Psychometric Properties of the CES-D-10 in a Psychiatric Sample

Thröstur Björgvinsson, Sarah J. Kertz, Joe S. Bigda-Peyton, Katrina L. McCoy, Idan M. Aderka

Research output: Contribution to journalArticlepeer-review


The 10-item Center for the Epidemiological Studies of Depression Short Form (CES-D-10) is a widely used measure to screen for depression in primary care settings. The 10-item measure has demonstrated strong psychometric properties, including predictive accuracy and high correlations with the original 20-item version, in community populations. However, clinical utility and psychometric properties have yet to be assessed in an acutely symptomatic psychiatric population. This study examined the psychometric properties of the CES-D-10 in a sample of 755 patients enrolled in a psychiatric partial hospital program. Participants completed a diagnostic interview and a battery of self-report measures on admission and discharge. Exploratory factor analysis and confirmatory factor analysis suggested that a one-factor structure provided a good fit to the data. High item-total correlations indicated high internal consistency, and the CES-D-10 demonstrated both convergent validity and divergent validity. Previously suggested cutoff scores of 8 and 10 resulted in good sensitivity (.91 and .89, respectively) but poor specificity (.35 and .47). These data suggest that although the CES-D-10 has generally strong psychometric properties in this psychiatric sample, the measure should be primarily used to assess depression symptom severity rather than as a diagnostic screening tool.

Original languageEnglish
Pages (from-to)429-436
Number of pages8
Issue number4
StatePublished - Aug 2013
Externally publishedYes


  • CES-D-10
  • Center for the Epidemiological Studies of Depression-Short Form
  • assessment
  • depression
  • diagnosis

ASJC Scopus subject areas

  • Clinical Psychology
  • Applied Psychology


Dive into the research topics of 'Psychometric Properties of the CES-D-10 in a Psychiatric Sample'. Together they form a unique fingerprint.

Cite this