Nothing sours a good week with a CGM quite like a sensor that fails on day three of a fifteen-day wear. Or one that peels off in the shower. Or one that spends every night crying wolf about lows that are not real. These complaints come up constantly, and most of them have specific, fixable causes.
This guide walks through the most common CGM problems, what actually causes each one, and what to do about it. None of these mean you are using your device wrong.
Start Here: When to Call for a Replacement
Before troubleshooting anything, know that faulty sensors are frequently replaced at no cost, and most people under-use this:
- Call the manufacturer, not your supplier, for a sensor that failed early or read inaccurately. They handle warranty replacements directly.
- Have your lot number ready, which is on the sensor packaging. Keep the box until the sensor completes its wear.
- Report it even if you are not sure, since reporting also feeds back into quality monitoring.
- Note the pattern. Multiple failures from the same lot is useful information for them and for you.
- Do not stretch a suspect sensor. If it is reading unreliably, replacing it is safer than working around it.
If a sensor fails and you are without data, that gap matters. Reliable interpretation of your reports generally depends on capturing at least 70 percent of readings over a two-week window, so repeated failures quietly erode the picture you and your provider are working from.
Complaint One: The Sensor Falls Off Early
Adhesive failure is the most common complaint and usually the easiest to fix, because it is mostly about what happens in the two minutes before insertion.
Skin oils, lotion, and moisture are the main culprits. Wash the site with plain soap and water, let it dry completely, and skip moisturizer that morning. Some people use an alcohol wipe and then wait for it to fully evaporate, since applying adhesive to damp skin defeats the purpose. Barrier or adhesive-prep products can help both with sticking and with reducing irritation.
Heat, sweat, and swimming all shorten adhesive life, which is why summer is when this complaint spikes. An overpatch or adhesive tape placed over the sensor, without covering any part that needs to breathe or transmit, extends wear considerably for active users.
Placement matters more than people expect. A sensor on the outer upper arm survives daily life better than one placed where a backpack strap, seat belt, or waistband rubs across it repeatedly. Our guide on extending sensor adhesion goes deeper on this.
Complaint Two: False Lows at Night
Being woken at 3 a.m. by a low alarm that turns out to be nothing is genuinely exhausting, and it has a specific cause with a specific name.
A compression low happens when sustained pressure on the sensor, usually from lying on it, restricts local blood flow and interstitial fluid around the sensor wire. With less fluid available to read, the sensor reports a value well below your actual blood glucose. Research on interstitial CGM has documented this susceptibility to pressure directly.
The pattern is recognizable: a sharp sudden drop from otherwise steady readings, forming a V shape, with no meal, dose, or exercise to explain it, and recovery within minutes of you shifting position. If you see that shape and feel fine, confirm with a fingerstick before treating, because treating a false low sends you high afterward.
Prevention is placement. When you insert a sensor, think about which side you sleep on and choose accordingly. Compression happens in the daytime too, from tight clothing, seat belts, and the inner surface of the upper arm.
Complaint Three: The Numbers Seem Wrong
Some difference between your sensor and your meter is expected, since they measure different fluids. Interstitial glucose lags behind blood glucose, so gaps widen when your glucose is moving fast.
Timing within the sensor's life matters too. Accuracy tends to be somewhat lower on the first day of wear and again toward the end of the wear period. If your numbers look erratic on day one, giving it 12 to 24 hours before drawing conclusions is reasonable.
Site selection genuinely affects accuracy, and studies have found differences between arm, abdomen, and other locations. Scar tissue, very thin skin, and previously overused sites all degrade readings. If one location consistently underperforms for you, moving is a legitimate fix.
Certain substances can interfere with some sensors, and manufacturers publish device-specific lists. This is worth looking up once for your particular sensor rather than assuming it applies or does not.
Why this matters: people often conclude their sensor is defective when the cause is placement, timing, or an interfering substance. Working through those first saves you from replacing sensors that would have been fine, and identifies the ones that genuinely were not.
Complaint Four: Skin Irritation Under the Adhesive
Redness, itching, or a rash under a sensor is common and worth addressing rather than tolerating, because irritated skin also limits where you can rotate to.
Barrier products applied before the sensor create a layer between the adhesive and your skin and help many people considerably. Thin transparent dressings can serve the same purpose. Rotating sites deliberately gives previously used skin time to recover, which matters more than it sounds when sensors wear for two weeks at a time.
