Dimensions of Normal and Abnormal Personality: Elucidating DSM-IV Personality Disorder Symptoms in Adolescents

Noor B. Tromp and Hans M. Koot

VU University Amsterdam

ABSTRACT The present study aimed to elucidate dimensions of normal and abnormal personality underlying DSM-IV personality disorder (PD) symptoms in 168 adolescents referred to mental health services. Dimensions derived from the Big Five of normal personality and from Livesleys (2006) conceptualization of personality pathology were regressed on interviewbased DSM-IV PD symptom counts. When examined independently, both models demonstrated signicant levels of predictive power at the higher order level. However, when added to the higher order Big Five dimensions, Livesleys higher and lower order dimensions afforded a supplementary contribution to the understanding of dysfunctional characteristics of adolescent PDs. In addition, they contributed to a better differentiation between adolescent PDs. The present ndings suggest that adolescent PDs are more than extreme, maladaptive variants of higher order normal personality traits. Adolescent PDs seem to encompass characteristics that may be more completely covered by dimensions of abnormal personality. Developmental issues and implications of the ndings are discussed.

There is growing interest in the developmental antecedents of adult personality pathology. Research increasingly indicates that personality pathology is not limited to adulthood, supporting its validity as a construct in both clinical and nonclinical adolescent populations (for a review, see Johnson, Bromley, Bornstein, & Sneed, 2006). Questions remain about the optimal way to conceptualize personality pathology. Dissatisfaction with the currently used categorical classication system of personality disorder (PD), the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV; American Psychiatric Association [APA], 2000), is widespread among researchers and
Correspondence concerning this article should be addressed to H. M. Koot Department of Developmental Psychology, VU University Amsterdam, Van der Boechorststraat 1, 1081 BT Amsterdam, The Netherlands. E-mail: hm.koot@psy.vu.nl.

Journal of Personality 78:3, June 2010 r 2010, Copyright the Authors Journal compilation r 2010, Wiley Periodicals, Inc. DOI: 10.1111/j.1467-6494.2010.00635.x

840

Tromp & Koot

clinicians alike (Bernstein, Iscan, & Maser, 2007). It has been argued repeatedly that personality pathology is better conceptualized with a dimensional model (Clark, 2007; Trull & Durrett, 2005). Although the debate on categories versus dimensions is mostly conned to adult psychiatry, the dimensional approach may also be valuable for describing adolescent personality pathology. First, relatively little is known on the structure of adolescent personality pathology (but see De Clercq, De Fruyt, Van Leeuwen, & Mervielde, 2006; Tromp & Koot, 2008). When the DSM criteria and thresholds are applied to adolescent populations, questions on this structure remain unanswered. Second, in applying the DSM classications to adolescents, symptomatology is most likely not interpreted in the context of adolescent psychological development. For example, behaviors that are considered pathological in adults may beat least temporarilypart of normal development in adolescents (e.g., shyness, identity problems, emotional instability). Finally, the DSM classication has frequently been criticized for its inability to classify subthreshold traits and symptoms. Especially for those adolescents who score less extremely on symptom criteria (but who could be at risk for developing full-blown pathology), the related subthreshold level of impairment may be overlooked in applying the DSM criteria. Dimensional models retain important information concealed in subthreshold symptom levels (Trull & Durrett, 2005).
Dimensions of Normal Personality and Personality Pathology

Dimensional models that have been employed frequently in an attempt to understand personality pathology are the Big Five ( John, Donahue, & Kentle, 1991) and the Five-Factor Model (FFM) of normal personality (Digman, 1990). Many studies have shown that normal personality dimensions (particularly high Neuroticism and low Agreeableness and Conscientiousness) are underlying PDs in clinical and nonclinical adult samples (Widiger & Costa, 2002; Saulsman & Page, 2004), and a few have done so in nonclinical adolescent samples (De Clercq & De Fruyt, 2003; De Clercq, De Fruyt, & Van Leeuwen, 2004). When the adolescent and meta-analytic adult data (Saulsman & Page, 2004) on correlation patterns between FFM dimensions and DSM-IV PDs are compared, a noteworthy difference can be observed. Adolescent PDs seem to have more normal personality dimensions in common than adult PDs, suggesting that adolescent PDs are less differentiated than those in adults. Such

Dimensions and DSM-IV Disorders of Personality

841

smaller degrees of differentiation in adolescent disordered personality possibly indicate that personality pathology, if present, is still in a developing stage at this age. This may suggest that personality traits in the normal range can contribute substantially to the description of disordered personality in adolescence. In addition to investigating its relations with DSM-dened PDs, studies have related dimensions of normal personality to dimensional models of abnormal personality. Findings in adult populations have indicated that dimensions of normal and abnormal personality are essentially congruent at the higher order level, suggesting a substantial common ground, with four broad factors that have been labeled Emotional (In)stability, Extraversion/Introversion, (Dis)agreeableness, and Conscientiousness/Compulsivity (Clark & Livesley, 2002; Larstone, Jang, Livesley, Vernon, & Wolf, 2002; Widiger & Simonsen, 2005). This structure has also been proposed as a common dimensional model to integrate adaptive and maladaptive individual differences in childhood and adolescence (Mervielde, De Clercq, De Fruyt, & Van Leeuwen, 2005). Research has thus underscored the relevance of dimensions of normal personality in describing personality pathology, despite its exclusive reliance on adaptive personality traits. Notwithstanding this evidence, an important issue must be addressed to determine whether personality pathology is sufciently covered by normal personality dimensions. In addition to dening PDs as involving inexible and maladaptive traits, the DSM also requires that these traits cause clinically signicant distress or impairment in social, occupational, or other important areas of functioning (APA, 2000, p. 689). Distress or dysfunction is inherent in the notion of disorder. The Big Five is a purely descriptive and structural framework of personality traits found within the natural language, indicating the relative presence of each trait. Although extreme scores on these normal traits may indicate abnormal personality, disordered personality encompasses more than pure statistical deviance. Extreme scores do not necessarily imply distress or dysfunction, required for disordered personality. A supplementary model may be needed to capture the dysfunctional characteristics inherent in disordered personality.
Considering a Supplement: Livesleys Model

A potential answer to this issue can be found within Livesleys (2006) conceptualization of personality pathology, which includes explicit

