Ethical Considerations in Treatment of Personality Dysfunction: Using Evidence, Principles, and Clinical Judgment

Ethical Considerations in Treatment of Personality Dysfunction: Using Evidence, Principles, and Clinical Judgment



Jeffrey  J. Magnavita

Independent practice, Glastonbury, CT

Kenneth  N. Levy

Pennsylvania  State University




Kenneth  L. Critchfield

University of Utah Neuropsychiatric  Institute

Jay  L. Lebow

The Family Institute at  Northwestern


JEFFREY J. MAGNAVITA received his PhD from the University of Con- necticut. He maintains an independent practice in Glastonbury, CT, and is an affiliate medical staff member at Hartford Hospital and Connecticut Children’s Medical Center. His areas of professional interests include unified clinical science and evidence-based treatment of personality dys- function. He is the founder of the Unified Psychotherapy Project, which is beginning to catalogue the methods and techniques of psychotherapy.

KENNETH N. LEVY received his PhD in clinical psychology from the City University of New York. He is an assistant professor in the Department of Psychology at the Pennsylvania State University and an adjunct assistant professor of psychology in psychiatry at the Joan and Sanford I. Weill Medical College of Cornell University. His research focuses on social cognition, attachment theory, emotion regulation, personality disorders, and psychotherapy process and outcome. He also maintains an independent practice in State College,  PA.

KENNETH L. CRITCHFIELD received his PhD in clinical psychology from the University of Utah, where he is now an adjunct assistant professor in the

Departments of Psychology and Psychiatry. He receives a salary from the University of Utah to co-direct the Interpersonal Reconstructive Therapy Clinic of the University Neuropsychiatric Institute. He also maintains a small independent practice. His research and training interests focus on use of interpersonal theory to inform psychotherapy case formulation, in- session relationship processes, and treatment outcome, especially for com- plex cases involving personality  disorder.

JAY L. LEBOW received his PhD from Northwestern University. He is a senior staff member at the Family Institute at Northwestern University, where he is involved in clinical practice, teaching, and research, and is clinical professor of psychology at Northwestern University. His areas of interest include integrative psychotherapy, couple and family therapy, the interface of research and practice, the assessment of progress in psycho- therapy, and intervention in high-conflict  divorce.

CORRESPONDENCE CONCERNING THIS ARTICLE should be addressed to Jeffrey J. Magnavita, Glastonbury Medical Arts Center, 300 Hebron Av- enue, Suite 215, Glastonbury, CT 06033. E-mail:

Professional Psychology: Research and Practice, 2010, Vol. 41, No. 1, 64 –74

© 2010 American Psychological Association        0735-7028/10/$12.00       DOI:  10.1037/a0017733



Clinical work with clients suffering from personality disorders can be among the most challenging for psychologists. These clients may have a wide range of clinical presentations, and many practitioners may lack the specialized training needed to provide successful treatment to these clients. Clinicians  are  faced  with several challenges in making treatment decisions that are ethically informed and based on available research findings. Because of the relative dearth of evidence-based treatments for these clients, clinicians are encour- aged to use a cost– benefit analysis approach when weighing the benefits versus disadvantages of specific interventions and treatment approaches. Recommendations for effective and ethical treatment of clients with personality dysfunction are provided that are based on an empirically grounded framework. Three expert commentators provide insights into the state-of-the-art of clinical work with these clients.



When the Evidence Base Is  Scant:

Some Considerations in the Ethical Treatment of  Personality Dysfunction

Jeffrey  J. Magnavita

Personality disorders are characterized by chronic patterns of dysfunction in multiple domains of the personality system, such as

cognitive–affective, perceptual, interpersonal, familial, and soci- etal. The classification of personality disorders is a complex en- deavor because these conditions present themselves in a variety of forms ranging from seemingly hidden to the untrained eye to clearly severe in manifestation. They often coexist with other personality disorders (Dimaggio & Norcross, 2008) and clinical syndromes (Magnavita, 1998), such as anxiety, depression, sub- stance use disorders, bipolar disorder, and conduct disorder, as well as the spectrum of relational disorders that cause distress, such as marital/couple disturbances, parent– child conflict, family dysfunction, and conflicts with society. Therefore, personality disorders represent a heterogeneous population and are very com- monly seen in mental health settings but are not necessarily the primary reason for seeking treatment. The treatment of these complex disorders requires an extensive knowledge base in theory and practice (Livesley, 2001; Magnavita, 2004), as well as per- sonality systematics, which is the study of the interrelationships among the various domains of the personality system from the micro- to macrolevel (Magnavita, 2006, in press).

It is likely then that most psychologists encounter people with personality dysfunction on a regular basis, regardless of the setting in which they work or the type of work conducted. In fact, epidemiological findings show that about half of those receiving mental health treatment and 1 out of 10 people warrant a diagnosis of personality disorder (Weismann, 1993). Psychologists are gen- erally well trained in the identification and diagnosis of personality disorders and whether one accepts personality disorder as a valid clinical entity or not, it is likely that many of these patients have been described as “treatment refractory,” “nonresponders,” “un- treatable,” and so on, and are encountered by and test the clinical and ethical capacities of all practitioners. In this article, I discuss how, at various clinical decision points in the treatment of patients with personality dysfunction, ethical issues need to be considered with a cost– benefit analysis and that this process can be a guide to ethical practice. There is accumulating clinical evidence of the efficacy of various approaches to the treatment of personality dysfunction (Magnavita, in press). Unfortunately, there currently is a dearth of empirical evidence with which to make many basic clinical decisions and, thus, ethical decision making is critical to patient care. At this point in our clinical science, there is no ideal approach to treating personality disorders (Stone, 2009) but there are many options with different  implications.


Clinical Decision Making at Critical Points in Assessment  and Treatment

A common clinical experience is that one is asked to provide professional psychological assistance to a child, adolescent, adult, couple, or family for some type of distress they are trying to cope with but feel stuck. After assessment a course of treatment is discussed and initiated with the patient system generally focusing on the clinical syndrome or DSM, Axis I condition (American Psychiatric Association, 1994). Evidence-based treatment is initi- ated, and the psychologist is hopeful that a positive outcome will be achieved in a reasonable amount of time. Treatment proceeds, but sometimes things seem to worsen, or not improve, and the psychologist begins to experience a spectrum of reactions from feelings of inadequacy to frustration toward the patient. At this point, the evidence suggests that a comorbid personality disorder might be complicating first-line treatment. The first series of ethical dilemmas are beginning to coalesce. If one is not trained in treating personality disorders, should a referral be made at this point to someone trained in personality systematics and specialized treatment for personality dysfunction? A risk– benefit analysis might suggest that continuing without addressing the personality dysfunction will lead to increased frustration in both patient and psychologist, and the evidence suggests that this misalliance and lack  of  a  collaborative  treatment  plan  are  likely  to  result  in a

premature termination of treatment (Hilsenroth & Cromer, 2007; Yeomans, Gutfreund, Selzer, Clarkin, Hull, & Smith, 1994). This is one outcome with limited apparent consequences to the psy- chologist. The patient may be angry and not pay the fee or fail to show up for the next scheduled appointment. Everyone loses, but no greater harm has occurred. Conversely, the clinician must weigh the potential harm of the patient terminating treatment and continuing with a self-destructive or self-defeating pattern. This may present a less acceptable risk in some cases. In other cases, the harm to self and others might be too high a risk.


