CGM Interpretation: Time in Range, Targets & What They Mean

The core question a CGM report answers

A CGM does not just tell you "your glucose is high" or "your glucose is low." It answers a more specific question: how much of your day sits inside a safe range, how far outside that range do you go, and how much does your glucose bounce around while getting there. Two people can have the identical average glucose and completely different risk profiles, because average glucose hides variability and hides hypoglycemia. The useful question when reading a CGM report is not "what is my average glucose" but "how much time am I spending outside range, and how volatile is my glucose while I'm in it."

That framing, not the single number on the home screen, is what this guide is built around.

What the device is measuring, and why there's a lag

A CGM measures glucose concentration in interstitial fluid (the fluid between cells), not directly in blood. Interstitial glucose trails capillary blood glucose by roughly 5 to 15 minutes, and that lag widens during rapid glucose changes such as a post-meal spike or an exercise-driven drop. A thin filament sensor under the skin generates an electrical signal proportional to glucose concentration; modern factory-calibrated sensors (for example, Dexcom G7 or Abbott FreeStyle Libre 3) do not require routine fingerstick calibration, while older models sometimes did.

Each sensor produces hundreds of readings per day, but no single reading is very informative on its own. The clinical value comes from the pattern across many days. International consensus recommendations on interpreting CGM data recommend a minimum of roughly 14 days of wear with adequate sensor active time before drawing conclusions about someone's typical glucose pattern. A single unusual day, an illness, a missed dose, a birthday dinner, can distort a shorter window.

Most CGM software converts raw readings into an Ambulatory Glucose Profile (AGP), a standardized report format used across the diabetes technology field that compresses two weeks of data into a single 24-hour overlay.

Time in Range: the metric with the clearest outcome link

Time in Range (TIR) is the percentage of the day glucose stays between 70 and 180 mg/dL. Among CGM-derived metrics, TIR has the most direct evidence connecting it to long-term complications: analyses of continuous glucose data linked to outcomes in people with type 1 diabetes have found that higher TIR is associated with lower rates of retinopathy progression and microalbuminuria over time. The exact magnitude of that association (for example, a specific percentage risk reduction per 10-point TIR increase) is a number worth verifying against the primary trial publication before it is quoted precisely in a clinical conversation; the qualitative relationship (more time in range, less microvascular complication risk) is well established, but citing a precise multiplier here without re-checking the source paper risks overstating precision.

International consensus targets for most adults with type 1 or type 2 diabetes are:

• TIR (70 to 180 mg/dL): above 70%
• Time below range (TBR, under 70 mg/dL): under 4%
• Time significantly below range (under 54 mg/dL): under 1%
• Time above range (TAR, above 180 mg/dL): under 25%
• Time significantly above range (above 250 mg/dL): under 5%

For older adults or anyone at elevated hypoglycemia risk, the TIR target relaxes to above 50%, and the time-below-range target tightens further. Pregnancy uses a different and tighter window (roughly 63 to 140 mg/dL) with its own TIR target. These relaxed and pregnancy-specific targets come from the same international consensus process and should be confirmed with a treating clinician rather than applied generically, since individual risk factors change which target applies.

The American Diabetes Association's Standards of Care treats time in range as a legitimate outcome measure for glycemic control, not merely a research curiosity, and current guidance (2024 edition) is available directly from the ADA's Standards of Care supplement. Because guideline language and target thresholds are periodically revised, anyone titrating therapy off these numbers should check the current edition rather than relying on a cached summary. (Diabetes Care, ADA Standards of Care supplement)

Reading the Ambulatory Glucose Profile (AGP)

The AGP compresses roughly two weeks of readings into one 24-hour graph: a median line and shaded bands showing the middle 50% and middle 90% of readings at each time of day. Reading it systematically:

Median line. A median that sits between roughly 100 and 140 mg/dL throughout the day suggests reasonable average control. A median consistently above 160 mg/dL suggests sustained hyperglycemia that likely warrants a conversation about therapy.

Interquartile band (25th to 75th percentile). This captures "typical" variability. If this band stays largely within 70 to 180 mg/dL, TIR targets are probably being met on most days.

Outer band (5th to 95th percentile). This captures extremes. A lower band dipping below 54 mg/dL means clinically significant hypoglycemia is occurring on at least some days, even if it is not every day. An upper band regularly exceeding 250 mg/dL means meaningful time in a range associated with acute metabolic stress.

The daily-traces view underneath the AGP shows each day separately, which is useful for isolating an outlier day (illness, a missed dose, an unusual meal) that might otherwise distort the aggregate statistics.

