Edited by: Bo Bach, Psychiatry Region Zealand, Denmark
Reviewed by: Włodzimierz Andrzej Strus, Cardinal Stefan Wyszyński University, Poland; Jens Thimm, University of Bergen, Norway
This article was submitted to Psychopathology, a section of the journal Frontiers in Psychiatry
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Both the Model for Personality Disorders (PDs) in the 11th edition of the International Classification of Diseases (ICD-11) and the Alternative DSM-5 Model for Personality Disorders (AMPD) use personality trait domains to specify individual manifestations of personality psychopathology beyond the evaluation of the overall personality impairment severity (
Although both ICD-11 and AMPD include Negative affectivity, Detachment, Antagonism/Dissociality, and Disinhibition among the five domains of personality traits, some differences can be noted between these models that make it impossible for a clinician to switch seamlessly between the two nomenclatures when describing a patient. In contrast to AMPD, the main differences in ICD-11 include the omission of personality traits qualifiers in relation to schizotypy and psychoticism; the inclusion of the Anankastia qualifier; and the absence of specific trait facets delineating individual qualifiers (
In an effort to synchronize these two models, an algorithm has recently been developed to evaluate the combined AMPD and ICD-11 personality traits model based on six higher-order domains (i.e., Negative affectivity, Detachment, Antagonism, Disinhibition, Anankastia and Psychoticism), covering 17 of the lower-order facets, and featuring a total number of 34 items. This algorithm is captured by the Personality Inventory for DSM-5 - Brief Form Plus (PID-5BF+) (
The aim of this study was to verify the psychometric properties of the independent 36-item version of PID5BF+M and to compare them with the extracted PID5BF+M version from the original PID-5 by testing them for invariance. Given that the focus of the study was to primarily verify the general factor structure of the independent version of the measure, a thorough analysis of the validity of the separate Anankastia domain was not among the main aims of this study. At the same time, we examined the associations between the Borderline Pattern qualifier and the qualifiers of personality traits with respect to the proven continuity of PID5BF+M with specific diagnoses of PDs (
Two samples of volunteers from the general population were used. They consisted of university students from various fields of study, working volunteers and pensioners. To be included in the study, volunteers in both groups needed to fulfill the criterion of being ≥ 18 years of age. Participation in the study was voluntary and anonymous for all respondents, and all participants were asked to give their informed consent to participate in the study, which they had the opportunity to withdraw at any time without stating the reason. Participants were not rewarded for their participation in the study; however, if they were interested in feedback, they could provide us with their email address. The ethics committee of the General university Hospital in Prague approved the study protocol and the informed consent form.
After removing participants based on their PID-5 Response Inconsistency Scale (PID-5-RIS) score (
After removing individuals with patterned responses (
Questionnaires were administered to Sample 2 individually, to be filled either by the paper-and-pencil method or online. In the case of paper-and-pencil administration, respondents were asked to carefully read the instructions before starting the questionnaire. Trained administrators were present during the administration to respond to the possible technical queries of respondents. Online data collection was limited to 3 months. During this period, respondents were addressed anonymously through adverts on social media and relevant websites. They would complete the questionnaires upon accessing a link provided in the advert and were asked to answer all items. Some respondents (
Given that Sample 1 and Sample 2 were drafted from different populations, they showed differences when compared on the demographic characteristics. Specifically, the samples differed on sex, as the proportion of females was higher in Sample 2 than in Sample 1 (67 vs. 51%),
We used the self-report PID5BF+M to operationalize the ICD-11 and DSM-5 domains of personality traits. The complete PID5BF+M consists of 18 facets assessed through 36 items (2 items per facet), rated on a 4-point Likert scale (0 = very untrue or often untrue; 1 = sometimes or somewhat untrue; 2 = sometimes or somewhat true; and 3 = very true or often true). The 6 domains have been calculated based on the average scores of the three primary facets of each particular domain: Negative affectivity has been calculated from the average scores of the facets of emotional lability, anxiousness, separation insecurity; Detachment from the facets of withdrawal, anhedonia, intimacy avoidance; Antagonism from the facets of manipulativeness, deceitfulness, grandiosity; Disinhibition from the facets of irresponsibility, impulsivity, distractibility; Psychoticism from the facets of unusual beliefs and experiences, eccentricity, perceptual dysregulation; and Anankastia from the facets of perfectionism, rigidity, and orderliness. An independent version of the PID5BF+M was administered to Sample 2, while Sample 1 assessment included the PID5BF+M extracted from the 220-item version of PID-5, in accordance with previous studies (
We administered selected items of the self-reported International Personality Disorder Examination Questionnaire (IPDEQ) (
To ensure the validity of data in Sample 1, we used the PID-5-RIS developed by Keeley et al. (
As the first step, we estimated the fit of the six-factor model in Sample 1, which was administered the full 220-item version of the PID-5, using exploratory structural equation modeling (ESEM). This was first estimated in a multigroup model to evaluate whether there were any differences in the subgroups, i.e., the Police applicants and the community sample. As no substantial differences were found, we merged the subgroups into a single group to increase the sample size. Subsequently, the model was re-evaluated in terms of model fit, and the pattern of factor loadings.
