Spanish Clinical Interview Proves Valid for Evaluating Structural Personality Organization

A cross-sectional clinical validation study conducted across outpatient and community cohorts in Barcelona has demonstrated that the Spanish version of the Structured Interview of Personality Organization Revised, known as the STIPO-R, exhibits robust reliability and multi-faceted construct validity for assessing normal and pathological personality organization. Developed to operationalize Otto Kernberg’s structural model of personality pathology, the instrument addresses enduring challenges in diagnostic practice by offering a comprehensive dimensional evaluation that captures intrapsychic mechanisms alongside behavioral manifestations. While contemporary diagnostic systems, including the Alternative Model for Personality Disorders introduced in Section Three of the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, Text Revision, have increasingly emphasized dimensional impairments in personality functioning, their operationalization through measures like the Level of Personality Functioning Scale captures only a subset of Kernberg’s original structural determinants. In particular, defensive operations remain absent from the Alternative Model framework, while aggression and moral functioning are only partially represented within descriptive trait domains. By empirically confirming the psychometric viability of the Spanish STIPO-R, the investigation demonstrates that clinicians can systematically assess defensive operations, aggression, and moral values alongside identity and object relations within a unified, standardized semi-structured interview protocol.

The theoretical framework underpinning the STIPO-R posits that personality organization exists along a developmental continuum extending from healthy, integrated functioning through neurotic structures to borderline and psychotic personality organization. Within this multidimensional paradigm, the organizing core of personality pathology resides in motivational determinants and intrapsychic structures rather than mere descriptive signs and symptoms, anticipating the dimensional paradigm shift formalized in contemporary nosological frameworks. To rigorously translate and adapt the original English 55-item semi-structured interview developed by John Clarkin, Barry Stern, and colleagues, the research team instituted an extensive linguistic and conceptual adaptation protocol following international methodological guidelines. Two independent native Spanish speakers familiar with clinical personality constructs produced an initial forward translation, after which a consensus version underwent blind back-translation into English by an independent native English speaker. Discrepancies were resolved collaboratively between the translators and external clinical experts, with subsequent direct oversight and approval from the original instrument developers. Following initial piloting with twenty individuals from the target population to ensure semantic equivalence, comprehensibility, and clinical feasibility, the Spanish adaptation established an empirical foundation for formal psychometric evaluation within clinical healthcare settings.

The study protocol evaluated a total sample of 206 adult participants recruited between September 2018 and December 2023 from public healthcare and university settings in Barcelona, Spain. The clinical cohort comprised 134 outpatient psychiatric patients with confirmed personality disorder diagnoses treated in a community mental health unit, while the comparison control cohort comprised 72 university students without any personality disorder diagnosis in the preceding ten years. Clinical recruitment was executed across three sequential visits that incorporated rigorous screening to exclude confounding conditions such as schizophrenia, bipolar disorder type I, active substance use disorder, and organic medical conditions. Diagnostic status within the clinical cohort was established through individual administration of the Structured Clinical Interview for DSM-IV Axis II Personality Disorders, which served as the diagnostic reference standard, revealing that 86.51 percent of clinical participants met criteria for Cluster B disorders, 63.71 percent for Cluster C, and 33.6 percent for Cluster A, with borderline personality disorder representing the single most frequent presentation. Across all demographic markers aside from age, the clinical group exhibited pronounced indicators of psychosocial vulnerability, including elevated rates of prior psychiatric hospitalization, active outpatient engagement, unemployment, and documented history of suicidal behavior.

Content validity was formally evaluated through a blinded, quantitative expert judgment methodology designed to confirm whether the adapted items accurately represented their theoretical domains. The research team convened an independent panel of four licensed female clinical psychologists possessing between 8 and 17 years of specialized professional experience in personality disorder diagnosis and treatment within the Spanish national health system. The judges, who were thoroughly versed in Kernberg’s structural model but completely unfamiliar with the specific interview questions, received uniquely randomized 58-item versions of the Spanish STIPO-R, which expanded the 55 parent items to separate three sub-items within the opening interview query. Working independently, each expert classified every item into its corresponding clinical domain, secondary subdomain, and the embedded eleven-item Narcissism scale. Inter-rater concordance analysis yielded strong to near-perfect global agreement, quantified by an overall Cohen’s Kappa of 0.85 for domain assignments, 0.94 for subdomain allocations, and 0.86 for the Narcissism scale. Chi-square tests of Kappa equality confirmed no statistically significant discordance between expert assignments and the predefined instrument structure for subdomains and the Narcissism scale, establishing robust empirical evidence that the Spanish translation preserves theoretical fidelity and content representativeness.

