What a Continuous Glucose Monitor Measures
What a continuous glucose monitor actually senses, what it estimates, why readings lag behind blood sugar, and how accurate over-the-counter sensors were in one independent study.
Content last updated . This page has not been medically reviewed.
Direct reading, estimate and calculated display
- Direct (sensor). The sensor produces a signal tied to glucose in interstitial fluid, the fluid around cells. The NIDDK says CGM sensors estimate the glucose level in the fluid between your cells, which is very similar to the glucose level in your blood. [1]
- Estimated (software). The system converts that signal to a blood glucose estimate, shows trends and sends them to an app. [1] One FDA announcement described an over-the-counter system that presented measurements and trends every 15 minutes, with each sensor worn up to 15 days. [2]
- Calculated by the app. Charts, trend arrows and any extra scores are built on the estimate. We did not find a source we read that gives proprietary "scores" a validated clinical meaning: Proper evidence not found.
The NIDDK adds that for some models a finger-stick test is needed to calibrate, and that disposable sensors are replaced every 7 to 14 days depending on the model. [1]
Why readings lag
A review of CGM technology explains that the delay between a blood glucose reading and the value a CGM displays is the sum of the time lag between interstitial fluid and plasma glucose, plus the sensor's reaction delay and any signal processing needed to smooth the trace. [3] That review dates from 2009, so it explains the concept rather than describing current devices. In practice it means a CGM line can trail a quick rise or fall in blood sugar.How accurate are over-the-counter sensors?
Evidence is thin. A 2026 study of 39 young adults without diabetes compared over-the-counter CGMs with capillary meter (finger-stick) glucose during rest or exercise combined with a placebo or sugar-sweetened drink. [4] The study reported:- mean absolute relative difference (MARD, the average percentage gap) of 12.9%;
- 66.5% of readings within 15% or 15 mg/dL of the meter;
- worse agreement after the sugar drink (MARD 15.3%) than placebo (10.6%);
- a population lag of about 6.2 minutes;
- 13.8% of trend arrows classed as moderate to extreme risk;
- differences between participants that measured biological variables did not explain. [4]
The authors concluded these sensors seemingly agreed with overall trends but were less reliable for exact values, and urged caution for diagnostic thresholds. [4] The comparison was with a finger-stick meter rather than a laboratory reference, in young adults, and the abstract does not name the device models, so do not generalize to other devices, ages or conditions. Label: promising but limited for trend-watching, insufficient evidence for exact values.
Accuracy is not benefit
Even a reasonably accurate sensor does not prove that using it improves health. A 2026 review found only limited and indirect evidence for CGM in people with type 2 diabetes not on glucose-lowering treatment, prediabetes or obesity. [6] See CGMs in people without diabetes, and for blood versus interstitial readings, interstitial vs blood glucose.Frequently asked questions
Does a CGM measure blood glucose?
Not directly. It estimates the glucose level in the fluid between cells, which the NIDDK describes as very similar to blood glucose. [1]
Why does my CGM reading differ from a finger-stick?
Part of the reason is a delay between blood and interstitial glucose plus sensor and processing delays. [3] In one study, agreement was worse after a sugar-sweetened drink. [4]
How accurate are over-the-counter CGMs in people without diabetes?
In one independent study of 39 young adults, average difference from a finger-stick meter was 12.9%, with better agreement for trends than for exact values. [4]
Can a smartwatch measure my glucose?
The FDA has not authorized, cleared or approved any smartwatch or smart ring intended to measure or estimate blood glucose on its own. [5]
Should I use a CGM to decide what to eat?
There is no strong evidence of benefit for people without diabetes, and the FDA advises not making medical decisions from the output without talking to a provider. [6][2]
More details: evidence, limits and cautions
The details behind this page: cautions, limits of the evidence and where sources disagree. Open any note to read it.
Regulation: one device, one function
Devices that claim to read glucose without piercing the skin are a different matter: in February 2024 the FDA said it had not authorized, cleared or approved any smartwatch or smart ring intended to measure or estimate blood glucose on its own. [5]
Who should check first, and safety limits
References
- National Institute of Diabetes and Digestive and Kidney Diseases. Continuous Glucose Monitoring. Read 2026-10-22.
- U.S. Food and Drug Administration. FDA Clears First Over-the-Counter Continuous Glucose Monitor. Press announcement, March 5, 2024. Read 2026-10-22.
- Keenan DB, et al. Delays in minimally invasive continuous glucose monitoring devices: a review of current technology. J Diabetes Sci Technol, 2009. Read 2026-10-22.
- Prakash N, et al. Agreement of Over-the-Counter Continuous Glucose Monitors With Capillary-Meter Glucose in Young Adults Without Diabetes. J Diabetes Sci Technol, 2026. Read 2026-10-22.
- U.S. Food and Drug Administration. Do Not Use Smartwatches or Smart Rings to Measure Blood Glucose Levels: FDA Safety Communication. Issued February 21, 2024. Read 2026-10-22.
- Dower JA, et al. Continuous Glucose Monitoring in Type 2 Diabetes and Beyond: A Review. JAMA Intern Med, 2026. Read 2026-10-22.