Eating disorders: lessons from a cross-cultural study.

Five hundred and eighty schoolgirls living in a small North Indian industrial town were screened using the EAT-26. Although twenty-nine per cent scored above the threshold for the questionnaire (greater than or equal to 20), closer scrutiny of the pattern of responses revealed misinterpretations of a large number of questions possibly on a conceptual basis. Linguistic and conceptual pitfalls inherent to cross-cultural epidemiological research are discussed with reference to previous work.