If irritation is severe, spreading, or accompanied by blistering, that is worth raising with your care team rather than managing at home, since a true adhesive allergy needs a different approach than simple irritation. Warmth, swelling, or drainage at a site should be evaluated promptly.
Why It Matters
Sensor problems do not just cost you data. They cost you sleep, they cost you trust in the device, and enough of them cost you the willingness to keep wearing it. People abandon CGMs over adhesive failures and nighttime false alarms far more often than over anything clinical, which is a genuinely bad outcome given what continuous data does for glucose management.
Most of these have fixes that take two minutes to implement. Knowing that a 3 a.m. alarm is probably a compression low, and that a fingerstick will confirm it, turns a recurring crisis into a minor annoyance.
At AdaptHealth Diabetes, our mission is to make diabetes management simpler and more accessible, and that includes making sure a failed sensor is replaced quickly rather than leaving you without data. We supply sensors across every major brand and can help you troubleshoot supply-side issues. Read more in our education center, or see our guide to CGM site selection and rotation.
Frequently Asked Questions About CGM Problems
Will the manufacturer replace a sensor that failed early?
In most cases yes, and this is more routine than people assume. Contact the manufacturer's support line directly rather than your supplier, and have the sensor's lot number from the packaging ready along with roughly when and how it failed. Keep the box until the sensor finishes its wear so you have that information available. Replacements are typically shipped without charge when the failure is within warranty.
How do I know if it is a compression low or a real low?
Look at the shape and the context. A compression low shows up as a sudden sharp drop from steady readings, usually overnight, with no meal, insulin dose, or exercise to explain it, and it recovers on its own within minutes of you moving. A real low usually has a cause and a more gradual approach. When in doubt, confirm with a fingerstick before treating, and always treat if you have symptoms regardless of what the sensor says.
Can I put a sensor back on if it comes loose?
Not reliably. Once the adhesive has lifted, the sensor wire underneath has usually shifted enough that readings become unreliable even if you tape it back down. Taping over a partially detached sensor risks trusting bad data, which is worse than having none. Replace it and contact the manufacturer about a warranty replacement if it failed well before its expected wear time.
Why is my sensor less accurate on the first day?
The tissue around a newly inserted sensor is briefly disturbed by the insertion itself, and it takes time for the local environment to settle. This is why sensors have a warm-up period, and why accuracy often continues improving for several hours after warm-up ends. If day-one readings look erratic but plausible, giving it 12 to 24 hours before judging is reasonable. If they are wildly off or the sensor errors repeatedly, call for a replacement.
Is it normal to go through this much trial and error?
Yes, and it is worth saying plainly. Finding the sites, prep routine, and adhesive approach that work for your skin and your life takes most people several sensors and sometimes several months. Bodies differ, and so do routines. None of this reflects on how well you are managing your diabetes, and the people who seem to have it effortlessly figured out mostly went through the same process earlier.
Keep Sensors Coming, Even When One Fails
Sensor failures are less disruptive when you have a spare on hand and a supplier who can move quickly. AdaptHealth Diabetes keeps CGM sensors stocked and shipped on schedule across every major brand, and can help you sort out coverage for replacements. Reach out through our contact page, or visit our resources page for help with reordering.
For general guidance on glucose monitoring and managing blood sugar, the CDC offers patient information at cdc.gov/diabetes.
Medical Disclaimer: This article is for educational purposes only and is not a substitute for professional medical advice, diagnosis, or treatment. Always follow the instructions provided with your specific CGM and consult your diabetes care team about persistent accuracy problems, skin reactions, or site issues. Signs of infection at a site, including warmth, swelling, or drainage, should be evaluated promptly. If you feel unwell or your symptoms do not match your readings, use a blood glucose meter and seek care as needed.
Sources
- Susceptibility of Interstitial Continuous Glucose Monitor Performance to Sleeping Position (NIH/PMC): https://pmc.ncbi.nlm.nih.gov/articles/PMC3879750/
- Accuracy and Safety of Dexcom G7 Continuous Glucose Monitoring in Adults with Diabetes (NIH/PMC): https://pmc.ncbi.nlm.nih.gov/articles/PMC9208857/
- Expert recommendations on CGM data interpretation and minimum data capture (NIH/PMC): https://pmc.ncbi.nlm.nih.gov/articles/PMC12060294/
- Centers for Disease Control and Prevention (CDC), Manage Blood Sugar: https://www.cdc.gov/diabetes/treatment/index.html
- American Diabetes Association (ADA), CGM and Time in Range: https://diabetes.org/about-diabetes/devices-technology/cgm-time-in-range
