discussion

Pathological Personality Traits Can Capture DSM–IV Personality Disorder Types

Joshua D. Miller and Lauren R. Few University of Georgia

Donald R. Lynam Purdue University

James MacKillop University of Georgia and Brown University

The Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) includes an alter- native diagnostic approach to the assessment of personality disorders (PDs) in Section III with the aim of stimulating further research. Diagnosis of a PD using this approach is predicated on the presence of personality impairment and pathological personality traits. The types of traits present (e.g., callousness vs. emotional lability) are used to derive DSM–IV PD scores. Concerns have been raised, however, that such a trait-based approach will yield PD constructs that differ substantially from those generated using the approaches articulated in previous iterations of the DSM. We empirically examined this issue in a sample of 109 adults who were currently receiving mental health treatment. More specifically, we examined the correlations between interview-based PD scores derived from DSM–IV to DSM-5 PD trait counts, and tested them in relation to the 30 specific facets of the five-factor model, as well as internalizing and externalizing symptoms. Overall, the DSM–IV PD scores and DSM-5 PD trait counts correlated significantly with one another (Mr � .63), demonstrated similar patterns of interrelations among the PDs, and manifested highly similar patterns of correlations with general personality traits and symptoms of psychopathology. These results indicate that the DSM-5 PD trait counts specified in the alternative DSM-5 PD diagnostic approach capture the same constructs as those measured using the more traditional DSM–IV diagnostic system.

Keywords: DSM-5, personality disorder, assessment, personality traits

A host of problems are associated with personality disorders (PDs) in the Diagnostic and Statistical Manual of Mental Disor- ders, Fourth Edition (DSM–IV), including a lack of adequate coverage, widespread comorbidity, and difficulty distinguishing normality from abnormality (Clark, 2007; Widiger & Trull, 2007). In response to these problems, the DSM-5 Personality and Person- ality Disorder Work Group (DSM-5 PPD Work Group; American Psychiatric Association [APA], 2013) proposed a diagnostic sys- tem that differed radically from previous iterations. The two cen- tral components of this PD proposal are the assessment of self and interpersonal functioning, and the use of pathological traits to describe six DSM–IV types (i.e., schizotypal, antisocial, bor- derline [BPD], narcissistic, avoidant, and obsessive�compul- sive [OCPD]). For example, in this proposal, the DSM-5 narcis- sistic PD would be diagnosed if an individual manifested self and identity impairment and elevated scores on the traits of grandiosity

and attention seeking. The proposal also called for this system to describe PDs that are not covered by these “types,” including four DSM–IV PDs that were to be eliminated (i.e., paranoid, schizoid, histrionic, and dependent), using a PD�trait specified diagnosis. The pathological traits proposed for use in DSM-5 were derived from a new dimensional model of personality pathology compris- ing five higher order domains and 25 lower order traits that is thought to be an extension of the five-factor model (FFM) of personality (APA, 2013).

This proposal by the DSM-5 PPD Work Group was met with substantial criticism on a host of issues, including the decision to delete certain DSM–IV PDs (Bornstein, 2011), the mixing of dimensional and typal systems (Livesley, 2012), and the overall complexity of the proposal that led to concerns about its clinical utility (Shedler et al., 2010). In the current study, we addressed specific criticisms of the use of traits to assess PDs. For instance, Gunderson (2010), chair of the DSM–IV PD Work Group, ex- pressed significant concerns with the proposal for the diagnosis of BPD, suggesting that the “trait definition proposed for BPD . . . is not empirically based and is completely divorced from clinical concepts and literature” (p. 699). Even trait advocates such as Livesley (2012) argued against use of traits to capture DSM–IV PD constructs, suggesting “these diagnoses differ from their DSM–IV equivalents so substantially that they should be considered new diagnostic constructs” (p. 84). Due in part to the vocal opposition toward this diagnostic approach, the DSM-5 PPD Work Group

This article was published Online First February 10, 2014. Joshua D. Miller and Lauren R. Few, Department of Psychology, Uni-

versity of Georgia; Donald R. Lynam, Department of Psychological Sci- ences, Purdue University; and James MacKillop, Department of Psychol- ogy, University of Georgia, and Department of Behavioral and Social Sciences, Brown University.

Correspondence concerning this article should be addressed to Joshua D. Miller, Department of Psychology, University of Georgia, 125 Baldwin Street, Athens, GA 30602-3013. E-mail: [email protected]

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

Personality Disorders: Theory, Research, and Treatment © 2014 American Psychological Association 2015, Vol. 6, No. 1, 32– 40 1949-2715/15/$12.00 DOI: 10.1037/per0000064

32

proposal was included in Section III of DSM-5 to encourage further study, while the official DSM-5 PD diagnoses are made using the diagnostic approach articulated in DSM–IV.

Use of pathological traits in the diagnosis of PDs is neither as radical nor as exploratory as suggested by critics. A sizable literature exists on this topic, mostly from the perspective of the FFM, which has suggested that general traits can be used to conceptualize and assess PDs (see Miller, 2012, for a review). Within this approach (e.g., Miller, Bagby, Pilkonis, Reynolds, & Lynam, 2005), an individual can be scored on a given DSM–IV PD on the basis of a summation (or count) of scores on a smaller number of relevant FFM traits. Research from this perspective has demonstrated that FFM PD trait count scores correlated significantly with DSM–IV PDs, create personality profiles that are in line with those created by more traditional DSM–IV PD measures, and account for unique variance in indices of functioning, above and beyond that explained by DSM–IV PD constructs (e.g., Miller et al., 2010). In addition, specific to Gunderson’s concerns, FFM trait approaches to the scoring of BPD have demonstrated substantial success in rec- reating its nomological network (Miller, Morse, Nolf, Stepp, & Pilkonis, 2012; Trull et al., 2003).

Few studies exist, however, addressing the success of this count approach using the novel trait model proposed for DSM-5. Yalch, Thomas, and Hopwood (2012) compared the criterion validity of the DSM-5 trait count approach, a prototype matching approach (described in the initial DSM-5 PD proposal, but rejected early in the process), and a DSM–IV symptom- based approach for BPD and antisocial personality disorder in a sample of undergraduates. The trait approach manifested good convergent validity with the other approaches and superior criterion-related validity in relation to measures of functioning. Similarly, Samuel et al. (2013) examined the relations between DSM-5 PD trait counts and DSM–IV PDs using self-report data collected from a large undergraduate sample; they found that DSM-5 trait counts correlated significantly with DSM–IV PDs (Mdnr � .61) and reproduced the patterns of comorbidity found among DSM–IV PDs.

