Objectives. The psychometric properties of the Depression Anxiety Stress Scales (DASS; Lovibond & Lovibond, 1995a) were examined in depressed psychiatric hospital samples. Design and Methods. Three studies administered the DASS and other symptom measures at admission and discharge to consecutive adult hospital patients with a primary diagnosis of depression. Study 3 aimed to address problems with the DASS by extending the response options. Results. Study I found that the DASS had good reliability and validity, was moderately sensitive to change, but the Depression Scale exhibited a ceiling effect. In Study 2, confirmatory factor analysis supported a three-factor structure and the DASS continued to demonstrate good psychometric properties, but the ceiling effect was replicated. Study 3 found that by extending the response scale to include an additional option, the factor structure of the instrument as a whole was maintained, the sensitivity to treatment was increased, but the ceiling effect was only marginally reduced. Conclusions. The psychometric properties of the DASS were sound in clinically depressed samples, but the Depression Scale exhibited a ceiling effect that could not be resolved with minor changes to the scale. Suggestions for revisions of the DASS are made.