Validity and reliability of the Persian version of the Cornell Scale for Depression in Dementia

Abstract

The current study aimed to evaluate the validity and reliability of the Persian version of the Cornell Scale for Depression in Dementia (CSDD) in 100 patients aged $\ge$60 with dementia in Tehran, Iran. Test–retest reliability and internal consistency were analyzed through intraclass correlation coefficient and Cronbach’s alpha tests, respectively. The validity was assessed in the areas of face, content and structural validity. The Geriatric Depression Scale (GDS) and The Montgomery–Åsberg Depression Rating Scale (MADRS) were also used to determine the convergent validity. The intraclass correlation coefficient (ICC) was equal to or more than 0.8 between the results of determining depression twice via the CSDD. The internal consistency (Cronbach’s α = 0.90) was excellent. The factor analysis, identified five main factors (mood-related signs, behavioral disturbance, physical signs, cyclic functions and ideational disturbance), aside from depression, demonstrated the variance to be 37.88%. Evaluating convergent validity showed satisfactory results (p < 0001). Using the MADRS as standard. the best cut-points for the CSDD via the use of the Youden index were scores of 5.5, 10.5 and 16.5 for mild, moderate and severe depression, respectively. The Persian version of the CSDD is sufficient in terms of validity and reliability for screening depression in the elderly with dementia in Iran.

Keywords:

• Dementia
• depression
• validity
• reliability
• Iran

Acknowledgement

We would like to express our gratitude to Professor George S. Alexopoulos for his kind help in providing the original questionnaire and the translation process. We would also like to thank Ms. Bahar Darabi for her help in preparing the English version of the text.

Disclosure statement

No potential competing interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available on request from the corresponding author.