Key Points
Question
Is continuous glucose monitoring effective in improving glycemic control compared with standard blood glucose monitoring in adolescents and young adults with type 1 diabetes?
Findings
In this randomized clinical trial that included 153 participants aged 14 to 24 years with type 1 diabetes, treatment with continuous glucose monitoring compared with standard blood glucose monitoring resulted in a significantly lower hemoglobin A1c level after 26 weeks (adjusted difference, 0.37%).
Meaning
Among adolescents and young adults with type 1 diabetes, continuous glucose monitoring resulted in a small but statistically significant improvement in glycemic control over 26 weeks.
Abstract
Importance
Adolescents and young adults with type 1 diabetes exhibit the worst glycemic control among individuals with type 1 diabetes across the lifespan. Although continuous glucose monitoring (CGM) has been shown to improve glycemic control in adults, its benefit in adolescents and young adults has not been demonstrated.
Objective
To determine the effect of CGM on glycemic control in adolescents and young adults with type 1 diabetes.
Design, Setting, and Participants
Randomized clinical trial conducted between January 2018 and May 2019 at 14 endocrinology practices in the US including 153 individuals aged 14 to 24 years with type 1 diabetes and screening hemoglobin A1c (HbA1c) of 7.5% to 10.9%.
Interventions
Participants were randomized 1:1 to undergo CGM (CGM group; n = 74) or usual care using a blood glucose meter for glucose monitoring (blood glucose monitoring [BGM] group; n = 79).
Main Outcomes and Measures
The primary outcome was change in HbA1c from baseline to 26 weeks. There were 20 secondary outcomes, including additional HbA1c outcomes, CGM glucose metrics, and patient-reported outcomes with adjustment for multiple comparisons to control for the false discovery rate.
Results
Among the 153 participants (mean [SD] age, 17 [3] years; 76 [50%] were female; mean [SD] diabetes duration, 9 [5] years), 142 (93%) completed the study. In the CGM group, 68% of participants used CGM at least 5 days per week in month 6. Mean HbA1c was 8.9% at baseline and 8.5% at 26 weeks in the CGM group and 8.9% at both baseline and 26 weeks in the BGM group (adjusted between-group difference, −0.37% [95% CI, −0.66% to −0.08%]; P = .01). Of 20 prespecified secondary outcomes, there were statistically significant differences in 3 of 7 binary HbA1c outcomes, 8 of 9 CGM metrics, and 1 of 4 patient-reported outcomes. The most commonly reported adverse events in the CGM and BGM groups were severe hypoglycemia (3 participants with an event in the CGM group and 2 in the BGM group), hyperglycemia/ketosis (1 participant with an event in CGM group and 4 in the BGM group), and diabetic ketoacidosis (3 participants with an event in the CGM group and 1 in the BGM group).
Conclusions and Relevance
Among adolescents and young adults with type 1 diabetes, continuous glucose monitoring compared with standard blood glucose monitoring resulted in a small but statistically significant improvement in glycemic control over 26 weeks. Further research is needed to understand the clinical importance of the findings.
Trial Registration
ClinicalTrials.gov Identifier: NCT03263494
This randomized clinical trial examines the effect of continuous glucose monitoring vs standard blood glucose monitoring on glycemic outcomes among adolescents and adults with type 1 diabetes and suboptimal glycemic control.
Introduction
Glycemic control remains suboptimal in the majority of adolescents and young adults with type 1 diabetes, with only 17% attaining the 2019 American Diabetes Association’s hemoglobin A1c (HbA1c) target of less than 7.5% and 14% attaining the target of less than 7% in the T1D Exchange clinic registry.1,2
Continuous glucose monitoring (CGM) devices provide glucose readings, trends, and alerts to the user in real time to inform diabetes treatment decisions. Although CGM has been shown to improve glycemic control in adults,3,4 studies have not shown overall benefit in adolescents and young adults (although for the minority of adolescents and young adults who used CGM regularly, a benefit was observed).3 These studies used older-generation CGM devices.
Substantial improvements in CGM technology have led to greater accuracy and convenience, including approval by the US Food and Drug Administration in 2016 to use CGM for diabetes management without confirmatory blood glucose monitoring (BGM).5 Considering the improvements in CGM technology, a randomized trial was conducted to evaluate the ability of CGM to improve glycemic outcomes in adolescents and young adults with type 1 diabetes with suboptimal glycemic control.
Methods
Study Conduct and Oversight
This randomized clinical trial was conducted at 14 endocrinology practices in the US. The protocol and Health Insurance Portability and Accountability Act–compliant informed consent/assent forms were approved by institutional review boards. Written informed consent with or without assent was obtained from each participant and parent/legal guardian, as applicable, prior to enrollment. An independent data and safety monitoring board provided trial oversight for review of safety data. The protocol and the statistical analysis plan are available in Supplement 1.
