Does Continuous Glucose Monitoring Use Prompt Greater Engagement in Self-Management? A Randomized Controlled Trial Focusing on Adults With Type 2 Diabetes

Abstract

Background:

Continuous glucose monitoring (CGM) promotes glycemic benefits in adults with type 2 diabetes (T2D), including insulin users as well as noninsulin users, often with minimal professional support. To investigate whether these benefits may stem from increased user engagement in self-management, we conducted a randomized controlled trial comparing the impact of CGM versus self-monitoring of blood glucose (SMBG) on self-reported engagement and HbA1c in CGM-naïve adults with T2D.

Methods:

Potential participants completed the Impact of Glucose Monitoring on Self-Management Scale (IGMSS) and an HbA1c home test to confirm eligibility (>7.5%). N = 110 eligible participants were randomized to receive a FreeStyle Libre 3 (CGM arm) or a FreeStyle Precision Neo Blood Glucose Monitoring System (SMBG arm). The IGMSS and HbA1c home test were repeated after three months. Latent change score models estimated group differences in outcomes over time.

Results:

CGM users reported significantly greater engagement with T2D self-management than SMBG users (IGMSS total b = 0.61, P < .001), including greater gains on all three major subscales, capability (b = 0.76, P < .001), opportunity (b = 0.46, P = .001), and motivation (b = 0.66, P < .001). CGM users also saw a significant HbA1c drop of ~1% (9.2% to 8.3%, P < .001, d = .65), with less than half the reduction in SMBG users (8.9% to 8.4%, P = .065, d = .30). However, the effect of group on HbA1c change did not reach statistical significance (P = .170), likely due to limited sample size.

Conclusions:

These findings suggest that introducing CGM to adults with T2D heightens users’ engagement with their own diabetes care and also improves glycemic control more than providing SMBG.

Keywords

CGM type 2 diabetes self-management engagement

Introduction

In populations with type 2 diabetes (T2D), including both insulin users as well as noninsulin users, research investigations have documented that introducing continuous glucose monitoring (CGM) contributes to significant glycemic benefits over time.^1-6 In several studies, these positive outcomes have been observed to occur with minimal or no health care professional (HCP) intervention, pointing to the likely mediational influence of changes in self-management behavior.^5,7,8 In essence, CGM may exert many of its glycemic gains by enhancing users’ interest, motivation and sense of engagement with their own diabetes care. This may be of critical importance, since it is well-recognized that adults with T2D are often suboptimally engaged with their own care.⁹ However, data directly supporting this hypothesized influence of CGM on user engagement are rather sparse. Continuous glucose monitoring use in people with T2D has been linked with increases in treatment satisfaction^1,10,11 as well as positive behavior change over time.^12,13 From these findings, we may infer that users’ attitudes have shifted toward greater engagement, but this has yet to be directly tested. Qualitative studies also provide support for this premise, with one recent interview study of 34 T2D study participants highlighting how interest and engagement with self-management rose markedly after three months on CGM.¹⁴ In total, though these preliminary data are intriguing, we still lack solid quantitative evidence regarding how CGM may impact on these key diabetes-related attitudes of individuals with T2D.

To address this gap, we designed a three-month, randomized controlled intervention trial for CGM-naïve adults with T2D, where half of the sample received a new CGM system and half received a conventional fingerstick blood glucose monitor. Assessing user engagement at study start and end with the Impact of Glucose Monitoring on Self-Management Scale,¹⁵ we hypothesized that CGM users would demonstrate significantly greater gains in self-reported engagement in diabetes self-care — as well as greater glycemic benefits— than those using the more traditional SMBG (self-monitoring of blood glucose) method.

Research Design and Methods

Participants

Adults with T2D from across the United States, recruited through Taking Control of Your Diabetes (TCOYD, a nonprofit organization devoted primarily to remote diabetes education and support), Thrivable (a for-profit company connecting people with diabetes and research teams), and a number of social media sites, were informed about the study and invited to complete an online screening survey. Inclusion criteria were T2D duration >12 months, current use of ≥1 antihyperglycemic agent, age ≥ 21 years, HbA1c > 7.5%, no prior experience with CGM, English-speaking, and access to an appropriate smartphone that would allow connection to Libre 3. Individuals were excluded from participating if they were currently using fast-acting insulin (i.e. on multiple daily injection therapy or using an insulin pump). A total of 1722 individuals responded to the invitation and completed the online screening survey, with 307 found to be initially eligible (17.8%). The main reasons for ineligibility were self-reported HbA1c ≤ 7.5%, prior CGM use and current use of fast-acting insulin (Figure 1).

