Development and Preliminary Validation of a Narcissism Assessment Scale Based on the Hierarchical Taxonomy of Psychopathology (HiTOP) Model

Abstract

Introduction

A sense of entitlement, thoughts of deserving more than others, and belief of having superior abilities compared to other people characterizes narcissistic personality disorder (NPD). Encompassing this personality disorder and other mental conditions, the Hierarchical Taxonomy of Psychopathology (HiTOP) is an evidence-based, dimensional model covering not only clinical symptoms but also pathological traits.

Material and methods

Our study aimed to develop a self-report scale, the IDCP-NPD, for screening pathological traits of NPD from the perspective of the HiTOP facets of the Personality Inventory for DSM-5 (PID-5), factors of the Computerized Adaptive Assessment of Personality Disorder-Static Form (CAT-PD-SF), scales of the Five-Factor Narcissism Inventory (FFNI), and the Hypersensitive Narcissism Scale (HSNS).

Results

The IDCP-NPD comprised 65 items in two factors: Antisocial Grandiosity and Interpersonal Dominance. Internal structure reliability was good (>.80). The factors showed associations with the expected external measures, and the groups based on the scores in the NPD external measures showed big to huge differences.

Conclusions

Our findings suggested the IDCP-NPD is a helpful measure to screen the NPD traits in the clinical context. Additionally, the structure observed for the IDCP-STPD confirms the spectrum level of the HiTOP.

Resumen

Introducción

Un sentimiento de derecho, pensamientos de merecer más que los demás y la creencia de tener habilidades superiores en comparación con otras personas caracterizan el trastorno narcisista de la personalidad (NPD). La taxonomía jerárquica de la psicopatología (HiTOP), que abarca este trastorno de la personalidad y otras afecciones mentales, es un modelo dimensional basado en la evidencia que cubre los rasgos patológicos en su rango inferior.

Material y métodos

Nuestro estudio tuvo como objetivo desarrollar una escala de autoinforme, el IDCP-NPD, para detectar rasgos patológicos de NPD desde la perspectiva del HiTOP, de facetas del Inventario de Personalidad para el DSM-5 (PID-5), de factores de la Evaluación Adaptativa Computarizada del Trastorno de Personalidad Forma Estática (CAT-PD-SF), de escalas del Inventario de Narcisismo de Cinco Factores (FFNI) y de la Escala de Narcisismo Hipersensible (HSNS).

Resultados

El IDCP-NPD estuvo compuesto por 65 ítems distribuidos en dos factores: grandiosidad antisocial y dominancia interpersonal. La confiabilidad de la estructura interna fue buena (>0.80). Los factores mostraron asociaciones con las medidas externas esperadas, y los grupos basados en las puntuaciones de las medidas externas del NPD mostraron diferencias entre grandes y enormes.

Conclusiones

Nuestros hallazgos sugirieron que el IDCP-NPD es una medida útil en el contexto clínico para detectar los rasgos de NPD. Además, la estructura observada para el IDCP-STPD confirma el nivel espectral del HiTOP.

Introduction

A heightened sense of entitlement often describes narcissistic personality disorder (NPD), as well as thoughts of deserving more than others and having superior abilities compared to other people. However, these beliefs are rarely grounded in reality.¹ Many researchers advocate that there are two subtypes of NDP: the overt type (grandiosity, aggression, and dominance) and the covert type (vulnerability, feelings of inadequacy, incompetence, and negative affect), although there is debate on this division.^2–4 Regardless of how narcissistic traits are manifested, the harm caused by NPD is well-documented, including psychological distress, functional impairment, and interpersonal conflicts.^5–7

According to Section II of the DSM-5,⁵ among the main aspects of diagnosing NPD are a pervasive pattern of grandiosity, a constant need for admiration, a lack of empathy, and fantasies of unlimited success, power, brilliance, or beauty. Traditional systems follow these guidelines that consider a categorical diagnostic form and have been criticized, focusing excessively on grandiose aspects and not paying enough attention to the vulnerable traits of NPD.⁸ Recognizing and addressing the vulnerable aspect of NPD is crucial for a comprehensive understanding of the disorder, as it sheds light on the underlying emotional fragility, fear of rejection, and potential for significant distress that coexists with the outward manifestations of grandiosity. Besides, understanding NPD through the lens of the categorical model may oversimplify the complexity of personality, lack sufficient reliability and validity, and may lead to arbitrary diagnostic thresholds, hindering a more nuanced understanding of personality pathology's diverse and dimensional nature.

