Expanding Use of Continuous Glucose Monitoring in Type 2 Diabetes

The Centers for Disease Control and Prevention estimates that approximately 38 million individuals across all age groups, 11% of the US population, are affected by diabetes, with more than 90% of diagnoses being type 2 diabetes (T2DM). ¹ The 2018 diabetes report from the National Institute of Diabetes and Digestive and Kidney Diseases revealed that among adults diagnosed with diabetes, 14% depend solely on insulin, 13% utilize a combination of insulin and oral medication, 57% use oral medication only, and 16% manage their condition through diet and exercise alone. ²

Treatment regimens for T2DM typically incorporate pharmacologic therapies, such as oral medications and injectable agents, alongside lifestyle interventions, such as diet modifications, exercise routines, and weight management. Self-monitoring of blood glucose (SMBG) is a cornerstone of management. Although valuable, the traditional SMBG method of fingerstick testing with a home glucose meter provides only snapshots of glucose levels, potentially missing critical fluctuations, and trends.

Continuous glucose monitor (CGM) technology has emerged as a pivotal tool in diabetes management, providing an alternative way to monitor glucose levels and track trends over time. The 2024 American Diabetes Association guideline recommends CGM use among people with type 1 diabetes and insulin-treated T2DM. ³ However, CGMs' efficacy and potential benefits in optimizing glycemic control for individuals with T2DM not on insulin therapy remain inconclusive and underexplored.

The rationale for CGM use in T2DM lies in its ability to facilitate a comprehensive understanding of glucose levels in response to various factors, such as meals, exercise, stress, and medications, allowing for personalized and timely adjustments to treatment regimens, enhancing glycemic control, empowering individuals to make informed health decisions and engage in their care, and potentially reducing the risk of long-term complications associated with diabetes. With the majority of persons with diabetes (PWD) not on insulin therapy, there is potential to extend the benefits of CGMs to a wider population.

This narrative review summarizes the current evidence on the benefits of CGM use in individuals with non-insulin-treated T2DM.

Methods

A comprehensive literature search from 2008 to 2024 was conducted using PubMed, using key terms including "type 2 diabetes," "continuous glucose monitoring," and "non-insulin." The search included high-impact and peer-reviewed journals dedicated to diabetes research and technology, such as Diabetes Technology & Therapeutics and Diabetes Spectrum. Studies that assessed CGM use in individuals with T2DM on oral or noninsulin injectable medications were included. Research that focused solely on type 1 diabetes or insulin-treated were excluded unless non-insulin-treated individuals were evaluated as a separate subgroup. Based on their relevance to the topic, 4 randomized controlled trials and 2 retrospective/observational studies were selected. The studies were evaluated for sample size, duration, demographics, and primary or secondary outcomes, including A1C levels, time in range (TIR), and behavioral modifications. The findings from each study were then analyzed to provide insights into the effectiveness of CGMs in non-insulin-treated T2DM.

Literature Review

Allen et al

Drawing from established behavioral change theory and recognizing the well-documented benefits of physical activity, Allen et al ⁴ conducted a randomized controlled pilot study to assess whether utilizing CGMs alongside counseling in individuals with T2DM could foster short-term improvements in physical activity and self-efficacy behavior. The authors integrated CGM feedback into a relatively brief (90 minute) personalized counseling session on physical activity to evaluate the impact CGM feedback has on physical activity. ⁴

Participants were enlisted from 2 health systems in western Massachusetts. Inclusion criteria encompassed individuals who were diagnosed with T2DM, were 18 years or older, engaged in physical activity for no more than 2 days per week, had an A1C >7.5%, were not on insulin therapy, and possessed English reading and speaking proficiency. Exclusion criteria included inability to walk 0.25 miles within 10 minutes, usage of glucocorticoids, or not passing the history and physical activity prescreening evaluation. ⁴

