Psychiatric Annals

TREATMENTS REDUCING PERSONALITY DISORDERS 

Dialectical Behavior Therapy for Borderline Personality Disorder

Clive J Robins, PhD

Abstract

Borderline personality disorder (BPD) presents serious problems not only for the people who meet criteria for the disorder, but also for their friends and family and for physicians who treat them. According to the OSM-IV,1 approximately 10% of psychiatric outpatients and approximately 20% of psychiatric inpatients meet criteria for BPD and the diagnosis is associated with high rates of both psychiatric and medical service use. Furthermore, BPD diagnosis is associated with high rates of parasiticide,2 (ie, any intentional self-injury resulting in tissue damage or risk of death, ranging from high lethality, high-intent suicide attempts to minor self-injuries with no suicidal intent). The lifetime suicide rate has been estimated at approximately 10%, and double that among those with a history of self-injury.3

DIAGNOSTIC CRITERIA

For a DSM-IV diagnosis of BPD, an individual needs to demonstrate five or more of nine criteria. Dialectical behavior therapy (DBT)4 conceptually organizes these nine criteria into five broad areas in which the person's experience and behavior may be dysregulated.

Emotion Dysregulation. The DSM-ÍV describes a criterion of a marked reactivity and instability of mood. This often is superimposed on a backdrop of chronic dysphoria. The DSM-IV also specifically singles anger out from other emotions as intense, inappropriate, and difncuit for the BPD patient to control My own experience has been that borderline patients are at least as likely to be underexpressive of anger as they are to be overexpressive of it. Either way, or with wild swings in both directions, certainly most patients with BPD cannot express anger appropriately. Patients' expressions of anger are probably more aversive to clinicians than expressions of shame, fear, guilt, or sadness, and so perhaps more salient. However, DBT views borderline patients as having equivalent difficulties regulating all emotions, at least aversive ones. Indeed, emotion dysregulation is viewed as the core difficulty of borderline patients. Other criteria are viewed as primarily either expressions of emotion dysregulation or maladaptive efforts to regulate mood.

Relationship Dysregulation. The DSM-I V describes unstable, intense relationships characterized by idealization and devaluation, and frantic efforts to avoid abandonment. Patients with BPD frequently have experienced a more than average amount of actual abandonment, either in childhood, or at least in adulthood, often as a result of their own behaviors. Idealization and devaluation can be viewed as extreme cognitive styles made more likely by both the patient's desperate need for connectedness that often is lacking in his or her life and by the normal effects of strong emotion on cognition.

Self-Dysregulation. The DSM-IV describes an unstable sense of self and chronic feelings of emptiness. Frequent extremes of emotion and associated behaviors could be expected to lead to difficulties in establishing a coherent sense of one's self. In addition, frequent invalidation of one's behaviors, thoughts and feelings, which patients with BPD typically have experienced, can lead to self-invalidation of one's own preferences, goals, perceptions, etc.

Behavior Dysregulation. Trie DSM-IV describes both impulsive and potentially harmful behaviors generally (eg, substance abuse, reckless driving, or binge eating) and also suicidal and other self-injurious behaviors specifically. These behaviors may serve a variety of functions. For example, an instance of parasiticide may reflect high intent to die. Alternatively, the function may be primarily to communicate distress or elicit an interpersonal response. Most frequently, however, the primary function seems to be to escape or otherwise regulate aversive emotions. Suicide attempts typically have this escape function, and patients typically report that, after other types of self-injury, they feel less afraid, ashamed, angry, etc. In my view, clinicians too frequently assume that parasinoidal acts of patients with BPD primarily have a communicative function, which is reflected in…

Borderline personality disorder (BPD) presents serious problems not only for the people who meet criteria for the disorder, but also for their friends and family and for physicians who treat them. According to the OSM-IV,1 approximately 10% of psychiatric outpatients and approximately 20% of psychiatric inpatients meet criteria for BPD and the diagnosis is associated with high rates of both psychiatric and medical service use. Furthermore, BPD diagnosis is associated with high rates of parasiticide,2 (ie, any intentional self-injury resulting in tissue damage or risk of death, ranging from high lethality, high-intent suicide attempts to minor self-injuries with no suicidal intent). The lifetime suicide rate has been estimated at approximately 10%, and double that among those with a history of self-injury.3

DIAGNOSTIC CRITERIA

For a DSM-IV diagnosis of BPD, an individual needs to demonstrate five or more of nine criteria. Dialectical behavior therapy (DBT)4 conceptually organizes these nine criteria into five broad areas in which the person's experience and behavior may be dysregulated.

