20b
Brandon T. Unruh
Given my day job as a personality disorder (PD) specialist delivering the major long-term borderline PD treatments in adapted forms in a short-term residential PD treatment program, Dixon-Gordon, Conkey, and Woods’ exhaustive review (this volume) of studies on PD treatments with a duration of less than one year poses important questions about how best to shape and reshape the aims, structure, and techniques of PD treatments. This commentary aims to show that the growing work on short-term interventions summarized here pushes the state of PD treatment forward in two directions that are increasingly important given the limitations of long-term treatments: enhancing treatment accessibility and improving our understanding of how PDs change.
Short-Term PD Treatment Studies Will Improve Treatment Accessibility
Dixon-Gordon and colleagues contextualize their review of short-term PD treatment studies by summarizing how the implementation of empirically validated long-term PD treatments has been limited by their duration and cost, as well as the magnitude of time required by both clinicians (to establish and maintain adherence) and patients (to attend multiple treatment sessions per week). The authors are right to remind us that the public health demand is not being met by our current approach to disseminating PD treatments.
The authors suggest several ways in which the best of the short-term PD treatment studies can pave the way for improvements in accessibility. These strategies include: emphasizing more cost-effective group-based interventions, empowering an expanded workforce inclusive of clinicians with less specialized or advanced training, and honing in on the areas of emotion regulation and interpersonal functioning that appear to be particularly relevant targets for PDs. The authors suggest that future research directions should include testing incrementally smaller durations or “doses” of treatments and identifying moderators predictive of which patients will do well with shorter treatment durations and which will require longer or more specialized treatments.
The conclusions and suggestions made by Dixon-Gordon and colleagues match up well with evolving principles of generalist PD treatment as outlined in Good Psychiatric Management (GPM; Gunderson & Links, 2014) and emerging GPM-oriented stepped-care models (Choi-Kain, Albert, & Gunderson, 2016). The GPM model – itself an eclectic blend of psychodynamic, behavioral, and supportive case management techniques with empirical understandings of the origins and features of borderline PD – offers a clear conceptual framework for the authors’ proposal of how short-term PD interventions may be codified and sequenced into an algorithm guiding the selection of “first-line” PD treatments.
For example, in GPM, the length and intensity of treatment is titrated according to the patient’s pace of functional progress. Adjunctive treatments may be added, or primary treatment sessions may become more frequent, when a specific goal is being productively addressed. When the patient does not progress, GPM recommends a consultation to evaluate barriers to progress, as well as a possible referral to one of the longer-term specialist treatments, such as Dialectical Behavior Therapy (DBT), Mentalization-Based Treatment (MBT), or Transference-Focused Psychotherapy (TFP), if indicated. These longer-term specialist PD treatments take the treatment out of the hands of generalist clinicians, as they require implementing a more elaborated theoretical orientation, structural organization, and set of techniques over an extended period of time. Learning and delivering these treatments adherently requires a level of investment in training, supervision, and infrastructure that is unrealistic for most clinicians treating borderline PD and, thus, is best reserved for aspiring borderline PD specialists (Unruh & Gunderson, 2016).
In contrast, GPM employs simpler, more time-limited interventions that are increased or decreased across various intensities and levels of care based on the patient’s symptoms, functioning, level of engagement and progress, and comorbidities. However, GPM currently lacks precise, empirically driven rationales for when to recommend specific adjunctive interventions or shifts within the focus of the treatment over time. For example, when should a patient who started weekly case management sessions three months ago be referred to an adjunctive DBT skills group, mentalization group, or psychodynamic interpersonal group? Should this patient remain in the adjunctive group for three, six, or nine months (or longer) if progress is not being made before they are stepped up to a more intense or longer-term treatment modality? GPM leaves these decisions up to clinician judgment and accumulated anecdotal wisdom, emphasizing the pragmatism of “doing what works” and the flexibility to run with the strengths of the clinician and opportunities within the clinical system at hand.
Of the major longer-term treatments, GPM is thus the most accessible for most clinicians, clinical systems, and patients. This is important because longitudinal studies suggest that most PD patients receive a great deal of treatment over many years that is not empirically validated (Zanarini, Frankenburg, Reich, Conkey, & Fitzmaurice, 2015) or provided by PD specialists.
For these reasons, further studies on longer-term specialized treatments are not likely to yield improved guidelines for the kinds of interventions most PD patients will receive. “Dismantling” studies, wherein various elements of long-term treatments are isolated and evaluated in comparison with one another, are helpful but rare. And, when longer-term treatments are compared head-to-head, the comparison is between two elaborate systems of theory and technique, rather than individual interventions accessible to generalist clinicians and their patients.
