Summary: In 30 adults with prediabetes and overweight or obesity, adding real-time continuous glucose monitoring (CGM) feedback to individualized nutrition therapy increased whole-grain (p=0.02) and plant-based protein (p=0.02) intake and improved sleep efficiency by about 5% (p=0.02) versus a CGM-blinded control arm, in a small open-label pilot randomised trial. Fruit intake and carbohydrate reduction showed only non-significant trends, and no glycaemic or weight outcomes were reported.
PICO Summary
| Element | Detail |
|---|---|
| Population | 30 adults with prediabetes and overweight or obesity; single-centre randomised controlled trial, United States. |
| Intervention | Individualized nutrition therapy plus real-time CGM feedback (n=15); both arms received dietitian-led nutrition recommendations. |
| Comparison | Same individualized nutrition therapy with CGM worn but blinded to glucose data (n=15). |
| Outcome | Whole-grain intake increased (p=0.02) and plant-based protein increased (p=0.02) in the CGM-feedback arm. Sleep efficiency improved by approximately 5% (p=0.02). Fruit intake (p=0.07) and percentage of calories from carbohydrate (p=0.08) showed non-significant trends. No 95% CIs, ARR, or NNT were reported; no glycaemic or weight outcomes were presented in the abstract. |
Expert Commentary
This is a small, single-centre, exploratory randomised trial, and its findings should be read as hypothesis-generating rather than as established efficacy. With only 15 participants per arm, the study is best understood as a feasibility and signal-detection exercise: the significant increases in whole-grain and plant-based protein intake and the roughly 5% gain in sleep efficiency are plausible, but they rest on multiple dietary comparisons in a very small sample, so the risk of chance findings is real and effect sizes are likely to be unstable. The trial is unavoidably open-label, because participants cannot be blinded to their own real-time glucose data, which leaves behaviour change open to expectation and engagement effects. No glycaemic, weight, or adverse-event outcomes are reported in the abstract, so any benefit on the metrics that ultimately define diabetes prevention remains untested here. The dominant limitation is sample size: confidence intervals were not reported, and the analysis cannot exclude clinically trivial differences. Can I use this with my patients? Not yet as a reason to add CGM to dietary counselling for prediabetes outside of research or shared-decision pilots, though it is reasonable to keep the question open for motivated patients. Larger, longer trials reporting glycaemic and weight endpoints with confidence intervals are needed before this approach is recommended.
References
Basiri R, Rajanala Y. Effects of Individualized Nutrition Therapy and Continuous Glucose Monitoring on Dietary and Sleep Quality in Individuals with Prediabetes and Overweight or Obesity. Nutrients. 2025;17(9):1507. doi:10.3390/nu17091507