During the study, participants underwent evaluations for various outcome measures, including physical activity, physical activity self-efficacy, A1C, blood pressure, and body mass index (BMI). Data were collected at baseline and at the end of week 8. All participants received 90 minutes of personalized diabetes education based on the International Diabetes Center curriculum and additional counseling via phone at week 4. Both the control and intervention groups wore physical activity monitors for 7-day periods at baseline and week 7. The intervention group wore a CGM for 3 days and received an additional 90 minutes of personalized CGM counseling that involved reviewing CGM graphs to illustrate the impact of physical activity on glucose reductions. ⁴

A total of 52 participants completed the study, 25 in the control group and 27 in the intervention group. The intervention group showed statistically significant (P < .05) improvements in A1C, BMI, physical activity self-efficacy, and physical activity levels. Specifically, mean A1C levels decreased from 8.9% to 7.7% in the intervention group, compared to the control group, which decreased from 8.4% to 8.1%. Additionally, BMI decreased in the intervention group from 37.11 kg/m² to 36.58 kg/m², whereas in the control group, it slightly increased from 33.81 kg/m² to 33.93 kg/m². Moreover, the intervention group demonstrated an increase in physical activity self-efficacy and physical activity levels, whereas these categories decreased in the control group. ⁴

The investigators suggest that incorporating CGM feedback into personalized counseling sessions on physical activity is effective and led to reductions in A1C and BMI and improvements in physical activity levels. ⁴

Wada et al

Wada et al ⁵ conducted a multicenter, open-label, parallel-group, randomized controlled trial to evaluate the effects of CGMs compared to SMBG on glycemic control among Japanese individuals with non-insulin-treated T2DM. The study included adults with diagnosis of T2DM, A1C between 7.5% and 8.5%, and ages between 20 and 70 years. Exclusion criteria comprised those on insulin therapy, prior use of CGM or SMBG, individuals on dialysis, severe renal impairment (estimated glomerular filtration rate <30 mL/ min/1.73 m²), history or current proliferative diabetic retinopathy, inability to operate the CGM device, or deemed unsuitable study candidates by health care providers. All participants wore a CGM sensor for at least 7 days, received education on device usage based on their assigned group, and were advised on lifestyle/diet modifications according to glucose levels. CGM or SMBG devices were then provided for 12 weeks, with the SMBG group wearing blinded sensors during the last 2 weeks of this period. The primary outcome measured was the change in A1C levels, and secondary outcomes included changes in mean glucose levels, glucose variability indices, and TIR. ⁵

A total of 100 participants were randomized into either the CGM (n = 49) or SMBG (n = 51) group. Both groups exhibited a significant reduction in A1C levels at 12 weeks (CGM -0.43%, P <.001; SMBG -0.30%, P = .001) compared to baseline, with no significant differences between the groups. However, at 24 weeks, the CGM group demonstrated a significant decrease in A1C (-0.46%, P < .001) compared to the SMBG group (-0.17%, P = .124), showing a -0.29% difference (P = .022). Changes in BMI, blood pressure, fasting plasma glucose, and other laboratory data did not differ significantly between the 2 groups from baseline to 24 weeks, except for a significantly higher HDL in the CGM group compared to the SMBG group. ⁵

Glycemic outcomes derived from sensor data revealed significant improvements in mean glucose levels, glucose variability, TIR, and time spent in hyperglycemia in the CGM group, with no significant differences in glucose coefficient of variation and time spent in hypoglycemia. The duration of TIR within a 24-hour period increased significantly in the CGM group compared to the SMBG group, with an adjusted mean difference of 2.36 hours (95% CI, 1.21-3.51; P < .001). Adverse events were minimal and primarily related to skin irritation with the sensor, all of which resolved without serious consequences, indicating the safety and tolerability of both monitoring methods. The authors concluded CGM use is effective in achieving glycemic control in non-insulin-treated T2DM. ⁵

Cox et al

Another randomized clinical trial, conducted by Cox et al, ⁶ evaluated the efficacy of a novel lifestyle intervention, glycemic excursion minimization (GEM), combined with CGM feedback (GEM^CGM), in optimizing glycemic control among adults with T2DM who were not using insulin. Unlike traditional approaches that use weight loss as a mechanism to improve A1C, GEM focuses on minimizing postprandial glucose excursions, which are a primary contributor to A1C, including meals, snacks, and drinks, through education on physical activity and dietary choices, emphasizing glycemic load or index. ⁶