Above All, Do No Harm and Where Possible, Do Good

Ethical treatment of patients manifesting a personality disorder can be guided by questions addressing the risk– benefit of various clinical decisions to safeguard the treatment process and optimize outcome. No definitive answers will be forthcoming, but the path to ethical treatment will be safeguarded and the patient’s care will be honored. It is evident that clinicians who work in this area must be able to manage working with incomplete information and uncertainty. In the aforementioned example, the clinician can ask two sets of risk– benefit analysis questions concerning treatment decision making with a focus on analyzing potential risk and harm. The first two questions are, “What are the risks if I continue to treat the patient without possessing specialized competence in the treat- ment of personality disorders?” and the opposite, “What are the risks if I refer to someone else whom I know has competence in treating personality disorders?” In answering the first series of questions, the psychologist might respond, “The risk of continuing to treat is that the results might be less than optimal, and it is likely that the patient will drop out of treatment because of a lack of progress.” The answer to the second question might be, “The risk of referring to someone else might engender a feeling of abandon- ment, which might worsen the patient’s condition and possibly make the patient feel that he or she is untreatable, leading to potential demoralization and worsening of the condition.”

Proceeding in our risk– benefit analysis, the next set of questions then considers the alternative choice. “What are the benefits of maintaining the patient in treatment with me?” and the corollary, “What are the benefits of referring to another clinician with po- tentially greater competence to deliver an evidence-based treat- ment?” In answering the second two questions, the psychologist might respond, “Keeping the patient in treatment with me might lead to a better outcome if there has been a positive therapeutic alliance and I seek consultation from someone more expert,” and, in addressing the final question, “The potential benefit of referring the patient to an ‘expert’ could lead to the possibility of a better outcome.”

This type of risk– benefit analysis looking at the possible results of each question in this quadratic formulation can be used as a guide where insufficient empirical data exists to enhance ethical clinical decision making in the treatment of personality dysfunc- tion. We are many decades away from having clinical algorithms to make these complex treatment  decisions.


Shattering Sacred Icons and Challenging the Reification of Ritualized Clinical  Practices

Although there is burgeoning research evidence to guide the treatment of personality disorders, in many regards psychologists have surprisingly little empirical evidence that can be used to inform their practices and clinical decision making (Magnavita, in press). There are a number of treatment approaches that have a reasonable research base, such as transference-focused therapy, dialectic behavior therapy, short-term psychodynamic psychother- apy, and schema-focused therapy. However, there is no ideal approach for every patient. Blindly following a protocol for an empirically based approach may in itself be unethical. In one case (in which I was the therapist), the patient asked for hypnosis to help him stop drinking. After being told there was no empirical evidence that this would be productive, the patient, at the end of what appeared to be a very productive session, announced she wasn’t coming back. In this case, one might quote the old adage, “The operation was a success, but the patient died,” as one of my early supervisors used to remind us fledgling therapists. Rather, the complexity of the disorder may require an eclectic approach (Stone, 2009). Much of what we accept as standard practice, such as length of sessions, frequency of sessions, boundary issues, and multiple therapeutic relationships (when the same psychologist combines multiple treatment modalities) are best characterized as clinical lore and ritual, often without any substantive body of evidence supporting the continuation of these practices. In the era of evidence-based treatment for the more severe personality dis- orders, it seems unlikely that any one approach might fit all.

In fact, there are many practices that are reified by psychother- apists and yet may be contraindicated. For example, the 50-min session, which has been the mainstay of most psychotherapy, may not be optimal for those whose defenses are entrenched, as we see in many patients with personality disorders. Longer sessions may allow sufficient time to restructure defenses and address core issues. Flexibility might necessitate experimenting with the most efficacious treatment format and then altering it, depending on patient response to treatment. The practice of treating personality dysfunction is based on a foundation of empirically determined principles, such as clinician flexibility and creativity (Castonguay & Beutler, 2006), combined with continual ethical informed deci- sion making, even when there may be little in the way of empirical data on which to base these clinical decisions.

Some of the central challenges that are likely to be encountered when treating personality dysfunction and ways to ensure ethical treatment are now discussed. First, the topic of clinical competence is addressed.


Clinical Competence to Practice Treatment of Complex  Clinical Conditions

In considering a practice devoted to or emphasizing treating personality dysfunction, or even in effectively treating those who are in one’s practice, it is fundamental that one has sufficient education, training, supervision, and experience. The first issue to consider is an honest self-evaluation of one’s knowledge base and experience. If one has primarily spent time as an administrator or doing career counseling, it might be an ethical breach to undertake treating such complex disorders. On the other hand, most psychol- ogists have at least basic training in personality theory, psychopa- thology, and advanced psychotherapeutic skills to theoretically enable them to treat the less severe of the personality disorders and with ongoing supervision and training more challenging cases.

Standards for clinical competence do not yet exist for this “specialty” other than the general ethical guidelines for practicing in any area. Essential components of developing treatment com- petence are familiarity with the relevant research literature on etiology, prevalence, principles of therapeutic change (Castonguay & Beutler, 2006); psychodiagnostic and interviewing skills; train- ing in one or more models of treatment (psychodynamic, cognitive– behavioral, schema-focused, transference-focused ther- apy); training in more than one modality of treatment (individual, group, family, couples, pharmacological); or at least access to others with whom one can work as a team, as well as supervi- sion—ideally, audiovisual supervision by one trained in one or more approaches to the treatment of personality dysfunction.

One must consider the implications for treating personality dysfunction without adequate training and supervision. The most important caveat, as there is a higher suicide rate in some patients with personality disorders, (especially with comorbid syndromes such as bipolar, substance abuse, and depressive disorders), is the increased risk of suicide. Because of the fact that, for some personality disorders, there is a high lethality risk, even those psychologists who are expert will likely encounter near lethal attempts or completed suicide (Soloff, Fabio, Kelly, Malone, & Mann, 2005). Another potential hazard has to do with the lack of treatment potency that might result in a continuation of patterns of behavior that may be harmful to family members or the individual himself or herself. For example, untreated personality dysfunction may fail to stop patterns of child maltreatment, high-risk behavior, and a continuation of the multigenerational transmission process (patterns of dysfunctional behavior passed on from one generation to another). Personality dysfunction that is not identified early in children and adolescents may lead to severely harmful patterns in adulthood. For example, there is accruing evidence that conduct disorder in children may be a precursor to adult antisocial person- ality disorder. It is therefore critical to be competent in diagnosis and treatment, at least at a basic level with ongoing training and supervision.

Psychologists specializing in treating personality disorders need to examine their level of commitment, resources, and time. A psychologist in part-time practice might not have the availability necessary to handle inevitable crises that emerge. An honest self- evaluation is critical. Do you have the available time, back up, access to psychopharmacology, hospital privileges, and the emo- tional energy to devote to complex cases?


The Challenge of Delivering Multiple Modalities of Treatment

One of the “sacred” clinical taboos that came from psychody- namic tradition and still permeates the field is the issue of thera- peutic fidelity, which required that the sanctity of the therapeutic relationship not be “corrupted” by outsiders. The individual ther- apeutic frame above all needed to be preserved so that the uncon- scious was not contaminated. Family systems theory challenged this notion with the assumption that we are all part of a system (Melito, 2006) and that pretending that this is not the case limits the range of possible clinical  interventions.

Many individuals with personality dysfunction are notoriously difficult to treat, and no single modality or approach can be expected to be good for all. An individualized approach is necessary, and therapeutic flexibility is a well-grounded principle (Critchfield & Benjamin, 2006). It is a common occurrence in small towns and for those who use a family practice model of psychology for a psychologist to treat multiple members of a family. This might include treating multiple family members with various modalities of treatment. There are surprisingly few stan- dards with sufficient evidence base to inform practice or even a database to show practice patterns. It has been my experience in conducting a practice with a specialty in treating personality dys- function in children, adolescents, and adults, that a family-based model is highly valued and leads to positive outcomes. In fact, I suggest that the ability to follow families and individuals, treating them over the family life cycle with modalities in which one has been trained, enhances treatment  outcome.

