CADTH Health Technology Review

Real-Time Continuous Glucose Monitoring for People Living With Type 2 Diabetes

Rapid Review

Authors: Anusree Subramonian, Aleksandra Grobelna

Key Messages

• The evidence regarding clinical and cost-effectiveness of real-time continuous glucose monitoring (rtCGM) compared to self-monitoring of blood glucose (SMBG) in patients with type 2 diabetes mellitus is summarized in this report.

• Results from 5 systematic reviews and 1 randomized controlled study suggest that in adult patients, rtCGM may be favoured over SMBG in improving glycated hemoglobin levels, and in lowering time with extreme low or high blood glucose levels. However, the evidence is uncertain due to limited quality evidence.

• In adults, limited safety evidence suggests that rtCGM is safe with low rates of adverse events.

• A cost-effectiveness analysis conducted in Spain found that rtCGM is not a cost-effective option compared to SMBG in adults with type 2 diabetes mellitus.

• No evidence was identified regarding clinical or cost-effectiveness of rtCGM in pediatric patients with type 2 diabetes mellitus.

Context and Policy Issues

Diabetes mellitus (DM) is a serious public health concern affecting 1 in 10 Canadians.¹ Globally, it has been estimated that approximately 9.3% of the population lives with diabetes.² Type 2 DM (T2D), also known as non-insulin dependent DM, accounts for about 90% to 95% of all patients with diabetes.³ T2D is a metabolic disorder characterized by chronic hyperglycemia and impaired metabolism of carbohydrates, lipids, and proteins resulting from insulin insufficiency and action.⁴ It is associated with microvascular (e.g., diabetic retinopathy, nephropathy) and macrovascular (e.g., cardiovascular disease, strokes) complications that significantly influence the morbidity and mortality of patients. Thus, glycemic control in T2D patients is of high importance. Monitoring of glucose levels (glycemic monitoring) is recommended to ensure stable blood glucose levels, manage symptoms, and determine appropriate medications and insulin.

As the name indicates, self-monitoring blood of glucose (SMBG) refers to patients monitoring their own blood glucose levels using glucose test strips, requiring frequent finger pricks. Continuous glucose monitoring (CGM) involves a needle-like sensor inserted into the abdomen or upper arm, a transmitter, and a monitor to show results. CGM systems continuously measure glucose values from the interstitial fluid. Real-time CGM (rtCGM), also known as personal CGM, measures glucose values as frequently as every 5 minutes and displays measurements in real time.⁵ In addition, device wearers are alerted in case the values go above or below the calibrated target range. Examples of rtCGM devices include Dexcom, Medtronic, and MiniMed systems, and are available in several models.^6,7

Although CGM, including rtCGM, is used mostly in patients with type 1 DM (T1D), its effectiveness in T2D has been researched.⁶ A recently published CADTH report did not identify any evidence regarding the comparative effectiveness of rtCGM and intermittently scanned CGM devices in people living with T2D.⁸ The objective of this report is to summarize the evidence regarding the clinical effectiveness and cost-effectiveness of glycemic monitoring with rtCGM compared to SMBG in adult and pediatric individuals with T2D.

Research Questions

1. What is the clinical effectiveness of monitoring glycemia with real-time continuous glucose monitoring for adults living with type 2 diabetes?

2. What is the clinical effectiveness of monitoring glycemia with real-time continuous glucose monitoring for pediatric patients living with type 2 diabetes?

3. What is the cost-effectiveness of monitoring glycemia with real-time continuous glucose monitoring for people living with type 2 diabetes?

Methods

Literature Search Methods

A limited literature search was conducted by an information specialist on key resources including MEDLINE, Embase, the Cochrane Database of Systematic Reviews, the International HTA Database, and the websites of Canadian and major international health technology agencies, as well as a focused internet search. The search strategy comprised both controlled vocabulary, such as the National Library of Medicine’s MeSH (Medical Subject Headings), and keywords. The main search concepts were real time continuous glucose monitoring (rtCGM) and type 2 diabetes (T2D). No filters were applied to limit the retrieval by study type. Comments, newspaper articles, editorials, letters, and conference abstracts were excluded.

Where possible, retrieval was limited to the human population. The search was completed on June 9, 2022, and limited to English-language documents published since January 1, 2017.

Selection Criteria and Methods

One reviewer screened citations and selected studies. In the first level of screening, titles and abstracts were reviewed and potentially relevant articles were retrieved and assessed for inclusion. The final selection of full-text articles was based on the inclusion criteria presented in Table 1.

