Extension of the hierarchical model of anxiety and depression to the personality domain

Research on the empirical structure of distress (mood and anxiety) disorders has yielded many valuable insights. These efforts have culminated in the integrative hierarchical model (Mineka, Watson, & Clark, 1998), which specified general, specific, and unique components of distress disorders. However, some components have not yet been identified, and the model cannot account for temporal relations among the syndromes. Clark, Watson, and Mineka (1994) suggested that personality traits may help us fill these gaps.

Accordingly, I proposed a hierarchical-vulnerability model of distress disorders, which posits that psychiatric syndromes are products of interactions between personality traits and environmental stressors. This model is patterned after the integrative hierarchical model, as it specifies various levels of generality among personality-psychopathology links. The aim of this study was to develop the personality component of the model by identifying traits relevant to four target syndromes: major depression, panic disorder, social phobia, and OCD. Building on previous research, I hypothesized several trait contributors for each of the disorders.

I administered an extensive personality battery and an interview measure of the target syndromes to two samples: 385 undergraduates and 188 psychiatric patients. First, I evaluated the associations among the personality measures. Next, I tested study hypotheses using correlational and multiple regression analyses. I also examined the robustness of results across samples.

The results confirmed the central role of negative emotionality as the shared trait vulnerability. I also identified two specific trait contributors (linked to two disorders each), and seven unique contributors. Notably, all unique associations involved either depression or social anxiety. The model was able to explain roughly half of variance in these two syndromes but only approximately a quarter of the variance in panic and OCD. On the other hand, the traits consistently accounted for 75% of the covariance among the syndromes.

In sum, the hierarchical-vulnerability model offers a useful approach to conceptualizing personality-psychopathology relations. However, the stress component of the model is yet to be developed. Furthermore, the present findings need to be replicated in new samples and evaluated in longitudinal studies. Finally, the model needs to be extended to include other disorders and traits.