In the next step, we fit the model in Sample 2, which was administered the independent 36-item PID5BF+M. Again, this was first assessed as a multigroup model to evaluate the differences in the methods of administration (paper-pencil vs. online). These methods have been previously found to be invariant in Sample 1 (
Subsequently, we estimated the internal consistency of each facet using polychoric correlations (as there are only two items per facet) and McDonald's omega for the reliability of the domain scores. We also examined the test-retest reliability in Sample 2. The convergent validity of the shortened 36-item version extracted from the full version was assessed by comparing correlations of 15 facets and 5 domains defined by the 220-item version in Sample 1 (we omitted the Anankastia domain and its relevant facets of perfectionism, rigidity, and orderliness, which was not part of the original PID-5).
Moreover, we directly compared the model fit of the extracted and the independent version PID5BF+M by combining Sample 1 and 2 in a multigroup model and testing whether there was support for measurement invariance. Finally, the predictive validity of the independent 36-item PID5BF+M was tested by examining the associations of the six hypothesized domains and the two types of emotionally unstable PDs (i.e., borderline and impulsive) indexed by the IPDEQ, as well as of the emotionally unstable PD as a whole.
All ESEM models were estimated in Mplus 8 (
First, we compared the model fit of the community and the Police applicants subsamples, comprising Sample 1. The six-factor ESEM extracted from the original 220-item PID-5 was estimated in a multigroup model within each group. There were minimal differences between the configural (loadings and intercepts freely estimated in each group) and the scalar (loadings and intercepts fixed to equality across groups) model, S-B Δ
Loadings patterns of the facets derived from the extracted version of PID5BF+M.
Emotional lability | −0.04 | 0.01 | 0.29 | 0.06 | 0.14 | |
Anxiousness | 0.11 | 0.03 | −0.04 | 0.00 | 0.03 | |
Separation insecurity | 0.00 | 0.06 | 0.16 | −0.06 | ||
Withdrawal | 0.12 | 0.05 | −0.12 | −0.01 | 0.08 | |
Anhedonia | 0.09 | 0.09 | 0.27 | 0.04 | −0.10 | |
Intimacy avoidance | −0.07 | 0.04 | 0.05 | 0.20 | 0.07 | |
Manipulativeness | 0.00 | 0.09 | −0.04 | −0.05 | 0.07 | |
Deceitfulness | 0.10 | 0.01 | 0.22 | 0.10 | 0.02 | |
Grandiosity | 0.12 | 0.08 | −0.01 | 0.18 | 0.17 | |
Irresponsibility | −0.06 | 0.14 | 0.23 | −0.11 | ||
Impulsivity | 0.05 | 0.03 | 0.07 | 0.01 | 0.07 | |
Distractibility | 0.25 | 0.21 | −0.07 | −0.10 | 0.27 | |
Perfectionism | −0.01 | 0.03 | −0.01 | 0.04 | 0.05 | |
Rigidity | 0.02 | 0.09 | 0.07 | −0.10 | 0.03 | |
Orderliness | 0.19 | 0.06 | 0.03 | 0.02 | 0.02 | |
Unusual beliefs and experiences | 0.10 | 0.05 | 0.17 | −0.01 | 0.05 | |
Eccentricity | 0.08 | 0.07 | 0.08 | 0.04 | 0.07 | |
Perceptual dysregulation | 0.06 | 0.05 | 0.09 | 0.21 | 0.15 |
In the next step, we tested the same model for individuals who were administered the independent 36-item PID5BF+M. First, we tested whether the type of administration (paper-and-pencil vs. online) affected the results in a multigroup model. The results for the multigroup model showed a minimal difference between the configural and the scalar model, S-B Δ
Loadings patterns of the facets derived from the independent version of PID5BF+M.
