15c
Janine D. Flory
The commentaries by Bornstein (this volume) and Chmielewski and Trujillo (this volume) of my review and critique of categorical approaches for diagnosing personality disorders (PDs) are appreciated. Both respectfully take issue with the continued singular conceptualization of PDs as discrete categories. This is not at odds with my review as my intent was not to “take sides” on a contentious debate but to provide an overview of existing methods for diagnosing PDs as categories and to make recommendations for future research given the reality of DSM-5.
Bornstein (this volume) expands upon the observation that models and measures are often conflated and critiques of instruments (e.g., a structured interview) might be better characterized as a critique of a categorical model of pathology (i.e., DSM) and vice versa. I frequently remind myself not to make the same error. Bornstein also notes the limitations of conclusions drawn from research that is solely based on self-report, as people generally lack context and insight when rating their own behaviors and traits. The recommendation to use specific language indicating the source of information when reporting results about personality pathology is sound. I would add that large-scale personality research that uses self-rated constructs is often conducted in university settings with college students, calling into question the generalizability of the findings to clinical and community samples.
Bornstein echoes a recurring call for multi-modal assessment of PDs, including informant reports, expressed behavior and performance-based test data. He provides some intriguing data using such methods to advance the study of borderline PD and dependent PD. Bornstein ends with the sanguine observation that the dimensional and categorical perspectives have much in common and that an integrative approach will advance research and clinical practice. I share this hopeful view.
Chmielewski and Trujillo (this volume) take a more pessimistic view of the continued use of diagnostic categories and call for greater rigor in establishing reliable methods for diagnosis. I could not agree more with this appeal as it aligns with my position in the chapter that the state of the science in establishing reliable methods of assessment using interviews lags other areas of PD research. Research that uses multi-modal assessments to establish reliable diagnostic methods, including interviews and self-reports as well as informant reports, biological markers, and expressed behaviors is almost non-existent. To quote Clark, “single-point-in-time and single-source-of-information assessment should not be expected to yield entirely valid PD diagnoses” (Clark, 2007, p. 244). And yet, the type of work that is almost universally recommended for advancing the field is not an easy or inexpensive research endeavor. Neither is research designed to conceptualize treatable conditions based on multiple dimensions of personality and functioning, which requires large-n research, especially if biological, ecological, and informant measures are included. Collaboration and open communication among research groups is essential to carry out this important work.
Chmielewski & Trujillo (this volume) describe ways to increase rigor when establishing reliability including using the test-retest method of rater reliability rather than the audio/video recording method in which a second-rater listens to or watches a previously recorded interview. The argument is that the listener is tipped off by skip-out questions and knows when the original interviewer has made a yes/no decision about a particular diagnostic criterion. The use of retest interviews can be burdensome to interviewees, however, and increases the expense of a research study. A cheaper, less burdensome alternative is to audio/video record an interview asking all questions with no skip-outs. The authors also helpfully suggest setting benchmarks for reliability based on estimates for normative personality traits. Chmielewski & Trujillo end with a less hopeful view of the future utility of establishing reliability for categorical diagnoses and firmly advocate for continued shift to dimensional conceptualization and assessment of PDs.
Although not a specific focus of the chapter, I would add that the process of refining and understanding how to diagnose PDs will also be advanced by treatment development. At present, it is rare that dimensions or categories direct treatment selection or planning for PDs. In contrast, measurement-based care is being implemented in treatment protocols for mood and anxiety disorders and substance use disorders (Lewis et al., 2019). The need for consensus on assessment of key dimensional constructs that inform PD diagnosis should include consideration of how assessments can inform treatment selection and track recovery and potential for relapse.
In closing, the process of identifying optimal methods for diagnosing PDs is still a work in progress. As I have noted in the chapter, researchers and clinicians who work with other DSM categories have also struggled with how to define and assess disorder. For example, the diagnosis of PTSD has expanded from 12 diagnostic criteria in DSM-III to 20 in DSM-5, shifting the relative importance of the symptom clusters along the way. And while the Clinician Administered PTSD Scale (CAPS) is universally accepted as the gold-standard interview for the diagnosis of PTSD, the DSM-IV version of the interview had more than nine scoring rules for determining whether a symptom was a symptom (Weathers, Keane, & Davidson, 2001). The interview has been revised for DSM-5 and scoring has been simplified (Weathers et al., 2018), but there is a generation of research using the earlier definitions of the disorder and interview with established cut points that no longer apply.
The shift from a categorical to a dimensional paradigm is an iterative journey that requires cooperation between camps who have different conceptual views and use different methodologies. We have come a long way from conceptualization of personality and PDs as immutable traits and disorders that are environmentally mediated. There is widespread recognition that PDs develop from a complex interplay between environmental, cultural, and inherited and biological factors. From the ground, it is hard see this as a shifting of the tectonic plates (Widiger & Trull, 2007), but I believe there is reason to be hopeful that shifting will continue in a productive direction.
References
Clark, L. A. (2007). Assessment and diagnosis of personality disorder: Perennial issues and an emerging reconceptualization. Annual Review of Psychology, 58, 227–257.
Lewis, C. C., Boyd, M., Puspitasari, A., Navarro, E., Howard, J., Kassab, H., … Kroenke, K. (2019). Implementing measurement-based care in behavioral health: A review. JAMA Psychiatry, 76, 324–335.
Weathers, F. W., Bovin, M. J., Lee, D. J., Sloan, D. M., Schnurr, P. P., Kaloupek, D. G., … Marx, B. P. (2018). The Clinician-Administered PTSD Scale for DSM-5 (CAPS-5): Development and initial psychometric evaluation in military veterans. Psychological Assessment, 30(3), 383–395.
Weathers, F. W., Keane, T. M., & Davidson, J. R. (2001). Clinician-administered PTSD scale: A review of the first ten years of research. Depression and Anxiety, 13(3), 132–156.
Widiger, T. A., & Trull, T. J. (2007). Plate tectonics in the classification of personality disorder: Shifting to a dimensional model. American Psychologist, 62(2), 71–83.