Högskolan i Skövde

Numerous epidemiological studies show that processed meat consumption is associated to type 2 diabetes (T2D), but the results are less consistent for unprocessed meat and poultry intake. However, in the studies the meat consumption has been estimated through self-reporting using questionnaires. This method is known to suffer from measurement errors, which may lead to dubious observed diet-disease associations. Dietary biomarkers may be used as more objective instruments of specific meat intake. Therefore, the aim in this study was to (1) identify plasma metabolites associated to total, processed, unprocessed meat and poultry intake, and (2) to investigate their associations with the risk of developing T2D. Furthermore, a correlation analysis was performed to elucidate dietary patterns related to the reported meat intakes and their metabolites. Through multivariate modelling, panels of 22, 22, 15 and 19 metabolites where selected, that mirrors total, processed, unprocessed meat and poultry intake, respectively. Ridge predictive modelling was performed generating metabolite scores reflective of the diverse meat consumptions. These scores and the reported meat intakes were used in conditional logistic regression for the assessment of their association with risk of developing T2D. Both the processed metabolite scores and reported intakes were associated with higher risk of developing T2D, whereas the unprocessed meat, poultry scores and associated intakes were not. In conclusion, assessments of plasma metabolite scores appears promising as objective measurements of dietary meat intakes and could be used as a complement to the questionnaires. Results need to be validated in independent cohorts.