Carroll Depression Scale

The Carroll Depression Scale is one of three measures of depression developed by Bernard Carroll. The Carroll Depression Scales include the Carroll Depression Scale (CDS), the Carroll Depression Scale Revised, and the Brief Carroll Depression Scale. The CDS is published by Multi-Health Systems (http://www.mhs.com).

The CDS is a self-report inventory that takes 10–20 minutes to complete. It is used to measure depression symptomatology and symptom severity. The CDS is appropriate for adults age 18 years and older. There is no upper age limit for the CDS. The publishers suggest it is particularly useful for older adults and severely depressed people because of the cognitive simplicity of the yes-no response format.

The self-report items were designed to match the content of the Hamilton Depression Rating Scales. The CDS includes 52 items. The Revised CDS adds 9 items that assess subtypes of depression defined in the fourth edition of the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders. The Brief CDS consists of 12 of the 52 CDS items. Items assess central features of depression, such as appetite, energy, crying, and sexual interest.

Field trials for the CDS were conducted at the University of Michigan, Ann Arbor, in the 1970s; the CDS has also been used in trials at Duke University Medical Center. The participant pool from these two settings includes 959 depressed patients and 248 non-depressed people and has been used in a variety of reliability and validity studies. The participant pool is 100% White; other information about it is very limited. Given the test's norm group, it may have limited utility with non-White populations.

Validity data suggest good face, convergent, and discriminant validity. Face validity is demonstrated by a .80 correlation between scores from the CDS and scores from the Hamilton Depression Rating Scales. Convergent validity is demonstrated by moderate to high correlations with the Clinical Global Rating of Depression (.63), the Montgomery-Asberg Depression Rating Scale (.71), the Beck Depression Inventory (.86), and the Center for Epidemiological Studies of Depression Scale (.67). Discriminant validity is shown by the CDS's ability to differentiate depressed from anxious patients. The State-Trait Anxiety Inventory and the CDS correlate at a rate of .26.

Reliability data suggest strong split-half and test-retest reliability. Cronbach's alpha for the CDS is .95. Split-half reliability between odd and even items is .87. The Pearson correlation coefficient is .96. However, test-retest reliability was measured only on those patients whose Hamilton Depression Rating Scales scores did not vary more than two points between administrations. This restriction may have inflated reliability estimates. The authors state they chose to restrict the data in this way in their belief that the CDS is a state rather than a trait measure.