Borderline personality disorder affects 1% to 2% of the population, is present in approximately 10% of psychiatric outpatients, and is associated with significant medical comorbidity and societal economic costs (1, 2). The most empirically supported treatment for borderline personality disorder is dialectical behavior therapy (DBT) (3). However, anecdotal reports suggest that most psychiatry residents receive little formal training in DBT during their residencies and graduate with limited facility in the use of the technique. These factors likely contribute to the underutilization of DBT in community settings.
Limited access to DBT treatment for community patients with borderline personality disorder is a significant concern, as this form of psychotherapy has shown the greatest efficacy in treating some manifestations of the condition. To date, DBT treatment of borderline personality disorder has been compared with “treatment as usual” in seven published controlled studies that provided treatment for at least 12 weeks with 10 or more patients per treatment arm (4–10). The most consistent finding to emerge from these studies was the superiority of DBT in reducing parasuicidal behavior and suicidal ideation (4, 7, 9, 10). The least-powered of the published studies found significantly lower rates of suicidal ideation, but only a strong trend for reduced parasuicidality in the DBT-treated group (8). The only study finding no difference in parasuicide rates between DBT and treatment as usual employed an abbreviated version of DBT, both in terms of the number of sessions and array of skills taught (6).
Other aspects of borderline personality disorder have been found to respond to DBT, particularly increased treatment retention (4, 5, 9) and reduced frequency of inpatient hospitalization (4, 6, 7). There is also some evidence that DBT reduces levels of depression in patients with borderline personality disorder more than treatment as usual (7–9). Thus, several independent groups have now replicated the initial findings of Marsha Linehan’s studies, suggesting that the early findings supporting the efficacy of DBT are not simply the result of the treatment founder’s enthusiasm or interpersonal style.
This article presents the case of a patient with borderline personality disorder treated with DBT by one of us (B.S.) during her psychiatry residency. We employ the case report and surveys of residency programs to explore the challenges that arise in transferring principles of DBT to treatment of individuals during residency training.
The patient’s written and verbal consent were obtained for the case presentation. Self-report surveys (available upon request) regarding DBT exposure during residency training were sent to a total of 157 program directors of general psychiatry residency programs across the United States with available e-mail addresses on the Fellowship and Residency Electronic Interactive Database Access (FREIDA) system online. The surveys inquired about the attitudes and experiences of residency directors and senior residents towards DBT. The program directors were asked to forward an attached similar survey to either the chief resident or, if there was not one, a PGY-4 resident in their program.
“Mr. N” was a 41-year-old Caucasian homosexual man with a history of obsessive-compulsive disorder and depression. He had presented with the chief complaint of “I feel kind of desperate.” He had recurrent intrusive thoughts about going to “hell” as punishment for “being gay.” Mr. N also reported other ego-dystonic intrusive thoughts and compulsively performed religious acts in order to lessen his anxiety about going to hell. He also had been endorsing symptoms consistent with a major depressive episode for the previous 6 months. There was no history of mania or hypomania.
Exploration of more remote history revealed that the patient had a pervasive pattern of affective instability. Difficulty controlling his anger was an ongoing problem, even when he was not experiencing a sustained depressed mood. When distress felt intolerable to him, in addition to reacting with anger, he would binge eat, self-mutilate, and/or sleep excessively. These episodes typically lasted for several minutes to hours. He typically cut himself and threatened suicide when he wanted his partner to soothe him. Though he never made a serious attempt, Mr. N reported that he’d intermittently felt passively suicidal for years. He also reported chronic feelings of emptiness.
Mr. N had been hospitalized four times in 3 years. Psychiatrists previously treated him with numerous medications, including antidepressants of different classes, atypical antipsychotics, lithium, anticonvulsants, and alprazolam. Electroconvulsive therapy, the most recent treatment, produced no improvement. Mr. N had received psychotherapy in the past but always terminated it prematurely, believing that it was not helping him.
