14b
Michael Carnovale and R. Michael Bagby
Evans, Williams, and Simms (this volume) have written an impressively comprehensive and contemporary review of the various instruments and methods designed to assess dimensional personality pathology. These authors review a wide range of tests and cover most of the issues that are critical areas of concern or discussion in the field. This chapter will surely serve as a key reference for clinicians and researchers interested in the assessment of personality pathology. It should be emphasized, however, that the field of dimensional personality pathology is moving and evolving at a breathtaking pace, which is especially the case since the introduction of the Alternative Model of Personality Disorders (AMPD) in Section III of the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) (American Psychiatric Association, 2013), and we would anticipate yet another review will be needed in short order.
In this commentary, we: (a) expand on several topics covered by Evans et al. that we believe are especially relevant to clinical applicability/utility (i.e., multisource assessment, validity scales, and norms), and (b) identify and discuss other topics that we believe are relevant to the assessment of dimensional personality pathology that were not directly addressed in their chapter (i.e., longitudinal assessment, other newer measures).
Multiple Sources of Information
The use of multiple sources of information in the assessment of personality pathology is clearly an important issue. Specifically, Evans et al. point out the potential of self-reports’ validity being compromised by a lack of insight, and the potential of informant reports and clinical interviews providing incremental validity. There are, however, many questions that we believe future research should address within this topic.
First, with the use of informant-reports in clinical practice, is merely one informant adequate, or is it more advantageous to have multiple informants? At a psychometric level, it is likely the case that the use of multiple informants increases the reliability of the assessment of a target’s dimensional personality pathology, but it may be less pragmatic to actually solicit reports from multiple informants (or even one informant) in practice. Further, is there a specific kind of an informant that may provide the most useful information regarding a target’s personality pathology (e.g., a romantic partner compared to a parent or close friend)?
Second, and relatedly, how can systematic measurement error (i.e., response biases) in informant-reports be taken into account? For example, with the use of one particular informant, it may be the case that this informant especially likes the target, which may therefore compromise the validity of this informant’s ratings of the target (e.g., presenting the target in a favorable light; Leising, Erbs, & Fritz, 2010).
Third, are clinician-rated instruments compromised if there is a lack of insight from a patient, or if there is evidence that a patient is engaging in non-credible responding on self-report instruments? Specifically, as certain clinician-rated instruments heavily rely on the responses of the patients themselves, it is reasonable to expect that the information provided by the clinician-rated instruments may be accordingly affected.
Fourth, there is some uncertainty if meta-perception-based reports (i.e., asking a target how they believe other people see them) can be useful in the assessment of personality pathology (e.g., Carlson & Barranti, 2016). It may be the case that framing assessment items from a different point of view may overcome response biases inherent in typically framed self-reported items (i.e., worded in first-person), and may provide a more pragmatic alternative to informant-reports. Specifically, if there is evidence to suggest that the use of meta-perceptions provides equivalent incremental validity as informant-reports, it may be easier for a clinician to administer a meta-perception-based measure than to solicit an informant to complete a measure.
Finally, as mentioned by Evans et al., what are the patterns of discrepancy between self-reports, informant-reports, and clinician-rated instruments that could be expected in the context of assessing personality pathology? Is it the case that self-reports tend to have higher mean scores than informant-reports (e.g., Sleep, Lamkin, Lynam, Campbell, & Miller, 2018)? Is it the case that those with more severe personality pathology tend to diverge from informant-reports more than those with less severe personality pathology? It is clear that much more research needs to be done in this area in order to potentially inform clinical practice in the assessment of personality pathology.
Absence of Validity Scales
A second issue of clinical relevance noted by Evans et al. is the absence of validity scales in many of the instruments assessing personality pathology. Validity scales typically assess three types of test protocol validity: inconsistent responding, non-credible over-reporting, and non-credible under-reporting. Out of the more than 20 scales listed in Table 14.1 in Evans et al., only six include validity scales, and even fewer have validity scales that assess both forms of non-credible reporting and inconsistent responding. We believe this a remarkable shortcoming. Although an over-reporting (Sellbom, Dhillon, & Bagby, 2018) and response inconsistency scales (Bagby & Sellbom, 2018; Keeley, Webb, Peterson, Roussin, & Flanagan, 2016) have been developed for the Personality Inventory for DSM-5 (PID-5), an under-reporting scale has yet to be developed and validated for this measure. While many personality and some clinical researchers may eschew the notion of validity scales for a variety of different reasons, clinicians expect them and the clinical utility of any measure that includes validity scales is enhanced (Hopwood & Sellbom, 2013). As suggested in the previous section on multisource assessment, it may also be useful to investigate the inclusion of validity scales for informant-based measures.
Need for Normative Data
We are also in agreement with Evans et al. that, for any instrument assessing personality pathology, it would be useful to gather normative data in order to generate T-scores for a given examinee, and therefore increase clinical utility. Although a number of the measures outlined by Evans et al. have associated normative data, there are also many measures outlined that have yet to gather normative data. With respect to the PID-5 scoring of the Personality Assessment Inventory (PAI) in particular, Evans et al. suggest that an advantage of using the PAI relative to many other instruments discussed is that the PAI has strong norms. We agree that the PAI does contain strong norms for its established subscales, but at the same time, we would argue that it is premature to promote the PID-5 scoring of the PAI based on this reason alone. That is, we believe new normative data must be collected based on responses to these new PID-5-based subscales in order to realize its clinical utility with respect to the ability to derive T-scores.
