CGMs gaining non‑drug safety role

Continuous glucose monitors were built for insulin therapy. Now they are drifting into a different job. The same sensor that helps a person with diabetes avoid dangerous highs and lows is starting to look like a safety instrument for a much larger group: people on modern metabolic drugs, people with type 2 diabetes who do not use insulin, and even people without diabetes who want to watch how food, exercise, and medication change their glucose in real time. That shift became easier to imagine after the American Diabetes Association’s 2025 standards kept pushing CGM deeper into routine diabetes care, while a separate FDA decision in March 2024 cleared Dexcom’s Stelo as the first over-the-counter CGM for adults 18 and older who do not use insulin. (diabetesjournals.org) That matters because the center of gravity in metabolic medicine has moved. Care is no longer defined only by insulin dosing. It is increasingly shaped by GLP-1 drugs, dual agonists, weight-loss treatment, and earlier intervention in type 2 diabetes and prediabetes. In that world, a glucose trace can become a running readout of whether therapy is working, whether meals are landing differently, and whether a patient is drifting toward a problem that would never show up in a quarterly A1C test. The evidence for using CGM this way is still uneven, but it is real enough that clinical trials have already started treating CGM metrics as useful endpoints alongside older lab measures. (nature.com) The OTC clearance made the change visible. Stelo was cleared for adults who do not use insulin, including people with diabetes managed with oral drugs and people without diabetes who want to understand how diet and exercise affect blood sugar. It gives a reading every 15 minutes, each sensor can be worn for up to 15 days, and the FDA was explicit about what it is not for: people with problematic hypoglycemia should not rely on it, and users should not make medical decisions from the app without talking to a clinician. Even in that narrow label, you can see the new role emerging. A sensor no longer has to be part of drug delivery to become part of drug surveillance. (fda.gov) That is where pharmacovigilance enters the story. Traditional drug safety systems still depend heavily on adverse-event reports filed after something goes wrong. FAERS is built around those reports, using the ICH E2B data standard to structure individual case safety records for drugs and biologics. CGMs offer a different kind of evidence. They can capture a physiologic pattern before anyone files a report at all. A sudden rise in post-meal glucose after a medication change, repeated overnight dips, or a new stretch of instability could all become part of a richer safety picture if the data are linked to treatment history and outcomes. FDA’s digital health guidance already describes how sensor-based tools can be used for remote data acquisition in studies of medical products. The missing piece is not the sensor. It is the pipeline that turns a glucose stream into usable safety evidence. (fda.gov) That pipeline is harder than it sounds. CGMs are noisy in the ordinary ways that consumer health tech is noisy. People stop wearing them. Adhesives irritate skin. Phones disconnect. Users scan the graph, change what they eat for a few days, then drift back to baseline. FDA’s own Stelo announcement lists local infection, skin irritation, and pain or discomfort among reported adverse events, and MAUDE reports for other CGMs show the familiar device-level problems, from rash to inflammation to sensor trouble. Those issues do not make the data useless. They do mean that any safety system built on CGMs will have to sort drug effects from device effects and biology from behavior. (fda.gov) The other problem is who gets counted. FDA framed OTC CGM access as a health-equity step because it removes the prescription barrier. That is true as far as it goes. But access without reimbursement is still selective access. The people most likely to generate dense streams of glucose data may also be the people most able to pay out of pocket, own compatible phones, and tolerate the constant feedback of a body sensor. If CGMs start feeding post-market safety work around metabolic therapy, the resulting evidence could be rich and skewed at the same time. (fda.gov) That is why this story is bigger than one device clearance or one set of diabetes guidelines. CGMs are turning into general-purpose metabolic sensors just as obesity and diabetes treatment is becoming more continuous, more consumer-facing, and more data-hungry. The surprising part is not that people want to watch their glucose. It is that a patch on the back of the arm, replaced every 15 days and pinging a phone every quarter hour, may end up telling regulators and drug-safety teams almost as much as it tells the wearer.