Emotion Dysregulation. The DSM-ÍV describes a criterion of a marked reactivity and instability of mood. This often is superimposed on a backdrop of chronic dysphoria. The DSM-IV also specifically singles anger out from other emotions as intense, inappropriate, and difncuit for the BPD patient to control My own experience has been that borderline patients are at least as likely to be underexpressive of anger as they are to be overexpressive of it. Either way, or with wild swings in both directions, certainly most patients with BPD cannot express anger appropriately. Patients' expressions of anger are probably more aversive to clinicians than expressions of shame, fear, guilt, or sadness, and so perhaps more salient. However, DBT views borderline patients as having equivalent difficulties regulating all emotions, at least aversive ones. Indeed, emotion dysregulation is viewed as the core difficulty of borderline patients. Other criteria are viewed as primarily either expressions of emotion dysregulation or maladaptive efforts to regulate mood.

Relationship Dysregulation. The DSM-I V describes unstable, intense relationships characterized by idealization and devaluation, and frantic efforts to avoid abandonment. Patients with BPD frequently have experienced a more than average amount of actual abandonment, either in childhood, or at least in adulthood, often as a result of their own behaviors. Idealization and devaluation can be viewed as extreme cognitive styles made more likely by both the patient's desperate need for connectedness that often is lacking in his or her life and by the normal effects of strong emotion on cognition.

Self-Dysregulation. The DSM-IV describes an unstable sense of self and chronic feelings of emptiness. Frequent extremes of emotion and associated behaviors could be expected to lead to difficulties in establishing a coherent sense of one's self. In addition, frequent invalidation of one's behaviors, thoughts and feelings, which patients with BPD typically have experienced, can lead to self-invalidation of one's own preferences, goals, perceptions, etc.

Behavior Dysregulation. Trie DSM-IV describes both impulsive and potentially harmful behaviors generally (eg, substance abuse, reckless driving, or binge eating) and also suicidal and other self-injurious behaviors specifically. These behaviors may serve a variety of functions. For example, an instance of parasiticide may reflect high intent to die. Alternatively, the function may be primarily to communicate distress or elicit an interpersonal response. Most frequently, however, the primary function seems to be to escape or otherwise regulate aversive emotions. Suicide attempts typically have this escape function, and patients typically report that, after other types of self-injury, they feel less afraid, ashamed, angry, etc. In my view, clinicians too frequently assume that parasinoidal acts of patients with BPD primarily have a communicative function, which is reflected in terms such as suicide "gestures" and "manipulative suicide attempts." It is important to assess the patienf s likely intent rather than to assume it based solely on the form of the behavior.

Cognitive Dysregulation. A subset of these patients may experience transient paranoia, dissociation, or hallucinations when under stress. These phenomena typically remit as the stress level decreases.

COURSE AND TREATMENT OUTCOMES

The disorder tends to be chronic, with about 50% of patients continuing to meet criteria at a 7year follow-up.5 The clinician's possible resulting feelings of frustration, incompetence, or hopelessness may be joined by feelings of anxiety, fear, or anger if the patient also engages in one of the three behaviors most stressful to mental health clinicians:6 suicide attempts, suicide threats, and expressing anger toward the clinician. Unsurprisingly, it is not uncommon for clinicians to hold pejorative or otherwise negative attitudes about individuals with BPD and to prefer not to work with them.

Treatment outcomes for BPD traditionally have been poor. Some forms of treatment may be iatrogenic for many with BPD. To cite two examples, psychoanalysts anecdotally have noted that these patients often became worse in response to classic analysis, and psychiatric inpatient units (particularly in the premanaged care era of longer stays) often observed an increase in problem behaviors of some BPD patients over the course of admission. Until very recently, there were almost no data strongly supporting efficacy of any class of medications and no such data for any form of psychotherapy or psychosocial program. Since DBT became the first psychosocial treatment with any supporting efficacy data from a randomized comparison study/ one other form of psychosocial treatment has received such support. Bateman and Fonagy8 reported significantly more favorable outcomes for patients diagnosed with BPD who were randomly assigned to a long-term, psychodynamically oriented partial hospital program (average length of stay = 1.45 years) than for those randomly assigned to standard outpatient treatment that included no psychotherapy. However, it should be noted that current managed care reimbursement trends in the United States are not consistent with a long-term partial hospital program.

