Reliability, validity and factor structure of the GDS-15

Objectives Depressive disorders are a public health problem even in developing countries. Access to valid and reliable screening instruments is needed for conducting community surveys. The main objective of this study is to provide the Iranian version of the Geriatric Depression Scale-15 (GDS). Methods The GDS-15 Farsi version was developed by translation and back translation. Two hundred and four subjects aged 59 years or older, who were chosen randomly from residents of the Ekbatan district of Tehran, the capital city of Iran, completed the GDS-15. The Composite International Diagnostic Interview (CIDI) was used to establish a gold standard diagnosis of major depressive disorders. Results The GDS was found to be an internally consistent measure. Alpha, split-half coefficients and test–retest reliability were 0.9, 0.89 and 0.58 respectively. Two factors were extracted by using factor analysis and the principle component analysis (varimax rotation): ‘depression ’ and ‘psychosocial activity’. The Depression factor (omitting items 2, 9, 10, 13), which could be considered as a short form of the scale (a 0.92), has significant correlation with the main scale (r 0.58). Using receiver operating curve (ROC) analysis, the optimum cutoff score for GDS-15 is 7/8, yielding a sensitivity of 0.9 and a specificity of 0.84. The optimum cutoff score for GDS-11 is 6, yielding a sensitivity of 0.9 and a specificity of 0.83. Conclusion The long and short forms of the GDS have excellent properties as screening instruments for major depression in older dwellers in Iran, particularly in urban areas, as presented in our findings. Copyright # 2006 JohnWiley & Sons, Ltd. key words—depression; elderly; screening; geriatric depression scale-1