Emotional lability | 0.01 | 0.01 | 0.02 | 0.07 | 0.05 | |
Anxiousness | 0.29 | −0.05 | 0.28 | 0.07 | −0.02 | |
Separation insecurity | −0.14 | 0.07 | 0.04 | −0.07 | ||
Withdrawal | 0.05 | 0.04 | −0.13 | 0.07 | 0.01 | |
Anhedonia | −0.08 | 0.08 | 0.29 | −0.05 | −0.07 | |
Intimacy avoidance | −0.09 | 0.00 | 0.20 | 0.03 | 0.18 | |
Manipulativeness | −0.03 | 0.02 | −0.01 | 0.01 | 0.11 | |
Deceitfulness | 0.12 | 0.06 | 0.05 | 0.05 | −0.02 | |
Grandiosity | −0.03 | 0.04 | 0.12 | 0.06 | ||
Irresponsibility | −0.02 | 0.09 | 0.28 | 0.27 | ||
Impulsivity | 0.45 | 0.00 | 0.20 | −0.05 | 0.09 | |
Distractibility | 0.24 | 0.19 | 0.11 | −0.19 | 0.25 | |
Perfectionism | 0.08 | 0.04 | 0.08 | 0.03 | 0.08 | |
Rigidity | 0.04 | 0.11 | 0.05 | −0.02 | 0.03 | |
Orderliness | 0.00 | 0.06 | −0.07 | 0.17 | ||
Unusual beliefs and experiences | 0.20 | 0.05 | 0.10 | −0.08 | 0.07 | |
Eccentricity | 0.03 | 0.05 | 0.10 | 0.04 | 0.14 | |
Perceptual dysregulation | 0.04 | 0.07 | −0.01 | 0.26 | 0.02 |
Given that the six-factor solution of the independent version was less stable due to the lack of substantial facet loadings for the Disinhibition domain, we decided to explore this issue further by estimating exploratory factor analyses with varying number of factors (from one to eight) to see whether a different factor solution would provide a clearer pattern. The solutions with fewer than five factors showed poor model fit, suggesting that such factor structures did not accurately represent the data. The five-factor solution provided a clearer factor solution with regards to facet loadings then the six-factor version for the five PID domains sans Disinhibition while the facets of the Disinhibition domain loaded on other domains (impulsivity and withdrawal on Negative affectivity, irresponsibility on Antagonism). The seven- and eight-factor solutions provided an incremental improvement in model fit; however, they have not provided a clearer factor solution with regards to the pattern of facet loadings then the five- or six-factor solutions. The loading patterns for the five-, seven-, and eight-factor solutions and their model fit indices are provided in
Furthermore, we estimated a five-domain model in accordance with the five domains qualifiers defined in ICD-11. The fit of the model was good,
Loadings patterns of a 5-factor ICD-11 structure derived from the independent version of PID5BF+M.
Emotional lability | −0.04 | −0.04 | 0.12 | 0.10 | |
Anxiousness | −0.11 | 0.27 | 0.07 | ||
Separation insecurity | −0.09 | −0.01 | −0.01 | ||
Withdrawal | 0.09 | 0.04 | −0.16 | 0.13 | |
Anhedonia | −0.10 | 0.05 | 0.21 | −0.05 | |
Intimacy avoidance | −0.01 | 0.07 | 0.07 | 0.08 | |
Manipulativeness | 0.00 | 0.01 | 0.02 | 0.05 | |
Deceitfulness | 0.14 | 0.03 | 0.08 | 0.07 | |
Grandiosity | −0.04 | 0.10 | 0.14 | ||
Irresponsibility | 0.10 | 0.18 | 0.40 | −0.24 | |
Impulsivity | −0.01 | 0.21 | 0.00 | ||
Distractibility | 0.26 | 0.20 | −0.12 | ||
Perfectionism | 0.07 | 0.00 | 0.06 | 0.04 | |
Rigidity | 0.04 | 0.07 | 0.02 | −0.01 | |
Orderliness | 0.05 | 0.15 | 0.02 | 0.20 |
Internal reliabilities of the PID5BF+M scales in both samples are shown in
Internal reliabilities of PID-5 facets and domains across two versions of PID5BF+M.