Family and Social History
Mr. N reported that his father had been physically and verbally abusive to him, particularly by ridiculing the patient when he expressed emotions. He described a trusting relationship with his partner, stating, “the greatest part of our relationship is that I can depend on my partner.” Mr. N was unemployed at initial presentation. He had a past history of alprazolam and marijuana abuse. At the time of presentation, he reported smoking marijuana approximately once a month and drinking two or three glasses of wine once or twice a week.
Mr. N was usually cooperative, albeit irritable at times. His psychomotor speed appeared mildly retarded and he had decreased prosody of speech and mild thought blocking, all of which improved over time. His mood was usually depressed but did vary over the course of treatment. His affect was typically congruent with mood. He reported intermittent passive suicidal thoughts but never had any active suicidal ideation. Obsessions were present, as previously described, and his judgment varied, becoming more self-destructive with increasing levels of distress.
We diagnosed Mr. N according to DSM-IV’s multiaxial assessment:
At initial presentation, Mr. N was taking clonazepam, which he continued at a final dose of 0.5 mg b.i.d. He was gradually switched from his other medications to a regimen of sertraline, 200 mg/day, and quetiapine, 300 mg at bedtime, for the depressive and obsessive symptoms. Because the OCD symptoms were the prominent issue during the initial evaluation, early sessions focused on exposure and response prevention (ERP). The patient did well initially but soon revealed he was engaging in compulsions and taking extra clonazepam to relieve his anxiety. Mr. N also reported increased anger outbursts when frustrated by the ERP exercises. When his partner told him he was “acting like a baby,” Mr. N intentionally applied a hot iron to himself. Symptoms of affective instability, self-mutilation, and passive suicidal ideation emerged quickly. ERP was discontinued as Mr. N’s safety could not be ensured. Gathering more extensive psychosocial history confirmed the diagnosis of borderline personality disorder.
Seven months after the ECT treatment, Mr. N and the resident decided to start DBT. He was given reading materials obtained from a Web site on DBT (www.behavioraltech.com). The DBT was composed of three parts: one 45-minute individual therapy session per week and one 45-minute session per week for skills training, both with Dr. Sharma, and one 45-minute session per week of supervision of Dr. Sharma by her DBT supervisor (R.B.). Mr. N was advised to call Dr. Sharma as often as he needed and was encouraged to call prior to (rather than after) hurting himself. In order to prevent “burn out” of a single therapist, Mr. N was instructed to talk to the on-call resident after 10 p.m. and on weekends. The on-call resident was given instructions by Dr. Sharma beforehand on how to handle the patient’s calls. If the patient were to call after he had hurt himself, he would be advised to go to the emergency room if medically necessary; otherwise, he and Dr. Sharma would discuss the incident at the next nonskills training session. This protocol was explicitly discussed with the patient at the start of DBT, and the hope in implementing it was to avoid reinforcing the patient’s parasuicidal behavior.
Crises in the patient’s life were discussed as they occurred and, as he learned new ways of coping through skills training, Dr. Sharma encouraged him to apply them to his daily life. In the skills training sessions as outlined in Linehan’s treatment manual (11), Dr. Sharma started with the overview, proceeded to the mindfulness module, and then began the interpersonal module. However, due to Mr. N’s difficulty coping with his life circumstances, she switched from the interpersonal module to the distress tolerance module (usually taught last). Consistent with the treatment manual (11), Mr. N was instructed to record his affect, suicidal ideation, parasuicidal behavior, and skills that he practiced on weekly diary cards. If Mr. N did not bring in the diary card, part of the session would be devoted to filling one out together to reinforce the importance of completing the diary cards.