Alternatively, dimensional models propose an understanding and continuous assessment of psychopathological manifestations. Haslam et al.⁹ conducted a meta-analysis of taxometric studies, observing most personality and psychopathology constructs are better explained by latent dimensions rather than categories. A contemporary example of a dimensional model is the Hierarchical Taxonomy of Psychopathology (HiTOP), set to improve the measures through empirical research and assessment of the symptoms, combining them into homogeneous traits and grouping them into psychopathology spectra.¹⁰ The HiTOP proposes understanding mental disorders from six hierarchical levels: super spectra, spectra, subfactors, syndromes/disorders, components, and the lowest level comprising specific signs, symptoms, and features of psychopathology.^11,12 The empirical focus strongly supports a continuum between adaptive and maladaptive functioning, a common impairment dimension, and specific impairment expressions that require clinical attention.¹³ This study focuses on narcissistic personality disorder (NPD) under the HiTOP.

The HiTOP model allocates the traditional understanding of narcissistic personality disorder (NPD) as part of what we measure on the antagonistic externalizing spectrum,¹⁰ which has been consistently seen when approaching NPD from the HiTOP model.^14,15 Grandiosity, rudeness, egocentricity, attention seeking, callousness, deceitfulness, manipulativeness, and dominance are the lower-level pathological traits of this spectrum related to the NPD's pathological pattern. A recent meta-analysis confirmed NPD fits the antagonistic externalizing spectrum.¹⁶

While the Hierarchical Taxonomy of Psychopathology (HiTOP) has accumulated empirical evidence supporting its scientific and clinical utility,^11,16 there remains a paucity of assessment scales designed to operationalize this model effectively. Notably, measures that evaluate pathological traits may be well-suited to appraise the lower echelons of the HiTOP hierarchy, as outlined in the recommended scales proposed by Kotov et al.¹⁰ We can also identify investigations on prominent self-reports, built under different taxonomic aspects, efficiently applicable to the HiTOP model.^15,17 Although several NPD measures are available (e.g., Pathological Narcissism Inventory (PNI¹⁸), Five-Factor Narcissism Inventory (FFNI¹⁹)), we could not find specific measures of NPD traits implemented to incorporate the HiTOP.

Drawing on the postulation of Kotov et al.¹⁰ that measures appraising pathological traits may be suitable for evaluating the lower levels of the HiTOP hierarchy, our investigation centres on the employment of the Dimensional Clinical Personality Inventory (IDCP-2²⁰) as a promising self-report scale to operationalize the HiTOP. The IDCP-2 assesses pathological traits, comprising 210 items that assess 47 pathological personality traits grouped into 12 dimensions. Recent studies have suggested the suitability of IDCP-2 to the HiTOP model,²¹ as well as its usefulness in assessing specific disorders based on this model, such as the Avoidant PD²² and the schizotypal PD.²³ Previous empirical evidence suggests the Attention Seeking, Seduction and manipulation, Need for recognition, Superiority, Dominance, and Indifference IDCP-2 factors as relevant for NPD traits assessment.^24,25

In light of the emergence of the HiTOP model as a viable base for developing screening and diagnostic tools, there is a pressing need for scales that align with the HiTOP's underlying assumptions.^11,25 These scales must effectively capture NPD traits’ clinical significance and associated impairments.^6,7 To this end, our study aimed to construct and validate a dedicated version of the IDCP-2 that focuses specifically on NPD traits, thereby adhering to the HiTOP model. Our study thus employs the HiTOP model as the conceptual framework for comprehending and operationalizing NPD.

