What Time in Range Means in Glucose Data

Time in range converts thousands of continuous glucose monitor readings into one percentage: the share that fell inside a specified interval. If a 24-hour record puts 70 percent of your readings between 70 and 180 mg/dL, you spent about 16 hours and 48 minutes in that range, assuming the data are complete and evenly sampled.

The number is useful because two people with the same average glucose can have very different patterns. One may stay near the middle, while another alternates between highs and lows. Time in range shows part of that difference, but time below range, time above range, variability, and the daily trace remain necessary.

This article explains CGM metrics rather than setting a goal for you. Your glucose targets and device actions should be individualized with a qualified care team. Severe low-glucose symptoms, confusion, seizure, loss of consciousness, persistent vomiting, or signs of ketoacidosis require urgent action under an established medical plan.

From isolated checks to a glucose movie#

A capillary blood glucose measurement captures one moment. A CGM estimates glucose in interstitial fluid repeatedly, often every few minutes. The result is closer to a movie than a snapshot.

That movie can show overnight trends, post-meal rises, exercise-related changes, and the duration of lows. It can also create information overload. Standardized metrics help clinicians and users summarize the same dataset in a consistent way.

Time in range is one of those metrics. Time above range and time below range divide readings outside the target into clinically relevant levels. Mean glucose, glucose management indicator, and coefficient of variation describe average and spread. The ambulatory glucose profile displays the distribution across a standardized day. Every one of these summaries compresses information, and what the percentage cannot tell you is whether a pattern follows your meals, your dosing, an illness, a sensor problem, your sleep, or exercise. For that you need the trace and the context.

The denominator matters#

Time in range equals readings inside the defined interval divided by all valid readings in the chosen period. Missing data can distort the result if they occur systematically. A sensor that is removed during exercise or fails overnight may omit exactly the periods most likely to differ.

The 2026 ADA Standards present 14 days and at least 70 percent active data as a basis for pattern management. Fourteen days can capture weekday and weekend variation without waiting months, and 70 percent limits the risk that a sparse sample drives the interpretation.

Those are not magic guarantees. A two-week period during illness, travel, fasting, a medication transition, or an unusual work schedule may not represent ordinary life. Longer windows can be helpful when glucose patterns vary across menstrual cycles or other recurring events. So a report should make clear the dates, the wear time, the active percentage, and whether your settings or routines changed inside the window.

The standard adult range#

For many nonpregnant adults with type 1 or type 2 diabetes, consensus documents use 70 to 180 mg/dL, equivalent to 3.9 to 10.0 mmol/L. The 2026 ADA table lists a common goal above 70 percent for many adults.

That goal is a population benchmark rather than a pass-fail line. One percentage point represents about 14 minutes per day. Movement from 55 to 60 percent may be meaningful even though it remains below 70. A person avoiding recurrent severe lows may appropriately have a more permissive upper pattern.

Ranges differ in pregnancy, where tighter targets and specialized interpretation apply. Older adults with complex health or higher treatment risk may use less stringent targets. Children, people with impaired awareness of low glucose, and those with serious comorbidities require individualized plans. Always read the range printed on your own report. Software can calculate time in a custom interval that is not the standard consensus interval.

Time below range is the safety signal#

Consensus reporting separates time below 70 mg/dL from time below 54 mg/dL. The lower level represents clinically significant hypoglycemia requiring immediate attention. For many adults, the common targets are less than 4 percent below 70 and less than 1 percent below 54, with stricter low-glucose limits in higher-risk groups.

Improving time in range by trading highs for lows is not a success. Hypoglycemia can impair thinking, coordination, driving, sleep, and heart rhythm, and severe events can cause seizure, injury, or death, so the immediate priority is often reducing lows before tightening the rest of the profile.

The timing matters. Three percent below range spread as brief borderline readings is different from a repeated prolonged overnight low. Sensor compression during sleep, lag during rapid change, and device error can mimic or distort episodes. Symptoms, confirmatory checks when instructed, and the trace help distinguish them.

Time above range has two levels#

Reports commonly separate readings above 180 mg/dL from those above 250 mg/dL. For many adults, consensus targets aim for less than 25 percent above 180 and less than 5 percent above 250.

Again, the distribution matters. A brief post-meal rise and a sustained overnight elevation can produce similar percentages but point to different questions. Persistent very high glucose may raise acute concerns, especially with illness, insufficient insulin, or ketones.

CGM does not diagnose diabetic ketoacidosis. A safety plan may include blood or urine ketone testing and specific instructions for insulin, hydration, and emergency care. Follow your own prescribed plan rather than inferring treatment from a generic target. Time above range can also reflect sensor calibration or placement issues, medication effects, stress hormones, or changes in routine. Pattern review should be curious before it is corrective.

A1C and time in range answer different questions#

A1C reflects glucose-related changes to hemoglobin over roughly two to three months, weighted toward more recent weeks. It has a large evidence base linking levels with complications in diabetes trials. It is convenient and does not require wearing a sensor.

A1C is an average and cannot show variability or timing. It can also be affected by red-cell lifespan, anemia, hemoglobin variants, kidney disease, pregnancy, transfusion, and some medicines.

