Abstract:
Objective: To evaluate the impact of continuous glucose monitoring (CGM)-guided behavioral interventions
on glycemic control, metabolic parameters, and behavioral outcomes in individuals with prediabetes and
obesity.
Design: Systematic review and meta-analysis.
Data sources: PubMed, Embase, Cochrane Library, Web of Science, and ClinicalTrials.gov from inception to December 2025.
Eligibility criteria: Randomized controlled trials, pilot studies, and observational studies examining CGM-integrated nutritional and lifestyle interventions in adults and adolescents with prediabetes, intermediate hyperglycemia, impaired glucose tolerance, or obesity. Studies combining CGM with individualized nutrition therapy, time-restricted eating, digital health platforms, wearable devices, or exercise interventions were included.
Data extraction and synthesis: Two independent reviewers screened studies and extracted data on glycemic metrics (time in range, glycemic variability, mean glucose, HbA1c), body weight, body composition, dietary quality, eating patterns, sleep parameters, and behavioral adherence. Risk of bias was assessed using the Cochrane Risk of Bias 2 tool for randomized trials and ROBINS-I for non-randomized studies. Meta-analysis used random-effects models with heterogeneity assessed through I² statistics.
Results: Real-time CGM feedback integrated with individualized nutrition therapy significantly improved glycemic control in individuals with prediabetes and overweight/obesity. Time in range increased significantly (p=0.02), while mean glucose, glucose management indicator, and coefficient of variation decreased (p=0.01). Digital health programs incorporating CGM and wearable sensors (n=2217) demonstrated substantial reductions in hyperglycemia and glycemic variability, with particularly pronounced effects in non-diabetic individuals. Weight reduction occurred across all metabolic categories, especially in overweight/obese participants. CGM-enhanced interventions facilitated meaningful dietary modifications, including significantly increased whole grain consumption (p=0.02) and plant-based protein intake (p=0.02). Non-glycemic benefits included 5% improvement in sleep efficiency (p=0.02). Scanning frequency positively correlated with increased protein intake, suggesting heightened awareness of glycemic fluctuations drives behavioral change. Pilot studies demonstrated feasibility of time-limited eating guided by CGM in adolescents with obesity, showing promise for body composition and glycemic pattern improvement. Methodological advances combining CGM with wrist motion sensors significantly improved objective eating occasion detection compared with self-report alone. Meta-analysis revealed heterogeneity based on intervention intensity and CGM technology type. Real-time CGM with personalized coaching demonstrated superior outcomes compared with intermittent scanning alone. Risk of bias assessment showed most included randomised trials had low to moderate risk, with primary concerns being performance bias and attrition in longer term studies.
Conclusions: CGM-guided behavioral interventions represent an effective strategy for improving glycemic control and reducing type 2 diabetes progression risk in individuals with prediabetes and obesity. Real-time feedback facilitates meaningful behavioral changes in diet quality, eating timing, and lifestyle patterns. Integration of CGM with complementary wearable technologies enhances intervention precision. Evidence supports implementation of CGM-guided interventions as a preventive strategy in high-risk populations, though larger, longer term trials with standardised protocols are needed. Future research should address optimal intervention duration and intensity, cost effectiveness, health equity in access, visceral adiposity reduction, and policy frameworks for population-level implementation.
Keywords: Continuous glucose monitoring; Type 2 Diabetes; Prediabetes; Obesity; Behavioral intervention
