CGMs without diabetes: what the research shows so far

In a 2026 analysis of 3,634 adults without diabetes, the share of time spent in a tight glucose range showed moderate discrimination for predicted 10-year cardiovascular risk (AUC 0.75). The authors also note that longer-term outcome data are still needed. Consumer glucose monitors, in other words, are ahead of the evidence for people without diabetes.

What does continuous glucose monitoring show in people without diabetes?

The study, published in Nature Communications in 2026, drew on three PREDICT studies of adults without diabetes or prediabetes (n = 616, 661 and 2,357). Participants had a mean age of 46 and 83% were women.

The main findings:

• Median time spent between 3.9 and 5.6 mmol/L was 73.3%, and time between 3.9 and 7.8 mmol/L was 95.3%.
• Time in the tighter range (3.9 to 5.6) gave moderate discrimination for predicted 10-year atherosclerotic cardiovascular disease risk (AUC 0.75).
• More time in the tighter range was associated with lower HbA1c, lower oral glucose tolerance test glucose, lower carbohydrate intake and higher protein intake. Sleep duration was inversely correlated with mean glucose.

Note what “predicted risk” means here: a calculated 10-year estimate from standard risk factors, not people who actually went on to have heart attacks. The authors conclude that longer-term health outcomes are required to show whether CGM monitoring has utility for health management.

Do glucose monitors improve health in people without diabetes?

That has not been shown. Researchers at Johns Hopkins Bloomberg School of Public Health, writing in 2026, describe the evidence as scant and say no major clinical trials have demonstrated benefits such as weight loss or disease prevention in this group.

Two quotes capture the debate. Dr Elizabeth Selvin says regular lab tests such as glucose and HbA1c remain the best way to understand risk, and that monitoring too closely can cause unnecessary anxiety about occasional high readings. Dr Michael Fang notes that, in people without diabetes, “we don’t really know how to act on differing glucose patterns,” although some small trials show short-term benefits.

Interest is rising regardless: Glimpse’s trend data show search for continuous glucose monitors up 26% year over year as of January 2026. And the PREDICT data suggest glucose patterns carry signal.

The gap is interpretation. A single spike after a meal is normal physiology. A trend across weeks, set beside sleep, diet and lab results, is a different kind of information.

The bottom line

• In 3,634 adults without diabetes, time between 3.9 and 5.6 mmol/L had moderate discrimination for predicted 10-year cardiovascular risk (AUC 0.75).
• That is a predicted-risk association, not proof that CGM use changes outcomes; longer-term data are needed.
• Johns Hopkins researchers say no major trials show health benefits from CGM in people without diabetes, and warn of anxiety over normal spikes.