CGM without diabetes

What a CGM measures

A continuous glucose monitor (CGM) is a small sensor worn on the skin. A thin wire sits under the skin and measures glucose in the fluid around cells, then sends readings to a phone app.[1] This description comes from a manufacturer's summary of one sensor, so it may not describe every sensor.[1]

What FDA's 510(k) records say

In its clearance letters for K262472 and K233655, FDA says it "determined the device is substantially equivalent (for the indications for use stated in the enclosure) to legally marketed predicate devices".[3][4] A 510(k) summary is submitted by the company that applies for clearance, so its statements are that company's own.[1][4]

Dexcom's Stelo Glucose Biosensor System was cleared by FDA through the 510(k) process. In clearance K262472, dated August 11, 2026, the indications for use read: "The Stelo Glucose Biosensor System is an over-the-counter (OTC) integrated Continuous Glucose Monitor (iCGM) intended to continuously measure, record, analyze, and display glucose values in people 2 years and older not on insulin. The Stelo Glucose Biosensor System helps to detect normal (euglycemic) and low or high (dysglycemic) glucose levels. The Stelo Glucose Biosensor System may also help the user better understand how lifestyle and behavior modification, including diet and exercise, impact glucose excursion."[3]

Dexcom's summary for Stelo clearance K260935, filed with FDA in 2026, says: "The Stelo System is for people who do not use insulin, ages 2 years and older (caregiver supervision required for those under 18). The Stelo System is not for people on dialysis, with problematic hypoglycemia, are modifying medication without HCP consultation, or have a history of eating disorders." HCP stands for healthcare professional.[5]

Abbott's Lingo Glucose System was cleared by FDA through the 510(k) process. In clearance K233655, dated May 29, 2024, the indications for use read: "The Lingo Glucose System is an over-the-counter (OTC) integrated Continuous Glucose Monitor (iCGM) intended to continuously measure, record, analyze and display glucose values in people 18 years and older not on insulin. The Lingo Glucose System helps to detect euglycemic and dysglycemic glucose levels. The Lingo Glucose System may also help the user better understand how lifestyle and behavior modification, including diet and exercise, impact glucose excursion. The user is not intended to take medical action based on the device output without consultation with a qualified healthcare professional."[4]

Neither of these two cleared indications mentions diabetes. Both describe the intended users as people "not on insulin", with a lower age limit of 2 years for Stelo and 18 years for Lingo.

When FDA announced its March 5, 2024 clearance of Stelo, it described the system as "intended for anyone 18 years and older who does not use insulin, such as individuals with diabetes treating their condition with oral medications, or those without diabetes who want to better understand how diet and exercise may impact blood sugar levels." The announcement added: "Importantly, this system is not for individuals with problematic hypoglycemia (low blood sugar) as the system is not designed to alert the user to this potentially dangerous condition."[6]

Dexcom's 2024 summary for the Stelo system, filed with FDA, says its clinical study enrolled adults with diabetes and that the results meet FDA's performance requirements for this type of sensor (the iCGM special controls). Abbott's summary for the Lingo says its accuracy data came from a study of the FreeStyle Libre 2 sensor in people with diabetes. Abbott argues that glucose varies more in people with diabetes, so that study is a worst case for the sensor, and the summary states that system accuracy met the iCGM special controls.[1][4]

Dexcom's 2024 summary for the Stelo system, filed with FDA, says the glucose value shown in the app might be up to 15 minutes old.[1] Dexcom's 2024 summary for the Stelo system, filed with FDA, says its clinical study enrolled adults with diabetes and lists local infection, skin irritation (edema) and pain or discomfort among the device-related adverse events. The summary's list gives no counts.[1]

How far readings can differ

Three small studies that we read of glucose sensors in healthy adults found errors that were not in the same direction. In 44 healthy young men given a glucose drink, a FreeStyle Libre 2 sensor read lower than finger-prick glucose at every time point and differed from it by 12.9 percent on average (95% CI 11.8 to 14.0).[7]

In 29 healthy women after a standard breakfast, the same type of sensor read higher than venous plasma glucose, by 1.14 mmol/L on average, and the average difference was 27.5 percent (95% CI 20.8 to 34.2).[8] In 39 adults aged 18 to 28, an Abbott Lingo sensor read 4.8 mg/dL lower than a finger-prick meter on average, and the average difference was 12.9 percent.[9] The studies differ in sensor, reference method and participants.