Participants
Major eligibility criteria included clinical diagnosis of type 1 diabetes, age of 14 to 24 years, diabetes duration of at least 1 year, use of either an insulin pump or multiple daily insulin injections, total daily insulin of at least 0.4 units/kg/d, no use of real-time CGM in the 3 months prior to enrollment, and HbA1c of 7.5% to less than 11.0% (see eTable 1 in Supplement 2 for a complete listing of the inclusion and exclusion criteria). The study aimed to enroll at least 33% of participants in the following categories: multiple daily insulin injection users, insulin pump therapy users, individuals with HbA1c of at least 9.0%, and young adults aged 19 to 24 years.
Each participant was required to complete a 14- to 21-day prerandomization period using a masked CGM device in which sensor glucose values were not visible to participants. To be eligible for randomization, the participant needed at least 200 hours (equivalent to 8.3 days) of masked CGM glucose data during the prerandomization period. In addition, the individual had to perform a mean of at least 2 daily fingerstick blood glucose meter checks confirmed from download of home meter and calibration of the masked CGM device a mean of 1.8 times daily (per the manufacturer’s recommendation of 2 daily calibrations).
Randomization
Eligible participants were randomly assigned on the study website, via a computer-generated sequence, to use CGM (Dexcom G5, Dexcom, Inc.) with fingerstick blood glucose meter checks as needed or to continue BGM with a blood glucose meter without CGM in a 1:1 ratio, using a permuted block design (block sizes of 2 and 4) stratified by site. Study investigators and personnel were masked to randomization sequence created by a coordinating center statistician, but not to treatment assignment. Participants were not masked to treatment assignment. The central laboratory was masked to treatment assignment. CGM outcomes were analyzed by a statistician at the coordinating center who was not masked to treatment assignment.
Intervention and Procedures
The CGM system included a transmitter, receiver, and disposable sensor that was inserted under the skin for 7 days (and then replaced), with glucose concentrations measured from interstitial fluid every 5 minutes. The CGM system required 2 daily calibrations from BGM.
Participants with compatible mobile phones were given the option to use either a study-provided CGM receiver or the CGM smartphone application on their mobile phone. Training on real-time CGM was provided using standardized materials developed for the study (eAppendix in Supplement 2). Additionally, participants in the CGM group received a handout at each study visit highlighting the benefits and features of CGM, such as the reduced need for fingerstick blood glucose meter measurements and the utility of the smartphone application (eAppendix in Supplement 2).
Participants in both groups received general diabetes management education and were provided a study blood glucose meter and test strips (Bayer Contour Next USB, Ascensia Diabetes Care) if they did not have their own downloadable meter (8% of participants received a study meter). Clinicians were encouraged to review downloaded glucose data (CGM and BGM data) at each visit to inform treatment recommendations, which were at the clinician’s discretion. The BGM group was asked to perform fingerstick blood glucose meter checks at least 4 times daily.
Both study groups had scheduled in-clinic visits at 4, 6, 13, and 26 weeks and contacts (via phone or video conference) at 1, 2 (in-clinic or remotely for the CGM group), and 19 weeks following randomization. The BGM group wore a masked CGM device for 1 week following the 13-week visit and for 2 weeks prior to the 26-week visit (clinic visit at 24 weeks for CGM device placement).
Central laboratory HbA1c was measured at randomization and 13 and 26 weeks at the University of Minnesota using the Tosoh A1c 2.2 Plus Glycohemoglobin Analyzer method. Participants completed patient-reported outcome assessments prior to randomization and at 13 and 26 weeks.
Data Collection and Outcomes
Participant sociodemographic data, including fixed categories for race/ethnicity, were collected from medical records and confirmed by the participants to describe the study cohort and provide information to inform generalizability.
Primary Outcome
The primary outcome was change in central laboratory–measured HbA1c from baseline to 26 weeks, adjusted for baseline value.
Prespecified Secondary Outcomes
Prespecified secondary HbA1c outcomes included the percentages of participants with HbA1c less than 7.0%, HbA1c less than 7.5%, HbA1c target for their age group met (<7.5% for age <19 years and <7.0% for age ≥19 years), relative reduction in HbA1c of at least 10%, absolute reduction in HbA1c of at least 0.5%, absolute reduction in HbA1c of at least 1%, and absolute reduction in HbA1c at least 0.5% or HbA1c less than 7.0%.
