Robert F. Bornstein, Ph.D.: DSM, PDM, and the Future of Personality Pathology

In May 2012 I took part in a symposium entitled Personality Disorders: DSM-5 and Beyond at the American Psychiatric Association meeting in Philadelphia. In general the speakers were quite critical of the DSM-5 proposal, with Mark Zimmerman, John Livesley, and Steve Huprich offering particularly strong counterpoints. There were a number of probing questions from the audience, and opinions were strong, both pro and con. Voices were raised; feelings were hurt. The words “ridiculous”, “incoherent”, and “embarrassing” appeared in various presenters’ PowerPoint slides. Two members of the DSM-5 Personality and Personality Disorders (PPD) work group—John Livesley and Roel Verheul—actually resigned from the work group right there in the room, during the symposium.

Personality Pathology in DSM-5 and PDM-2

We’re in the midst of a second major DSM paradigm shift. The first took place in 1980 when DSM-III introduced the multiaxial framework, with personality disorders separated from clinical disorders and coded on Axis II. It’s clear that DSM-5 will represent a paradigm shift as well: As Andrew Skodol noted in his recent article on the DSM-5 revision process, during the initial planning meetings participants were encouraged to think outside the box, consider making sweeping changes (rather than modest adjustments), and—in Skodol’s (2012, p. 319) words—to avoid “slavish adherence” to DSM-IV syndromes and constructs. (In this context Joel Paris’s comments during the Philadelphia symposium were instructive: He noted that back in 1980 Robert Spitzer, architect of DSM-III, received blowback from colleagues quite similar to that now directed at Skodol, with more than a few clinicians contending that DSM-III was a monumental bungle that would “destroy psychiatry”.)

Historical precedent notwithstanding, the DSM-5 is moving ahead rapidly, and one message that came through clearly during our Philadelphia symposium is that— controversial or not—the manual is going to be published in 2013. This makes sense. Because the last revision of DSM symptoms took place in 1994, the rubrics we use to diagnose patients today are based on empirical evidence that is, at best, two decades old. When DSM-IV was published on January 15th, 1994 a first class stamp cost 29 cents, Nelson Mandela had not yet been elected President of South Africa (that took place on April 29th), and O. J. Simpson was still just a retired football player (that changed on June 13th). As I tell students in my undergraduate personality disorders seminar, the symptom criteria we discuss in this class have been around longer than most of the people in the room.

Looking ahead, it is clear that DSM-5 will emphasize the neurophysiological underpinnings of personality disorders more strongly than did recent versions of the manual, with the ultimate goal of identifying endophenotypes for major syndromes. Trait and circumplex models have played a central role in shaping the DSM-5, illuminating core dimensions of personality pathology and providing much of the manual’s methodological infrastructure (see Bender, Morey, & Skodol, 2011).

The development of PDM-2 is now underway as well, though in a much earlier, formative stage. Conversations have begun, discussions with publishers are underway, and individuals willing to take on leadership roles are beginning to emerge (albeit in some instances reluctantly). Initial plans suggest that—like the PDM-1—PDM-2 will be more process-focused than DSM-5, with greater attention to underlying dynamics, motives, conflicts, and defenses. There will likely be explicit discussion of etiological factors and treatment dynamics for each syndrome. One goal of the PDM-2 is to develop more precise, operational symptom criteria, and to frame the manual (both broadly, and at the syndrome and symptom level) in a way that makes it accessible to clinicians of varied theoretical orientations (see Gordon, 2009, for preliminary findings in this area).

The Future of Personality Pathology: Implications and Opportunities

Given the contrasting structures of DSM and PDM, there has been a natural tendency for researchers who value trait and circumplex models to gravitate toward the DSM, and document convergences between psychological test results obtained from different sources (e.g., self-reports and reports by knowledgeable informants); factor analytic methods and structural equation modeling (SEM) procedures have also played a central role. More psychodynamically oriented researchers, who tend to favor multi-modal assessment and performance-based testing, have gravitated toward the PDM, seeking to document meaningful divergences that emerge when different test modalities are used to quantify features of a construct (e.g., self-report versus performance-based indices of narcissism or dependency). Both of these reactions are understandable, and from a professional vantage point, quite reasonable: We all tend to affiliate with colleagues who think like we do.

To make future versions of the DSM and PDM as good as they can be I suggest we resist our natural urge to interact primarily with those with whom we already agree. To do this risks accentuating a growing intellectual divide that could, if left unchecked, undermine personality disorder research during the coming years—the sort of divide that has been reified in the overblown schism between “empirically validated treatments” and interventions that emerge from a more psychodynamic tradition. Put another way, if we as individuals drift mindlessly toward the diagnostic system with which we are most comfortable a priori, we may inadvertently help create ingroups and outgroups that reflect two contrasting (but potentially complementary) perspectives on diagnosis and assessment. Over time we will find it more and more difficult to carry on productive dialogue with colleagues whose views differ from our own.

To strengthen the DSM-6 (DSM-5 is largely a done deal), those of us with expertise in psychodynamic frameworks and multi-modal assessment should make a point of connecting with—collaborating with—those whose work emphasizes trait models and SEM methods, so that we may test DSM hypotheses from a more integrative perspective. Given personality disordered patients’ limited insight and distorted self-perception, multi-modal assessment (including performance-based assessment) is crucial for the delineation of heuristic, clinically useful personality disorder symptom criteria in future versions of the DSM. In my view this represents a once-in-a-lifetime opportunity for psychoanalytic clinicians and researchers—including those of us here at Derner—to have a substantive impact on future diagnostic systems.

References

Bender, D. S., Morey, L. C., & Skodol, A. E. (2011). Toward a model for assessing level of personality functioning in DSM-5, Part 1: A review of theory and methods. Journal of Personality Assessment, 93, 332-346.

Gordon, R. M. (2009). Reactions to the Psychodynamic Diagnostic Manual (PDM) by psychodynamic, CBT, and other nonpsychodynamic psychologists. Issues in Psychoanalytic Psychology, 31, 55-62.

Skodol, A. E. (2012). Personality disorders in DSM-5. Annual Review of Clinical Psychology, 8, 317-344.