Seven basic dimensions of personality pathology and their clinical consequences: Are all personalities equally harmful?

OBJECTIVES: Dimensional pathology models are increasingly being accepted for the assessment of disordered personalities, but their ability to predict negative outcomes is yet to be studied. We examine the relative clinical impact of seven basic dimensions of personality pathology through their associations with a wide range of clinical outcomes.

METHODS: A sample of 960 outpatients was assessed through a 7-factor model integrating the Cloninger, the Livesley, and the DSM taxonomies. Thirty-six indicators of clinical outcome covering three areas - dissatisfaction, functional difficulties, and clinical severity - were also assessed. The unique contribution of each personality dimension to clinical outcome was estimated through multiple regressions.

RESULTS: Overall, personality dimensions explained 17.6% of the variance of clinical outcome, but varied substantially in terms of their unique contributions. Negative Emotionality had the greatest impact in all areas, contributing 43.9% of the explained variance. The remaining dimensions led to idiosyncratic patterns of clinical outcomes but had a comparatively minor clinical impact. A certain effect was also found for combinations of dimensions such as Negative Emotionality × Impulsive Sensation Seeking, but most interactions were clinically irrelevant.

CONCLUSIONS: Our findings suggest that the most relevant dimensions of personality pathology are associated with very different clinical consequences and levels of harmfulness.

PRACTITIONER POINTS: The relative clinical impact of seven basic dimensions of personality pathology is examined. Negative Emotionality (Neuroticism) is 6-14 times as harmful as other pathological dimensions. The remaining dimensions and their interactions have very specific and comparatively minor clinical consequences.

LIMITATIONS: We examine only a handful of clinical outcomes. Our results may not be generalizable to other clinical or life outcomes. Our variables are self-reported and hence susceptible to bias. Our design does not allow us to establish causal relationships between personality and clinical outcomes.