In contrast, Dixon-Gordon and colleagues’ review highlights that shorter-term treatment studies are moving forward by testing specific interventions of limited duration and intensity, which could be readily appropriated by generalists as part of GPM’s stepped-care model. Briefer permutations of the long-term specialized treatments are being repurposed, combined, and sequenced, and then tested in novel shorter-term configurations: a DBT skills training group is compared with DBT skills group plus individual DBT; GPM is compared with GPM plus a DBT mindfulness group; a mentalizing group is offered following a DBT skills group; and various “doses” of DBT formats are tried at differing session frequencies and overall treatment durations. These shorter-term treatment studies may answer questions long posed and left unanswered by long-term treatment studies, including which particular DBT elements are most effective, whether elements of DBT and MBT can be effectively combined, and which group interventions are best across modalities.
Short-term treatment studies may thus be an ideal experimental “cauldron” in which to isolate, mix, compare, and contrast specific, relatively simple, time-limited interventions that can be sequenced into empirical algorithms within a “first-line” generalist framework such as GPM.
Short-Term PD Treatment Studies Will Help Us Better Understand Early Change Processes
Dixon-Gordon and colleagues highlight certain short-term PD treatments as particularly effective, noting that some have effect sizes similar to those achieved by some longer-term PD treatments. Notably, the most effective short-term treatments bear the same kinds of ties as longer-term treatments to particular theoretical orientations, technical approaches, and models of change. For example, emotion regulation group therapy was developed to enhance emotion regulation “based on empirical research suggesting self-injury primarily functions to regulate intense distress”; short-term dynamic psychotherapy bears strong marks of psychoanalytic theory and technique; manual-assisted cognitive therapy relies heavily on cognitive interventions; and six-month DBT shares clear-cut theoretical and technical features with its parent treatment. All major long-term specialist treatments for BPD are similarly linked with differing theories of how BPD originates and is perpetuated, and with different treatment aims and purported mechanisms of change (Gunderson, Fruzzetti, Unruh, & Choi-Kain, 2018).
However, if these short-term interventions are used to flesh out a “first-line” generalist treatment program that would precede referral to long-term specialist treatments, can we really expect generalist clinicians to learn and effectively implement them? How many different packages of theory, structure, and technique can be learned and delivered by the average clinician not specializing in PDs? Indeed, Dixon-Gordon and colleagues rightly argue that future directions in short-term PD treatment research should attempt to identify effective transtheoretical clinical strategies that can “cut across theoretical orientations.”
Here is where the process of fine-tuning short-term PD treatments may benefit from conclusions drawn from research on longer-term approaches. First, we know that most patients benefit similarly from generalist and specialist treatments. Head-to-head comparisons of specialized PD treatments with manualized generalist treatments (such as DBT versus good psychiatric management or MBT versus structured clinical management) show little substantive difference in outcome for most patients (Cristea et al., 2017). This implies that the more specific elements of each treatment are less important than shared non-specific structural or process elements (Fonagy, Luyten, & Bateman, 2017).
What do we know about common active ingredients in long-term PD treatments? Current hypotheses about shared change processes invoke commonality at two levels. Some studies highlight systemic and structural factors, such as basic elements of the treatment frame or therapist stance. Weinberg, Ronningstam, Goldblatt, Schechter, and Maltsberger (2010) concluded that major evidence-based treatments for PDs share a clear treatment framework, high attention to affect, a focus on the treatment relationship, an active therapist, and exploratory and change-oriented interventions. Further, since their analysis, the evolution of most major PD treatments has resulted in increasing overlap in their structural factors. Specifically, most major PD treatments now explicitly encourage support for the therapist by way of team consultation or peer supervision; crisis planning and specific protocols for addressing suicidal, self-harming, and treatment-interfering behavior; delivery of psychoeducation; a designated pre-treatment or contracting phase; and attention to functioning outside of therapy. Although the mechanisms by which these structural elements address borderline PD psychopathology are unclear, one possibility is that they work through containment of typical challenging emotional experiences arising in both patients and clinicians working together in the face of borderline PD-related interpersonal difficulties.
Shared change processes are also posited to occur at the level of mental state factors, such as in emotion regulation or reflective functioning. One example is the process of “epistemic trust” that has become integrated into MBT as its conceptualization of borderline PD has broadened to incorporate research on social cognition and learning theory. This model posits that successful borderline PD treatment within any modality requires patients to come to believe in the trustworthiness of clinicians and in the personal relevance of what is being communicated within the treatment. Arriving at this position of epistemic trust is considered a necessary precondition for any new learning about the social world to take hold and then generalize beyond the treatment context. According to this model, epistemic trust is ideally established through clinicians’ attention to mentalizing processes within psychotherapy no matter what particular technique or theoretical school is followed (Fonagy, Luyten, & Allison, 2015).