In the current study, we tested the validity of the PD trait count portion of the diagnostic approach for the PDs articulated in Section III of the DSM-5 PD by examining the correlations between DSM-5 PD trait counts and DSM–IV PDs (now the official PD diagnostic system in DSM-5 as well) and testing whether DSM-5 PD trait counts yield personality and psycho- pathology profiles that are consistent with those created by DSM–IV PDs. Ours is the first study to address these issues in a clinical sample and using DSM–IV PD symptoms and DSM-5 PD traits that were rated by research personnel after a semi- structured interview. We predicted that DSM-5 PD counts would exhibit (a) substantial correlations with traditional DSM–IV PD scores, (b) similar personality profiles as the DSM–IV PDs using the 30 facets of the FFM as the criteria, and (c) similar correlations with internalizing and externalizing symptoms. We also hypothesized that DSM-5 PD trait counts would successfully reproduce the pattern of intercorrelations found among DSM–IV PDs, and that this pattern of overlap would be due in large part to the use of overlapping traits.

Method

Participants and Procedure

Participants included 109 community adults (70% women; 90% White, 6% African American; Mage � 35.8 years, SD � 12.6) who were currently receiving psychological or psychiatric treatment. To participate, individuals had to be currently receiving psycho- logical care, between the ages of 18 and 65, have a minimum of a Grade 8 education, use a computer 3� days a week (to ensure that they could complete computerized assessments), and not be expe- riencing psychotic symptoms. Individuals were administered a semistructured interview for DSM–IV PD symptoms and com- pleted a number of self-report measures across a single 3– 4 hr protocol. Interviews were conducted by graduate students enrolled in a doctoral program in clinical psychology. Individuals were compensated $30 for participation. Institutional review board ap- proval was obtained for all aspects of the study.

Measures

Structured Clinical Interview for DSM–IV Axis II Person- ality Disorders. The Structured Clinical Interview for DSM–IV Axis II Personality Disorders (SCID-II; First, Gibbon, Spitzer, Williams, & Benjamin, 1997) is a semistructured interview that assesses the 10 DSM–IV PDs. Each symptom is scored using a 1�3 rating. Symptom ratings were completed by the interviewer (i.e., a doctoral student in clinical psychology) and a second rating was completed (n � 103) via a review of the videotaped interview by trained graduate students who were blind to the primary ratings.

Interrater reliability of the SCID-II ratings ranged from .79�.92 and alphas for the DSM–IV PD scores ranged from .68�.84. In terms of diagnoses, 37.6% of the sample met criteria for at least one PD; the most common diagnoses were avoidant (19.3%) and BPD (11%).

DSM-5 personality disorder traits. On completion of the SCID-II, the interviewer rated each participant on the 25 DSM-5 PD traits (e.g., callousness) using the official clinician rating guide provided by the DSM-5 PPD Work Group, which uses single-item ratings for each trait (0�3). Secondary ratings were generated after a review of the SCID-II interview. Interrater reliabilities for these 25 traits ranged from .12 (perseveration) to .83 (impulsivity), with a median of .55. DSM-5 PD traits counts were created by summing the ratings for each PD using the traits specified by the DSM-5 PPD Work Group (e.g., DSM-5 narcissistic PD traits � attention seeking � grandiosity). Despite not being included in Section III of DSM-5, we included the DSM-5 PPD Work Group’s previous specifications for scoring the four PDs specified for deletion on the basis of the new DSM-5 trait model (paranoid � suspiciousness � hostility � unusual beliefs and experiences � intimacy avoidance; schizoid � withdrawal � intimacy avoidance � restricted affec- tivity � anhedonia; histrionic � emotional lability � manipula- tiveness � attention seeking; dependent � submissiveness � anxiousness � separation insecurity).

Revised NEO Personality Inventory. The Revised NEO Per- sonality Inventory (NEO PI-R; Costa & McCrae, 1992) is a 240- item self-report measure of the FFM. Each of the five domains is comprised of six more specific facets. Alphas for the facets ranged from .58�.90, with a median of .81. Analyses with the NEO PI-R were limited to 106 participants.

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

33DSM-5 PERSONALITY DISORDER TRAITS

Patient-Reported Outcomes Measurement Information. The short forms of the Emotional Distress � Anxiety and Emo- tional Distress � Depression domains of the Patient-Reported Outcomes Measurement Information System (Pilkonis et al., 2011) are brief self-report questionnaires (i.e., seven and eight items for the Anxiety and Depression scales, respectively) that assess the experience of a particular emotion over the past 7 days. Both variables were normally distributed (Anxiety: M � 22.15, SD � 6.50, � � .94; Depression: M � 22.20, SD � 9.17, � � .97). The two scores correlated highly with one another (r � .81) and, thus, were combined to create a single internalizing variable.

Brief Symptom Inventory. The Brief Symptom Inventory (Derogatis, 1993) is a 53-item self-report inventory designed to assess psychiatric symptoms, and it provides scores on nine symp- tom scales (e.g., Somatization, Obsessive�Compulsive, Depres- sion, Hostility, Phobic Anxiety, and Psychoticism), as well as a Global Severity Index (GSI). The GSI, which we used here (� � .97), is computed by generating an average of all 53 items.

Crime and Analogous Behavior scale. The Crime and Anal- ogous Behavior scale (Miller & Lynam, 2003) is a self-report inventory that assesses externalizing behaviors such as substance use and antisocial behavior. A lifetime antisocial behavior variable (10 items) was created by giving participants 1 point for every relevant act endorsed (e.g., stealing; � � .77; M � 2.31, SD � 2.16). A lifetime drug use variable (eight items) was created by giving participants 1 point for every drug endorsed (e.g., cocaine; � � .79; M � 2.88, SD � 2.21).

