Personality Disorders: Theory, Research, and Treatment 2014, Vol. 5, No. 3, 305–313

© 2014 American Psychological Association 1949-2715/14/$12.00 DOI: 10.1037/per0000058

BRIEF REPORT

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Stability of Narcissistic Personality Disorder: Tracking Categorical and Dimensional Rating Systems Over a Two-Year Period Aline Vater

Kathrin Ritter

Free University of Berlin, Charité–Universitätsmedizin Berlin, and Technische Universität Darmstadt

Charité–Universitätsmedizin Berlin and Free University of Berlin

Sandra Strunz

Elsa F. Ronningstam

Charité–Universitätsmedizin Berlin

Harvard Medical School/McLean Hospital

Babette Renneberg

Stefan Roepke

Free University of Berlin

Free University of Berlin and Charité–Universitätsmedizin Berlin

Personality disorders are characterized as temporally stable patterns of symptoms (APA, 2000). However, evidence on the stability of narcissistic personality disorder (NPD) is generally lacking. This study tracked the prevalence and remission rates of individual criteria for NPD over the course of 2 years. In addition, the stability of dimensional personality pathology in patients with NPD (assessed with the Dimensional Assessment of Personality Pathology, DAPP-BQ) was assessed over time. A sample of 96 patients with a diagnosis of NPD was recruited at baseline. Forty patients participated in the follow-up assessment 2 years later. Our results indicate a moderate remission rate (53%) for NPD as a categorical diagnosis. However, single NPD criteria differed in their prevalence and temporal stability, similar to findings for other personality disorders. Moreover, scores on dimensional subscales of the DAPP-BQ remained stable over time. Theoretical implications are discussed. Keywords: narcissistic personality disorder, stability, categorical diagnosis, dimensional rating

deny their problems (Kernberg, 1975); however, empirical data supporting this assumption is generally lacking. The aim of the following study was to provide data on the stability of NPD over 2 years. Only one previous study has addressed the 3-year stability of NPD with patients meeting full diagnostic criteria for the disorder (Ronningstam, Gunderson, & Lyons, 1995). Ronningstam, Gunderson, and Lyons (1995) used the LEAD diagnostic standard (i.e., longitudinal, expert, all data) that integrates information from multiple sources such as psychological and neurological testing, psychotherapists, psychiatrists, and senior consultants. Diagnostic criteria according to DSM–III–R (APA, 1987) and DSM–IV (APA, 1994) were applied using the Diagnostic Interview for Narcissism (Gunderson, Ronningstam, & Bodkin, 1990). Results of this study showed that six (50%) of the 12 subjects who met the DSM–III–R criteria for NPD at baseline still qualified for this diagnosis at the 3-year follow-up. Moreover, of the 13 subjects who met the DSM–IV criteria for NPD at baseline, six (46%) continued to meet diagnostic criteria for NPD. Even though Ronningstam et al. (1995) did not provide remission rates for each diagnostic criterion, they reported significant reductions on all scales of the Diagnostic Interview for Narcissism, namely the grandiosity, interpersonal relations, reactiveness, mood states, and social/moral adaption subscales.

Narcissistic personality disorder (NPD) is characterized as a “pervasive pattern of grandiosity, need for admiration, and lack of empathy” (APA, 2000). Several authors stated that patients with NPD are resistant to change over time due to their propensity to

This article was published Online First February 10, 2014. Aline Vater, Cluster of Excellence–Languages of Emotion, Free University of Berlin, Charité–Universitätsmedizin Berlin, Campus Benjamin Franklin, and Department of Psychology, Technische Universität Darmstadt; Kathrin Ritter, Charité–Universitätsmedizin Berlin, Campus Benjamin Franklin and Department for Educational Science and Psychology, Free University of Berlin; Sandra Strunz, Charité–Universitätsmedizin Berlin, Campus Benjamin Franklin; Elsa F. Ronningstam, Harvard Medical School/McLean Hospital; Babette Renneberg, Department for Educational Science and Psychology, Free University of Berlin; Stefan Roepke, Cluster of Excellence–Languages of Emotion, Free University of Berlin and Charité–Universitätsmedizin Berlin, Campus Benjamin Franklin. Aline Vater and Kathrin Ritter contributed equally to this project and should be considered co-first authors. Correspondence concerning this article should be addressed to Aline Vater, Free University of Berlin, Cluster of Excellence–Languages of Emotion, Habelschwerdter Allee 45, 14195 Berlin, Germany. E-mail: [email protected] 305

VATER ET AL.

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

306

Four other studies explored the stability of NPD criteria including patients with mixed diagnoses or nonclinical samples. None of these studies included patients meeting diagnostic criteria for NPD. Ball and colleagues investigated the stability of DSM–III–R narcissistic features in patients with substance abuse (Ball, Rounsaville, Tennen, & Kranzler, 2001). Results of this study suggest a moderately high stability (r ⫽ .48 between baseline and 1-year follow-up) for DSM–III–R NPD features. Lenzenweger (1999) provided information on remission of personality disorders (according to DSM–III–R) within a sample of 250 nonclinical individuals (using data from the Longitudinal Study of Personality Disorders, LSPD). The data suggests moderate to high stability of personality disorders over time. Using individual growth trajectories, Lenzenweger, Johnson, and Willett (2004) reanalyzed this data and observed variable stability of personality disorder features over time, as individuals showed stable, increasing, and decreasing symptom trajectories. According to these data, DSM criteria do a poor job of approximating the clinical reality with respect to the stability of impairment among those with personality disorder (PD). Moreover, remission rates for narcissistic features were comparable with those of other personality disorders (r ⫽ .39). Samuel et al. (2011) used a large sample of 668 patients (from the Collaborative Longitudinal Personality disorder Study, CLPS) with mixed diagnoses and found evidence for moderate temporal stability of narcissistic features (Kappa ␬ ⫽ .36) (according to DSM–IV–TR) over 2 years (Samuel et al., 2011). Moreover, dimensional pathological personality traits (Schedule for Nonadaptive and Adaptive Personality-2; SNAP-2, Clark, Simms, Wu, & Casillas, in press) appeared to be more stable (r ⫽ .63) than the categorical diagnosis of NPD (Diagnostic Interview for DSM–IV Personality Disorders, DIPD-V, Zanarini, Frankenburg, Sickel, & Yong, 1996). Finally, a recent study (using data from the Collaborative Longitudinal Personality disorder Study, CLPS) investigated the stability of narcissistic features (according to DSM–IV–TR) in patients with mixed diagnoses (N ⫽ 266) over 10 years (Hopwood et al., 2013). Hopwood et al. found rather low stability rates for NPD features (r ⫽ .24). In sum, the results of these studies challenge the theoretical assumption that the diagnostic category of NPD and its related features are temporally stable constructs.

