Continuous glucose monitoring (CGM) is well established for insulin‑treated diabetes, but its role in patients with non‑insulin‑treated type 2 diabetes (T2D) is less defined. The investigators designed a prospective pre–post evaluation to assess short‑term clinical and economic outcomes after initiating personal use of the G7 CGM system in adults with poorly controlled T2D (hemoglobin A1c ≥ 8%).
This was a prospective pre–post study enrolling adults with non‑insulin‑treated T2D and A1c ≥ 8%. The target enrollment was 200; 217 patients were ultimately enrolled. Participants initiated the G7 CGM device and were observed for 6 months. A claims‑data subgroup of 76 patients was used for analyses of health care utilization, glycemic events, and medication use. Baseline was defined as the 3 months preceding CGM initiation, and follow‑up assessments occurred at 3 and 6 months.
Primary endpoints were change in A1c and scores on the Audit of Diabetes‑Dependent Quality of Life (ADDQoL). Secondary assessments included glycemic events, health care utilization, and medication outcomes in the claims subgroup. Changes at 3 and 6 months were evaluated using the Wilcoxon signed‑rank test. Multivariable analysis of covariance models adjusted for baseline A1c and clinical covariates to evaluate A1c change. ADDQoL was analyzed with linear mixed‑effects models including random intercepts and fixed effects for time; Cuzick's test assessed quality‑of‑life trends.
Among the enrolled cohort, mean age was 58.3 ± 11.5 years. The sample was 75% White and 58% male. Baseline median A1c was 8.60% (IQR 8.20–9.10) and mean A1c was 8.96% ± 1.24%. Common comorbidities at baseline included hyperlipidemia (73%), hypertension (70%), and obesity (44%). The claims subgroup comprised 76 participants for whom utilization and event data were available.
Significant reductions in A1c were observed at both 3 and 6 months after initiation of CGM. Median A1c changes were −1.10% at 3 months and −1.40% at 6 months; mean changes were −1.28% and −1.42%, respectively (P < 0.001). In adjusted multivariable models the adjusted mean change from baseline was −1.32 percentage points at 3 months (95% CI −1.48 to −1.15; P < 0.001; n = 153) and −1.48 percentage points at 6 months (95% CI −1.68 to −1.28; P < 0.001; n = 143).
Achievement of guideline‑based A1c thresholds improved markedly. The proportion meeting the ADA target (<7%) increased from 0% at baseline to 26.4% at 6 months. The proportion meeting the HEDIS target (<8%) increased from 0% at baseline to 50.9% at 6 months (both changes P < 0.001).
In the claims subgroup, the proportion with documented hyperglycemic events decreased from 53.7% at baseline to 16.3% at 6 months; this decline was reported but not statistically significant within the subgroup analyses as presented in the source.
Across the study period, no significant changes were detected in health care utilization, medication adherence, or overall medication burden based on the claims subgroup analysis. The source reports stable utilization and therapy patterns despite the observed glycemic improvements.
Patient‑reported perceived quality of life improved on a categorical measure: the proportion reporting good‑to‑excellent quality of life rose from 24.1% at baseline to 45.8% at 6 months. However, weighted ADDQoL scores remained stable over time according to the mixed‑effects analyses reported.
Higher CGM use and greater time in range were associated with better glycemic outcomes, including improved A1c and glucose management indicator measures. The source indicates these device‑usage metrics correlated with the magnitude of A1c improvement.
In this 6‑month prospective study of G7 CGM in adults with poorly controlled non‑insulin‑treated T2D, CGM use was associated with significant reductions in A1c, increased attainment of ADA and HEDIS targets, and improved patient‑reported perceived quality of life. Health care utilization and medication therapy remained stable during follow‑up. The authors conclude these findings support broader implementation and reimbursement of CGM in this population but note that further research is needed to determine long‑term sustainability and to disentangle the specific contributions of diet, medication changes, and adherence to the observed glycemic improvements.
Conflict of interest and funding: the authors reported no disclosures. The study was funded by Dexcom, Inc. The source did not report additional long‑term follow‑up data beyond the 6‑month period or more granular breakdowns of medication changes beyond the statements provided.