Measurement of depression and anxiety for hospitalized depressed patients

Standard scales for measuring anxiety and depression did not distinguish between these two psychological constructs in a sample of 295 inpatients with major depressive disorder. Items from these scales were used to form new measures, based on the results of a factor analysis. The new depression and anxiety subscales were internally consistent and only moderately correlated with one another, compared with the standard measures, which were highly correlated. When the factorial procedure was repeated with a subsample of patients with only mild to moderate symptoms, there was no discrimination between depression and anxiety. This finding suggests that when measured in a sample with a restricted range of symptom severity, anxiety and depression have poor discriminant validity.