Methods

We divided the method into two stages, considering the aim of this study. The first (Stage I) referred to the procedures for the revision of the IDCP-2 based on the HiTOP, focusing on the evaluation of typical NPD traits; the second (Stage II) aimed to verify the psychometric properties of the new version of the IDCP-2 for NPD traits, the IDCP-NPD, in a community sample.

Stage I

Firstly, we verified the spectra of the HiTOP that encompass the NPD, focusing on the selection and definition of the traits relevant to the disorder. We deliberately attempted to maintain as many traits as possible, aiming for the representativeness of typical NPD traits. Based on this, we searched for definitions in the literature, mainly according to the measurement tests mentioned in Table 1 of Kotov et al.,¹⁰ such as the Personality Inventory for DSM-5 (PID-5²⁶), Five-Factor Narcissism Inventory (FFNI¹⁹), Computerized Adaptive Test of Personality Disorder-Static Form (CAT-PD-SF²⁷), and Hypersensitive Narcissism Scale (HSNS²⁸).

Table 1.

Selected traits, IDCP-2 factors, developed measures, and external measures based on the NPD spectrum from HiTOP.

Spectrum	HiTOP traits	Selected traits	IDCP-2 factors and developed measures	Respective external measures (hypothesized positive correlation)
Antagonistic Externalizing	Attention Seeking	Attention Seeking	Attention Seeking (4 items)	Attention Seeking (PID-5)
	Callousness	Callousness	Callousness (4 items)	Callousness (PID-5)
	Deceitfulness	Deceitfulness	Deceitfulness (6 items)	Deceitfulness (PID-5)
	Grandiosity	Grandiosity	Need for recognition (4 items); Superiority (5 items)	Grandiosity (PID-5)
	Manipulativeness	Manipulativeness	Seduction and manipulation (3 items);	Manipulativeness (PID-5)
	Rudeness	Rudeness	Antagonism (7 items); Rudeness (10 items)	Rudeness (CAT-PD-SF)^37
	Egocentricity	Egocentricity	Egocentricity (13 items)	Exploitativeness (FFNI)^19;Lack of empathy (FFNI)^19
	Dominance	Dominance	Dominance (4 items); Control (3 items)	Authoritativeness (FFNI)^19
	Flirtatiousness	–	–	–
	(low) Timorousness	–	–	–
Internalizing	Vulnerability	Vulnerability	Vulnerable Grandiosity (10 items)	Vulnerability (HSNS)^28

Note. New factors are bolded; the number of items developed to the new factor, and the final number are in brackets.

As the next step, we selected the IDCP-2 factors representing the target traits, starting from the selected HiTOP traits related to NPD and its definitions. This procedure identified gaps in the IDCP-2 assessment of the NDP traits based on HiTOP. In other words, we observed that specific aspects considered relevant for assessing NPD traits (e.g., grandiose vulnerability) were not covered by IDCP-2 factors. The three authors of this study operationalized the traits not covered by the IDCP-2. In other words, we developed items to represent all the unrepresented or not well-represented traits from the HiTOP.

The development of the self-report items was conducted independently by the authors. The authors, by consensus, selected the best items from the conceptual point of view based on content and semantic issues. Specifically, the items’ clarity, consistency, and redundancy were verified, the latter regarding non-repetition of the characteristics already evaluated by the IDCP-2. We reached the first version of the IDCP-NPD, which we used in stage II of the study.

Stage II – verification of the psychometric properties of the IDCP-NPD

Participants

Data collection was by convenience sampling, a procedure primarily used with community samples in studies on pathological traits.²⁹ The sample was composed of 449 people from the community. The participants aged between 18 and 70 years (M = 27.82, SD = 9.08), and were mainly female (70.6%) and Caucasian (47%). Most of the participants were single (54.1%), reported completed high school (39.2%), or had incomplete higher education (24.5%). Regarding their psychiatric/psychological treatment history, 15.4% reported attending psychotherapy, and 12.9% to psychiatric treatment. Some participants reported past suicidal ideation (47%), and suicide attempts (38%).