CGM-measured outcomes were calculated at follow-up using data pooled from up to 7 days before or after the 13-week visit and 14 days prior to the 26-week visit. Prespecified secondary CGM outcomes included percentage of time in which glucose level was in the target range (70-180 mg/dL), greater than 180 mg/dL, greater than 250 mg/dL, greater than 300 mg/dL, less than 70 mg/dL, and less than 54 mg/dL; mean glucose; coefficient of variation; and rate of CGM-measured hypoglycemic episodes.
Prespecified secondary patient-reported outcomes described herein were measured using the following instruments: Problem Areas in Diabetes-Pediatric survey,6 Glucose Monitoring Satisfaction Survey,7 Hypoglycemia Confidence Scale,8 and Pittsburgh Sleep Quality Index9 (see eTable 12a in Supplement 2 for descriptions). Additional questionnaire outcomes on CGM efficacy and technology attitudes will be reported separately given their application mainly to the CGM group.
Exploratory Outcomes
Prespecified exploratory outcomes included the mean number of blood glucose meter checks per day, total daily insulin dose per kilogram, number of short-acting injections for injection users, and number of bolus doses for pump users.
Safety Outcomes
Reportable adverse events included severe hypoglycemia (defined as an event that required assistance from another person due to altered consciousness), hyperglycemia resulting in evaluation or treatment at a health care provider facility or that involved diabetic ketoacidosis (as defined by the Diabetes Control and Complications Trial10), device-related events with potential effects on participant safety, and all serious adverse events regardless of causality.
Statistical Methods
A sample size of 140 participants was determined to have 90% power to detect a between-group difference in mean HbA1c, assuming a population difference of 0.5%, SD of 0.9%, and a 2-sided type I error rate of 5%. This number was increased to 150 participants to account for missing follow-up data.
All participants were analyzed according to their randomization group and included in the primary analysis. For the primary analysis, the difference in change in HbA1c from baseline to 26 weeks between the 2 treatment groups was assessed in a longitudinal linear regression model including the HbA1c value at baseline, 13 weeks, and 26 weeks and clinical center as a random effect. Missing data were handled by direct likelihood, which maximizes the likelihood function integrated over possible values of the missing data.11 Analyses of prespecified secondary and exploratory continuous outcomes paralleled those for the primary outcome (CGM data were pooled across follow-up time points). Binary secondary HbA1c outcomes were compared between treatment groups using available cases only in a logistic regression model adjusting for baseline HbA1c and clinical center as a random effect.
Additional analyses for select CGM glucose outcomes were performed separately for daytime (6:00 am to 11:59 pm) and nighttime (12:00 am to 5:59 am) hours. Additional analyses were performed on HbA1c and select CGM outcomes with data obtained through 13 weeks using the same methods as those used during the entire follow-up period of 26 weeks.
Modification of the treatment effect by baseline variables was assessed by including an interaction term in the primary model. Sensitivity analyses (with adjustment for potential confounding and including only participants who met per-protocol criteria) were performed as described in the statistical analysis plan (Supplement 1).
For all secondary and exploratory analyses (65 comparisons total), 2-sided P values and 95% CIs were adjusted for multiple comparisons to control the false discovery rate using the adaptive Benjamini-Hochberg procedure12 (eTable 2 in Supplement 1). The choice of summary statistics for all outcomes was based on the distribution. Mean and SD were used if the outcome was approximately normal and median and interquartile range (IQR) were used if the outcome was skewed. Analyses were conducted with SAS software, version 9.4 (SAS Institute Inc).
Results
Between February 2018 and November 2018, 153 participants were randomly assigned to the CGM group (n = 74) or BGM group (n = 79). Thirty-one participants were consented for the study but not randomized (Figure 1). Participant characteristics are shown in Table 1. The 26-week visit was completed by 71 participants (96%) in the CGM group and 71 participants (90%) in the BGM group (Figure 1 and eFigure 1 in Supplement 2). Unscheduled visits and contacts are reported in eTable 3 in Supplement 2.
Figure 1. Flow of Participants in a Study of the Effect of Continuous Glucose Monitoring (CGM) vs Standard Blood Glucose Monitoring (BGM) on Glycemic Control in Adolescents and Young Adults With Type 1 Diabetes.
Information on patients screened but not enrolled was not collected for this study. One participant in the BGM group and 1 participant in the CGM group were determined to be ineligible following randomization (inadequate hours of data during CGM run-in for the CGM participant and real-time CGM used within 3 months prior for the BGM participant). One participant in the BGM group initiated real-time CGM before completing the 26-week visit and was analyzed as randomized. Missing data for the primary outcome were handled by direct likelihood, which maximizes the likelihood function integrated over possible values of the missing data. All participants had data for at least 1 point and were included in the model.