Negative affectivity | Emotional lability | 0.59 | 0.72 | 0.54 | 0.67 |
Anxiousness | 0.86 | 0.78 | |||
Separation insecurity | 0.57 | 0.49 | |||
Detachment | Withdrawal | 0.60 | 0.66 | 0.51 | 0.59 |
Anhedonia | 0.51 | 0.54 | |||
Intimacy avoidance | 0.44 | 0.57 | |||
Antagonism | Manipulativeness | 0.71 | 0.74 | 0.68 | 0.70 |
Deceitfulness | 0.64 | 0.49 | |||
Grandiosity | 0.54 | 0.32 | |||
Disinhibition | Irresponsibility | 0.49 | 0.73 | 0.40 | 0.65 |
Impulsivity | 0.48 | 0.47 | |||
Distractibility | 0.53 | 0.60 | |||
Anankastia | Perfectionism | 0.58 | 0.69 | 0.50 | 0.72 |
Rigidity | 0.49 | 0.57 | |||
Orderliness | 0.58 | 0.64 | |||
Psychoticism | Unusual beliefs and experiences | 0.54 | 0.74 | 0.49 | 0.75 |
Eccentricity | 0.68 | 0.54 | |||
Perceptual dysregulation | 0.58 | 0.53 | |||
Average reliability | 0.58 | 0.71 | 0.54 | 0.68 |
For the independent 36-item version, lower correlations were found for more facets than in the extracted version; these included separation insecurity, deceitfulness, grandiosity, irresponsibility, impulsivity, and unusual beliefs and experiences. In the case of the independent 36-item version of PID5BF+M, all domain reliabilities were satisfactory. The reliability of domains was as follows: Negative affectivity (ω = 0.67), Detachment (ω = 0.59), Antagonism (ω = 0.70), Disinhibition (ω = 0.65), Anankastia (ω = 0.72), and Psychoticism (ω = 0.75), with the average domain trait scores reliability of 0.68.
The results also showed good test-retest reliability of the underlying domains of the independent 36-item PID5BF+M: Negative affectivity
The correlations between the facets derived from the original 220-item PID-5 and the facets defined by PID5BF+M are shown in
Correlations of selected facets and domains between the 220-item and the 36-item PID-5.
36-item version | EMO Emotional lability | 0.77 | 0.58 | 0.39 | 0.25 | 0.22 | 0.06 | 0.25 | 0.33 | 0.23 | 0.29 | 0.46 | 0.48 | 0.33 | 0.45 | 0.54 | 0.71 | 0.23 | 0.31 | 0.50 | 0.50 |
ANX Anxiousness | 0.59 | 0.88 | 0.41 | 0.38 | 0.40 | 0.11 | 0.21 | 0.33 | 0.24 | 0.22 | 0.31 | 0.52 | 0.30 | 0.46 | 0.52 | 0.76 | 0.37 | 0.30 | 0.44 | 0.49 | |
SEP Separation insecurity | 0.43 | 0.47 | 0.82 | 0.17 | 0.35 | 0.11 | 0.22 | 0.27 | 0.19 | 0.29 | 0.36 | 0.39 | 0.24 | 0.34 | 0.38 | 0.69 | 0.25 | 0.26 | 0.42 | 0.36 | |
WIT Withdrawal | 0.26 | 0.44 | 0.07 | 0.81 | 0.45 | 0.33 | 0.24 | 0.31 | 0.23 | 0.23 | 0.17 | 0.38 | 0.21 | 0.39 | 0.34 | 0.31 | 0.69 | 0.31 | 0.32 | 0.37 | |
ANH Anhedonia | 0.36 | 0.44 | 0.30 | 0.51 | 0.75 | 0.26 | 0.25 | 0.36 | 0.21 | 0.36 | 0.36 | 0.53 | 0.26 | 0.38 | 0.43 | 0.45 | 0.63 | 0.32 | 0.51 | 0.41 | |
INT Intimacy avoidance | 0.18 | 0.23 | 0.12 | 0.42 | 0.34 | 0.78 | 0.22 | 0.26 | 0.24 | 0.23 | 0.18 | 0.28 | 0.22 | 0.29 | 0.35 | 0.21 | 0.65 | 0.28 | 0.28 | 0.32 | |
MAN Manipulativeness | 0.24 | 0.27 | 0.16 | 0.34 | 0.25 | 0.17 | 0.82 | 0.75 | 0.49 | 0.48 | 0.29 | 0.31 | 0.39 | 0.43 | 0.41 | 0.27 | 0.33 | 0.80 | 0.41 | 0.47 | |
