Personality disorders have often been relegated to stepchild status within psychiatry. Insurance and managed care companies may incorrectly assert that they are not treatable and, therefore, that treatment of these patients is not reimbursable. Psychiatrists themselves often confine their diagnoses to axis I syndromes. Research dollars for randomized, controlled trials of personality disorder treatments have been hard to come by.
A quarter-century after the creation of the DSM axis II, however, personality disorders have come of age. They have their own international organization devoted to studying them, and treatments of proven efficacy have been developed (1,2). A respected personality disorders journal has been in press for nearly two decades. Intellectual ferment has never been more active in the personality disorders field.
Three articles in this issue of the Journal reflect this ferment and contribute to the ongoing dialogue about the future direction of personality disorders, especially in light of the anticipation of major changes in DSM-V. McGlashan and colleagues provide yet another significant contribution from the Collaborative Longitudinal Personality Disorders Study on the fate of four DSM-IV personality disorders: borderline, schizotypal, avoidant, and obsessive-compulsive. In 24-month blind follow-up assessments, the investigators were able to identify certain traits that were relatively fixed, whereas other criteria appeared to be more reactive and behavioral. In borderline personality disorder, for example, the authors suggested that the more stable criteria, such as anger and impulsivity, may represent the biogenetic core of borderline personality disorder, and the identification of these features may help with the modification of diagnostic criteria in future renditions of the DSM. They also speculated that the least stable criteria, such as self-injury and abandonment concerns, may be better targets for psychosocial interventions, while the core biological criteria may be the best targets for biological treatments. These suggestions have to be considered tentative, however, because, despite the authors’ efforts to take treatment into account (3), the details of which patients received which treatments were not specified.
Zittel Conklin and Westen provide another type of data about borderline personality disorder. In a continuation of previous work (4–6), these investigators sought to characterize borderline personality disorder patients in the community, compared to those who are studied in academic centers. Using the Q-sort method of providing personality descriptions, they found that borderline personality disorder patients seen in everyday practice appear to have more distress and emotional dysregulation than what is captured by the DSM-IV criteria. The two items most descriptive of the borderline personality disorder patients in their study were "tends to feel unhappy, depressed, or despondent" and "emotions tend to spiral out of control."
These findings will resonate with psychiatrists who attempt to treat this group of patients. Often people with borderline personality disorder are dismissed as "manipulators" or regarded pejoratively as "splitters." What these findings underscore is that these people are in pain and feel that they are at the mercy of a maelstrom at the core of their being. Clinicians must be trained to recognize this pain and to get beyond the negative and alienating features of borderline personality disorder patients in order to endure the emotional roller coaster ride that often accompanies the treatment.
The investigators emphasize the value of data provided by experienced clinical observers who see a patient over time. Research instruments that assess an individual at one snapshot in time are fraught with problems in the assessment of personality disorders (7). Patients with borderline personality disorder may be kaleidoscopically different from one week to the next based on their affective state and the vicissitudes of their object relationships (8).
A classic New Yorker cartoon from the early 1960s depicts a peacock with its spectacular tail in full splendor saying to a smaller bird with no tail whatsoever, "Now let’s talk about you." The humor in the cartoon derives from the fact that every reader knows how it feels to be on the receiving end of a narcissistic display of self-importance. Indeed, in the third contribution on personality disorders featured in this issue of the Journal, Betan et al. report an empirically based description of countertransference responses to narcissistic patients that strongly resembles theoretical and clinical accounts. In a random sample of 181 psychiatrists and clinical psychologists from North America, the investigators tested a new questionnaire and found that it yielded eight clinically and conceptually coherent factors that were independent of the clinicians’ theoretical orientation. As one might anticipate, the eight factors were associated in predictable ways with axis II pathology. As part of their data analysis, they created a composite description of countertransference patterns in the treatment of patients who met the criteria for narcissistic personality disorder. They found that clinicians reported feeling resentment, anger, and dread in their interactions with the patient and tended to feel devalued and criticized by the patient. During their appointments with such patients, they felt distracted and avoidant and wished to end the treatment.
Clinicians have long known that patients with personality disorders re-create their characteristic mode of relatedness in their relationship with the clinician and impose a certain way of thinking, feeling, and reacting on the clinician. A problem for clinicians in systematically using this information diagnostically is that countertransference draws from the clinician’s own conflicts and past experiences as well as from the feelings induced by the patient (9,10). Nevertheless, what the data from this investigation illustrate is that there is an "average expectable countertransference" that may transcend the highly specific individual feelings brought to the clinical setting based on one’s own personal background. Professionals who work in group treatment settings, such as day treatment units or day hospitals, know that there are consistent reactions to certain types of patients, reflecting potential problems in the treatment.
A dilemma posed by the contemplation of including countertransference responses as an aid to the diagnosis of personality disorders, however, is that some forms of countertransference are largely unconscious. Often clinicians become aware of their feelings toward the patient only through small enactments, such as starting appointments late, getting sleepy to the point where the patient notices it, or making sarcastic comments in the guise of confrontation. Hence countertransference may be a discovery based on careful self-scrutiny that emerges in the course of the treatment.
Finally, if, as Betan et al. suggest, countertransference should be given the position of importance that it so richly deserves in the understanding and treatment of psychiatric patients, a formidable obstacle must be addressed. Self-reflection is no longer emphasized in residency training programs as it was in decades past. Trainees are not necessarily encouraged to have a personal psychotherapy experience to examine their own conflicts. Hence the realm of countertransference may be an unexplored continent. Psychiatry has long distinguished itself from other medical specialties by its attention to the clinician’s feelings as an important diagnostic and therapeutic tool in its armamentarium. To effectively treat patients with personality disorders, that tool must not disappear through disuse atrophy.
Address correspondence and reprint requests to Dr. Gabbard, Baylor College of Medicine, 6655 Travis, Suite 500, Houston, TX 77030; firstname.lastname@example.org (e-mail).