Glucose Management Indicator (GMI): an estimate, not a lab value

GMI is a formula that converts mean CGM glucose into an A1c-equivalent estimate:

GMI (%) = 3.31 + (0.02392 × mean glucose in mg/dL)

A mean CGM glucose of 154 mg/dL yields a GMI of roughly 7.0%. GMI and lab-measured HbA1c commonly track each other but do not always agree closely, and the gap is not random noise; it reflects real physiology. Red blood cell lifespan, hemoglobin variants, iron deficiency, and chronic kidney disease can all shift lab HbA1c away from what glucose exposure alone would predict. When GMI and a lab HbA1c disagree by a meaningful margin, that disagreement itself is diagnostically useful and worth raising with a clinician rather than assuming one number is simply wrong.

A decision framework for reading your own CGM report

When you check your CGM readings, a single percentage or pair of numbers can be difficult to interpret on its own. Should you continue your current approach, or is it time to reach out to your care team? The patterns and actions outlined here connect typical CGM metrics to practical next steps. These suggestions serve as a foundation for discussing your results with a clinician rather than replacing that conversation, and they're most meaningful when based on at least 10 to 14 days of collected readings.

The pattern to avoid: looking only at the single "average glucose" or "GMI" number on the home screen and concluding things are fine. A stable-looking average can sit on top of dangerous swings, and a slightly elevated average with low variability is a very different clinical situation than the same average built from wide daily swings between hypoglycemia and hyperglycemia.

Coefficient of variation: measuring how much glucose swings, not just where it sits

Coefficient of variation (CV) is standard deviation divided by mean glucose, expressed as a percentage. Consensus recommendations commonly cite roughly 36% as a threshold: at or below that level, glycemia is considered relatively stable; above it, variability is considered high and is associated with greater hypoglycemia risk in people using insulin or insulin secretagogues. This threshold comes from analyses of CGM datasets in people with diabetes and should be treated as a useful reference point rather than a hard biological cutoff specific to any one individual.

For comparison, published population data on adults without diabetes wearing CGM has generally shown mean glucose in the high 90s mg/dL with a CV meaningfully below the 36% threshold used in diabetes care, though formal CV targets for people without diabetes have not been established by major endocrine societies. Anyone using CGM purely for wellness tracking without a diabetes diagnosis should read that comparison as descriptive context, not a clinical target.

What sustained high readings usually mean, and what doesn't require action

Sustained readings above 180 mg/dL indicate hyperglycemia. An isolated post-meal spike to 180 or 190 mg/dL that resolves within roughly two hours is common and not itself alarming, including in people without diabetes. The concern is sustained elevation, or peaks that regularly exceed 250 mg/dL.

Common, evidence-consistent causes of persistent high readings include:

• Insufficient medication dosing. In type 2 diabetes this is often the most common driver, and a TAR consistently above 25% is a reasonable trigger to review medication with a clinician.
• Dawn phenomenon. A pre-waking hormonal surge (cortisol, growth hormone) can raise fasting glucose. CGM often makes this visible for the first time.
• Meal composition and timing. Refined-carbohydrate-heavy meals eaten without protein, fat, or fiber tend to produce sharper spikes; some trial data suggests eating order (protein/vegetables before refined carbohydrate) can blunt the post-meal peak, though the exact magnitude reported in any single study should be checked against the primary paper before being quoted as a fixed percentage.
• Stress and illness. Cortisol and inflammatory signaling raise glucose independent of food intake.
• Sensor compression artifact. Lying on the sensor overnight can produce a false low followed by a rebound "spike" that is an artifact of pressure on the sensor, not a true glucose excursion; it typically shows as a sharp V-shape.

Lifestyle interventions with reasonably consistent supporting evidence, most notably a short walk after meals, have been associated with meaningfully lower post-meal glucose peaks in trial data; the precise percentage reduction varies by study population and should not be treated as a guaranteed personal result. Medication changes, meal composition adjustments, and sleep are the next tiers, ideally discussed with a clinician rather than self-titrated.

What low readings mean, and when it's an emergency

Glucose below 70 mg/dL is classified as Level 1 hypoglycemia. Below 54 mg/dL is Level 2, clinically significant hypoglycemia requiring prompt treatment. Below 40 mg/dL with altered consciousness, confusion, or inability to safely swallow is Level 3, a medical emergency requiring urgent care or emergency services, not a wait-and-see approach.