Aims of This Study In summary, empirical data on the prevalence and stability of NPD for clinical patients meeting diagnostic criteria is generally lacking. Notably, we have three subordinate aims. First, we focused on examining prevalence and remission rates of NPD as a diagnostic category over a 2-year period. Consistent with the studies reviewed above (Ball et al., 2001; Hopwood et al., 2013; Lenzenweger et al., 2004; Ronningstam et al., 1995; Samuel et al., 2011), we hypothesized that NPD as a diagnostic category would exhibit moderate temporal stability. In contrast to most existing studies, we included patients diagnosed with NPD. Second, we aimed to provide evidence for the temporal stability of each NPD criterion. We followed an analytical strategy by McGlashan et al. (2005) who performed rank-order hierarchies of each criterion in terms of prevalence and remission rates for various personality disorders. McGlashan et al. showed that indi-

vidual criteria for personality disorders vary in their stability. As this is the first study that analyzed the stability of individual NPD criteria, we had no empirical rationale for formulating an a priori hypothesis. Third, we aimed to provide evidence for the stability of a dimensional rating system. By doing so, we aimed to contribute to the discussion about whether dimensional rating systems are valid complementary approaches to the categorical diagnostic approach. We chose the Dimensional Assessment of Personality Pathology– Basic Questionnaire (DAPP-BQ, Livesley & Jackson, 2009). Prior to the present study, no study has assessed the stability of the DAPP-BQ in patients with NPD over time. A recent study with patients with various psychiatric disorders suggests that dimensional pathological personality traits are more stable than the categorical diagnosis over the course of 2 years (Samuel et al., 2011). There is no previous research on the stability of dimensional pathological personality traits in patients with NPD. Thus, we explored whether there is a clinically meaningful change on DAPP-BQ subscales.

Method Participants and Procedure Baseline. A total of 96 patients with NPD were recruited from the department of psychiatry at Charité Berlin and cooperating German hospitals at baseline. During their hospitalization, NPD patients were enrolled in a broad multicenter clinical study on NPD. All NPD patients agreed to be contacted for a follow-up study. Procedures were approved by the Ethics Committee of Charité Berlin. Written informed consent was obtained. Axis I diagnoses were assessed with the Structured Clinical Interview for DSM–IV for Axis I Disorders (First, Spitzer, Gibbon, & Williams, 1996; German version: Wittchen, Zaudig, & Fydrich, 1997). Axis II diagnoses of patients were assessed with the Structured Clinical Interview for DSM–IV for Personality Disorders (SCID-II, First, Gibbon, Spitzer, Williams, & Benjamin, 1997; German version: Fydrich, Renneberg, Schmitz, & Wittchen, 1997) by trained psychiatrists or psychologists. During their inpatient treatment, patients completed self-report questionnaires (see Measures section). Follow-up. After 24 months, we initially contacted participants via telephone. Patients who were unreachable via telephone, were contacted via e-mail or postal mail. A total of 40 out of 96 patients participated in the follow-up study (41.7% of all patients contacted). Of the remaining 56 (58.3% of all patients contacted) patients who did not complete the follow-up study, 33 did not respond to a letter and their telephone number was unavailable (58.9% of noncompleters), five did not answer the telephone (8.9% of noncompleters), nine rejected participation (16.1% of noncompleters), eight had moved away (14.3% of noncompleters), and one participant had committed suicide (1.8% of noncompleters). At follow-up, participants were assessed with the SCID-I and -II again. Two independent interviewers who were familiar with personality disorder diagnosis were blind to prior diagnosis at baseline. Patients were randomly assigned to interviewers and both interviewers diagnosed an equal amount of patients. For a significant number of patients it was only possible to participate in one interview session at follow-up (e.g., for approximately one third of

STABILITY OF NPD

the patients, state of residence was outside of Berlin). Moreover, the emphasis of the follow-up was placed on evaluating the stability of the categorical diagnosis and the stability of dimensional pathological personality traits. Thus, only the most prevalent SCID-I diagnoses (i.e., diagnoses with more than four patients meeting the SCID-I diagnosis at baseline) were reevaluated at follow-up (see Results section). Another reason for this procedure was to keep the length and structure of the interview constant across all participants. Patients completed self-report questionnaires before the diagnostic interviews.

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Measures Categorical diagnosis. Interrater reliability of SCID-II diagnoses was assessed with a pairwise interview design with three raters and eight patients (Kappa ␬ ⫽ 0.797) for NPD diagnosis. Moreover, interrater reliabilities for each diagnostic criterion ranged from Kappa ␬ ⫽ 0.535 minimum) to Kappa ␬ ⫽ 1.000 (maximum). Raters at follow-up were blind to prior diagnostic criteria at baseline. Dimensional pathological personality. The Dimensional Assessment of Personality Pathology (DAPP-BQ; Livesley & Jackson, 2009; German version: Pukrop, Gentil, Steinbring, & Steinmeyer, 2001) is a reliable and valid dimensional measure of personality pathology (e.g., Pukrop et al., 2009). The 5-point response scale ranges from 1 (strongly disagree) to 5 (strongly agree). In this study, the internal consistency (Cronbach’s alpha) of the narcissism scale was ␣ ⫽ .92 (baseline) and ␣ ⫽ .89 (follow-up). Moreover, internal consistencies for subscales ranged from ␣ ⫽ .75 (rejection) to ␣ ⫽ .911 (anxiousness) at baseline and from ␣ ⫽ .77 (low affiliation) to ␣ ⫽ .93 (anxiousness) at follow-up. Beck Depression Inventory (BDI). The BDI (Beck, Ward, Mendelson, Mock, & Erbaugh, 1961; German version: Hautzinger, Bailer, Worall, & Keller, 1995) was utilized to assess severity of depression. The BDI is a widely used and well-validated selfreport measure for depression and reflects the individual’s experience of specific symptoms over the past week. Participants responded to 21 items on 4-point scales which were represented by four different response statement. The internal consistency (Cronbach’s alpha) of the scale was ␣ ⫽ .87 at baseline and ␣ ⫽ .88 at follow-up.

