Estimated Average Glucose And Time In Range: Two Different Summaries

Two Summaries, One Bloodstream

Estimated average glucose and time in range are not rivals, and they are not two names for one number. The first squeezes roughly three months of glucose into a single average. The second counts how much of the measured time your glucose sat inside a set range, usually 70 to 180 mg/dL. One gives you a level. The other gives you a shape. A report that carries only one of them is telling you half of a true story.

The question arrived in my family as a phone screenshot. My cousin, who lives abroad and wears a continuous glucose monitor on his arm, sent me his clinic report: page after page of percentages, a green band, a figure of 68 percent marked as his time in range. His question was plain. His clinic abroad talked about this green band. His father's clinic at home talked about one percentage called HbA1c. Were the two clinics even describing the same disease?

They were. They were summarizing the same blood on different clocks. This article puts the two summaries side by side, states what each one counts, and then runs one single day of readings through both so the difference stops being abstract.

Educational information only, not medical advice. This article explains two ways of summarizing glucose. What any summary means for one person's treatment belongs to that person's own doctor.

What Estimated Average Glucose Counts

Estimated average glucose, or eAG, is a translation. A laboratory measures HbA1c, the share of hemoglobin that carries attached glucose, and the eAG restates that percentage in the units a home meter uses. The translation formula comes from the A1c-Derived Average Glucose study, which matched HbA1c values against thousands of meter readings: eAG in mg/dL equals 28.7 times the HbA1c, minus 46.7. An HbA1c of 7.0 percent therefore reads as an eAG of 154 mg/dL, and you can run the same conversion on the HbA1c to eAG calculator on this site.

The strength of eAG is its reach. It folds in every hour of about three months: nights, weekends, the meals you remember and the ones you forgot. No diary discipline can match that coverage, because the blood itself kept the record. Its weakness is the mirror image of the same fact. An average of 154 mg/dL can come from a calm line that rarely leaves the 140s and 160s, or from a rough line that swings between 60 and 260 and passes through 154 on its way up and down. The eAG prints the same figure for both lives.

An average tells you the level of the river. It says nothing about how often the river flooded.

Families meet this weakness the first time a diary and a lab report disagree, a puzzle I worked through with my father's notebook in My Comparison Of HbA1c And Daily Meter Readings. The eAG was not wrong there. It was simply silent about the hours the notebook never saw.

What Time In Range Counts

Time in range starts from the other end. A continuous glucose monitor samples glucose every few minutes, day and night, and the report counts what share of those readings fell between 70 and 180 mg/dL. An international consensus report published in Diabetes Care in 2019 fixed that range for most adults with diabetes and set a goal of more than 70 percent of readings inside it, with under 4 percent below 70 mg/dL.

The strength of time in range is the shape. A reader sees at once how much of the day ran high, how much ran low, and how much sat in the band. Lows, which an average quietly cancels against highs, stand in their own column and cannot hide. The weakness is reach and access. The report only covers the days the sensor was worn, usually the last two weeks or so, and it depends on a device many patients do not have and many clinics cannot supply.

The percentage is built from volume. A sensor sampling every five minutes records 288 readings in a day, and a two-week report rests on roughly four thousand of them. That density is what lets the summary speak about hours instead of moments. It also explains the usual reporting window: the report describes the fortnight the sensor actually witnessed, not a season it has to infer. When my cousin's clinic prints his 68 percent, that figure is a census of thousands of small measurements, each one stamped with its hour.

Time in range also carries a blind spot of its own. A reading of 181 mg/dL and a reading of 340 mg/dL both land in the same above-range bucket. The percentage knows the day went high; it does not say how high. Two reports can show the same time in range while one person's peaks barely cross the line and the other person's peaks double it.

One Day, Read Both Ways

Take one ordinary day and ten checks spread across it: 112 on waking, 148 after breakfast, 196 mid-morning, 171 before lunch, 139 in the afternoon, 155 before dinner, 208 after dinner, 183 in the evening, 146 later on, and 128 at bedtime. The ten readings add up to 1,586. Divide by ten and the day's mean is 159 mg/dL, one point below the eAG that belongs to an HbA1c of 7.2 percent, which is 160 mg/dL.

Now read the same day as a sensor report would. Seven of the ten checks fall inside 70 to 180, so the day scores 70 percent in range, exactly on the consensus goal. Three checks, 30 percent, ran above range, and none ran below. The two summaries agree that the day was broadly controlled, and they disagree about what deserves attention. The average reader shrugs at 159. The range reader asks about the two hours after dinner.

The difference between the two pages is a difference of clocks. My cousin's sensor counted every hour of a fortnight. My father's report counts every season in one number. Both are true, and neither is the whole day: his report described two weeks in fine hourly detail, my father's described three months at a glance, and each man envied the other's page for exactly the thing his own page lacked.

What Each Summary Hides

The table below sets the two side by side on the four questions families actually ask.

Read the last row twice, because it decides how the two should be used together. The average hides variation. The percentage hides distance. A high time in range next to a high HbA1c is not a contradiction; it can mean the peaks outside the band were tall ones.

A good HbA1c next to a poor time in range can mean lows and highs are cancelling each other out. When the two disagree, the disagreement itself is the finding, and it belongs in the next appointment conversation. The post-meal hours are usually where the answer sits, as I found when I traced A Normal Fasting Pattern With A High HbA1c.

The Family Call Where This Becomes Useful

Here is the use case this comparison serves most often in my own family. A relative abroad reads out sensor percentages on a video call. Someone at home holds a lab report with one HbA1c line. Without a shared frame, the call produces two worries instead of one understanding: the relative thinks the home clinic is out of date, and the family at home thinks the sensor report is a different illness.

I now ask one question first on calls like this: which clock did that number come from? With the frame from this article, the call then takes two minutes. The HbA1c of 7.2 percent converts to an eAG of 160 mg/dL. The sensor report says 70 percent of readings sat between 70 and 180.

Both describe glucose that spends most of its time near the top of the band with excursions above it. The numbers now agree in plain language, and the remaining question, what to do about the after-dinner peak, is one question instead of an argument between two clinics.

A Note Before You Use Any Of This

Educational information only, not medical advice. This article compares two summaries of glucose data. It does not set a target for any reader, and it does not judge any single result. Time in range goals, HbA1c goals, and treatment decisions belong to the reader's own doctor, who knows the full record behind the numbers.