P776 - IS ERRONEOUS SELF-REPORTED ENERGY INTAKE LESS COMMON IN VEGETARIANS AND VEGANS? INSIGHTS FROM THREE CROSS-SECTIONAL STUDIES

P776

IS ERRONEOUS SELF-REPORTED ENERGY INTAKE LESS COMMON IN VEGETARIANS AND VEGANS? INSIGHTS FROM THREE CROSS-SECTIONAL STUDIES

M. A. Storz^1,*, J. Herter¹, J. S. Dürrschnabel ¹, L. Hannibal², R. Huber¹

¹Centre for Complementary Medicine, ²Laboratory of Clinical Biochemistry and Metabolism, Department of General Pediatrics, Adolescent Medicine and Neonatology, Medical Center - University of Freiburg, Freiburg, Germany

Rationale: The problem of self-reported erroneous energy intake (EI) is omnipresent in the dietetic literature, leading to false associations between dietary intakes and health outcomes. Vegetarians and vegans have been characterized as particularly health conscious, emphasizing dietary identity and a strict eating behavior; thus, we hypothesized that both are less likely than omnivores to erroneously underreport EI.

Methods: Based on data from 3 cross-sectional studies, we explored the association between vegetarian diet status and EI underreporting. We pooled data from two German studies (the Freiburg-METABO study & the LAKE study), which used a convenience sampling technique. In addition, we analyzed National Health and Nutrition Examination Surveys (NHANES) data based on a complex, stratified, multistage probability cluster sampling design. Crude and multivariate logistic regression models were constructed. Under-reporter status was assessed using a predictive equation based on doubly labelled water measurements.

Results: The analysis of both German studies (n=206 participants) revealed no association between vegetarian status and underreporting. The number of EI under-reporters was <10% in all diet groups (omnivores, vegetarians, vegans) and the lowest proportion was found in vegans. In the NHANES (n=8,445 observations), an association between vegetarian status and underreporting was found only after adjusting for BMI – a potential mediator in this context. After adjustment, semi-vegetarianism and lacto-ovo-vegetarianism were both associated with significantly increased odds for underreporting (OR: 1.91 (1.05-3.47) and 1.77 (1.03-3.43), with p-values of 0.036 and 0.038, respectively).

Conclusion: We found no association between vegetarian status and EI under-reporting, which has important implications for the ongoing debate on dietary intake and nutrient adequacy in vegetarian diets.

Disclosure of Interest: None declared