Continuous Glucose Monitoring With Low-Carbohydrate Nutritional Coaching to Improve Type 2 Diabetes Control: Randomized Quality Improvement Program

BACKGROUND: Type 2 diabetes mellitus (T2DM) is a leading cause of morbidity and mortality globally, with adverse health consequences largely related to hyperglycemia. Despite clinical practice guideline recommendations, effective pharmacotherapy, and interventions to support patients and providers, up to 60% of patients diagnosed with T2DM are estimated to have hemoglobin A(1c) (HbA(1c)) levels above the recommended targets owing to multilevel barriers hindering optimal glycemic control. OBJECTIVE: The aim of this study is to compare changes in HbA(1c) levels among patients with suboptimally controlled T2DM who were offered the opportunity to use an intermittently viewed continuous glucose monitor and receive personalized low-carbohydrate nutrition counseling (<100 g/day) versus those who received usual care (UC). METHODS: This was a 12-month, pragmatic, randomized quality improvement program. All adult patients with T2DM who received primary care at a university-affiliated primary care clinic (N=1584) were randomized to either the UC or the enhanced care (EC) group. Within each program arm, we identified individuals with HbA(1c) >7.5% (58 mmol/mol) who were medically eligible for tighter glycemic control, and we defined these subgroups as UC–high risk (UC-HR) or EC-HR. UC-HR participants (n=197) received routine primary care. EC-HR participants (n=185) were invited to use an intermittently viewed continuous glucose monitor and receive low-carbohydrate nutrition counseling. The primary outcome was mean change in HbA(1c) levels from baseline to 12 months using an intention-to-treat difference-in-differences analysis comparing EC-HR with UC-HR groups. We conducted follow-up semistructured interviews to understand EC-HR participant experiences with the intervention. RESULTS: HbA(1c) decreased by 0.41% (4.5 mmol/mol; P=.04) more from baseline to 12 months among participants in the EC-HR group than among those in UC-HR; however, only 61 (32.9%) of 185 EC-HR participants engaged in the program. Among the EC-HR participants who wore continuous glucose monitors (61/185, 32.9%), HbA(1c) was 1.1% lower at 12 months compared with baseline (P<.001). Interviews revealed themes related to EC-HR participants’ program engagement and continuous glucose monitor use. CONCLUSIONS: Among patients with suboptimally controlled T2DM, a combined approach that includes continuous glucose monitoring and low-carbohydrate nutrition counseling can improve glycemic control compared with the standard of care.