Table 1. Baseline Characteristics of Participants in a Study of the Effect of Continuous Glucose Monitoring (CGM) on Glycemic Control in Adolescents and Young Adults With Type 1 Diabetes.
| Characteristic | No. (%) | |
|---|---|---|
| Continuous glucose monitoring (n = 74) | Blood glucose monitoring (n = 79) | |
| Age, y | ||
| 14-<19 | 48 (65) | 53 (67) |
| 19-<25 | 26 (35) | 26 (33) |
| Mean (SD) [range] | 17 (3) [14-24] | 18 (3) [14-24] |
| Diabetes duration, mean (SD) [range], y | 9 (5) [1-21] | 10 (5) [1-21] |
| Sex | ||
| Female | 33 (45) | 43 (54) |
| Male | 41 (55) | 36 (46) |
| Race/ethnicity | (n = 73) | (n = 79) |
| White, non-Hispanic | 48 (66) | 47 (59) |
| Black, non-Hispanic | 3 (4) | 9 (11) |
| Hispanic or Latino | 18 (25) | 15 (19) |
| Asian | 1 (1) | 5 (6) |
| American Indian/Alaskan Native | 0 | 1 (1) |
| More than 1 race | 3 (4) | 2 (3) |
| Health insurance | (n = 73) | (n = 79) |
| Private | 43 (59) | 47 (59) |
| Public | 30 (41) | 32 (41) |
| CGM use | ||
| Past but not current | 24 (32) | 30 (38) |
| Never | 50 (68) | 49 (62) |
| Insulin pump use | 36 (49) | 47 (59) |
| HbA1c at screeninga | ||
| <9% | 34 (46) | 33 (42) |
| ≥9% | 40 (54) | 46 (58) |
| Mean (SD) [range], % | 9.1 (1.0) [7.5-10.9] | 9.1 (1.0) [7.5-10.9] |
| HbA1c at randomizationb | ||
| <9% | 42 (57) | 43 (54) |
| ≥9% | 32 (43) | 36 (46) |
| Mean (SD) [range], % | 8.9 (1.0) [6.8-10.8] | 8.9 (1.0) [6.4-10.9] |
| C-peptide | (n = 74) | (n = 78) |
| Detectable C-peptidec | 41 (55) | 30 (38) |
| C-peptide >0.2 nmol/Lc | 13 (18) | 6 (8) |
| Total daily insulin dose/kg, mean (SD) | 0.92 (0.36) | 0.91 (0.25) |
| ≥1 Severe hypoglycemia event in the past 12 mod | 7 (9) | 2 (3) |
| ≥1 Diabetic ketoacidosis event in the past 12 moe | 5 (7) | 5 (6) |
Screening hemoglobin A1c (HbA1c) was measured by point-of-care device or local laboratory and used to determine eligibility.
Randomization HbA1c was measured by central laboratory.
Random C-peptide. The detection limit of the assay was 0.003 nmol/L. Presence of C-peptide suggests some insulin production by the β cells in the pancreas.
Severe hypoglycemia was defined as an event that required assistance from another person to administer carbohydrate, glucagon, or other resuscitative actions.
Diabetic ketoacidosis was defined as an episode when the participant had ketosis that necessitated treatment in a health care facility.
Device use in the CGM group was initially high, with 82% of participants using CGM for a mean of at least 5 days per week in the 28 days prior to the 6-week visit (eTable 4 in Supplement 2). This dropped by week 26, with 68% of participants using CGM for a mean of at least 5 days per week. At 26 weeks, 10 participants (14%) in the CGM group had no CGM use, including 3 participants who dropped out. Nine CGM device issues were reported over the 26-week study period, none of which were related to an adverse event (eTable 5 in Supplement 2). One participant in the BGM group initiated CGM use prior to the 26-week visit.
In the CGM group, the median (25th, 75th percentile) of each individual’s mean number of BGM checks per day was 3.9 (3.0, 5.0) at baseline and 2.3 (1.9, 3.0) at follow-up compared with 3.5 (3.0, 4.5) at baseline and 3.0 (2.5, 4.3) at follow-up in the BGM group (adjusted bewteen-group difference, −0.8 [95% CI, −1.4 to −0.4]; P < .001). Among participants in the CGM group who were actively using CGM, the percentage who reported using CGM to dose insulin without blood glucose meter confirmation was 92% at the 2-week visit and increased to 98% at the 26-week visit. Insulin data are reported in eTable 6 in Supplement 2.
Primary Outcome: Hemoglobin A1c
Mean HbA1c was 8.9% at baseline and 8.5% at 26 weeks in the CGM group and was 8.9% at both baseline and 26 weeks in the BGM group (adjusted between-group difference, −0.37% [95% CI, −0.66% to −0.08%]; P = .01) (Table 2, Figure 2, and eFigure 2 in Supplement 2). Significant improvement in glucose control was observed by the 13-week visit, with a mean HbA1c of 8.4% in the CGM group and 8.9% in the BGM group (adjusted between-group difference, −0.50% [95% CI, −0.79% to −0.21%]; P < .001) (Figure 2 and eTable 7 in Supplement 2).
