Glucose monitors are being sold to healthy people. An endocrinologist's answer: most don't need one.

The gadget, and the fine print

In March 2024 the FDA cleared the first glucose sensor Americans can buy without a prescription: Dexcom's Stelo. Abbott's Lingo arrived the same year. You wear the coin-sized sensor on the back of your arm for about two weeks, it samples the glucose in the fluid under your skin every few minutes, and it streams the numbers to your phone. Stelo runs about $89 a month on subscription.

None of that makes it a medical test. The FDA cleared Stelo only for adults who do not use insulin, and warned it is not built to catch dangerous lows, so it is wrong for anyone prone to hypoglycaemia. Lingo is sold purely as a "wellness" product. You are buying a consumer gadget on a rolling subscription, and the pitch to healthy people is far newer than the sensor inside it.

A spike is digestion, not damage

Here is what the red alerts leave out. In a person without diabetes, glucose is meant to climb after a meal. A recent scoping review found it typically peaks about an hour after eating and settles back within two to three hours, and that healthy people spend roughly 96% of the day inside the normal range.

Archana Sadhu, an endocrinologist at Houston Methodist, goes further on what "normal" even looks like. "It's completely normal for a healthy 20- or 30-year-old woman to have fasting blood sugars in the 60s," she says, a reading that would alarm someone with diabetes on insulin. A bagel nudging your line upward is not your metabolism breaking. It is your metabolism working.

The evidence for healthy users barely exists

This part is worth being blunt about. For people with diabetes, continuous monitors have been a genuine breakthrough. For everyone else, the benefit is mostly unproven. Johns Hopkins epidemiologists Michael Fang and Elizabeth Selvin have argued the point repeatedly: there are no major clinical trials showing that monitoring improves the health of people without diabetes, and every rule for reading the data was written for diabetes care. Sadhu is blunter still on the design question. "These devices aren't designed for persons without diabetes," she says. The claims stacked on top, about weight loss, energy and "optimising" your metabolism, are running well ahead of the science.

When the data starts to cost you

And the downside is not just a wasted $89. Sadhu's real worry is behavioural: "My biggest concern about people using these devices in unintended ways is the anxiety it can cause and, because of that anxiety, the inappropriate actions that may be taken." Watch a graph all day and eating turns into a test you are trying to pass, which is a short road to a narrower, more anxious diet. The same review found wellness and marketing sources tying glucose spikes to cancer, mood and sleep in ways the medical literature does not support. If a device makes you frightened of an apple, it is not making you healthier.

What this means for you

• If you're 25-34: You are the core marketing target, and most of you have no diabetes risk factors, so a monitor will mostly show you normal spikes. If you are curious, wear one for a week or two to learn how your meals behave, then take it off. Do not let a graph rewrite a healthy relationship with food.

• If you're 35-49: This is when prediabetes risk starts to climb, especially with a family history or a past gestational diabetes diagnosis. A one-off fasting glucose or HbA1c blood test tells you more than any wearable, and unlike the sensor it is an actual diagnostic.

• If you're 50-64: If you have risk factors, ask your clinician for standard screening rather than self-diagnosing from an app. Used for a few weeks with a professional reading it with you, a sensor can teach you something, but it does not replace a blood test.

The blood test you actually want

None of this makes a sensor useless, only oversold. Wear one for a fortnight and you will see, correctly, that white rice moves your glucose more than the same calories of beans, and that a ten-minute walk after dinner flattens the curve. You already knew both were the healthy move. Sadhu allows one real exception, prediabetes, where a sensor used under guidance can occasionally earn its place. For everyone else chasing their actual risk, a single fasting glucose or HbA1c test at a clinic tells a person without diabetes far more than weeks of graphs.

Our take

For diabetes, continuous monitors are a real advance. For healthy people, they are mostly an expensive way to watch your body do exactly what it is meant to. Used for two curious weeks, a sensor can teach you a little about your own meals. Worn anxiously and forever, it sells worry dressed as insight. If it is your risk you are after, the $89 is better spent on the blood test.