CGM Variability Metrics: TIR, CV, and MAGE Interpretation

Continuous glucose monitoring (CGM) doesn’t just show your average glucose. It shows how your glucose behaves over time—how much it swings, how often it stays in range, and how quickly it changes. Those details matter because variability can drive symptoms, affect long-term outcomes, and influence how you adjust meals, insulin, or other diabetes therapies.

This guide helps you interpret three widely used CGM variability metrics: TIR (Time in Range), CV (Coefficient of Variation), and MAGE (Mean Amplitude of Glycemic Excursions). You’ll learn what each metric reflects, how to read them in real-world situations, and how to connect them to practical decisions. If you’re reviewing CGM reports—whether weekly downloads, clinic reports, or app summaries—this will help you understand what the numbers are actually saying.

Throughout, you’ll see example scenarios using realistic glucose patterns (including post-meal spikes and overnight swings). You’ll also get practical guidance on what to check next when metrics look concerning.

What CGM variability metrics are actually measuring

Before interpreting numbers, it helps to understand the logic behind them. CGM metrics are derived from thousands of interstitial glucose readings collected every 5 minutes (or sometimes more frequently depending on the device and settings). From those readings, metrics summarize different aspects of glucose control.

TIR focuses on “where” your glucose spends time—typically the percentage of time your glucose is within a target range (commonly 70–180 mg/dL, or 3.9–10.0 mmol/L). It is a time-based measure.

CV focuses on “how spread out” your glucose values are relative to the mean. It is calculated as the standard deviation divided by the mean, expressed as a percentage. It is a variability measure.

MAGE focuses on “how big” the swings are—specifically the average of the major glycemic excursions. It attempts to capture the amplitude of the largest ups and downs, filtering out minor fluctuations that may not be clinically meaningful.

When you interpret these together, you get a more complete picture: TIR tells you how much time you’re in range, CV tells you how inconsistent your glucose is overall, and MAGE tells you whether the biggest excursions are large enough to matter.

TIR (Time in Range): interpretation, targets, and common pitfalls

How TIR is calculated

TIR is expressed as a percentage of time your CGM readings fall within a defined glucose interval. The most commonly used adult target range is 70–180 mg/dL (3.9–10.0 mmol/L). Many systems also report time below range (e.g., <70 mg/dL) and time above range (e.g., >180 mg/dL), sometimes with additional thresholds such as <54 mg/dL and >250 mg/dL.

The exact calculation depends on the CGM system’s reporting rules, but conceptually it counts the proportion of readings (or time intervals) within the target band over a chosen period (often 14 days, 30 days, or a clinician-specified window).

What “good” TIR often means

Consensus targets frequently used in clinical settings include:

• ≥70% time in the 70–180 mg/dL range for many non-pregnant adults
• <4% time below 70 mg/dL
• <1% time below 54 mg/dL

For pregnancy and some specialized populations, targets differ and are often tighter.

How to interpret TIR alongside variability

TIR is not the same as “stable glucose.” You can have a moderate TIR while still experiencing large swings—especially if you spend time both above and below target. Conversely, you can have high TIR with relatively low variability if your glucose stays close to the band most of the day.

Here are several patterns you might recognize:

• High TIR, low CV: Your glucose is both mostly in range and not swinging widely. This generally aligns with good overall control.
• High TIR, higher CV: You may be in range most of the time, but your glucose is still oscillating within and near the band. This can happen with frequent meal-time peaks that still remain below 180 mg/dL.
• Moderate TIR, high MAGE: You may spend less time in range because the biggest excursions push you above 180 mg/dL or below 70 mg/dL. MAGE suggests those swings are large enough to be a key driver.
• Low TIR with frequent time below range: Hypoglycemia may be reducing TIR and also increasing variability. This requires attention to safety first.

Common pitfalls when reading TIR

Several issues can distort TIR interpretation:

• CGM wear time: If you don’t wear the sensor enough (for example, fewer than ~70% of the time over the measurement window), TIR may be less reliable. A report based on partial data can miss periods of high variability.
• Data gaps: Signal loss or calibration issues can create missing intervals. Some apps handle gaps differently.
• Different target ranges: Some reports use alternate ranges (e.g., narrower targets). Always confirm the range used for the TIR percentage.
• Time window mismatch: A 3-day snapshot can look very different from a 14-day average, especially if your routine changes or you had an illness, travel, or unusual meals.

CV (Coefficient of Variation): what it tells you and how to use it

How CV is calculated

CV is defined as:

CV (%) = (Standard deviation ÷ Mean glucose) × 100

Because it uses the mean, CV is scale-dependent. Two people can have the same standard deviation but different means, producing different CVs. That’s one reason CV is often used to gauge relative variability rather than absolute spread.