Abbott's Lingo Glucose System was cleared by FDA through the 510(k) process. In clearance K233655, dated May 29, 2024, the indications for use read: "The Lingo Glucose System is an over-the-counter (OTC) integrated Continuous Glucose Monitor (iCGM) intended to continuously measure, record, analyze and display glucose values in people 18 years and older not on insulin. The Lingo Glucose System helps to detect euglycemic and dysglycemic glucose levels. The Lingo Glucose System may also help the user better understand how lifestyle and behavior modification, including diet and exercise, impact glucose excursion. The user is not intended to take medical action based on the device output without consultation with a qualified healthcare professional."[4]

The study of a FreeStyle Libre 2 sensor in 44 healthy men was supported by Universität der Bundeswehr München, and its authors report no conflicts of interest.[7]

The study of a FreeStyle Libre 2 sensor in 29 healthy women was supported by the National Institute for Health and Care Research (NIHR) Leicester Biomedical Research Centre, and its authors report no conflict of interest.[8]

In a 2026 study of 39 adults aged 18 to 28 without diabetes, 66.5 percent of the Abbott Lingo sensor's readings fell within 15 percent or 15 mg/dL of the reference readings.[9] The study used a finger-prick meter as the reference, not a laboratory method.[9] The difference from the meter was larger after a sugary drink than after placebo (average difference 15.3 percent versus 10.6 percent).[9] One study of 44 healthy men found a FreeStyle Libre 2 reading lower than finger-prick glucose, and one study of 29 healthy women found it reading higher than venous plasma glucose, so these Lingo figures do not show the direction or size of error for other sensors or other people.[7][8][9]

The 2026 study of the Abbott Lingo sensor in 39 adults was supported by a grant from the National Institute of Diabetes and Digestive and Kidney Diseases. One of its authors reports being a scientific advisor to two companies, receiving professional services from three companies for studies that author leads, and having a provisional patent filing for a digital system for lifestyle medicine outside the submitted work.[9]

In a 2026 study of 39 adults aged 18 to 28 without diabetes, the Abbott Lingo sensor ran an estimated 6.2 minutes behind finger-prick meter readings. The authors say this estimate is imprecise, because each test session had only 6 finger-prick readings. The estimate's 95% confidence interval was 4.1 to 10.1 minutes. Five of the 39 had fasting glucose in the prediabetic range.[9]

What readings looked like in one study of adults without diabetes

In the Framingham Heart Study, 560 community-dwelling adults with normal glucose (average age 58.5 years, 64.5 percent women, 93.3 percent non-Hispanic White) wore a Dexcom G6 Pro sensor for at least 7 days. They spent 87.0 percent of the time between 70 and 140 mg/dL on average, and about 3 hours a day above 140 mg/dL.[10]

The Framingham Heart Study analysis of sensor readings in adults was funded by contracts from the National Heart, Lung, and Blood Institute and a grant from the National Institute of Diabetes and Digestive and Kidney Diseases, and Dexcom provided the sensors at a discounted rate. Two of its authors report receiving a research grant from a pharmaceutical company, unrelated to the project.[10]

What studies of adults without diabetes found

A 2026 systematic review of studies in adults without diabetes found two randomized dietary comparisons. In one, an 8-week trial in 40 overweight or obese adults aged 18 to 35, adding a CGM to education on low glycemic index eating was followed by carbohydrate intake 15.2 g lower than with education alone (95% CI 9.5 to 20.9 g lower).[2]

Physical activity showed no clear difference. The review notes that this trial tested the CGM together with education, so it does not show the effect of a CGM used alone, and it rated certainty low for diet outcomes and very low for other outcomes.[2]