842

Tromp & Koot

referral to dysfunction. Livesley and colleagues proposed a denition of PD as involving the failure to achieve adaptive solutions to major life tasks relating to identity or self (intrapersonal), intimacy and attachment (interpersonal), and prosocial behavior (Livesley, 2007; Livesley & Jang, 2000; Livesley, Schroeder, Jackson, & Jang, 1994). It can be argued that the complexity of disordered personality is conceptually more completely encompassed in Livesley and colleagues denition of PD, possibly when used in addition to normal personality traits. For example, intrapersonal characteristics highly relevant to disordered personality such as identity problems (Marcia, 2006) are not covered by the dimensions of normal personality. In addition, interpersonal characteristics such as disturbances in attachment and autonomy receive little attention within the normal personality traits, although interpersonal impairment is an important aspect of disordered personality (APA, 2000; Clark, Livesley, Schroeder, & Irish, 1996). Also, dysfunctional behavioral characteristics associated with disordered personality, such as self-harm (Klonsky, Oltmanns, & Turkheimer, 2003) and cognitive distortion (Livesley & Schroeder, 1990), are not represented in models of normal personality (Clark, Vorhies, & McEwen, 2002; Costa & McCrae, 1990; Schroeder, Wormworth, & Livesley, 2002). Empirical studies in adult populations have supported the additional contribution of models of abnormal personality to an adequate description of the dysfunction associated with disordered personality. These studies have demonstrated that higher order dimensions of abnormal personality provided a signicant incremental contribution to the variance accounted for in DSM-IV PD symptoms, above and beyond variance accounted for by the Five-Factor domains (Bagby, Marshall, & Georgiades, 2005; Reynolds & Clark, 2001). In a recent study, Nestadt et al. (2008) concluded that the breadth of PD pathology is not completely captured by the FFM domains, although the lower order facets of the FFM improved the extent to which the FFM explained PD pathology. The authors further concluded that additional clinical material may be necessary to describe PD pathology, proposing the Dimensional Assessment of Personality Pathology Basic Questionnaire (DAPP-BQ; Livesley & Jackson, 2009), the Schedule for Nonadaptive and Adaptive Personality (SNAP; Clark, 1993), or the Structured Interview for the FFM (SIFFM; Trull, Widiger, & Burr, 2001) as potential useful alternatives. From the viewpoint of these conceptual and empirical issues, the most valuable model for

Dimensions and DSM-IV Disorders of Personality

843

understanding the dimensional representation of disordered personality may be one that supplements dimensions of normal personality with those of abnormal personality. The evidence that normal personality dimensions are underlying PDs in nonclinical adolescents (De Clercq & De Fruyt, 2003; De Clercq et al., 2004) has, to our knowledge, not been replicated in a clinical adolescent sample, nor have dimensional models of normal and abnormal personality been investigated in parallel in terms of their relations with adolescent PDs. In view of the growing interest in the developmental antecedents of adult personality pathology and the substantial evidence favoring dimensional over categorical models, it may be interesting to study the relations of dimensional models to DSM-IV PD symptoms in an adolescent sample. During this age period, when personality pathology may still be in development and thus show a more diffuse structure than during adulthood, a dimensional approach may provide differentiation and understanding in addition to that provided by the DSM categorical classication system. Recently, the promising psychometric qualities of an age-appropriate operationalization of Livesleys model of personality pathology (DAPP-BQ for Adolescents [DAPP-BQ-A]) were demonstrated for use in adolescent populations (Tromp & Koot, 2008). Moreover, in an earlier report using the same sample as the current study (Tromp & Koot, 2009), it was demonstrated that higher and lower order dimensions of the DAPP-BQ-A correlated as predicted with PD symptoms assessed with a semistructured interview. In addition, these dimensions predicted specic PD symptomatology above and beyond gender, age, and co-occurring PD symptoms. Elaborating on these previous studies, the present study includes an assessment of normal personality dimensions, which have been shown to relate to PDs in adults and nonclinical adolescents (De Clercq & De Fruyt, 2003; De Clercq et al., 2004; Saulsman & Page, 2004). Specically, this is the rst study to elucidate dimensions derived from the Big Five and Livesleys model underlying interview-based DSM-IV PD symptoms in adolescents referred to mental health services. It is hypothesized that both models include dimensions relevant to disordered personality and capture substantial amounts of variance in adolescent PD symptoms. It is further hypothesized that the full complexity of disordered personality is more completely captured by a supplementary model including, in addition to the higher order Big Five dimensions of normal personality, the dimensions of abnormal

844

Tromp & Koot

personality proposed by Livesleys model. Subsequently, the present study investigates which lower order dimensions within Livesleys model may provide differentiation of adolescent personality pathology above and beyond individual differences accounted for by the higher order dimensions of normal personality.
METHOD Participants
The sample, as previously described by Tromp and Koot (2009), consisted of 168 adolescents (35% male), with a mean age of 15.9 years (SD 5 2.3; range 1222 years), referred to youth mental health services in one of four collaborating centers for in- and outpatient treatment in The Netherlands. The adolescents were referred for various reasons, not specically for personality pathology. Axis I disorders, assessed with the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-I; First, Spitzer, Gibbon, & Williams, 1996) and the attention-decit/hyperactivity, oppositional deant, and conduct disorder modules of the Schedule for Affective Disorders and Schizophrenia for School-Age Children Present and Lifetime Version (Kaufman et al., 1997), were present in 77% of the sample at the time of assessment (Tromp & Koot, 2009). The prevalence rates for Axis I disorders were as follows: attention-decit/ hyperactivity disorder 7%, oppositional deant disorder 9%, conduct disorder 7%, mood disorder 49%, anxiety disorder 39%, psychotic disorder 15%, substance use disorder 10%, somatoform disorder 6%, and eating disorder 15%. Gender was equally distributed across the inpatient (n 5 89) and outpatient (n 5 79) subsamples. On average, outpatients were slightly older than inpatients (Cohens d 5 .44). Inpatients scored signicantly higher than outpatients (po.05) on several dimensions of normal and abnormal personality and on all but one (ObsessiveCompulsive) PD symptom scales. Outpatients scored signicantly higher only on Big Five Inventory (BFI) Agreeableness. The proportions of DSM-IV PD diagnoses, as assessed with the Structured Clinical Interview for DSM-IV Personality Disorders (SCID-II; First, Spitzer, Gibbon, & Williams, 1997), in the sample were as follows: Paranoid 11%, Schizoid 3%, Schizotypal 1%, Antisocial 14%, Borderline 17%, Histrionic 0%, Narcissistic 1%, Avoidant 15%, Dependent 4%, Obsessive-Compulsive 8%, Depressive 20%, and Passive-Aggressive 7%. The distribution of the number of PD diagnoses in this sample was as follows: 58% had no diagnosis, 18% had one, 8% had two, 5% had three, 6% had four, 1% had ve, 2% had six, and 1% had seven PD diagnoses. Finally, with regard to co-occurrence of Axis I and Axis II disorders, 17%

Dimensions and DSM-IV Disorders of Personality

845

of the sample did not qualify for a diagnosis on Axis I or II, 38% qualied for diagnoses on both Axis I and II, 41% qualied for a diagnosis on Axis I but not on Axis II, and 6% qualied for a diagnosis on Axis II but not on Axis I (Tromp & Koot, 2009).