Figure 1.

CONSORT flow diagram.

Procedures

Subjects who met the initial screening criteria were asked to provide consent and complete an online questionnaire battery. When these were successfully completed, subjects were mailed an HbA1c home test kit to determine whether subjects met the final study criterion (HbA1c > 7.5%). Qualifying individuals were then randomized either to the CGM or SMBG study arm. This was accomplished by means of a randomization scheme created by a blinded statistician who used a random sequence generator. Both participants and staff were blinded to the group allocation until after the consent form was signed. Continuous glucose monitoring arm participants were sent a three-month supply of FreeStyle Libre 3 sensors, while SMBG arm participants were sent a FreeStyle Precision Neo Blood Glucose Monitoring System and a three-month supply of glucose test strips. To get started, all subjects were informed that instructions and support were available at the respective FreeStyle websites, and they were encouraged to use their devices for daily treatment and management decisions. If requested, research assistants were available by phone if participants had questions or concerns about using their glucose monitoring systems.

At three months, participants were re-contacted and requested to complete a second online survey and home HbA1c test kit. A $50 electronic gift card was sent to subjects after study completion. All data were kept in a central database using a HIPAA-protected server, with no linkages to personal health information or personal identifiers. The research protocol was approved by Advarra, an independent Institutional Review Board. This trial was not subject to clinical trial registration requirements and was exploratory in nature; however, all study procedures followed a prespecified protocol (available from authors upon request), and no major changes aside from sample size (described in Data Analysis) were made to methods or outcomes after study start.

Outcome Assessment

The baseline and three-month surveys included items to identify sociodemographic and health characteristics, including age, gender, ethnicity, current diabetes medication and current mode of glucose monitoring (if any). Also included was the newly developed Impact of Glucose Monitoring on Self-Management Scale (IGMSS),¹⁵ which assesses user engagement with 18 items rated from one (strongly disagree) to five (strongly agree). The IGMSS forms a total score and three major subscale scores: Capability (indicating how glucose monitoring was providing the user with needed knowledge and supporting better health, sample item: “helped me to understand how specific foods, exercises and stress directly affects my glucose levels”), Opportunity (pointing to how glucose monitoring opened the door to more productive interactions with HCPs, friend and family; sample item: “helped me to have more useful conversations with my health care provider about diabetes management”), and Motivation (signifying how glucose monitoring impact on motivation for self-management; sample item: “helped me to become more optimistic about living with diabetes”). Further information regarding the internal consistency, factor structure and construct validity of the IGMSS can be found in the recent paper by Vallis and colleagues.¹⁵

At baseline and three months, all subjects were also mailed an HbA1c home test kit (University of Minnesota Advanced Research Diagnostic Laboratory) to complete and return to the research team.

Data Analysis

All analyses were conducted in IBMM SPSS Statistics and Mplus V8.11. Two-tailed p-values (alpha = .05) were used. Analyses used an intention-to-treat (ITT) approach, including data from all participants in their assigned groups, regardless of adherence to protocol. Descriptive statistics were computed to summarize baseline characteristics, including sociodemographic and other self-report variables. Differences between the two intervention groups at baseline were tested with independent samples t tests (continuous variables) or chi-square tests (categorical variables).

This pilot trial was not powered to detect between-group differences in glycemic outcomes. However, sample size estimates for the primary outcome (IGMSS) were informed by prior psychometric data from the scale’s development. Based on standard deviations of 0.60 to 0.70, a minimally important difference was defined as 0.5 SD. Detecting this difference with 80% power and a 95% confidence level required a sample of 63 participants per group. To account for expected attrition, the protocol targeted 75 participants per group (N = 150). Due to slower-than-expected enrollment, the final randomized sample was 110 (n = 54 CGM, n = 56 SMBG), which limited statistical power for detecting effects. Therefore, effect sizes are estimated and emphasized in addition to statistical significance. Since a clinically meaningful level of change on the IGMSS has not yet been defined, standardized effect size estimates (Cohen’s d, standardized regression coefficient β) are used. In addition, HbA1c results are examined with respect to the previously established definition of clinically significance of 0.3% differential change over time between groups.¹⁶

