Depressive symptoms in elderly women with chronic conditions: measurement issues.

OBJECTIVE: Depression is highly prevalent in elderly women with chronic conditions and measuring depressive symptoms is complicated by co-occurring chronic illnesses. The 20-item Center for Epidemiological Studies Depression (CES-D) Scale is commonly used with elders, but its length may result in missing data. Twelve short four to 16 item CES-D forms exist, but their psychometric properties have not been systematically evaluated.
METHOD: This study of 250 elderly women compared reliability and validity estimates for the 20-item CES-D and its 12 short forms; all scales used four response alternatives. The study also investigated whether women with varying numbers of chronic conditions would be similarly classified as depressed based on standardized cut scores.
RESULTS: Cronbach's alpha was 0.84 for the 20-item CES-D and alphas ranged from 0.53 to 0.83 for the short forms. Correlations between the 20-item CES-D and the short forms ranged from 0.77 to 0.96. Using the established 20-item CES-D cut score, 12% of the elderly women would be classified with clinically significant depressive symptoms. The shortest scales (four and five item) identified the greatest percentages of women as depressed while the 16-item scale identified the lowest. All 12 short forms showed a trend where more women were classified as depressed as their number of chronic conditions increased.
CONCLUSION: The findings indicate a need for cautious interpretation of findings when shorter forms of the CES-D are used, particularly since shorter forms are less reliable and appear to over-identify women with chronic conditions as having clinically relevant depressive symptoms.