(OR24-04-23) The Day to Day Variability in Postprandial Glucose Dips Predicts Daily Energy Intake: The ZOE PREDICT STUDY

Objectives: Individuals with large postprandial glucose dips have greater hunger and energy intake compared to individuals with small glucose dips. However, the effect of day-to-day variability in dip size within individuals is unknown. This research aims to understand day-to-day variability in glycaemic dips and associations with subsequent dietary intake within an individual.

Methods: Demographic information, daily energy intake (logged diet), hunger and postprandial glycaemic responses (continuous glucose monitors (CGMs)) to standardised and free-living meals were assessed in ZOE PREDICT 1 participants (NCT03479866). Glycaemic dips were calculated as the difference between the lowest glucose concentration in the 2–3 h postprandial period and the average baseline concentration. Participants were classified into 3 dipping categories ( < 5%, 5-10%, ≥10% below baseline) based on their average response across 5 standardised meals. Mixed modelling was performed with person-specific random effects (covariates; sleep, physical activity status, set meal type).

Results: In UK adults (n=654, 71% female, mean age 47±12y), 23% of participants did not dip, 24% dipped < 5%, 28% 5-10% and 23% dipped ≥10% (big dippers). There was a linear relationship between the dip category and both BMI and energy requirements (big vs non dippers; BMI 27 vs 25 kg/m2; basal metabolic rate 1492 vs 1395 kcal). Over the study (14 days), the number of dips per/d were highly variable within an individual (mean intra-individual CV 0.44%, population range 0-13 dips per/d) and meals with higher % of energy from sugar and lower % from protein or fibre were associated with larger dips (p < 0.05 for all). Within an individual, a big dip (≥10%) post breakfast predicted greater daily energy intake (β = 166 kcal (SE 44)) compared to a day where no dip occurred. This association was not replicated for glucose 2hiAUC. In individuals (n=374) simultaneously wearing 2 CGMs there was high reliability between CGMs for capturing glycaemic dip size (r=0.93).

Conclusions: Findings highlight glucose dips as an important feature of the glucose response curve from CGMs in relation to energy intake. Further, the inter- and intra-individual variability in glucose dips as a predictor of energy intake highlights identifying individuals with regular dips as a potential target for personalised dietary advice.

Funding Sources: ZOE Ltd