Measures
All participants completed the BFI ( John et al., 1991), an instrument designed to measure the Big Five higher order dimensions of normal personality: Neuroticism, Extraversion, Openness/Intellect, Agreeableness, and Conscientiousness. Each dimension was assessed by seven items. The resulting 35 items are scored on a Likert-type scale, ranging from 1 (disagree strongly) to 5 (agree strongly). The 35-item BFI is a slightly shortened version of the Dutch BFI-44 for which good psychometric properties were recently reported, including factorial equivalence to the English original and other BFI translations and relative independence and internal consistency of the ve dimensions (Denissen, Geenen, Van Aken, Gosling, & Potter, 2008). Factor analysis in a Dutch population (N 5 233) using the 35-item version replicated the original ve-factor structure, and yielded alpha coefcients ranging from .48 for Agreeableness to .81 for Neuroticism (mean 5 .69; Eileen M. Donahue, personal communication, April 3, 1995). For the present study, the wording of the items was slightly modied by the researchers to ensure age-appropriate assessment. Words that were deemed difcult for adolescents (e.g., aesthetic, efcient, sophisticated ) were replaced with more comprehensible wordings, and all items were altered into full sentences in rst-person form. In the present sample, principal components analysis with varimax rotation extracting ve factors yielded a structure that closely resembled the original ve-factor structure. Reliability coefcients (Cronbachs alphas) found in the present study were .87 for Neuroticism, .76 for Extraversion, .70 for Openness/Intellect, .62 for Agreeableness, and .55 for Conscientiousness. Research has shown that the BFI items captured the same ve factors as the NEO-PI-R (Costa & McCrae, 1992), the predominant measure of the FFM (Reynolds & Clark, 2001; see also Goldberg, 1993). In addition, participants completed the DAPP-BQ-A (Tromp & Koot, 2008), an age-appropriate operationalization of Livesleys model of personality pathology. This self-report questionnaire assesses four higher order dimensions, which are labeled Emotional Dysregulation (170 items, mean interitem correlation [MIIC ] 5 .25, Cronbachs a in the present sample 5 .98), Dissocial Behavior (64 items, MIIC 5 .20, a 5 .84), Inhibitedness (32 items, MIIC 5 .08, a 5 .73), and Compulsivity (16 items, MIIC 5 .29, a 5 .87). In addition, the DAPP-BQ-A assesses 18 lower order dimensions, with mean interitem correlations ranging from .11 for Restricted Expression to .72 for Self-Harm (median 5 .33): Submissiveness (a 5 .91), Cognitive Distortion

846

Tromp & Koot

(a 5 .91), Identity Problems (a 5 .78), Affective Instability (a 5 .90), Oppositionality (a 5 .87), Anxiety (a 5 .94), Social Avoidance (a 5 .91), Suspiciousness (a 5 .88), Insecure Attachment (a 5 .93), Narcissism (a 5 .88), Self-Harm (a 5 .97), Stimulus Seeking (a 5 .86), Callousness (a 5 .86), Rejection (a 5 .87), Conduct Problems (a 5 .89), Restricted Expression (a 5 .67), Intimacy Problems (a 5 .71), and Compulsivity (a 5 .87). Each of the lower order dimensions is measured by 16 items, except the scales for SelfHarm and Suspiciousness, which contain 12 and 14 items, respectively. In addition, eight items are included to measure social desirability. The 290 items are scored on a Likert-type scale, ranging from 1 (very unlike me) to 5 (very like me). The DAPP-BQ-A was translated and adapted from its adult predecessor, the DAPP-BQ (Livesley & Jackson, 2009). The translation and adaptation procedures are reported in more detail elsewhere (Tromp & Koot, 2008). Subsequently, all participants were interviewed using the Dutch version of the SCID-II (Weertman, Arntz, Dreessen, Van Velzen, & Vertommen, 2003; Weertman, Arntz, & Kerkhofs, 2000) by one of two trained research psychologists, who were blind to adolescents BFI and DAPP-BQ-A scores. Although the SCID-II was primarily designed for the assessment of PDs in adults, empirical ndings on the reliability and validity of structured clinical interviews in adolescent samples suggest that the SCID-II is a useful instrument at this age (Brent, Zelenak, Bukstein, & Brown, 1990; Brent et al., 1993; Grilo, Becker, Edell, & McGlashan, 2001). PD symptom counts were obtained by computing for each PD the total number of criteria that were met. The reliability coefcients for the 12 PD scales were .73 for Paranoid, .59 for Schizoid, .51 for Schizotypal, .78 for Antisocial, .84 for Borderline, .40 for Histrionic, .53 for Narcissistic, .76 for Avoidant, .76 for Dependent, .59 for ObsessiveCompulsive, .82 for Depressive, and .61 for Passive-Aggressive.

Procedure
As was described previously (Tromp & Koot, 2009), participants completed the questionnaires at home or at the mental health center in paperand-pencil format (68%) or via the Internet (32%). These groups did not differ on either age or gender. However, they did differ signicantly on the BFI dimension Agreeableness (Cohens d 5 .54), indicating higher scores for the adolescents in the Internet group, and on the DAPP-BQ-A dimensions Dissocial Behavior (d 5 .37) and Inhibitedness (d 5 .35), indicating higher scores for the paper-and-pencil group. These differences did not seem to represent systematic methodological effects, as analyses of variance (ANOVAs) showed that the effects were attributable to the scores of inpatients, the large majority of whom (87%) used

Dimensions and DSM-IV Disorders of Personality

847

paper-and-pencil format. Only for Dissocial Behavior was a small but signicant interaction effect of Assessment Format Referral Status found (Z2 5 .03), indicating that the effect of assessment format was only signicant for inpatients. Subsequently, the SCID-II was administered at the mental health center. In addition, instruments to assess demographic characteristics, Axis I disorders, IQ, and childhood trauma were administered but not used for the present study. After completing all assessments, adolescents received a voucher worth 15 Euro as acknowledgment of their participation. Study procedures were approved in accordance with appropriate Dutch national guidelines by the Central Committee on Research involving Human Subjects.

Statistical Analyses
Because several cross-loadings appeared in the factor structure of the DAPP-BQ-A, factor scores were computed on each higher order dimension for each participant based on the factor structure in a sample of 1,798 clinical and nonclinical adolescents (Tromp & Koot, 2008). Because considerable gender and some age differences in the number of PD symptoms and the scores on dimensions of normal and abnormal personality were observed, all analyses controlled for gender and age. Partial correlations were computed between the ve BFI and four DAPP-BQ-A higher order dimensions. Next, three series of multiple hierarchical regressions were conducted following analytic procedures similar to those used previously by Tromp and Koot (2009) that evaluated the incremental variance in PD symptom counts accounted for by DAPP-BQ-A dimensions over and above gender, age, and general PD severity. Specically, in the rst series, after entering gender and age into the model, the DAPPBQ-A higher order dimensions were entered as one block, followed by the BFI higher order dimensions, to determine if the latter explained an incremental proportion of variance in PD symptom counts. In the second series, the entry sequence was reversed to test the incremental contribution of the DAPP-BQ-A higher order dimensions. Gender and age were entered rst, followed by a block including the BFI higher order dimensions, which in turn was followed by a block including the DAPP-BQ-A higher order dimensions. In the nal series of multiple hierarchical regressions, after entering gender and age into the model, the BFI higher order dimensions were entered as one block, followed by a block including the DAPP-BQ-A lower order dimensions that showed signicant univariate correlations with the symptom counts for each respective PD (cf. Tromp & Koot, 2009). Individual beta weights of the BFI and DAPP-BQ-A dimensions were evaluated to see which predictors uniquely (po.05) contributed to the regression models.