Interpreting typical CV thresholds

Many clinicians use CV as a practical indicator of overall stability. While cutoffs can vary by population and clinical context, commonly referenced adult guidance includes:

• CV < 36%: often considered relatively low variability in many CGM-based frameworks
• CV ≥ 36%: suggests higher variability and a greater risk of glycemic excursions

Some reports also flag higher CV values (e.g., 40% or above) as meaningfully unstable glucose patterns. If your CV is high, it’s not automatically a “diagnosis,” but it is a strong signal that your glucose is not staying consistently predictable.

Why CV can look “good” even when you have trouble

CV summarizes spread around the mean, not the direction or clinical significance of excursions. A few scenarios illustrate why CV can be misleading:

• Frequent small oscillations: You may have a high CV because the mean is relatively low or because small fluctuations occur often. But those fluctuations might not cross into dangerous ranges.
• One or two extreme days: A short period with a big event (illness, steroid use, missed insulin, or travel) can raise CV. If those days are not representative, you may see a misleadingly high number.
• Overcorrection cycles: If your glucose repeatedly rises and then falls due to insulin adjustments, CV may rise. But CV alone doesn’t tell you whether the major excursions are primarily high, primarily low, or both.

How to use CV with TIR and MAGE

To interpret CV effectively, you should connect it to TIR and MAGE:

• If CV is high and TIR is low, variability is likely driving time outside target. MAGE can help identify whether the biggest swings are the main issue.
• If CV is high but TIR is decent, you may be oscillating near the edges of range. MAGE can help determine whether excursions are “major” even if they don’t push far outside targets.
• If CV is moderate but MAGE is high, your overall spread may look acceptable, yet the biggest excursions are significant. This can happen when you have a generally stable baseline with occasional large spikes.

MAGE (Mean Amplitude of Glycemic Excursions): understanding major swings

What MAGE is designed to capture

MAGE aims to quantify the average amplitude of “major” glycemic excursions—meaning the largest upward and downward movements that exceed a threshold based on variability. In practical terms, MAGE tries to measure the size of the most clinically relevant swings while ignoring smaller fluctuations that may not reflect meaningful instability.

Different algorithms and software implementations can vary slightly, but the guiding principle is consistent: MAGE focuses on major changes rather than every minor rise and fall.

How to interpret MAGE values in context

MAGE is often reported in mg/dL (or mmol/L). Higher values indicate larger major excursions. If your MAGE is elevated, it suggests that your glucose is experiencing substantial swings—often tied to predictable triggers such as:

• carbohydrate-heavy meals
• missed insulin or delayed bolusing
• insulin dosing that doesn’t match absorption timing
• exercise patterns that cause post-exertion glucose changes
• overnight insulin timing issues
• stress, illness, or menstrual cycle effects (for those who experience them)

However, MAGE should not be read alone. A high MAGE can coexist with moderate TIR if you spend only limited time outside target but those times involve very large excursions. Conversely, a lower MAGE might still produce low TIR if your glucose stays just above or below target for extended periods.

Real-world scenario: interpreting a “spike day”

Imagine you review a 14-day CGM report. Your TIR is around 65% (below the common ≥70% target), your CV is 38%, and your MAGE is high for your typical baseline. When you inspect the daily graphs, you notice one pattern: several afternoons show glucose rising from ~130 mg/dL to ~230–260 mg/dL within 1–2 hours after lunch.

Even if the rest of the day is relatively steady, those large post-lunch peaks can drive:

• lower TIR (time above 180 mg/dL)
• higher CV (greater spread around the mean)
• higher MAGE (major excursions with large amplitude)

In this scenario, MAGE is especially helpful because it points to the magnitude of the biggest excursions. The next step is not “chasing a number.” It’s identifying the lunch pattern: portion size, food type (fast vs slow carbs), bolus timing, and whether your insulin-to-carbohydrate ratio or correction factor needs adjustment for that time of day.

Putting TIR, CV, and MAGE together: a practical interpretation framework

Using multiple metrics prevents you from overreacting to a single number. Here’s a practical way to interpret the combination without getting lost in math.

Step 1: Start with safety and time outside range (TIR)

Begin by checking TIR and the time below range. If you have meaningful time below 70 mg/dL or especially below 54 mg/dL, your priority is reducing hypoglycemia risk. Variability metrics often worsen when lows occur, but the clinical concern is the low glucose itself.

Step 2: Check overall instability (CV)

Next, look at CV. If CV is high, it suggests your glucose is not consistently predictable. This supports investigating factors that cause day-to-day inconsistency (sleep, meal timing, exercise variation, site issues, insulin timing, or absorption differences).