Of course, the psychologist must weigh each decision in terms of the potential risk of harm. A common problem often emerges when treating a couple and determining that one or both needs a phase of individual treatment to address their personality patterns in a more focused manner. Can the same psychologist see both? What are the ethical considerations? What happens if they have both developed a collaborative therapeutic alliance with the psy- chologist? Which one should be referred to someone else? How does the decision making change if they both seem to need only brief focused treatment, or if one clearly needs long-term treatment and the other ideally needs to be in a supportive role. What if one member of the family has a clearly diagnosable personality disor- der? Should this be identified, and who would benefit from this information? The most ethical manner in which to handle these complex clinical decisions is to use informed consent. This entails an open discussion with the patient(s) in which a presentation of any evidence that would inform any decision is shared and, more important, the possible benefits and risks are openly explored. This is often a very therapeutic intervention in itself which honors the patient(s) and encourages a collaborative treatment model essential in positive alliance maintenance.

In my experience, there are no certainties in what is “ethical,” except for what serves the developmental needs and clinical re- quirements of the individual and is within extant ethical standards. Regarding the issue of providing multimodal treatment, in some cases it seems best to refer to another psychotherapist, and in others it is preferable for the psychologist to continue to work with both. These ethical issues are routinely encountered by psychologists who practice in rural areas, but virtually all psychologists in clinical practice will have to make decisions based on a consideration of the benefit and risk. For example, in many communities, and even in larger metropolitan areas, a psychologist is considered a vital member of the professional community and will often be asked to treat people with whom they might share various roles. This is inevitable, and the stance you need to take needs to be contem- plated. The primary consideration is to be very clear about bound- aries so that each person is respected in how they wanted to be treated outside of the treatment office. I often ask how they want me to handle outside the office contacts at the beginning of the professional relationship. Until we have a solid evidence base to inform our practice, we need to use what we have and focus our attention on maintaining a therapeutic alliance with clear goals and expectations for roles and  treatment.

The Dearth of Evidence Base for Developing Treatment Packages

A treatment package is the way in which various aspects of treatment are combined, including format (brief, short term, long term, intermittent), length and frequency of sessions, therapeutic approach (psychodynamic, cognitive– behavioral, interpersonal, integrative, etc.), modalities (individual, couples, family, group, etc.), and how treatment modalities are delivered (sequentially or combined; Livesley, 2003; Magnavita, 2008). For example, when should individual psychotherapy be augmented with couples, group, pharmacological, or family  treatment?

Developing a collaborative treatment package that will enhance outcome has little in the way of empirical basis. We do not have the answers to many of the basic questions that we take for granted. For example, the standard 45–50-min format for psycho- therapy sessions has no known empirical basis. It might be that longer sessions are more potent or that, for some individuals, shorter sessions on a more frequent basis are optimal. We also have little data on how treatment is enhanced when we combine modalities. It seems that, for some disorders such as anxiety and depression, the outcome is optimized. It may be so for personality disorders, but the data are limited. We also do not know whether  it is more effective to have one psychotherapist who delivers multiple modalities of treatment or multiple psychotherapists, or what circumstances make differential decisions  sound.


Ethical  Boundary Management

Probably the aspect of treating personality dysfunction requiring the most clinical consideration is the awareness of and manage- ment of boundary issues: Flexibility is key (Critchfield & Ben- jamin, 2006). This may sound paradoxical, but a balance between flexibly responding to the genuine needs of the patient and main- taining firm boundaries in critical areas is central to competent and ethical treatment. Boundary issues take many forms, from exces- sive phone calls to unreasonable demands on the therapeutic relationship for availability and accessibility. Each psychologist needs to evaluate his or her own capacities and ascertain how each patient should be ethically responded to when there is an issue that taxes the therapeutic boundary. There are so many issues to ex- plore that are beyond the scope of this brief article, such as the use of touch, self-disclosure, dual roles, and the like, that need to be evaluated, always erring on the side of the best interest of the patient and the dictum “Do no harm” (and, where possible, do as much good as you can to reduce the suffering of those who seek our  expert assistance).



The ethical treatment of those presenting with personality dys- function is an area in which relatively few guidelines based on evidence are available to assist in the multiple clinical decisions that are required. Ethical treatment of anyone suffering from a complex disorder is going to require careful risk– benefit analysis in collaboration with the patient on multiple issues that affect the course and outcome of treatment. It is essential that a collaborative relationship be continually honored, informed consent be given, and (where available) empirical evidence utilized to make optimal treatment decisions. We now know that personality disorders are responsive to treatment and that there are several approaches that have a solid empirical foundation available to the psychologist. There are dialectics among multiple roles that one must hold when operating within any community in the role of therapeutic change agent. One must be able to handle the paradox of being both professional and “real” person.

As the reader will have surmised, the treatment of personality dysfunction entails managing complexity while maintaining a strong sense of purpose aimed toward enhancing functioning, improving quality of life, and preventing negative outcomes where possible. Here are some of the issues identified in this article that face psychologists who treat patients with personality dysfunction and those considering establishing this as an area of specialization:

  • A central issue that is facing the psychology is the question of what constitutes an acceptable level of competence to treat per- sonality disorders. We have not yet established specific guidelines that represent a sufficient level of education and training. Is it necessary to do so? On the other hand, are psychologists compe- tent by the nature and extent of their training to treat this ubiqui- tous condition?
  • Another issue has to do with models of clinical decision making and what is needed by clinicians. Clinical decision making is what separates psychological professionals from untrained or inadequately trained practitioners. It is clear that clinical decision making requires empirical evidence to guide the process. How- ever, for many basic decisions, psychologists have little in the way of empirical evidence to guide the process. What are clinicians supposed to do while awaiting the research? Many of us face these clinical presentations daily in our practices and out of necessity make decisions about multiple aspects of
  • Another issue has to do with the multiple treatment needs that patients with personality dysfunction often have and potential conflict of interest due to different perspectives and training of other professionals as well as other psychologists. There are often multiple clinicians with differing perspectives who work with the same patient, sometimes at cross-purposes. A psychopharmacolo- gist might have a strong belief that medication is essential to treatment, and the psychologist might believe that minimal amounts allow for optimal treatment response. This tension among various health professionals is inherent when treating personality disorders and needs to be addressed in open communication among all parties. It is important to understand the nature of a multidisciplinary treatment team and work together for the pa- tient’s best interest, but this is not always smooth, and it might be that the tension and resolution is best for optimal outcome when there is uncertainty about what is the “best”
  • Another issue has to do with boundaries at multiple levels. There are no clear-cut “cookbook” answers. Issues about boundary management must be considered on a case-by-case basis. Some- times flexible boundaries are optimal and at other times might even be harmful. Each situation needs to be considered using risk– benefit analysis and discussion with the patient. The issue of what information should be shared when working with couples is another complicated boundary issue. In some cases, information sharing allows for a better functioning treatment team. At other times, boundaries should not be fluid and information needs to be safeguarded. For example, when personal information is requested

by someone evaluating the patient, a decision might be made to withhold information in the patient’s best interest.

These and other issues relevant to the treatment of patients with personality dysfunction and comorbid conditions are addressed in the expert commentaries to  follow.



American Psychiatric Association. (1994). Diagnostic and statistical man- ual of mental disorders (4th ed.). Washington, DC: Author.

Castonguay, L. G., & Beutler, L. E. (2006). Principles of therapeutic change that work. New York: Oxford Press.

Critchfield, K. K., & Benjamin, L. S. (2006). Principles for psychosocial treatment of personality disorders: Summary of the APA Division 12 Task Force/NASPR Review. Journal of Clinical Psychology, 62, 661– 674.

Dimaggio, G., & Norcross, J. C. (2008). Treating patients with two or more personality disorders: An introduction. Journal of Clinical Psychology: In Session, 64, 127–138.

