Browse all articles

Home Get the app
Glucose Management Indicator (GMI) and Time in Range

Biomarker Published Aug 8, 2026

Glucose Management Indicator (GMI) and Time in Range

CGM report numbers that show average glucose and time in range

Also known as

GMI · glucose management indicator · TIR · time in range · CGM metrics · ambulatory glucose profile · AGP report · estimated A1C · eA1C

Use them to spot hidden lows and spikes that a single lab average can miss.

4 min read · 836 words · 4 sources

In brief

In brief

Glucose Management Indicator (GMI) estimates an A1C-like value from continuous glucose monitor data, while Time in Range measures the share of time glucose stays within a target band, usually 70 to 180 mg/dL for nonpregnant adults with diabetes.

  • GMI converts average CGM glucose into an estimated A1C, not a laboratory blood test.2
  • Time in Range counts the proportion of the day glucose stays inside the chosen target range.
  • GMI and laboratory A1C can disagree when red blood cell turnover is unusual.

Deep dive

How it works

GMI is built from mean interstitial glucose, which is glucose measured in fluid between cells rather than directly inside a vein. CGM devices sample this fluid repeatedly, then software averages the readings. Time in Range uses the same readings but does not average them. It sorts each reading into below range, in range, or above range, then reports the percentage of recorded time in each bucket.

When you'll see this

The term in the wild

Scenario

You open Dexcom Clarity and see average glucose 154 mg/dL, GMI 7.0 percent, and Time in Range 72 percent from 14 days of data.

What to notice

The GMI says your sensor average lines up with an A1C shaped value near 7 percent. The Time in Range says roughly 17 hours per day were between 70 and 180 mg/dL.

Why it matters

This pattern is close to the common consensus target for many nonpregnant adults, but the next safety check is whether time below 70 mg/dL stayed under 4 percent.

Scenario

Your lab A1C is 6.8 percent, but LibreView shows GMI 7.4 percent and only 55 percent Time in Range.

What to notice

The lab and sensor are answering different questions. The lab A1C reflects glucose attached to red blood cells, while GMI comes from recent sensor average glucose.

Why it matters

The mismatch is not automatically a sensor failure. It is a reason to review CGM wear time, recent glucose changes, anemia, kidney disease, pregnancy status, or other factors with a clinician.

Scenario

A supplement forum claims berberine improved someone’s Time in Range from 48 percent to 65 percent in two weeks.

What to notice

Time in Range can show a real pattern change, but the claim needs context: medication changes, diet, weight change, illness, sensor wear, and low glucose time can all move the number.

Why it matters

For glucose active supplements such as berberine, the safety issue is not just higher Time in Range. It is whether time below 70 mg/dL increased, especially if the person also uses insulin or a sulfonylurea.

Scenario

A pregnancy CGM report shows 70 to 180 mg/dL targets, but the endocrinology team talks about 63 to 140 mg/dL.

What to notice

Pregnancy uses different CGM targets than the common adult diabetes range. The tighter range changes what counts as in range.

Why it matters

Using the wrong range can make a report look better than it is for pregnancy specific care.

The full picture

The same A1C can hide two different days

A continuous glucose monitor report can show two headline numbers that look redundant: GMI and Time in Range. They are not duplicates. One turns your sensor average into an A1C shaped number. The other counts minutes.

Here is the surprise: two people can have a similar GMI, yet one spends most of the day between 70 and 180 mg/dL while the other swings from low to high and back again. Their averages can match, but their lived days are different. That is why modern CGM reports show both numbers instead of replacing one with the other.

What GMI actually calculates

Glucose Management Indicator, or GMI, is calculated from the average glucose recorded by a continuous glucose monitor. In the United States, the formula is: GMI (%) = 3.31 + 0.02392 x mean glucose in mg/dL. If your CGM average is about 154 mg/dL, the GMI is about 7.0 percent. If your average is 180 mg/dL, the GMI is about 7.6 percent.

GMI was created because the older phrase, estimated A1C, caused a specific problem: people expected it to equal the lab A1C drawn from blood. The U.S. Food and Drug Administration objected to the older name because it could imply that a sensor calculation was the same thing as a laboratory A1C. The new name keeps the useful estimate while making the difference clearer.

A lab A1C measures how much glucose has attached to hemoglobin, the oxygen carrying protein inside red blood cells. GMI does not measure red blood cells. It is math from sensor glucose. That means GMI may be higher or lower than lab A1C when red blood cells live for a shorter or longer time, after recent glucose changes, during pregnancy, with kidney disease, anemia, blood loss, transfusion, or some hemoglobin variants.

