t(93) = 5.02, p < 0.001), Object Relations (β = 0.20, t(93) = 3.76, p < 0.001) and Identity (β = 0.19, t(93) = 2.69, p < 0.001) produced an R2 = 0.89, indicating that the four components accounted for 89% of the variance in Overall Personality Organization. We used these four components of Personality Organization (level of defense, reality testing, object relations and identity integration) in this research for a detailed analysis rather than using the more global categories of PO (Neurotic, Borderline and Psychotic).
1. Since the research suggests that lower functioning patients need ongoing clarifications, we hypothesized that the more severe the level of personality organization traits, the more need for ongoing clarifications throughout the course of treatment; 2. Since the research suggests that lower functioning patients tend to be more confused about boundaries, we hypothesized the more severe level of personality organization traits, the more practitioners' expectation of their patients being confused about boundaries; 3. Since the research suggests that lower functioning patient evoke a good deal of what is commonly consider by clinicians as CT, we hypothesized the more severe the level of personality organization traits, the more expected countertransference; 4. Since the research suggests that lower functioning patients may require more supportive interventions, we hypothesized the more severe the level of personality organization traits, the more expected use of supportive interventions.
The IRB of Chestnut Hill College determined that this project adequately protects the welfare, rights, and privacy of human subjects. The first author collected most the data from continuing education workshops on the topic of comparing diagnostic systems, ethical considerations and risk management. This was often a mandatory requirement of the state or agency that helped to produce a large sample of participants from the most common theoretical orientations. That is, the workshops were not biased toward a particular theoretical orientation. Participants were asked to assess a patient they have recently seen (within a week or so) for at least 3 sessions, who is 18 or older, and who was not actively psychotic or neurologically impaired at the time of treatment. Participants were provided with written definitions of the operationalized constructs used in the study. The PO components were defined on the Psychodiagnostic Chart (PDC) (see below). The Diagnostic Considerations Survey (DCS) constructs were already familiar to the participants (i.e., roles, tasks, boundaries and supportive therapy), but we provided a definition of "countertransference" which is often defined in differing and complex ways. We used a definition that would be accepted by clinicians of various theoretical orientations and more relevant to risk management than to any one type of therapy (see below). Participants first rated their patient on each of the 4 personality organization components (level of defense, level of identity integration, object relations, and level of reality testing) on a scale (1 = severe level and 10 = healthy level). Then they were asked to respond to the DCS, which inquired about likely treatment considerations.
Participants were not aware of the hypotheses of the investigation other than that the investigators wanted help in understanding diagnostic and ethical issues. Participation was voluntary. Of the workshop participants, 509 completed the ratings on the PDC; 508 completed all the questions including the scales for defensiveness; 509 completed all the questions including the scales for identity, object relations and reality testing. Four hundred and eighty clinicians completed the Diagnostic Considerations Survey (DCS). Forty-six percent held doctoral degrees, 67% female, the mean age was 43.6 (SD = 13.4), and their stated primary orientations were 26% Psychodynamic, 33% CBT and 41% Other (e.g., Family Systems, Humanistic/Existential, and Eclectic).