The study included adults diagnosed with T2DM for fewer than 11 years, ages between 30 and 80 years, with A1C greater than or equal to 7%, not using insulin, capable of walking for 30 minutes, and interested in CGMs. Individuals using insulin or nondiabetic medications affecting blood glucose control were excluded. Over 5 months, all participants continued their routine care with their respective health care providers, who adjusted medications as deemed appropriate based on clinical needs. The GEM^CGM group attended 4 group sessions led by a diabetes nurse educator, focusing on food and activity choices, carbohydrate reduction, increased physical activity, and lifelong management strategies. Each participant received a 7-day sensor to insert at each session, with an additional sensor provided 1 month after the fourth session. Follow-up assessments were conducted 3 months after the final session, and analyses of covariance were utilized to compare preintervention and postintervention measures between the control and intervention groups. The primary outcomes included A1C and the medication effect score (MES), reflecting changes in the overall intensity of pharmacologic management of diabetes. ⁷ Other variables assessed included diabetes knowledge, carbohydrate intake, and physical activity, and secondary outcomes encompassed quality of life (QoL) and diabetes empowerment. ⁶

A total of 30 adults were randomly assigned to the GEM^CGM (n = 20) or routine care (RC; n = 10) groups. A1C was reduced by 1.3% in the GEM^CGM group versus 0.19% in the RC group (P = .03), and MES decreased by 0.02 in the GEM^CGM group compared to an increase of 0.81 in the RC group (P = .009), indicating GEM^CGM resulted in greater glycemic control improvement and a lower overall intensity of diabetes medications compared to RC. TIR increased from 44% to 50% in the GEM^CGM group, and the RC group increased from 42% to 43%. The GEM^CGM participants also increased in diabetes knowledge, the World Health Organization's psychological QoL score, and diabetes empowerment while reducing carbohydrate intake significantly more than RC, without adverse effects on hypoglycemia or lipid profiles. ⁶

Overall, this study demonstrates the potential benefits of combining CGM use with the GEM lifestyle intervention for improved glycemic control among individuals with non-insulin-treated T2DM, as evidenced by reductions in A1C levels and improvements in TIR compared to RC. Although the exact contribution of CGM use, GEM, or their combination to the observed benefits in glycemic control, physical activity, and dietary choices is challenging to isolate, the results suggest a potential synergistic effect when integrating CGM use with lifestyle interventions like GEM to maximize the effectiveness of CGM use and enhance diabetes outcomes. ⁶

Wright et al

A retrospective, observational database (IBM Explorys) analysis by Wright et al ⁸ assessed the impact of a flash CGM prescription in a large group of individuals with T2DM and suboptimal glycemic control who were either treated with basal insulin or noninsulin therapies. Inclusion criteria included the following: T2DM diagnosis, age <65 years, no prior CGM use, a baseline A1C of greater than or equal to 8.0%, flash CGM prescription between October 2017 and February 2020, not utilizing short- or rapid-acting insulin, baseline A1C test within 180 days of flash CGM prescription, and a postassessment A1C value between 60 and 300 days after flash CGM was prescribed. The primary outcome was change in A1C after flash CGM was prescribed. Results were stratified based on treatment (insulin vs. noninsulin) and baseline A1C (8.0% to <10.0%, 10.0% to <12.0%, and 12.0%).

The analysis included 1034 individuals diagnosed with T2DM. The majority were over 50 years old, had baseline A1C levels ranging from 8.0% to 10.0%, and were undergoing noninsulin treatment. Additionally, most individuals had hypertension and dyslipidemia, with over half having a BMI exceeding 30 kg/m². ⁸

At the conclusion of the study (with an average follow-up of 159 days), a 1.5% decrease (± 2.2%) in A1C was noted across the entire cohort. In a subgroup analysis comparing insulin versus noninsulin therapy, individuals not on insulin exhibited a greater reduction in A1C (1.6% ± 2.3%) compared to those treated with basal insulin (1.1% ± 1.9%) despite similar baseline A1C prior to the prescription of the flash CGM system. As expected, the most substantial reductions in A1C levels were observed in individuals with a baseline A1C of 12% or higher, followed by those with a baseline A1C ranging from 10% to less than 12%. ⁸