Table 2. Glycemic Outcomes in a Study of the Effect of Continuous Glucose Monitoring on Glycemic Control in Adolescents and Young Adults With Type 1 Diabetes.
| Outcome | Randomization | 26 weeksa | Adjusted between-group difference (95% CI)a | P valueb | ||
|---|---|---|---|---|---|---|
| Continuous glucose monitoring (n = 74) | Blood glucose monitoring (n = 79) | Continuous glucose monitoring (n = 71) | Blood glucose monitoring (n = 71) | |||
| Primary outcome | ||||||
| HbA1c, mean (SD), % | 8.9 (1.0) | 8.9 (1.0) | 8.5 (1.2) | 8.9 (1.2) | −0.37 (−0.66 to −0.08) | .01 |
| Change in HbA1c from baseline, mean (SD), % | −0.4 (1.0) | 0.1 (0.8) | ||||
| Secondary outcomes | ||||||
| HbA1c <7.0%, No. (%) | 1 (1) | 2 (3) | 6 (8) | 4 (6) | 4 (−4 to 11) | .30 |
| HbA1c <7.5%, No. (%) | 3 (4) | 7 (9) | 13 (18) | 8 (11) | 9 (−1 to 18) | .11 |
| Met HbA1c target (<7.5% for age <19 y and <7.0% for age ≥19 y), No. (%) | 2 (3) | 6 (8) | 9 (13) | 7 (10) | 4 (−6 to 12) | .42 |
| Relative reduction in HbA1c ≥10%, No. (%) | 20 (28) | 6 (8) | 19 (8 to 32) | .005 | ||
| Absolute reduction in HbA1c, No. (%) | ||||||
| ≥0.5% | 31 (44) | 15 (21) | 23 (7 to 37) | .005 | ||
| ≥1.0% | 18 (25) | 4 (6) | 19 (8 to 31) | .003 | ||
| ≥0.5% or HbA1c ≤7.0% | 19 (27) | 8 (11) | 15 (3 to 28) | .02 | ||
| Baseline | Follow-up (13 and 26 weeks pooled)c | |||||
| Secondary continuous glucose monitoring metrics | Continuous glucose monitoring (n = 73)c | Blood glucose monitoring (n = 79) | Continuous glucose monitoring (n = 68) | Blood glucose monitoring (n = 72) | ||
| Hours of continuous glucose monitoring data, median (IQR) | 302 (269 to 324) | 311 (268 to 378) | 376 (262 to 475) | 426 (371 to 477) | ||
| Time in target glucose range (70-180 mg/dL), mean (SD), % | 37 (13) | 36 (12) | 43 (15) | 35 (12) | 6.9 (3.1 to 10.7) | <.001 |
| Glucose, mean (SD), mg/dL | 209 (36) | 212 (36) | 199 (36) | 217 (35) | −14.3 (−23.6 to −5.1) | .003 |
| Coefficient of variation, mean (SD), %d | 42 (7) | 42 (7) | 39 (6) | 42 (7) | −2.2 (−3.9 to −0.5) | .01 |
| Hyperglycemia | ||||||
| Time with glucose >180 mg/dL (hyperglycemia), mean (SD), % | 58 (15) | 59 (15) | 54 (18) | 61 (14) | −5.8 (−10.0 to −1.7) | .007 |
| Time with glucose >250 mg/dL, mean (SD), % | 32 (15) | 34 (15) | 26 (15) | 35 (14) | −7.9 (−12.3 to −3.4) | <.001 |
| Time with glucose >300 mg/dL, median (IQR), % | 15 (9 to 26) | 17 (11 to 28) | 11 (5 to 19) | 20 (12 to 26) | −5.1 (−8.2 to −2.3) | <.001 |
| Hypoglycemia | ||||||
| Time with glucose <70 mg/dL, median (IQR), % | 3.2 (1.3 to 7.7) | 3.7 (1.7 to 6.7) | 2.2 (1.0 to 5.0) | 3.2 (1.9 to 6.2) | −0.7 (−1.5 to −0.1) | .02 |
| Time with glucose <54 mg/dL, median (IQR), % | 1.0 (0.4 to 3.2) | 1.3 (0.3 to 3.0) | 0.7 (0.2 to 1.4) | 1.3 (0.5 to 2.5) | −0.4 (−0.7 to −0.1) | .002 |
| Rate of hypoglycemic events per weeke | 1.5 (0.6 to 3.3) | 1.7 (0.6 to 3.2) | 1.4 (0.4 to 2.6) | 1.7 (1.0 to 3.1) | −0.3 (−0.7 to 0.1) | .11 |
Abbreviation: IQR, interquartile range.
