Users Report Issues With New Glucose Monitor Model

Technology has come a long way in recent years. In the 2010s, these devices required blood glucose calibration twice daily and were still not reliable enough for dosing. insulin based on evidence. Some insulin pumps now use near real-time data to make adjustments automatically. These improvements have brought greater trust in the data users receive and higher standards. Incorrect readings can result in a dangerous dose of insulin.

According to Simpson, the G7 made several changes to Dexcom's earlier designs, including a much smaller footprint, and also updated the algorithm used to translate sensor signals into glucose readings for greater accuracy. “From a performance standpoint, we have demonstrated in clinical trials that the G7 is significantly more accurate than the G6,” he says.

So Heller and others were surprised when the new Dexcom sensor began to perform worse. For some batches of sensors, it is possible that the problem was partly due to an untested change to a component used in the resistive layer of the sensors. The new component showed worse performance, according to warning letter released by the US Food and Drug Administration in March 2025 following an inspection of two US manufacturing sites. According to Simpson, this material has since been removed from all G7 sensors and the company continues to work with FDA to solve problems. (“The Warning Letter does not limit Dexcom's ability to manufacture, sell, manufacture or distribute products, seek a recall of any products, and does not limit our ability to seek approval for new products,” Dexcom added in a statement.)

“There is a distribution of accuracy related to the physiology of people as well as the devices themselves. Even in our clinical studies, we've seen some that were really accurate and some that had some imprecision,” Simpson says. “But overall our sensor is very accurate.”

At the end of November, Abbott – one of Dexcom's main competitors –recalled some of its CGMs due to inaccurate low glucose readings. The recall affected approximately 3 million sensors and was caused by a problem on one of Abbott's production lines.

The discrepancy between the stated accuracy and user experienceHowever, this goes beyond any one company's manufacturing errors.

Does MARD matter?

The accuracy of CGM systems is often measured using the “mean absolute relative difference” or MARD, a percentage that compares sensor readings to laboratory blood glucose measurements. The lower the MARD, the more accurate the sensor.

This number is often used in advertising and marketing, and it has historical significance, says Manuel Eichenlaubbiomedical engineer Institute for Diabetes Technology Ulm V Germanywhere he and his colleagues conduct independent research into the effectiveness of CGM. For many years there was general belief that a MARD below 10 percent meant the system would be accurate enough to be used for insulin dosing. In 2018, the FDA established a specific set of accuracy. requirements outside of MARD for glucose monitors with insulin control, including Dexcom. But manufacturers are developing clinical trials that determine accuracy rates, and how the studies are conducted can make a big difference.

When Dan Heller wore the Dexcom G6 and G7 at the same time, he says he noticed that the G7's readings were more erratic, making it more difficult to properly control his blood sugar. Dan Heller

For example, blood glucose serves as “the primary metric for comparing CGM values,” Eichenlaub says. But glucose levels vary depending on the blood section in the body; Blood collected from capillaries using a finger prick fluctuates more and may have glucose levels that are about 5 to 10 percent higher than venous blood. (Dexcom is testing a gold-standard venous blood analyzer. When users see inaccuracies in home meters that use capillary blood, it may partly be a reflection of the meter's own inaccuracy, Simpson says, although he acknowledges real inaccuracies in the CGM.)

In addition, the sample distribution is not standardized. MGAs are known to be less accurate at the beginning and end of use, and when glucose levels are out of range or changing rapidly. This means that the accuracy of the measurements may be skewed by taking fewer samples immediately after a meal or at the end of the CGM's life.

Dexcom's testing protocol meets FDA expectations and tests devices across different blood sugar ranges over the life of the sensor, Simpson said. “In these clinical trials, we stress the sensors to try to mimic real-world conditions,” he says.

Dexcom and other companies advertise a MARD of about 8 percent. But some independent studies are more demanding and show higher numbers; A direct study of three popular CGMs under Eichenlaub's leadership found MARD values ​​approaching 10 percent or higher.

Eichenlaub and other CGM experts believe greater standardization of testing and expanded FDA requirements are needed, so they recently proposed comprehensive recommendations on CGM performance testing. In the USA and Europethe market is currently dominated by a few manufacturers. But New players are entering the growing market, especially in Europe. may not be held to the same standards as legacy manufacturers, he says.. “It’s important to have a standardized way to evaluate the performance of these systems.”

However, for users like Heller, greater accuracy only matters if it leads to better diabetes control. “I don't care about MARD. I need reliable data. effective” says Heller. He encourages engineers working on these devices to think like a patient. “At some point there is quantitative data, but you need qualitative data.”