Results Descriptive Statistics and Group Comparisons Drop-out. Sociodemographic variables are depicted in Table 1. We analyzed whether patients who participated at baseline, but not the follow-up study session (N ⫽ 56) differed significantly from patients who participated at both interview sessions (N ⫽ 40). Regarding the severity of narcissism, no significant differences between the two groups (i.e., completers vs. noncompleters) were observed in the prevalence of NPD criteria, t(93) ⫽ .530, p ⫽ .597, or scores on the narcissism subscale of the DAPP-BQ, t(93) ⫽ 4.320, p ⫽ .667, at baseline. Regarding descriptive statistics at baseline, participants who did not complete the follow-up interview session were significantly older (M ⫽ 34.8; SD ⫽ 10.2) than patients who completed both interview sessions (M ⫽ 30.2; SD ⫽ 7.0), t(93.90) ⫽ 2.647, p ⫽

307

.010. No significant differences emerged for gender (Baseline: N-male ⫽ 31, N-female ⫽ 25; follow-up: N-male ⫽ 17; N-female ⫽ 23; ␹2 ⫽ 1.543, df ⫽ 1, p ⫽ .214). Completers and noncompleters did not significantly differ in any of the comorbidity rates, with the exception of agoraphobia. None of the noncompleters had agoraphobia. However, four out of 40 of the completers received a diagnosis of agoraphobia at baseline, Fisher’s exact tests: ␹2(1) ⫽ 5.843, p ⫽ .028. General symptom reduction. According to our data, patients at baseline (M ⫽ 29.4; SD ⫽ 11.1) scored higher on depression than patients at follow-up (M ⫽ 22.1; SD ⫽ 11.7), BDI, t(35) ⫽ 4.084, p ⬍ .001, d ⫽ .66. Moreover, patients showed significantly lower comorbidity rates for dysthymia (McNemar: ␹2 ⫽ 1.77, p ⫽ .002) and trending lower rates of alcohol abuse (McNemar: ␹2 ⫽ 12.9, p ⫽ .063) at follow-up.

Prevalence of the Categorical Diagnosis and Diagnostic Criteria After 2 years, 19 of the initial 40 NPD patients still met five or more criteria for NPD according to the DSM-VI-TR. Thus, the 2-year remission rate for NPD was 52.5%. The average decrease in NPD criteria from baseline (M ⫽ 5.66, SD ⫽ 1.07) to follow-up (M ⫽ 4.29, SD ⫽ 2.12) was 1.37 criteria, t(37) ⫽ 3.970, p ⬍ .001, d ⫽ .83. Table 2 provides descriptive statistics on the direction of change, that is, information on the number of NPD patients who lost, maintained, or gained a criterion. We further examined each individual criterion at baseline and follow-up. We used the analytical strategy introduced by McGlashan et al. (2005) who provided information not only on prevalence rates of each criterion (at baseline and follow-up), but also rank-ordered remission rates. The frequency (percent) of personality disorder criteria (present and significant) at baseline (Column 1) based on SCID-II interview ratings is displayed in Table 3. Table 3 also presents the rank order of criteria from most to least prevalent. According to the data, need for admiration, fantasies of unlimited success and envy were the most prevalent criteria at baseline. Arrogance, lack of empathy and belief in uniqueness were the least prevalent criteria at baseline. Column 2 of Table 3 presents the rank order by prevalence of all criteria after the 2-year follow-up. Within 2 years, the prevalence of all criteria decreased, with the exception of belief in uniqueness, which was more prevalent after 2 years (increase of 2.6%). Envy, need for admiration, and grandiosity were the most prevalent criteria at the 2-year follow-up. Entitlement, exploitativeness, and lack of empathy were the least prevalent criteria after 2 years.

Remission Rates of Diagnostic Criteria The frequencies (percent) of criteria that remitted (i.e., that were present at baseline, but were not present at follow-up) are presented in Column 3 of Table 3. It is important to note that values in Column 3 do not represent the difference between values in Column 1 and Column 2 (for further information see McGlashan et al., 2005). Although Column 2 shows the mere presence of criteria (i.e., including criteria that were not present at baseline, but were present at follow-up), Column 3 provides information on remission of criteria that were present at baseline and were not present 2-years later. Therefore, a criterion a

VATER ET AL.

308

Table 1 Descriptive Statistics, Comorbidities and Medication of all NPD Patients Data at baseline

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

(1) NPD patients who participated baseline only (N ⫽ 56)

Data at follow-up (2) NPD patients who participated at baseline and follow-up (N ⫽ 40) SD

(3) NPD patients who participated at baseline and follow-up (N ⫽ 40)

M

SD

M

Age Years of education

34.80 11.02 N

10.15 1.57 %

30.18 11.13 N

Gender (female) Any affective disorder Major depression current Major depression lifetime Dysthymia Alcohol dependency Alcohol abuse Any anxiety disorder Panic disorder with agoraphobia Social phobia Generalized anxiety disorder Posttraumatic stress disorder Somatization disorder Any eating disorder Anorexia nervosa Bulimia nervosa Any cluster A PD Any cluster B PD NPD Any cluster C PD Without psychotropic medication Antipsychotic Antidepressant Mood stabilizer

25

44.6

23

57.5

23

57.5

20 24 14 11 12

35.7 42.9 25.0 19.6 21.4

13 16 15 4 8

32.5 40.0 37.5 10.0 20.0

11 16 3 2 3

27.5 40.0 7.5 5.0 7.5

5 3 2 5 3

8.9 5.4 3.6 8.9 5.4

2 2 3 6 1

5.0 5.0 7.5 15.0 2.5

3 n.a. n.a. n.a. n.a.

