Association between dietary intake of foods estimated to contain live microbes and health indicators in Australian adults: An exploratory analysis

• Live microbes (LM) are naturally present in some commonly consumed foods • LM categories were assigned to >200 food codes of the AUSNUT-database • Frequent intake of foods with higher estimated LM content were linked to lower chronic disease risk factor levels • More research is needed on estimated live dietary microbe intake and health To develop a database for estimating and categorising live microbe (LM) content in common foods and beverages, compatible with data from the Australian Eating Survey® food frequency questionnaire (AES®-FFQ). A secondary aim was to apply this data to examine the relationship between estimated live dietary microbe intake (derived from self-reported LM-containing food consumption), cardiometabolic health indicators, and inflammatory markers. We hypothesised that greater consumption of foods with higher estimated live microbe content would be associated with better cardiometabolic health in adults. Foods were categorised by potential viable microorganism concentration based on published methods: low (10 7 CFU/g). These LM categories were mapped to a subset of AUSNUT food codes corresponding to AES®-FFQ items. Using cross-sectional baseline data from healthy adults, estimated live dietary microbe intakes were calculated, and relationships with demographic, anthropometric, and clinical chemistry data were analysed using adjusted weighted-least squares regression. Foods were assigned to LM categories for over 200 foods from the AUSNUT database corresponding with AES®-FFQ food items. Dietary data from 58 adults were analysed to assess associations between estimated live dietary microbe intake and health outcomes. Higher self-reported dietary intake of foods with estimated medium-to-high live dietary microbe content was inversely associated with body weight ( p =0.008), BMI ( p =0.005), waist circumference ( p