EFFICACY OF DRT

Linehan et al.7 randomly assigned 44 parasuicidal women diagnosed with BPD to 1 year of DBT or to treatment-as-usual in the community. Women receiving DBT had significantly greater reductions in self-harming behaviors (including suicide attempts), in the medical risk of those behaviors, and in the frequency of psychiatric hospitalizations and lengths of stay, and lower treatment dropout rates than women receiving treatment as usual. Dialectical behavior therapy also showed superior efficacy in reducing trait anger and improving Global Assessment Scale scores and social adjustment.9 On measures of depression, hopelessness, reasons for living, and suicidal ideation/ however, the treatment conditions did not significantly differ (although patients in both conditions showed improvement during the 12 months of treatment). Improved symptoms and functioning of patients were generally maintained at 6- and 12-month follow-ups.10

My colleagues and I recently conducted an independent randomized study of DBT.11 Twenty women veterans who met criteria for BPD were randomly assigned to DBT or to treatment as usual for 6 months. Compared with patients in treatment as usual, those in DBT reported significantly greater decreases in suicidal ideation, hopelessness, depression, and anger expression. In addition, only patients in DBT demonstrated significant decreases in number of parasuicidal acts, anger experienced but not expressed, and dissociation, and a strong trend on number of hospitalizations, although treatment group differences were not statistically significant on these variables. Patients in both conditions reported significant decreases in depressive symptoms and in number of BPD criterion behavior patterns, but no decrease in anxiety.

Dialectical behavior therapy also has been adapted for treatment of other multi-problem patients. Randomized studies support the efficacy of DBT for reducing substance use in substance abusing patients with BPD,12 for binge eating disorder,13 and for bulimia,14 and a nonrandomized but controlled study suggests likely efficacy for suicidal adolescents.15 Other populations for whom this treatment model is being adapted and studied include couples in which there is domestic violence,16 men in forensic settings,17 and depressed elders with personality disorders.18

A BIOSOCIAL THEORY OF BPD

Dialectical behavior therapy is guided by a biosocial theory of the development and maintenance of BPD behaviors. The theory contains two major elements, one biological and the other socialenvironmental. An individual diagnosed with BPD may have a dysfunction of the emotion regulation system. The brain systems involved in eliciting and modulating emotions may be different than those in the average person, possibly because of genetics, events during fetal development, and /or early life trauma. The environmental aspect Linehan refers to as the "invalidating environment" is one in which individuals' communications regarding their private experiences frequently are met with responses that suggest they are invalid, faulty, or inappropriate, or that oversimplifies the ease of solving the problem.

Linehan4 suggests that, in addition to BPD resulting from a combination of these two factors, there is a transaction between the two, such that emotion dysregulation tends to lead to invalidation and vice versa. The responses of an individual who is particularly emotionally sensitive are likely to be puzzling to others who do not share this emotionality. They may then conclude that the person is faking his or her response to manipulate a situation, or is being entirely unreasonable and "crazy," or is not at all trying to control his or her behavior. If this belief is communicated, explicitly or implicitly, the sensitive individual is likely to feel even more emotionally vulnerable. Furthermore, if an individual's emotional state, his or her thoughts related to it, and his or her difficulty in changing his or her emotions are not taken seriously or are punished, and if this occurs during the course of development, then the individual may not learn how to accurately recognize or communicate different emotions.

Over time, as the individual's behavior becomes more extreme, either in attempts to regulate emotion in the absence of more adaptive skills or in attempts to communicate, he or she is more likely to experience invalidation from his or her environment, including from the mental health system. Thus, in this transactional model, the individual and those in his or her interpersonal environment continuously change one another. Similarly, the individual is not viewed in DBT as "having" a disorder, but as acting, at times, in disordered ways. It is this individual in this particular situation whose behavior is ineffective and dysfunctional. It is quite possible that the individual might behave quite functionally in a radically different context. Consistent with this systemic view, roadblocks in treatment are not automatically attributed to the patient, but to some transaction among the patient, therapist, the consultation team, the institutional environment in which treatment occurs, and the patient's home environment (any combination of which may be targeted for intervention).

STAGES OF TREATMENT

One of the difficulties clinicians face in helping individuals with BPD is the number of domains in which they have problems. It is not uncommon for a BPD patient to meet criteria for other disorders such as major depression, panic disorder, posttraumatic stress disorder, obsessive-compulsive disorder, eating disorders, or substance abuse, as well as to have various psychosocial problems involving relationships, finances, school or work, housing, etc. Their difficulties also may adversely affect the treatment process and therapeutic relationship. The sheer number of problems to address can be overwhelming to patient and therapist alike. In DBT, this is addressed by conceptualizing treatment as following a sequence of stages that are determined by the level of dysfunction, and by having a clear hierarchy for prioritizing treatment targets within each stage.