848 RESULTS Correlation Results

Tromp & Koot

The correlations between the BFI and DAPP-BQ-A higher order dimensions are presented in Table 1. The correlations for conceptually related dimensions were in the predicted directions. In terms of Cohens (1988) criteria, a high correlation (4|.50|) was found between BFI Neuroticism and DAPP-BQ-A Emotional Dysregulation (.69), and medium (4|.30|) correlations were found between BFI Extraversion and DAPP-BQ-A Inhibitedness ( .49), BFI Agreeableness and DAPP-BQ-A Dissocial Behavior ( .39), and BFI Conscientiousness and DAPP-BQ-A Compulsivity (.36). Table 1 also shows two additional medium correlations: BFI Agreeableness and DAPP-BQ-A Inhibitedness ( .34) and BFI Conscientiousness and DAPP-BQ-A Emotional Dysregulation and Dissocial Behavior (both .35). In terms of Cohens criteria, all other correlations were small (  |.30|).
Regression Results

First, regression analyses were conducted at the higher order level. Table 2 reports (in the rst column) the proportion of variance in PD symptom counts explained by the full model, including gender, age, and BFI and DAPP-BQ-A higher order dimensions. As the

Table 1 Partial Correlations Between BFI and DAPP-BQ-A Higher Order Dimensions
BFI ExtraAgreeNeuroticism version Openness ableness Conscientiousness .69nnn .06 .17n .05
nnn

DAPP-BQ-A Emotional Dysregulation Dissocial Behavior Inhibitedness Compulsivity

n

.30nnn .23nn .49nnn .13

.00 .07 .01 .20nn

.08 .39nnn .34nnn .01

.35nnn .35nnn .02 .36nnn

po.05,

nn

po.01,

po.001.

Dimensions and DSM-IV Disorders of Personality

849

multicollinearity statistics yielded nonsignicant results (all variance ination factors [VIF]o10, and mean VIF 5 1.87; Bowerman & OConnell, 1990; Myers, 1990), all higher order dimensions were included as predictor variables. On average, all predictors together accounted for a total adjusted R2 of .30 (range .07 for Histrionic to .55 for Depressive). The incremental contribution of the BFI higher order dimensions over and above the higher order dimensions of the DAPP-BQ-A was examined. The DAPP-BQ-A higher order dimensions accounted for a signicant proportion of variance in all PDs, with a mean value for adjusted R2 change of .26 (range .07 for Schizoid to .44 for Depressive). The BFI higher order dimensions provided a signicant incremental contribution for 4 of the 12 PDs. The mean adjusted R2 change was .01, with signicant increments for Schizoid (.03), Avoidant (.02), Obsessive-Compulsive (.04), and Depressive (.04) PDs. Examination of the individual beta weights of the BFI higher order dimensions showed that Neuroticism uniquely (po.05) contributed to explained variance in Obsessive-Compulsive (b 5 .25) and Depressive PD (.19) symptoms, Extraversion to Avoidant ( .16), Conscientiousness to Schizoid (.21) and Obsessive-Compulsive (.20), and Openness to Depressive (.13). Subsequently, the entry sequence was reversed. Table 2 shows that the BFI higher order dimensions accounted for a signicant proportion of variance in all but 1 (Histrionic) of the 12 PDs. The mean value for adjusted R2 change was .16 (range .01 for Histrionic to .32 for Depressive). Table 2 further shows that the DAPP-BQ-A higher order dimensions provided a signicant incremental contribution for all but one PD (Schizoid). The mean adjusted R2 change was .11 (range .02 for Schizoid to .25 for Antisocial). Examination of the individual beta weights of the DAPP-BQ-A higher order dimensions showed that Emotional Dysregulation uniquely (po.05) contributed to explained variance in Paranoid, Schizotypal, Borderline, Avoidant, Dependent, Depressive, and Passive-Aggressive PD symptoms; Dissocial Behavior to Antisocial, Borderline, Histrionic, Narcissistic, Obsessive Compulsive, and Passive-Aggressive; Inhibitedness to Paranoid, Antisocial, Avoidant, and Depressive; and Compulsivity to Antisocial, Borderline, and Depressive (all negatively). The third series of regression analyses included DAPP-BQ-A lower order dimensions as predictor variables. As the multicollinearity statistics yielded nonsignicant results (all VIF o10, and

Table 2 Hierarchical Multiple Regressions of DSM-IV Personality Disorder (PD) Symptom Counts: DAPP-BQ-A Higher Order and Lower Order Dimensions over BFI Higher Order Dimensions
I: DAPP-BQ-A Higher Order DR2adj DF df 17,143 3.27nnn .10 6.73nnn ED (.39), IH (.18) .00 0.67 5,155 .16 DF b Predictors (b) DR2adj II: DAPP-BQ-A Lower Order Predictors (b) Suspiciousness (.45)

BFI

PD Symptom Counts (Total R2adj) DR2adj

DF a

Predictors (b)

PAR (.29) .02 1.74

.20

8.99nnn

N (.33), A ( .25)

SZD (.15)

.09

4.33nn

E ( .19), A ( .18), C (.16) .07 4.82nn ED (.37) .10 2.41nn

SZT (.25)

.16

7.50nnn

N (.27), C ( .19), E ( .15) .25 18.26nnn DB (.51), CP ( .24), IH (.16) ED (.44), .28

16,144

Cognitive Distortion (.37), Suspiciousness (.24)

ANT (.42)

.12

5.65nnn

A ( .27), C ( .20), E (.18) .20 15.68nnn

7.27nnn

13,147

Conduct Problems (.45), Callousness (.19)

BPD (.47)

.24

11.97nnn

N (.39),

.26

6.50nnn

16,144

Self-Harm (.34), Stimulus Seeking (.23),

(Continued)

Table 2 (Cont.)
I: DAPP-BQ-A Higher Order DR2adj df DB (.27), CP ( .21) .07 DB (.23) .17 3.48nnn 4.10nn 13,147 DF b Predictors (b ) DR2adj DF II: DAPP-BQ-A Lower Order Predictors (b) Submissiveness (.22), Anxiety ( .35) Self-Harm (.32), Submissiveness (.28), Callousness (.25), Narcissism (.23) 15,145 Self-Harm (.29), Narcissism (.26), Identity Problems ( .39) .13 3.89nnn 14,146 Social Avoidance (.36), Self-Harm (.27) .19 4.29nnn 15,145 Submissiveness (.45), Self-Harm (.23) DB (.33) .07 1.84n 16,144 Rejection (.24)

BFI

PD Symptom Counts (Total R2adj) DR2adj

DF a

Predictors (b )

C ( .21), A ( .17)

HIS (.07)

.01

1.18

NAR (.22) .10 5.88nnn DB (.35) .15

.10

4.84nnn

A ( .30)

3.28nnn

AVD (.43) .10 7.99nnn ED (.53), IH (.15) ED (.56) .10 6.42nnn

.25

13.51nnn

N (.35), E ( .33)

DEPT (.28) .07 4.75nn

.11

5.40nnn

N (.27), C ( .23)

OC (.22)

.12

5.53nnn

N (.36), A ( .15)

(Continued)

Table 2 (Cont.)
I: DAPP-BQ-A Higher Order DR2adj df 16,144 .16 15.02nnn .18 5.17nnn DF b Predictors (b) DR2adj DF II: DAPP-BQ-A Lower Order Predictors (b) Self-Harm (.43), Suspiciousness (.19)

BFI

PD Symptom Counts (Total R2adj) DR2adj

DF a

Predictors (b)