Data Analytic Plan

First, the convergent and discriminant validity of DSM-5 PD trait counts were examined in relation to dimensional DSM–IV PD scores.1 Second, we tested whether the comorbidity manifested by DSM-5 PD trait counts reproduces that found among the DSM–IV PDs and whether use of overlapping traits among DSM-5 PD trait counts explained the comorbidity among DSM-5 PD trait counts. Third, the two sets of DSM PD scores— dimensional DSM–IV PD scores and DSM-5 PD trait counts—were correlated with the 30 personality traits from the NEO PI-R. We then examined the overall similarity of these sets of FFM profiles by calculating second-order intraclass correlation coefficients (ICCs) using double-entry q-correlations, which are more stringent than Pearson correlations as measures of agreement (i.e., this method measures absolute rather than relative agreement). Fourth, we examined these two sets of PD scores in relation to internalizing and exter- nalizing symptoms; because there were too few criteria to compute similarity indices for these results, we tested whether the correla- tions for the two PD scoring approaches differed significantly (tests of dependent rs).

Results

Convergence and Discriminant Validity Correlations of DSM–IV PDs Scores and DSM-5 PD Trait Counts

We first examined the convergent correlations between DSM–IV PDs scores and DSM-5 PD trait counts (see Table 1), which ranged from .43 (OCPD) to .81 (BPD, antisocial), with a mean of .63. In seven of 10 instances, DSM-5 PD trait counts manifested their

largest correlation (or tied for the largest in the case of the histrionic PD count) with their respective DSM–IV PD counterpart. In two of the other three cases, paranoid and narcissistic, the DSM-5 PD trait count manifested a slightly larger correlation with a “neighbor” PD from the same cluster (e.g., the DSM-5 paranoid PD trait count manifested a slightly larger correlation with the DSM–IV schizotypal PD than with the DSM–IV paranoid PD).

Comorbidity Within DSM–IV PD Scores and DSM-5 PD Trait Counts

Next, we examined the interrelations of the PD scores individ- ually within each paradigm (i.e., separately for DSM–IV PD and DSM-5 PD trait counts; see Table 2) and tested whether DSM–IV PDs and DSM-5 PD trait counts produced similar patterns of relations by calculating a second-order correlation (i.e., we corre- lated the correlations presented above and below the diagonal in Table 2). As expected, the two sets of PDs manifested similar patterns of relations among the sets of PD scores (r � .78). Although some may question why one would want to duplicate problematic patterns of comorbidity, we would note that comor- bidity is expected within the dimensional trait approach to PDs to the extent that PDs share overlapping traits. As a demonstration of this, we examined the degree to which trait overlap could account for the observed comorbidity. The number of shared traits between the individual DSM-5 PD counts (see Table 2) was substantially correlated with the pattern of comorbidity found for the DSM-5 PD trait counts, r � .76.

Profile Similarity of DSM–IV and DSM-5 PD Scores in Relation to the FFM

Next, we examined the similarity of the correlations generated by the two sets of DSM PD scores with the 30 facets of the FFM by calculating ICCs between the two sets of correlations. As can be seen in Tables 3–5, the ICCs ranged from .59 (OCPD) to .98 (BPD), with a mean of .90; all ICCs were statistically significant. For illustrative purposes, we review the findings for the PDs that manifested the strongest and weakest convergence: BPD and OCPD. The two BPD scores manifested nearly identical sets of correlations with the 30 traits of the FFM, with both approaches yielding moderate to strong positive correlations with facets of neuroticism (e.g., angry hostility, depression), and generally small to moderate negative correlations with facets of agreeableness (e.g., straightforwardness, compliance) and conscientiousness (e.g., competence, deliberation). The correlations for OCPD were the least similar; the DSM-5 OCPD trait count manifested larger negative correlations with the facets of extraversion (e.g., warmth, gregariousness, positive emotions) and agreeableness (e.g., trust, altruism).

1 To reduce concerns that common method variance might have inflated convergent correlations or similarity scores, different raters were used for DSM-5 trait scores (primary rater) and DSM–IV PD scores (secondary rater), except for six cases in which only the primary rater’s scores were available for both sets of scores. The results were virtually identical (mean correlation between DSM–IV PD scores and DSM-5 trait counts � .61; mean intraclass correlation coefficient between trait profiles generated by DSM–IV PD scores and DSM-5 trait counts � .88) if the raters are reversed (i.e., DSM–IV PDs rated by primary rater; DSM-5 traits rated by secondary rater).

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

34 MILLER, FEW, LYNAM, AND MACKILLOP

Relations Manifested by DSM–IV PDs and DSM-5 Trait Counts With Internalizing and Externalizing Symptoms

Next, we compared the correlations manifested by DSM–IV PD scores and DSM-5 PD trait counts in relation to internalizing and externalizing symptoms (see Table 6). Across 40 comparisons, only four correlations differed significantly such that both DSM-5 histrionic and dependent PD trait counts were more strongly re- lated to symptoms of anxiety and depression than were their DSM–IV counterparts.

Discussion

The diagnostic approach articulated for the PDs in Section III of DSM-5 represents a substantial departure from previous DSM

systems. Not surprisingly, this approach met with considerable resistance in relation to almost every aspect (see special issues in the Journal of Personality Assessment, Journal of Personality Disorders, and Personality Disorders: Theory, Research, and Treatment). In the current study, we focused specifically on con- cerns that dimensional traits cannot successfully recreate DSM–IV PDs (e.g., Gunderson, 2010; Livesley, 2012). Given the existence of a significant body of previous research that has demonstrated the success of such an approach using general personality traits (see Miller, 2012 for a review), we hypothesized that DSM-5 pathological trait counts would prove successful at capturing DSM–IV PD constructs measured using the traditional symptom- based approach.