Stage 1 treatment is for individuals who are demonstrating severe behavioral dyscontrol, such as self-injury, severe eating disorder or substance abuse, or repeated hospitalizations. Most of these behaviors can be conceptualized as escape behaviors. The goal of this stage 1 treatment is simply to block extreme escape behaviors and develop greater behavioral control. This stage may be brief or protracted, or may not be needed at all. Stage 2 treatment focuses on avoidance rather than escape behaviors. Many patients with BPD have experienced earlier traumata (about 75% report a history of childhood sexual abuse, and physical abuse may be equally common) and have posttraumatic stress disorder symptoms or otherwise avoid certain situations or emotions. The goal of stage 2 therefore is to increase appropriate experiencing of emotions. Ln the case of trauma history, one of the more common mistakes therapists make with patients with BPD is to focus on the trauma history when the patient is still engaging in stage 1 behaviors. If these are the patient's maladaptive attempts to regulate emotion, focusing discussion on the most emotionally disturbing events of the individual's life is likely to lead to an increase in those behaviors, including suicide attempts. Exposure-based strategies for treating posttraumatic stress disorder therefore need to be conducted only when the patient has demonstrated reasonable stability regarding self-injury and other serious, potentially harmful behaviors. Stage 3 treatment has the goals of ordinary happiness and unhappiness, improved relationships and self-esteem, and stage 4 treatment moves away from amelioration of problems to promotion of an increased capacity for joy and sense of connection. The current research on DBT and the remainder of this article both focus on stage 1 treatment.

Prior to beginning DBT, in a pretreatment stage, it is important that the therapist and patient reach an explicit agreement about the goals of the treatment and its methods. For example, if patients engage in suicide attempts or other self-injury, we require a commitment from them that stopping such behavior is their goal and that they commit to stopping the behavior immediately. This does not, of course, mean that patients will never again selfinjure. If it were so simple, they would not need treatment. However, it does mean a strong commitment about their intent. If self-injury occurs during the course of treatment, then the therapist and patient will analyze the incident in detail and develop alternative solutions to whatever factors were associated with the behavior occurring at that time. Patients also must agree to address any behaviors that interfere with the treatment, and to attend the different treatment modes consistently. Explicit commitments are more likely than vague or implied agreement to influence actual behavior. Verbal agreements are more natural and therefore preferable, in my opinion, to more legalistic written agreements.

PRIORITIZING TREATMENT TARGETS

Life-Threatening and Other Self-Injurious Behaviors. The highest priority is given to this class of behaviors, which includes suicide attempts, suicide threats, non-suicidal self-injury, urges to engage in these behaviors, and changes in suicidal ideation. Some clinicians do not address such behaviors directly because they view them merely as symptoms of an underlying personality problem that their treatment instead focuses on, or because they fear that talking about them will make them more likely. In DBT, we take seriously even relatively minor self-injuries, as they can otherwise tend to escalate over time, like addictive behaviors. Whenever the patient has engaged in such behavior between sessions, it will be the primary focus of the next session, in which the patient and therapist will conduct a detailed behavioral analysis of the incident and develop solutions for the patient to use in future similar situations. I describe in a later section how such a behavioral analysis is conducted.

How does the clinician know that the patient has engaged in self-injury or any other behavior that is being targeted in therapy? Many of us have had the experience of learning about such an incident from the patient just as they are leaving the session, or even much later. In DBT, patients complete diary cards or sheets on a daily basis. These forms are used to collect information about occurrence or level of targeted behaviors, such as selfinjury, suicidal ideation, purging, substance use, levels of certain emotions, etc. At the start of each session the patient gives the completed form to the therapist, who then uses it, in combination with the hierarchy of targets for this patient, to determine the priority of topics for that session. In my experience, almost all patients will complete these forms, so long as any initial noncompliance is appropriately addressed.

Therapy-Interfering Behaviors. These are any behaviors, of patient, therapist or others, that seriously interfere with the progress of therapy. So this would include many behaviors that clinicians of other orientations would refer to as resistance, but also other behaviors. One reason for the high priority given to these behaviors is that approximately 50% of patients with BPD drop out of most treatments within 1 year, often because of their frustration or dissatisfaction with the provider or their provider's frustration with them. If the patient does not attend treatment for very long, it is unlikely we will be able to help him or her. In the Linehan et al.7 study, only 17% of patients in DBT dropped out of treatment with their initially assigned therapist, compared with 58% of those in treatment as usual. In our own study,11 the dropout rate for DBT was 23%.