DEC Deceitfulness | 0.36 | 0.37 | 0.31 | 0.32 | 0.26 | 0.14 | 0.63 | 0.81 | 0.43 | 0.51 | 0.37 | 0.44 | 0.38 | 0.44 | 0.45 | 0.42 | 0.31 | 0.73 | 0.52 | 0.48 | |
GRAN Grandiosity | 0.30 | 0.37 | 0.23 | 0.32 | 0.28 | 0.21 | 0.44 | 0.49 | 0.79 | 0.37 | 0.27 | 0.31 | 0.40 | 0.41 | 0.40 | 0.36 | 0.34 | 0.67 | 0.37 | 0.46 | |
IRR Irresponsibility | 0.41 | 0.34 | 0.25 | 0.35 | 0.28 | 0.18 | 0.42 | 0.52 | 0.39 | 0.76 | 0.53 | 0.57 | 0.42 | 0.56 | 0.52 | 0.41 | 0.35 | 0.52 | 0.72 | 0.58 | |
IMP Impulsivity | 0.53 | 0.40 | 0.39 | 0.27 | 0.32 | 0.10 | 0.31 | 0.42 | 0.27 | 0.47 | 0.94 | 0.55 | 0.30 | 0.46 | 0.49 | 0.54 | 0.29 | 0.39 | 0.80 | 0.48 | |
DIST Distractibility | 0.54 | 0.54 | 0.28 | 0.39 | 0.35 | 0.11 | 0.23 | 0.36 | 0.20 | 0.45 | 0.50 | 0.84 | 0.29 | 0.55 | 0.54 | 0.55 | 0.36 | 0.31 | 0.73 | 0.54 | |
UNU Unusual Beliefs and experiences | 0.39 | 0.40 | 0.21 | 0.28 | 0.23 | 0.18 | 0.40 | 0.44 | 0.38 | 0.38 | 0.30 | 0.40 | 0.80 | 0.61 | 0.60 | 0.41 | 0.30 | 0.48 | 0.43 | 0.76 | |
ECC Eccentricity | 0.43 | 0.44 | 0.25 | 0.38 | 0.27 | 0.17 | 0.38 | 0.45 | 0.41 | 0.47 | 0.39 | 0.49 | 0.49 | 0.87 | 0.60 | 0.45 | 0.35 | 0.48 | 0.53 | 0.78 | |
PCD Perceptual dysregulation | 0.34 | 0.34 | 0.23 | 0.32 | 0.32 | 0.30 | 0.30 | 0.37 | 0.33 | 0.37 | 0.39 | 0.39 | 0.49 | 0.47 | 0.70 | 0.37 | 0.39 | 0.39 | 0.46 | 0.61 | |
NA Negative affectivity | 0.75 | 0.81 | 0.67 | 0.33 | 0.40 | 0.11 | 0.29 | 0.39 | 0.28 | 0.34 | 0.48 | 0.58 | 0.37 | 0.52 | 0.60 | 0.91 | 0.35 | 0.37 | 0.57 | 0.57 | |
DE Detachment | 0.35 | 0.49 | 0.21 | 0.77 | 0.67 | 0.58 | 0.31 | 0.40 | 0.30 | 0.36 | 0.31 | 0.52 | 0.30 | 0.47 | 0.48 | 0.43 | 0.86 | 0.39 | 0.48 | 0.48 | |
AN Antagonism | 0.37 | 0.41 | 0.30 | 0.40 | 0.32 | 0.21 | 0.77 | 0.85 | 0.69 | 0.56 | 0.39 | 0.45 | 0.47 | 0.53 | 0.51 | 0.44 | 0.39 | 0.90 | 0.55 | 0.58 | |
DI Disinhibition | 0.62 | 0.53 | 0.38 | 0.41 | 0.39 | 0.15 | 0.38 | 0.53 | 0.35 | 0.67 | 0.82 | 0.81 | 0.41 | 0.64 | 0.64 | 0.62 | 0.40 | 0.49 | 0.93 | 0.65 | |
PS Psychoticism | 0.48 | 0.50 | 0.29 | 0.41 | 0.34 | 0.26 | 0.46 | 0.52 | 0.47 | 0.52 | 0.44 | 0.53 | 0.75 | 0.83 | 0.78 | 0.51 | 0.42 | 0.57 | 0.59 | 0.91 |
In order to test the similarity of both versions of PID5BF+M, the Sample 1 with the extracted version and the Sample 2 with the independent 36-item version were combined in a multigroup model. The fit of the configural model was
We tested the predictive validity of the independent 36-item PID5BF+M to determine borderline pathology according to ICD-11 by assessing associations with the IPDEQ criteria for emotionally unstable PDs of both types (i.e., borderline, and impulsive). The results showed that the highest positive correlation for the borderline personality subdomain was found with Negative affectivity (
Correlations of 36-item PID5BF+M domains and IPDE-defined emotionally unstable PD.
