Continuous Glucose Monitor (CGM) Longevity-Medicine Target Ranges

Updated Jan 28, 2025

Image: HealthRX.com clinical illustration

A continuous glucose monitor (CGM) is a wearable sensor, such as the Dexcom G7, Abbott Libre 2/3, Dexcom Stelo, or Abbott Lingo, that measures interstitial fluid glucose every few minutes. Dexcom G7 and Libre 2/3 are prescription devices approved for people with diagnosed diabetes. Stelo and Lingo are over-the-counter (OTC) sensors the FDA cleared in 2024 for adults without diabetes, primarily for general wellness and pattern tracking rather than diagnosis (FDA, 2024) (source).

This is the question this page actually answers: not "what is the optimal CGM number," but "how much certainty backs each commonly cited number, and which of them should change what a healthy adult actually does." No FDA label, ADA guideline, or published consensus statement defines an official CGM target range for longevity optimization in people without diabetes. The numeric targets circulating in longevity-medicine practice (time-in-range 70-140 mg/dL, coefficient of variation below 36% or 30%, fasting glucose 72-85 mg/dL) are extrapolations from diabetes-guideline thresholds and population glucose-mortality associations, not numbers validated by an outcome trial in metabolically healthy adults. That distinction should shape how much weight a reader puts on any single CGM number.

What is actually established

The American Diabetes Association's Standards of Care define time-in-range for people with diagnosed diabetes as greater than 70% of readings between 70 and 180 mg/dL, and set fasting glucose diagnostic categories (normal, prediabetes 100-125 mg/dL, diabetes above 126 mg/dL) (ADA Standards of Care). These are diagnostic and diabetes-management thresholds, not longevity-optimization targets, and they do not apply directly to someone without diabetes.
The FDA has cleared OTC CGM sensors for use by adults without diabetes as a general wellness tool. The clearance addresses device safety and measurement performance, not a specific target range for non-diabetic users (FDA, 2024).
HbA1c reflects an average glucose over roughly three months and can be distorted by red-cell turnover, hemoglobin variants, anemia, and kidney disease. CGM captures moment-to-moment variation that a single fasting draw or an HbA1c cannot, which is why postprandial spikes and nocturnal patterns can be present even when HbA1c is normal. This is a well-accepted physiological point, but it is a mechanistic explanation, not proof that correcting a given CGM number extends lifespan.
Glycemic variability (large swings in glucose independent of the average) is a well-studied index, and its association with vascular stress markers is a recurring theme in the mechanistic and observational literature. Whether reducing variability by a specific amount changes hard outcomes such as cardiovascular events or mortality in a person without diabetes has not been established by trial evidence.

What newer research adds, and its real population

Two recent studies are directly relevant to the longevity-and-glucose-variability question, but neither was conducted in the population most CGM-for-wellness users represent (a healthy, non-diabetic adult tracking day-to-day metabolic health).

A 2026 secondary analysis of the HYPOAGE study examined CGM-based metrics and mortality in older adults with type 2 diabetes treated with insulin (HYPOAGE secondary analysis). This population already has diagnosed diabetes and insulin-related hypoglycemia risk, so its findings about which CGM metrics track mortality risk in that group cannot be assumed to transfer to a 40-year-old without diabetes optimizing glucose for longevity. It is nonetheless useful evidence that CGM-derived metrics, not just HbA1c, carry independent mortality signal in an older, insulin-treated diabetic cohort, which supports the broader premise that CGM captures something HbA1c misses, without validating any specific non-diabetic target number.

A 2022 study on glycemic variability and longevity in Chinese centenarians reported associations between glycemic variability indices and survival to extreme old age (centenarian study). This is a descriptive, observational study in an exceptional population selected for having already reached age 100; it demonstrates that variability indices differ in centenarians compared with younger comparison groups, but it cannot be used to derive an actionable CGM target for a middle-aged adult today. Both studies should be read as directional support for "variability matters, not just the average," not as the source of a specific cutoff.

Reading a CGM report: the core metrics

A standard CGM report (often called an Ambulatory Glucose Profile) summarizes several metrics over a wear period:

Time-in-range (TIR): percentage of readings within a chosen glucose window
Mean glucose: the average of all sensor readings over the wear period
Coefficient of variation (CV): standard deviation divided by the mean, expressed as a percentage; a measure of how much glucose swings around its average
Time above range / time below range: percentage of readings above or below the chosen window, with time below 70 mg/dL flagged separately because of hypoglycemia risk

For people with diagnosed diabetes, ADA guidance frames these around the 70-180 mg/dL window with a TIR goal above 70%. Longevity-medicine practitioners commonly tighten the window to 70-140 mg/dL and aim for TIR above 90%, and commonly cite a CV below 36% (a threshold used in diabetes-management consensus documents) or a stricter unofficial target below 30%. These tighter numbers are reasonable extrapolations, not independently validated non-diabetic targets, and a reader should treat them as practice heuristics rather than diagnostic cutoffs.