DEPR (.55) .10 6.58nnn .09 2.18nn 16,144

.32

18.52nnn

PA (.28) .11 .15

.17

7.90nnn

N (.52), E ( .15), O (.14) N (.30), A ( .25)

ED (.53), IH (.15), CP ( .13) ED (.36), DB (.29)

Mean DR2adj (.30)

.16

Note. Analyses controlled for gender and age. DAPP-BQ-A 5 Dimensional Assessment of Personality Pathology Basic Questionnaire for Adolescents (Tromp & Koot, 2008); BFI 5 Big Five Inventory ( John et al., 1991). I 5 incremental contribution of DAPP-BQ-A higher order dimensions over BFI higher order dimensions. II 5 incremental contribution of DAPP-BQ-A lower order dimensions over BFI higher order dimensions. PAR 5 Paranoid, SZD 5 Schizoid, SZT 5 Schizotypal, ANT 5 Antisocial (A-criteria), BPD 5 Borderline, HIS 5 Histrionic, NAR 5 Narcissistic, AVD 5 Avoidant, DEPT 5 Dependent, OC 5 Obsessive Compulsive, DEPR 5 Depressive, PA 5 PassiveAggressive. N 5 Neuroticism, E 5 Extraversion, O 5 Openness, A 5 Agreeableness, C 5 Conscientiousness, ED 5 Emotional Dysregulation, DB 5 Dissocial Behavior, IH 5 Inhibitedness, CP 5 Compulsivity, b 5 standardized regression coefcient. All bs signicant at po.05. a Degrees of freedom 5, 160. bDegrees of freedom 4, 156. n po.05, nnpo.01, nnnpo.001.

Dimensions and DSM-IV Disorders of Personality

853

mean VIF ranged from 3.10 to 3.74; Bowerman & OConnell, 1990; Myers, 1990), all lower order dimensions that showed signicant univariate correlations with the symptom counts of each respective PD (cf. Tromp & Koot, 2009) were included as predictor variables. The right panel of Table 2 shows that the DAPP-BQ-A lower order dimensions provided a signicant incremental contribution to variance in symptom counts for all but one PD (Schizoid). The mean adjusted R2 change was .15 (range .00 for Schizoid to .28 for Antisocial). Examination of the individual beta weights of the DAPPBQ-A lower order dimensions showed that Suspiciousness uniquely (po.05) contributed to variance in Paranoid, Schizotypal, and Depressive PD; Cognitive Distortion to Schizotypal; Conduct Problems to Antisocial; Callousness to Antisocial and Histrionic; Self-Harm to Borderline, Histrionic, Narcissistic, Dependent, Avoidant, and Depressive; Stimulus Seeking to Borderline; Submissiveness to Borderline, Histrionic, and Dependent; Anxiety to Borderline (negatively); Narcissism to Histrionic and Narcissistic; Identity Problems to Narcissistic (negatively); Social Avoidance to Avoidant; and Rejection to Obsessive-Compulsive.
DISCUSSION

The aim of the present study was to elucidate dimensions derived from two modelsthe Big Five of normal personality and Livesleys conceptualization of personality pathologyunderlying DSM-dened PD symptoms in a clinical sample of adolescents referred for inpatient and outpatient mental health services. In a previous report (Tromp & Koot, 2009), it was demonstrated that dimensions derived from Livesleys model showed signicant, substantial, and conceptually meaningful relations to DSM-IV PDs. The present study expands this focus and includes a parallel examination of the Big Five and Livesleys model, as assessed by the BFI and DAPP-BQ-A, respectively. The results showed that higher order dimensions of both models contributed signicantly to variance in adolescent PDs. However, at the higher order level, Livesleys model afforded signicant and substantial improvement over the Big Five in predicting symptoms of almost all PDs. More interestingly, lower order dimensions within Livesleys model showed unique and conceptually meaningful relations to adolescent PD symptoms, over and above the effects of higher order dimensions of normal personality. The ndings suggest that adolescent PDs are

854

Tromp & Koot

more than extreme, maladaptive variants of higher order normal personality traits and encompass characteristics that may be more completely covered by dimensions of abnormal personality. Relations between the higher order dimensions of normal and abnormal personality showed meaningful patterns largely consistent with the four higher order dimensions reported in the literature (Mervielde et al., 2005; Widiger & Simonsen, 2005). However, according to Cohens (1988) criteria, most correlations were moderate, suggesting that each model captures substantial unique variance not captured by the other model. Nevertheless, regression analyses did not support this contention. Although Livesleys model afforded additional explained variance over the Big Five for almost all PDs, the reverse did not hold, with increments in explained variance afforded by the Big Five dimensions minimal both in number and in magnitude. The correlations between the higher order dimensions of normal and abnormal personality showed one value that merits special attention. BFI Neuroticism and DAPP-BQ-A Emotional Dysregulation correlated .69, which seems to suggest that the constructs are overlapping considerably. However, in the regression analyses, Emotional Dysregulation appeared as a unique predictor, after accounting for the effects of Neuroticism and the other BFI dimensions, for symptoms of 7 out of 12 PDs. Neuroticism uniquely contributed to symptoms of only 2 PDs. Thus, Emotional Dysregulation seems to capture core features of disordered personality that are not included in Neuroticism or any of the other BFI dimensions. This conclusion is in agreement with results from an examination of the phenotypic and genetic higher order structure of PD in adults, which also suggested that Emotional Dysregulation, although similar, has more extensive coverage than Neuroticism (Livesley, Jang, & Vernon, 1998). Similar to a previous study in adolescents (De Clercq & De Fruyt, 2003) and to meta-analytic ndings in adults (Saulsman & Page, 2004), the present regression analyses yielded two prominent dimensions common across adolescent PDs: high BFI Neuroticism/DAPPBQ-A Emotional Dysregulation and low BFI Agreeableness/high DAPP-BQ-A Dissocial Behavior. Also consistent with the two previous studies is the lack of relations between the normal personality dimension Openness and symptoms of any of the PDs, with the exception of Depressive PD, for which Openness was found to be a small positive predictor. The value of BFI Openness as a dimension of disordered personality thus seems negligible. The consistency in

Dimensions and DSM-IV Disorders of Personality

855

ndings between the previous studies (De Clercq & De Fruyt, 2003; Saulsman & Page, 2004) and the present study is particularly notable considering the differences between these studies in sample (nonclinical vs. clinical), measure of normal personality dimensions (NEOPI-R and others vs. BFI), as well as in the explanatory model (Big Five vs. Livesleys model). Despite the consistencies between the meta-analytic ndings reported for adults (Saulsman & Page, 2004) and the present adolescent ndings, there also was a difference. The results of the regression analyses showed that low BFI Extraversion/high DAPPBQ-A Inhibitedness was characteristic of several PDs, with the exception of Antisocial PD. In contrast, adult correlation ndings showed directionally mixed relations for Extraversion (e.g., positive for Histrionic and negative for Schizoid PD). Interestingly, the present pattern did align with previous ndings in nonclinical adolescents, which also demonstrated consistently negative correlations of PDs with Extraversion (De Clercq & De Fruyt, 2003; De Clercq et al., 2004). In sum, in addition to high Neuroticism/Emotional Dysregulation and low Agreeableness/high Dissocial Behavior, the higher order dimension low Extraversion/high Inhibitedness was common across many PDs in adolescents. From a developmental perspective, the adolescent ndings indicate smaller degrees of differentiation and specialization of disordered personality compared to adult populations. In a study comparing within-category cohesiveness of PD criteria and between-category criterion overlap of PD categories in small samples of adolescent and adult inpatients, Becker and colleagues (1999) also concluded that PD diagnoses seem to be less sharply differentiated among adolescents. As indicated in the introduction, the more diffuse structure of personality pathology in adolescence may be partly due to the possibility that disordered personality, if present, still is in a developing stage at this age. Also, some behavioral features that are considered indicative of specic types of pathology in adulthood may reect normative behavior at this developmental stage (e.g., Identity Problems as a feature of Borderline PD). Research has indeed shown that PD symptoms may be more numerous during adolescence than during adulthood (Johnson et al., 2006). Higher overall PD symptom levels may result in higher overlap of these symptoms in adolescents. Whereas examination of the regression results across the PDs showed uniformity, examination within the PDs provided some

