LAUSR

BACKGROUND Individual temperament is associated with psychiatric morbidity and could explain differences in psychiatric comorbidities. We investigated the association of temperament profile clusters with anxiety disorder comorbidity in patients with depression. METHODS We assessed the temperament of 204 specialized care-treated depressed patients with the Temperament and Character Inventory (TCI-R) and their diagnoses with the Mini International Neuropsychiatric Interview. Two-step cluster analysis was used for defining patients' temperament profiles and logistic regression analysis was used for predicting different anxiety disorders for various temperament profiles. RESULTS Four temperament clusters were found: 1) Novelty seekers with highest Novelty Seeking scores (n = 56),2) Persistent with highest Persistence scores (n = 36), 3) Reserved with lowest Novelty Seeking scores (n = 66) and 4) Wearied with highest Harm avoidance, lowest Reward Dependence and lowest Persistence scores (n = 58). After adjusting for clinical variables, panic disorder and/or agoraphobia were predicted by Novelty seekers' temperament profile with odds ratio [OR] = 3.5 (95% confidence interval [CI] = 1.8 - 6.9, p < 0.001), social anxiety disorder was predicted by Wearied temperament profile with OR = 3.4 (95% CI = 1.6 - 7.5, p = 0.002), and generalized anxiety disorder was predicted by Reserved temperament profile with OR = 2.6 (95% CI = 1.2 - 5.3, p = 0.01). LIMITATIONS The patients' temperament profiles were assessed while displaying depressive symptoms, which may have affected results. CONCLUSIONS Temperament clusters with unique dimensional profiles were specifically associated with different anxiety disorders in this study. These results suggest that TCI-R could offer a valuable dimensional method for predicting the risk of anxiety disorders in diverse depressed patients.