The structure of common and severe psychopathology: analyses of syndromes and symptoms

One of the main goals of research in clinical psychology is to enhance the understanding and conceptualization of psychopathology. As such, it is essential that the model used to classify mental illness be as valid as possible. The Diagnostic and Statistical Manual of Mental Disorders (4th ed., text rev; DSM-IV-TR, American Psychiatric Association, 2000) provides the current model of psychopathology. However, there has been growing dissatisfaction with the current version of the DSM and there is an increasing view that the DSM taxonomy is scientifically unsound (Brown & Barlow, 2005; Watson & Clark, 2006; Widiger & Samuel, 2005). Quantitative structural models of psychopathology (e.g., Internalizing/Externalizing; Krueger, 1999) have provided an alternative to the DSM model and have greatly advanced the conceptualization of psychopathology. Although these models represent a significant improvement they still have several limitations. First, they exclude many of the "severe" disorders (e.g., psychotic disorders). Second, the placement of some disorders (e.g., OCD and PTSD) in the model has not been consistent. Finally, they are based on categorical diagnoses, which are less valid and reliable than dimensional alternatives (Markon, Chmielewski, & Miller, 2011). Thus, the use of categorical diagnosis as the foundation for empirical models of psychopathology likely results in suboptimal models. Moreover, it has been argued that many diagnoses are excessively heterogeneous, suggesting that analyzing the symptoms that underlie the disorders could result in more fine-grained models that more closely "carves nature at its joints."

The current study addresses the limitations of current quantitative models to create an expanded model of psychopathology that likely has increased validity. Structural analyses were conducted at both the syndromal level (to serve as a proxy for dimensional versions of DSM diagnosis) and the symptom level in a large patient sample using both self-report and interview data. A three-factor model, containing Internalizing, Externalizing, and Psychoticism/Oddity dimensions emerged across both sets of analyses The emergence of this model at both the syndromal and symptom level provides strong evidence that a third higher order Psychoticism/Oddity dimension is necessary for a more complete model of psychopathology. The association of these dimensions with normal personality traits and other external correlates also was examined and implications for the conceptualization of psychopathology are discussed.