Step 3: Identify whether the biggest swings are the problem (MAGE)

Then evaluate MAGE. High MAGE suggests major excursions—large rises or drops. Those often have clearer triggers (specific meals, specific times of day, or specific insulin timing issues). If MAGE is low but TIR is still low, you may have prolonged time just above or below range rather than rare major swings.

Step 4: Use the “shape” of your CGM to connect metrics to causes

Metrics are summaries. The graphs provide mechanism. Use the following visual cues:

• Frequent sawtooth pattern: can indicate frequent small corrections, mismatched bolus timing, or rapid carb absorption.
• Post-meal peaks: often point to bolus timing (late vs early), carbohydrate counting errors, or insulin dosing mismatch.
• Overnight drift: suggests basal timing issues, bedtime snack effects, or dawn phenomenon.
• Exercise-related dips: may indicate insufficient adjustment for activity intensity and timing.
• Recovery bumps: can reflect counter-regulatory responses or overcorrection after lows.

How to interpret metrics by time of day

Many CGM reports can segment metrics by time of day (morning, afternoon, overnight) or show day/night variability. This is where TIR, CV, and MAGE become more actionable.

Morning and afternoon patterns

If your morning TIR is low, consider dawn phenomenon, breakfast composition, or overnight basal adequacy. If afternoon TIR is low, post-lunch carbohydrate load or insulin timing is a frequent culprit.

High MAGE in the afternoon often aligns with larger post-meal spikes. High CV across the entire day suggests broader inconsistency rather than a single meal-time problem.

Overnight patterns

Overnight is a special case because you are not actively managing meals, and insulin adjustments may be the main lever. If your overnight time below range is notable, that can reduce safety and also inflate variability measures.

Look for:

• gradual declines (basal too strong)
• sharp drops (insulin stacking, delayed carbs, or unexpected activity)
• rises near dawn (dawn phenomenon or insufficient late-night basal)

If overnight CV is high, it suggests glucose is changing substantially without a clear behavioral trigger. That often points to insulin timing, absorption differences, or sensor wear issues.

Common reasons variability metrics worsen (and what to check first)

When your metrics indicate instability, it’s tempting to assume something is “wrong” with your diabetes management. Often, the issue is specific and fixable. Use this checklist to interpret what might be driving TIR, CV, and MAGE changes.

Insulin timing and meal composition

Many people see post-meal spikes when bolus timing is late relative to carbohydrate absorption. Conversely, bolusing too early for slow-digesting meals can increase hypoglycemia risk.

Consider meal composition:

• fast carbs (juice, white bread) tend to raise glucose earlier
• fat and fiber slow absorption and can create delayed peaks
• mixed meals can produce multi-phase glucose curves

Correction dosing and “chasing” glucose

Frequent corrections can create cycles of rise and fall. Those cycles can increase CV and MAGE by increasing the amplitude of swings. If you see repeated high-to-low-to-high patterns, review whether corrections are being applied too aggressively or too frequently.

Sensor and data quality issues

CGM variability metrics depend on data quality. If your sensor experiences frequent signal loss, compression lows, or calibration-related drift, variability metrics can be artificially elevated.

Practical checks include:

• confirming sensor placement and avoiding pressure during sleep
• checking for consistent reading behavior compared with fingerstick checks when recommended
• ensuring adequate wear time for the reporting window

Physical activity and stress

Exercise can lower glucose during activity and also later through glycogen changes. Stress hormones can increase glucose and reduce insulin sensitivity. If you notice that variability spikes coincide with certain days, these factors can explain the changes in TIR, CV, and MAGE.

Illness, medications, and hormonal factors

Steroids, infections, and hormonal cycles can significantly affect glucose dynamics. If your metrics worsen during a specific timeframe (e.g., the week you started a new medication), interpret the metrics as reflecting those external drivers—not just day-to-day technique.

Real-world example: interpreting a conflicting set of metrics

Let’s look at a scenario where metrics don’t “agree” at first glance.

You review your CGM report and see:

• TIR (70–180 mg/dL): 72%
• CV: 34%
• MAGE: 120 mg/dL

At first, TIR and CV look reassuring. But MAGE is high, suggesting major excursions. What might that mean?

When you examine the trace, you notice that most of the time your glucose is around 110–160 mg/dL, keeping TIR high. However, once or twice per day you experience very large spikes—perhaps after dinner—where glucose rapidly rises to 240–280 mg/dL and then returns to range within a few hours.

This pattern produces:

• high TIR because the time outside range is limited
• moderate CV because the overall distribution is still centered
• high MAGE because the biggest swings are large

In this case, the practical response is to target those specific spikes—often by reviewing dinner carbohydrate estimation, bolus timing, and whether the insulin action profile matches the meal. The key point is that you shouldn’t ignore MAGE just because TIR looks good.

Guidance for action: how to respond to high variability metrics

CGM metrics are feedback. Your next steps should be systematic and safe. You may not be able to interpret a single number without context, and you should involve your clinical team when making medication changes.