Exclusion Criteria

Articles were excluded if they did not meet the selection criteria outlined in Table 1, they were duplicate publications, or they were published before 2017. Systematic reviews in which all relevant studies were captured in other more recent or more comprehensive systematic reviews were excluded. Primary studies retrieved by the search were excluded if they were captured in 1 or more of the included systematic reviews.

Critical Appraisal of Individual Studies

The included publications were critically appraised by 1 reviewer using the following tools as a guide: A MeaSurement Tool to Assess systematic Reviews 2 (AMSTAR 2)⁹ for systematic reviews, the Downs and Black checklist¹⁰ for randomized and non-randomized studies, and the Drummond checklist¹¹ for economic evaluations. Summary scores were not calculated for the included studies; rather, the strengths and limitations of each included publication were described narratively.

Table 1: Selection Criteria

ICER = incremental cost-effectiveness ratio; Q = question; QALY = quality-adjusted life-year.

Summary of Evidence

Quantity of Research Available

A total of 879 citations were identified in the literature search. Following screening of titles and abstracts, 803 citations were excluded and 76 potentially relevant reports from the electronic search were retrieved for full-text review. No potentially relevant publications were retrieved from the grey literature search for full-text review. Of these potentially relevant articles, 69 publications were excluded for various reasons, and 7 publications met the inclusion criteria and were included in this report. These comprised 5 systematic reviews (SRs),^12-16 1 randomized controlled trial (RCT),¹⁷ and 1 economic evaluation.¹⁸ Appendix 1 presents the PRISMA¹⁹ flow chart of the study selection.

References of potential interest that did not meet the inclusion criteria but provided real-world evidence about the use of rtCGM for people living with T2D are provided in Appendix 6. Additional references of potential interest are provided in Appendix 7.

Summary of Study Characteristics

Five SRs,^12-16 1 RCT,¹⁷ and 1 economic evaluation¹⁸ were included in this report. Three SRs^12,13,16 were published in 2022, and 2^14,15 were published in 2019. The RCT¹⁷ was published in 2022, and the economic evaluation was published in 2018.¹⁸

All included SRs had broader inclusion criteria than the present review in terms of interventions of interest. All SRs^12-16 included studies on rtCGM as well as intermittent CGM or flash glucose monitoring. One SR was part of a larger review evaluating multiple interventions for diabetes.¹⁴ The publication included in this report summarized 2 interventions in T2D, namely CGM and continuous subcutaneous insulin infusion. As for populations considered in the SRs, 3 of them had a broader scope than the present review.^12,13,16 They all included studies on patients with T1D, T2D, and gestational diabetes.^12,13,16

Additionally, 2 SRs had specific inclusion criteria regarding population.^13,16 Specifically, Chang et al.¹³ limited the study population to perinatal women with diabetes (aged 18 years or older), and Kieu et al.¹⁶ limited the population to patients with diabetes managed by a primary care provider (with or without co-management by endocrinologists). None of the included studies from these 2 SRs met the inclusion criteria for the current report. These SRs included studies published up to January 2021¹³ and June 2021,¹⁶ respectively.

Thus, 3 of the included SRs identified primary studies that are relevant to the current report.^12,14,15 Only the characteristics and results of the subset of relevant studies will be described in this report. Additional details regarding the characteristics of included publications are provided in Appendix 2.

Study Design

Among the 3 SRs that included primary studies relevant to the current report,^12,14,15 1 SR¹² included RCTs and observational studies, and the other 2^14,15 included RCTs only. Two SRs^14,15 reported quantitative syntheses of results by conducting a meta-analysis (MA), 1 using a fixed effect model¹⁵ and the other using both fixed and random effect models.¹⁴

The SR by Aggarwal et al. (2022)¹² searched multiple electronic databases for studies published between 2018 and July 2021. They included 26 studies, of which 3 (2 RCTs and 1 non-randomized study) were relevant to the current report. The SR with MA by Dicembrini et al. (2019)¹⁴ searched for RCTs published up to 2018. Among the 10 RCTs included, 5 were relevant to the current report. However, 1 of the relevant RCTs¹⁷ was unpublished at the time of the publication of the SR, and partial information about the trial was obtained from the clinical trials registry. Since the results from the RCT were not included in the MA or reported in the SR, it is included in the current report separately to capture relevant evidence.¹⁷ Lastly, the SR with MA by Janapala et al. (2019)¹⁵ searched for studies published over the previous 10 years (the search date was not reported). They included 5 RCTs, of which 4 were relevant to the current report. There was some overlap of relevant included studies between the SRs by Dicembrini et al.¹⁴ and Janapala et al.¹⁵ A detailed overlap table is provided in Appendix 5.