The expectation for NPD occurrence in community samples ranges from 0.5% to 6.2%.⁵ In an epidemiologic study conducted with a community representative sample in the São Paulo Metropolitan Area, Brazil, 2.7% of cases of PD from Cluster B were observed.³⁰ According to this, we expected people with elevation in typical traits of the NPD to be found in our sample. Moreover, a sensitivity analysis using G*Power (Faul et al., 2007) indicated that with n = 449, we have power = .93 to detect effect sizes ≥.16 in Pearson's correlations analysis (p = .05).

Measures

Dimensional Clinical Personality Inventory – narcissistic personality disorder version (IDCP-NPD)

The IDCP-NPD is a self-report scale originated from de Dimensional Clinical Personality Inventory 2 (IDCP-2³¹) based on the traits of the NPD according to the HiTOP.^10,11 The IDCP-2 consists of a self-report measure developed for assessing pathological personality traits, comprising 210 items on a Likert 4-point scale, from “has nothing to do with me” to “everything to do with me”. The IDCP-2 covers 47 factors on 12 dimensions (Dependency, Aggressiveness, Mood Instability, Eccentricity, Attention Seeking, Distrust, Grandiosity, Isolation, Criticism Avoidance, Self-sacrifice, Conscientiousness, and Inconsequence). For this study, according to the traits assessed by each IDCP-2 factor, we selected the following factors: Attention Seeking (α = 0.78, ω = 0.87), Seduction and manipulation (α = 0.83, ω = 0.83),²⁴ Antagonism (α = 0.86, ω = 0.89),³² Indifference (α = 0.77, ω = 0.81),³³ Need for recognition (α = 0.87, ω = 0.89), Superiority (α = 0.87, ω = 0.91), Callousness (α = 0.88, ω = 0.92), Dominance (α = 0.81, ω = 0.85),³⁴ and Control (α = 0.83, ω = 0.83).³⁵ We also administered the 23 items developed and selected in Stage 1 of this study. These items represent the traits of Rudeness, Egocentricity, and Vulnerable Grandiosity.

Personality Inventory Disorder for DSM-5 (PID-5)

The PID-5²⁶ is a self-report inventory for assessing 25 facets representing pathological traits according to the Alternative Model for Personality Disorders (AMPD) of the DSM-5.⁵ It contains 220 items on a Likert 4-point scale, from “false or often false” to “true or often true”. The 25 facets are grouped into five dimensions (i.e., Negative Affect, Detachment, Antagonism, Disinhibition, and Psychoticism). For this study, we selected the following facets: Attention Seeking (α = 0.86, ω = 0.90), Callousness (α = 0.91, ω = 0.93), Deceitfulness (α = 0.89, ω = 0.92), Grandiosity (α = 0.85, ω = 0.87), and Manipulativeness (α = 0.89, ω = 0.90). Satisfactory psychometric properties for PID-5 were previously published.^14,36

Comprehensive Assessment of Traits Relevant to Personality Disorder-Static Form (CAT-PD-SF)

The CAT-PD-SF³⁷ contains 216 items assessing 33 maladaptive personality traits organized within five domains: Negative Emotionality, Detachment, Antagonism, Disconstraint, and Psychoticism. Participants rate how well the statements describe themselves using a 5-point Likert-type scale ranging from 1 for very untrue of me to 5 for very true of me. For this study, we selected the Rudeness factor (α = 0.88, ω = 0.92). Satisfactory psychometric properties for the CAT-PD-SF were previously published.^37,38

Five-Factor Narcissism Inventory (FFNI)

The FFNI¹⁹ is a 148-item self-report inventory of 15 traits related to vulnerable and grandiose narcissism. Each scale was created to assess a more maladaptive variant found to be an essential component of either or both narcissism dimensions. Participants rate how well the statements describe themselves using a 5-point Likert-type scale ranging from 1 for very untrue of me to 5 for very true of me. For this study, we selected the following scales: Exploitativeness (α = 0.91, ω = 0.94), Lack of empathy (α = 0.88, ω = 0.91), and Authoritativeness (α = 0.86, ω = 0.90). Satisfactory psychometric properties for the FFNI were previously published.³