The authors of the 2026 systematic review of adults without diabetes conclude: "Participant-visible CGM may support selected dietary changes as a biofeedback adjunct within structured education or lifestyle interventions, but large, durable benefits and stand-alone efficacy are not established."[2]

The 2026 systematic review of adults without diabetes describes its studies' entry rules as "test-specific criteria rather than a uniform confirmation of normal glucose tolerance across studies".[2]

The authors of the 2026 systematic review of 12 studies in adults without diabetes declare no conflicts of interest, and the review received no external funding.[2]

A 2026 review with meta-analysis pooled 23 studies of mixed designs in 1,074 people without diabetes. It reports lower mean glucose with CGM use overall, a benefit in people with prediabetes, and "no appreciable glycemic benefit" in healthy people with normal glucose. It found no significant difference in body mass index.[11]

The authors of the 2026 review of 23 studies in people without diabetes declare no competing interests. They report grants from the National Key Research and Development Program of China, a Guangdong provincial science and technology program and the TZI-ZJU Industrial Program.[11]

A 2024 meta-analysis of 25 randomized trials of CGM feedback found HbA1c 0.28 percentage points lower and time in range 7.4 percent higher in the CGM feedback groups, with no significant effect on time above range, body mass index or weight. Seventeen of the 25 trials were in type 2 diabetes, and the authors say there were too few studies to report results by diabetes status.[12]

In the 2024 meta-analysis of 25 randomized trials of CGM feedback, 11 trials (44 percent) reported CGM-affiliated conflicts of interest. Three of the meta-analysis's authors also report consulting relationships with companies, one of them former and one unpaid.[12]

An Epic Research report, posted on the Epic Research website rather than published in a journal, looked at 9,675 adults without evidence of diabetes who had a baseline HbA1c between 5.0 and 5.9 percent. Six to 18 months later, average HbA1c in the group prescribed a CGM had gone from 5.49 percent to 5.50 percent. It went from 5.55 percent to 5.62 percent in the group prescribed test strips, and from 5.55 percent to 5.57 percent in the group prescribed neither.[13]

The Epic Research report groups people by what they were prescribed, so it does not show whether they wore a sensor.[13]

What diabetes organizations say

The American Diabetes Association's 2026 Standards of Care describe over-the-counter CGM as lacking alarms and alerts, and say anyone can buy the devices, including people without diabetes or with prediabetes.[14] The same Standards of Care say over-the-counter CGM can be used by people who are not taking diabetes medications that increase the risk of hypoglycemia.[14]

The American Diabetes Association's 2026 Standards of Care describe people without diabetes or with prediabetes who buy over-the-counter CGM as those "who wish to assess their glycemic responses to their lifestyles, including the effects of food choices and exercise."[14]

The Endocrine Society's page on continuous glucose monitoring, in its clinical practice guidelines section, is dated October 1, 2011 and is marked "RETIRED". It lists three settings: "real-time CGM in adult hospital settings", "real-time CGM in children and adolescent outpatients" and "real-time CGM in adult outpatients".[15]

Talk to a licensed clinician about whether a CGM is appropriate for you.

Possible downsides

A 2026 systematic review of 12 studies in adults without diabetes concluded that "the available evidence neither establishes that participant-visible CGM causes psychological harm nor adequately rules such harm out." In the 2026 systematic review of adults without diabetes, the randomized reports on well-being, mood or anxiety "did not provide usable between-group numerical estimates".[2]

In the studies we read (database searches run on October 8, 2026), the data on skin reactions in adults without diabetes are thin. A 2007 series of 22 healthy volunteers who wore an older sensor model for 9 days recorded only minor local events such as itching, pain, redness and burning, and 12 of the 22 had no complications.[16]

A 2026 review reports that in one trial some participants described irritation, that a pilot study of six healthy people found no skin irritation or pain, and that another study described problems with the sensor staying stuck on, "without reliable event counts". None of these reports gives how often skin reactions happen with a current sensor.[2]

If you have a medical emergency, call 911. Problems with medical devices, including imaging equipment, can be reported to FDA through MedWatch.

Provider directory

Find a Provider