Throughout the therapy, Dr. Sharma employed the dialectical stance of balancing validation/acceptance with promoting change. At the initiation of therapy, Mr. N called several times a day and engaged in self-cutting. Within 2 weeks of DBT, however, his mood improved, the parasuicidal behavior stopped, and the number of phone calls decreased. Furthermore, initially in response to distress, he would react impulsively, closing his line at work while customers were waiting (he had obtained a job as a grocery cashier) or suddenly quitting work and then regretting it. This behavior improved as he started to employ DBT skills, particularly mindfulness. A few months later, Mr. N changed jobs and went to work at a floral design business. Soon after, however, the self-cutting and phone calls to Dr. Sharma resumed; he ended up quitting this job in less than 2 weeks.
Over the following months, Mr. N started to improve again with a period of no parasuicidal behavior. He was compliant with appointments but only filled out his diary card once; much work within the sessions focused on eliminating such therapy-interfering behavior. During this period, Dr. Sharma also held family meetings to provide psycho-education and discuss the family members’ roles in relation to Mr. N’s behavior. The patient gradually expressed more motivation to change and for discontinuing his “sick role.”
In accord with the DBT model, Dr. Sharma completed behavior analyses and made recommendations regarding changing reinforcement contingencies. For example, Mr. N was asked to make a list of alternatives to cutting himself, for which he included eating donuts and sleeping. The patient was encouraged to eat a donut or take a nap whenever he felt like cutting himself. When unemployed and needing to look for a job, he expressed a desire to sleep all day instead; therefore, Dr. Sharma recommended that he call job contacts for 30 minutes and then take a nap as a reward. Again, the acceptance (taking a nap) and the change (calling job contacts) attempted to create a dialectical balance.
The patient started the new year relatively well but gradually worsened through the month of January. He slipped back into self-cutting, and his functioning deteriorated to the point that his partner took him to be hospitalized in late January. Despite sending letters to Mr. N, Dr. Sharma never received a reply, which ultimately resulted in the termination of his treatment.
Surveys were sent to the directors of 157 residency training programs. Responses were received from only 22 residency program directors and from the senior residents of only 13 programs, despite efforts to increase response rates through follow-up e-mail contact. From one program, several senior residents submitted completed surveys, from which the modal response was derived and reported. Appendix 1 presents the results of the survey. Thirty-six percent of the psychiatry residency programs have no lectures on DBT and 32% provide no DBT supervision. Furthermore, the majority of program directors surveyed (68%) believe that DBT exposure should be increased during training. Likewise, only 54% of the residents feel they have received sufficient DBT training.
This report focused on the use of DBT to treat borderline personality disorder, but other treatments for this disorder are available. The treatment of Mr. N was supervised by experts of various disciplines from the Department of Psychiatry at Emory University, providing the resident (B.S.) with different perspectives regarding treatment. The psychoanalytic supervisor primarily conceptualized the patient to be masochistic with self-directed rage. From a Kohutian perspective, she emphasized the importance of the therapist’s use of mirroring in the beginning of treatment with this patient. The psychopharmacologist (B.W.D.) focused on simplifying the medication regimen and questioned why the patient’s previously effective medication appeared to stop working, suspecting that psychological rather than neurobiological factors were contributing to the patient’s decline. The family therapist emphasized the contribution of the patient and partner roles (sick role and caretaker, respectively) to the current psychological state of the patient. All of the experts agreed that couples therapy would be beneficial. From the perspective of the resident, working with this patient resulted in one of the most valuable experiences of her training, in turn resulting in an in-depth knowledge about DBT, its practice and pitfalls. Discussing the case with the various supervisors challenged her ability to integrate the different schools of thought.
Mr. N’s homosexuality also challenged the resident’s skills for the treatment of sexual minorities. The therapist and her supervisors were mindful of evidence suggesting that borderline personality disorder may be overdiagnosed in men struggling with sexual identity, as well as the associated poorer therapist expectations for treatment (12). However, the chronicity of his symptoms and their presence in nonsexual aspects of his life warranted the diagnosis of borderline personality disorder.
