Advances in continuous glucose monitoring and their impact on glycemic control in diabetes.
Journal
Journal of diabetes investigation
ISSN
2040-1124
Date Issued
2026-02-25
Author(s)
Abstract
Achieving optimal glycemic control remains a central challenge in diabetes care. Despite pharmacologic advances, substantial proportions of patients fail to reach recommended targets, contributing to the ongoing complications. The need for more precise, dynamic glucose data has accelerated continuous glucose monitoring (CGM) adoption. Recent technological advances have dramatically improved sensor accuracy and clinical utility, making CGM central to modern diabetes management. Current guidelines recommend CGM soon after diagnosis for most insulin-treated individuals and broader consideration regardless of diabetes type or A1c level1. CGM-derived metrics—particularly time in range (TIR), glycemic variability (GV), and time below range (TBR)—correlate with clinical outcomes2, 3, with TIR demonstrating strong correlation with HbA1c4. This review summarizes recent progress in CGM technology, highlights clinical evidence, and discusses future directions for precision diabetes care. Compared with self-monitoring of blood glucose (SMBG), CGM offers superior detection of glycemic variability, nocturnal hypoglycemia, and postprandial excursions. CGM technology has evolved dramatically over the past two decades (Table 1), transitioning from early devices requiring frequent fingerstick calibration (2–4 times daily) with mean absolute relative difference (MARD) of 15–20% to contemporary factory-calibrated sensors achieving MARD below 9%. Modern real-time CGM systems now offer 10–14 days wear time, Bluetooth connectivity, and seamless integration with smartphones, while automated insulin delivery (AID) systems have further advanced with closed-network architecture enabling real-time algorithmic insulin adjustments. Recent studies demonstrate AID systems achieved an 11% TIR increase, a 0.95% TBR reduction, and a 12% decrease in time above range versus multiple daily injections or traditional pumps5, with meta-analyses confirming mean HbA1c reductions of 0.36%6. Current guidelines recommend AID systems as preferred therapy over multiple daily injections in type 1 diabetes, and consideration for type 2 diabetes patients on basal insulin not achieving glycemic targets. The 2026 ADA Standards delineate the CGM technologies—including real-time CGM (rtCGM), intermittently scanned CGM (isCGM), over-the-counter CGM systems, and clinic-owned professional CGM—emphasizing individualized device selection according to patients' clinical needs, personal preferences, and digital literacy1. While both rtCGM and isCGM demonstrate superiority over SMBG, rtCGM may provide greater HbA1c reductions in insulin-treated populations through real-time alerts facilitating timely treatment adjustments and hypoglycemia intervention7. CGM is essential for type 1 diabetes, offering high-resolution glucose data that improves TIR, reduces hypoglycemia, and enhances patient confidence. Guidelines recommend early adoption, even at diagnosis1. The clinical data showed CGM use is associated with a reduction of approximately 30 min in daily hypoglycemia8. The landmark Juvenile Diabetes Research Foundation study demonstrated age-related differences: adults ≥25 years showed excellent adherence with ~0.5% HbA1c reductions, while pediatric populations exhibited lower wear time and smaller improvements9. These findings underscore the importance of age-appropriate education and support strategies to optimize CGM utilization across all age groups. Beyond physiological metrics, the psychosocial impact is also profound. Quality of life (QoL) assessments indicate 15–20% gains in adults, primarily reflecting a marked reduction in glucose-related anxiety10. CGM use has expanded beyond intensive insulin therapy, with TIR associated with all-cause mortality, cognitive function, and long-term complications2, 3. A 2024 meta-analysis demonstrated mean HbA1c reductions of approximately 0.3% across diverse treatment modalities, including non-insulin therapies7. CGM users achieve substantially higher TIR—typically a 10–15% improvement, translating to approximately 2.4–3.6 additional hours per day in range—alongside a significant reduction in TBR1. These clinical gains are further characterized by a 3–5% decrease in glycemic variability and a 10–minute reduction in daily hypoglycemia, particularly during nocturnal periods8. Emerging evidence also shows that transitioning from conventional therapy to AID systems can achieve 0.6% HbA1c reductions and 16% TIR increases11, suggesting that AID may effectively overcome therapeutic inertia in appropriate candidates. The comparative improvements in HbA1c levels, glycemic variability, TIR, QoL, and hypoglycemia duration are summarized in Figure 1 1, 6, 8, 10. Evidence for CGM in gestational diabetes mellitus (GDM) is emerging. Both CGM and AID improve glucose variability and patient satisfaction, with a trend toward better maternal and neonatal outcomes, including reducing preterm deliveries and large-for-gestational-age neonates, although the risk of small-for-gestational-age infants requires further investigation12. The 2026 American Diabetes Association (ADA) Standards recommend CGM and AID for pregnant women with type 1 diabetes, with consideration for type 2 diabetes; however, no Food and Drug Administration-approved AID systems currently have pregnancy-specific algorithms, representing an important unmet need1. In hospital settings, better in-hospital glucose predicts lower post-discharge HbA1c13. CGM identifies unstable glycemic patterns independent of HbA1c and quantifies GV, a complications predictor14. Emerging evidence supports CGM for lifestyle modification in diabetes, prediabetes with overweight and obesity. CGM consistently improves quality of life, treatment satisfaction, and confidence in self-care decisions. Barriers include cost, reimbursement variability, digital literacy, and cultural factors15. Technical limitations include sensor interference from substances. In vitro studies identified potential environmental challenges, including aviation conditions where altitude-related atmospheric pressure changes may affect both pump insulin delivery rates and CGM accuracy16. Guidelines emphasize education without creating access barriers, with certified diabetes educator training recommended for pumps and AID systems. Early alarm settings training minimizes fatigue, and sustained access is essential1. Standardized One-page reports supporting shared decision-making was encouraged. Ethical concerns regarding AI algorithms—particularly transparency, bias, and autonomy—require evaluation17. AI foundation models represent a paradigm shift, transforming CGM from a monitoring tool to a predictive digital biomarker. GluFormer, trained on millions of glucose measurements, captured individual-specific glycemic signatures and identified 66% of incident diabetes cases and 69% of cardiovascular deaths in the top risk quartile versus 7% and 0% in the bottom quartile over an 11-year follow-up18. Future directions include fully closed-loop systems, non-invasive glucose sensors, enhanced AI-driven decision support, and precision-medicine approaches integrating CGM data with genomics and behavioral patterns. CGM has fundamentally reshaped diabetes management through real-time glycemic patterns assessment enabling precise clinical decisions. Integration with AID systems and standardized metrics like TIR—now recognized as robust outcome predictors—provides critical information beyond HbA1c. With increasing accessibility and diminishing technological limitations, CGM is becoming central to personalized diabetes care, with forthcoming innovations further improving accuracy, automation, and clinical impact across diverse populations. The authors declare no conflict of interest. Approval of the research protocol: N/A. Informed Consent: N/A. Approval date of Registry and the Registration No. of the study/trial: N/A. Animal Studies: N/A. Yi-Der Jiang is an Editorial Board member of Journal of Diabetes Investigation and a co-author of this article. To minimize bias, he was excluded from all editorial decision-making related to the acceptance of this article for publication. The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
Subjects
continuous glucose monitoring
diabetes mellitus
glycemic control
Type
journal article