Structural validity was tested through Confirmatory Factor Analysis utilizing the Diagonally Weighted Least Squares estimation method in the Lavaan software framework, modeling 55 item indicators, 171 freely estimated parameters, and 1,424 degrees of freedom across the 206 participants. Statistical power calculations confirmed that the sample size provided 82 percent power to detect an effect size variance of 0.006 in terms of the Root Mean Square Error of Approximation. The hypothesized six-factor dimensional model demonstrated adequate global fit without post-hoc respecification, yielding an overall chi-square value of 2484.2 with fit indices satisfying established acceptance thresholds, including Comparative Fit Index and Tucker-Lewis Index values above 0.90, as well as Root Mean Square Error of Approximation and Standardized Root Mean Square Residual values below 0.08. Factor saturation values across the latent constructs were exceptionally high, registering at 1.00 for Identity, 0.99 for Object Relations, 0.98 for Primitive Defenses, 0.91 for Advanced Defenses, 0.97 for Aggression, and 0.83 for Moral Values. Significant parameter variability was confirmed across nearly all clinical indicators, although the Primitive Defenses domain demonstrated low item-level variance that suggested potential opportunities for future psychometric refinement.

Evaluation of internal consistency across the confirmed dimensional scales demonstrated robust measurement reliability, satisfying conventional psychometric benchmarks. Reliability estimation using McDonald’s Omega coefficient yielded values well at or above the acceptable threshold of 0.70 across all six evaluated dimensions. Specifically, the Identity domain achieved an Omega coefficient of 1.00, Object Relations demonstrated an Omega of 0.93, Primitive Defenses registered at 0.89, Advanced Defenses yielded an Omega of 0.77, Aggression recorded an Omega of 0.91, and Moral Values produced an Omega coefficient of 0.79. Analysis of monotonic inter-domain associations via Spearman rank correlations indicated that most dimensional scales and the composite Narcissism scale yielded coefficients ranging between 0.60 and 0.80. The Moral Values domain demonstrated monotonic correlations ranging from 0.42 with Advanced Defenses to 0.70 with the overall instrument mean. These findings confirm that while a strong higher-order dimension of general personality pathology severity accounts for considerable shared variance across domains, the individual subscales preserve meaningful domain-specific variance, reinforcing Kernberg’s conceptualization of personality organization as comprising distinct yet structurally interrelated functional systems.

Concurrent validity was examined by evaluating systematic associations between Spanish STIPO-R domain ratings and categorical personality disorder diagnostic clusters derived from the Structured Clinical Interview for DSM-IV Axis II Personality Disorders. In accordance with theoretical expectations, the Identity, Primitive Defenses, Advanced Defenses, and Aggression domains, alongside the cross-cutting Narcissism scale, exhibited their strongest associations with Cluster B diagnoses, which encompass antisocial, borderline, histrionic, and narcissistic presentations. Significant secondary associations were observed between Identity and Primitive Defenses with Cluster A pathology, consistent with the primitive defensive organization and identity fragmentation characteristic of paranoid, schizoid, and schizotypal structures. Advanced Defenses demonstrated significant associations with both Cluster B and Cluster C, reflecting the mixed defensive repertoire frequently observed among outpatient clinical populations where neurotic coping strategies interface with characterological distress. Notably, the embedded Narcissism scale exhibited significant associations across all three diagnostic clusters, supporting Kernberg’s structural observation that narcissistic pathology operates across a wide developmental spectrum ranging from higher-functioning neurotic organization to severe borderline personality pathology.