The Psychodiagnostic Chart (PDC)a: Gordon and Bornstein [34,35] developed the Psychodiagnostic Chart as a brief clinicians rating form based on the Adult section of the Psychodynamic Diagnostic Manual (PDM) and later with the PDM-2. Gordon and Stoffey  found good construct validity and good two-week test-retest reliability with the PDC. As a measure of scale internal consistency, Cronbach's alpha was calculated for the seven components of the Overall Personality Organization Scales. Coefficient alpha was 0.94, indicating a high degree of internal consistency among the scales; mean interscale correlation was 0.76. Retest reliabilities for the component scales of Overall Personality Organization ranged from 0.69 to 0.90 (all p's < 0.001) indicating moderate to high levels of stability across the 2-week interval. Retest reliability for the Overall Personality Organization Scale was 0.92 (p < 0.001), while retest reliability for Overall Severity of Personality Disorder ratings was 0.89, (p < 0.001).Retest reliabilities for the nine Mental Functioning scales of the PDC ranged from 0.77 to 0.89 (all p's < 0.001), while retest reliability for Severity of Symptoms ratings was 0.87 (p < 0.001). Gordon and Stoffey  also found good convergent and discriminant validity for the PDC scales with respect to scores on the MMPI-2 clinical scales, the OPD Axis IV Psychic Structure/Mental Functioning scales, and the KAPP. They had predicted significant negative correlations between the PDC Severity of Symptoms scale and scores from the MMPI-2 A scale (severity of symptoms), KAPP-18, and OPD-7, and a significant positive correlation with current GAF. All correlations were significant (p = 0.001) and in the predicted direction: MMPI-2 A (r = -0.46), K18 (r = -0.80), OPD-7 (r = -0.90), and GAF (r = 0.75). Gordon and Bornstein  found that the PDC and PDC-2 had excellent clinical utility and relations to external criteria.
The dimensions on the PDC are: Personality Organization (healthy, neurotic, borderline, psychotic), Personality Syndromes (e.g. Schizoid, Histrionic, Narcissistic, etc.), Mental Functioning (e.g. Capacity for Intimacy, Defensive Level, Self Observing Capacity, etc.), ICD or DSM symptom diagnoses (e.g. Mood Disorder, Anxiety Disorder, etc.) and Cultural/Contextual Dimension (e.g. Immigration trauma, divorce, etc.). In this study, we are using the four components used to compute the over-all level of Personality Organization: level of defense, level of identity integration, object relations, and level of reality testing. The clinician is asked to rate each PO component on a 1-10 scale, where 1-2 is at the psychotic level and 9-10 is at the healthy level of P.O. The P.O. component traits are defined on the PDC as:
1. Identity: Ability to view self in complex, stable, and accurate ways
2. Object Relations: Ability to maintain intimate, stable, and satisfying relationships
3. Level of Defenses: 1-2: Psychotic level (delusional projection, psychotic denial, psychotic distortion); 3-5: Borderline level (splitting, projective identification, idealization/devaluation, denial, acting out); 6-8: Neurotic level (repression, reaction formation, intellectualization, displacement, undoing); 9-10: Healthy level (anticipation, self-assertion, sublimation, suppression, altruism, and humor).
4. Reality Testing: Ability to appreciate conventional notions of what is realistic.
Diagnostic Considerations Survey (DCS): The Diagnostic Considerations Survey (DCS) is an ad hoc face valid survey. Clinicians rate each question on a 7 point Likert scale. The DCS questions assessed in this study are: 1. How important do you think it will be to have ongoing discussions about clarifying roles, tasks and boundaries with this patient throughout the course of treatment? (1 = Not at all important, 7 = Very important); 2. How much confusion do you expect this patient to have regarding boundaries in the professional relationship? (1 = None 7 = A Great Deal); 3. How strong a countertransference reaction might you expect to experience with this patient? (1 = None 7 = Very Strong); 4. How much more supportive treatment would you do with this patient as compared to other patients? (1 = Very Little 7 = A Great Deal).
The constructs such as roles, tasks, boundaries, and supportive treatment were common to clinicians, however the definition of CT required operationalization for the purposes of this study. The definition of CT used in this study is the same as we used in Gordon, et al. . We wrote that CT was operationally defined "...in simple language assessable to all theoretical orientations". Though we acknowledge that CT is complex, for the purposes of this research, CT is defined simply and limited to those practitioner reactions that are likely to interfere with treatment and possibly lead to ethical dilemmas and risk management problems. This simplified definition is often consistent with how countertransference is used in ethics education. It asks the practitioner, "Every therapist has at times problematic countertransference reactions (anger, fear, boredom, too much sexual attraction, frustration and dislike)". (p. 239-240).
1. How important do you think it will be to have ongoing discussions about clarifying roles, tasks and boundaries with this patient throughout the course of treatment?