Some common therapy-interfering behaviors of patients are those that interfere with receiving the treatment, such as not attending regularly, repeatedly being very late, and dissociating during sessions. Another class is behaviors that cross the therapist's personal limits or reduce his or her motivation to treat the patient, such as repeated hostility toward the therapist, not completing diary cards or doing other agreed-on between-session assignments, calling the therapist too frequently, not calling when it would have been appropriate to do so, etc. What is considered therapy-interfering thus will depend on the particular limits, needs, and tolerances of a particular therapist. Dialectical behavior therapy does not suggest what those limits should be, just that therapists need to observe their limits and directly address any behaviors that are interfering with therapy. Just like other targeted behaviors, a therapy-interfering behavior may be the focus of an in-session behavioral analysis and solution analysis.

Therapists too can and do engage in therapyinterfering behaviors, such as being either too oriented toward change or too oriented toward acceptance, being late for appointments, distracted or sleepy during sessions, not returning phone calls, disrespectful of the patient, etc. The patient with BPD tends to be vigilant about every aspect of the therapist's conduct toward him or her, and the therapist needs to be alert to and willing to acknowledge and attempt to change his or her own behaviors that are interfering with the therapy.

Quality-of-Life-Interfering Behaviors. Borderline personality disorder patients may engage in behaviors that seriously interfere with experiencing an adequate quality of life, including substance abuse, bingeing and /or purging, unsafe sexual behaviors, shoplifting, homelessness, not having any activities or social contacts beyond therapy, etc. Quality-of-life-interfering behaviors to be targeted are agreed on and noted on the diary card. Should these occur in a given week, behavioral analysis of them is conducted as a third priority.

Skilled Behaviors. The fourth and final priority is the acquisition and strengthening of skills that can be used to replace the problem behaviors. The individual therapist typically is unable to spend a lot of time systematically on skill development in stage 1 treatment, because the patient so frequently is in crisis or has engaged in some high priority behavior that must be addressed. Another mode of the treatment model, the skiUs-training group, primarily serves this function.

TREATMENT MODES AND THEIR FUNCTIONS

Dialectical behavior therapy proposes that a comprehensive treatment for patients with BPD needs to do at least four things: (1) help the patient develop new skills; (2) address motivational obstacles to skills use; (3) help the patient generalize what they learn to their daily lives; and (4) keep therapists motivated and skilled in treating a difficult-to-treat population. In standard outpatient DBT, these four functions are addressed through four different modes of treatment, described below. In other settings, such as inpatient or partial hospital, other modalities of treatment could be used to address the same four functions.

Developing Skills: The Skills-Training Group. Patients with BPD usually have deficits in knowledge or use of skilled behaviors in many areas, including tolerating distress, regulating emotions, interacting assertively, and being mindfully aware of the present moment. So DBT teaches these four sets of skills systematically, usually in a skillstraining group, and guides patients in practicing them. The group is much like a seminar. The first part of each session consists of a review of homework assigned the previous week and troubleshooting obstacles, and the second part consists of presentation of new material, following a curriculum and manual,18 with lecture and discussion, role-play and rehearsal, and homework assignments. The entire curriculum can be covered in about 6 months of weekly 2-hour sessions, and for most patients it is recommended that they go through the sequence twice. Because of the emotional sensitivity of the patients, "process" discussions are discouraged, and the focus is maintained on the skills to be learned.

Problem Solving Motivational Obstacles: Individual Therapy. A patient may know and have the ability to employ a skill, but not use it in a given situation. So the treatment also must address motivational factors that may be mamtaining problem behaviors and motivational obstacles to skills use. For our purposes, we can consider motivational to mean any factors that increase or decrease the likelihood of the behavior. These include interfering emotions, thoughts or beliefs of the patient, and the internal and environmental consequences of their behaviors. One of the primary functions of individual therapy in DBT is to address emotions and beliefs that lead to problem behavior or interfere with use of skills, and consequences mat may be reinforcing problem behavior or pumshing skilled behavior. The individual therapist may also teach skills in session when needed, but his or her primary role is to help the patient use whatever skills he or she has to navigate the crises that inevitably arise in his or her life in an adaptive and effective way. Frequently, this is in the context of a behavioral analysis of a recent incident of problem behavior. The individual therapist is considered the primary therapist for the patient, regarding the other treatment modes.

Generalizing L·arning: Telephone Coaching. We assume that anything learned in the treatment setting may suffer some loss of generalization to the patient's natural environment. Telephone coaching between sessions can promote the use of skilled behavior while the patient is in the crisis or situation to which he or she needs to respond. The purpose is not for the therapist to provide emotional support or to analyze the situation. It is to coach the patient, in a brief period (10 minutes) on how to respond to a situation or to an urge to self-injure, or drink alcohol, or engage in some other problem behavior. Patients therefore are instructed to call before, rather than after, engaging in high-priority target behaviors, when they have already "solved the problem" in their own way. In particular, they are not to call their therapist within 24 hours of self-injury, as therapist attention contingent on mat behavior may reinforce it. If the patient does call in that circumstance, the therapist only assesses and responds to medical risk and does not engage in problem solving or skills coaching. Other than that rule, therapists need to determine and observe their own limits regarding patients' use of telephone contacts. Calls to skills group leaders should be only about the group itself (eg, attendance or the homework).