7.5 n.a. n.a. n.a. n.a.

4 4 20 25 56 20 32 10 25 3

7.1 7.1 35.7 44.6 100.0 35.7 57.1 17.7 44.6 5.7

3 5 11 30 40 12 29 11 24 3

7.5 12.5 27.5 75.0 100.0 30.0 72.5 27.5 60.0 7.5

n.a. n.a. 12 30 19 8 11 5 19 3

n.a. n.a. 30.0 75.0 52.5 20.0 27.5 12.5 47.5 7.5

6.98 1.67 %

M

SD

31.70 11.13 N

6.99 1.67 %

Note. Any Cluster B PD only includes Histrionic and Antisocial Personality Disorder. Group 1 includes patients who participated at baseline, but were unreachable at follow-up; it shows comorbidities at baseline. Group 2 includes patients who participated at baseline and follow-up; it shows comorbidities at baseline. Group 3 includes patients who participated at baseline and follow-up; it shows comorbidities at follow-up.

patient gained over the course of 2 years is not represented in Column 3 of Table 3. The rank ordering in Column 3 shows that envy, arrogance, and need for admiration were the least remitted criteria. Grandiosity, fantasies of unlimited success, and exploitativeness were the criteria with the highest remission rates.

Stability of Dimensional Personality Traits Table 4 provides descriptive statistics for all four domains (marked in gray) and all subscales of the DAPP-BQ. The t test results reveal that patients with NPD reported significantly lower scores on nine DAPP-BQ subscales (see Table 4); however, effect sizes were small to medium (all Cohen’s d ⬍ .50), with the exception of insecure attachment (Cohen’s d ⫽ .66). In order to provide data regarding whether the changes on DAPP-BQ subscales were clinically meaningful, we calculated reliable change indices (RCI). The RCI is an individual level statistic that calculates whether there is reliable change in an individual’s score. We used the formula provided by Jacobson and Truax (1991). Thus, the RCI is equal to the individual’s

score before the intervention minus their score after the intervention then divided by the standard error of the difference of the test. If the RCI is 1.96 or greater, then the difference is statistically significant (1.96 equates to the 95% confidence interval). If the RCI is less than 1.96, then the difference is not significant. According to our data, one out of 40 patients (2.6%) improved on the narcissism subscale of the DAPP-BQ (see Table 4).

Discussion This study examined the temporal stability of the categorical diagnosis of NPD, the stability of each individual diagnostic criterion of NPD as well as the stability of a dimensional rating system for pathological personality over a 2-year period. Our results suggest a relatively moderate remission rate for the categorical diagnosis of NPD over the course of 2 years. Moreover, individual NPD criteria differed in their prevalence and stability. No clinically meaningful change in narcissistic traits emerged on a dimensional rating system for personality pathology (DAPP-BQ).

STABILITY OF NPD

Table 2 Descriptive Statistics of Direction of Change of Diagnostic Criteria of NPD According to DSM–IV–TR (N ⫽ 40) Criteria Envy Arrogance

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Need for admiration Lack of empathy Belief in uniqueness Entitlement Grandiosity Fantasies of unlimited success Exploitativeness

Direction of change

N

%

⫹1 0 ⫺1 ⫹1 0 ⫺1 ⫹1 0 ⫺1 ⫹1 0 ⫺1 ⫹1 0 ⫺1 ⫹1 0 ⫺1 ⫹1 0 ⫺1 ⫹1 0 ⫺1 ⫹1 0 ⫺1

4 23 11 5 26 7 2 24 12 5 26 7 8 23 7 5 23 10 8 19 11 5 17 16 4 17 17

10.5 60.5 28.9 13.2 68.4 18.4 5.3 63.2 31.6 13.2 68.4 18.4 21.1 50.5 18.4 13.2 60.5 26.3 21.1 50.0 28.9 13.2 44.7 42.1 10.5 44.7 44.7

Note. Direction of change: ⫹1 ⫽ gain of criterion (patients who did not meet the criterion at baseline, but met it at follow-up); 0 ⫽ neither loss nor gain of criterion (absence or presence of criterion remained stable), ⫺1 ⫽ loss of criterion (patients who met the criterion at baseline, but did not meet it at follow-up); N ⫽ number of individuals who gained, neither gained nor lost; or lost a criterion.

Stability of NPD as Diagnostic Category According to our results, the 2-year remission rate for NPD was 52.5%. This finding is in line with a previous study by Ronningstam et al. (1995) that provided the first empirical evidence for moderate 3-year stability of the DSM–III–R (APA, 1987) and

309

DSM–IV (APA, 1994) NPD diagnosis. The remission rate that we found here is comparable to the remission rates of other personality disorders (Grilo et al., 2004; Shea et al., 2002; Zimmerman, 1994). For instance, Grilo et al. (2004) reported remission rates of 61% for schizotypal PD, 56% for borderline PD, 50% for avoidant PD, and 60% for obsessive– compulsive PD within 2 years. If only the categorical diagnosis is taken into consideration, the conclusion from these data is an NPD remission rate of 50%. However, this rate also includes changes from five criteria to just below threshold for NPD; thus, individuals whose NPD diagnosis “remitted” still met several criteria, but failed to meet arbitrary categorical thresholds. At present, there is no empirical basis for selecting a cut-off score of five criteria for NPD. It is, therefore, reasonable to suspect that a majority of the NPD patients who “remitted” are still struggling with personality pathology symptoms after 2 years. This argument highlights the loss of information that occurs when interpreting characteristics of NPD categorically. Future studies should explore whether remission of the categorical diagnosis is associated with higher social and psychological functioning.

