Psychological Medicine

Original Articles

The Inventory of Depressive Symptomatology (IDS): psychometric properties

A. J. Rusha1 c1, C. M. Gulliona1, M. R. Bascoa1, R. B. Jarretta1 and M. H. Trivedia1

a1 Department of Psychiatry and Mental Health Clinical Research Center, University of Texas Southwestern Medical Center, Dallas, TX., USA

Synopsis

The psychometric properties of the 28- and 30-item versions of the Inventory of Depressive Symptomatology, Clinician-Rated (IDS-C) and Self-Report (IDS-SR) are reported in a total of 434 (28-item) and 337 (30-item) adult out-patients with current major depressive disorder and 118 adult euthymic subjects (15 remitted depressed and 103 normal controls). Cronbach's α ranged from 0·92 to 0·94 for the total sample and from 0·76 to 0·82 for those with current depression.

Item total correlations, as well as several tests of concurrent and discriminant validity are reported. Factor analysis revealed three dimensions (cognitive/mood, anxiety/arousal and vegetative) for each scale. Analysis of sensitivity to change in symptom severity in an open-label trial of fluoxetine (N = 58) showed that the IDS-C and IDS-SR were highly related to the 17-item Hamilton Rating Scale for Depression. Given the more complete item coverage, satisfactory psychometric properties, and high correlations with the above standard ratings, the 30-item IDS-C and IDS-SR can be used to evaluate depressive symptom severity. The availability of similar item content for clinician-rated and self-reported versions allows more direct evaluations of these two perspectives.

Correspondence:

c1 Address for correspondence: Dr A. John Rush, Department of Psychiatry at St. Paul, POB 1. Suite 600, 5969 Harry Hines Boulevard., Dallas, Texas 75235-9101, USA.

Metrics