Negative affectivity | 0.51 | 0.42 | 0.55 |
Detachment | 0.35 | 0.20 | 0.33 |
Antagonism | 0.32 | 0.39 | 0.42 |
Disinhibition | 0.42 | 0.47 | 0.53 |
Ananakastia | 0.12 | 0.11 | 0.14 |
Psychoticism | 0.37 | 0.36 | 0.43 |
The current study aimed to assess the validity of the shortened version of PID-5, the PIDBF+M, using two non-clinical Czech samples. The validity of the measure was assessed in two forms: first, as extracted from the original 220-item PID-5; second, when the PIDBF+M was administered as a stand-alone measure. Evaluation of psychometric properties of the independent PID5BF+M seems to be an important step toward disseminating the dimensional diagnostic approach to a broad range of clinicians who, with international adaptations of ICD-11, urgently need short but reliable instruments for PD diagnostics within the new system (
In terms of maintaining continuity with the previous research (
However, we noted some problems with the Disinhibition domain that deserve a more detailed comment. First, there was the issue of irresponsibility facet loading primarily on Psychoticism and only secondarily on Disinhibition, its respective domain. Although there was a relatively small difference between loadings found in this study (λ = 0.37 vs. λ = 0.36), it can be seen as further evidence of proneness of the irresponsibility factor to cross-loadings, previously demonstrated in the extracted PID5BF+M by other studies using the original 220-item version of PID-5 [e.g., (
Despite the fact that the separation insecurity facet from the extracted PID5BF+M showed a substantial primary loading on the Negative affectivity domain across international samples (
The proposed 6-factor structure of PID5BF+M was also replicated in the 36-item independent version. Nevertheless, the issue of problematic loadings of the three primary facets of the Disinhibition domain was even more prominent in this version. Although our results confirmed the existence of negative cross-loading of the irresponsibility facet on the Anankastia domain and the cross-loading of distractibility on the Negative affectivity and Psychoticism domains (
The loading pattern in the 5-factor ICD-11 version was very similar to the original, six-factor version of the PID5BF+M, suggesting that the above-mentioned issues with low loadings especially for the Disinhibition domain were not alleviated by the omission of the Psychoticism domain.
