The Structure of DSM-IV Borderline Personality Disorder and Its Implications for Treatment

Abstract

Since Stern’s observations (1938), the term borderline gained widespread attention in psychiatric practice. For a long time, this term was used in a confusing way, identifying states, syndromes, personality types, and schizophrenia subgroups. In the ’60s, Kernberg (1967) described Borderline personality as a distinct personality organization (BPO), different from Neurotic and Psychotic personality organizations, as well as from classic psychiatric syndromes (e.g., schizophrenia, mood disorders, etc.). With the introduction of the Diagnostic and Statistical Manual of Mental Disorder, 3rd Edition (DSM-III) (APA, 1980) and, later, 3rd Edition, Revised (DSM-III-R) (APA, 1987) the Borderline Personality Disorder (BPD) was described as one of the Axis II Personality Disorders (PDs). While Kernberg’s BPO heavily relied on a psychoanalytically-oriented diagnostic framework (i.e., based on the identification of inferred psychic functioning), DSM-III/-R tried to convert the previous clinical pictures in an atheoretical, operationalized set of diagnostic criteria for BPD. In agreement with the neo-kraepelinian point of view, BPD was considered as a unidimensional, categorial diagnosis. However, in recent years clinical psychologists built many models normal and abnormal personality functioning based on dimensional constructs. This raised a controversy between models considering BPD as a distinct psychiatric disorder and those looking at BPD as an extreme variant of a normally distributed personality dimension. For the former BPD is a complex intermingling of several stable characteristics of personality (behavioral, cognitive, affective, interpersonal, and temperamental), while for the latter BPD is the extreme expression of a temperamental disposition, such as Novelty Seeking or Neuroticism. This controversy refers to a deeply rooted debate, raised in clinical practice, stressing the pros and cons of each approach. Categorial models are economic, easily transmissible, widely used, and provided with external validity. However, they are stereotyped in non-prototypical and boundary cases, and do not explain comorbidity. Dimensional models provide more vivid pictures, are flexible, and could be converted to categories whenever needed, while the reverse could not be done. However, they are theoretically-based and there is no agreement between theorists on the number of dimensions. While in medicine research showed that some pathologies are better described by categories (e.g., diabetes mellitus), while others by dimensions (e.g., blood hypertension), up to now no definitive data are available in psichiatric and psychological research on PDs.