Hilsenroth, M. J., & Cromer, T. D. (2007). Clinical interventions related to alliance during the initial interview and psychological assessment. Psy- chotherapy: Theory, Research, Practice, Training, 44, 205–218.

Livesley, W. J. (Ed.). (2001). Handbook of personality disorders: Theory, research, and treatment. New York: Guilford Press.

Livesley,  W.  J.  (2003).  Practical  management  of  personality disorder.

New York: Guilford Press.

Magnavita, J. J. (1998). Methods of restructuring personality disorders with comorbid syndromes, Journal of Clinical Psychology: In Session, 4, 73– 89.

Magnavita, J. J. (Ed.). (2004). Handbook of personality disorders: Theory and practice. Hoboken, NJ:  Wiley.

Magnavita, J. J. (2006). In search of the unifying principles of psychother- apy: Conceptual, empirical, and clinical convergence. American Psy- chologist,  61, 882–982.

Magnavita, J. J. (2008). Using the MCMI in general treatment planning. In

  1. Millon & C. Bloom (Eds.), The Millon inventories (pp. 327–346). New York: Guilford Press.

Magnavita, J. J. (Ed.). (in press). Evidence-based treatment of personality dysfunction: Principles, processes, and strategies. Washington, DC: American  Psychological Association.

Melito, R. (2006). Integrating individual and family therapies: A structural- developmental approach. Journal of Psychotherapy Integration,  16,  346 –381.

Soloff, P. H., Fabio, A., Kelly, T. M., Malone, K. M., & Mann, J. J. (2005). High-lethality status in patients with borderline personality disorder. Journal of Personality Disorders, 19,  386 –399.

Stone, M. H. (2009). Perspective: Pitfalls in the psychotherapy of border- line personality or the triumph of faith over fact. Journal of Personality Disorders,  23, 3–5.

Weismann, M. M. (1993). The epidemiology of personality disorders: A 1990 update. Journal of Personality Disorder, 7(Suppl.),  44 – 62.

Yeomans, F., Gutfreund, J., Selzer, M., Clarkin, J., Hull, J., & Smith, T. (1994). Factors related to drop-outs by borderline patients. The Journal of Psychotherapy Practice and Research, 3,  16 –24.


Practice at the Border: The Art of Integrating the Science on Personality Disorders  for Clinical Practice

Kenneth  N. Levy

Magnavita provides a great service by highlighting a number of important  issues  in  the  treatment  of  patients  with  personality disorders (PDs). He stresses that most psychologists, regardless of setting, encounter people with PDs on a regular basis. He then elaborates the ethical quandary resulting from the evidence on best practices for PDs being relatively scant at this time. He then interprets the existing literature and explicates important recom- mendations to help guide clinicians. All in all, this paper represents a significant contribution, and I want simply to amplify and clarify some of the points  made.

As Magnavita contends, all too often the definition of “evi- dence” is limited to results from randomized controlled trials (RCTs; Task Force on Promotion and Dissemination of Psycho- logical Procedures, 1995). Despite the importance of RCT designs, to focus only on data from RCTs overlooks broader evidence that is also relevant for clinical practice. Along these lines, a number of authors (e.g., Clarke & Oxman, 1999; Gabbard, Gunderson, & Fonagy, 2002; Leichsenring, 2004; Levy & Scott, 2007; Nathan & Gorman, 2002) have suggested that integrating evidence from multiple sources is necessary to build an empirically grounded framework. Similarly, others have focused on empirically sup- ported principles of treatment rather than credentialed, trade- marked, brand-name, or evidence-based treatment packages (Cas- tonguay & Beutler, 2006; Rosen & Davidson, 2003). Others have noted that naturalistic and practice research network studies can bridge the gap between practice and research (Borkovec, 2004; Seligman, 1996). Limiting findings to those from RCTs may impede the exploration of potentially fruitful and novel avenues of study and neglects important findings from other sources that might be useful in guiding clinical practice.

The importance of training in PDs is evinced by epidemiological data, which indicate that PDs have a high overall lifetime preva- lence ranging between 10% and 14% in the community (Skodol et al., 2002). In an outpatient sample, Zimmerman, Rothschild, and Chelminski (2005) found that 45.5% of patients met criteria for a PD. At The Pennsylvania State University clinic, a large commu- nity mental health center, we found a similar prevalence rate of 46%. Common and serious PDs include borderline personality disorder (BPD), PD not otherwise specified, antisocial PD, and narcissistic PD.

In addition to being highly prevalent, PDs are commonly co- morbid with a range of disorders, particularly major depression, dysthymia, bipolar disorder, anxiety disorders, posttraumatic stress disorder (PTSD), eating disorders, and substance abuse (Zanarini, Frankenberg, Hennen, Reich, & Silk, 1998). Zanarini refers to this pattern of comorbity as complex comorbidity because of the high number of comorbid diagnoses and the co-occurrence of both internalizing (e.g., depression) and externalizing (e.g., substance use) disorders.

were not followed by improvements in BPD (Gunderson, Morey, Stout, et al., 2004; Klein & Schwartz, 2002; Links et al, 1995). Finally, several studies have shown that the efficacy of treatment of PTSD is significantly reduced when the patient has comorbid BPD (Cloitre & Koenen, 2001; Feeny, Zoellner, & Foa, 2002).

These findings strongly suggest at least two evidence-based principles. The first is that given the prevalence of PDs, all clinicians need to develop specific expertise in at least identifying and diagnosing patients with these disorders. Second, given that PDs are so frequently comorbid with a range of Axis I disorders, whenever a clinician determines that a patient meets criteria for one of these common comorbid Axis I disorders, it is incumbent upon that clinician to assess for PDs because it would likely affect the course and treatment of the disorder. Failing to do so could be considered derelict.

The good news is that once a PD is identified,  there  are  several treatment options, both cognitive-behavioral and psy- chodynamic, that show potential. For BPD, these include dia- lectical behavior therapy (DBT; Linehan, Armstrong, Suarez, Allmon, & Heard, 1991, Linehan et al., 2006), mentalization-based therapy (Bateman & Fonagy, 1999, 2001, 2008, in press), schema- focused psychotherapy (Giesen-Bloo et al., 2006), transference- focused psychotherapy (Clarkin, Levy, Lenzenweger, & Kernberg, 2007; Levy et al., 2006), a specific form of CBT (Davidson et al.), and STEPPS (Blum, Pfohl, St. John, Monahan, & Black, 2002). Additionally, a number of treatments have shown efficacy in treating PDs generally (Abbass, Sheldon, Gyra, & Kalpin, 2008; Muran, Safran, Samstag, & Winston, 2005; Safran, Muran, Samstag, & Win- ston, 2005; Svartberg, Stiles, & Seltzer, 2004; Winston, Laikin, Pol- lack, Samstag, McCullough, & Muran, 1994; Winston, Pollach, McCullough, Flegenheimer, Kestenbaum, & Trujullo,  1991).