Time in range provides granular recent data but covers only the wear period and depends on sensor performance. Studies, including reanalysis of Diabetes Control and Complications Trial glucose profiles, associate lower time in range with microvascular outcomes. Association and validation support the metric, but they do not make every percentage change equivalent across all populations. When A1C and CGM estimates disagree, the discrepancy is information, and it should prompt review of data quality, biology, timeframe, and assay conditions rather than forcing one value to match the other.

Glucose management indicator is an estimate#

Glucose management indicator, or GMI, converts mean CGM glucose into an A1C-like percentage through a population equation. The word indicator is important. GMI is not a laboratory A1C and need not equal it.

Differences can be consistent within an individual. One person may routinely have a laboratory A1C higher than GMI, another lower. Red-cell factors affect A1C, while sensor bias and sampling affect GMI.

GMI can help discuss what the recent CGM average would predict on average in the equation's source population, but it should not be presented as a future laboratory result or a diagnosis. Read it with its observation period attached, because a GMI drawn from two unusual weeks cannot summarize the three months behind them.

Variability adds another dimension#

Coefficient of variation expresses standard deviation as a percentage of mean glucose. Consensus reports commonly use 36 percent or lower as a marker of more stable glucose for many people, with some circumstances favoring a lower value.

Two datasets can have identical time in range yet different variability: one may cluster just inside the range; another may swing from lows to highs while spending the same total duration within it. Variability can affect hypoglycemia risk and daily burden.

Standard deviation rises with mean glucose, which is why coefficient of variation can be easier to compare across different averages. Neither statistic shows when swings occur. The ambulatory glucose profile and daily traces provide timing.

The goal is not a perfectly flat line. Meals, movement, hormones, and sleep produce normal dynamics. The clinical question is whether variation is unsafe, burdensome, or actionable.

The ambulatory glucose profile#

An ambulatory glucose profile overlays multiple days on a 24-hour clock. A median line shows the central pattern, and shaded bands show percentiles. Wide bands indicate greater variation at that time of day.

The display can reveal a repeated dawn rise, post-meal excursions, or overnight lows. It can also conceal day-specific events because several days are compressed together. Reviewing individual daily plots prevents a one-off event from being mistaken for a stable pattern.

Interpretation usually starts with data sufficiency, then lows, then recurring patterns, then variability and highs. Notes about meals, activity, medicines, sleep, and symptoms can add context, provided tracking does not become an unreasonable burden. What you want out of that review is a small number of testable questions, not an attempt to correct every fluctuation at once.

Sensor values are estimates#

CGMs measure glucose in interstitial fluid rather than directly in blood. During rapid change, interstitial and blood values can differ in timing. Algorithms translate the sensor signal, and accuracy varies across glucose levels and operating conditions.

Pressure on a sensor can cause falsely low readings. Dehydration, insertion issues, interfering substances for some devices, expired components, and connectivity loss may affect performance. Product instructions identify when a blood glucose check is needed and which substances interfere.

Regulatory authorization applies to a particular device, intended population, and instructions. Features such as dosing decisions, alerts, automated insulin delivery, and over-the-counter use differ between models and jurisdictions. A reading that conflicts with symptoms should not be ignored or automatically trusted. Follow the device's instructions and your own safety plan.

Time in range outside diabetes care#

CGM use among people without diabetes has increased, but standardized targets and proven benefits are less established. Healthy glucose varies with food, exercise, stress, sleep, and measurement noise. Applying diabetes thresholds can medicalize ordinary variation.

A transient rise after eating does not diagnose insulin resistance or predict disease by itself. Consumer applications may assign proprietary “scores” without prospective evidence that optimizing the score improves patient-important outcomes.

Research may clarify roles in selected populations, but an appealing graph is not proof of benefit. Ask whether the device is authorized for the use you have in mind, whether the thresholds are validated, whether acting on them improves outcomes, and whether the watching is costing you anxiety, restriction, or unnecessary care. And screening for diabetes still follows validated clinical criteria, not a sensor trace you interpreted yourself.

Turning data into safer questions#

Begin with whether you have enough representative data. Confirm the target range and the units. Review time below range and any severe episodes first, then look at the ambulatory profile and the individual days for periods that keep recurring.

Ask what changed: medication timing, missed or delayed doses, meals, alcohol, activity, sleep, menstrual cycle, illness, travel, sensor placement, or device settings. Treat these as hypotheses, not explanations proven by correlation.

Shared review can turn a pattern into a bounded experiment with a follow-up date. Medication changes, especially insulin or drugs that can cause hypoglycemia, should follow qualified clinical guidance. And count the right kind of success: fewer severe lows, less alarm burden, more predictable nights, or less distress, not only a higher time-in-range number.

A percentage with a purpose#

Time in range makes dense CGM data easier to discuss and compare, and its value comes from standard definitions and its connection to the underlying trace, not from turning a day into a score.

The safest interpretation names the range, period, data completeness, lows, highs, variability, and clinical context. It combines CGM with symptoms, A1C, treatment risk, and your own priorities. Used that way, the metric supports better questions without pretending that one percentage can describe an entire life with diabetes.

Sources#

The metadata sources include the 2026 ADA Standards, international CGM consensus reports, and validation studies. Device-specific instructions and individualized clinical guidance remain essential.