Postprandial Glucose Level Decreases and Appetite in Adults Without Diabetes.

Biomarkers that can predict appetite and be monitored in daily life could guide appetite and weight management. Glucose level declines induce hunger in laboratory settings, but free-living data are limited. To characterize postprandial glucose level decreases (PGDs) 2 to 3 hours following free-choice meals and estimate the association of PGDs with appetite. This cohort study, conducted from May 2021 to August 2024, used intensive longitudinal data from smartphone app-based ecologic momentary assessments (EMA) and continuous glucose monitoring (CGM) over 9 days among Singaporean adults without diabetes. Data were analyzed between November 2024 and August 2025. Two CGM-derived PGD measures per meal: PGD magnitude (percentage of glucose level reduction 2 to 3 hours after meals compared with the premeal baseline level) and PGD below the baseline level (binary indicator coded as yes when PGD magnitude was >0). The main outcomes were 6 EMA-derived appetite measures, including hunger level and hunger increase 2 to 3 hours and 3 to 4 hours after meals, time to the next meal, and time to the next meal or snack. Multivariable generalized estimating equation models estimated the adjusted association for each pair of PGD exposure (PGD magnitude, PGD below the baseline level) and the appetite measure. The analysis included 7650 meals from 895 participants (mean [SD] age, 40.1 [13.6] years; 561 [63%] female). Median time to the next meal was 380 (IQR, 285-660) minutes when PGD was below the baseline level (4121 meals [54%]) compared with 425 (IQR, 325-714) minutes when PGD below the baseline level was absent. Larger PGD magnitudes were associated with greater postprandial hunger levels at 2 to 3 hours (β = 0.05 [95% CI, 0.03-0.07] per 10% decrease) and 3 to 4 hours (β = 0.09 [95% CI, 0.06-0.13] per 10% decrease), greater hunger increases at 2 to 3 hours (β = 0.08 [95% CI, 0.04-0.12] per 10% decrease) and 3 to 4 hours (β = 0.07 [95% CI, 0.02-0.12] per 10% decrease), and shorter time to the next meal (β = -6.30 [95% CI, -8.88 to -3.72] minutes) and the next meal or snack (β = -6.54 [95% CI, -8.88 to -3.72] minutes). Similarly, PGD below the baseline level was associated with all appetite measures, including a shorter time to the next meal (β = -27.30 [95% CI, -36.90 to -17.71] minutes). In this cohort study of adults without diabetes, PGD was associated with greater perceived hunger and earlier subsequent eating. Our findings support PGD as a potential biomarker for appetite regulation and target for weight management.