856

Tromp & Koot

differentiation between PDs, showing somewhat unique patterns of associations with BFI and DAPP-BQ-A higher order dimensions. For example, whereas both Antisocial and Borderline PD symptoms displayed a positive relation with DAPP-BQ-A Dissocial Behavior and a negative relation with Compulsivity, Antisocial showed an additional positive relation with Inhibitedness, and Borderline showed an additional positive relation with Emotional Dysregulation. Nevertheless, several PDs that form distinct categories within the DSM system (and even fall within different DSM clusters) showed similar BFI and DAPP-BQ-A proles. For example, Paranoid, Obsessive-Compulsive, and Passive-Aggressive PDs were all characterized by high BFI Neuroticism and low Agreeableness. Similarly, Schizotypal, Avoidant, and Depressive PDs were characterized by high DAPP-BQ-A Emotional Dysregulation and high Inhibitedness. Thus, at the higher order level of the Big Five and Livesleys model, it seemed difcult to differentiate between adolescent PDs. Regression analyses at the lower order level of the DAPP-BQ-A yielded interesting ndings that could improve the differentiation and understanding of adolescent personality pathology. The analyses showed that specic dimensions of abnormal personality are uniquely and in conceptually meaningful ways related to adolescent PDs. In line with Livesleys denition of PDs, intrapersonal (e.g., Identity Problems) and interpersonal characteristics (e.g., Submissiveness), traits indicating disturbed prosocial behavior (e.g., Callousness), as well as dysfunctional behavioral traits (e.g., Self-Harm, Cognitive Distortion) appeared as unique predictors of adolescent PD symptoms, above and beyond variance accounted for by normal personality traits. As indicated before, DAPP-BQ-A Emotional Dysregulation seemed to capture core features of disordered personality not captured by BFI Neuroticism. Lower order dimensions within this DAPP-BQ-A dimension, such as Identity Problems, Submissiveness, Self-Harm, and Cognitive Distortion, may account for the additional value of Emotional Dysregulation. Self-Harm appeared as a signicant predictor for symptoms of six PDs, thereby possibly functioning as a more common indicator of the severity of personality pathology. However, Self-Harm was always accompanied by other more specic characteristics that seemed to be indicators of specic dysfunctional traits within the PDs (e.g., Social Avoidance of Avoidant PD and Submissiveness of Dependent PD). Similarly, whereas both Paranoid and Obsessive-Compulsive PDs were

Dimensions and DSM-IV Disorders of Personality

857

characterized by high Neuroticism and low Agreeableness within the Big Five, Paranoid was characterized by the Suspiciousness dimension and Obsessive-Compulsive by the Rejection dimension within the lower order level of Livesleys model. The lower order dimensions within Livesleys model thus seem to offer a valuable supplement to the higher order normal personality traits in the differentiation between adolescent PDs.
Limitations and Implications

Several limitations may qualify the ndings reported in this paper, the majority of which regard the operationalization of normal and abnormal personality. First, the BFI with its 35 items could have an a priori psychometric disadvantage compared to the 290-item DAPP-BQ-A. The limited number of items could result in restriction of range in a clinical sample due to uniformly high or low scores. However, post hoc ShapiroWilk tests of normality only showed negligible negative skewness for Neuroticism and Openness. Moreover, the incremental contribution of the DAPP-BQ has also been shown in relation to the 240-item NEO-PI-R, suggesting that the additional explained variance can not be ascribed to a larger number of items (Bagby, Marshall, et al., 2005). Another limitation regards the psychometric properties of the administered Dutch versions of the BFI and SCID-II. The moderate Cronbachs alphas for two of the BFI dimensions, Agreeableness and Conscientiousness, may qualify conclusions on the differences in predictive power between the DAPP-BQ-A and the BFI. In addition, the consequences of the modications applied to the BFI items to ensure age-appropriate assessment were not investigated. However, it should be noted that only slight modications were applied. Furthermore, it is noteworthy that the relations between the BFI dimensions and DSM-IV PD symptoms found in the present sample were comparable to those found previously between dimensions of FFM measures and DSM-IV PD symptoms in adolescents (De Clercq & De Fruyt, 2003; De Clercq et al., 2004). With regard to the SCID-II, interrater reliability estimates were not available. Both research psychologists who administered the interviews were trained extensively by the same experienced diagnostician before the data collection of the present study started. The rst 20 interviews were scored jointly by both interviewers to arrive at consensus scores

858

Tromp & Koot

at the symptom level. Unfortunately, separate scores were not preserved to examine interrater reliability. Third, it may be suggested that the incremental contribution of DAPP-BQ-A over the BFI dimensions in the prediction of PD symptoms could be attributable to criterion contamination. The research that led to the development of the DAPP-BQ was originally intended to validate DSM PDs. However, the DAPP-BQ was not designed to assess DSM syndromes, and its items do not correspond to DSM criteria. The DSM-III criteria, argued by Livesley (2006) to have been developed on the basis of informal and unsystematic samplings of clinical opinion rather than systematic denitions of diagnoses, were not adopted but only used to provide a preliminary theoretical taxonomy. A modication of the lexical approach, also used to explicate the Big Five, was used to develop denitions of each PD and included extensive content analyses of interviews with PD patients and the clinical literature (Livesley, 2006). A fourth limitation regards the operationalization of normal personality, which did not allow for analyses using the lower order dimensions of normal personality. Research has demonstrated that analyses at the lower order level provided substantial increase in specicity and discrimination between PDs as well as richer description of the PDs compared to analyses at the higher order level (Bagby, Costa, Widiger, Ryder, & Marshall, 2005; Reynolds & Clark, 2001). This may be particularly true for adolescent PDs, which show somewhat smaller degrees of differentiation than adult PDs. Specic PDs characterized by the same pattern of personality dimensions (e.g., in the present sample, Paranoid, Obsessive-Compulsive, and Passive-Aggressive PDs) may be more easily discriminated based on the composing lower order dimensions. Further research is needed to investigate the relevance of lower order dimensions of abnormal personality, above and beyond the lower order dimensions of normal personality. If such research can replicate the conclusions of the present study, the appropriateness of Livesleys (2006) conceptualization of personality pathology for describing PD symptoms will be further underscored. A nal limitation of the present study concerns the criterion against which the explanatory value of the dimensional models was tested. DSM-IV-based PD descriptions are increasingly criticized (Trull & Durrett, 2005). The prediction of DSM-dened PD symptoms should therefore not be the ultimate goal of dimensional

