Continuous Glucose Monitors for People Without Diabetes: Does the Evidence Support Them?

The short answer#

If you do not have diabetes and you feel well, there is no strong evidence that a continuous glucose monitor (CGM) improves your health, and no major guideline recommends routine use in healthy people. These sensors track glucose accurately enough, but in a healthy body the readings mostly confirm that your metabolism is working. A CGM can be a short, interesting teaching tool. It is not a diagnostic test, and a normal rise after a meal is not something you need to fix.

Key points#

• The first over-the-counter CGMs (Dexcom Stelo, Abbott Lingo) reached the market in 2024 for adults not using insulin, and they do not alarm for high or low glucose.
• In healthy adults, glucose stays between about 70 and 140 mg/dL roughly 96% of the day, with a mean near 98 to 99 mg/dL.
• A brief rise after eating is normal physiology, not injury; a "spike" on the app is usually your pancreas doing its job.
• No randomized trial has shown that wearing a CGM improves real health outcomes in people without diabetes or prediabetes.
• CGMs are least accurate in the low and low-normal range, which is exactly where false readings and needless worry cluster.
• Out-of-pocket costs run roughly $80 to $250 a month, and insurers rarely cover wellness use.
• The narrow case for value is short-term habit feedback in some people with prediabetes or obesity, best done with a clinician.

What do over-the-counter CGMs actually measure?#

A CGM is a small sensor worn on the arm or abdomen. A tiny filament sits just under the skin and estimates glucose in the interstitial fluid every few minutes, then sends the number to a phone. Because it samples the fluid around your cells rather than blood, the reading lags a finger-stick by several minutes, most noticeably when glucose is moving fast.

Until recently these devices needed a prescription. That changed in March 2024, when the FDA cleared Stelo by Dexcom as the first glucose biosensor sold without one. It is meant for adults 18 and older who do not use insulin, is worn for up to 15 days, and updates roughly every 15 minutes. Abbott's Lingo followed as a general wellness tracker, and Libre Rio for type 2 diabetes managed without insulin. One design detail matters for healthy users: none of these consumer sensors sound an alarm for dangerous highs or lows, and Stelo is explicitly not for people who get problematic hypoglycemia. If you want to understand what a lab test of long-term control actually captures, our guide on what an HbA1c really measures and its blind spots is a useful companion.

Do glucose spikes matter in healthy people?#

Mostly no. The best normative data come from a multicenter study of 153 people without diabetes wearing research-grade sensors. Their average glucose was about 98 to 99 mg/dL (a bit higher, near 104, over age 60), and they spent roughly 96% of the day between 70 and 140 mg/dL. Their glucose still moved: the within-person coefficient of variation averaged about 17%, and 28% of these healthy participants logged at least one reading under 70 mg/dL that meant nothing at all.

That last figure is the heart of the problem. A metabolically healthy person eats, glucose climbs for an hour or so, insulin brings it back down, and the curve on the app looks dramatic even though the physiology is ordinary. Wellness marketing often reframes these normal rises as "spikes" to be flattened. There is no evidence that blunting ordinary post-meal rises in a healthy person prevents disease, and no agreed threshold defines a "bad" spike outside of diabetes.

What does the evidence show for wellness CGMs?#

Thin, and mostly indirect. Endocrinologists who have reviewed the field describe the wellness case as largely hypothetical: the underlying theory is plausible, but the outcome data are missing. Trials in people without diabetes are small and short, and none demonstrate that CGM use lowers the risk of diabetes, heart disease, or any hard endpoint. The clearest signals come from people who already have prediabetes or obesity, and even there the CGM is one part of a coaching-plus-diet program, so the benefit is hard to pin on the device itself.

A compact way to see the gap:

If your real question is risk, a fasting glucose or HbA1c is cheaper, validated, and interpretable. Our explainer on prediabetes, what the number means and what actually helps, walks through the steps that do move risk.

Can a CGM help you lose weight?#

Sometimes, but probably not because of the sensor. In trials where people lost weight while wearing a CGM, they were also getting structured nutrition guidance, and the weight change tracks with the coaching more than the hardware. For a metabolically healthy person, there is no good evidence that a CGM beats familiar advice: eat more fiber and protein, move after meals, sleep enough, and watch refined carbohydrate and alcohol. Any device can spark a few weeks of attention to food. Whether that novelty lasts, or is worth $80 to $250 a month, is the honest question.

What are the potential harms?#

The cost is the obvious one, and it is rarely covered for wellness use. The subtler harms are behavioral. CGMs are least accurate at the bottom of the range, so a healthy person can see a "low" that is really a sensor artifact and start chasing it. Harvard's editors point out the irony: if biologically meaningless dips push you to snack more, you could end up with weight gain and higher risk, the opposite of the goal. Clinicians who study consumer use also flag anxiety and a drift toward rigid, orthorexic eating, where ordinary foods get banned for producing an ordinary curve. For a person already prone to worry about food or health metrics, a screen that flashes numbers all day can make things worse, not better.

Who might reasonably benefit, and how should you seek care?#

A short, purposeful trial can make sense for a few groups: people with prediabetes (especially at the higher end of the HbA1c range), people with obesity working on specific habits, or someone in pregnancy being followed for gestational diabetes. In each case the sensor is a coaching aid used with a clinician for a defined period, not a permanent gadget. If you are simply curious, two weeks can be a fine experiment, as long as you treat the graph as a curiosity rather than a verdict on your health. Before spending on a sensor, ask your clinician for a fasting glucose or HbA1c; those tests are validated for diagnosis, and a wellness CGM is not.

When to seek care#

See a clinician, rather than relying on a wellness sensor, if you notice classic signs of diabetes: unusual thirst, frequent urination, unexplained weight loss, blurred vision, or persistent fatigue. Get evaluated if repeated fasting readings are high, or if you have true hypoglycemia symptoms (shakiness, sweating, confusion, or fainting that eases with eating), which a consumer CGM is not designed to catch. And if tracking glucose is feeding anxiety or tight food rules, that is a reason to take the sensor off and talk to someone.

For decisions about testing, diagnosis, or your own health, talk with a licensed clinician who knows your history.