OBJECTIVE:

To establish the reliability and validity of a 10-item short form of the CES-D (CESD-10).

DESIGN:

Cross-sectional and longitudinal follow-up over 3 years. Standardized questionnaire and face-to-face interview were conducted.

SETTING:

Community and psychogeriatric assessment clinic of a voluntary organization.

PARTICIPANTS:

Five hundred and fifty-four elderly in the general community, 30 elderly from a community centre and 31 elderly patients with depressive symptoms.

MEASURES:

Instruments other than the CESD-10 included Activities of Daily Living (ADL), the Life Satisfaction Scale (LSS), the Lubben Social Network Scale, a single-item measure of self-rated health.

RESULTS:

Reliability of the CESD-10 in terms of internal consistency was satisfactory (Cronbach alpha = 0.78-0.79). Moderate consistency over a period of 3 years was also found to be significant (r = 0.44, p < 0.01). The CESD-10 showed comparable accuracy to the original CES-D in classifying cases with depressive symptoms (kappa = 0.84, p < 0.01). Significant relationships of the CESD-10 with impairments of daily functioning, life satisfaction, social support, and self-rated health were established. The CESD-10 also differentiated significantly between groups of 'normal' and clinically depressed elderly.

CONCLUSION:

The CESD-10 attained satisfactory content and temporal reliability. Its construct and concurrent validity were established. With its brevity, it should prove a useful mental health measure for the elderly.