Continuous glucose monitors (CGMs) have been a game changer in helping people successfully manage their diabetes. At the same time, they've also become popular wellness tools for the general public. This shift means that more affordable devices are entering the market which raises questions about their safety and reliability, especially for users living with diabetes.
There are currently no global safety standards in place for CGMs. The result is that some devices on the market have been found to be under-tested, and others are delivering massively inaccurate results.
"CGMs are incredible devices, freeing people with diabetes from constant fingerstick testing and guess work. But inaccurate devices can have life-threatening consequences," says Kirsten de Klerk, Co-founder of SA Diabetes Advocacy.
De Klerk adds that some CGM devices are resulting in families facing life-threatening situations due to inaccurate readings, parents left unsupported when devices fail, and doctors who are unsure which devices can be trusted.
"Think of your CGM like a car," she says. "You wouldn't drive one that hadn't undergone thorough testing across a range of conditions. A CGM should be no different!"
Why does CGM Accuracy matter?
People living with diabetes need to constantly monitor and manage their blood glucose levels. A good CGM will give readings which are very close to actual blood sugar levels, however an inaccurate reading can have significant consequences.
"Inaccurate CGM readings could result in a person missing or wrongly treating dangerous glucose events such as very high or very low glucose levels" says de Klerk. This in turn can lead to dangerous and potentially life-threatening cases of diabetic ketoacidosis (DKA) or severe hypoglycaemia.
Over time, inaccurate readings can increase the risk of long-term complications such as eye, kidney and nerve problems.
Dr Patrick Ngassa Piotie, Chairperson of the Diabetes Alliance South Africa, adds "As access to CGMs grows, so does our responsibility to safeguard the people who rely on them. The high cost of established devices is pushing many towards cheaper products with unverified accuracy, a clear issue of equity and patient safety. From a public health perspective, it is essential that all CGMs on the market meet rigorous quality and performance standards."
In the absence of any global safety or ISO standards for CGMs, international healthcare advocacy group FIND, in partnership with SA Diabetes Advocacy, has published a list of six questions to help users determine whether a device has been thoroughly and sufficiently tested.
These include:
This is important because people living with Type 1 diabetes (T1D) and those living with Type 2 diabetes (T2D) have vastly different experiences. A person with T1D will experience more high and low glucose events, so FIND recommends that you exercise caution when the study group consists of less than 75% of people with T1D.
"You want to know that your device has been tested under stringent and difficult conditions. If your car had only ever been tested at 60km/h on straight, well-kept roads in perfect weather conditions, you wouldn't drive it on winding roads during a storm" says De Klerk.
FIND recommends that the studies should have at least 50 participants with T1D or who are using insulin treatment.
"Larger test studies are better because they provide more reliable, statistically significant results and reduce the impact of random variation or bias. In short, studies with more participants give us a much clearer picture of what is really happening," De Klerk says.
Device manufacturers should also be able to explain why they chose that number of people to take part in that study, upon request.
CGMs can work differently at the start, middle and end of their wear time, and accuracy can vary. That's why it is so important to test it across the whole lifetime of the sensor, which for most CGMs is around 7-14 days.
De Klerk says: "If we go back to our car analogy, a car should be as safe on its first day out of the showroom as it is ten years down the line. Likewise, your CGM should have been tested so that it gives accurate results for the full, recommended wear time."
FIND recommends that studies should include readings at both high and low glucose levels. This is because some CGMs have been found to be less accurate when glucose is either high or low. FIND's recommendation is that studies should include at least 8% of readings below 4.4mmol/L and 5% above 16.7mmol/L.
"Good studies test the sensor in challenging glucose levels, not just the easy middle range," says.
A high carb meal or correction dose of insulin can result in quick changes to glucose levels and CGMs should be able to function as well in those conditions as when glucose levels are stable.
FIND recommends looking for peer-reviewed, published studies to ensure that they're objective and accurate.
"Relying on studies by device manufacturers is like buying your car based solely on a review by the car manufacturer itself. It might still be a good car, but you would definitely question the bias.When other scientists review a study, it helps to make sure it's honest, fair and reduces the risk of bias," says
Keep in mind though, that good studies can take time, and results might not have been published yet, so don't draw conclusions based on this question alone.
Beyond these six questions, FIND and SA Diabetes Advocacy also recommend discussing your CGM choice with your healthcare team, and using fingerstick tests or a reliable CGM device as a backup when first using a new device.