The results of these efficacy studies suggest important evidence- based principles. First, PDs (including BPD) are treatable disor- ders. Second, because PDs are chronic disorders, they require ongoing or long-term treatments. This seems especially true of BPD, for which all efficacious approaches conceptualize treatment as a multiyear process. Third, psychotherapists have a range of options across several orientations available to them, and it is premature to foreclose on any one of the available options that have been tested. Although there have been few direct compari- sons, enough data now exist to suggest that no one approach is superior to another (despite special endorsements of DBT).1 Most important, commonalities across these treatments suggest several important guidelines for clinicians, such as providing a structured, coherent treatment, being in supervision, paying particular atten-

When  comorbid,  PDs  negatively  affect  the  course  of  these  disorders and the outcome of otherwise efficacious treatments. Bipolar patients with comorbid PDs are less employed, use more medications, have increased rates of alcohol and substance use disorders, show poorer treatment response, and have significantly worse interepisode functioning than bipolar patients who are not afflicted with PDs (Bieling, Green, & Macqueen, 2007). It is interesting to note that the reverse is not true: A comorbid bipolar disorder does not affect the course or outcome for PD patients (Gunderson, Morey, Stout, et al., 2006). Similarly, several studies have found that improvements in BPD were often followed by improvements in depression but that improvements in depression

1 Several managed care companies now define special benefits for DBT. Several state departments of mental health (Illinois, Connecticut, Massa- chusetts, New Hampshire, North Carolina, and Maine) have now enthusi- astically endorsed and subsidized DBT as the treatment of choice for BPD. Although DBT appears to be a solid treatment and has marshaled a large amount of evidence for its efficacy—more evidence than any other treat- ment for BPD to date—it would be unfortunate to close off or limit access to the various other treatments which have also been shown to be effica- cious and promising. Only about 50 – 60% of patients improve in any of the treatments, and it is likely that there are patient characteristics that will predict which patients would do better in which treatment (Levy, 2008; Levy, Wasserman, Meehan, & Clarkin,  2008).

One point for clarification is Magnavita’s contention that psy- chologists are generally well trained in the assessment and diag- nosis of PDs. I question this assumption. It is my experience that clinicians across all disciplines, including psychology, are reluc- tant to give PD diagnoses and have difficulty making differential diagnoses between PDs (especially BPD) and a number of Axis I disorders, particularly bipolar spectrum disorders and complex PTSD. In fact, it seems that clinicians tend to prefer to confer these Axis I diagnoses over PDs, despite their questionable validity (Paris, 2007), or tend to hedge by using terms such as Axis II deferred, or personality disorder  traits.

Clinicians may be reluctant to diagnose PDs because Axis I diag- noses are more familiar and making Axis II diagnoses requires expe- rience and clinical judgment (Paris, 2007). Misconceptions that PDs are untreatable or are pejorative may contribute this reluctance. Stud- ies have found that mental health professionals report avoiding or actively disliking patients with BPD and have difficulty disclosing BPD diagnoses (McDonald-Scott et al., 1992; Pfohl et al., 1999). Other studies also suggest that clinicians do not recognize and diagnose personality disorders in ordinary clinical practice. Mey- erson (2009) recently reported that 74% of patients in a study for BPD  had  previously  been  misdiagnosed  despite  an  average of

10.44 years since their first psychiatric encounters. The most common false-positive diagnoses were bipolar disorder (17%), depression (13%), and anxiety disorders (10%). Zimmerman and Mattia (1999) found that clinicians left to their own judgments diagnosed BPD in 0.4% of almost 500 patients seen, compared with 14.4% by structured interview. They also found that provid- ing clinicians with the findings from the structured interviews significantly increased the likelihood of the BPD diagnosis. In the clinic at Penn State (Levy, 2009), before we implemented semi- structured interviews for PDs, the prevalence rate for BPD was 1.6%, and 4.2% after supervision. After implementing semistruc- tured interviews for PDs in September 2008, the prevalence rate increased to about 20% for BPD, and, as mentioned earlier, we found that 46% met criteria for a PD. Clearly, an evidence-based principle is that structured interviews identify many cases of PDs missed in ordinary practice.

A corollary implication of these findings is that there is a need for specialized training in PDs. I know of no studies that specifi- cally address this issue; however, Norcross, Sayette, and Mayne (2008) provided data. In surveying 319 clinical PhD, PsyD, and counseling programs, they found that only 24 (7.5%) of the pro- grams reported having a faculty member with expertise in PDs, and only 7 (2%) programs indicated that they specialize in the training and treatment of PDs. In contrast, 25% of programs have  a faculty member with stated expertise in panic disorder, and just under 10% of programs have a panic disorder specialty clinic. The disparity is shocking, considering the difference in prevalence (45% for PDs vs. 10% for panic disorder in outpatients).

In conclusion, despite the relative paucity of data, important principles have emerged. Clinicians should (1) be knowledgeable about the epidemiology and psychopathology of PDs; (2) assess for PDs, especially when a patient presents with depression, bipo- lar disorders, substance use, chronic PTSD, or complex comorbid- ity; (3) obtain training in and utilize a range of empirically sup- ported treatments or refer to clinicians who have specific  training;

(4) join a supervision or intervision group if treating a patient with a PD; (5) pay close attention to the structure and frame of the treatment; (6) and be vigilant for indications of colluding with the patient, acting out, or iatrogenic  behaviors.

One reason that PD research is relatively scant is that it is underfunded at the federal level. About $10 million annually is spent on PD research, which is only 2% of what is allocated for research on schizophrenia and 6% of that for bipolar disorder, although prevalence rates show the opposite trend (20% for PDs, 1.6% for bipolar disorder, and 0.4% for schizophrenia). In addition to limiting research, this lack of funds contributes to a critical shortage of young investigators.

The cliche´ that “more research is needed” is clearly true in the case of PDs; however, more awareness and training are also desperately needed if psychologists hope to help this large segment of  the population.



Abbass, A., Sheldon, A., Gyra, J., & Kalpin, A. (2008). Intensive short- term dynamic psychotherapy for DSM-IV personality disorders: A ran- domized controlled trial. The Journal of Nervous and Mental Disease, 196, 211–216.

Bateman, A., & Fonagy, P. (1999). Effectiveness of partial hospitalization in the treatment of borderline personality disorder: A randomized con- trolled trial. American Journal of Psychiatry, 156, 1563–1569.

Bateman, A., & Fonagy, P. (2001). Treatment of borderline personality disorder with psychoanalytically oriented partial hospitalization: An 18-month follow-up. American Journal of Psychiatry, 158,  36 – 42.

Bateman, A., & Fonagy, P. (2008). 8-year follow-up of patients treated for borderline personality disorder: Mentalization-based treatment versus treatment as usual. The American Journal of Psychiatry, 165, 631– 638. Batman, A., & Fonagy, P. (2009). Randomized controlled trial of outpa- tient mentalization-based treatment versus structured clinical manage- ment for borderline personality disorder. American Journal of Psychia-

try, epub ahead of  print.

Bieling, P. J., Green, S. M., & MacQueen, G. (2007). The impact of personality disorders on treatment outcome in bipolar disorder: A re- view. Personality and Mental Health, 1, 2–13.

Blum, N., Pfohl, B., St. John, D., Monahan, P., & Black, D. W. (2002). STEPPS: A cognitive-behavioral systems-based group treatment for outpatients with borderline personality disorder—A preliminary report. Comprehensive Psychiatry, 43, 301–310.

Borkovec, T. D. (2004). Research in training clinics and practice research networks: A route to the integration of science and practice. Clinical Psychology: Science and Practice, 11,  212–215.

Castonguay, L. G., & Beutler, L. E. (2006). Principles of therapeutic change: A task force on participants, relationships, and techniques factors. Journal of Clinical Psychology, 62,  631– 638.

Clarke, M., & Oxman, A. (1999). Cochrane reviews will be in Medline. British Medical Journal, 319, 1430 –1437.

Clarkin, J. F., Levy, K. N., Lenzenweger, M. F., & Kernberg, O. F. (2007). Evaluating three treatments for borderline personality disorder: A mul- tiwave study. American Journal of Psychiatry, 164, 922–928.

Cloitre, M., & Koenen, K. C. (2001). The impact of borderline personality disorder on process group outcome for women with posttraumatic stress disorder related to childhood abuse. International Journal of Group Psychotherapy, 51, 379 –398.