The primary analyses aimed to evaluate the effectiveness of CGM versus SMBG use on self-reported engagement (IGMSS scores) from baseline to three months. Secondary analyses assessed group differences in glycemic (HbA1c) change. Both outcomes were tested in a series of models, beginning with univariate (unadjusted) paired samples t tests of within-group change over time. These were followed by formal, adjusted tests of Group × Time differences in multivariate models. In multivariate analyses that tested the focal adjusted Group × Time interaction, the following covariates were included: race/ethnicity (non-Hispanic/Latino White = 1), gender (female = 1), age, and baseline insulin use. Change over time from baseline to three months was analyzed using a latent change score approach in the structural equation modeling framework of Mplus v8.11.¹⁷ This maximum-likelihood-based approach is optimal for two time point designs because it makes use of all available data, including those from participants with missing data at three months.¹⁸ Therefore, the full sample (N = 110) was included in these analyses. An error-free latent variable was created to represent change in the outcome over the three-month period while accounting for baseline level (intercept) and their covariance. Treatment group and covariates were included as predictors of the latent change score in a regression model. Unstandardized (b) and standardized (β) regression coefficients are both provided.

Results

As illustrated in the CONSORT diagram (Figure 1), of the 307 subjects found to be initially eligible via the online screening, an additional 106 were excluded from the study after failing to complete the online consent or deciding to not complete the online baseline survey. The remaining 201 were mailed an HbA1c test kit, which resulted in another 91 subjects excluded (63 did not return the kit, 28 were found to have an HbA1c ≤ 7.5%). This resulted in a final sample of 110 subjects to be randomized (35.8% of initially eligible participants), resulting in 54 receiving the FreeStyle Libre 3 and 56 receiving the FreeStyle Precision Neo. Of note, 87.0% of CGM users (n = 47) and 82.1% of SMBG users (n = 46) returned to complete the second survey and home HbA1c test kit at three months. All 110 subjects were enrolled and began their participation in the study between November 2023 and May 2024.

Participant Characteristics

Table 1 presents the baseline characteristics for the total sample and within each treatment group, as well as the results of baseline difference t tests. The majority of the sample was female (67.3%), non-Hispanic White (72.7%), well-educated (50.9% college educated) and reported annual household earnings ≥$50,000 (58.1%). Mean age was 57.7 years (SD = 12.0) and mean HbA1c was 9.0% (SD = 1.6). Almost all participants (95.5%) reported current use of ≥1 oral diabetes medication, while 34.5% were using a once weekly GLP-1 medication and 23.6% were on basal insulin. More than two-thirds of subjects (70.9%) reported they were regularly monitoring their glucose levels with a traditional glucose meter (SMBG). There were no significant differences in baseline characteristics between participants in the CGM arm versus the SMBG arm.

Table 1.

Participant Baseline Characteristics, N (%).

	Total sample(N = 110)	Libre CGM group(n = 54)	SMBG group(n = 56)	P (Libre CGM vs SMBG group)^a
Age, M (SD)	57.7 (12.0)	57.1 (12.2)	58.2 (11.9)	.621
Female	74 (67.3%)	40 (74.1%)	34 (60.7%)	.135
Race/ethnicity
Asian	6 (5.5%)	5 (9.3%)	1 (1.8%)	.315
Black/African American	13 (11.8%)	6 (11.1%)	7 (12.5%)
Latino/Hispanic/Chicano	8 (7.3%)	4 (7.4%)	4 (7.1%)
Pacific Islander	1 (0.9%)	1 (1.9%)	0 (0%)
Native American	0 (0%)	0 (0%)	0 (0%)
Non-Hispanic White/Caucasian	80 (72.7%)	38 (70.4%)	42 (75.0%)
Multiple race/ethnic backgrounds	2 (1.8%)	0 (0%)	2 (3.6%)
Other	0 (0%)	0 (0%)	0 (0%)
Education
Less than high school	0 (0%)	0 (0%)	0 (0%)	.374
High school degree or equivalent	22 (20.0%)	10 (18.5%)	12 (21.4%)
Some college, no degree	15 (13.6%)	7 (13.0%)	8 (14.3%)
Associate’s degree	17 (15.5%)	5 (9.3%)	12 (21.4%)
Bachelor’s degree	37 (33.6%)	21 (38.9%)	16 (28.6%)
Graduate degree or higher	19 (17.3%)	11 (20.4%)	8 (14.3%)
Income
Under $15 000	7 (6.4%)	5 (9.3%)	2 (3.6%)	.197
$15 000 to $24 999	8 (7.3%)	4 (7.4%)	4 (7.1%)
$25 000 to $49 999	28 (25.5%)	12 (22.2%)	16 (28.6%)
$50 000 to $74 999	23 (20.9%)	8 (14.8%)	15 (26.8%)
$75 000 to $99 999	21 (19.1%)	11 (20.4%)	10 (17.9%)
$100 000 to $149 000	11 (10.0%)	8 (14.8%)	3 (5.4%)
$150 000 to $200 000	5 (4.5%)	4 (7.4%)	1 (1.8%)
$200 000 and up	4 (3.6%)	2 (3.7%)	2 (3.6%)
Undisclosed	3 (2.7%)	0 (0%)	3 (5.4%)
Currently using SMBG	78 (70.9%)	36 (66.7%)	42 (75.0%)	.336
Diabetes medication
On oral medication	105 (95.5%)	53 (98.1%)	52 (92.9%)	.183
On basal insulin	26 (23.6%)	14 (25.9%)	12 (21.4%)	.579
On pre-mixed insulin	3 (2.7%)	1 (1.9%)	2 (3.6%)	.580
On weekly injectable	38 (34.5%)	18 (33.3%)	20 (35.7%)	.793
HbA1c (%), M (SD)	9.0% (1.6)	9.2% (1.6)	8.9% (1.6)	.377
HbA1c (mmol/mol), M (SD)	74.9 (17.5)	77.0 (17.5)	73.8 (17.5)	.377