Three participants in the CGM group and 8 participants in the BGM group were missing central laboratory hemoglobin A1c (HbA1c) values at 26 weeks. Missing data for continuous outcomes were handled using direct likelihood. Binary outcomes were analyzed using available cases only.
Continuous outcomes were analyzed in a longitudinal regression model including baseline and follow-up with adjustment for clinical center as a random effect. The reported number of participants is for those with available baseline and follow-up data, but all participants were included in the statistical model. The hypoglycemia metrics and percentage of time with glucose greater than 300 mg/dL had skewed distributions and were modeled using a rank-based transformation. For these skewed outcomes, point estimates and CIs for the treatment group difference were calculated using the technique described by Hodges and Lehmann.13 Binary HbA1c outcomes were analyzed in a logistic regression model adjusted for baseline HbA1c and clinical center as a random effect. For these outcomes, risk-adjusted differences were calculated according to the method of Kleinman and Norton.14 P values and 95% CIs for all secondary outcomes were adjusted for multiple comparisons to control the false discovery rate.
Baseline CGM data for 1 participant in the CGM group were lost after the site confirmed they met the eligibility criteria. Six participants in the CGM group and 7 in the BGM group were missing follow-up CGM data. All participants had data for at least 1 time point. Missing data were handled using direct likelihood.
Coefficient of variation is defined as SD divided by the mean.
A CGM-measured hypoglycemic event was defined as 15 consecutive minutes with a sensor glucose value less than 54 mg/dL. The end of the hypoglycemic event was defined as a minimum of 15 consecutive minutes with a sensor glucose concentration greater than 70 mg/dL.15
Figure 2. Hemoglobin A1c Levels During a Study of the Effect of Continuous Glucose Monitoring on Glycemic Control in Adolescents and Young Adults With Type 1 Diabetes.
The top and bottom of the boxes denote the 25th and 75th percentile, the line represents the median, and the dot represents the mean. The whiskers represent the minimum and maximum values after removing outliers.
An HbA1c reduction from baseline to 26 weeks of at least 0.5% was observed in 44% of the CGM group vs 21% of the BGM group (adjusted between-group difference, 23% [95% CI, 7%-37%]; P = .005) and a reduction of at least 1.0% was shown in 25% of participants in the CGM group vs 6% in the BGM group (adjusted between-group difference, 19% [95% CI, 8%-31%]; P = .003) (Table 2).
The significant treatment effect for HbA1c at 26 weeks remained when adjusting for duration of diabetes, sex, insulin delivery method, and C-peptide. There was no significant interaction of the effect of study treatment on 26-week HbA1c according to baseline age, sex, insulin delivery method, and baseline HbA1c (eTable 8 in Supplement 2). In a per-protocol analysis, the 26-week adjusted between-group difference for the CGM vs BGM group was −0.69% ([95% CI, −1.01% to −0.36%]; P < .001) (eTable 9 in Supplement 2).
CGM Metrics
The mean percentage of time in target glucose range of 70 to 180 mg/dL was 37% (9.0 h/d) at baseline and 43% (10.3 h/d) during follow-up in the CGM group and 36% (8.7 h/d) at baseline and 35% (8.3 h/d) during follow-up in the BGM group (adjusted between-group difference, 6.9% [1.7 h/d] [95% CI, 3.1%-10.7%]; P < .001) (Table 2 and eTable 10 and eFigure 3 in Supplement 2). The percentages of time in target glucose range during daytime and nighttime hours are provide in eTable 11 in Supplement 2. Mean time in hypoglycemia (glucose <70 mg/dL) was significantly lower in the CGM group than the BGM group (adjusted between-group difference, −0.7% [95% CI, −1.5% to −0.1%]; P = .002) (Table 2). Results for other CGM outcomes are provided in Table 2 and eTable 10 and eFigure 3 in Supplement 2. Results of CGM outcomes at 13 weeks are reported in eTable 7 in Supplement 2.
Adverse Events
Severe hypoglycemic events occurred in 3 participants (4%) in the CGM group and 2 (3%) in the BGM group. Diabetic ketoacidosis occurred in 3 participants (4%) in the CGM group and 1 (1%) in the BGM group (Table 3). Additional adverse events are shown in Table 3.