Tracking Personality Pathology across Time
One issue that was given limited attention by Evans et al. was the extent to which each measure could be used to track personality pathology traits across time (e.g., over the course of treatment). In general, only a relatively small amount of research has been conducted regarding the longitudinal assessment of personality pathology (for a review, see Morey & Hopwood, 2013), and regarding the capacity of most of the scales reviewed for longitudinal assessment (e.g., Wright, Hopwood, Skodol, & Morey, 2016). Moreover, most of this research has focused on more “panel” style longitudinal data (e.g., yearly assessments), and less research has focused on more intensive longitudinal data (i.e., daily diary methods, ecological momentary assessments; Roche, 2018; Wright & Simms, 2016), especially in the context of treatment.
Two possible measures that may be feasible to use to track personality pathology in the context of treatment are the brief 25-item version of the PID-5 (PID-5-BF; see Roche, 2018 for the use of the PID-5-BF in daily diary research) and the recently developed Personality Dynamics Diary (PDD; Zimmermann et al., 2019). The PID-5-BF retains the five broader domains of the full PID-5, whereas the PDD contains nine subscales that assess various trait expressions. Of note, Zimmermann et al. (2019) found some support for the clinical utility of the PDD from the perspective of patients and therapists.
It would also be interesting to extend the use of informant-reports in the context of intensive longitudinal assessments. For example, can informant-reports provide useful information regarding the symptom course of a patient during treatment, beyond self-reports? As well, and more broadly, would it be feasible and useful to collect information regarding a patient’s behaviors across different types of contexts, from both self-reports and informant-reports? Overall, it is clear that more research has to be done in order to examine the feasibility and clinical utility of certain scales for tracking personality pathology across time.
New Measures with Considerable Potential
While Evans et al. reviewed an extensive list of measures, there are two newer measures that were not reviewed that may hold significant potential to improve existing dimensional assessment. First, the Structured Clinical Interview for the DSM-5 Alternative Model for Personality Disorders (SCID-5-AMPD; First, Skodol, Bender, & Oldham, 2018), as the name suggests, is a structured interview that contains three modules designed to assess the AMPD. Module I involves a more structured assessment of the Level of Personality Functioning Scale (i.e., Criterion A of the AMPD), Module II involves the assessment of the 25 maladaptive personality facets and the five maladaptive personality domains of the AMPD (analogous to the scales that can be derived from the PID-5), and Module III involves the assessment of the six personality disorders that were retained and modified in the AMPD. As of the writing of this commentary, there have been two studies examining the reliability and validity of Module I of the SCID-5-AMPD (Buer Christensen et al., 2018; Kampe et al., 2018). It would be informative for future research to further examine the psychometric properties of the SCID-5-AMPD, as well as for the potential addition of norms for this measure to increase clinical utility.
Second, the recent Personality Inventory for ICD-11 (PiCD; Oltmanns & Widiger, 2018) is a self-reported measure designed to assess the dimensional trait model of personality pathology in the 11th edition of the World Health Organization’s International Classification of Diseases. Structurally, the PiCD contains five subscales: (a) Negative Affective, (b) Detachment, (c) Dissocial, (d) Disinhibition, and (e) Anankastic. It is evident that four of these five domains are conceptually similar to the Negative Affect, Detachment, Antagonism, and Disinhibition domains, respectively, from the AMPD (Oltmanns & Widiger, 2018). There is preliminary evidence for the reliability and validity of the PiCD (Oltmanns & Widiger, 2018), and as this classification for personality disorders has been accepted by the World Health Organization (Mulder & Tyrer, 2019), there is also likely to be more research and examinations of its clinical utility (Bagby & Widiger, in press).
Potential Downsides in the Expansive Growth of Measures
One final, and arguably the most important, concern we would like to discuss is the potentially problematic existence of such a large number of measures. In general, we agree with Evans et al. regarding the need for a consensual set of personality pathology features, and we believe it may be difficult to determine the relative utility of each measure and to ultimately decide on one or two measures in the context of research or in clinical practice. In the research context, we believe it’s important that the field of dimensional personality pathology accumulate data sets that are reasonably comparable across sites, and that results from published studies have a common metric to evaluate outcomes (which also has implications for evaluating replicability). This seems to be a challenging task given the proliferation of existing measures. In clinical practice, one should be using the most empirically supported approaches, but it may be difficult for the busy clinician to navigate the vast assessment literature to compare all of the measures at a nuanced level. Therefore, we believe much research has to be done comparing these measures in terms of their relative psychometric properties (e.g., Crego, Oltmanns, & Widiger, 2018), and their relative clinical utility (e.g., prediction of treatment course/outcomes, preferential treatment selection). Further, it may be useful to conduct multi-trait/multi-method studies in order to examine the presence of “jingle/jangle” fallacies among similarly conceptualized or differentially conceptualized subscales (i.e., to test convergent and discriminant validity, or conceptual overlap; Ziegler, Booth, & Bensch, 2013). As mentioned by Evans et al., perhaps a unified approach among a large number of researchers, such as through the HiTOP initiative, is a step forward toward construct clarification and consensus, and in turn, measurement consensus.