Abbreviations: CGM, continuous glucose monitoring; SMBG, self-monitoring blood glucose.

p value of group difference effect tested via independent samples t tests or chi-square tests.

Change in Engagement in Diabetes Care Over Time

The primary analyses examined changes in IGMSS scores over time. Results of the latent change score models indicated that, over the three-month study period, CGM users reported that their glucose monitoring systems had helped them to become significantly more engaged with T2D self-management than SMBG users (IGMSS total score, unstandardized b = 0.61, P < .001), which was a medium sized effect (β = 0.30) (Table 2). This included significantly greater gains in all three major IGMSS subscales: capability (b = 0.76, P < .001, β = 0.36), opportunity (b = 0.46, P = .001, β = 0.23), and motivation (b = 0.66, P < .001, β = 0.28). When examining changes in IGMSS within each intervention group via paired sample t tests, the CGM group showed statistically significant improvements in all IGMSS scores (including the Total as well as the Capability, Opportunity, and Motivation subscales) from baseline to three months (all Ps < .05), while SMBG users’ scores did not significantly change (all Ps > .05). Effect sizes measured by Cohen’s d supported these findings. Specifically, in the CGM group, the effect size for improvements in engagement (in the total score and the three major subscale scores) ranged from moderate to large (ds .47-.70), while effect sizes in the SMBG group were small (ds .03-.18).

Table 2.

Change Over Time in Outcomes: Paired Samples t Test and Latent Change Score Model Results.

	Total sample(N = 110)					Libre CGM group(n = 54)					SMBG group(n = 56)					Adjusted group difference in latent change^a(N = 110)
	Mean (SD)		Mean Change	Cohen’s d	P	Mean (SD)		Mean Change	Cohen’s d	P	Mean (SD)		Mean Change	Cohen’s d	p	b	β	P
	Baseline	3 Months	Mean Change	Cohen’s d	P	Baseline	3 Months	Mean Change	Cohen’s d	P	Baseline	3 Months	Mean Change	Cohen’s d	p	b	β	P
IGMSS total	3.7 (0.9)	4.0 (0.8)	0.30^**	.311	.002	3.9 (0.8)	4.4 (0.4)	0.50 ^**	.631	<.001	3.6 (0.9)	3.7 (0.9)	0.11	.099	.485	0.61 ^**	0.30	<.001
Capability	3.8 (0.9)	4.1 (0.8)	0.33^**	.317	.002	3.9 (0.9)	4.5 (0.5)	0.63 ^**	.697	< .001	3.7 (0.9)	3.7 (0.9)	0.04	.034	.812	0.76 ^**	0.36	<.001
Opportunity	3.6 (0.9)	3.9 (0.8)	0.27^**	.293	.004	3.8 (0.8)	4.2 (0.5)	0.36 ^**	.466	.002	3.5 (0.9)	3.7 (0.9)	0.18	.175	.216	0.46^**	0.23	.001
Motivation	3.7 (0.9)	4.1 (0.9)	0.30^**	.268	.009	3.9 (0.9)	4.4 (0.5)	0.24 ^*	.598	.037	3.6 (0.9)	3.7 (1.1)	0.08	.061	.664	0.66^**	0.28	<.001
HbA1c (%)	9.0 (1.6)	8.4 (1.6)	–0.70 ^**	–.452	<.001	9.2 (1.6)	8.3 (1.7)	–0.95 ^**	–.651	<.001	8.9 (1.6)	8.4 (1.5)	–0.46	–.279	.065	–0.39	–0.13	.170

