The Effect of Hunger on the Likelihood of Glucose Excursions in Adults with Overweight or Obesity: Continuous Glucose Monitoring and Ecological Momentary Assessment Observational Study.

Background: Maintaining stable glucose levels is important for metabolic health. Glucose excursions (GEs), which are marked increases in glucose following food intake, have been associated with a higher risk for cardiovascular disease and metabolic dysfunction. Individuals with overweight or obesity who do not have diabetes may still show impaired glucose regulation, as reflected in increased glucose variability. Hunger, as a real-time physiological cue, may be associated with subsequent glucose changes and represents a potential target for just-in-time adaptive interventions. Objective: This study aimed to investigate the temporal relationship between self-reported hunger and subsequent glucose dynamics, including the likelihood of GE occurring, in adults with overweight or obesity using continuous glucose monitoring and ecological momentary assessment. Methods: Data from 84 nondiabetic adults with overweight or obesity (mean age 47.7, SD 11.5 years) were analyzed over a 14-day period. Participants were recruited through community-based methods and enrolled between October 2022 and April 2023. Ecological momentary assessment reports of hunger were collected 6 times daily via smartphone prompts, and interstitial glucose levels were captured via continuous glucose monitoring. Three linear mixed effects models were applied to examine (1) the association between hunger and immediate glucose level, (2) glucose level 30 minutes after hunger, and (3) glucose trajectories from 30 to 180 minutes after hunger. In addition, a logistic mixed effects model was used to estimate the likelihood of a GE occurring within 30 to 180 minutes following a hunger report. Results: Hunger reports were associated with significantly lower glucose levels both at the time of reporting (mean difference [MD] 13.6 mg/dL, 95% CI -15.9 to -11.3; P<.001) and 30 minutes afterward (MD -11.0 mg/dL, 95% CI -13.4 to -8.6; P<.001). A significant time-by-hunger interaction revealed a progressive increase in glucose levels over 180 minutes after hunger (MD +6.7 mg/dL at 180 minutes, 95% CI +5.1 to +8.3; P<.001). Hunger was associated with increased odds of experiencing a GE by 23.7% at 30 to 60 minutes (odds ratio [OR] 1.24; 95% CI 1.07 to 2.38; P=.02), 36.1% at 60 to 90 minutes (OR 1.36; 95% CI 1.61 to 2.40; P<.001), and 13% at 90 to 120 minutes (OR 1.13; 95% CI 1.05 to 1.63; P=.01). No significant GE risk was observed beyond 120 minutes. Conclusions: Hunger was associated with an initial decrease in glucose followed by a delayed rise and heightened risk of GE within 30 to 120 minutes. These findings suggest that hunger may be a real-time behavioral signal associated with subsequent glucose changes. Although these findings should be interpreted with caution given the absence of direct measures of eating behavior, they highlight the potential relevance of hunger for future behavioral and digital health interventions.