Dimensions and DSM-IV Disorders of Personality

859

models. However, no good alternatives are currently available, and the DSM system still is the most widely used taxonomy in research and clinical practice. Despite these limitations, the present study has several important implications. The ndings suggest that a major point of critique on the categorical DSM system in adulthood, that is, the excessive diagnostic overlap among PDs, is equally, possibly even more, applicable in adolescence. Dimensional models of personality may provide a solution to this problem by allowing traits to be continuously distributed and by including assessment based on trait continua instead of diagnostic categories. Another implication concerns theories on the developmental trajectories of temperament, personality, and psychopathology. Shiner and Caspi (2003) presented a conceptual model of possible ways in which temperament/personality and psychopathology are associated. Their taxonomy of temperament/personality in children and adolescents, further elaborated in a subsequent article (Caspi, Roberts, & Shiner, 2005), shows clear correspondence to the Big Five dimensions. However, the present ndings suggest that efforts to nd empirical support for the developmental trajectories into psychopathology should incorporate dimensions of abnormal personality. This is in line with the proposal of Mervielde et al. (2005) that adaptive and maladaptive personality traits can be integrated in a common dimensional model that can be used to clarify the associations between personality and psychopathology in children and adolescents. Because of the cross-sectional design of the present study, the ndings did not provide insight into the developmental course of adolescent disordered personality. Future studies utilizing longitudinal designs and examining both normal and abnormal personality traits could offer empirical evidence for the developmental pathways into disordered personality. It seems likely that dimensions of normal personality are informative predictors of later-appearing disordered personality. From the perspective of the empirical ndings presented in the present paper, a supplementary model to describe adolescent disordered personality including higher order dimensions of normal and abnormal personality seems redundant, because adding dimensions of normal personality to those of abnormal personality offered little extra explanatory value to the model. A practical implication following from this is that diagnostic procedures to assess personality

860

Tromp & Koot

pathology in adolescent clinical settings could be limited to the assessment of abnormal personality. This will result in a smaller burden, valuable especially for patients of younger ages. On the other hand, it may be useful from a clinical perspective to include assessment of normal personality traits to enhance insight into the patients strengths, beyond the pathological trait characteristics. As Livesley and Jang (2000) stated, a denition of personality disorder as a harmful dysfunction requires an understanding of the adaptive functions of personality and how these functions are impaired (p. 143). Assessment of the abnormal dimensions within Livesleys model, in addition to assessment of normal personality, may improve diagnostic procedures as well as provide specic indications for treatment interventions through a more complete description of the deviant characteristics of adolescent personality pathology. In conclusion, the present ndings suggest that adolescent PDs are more than extreme, maladaptive variants of higher order normal personality traits and encompass characteristics that may be more completely covered by dimensions of abnormal personality. Future studies would be well served by applying longitudinal designs to shed light on the developmental pathways of normal into disordered personality. Finally, future studies may attempt to extend knowledge on adolescent personality pathology by investigating more comprehensive assessment methods, including independent assessment of dysfunction and using multiple informants (cf. Clark, 2007).

REFERENCES
American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders. (4th ed., text revision). Washington, DC: Author. Bagby, R. M., Costa, P. T., Widiger, T. A., Ryder, A. G., & Marshall, M. (2005). DSM-IV personality disorders and the Five-Factor Model of personality: A multi-method examination of domain- and facet-level predictions. European Journal of Personality, 19, 307324. Bagby, R. M., Marshall, M. B., & Georgiades, S. (2005). Dimensional personality traits and the prediction of DSM-IV personality disorder symptom counts in a nonclinical sample. Journal of Personality Disorders, 19, 5367. Becker, D. F., Grilo, C. M., Morey, L. C., Walker, M. L., Edell, W. S., & McGlashan, T. H. (1999). Applicability of personality disorder criteria to hospitalized adolescents: Evaluation of internal consistency and criterion overlap. Journal of the American Academy of Child and Adolescent Psychiatry, 38, 200205.

Dimensions and DSM-IV Disorders of Personality

861

Bernstein, D. P., Iscan, C., & Maser, J. (2007). Opinions of personality disorder experts regarding the DSM-IV personality disorder classication system. Journal of Personality Disorders, 21, 536551. Bowerman, B. L., & OConnell, R. T. (1990). Linear statistical models: An applied approach (2nd ed.). Belmont, CA: Duxbury. Brent, D. A., Johnson, B. A., Bartle, S., Bridge, J., Rather, C., Matta, J., et al. (1993). Personality disorder, tendency to impulsive violence, and suicidal behavior in adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 32, 6975. Brent, D. A., Zelenak, J. P., Bukstein, O., & Brown, R. V. (1990). Reliability and validity of the Structured Interview for Personality Disorders in adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 29, 349 354. Caspi, A., Roberts, B. W., & Shiner, R. L. (2005). Personality development: Stability and change. Annual Review of Psychology, 56, 453484. Clark, L. A. (1993). Manual for the Schedule for Non-adaptive and Adaptive Personality. Minneapolis: University of Minnesota Press. Clark, L. A. (2007). Assessment and diagnosis of personality disorder: Perennial issues and an emerging reconceptualization. Annual Review of Psychology, 58, 227257. Clark, L. A., & Livesley, W. J. (2002). Two approaches to identifying the dimensions of personality disorder: Convergence on the Five-Factor Model. In P. T. Costa & T. A. Widiger (Eds.), Personality disorders and the Five-Factor Model of personality (2nd ed., pp. 161176). Washington, DC: American Psychological Association. Clark, L. A., Livesley, W. J., Schroeder, M. L., & Irish, S. L. (1996). Convergence of two systems for assessing specic traits of personality disorder. Psychological Assessment, 8, 294303. Clark, L. A., Vorhies, L., & McEwen, J. L. (2002). Personality Disorder symptomatology from the Five-Factor Model perspective. In P. T. Costa & T. A. Widiger (Eds.), Personality disorders and the Five-Factor Model of personality (2nd ed., pp. 125147). Washington, DC: American Psychological Association. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum. Costa, P. T., & McCrae, R. R. (1990). Personality disorders and the ve-factor model of personality. Journal of Personality Disorders, 4, 362371. Costa, P. T., & McCrae, R. R. (1992). Revised NEO Personality Inventory (NEOPI-R) and NEO Five-Factor Inventory (NEO-FFI) professional manual. Odessa, FL: Psychological Assessment Resources. De Clercq, B., & De Fruyt, F. (2003). Personality disorder symptoms in adolescence: A ve-factor model perspective. Journal of Personality Disorders, 17, 269292. De Clercq, B., De Fruyt, F., & Van Leeuwen, K. (2004). A little ve lexically based perspective on personality disorder symptoms in adolescence. Journal of Personality Disorders, 18, 479499. De Clercq, B., De Fruyt, F., Van Leeuwen, K., & Mervielde, I. (2006). The structure of maladaptive personality traits in childhood: A step toward an

