NEJM publishes Dexcom-sponsored CONNECT trial of CGM in type 2 diabetes without insulin

On October 14, 2026, The New England Journal of Medicine published "Continuous Glucose Monitoring in Non–Insulin-Treated Type 2 Diabetes," the full report of the CONNECT randomized trial.[1] This is a Dexcom-sponsored study. The journal's metadata lists Dexcom as the funder,[1] and ClinicalTrials.gov names DexCom, Inc. as the lead sponsor, with the Jaeb Center for Health Research among its collaborators.[3] Dexcom announced the publication the same day.[2]

What the trial tested

CONNECT enrolled adults with type 2 diabetes who were not using insulin. The registry record excluded anyone who had used insulin in the 12 months before screening or planned to start it in the next 12 months.[3] Participants also needed an A1C of at least 7.5% at screening, a diagnosis at least six months old managed in primary care, and no personal CGM use in the previous six months.[3]

People were randomized 1:1 to real-time CGM or to routine care for 26 weeks.[3] The routine-care group wore a blinded CGM at intervals, meaning the sensor recorded glucose but they couldn't see the readings.[3] The main outcome was A1C at 26 weeks.[3] The trial was open-label, so everyone knew which group a participant was in.[3]

Dexcom said the study screened 440 people at 22 primary care sites in the United States.[2] ClinicalTrials.gov lists an actual enrollment of 283.[3] We did not find a primary source we could open that gives the number randomized to each group, so we haven't reported one.

What Dexcom reported

The figures below come from Dexcom's release, which credits them to the trial's presentation; we couldn't open the full NEJM text to check them against the paper.[2] According to Dexcom, people using the Dexcom G7 lowered their A1C by an average of 1.6 percentage points from a mean starting level of 8.8%. That was 0.9 points more than the routine-care group.[2] G7 users also spent about five more hours per day with glucose between 70 and 180 mg/dL than people in routine care, and Dexcom said that gap showed up within the first week.[2]

At 26 weeks, 68% of G7 users had an A1C below 7.5% and 46% had one below 7.0%, the company said.[2] For participants taking GLP-1 medicines, Dexcom reported an additional 0.9-point A1C reduction with CGM compared with routine care.[2] Dexcom also reported that 97% of participants were still using CGM after 26 weeks, most of them daily.[2]

The release called CONNECT a "landmark" trial. Its chief executive, Jake Leach, said "the evidence is undeniable."[2] Those are the sponsor's words. Roy W. Beck of the Jaeb Center, the paper's senior author, said in the same release that the consistent benefit across groups "supports consideration of the use of a CGM much earlier in Type 2 diabetes management."[2] He is listed among the NEJM authors.[1]

When the results came out

This was not the first public look at the data. Dexcom's release cites a presentation at the American Diabetes Association's 2026 Scientific Sessions in New Orleans on June 6, 2026, with Tamara Oser as first author.[2] The NEJM paper lists Tamara K. Oser of the University of Colorado Anschutz School of Medicine first.[1]

The trial isn't over. The registry describes a second 26-week extension phase in which both groups use CGM, lists the trial as active but not recruiting, and gives an estimated completion date of December 16, 2026.[3] As of October 15, 2026, no results had been posted to the registry record itself.[3]

What this means for you

Dexcom said the findings support using CGM earlier in type 2 diabetes care, including for people over 65.[2] The company did not announce any change to the Dexcom G7's labeling, availability or coverage in its release.[2] Whether a CGM fits your own diabetes care is a question to bring to your care team or pharmacist.