Use a short, focused review window

If your report is monthly, try narrowing to 3–7 days that show the issue clearly. Look for:

• specific days with outlier spikes or lows
• consistent times of day where MAGE seems driven
• repeated meal patterns associated with excursions

Link excursions to events you can verify

When you see a major rise or drop, ask: what happened within the previous 0–3 hours? Examples include:

• What did you eat and when?
• Did you bolus on time, and was the dose what you intended?
• Was there exercise or a long walk?
• Was sleep short or disrupted?
• Any illness or stress that day?

Even if you don’t have perfect logs, patterns often emerge quickly.

Adjust one variable at a time

If you change meal size, bolus timing, and correction strategy all at once, it becomes hard to interpret whether improvements in TIR, CV, or MAGE are real or coincidental. Choose one lever to test. For example, if MAGE is driven by post-lunch spikes, focus on bolus timing for lunch for several days, then re-evaluate.

Don’t ignore hypoglycemia time

If time below range is elevated, prioritize safety. Reducing hypoglycemia often improves overall variability because lows can trigger counter-regulatory responses and correction behaviors that increase swings. This is especially relevant when CV and MAGE rise due to repeated lows and rebounds.

Consider sensor placement and wear consistency

If your CGM wear time is inconsistent, variability metrics can be less stable. A sensor that frequently loses signal may miss the start of a spike or the depth of a dip, altering TIR and variability calculations. Consistent wear improves interpretability.

Special considerations: diabetes type, therapy, and clinical context

Interpretation can differ depending on your treatment and physiological context.

Type 1 vs type 2

People using insulin—especially multiple daily injections or pumps—often see variability patterns tied to insulin timing and dosing. People using non-insulin therapies may have different variability drivers. Regardless, TIR, CV, and MAGE remain useful summaries, but the “why” behind them can vary.

Automated insulin delivery and closed-loop systems

With automated insulin delivery, CGM metrics can reflect how the algorithm responds to meal patterns and glucose dynamics. You might see improved TIR but still elevated MAGE if meals are consistently mis-timed or if absorption differs significantly from typical patterns. In such cases, reviewing how meals are entered (where applicable), and whether pre-bolus timing is appropriate, can help.

Pregnancy and tighter targets

During pregnancy, target ranges are commonly tighter and time-in-range thresholds differ. CV and MAGE can still be informative, but you should interpret them using pregnancy-specific clinical guidance.

Summary: how to interpret CGM variability metrics effectively

When you interpret CGM variability metrics TIR CV MAGE interpretation, the most useful approach is to treat each metric as a different lens:

• TIR tells you how much time your glucose is within a clinically defined target range. Start with safety by checking time below range.
• CV tells you how spread out your glucose values are relative to your mean. High CV suggests overall instability, not just one problem time.
• MAGE tells you whether the biggest excursions are large. It helps identify whether major spikes or drops are the primary driver of your variability.

In practice, you’ll get better results by connecting metric patterns to the shape of your glucose trace and to events you can verify: meals, insulin timing, exercise, sleep, stress, and illness. If you do that consistently, you can often pinpoint why variability is rising—and choose targeted, measurable changes.

Finally, remember that CGM metrics are feedback, not judgment. A single report can reflect unusual days. Look at trends over 14 days or longer when possible, ensure consistent sensor wear, and involve your clinical team when adjusting therapy.

FAQ: CGM variability metrics TIR, CV, and MAGE

What is the main difference between TIR and CV?

TIR measures time spent within a glucose range (for example, 70–180 mg/dL). CV measures variability by expressing how spread out your glucose values are relative to the mean.

Is a high CV always associated with low TIR?

Not necessarily. You can have a higher CV from frequent fluctuations that remain near the target range, keeping TIR relatively good. Conversely, low TIR can occur even with moderate CV if glucose stays just outside target for extended periods.

What does a high MAGE usually mean?

High MAGE indicates that the major glucose excursions (the largest swings) are large. This often points to specific triggers such as post-meal spikes, overnight basal mismatch, missed or mistimed insulin, or other consistent factors.

How long should you look at CGM data to interpret these metrics?

Many clinical summaries use 14 days because it smooths out day-to-day variation. Shorter windows can be informative but may be less reliable if your routine changes or you had unusual events.

Can sensor wear time change these metrics?

Yes. Lower wear time or frequent data gaps can distort TIR and variability calculations because the report may miss excursions. Consistent wear improves interpretability.

Should you focus on improving one metric only?

Usually not. The most informative strategy is to interpret metrics together: start with safety (time below range), then assess overall instability (CV), and identify whether major swings are driving the issue (MAGE). That combination helps you target the most meaningful cause.

08.01.2026. 22:15