Tirzepatide and Continuous Glucose Monitors: What CGM Data Can and Cannot Tell People on Treatment

Continuous glucose monitors, usually shortened to CGMs, were once reserved for people with diabetes who used insulin, but since 2024 two sensors have been cleared by the FDA for over the counter sale to adults, and many people taking tirzepatide now wear one out of curiosity or at a clinician's suggestion. A CGM shows how glucose responds to meals, sleep and exercise across an entire day, although the readings are easy to misread without context.

This article explains what CGM data tends to show during tirzepatide treatment, who is most likely to benefit from a sensor and how to read metrics such as time in range. It is educational information rather than medical advice.

What is a continuous glucose monitor and how does it differ from a fingerstick?

A continuous glucose monitor is a small wearable sensor, usually placed on the back of the upper arm, that estimates glucose every few minutes through a thin flexible filament under the skin and sends the readings to a phone app. A fingerstick meter measures glucose in a drop of capillary blood at a single moment, whereas a CGM builds a curve from hundreds of readings per day that shows how glucose rises and falls around meals and during sleep.

Because the filament sits in the fluid that surrounds cells, which glucose reaches a few minutes after it appears in the blood, a CGM reading taken while glucose is climbing after a meal or falling during exercise can differ from a fingerstick taken at the same moment, and a surprising reading should be confirmed with a fingerstick.

Which CGMs can people buy without a prescription?

Two sensors are currently cleared by the FDA for over the counter use by adults who do not take insulin: the Dexcom Stelo, cleared in March 2024 with a wear time of up to 15 days, and the Abbott Lingo, cleared in June 2024 with a wear time of up to 14 days. Both are designed for people without diabetes or with type 2 diabetes managed without insulin, and neither is intended to guide insulin dosing or to diagnose diabetes.

What does glucose normally look like in people without diabetes?

Glucose in adults without diabetes normally stays between 70 and 140 mg/dL for almost the entire day, with short rises after meals. In the multicenter study by Shah and colleagues, which fitted CGMs to 153 healthy participants aged 7 to 80, median time between 70 and 140 mg/dL was 96 percent, median time above 140 mg/dL was about 2 percent (roughly 30 minutes per day), and median time below 70 mg/dL was about 1 percent (roughly 15 minutes per day).

Brief readings in the 140s after a large meal are common in healthy people, so a single post meal peak on a CGM does not by itself indicate a problem.

How does tirzepatide change CGM readings?

Tirzepatide tends to lower average glucose and flatten post meal peaks, and published research suggests that several mechanisms contribute. Tirzepatide activates both GIP and GLP-1 receptors, which increases insulin release when glucose is elevated and reduces glucagon release from the pancreas, and because this insulin response depends on glucose being high, tirzepatide alone seldom pushes glucose too low.

Tirzepatide also slows stomach emptying, which spreads the absorption of a meal over a longer period and blunts the rise that follows eating. This effect is strongest in the first weeks after a dose change and appears to lessen over time, while improved insulin sensitivity from weight loss becomes more important as treatment continues.

The SURPASS-3 CGM substudy enrolled 243 adults with type 2 diabetes who were already taking metformin, with or without an SGLT2 inhibitor, and compared weekly tirzepatide with daily insulin degludec. Tirzepatide groups also showed lower glucose variability and numerically less time below 70 mg/dL, although these results cannot be assumed for people without diabetes.

What happens to blood sugar in people with prediabetes on tirzepatide?

In SURMOUNT-1, a 72 week trial of tirzepatide for weight management in adults without diabetes, most participants who entered with prediabetes had glucose in the normal range by the end of the study. About 95 percent of participants with prediabetes at baseline who received tirzepatide reached normal glucose status at week 72, compared with about 62 percent of those who received placebo, according to the trial report by Jastreboff and colleagues in the New England Journal of Medicine.

Laboratory tests such as fasting glucose and HbA1c remain the standard way to track prediabetes, because CGM averages are not validated for diagnosis.

Who is most likely to benefit from wearing a CGM on tirzepatide?

People taking tirzepatide together with insulin or a sulfonylurea are the group most likely to benefit from continuous monitoring, because the tirzepatide prescribing information notes a higher risk of hypoglycemia with these combinations and advises that a reduction in the insulin or sulfonylurea dose may be needed. A recent clinical review of CGM use during tirzepatide initiation and dose increases described how sensor data can help clinicians spot falling glucose early and adjust the other medications.

For people without diabetes, a CGM is optional. Some people find that seeing their response to meals helps them plan eating patterns, others find the stream of numbers stressful, and evidence that CGM use improves outcomes in people without diabetes is limited.

How should time in range and other CGM metrics be interpreted?

Time in range is the share of the day that glucose spends between 70 and 180 mg/dL, and a 2019 international consensus published in Diabetes Care recommends that most adults with type 1 or type 2 diabetes aim for more than 70 percent time in range. The same consensus sets limits for time spent above and below that range, shown in Figure 4, with looser targets for older adults and people at high risk of hypoglycemia.

CGM apps also report a glucose management indicator, or GMI, which estimates the HbA1c that the average sensor glucose would predict. GMI is calculated from mean glucose using a formula published by Bergenstal and colleagues in 2018, and it can differ from a laboratory HbA1c by half a percentage point or more because HbA1c also depends on how long red blood cells survive.

What are the common pitfalls when reading CGM data on tirzepatide?

The most common pitfall is treating a single reading as a diagnosis, when CGM data is most useful as a pattern across one to two weeks.

Compression lows. Lying on the arm that carries the sensor can squeeze the surrounding tissue and produce a false low reading overnight, which usually recovers as soon as the person changes position.

Low intake days. Tirzepatide can reduce appetite substantially, and a day with very little food or fluid may show lower readings, particularly in people taking insulin or a sulfonylurea, which is one reason regular meals and adequate hydration matter during treatment.

Food anxiety. Some people begin avoiding nutritious foods such as fruit, beans or whole grains because those foods produce a visible rise, even though a moderate rise that returns to baseline within two to three hours is a normal response.

Can a CGM be used to adjust the tirzepatide dose?

A CGM should not be used to change the tirzepatide dose without a prescriber, because tirzepatide doses are set by the titration schedule on the label, by tolerability and by overall clinical goals rather than by daily glucose readings. A CGM can still give the prescriber a clearer picture, for example by showing low readings that suggest an insulin dose needs lowering, and sharing the 14 day summary report that most apps export is more useful than describing individual readings.

The bottom line

CGM data from clinical trials shows that tirzepatide tends to raise time in range and flatten glucose peaks in people with type 2 diabetes, and most SURMOUNT-1 participants with prediabetes reached normal glucose status. A CGM is most valuable for people who combine tirzepatide with insulin or a sulfonylurea, while for others it is an optional learning tool whose readings are best reviewed as patterns and shared with a clinician.