Motivating Therapists and Enhancing Their Skill: Consultation Team Meeting. Therapists working with patients with BPD may easily experience frustration, hopelessness, anger, fear, and other emotions that interfere with their ability to help the patient. Therapists also may respond easily to extreme patient behavior by extreme or unbalanced use of treatment strategies. The purpose of the DBT consultation team is to help motivate therapists and to provide guidance in conducting the treatment. Dialectical oppositions are encouraged in the service of problem solving.

CORE TREATMENT STRATEGIES: PROBLEM SOLVING AND VALIDATION

In all modes of treatment, individual therapists and skiUs-tiaining group leaders employ the same sets of strategies. Some of these strategies are designed more to promote change and others more to convey acceptance of the patient. The patient very likely needs to make many major changes in his or her life, a need that can seem desperate because of the high level of distress. However, if the therapist focuses only on change strategies, the patient is likely to feel misunderstood and angry, or to blame himself or herself for not being different. Alternatively, the therapist may focus primarily on validating the patient's pain and associated behaviors. Once again, the patient may experience this as unhelpful because his or her experience that his or her life desperately needs to change is not addressed. In DBT, change and acceptance strategies therefore are woven together, integrated throughout the treatment, always in an effort to achieve a dialectical balance. The core strategies that are balanced are problem solving and validation.

Problem Solving. When a targeted behavior has occurred, usually indicated by the diary card, the therapist and patient attempt to determine what factors influenced its occurrence and how they could be dealt with in the future. The first step is to conduct a behavioral analysis. The goal of a behavioral analysis is to uncover a sequential rendering of the preceding vulnerabilities, prompting events, interpretations, feelings, urges, and observable behaviors that came before, and followed, a specifically defined problem behavior. The first step of a behavioral analysis is to clearly describe the problem behavior. This description should be nonjudgmental, objective, and specific. An example is: "Friday evening, between 11 and 11:30, scratched ankles repeatedly with fingernails, enough to draw blood but not requiring stitches." The therapist then conducts a chain analysis of the behavior's immediate antecedents and its consequences. In DBT, we first try to identify a prompting event, something external to the patient that precipitated the chain of events. The patient may initially be unable to identify a prompting event as in, "I always feel suicidal." A useful strategy may then be to pinpoint more exactly the point in time at which the urge increased. It is also often helpful to identify vulnerability factors that may have preceded the prompting event, and it is always helpful to establish the links in the chain that led from the prompting event to the actual problem behavior. These links may involve thoughts that the patient had about the event, feelings associated with those thoughts, subsequent behaviors, reactions to those behaviors by others and by the patient himself or herself, and so on. The therapist also inquires about consequences of the problem behavior, including its affective consequences for the patient, interpersonal responses of others, and environmental changes that resulted. This may help to identify possible reinforcing factors and provide the therapist with opportunities to highlight negative consequences.

In conducting a chain analysis, the object is to develop a chain with as many links as possible. The more links in the chain, the more places there are that something different could occur. Perhaps the patient needs to avoid certain situations in which this type of prompting event is likely, or perhaps he or she needs to develop greater skills in changing that situation, or in tolerating it. Perhaps changes are indicated at the level of cognitive interpretations. Or at the very last link prior to the actual behavior, the patient could engage in one or more distress tolerance strategies that would help him or her to tolerate the urge to engage in the behavior without doing so.

Once a behavior in a situation has been analyzed and understood, and possible solutions evaluated by the therapist and patient, implementing those solutions may require some combination of four sets of behavior therapy procedures: skills training; contingency management; cognitive modification; and exposure. If the patient simply does not know how to behave more skillfully, skills training is indicated. If, however, the patient knows what to do, but is punished or not reinforced for doing so, or is reinforced for doing otherwise, contingency management is called for. If the patient's beliefs, attitudes, and thoughts interfere with behavior, cognitive modification procedures may be helpful. If the patient is unable to act more skillfully because of strong emotional reactions, exposure procedures that allow those reactions to habituate may be useful. Space precludes detailed description of such procedures, but they are standard cognitive and behavioral therapy procedures.

