A Number Every Few Minutes: What Continuous Glucose Monitors Proved — and What They Haven't

An inventory of what a modern continuous glucose monitor delivers: a current glucose value, read from the interstitial fluid just beneath the skin of the upper arm. A trend arrow — rising, falling, steady. A history curve of every meal's aftermath. Time in range. Time below range. Time above range. A variability score. And, since 2024, a place on the American drugstore shelf: several CGM systems — including ones built on Abbott's FreeStyle Libre platform — cleared for over-the-counter sale, no prescription, no diabetes diagnosis required. The FreeStyle Libre itself is factory-calibrated and intermittently scanned — a 'flash' monitor, worn on the back of the upper arm, read by phone or scanner on demand.

Also in the inventory: the ritual the device replaced. Fingerstick self-monitoring — the lancet, the strip, the several-times-a-day prick. Against that baseline, the randomized trials and meta-analyses count precisely what the sensor changes: about 0.2 to 0.3 percentage points off HbA1c, on average. Thirty-eight percent less time spent in hypoglycemia in type 1 diabetes. Roughly half as much hypoglycemia in insulin-treated type 2 diabetes. About 10 percent more time in range, over education alone, in non-insulin-treated type 2 diabetes.

The question the wellness market is actually asking — should people without diabetes wear one? — sits on top of a literature built almost entirely for people who have it. The defensible answer: for diabetes management, the evidence is solid and specific; for everyone else, it is thin.

What the diabetes trials established

A 2020 Diabetes Care meta-analysis of randomized trials, a 2024 meta-analysis focused on type 2 diabetes, and a 2018 systematic review converge: CGM modestly improves glycemic metrics over fingerstick testing, and its signature win is hypoglycemia avoidance — a safety outcome, the kind that matters at 3 a.m. Nor are the numbers mere novelty: the derived metric time in range is validated against long-term microvascular complications such as retinopathy, which gives the endless stream of readings real prognostic meaning. A 2022 review in the Journal of Clinical Endocrinology & Metabolism lays out exactly how these CGM-derived metrics map onto the complications of diabetes.

The fine print of the same trials

Also inventoried, because honesty requires it: in the insulin-treated type 2 trial, HbA1c did not improve. Most trials were short and open-label — you cannot blind a sensor, and the person watching their own curve is part of the intervention. Hard-endpoint data are scarce. And regulators have recalled certain sensor lots for inaccurate readings, which is worth knowing before treating any single number on the phone as gospel.

The biohacker column

For people without diabetes, the evidence file is about one meta-analysis thick: a 2024 review of CGM as a behavior-change tool found only modest short-term glycemic changes — driven mostly by participants who had diabetes — and no significant effect on weight. No trial shows disease prevention or lifespan benefit. A 2022 review of CGM and physical activity sketches the plausible uses, but plausibility is the ceiling. These are medical tools for diabetes management, not proven longevity interventions.

Practical meaning

For someone with diabetes — especially insulin-treated — CGM is among the best-evidenced tools in the entire consumer health catalog, and a conversation for their care team. For the healthy optimizer, the device reliably generates data; whether that data changes anything that matters is the unproven part — and an unexpected low reading in a person without diabetes is a reason to see a clinician, not to self-titrate snacks.

One last item for the inventory: the trend arrow. It knows which way your glucose is moving. What it cannot tell you is where, exactly, you are going. Educational, not medical advice.