Evaluation of the Hospital Anxiety and Depression Scale in Non-clinic Turkish Sample Using Item Response Theory Analysis

Abstract

In this study, it was aimed to review the structural validity of the Turkish version of the Hospital Anxiety and Depression Scale in the community sample and to examine its items with Item Response Theory. A total of 965 people, 480 males (49.7%) and 485 females (50.3%), between the ages of 18-64 (mean = 33.82 +/- 13.99 years) participated in the study via the web application. According to the Explanatory Factor Analysis results of the scale, the two-factor structure consisting of Anxiety and Depression sub-scales were observed suitable for its original study structure. It was determined that the scale explains 47.14% of the total variance with two factors. According to the Confirmatory Factor Analysis result, it showed a good fit with all goodness of fit indicators. The correlation coefficient between the sub-scales was calculated as 0.64, and the shared common variance was 0.41. In the whole group, the Cronbach's alpha internal consistency reliability coefficient was determined 0.83 for the Anxiety sub-scale and 0.77 for the Depression sub-scale. Item parameters estimate with the two-parameter logistic Item Response Theory model, and it was observed that items 4 and 7 had low discrimination values and items 2 and 13 had the highest discrimination values. The study has provided additional information on the psychometric properties of the Turkish form scale and its usage Examining the scale with Item Response Theory helps to understand anxiety and depression while providing a methodical contribution to the use of the scale for evaluating psychiatric comorbidities.