In the current study, DSM-5 PD trait counts demonstrated sub- stantial convergence with DSM–IV PDs. The convergent validity

Table 1 Convergent and Discriminant Validity Correlations Among the DSM-5 PD Trait Counts and DSM–IV PDs

DSM-5 PD trait counts

PAR SZD SCT ASPD BPD HIS NAR AVD DEP OCPD Mdn disc. r

DSM–IV PDs PAR .59�� .25�� .44�� .45�� .51�� .38�� .15 .31�� .34�� .21� .34 SZD .46�� .67�� .48�� .10 .16 �.04 �.10 .64�� .19 .43� .19 SCT .60�� .22� .56�� .43�� .55�� .40�� .16 .29�� .41�� .17 .40 ASPD .32�� .01 .19� .81�� .61�� .58�� .32�� .05 .27�� �.18 .27 BPD .49�� .14 .32�� .68�� .81�� .60�� .20� .26�� .58�� .04 .32 HIS .15 �.22� .02 .49�� .33�� .60�� .58�� �.22� .06 �.16 .06 NAR .24� �.08 .13 .42�� .32�� .46�� .53�� �.05 .00 �.04 .13 AVD .24� .46�� .29�� .07 .29�� .00 �.25�� .55�� .57�� .40�� .29 DEP .19 .25�� .19 .23� .42�� .18 �.12 .35�� .59�� .13 .19 OCPD .13 .17 .12 �.06 .08 .00 �.04 .19� .21� .43�� .12 Mdn disc. r .24 .17 .19 .42 .33 .38 .15 .26 .27 .13

Note. Bold correlations represent the convergent validity correlations. Disc. r � discriminant validity correlation; DSM � Diagnostic and Statistical Manual of Mental Disorders; PD � personality disorder; PAR � paranoid PD; SZD � schizoid PD; SCT � schizotypal PD; ASPD � antisocial PD; BPD � borderline PD; HIS � histrionic PD; NAR � narcissistic PD; AVD � avoidant PD; DEP � dependent PD; OCPD � obsessive�compulsive PD. � p � .05. �� p � .01.

Table 2 Comorbidity Within the DSM-5 PD Trait Counts and DSM–IV PDs

PDs

PAR SZD SCT ASPD BPD HIS NAR AVD DEP OCPD

PAR — .55 (1) .84 (2) .51 (1) .64 (1) .44 (0) .31 (0) .61 (1) .49 (0) .48 (1) SZD .29 — .70 (2) .03 (0) .18 (0) �.13 (0) �.13 (0) .94 (3) .36 (0) .74 (2) SCT .52 .33 — .27 (0) .45 (0) .19 (0) .13 (0) .72 (1) .50 (0) .57 (1) ASPD .38 .11 .33 — .81 (3) .81 (1) .58 (0) .05 (0) .27 (0) �.07 (0) BPD .55 .10 .46 .66 — .77 (1) .40 (0) .32 (1) .66 (2) .09 (0) HIS .18 �.19 .13 .48 .38 — .71 (1) �.07 (0) .29 (0) �.12 (0) NAR .27 .05 .20 .33 .29 .42 — �.15 (0) .00 (0) �.01 (0) AVD .35 .33 .32 .15 .38 �.16 �.11 — .58 (1) .69 (1) DEP .36 .21 .37 .30 .51 .04 �.01 .66 — .32 (0) OCPD .20 .11 .12 �.04 .18 .09 .08 .32 .18 — r .78�

Note. Discriminant correlations among the DSM–IV PDs are listed below the diagonal; discriminant correlations among the DSM-5 PD trait counts are listed above the diagonal. Numbers in parentheses above the diagonal represent the number of shared traits used in the DSM-5 Section III PD counts (e.g., three of the same traits are used in the diagnosis of ASPD and BPD). DSM � Diagnostic and Statistical Manual of Mental Disorders; PD � personality disorder; PAR � paranoid PD; SZD � schizoid PD; SCT � schizotypal PD; ASPD � antisocial PD; BPD � borderline PD; HIS � histrionic PD; NAR � narcissistic PD; AVD � avoidant PD; DEP � dependent PD; OCPD � obsessive�compulsive PD. � p � .01.

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

35DSM-5 PERSONALITY DISORDER TRAITS

coefficients found for the DSM-5 PD trait count with DSM–IV PD scores (Mr � .63) are nearly identical to those found by Samuel et al. (2013; Mdnconvergent r � .61) who studied this issue using self-report data collected in an undergraduate study and are similar to the convergence found when using different PD measures designed to assess the exact same version of DSM constructs (see Miller, Few, & Widiger, 2012, for a review). It is interesting that two of the three DSM-5 PD trait counts that demonstrated more moderate convergence with their DSM–IV PD counterparts— his- trionic and narcissistic—are diagnosed using the fewest traits (i.e., 2–3 traits).2 It is possible that these DSM-5 PDs could be made to be more convergent with their DSM–IV counterparts if additional traits were employed. Decisions as to which traits to add could be made on the basis of an accumulating empirical literature or expert ratings (Samuel, Lynam, Widiger, & Ball, 2012). For instance, Samuel et al. collected expert ratings of DSM-5 traits thought to be most prototypical of DSM–IV PDs. Based on these data, narcis- sistic PD should also include the traits of callousness and manip- ulation. Studies are necessary to see which traits, including both those included and those not included in the DSM-5 trait model, are necessary to have a sufficiently comprehensive taxonomy of personality pathology.

Convergence can also be judged via the similarity of the pattern of correlates generated by these scores. In the current study, we

examined convergence by comparing the FFM trait profiles gen- erated by the two diagnostic approaches to one another and com- paring the correlations of these two approaches with internalizing and externalizing symptoms. In general, DSM-5 PD trait counts yielded FFM personality profiles that were closely aligned with those manifested by DSM–IV PDs (MrICC � .90), suggesting that DSM-5 PD counts, like general trait PD counts (e.g., Miller et al., 2010), can effectively capture DSM–IV PD constructs.3 DSM–IV PDs and DSM-5 trait counts also manifested similar sets of corre-

2 In fact, the number of traits used per DSM-5 PD trait count correlated significantly with the level of convergent validity (i.e., correlation between DSM–IV PDs and DSM-5 trait counts) manifested in the current study (r � .69).

3 The success of DSM-5 PD trait counts is not limited to the use of clinician-rated traits. Using self-reported traits from the Personality Inven- tory for DSM-5 (PID-5; Krueger, Derringer, Markon, Watson, & Skodol, 2012) to generate DSM-5 PD trait counts in this same data set, we found convergent correlations with the interview-based ratings of DSM–IV PDs that ranged from .46 (histrionic) to .69 (BPD), with a median of .56. Similarly, the profile similarities of the PID-based DSM-5 PD trait counts with the interview-based DSM–IV PDs with regard to the 30 FFM facets ranged from .47 (histrionic) to .96 (paranoid), with a median rICC of .89. In sum, DSM-5 PD trait counts assess constructs similar to their DSM–IV counterparts whether or not they are assessed using interviewers’ ratings or self-reports.