Validation. The other core strategy, used to balance problem solving, is validation. This simply means communicating to the patient that his or her responses make sense, are understood, or are reasonable. One might note that emotional responses always are valid in some sense, in that there is something that provokes them. It may be that an individual's perception that another person is being hostile is invalid, but if that is their perception, then experiencing fear makes sense. Some things clearly are invalid, such as a belief that the government has been taken over by Martians. Many other things, however, can be valid in some way but not valid in another

For example, self-injury may regulate a patient's emotions. The behavior therefore is valid in terms of a short-term consequence. It makes sense. On the other hand, the behavior probably has various negative consequences and is not effective in helping the patient reach his or her long-term goals in life. Early in treatment, it may be helpful to validate self-injury in the sense of communicating that it is understandable. Validation does not mean approval, and such a statement usually should be accompanied by statements about the necessity of replacing the behavior with others to be learned and practiced.

Validation can occur at a number of levels. The most basic level involves simply unbiased listening and observing. This communicates to the patient that he or she is important, and worth listening to and taking seriously. The second level of validation involves accurate reflection of the patient's communications. Summarizing and paraphrasing communicates to the patient that he or she has been understood. A more advanced level of validation involves articulating unverbalized emotions, thoughts, and behavior patterns with statements like, "I see I made you angry" or "If I were in that situation, I'd be really mad." The patient may feel particularly understood if he or she did not even have to communicate his or her reaction. Of course, if the therapist's inference is incorrect, it is likely to be experienced as invalidating, so it is important not to stray far from the observable data. A fourth level of validation is in terms of the patient's past learning history or biological dysfunction.

For example, the therapist might state, "I think it is understandable that you often find it difficult to focus because of your diagnosed attentiondeficit /hyperactivity disorder" or "It makes sense that you would have difficulty trusting me, because I'm a man and you have been treated very badly by men in the past." However, this last example also illustrates that implicit in the therapist's statement is the notion that the patient's reaction involves a distortion (ie, is a transference reaction). At times it may be more helpful to validate in terms of the present context or normative functioning, such as, "It makes sense that you would be having difficulty trusting me. After all, we have only met a few times and it usually takes some time for most people to come to trust their therapist." The highest level of validation described in DBT is called radical genuineness. This involves the therapist responding as his or her natural self, rather than with role-prescribed behavior, not treating the patient as overly fragile, but instead as able to tolerate one's natural reactions. It therefore validates the patient's strength.

COMMUNICATION STYLE STRATEGIES: RECIPROCAL AND IRREVERENT

Dialectical behavior therapists strive to balance their communication style between being reciprocal and being irreverent. A reciprocal style is the modal style. It involves being warm, genuine, and empathie. It may include self-disclosure. Self-disclosure is used in behavior therapies in several ways. One is simply to respond to the patient's queries regarding professional or personal information. It is reasonable for patients to want to know about their therapist's qualifications, and many patients want to have some sense of their therapist as a person, not just as a role figure. In DBT, we do not take the position that disclosing such information is inherently a "boundary violation." Rather than being guided by an arbitrary set of rules, it makes more sense to determine in each situation whether the information might be helpful to the patient, whether it might be harmful to him or her, and whether the therapist personally is comfortable with revealing the information. I believe therapists are often too withholding for their patient's good, and do so only because that is what they have been taught in training.

Another form self-disclosure can take is modeling. The therapist can describe how he or she coped with a situation in their own life to serve as a model. This is usually most effective when a coping model rather than expert mastery model is used. Finally, DBT therapists frequently disclose their reactions to the patient ("self-involving self-disclosure") to reinforce or punish a behavior, or because the patient does not get useful feedback from others about the impact of his or her behavior.

Irreverence involves deliberately being out of synch with the patient, saying or doing something unexpected, humorous, or confrontational. Used too frequently, it runs the risk of alienating the patient, who may perceive the therapist as sarcastic or mean. Used judiciously, it can help a patient look at a situation from a new perspective, or back away from a threat. For example, a patient tells her individual therapist that she is going to quit the skillstraining group. Both therapist and patient know that she must be in the group to stay in this individual therapy. The therapist might simply state, "I'm going to miss working with you." With a patient I know well and have a good relationship, I might say, "Thaf s one of the craziest things I've heard you say," ramer than the more reciprocal, "You must be feeling very upset to be thinking that."

CASE MANAGEMENT STRATEGIES: CONSULTATION TO THE PATIENT AND ENVIRONMENTAL INTERVENTION

A third level of strategies to be balanced is the level of case management. Borderline patients often have come to adopt a passive problem-solving style. They may be very good at getting others to solve problems for them but poor at doing it themselves. The primary strategy of the DBT therapist therefore is to consult to patients about how to manage their social or professional networks, not to consult with the network about how to manage the patient or intervene directly on behalf of the patient. In the long run, it is more helpful to teach people to fish than to give them a fish. Only if the outcome is very important and the patient does not yet have the ability to be effective, would the therapist intervene on his or her behalf. Whenever possible, this would be done in the patient's presence.