The concurrent validity analyses also identified critical discriminant distinctions that illuminate differences between categorical nosology and dimensional structural assessment. The Aggression domain demonstrated no statistically significant associations with either Cluster A or Cluster C, consistent with previous empirical literature indicating that overt destructive behavior and severe aggressive impulses are disproportionately concentrated within Cluster B conditions. The Object Relations domain exhibited no significant associations with any of the three categorical diagnostic clusters. The authors attributed this absence of association to fundamental conceptual divergences between diagnostic frameworks: while categorical criteria focus primarily on descriptive maladaptive traits and overt behaviors, the STIPO-R Object Relations domain evaluates the structural quality of internal representational worlds, empathic capacity, relational commitment, and interpersonal depth, aligning more directly with dimensional Criterion A functional impairment. Furthermore, the Moral Values domain demonstrated significant associations with Clusters A and C but failed to associate with Cluster B, a pattern explained by the low prevalence of antisocial personality disorder within the outpatient sample and the STIPO-R’s focus on internal guilt, ethical standards, and moral reflection rather than overt behavioral violations alone.

Criterion validity was established through non-parametric group comparisons and formal Receiver Operating Characteristic curve analyses contrasting the 134 clinical outpatients against the 72 healthy community controls. Mann-Whitney-Wilcoxon testing confirmed that the clinical cohort scored significantly higher across all individual domains, all secondary subdomains, the composite Narcissism scale, and the global interview mean, with between-group distributions differing significantly at p less than 0.01. In Receiver Operating Characteristic analyses modeling group classification from logistic regressions, the global STIPO-R mean achieved an Area Under the Curve of 0.99, accompanied by a Wald statistic of 20.86. Individual domain analyses yielded Area Under the Curve values of 0.99 for Identity, 0.98 for Aggression, 0.96 for Object Relations, 0.96 for the Narcissism scale, 0.95 for Primitive Defenses, 0.94 for Advanced Defenses, and 0.81 for Moral Values. Applying the Youden Index identified an optimal diagnostic cut-off point of 0.52, yielding an empirical sensitivity of 0.96, a specificity of 0.99, and an overall correct classification accuracy of 97.09 percent, while the Accuracy Ratio criterion defined a threshold of 0.40 that achieved 97.57 percent diagnostic accuracy, demonstrating exceptional capacity to differentiate psychiatric patients from healthy individuals.

Notwithstanding these robust psychometric properties, the investigators identified several methodological limitations that delineate the scope of clinical interpretation and establish priorities for future inquiry. Because a single clinical investigator administered both the diagnostic reference standard and the Spanish STIPO-R, shared method variance and rater confirmation bias cannot be entirely excluded, highlighting the necessity of future multi-rater studies to formally quantify inter-rater reliability. Additionally, formal measurement invariance testing across study cohorts could not be executed due to complete absence of response variability in the Aggression domain among the healthy control group. The utilization of an extreme-groups case-control design comparing diagnosed psychiatric outpatients directly against healthy university students represents an additional constraint, as it does not capture the full clinical spectrum of subthreshold or mixed-pathology presentations and may yield inflated discrimination estimates. Nevertheless, the empirical findings confirm that the Spanish STIPO-R is a valid, reliable, and clinically informative instrument that enriches clinical practice by evaluating the full theoretical spectrum of personality organization, equipping clinicians to formulate nuanced structural diagnoses, calibrate level-of-care allocations, and establish prioritized therapeutic interventions for personality pathology.

Subject of Research: Psychometric validation of the Spanish Structured Interview of Personality Organization Revised

Article Title: Psychometric properties of the Spanish structured interview of personality organization revised (STIPO-R) in a clinical and community sample

Article References: Rufat, M. J., Matalí, J. L., Gutiérrez, F., Ferrer, M., Sánchez, E., Zumel, C., Ripoll, E., Vegué, J., & Calvo, N. (2026). Psychometric properties of the Spanish structured interview of personality organization revised (STIPO-R) in a clinical and community sample. Borderline Personality Disorder and Emotion Dysregulation, 13(1), Article 12. https://doi.org/10.1186/s40479-026-00342-6

Image Credits: AI Generated

DOI: 10.1186/s40479-026-00342-6

Keywords: personality disorders, psychometrics, clinical psychology, STIPO-R, personality organization, diagnostic assessment, confirmatory factor analysis, Otto Kernberg, Psychometric, properties, Spanish, structured