Table 3 DSM–IV and Five Cluster A Personality Disorders in Relation to the Five-Factor Model

NEO PI-R traits

Paranoid Schizoid Schizotypal

DSM–IV DSM-5 TC DSM–IV DSM-5 TC DSM–IV DSM-5 TC

Anxiety .29 .21 .16 .27 .16 .14 Angry hostility .61 .49 .23 .16 .33 .28 Depression .30 .36 .27 .36 .31 .28 Self-consciousness .17 .21 .16 .37 .05 .23 Impulsiveness .29 .19 �.02 �.02 .26 .14 Vulnerability .32 .27 .24 .30 .29 .23 Warmth �.28 �.39 �.60 �.73 �.17 �.40 Gregariousness �.28 �.34 �.50 �.57 �.12 �.34 Assertiveness �.01 �.09 �.36 �.45 �.10 �.23 Activity �.02 �.13 �.24 �.45 .05 �.17 Excitement seeking .05 �.02 �.23 �.17 .13 .01 Positive emotions �.22 �.30 �.50 �.60 �.06 �.28 Fantasy �.07 �.24 �.23 �.13 �.10 �.09 Aesthetics �.03 �.09 �.25 �.15 �.10 �.03 Feelings �.02 �.09 �.35 �.36 �.02 �.10 Actions �.23 �.36 �.34 �.32 �.25 �.26 Ideas �.10 �.18 �.27 �.11 �.10 �.05 Values �.20 �.37 �.24 �.21 �.22 �.27 Trust �.57 �.59 �.35 �.46 �.41 �.52 Straightforwardness �.43 �.33 �.16 �.06 �.31 �.25 Altruism �.33 �.35 �.31 �.34 �.20 �.31 Compliance �.47 �.38 �.04 .03 �.31 �.16 Modesty �.11 �.05 .03 .13 �.12 .04 Tendermindedness �.18 �.22 �.27 �.20 �.06 �.17 Competence �.26 �.29 �.23 �.27 �.34 �.29 Order .04 .03 .06 �.09 .10 .02 Dutifulness �.21 �.14 �.10 �.03 �.21 �.15 Achievement striving �.03 �.05 �.12 �.20 �.15 �.11 Self-discipline .01 .03 �.01 �.10 .02 �.04 Deliberation �.24 �.23 �.06 .07 �.32 �.13 Profile match .95� .92� .82�

Note. Profile matches based on double-entry q-correlation, which is an intraclass correlation. DSM � Diagnostic and Statistical Manual of Mental Disorders; TC � trait count. � p � .01.

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

36 MILLER, FEW, LYNAM, AND MACKILLOP

lations with internalizing and externalizing symptoms such that, of 40 comparisons, only four differed significantly (two of which occurred in relation to histrionic PD). In general, the PDs expected to be most strongly related to internalizing symptoms (e.g., avoidant, dependent, BPD) were most strongly related to symp- toms of depression and anxiety, while the PDs expected to be related to externalizing behaviors (e.g., antisocial, BPD) were most strongly related to antisocial behavior and substance use (Kotov et al., 2011; Røysamb et al., 2011).

Despite the general convergence of these two sets of PDs scores with the criterion variables, DSM-5 trait counts for OCPD and histrionic PD exhibited more limited profile agreement with the FFM traits, albeit still fairly large and significant (rICC � .59 and .71, respectively). The DSM-5 OCPD trait count includes the traits of rigid perfectionism, perseveration, intimacy avoidance, and restricted affectivity. The inclusion of the latter two traits, which were not in the original OCPD count proposed by the DSM-5 PPD Work Group, are likely partially responsible for the smaller degree of convergence found with its DSM–IV counterpart because such traits were not included in DSM–IV OCPD criteria (although these traits are discussed to some extent in the accompanying text). In the current data, the inclusion of these two additional traits served to reduce the DSM-5 OCPD trait count’s correlation with DSM–IV

OCPD scores (two traits: r � .56; four traits: r � .43) and its similarity with regard to its general trait profile (two traits: rICC � .78; four traits: rICC � .59). Although traits similar to intimacy avoidance and restricted affectivity are included in other trait- based assessments of OCPD (e.g., Samuel, Riddell, Lynam, Miller, & Widiger, 2012), this representation in the DSM-5 OCPD trait count constitutes some degree of shift away from the construct as it was assessed in DSM–IV (and now in the main text of DSM-5). The lack of strong convergence between OCPD scores from DSM–IV and the DSM-5 OCPD trait count is also likely due in part to an omission of traits that are central to this disorder. Using the original 37-trait model proposed by the DSM-5 PPD Work Group, experts (see Samuel, Lynam, et al., 2012) rated four traits as being particularly central to OCPD: perseveration, perfectionism, rigid- ity, and orderliness. Three of these four traits—all but persevera- tion—were combined, however, into a single trait of rigid perfec- tionism in the final 25-trait model included in Section III of DSM-5. We suspect that inclusion of a broader array of relevant traits would strengthen the DSM-5 trait model’s ability to success- fully capture OCPD.

Similar divergences were found between DSM–IV and the DSM-5 histrionic PD trait count such that the latter manifested larger positive correlations with facets of neuroticism and inter-

Table 4 DSM–IV and Five Cluster B Personality Disorders in Relation to the Five-Factor Model

