October 2026: NEJM trial finds Dexcom G7 lowered A1c in type 2 diabetes without insulin
The CONNECT trial, published in the New England Journal of Medicine in October, found adults with type 2 diabetes who do not use insulin cut their A1c more with a Dexcom G7 sensor than with fingersticks.
Pulse, covering wearables and health sensors
Published , 3 min read
0.9 ptsextra A1c drop with G7 versus routine care at 26 weeks
This is a look back at an event from . We wrote it on November 15, 2026, from the sources listed at the end.
What to know
- The randomized CONNECT trial enrolled 283 adults with type 2 diabetes who do not use insulin at 22 US primary care practices.
- Average A1c fell from 8.8% to 7.2% with the Dexcom G7, versus 8.7% to 8.0% with routine fingerstick testing, over 26 weeks.
- The G7 group spent about five more hours a day in the 70 to 180 mg/dL glucose range.
Source: Harvard Health Publishing; Managed Healthcare Executive
A randomized trial of Dexcom's G7 continuous glucose monitor (CGM) in adults with type 2 diabetes who do not take insulin was published in the New England Journal of Medicine on October 14, 2026, according to Harvard Health Publishing. The study, called CONNECT, found that people who wore the sensor lowered their A1c, a measure of average blood sugar over about three months, more than people who kept testing with a fingerstick meter.
The results were first presented at the American Diabetes Association's Scientific Sessions in New Orleans in June, Medical Device Network and Zacks reported at the time. Dexcom mentioned the journal paper in its third-quarter results. Kivorent News had not covered the trial itself.
What the trial did
CONNECT enrolled 283 adults aged 21 to 84 at 22 US primary care practices, according to Managed Healthcare Executive and Zacks. 145 were assigned to the G7 and 138 to routine care with a standard blood glucose meter. Both groups kept their existing medications and received standardized diet and exercise education, Zacks reported. The main comparison ran for 26 weeks, and 265 people completed that period.
Most participants took other diabetes drugs. About 37% used an SGLT2 inhibitor and about 40% an incretin-based drug such as a GLP-1, Medical Device Network and Managed Healthcare Executive reported. The average age was 60, and 32% came from racial or ethnic minority groups. For how randomized studies like this one separate a device's effect from chance, see our explainer on randomized controlled trials.
The results
Average A1c in the G7 group fell from 8.8% to 7.2%. In the routine-care group it fell from 8.7% to 8.0%, according to Harvard Health and Managed Healthcare Executive. That is a difference of 0.9 percentage points in favor of the sensor.
The G7 group spent about five more hours a day with glucose between 70 and 180 mg/dL. Medical Device Network put time in range at 62% for the G7 group and 41% for the control group at 26 weeks. Zacks reported that 82% of G7 users cut their A1c by at least 0.5 points, and that people who started above 10% saw an average drop of 3.1 points. Median daily sensor wear was 97%.
Managed Healthcare Executive reported less time below 70 mg/dL with the G7 and no severe low blood sugar events. A six-month extension is running to follow participants for a full year. The senior author was Roy Beck of the Jaeb Center for Health Research, according to Medical Device Network. Kivorent News could not open the journal paper itself, so funding and the full safety tables are not reported here. Dexcom makes the G7, and the trial's results have been presented publicly by Dexcom executives.
How the sensor fits in
The G7 is a small sensor worn on the arm that reads glucose in the fluid under the skin and sends it to a phone. Our explainer on how continuous glucose monitors work covers the method. Dexcom also sells Stelo, an over-the-counter sensor aimed at people who do not use insulin. Abbott's competing Libre Duo glucose and ketone sensor is due in the US in December.
Why it matters
About 20 million US adults with type 2 diabetes do not use insulin, Harvard Health estimates. The American Diabetes Association's 2026 guidelines already recommend CGMs for this group, Harvard Health noted, and CONNECT adds randomized evidence from ordinary primary care practices. Dexcom has said it expects Medicare to decide on CGM coverage for this group this year, according to MedTech Dive. That decision would shape how many of these patients can get a sensor paid for, and how big a market Dexcom and Abbott are competing for.
This article is general information about how technology works. It is not medical advice. Talk to a clinician about your own health.
Sources
- Continuous glucose monitors benefit people who don't need insulin, Harvard Health Publishing, November 2, 2026
- Jessica Castle, M.D., discusses CGM's A1c impact in non-insulin type 2 diabetes, Managed Healthcare Executive, October 21, 2026
- Dexcom G7 CGM helps T2D diabetics not on insulin manage glucose levels, Medical Device Network, June 8, 2026
- DexCom Reports Positive CONNECT Data for Non-Insulin Diabetes, Zacks Equity Research via Nasdaq, June 8, 2026
- Abbott, Dexcom grapple with diabetes coverage ahead of planned Medicare expansion, MedTech Dive, November 2, 2026
Glucose and metabolic techWearablesDigital health and AI in care
Written by an AI correspondent under standards set by editor Jacob Knight. Spotted an error? Tell us and we will fix it.