The Tamil version of Eating Attitudes Test-26: Reliability and factor structure among persons with schizophrenia

Abstract

Background: Disordered eating behaviors (DEBs) among persons with schizophrenia (PwS) have been reported widely in the literature, with very few studies in India. Robust tools to assess DEB are needed in the vernacular language to capture symptoms of disordered eating accurately. No such tools are available in the Tamil language. Globally, the Eating Attitudes Test (EAT-26) is widely used to assess DEB among PwS. Aim: This study aimed to translate and study the factor structure and reliability of EAT-26 among Tamil-speaking PwS. Materials and Methods: EAT-26 was translated into Tamil following the Oxford linguistic validation process. Experts evaluated its face validity and content validity. One hundred and fifty PwS, aged between 18 and 65 years, who attended the outpatient department of a psychiatric facility, and consented to participate, completed the Tamil version of EAT-26. Test-retest reliability of EAT-26 was assessed by readministering the tool to 30 PwS after two weeks. Data were analyzed using Stata 16.1. Internal consistency and test-retest reliability were computed using Cronbach's alpha and intraclass coefficients, respectively. The factor structure of EAT-26 was explored using principal component analysis (PCA). Spearman's rho was calculated to understand the correlation between the factors. Results: EAT-26 had an internal consistency of 0.71 and test-retest reliability of 0.896. Factor analysis revealed nine latent factors consisting of 21 of the original 26 items on EAT-26. These 21 items could explain a variance of 63.63%. Conclusions: The Tamil version of the EAT-26 is a reliable tool to assess DEB among Tamil-speaking PwS. It can be used to screen PwS for eating disorder risk.