If your A1C and CGM do not match, it does not automatically mean either number is wrong. A1C reflects glucose attached to hemoglobin over roughly two to three months, while your CGM estimates glucose from interstitial fluid over its wear period; bring both reports to your next visit so the pattern can be interpreted together.
The useful question is not which number wins. It is what each number adds: your CGM can show daily highs, lows, and patterns, while A1C gives a longer-view blood-test result.
Why don’t my A1C and CGM match?
A1C and CGM measure related things in different ways and over different windows of time. A CGM measures glucose in the fluid around cells, while A1C is based on glucose attached to hemoglobin in red blood cells.
A 2025 Diabetes Technology & Therapeutics study of people with prediabetes or normal glucose found that standard CGM metrics did not track A1C, and concluded they should not be read as a stand-in for it. Another analysis of long-term CGM users describes several reasons for a gap: CGM sensor bias, a shorter CGM sampling period, missing data, A1C assay variation, and non-glucose influences on red blood cells.
Red-blood-cell conditions can matter because they change how A1C is formed or interpreted. The National Institute of Diabetes and Digestive and Kidney Diseases A1C guide explains that hemoglobin variants and conditions affecting red blood cells can affect A1C results. Anemia, kidney disease, blood loss, a recent transfusion, and some hemoglobin conditions are examples worth putting on your visit list if they apply to you.
What is the difference between GMI and A1C?
GMI means glucose management indicator. It is a percentage calculated from your CGM’s average glucose over the selected reporting period. A1C is a laboratory blood test. They can look similar, but GMI is not a replacement name for A1C.
| Number | What it is based on | What it helps you discuss |
|---|---|---|
| GMI | Your CGM’s mean glucose during the report period | Recent glucose patterns alongside time in range |
| A1C | Glucose attached to hemoglobin | Longer-term glucose exposure over about two to three months |
| Finger-stick reading | Capillary blood glucose at that moment | A point-in-time comparison when a CGM reading seems unexpected |
A GMI of 5.5 means the CGM average during that report period produced a GMI estimate of 5.5%. It does not tell the whole story by itself: check the report period, active-data percentage, time in range, and whether there were meaningful highs or lows.
For a reference point, the formula reported for GMI is 3.31 + 0.02392 multiplied by mean glucose in mg/dL. That means an average of 154 mg/dL corresponds to a GMI of about 7.0%.
Is there a correlation between A1C and glucose?
Yes. A1C is related to average glucose, which is why it is used in diabetes testing and follow-up. The CDC’s diabetes testing guidance uses A1C categories for testing, including 6.5% or higher on two separate tests for diabetes.
For an A1C of 7.0%, the commonly used estimated average glucose is about 154 mg/dL. Treat that as a comparison point, not a personal conversion calculator. In the retrospective cohort, CGM-based average glucose differed from the A1C-derived estimate by about 12 mg/dL on average.
Can I have a high A1C but normal CGM glucose?
Yes, a higher A1C alongside a lower CGM average can happen. One explanation is timing: the lab result reflects a much longer period than a recent CGM report. Another is that factors affecting red blood cells can move A1C higher or lower than the glucose pattern would suggest.
The reverse can happen too. A recent CGM period with higher readings may not yet be fully reflected in an A1C that includes earlier weeks. Bring the lab date and the exact CGM date range together rather than comparing an A1C to a single day or one isolated sensor reading.
Why is my CGM higher or lower than a finger stick?
A CGM measures interstitial glucose, not blood glucose from a finger stick, so a difference can be more noticeable when glucose is changing. Sensor insertion and placement, sensor expiration, software issues, and missing or incomplete data can also affect the number you see.
If a CGM result does not fit how you feel or keeps disagreeing with a finger-stick result, note the time, both readings, food, activity, and any alerts. Then use your device’s instructions and support route for model-specific troubleshooting, and take that short log to your clinician. For urgent-pattern context, keep these guides handy: high blood sugar symptoms and low blood sugar symptoms and treatment.
Where should I place a CGM sensor?
CGM sensors are commonly worn on the arm or abdomen, but the approved placement depends on the exact model. Use the placement instructions for your own device rather than borrowing a location from another CGM.
For an arm sensor, choose the model-approved area where the patch can lie flat and is less likely to be rubbed, bumped, or compressed. For a stomach sensor, use only a model-approved abdominal area. If you sleep on your side, choose an approved spot that is less likely to sit under your body weight; if every approved spot gets compressed overnight, ask the device maker for placement help.
A blood sugar monitor without finger pricks guide can help you compare the practical role of a CGM with other monitoring options.
Should I trust GMI or A1C more?
Use the question you are trying to answer. GMI and CGM trends are useful for seeing recent day-to-day glucose patterns. A1C is the laboratory measure used for diabetes testing and longer-term follow-up. When they disagree repeatedly, the mismatch itself is useful visit information because it can point to timing, CGM data quality, or an A1C factor that deserves discussion.
In an analysis of long-term CGM users, averaging A1C-based and CGM-based estimates was more accurate when CGM covered fewer than about 26 days. That is a good reason to take both, rather than trying to force one number to become the other.
What should I bring to my appointment?
- Your lab A1C result and the date it was drawn.
- Your CGM report, including the exact date range, average glucose, GMI, time in range, and active-data percentage.
- Any finger-stick comparisons that seemed surprising, with the time of each reading.
- A list of recent illness, blood loss, transfusion, anemia, kidney concerns, or known hemoglobin conditions.
- Your CGM model, sensor location, and any repeated placement or adhesion problem.
A simple script to use
“My A1C was ___ on ___, and my CGM GMI was ___ for ___ through ___. Can we look at whether the date range, sensor data, or anything affecting A1C could explain the difference?”
Before you leave, ask which number or report should be repeated, what date range to use next time, and whether your sensor placement needs a model-specific review.
By the Primary Care Visit Guide Team
Leave a Reply