This model posits an interpersonal and intra-psychic process of change common to all long-term borderline PD treatments, despite divergent elements within each treatment that appear to have greater specificity. It may be that the elements most specific to each treatment send ostensive cues that effectively engage certain individuals and foment disconnection for others. For example, the didactic classroom elements of DBT skills group and the mnemonics used for skills acquisition may appeal to the desire of some patients for tangible skills and solutions from treatment, whereas other patients may experience DBT as too structured or directive and feel more comfortable with MBT’s more exploratory group culture. Other patients may be drawn to TFP’s frank portrayals of less experience-near elements of interpersonal process (such as hypothesized split-off aggression). It is possible to see how different patients might arrive through divergent treatment modalities at a similar position of trust in, regard for, and commitment to a treatment that has become a personally relevant and trustworthy source of knowledge about self and others. Could this model derived from long-term treatment studies be applied to understanding mechanisms of short-term treatments?
Dixon-Gordon and colleagues appropriately couch their conclusions about short-term treatments as a whole tentatively, pointing out that only a few are proven effective according to accepted standards of empirical validation. However, I believe the comprehensiveness of their review positions them to venture even bolder hypotheses that may guide future research. Among the short-term treatments they have identified as most effective, what elements are actually shared at the structural and organizational level? More speculatively, what common processes of change might exist at the level of mental states? Do the treatments that work do so through containment – i.e., by providing a sense of safety and structure that prevents treatment-interfering behavior? Do they more effectively establish epistemic trust by better capturing the attention and trust of the individuals they intend to treat? Are they better at helping patients understand the mental states underlying their own behavior and that of others? Many more hypotheses could be proposed and considered. I am interested in hearing Dixon-Gordon and colleagues’ further speculation about what transtheoretical change processes could underlie the approaches that appear to work most effectively in the short term, despite their varying theoretical underpinnings and techniques.
For PD treatments that work in any duration and at any level of care, improving their accessibility to patients and our understanding of how they work will depend on continued efforts to investigate both longer-term and short-term treatments. Clinicians and researchers investigating these treatments on either side of the duration divide must remain united in dialogue by the shared quest to identify elements that tie all treatments that work together. Whereas longer-term treatment studies lend themselves to the analysis of larger-scale structural and technical elements leading to fuller remission of symptoms, it may be that changes observed within shorter-term treatment studies more directly enhance our understanding of how most patients with PDs can begin to recover.
References
Choi-Kain, L. W., Albert, E. B., & Gunderson, J. G. (2016). Evidence-based treatments for borderline personality disorder: Implementation, integration, and stepped care. Harvard Review of Psychiatry, 24(5), 342–356.
Cristea, I. A., Gentili, C., Cotet, C. D., Palomba, D., Barbui, C., & Cuijpers, P. (2017). Efficacy of psychotherapies for borderline personality disorder: A systematic review and meta-analysis. JAMA Psychiatry, 74(4), 319–328.
Fonagy, P., Luyten, P., & Allison, E. (2015). Epistemic petrification and the restoration of epistemic trust: A new conceptualization of borderline personality disorder and its psychosocial treatment. Journal of Personality Disorders, 29(5), 575–609.
Fonagy, P., Luyten, P., & Bateman, A. (2017). Treating borderline personality disorder with psychotherapy: Where do we go from here? JAMA Psychiatry, 74(4), 316–317.
Gunderson, J. G., Fruzzetti, A., Unruh, B., & Choi-Kain L. (2018). Competing theories of borderline personality disorder. Journal of Personality Disorders, 32(2), 148–167.
Gunderson, J. G., & Links, P. (2014). Handbook of Good Psychiatric Management for Borderline Personality Disorder. Arlington, VA: American Psychiatric Publishing.
Unruh, B. T., & Gunderson, J. G. (2016). “Good enough” psychiatric residency training in borderline personality disorder: Challenges, choice points, and a model generalist curriculum. Harvard Review of Psychiatry, 24(5), 367–377.
Weinberg, I., Ronningstam, E., Goldblatt, M. J., Schechter, M., & Maltsberger, J. T. (2010). Common factors in empirically supported treatments of borderline personality disorder. Current Psychiatry Reports, 13(1), 60–68.
Zanarini, M. C., Frankenburg, F. R., Reich, D. B., Conkey, L. C., & Fitzmaurice, G. M. (2015). Treatment rates for patients with borderline personality disorder and other personality disorders: A 16-year study. Psychiatric Services, 66(1), 15–20.