A note on sensor accuracy

OTC and prescription CGM sensors measure interstitial fluid glucose, which lags blood glucose by several minutes, and every sensor carries a margin of error relative to a laboratory blood draw (commonly summarized as mean absolute relative difference, or MARD). Readings a few points above or below a stated target should be interpreted as estimates, not precise diagnostic values. Known interference sources include acetaminophen and high-dose vitamin C with some sensor chemistries, and "compression lows," falsely low readings from sleeping on an arm-worn sensor, which can be mistaken for real hypoglycemia.

CGM signal triage framework: what pattern, what it plausibly means, what to do next

This framework is meant to help a non-diabetic CGM user (or their clinician) decide whether a pattern on a report is likely benign, worth a dietary adjustment, or worth an actual lab workup. It does not replace clinical judgment, and it assumes at least 14 days of wear with reasonable data coverage; shorter or gappy wear periods make every metric below less trustworthy.

Observed pattern	Plausible explanation	Certainty	Reasonable next step
Wear period under 70% data coverage	Sensor errors, early removal, poor adhesion	Established data-quality issue	Re-wear before drawing any conclusion from TIR, CV, or mean glucose
Mean glucose consistently above ~105 mg/dL over 14 days	Reduced insulin sensitivity, high-glycemic diet, or early beta-cell change	Plausible, needs confirmation	Fasting insulin and HOMA-IR; consider a formal oral glucose tolerance test
Postprandial peaks above 140 mg/dL on a large share of meals	Meal composition, carbohydrate load, or delayed insulin response	Plausible, meal-context dependent	Review meal order and composition first; if it persists across varied meals, discuss OGTT with insulin levels
CV persistently above 36% with adequate, varied food intake	Sleep disruption, alcohol, undiagnosed sleep apnea, cortisol dysregulation	Plausible, several competing causes	Review sleep and alcohol first; if unresolved, consider sleep apnea screening and cortisol assessment before assuming a glucose-specific problem
Time below 70 mg/dL above roughly 1-2% of readings, no diabetes diagnosis	Aggressive carbohydrate restriction, alcohol, missed meals	Established hypoglycemia signal at the device level	Do not treat low readings as automatically "good"; review diet pattern and consider medical evaluation if lows recur, especially overnight
Single post-exercise spike to 130-160 mg/dL resolving within 60-90 minutes	Catecholamine-driven glycogenolysis after resistance training or HIIT	Established, generally benign	No action needed; this is an expected response in insulin-sensitive people
Post-exercise spike still above ~140 mg/dL after 2+ hours	Possible impaired post-exercise insulin action	Plausible, not diagnostic on its own	Track recurrence across sessions before treating as meaningful
Fasting reading above 95 mg/dL on most nights of a 14-day wear	Hepatic insulin resistance, sleep apnea, dawn phenomenon	Plausible, requires lab confirmation	Fasting insulin/HOMA-IR; ask about sleep quality and snoring; do not self-diagnose prediabetes from CGM alone
Any reading pattern accompanied by symptoms (confusion, palpitations, sweating, chest pain)	Could reflect true hypoglycemia or an unrelated acute problem	Clinical judgment required	Seek medical evaluation; a CGM number should never override how a person actually feels

The recurring rule underneath this table: CGM is a screening and pattern-recognition tool, not a diagnostic test. A single number, even one repeated across several days, is a prompt to look further (diet, sleep, a confirmatory lab test), not a stand-alone diagnosis of insulin resistance or prediabetes.

Where CGM data should prompt further testing, and where it should not

Reasonable, conventional next steps when a longevity-focused non-diabetic CGM report looks abnormal include fasting insulin and HOMA-IR, and a two-hour oral glucose tolerance test with insulin levels when postprandial peaks are frequent or fasting values are persistently elevated. These are standard endocrinology tools, not CGM-specific inventions. What CGM data should not be used for is a stand-alone diagnosis of prediabetes or diabetes; diagnosis still requires laboratory-grade testing (fasting plasma glucose, OGTT, or HbA1c) interpreted by a clinician.

Anyone using CGM alongside a GLP-1 receptor agonist (semaglutide, tirzepatide) or an SGLT-2 inhibitor should know these medications change glucose patterns in ways that can make CGM trends harder to interpret without clinical guidance, and should not adjust dosing based on CGM numbers alone.

