In adults with type 2 diabetes (T2D) who are managed in primary care and are not receiving insulin, the use of continuous glucose monitoring (CGM) helps improve glycaemic control and reduce the risk of hyperglycaemia, according to the CONNECT study.
The primary outcome of glycated haemoglobin level decreased from 8.8 percent at baseline to 7.2 percent at 26 weeks among participants using the CGM and from 8.7 percent to 8.0 percent among participants who performed monitoring with a blood glucose meter at a frequency advised by their healthcare provider (routine-care group) (mean change, −1.6 vs −0.7 percentage points; p<0.001), reported Dr Tamara Oser from the University of Colorado Anschutz School of Medicine, Aurora, Colorado, US, and colleagues. [N Engl J Med 2026;395:1717-1727]
A higher proportion of patients in the CGM than in the routine care group achieved a reduction of ≥0.5 percentage points in glycated haemoglobin level (82 percent vs 56 percent) and had a glycated haemoglobin level of <7 percent (46 percent vs 24 percent).
“Key secondary multiplicity-controlled CGM-measured outcomes that were reflective of hyperglycaemia were all significantly better in the CGM group than in the routine care group,” noted Oser and colleagues.
The percentage of time in the glucose-level range of 70–180 mg/dl increased from 29 percent at baseline to 62 percent at 26 weeks in the CGM group and from 31 percent to 42 percent, respectively, in the routine care group (mean difference, 21 percentage points; p<0.001).
The mean glucose level at week 26 was significantly lower in the CGM group vs the routine care group (175 vs 207 mg/dl; p<0.001), as was the percentage of time with the glucose level >180 mg/dl (38 percent vs 58 percent; p<0.001) or >250 mg/dl (7 percent vs 22 percent; p=0.002).
There were no documented cases of severe hypoglycaemia or hyperglycaemia-related serious adverse events.
The findings of CONNECT provide evidence on the effectiveness of CGM for patients with T2D not being treated with insulin, according to Oser and colleagues, adding that the inclusion of only patients receiving diabetes care in primary care practices “maximizes the generalizability of the results.”
CONNECT included 283 participants enrolled at 22 primary care practices in the US. These participants were randomly allocated to either the CGM group (n=145; mean age 59 years, 43 percent female, mean BMI 33 kg/m2) or the routine care group (n=138; mean age 60 years, 41 percent female, mean BMI 34 kg/m2). The median duration of diabetes was 9–10 years. Overall, 40 percent of participants were using a GLP-1 RA, 37 percent were using an SGLT2 inhibitor, and 18 percent were using both medications.
“During follow-up, either insulin or additional non-insulin glucose-lowering medications were prescribed for a minority of participants (31 percent in the CGM group and 17 percent in the routine-care group), which suggests that the improvement observed with CGM cannot be explained by medication changes alone but may have been due to modifications in behavioural habits that could affect glucose levels,” Oser and colleagues noted.
CGM use was very high over the 26-week study period, with a median use of 97 percent of the time and 86 percent of the participants using CGM at least 80 percent of the time.