NEO PI-R traits

Antisocial Borderline Histrionic Narcissistic

DSM–IV DSM-5 TC DSM–IV DSM-5 TC DSM–IV DSM-5 TC DSM–IV DSM-5 TC

Anxiety .11 .14 .41 .39 �.10 .12 �.12 �.05 Angry hostility .28 .43 .47 .53 .18 .40 .26 .28 Depression .24 .23 .50 .50 .01 .21 �.01 �.04 Self-consciousness .17 .12 .39 .36 �.01 .06 �.01 �.12 Impulsiveness .37 .44 .47 .48 .31 .42 .24 .27 Vulnerability .19 .24 .50 .52 �.05 .18 �.01 �.10 Warmth �.10 �.13 �.19 �.21 .26 .11 �.07 .14 Gregariousness �.05 .00 �.18 �.14 .19 .08 �.10 .15 Assertiveness �.06 .13 �.12 �.11 .26 .20 .29 .38 Activity �.02 .04 �.15 �.15 .16 .17 .15 .29 Excitement seeking .18 .29 .11 .17 .23 .26 .05 .27 Positive emotions .02 .01 �.16 �.20 .31 .16 .05 .21 Fantasy .07 �.01 .06 .00 .21 .09 .24 .13 Aesthetics .04 .02 .13 .04 .24 .11 .06 .08 Feelings �.05 .01 .08 .08 .23 .18 .07 .15 Actions �.02 �.08 �.09 �.20 .19 �.07 �.02 .09 Ideas �.05 �.08 �.05 �.13 .18 �.04 .02 .00 Values �.15 �.20 �.05 �.19 �.14 �.18 �.03 �.12 Trust �.14 �.29 �.29 �.40 .09 �.16 �.09 �.08 Straightforwardness �.35 �.45 �.40 �.38 �.17 �.32 �.42 �.27 Altruism �.28 �.35 �.19 �.22 .02 �.17 �.34 �.21 Compliance �.18 �.44 �.26 �.33 �.15 �.32 �.30 �.33 Modesty �.06 �.18 .04 .00 �.24 �.21 �.41 �.36 Tendermindedness �.08 �.10 �.13 �.11 .03 �.12 �.24 �.08 Competence �.39 �.39 �.46 �.50 �.12 �.26 �.05 �.03 Order �.19 �.14 �.24 �.23 �.12 �.14 �.10 .01 Dutifulness �.50 �.48 �.47 �.44 �.20 �.36 �.19 �.12 Achievement striving �.29 �.18 �.33 �.31 .04 �.10 .07 .19 Self-discipline �.22 �.19 �.31 �.26 �.08 �.12 �.03 .02 Deliberation �.52 �.57 �.45 �.50 �.24 �.41 �.23 �.31 Profile match .93� .98� .71� .83�

Note. Profile matches based on double-entry q-correlation, which is an intraclass correlation. DSM � Diagnostic and Statistical Manual of Mental Disorders; TC � trait count. � p � .01.

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

37DSM-5 PERSONALITY DISORDER TRAITS

nalizing symptoms, which may be due to the inclusion of the emotional lability facet in the DSM-5 histrionic PD count (along with manipulativeness and attention seeking). DSM-5 emotional lability is defined as “instability of emotional experiences and mood; emotions that are easily aroused, intense, and/or out of proportion to events and circumstances” (APA, 2013, p. 779)—a definition that is not entirely consistent with the emotional expe- rience described for histrionic PD in DSM–IV. In DSM–IV, indi- viduals with this disorder were described as having “rapidly shift- ing and shallow emotions” and demonstrating “self-dramatization, theatricality, and exaggerated expression of emotion” (APA, 2000, p. 714). The DSM-5 histrionic PD trait count may overemphasize the role of genuine negative affectivity and fail to note that these emotional changes may be superficial and displayed in the service of interpersonal manipulation.

We also examined the discriminant validity of DSM-5 PD trait counts and found that, in general, DSM-5 Section III trait counts manifested larger convergent validity correlations (Mr � .63) than discriminant correlations (Mr � .25). There were some cases in which a DSM-5 trait count manifested a larger discriminant cor- relation with a DSM–IV PD other than its counterpart (e.g., DSM-5 narcissistic PD with DSM–IV histrionic), but the differences were small in nature and were typically found with near neighbor

disorders (e.g., those from the same cluster). We also demonstrated that the relations found among DSM-5 PD trait counts largely reproduced the pattern of interrelations found among DSM–IV PDs; the correlations among the two sets of correlation matrices manifested a second-order correlation of .78. In addition, we were able to show that the correlations among DSM-5 PD trait counts were largely predictable by the number traits shared among the PDs. Although the trait approach does not necessarily diminish the issue of overlap among the PDs, it provides a simple and parsi- monious explanation—PDs will co-occur to the extent that they share the same or similar personality traits (e.g., Lynam & Widi- ger, 2001).

Ultimately, the current data suggest that pathological personality traits can be used to score DSM–IV PDs, although certain DSM-5 PD trait counts may require further modification if the goal is to replicate perfectly their DSM–IV counterparts. These data do not, however, speak to the broader issue of whether traits should be used to recreate these DSM–IV PD types, because with that rec- reation comes many of the problems that are associated with the DSM–IV PD diagnostic approach. For instance, Clark (2007), in discussing trait approaches like this (i.e., using the FFM), sug- gested “the DSM diagnoses are much too flawed to warrant emulation” and that trait approaches like the FFM have “great

Table 5 DSM–IV and 5 Cluster C Personality Disorders in Relation to the Five-Factor Model

Avoidant Dependent OCPD

NEO PI-R traits DSM–IV DSM-5 TC DSM–IV DSM-5 TC DSM–IV DSM-5 TC

Anxiety .47 .39 .42 .45 .34 .25 Angry hostility .22 .23 .24 .31 .24 .25 Depression .56 .47 .45 .50 .26 .27 Self-consciousness .59 .47 .42 .49 .23 .29 Impulsiveness .26 .03 .30 .27 .15 �.03 Vulnerability .57 .43 .56 .55 .28 .23 Warmth �.49 �.71 �.26 �.27 �.09 �.58 Gregariousness �.44 �.60 �.25 �.29 �.19 �.45 Assertiveness �.57 �.52 �.44 �.47 �.06 �.30 Activity �.37 �.48 �.22 �.31 �.04 �.30 Excitement seeking �.10 �.19 �.04 �.02 �.09 �.20 Positive emotions �.41 �.63 �.24 �.30 �.21 �.49 Fantasy .05 �.12 .05 .01 .07 �.19 Aesthetics �.17 �.11 �.09 .08 �.08 �.22 Feelings �.19 �.30 �.18 �.03 .15 �.21 Actions �.25 �.35 �.22 �.22 �.21 �.37 Ideas �.14 �.14 �.17 �.11 .08 �.15 Values �.04 �.21 �.15 �.18 �.12 �.20 Trust �.29 �.50 �.28 �.32 �.11 �.41 Straightforwardness �.12 �.06 �.19 �.15 �.10 .02 Altruism �.21 �.29 �.20 �.04 .02 �.23 Compliance .08 .05 .00 .05 �.06 �.04 Modesty .24 .18 .17 .23 .11 .10 Tendermindedness �.12 �.19 �.14 �.08 �.07 �.19 Competence �.42 �.32 �.49 �.40 �.02 �.12 Order �.17 �.10 �.19 �.14 .16 .14 Dutifulness �.29 �.04 �.39 �.20 .09 .12 Achievement striving �.44 �.26 �.45 �.39 .04 �.02 Self-discipline �.33 �.11 �.29 ��.20 �.14 �.07 Deliberation �.07 .06 �.27 �.13 .07 .21 Profile match .91� .96� .59�

Note. Profile matches based on double-entry q-correlation, which is an intraclass correlation coefficient (i.e., rICC). DSM � Diagnostic and Statistical Manual of Mental Disorders; TC � trait count. � p � .01.

