Should I Use a CGM? Blood Sugar Targets, Danger Zones, and Morning Highs Explained

What Is a CGM and How Does It Work?

A continuous glucose monitor (CGM) is a small wearable sensor, usually placed on the upper arm or abdomen, that measures interstitial glucose every one to five minutes and transmits readings wirelessly to a phone or receiver. Unlike a fingerstick that gives one data point, a CGM generates a trend line showing exactly how fast glucose is rising or falling. The FDA cleared the first CGM in 1999, and the technology has since advanced through multiple generations. [1]

Current options include the Dexcom G7 (30-minute warm-up, 10-day wear, approved for use without fingerstick confirmation in most situations), the Abbott FreeStyle Libre 3 (60-minute warm-up, 14-day wear, factory-calibrated), and the Medtronic Guardian 4 (used primarily with Medtronic insulin pumps in closed-loop systems). [2] Each device measures glucose in the fluid between cells rather than directly in blood, which means readings may lag whole-blood capillary glucose by roughly five to ten minutes during rapid excursions. The FDA notes this lag is clinically meaningful during exercise or after rapid carbohydrate intake. [3]

Sensor accuracy is reported as mean absolute relative difference (MARD). The Dexcom G7 achieved a MARD of 8.2% in its key trial (N=116), which meets the FDA threshold of <10%. [4] The Libre 3 reported a MARD of 7.8% in adults. [5]

Should I Use a CGM? Who Benefits Most

The 2024 American Diabetes Association Standards of Care state that real-time CGM "should be offered" to all adults with type 1 diabetes regardless of insulin regimen, and to all adults with type 2 diabetes using multiple daily injections or insulin pump therapy. [6] The ADA also notes that CGM "may be beneficial" for people with type 2 diabetes on basal insulin only or non-insulin therapies, particularly those with recurrent hypoglycemia or suboptimal glycemic control.

Beyond diagnosed diabetes, the picture is changing. The SIGNAL trial (N=153, published 2023) found that people with prediabetes who wore a CGM for 12 weeks and received coaching based on their data achieved significantly greater reductions in post-meal glucose excursions compared with standard care, with a between-group difference of 14.6 mg/dL at two hours after meals. [7] A separate 2022 analysis in JAMA Internal Medicine (N=40 adults without diabetes) documented that 15% of participants spent more than 15% of the 24-hour period above 140 mg/dL on a standard Western diet, a threshold most clinicians would consider abnormal. [8]

CGM use also appears to reduce hypoglycemic events. The GOLD trial (N=161 adults with type 1 diabetes) reported that real-time CGM reduced severe hypoglycemia by 72% compared with self-monitoring of blood glucose alone (3.0 vs. 10.8 events per 100 person-years, P<0.001). [9]

People who tend to benefit most include those who are insulin-dependent, those with hypoglycemia unawareness (the inability to feel low blood sugar symptoms until glucose drops below 54 mg/dL), those managing gestational diabetes, and anyone trying to understand how specific foods or exercise patterns affect their glucose curve. People with type 2 diabetes on oral medications only who have consistently stable A1c readings at goal may find fingerstick testing sufficient, though the convenience of real-time trending is undeniable even in that group.

What Is a Normal Blood Sugar Level?

The ADA defines normal fasting plasma glucose as 70 to 99 mg/dL, prediabetes as 100 to 125 mg/dL, and diabetes as a fasting level at or above 126 mg/dL confirmed on two separate occasions. [10] Two-hour post-meal (postprandial) glucose below 140 mg/dL is considered normal; 140 to 199 mg/dL indicates impaired glucose tolerance; at or above 200 mg/dL is consistent with diabetes. [10]

For people already managing diabetes, the ADA's 2024 Standards set individualized targets but note that a reasonable A1c goal for most non-pregnant adults is <7.0%, which corresponds to an estimated average glucose (eAG) of approximately 154 mg/dL. [6] More stringent targets (A1c <6.5%) may be appropriate for selected patients with short disease duration, long life expectancy, and no significant cardiovascular disease, provided they can be achieved without substantial hypoglycemia. Looser targets (<8.0%) are appropriate for older adults with multiple comorbidities or limited life expectancy.

CGM-based targets differ from A1c targets. The ADA recommends that most adults with type 1 diabetes spend more than 70% of the 24-hour period in the Time in Range of 70 to 180 mg/dL, less than 4% below 70 mg/dL, and less than 1% below 54 mg/dL. [6] These thresholds were validated against A1c in the large CLARITY study (N=2,571), which showed that each 5% increase in Time in Range correlated with an approximately 0.33% decrease in A1c. [11]

What Is a Normal A1c?