Davidson, K., Norrie, J., Tyrer, P., Gumley, A., Tata, P., Murray, H., & Palmer, S. (2006). The effectiveness of cognitive behavior therapy for borderline personality disorder: Results from the Borderline Personality Disorder Study of Cognitive Therapy (BOSCOT) trial. Journal of Per- sonality Disorders, 20, 450 – 465.

Feeny, N. C., Zoellner, L. A., & Foa, E. B. (2002). Treatment outcome for chronic PTSD among female assault victims with borderline personality characteristics: Preliminary examination. Journal of Personality Disor- ders, 16, 30 – 40.

Gabbard, G. O., Gunderson, J. G., & Fonagy, P. (2002). The place of psychoanalytic treatments within psychiatry. Archives of General Psy- chiatry,  59, 505–510.

Giesen-Bloo, J., Van Dyck, R., Spinhoven, P., Van Tilburg, W., Dirksen, C., Van Asselt., Kremers, I., et al. (2006). Outpatient psychotherapy for borderline personality disorder: A randomized trial of schema-focused therapy vs transference-focused psychotherapy. Archives of General Psychiatry, 63, 649 – 658.

Gunderson, J. G., Morey, L. C., Stout, R. L., Skodol, A. E., Shea, M. T., McGlashan, T. H., . . . Bender, D. S. (2004). Major depressive disorder and borderline personality disorder revisited: Longitudinal interactions. Journal of Clinical Psychiatry, 65,  1049 –1056.

Gunderson, J. G., Weinberg, I., Daversa, M. T., Kueppenbender, K. D., Zanarini, M. C., Shea, M. T., . . . Dyck, I. (2006). Descriptive and longitudinal observations on the relationship of borderline personality disorder and bipolar disorder. American Journal of Psychiatry, 163, 1173–1178.

Klein, D. N., & Schwartz, J. E. (2002). The relation between depressive sympotoms and borderline personality disorder features over time in dysthymic disorder. Journal of Personality Disorders, 16, 523–535.

Leichsenring, F. (2004). Quality of depressive experiences in borderline personality disorders: Differences between patients with borderline per- sonality disorder and patients with higher levels of personality organi- zation. Bulletin of the Menninger Clinic, 68, 9 –22.

Levy, K. N. (2008). Psychotherapies and lasting change. American Journal of Psychiatry, 165, 556 –559.

Levy, K. N. (2009). [Prevalence and reliability of Axis I and II diagnoses in a community mental health university based training clinic]. Unpub- lished  raw data.

Levy, K. N., Meehan, K. B., Kelly, K. M., Reynoso, J. S., Weber, M., Clarkin, J. F., & Kernberg, O. F. (2006). Change in attachment patterns and reflective function in a randomized control trial of transference- focused psychotherapy for borderline personality disorder. Journal of Consulting and Clinical Psychology, 74,  1027–1040.

Levy, K. N., & Scott, L. N. (2007). The “art” of interpreting the “science” and the “science” of interpreting the “art” of the treatment of borderline personality disorder. In S. Hoffman & J. Weinberger (Eds.), The art and science of psychotherapy (pp. 269 –299). London:  Brunner-Routledge.

Levy, K. N., Wasserman, R. H., Meehan, K. B., & Clarkin, J. F. (2008, September). Reflective function as a moderator of dropout in three treatments for BPD. K. Levy (Chair), Predictors and Processes in Psy- chotherapy for Borderline Personality Disorder. Symposium conducted at the meeting of the North American Society for Psychotherapy Re- search, New Haven, CT.

Linehan, M. M., Armstrong, H. E., Suarez, A., Allmon, D., & Heard, H. L. (1991). Cognitive-behavioral treatment of chronically parasuicidal bor- derline patients. Archives of General Psychiatry, 48, 1060 –1064.

Linehan, M. M., Comtois, K. A., Murray, A. M., Brown, M. Z., Gallop,

  1. J., Heard, H. L., Korslund, K. E., et al. (2006). Two-year randomized controlled trial and follow-up of dialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Archives of General Psychiatry, 63, 757–766.

Links, P. S., Heslegrave, R. J., Mitton, J. E., van Reekum, R., & Patrick,

  1. (1995). Borderline psychopathology and recurrences of clinical dis- orders. Journal of Nervous and Mental Disease, 183, 582–586.

McDonald-Scott, P., Machizawa, S., & Satoh, H. (1992). Diagnostic dis- closure: A tale in two cultures. Psychological Medicine, 22, 147–157.

Meyerson, D. (2009, May). Is borderline personality disorder underdiag- nosed? Paper presented at the 162nd annual meeting of the American Psychiatric Association, San Francisco,  CA.

Muran, J. C., Safran, J. D., Samstag, L. W., & Winston, A. (2005). Evaluating an alliance-focused treatment for personality disorders. Psy- chotherapy: Theory, Research, Practice, Training, 42, 532–545.

Nathan, P. E., & Gorman, J. M. (Eds.). (2002). A guide to treatments that work (2nd ed.). New York: Oxford University Press.

Norcross, J. C., Sayette, M. A., & Mayne, T. J. (2008). Insider’s guide to graduate programs in clinical and counseling psychology (2008/2009 ed.). New York: Guilford  Press.

Paris, J. (2007). Why psychiatrists are reluctant to diagnose borderline personality disorder. Psychiatry, 4,  35–39.

Pfohl, B. M., Gunderson, J. G., Silk, K. R., Zimmerman, M., Williams,

  1. B., Phillips, K. A., et al. (1999, May). Survey of clinician attitudes towards BPD. Paper presented at the 152nd annual meeting of the American Psychiatric Association, Washington, DC.

Rosen, G. M., & Davison, G. C. (2003). Psychology should list empirically supported principles of change (ESPs) and not credential trademarked therapies or other treatment packages. Behavior Modification, 27, 300 – 312.

Safran, J. D., Muran, J. C., Samstag, L. W., & Winston, A. (2005). Evaluating alliance-focused intervention for potential treatment failures: A feasibility study and descriptive analysis. Psychotherapy: Theory, Research, Practice, Training, 42,  512–531.

Seligman, M. E. P. (1996). Science as an ally of practice. American Psychologist,  51, 1072–1079.

Skodol, A. E., Gunderson, J. G., McGlashan, T. H., Dyck, I. R., Stout,

  1. L., Bender, D. S., Grilo, C. M., et al. (2002). Functional impairment in patients with schizotypal, borderline, avoidant, or obsessive- compulsive personality disorder. American Journal of Psychiatry, 159, 276 –283.

Svartberg, M., Stiles, T. C., & Seltzer, M. H. (2004). Randomized, con- trolled trial of the effectiveness of short-term dynamic psychotherapy and cognitive therapy for cluster C personality disorders. American Journal of Psychiatry, 161, 810 – 817.

Task Force on Promotion and Dissemination of Psychological Procedures. (1995). Training and dissemination of empirically validated psycholog- ical treatments: Report and recommendations. The Clinical Psycholo- gist,  48, 3–23.

Winston, A., Laikin, M., Pollack, J., Samstag, L. W., McCullough, L., & Muran, J. C. (1994). Short-term psychotherapy of personality disorders. American Journal of Psychiatry, 151,  190 –194.

Winston, A., Pollack, J., McCullough, L., Flegenheimer, W., Kestenbaum, R., & Trujillo, M. (1991). Brief psychotherapy of personality disorders. Journal of Nervous and Mental Disease, 179, 188 –193.

Zanarini, M., Frankenberg, F., Hennen, J., Reich, D., & Silk, K. (2004). Axis I comorbidity in patients with borderline personality disorder: 6-year follow-up and prediction of time to remission. American Journal of Psychiatry, 161, 2108 –2114.

Zimmerman, M., & Mattia, J. L. (1999). Differences between clinical and research practices in diagnosing borderline personality disorder. Amer- ican Journal of Psychiatry, 156,  1570 –1574.