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

38 MILLER, FEW, LYNAM, AND MACKILLOP

value in PD assessment, but it lies in the dimensions themselves and their potential for deepening our understanding of PD traits, not in their ability to approximate demonstrably inadequate cate- gories” (p. 232). Similarly, Livesley (2012) noted that the DSM-5 proposal “perpetrates the myth of discrete categories of personality disorder despite voluminous evidence to the contrary” (p. 84). We believe, however, that these trait recreations of DSM–IV PDs may help clinicians transition from these well-known but flawed PD categories to the more valid yet unfamiliar dimensional trait ap- proaches (e.g., Miller et al., 2008). Although some trait proponents object to such a compromise (i.e., Livesley, 2012), we believe it may be the only way that the broader mental health field will successfully move from a typal approach to a dimensional trait approach (cf., Tyrer, Crawford, & Mulder, 2011).

Limitations and Conclusions

The current study is the first to address the validity of these DSM-5 PD counts using data collected from a clinical sample and using interview-based ratings of the PD constructs (DSM–IV PDs and/or DSM-5 traits). Despite the novelty of these data, it is

important to consider that the current findings are derived from a relatively small sample and, thus, it will be important to replicate these analyses in larger, more diverse samples. In addition, we compared DSM–IV and DSM-5 PD scores to a relatively limited number of external correlates (i.e., general personality traits, and internalizing and externalizing symptoms), and it is possible that results might differ in relation to other meaningful constructs (e.g., etiological factors, treatment utilization/satisfaction). Some of DSM-5 traits demonstrated limited interrater reliability, which may have attenuated the relations manifested by DSM-5 PD trait counts that included these traits (e.g., perseveration, which feeds into the creation of the DSM-5 OCPD trait count) with their DSM–IV counterparts. In the current study, DSM-5 trait ratings were derived from information gleaned from an interview designed for the assessment of traditional DSM–IV PDs (e.g., BPD), which may have affected the interrater reliability and validity of these trait ratings, particularly for traits that may not be as well covered by an assessment of DSM–IV PDs. This methodological limitation was unavoidable, however, because a specific semistructured interview for the Section III DSM-5 traits had not been developed at the time of this study. We would note that the general reliability and validity of these ratings, despite this issue, may be cause for optimism in that these traits can be relatively successfully assessed by existing measures and will likely perform even more robustly when assessed using an interview designed specifically for this purpose.

In sum, the current results suggest that DSM-5 PD trait counts yield constructs that are substantively similar to their DSM–IV counterparts. This should allay concerns that trait-based ap- proaches will fundamentally alter the nature of the PD constructs. With tensions high among researchers with regard to the DSM-5 Section III PD proposal, we believe it will prove critically impor- tant that decisions regarding the diagnostic approach used for the PDs in future iterations of the DSM and criticisms of these deci- sions be empirically driven.

References

American Psychiatric Association. (2000). Diagnostic and statistical man- ual of mental disorders (4th ed., text rev.). Washington, DC: Author.

American Psychiatric Association. (2013). Diagnostic and statistical man- ual of mental disorders (5th ed.). Washington, DC: Author.

Bornstein, R. F. (2011). Reconceptualizing personality pathology in DSM-5. Limitations in evidence for eliminating dependent personality disorder and other DSM–IV syndromes. Journal of Personality Disor- ders, 25, 235–247. doi:10.1521/pedi.2011.25.2.235

Clark, L. A. (2007). Assessment and diagnosis of personality disorder: Perennial issues and an emerging reconceptualization. Annual Review of Psychology, 58, 227–257. doi:10.1146/annurev.psych.57.102904 .190200

Costa, P. T., & McCrae, R. R. (1992). Revised NEO Personality Inventory (NEO PI-R) and NEO Five-Factor Inventory (NEO-FFI) professional manual. Odessa, FL: Psychological Assessment Resources.

Derogatis, L. R. (1993). Brief Symptom Inventory. Administration, Scoring, and Procedures Manual (4th Ed.). Minneapolis, MN: National Com- puter Systems.

First, M. B., Gibbon, M., Spitzer, R. L., Williams, J. B. W., & Benjamin, L. S. (1997). Structured Clinical Interview for DSM–IV Axis II Disor- ders (SCID-II). Washington, DC: American Psychiatric Press.

Table 6 Relations Between DSM–IV and Five Personality Disorder Scores and Internalizing and Externalizing Symptoms

DSM PDs

Internalizing symptoms Externalizing behaviors

Depression- anxiety

Global severity ASB

Substance use

Paranoid DSM–IV .36 .56 .29 .12 DSM-5 TC .43 .56 .37 .30

Schizoid DSM–IV .32 .31 .08 .16 DSM-5 TC .44 .42 .07 .17

Schizotypal DSM–IV .33 .50 .39 .35 DSM-5 TC .40 .49 .28 .30

Antisocial DSM–IV .22 .39 .57 .49 DSM-5 TC .26 .45 .55 .51

Borderline DSM–IV .49 .63 .34 .32 DSM-5 TC .51 .68 .40 .41

Histrionic DSM–IV �.10a .02a .20 .16 DSM-5 TC .21b .37b .31 .25

Narcissistic DSM–IV .06 .11 .14 .28 DSM-5 TC �.02 .05 .18 .22

Avoidant DSM–IV .46 .49 .12 .09 DSM-5 TC .57 .55 .04 .15

Dependent DSM–IV .36a .42a .16 .06 DSM-5 TC .55b .61b .20 .16

OCPD DSM–IV .14 .21 �.15 �.04 DSM-5 TC .30 .28 �.04 .01

Note. Within each personality disorder and outcome, correlations with different superscripts (a, b) differ significantly across the two DSM per- sonality disorder scoring methodologies (p � .05). DSM � Diagnostic and Statistical Manual of Mental Disorders; ASB � antisocial behavior; TC � trait count; OCPD � obsessive�compulsive personality disorder.