Unadjusted means shown for each time point, with changes from baseline to three months shown for matched pairs only. Cohen’s d effect size can be interpreted as follows: d ~ .2: small effect; d ~ .5: medium effect, d ~ .8: large effect. Medium effect sizes are shown in boldface. Large effect sizes are shown in boldface, underlined.

Abbreviations: CGM, continuous glucose monitoring; SMBG, self-monitoring blood glucose; IGMSS, impact of glucose monitoring on self-management scale.

Covariates included race/ethnicity, gender, age, insulin use. Unstandardized regression coefficient b is adjusted difference in change between groups, while β is the standardized regression coefficient. β can be interpreted as: .1 to .29 ~ small effect; .3 to .49 = medium effect; ≥5 ~ large effect.

P < .01, *P < .05.

Change in Glycemic Control (HbA1c) Over Time

Secondary analyses focused on changes in HbA1c over time, using the same series of models. The effect of group on HbA1c change was not statistically significant (P = .170), but we did observe significant within-group results. Specifically, only CGM users evidenced a statistically and clinically significant A1c drop of nearly 1% (9.2%-8.3%, P < .001, d = .65). In contrast, SMBG users evidenced less than half the reduction in HbA1c that was found for CGM users (0.46%), which was not statistically significant (8.9%-8.4%, P = .065, d = .30). The within-group results therefore showed an unadjusted group differences in change in HbA1c of 0.5%, well past the threshold of a clinically significant group difference. The latent change score model results revealed that the mean difference in latent A1c change over time between groups, adjusting for baseline A1c and covariates, was slightly lower at .39% (unstandardized b). Despite this clinically meaningful pattern of results, it must be re-emphasized that we did not observe significant between-group difference in glycemic outcomes.

Discussion

Our findings suggest that supplying CGM to adults with T2D can enhance user’s interest and engagement with their own diabetes care. These quantitative results build upon previous qualitative research on CGM in adults with T2D¹⁴ and T1D.¹⁹ Specifically, in this current sample of CGM-naïve participants, we found that subjects randomly assigned to receive CGM (FreeStyle Libre 3) for a three-month period reported significantly greater improvement in engagement with diabetes self-care, as assessed by the IGMSS, than subjects provided with an SMBG system (FreeStyle Precision Neo). Significantly greater gains in engagement were observed in the total score on the IGMSS as well as all three major subscales, indicating that CGM participants compared with SMBG participants perceived greater capability to manage diabetes (eg, now having more knowledge regarding how glucose levels can be influenced), greater opportunity to take positive action (eg, now able to have more valuable conversations with health care providers), and greater motivation to focus on diabetes care (eg, now feeling more hopeful that living well with diabetes was possible). Furthermore, within-group analyses revealed that only the CGM group evidence significant gains in IGMSS total and subscale scores over time.

In concert with the observed changes in diabetes attitudes, parallel improvements in glycemic outcomes were also apparent. Only the CGM group demonstrated a significant drop in HbA1c over the three-month period, falling from 9.2% to 8.3%. In contrast, the SMBG group evidenced a smaller and nonsignificant glycemic benefit, dropping from 8.9% to 8.4%. Although the unadjusted group means point to a clinically significant difference in glycemic change between the two groups (~0.5%), statistical significance was not achieved. This was likely due to the limited sample size, leading to the study being underpowered to observe significant between-group glycemic differences.