According to the authors, the findings of this study are generalizable to most individuals with T2DM treated with noninsulin therapy and with inadequately controlled A1C and justify the use of flash CGM due to significant A1C reductions. ⁸

Aronson et al

Aronson et al ⁹ sought to evaluate the effectiveness of intermittently scanned CGM (isCGM) integration with diabetes self-management education (DSME) in T2DM in individuals not on insulin therapy. This multisite, 16-week, open-label, randomized controlled trial included 116 participants with T2DM, diagnosed for at least 6 months, with an A1C of 7.5% or higher, and no previous experience with CGM. Participants were divided into 1 groups: one received isCGM + DSME and the other DSME alone. All participants in the study were equipped with a blinded CGM device, specifically, Freestyle Libre Pro, for 14 days at baseline and week 14. Both groups attended 6 follow-up visits with diabetes educators for personalized feedback. ⁹

Participants were primarily on metformin, SGLT-2 inhibitors, DPP-4 inhibitors, and GLP-1 receptor agonists. The primary outcome was the difference in TIR during the final 2 weeks. Secondary outcomes included average A1C and percentage of time spent below the target range (<70 mg/dl) and above target range (>180 mg/dl). The isCGM + DSME group had a significant increase in TIR from 56.3% to 76.3%, compared to the DSME-only group's increase from 57.5% to 65.6% (adjusted mean difference: 9.9%, P = .009). Both groups showed improvements in A1C levels, with the CGM + DSME group from 8.5% to 7.6% and the DSME-only group from 8.7% to 8.1% (adjusted mean difference: 0.3%, P = .048). The most pronounced benefits were observed in individuals with higher baseline A1C levels. However, factors such as the duration of diabetes, the number of antihyperglycemic agents used, and the use of GLP-1 RA did not affect the primary outcome. ⁹

The investigators found that combining CGM with DSME significantly increased TIR and modestly reduced A1C in individuals with T2DM not using insulin. The most notable improvements were observed in individuals with higher initial A1C levels. These results underscore the value of combining CGM use with structured diabetes education to enhance glycemic control in this population. ⁹

Powell et al

Most recently, Powell et al ¹⁰ conducted a retrospective study comparing glycemic control before and after implementing CGM use in individuals with T2DM. ¹⁰ Additionally, the authors examined how a collaborative care model influences the management of diabetes. The primary outcome was difference in A1C after CGM placement. Secondary outcomes were changes in subsequent A1C, utilization of a clinical pharmacist intervention, and variations in A1C among individuals treated with and without insulin.

A total of 45 participants, with an average age of 52 years and average A1C of 9.8%, were enrolled and stratified by insulin usage and interaction with clinical pharmacists. Among them, 11 were not on insulin therapy, and 26 attended follow-up appointments with a clinical pharmacist. A1C was measured before and up to 18 months after CGM placement, with the first measurement occurring approximately 3 months after placement. In addition, 41 participants completed an A1C measurement after CGM placement. ¹⁰

There was a significant decrease in A1C across all groups; mean decrease was 1.9% (P < .001). This reduction was noticeable among individuals who were taking insulin (1.8%) and those who were not on insulin (2%). The most substantial A1C decrease was observed in those receiving collaborative care with a clinical pharmacist, with a mean reduction of 2.5% (P <. 001). ¹⁰

The findings from this study confirm a decrease in A1C among PWD utilizing CGMs regardless of insulin use. Notably, those who received care through a collaborative care model involving a clinical pharmacist experienced the most substantial decrease in A1C. Although the reduction in A1C among those not comanaged by a pharmacist was not statistically significant (mean reduction of 0.8%, P = .07), it may be considered clinically meaningful, suggesting that CGM use alone can improve glycemic control, although a multidisciplinary team may enhance outcomes further. ¹⁰

Discussion

Currently, there is uncertainty regarding the benefits of CGM use for individuals with non-insulin-treated T2DM. This narrative review synthesized and evaluated findings from 6 distinct studies to comprehensively assess the impact of CGM on A1C, TIR, and behavioral modification in this population.