Table 3. Safety Outcomes in a Study of the Effect of Continuous Glucose Monitoring on Glycemic Control in Adolescents and Young Adults With Type 1 Diabetes.
| Outcome | Participants with ≥1 event, No. | |
|---|---|---|
| Continuous glucose monitoring group (n = 74) | Blood glucose monitoring group (n = 79) | |
| Severe hypoglycemiaa | 3 | 2 |
| Incidence rate (events per 100 person-years) | 8.3 | 7.8 |
| Diabetic ketoacidosisb | 3 | 1 |
| Incidence rate (events per 100 person years) | 8.3 | 2.6 |
| Other serious adverse events | ||
| Overall | 2 | 2 |
| Appendicitis | 1 | 0 |
| Fainting | 1 | 0 |
| Hyperglycemia | 0 | 1 |
| Suicidal ideation | 0 | 1 |
| Syncope | 1 | 0 |
| Nonserious adverse events | ||
| Overall | 3 | 4 |
| Ketosis | 0 | 3 |
| Hyperglycemia | 1 | 0 |
| Lightheadedness | 1 | 0 |
| Panic attack | 0 | 1 |
| Vomiting | 1 | 0 |
Severe hypoglycemia was defined as an event that required assistance from another person to administer carbohydrate, glucagon, or other resuscitative actions.
Diabetic ketoacidosis was defined as an episode when the participant had ketosis that necessitated treatment in a health care facility.
Patient-Reported Outcomes
The CGM group reported significantly higher glucose monitoring satisfaction, measured via the Glucose Monitoring Satisfaction Survey score, at 26 weeks than the BGM group (adjusted between-group difference, 0.27 [95% CI, 0.06-0.54]; P = .003; eTable 12b in Supplement 2). No statistically significant between-group differences were observed for problem areas in diabetes, hypoglycemia confidence, or sleep quality (eTable 12b in Supplement 2).
Discussion
This randomized trial among adolescents and young adults with type 1 diabetes showed a small but statistically significant lowering of HbA1c over 26 weeks of CGM use compared with standard BGM. This finding offers potential for clinical importance with a meaningful shift in the HbA1c distribution toward improved glycemic control; however, further research of longer duration and with clinical outcomes is needed before reaching definitive conclusions about the clinical value of the study’s findings.
The largest and most referenced randomized trial that examined CGM use in this age group was the JDRF (Juvenile Diabetes Research Foundation) CGM randomized clinical trial conducted more than 10 years ago.3 In that trial, a benefit of CGM was not seen in adolescents and young adults. However, only 30% used CGM regularly (6-7 d/wk), which is substantially less than observed in the current trial. Enhancements in CGM technology over the past 10 years have reduced the burden of using CGM, which likely accounts for the greater usage found in the current trial. This is evidenced by the improvement observed in patient-reported outcomes related to technology satisfaction and no reported increase in burden, which is noted by no difference in reported diabetes problem areas. Although CGM use in the current trial was higher than that in the JDRF CGM trial,3 it was substantially lower than the usage rate found in adults with type 1 diabetes using a similar CGM system.4 This emphasizes the greater challenges faced in managing diabetes in adolescents and young adults compared with older adults.
The strengths of the study include enrollment of a geographically and ethnically diverse sample of adolescents and young adults with type 1 diabetes from 14 diabetes centers and high participant retention, particularly given the recognized life changes that affect older teens and young adults.16 More than one-third of the cohort were racial and ethnic minority participants and more than 40% had public insurance, providing a pathway to CGM use for this underserved population. These data support the need for expanded reimbursement for CGM, especially for teens and young adults whose private or publicly funded insurance varies widely.
Limitations
There are several limitations of the study. First, CGM used in the trial required twice-daily calibrations with blood glucose measurements, whereas this is no longer required with the current generation of the factory-calibrated CGM devices. Second, in view of the eligibility criteria, the results may not apply to individuals with type 1 diabetes and HbA1c outside the eligibility range of HbA1c of 7.5% to 10.9%. Third, the informed consent process and the run-in phase had the potential to exclude individuals who might be less adherent to CGM use than the cohort that was studied. Fourth, the study included a relatively short intervention period of 6 months. This study also included an extension phase in which the CGM group continued using CGM through 12 months and the BGM group initiated CGM. Results of the extension phase may provide insight into longer-term use of CGM.
Conclusions
Among adolescents and young adults with type 1 diabetes, CGM compared with standard BGM resulted in a small but statistically significant improvement in glycemia over 26 weeks. Further research is needed to understand the clinical importance of this finding.