In total, the results from this study provide some of the first quantitative evidence highlighting how CGM influences how people with T2D think and feel about diabetes and diabetes care, leading to a greater sense of personal engagement with the disease and potentially contributing via behavior change to the observed glycemic gains.¹⁴ Importantly, aside from providing participants with monitoring equipment and supplies, no other instruction, guidance or meaningful support were provided by study staff. This suggests that such psychological and glycemic benefits may accrue in response to the user’s own personal experience alone with CGM. We would surmise that had intensive guidance and coaching accompanied the provision of CGM, even greater gains would have been observed. However, prior qualitative evidence suggests that personal CGM experience alone can lead to greater engagement and improve glycemic outcomes.¹⁴

The major strengths of this study are that subjects were randomly assigned to CGM or SMBG and that all participants in both groups received a new monitoring device, thus addressing two importance sources of potential bias. In addition, there were broad inclusion criteria, and the study emphasized real-world use of glucose monitoring, with no constraints on use and or any additional guidance or support provided. Still, significant limitations are apparent. First, users’ sense of engagement with diabetes was assessed solely via self-report (IGMSS), so it remains uncertain how this reflects or influences actual self-care behavior (eg, changes in physical activity, medication adherence, and frequency of glucose monitoring). While we were unable to collect objective data regarding behavior change (eg, accelerometry to track physical activity, remote glucose device monitoring to document CGM wear) in the current trial, we hope to do so in future studies. Second, as is typical in device trials, participants could not be blinded to their assigned monitoring method, which may have influenced self-reported engagement outcomes. Third, we do not know to what degree participants may have received support from their own HCPs during the study period, which may have influenced outcomes. Fourth, the study period was only three months; whether greater psychological or glycemic benefits might have been gained if monitoring supplies had been extended over a longer period remains unknown. Fifth, the limited study size precluded the examination of the potential impact of CGM on subgroups (eg, insulin vs noninsulin users). Finally, the majority of the study sample was non-Hispanic White, therefore not representative of the broader T2D population. While we are unaware of any evidence that the impact of CGM use will differ in historically underserved T2D populations, it is notable that T2D prevalence is markedly higher in African American and Hispanic communities compared with non-Hispanic White communities,²⁰ and that there are significant ethnic disparities in the prescription and uptake of CGM use among people with T2D.²¹

Conclusions

Evidence from this randomized controlled trial suggests that introducing CGM to adults with T2D, including insulin users as well as noninsulin users, heightens individuals’ engagement with their own diabetes care and subsequently improves glycemic control more than providing a standard fingerstick monitor and supplies (SMBG). By recognizing and supporting the enhanced sense of engagement with one’s diabetes care that CGM can unleash, HCPs may provide their patients with the needed impetus and resources for even larger and more long-lasting benefits. Future research is needed to document that these positive results extend beyond the brief, three-month period of this study and to clarify how enhanced engagement may contribute to positive change in T2D self-care behaviors.

Footnotes

Acknowledgements

We gratefully acknowledge the critical contributions of our Scripps Health Research & Development team to the execution of this study.

Abbreviations

CGM, continuous glucose monitoring; HbA1c, glycosylated hemoglobin; HCP, health care professional; IGMSS, impact of glucose monitoring on self-management scale; SMBG, self-monitoring of blood glucose; T2D, type 2 diabetes; TCOYD, taking control of your diabetes.

Authors’ Note

Portions of these findings will be presented at the Advanced Technologies and Treatments for Diabetes Annual Conference (March 2025, Amsterdam).

Author Contributions

Study conception/protocol: FLG and WP. Statistical analysis: ES. Interpretation of Data: WP, ES, FLG, and AK. Manuscript development: WP and ES. All authors read and approved the final manuscript.

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: WHP has served as a consultant for Abbott Diabetes Care, Dexcom, and Senseonics. MV has served as a consultant for Abbott Diabetes Care and Medtronic. FLG and AK are employees at Abbott Diabetes Care.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by Abbott Diabetes Care.

ORCID iDs

William H. Polonsky

Emily C. Soriano

Data availability

Data may be made available upon reasonable request.

References

Ferreira

ROM

Trevisan

Pasqualotto

, et al. Continuous glucose monitoring systems in noninsulin-treated people with type 2 diabetes: a systematic review and meta-analysis of randomized controlled trials. Diabetes Technol Ther. 2024;26:252-262.

Lind

Christensen

Hansen

Nørgaard

Comparing continuous glucose monitoring and blood glucose monitoring in adults with inadequately controlled, insulin-treated type 2 diabetes (Steno2tech study): a 12-month, single-center, randomized controlled trial. Diabetes Care. 2024;47:881-889.

