What Is a Dexcom G6 Continuous Glucose Monitor?

The Dexcom G6 is a small, wearable medical device that measures your glucose levels around the clock and sends readings to your smartphone or a dedicated receiver every five minutes, without requiring routine fingerstick blood tests. It belongs to a category called real-time continuous glucose monitors (CGMs), and it was one of the first systems to eliminate the need for users to calibrate the sensor with finger-prick blood samples. The device has become one of the most widely used CGMs for people with type 1 and type 2 diabetes, though its applications have expanded into pregnancy monitoring and even research in people without diabetes.

How the Sensor Actually Measures Glucose

The Dexcom G6 consists of three parts: a tiny sensor filament inserted just under the skin, a transmitter that clips onto the sensor, and a display device (either a smartphone app or a standalone receiver). The sensor itself is an electrochemical device built around a two-electrode system. One electrode, the working electrode, is coated with an enzyme called glucose oxidase. When glucose in the fluid beneath your skin comes into contact with that enzyme layer, a chemical reaction produces an electrical signal proportional to the glucose concentration. The second electrode serves as a reference point so the system can interpret that signal accurately.

The glucose readings come from interstitial fluid, the thin layer of fluid surrounding your cells, rather than directly from your blood. This is a meaningful distinction because glucose has to travel from your bloodstream into that interstitial space before the sensor detects it. In healthy adults studied overnight in a fasting state, the delay for glucose to appear in interstitial fluid after entering the bloodstream was about five to six minutes on average, with a conservative upper estimate of roughly ten minutes.¹ A follow-up study in adults with type 1 diabetes found a similar range, with a median delay of about seven minutes and all participants showing detectable values by ten minutes.² In practical terms, if your blood sugar is dropping quickly after a dose of fast-acting insulin, your CGM reading may be a few minutes behind reality. This is why Dexcom and other manufacturers include trend arrows showing whether glucose is rising, falling, or stable, so you can anticipate where your levels are heading rather than reacting only to the current number.

No Fingerstick Calibration Required

Earlier generations of CGMs required users to prick their finger and enter a blood glucose reading at least twice a day so the sensor could recalibrate itself. The G6 eliminated that requirement through what the company calls factory calibration: each sensor is manufactured with a calibration code, and the algorithm built into the transmitter uses that code to interpret the raw electrical signals without needing external reference points from fingerstick tests.³ Sensors read interstitial glucose at five-minute intervals and store the values on the connected receiver or phone app.⁴

That said, the G6 still allows optional calibration if a user feels the readings are off, and there are times when a confirmatory fingerstick is recommended. If your symptoms do not match what the CGM is showing, or if the sensor is giving a reading during a period of rapid glucose change, a traditional blood glucose meter remains the gold standard for making immediate treatment decisions.

How Accurate Is the Dexcom G6?

Accuracy for a CGM is typically described using a metric called mean absolute relative difference, or MARD. Think of MARD as the average percentage by which the sensor reading differs from a reference lab measurement: lower numbers mean closer agreement. In a study of pregnant women wearing the G6 on different body sites, the overall MARD was about 10%, and sensors worn on the back of the upper arm performed best at around 9%.⁵ Those figures are broadly in line with Dexcom’s published clinical trial data for adults, which places the G6’s MARD in the range of 9 to 10%.

Accuracy can shift in certain settings. A study of children with type 1 diabetes during hospital admissions found the MARD against lab glucose values was about 7%, while comparisons with point-of-care glucose meters at the bedside gave a somewhat higher MARD of about 12%.⁶ Another pediatric study using point-of-care comparisons reported an MARD of about 16%, though roughly 97% of paired readings still fell within clinically acceptable zones.⁷ The takeaway: the G6 is accurate enough for day-to-day diabetes management for most people, but extreme situations like critical illness, very rapid glucose swings, or the first hours after sensor insertion can temporarily widen the gap between the CGM and a lab value.

Alerts That Warn You Before Lows Happen

One of the features that distinguishes real-time CGMs from simpler glucose-tracking systems is the ability to set customizable alerts. The G6 can buzz your phone when your glucose crosses a high or low threshold you have chosen. It also has a “predictive urgent low” alert: an algorithm watches your glucose trend and, if it predicts you will drop to roughly 55 mg/dL or below within 20 minutes, it sends a warning while your glucose is still high enough to act.⁸

Real-world data suggests these alerts make a measurable difference. When users who had already been using a prior Dexcom model switched to the G6 with its predictive alert, the percentage of time spent in seriously low glucose ranges dropped significantly.⁸ A separate study found that adults with type 1 diabetes who switched from a system that required manual scanning to the G6’s real-time alerts spent less time below 70 mg/dL and showed less glucose variability overall.⁹ For anyone who has experienced the confusion or danger of a nighttime low, having a device that can essentially tap you on the shoulder before things get serious is a substantial safety net.

