Should I get a continuous glucose monitor if I don't have diabetes?

No — and the most important reason is that the readings you'd be reacting to are substantially wrong.

The accuracy problem is the whole story

CGMs are validated against the glucose ranges of people with diabetes. In healthy people, whose glucose barely moves by comparison, the picture is different.

Measured against real venous blood in 34 healthy adults, the CGM's error was 17.6% — roughly double the ~9-10% these devices achieve in the diabetic range.

A randomised crossover in 15 healthy adults made it concrete. Against a fingerprick blood reference, the CGM:

• read 0.9 mmol/L too high, fasting and after meals
• overestimated time above the "spike" threshold by about four-fold — roughly two-fold once you subtract the baseline offset
• reclassified a fruit smoothie from a low glycaemic index of 53 to a medium-high 69

And the error isn't a fixed offset you could subtract out — it varies from person to person.

So the "spikes" you'd be tracking, screenshotting, and reorganising your diet around are, in substantial part, artefacts of the sensor. (Worth knowing: that crossover was funded by a smoothie manufacturer, whose interests the finding happens to serve. The 17.6% error figure is independent of it.)

Does it make you healthier?

Only one RCT in people without diabetes has measured body composition: 39 completers, 8 weeks, CGM added to an already-active low-GI diet programme. It found 0.8 kg extra weight loss and no change in BMI or HbA1c.

A meta-analysis of 25 trials found no weight or BMI benefit at all — and its HbA1c result comes from diabetic populations, since only 3 of the 25 trials enrolled anyone without diabetes. No study on the page shows a disease outcome improving.

The 2026 ADA standards expanded CGM use considerably — and for people without diabetes issued no recommendation, no target range, and no way to interpret the numbers.

What about "personalised nutrition"?

The famous study behind that pitch showed people's glucose responses to the same food differ. True, interesting — and not the same claim. It never tested whether a healthy person who buys a CGM and interprets it themselves ends up healthier.

The part nobody advertises

In one survey, 68% of CGM users became afraid they were developing type 2 diabetes after seeing a high reading — regardless of whether they actually were. Distress tracked with disordered-eating symptoms.

You'd be paying to be alarmed by numbers that are partly wrong, about a condition you probably don't have, with no evidence the exercise makes you healthier.