Skip to main content

Body · product guide

Glucose monitor: does it work?

Continuous sensors and finger-prick meters. Proven to improve glucose control in diabetes; for people without diabetes, little evidence that wearing one helps.

Moderate evidenceReviewed September 2026

What it is

A continuous glucose monitor (CGM) is a small sensor with a thin filament under the skin of your arm or belly that reads glucose in the fluid between cells every few minutes, for 10–15 days per sensor. Its readings trail blood glucose by several minutes. A blood glucose meter reads a drop of fingertip blood on a test strip. Both are standard tools in diabetes care.

Since 2024, some CGMs have been sold over the counter in the US to adults who don't use insulin, and they're marketed to people without diabetes as a way to 'see your spikes'. Glucose normally rises after meals and falls back: in healthy adults, sensors show glucose between 70 and 140 mg/dL (3.9–7.8 mmol/L) about 96% of the time.

How strong is the evidence?

Moderate evidence

In randomized trials, CGM lowered HbA1c by about 0.4–0.6 percentage points more than usual care in adults with type 1 diabetes on injections and type 2 diabetes on basal insulin, and cut time spent with low glucose. For people without diabetes there are no outcome trials: no evidence that tracking or flattening post-meal rises prevents diabetes, helps weight loss or improves health. Finger-prick meters vary too: in an independent test of 18 cleared meters, only 6 met the accuracy standard in all three rounds.

The claims you'll see, checked

Improves glucose control in diabetes

Holds up

Randomized trials in insulin-treated type 1 and type 2 diabetes showed lower HbA1c and less time low.

Helps people without diabetes lose weight or prevent diabetes

Unproven

No outcome trials exist in people without diabetes.

Post-meal 'spikes' are harming healthy people

Unproven

Brief rises after meals are normal; healthy adults spent a median of about 30 minutes a day above 140 mg/dL. There's no evidence that flattening them improves health.

A CGM can diagnose diabetes or prediabetes

False

Diagnosis uses HbA1c, fasting glucose or a glucose tolerance test from a lab.

A smartwatch or ring can measure blood sugar without a needle

False

In 2024 the FDA warned that no such device is authorized and that wrong readings could lead to dangerous dosing errors.

All finger-prick meters are accurate

Mixed

Cleared meters don't always meet accuracy standards in independent testing; some did every time, some never did.

What to look for

  • Clearance for your use. If you take insulin, you need a system cleared for treatment decisions, with low and high glucose alarms.
  • Accuracy data: MARD (mean absolute relative difference), where lower is better and current sensors are roughly 8–10%, and whether finger-prick calibration is needed.
  • Wear length, warm-up time and phone compatibility.
  • The yearly cost of sensors, and whether insurance covers them for your situation.
  • For meters: compliance with the ISO 15197:2013 accuracy standard, and affordable strips, since you'll buy far more strips than meters.
  • Data sharing with your clinician, or with a caregiver for alarms.

What to skip

  • Needle-free glucose watches and rings.
  • 'Metabolic health' subscriptions that score your foods by your spikes; there's no evidence this improves health in people without diabetes.
  • Buying a CGM to diagnose yourself. Get an HbA1c test instead.

How to use it

  • Curious and without diabetes? A two-week trial is plenty. Compare with healthy reference ranges rather than chasing a flat line.
  • Allow for the lag: when glucose is changing fast, confirm with a finger-prick before acting.

Who should be careful

  • Don't change diabetes medicines, especially insulin, based on a wellness CGM or a single reading without your clinician.
  • Lying on the sensor can cause falsely low overnight readings; confirm lows with a finger-prick if the number doesn't match how you feel.
  • Some substances can distort readings on certain sensors, such as high-dose vitamin C, paracetamol (acetaminophen) and hydroxyurea; check your sensor's label.
  • Adhesive can irritate skin, so rotate sites. Remove the sensor before MRI or CT scans unless the maker says otherwise.
  • If watching your food responses makes you anxious or restrictive, stop; it can feed disordered eating.
  • Symptoms of very high or low glucose (confusion, sweating, shaking, extreme thirst) need action or care, whatever the device shows.

The studies

Every citation is checked against its PubMed record.

  1. Effect of Continuous Glucose Monitoring on Glycemic Control in Adults With Type 1 Diabetes Using Insulin Injections: The DIAMOND Randomized Clinical Trial
    Beck RW, Riddlesworth T, Ruedy K, Ahmann A, Bergenstal R, Haller S, et al. · JAMA · 2017 · Randomized controlled trial

    In 158 adults with type 1 diabetes on multiple daily injections, 24 weeks of CGM lowered HbA1c by 0.6 percentage points more than usual care and cut time spent below 70 mg/dL from a median of 80 to 43 minutes a day.

  2. Effect of Continuous Glucose Monitoring on Glycemic Control in Patients With Type 2 Diabetes Treated With Basal Insulin: A Randomized Clinical Trial
    Martens T, Beck RW, Bailey R, Ruedy KJ, Calhoun P, Peters AL, et al. · JAMA · 2021 · Randomized controlled trial

    In 175 adults with poorly controlled type 2 diabetes on basal insulin in primary care, 8 months of CGM lowered HbA1c by 0.4 percentage points more than finger-prick monitoring and raised time in range (70–180 mg/dL) from 43% to 59%.

  3. Continuous Glucose Monitoring Profiles in Healthy Nondiabetic Participants: A Multicenter Prospective Study
    Shah VN, DuBose SN, Li Z, Beck RW, Peters AL, Weinstock RS, et al. · Journal of Clinical Endocrinology & Metabolism · 2019 · Prospective observational study

    In 153 healthy people aged 7–80 without diabetes wearing blinded sensors for up to 10 days, mean glucose was about 98–99 mg/dL (104 over age 60); median time between 70 and 140 mg/dL was 96%, with about 30 minutes a day above 140.

  4. Investigation of the Accuracy of 18 Marketed Blood Glucose Monitors
    Klonoff DC, Parkes JL, Kovatchev BP, Kerr D, Bevier WC, Brazg RL, et al. · Diabetes Care · 2018 · Independent accuracy surveillance study

    Testing 18 FDA-cleared finger-prick meters in 1,035 people against laboratory reference values, only 6 met the accuracy standard in all three rounds of testing, and 4 met it in none.