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

39DSM-5 PERSONALITY DISORDER TRAITS

Gunderson, J. G. (2010). Revising the borderline diagnosis for DSM-V: An alternative proposal. Journal of Personality Disorders, 24, 694 –708. doi:10.1521/pedi.2010.24.6.694

Kotov, R., Ruggero, C., Krueger, R. F., Watson, D., Yuan, Q., & Zimmer- man, M. (2011). New dimensions in the quantitative classification of mental illness. Archives of General Psychiatry, 68, 1003–1011. doi: 10.1001/archgenpsychiatry.2011.107

Krueger, R. F., Derringer, J., Markon, K. E., Watson, D., & Skodol, A. V. (2012). Initial construction of a maladaptive personality trait model and inventory for DSM-5. Psychological Medicine, 42, 1879 –1890. doi: 10.1017/S0033291711002674

Livesley, J. (2012). Tradition versus empiricism in the current DSM-5 proposal for revising the classification of personality disorders. Criminal Behaviour and Mental Health, 22, 81–90. doi:10.1002/cbm.1826

Lynam, D. R., & Widiger, T. A. (2001). Using the five-factor model to represent the DSM–IV personality disorders: An expert consensus ap- proach. Journal of Abnormal Psychology, 110, 401– 412. doi:10.1037/ 0021-843X.110.3.401

Miller, J. D. (2012). Five-factor model personality disorder prototypes: A review of their development, validity, and comparison with alternative approaches. Journal of Personality, 80, 1565–1591. doi:10.1111/j.1467- 6494.2012.00773.x

Miller, J. D., Bagby, R. M., Pilkonis, P. A., Reynolds, S. K., & Lynam, D. R. (2005). A simplified technique for scoring the DSM–IV personality disorders with the five-factor model. Assessment, 12, 404 – 415. doi: 10.1177/1073191105280987

Miller, J. D., Few, L. R., & Widiger, T. A. (2012). Assessment of person- ality disorders and related traits: Bridging DSM–IV–TR and DSM-5. In T. A. Widiger (Ed.), Oxford handbook of personality disorders (pp. 108 –140). New York, NY: Oxford University Press. doi:10.1093/ oxfordhb/9780199735013.013.0006

Miller, J. D., & Lynam, D. R. (2003). Psychopathy and the five-factor model of personality: A replication and extension. Journal of Person- ality Assessment, 81, 168 –178. doi:10.1207/S15327752JPA8102_08

Miller, J. D., Lynam, D. R., Rolland, J. P., De Fruyt, F., Reynolds, S. K., Pham-Scottez, A., . . . Bagby, R. M. (2008). Scoring the DSM-IV Personality Disorders using the Five-Factor Model: Development and validation of normative scores for North American, French and Dutch- Flemish samples. Journal of Personality Disorders, 22, 433– 450.

Miller, J. D., Maples, J., Few, L. R., Morse, J. Q., Yaggi, K. E., & Pilkonis, P. A. (2010). Using clinician-rated five-factor model data to score the DSM–IV personality disorders. Journal of Personality Assessment, 92, 296 –305. doi:10.1080/00223891.2010.481984

Miller, J. D., Morse, J. Q., Nolf, K., Stepp, S. D., & Pilkonis, P. A. (2012). Can DSM–IV borderline personality disorder be diagnosed via dimensional personality traits: Implications for the DSM-5 personality disorder proposal. Journal of Abnormal Psychology, 121, 944 –950. doi:10.1037/a0027410

Pilkonis, P. A., Choi, S. W., Reise, S. P., Stover, A. M., Riley, W. T., & Cella, D., on behalf of the PROMIS Cooperative Group. (2011). Item banks for measuring emotional distress from the Patient-Reported Out- comes Measurement Information System (PROMIS®): Depression, anxiety, and anger. Assessment, 18, 263–283. doi:10.1177/1073191111411667

Røysamb, E., Tambs, K., Ørstavik, R. E., Torgersen, S., Kendler, K. S., Neale, M. C., . . . Reichborn-Kjennerud, T. (2011). The joint structure of DSM–IV Axis I and Axis II disorders. Journal of Abnormal Psychology, 120, 198 –209. doi:10.1037/a0021660

Samuel, D. B., Hopwood, C. J., Krueger, R. F., Thomas, K. M., & Ruggero, C. (2013). Comparing methods for scoring personality disorder types using maladaptive traits in DSM-5. Assessment, 20, 353–361. doi:10.1177/1073191113486182

Samuel, D. B., Lynam, D. R., Widiger, T. A., & Ball, S. A. (2012). An expert consensus approach to relating the proposed DSM-5 types and traits. Personality Disorders: Theory, Research, and Treatment, 3, 1–16. doi:10.1037/a0023787

Samuel, D. B., Riddell, A. D. B., Lynam, D. R., Miller, J. D., & Widiger, T. A. (2012). A five-factor measure of obsessive-compulsive personality traits. Journal of Personality Assessment, 94, 456 – 465. doi:10.1080/ 00223891.2012.677885

Shedler, J., Beck, A., Fonagy, P., Gabbard, G. O., Gunderson, J., Kernberg, O., . . . Westen, D. (2010). Personality disorders in DSM-5. The American Journal of Psychiatry, 167, 1026 –1028. doi:10.1176/appi.ajp .2010.10050746

Trull, T. J., Widiger, T. A., Lynam, D. R., & Costa, P. T., Jr. (2003). Borderline personality disorder from the perspective of general person- ality functioning. Journal of Abnormal Psychology, 112, 193–202. doi: 10.1037/0021-843X.112.2.193

Tyrer, P., Crawford, M., & Mulder, R., on behalf of the ICD-11 Working Group for the Revision of Classification of Personality Disorders. (2011). Reclassifying personality disorders. The Lancet, 377, 1814 – 1815. doi:10.1016/S0140-6736(10)61926-5

Widiger, T. A., & Trull, T. J. (2007). Plate tectonics in the classification of personality disorder: Shifting to a dimensional model. American Psy- chologist, 62, 71– 83. doi:10.1037/0003-066X.62.2.71

Yalch, M. M., Thomas, K. M., & Hopwood, C. J. (2012). The validity of trait, symptom, and prototype approaches for describing borderline and antisocial personality disorders. Personality and Mental Health, 6, 207– 216. doi:10.1002/pmh.1184

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

40 MILLER, FEW, LYNAM, AND MACKILLOP