Hemoglobin A1c measures the percentage of hemoglobin molecules that have glucose attached, reflecting average blood glucose over the preceding two to three months. A normal A1c is below 5.7%. An A1c between 5.7% and 6.4% defines prediabetes, and an A1c at or above 6.5% on two occasions confirms a diagnosis of type 2 diabetes. [10]

A1c has practical limitations worth knowing. Conditions that shorten red blood cell lifespan, including hemolytic anemia, iron deficiency anemia treated with iron infusions, and certain hemoglobin variants (HbS, HbC, HbE), can falsely lower or raise A1c. In these situations, fructosamine or continuous glucose monitoring data provide better glycemic assessment. [12] Pregnancy also falsely lowers A1c because red blood cell turnover accelerates; the ADA recommends a target of <6.0% in the first trimester and <6.5% later in pregnancy, relying heavily on CGM data rather than A1c alone. [6]

The DCCT trial (N=1,441) established the foundational evidence: reducing A1c from approximately 9.0% to 7.0% over 6.5 years reduced the risk of diabetic retinopathy by 76%, nephropathy by 50%, and neuropathy by 60% in people with type 1 diabetes. [13] The UKPDS (N=5,102) demonstrated similar microvascular protection in type 2 diabetes, with each 1% reduction in A1c associated with a 37% decrease in microvascular complications. [14]

What Is a Dangerous Blood Sugar Level?

"Dangerous" depends on both direction and rate of change. On the low end, the ADA classifies hypoglycemia in three levels. Level 1 is glucose below 70 mg/dL (requires carbohydrate treatment). Level 2 is below 54 mg/dL (requires immediate action). Level 3 is severe hypoglycemia, defined as any episode requiring assistance from another person, regardless of glucose value. [6] Symptoms at Level 2 include confusion, seizure, and loss of consciousness. Untreated severe hypoglycemia can cause cardiac arrhythmia and death. [15]

On the high end, glucose above 250 mg/dL in someone with type 1 diabetes (or insulin-deficient type 2) warrants checking for urine or blood ketones. If large ketones are present, diabetic ketoacidosis (DKA) is likely and requires emergency care. [16] Glucose above 600 mg/dL with severe dehydration, altered consciousness, and an absence of significant ketones points toward hyperosmolar hyperglycemic state (HHS), which carries a hospital mortality rate between 5% and 20% in older adults. [17] Any glucose reading above 400 mg/dL should prompt a call to a clinician or an urgent care visit, even without obvious symptoms.

For people without diabetes, glucose rarely rises above 140 mg/dL postprandially, and the glucose nadir during an overnight fast is generally above 65 mg/dL. Readings outside those ranges in someone not on glucose-lowering medications warrant formal evaluation.

Why Do I Get Morning Highs?

Morning glucose elevations occur through at least two distinct mechanisms, and many people experience both simultaneously. Understanding which one is driving your morning numbers changes the management strategy entirely.

The first mechanism is the dawn phenomenon. Between approximately 3 a.m. and 8 a.m., the body releases a surge of counter-regulatory hormones, primarily growth hormone and cortisol, along with smaller contributions from glucagon and catecholamines. These hormones signal the liver to release stored glucose (glycogenolysis) and simultaneously reduce peripheral insulin sensitivity. In people with adequate insulin secretion, the pancreas compensates automatically. In those with diabetes or reduced beta-cell reserve, the hepatic glucose output goes largely unchecked, and fasting glucose rises 10 to 50 mg/dL higher than the 3 a.m. nadir. [18]

The second mechanism is the Somogyi effect, also called rebound hyperglycemia, though its clinical relevance is debated. The hypothesis holds that nocturnal hypoglycemia (often caused by too much evening insulin) triggers a counter-regulatory hormone surge that overshoots and produces high morning glucose. A 2007 CGM study (N=64) found evidence of Somogyi-type rebound in only 13% of participants who had high morning readings, suggesting the dawn phenomenon is far more common. [18]

A third and often overlooked driver is late-night snacking or inadequate evening medication. Meals high in fat and refined carbohydrates can sustain glucose elevations for four to six hours, meaning a 10 p.m. dinner can contribute directly to a 6 a.m. reading.