Zimmerman, M., Rothschild, L., & Chelminski, I. (2005). The prevalence of DSM-IV personality disorders in psychiatric outpatients. American Journal of Psychiatry, 62,  1911–1918.


Calibrating Research and Training Paradigms To Enhance Ethical Decision Making in Treatment of Personality Disorder

Kenneth  L. Critchfield

Magnavita raises important points about ethical provision of treatment to patients diagnosed with personality disorder (PD). It is well known that PDs pose considerable challenges to clinicians because of their severity, chronicity, and complexity. Stakes are often high and can include self-harm, suicidality, homicidality, or neglect and abuse of others. Magnavita notes correctly that little direct research is available to guide most clinical decision-making around personality disorder treatment at the level of specific in- tervention choices. He makes a welcome and straightforward case about the need for clinicians to consider ethical issues and weigh risks versus benefits for individual  patients.

The fact that there is little empirical data about many basic treatment decisions is itself alarming, given that roughly half of patients seen in community treatment settings qualify for PD diagnosis (Keown, Holloway, & Kuipers, 2002). Despite obvious need, research funding priorities have not focused much on devel- oping psychosocial treatments for PD, and research has accumu- lated only slowly. As a result, risks associated with the kinds of problems posed by Magnavita are left for clinicians to address with clinical experience, consultation with experienced others, and the rich but largely untested body of clinical theory focused on per- sonality. In this commentary, I extend Magnavita’s emphasis on the ethical responsibilities of clinicians to note that a related burden also rests more generally on our field. We need to pursue more vigorously the kinds of data, and offer the kinds of training, that will best prepare clinicians to work productively with this ubiquitous and challenging set of  disorders.


State of the Art in PD Treatment Research

Randomized control trial (RCT) methodology has been essential to provide clear demonstration that personality disorder can be treated successfully with psychosocial treatments. Other designs are available, but RCTs are the gold standard for treatment re- search because of the use of random assignment as a key meth- odological control. Internal validity is maximized through use of a treatment manual and related measures that verify adherent admin- istration. RCT methodology directly answers the question of whether a given treatment package produces a statistically signif- icant average effect on outcome. RCTs usually focus on a homog- enous group of patients defined by presence of only a single disorder, but PD treatment research has tended to allow more comorbidity on both axes of the Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM–IV; American Psychiatric Association, 1994). This has enhanced the ability to apply results in typical practice settings. Borderline personality disorder has received the most research attention, including several recently published RCTs (e.g., Bateman & Fonagy, 2008; Clarkin, Levy, Lenzenweger, & Kernberg, 2007; Giesen-Bloo et al., 2006; Line- han et al., 2006). A global view of results to date suggests that multiple approaches to treatment produce beneficial effects, de- spite diverse conceptualizations of personality and related mech- anisms  of change.

Principles have been identified that seem to apply across these treatment approaches. A Task Force convened by Castonguay and Beutler (2006) recently summarized principles having empirical support. For treatment of PD, these include the need to establish a strong therapeutic relationship. In addition, psychotherapists are instructed to be patient, flexible, creative, and relatively active in sessions, placing primary focus directly on presenting problems and concerns. Early formulation of problems and transparent discussion with patients about the treatment approach and rationale are also recommended. Additional principles and related discus- sion of details are available in summary form (Critchfield & Benjamin, 2006).

Despite important progress, many substantial challenges remain. For example, a strong therapeutic alliance has the best empirical support in its connection with outcome, but little is known about how to achieve it with diverse PD patients whose problems already involve difficulties relating well with others. Heterogeneity within each PD criteria set combines with comorbidity to produce a bewildering array of clinical profiles unlikely to be directly ad- dressed in RCT research, even if they are eventually organized according to one of several proposed alternative taxonomic frame- works for Axis II of the DSM–IV.


Example of an Alternative Research Paradigm Focused on Tailored Treatment of  Individuals

Clinicians need guidance for how to tailor psychotherapy in the face of diverse patient presentations. The challenge to researchers is how to supplement RCT results and provide information to guide such “flexibility.” One example of an approach that focused on “prescribed treatment variability” is found in work that my colleagues and I have been doing among a sample of patients referred in an inpatient setting for nonresponse to past interven- tions. Our referrals show high levels of diagnostic comorbidity, frequent rehospitalization, and high levels of dysfunction and suicidality. Axis II disorders are assessed with structured research interviews and characterize nearly all these referrals. The approach to treatment is Interpersonal Reconstructive Therapy (IRT; Ben- jamin, 2003). IRT offers a theory of psychopathology that is rooted in principles of attachment and emphasizes an individual’s inter- personal learning history as a basis for understanding how and why current problems persist despite repeated attempts to change. Psy- chotherapists can only be judged as “adherent” to IRT if interven- tions and in-session relationships are consistent with the particular patient’s case formulation and related treatment goals. Any inter- vention is considered appropriate only if it is adherent to IRT principles, makes sense in light of the case formulation, and supports progress toward agreed-upon therapy goals. Methods for handling crisis are also tailored relative to the case formulation. Thus, clinicians are encouraged to be “flexible and creative,” but not in random or capricious ways. The adherence measure reflects consistency of intervention choice with respect to underlying prin- ciples for change rather than specifying uniform use of particular techniques. Variability in adherence, operationalized in this man- ner, is expected to covary precisely with therapy outcome.

IRT research results have been promising so far. In pilot work, trained clinicians show remarkable ability to agree on central elements of the case formulation, and reliably to rate sessions for adherence to IRT principles (Critchfield, Davis, Gunn, & Ben- jamin, 2008). In a small sample of cases studied, adherence ap- pears highly correlated with retention in therapy, progress address- ing motivational elements associated with IRT, and change in symptoms, even after controlling for general empathy and collab- oration/alliance. As we expand our sample of treated cases, this approach should allow us to determine whether specific governing principles for intervention choice in IRT are associated with ex- pected changes in treatment across diverse patient profiles.


Training Ethical Application of Treatment Principles for Individuals

Most training models correctly emphasize empirically based research. However, knowledge of data points and how to follow a manual are no substitute for skills at critical thinking, conceptual integration, or ethical discernment. Optimally, data serve to inform and enhance those more central clinical skills. Training in the IRT clinic focuses almost relentlessly on critical and integrative think- ing applied with specific cases. The supervisory approach includes one-on-one videotape review and clinician self-rating of adherence measures, so that moment-by-moment intervention choices and particular case features are all considered in light of the broader conceptual model. I believe that training models such as these, which focus on skills in case formulation and empathic relating, in translating global principles into specific actions, and in appropri- ately generalizing familiar principles to new challenges, are needed to best serve our patients. These skills are necessary to make best use of the accumulating database surrounding PD treat- ment research. Magnavita’s emphasis on risk– benefit ratios, psy- chotherapist responsibility, flexibility, and collaboration with pa- tients demonstrates these deeper clinical skills well and is a welcome contribution.




Bateman, A., & Fonagy, P. (2008). 8-year follow-up of patients treated for borderline personality disorder: Mentalization-based treatment versus treatment as usual. American Journal of Psychiatry, 165, 631– 638.

Benjamin, L. S. (2003). Interpersonal diagnosis and treatment of person- ality disorder. New York: Guilford  Press.

Castonguay, L. G., & Beutler, L. E. (2006). Principles of therapeutic change that work. New York: Oxford University Press.

Clarkin, J. F., Levy, K. N., Lenzenweger, M. F., & Kernberg., O. K. (2007). Evaluating three treatments for borderline personality disorder: A multiwave study. American Journal of Psychiatry, 164, 922–928.