References
American Psychiatric Association. (2013). Diagnostic and Statistical Manual of Mental Disorders (5th ed.). Arlington, VA: American Psychiatric Publishing.
Bagby, R. M., & Sellbom, M. (2018). The validity and clinical utility of the Personality Inventory for DSM-5 Response Inconsistency Scale. Journal of Personality Assessment, 100(4), 398–405.
Bagby, R. M., & Widiger, T. A. (in press). The assessment of ICD-11 personality disorders: Introduction to the special section. Psychological Assessment.
Buer Christensen, T., Paap, M. C. S., Arnesen, M., Koritzinsky, K., Nysaeter, T.-E., Eikenaes, I., … Hummelen, B. (2018). Interrater reliability of the Structured Clinical Interview for the DSM-5 Alternative Model of Personality Disorders Module I: Level of Personality Functioning Scale. Journal of Personality Assessment, 100(6), 630–641.
Carlson, E. N., & Barranti, M. (2016). Metaperceptions: Do people know how others perceive them? In J. A. Hall, M. S. Mast, & T. V. West (Eds.), The Social Psychology of Perceiving Others Accurately (pp. 165–182). New York: Cambridge University Press.
Crego, C., Oltmanns, J. R., & Widiger, T. A. (2018). FFMPD scales: Comparisons with the FFM, PID-5, and CAT-PD-SF. Psychological Assessment, 30(1), 62–73.
First, M. B., Skodol, A. E., Bender, D. S., & Oldham, J. M. (2018). User’s Guide for the Structured Clinical Interview for the DSM-5 Alternative Model for Personality Disorders (SCID-5-AMPD). Arlington, VA: American Psychiatric Association Publishing.
Hopwood, C. J., & Sellbom, M. (2013). Implications of DSM-5 personality traits for forensic psychology. Psychological Injury and Law, 6(4), 314–323.
Kampe, L., Zimmermann, J., Bender, D., Caligor, E., Borowski, A.-L., Ehrenthal, J. C., … Hörz-Sagstetter, S. (2018). Comparison of the Structured DSM-5 Clinical Interview for the Level of Personality Functioning Scale with the Structured Interview of Personality Organization. Journal of Personality Assessment, 100(6), 642–649.
Keeley, J. W., Webb, C., Peterson, D., Roussin, L., & Flanagan, E. H. (2016). Development of a Response Inconsistency Scale for the Personality Inventory for DSM-5. Journal of Personality Assessment, 98(4), 351–359.
Leising, D., Erbs, J., & Fritz, U. (2010). The letter of recommendation effect in informant ratings of personality. Journal of Personality and Social Psychology, 98(4), 668–682.
Morey, L. C., & Hopwood, C. J. (2013). Stability and change in personality disorders. Annual Review of Clinical Psychology, 9, 499–528.
Mulder, R., & Tyrer, P. (2019). Diagnosis and classification of personality disorders: Novel approaches. Current Opinion in Psychiatry, 32(1), 27–31.
Oltmanns, J. R., & Widiger, T. A. (2018). A self-report measure for the ICD-11 dimensional trait model proposal: The personality inventory for ICD-11. Psychological Assessment, 30(2), 154–169.
Roche, M. J. (2018). Examining the alternative model for personality disorder in daily life: Evidence for incremental validity. Personality Disorders, 9(6), 574–583.
Sellbom, M., Dhillon, S., & Bagby, R. M. (2018). Development and validation of an Overreporting Scale for the Personality Inventory for DSM-5 (PID-5). Psychological Assessment, 30(5), 582–593.
Sleep, C. E., Lamkin, J., Lynam, D. R., Campbell, W. K., & Miller, J. D. (2018). Personality disorder traits: Testing insight regarding presence of traits, impairment, and desire for change. Personality Disorders: Theory, Research, and Treatment. Advance online publication.
Wright, A. G. C., Hopwood, C. J., Skodol, A. E., & Morey, L. C. (2016). Longitudinal validation of general and specific structural features of personality pathology. Journal of Abnormal Psychology, 125(8), 1120–1134.
Wright, A. G. C., & Simms, L. J. (2016). Stability and fluctuation of personality disorder features in daily life. Journal of Abnormal Psychology, 125(5), 641–656.
Ziegler, M., Booth, T., & Bensch, D. (2013). Getting entangled in the nomological net. European Journal of Psychological Assessment, 29(3), 157–161.
Zimmermann, J., Woods, W. C., Ritter, S., Happel, M., Masuhr, O. Jaeger, U., … Wright, A. G. C. (2019). Integrating structure and dynamics in personality assessment: First steps toward the development and validation of a personality dynamics diary. Psychological Assessment, 31(4), 516–531.