Temporal Stability of Dimensional Rating Systems Whereas previous studies utilizing longitudinal assessment of NPD have mainly relied upon scores from semistructured interviews, the current study additionally investigated the temporal stability of scores from a self-report questionnaire, the DAPP-BQ. The observation in the present study that dimensional ratings of NPD are highly stable is encouraging, as it provides stronger evidence for a more enduring pattern in patients with NPD than can be observed using the less stable categorical approach. Samuel et al. (2011) argue that multiple items assessing the nuances of a given construct may yield greater measurement precision and may perhaps be to some extent superior to diagnostic interviews (also see Sanislow et al., 2009). Moreover, diagnostic interviews assess behaviorally specific content. In contrast, the DAPP-BQ may assess broader and more general pathological personality traits and, as a consequence, appear to be more stable (Samuel et al., 2011). Additional research is clearly needed to better understand the temporal stability of dimensional and categorical diagnostic systems. Moreover, multiple follow-up assessments and shorter

Table 3 2-Year Follow-Up of Diagnostic Criteria of NPD According to DSM–IV–TR. Rank Order of Frequency of Prevalence, Change, Remission and Direction of Change Criteria present at baseline (most to least frequent)

Criteria present at 2 years (most to least frequent)

Criteria remitted at 2 years (least to most frequent)

Criteria

%

Criteria

%

Criteria

%

Need for admiration Fantasies of unlimited success Envy Exploitativeness Grandiosity Entitlement Arrogance Lack of empathy Belief in uniqueness

81.58 78.95 78.95 73.68 60.53 57.89 50.00 44.74 42.11

Envy Need for admiration Grandiosity Fantasies of unlimited success Belief in uniqueness Arrogance Entitlement Exploitativeness Lack of empathy

60.53 55.26 52.63 50.00 44.74 44.74 42.11 39.47 39.47

Envy Arrogance Need for admiration Lack of empathy Belief in uniqueness Entitlement Grandiosity Fantasies of unlimited success Exploitativeness

36.67 36.84 38.70 41.18 43.75 45.45 47.83 53.33 60.71

Note.

N ⫽ 40.

VATER ET AL.

310

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Table 4 2-Year Follow-Up of Dimensional Personality Pathology According to DSM–IV–TR

Identity problems Insecure attachment Emotional dysregulation Anxiousness Cognitive dysregulation Affective lability Stimulus seeking Rejection Self-harm Dissocial behavior Restricted expression Compulsivity Social avoidance Suspiciousness Conduct problems Submissiveness Narcissism Oppositionality Low affiliation Intimacy problems Callousness

NPD baseline

NPD 2-year follow-up

Correlation baseline and 2-year follow-up

M

SD

M

SD

r

t(37)

p

d

Percentage of RCI ⬎ 1.96

61.70 59.77 406.29 62.08 53.65 64.04 52.56 47.99 45.57 238.89 56.61 51.86 156.36 46.39 43.80 50.99 53.59 54.04 57.53 42.22 40.96

9.20 11.40 45.47 11.44 8.91 8.26 10.28 7.62 11.40 33.87 9.44 11.20 22.95 8.56 10.38 10.25 11.73 10.08 11.06 12.30 9.32

57.19 52.55 381.07 58.11 51.39 60.45 49.38 46.53 41.13 229.72 54.79 50.42 153.99 46.42 40.66 48.68 51.26 52.69 56.29 42.91 41.89

12.54 10.69 56.17 12.78 11.50 9.94 10.54 10.36 11.95 38.02 10.02 11.22 24.70 10.77 9.97 9.04 10.96 10.59 11.30 12.65 9.33

.677 .610 .698 .685 .710 .470 .789 .760 .721 .779 .748 .251 .747 .544 .598 .646 .680 .627 .767 .702 .686

3.00 4.56 3.82 2.53 1.71 2.34 2.89 1.34 3.13 2.33 1.62 .64 .86 ⫺.02 2.12 1.74 1.58 .93 1.00 ⫺.44 ⫺.78

.005 .000 .000 .016 .095 .025 .006 .189 .003 .025 .115 .523 .397 .982 .041 .091 .123 .358 .322 .662 .441

.42 .66 .50 .33 .22 .40 .31 .16 .39 .26 .19 .13 .10 .00 .31 .24 .21 .13 .11 ⫺.06 ⫺.100

15.8 13.2 10.5 10.5 10.5 7.9 7.9 7.9 5.3 5.3 5.3 5.3 5.3 5.3 2.6 2.6 2.6 2.6 2.6 2.6 2.6

T-test

Note. N ⫽ 40. M ⫽ mean score; SD ⫽ standard deviation; bold indicates scales that comprise the subscales below: emotional dysregulation ⫽ subscales affective lability, anxiousness, cognitive dysregulation, identity problems, insecure attachment, oppositionality, submissiveness; dissocial behavior ⫽ subscales callousness, conduct problems, narcissism, rejection, stimulus seeking; social avoidance ⫽ subscales intimacy problems, low affiliation, restricted expression; compulsivity ⫽ subscales suspiciousness, self-harm.

time intervals between assessments could be utilized as additional practical approaches to improve measurement.