CGM's biggest practical advantage over fingersticks is the trend arrow: it can show glucose falling before it becomes dangerous. A single down arrow generally indicates a moderate rate of fall; two down arrows indicate a faster fall that could produce a large drop within roughly 30 minutes if untreated. Randomized trial evidence in type 1 diabetes has shown that access to real-time CGM trend data reduces time spent in hypoglycemia compared with fingerstick-only monitoring, because people can intervene before reaching a dangerous level rather than after.

Common causes of low readings:

• Excess insulin or sulfonylurea dosing relative to carbohydrate intake or activity
• Delayed meals after rapid-acting insulin
• Exercise, especially prolonged aerobic activity, which can raise insulin sensitivity for many hours afterward
• Alcohol, which suppresses the liver's glucose output

Do not adjust insulin or sulfonylurea doses based on CGM patterns without clinician guidance. Nighttime hypoglycemia in particular carries real safety risk and a pattern of overnight lows is worth an unscheduled conversation with your care team, not a self-directed dose change.

What "normal" looks like without a diabetes diagnosis

Published CGM studies in adults without diabetes have generally reported the large majority of readings falling between roughly 70 and 140 mg/dL, with mean glucose in the high 90s mg/dL and post-meal peaks rarely exceeding 140 mg/dL. No major endocrine society, including the Endocrine Society, has published formal CGM interpretation targets for people without diabetes. Descriptive ranges sometimes used informally for wellness tracking, fasting glucose 70 to 100 mg/dL, post-meal peak under 140 mg/dL returning to baseline within two hours, TIR (70 to 140 mg/dL) above 90%, come from observational data in healthy populations. They are not clinical guidelines, and treating them as diagnostic cutoffs for someone without diabetes risks medicalizing normal glucose variation. A single elevated reading in a non-diabetic CGM user is far more often normal physiology than early disease.

When to bring your data to a clinician versus manage it yourself

A single spike after an unusual meal is not a clinical event. A repeating pattern, post-lunch readings above 200 mg/dL for several days running, recurring overnight lows, a CV that stays above 36% despite an acceptable average, is a signal worth a scheduled conversation.

Clinical practice guidance from diabetes technology societies generally recommends reviewing CGM data with a clinician every few weeks during active medication titration and at longer intervals once targets are stable. Bring at least 14 days of AGP data to that conversation, and ask specifically about TIR, time below range, time above range, and CV, not just the average glucose or GMI, since the average can look acceptable while masking dangerous swings.

For people using CGM as part of an automated insulin delivery (hybrid closed-loop) system, the algorithm adjusts insulin delivery using CGM trend data in real time, and trial evidence in type 1 diabetes has shown meaningful TIR improvement with these systems compared with sensor-augmented pump therapy without automation. For adults with type 2 diabetes not on insulin, CGM data can still usefully guide lifestyle and medication discussions even from a short trial period, though insurance coverage for CGM in this population varies by payer and has been changing; anyone relying on coverage details should verify current policy with their plan rather than assume prior-year rules still apply.

Evidence boundary: what's established, what's plausible, what isn't

Established: Time in range (70 to 180 mg/dL) is a validated, guideline-endorsed metric for people with diabetes, with international consensus targets for TIR, time below range, and time above range. The GMI formula is a defined, consensus-derived calculation, not a marketing claim. Real-time CGM trend data measurably helps people avoid hypoglycemia compared with fingerstick-only monitoring.

Plausible but not fully quantified on this page: The exact size of the relationship between a given TIR increase and a specific reduction in retinopathy or microalbuminuria risk; the precise percentage benefit of specific eating-order or post-meal-walking interventions. The qualitative direction of these effects is well supported in the literature; the precise multipliers vary by study and population and should be verified against the specific primary paper before being used in patient-facing dosing or lifestyle counseling.

Not established: Formal CGM interpretation targets for people without diabetes. Descriptive "normal ranges" for non-diabetic CGM users are observational reference points, not clinical thresholds, and should not be used to diagnose or medicate.

References

American Diabetes Association Professional Practice Committee. Standards of Care in Diabetes. Diabetes Care. 2024;47(Suppl 1). https://diabetesjournals.org/care/issue/47/Supplement_1

Other studies referenced in this article (international consensus on time in range, GMI derivation, CV variability thresholds, IMPACT and Control-IQ trial data, and CGM profiles in adults without diabetes) are well known in the diabetes technology literature, but their specific identifiers were not independently re-verified for this draft. An editor or reviewer should confirm each citation against the primary paper before publication and attach verified links.