Mr. N initially responded well to DBT but eventually discontinued treatment. It is difficult to say whether this resulted from the patient’s fear of improving and accepting the associated responsibility, the therapist’s inexperience, the DBT itself, or another factor. The patient’s history of childhood trauma may well have had an impact on his adult attachment style and increased the likelihood of early termination (13). Another consideration is that the DBT, for logistical reasons, was modified to include skills training in an individual therapy setting rather than the usual group therapy setting. If a residency insufficiently emphasizes DBT, with few residents employing the technique, there will be too few patients in DBT treatment to warrant the formation of a DBT skills training group, as was the situation in this case. Thus, unless DBT education reaches a certain level of intensity, either within the residency program itself or through collaboration with community DBT providers, residents’ exposure to DBT will be incomplete. Currently, there are no published studies that specifically examine whether group skills training sessions (standard DBT) are more effective than individual skills training sessions (this case). These types of studies are needed to determine appropriate limits in modifying standard DBT and to clarify how DBT may best be taught in residency training programs.
Although it is difficult to know the exact reason for why the patient in this case discontinued treatment, it is not an uncommon occurrence among patients with borderline personality disorder. Surveys of patients with borderline personality disorder or borderline personality organization have found that only 54% continued therapy beyond 6 months and only 33% completed treatment (14). Mr. N’s discontinuation of therapy simultaneously reinforces the limitations of DBT and the need for approaches that focus on increasing treatment retention.
The low response rate to the e-mailed surveys frustrated our intention to obtain a current measure of DBT education in residency, allowing no definitive conclusions to be drawn. It is possible that residency programs that returned the surveys have a greater focus on DBT and offer more DBT training compared with programs that chose not to respond, which would imply that the findings of the surveys represented an overestimation of the degree of DBT education occurring in psychiatry residencies for the nation as a whole. The majority of residents surveyed indicated uncertainty about whether they would use DBT after completing residency. This reluctance likely originates from residents’ low comfort levels in using DBT, which could be assessed through future investigations. Overall, the survey results suggest a need for more research and discussion about the role of DBT in psychiatry residency training.
DBT has been empirically proven to reduce parasuicide (4, 7, 9), decrease the need for hospitalization (4, 6, 7), and increase treatment retention (4, 5, 9) in individuals with borderline personality disorder. However, in a male patient as described here, with comorbidities of OCD and major depressive disorder, how transferable were the findings? What about the effects of medication and DBT together? Research exploring the efficacy of combined treatment with medication and DBT and the utility of DBT in treating patients with multiple comorbidities would be highly clinically relevant. A significant concern about DBT is that using a standard protocol is more useful for treating a diagnosis than a patient. Such protocols oversimplify the mental health practitioner’s attempt to treat an illness or behavior and may de-emphasize the importance of acknowledging each patient as a unique person. No form of therapy, including DBT, has magical potency, and the therapeutic value of a unique and trusting therapist-patient relationship should not be underestimated. Most practitioners with experience in treating patients with borderline personality disorder would agree that strength in this relationship is a major factor in promoting patients’ adherence to treatment. Though DBT has demonstrated efficacy for certain components of borderline personality disorder, it is not designed to address all aspects of the condition.
Despite its limitations, DBT has the most demonstrated efficacy in treating borderline personality disorder (over any other treatment) and, therefore, increasing residents’ knowledge and application of DBT during residency may substantially improve the quality of care of patients with borderline personality disorder (1). Numerous possibilities exist to incorporate DBT exposure in psychiatry residency programs. Potential options include: minimum requirements for didactic instruction in DBT or, as suggested by a survey respondent, incorporation of DBT into the cognitive behavior therapy curriculum; requirements for residents to conduct DBT by leading skills training groups and/or individual therapy in conjunction with a required number of supervision hours; reading requirements; electives; and clinical case conferences. For programs with little or no faculty experienced in DBT, Marsha Linehan’s Web site, www.behavioraltech.com, provides information on training opportunities. In considering making such changes to the curriculum, further studies are warranted to more precisely determine what end points are justified. Beyond these academic goals, the ultimate aim of these efforts would be to reduce the suffering of patients with borderline personality disorder and the burden they place on their families and society.