Prevalence and Change of Each Diagnostic Criterion We found that the categorical diagnostic system utilized by the DSM often does not produce temporally stable diagnoses for NPD. This finding is not consistent with the conventional understanding of NPD as a stable pattern of symptoms. However, it is consistent with research suggesting that NPD patients experience both grandiosity and vulnerability, a combination that is associated with short-term and long-term shifts in affect, behavior, and ultimately, symptom expression (e.g., Levy, 2012; Morf & Rhodewalt, 2001; Pincus, 2011; Pincus & Lukowitzky, 2010; Roche, Pincus, Conroy, Hyde, & Ram, 2013; Ronningstam, 2009). However, it is important to note that it remains unclear whether grandiose and vulnerable narcissism are distinct subtypes of narcissism or whether they represent two sides of one coin (see Bosson et al., 2008; Campbell, Bosson, Goheen, Lakey, & Kernis, 2007; Miller et al., 2011). For instance, Ronningstam (2009) stated that patients with NPD fluctuate between assertive grandiosity and vulnerability. Similarly, Kernberg (2009) noted that a sense of specialness and grandiosity is disrupted by feelings of insecurity. According to the Dynamic Self-Regulatory Processing Model of Narcissism (Morf & Rhodewalt, 2001), narcissistic individuals attempt to maintain a grandiose (but unrealistically ideal) self-concept. However, an underlying (not necessarily conscious) insecurity elicits self-serving intrapersonal processes and interpersonal strategies. When external or internal triggers provoke vulnerability, narcis-

sistic patients may use grandiose self-enhancement in order to defend the depleted self. We argue that the regulatory function of self-enhancement cannot be explained in terms of vulnerability alone. NPD patients primarily engage in self-enhancement in order to gain admiration and attention; however, underlying vulnerability may elicit more extreme self-enhancing strategies. We further argue that NPD patients do not simply have limited selfregulatory abilities, but also that their self-regulation is strongly, and even sometimes, effectively influenced by a number of factors (e.g., high ideals, high perfectionism). Following this line of argumentation, the stability of individual diagnostic criteria in patients with NPD should be interpreted in light of their self-regulatory functions. This is the first study that analyzed the stability and remission of individual criteria in NPD; thus, the hypotheses generated from the following post hoc discussion of our results must be subjected to empirical inquiry in future studies. Based on our results, need for admiration (Criterion 5) was among the most prevalent diagnostic criteria at baseline and it also was the most stable. This result alone argues that self-enhancement is a strong component of narcissistic self-regulation (i.e., not just as a defense against vulnerability). In line with the self-regulatory conceptualization of narcissism, need for admiration may reflect a stable, internally based component of self-enhancement, in contrast to behavioral features that tend to fluctuate. Following this line of reasoning, NPD may be tied to an enduring need for admiration as a basic intrinsic motivational force.

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

STABILITY OF NPD

Lack of empathy (Criterion 7) appears to be stable, but less prevalent at baseline. We argue that the low prevalence rate of a lack of empathy might be due to its conceptualization in DSM– IV–TR. Recent research suggests that lack of empathy indeed is central to NPD (Ritter et al., 2011; Schulze et al., 2013): NPD patients have no significant impairment in cognitive empathy (they are able to identify others’ states of mind), but lower emotional empathy (i.e., they show lower affective responses to others’ states) in comparison with nonclinical individuals. The distinction between these two different facets of empathy is not captured in the current SCID-II interview for NPD. If the presence of both cognitive and emotional facets of empathy were assessed, the criterion might reach higher prevalence rates. Furthermore, variability in empathic ability should be evaluated with consideration of the motivational and self-regulatory context, given that individuals with NPD are hypothesized to access cognitive empathic ability under certain circumstances, but might be unable to do so in less self-serving or more threatening situations. Consistent with this reasoning, Ritter et al. (2011) provide initial evidence that NPD patients only exhibit a motivational deficit for cognitive empathy, whereas empathic capacity is unaffected. Envy (Criterion 8) appeared to be a prevalent and stable criterion. In line with a recent study (Krizan & Johar, 2012), individuals with narcissistic traits who are in a vulnerable state may fail to meet their own standards, and as a consequence, experience envy. In contrast, individuals with grandiose narcissism may project feelings of envy onto others in order to buttress their own feelings of superiority. High prevalence rates of envy should be discussed in light of narcissistic patients’ motivation to selfenhance and a reluctance to self-disclose. As patients may hesitate to openly report feelings of envy (e.g., in order to avoid feelings of shame), feelings of envy may be a less useful descriptor relative to other diagnostic criteria. Moreover, one might speculate that feelings of envy represent a proximal criterion that is driven by more distal diagnostic criteria. For instance, envy might be particularly pronounced when the need for admiration is not satisfied. Other criteria were less stable, but highly prevalent at baseline, such as fantasies of unlimited success (Criterion 2). We assume that fantasizing about success might function as a coping mechanism to protect against temporary vulnerability experienced during hospitalization. If a person is unsatisfied with the current reality, fantasies may be a strategy for increasing positive affect in the short term. Indeed, research provides evidence that delusions about the self and unrealistic fantasies are correlated with subjective well-being and buffer against stress (Raskin & Novacek, 1991). Our data indicate that exploitativeness (Criterion 6) was the criterion with the highest remission rate. We contend that exploitativeness is a complex criterion as it can refer to concrete material (e.g., financial goods), as well as emotional exploitation and exploitation of social resources (e.g., status, prestige, affiliation). Emotional exploitativeness is supposedly more typical for NPD, although material exploitativeness may be related to psychopathy. Following this line of reasoning, exploitativeness in patients with NPD has to be understood within the framework of a selfregulatory model of NPD. For example, emotional exploitativeness may be used as a strategy to ensure constant attention from others in order to sustain a grandiose sense of self. Therefore, the wording “exploitative” may be less useful to describe this diagnostic feature. In line with theories regarding the self-regulatory

311

function of narcissistic traits (Morf & Rhodewalt, 2001), the diagnostic formulation “need for personal gain” may more adequately describe the self-sustaining core of emotionally exploitative behavior of patients with NPD. Belief in uniqueness (Criterion 3) was the least prevalent criterion at baseline in our sample, but became more prevalent after 2 years. One may assume that patients in our study suffered from a crisis and they initially presented in a vulnerable state that temporarily suppressed beliefs in uniqueness. In contrast to belief in uniqueness (Criterion 3), arrogance (Criterion 9), grandiosity (Criterion 1), and entitlement (Criterion 5) may more reliably capture grandiose content in NPD, as those criteria were prevalent at baseline and stable over the course of 2 years. However, we argue that those criteria all represent self-enhancing behaviors and attitudes that may fluctuate in the short term depending upon circumstances and the presence (or absence) of self-esteem supporting conditions. Moreover, all of these features have adaptive aspects in certain conditions, cultures, age groups, or developmental phases. Some of these features are extremely important for normal identity development (e.g., Hill & Roberts, 2012). If grandiose features are threatened by external information, decreased self-esteem may result and compensatory efforts may be activated. It is yet to be determined where normal self-regulatory efforts end and pathological self-enhancement begins. Further research is also needed to determine the role that self-agency, competence, and external support play in narcissistic self-regulation.