Trial protocol and statistical analysis plan
eTables and eAppendix
Data sharing statement
References
- 1.American Diabetes Association 6. Glycemic targets: Standards of Medical Care in Diabetes-2019. Diabetes Care. 2019;42(suppl 1):S61-S70. doi: 10.2337/dc19-S006 [DOI] [PubMed] [Google Scholar]
- 2.Foster NC, Beck RW, Miller KM, et al. State of type 1 diabetes management and outcomes from the T1D exchange in 2016-2018. Diabetes Technol Ther. 2019;21(2):66-72. doi: 10.1089/dia.2018.0384 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Tamborlane WV, Beck RW, Bode BW, et al. ; Juvenile Diabetes Research Foundation Continuous Glucose Monitoring Study Group . Continuous glucose monitoring and intensive treatment of type 1 diabetes. N Engl J Med. 2008;359(14):1464-1476. doi: 10.1056/NEJMoa0805017 [DOI] [PubMed] [Google Scholar]
- 4.Ruedy KJ, Parkin CG, Riddlesworth TD, Graham C; DIAMOND Study Group . Continuous glucose monitoring in older adults with type 1 and type 2 diabetes using multiple daily injections of insulin: results from the DIAMOND Trial. J Diabetes Sci Technol. 2017;11(6):1138-1146. doi: 10.1177/1932296817704445 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.US Food and Drug Administration Premarket approval: Dexcom G5 Mobile Continuous Glucose Monitoring System. Updated May 11, 2020. Accessed May 15, 2020. https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpma/pma.cfm?id=P120005S041
- 6.Markowitz JT, Volkening LK, Butler DA, Laffel LM. Youth-perceived burden of type 1 diabetes: Problem Areas in Diabetes Survey-Pediatric version (PAID-Peds). J Diabetes Sci Technol. 2015;9(5):1080-1085. doi: 10.1177/1932296815583506 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Polonsky WH, Fisher L, Hessler D, Edelman SV. Development of a new measure for assessing glucose monitoring device-related treatment satisfaction and quality of life. Diabetes Technol Ther. 2015;17(9):657-663. doi: 10.1089/dia.2014.0417 [DOI] [PubMed] [Google Scholar]
- 8.Polonsky WH, Fisher L, Hessler D, Edelman SV. Investigating hypoglycemic confidence in type 1 and type 2 diabetes. Diabetes Technol Ther. 2017;19(2):131-136. doi: 10.1089/dia.2016.0366 [DOI] [PubMed] [Google Scholar]
- 9.Buysse DJ, Reynolds CF III, Monk TH, Berman SR, Kupfer DJ. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res. 1989;28(2):193-213. doi: 10.1016/0165-1781(89)90047-4 [DOI] [PubMed] [Google Scholar]
- 10.Nathan DM, Genuth S, Lachin J, et al. ; Diabetes Control and Complications Trial Research Group . The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med. 1993;329(14):977-986. doi: 10.1056/NEJM199309303291401 [DOI] [PubMed] [Google Scholar]
- 11.Beunckens C, Molenberghs G, Kenward MG. Direct likelihood analysis versus simple forms of imputation for missing data in randomized clinical trials. Clin Trials. 2005;2(5):379-386. doi: 10.1191/1740774505cn119oa [DOI] [PubMed] [Google Scholar]
- 12.Benjamini Y, Hochberg Y. On the adaptive control of the false discovery rate in multiple testing with independent statistics. J Educ Behav Stat. 2000;25(1):60-83. doi: 10.3102/10769986025001060 [DOI] [Google Scholar]
- 13.Hodges JL, Lehmann EL. Estimates of location based on rank tests. Ann Math Stat. 1963;34(2):598-611. doi: 10.1214/aoms/1177704172 [DOI] [Google Scholar]
- 14.Kleinman LC, Norton EC. What’s the risk? a simple approach for estimating adjusted risk measures from nonlinear models including logistic regression. Health Serv Res. 2009;44(1):288-302. doi: 10.1111/j.1475-6773.2008.00900.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Battelino T, Danne T, Bergenstal RM, et al. Clinical targets for continuous glucose monitoring data interpretation: recommendations from the international consensus on time in range. Diabetes Care. 2019;42(8):1593-1603. doi: 10.2337/dci19-0028 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Peters A, Laffel L; American Diabetes Association Transitions Working Group . Diabetes care for emerging adults: recommendations for transition from pediatric to adult diabetes care systems: a position statement of the American Diabetes Association, with representation by the American College of Osteopathic Family Physicians, the American Academy of Pediatrics, the American Association of Clinical Endocrinologists, the American Osteopathic Association, the Centers for Disease Control and Prevention, Children with Diabetes, the Endocrine Society, the International Society for Pediatric and Adolescent Diabetes, Juvenile Diabetes Research Foundation International, the National Diabetes Education Program, and the Pediatric Endocrine Society (formerly Lawson Wilkins Pediatric Endocrine Society). Diabetes Care. 2011;34(11):2477-2485. doi: 10.2337/dc11-1723 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Trial protocol and statistical analysis plan
eTables and eAppendix
Data sharing statement