What Time in Range counts

Time in Range, or TIR, is the percentage of CGM readings between a chosen lower and upper glucose limit. For most nonpregnant adults with type 1 or type 2 diabetes, the standard target range is 70 to 180 mg/dL. A common goal is more than 70 percent of the day in range, which is about 17 hours per day. The same consensus targets also aim for less than 4 percent below 70 mg/dL, less than 1 percent below 54 mg/dL, less than 25 percent above 180 mg/dL, and less than 5 percent above 250 mg/dL.

Pregnancy uses tighter CGM ranges. For type 1 diabetes in pregnancy, the consensus target range is 63 to 140 mg/dL, with different goals because fetal exposure to glucose matters and low glucose risk must also be watched carefully.

The decision to make today

If you are looking at a CGM report, do not react to GMI first. First check Time Below Range. If lows are above target, bring that pattern to your clinician before trying to lower GMI. A lower average is not progress if it is being bought with unsafe lows. Once lows are controlled, Time in Range and GMI become more useful for judging meals, medication timing, activity, sleep, and whether a change is helping.

Myths vs reality

What people get wrong

Myth

GMI is just your A1C before the lab confirms it.

Reality

GMI is a sensor based calculation. Lab A1C is a blood measurement involving red blood cells. They often move together, but they are not the same test.

Why people believe this

The older label, estimated A1C or eA1C, made the number sound interchangeable with a lab A1C. The 2018 Diabetes Care paper explains that FDA concern about this wording helped drive the name change to GMI.

Myth

A good GMI means glucose control is safe.

Reality

A good average can hide too many lows. Time Below Range shows whether the average is being pulled down by glucose values under 70 mg/dL or under 54 mg/dL.

Why people believe this

A1C centered care trained many people to treat a single average as the main scoreboard, while CGM reports separate average glucose from low glucose exposure.

Myth

Time in Range always means 70 to 180 mg/dL.

Reality

That is the common range for many nonpregnant adults with diabetes, but pregnancy and some high risk situations use different targets.

Why people believe this

CGM apps and articles often display the adult default range first, so readers may not realize the range is adjustable and population specific.

Why this keeps coming up

These numbers keep showing up because many food, supplement, and habit changes are judged by whether they make glucose steadier, not just lower.

Berberinemealsactivitysleepmedication timing

How to use this knowledge

A specific failure mode is comparing two CGM reports when one has poor sensor coverage. Consensus reporting uses at least 14 days with about 70 percent active CGM data for a reliable snapshot. If a report has large gaps, treat the GMI and Time in Range as clues, not a verdict.

What to do with this

  • Check time below range before you try to lower GMI.
  • Use Time in Range to judge whether a change makes glucose steadier, not just lower.
  • Compare GMI with lab A1C as separate numbers, especially if red blood cell turnover is unusual.
  • Make sure you have enough CGM data before acting on the summary.
  • Use the pregnancy target range if you are reviewing a pregnancy CGM report.

Frequently asked

Common questions

How many days of CGM data do I need before GMI and Time in Range are meaningful?

A common reporting standard is about 14 days of CGM data with at least 70 percent active sensor time. Fewer days can still be useful for spotting obvious patterns, but the summary numbers are less stable.

Which number should I bring up first at a diabetes visit?

Start with Time Below Range if you are having lows. After safety is addressed, Time in Range, Time Above Range, and GMI can help guide changes in meals, medication timing, or activity.

Can GMI be used to diagnose diabetes?

No. Diagnosis still relies on accepted lab tests such as A1C, fasting plasma glucose, or an oral glucose tolerance test. GMI is a CGM summary tool, not a diagnostic lab result.

Why did my GMI improve before my lab A1C changed?

GMI reflects the recent CGM period, often the last 14 days. Lab A1C changes more slowly because it reflects glucose attached to red blood cells over a longer window.

Do supplements that affect glucose show up in Time in Range?

They can, but CGM numbers cannot prove the supplement caused the change by themselves. If you use insulin or glucose lowering medication, watch Time Below Range carefully when adding glucose active supplements.

Sources

  1. 1. Clinical Targets for Continuous Glucose Monitoring Data Interpretation: Recommendations From the International Consensus on Time in Range (2019)
  2. 2. Glucose Management Indicator (GMI): A New Term for Estimating A1C From Continuous Glucose Monitoring (2018)
  3. 3. Glucose Management Indicator (GMI)
  4. 4. Clinical Targets for Continuous Glucose Monitoring Data (2020)

Want personalized recommendations?

Show me what works for me