Impact on Blood Sugar Control

The constant stream of glucose data changes behavior. When you can see in real time what a particular meal, exercise session, or stressful morning does to your glucose, you tend to make different decisions. Clinical evidence reflects this. A cost-effectiveness model based on clinical trial data assumed that people using real-time CGM achieved about a one-percentage-point drop in HbA1c compared to a smaller drop in those relying on traditional fingerstick monitoring alone.¹⁰

The picture is not limited to type 1 diabetes. A randomized trial in adults with type 2 diabetes on basal insulin found that CGM users spent about 59% of their time in the target glucose range of 70 to 180 mg/dL, compared to 43% in the fingerstick-only group, and their average HbA1c dropped further as well.¹¹ Another study focused on people with poorly controlled type 2 diabetes found that pairing the G6 with structured education brought HbA1c from roughly 9% down to about 7% over 16 weeks, with time in range increasing by more than six hours per day.¹² The improvements were comparable whether people were younger or older than 60, which is reassuring given that type 2 diabetes is heavily concentrated in older adults.

Where You Wear It and How Long It Lasts

Each G6 sensor is approved for up to 10 days of wear. The sensor filament sits just under the skin, and an adhesive patch holds the assembly flat against your body. The FDA-approved sites are the abdomen (for adults) and the back of the upper arm or upper buttocks (for children aged 2 to 17), though studies have evaluated placement on the upper arm and buttocks in adults as well. In pregnant women, the posterior upper arm gave the best accuracy, while the abdomen and upper buttock were comparable to each other.⁵ In that study, no device-related adverse events occurred and skin reactions at insertion sites were absent or minor.

However, adhesive reactions are a known issue for a minority of users. The G6’s adhesive contains acrylate-based chemicals, and case reports have identified specific sensitizers. One compound, abbreviated MBPA, has been confirmed as a contact allergen in multiple patients who developed reactions to the G6 adhesive patch.¹³¹⁴ An earlier adhesive ingredient, isobornyl acrylate, was also identified as a sensitizer in some users.¹⁵ If you develop persistent redness, itching, or blistering under the sensor patch, that is worth discussing with a dermatologist, because true contact allergy tends to worsen with repeated exposure. Many users with mild irritation find that barrier films or skin-prep wipes applied before sensor insertion are enough to prevent problems.

Substances That Can Interfere with Readings

Because the G6 uses an enzymatic electrochemical reaction, certain substances circulating in your body can interfere with the electrical signal and produce falsely high or low readings. One of the biggest improvements the G6 brought over its predecessors was resistance to acetaminophen (the active ingredient in Tylenol), which had historically been a significant source of false highs in CGM users. In a study of 66 people with diabetes, the average interference from a standard one-gram dose of acetaminophen was only about 3 mg/dL, well below the threshold that would be considered clinically meaningful.¹⁶

That does not mean the G6 is immune to all interference. A more comprehensive lab-based analysis found that the device showed meaningful bias with several other substances, including ethyl alcohol, hydroxyurea (a medication used in sickle cell disease and some cancers), high-dose vitamin C (as N-acetyl-cysteine and ascorbic acid precursors), and certain amino acids and sugars like mannose and galactose.¹⁷ Most of these are not a concern for the average user at everyday doses, but people receiving intravenous vitamin C therapy, taking hydroxyurea, or consuming very large amounts of alcohol should be aware that their CGM readings may not be fully reliable during those periods.

The Sharing and Follower Ecosystem

The G6 can transmit your glucose data in real time to up to ten “followers” via the Dexcom Share feature and a companion Follow app. This is especially popular among families with children who have type 1 diabetes: a parent at work can watch their child’s glucose numbers on their own phone and receive the same alerts the child does. In a study of young CGM users, about 95% used the sharing feature, with each child typically having two to three followers. Having at least one follower was associated with lower average glucose, more time in the target range, and more consistent device use across all age groups studied.¹⁸

Sharing is not limited to pediatric use. A survey of both parents and partners of adults with type 1 diabetes found that caregivers actively used the real-time data to support diabetes management decisions.¹⁹ Dexcom also offers Clarity, a cloud-based software tool that aggregates glucose data over weeks and months into reports and trend visualizations that users can review themselves or share with their healthcare provider. The combination of real-time alerts for acute management and retrospective pattern analysis for longer-term tuning is a large part of why CGM tends to outperform fingerstick monitoring: you are not just getting one snapshot per test, you are getting the full movie of your glucose day.²⁰

Use in Pregnancy and Gestational Diabetes

The G6 has been studied in pregnant women, both those with preexisting diabetes and those undergoing screening for gestational diabetes. In the study of pregnant women with diabetes mentioned earlier, the device performed safely across multiple body sites with no device-related adverse events.⁵ In a separate study of women going through gestational diabetes screening, participants wore a blinded G6 sensor for 10 days around the time of their glucose challenge test. Those who spent at least 10% of their time above the target glucose range on CGM were significantly more likely to experience adverse neonatal outcomes than those who stayed mostly within range.²¹

This is an area where CGM may prove more informative than traditional screening methods, which rely on a single two-hour blood draw. A sensor worn for 10 days captures glucose behavior across meals, sleep, and activity in a way that a one-time lab test simply cannot. Research in this space is still evolving, and CGM is not yet part of standard gestational diabetes screening guidelines, but the data is moving in that direction.

