ADA 2026: A1c Fell 1.6% on Average in Non-Insulin T2D

Originally surfaced June 12, 2026, drawing on American Diabetes Association and Dexcom materials released during the ADA 2026 Scientific Sessions.

A clinical trial presented at the American Diabetes Association's 2026 Scientific Sessions in New Orleans found that adults with type 2 diabetes who were not using insulin had larger hemoglobin A1c reductions over 26 weeks when continuous glucose monitoring was added to routine care.

The CONNECT randomized controlled trial enrolled 283 adults with type 2 diabetes not on insulin therapy across 22 primary care practices in the United States. The ADA release reported that 265 participants completed the study, with a mean age of 60 and about 32 percent identifying as racial or ethnic minorities. Baseline A1c values ranged from 7.1 percent to 14.9 percent. A1c reduction was the primary endpoint.

Participants in the continuous glucose monitoring group had a mean A1c reduction of 1.6 percent, compared with 0.7 percent in the routine-care group, for a 0.9 percentage-point difference. Among participants who began the study with A1c above 10 percent, the monitoring group's mean reduction was 3.1 percent, more than 2 percentage points greater than the control group. At week 26, 68 percent of the monitoring group had reached an A1c below 7.5 percent, and 46 percent had reached below 7.0 percent.

Dexcom's release quoted senior author Roy Beck, MD, PhD, describing the results as the first Level A evidence supporting this management approach in people with type 2 diabetes not using insulin. That is an attributed characterization, not an independent journal conclusion. Public ADA and Dexcom materials did not report p-values or 95 percent confidence intervals for the primary A1c outcome.

What A1c represents

Hemoglobin A1c reflects average blood glucose over roughly two to three months. A value of 7.0 percent corresponds to approximately 154 mg/dL average blood glucose, and 8.8 percent corresponds to approximately 206 mg/dL using the commonly cited estimated average glucose formula. The ADA defines diabetes at A1c of 6.5 percent or above and prediabetes at 5.7 percent to 6.4 percent in its 2026 Standards of Care.

The CONNECT trial enrolled people across a broad range of baseline A1c values, which gives the outcome data meaningful real-world context. It also reinforces why A1c remains central in diabetes research: the trial's main clinical outcome was not a device engagement metric, but the same lab marker clinicians commonly use to track longer-term glycemic control.

What this study does and does not show

The CONNECT trial was funded by Dexcom, the manufacturer of the device used in the monitoring arm. Results were presented as a conference oral abstract, and a peer-reviewed journal publication has not yet appeared. Conference data undergoes different scrutiny than a published paper, and independent replication would strengthen the conclusions. The trial ran for 26 weeks, so longer-term A1c trajectories are not captured.

The ADA release noted that glycemic improvements appeared additive across drug classes, including GLP-1-based therapies and SGLT2 inhibitors. That does not mean the study answers every medication question. It means the reported A1c improvement was observed in a population already receiving real-world diabetes care.

The testing context

Lab-based A1c testing is one of the standard ways people with diabetes or prediabetes establish where they stand with a clinician. Unlike a point-in-time glucose reading, an A1c provides a longer-window estimate that guidelines use for both diagnosis and ongoing monitoring. The CONNECT trial used A1c as its primary clinical outcome, reflecting how central the test remains in structured diabetes research and care.

LabTestSuperstore's Hemoglobin A1C test page and diabetes and prediabetes section cover how A1c and related markers fit into diagnostic ranges.

What not to read into this

The CONNECT trial enrolled adults who already had a type 2 diabetes diagnosis. The findings do not apply to the general population without a diabetes history, and the outcomes do not suggest that any A1c number automatically triggers a specific clinical action. How any lab result connects to management decisions requires the context of a treating clinician.

This article is editorial commentary and is not medical advice. It has not been reviewed by a physician and should not be used to make decisions about testing, treatment, medication, or glucose monitoring devices.