Hypersensitive Narcissism Scale (HSNS)

The Hypersensitive Narcissism Scale (HSNS)²⁸ is a 10-item self-report measure that reflects hypersensitivity, vulnerability, and entitlement. Participants rate how well the statements describe themselves using a 5-point Likert-type scale ranging from 1 for very untrue of me to 5 for very true of me (α = 0.79, ω = 0.83). Satisfactory psychometric properties for the HSNS were previously published.^28,39

Procedure

The procedures of this study complied with the provisions of the Declaration of Helsinki regarding research on Human participants (World Medical Association [WMA]).⁴⁰ All participants digitally consented to participate in the study. The online survey conformed to the recommended standards for conducting and reporting web-based surveys, the Checklist for Reporting Results of Internet E-surveys (CHERRIES).⁴¹ Open data collection was performed using Google Forms resources and the link available to respondents from January to March 2021. We tested the functionality before making the survey available to participants to ensure it works best. We shared the research link on the social media website Facebook and WhatsApp, inviting individuals to participate and relying on the snowball strategy to reach a more substantial number of participants. No incentives were given to the participants.

Data analysis

To check the number of factors to be retained, we ran a parallel analysis with a polychoric correlation in MPlus 7.11. This procedure is known in the literature as a statistical procedure based on the Monte Carlo simulation through a randomized construction of a hypothetical set of variable matrices with the same dimensionality as the observable variable set.⁴² Then, we conducted an exploratory factor analysis to investigate the factorial solution of the IDCP-NPD with maximum likelihood extraction and geomin rotation. Focusing on external validity, we examined correlations between the factors of the IDCP-NPD and external measures. Besides, we compared the group's means in the IDCP-NPD factors. The groups were calculated based on the sum of the external measures, standardized in z distribution. The so-called healthy group (n = 71) included participants with one or more standard deviations below the mean, and the so-called pathological group (n = 70) included one or more standard deviations above the mean.

Results

Stage I

As a result of stage I, Table 1 presents the antagonistic externalizing spectrum, in which NPD is located, as well as the pathological traits covered by this spectrum. We also indicate the traits we selected as pertinent to NPD, listing the IDCP-2 factors whose definitions are consistent with these features. Additionally, we show the external measures we administered in this study per trait.

We selected eight pathological traits of the HiTOP antagonistic externalizing spectrum as relevant to NPD. To conceptually cover the manifestation of these traits, we selected nine factors from the IDCP-2 and created two new factors. Moreover, we chose to keep vulnerability as a crucial trait of NPD because the prominent literature focused on narcissistic vulnerability,^43–45 so we produced a factor to cover the vulnerability linked to grandiosity.

Stage II

The parallel analysis suggested two factors presenting expressive non-random values. Best fit indexes for this two-factor solution were RMSEA = 0.15, CFI = 0.90, TLI = 0.78, SRMR = 0.06, χ² = 140.90, AIC = 5725.28, BIC = 5850.60. Table 2 presents factor loadings, coefficients of internal consistency (Cronbach's alpha and McDonald's omega), and the correlation between factors.

Table 2.

Factor loadings internal consistency and correlation for the IDCP-NPD factors.

IDCP-NPD	Antisocial Grandiosity	Interpersonal Dominance	α	Ω
Callousness	0.80	−0.21	0.76	0.78
Superiority	0.35	0.35	0.86	0.87
Dominance	0.25	0.68	0.85	0.85
Vulnerability Grandiosity	0.47	0.00	0.82	0.83
Rudeness	0.60	−0.01	0.78	0.79
Egocentricity	0.79	0.10	0.81	0.83
Seduction and manipulation	−0.003	0.96	0.84	0.84
Deceitfulness	0.61	0.30	0.89	0.89
αaverage	0.81	0.82	0.86
ωaverage	0.81	0.83		0.87
raverage		0.74	0.76	0.89

Note. Values in bold indicate the allocation of variables in the two factors.

