Distinguishing symptom dimensions of depression and anxiety: An integrative approach


BACKGROUND: Clark and Watson developed the tripartite model in which a symptom dimension of ‘negative affect’ covers common psychological distress that is typically seen in anxious and depressed patients. The ‘positive affect’ and ‘somatic arousal’ dimensions cover more specific symptoms. Although the model has met much support, it does not cover all relevant anxiety symptoms and its negative affect dimension is rather unspecific. Therefore, we aimed to extend the tripartite model in order to describe more specific symptom patterns with unidimensional measurement scales. METHOD: 1333 outpatients provided self report data. To develop an extended factor model, exploratory factor analysis (EFA) was conducted in one part of the data (n=578). Confirmatory factor analysis (CFA) was conducted in the second part (n=755), to assess model-fit and comparison with other models. Rasch analyses were done to investigate the unidimensionality of the factors. RESULTS: EFA resulted in a 6-factor model: feelings of worthlessness, fatigue, somatic arousal, anxious apprehension, phobic fear and tension. CFA in the second sample showed that a 6-factor model with a hierarchical common severity factor fits the data better than alternative 1- and 3-factor models. Rasch analyses showed that each of the factors and the total of factors can be regarded as unidimensional measurement scales. LIMITATIONS: The model is based on a restricted symptom-pool: more dimensions are likely to exist. CONCLUSION: The extended tripartite model describes the clinical state of patients more specifically. This is relevant for both clinical practice and research. CI: Copyright A(c) 2011; JID: 7906073; 2011/03/17 [received]; 2011/10/04 [revised]; 2011/10/04 [accepted]; aheadofprint

Journal of Affective Disorders
Stef van Buuren
Stef van Buuren

My research interests include data science, missing data, child growth and development, and measurement.