862

Tromp & Koot

integrative developmental perspective for DSM-V. Journal of Abnormal Personality, 115, 639657. Denissen, J. J. A., Geenen, R., Van Aken, M. A. G., Gosling, S. D., & Potter, J. (2008). Development and validation of a Dutch translation of the Big Five Inventory (BFI). Journal of Personality Assessment, 90, 152157. Digman, J. M. (1990). Personality structure: Emergence of the ve-factor model. Annual Review of Psychology, 41, 417440. First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (1996). Structured Clinical Interview for Axis I DSM-IV DisordersPatient Edition (SCID-I/P). New York: Biometrics Research Department, New York State Psychiatric Institute. First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (1997). Structured Clinical Interview for DSM-IV Personality Disorders (SCID-II). Washington, DC: American Psychiatric Press. Goldberg, L. R. (1993). The structure of phenotypic personality traits. American Psychologist, 48, 2634. Grilo, C. M., Becker, D. F., Edell, W. S., & McGlashan, T. H. (2001). Stability and change of DSM-III-R personality disorder dimensions in adolescents followed up 2 years after psychiatric hospitalization. Comprehensive Psychiatry, 42, 364368. John, O. P., Donahue, E. M., & Kentle, R. (1991). The Big Five Inventory Versions 4a and 54, Technical Report. Berkeley: University of California, Institute of Personality and Social Psychology. Johnson, J. G., Bromley, E., Bornstein, R. F., & Sneed, J. R. (2006). Adolescent personality disorders. In D. A. Wolfe & E. J. Mash (Eds.), Behavioral and emotional disorders in children and adolescents: Nature, assessment, and treatment (pp. 463484). New York: Guilford Press. Kaufman, J., Birmaher, B., Brent, D., Rao, U., Flynn, C., Moreci, P., et al. (1997). Schedule for Affective and Schizophrenia for School-Age Children-Present and Lifetime Version (K-SADS-PL): Initial reliability and validity data. Journal of the American Academy of Child and Adolescent Psychiatry, 36, 980988. Klonsky, E. D., Oltmanns, T. F., & Turkheimer, E. (2003). Deliberate self-harm in a nonclinical population: Prevalence and psychological correlates. American Journal of Psychiatry, 160, 15011508. Larstone, R. M., Jang, K. L., Livesley, W. J., Vernon, P. A., & Wolf, H. (2002). The relationship between Eysencks P-E-N model of personality, the ve-factor model of personality, and traits delineating personality dysfunction. Personality and Individual Differences, 33, 2537. Livesley, W. J. (2006). The Dimensional Assessment of Personality Pathology (DAPP) approach to personality disorder. In S. Strack (Ed.), Differentiating normal and abnormal personality (2nd ed., pp. 401425). New York: Springer. Livesley, W. J. (2007). A framework for integrating dimensional and categorical classications of personality disorder. Journal of Personality Disorders, 21, 199224. Livesley, W. J., & Jackson, D. N. (2009). Manual for the Dimensional Assessment of Personality PathologyBasic Questionnaire (DAPP-BQ). Port Huron, MI: Sigma Assessment Systems. Livesley, W. J., & Jang, K. L. (2000). Toward an empirically based classication of personality disorder. Journal of Personality Disorders, 14, 137151.

Dimensions and DSM-IV Disorders of Personality

863

Livesley, W. J., Jang, K. L., & Vernon, P. A. (1998). Phenotypic and genetic structure of traits delineating personality disorder. Archives of General Psychiatry, 55, 941948. Livesley, W. J., & Schroeder, M. L. (1990). Dimensions of personality disorder: The DSM-III-R Cluster A diagnoses. Journal of Nervous and Mental Disease, 178, 627635. Livesley, W. J., Schroeder, M. L., Jackson, D. N., & Jang, K. L. (1994). Categorical distinctions in the study of personality disorder implications for classication. Journal of Abnormal Psychology, 103, 617. Marcia, J. E. (2006). Ego identity and personality disorders. Journal of Personality Disorders, 20, 577596. Mervielde, I., De Clercq, B., De Fruyt, F., & Van Leeuwen, K. (2005). Temperament, personality, and developmental psychopathology as childhood antecedents of personality disorders. Journal of Personality Disorders, 19, 171201. Myers, R. (1990). Classical and modern regression with applications (2nd ed.). Boston: Duxbury. Nestadt, G., Costa, P. T., Hsu, F. C., Samuels, J., Bienvenu, O. J., & Eaton, W. W. (2008). The relationship between the ve-factor model and latent Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition personality disorder dimensions. Comprehensive Psychiatry, 49, 98105. Reynolds, S. K., & Clark, L. A. (2001). Predicting dimensions of personality disorder from domains and facets of the Five-Factor Model. Journal of Personality, 69, 199222. Saulsman, L. M., & Page, A. C. (2004). The ve-factor model and personality disorder empirical literature: A meta-analytic review. Clinical Psychology Review, 23, 10551085. Schroeder, M. L., Wormworth, J. A., & Livesley, W. J. (2002). Dimensions of personality disorder and the ve-factor model of personality. In P. T. Costa & T. A. Widiger (Eds.), Personality disorders and the ve-factor model of personality (2nd ed., pp. 149160). Washington, DC: American Psychological Association. Shiner, R., & Caspi, A. (2003). Personality differences in childhood and adolescence: Measurement, development, and consequences. Journal of Child Psychology and Psychiatry, 44, 232. Tromp, N. B., & Koot, H. M. (2008). Dimensions of personality pathology in adolescents: Psychometric properties of the DAPP-BQ-A. Journal of Personality Disorders, 22, 623638. Tromp, N. B., & Koot, H. M. (2009). Dimensions of personality pathology in adolescents: Relations to DSM-IV personality disorder symptoms. Journal of Personality Disorders, 23, 514527. Trull, T. J., & Durrett, C. A. (2005). Categorical and dimensional models of personality disorder. Annual Review of Clinical Psychology, 1, 355380. Trull, T. J., Widiger, T. A., & Burr, R. (2001). A structured interview for the assessment of the Five-Factor Model of personality: Facet-level relations to the Axis II personality disorders. Journal of Personality, 69, 175198. Weertman, A., Arntz, A., Dreessen, L., van Velzen, C., & Vertommen, S. (2003). Short-interval test-retest interrater reliability of the Dutch version of the

864

Tromp & Koot

Structured Clinical Interview for DSM-IV Personality Disorders SCID-II. Journal of Personality Disorders, 17, 562567. Weertman, A., Arntz, A., & Kerkhofs, M. (2000). Handleiding gestructureerd klinisch interview voor DSM-IV As-II persoonlijkheidsstoornissen [Manual for the Structured Clinical Interview for DSM-IV Axis II personality disorders]. Lisse, The Netherlands: Swets Test Publishers. Widiger, T. A., & Costa, P. T. (2002). Five-factor model personality disorder research. In P. T. Costa & T. A. Widiger (Eds.), Personality disorders and the vefactor model of personality (2nd ed., pp. 5987). Washington, DC: American Psychological Association. Widiger, T. A., & Simonsen, E. (2005). Alternative dimensional models of personality disorder: Finding a common ground. Journal of Personality Disorders, 19, 110130.

This document is a scanned copy of a printed document. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material.