Lau

Manca

Singh

, et al. The effectiveness of continuous glucose monitoring with remote telemonitoring-enabled virtual educator visits in adults with non-insulin dependent type 2 diabetes: a randomized trial. Diabetes Res Clin Pract. 2024;217:111899. doi:10.1016/j.diabres.2024.111899.

Layne

Jepson

Carite

Parkin

Bergenstal

RM.

Long-term improvements in glycemic control with Dexcom CGM use in adults with noninsulin-treated type 2 diabetes. Diabetes Technol Ther. 2024;26(12):925-931.

Martens

Beck

Bailey

, et al. Effect of continuous glucose monitoring on glycemic control in patients with type 2 diabetes treated with basal insulin: a randomized clinical trial. JAMA. 2021;325:2262-2272.

Uhl

Choure

Rouse

Loblack

Reaven

Effectiveness of continuous glucose monitoring on metrics of glycemic control in type 2 diabetes mellitus: a systematic review and meta-analysis of randomized controlled trials. J Clin Endocrinol Metab. 2024;109:1119-1131.

Price

Deng

Kipnes

Beck

SE.

Episodic real-time CGM use in adults with type 2 diabetes: results of a pilot randomized controlled trial. Diabetes Ther. 2021;12(7):2089-2099.

Martens

Willis

Bergenstal

Kruger

Karslioglu-French

Steenkamp

DW.

A randomized controlled trial using continuous glucose monitoring to guide food choices and diabetes self-care in people with type 2 diabetes not taking insulin. Diabetes Technol Ther. 2025;27:261-270. doi:10.1089/dia.2024.0579.

Peyrot

Rubin

Lauritzen

Snoek

Matthews

Skovlund

SE.

Psychosocial problems and barriers to improved diabetes management: results of the cross-national diabetes attitudes, wishes and needs (DAWN) study. Diabet Med. 2005;22(10):1379-1385.

10.

Yaron

Roitman

Aharon-Hananel

, et al. Effect of flash glucose monitoring technology on glycemic control and treatment satisfaction in patients with type 2 diabetes. Diabetes Care. 2019;42(7):1178-1184.

11.

Aronson

Brown

Chu

, et al. Impact of flash glucose Monitoring in people with type 2 diabetes inadequately controlled with non-insulin antihyperglycaemic therapy (IMMEDIATE): a randomized controlled trial. Diabetes Obes Metab. 2023;25(4):1024-1031.

12.

Litchman

Allen

Sanchez-Birkhead

, et al. Continuous glucose monitoring plus an online peer support community reinforces healthy behaviors in Hispanic adults with type 2 diabetes. Diabetes Spectr. 2022;35(4):452-460.

13.

Kumbara

Iyer

Green

, et al. Impact of a combined continuous glucose monitoring-digital health solution on glucose metrics and self-management behavior for adults with type 2 diabetes: real-world, observational study. JMIR Diabetes. 2023:8:e47638. doi:10.2196/47638.

14.

Clark

Polonsky

Soriano

EC.

The potential impact of continuous glucose monitoring use on diabetes-related attitudes and behaviors in adults with type 2 diabetes: a qualitative investigation of the patient experience. Diabetes Technol Ther. 2024;26(10):700-708.

15.

Vallis

Berard

Cosson

, et al. Promoting self-management in adults with type 2 diabetes: development of the impact of glucose monitoring on self-management scale. Can J Diabetes. 2024;48:452-461.

16.

Lind

Odén

Fahlén

Eliasson

A systematic review of HbA1c variables used in the study of diabetic complications. Diabetes Metab Syndr. 2008;2:282-293.

17.

Muthén

BO.

Mplus User’s Guide. 8th ed. Los Angeles, CA: Muthén & Muthén; 2017.

18.

Kievit

Brandmaier

Ziegler

, et al. Developmental cognitive neuroscience using latent change score models: a tutorial and applications. Dev Cogn Neurosci. 2018;33:99-117.

19.

Messer

Johnson

Driscoll

Jones

Best friend or spy: a qualitative meta-synthesis on the impact of continuous glucose monitoring on life with type 1 diabetes. Diabet Med. 2018;35:409-418.

20.

Spanakis

Golden

SH.

Race/ethnic difference in diabetes and diabetic complications. Curr Diab Rep. 2013;13(6):814-823.

21.

Wallia

Agarwal

Owen

, et al. Disparities in continuous glucose monitoring among patients receiving care in federally qualified health centers. JAMA Netw Open. 2024;7:e2445316. doi:10.1001/jamanetworkopen.2024.45316.