Factor 1 was named Antisocial Grandiosity and concentrated higher factor loadings from five subscales of IDCP-NPD, ranging from 0.47 to 0.80. We called the second factor after Interpersonal Dominance, and factor loadings ranged from 0.35 to 0.96. Both alpha and omega estimates for Antisocial Grandiosity were 0.81, whereas the indexes for Interpersonal Dominance were 0.82 and 0.83, respectively. Notably, the Superiority factor showed a factor loading of 0.35 for both Antisocial Grandiosity and Interpersonal Dominance. However, we considered it in the second factor (i.e., Interpersonal Dominance) due to the items’ content and fit to the model. Table 3 presents the correlations with the IDCP-NPD factors and external measures.

Table 3.

Correlations between IDCP-2 factors for narcissistic personality disorder (IDCP-NPD) (lines) and external measures (columns).

	Attention Seeking	Callousness	Grandiosity	Deceitfulness	Rudeness	Authoritativeness	Exploitativeness	Lack of empathy	Vulnerability
Callousness	0.03	0.65**	0.39**	0.48**	0.36**	0.22**	0.49**	0.68**	0.40*
Superiority	0.36**	0.37**	0.64**	0.34**	0.29**	0.42**	0.37**	0.26**	0.39**
Dominance	0.39**	0.49**	0.51**	0.57**	0.27**	0.68**	0.48**	0.29**	0.26**
Vulnerability Grandiosity	0.48**	0.26**	0.36**	0.34**	0.41**	0.09	0.21**	0.20**	0.65**
Rudeness	0.39**	0.44**	0.42**	0.33**	0.48**	0.24**	0.29**	0.32**	0.54**
Egocentricity	0.30**	0.79**	0.61**	0.68**	0.41**	0.48**	0.73**	0.64**	0.45**
Seduction and manipulation	0.45**	0.44**	0.46**	0.56**	0.25**	0.53**	0.49**	0.21**	0.26**
Deceitfulness	0.38**	0.65**	0.50**	0.82**	0.41**	0.51**	0.73**	0.50**	0.44**
Antisocial Grandiosity	0.42**	0.74**	0.60**	0.71**	0.54**	0.41**	0.65**	0.62**	0.66**
Interpersonal Dominance	.47**	.51**	.62**	.57**	.32**	.63**	.52**	.29**	.32**
IDCP-NPD total score	.50**	.68**	.68**	.70**	.48**	.59**	.64**	.49**	.57**

Note. In bold, the correlations between NPD-HiTOP factors and the respective external measure, according to Table 1.

**

p-Value <0.01.

Overall, we observed correlations as expected, although we found some non-expected larger effect sizes, as is the case of Rudeness's higher correlation with Vulnerability. Callousness presented the strongest correlation with the Lack of empathy factor (.68), and not with the PID-5 facet Callousness (.65). We must observe that the Lack of empathy and Callousness factors evaluate analogous behavioural tendencies, pertinent for what we intended for IDCP-2 Callousness content. Besides, the effects (.68 and .65) are virtually equal.

Table 4 presents findings on the capability of IDCP-NPD to discriminate between healthy and pathological groups.

Table 4.

Student's t-test for group comparison and Cohen's d.