A CGM distinguishes these three causes precisely. If glucose is stable at midnight, drops gradually through 3 a.m., then rises steeply before waking, that pattern is the dawn phenomenon. If glucose crashes below 70 mg/dL at 2 a.m. then spikes by 6 a.m., Somogyi rebound is the candidate. If glucose stays elevated from dinner through morning without falling, the meal or medication timing is the culprit.

What Is the Dawn Phenomenon?

The dawn phenomenon is the physiologically normal, circadian-driven rise in blood glucose that begins in the early morning hours, typically 3 a.m. to 8 a.m., caused by the pulsatile secretion of growth hormone overnight and the cortisol awakening response. [19] It is present in virtually everyone. In people without diabetes, a compensatory insulin surge keeps glucose levels within the normal range. In type 1 diabetes, insulin deficiency means the liver glucose output goes uncorrected. In type 2 diabetes, insulin resistance plus reduced secretory capacity produces variable degrees of morning elevation.

Magnitude varies considerably. A 2016 analysis in Diabetes Care (N=248 adults with type 1 diabetes using CGM) found that dawn phenomenon raised morning glucose by a median of 28 mg/dL, with a range from less than 5 mg/dL to more than 80 mg/dL. [20] About 54% of participants showed a clinically significant rise (>15 mg/dL). The same study found that higher baseline A1c correlated with a larger dawn rise, suggesting that managing the dawn phenomenon specifically may account for a meaningful portion of overall A1c improvement.

Management options include adjusting basal insulin timing (taking glargine or degludec in the evening rather than the morning), using a closed-loop insulin delivery system, adding metformin 500 mg at bedtime (which suppresses hepatic glucose output), or for people with type 2 diabetes not on insulin, adding a GLP-1 receptor agonist such as semaglutide 0.5 to 2 mg weekly, which blunts post-meal and fasting glucose through complementary mechanisms. [6] The ADA's 2024 Standards specifically recommend closed-loop systems as preferred insulin delivery for most adults with type 1 diabetes because they automatically increase basal insulin during the dawn window. [6]

How to Use CGM Data to Take Action

Raw CGM numbers mean little without a framework for response. The three metrics that carry the most clinical weight are Time in Range (TIR), Time Below Range (TBR), and glucose variability expressed as coefficient of variation (CV).

TIR of 70 to 180 mg/dL above 70% is the primary CGM target for most adults with diabetes. Each 5-percentage-point improvement in TIR corresponds to roughly a 0.33-point A1c reduction, validated in the CLARITY dataset (N=2,571). [11] TBR below 70 mg/dL should stay under 4% of the day, and TBR below 54 mg/dL should be under 1%. CV below 36% indicates acceptable glucose variability; a CV above 36% signals instability that raises cardiovascular risk independent of average glucose. [21]

For people without diabetes using a CGM for metabolic insight, the most actionable signals are post-meal spikes above 140 mg/dL and glucose nadirs below 70 mg/dL after exercise. Both indicate that dietary composition or exercise timing deserves adjustment. A 30-minute walk after meals, for instance, can reduce the two-hour postprandial peak by 30 to 50 mg/dL in people with insulin resistance. [22]

Reviewing CGM data weekly with a clinician or certified diabetes care and education specialist (CDCES) generates better outcomes than wearing the device alone. The DIAMOND trial (N=158 adults with type 2 diabetes on multiple daily injections) found that adding real-time CGM to structured education reduced A1c by 1.0% more than fingerstick monitoring at 24 weeks (P<0.001). [23]

Practical Steps Before Getting a CGM

Before ordering a sensor, confirm insurance coverage. Medicare covers CGM for all beneficiaries with diabetes who take insulin at least three times daily or use an insulin pump, following the Dexcom G6 LCD finalization in 2021. [24] Many commercial plans now cover CGM for type 2 diabetes on any insulin regimen. People without diabetes pay out of pocket; the FreeStyle Libre 2 retails for approximately $50 to $75 per 14-day sensor without insurance.

A baseline fasting glucose and A1c drawn at the same time as CGM initiation provides a reference point for interpreting trends. Your prescribing clinician should also review your medication list for drugs that interfere with CGM accuracy, including acetaminophen (relevant for older Libre models at doses above 1 g), hydroxyurea, and mannitol. [3]

Share your CGM data in a standardized Ambulatory Glucose Profile (AGP) report at every clinic visit. The AGP format, endorsed by the International Diabetes Center and the ADA, condenses 14 days of readings into a single-page visual that shows median glucose, interquartile range, and the 10th to 90th percentile bands across a 24-hour clock. [6] Most CGM apps generate this report automatically.