Although the obtained values of internal consistency are significantly lower in comparison with the original version of PID-5 for both versions of PID-5BF+M, they were still largely satisfactory, both at the level of trait domains and the majority of individual trait facets. This indicates good reliability of the measure despite the substantial reduction in the number of items compared to the previously performed studies [e.g., (
Regarding the stability of the independent version of PID5BF+M over time, although the test-retest period in our study varied (ranging from 1 to 15 weeks), the average test-retest interval of ~1.5 months seems to be sufficient compared to other studies [e.g., (
In assessing the convergent validity of PID5BF+M, we compared the facets and domains of the 36-item version of the questionnaire with the 220-item PID-5 in this study. Because we administered the full version of PID-5 only to Sample 1, an independent version of PID5BF+M could not be included in these analyses. Another issue was the Anankastia domain, which is defined in PID5BF+M by a modified triplet of facets-namely by rigidity, perfectionism and orderliness-unlike PID5BF+, where Anankastia is defined by the facets of rigid perfectionism and perseveration, i.e., facets included in the original version of the PID-5. Given the fact that Anankastia is an integral part of the personality trait model in ICD-11, a more thorough assessment of the convergent and predictive validity of this domain would be appropriate. Nevertheless, the average correlations of 0.81 at the facet level of five of the original PID-5 domains (i.e., Negative affectivity, Disinhibition, Detachment, Antagonism, and Psychoticism) and 0.90 at the domain level confirmed good convergent validity in line with previous studies with PID5BF+ (
In terms of using PID5BF+M as an independent 36-item measure in a routine clinical practice, an important step was to verify the potential differences between the independent and the extracted version of the measure. In our study, the two models were largely invariant, as we found minimal differences in terms of model fit between the configural and the scalar model (all loadings and intercepts constrained to be equal across the samples). However, given that the two samples came from different populations, we urge caution in interpreting this finding as a definite proof of the invariance of the two measures (see Limitations for further discussion of this issue). What our study indicates is that the 220-item version and the 36-item version of PID5BF+M, each estimated in a specific sample, were not dissimilar in terms of factor structure, item loadings, and item intercepts, at least based on change in relative model fit indices.
These results support the consideration of an independent version of the 36-item PID5BF+M as a valid diagnostic tool for assessing maladaptive personality traits according to both DSM-5 and ICD-11. Although this conclusion is consistent with the clinicians' demand for short and valid tools that minimize the patient burden while providing reliable data, it should be borne in mind that the 220-item PID-5 was designed to achieve the most detailed description of patient's strengths and weaknesses (
The main goal of introducing the Borderline Pattern qualifier within the ICD-11 model for PDs, which virtually reflects the diagnostic criteria for the borderline PD in DSM-5 section II, was to maintain diagnostic continuity between the categorical and dimensional models while ensuring the smallest possible overlap of borderline PD with other PDs (
The results of our study contribute to the current state of knowledge in several ways. First, we have confirmed the independent 36-item PID5BF+M as a reliable and valid tool to generally assess personality psychopathology, in accordance with the proposed dimensional model of maladaptive personality traits according to ICD-11, as well as DSM-5 AMPD. We have also demonstrated the predictive validity of this measure in relation to the assessment of borderline psychopathology in line with the transition from categorical to a dimensional diagnosis of PDs. Nevertheless, our findings need to be considered with respect to certain limitations that may inspire future research. We consider the absence of a clinical group of patients to be the main limitation of this study. The inclusion of a clinical cohort could have helped answer the question of the problematic factor loading of the Disinhibition domain. In addition, the disintegration of this domain in the case of the PID5BF+M independent version makes it impossible to establish normative values. Relatedly, instead of “Disinhibition,” the pattern of loadings for this domain in the 36-item extracted version could have warranted using a different label. However, it is not clear whether this finding truly reflects a different factor structure or whether this finding is idiosyncratic to the current data. In this respect, the independent version of the tool should be further examined in international samples to verify the possible impact of translation or cultural specificity. It is also necessary to point out that Sample 1 and Sample 2 differed in terms of sex ratio, average age, or highest attained education, as they were convenience samples drafted from different populations. As such, we cannot rule out the possibility that the lack of substantial differences between the independent and the extracted version of PID5BF+M might be due to differences in the characteristics of these samples. However, past research on the PID5BF+M does not suggest that these characteristics would substantially affect the structure of the model (
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
The studies involving human participants were reviewed and approved by Ethics Committee of the General University Hospital in Prague Na Bojišti 1, 128 08 Prague 2. The patients/participants provided their written informed consent to participate in this study.
KR drafted this paper and was responsible for its final version. AK conducted all data analyses. LS was responsible for data collection. All authors contributed to the article and approved the submitted version.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The handling editor declared a past co-authorship with one of the authors KR.
The Supplementary Material for this article can be found online at:
1In the remaining part of the manuscript, we employed the label “Disinhibition” and used its original facets (i.e., impulsivity, irresponsibility, distractibility) to compute this domain score in the subsequent analyses. This decision was made for the purpose of consistency and comparison. However, we acknowledge that this domain was not well-defined by its respective three facets in Sample 2 (see more in Discussion).