Background and aims: The influence of diet and sleep and their combined effect on daily glucose control in individuals with gestational diabetes mellitus (GDM) remains unclear. This study investigated real-time associations between eating and sleep patterns with overnight and morning glucose using ecological momentary assessment.

Materials and methods: Thirty-four untreated patients with GDM wore a continuous glucose monitoring at 28 ± 2 weeks of gestation for 6 days; data from the first day were excluded. Five-day average and daily measures of overnight mean glucose, variability (SD), glycemic area under the curve (AUC), and glucose upon waking were calculated. Predictors included daily food diaries, sleep questionnaires, and sensor day. Multivariate stepwise regressions identified key predictors, and ANOVAs assessed temporal differences in glucose outcomes.

Results: Participants were 34 ± 6 years old with a pre-pregnancy BMI of 28 ± 6 kg/m². Greater day-to-day variability in snack intake was associated with increased 5-day average overnight mean glucose and AUC (p ≤ 0.03). Later dinner was associated with greater overnight glucose variability (p ≤ 0.01). Sleep parameters were not associated with glucose outcomes. Glucose measures differed significantly across days (p ≤ 0.03).

Conclusion: Daily variations in overnight and morning glucose should be considered. Irregular snacking and delayed meal timing could disrupt circadian glucose regulation; promoting consistent and earlier evening meals could support improved nocturnal glycemic control in patients with GDM.