Limitations A strength of this study is the recruitment of a relatively large sample size of individuals diagnosed with NPD at baseline. However, this study has several limitations. First, we cannot use our data to draw any conclusions regarding predictors of change. Second, NPD exhibited high comorbidity rates in our study. Third, this study includes a sample of individuals with NPD that presents with a high amount of vulnerability (e.g., high levels of depression). Although our sample represents a good approximation of the population of psychiatric inpatients with NPD, a study of outpatients with NPD might yield different results. Fourth, we focused on long-term changes in NPD. As already mentioned above, narcissistic symptomatology may also fluctuate in the short term. Thus, future studies should analyze not only long-term, but also short-term fluctuations of affective, cognitive, and behavioral patterns.

References American Psychiatric Association. (1987). Diagnostic and Statistical Manual of Mental Disorders, 3rd ed., revision. Washington, DC: American Psychiatric Association. American Psychiatric Association. (1994). Diagnostic and Statistical Manual of Mental Disorders, 4th ed. Washington, DC: American Psychiatric Association. American Psychiatric Association. (2000). Diagnostic and Statistical Manual of Mental Disorders, 4th ed., text revision. Washington, DC: American Psychiatric Association. Ball, S. A., Rounsaville, B. J., Tennen, H., & Kranzler, H. R. (2001). Reliability of personality disorder symptoms and personality traits in substance-dependent inpatients. Journal of Abnormal Psychology, 110, 341–352. doi:10.1037/0021-843X.110.2.341

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

312

VATER ET AL.

Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J. (1961). An inventory for measuring depression. Archives of General Psychiatry, 4, 561–571. doi:10.1001/archpsyc.1961.01710120031004 Bosson, J. K., Lakey, C. E., Campbell, W. K., Zeigler-Hill, V., Jordan, C. H., & Kernis, M. H. (2008). Untangling the Links between Narcissism and Self-esteem: A Theoretical and Empirical Review. Social and Personality Psychology Compass, 2, 1415–1439. doi:10.1111/j.17519004.2008.00089.x Campbell, W. K., Bosson, J. K., Goheen, T. W., Lakey, C. E., & Kernis, M. H. (2007). Do narcissists dislike themselves “deep down inside?” Psychological Science, 18, 227–229. doi:10.1111/j.1467-9280.2007 .01880.x Clark, L. A., Simms, L. J., Wu, K. D., & Casillas, A. (in press). Manual for the schedule for nonadaptive and adaptive personality (SNAP-2). Minneapolis, MN: University of Minnesota Press. Shea, M. T., Stout, R., Gunderson, J., Morey, L. C., McGlashan, T., . . . Keller, M. B. (2002). Short-term diagnostic stability of schizotypal, borderline, avoidant, and obsessive-compulsive personality disorders. American Journal of Psychiatry, 159, 2036 –2041. First, M. B., Gibbon, M., Spitzer, R. L., Williams, J. B. W., & Benjamin, L. (1997). Structured Clinical Interview for DSM–IV Personality Disorders (SCID-II). Washington, DC: American Psychiatric Press. First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (1996). Structured Clinical Interview for DSM–IV Axis I Disorders Research Version (SCID-I). New York, NY: New York State Psychiatric Institute, Biometrics Research. Fydrich, T., Renneberg, B., Schmitz, B., & Wittchen, H-U. (1997). SKIDII-Strukturiertes Klinisches Interview für DSM–IV Achse II: Persönlichkeitsstörungen [SCID-II, Structured clinical interview for DSM–IV Axis II: Personality disorders]. Göttingen, Germany: Hogrefe. Grilo, C. M., Shea, M. T., Sanislow, C. A., Skodol, A. E., Stout, R. L., Gunderson, J. G., . . . McGlashan, T. H. (2004). Two-year stability and change in schizotypal, borderline, avoidant, and obsessive-compulsive personality disorders. Journal of Consulting and Clinical Psychology, 72, 767–775. doi:10.1037/0022-006X.72.5.767 Gunderson, J. G., Ronningstam, E., & Bodkin, A. (1990). The diagnostic interview for narcissistic patients. Archives of General Psychiatry, 47, 676 – 680. doi:10.1001/archpsyc.1990.01810190076011 Hautzinger, M., Bailer, M., Worall, H., & Keller, F. (1995). BeckDepressions-Inventar (BDI). Testhandbuch [Beck Depression Inventory Manual]. Bern, Germany: Huber. Hill, P. L., & Roberts, B. W. (2012). Narcissism, well-being, and observer rated Personality across the lifespan. Social Psychological and Personality Science, 3, 216 –223. doi:10.1177/1948550611415867 Hopwood, C. J., Morey, L. C., Donnellan, M. B., Samuel, D. B., Grilo, C. M., McGlashan, T. H., . . . Skodol, A. E. (2013). Ten year rank order stability of personality traits and disorders in a clinical sample. Journal of Personality, 81, 335–344. Jacobson, N. S., & Truax, P. (1991). Clinical significance: A statistical approach to defining meaningful change in psychotherapy-research. Journal of Consulting and Clinical Psychology, 59, 12–19. doi:10.1037/ 0022-006X.59.1.12 Kernberg, O. F. (1975). Borderline conditions and pathological narcissism. New York, NY: Jason Aronson. Kernberg, O. F. (2009). Narcissistic personality disorder: Pt. 1. Psychiatric Annals, 39, 105–107. doi:10.3928/00485713-20090301-04 Krizan, Z., & Johar, O. (2012). Envy divides the two faces of narcissism. Journal of Personality, 80, 1415–1451. doi:10.1111/j.1467-6494.2012 .00767.x Lenzenweger, M. F. (1999). Stability and change in personality disorder features. Archives of General Psychiatry, 56, 1009 –1015. doi:10.1001/ archpsyc.56.11.1009 Lenzenweger, M. F., Johnson, M. D., & Willett, J. B. (2004). Individual growth curve analysis illustrates stability and change in personality disorder

