Reliability and validity of the Armenian version of the Hamilton Depression Rating Scale

Objective: The aims of the study were to investigate the reliability and validity of the Armenian version of the Hamilton Depression Rating Scales (HDRS). Method: Firstly, a pilot study was carried out. The sample of the pilot study; 20 patients who applied to the psychiatric clinic of the hospital with depressive complaints. Sociodemographic data form and 2 expert independently scored the HDRS were applied to all participants in the study. To test the reliability of the pilot study; Cronbach alpha (.80), split-half reliability (.88) and item total correlations (for expert1 .34 to .89; for expert2 .36 to .88) was calculated for the 17-item version of the HDRS. These values show the HDRS had a high reliability consistency. Inter-rater reliability was also high, with a Pearson's correlation coefficient of %99 for the HDRS and Kappa of .82. After yhe pilot study the main study was carried out. The sample of the research; 82 patients who applied to the psychiatric clinic of the hospital with depressive complaints and were evaluated by the psychiatrist and diagnosed with depression and 52 of the hospital staff as healthy controls. Sociodemographic data form, CES-D self-report scale and 2 expert independently scored the HDRS were applied to all participants in the study. Results: Examined the correlations between demographic variables (gender, marital status, education, Income rate, suicide attempt, family history of psychiatric illness etc.) with HDRS scores of depressive group. Females had significantly higher scores on HDRS. Convergent validity was assessed by examining the correlations of the total score of HDRS with CES-D total scores. The HDRS and CES-D scores had positive and statistically significant correlations (.81). Next, discriminant validity of the scale was assessed using t-tests and HDRS scores of the depressive group was found to be statistically higher than the healthy control group (p<0,05). The HDRS was shown to successfully discriminate between depressive and non-depressive groups. To test the reliability of the scale; Cronbach alpha (.97), split-half reliability (.96) and item total correlations (for expert1 .48 to .92; for expert2 .46 to .92) was calculated for the 17-item version of the HDRS. These values show the HDRS had a high reliability consistency. Inter rater reliability was also high, with a Pearson's correlation coefficient of %99 for the HDRS and Kappa of .95. ROC analysis revealed a cut-off score of> 6. Conculusion: The study provides initial support for the reliability and validity of The Armenian version of Hamilton Depression Rating Scale. Key words: Hamilton Depression Rating Scale, reliability, validity, depression, mental health.