CONCLUSION

Dialectical behavior therapy assumes that the core difficulty of these patients is one of emotion dysregulation and that this dysregulation is exacerbated and maintained in part by experiences of being invalidated by others. It deals with the large number of areas of difficulty of these patients by dividing treatment into stages and specifying a hierarchy for prioritizing behaviors targeted for treatment. The treatment attends to the patient's skills deficits as well as to motivational obstacles to skills use, to the need to assist patients in use of skills between sessions, and to the needs of the therapist. Dialectical behavior therapists balance use of more acceptance-oriented strategies of validation, a reciprocal communication style, and environmental intervention for the patient with more change-oriented strategies of problem solving, irreverent communication style, and consulting with the patient about how to interact with their environment. At this time, DBT is the only outpatient psychosocial treatment to have demonstrated efficacy in changing clinically important outcomes for patients diagnosed with BPD. Further studies will be needed to determine the limits of its efficacy and breadth of applicability across patient populations.

REFERENCES

1. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Washington, LXI: American Psychiatric Association; 1994.

2. Qarkin JF, Widiger TA, Frances A, Hurt SW, Gilmore M. Prototypic typology and the borderline personality disorder. ; Abnorm Psychol. 1983;92:263-275.

3. Stone MH. The course of borderline personality disorder. In: Tasman A, Hales RE, Frances AJ, eds. Review of Psychiatry. 1989;8:103-122.

4. Linehan MM. Cognitive-Behavioral Treatment of Borderline Personality Disorder. New York: Guilford Press; 1993.

5. Links PS, Heslegrave R, van Reekum R. Prospective followup study of borderline personality disorder: prognosis, prediction of outcome, and axis ? comorbidity. Gin / Psychiatry. 1998;43:251-259.

6. Hellman ID, Morrison TL, Abramowitz SI. The stresses of psychotherapeutic work: a replication and extension. / Clin Psychol. 1986;42:197-205.

7. Linehan MM, Armstrong HE, Suarez A, Alimón D, Heard HL. Cognitive-behavioral treatment of chronically suicidal borderline patients. Arch Gen Psychiatry. 1991;48:1060-1064.

8. Bateman A, Fonagy P. Effectiveness of partial hospitalization in the treatment of borderline personality disorder: a randomized controlled trial. Am f Psychiatry. 1999,156:1563-1569.

9. Linehan MM, Tutek D, Heard HL, Armstrong HE. Interpersonal outcome of cognitive-behavioral treatment for chronically suicidal borderline patients. Am } Psychiatry. 1994;51:1771-1776.

10. Linehan MM, Heard HL, Armstrong HE. Naturalistic follow-up of a behavioral treatment for chronically parasuicidal borderline patients. Arch Gen Psychiatry. 1993;50:157-158.

11. Koons CR, Robins CJ, Tweed JL, et'al. Efficacy of dialectical behavior therapy in women veterans with borderline personality disorder. Behavior Therapy. 2001;32:371-390.

12. Linehan MM, Schmidt HI, Dimeff LA, Craft JC, Kanter J, Comtois KA. Dialectical behavior therapy for patients with borderline personality disorder and drug-dependence. Am J Addictiotis. 1999;8:279-292.

13. Telch CF, Agras WS, Linehan MM. Dialectical behavior therapy for binge eating disorder. / Consult Clin Psychol. 2001;69:1061-1065.

14. Safer DL, Telch CF, Agras WS. Dialectical behavior therapy for bulimia nervosa. Am } Psychiatry. 2001;158:632-634.

15. Rathus JH, Miller AL. Dialectical behavior therapy adapted for suicidal adolescents. Suicide Life Threat Behav. 2002;32:146-157.

16. Fruzzetti AE, Levensky ER. Dialectical behavior therapy for domestic violence: rationale and procedures. Cognitive and Behavorial Practice. 2000;7:435-447.

17. McCann RA, Ball EM, Ivanoff A. DBT with an inpatient forensic population: the CMHD? forensic model. Cognitive and Behavorial Practice. 2000;7:447-456.

18. Lynch TR. Treatment of elderly depression with personality disorder comorbidity using dialectical behavior therapy. Cognitive and Behavorial Practice. 2000;7:468-477.

19. Linehan MM. SIdIIs Training Manual for Treating Borderline Personality Disorder. New York: Guilford Press. 1993.

10.3928/0048-5713-20021001-09

Sign up to receive

Journal E-contents