Urgent care or emergency evaluation is appropriate for symptomatic hypoglycemia (confusion, loss of coordination, seizure), not for an isolated low CGM reading without symptoms, which is more often a sensor artifact (compression low) or a benign response to fasting or exercise.

Evidence boundary

Established: ADA diagnostic thresholds and diabetes-management TIR targets exist and are guideline-backed. The FDA has cleared specific OTC CGM sensors for non-diabetic wellness use. CGM detects postprandial and nocturnal patterns that a single fasting draw or HbA1c cannot. Glycemic variability is measurable and is an active area of cardiometabolic research.

Plausible but unproven: That tightening TIR, CV, or fasting-glucose targets below ADA diabetic thresholds produces meaningfully lower cardiovascular, cognitive, or all-cause mortality risk in adults without diabetes. Specific numeric targets circulating in longevity-medicine practice (70-140 mg/dL TIR above 90%, CV below 30%, fasting 72-85 mg/dL) are reasonable clinical heuristics extrapolated from diabetic-guideline and epidemiological data, not numbers derived from an outcome trial in healthy adults.

Not established: That any single CGM metric, met or missed, predicts an individual's longevity outcome. That CGM-guided behavior change (food order, exercise timing, ketogenic diets) produces durable outcome benefits beyond glucose numbers themselves in non-diabetic populations. Readers should treat exact percentage or mg/dL claims from secondary sources with caution unless they can trace them to a specific guideline or trial.

Frequently asked questions

Is there an official CGM target range for people without diabetes?

No. The ADA's Standards of Care define target ranges for people with diagnosed diabetes. The FDA has cleared over-the-counter CGM sensors for non-diabetic wellness use but has not published a target range for that use. Numbers used in longevity-medicine practice, such as time-in-range 70-140 mg/dL or fasting glucose 72-85 mg/dL, are extrapolations rather than validated non-diabetic targets.

How is CGM different from a fasting glucose blood test?

A fasting blood draw is a single snapshot. CGM produces a continuous trace throughout the day and night, which can reveal postprandial spikes, nocturnal patterns, and variability that one blood draw cannot. This is a well-established capability of the device; it does not mean every pattern CGM reveals requires treatment.

What does coefficient of variation (CV) tell me that mean glucose does not?

Two people can have the same average glucose while one stays steady and the other swings widely. CV captures that swing. A CV below 36% is used in diabetes-management consensus documents as a marker of relatively stable glycemia; whether a stricter target changes outcomes in non-diabetic adults has not been established by trial evidence.

My fasting CGM reading is often around 95 mg/dL. Is that a problem?

A value in the mid-90s is within the ADA's normal fasting range but above the tighter targets some longevity clinicians prefer. Persistent elevation on most nights of a 14-day wear is a reasonable prompt to check fasting insulin and HOMA-IR with a clinician, not a diagnosis on its own.

How accurate are over-the-counter CGM sensors?

FDA-cleared OTC sensors have published accuracy data versus laboratory reference values, generally described as sufficient for trend and pattern monitoring rather than precise diagnostic thresholds. Readings should be treated as estimates with a real margin of error, particularly around exact cutoffs.

Does exercise affect CGM readings?

Yes. Aerobic exercise typically lowers glucose during and after activity. Resistance training and high-intensity intervals can transiently raise glucose through catecholamine release, which is a normal, generally benign response in insulin-sensitive people that resolves within roughly 60 to 90 minutes.

References

American Diabetes Association Professional Practice Committee. Standards of Care in Diabetes. Diabetes Care. https://diabetesjournals.org/care/article/47/Supplement_1/S1/153954
U.S. Food and Drug Administration. FDA Clears First Over-the-Counter Continuous Glucose Monitor. FDA News Release, 2024. https://www.fda.gov/news-events/press-announcements/fda-clears-first-over-counter-continuous-glucose-monitor
CGM-based metrics and mortality in older adults living with type 2 diabetes on insulin therapy: a secondary analysis of the HYPOAGE study (2026). https://pubmed.ncbi.nlm.nih.gov/42303126/
Effects of Variability in Glycemic Indices on Longevity in Chinese Centenarians (2022). https://pubmed.ncbi.nlm.nih.gov/35879983/

Several numeric claims common in longevity-medicine sources on this topic (exact percentages from specific trials on food order, time-restricted eating, and ketogenic-diet reversal rates) could not be verified against a confirmed primary source for this draft and have been removed or generalized. These should be reintroduced only after a qualified reviewer confirms the underlying paper and its population.