A few months ago, a friend showed me his Levels app. He'd been wearing a continuous glucose monitor — a tiny sensor on his arm — for three weeks and was obsessed with the data. "Look, rice spikes me to 160 but sweet potatoes only hit 120," he said, rearranging his entire diet around the numbers.

He's not diabetic. He's a healthy 35-year-old. And he's spending $200 a month on this.

CGMs are the hottest trend in biohacking. Companies like Levels, Nutrisense, and January AI have built entire businesses around selling glucose data to healthy people. The pitch: your blood sugar is silently damaging your health, and only continuous monitoring can reveal the problem.

But what does the science actually say?

A continuous glucose monitor is a small sensor (usually worn on the back of the upper arm or abdomen) with a tiny filament that sits just under the skin in the interstitial fluid. It measures glucose every 1-5 minutes and transmits readings to your phone.

The technology was developed for diabetes management, where it's genuinely life-changing. Type 1 diabetics using CGMs have significantly better glucose control and fewer dangerous hypoglycemic episodes (Beck et al., 2017, JAMA).

The biohacking play is different: take this medical-grade technology and use it to "optimize" glucose responses in people whose blood sugar is already normal.

1. Individual glucose responses vary wildly

The landmark Weizmann Institute study (Zeevi et al., 2015, Cell) tested 800 people eating identical meals and found enormous variability. Some people spiked to 140 mg/dL on a banana but barely moved on a cookie. Others were the opposite.

A 2020 follow-up from Stanford (Hall et al., PLOS Biology) confirmed this using CGMs on 57 non-diabetic participants: glucose responses to the same food varied by as much as 3-fold between individuals.

This is the strongest argument for wearing a CGM: population-level nutrition advice doesn't account for your individual response.

2. "Normal" fasting glucose may hide problems

A 2018 Stanford study monitored 57 participants with no diabetes diagnosis and found that some spent significant time above 140 mg/dL. About 25% of participants classified as "normal" by standard HbA1c criteria showed glucose patterns that resembled prediabetes during CGM monitoring.

3. Behavioral change is real

A 2024 study in Diabetes Technology & Therapeutics found that non-diabetic individuals wearing CGMs for 28 days significantly reduced their average glucose — primarily by modifying food choices and meal timing. The effect persisted at 3-month follow-up.

1. Normal glucose variability is... normal

A 2019 study put CGMs on 153 completely healthy adults. Even healthy people routinely hit 140+ mg/dL after meals. The biohacking companies have defined "optimal" ranges stricter than what healthy metabolisms actually do.

2. No evidence that optimizing glucose improves outcomes in healthy people

There is no long-term outcome data showing that minimizing glucose variability in people with normal HbA1c reduces cardiovascular disease, cognitive decline, or mortality.

3. The anxiety problem

A meaningful subset of non-diabetic CGM users developed "glucose anxiety" — obsessively checking readings, avoiding social meals, and restricting foods unnecessarily.

4. Accuracy limitations

CGMs measure interstitial glucose, which lags blood glucose by 5-15 minutes. The exact peak reading after a meal — the number most biohackers obsess over — is the least accurate measurement the device takes.

Strong case:

Moderate case:

Weak case:

Before spending $200/month on a CGM, consider what a $22 blood glucose meter can tell you. The G-425-1 Blood Glucose Monitor Kit (Amazon's Overall Pick) comes with 100 test strips, 100 lancets, and a lancing device. Eat your normal meals and check your glucose at 1-hour and 2-hour marks for two weeks.

For the deep dive on what your fasting labs should look like, check our Blood Panel Cheat Sheet.

1. Walk after meals. A 2022 meta-analysis found that just 2-5 minutes of light walking after eating reduced postmeal glucose spikes by 17-30%. Free.

2. Eat protein and fat before carbs. Eating vegetables and protein before carbohydrates reduced postmeal glucose by 29% and insulin by 37%. Meal sequencing — it works.

3. Build muscle. Skeletal muscle is the primary glucose disposal site. Each 10% increase in muscle mass was associated with 11% lower insulin resistance.

4. Prioritize sleep. A single night of 4-hour sleep increases insulin resistance by ~25% the next day.

5. Manage stress. Cortisol directly raises blood glucose via hepatic gluconeogenesis.

For a comprehensive supplement strategy, browse Thorne's full catalog with our partner discount.

CGMs are a genuinely useful technology that's been oversold to the wrong audience. If you're prediabetic or have metabolic risk factors, wearing one for a few months can be transformative. If you're metabolically healthy and curious, a short experiment (1-2 months) can teach you things about your body that no blood test can.

But wearing one indefinitely as a healthy person, chasing glucose scores, and fearing rice? That's optimization theater.

The fundamentals — walking after meals, eating protein first, building muscle, sleeping well, managing stress — will do more for your metabolic health than any amount of continuous glucose data.

As an Amazon Associate, I earn from qualifying purchases. All recommendations are based on research, not affiliate relationships. See our full disclosure.

Have questions? Reply to this email — I read every one.