This study will test whether glucose sensor data can be used to identify the time of day when adults with prediabetes are most likely to have high blood sugar after meals. Participants will first wear a continuous glucose monitor and wrist activity monitor and record meal times for 10 days. These data will be used to classify each participant's personal "glycemic vulnerability window," such as morning, evening, or generally variable patterns.

Participants will then be randomly assigned to either personalized meal timing plus a short walk after their most vulnerable meal, or to an attention-matched control group receiving sleep hygiene and general step-count advice. The main outcome will be the change in post-meal glucose exposure during each participant's vulnerable window after 4 weeks.

Prediabetes is a high-risk metabolic state in which postprandial glucose excursions and glycemic variability may contribute to progression toward type 2 diabetes. Although lifestyle modification can reduce diabetes risk, conventional advice is usually generic and does not account for individual differences in the timing of glucose intolerance across the day. Emerging evidence suggests that circadian biology, meal timing, sleep timing, and postprandial activity may influence glucose regulation, but it remains unclear whether continuous glucose monitoring can be used to personalize the timing of meals and brief activity in adults with prediabetes.

CLOCK-PRIME is a single-center, two-arm, randomized controlled trial in adults with prediabetes. Participants will undergo a 10-day blinded observational run-in period using continuous glucose monitoring, wrist actigraphy, and timestamped meal-photo logging. Run-in data will be used to classify participants into pre-specified circadian glycemic phenotypes based on postprandial glucose incremental area under the curve during morning and evening windows. Participants will be categorized as morning-vulnerable, evening-vulnerable, or globally variable.

After phenotype classification, participants will be randomized to either a phenotype-guided intervention or an attention-matched active control group. The intervention group will receive personalized guidance to shift the highest glycemic-load meal away from the participant's highest-vulnerability window and toward the lowest-vulnerability window. Participants will also be advised to perform a 10-minute brisk walk within 30 minutes after the meal occurring in their highest-vulnerability window. The control group will receive standardized sleep hygiene advice and a general step-count goal, without meal-timing or postprandial walking instructions.

The primary endpoint is the change from baseline to week 4 in vulnerable-window postprandial glucose incremental area under the curve measured by continuous glucose monitoring. Secondary endpoints include time in range, time above range, glucose coefficient of variation, mean postprandial peak glucose, nocturnal mean glucose, sleep regularity, social jetlag, and actigraphy-derived activity patterns. Exploratory mechanistic outcomes include fasting dried-blood-spot measures of cortisol:insulin ratio and selected primary bile acids to assess whether changes in neuroendocrine or enterohepatic metabolic pathways accompany improvement in glycemic vulnerability.

The study is designed to determine whether CGM-derived circadian glycemic vulnerability windows can support a feasible precision lifestyle strategy for reducing postprandial glycemic burden in adults with prediabetes.

Inclusion Criteria:

• Age 30 to 70 years
• Prediabetes, defined as either:
• HbA1c 5.7% to 6.4% within 3 months of screening, or
• Fasting plasma glucose 100 to 125 mg/dL on two separate occasions
• Body mass index 23 to 40 kg/m²
• Owns a smartphone compatible with study applications
• Willing to wear a continuous glucose monitor and wrist activity monitor during the study period
• Willing to record meals using timestamped meal-photo logging
• Able to provide written informed consent

Exclusion Criteria:

• Current or prior diagnosis of type 1 diabetes or type 2 diabetes
• Use of glucose-lowering medication within the past 3 months
• Use of systemic corticosteroid medication within the past 3 months
• Use of prescription weight-loss medication within the past 3 months
• Current shift work
• Transmeridian travel across more than 2 time zones within 4 weeks before enrollment
• Known untreated or unstable sleep disorder, including obstructive sleep apnea, narcolepsy, or insomnia disorder
• Pregnancy, planned pregnancy, or breastfeeding
• Gastrointestinal disease or surgery likely to affect nutrient absorption
• Current participation in a structured dietary or exercise intervention program
• Estimated glomerular filtration rate less than 60 mL/min/1.73 m²
• Inability or unwillingness to comply with continuous glucose monitoring, wrist actigraphy, meal logging, or study visits

De-identified individual participant data underlying the published results may be shared upon reasonable request after publication, subject to institutional approval, data-use agreement, and protection of participant confidentiality.

Participants and intervention staff cannot be masked because of the behavioral nature of the intervention. Outcome assessors responsible for CGM data extraction, dried-blood-spot biomarker analysis, and endpoint computation will remain masked to group allocation. The trial statistician will analyze the primary endpoint using coded treatment groups until the statistical analysis plan is finalized.