The reviewed studies consistently demonstrated significant reductions in A1C levels among non-insulin-treated T2DM individuals using CGM, with reductions ranging from 0.46% to 2.3%.^{4-6 ,8-10} The National Glycohemoglobin Standardization Program (NGSP) suggests a shift in A1C of 0.5% or more is both statistically and clinically meaningful. ¹¹ While only Wada et al ⁵ fell short of this threshold (0.46%), all 5 other studies exhibited reductions meeting or surpassing the 0.5% decrease, underscoring the observed changes in A1C associated with CGM use resulted in meaningful improvements in diabetes management and outcomes. ^4,6,8-10

Moreover, 3 studies found increases in TIR ranging from 6% to 30%.^5,6,9 The consensus from the International Conference on Advanced Technologies and Treatments for Diabetes suggests that every 5% increase in TIR is linked to meaningful clinical benefits for individuals with T2DM. ¹² Aronson et al ⁹ and Wada et al ⁵ reported a noticeable 20% and 30% increase in TIR among participants utilizing CGM devices, respectively. Although the 6% increase in TIR observed in the GEM^CGM group in the study by Cox et al ⁶ did not reach statistical significance when compared to the RC group (1%), individuals are still likely to be experiencing benefits from improved glycemic control. In addition to improving glycemic control, CGM use was found to positively influence behavioral changes in individuals with T2DM not on insulin therapy. Both Allen et al ⁴ and Cox et al ⁶ highlighted that CGM use facilitated increased physical activity, reduced carbohydrate intake, and promoted healthier lifestyle choices. These changes were achieved through mechanisms such as real-time feedback, increased awareness, and behavioral reinforcement. Real-time feedback provided by CGMs enabled individuals to understand how lifestyle factors impact blood glucose levels and to make informed decisions regarding their diet, exercise or activities, and overall diabetes management. The correlation between behavioral changes and glucose levels serves as a motivating factor for individuals to adopt healthier habits, as seen in these studies. Whereas CGM use alone has shown promising results in improving glycemic outcomes and facilitating behavioral modifications, several studies incorporated additional interventions, including counseling sessions, DSME, GEM, and collaboration with clinical pharmacists.^4-6,8 Integrating CGM with supplementary lifestyle or behavioral interventions may further enhance outcomes such as reduced A1C and increased TIR. However, the specific contribution of CGM versus other interventions or the combinations warrants further investigation.

Despite the positive findings, several limitations persist across all the reviewed studies, including small sample sizes, short-term follow-up, and potential confounding variables. Future research should aim to address these limitations by conducting real-world studies to confirm these findings and explore their implications in diverse populations. Additionally, cost-effectiveness and reported outcomes can further inform clinical practice.

Recently, Dexcom ¹³ announced the approval of Stelo, the first FDA-cleared glucose biosensor that will be available without a prescription. Stelo is indicated for individuals 18 years and older not on insulin and is expected to become accessible starting in the summer of 2024. Similarly, Abbott ¹⁴ recently announced the FDA clearance of Libre Rio, an over-the-counter CGM system, indicated for the same individuals. These approvals highlight the growing recognition of CGMs' role in diabetes management and its potential to positively impact the lives of individuals with non-insulin-treated T2DM by providing valuable insights into glucose levels, allowing for timely adjustments to treatment regimens. As CGM technology continues to evolve and become more accessible, it is important to understand how these advancements can be effectively integrated into patient care. Educating individuals on the proper use of these devices, interpreting CGM data, and translating findings into actionable steps are key responsibilities for diabetes care and education specialists. By leveraging these new tools, individuals can make informed decisions about their diabetes management, improving outcomes for those with non-insulin-treated T2DM.

Conclusion

This narrative review explores the impact of CGM use in individuals with non-insulin-treated T2DM, suggesting improvements in glycemic control and positive behavioral changes. Coupling CGM use with appropriate education, counseling, and lifestyle modifications may further enhance outcomes. However, more research is needed to better understand the ideal usage and long-term advantages of CGM use in this population. ■