Why Your CGM and Your Glucose Meter Disagree

A continuous glucose monitor (CGM) and a fingerstick meter don’t measure the same thing. The meter measures glucose in a drop of capillary blood; the CGM’s sensor sits under your skin and measures glucose in the interstitial fluid between your cells. Glucose reaches that fluid from the blood a few minutes later, so when your glucose is rising or falling, the two numbers are expected to differ, even if both devices are working perfectly.

This article is general information, not medical advice. If you use glucose readings to decide insulin doses, follow your care team’s guidance on when to check with a meter.

The Time Lag

Glucose moves from your blood into the interstitial fluid, and it takes time. Basu and colleagues measured the delay directly in healthy adults using tracers, and found it was 5 to 6 minutes after an overnight fast. Sensors then add their own processing, so in practice the delay can be longer.

When glucose is steady, the two should agree closely. After a meal, during exercise, or while a low is developing, they won’t: the CGM shows where your blood glucose was a few minutes ago. At a rise of 2 mg/dL (0.1 mmol/L) a minute, 10 minutes of lag is a 20 mg/dL gap.

How Accurate Each Has to Be

Neither device is exact:

• Meters. The international standard for home meters, ISO 15197:2013, requires 95% of readings to fall within ±15 mg/dL (0.83 mmol/L) of the laboratory value below 100 mg/dL (5.6 mmol/L), and within ±15% above it. Two meters, or two drops from the same finger, can differ within those limits.
• CGMs. Accuracy is usually reported as the mean absolute relative difference (MARD) from laboratory blood glucose; each maker publishes its sensor’s, and sensors are generally less accurate in the first day and at low glucose.

So a meter reading 112 mg/dL and a CGM reading 98 mg/dL can both be within their expected error, without any lag.

Getting Readings That Agree

• Compare when glucose is steady, such as before a meal or after an overnight fast, with the CGM’s trend arrow flat.
• Wash and dry your hands before a fingerstick; sugar on your skin from fruit or a drink raises the reading.
• Use the fingertip, not another site, when glucose may be changing.
• Expect a new sensor to settle. Readings in the first 24 hours are often less accurate.
• Watch for pressure on the sensor, such as sleeping on it, which can cause false lows at night.

For symptoms of a low, such as shaking, sweating, or confusion, check with a meter and treat by the meter.

Following It in Kaizora

Kaizora keeps these as two metrics, since they’re different quantities: Capillary Blood Glucose for meter readings and Interstitial Fluid Glucose for a CGM, both on the Vitals page. Apple Health’s export brings in either, depending on how the source app saved it, and you can log meter readings by hand in mg/dL or mmol/L. Putting both on one chart on Correlations shows how closely your sensor follows your meter. Laboratory glucose from a blood draw is a third metric, under Labs. See What Are Normal Glucose Levels After a Meal?

The Short Version

• A CGM measures interstitial fluid; a meter measures blood.
• Interstitial glucose trails blood by 5 to 6 minutes, and a sensor’s processing adds to that.
• Meters may be off by up to 15% and still meet the standard.
• Compare the two only when glucose is steady.

Sources

• Basu A et al., “Time lag of glucose from intravascular to interstitial compartment in humans,” Diabetes 62, no. 12 (2013): 4083–4087. doi:10.2337/db13-1132
• International Organization for Standardization, ISO 15197:2013, In vitro diagnostic test systems: requirements for blood-glucose monitoring systems for self-testing in managing diabetes mellitus.
• Battelino T et al., “Clinical targets for continuous glucose monitoring data interpretation: recommendations from the international consensus on time in range,” Diabetes Care 42, no. 8 (2019): 1593–1603. doi:10.2337/dci19-0028
• US Food and Drug Administration, Blood Glucose Monitoring Devices.