Critchfield, K. K., & Benjamin, L. S. (2006). Principles for psychosocial treatment of personality disorders: Summary of the APA Division 12 Task Force/NASPR Review. Journal of Clinical Psychology, 62, 661– 674.

Critchfield, K. L., Davis, M. J., Gunn, H. E., & Benjamin, L. S. (2008, June). Measuring therapist adherence in interpersonal reconstructive therapy (IRT): Conceptual framework, reliability, and validity. Poster presented to the International Society for Psychotherapy Research, Bar- celona, Spain.

Giesen-Bloo, J., Van Dyck, R., Spinhoven, P., Van Tilburg, W., Dirksen, C., Van Asselt, Kremers, I., et al. (2006). Outpatient psychotherapy for borderline personality disorder: A randomized trial of schema-focused therapy vs. transference-focused psychotherapy. Archives of General Psychiatry, 63, 649 – 658.

Keown, P., Holloway, F., & Kuipers, E. (2002). The prevalence of per- sonality disorders, psychotic disorders and affective disorders amongst the patients seen by a community mental health team in London. Social Psychiatry & Psychiatric Epidemiology, 37,  225–229.

Linehan, M. M., Comtois, K. A., Murray, A. M., Brown, M. Z., Gallop,

  1. J., Heard, H. L., Korslund, K. E., et al. (2006). Two-year randomized controlled trial and follow-up of dialectical behavior therapy versus therapy by experts for suicidal behaviors and borderline personality disorder. Archives of General Psychiatry, 63, 757–766.


The Effective Treatment of Personality Disorders: Easily Within Our  Grasp

Jay  L. Lebow

Magnavita’s thoughtful article calls our attention to the special issues that psychotherapists face in the treatment of those with personality disorders. The empirical evidence is clear about the relationship of the presence of comorbid personality disorder to the treatment of other co-occurring difficulties. When comorbid per- sonality disorder is present, regardless of type, the rates of success of all treatments diminish remarkably across a wide range of presenting problems. For example, in the National Institute of Mental Health Treatment of Depression Collaborative Research Program (Shea & Elkin, 1996), the presence of personality disor- der was a far better predictor of outcome than was the treatment received or any other client characteristic. This clearly is a special population necessitating special adaptations in  treatment.

Magnavita nicely orients us to the issues involved in treating this diverse population. Although there clearly is a need for special expertise in treating these individuals, the state of the science thus far suggests that we have yet to construct (and probably will never do so) the perfect treatment for those with personality disorders. Nonetheless, treatments that are effective do share a number of core characteristics (Castonguay & Beutler, 2006). Most espe- cially, a look at the treatments that have been effective points to the need to adapt typical psychotherapy procedures treatments to work with these individuals. It seems clear that the treatments that do turn out to be effective, such as dialectical behavior therapy (DBT), overhaul typical notions of practice, with session lengths that vary, availability for crisis intervention, and the merging of various psychotherapy formats such as group and individual psy- chotherapy. However, psychotherapist skillfulness is crucial; one person’s flexibility readily becomes another’s error. The need for good training and supervision and for ongoing assessment of psychotherapy progress is clear. The psychotherapist in the manner of what Stricker (see Stricker & Trierweiler, 1995) described as the local clinical scientist must weigh the data about the client that emerge in the process of making treatment decisions. The clinician is necessarily in the crucial role of deciding when to do what with these clients (and when to back off). Positive outcomes can only be expected to emerge over time so there is a need for patience, yet also a need to be able to recognize when what is being offered is not engaging or working. An initial poor alliance almost always leads to treatment failure.

Personality disorders typically also especially vexing are for those living in close connection with that individual, and families must be considered in conceptualizing treatment (Lebow & Ulias- zek, in press). Family and couple relationships of those with personality disorders most often are quite strained, as most per- sonality disorders are disorders in relating (Snyder & Whisman, 2003).

Couple and family psychotherapies often become the treatments for those with personality disorder, because these formats may be the only means in which to get the person with personality disorder (who is often in a precontemplation state of change) into treatment. Alternately, these formats may be the primary source of treatment when a systems problem is identified as most salient, as when severe marital distress or postdivorce conflict over children comes fully into focus. In these situations, the couple or the family become the focus of treatment because of distress in the couple or family relationships, but the psychotherapist must deal not only with those issues but with the special difficulties of working with the individual with personality  disorder.

Psychotherapists who work with such systems need multiple competencies. They must know how to deal with couple processes such as communication and problem solving, the special aspects of working with individuals with personality disorder, and the special aspects of systems such as attachments of individuals with such disorders. There are threats on all sides to such work. Easily, the person with the personality disorder might come to occupy too powerful a position in the psychotherapy system, in which they hold their presence in the psychotherapy system hostage to their demands. At the other side of the continuum, I have also witnessed many cases in which the presence of personality disorder becomes the occasion for discounting the needs of the person with person- ality disorder (whose needs paradoxically are likely to be greater than those of others). Each of these situations makes for the script in which psychotherapy stops being therapeutic and becomes toxic for all. Couple/family psychotherapists limited to methods and processes that work with other systems will rarely succeed with these couples and families. It also is essential to note that, although the choice of a partner with personality disorder as a mate is innately problematic, solutions to simply leave the relationship that are frequently generated in the individual psychotherapy of part- ners of those with personality disorders may make for what are far from the best of solutions, especially because these fragile families often descend into kin wars and child custody conflict (Lebow, 2003).

A discussion of treatments for those with personality disorder would also be incomplete without reference to person of the psychotherapist. Personality-disordered people call for the main- tenance of the best of the Rogerian principles of empathy, positive regard, and congruence, often under a good deal of duress. Psy- chotherapists low in these characteristics are poorly matched to these clients. The mindful practice that is at the center of DBT is not simply about practice; it is also about a psychotherapist being able to maintain a mindful attitude in the wake of the complex feelings of the clients they see. Some psychotherapists simply are better than others in maintaining a psychotherapeutic stance to those with personality disorder.

So I echo Magnavita’s key points. Working with individuals, couples, and families with personality disorders requires special expertise and treatment plans that incorporate this special exper- tise. I also emphasize that, because personality disorders almost invariably have a marked effect on the partners and other close family members involved, any treatment of personality-disordered individuals must include competence in understanding the needs of these people as well, regardless of whether they are in treatment. When families are directly involved in treatments, there remain complex questions about whether the same or different psycho- therapists should treat the individual and the system. Whether they are part of the family treatment or not, psychotherapists need to maintain a systemic perspective that includes a view of how the treatment relates to the needs of these others in the lives of the personality-disordered individual.



Castonguay, L. G., & Beutler, L. E. (2006). Principles of therapeutic change that work. New York: Oxford University Press.

Lebow, J. (2003). Integrative family therapy for disputes involving child custody and visitation. Journal of Family Psychology, 17, 181–192.

Lebow, J. L., & Uliaszek, A. W. (in press). Family therapy for personality disorders: An evidence-based approach. In J. J. Magnavita (Ed.). Evidence-based treatment of personality dysfunction. Washington, DC: American  Psychological Association.

Shea, M., & Elkin, I. (1996). The NIMH treatment of depression collab- orative research program. In C. Mundt, M. J. Goldstein, K. Hahlweg,  &

  1. Fiedler (Eds.), Interpersonal factors in the origin and course of affective disorders (pp. 316 –328). London: Gaskell/Royal College of Psychiatrists.

Snyder, D. K., & Whisman, M. A. (2003). Treating difficult couples: Helping clients with coexisting mental and relationship disorders. New York:  Guilford Press.

Stricker, G., & Trierweiler, S. J. (1995). The local clinical scientist: A bridge between science and practice. American Psychologist, 50, 995– 1002.


Received November 23, 2008

Revision received August 31,  2009

Accepted September 1, 2009 •

Comments are closed.
%d bloggers like this: