Is the relationship between anxious and depressive cognitions and symptoms linear or curvilinear?

Abstract

Two studies are reported that investigated the relationship between anxious and depressive cognitions and symptoms in clinical and nonclinical adult samples. In the first study, the Cognitions Checklist (CCL), Beck Depression Inventory, Beck Anxiety Inventory, and revised Hamilton Rating Scales of Anxiety and Depression were administered to 1,553 psychiatric outpatients. Hierarchical polynomial regression analyses found only a linear component underlying most of the relationships among cognition and symptom scales. In the second study, the CCL and measures of anxious and depressive symptoms were administered to 50 depressed psychiatric inpatients, 78 chronic medically ill patients, and 25 normal controls. The relationships of cognitions and symptoms differed across samples for both the clinical rating and self-report measures of symptoms. However, the results supported a linear relationship between frequency of negative cognitions and severity of anxious or depressive symptoms in the depressed inpatient, and to a lesser extent, medical samples.