Cost-Effectiveness Over the Long Term

CGM systems are not cheap. In the United States, the out-of-pocket cost for the G6 without insurance can run several hundred dollars per month, though most major insurance plans and Medicare now cover it for people on intensive insulin therapy. The question for health systems is whether the upfront expense is offset by better outcomes and fewer complications down the road.

Modeling studies suggest it is. A U.K. analysis found that over a patient’s lifetime, using the G6 instead of fingerstick monitoring added about 1.5 quality-adjusted life years at an additional cost of roughly £14,000, yielding a cost per quality-adjusted life year of about £9,600, well under the threshold most health systems consider good value.²² A similar analysis in France found a comparable gain in quality-adjusted life years, with a cost-effectiveness ratio of about €15,300 per year gained.²³ Both analyses noted that much of the value comes not just from better HbA1c but from the reduced fear of hypoglycemia and the quality-of-life improvement of not having to do multiple daily fingersticks. These are benefits that traditional clinical trials tend to undercount but that people living with diabetes rank highly.

Privacy and Data Considerations

The G6 generates a continuous stream of highly personal health data: your glucose level every five minutes, 24 hours a day, paired with timestamps that can reveal your eating patterns, sleep schedule, and activity habits. That data lives on Dexcom’s cloud servers, can be shared with followers and clinicians, and is technically accessible to the company itself for product improvement and analytics.

Researchers have raised concerns that the privacy protections around CGM data are not yet robust enough to match the sensitivity of what is being collected. A review of data privacy issues specific to CGMs concluded that patients have limited control over how their information is stored and used, and that cybersecurity measures in the broader connected-device ecosystem still lag behind the pace of adoption.²⁴ This does not mean the G6 itself has been the subject of a publicized data breach, but the broader point is that any internet-connected medical device that transmits health data wirelessly brings a level of exposure that a traditional glucose meter sitting in a drawer does not. If you are someone who values data minimalism, it is worth knowing that most of the G6’s core functionality works fine without the cloud-sharing features enabled, though you lose the ability to have followers and retrospective reporting through Clarity.

MRI and Medical Imaging Compatibility

A common practical question for CGM users is whether they need to remove the device before medical scans. The G6 is not MRI-safe per Dexcom’s labeling, and the company recommends removing the sensor and transmitter before entering an MRI suite. However, research that simulated the effects of a 3-Tesla MRI on the G6 found that the displacement force on the sensor and transmitter assembly was far too weak to dislodge it from the body, and that heating at the sensor site was minimal.²⁵ The practical concern is less about safety to the person and more about potential damage to the transmitter electronics or temporary inaccuracy of readings. If you need an urgent scan and cannot easily replace your sensor, discussing the situation with both your radiology team and your endocrinologist is the right move rather than simply skipping the scan.

The G6 in Context with Newer Models

The G6 is no longer the newest device in the Dexcom lineup. The G7, which received FDA clearance in late 2022, is smaller, has a shorter warm-up period, and integrates the sensor and transmitter into a single disposable unit. For many new users, the G7 is now the default recommendation. But the G6 remains widely used for several reasons. It is the version that has been most extensively studied in clinical trials, meaning the body of published evidence on its accuracy, real-world outcomes, and cost-effectiveness is larger. It also continues to be the version integrated into several automated insulin delivery systems, commonly called closed-loop or artificial pancreas systems, which use CGM data to automatically adjust insulin pump delivery. If you are currently using the G6 with an insulin pump and the system is working well, there may be no pressing reason to switch until your pump platform adds G7 compatibility.

For anyone considering a CGM for the first time, the G6 represents a mature and well-understood technology. Its accuracy has been validated across a wide range of populations, from young children to pregnant women to older adults with type 2 diabetes. The sensor chemistry, alert features, and sharing ecosystem have been refined over years of real-world use. Whether the G6 or a newer model is the right fit depends on your specific clinical situation, your insulin delivery setup, and what your insurance covers, but the core experience of continuous glucose monitoring that the G6 delivers has become one of the most impactful advances in everyday diabetes care in decades.