Factors	Group	M	SD	t (df = 101)	d
Callousness	Healthy	1.27	0.36	−9.36	0.78**
	Pathological	2.49	0.90
Superiority	Healthy	1.22	0.44	−8.30	1.54**
	Pathological	2.36	0.92
Dominance	Healthy	1.34	0.31	−15.71	2.92**
	Pathological	2.85	0.64
Vulnerability Grandiosity	Healthy	1.48	0.41	−8.97	1.67**
	Pathological	2.57	0.80
Rudeness	Healthy	1.39	0.31	−11.50	2.11**
	Pathological	2.67	0.77
Egocentricity	Healthy	1.14	0.16	−12.77	2.29**
	Pathological	2.55	0.81
Seduction and manipulation	Healthy	1.15	0.30	−13.28	2.42**
	Pathological	2.75	0.85
Deceitfulness	Healthy	1.11	0.15	−16.53	2.96**
	Pathological	2.97	0.83
Antisocial Grandiosity (factor)	Healthy	1.28	0.17	−18.42	3.35**
	Pathological	2.65	0.53
Interpersonal Dominance (factor)	Healthy	1.24	0.25	−15.17	2.78**
	Pathological	2.65	0.65
IDCP-NPD total score	Healthy	1.26	0.19	−19.28	3.53**
	Pathological	2.65	0.50

Note. df: degrees of freedom; 0: healthy group (n = 70); 1: pathological group (n = 71).

**

p-Value <0.01.

We observed higher means for the pathological group. All comparisons were significant, showing strong effect sizes. Deceitfulness and Dominance had the highest effects for the facets, and Callousness had the lowest. The Antisocial Grandiosity factor presented the most prominent effect, being the smallest only when compared to the total score.

Discussion

Operationalizing psychiatric diagnostic models is an emergency and central to the evaluative practice. We focused on it by developing a HiTOP-based specific version of the IDCP-2 to assess traits related to narcissistic personality disorder (NPD). We named this measure IDCP-NPD, which consists of a hybrid scale that allows clinicians to approach the HiTOP¹⁰ core traits to what we have understood as NPD in the traditional psychopathology taxonomies.⁵ To do so, we had an initial stage for developing the IDCP-NPD scale based on the current HiTOP Antagonistic Externalizing spectrum, paralleling it to the central dimensions and criteria for the NPD. Then, we investigated the scale's internal structure and external validity through the correlation between the IDCP-NPD factors and related external measures.

In the development stage, as summarized in Table 1, we gathered scales that are known to cover traits related to NPD. We consistently steered towards the HiTOP's antagonist externalizing spectrum, focusing on characteristics related to grandiosity, dominance, and manipulation.^10,14–16 As central to the current understanding of NPD on the categorical and alternative hybrid model,⁵ we conceptually selected eight antagonist externalizing traits and operationalized them in 11 specific factors. In addition, we noted the potential clinical importance of including vulnerability assessment linked to NPD.⁴⁶ In doing so, we stuck to covering overt and covert traits of NPD.⁴⁵

After revising and selecting the prominent HiTOP traits we intended to cover at the IDCP-NPD, we relied on conceptually identifying those on the preexistent IDCP-2 factors.²⁰ We extracted nine factors from IDCP-2 and identified content gaps pertinent to NPD-related behaviour, such as Rudeness, Egocentricity, and Vulnerability. For Rudeness, we created items focused on the lesser interpersonal preoccupation posture with people around. At the same time, we covered the self-centred expected behaviour for Egocentricity. For Vulnerability, we assessed instances when those grandiose and domineering antagonistic dispositions seem to hide feelings of vulnerability. Additionally, we analyzed the factors extracted from the IDCP-2 in light of the conceptual definitions we intend to measure and made revisions and item additions to factors with content partially represented by the existing items. The joint work of the authors, specialists in personality assessment, and scholars in the taxonomy of personality disorders concluded this stage with the proposition of a pilot version of the IDCP-NPD to be empirically tested, comprising 12 factors and 73 items.