We predicted the more severe the level of personality organization traits, the more need for ongoing clarifications throughout the course of treatment; Hypothesis 1 was supported. The patient's level of defense (r = -0.24), level of identity integration (r = -0.22), object relations (r = -0.25), and level of reality testing (r = -0.23) were all p < 0.0001.
2. How much confusion do you expect this patient to have regarding boundaries in the professional relationship? We predicted that the more severe the level of PO components, the more practitioners' expectation of their patients being confused about boundaries. Hypothesis 2 was supported. The patient's level of defense (r = -0.41), level of identity integration (r = -0.39), object relations (r = -0.39), and level of reality testing (r = -0.37) were all p < 0.0001.
3. How strong a countertransference reaction might you expect to experience with this patient? We predicted that the more severe the level of PO components should produce the most countertransference. Hypothesis 3 was supported. The patient's level of defense (r = -0.37), level of identity integration (r = -0.25), object relations (r = -0.25), and level of reality testing (r = -0.24) were all p < 0.0001.
4. How much more supportive treatment would you do with this patient as compared to other patients? We predicted that the more severe the level of PO components the more clinicians felt that they should use supportive therapy compared to their other patients. Hypothesis 4 was supported. The patient's level of defense (r = -0.35), level of identity integration (r = -0.30), object relations (r = -0.27), and level of reality testing (r = -0.21) were all p < 0.0001. (Table 1).
Table 1: Correlations between degree of defensiveness, identity diffusion, object relations, reality testing with countertransference and therapeutic interventions. View Table 1
The purpose of the present study was to better understand how practitioners' treatment considerations and countertransference expectations were related to their perceptions of their patients' levels of personality organization as measured by the four component traits (level of defense, level of identity integration, object relations, and level of reality testing). We found that a patient's personality organization component traits had a significant relationship to the practitioners' expected degree of countertransference and use of clarifications of roles, tasks, boundaries, as well as degree of supportive techniques. Specifically, the clinicians who participated in this study, who had a wide range of theoretical orientations, reported that they were more likely to review the ground rules about roles, tasks, and boundaries throughout the course of treatment with patients who had problems with primitive defenses, poor identity integration, poor object relations, and poor reality testing. Similarly, clinicians expected their patients to have more confusion about boundaries in relation to their degree of primitive defenses, poor identity integration, poor object relations, and poor reality testing. Therapists reported that they expected stronger countertransference reactions to patients at these lower levels of PO functioning. Therapists also reported that they were more likely to use supportive interventions with patients at more severe levels of these P.O. components.
These findings support the importance of explicitly discussing the ground rules for treatment and reviewing them in times of acting out and confusion . Work with patients who have less developed personality organizations requires that the therapist routinely monitors and provides clarification about the therapeutic process, as ruptures in the therapeutic alliance resulting from projections of hostility or paranoia can jeopardize the success of the treatment .
This study is based on responses of a large number of workshop participants from a wide range of theoretical orientations. Although the diversity of sampled clinicians is an important strength of the current investigation, the participants were recruited though workshops on diagnoses and ethics that were considered required CE credits by their state or by many agencies. As such, this was a sample of convenience rather than a truly random sample of clinicians. These findings need to be compared to other surveys, which use other sampling procedures. Eventually, these finding need to be validated with actual patients at different levels of personality organization and study the reactions and treatment recommendations of clinicians.
The current findings support the idea that the psychological constructs of clarification, boundary issues, countertransference and supportive techniques are relevant to any psychotherapy orientation. This research lends empirical support to the use of the four components of personality organization (defenses, identity integration, object relations, and reality testing) to the theoretical as well as the clinical significance of personality organization. These findings point out the importance for clinicians to accurately assess and understand their patients' level of personality organization in order to intervene appropriately. Diagnostic systems such as the PDM and PDM-2, as well as the PDC chart [34,35] provide a way to assist clinicians in identifying the PO level for their patients.