Hindi translation, Cultural adaptation and Validation of Eating Disorder Diagnostic Scale: A study from Bariatric Clinics, Maulana Azad Medical College & Lok Nayak Hospital, New Delhi, in North India

AIM: To translate, adapt and validate Eating Disorder Diagnostic Screen (EDDS) in Hindi. METHODOLOGY: World Health Organization (WHO) process of translation and adaptation of instruments was used. Hindi translation of EDDS was done by three psychiatrists, two of whom were familiar with western culture. All of them were literate in English, their mother tongue being Hindi. The bilingual panel of experts, including three surgeons, one Professor in English language, one morbidly-obese person [with International English Language Testing System (IELTS) score Band 8.5, familiar with both western and Indian cultures] together identified and resolved inadequate expressions of translations or discrepancies. Back translation to the English language was done by an independent translator, whose mother tongue was English. Pre-testing, focussed-group-sessions/cognitive-interviewing was done in morbidly-obese patients (n=55). Hindi-version so-developed was administered on students (n=120) after a week. Test-retest reliability was assessed. The original EDDS, Modified Weight Bias Internalization Scale (WBIS-M), and Fat Phobia Scale-Short Form (FPS-SF) were applied on 175 participants [students (n=120), patients (n=55)]. Psychometric properties of Hindi-EDDS were investigated. RESULTS: The mean age of students and patients were 18.60 and 43.38 years, respectively. The mean Body-Mass Index (BMI) of students and patients was 21.77 and 43.21Kg/m(2), respectively. Kaiser-Meyer-Olkin test and Bartlett’s Test of Sphericity for sample adequacy was 0.859 and χ2=3155.775; p=0.000, respectively. Spearman-Brown and Guttman Split-Half coefficient for Hindi-EDDS was 0.755 and 0.707, respectively. Internal consistency (Cronbach’s alpha and Kuder–Richardson 20) was α=0.898. Cross-language concordance had significant intra-class correlation coefficient (0.725-1.000). WBIS-M and FPS-SF showed a significant correlation with various items of Hindi-EDDS. DISCUSSION: A culturally appropriate tool to screen and diagnose EDs is lacking in India. The Hindi-version of EDDS is culturally sensitive and has acceptable psychometric properties on the basis of standard tests. CONCLUSION: Hindi-EDDS may be used to estimate various epidemiological parameters of EDs in India.