features: The longitudinal study of personality disorders. Archives of General Psychiatry, 61, 1015–1024. doi:10.1001/archpsyc.61.10.1015 Levy, K. N. (2012). Subtypes, dimensions, levels, and mental states in narcissism and narcissistic personality disorder. Journal of Clinical Psychology, 68, 886 – 897. Livesley, W. J., & Jackson, D. N. (2009). Manual for the dimensional assessment of personality pathology. Port Huron, MI: Sigma Press. McGlashan, T. H., Grilo, C. M., Ralevski, E., Morey, L. C., Gunderson, J. G., Skodol, A. E., . . . Pagano, M. E. (2005). Two-year prevalence and stability of individual DSM–IV criteria for schizotypal, borderline, avoidant, and obsessive-compulsive personality disorders: Toward a hybrid model of axis II disorders. The American Journal of Psychiatry, 162, 883– 889. doi:10.1176/appi.ajp.162.5.883 Miller, J. D., Hoffman, B. J., Gaughan, E. T., Gentile, B., Maples, J., & Keith Campbell, W. (2011). Grandiose and vulnerable narcissism: A nomological network analysis. Journal of Personality, 79, 1013–1042. doi:10.1111/j.1467-6494.2010.00711.x Morf, C. C., & Rhodewalt, F. (2001). Unraveling the paradoxes of narcissism: A dynamic self-regulatory processing model. Psychological Inquiry, 12, 177–196. doi:10.1207/S15327965PLI1204_1 Pincus, A. L. (2011). Some comments on nomology, diagnostic process, and narcissistic personality disorder in the DSM-5 proposal for personality and personality disorders. Personality Disorders: Theory, Research and Treatment, 2, 41–53. Pincus, A. L., & Lukowitzky, M. R. (2010). Pathological narcissism and narcissistic personality disorder. Annual Review of Clinical Psychology, 6, 421– 446. doi:10.1146/annurev.clinpsy.121208.131215 Pukrop, R., Gentil, I., Steinbring, I., & Steinmeyer, E. (2001). Factorial structure of the German version of the dimensional assessment of personality pathology— basic questionnaire. Journal of Personality Disorders, 15, 450 – 456. doi:10.1521/pedi.15.5.450.19195 Pukrop, R., Steinbring, I., Gentil, I., Schulte, C., Larstone, R., & Livesley, J. W. (2009). Clinical validity of the Dimensional Assessment of Personality Pathology (DAPP) for psychiatric patients with and without a personality disorder diagnosis. Journal of Personality Disorders, 23, 572–586. doi:10.1521/pedi.2009.23.6.572 Raskin, R., & Novacek, J. (1991). Narcissism and the use of fantasies. Journal of Clinical Psychology, 47, 490 – 499. doi:10.1002/10974679(199107)47:4⬍490::AID-JCLP2270470404⬎3.0.CO;2-J Ritter, K., Dziobek, I., Prei␤ler, S., Rüter, A., Vater, A., Fydrich, T., . . . Roepke, S. (2011). Lack of empathy in patients with narcissistic personality disorder. Psychiatry Research, 187, 241–247. doi:10.1016/j .psychres.2010.09.013 Roche, M. J., Pincus, A. L., Conroy, D. E., Hyde, A. L., & Ram, N. (2013). Pathological narcissism and interpersonal behavior in daily life. Personality disorders: Theory, research, and treatment. Journal of Personality Disorders, 27, 270 –295. Ronningstam, E. F. (2009). Narcissistic personality disorder: Facing DSM-V. Psychiatric Annals, 39, 111–121. doi:10.3928/0048571320090301-09 Ronningstam, E. F., Gunderson, J., & Lyons, M. (1995). Changes in pathological narcissism. The American Journal of Psychiatry, 152, 253– 257. Samuel, D. B., Hopwood, C. J., Ansell, E. B., Morey, L. C., Sanislow, C. A., Yen, S., . . . Grilo, C. M. (2011). Comparing the temporal stability of self-report and interview assessed personality disorder. Journal of Abnormal Psychology, 120, 670 – 680. doi:10.1037/a0022647 Sanislow, C. A., Little, T. D., Ansell, E. B., Grilo, C. M., Daversa, M., Markowitz, J. C., . . . McGlashan, T. H. (2009). Ten-year stability and latent structure of the DSM–IV schizotypal, borderline, avoidant, and obsessive-compulsive personality disorders. Journal of Abnormal Psychology, 118, 507–519. doi:10.1037/a0016478 Schulze, L., Dziobek, I., Vater, A., Heekeren, H. R., Bajbouj, M., Renneberg, B., . . . Roepke, S. (2013). Gray matter abnormalities in patients

STABILITY OF NPD

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

with narcissistic personality disorder. Journal of Psychiatric Research. Advance online publication. Shea, M. T., Stout, R., Gunderson, J., Morey, L. C., Grilo, C. M., McGlashan, T., . . . Fydrich, T. (1997). Strukturiertes Klinisches Interview für DSM–IV–TR - Achse I [Structured Clinical Interview for DSM–IV–TR - Axis I]. Göttingen, Germany: Hogrefe.

313

Zanarini, M. C., Frankenburg, F. R., Sickel, A. E., & Yong, L. (1996). The diagnostic interview for DSM–IV personality disorders (DIPD-IV). Belmont, MA: McLean Hospital. Zimmerman, M. (1994). Diagnosing personality disorders: A review of issues and research methods. Archives of General Psychiatry, 51, 225– 245. doi:10.1001/archpsyc.1994.03950030061006