Underreporting in nutritional surveys

Background: Underreporting has been recognized as one of the problems and constraints of nutritional studies in the last decade. Objective: To review available information regarding underreporting in Iran and the world. Methods: This article is a review of studies in which the problem of underreporting was assessed and evaluated. All papers from 1966 to 2001 were searched by medline using keywords of nutritional survey, underreporting, energy intake, BMI and validity. Several leading studies in this field were searched for and obtained from medline and the related articles were analyzed. Findings: The subjects participating in nutritional studies, inaccurately reported their energy and nutrient intakes. Most epidemiological studies use the “energy intake to basal metabolic rate” ratio (EI/BMR) to determine and evaluate underreporting. In most studies EI/BMR ratio below 1.27 is considered to be responsible for underreporting. The incidence of underreporting is between 20-30%, more prevalent in women. Underreporting increases with age, obesity, increasing levels of illiteracy, low socioeconomic status and smoking. The “Tehran Lipid and Glucose Study” (TLGS) is the only study in Iran to have documented underreporting of 1450 subjects, 32.6% women and 14.7% men aged between 20-50 years and 40.3% women and 17.1% men aged over 50 years. Conclusion: This article highlights the importance of the problem caused by underreporting