After collecting community sample data, we investigated the IDCP-NPD internal structure. We found a 2-factors solution, with good fit indices, named Antisocial Grandiosity and Interpersonal Dominance. Previous studies have found a 2-factors solution suitable for a set of traits related to superiority, entitlement, and need for praise (factor 1), and lack of empathy, arrogance, and exploitativeness (factor 2) (e.g., Supplementary Table S9 by Aslinger et al.⁴⁷), suggesting that they cover grandiosity and antagonism as two critical aspects dimensions of NPD. We can see these trends in our data, and besides the cross-content with Aslinger et al., we suggested a different name, keeping Antagonism as an identifier for spectra levels. Factor 1 seems related to grandiosity, comprising aspects of grandiose egocentricity, but also involving a propensity to antagonism by the presence of callousness stance, deceitfulness style, and rudeness behaviour. Factor 2 has an interpersonal bias strongly inclined towards dominance over the other, too attached to seductive and manipulative behaviour to control others for their selfish self-interest. We observed that the Superiority factor has low equal loadings on the two factors, although we approach its interpersonal features as more conceptually related to the Interpersonal Dominance factor. We recognize that superiority attributes are also encompassed by dominance, egocentrism, and vulnerability grandiosity subscales from IDCP-NPD.

Having settled the two factors’ structure, we carried out the investigation of their interpretive validity based on related external measures. The expected convergence between our hypotheses (see Table 1) and the associations observed (see Table 3) allows us to assume the usability of IDCP-NPD for clinical purposes. We found the expected associations between the IDCP-NPD factors and external measures,^10,19,28,37 according to the HiTOP specifications.¹⁰ An exception was the Rudeness factor, which was more associated with the Vulnerability facet of HSNS and not with the Rudeness factor of CAT-PD. Although we did not expect this association, it is consistent with the content evaluated by the factors Rudeness (i.e., generalized irritation and anger based on the belief that people always want to harm) and Vulnerability (i.e., grandiose and domineering antagonistic disposition seems to hide feelings when feeling vulnerable). The effect size of the association between these factors was similar to what we observed with the CAT-PD Rudeness factor. Furthermore, the discriminant capacity of IDCP-NPD was tested through a mean comparison between healthy and pathological groups. Although this type of finding adds little to what we already know about HiTOP, it is crucial for establishing the clinical usability of the test (IDCP-NPD). The comparison between groups indicated that the IDCP-NPD helps discriminate individuals high in the NPD traits from those low in these traits. In other words, the group with high scores had higher means in the factors and the total score of the IDCP-NPD. The Deceitfulness and Dominance factors presented the best performance to differentiate the groups. The Deceitfulness trait represents NPD's manipulation tendency (i.e., seeks to achieve his goals through lies, control and manipulation).¹⁶ The Dominance factor, which assesses the belief and use of manipulation to get your way, represents thoughts of deserving more than others, a typical trait of the NPD.¹ Considering the two factors, the Antisocial Grandiosity factor was the one that presented the best performance to differentiate the groups. This factor is related to grandiosity and egocentricity but also involves a propensity to antagonism, including callousness stance, deceitfulness style, and rude behaviour, which are core traits of NPD.¹⁶

Although we recommend using IDCP-NPD in clinical practice to screen NPD traits, our findings should be considered in light of methodological limitations. First, the sample consisted of individuals from the general population, not including people with a known diagnosis of NPD. Second, the external measures used in this study do not have cutoffs, precluding the establishment of a genuinely pathological group. Given these limitations, we suggest the conduction of further studies verifying the discriminative capacity of the IDCP-NPD for people with and without an NPD diagnosis, as well as seeking to establish a clinical cutoff for the scale.

In essence, using robust data analysis, IDCP-NPD showed good initial psychometric properties to assess traits from the narcissistic personality disorder, and our findings are consistent with previous empirical evidence with NPD,^2–4 and, more specifically, with the HiTOP model.^10,14–16 While we do not have a clinically established cutoff for the IDCP-NPD, we can carefully use the findings of this study for clinical care. For instance, we recommend using a score equal to 1.5 as clinically relevant for the total score, Antisocial Grandiosity, and Interpersonal Dominance factors. In other words, the professional can employ the IDCP-NPD in clinical settings, and patients with scores equal to or higher than 1.5 should be further investigated in a more detailed assessment. The factors from IDCP-2 composing the IDCP-NPD should be interpreted according to the norms and standardization of the scale.³¹

Conflicts of interest

The authors have no conflicts of interest to declare.