The study by Bergenstal et al. (2022)¹⁷ was an open-label, parallel group RCT (NCT01237301). After a 2-week run-in period to adjust medications, participants were randomized to 2 study groups using random block allocation. Randomized participants then entered a 14-day baseline period before the study, during which participants wore a blinded CGM device and used SMBG for calibration measurements. This was done to obtain baseline measurements for each outcome. Participants entered the open-label study after this 14-day baseline period.

The economic evaluation by García-Lorenzo et al. (2018)¹⁸ was a cost-effectiveness analysis conducted with a public payer perspective, using a lifetime horizon with annual cycles. A Markov model was used for the analysis. Health states used in the model were: no complications, retinopathy (leading to blindness), neuropathy (leading to amputation), cardiovascular disease (leading to amputation), and nephropathy (leading to end-stage renal disease). Costs were reported in 2017 Euro and a 3% discount rate was applied. The clinical data were sourced from an MA conducted by the study authors, cost data were obtained from the device manufacturers, and utility data and data about transition risks were obtained from the literature. Disutility values obtained from a US-based study were modified to apply to the Spanish population. The key assumptions used in the analysis were that the association between relative risk reduction for T2D complications and hemoglobin A1C were linear, patients diagnosed with T2D could develop only 1 comorbidity per cycle, and complications such as blindness or amputation could be developed after a specific complication. Costs of multiple conditions were also assumed to be the sum of cost of each condition.

Country of Origin

The authors of the 3 SRs that included relevant studies were from India,¹² Italy,¹⁴ and the US.¹⁵ They did not report the countries in which the primary studies were conducted.

The RCT by Bergenstal et al.¹⁷ was conducted in the US, and the economic evaluation was conducted in Spain.¹⁸

Patient Population

The relevant population in the SR by Aggarwal et al.¹² was patients of all ages with T2D. Across the 3 included relevant primary studies, there were 35,325 participants (35,080 from a single observational study), with mean age ranging from 42.4 years to 56 years in the rtCGM group and 59 years to 63.4 years in the control group. In the SR with MA by Dicembrini et al.,¹⁴ the eligible population was patients with T2D. Across the included relevant RCTs, there were 429 patients with a mean age ranging from 58 years to 63 years. Lastly, the SR by Janapala et al.¹⁵ included studies on adult patients (≥ 19 years of age) with T2D. There were 348 participants across the relevant primary studies. Mean age was not reported.

In the RCT, Bergenstal et al.¹⁷ enrolled adult patients (18 to 75 years of age) with T2D with hemoglobin A1C ≥ 7% treated with sulfonylurea, incretin, or insulin along with metformin. Patients on thiazolidinediones were excluded. There were 114 participants (rtCGM group, n = 59; SMBG group, n = 55). The mean age of the participants was 59.3 years in the rtCGM group and 58.8 years in the control group.

The target population for the cost-effectiveness analysis¹⁸ was people living with T2D. For the analysis, clinical data were obtained from an MA of 227 patients with a mean age of 57 years.

None of the included studies reported evidence on pediatric patients with T2D.

Interventions and Comparators

Relevant to the current report, the intervention and comparator of interest in all included studies were rtCGM and SMBG, respectively.^12-18

Across the primary studies included in the SRs, the length of study follow-up ranged from 12 weeks¹⁴ to 9 months.¹² The duration of rtCGM use was reported in 1 SR¹⁵; it ranged from 48 hours (in a period of 3 months) to daily use for 3 months.

In the RCT by Bergenstal et al.,¹⁷ the participants were followed up for 16 weeks (after a 14-day baseline period) with outcome assessments at weeks 8 and 16. Participants in the intervention group used rtCGM throughout the study period. Participants in the control group wore a blinded CGM device for 2 weeks before each follow-up assessment. This was done to obtain the ambulatory glucose profile. Participants in the rtCGM group were trained to self-adjust their medications, insulin, and diet based on the CGM data. Participants in the SMBG group were asked to “graph” the blood glucose values (method unclear) and use them along with clinician instructions to adjust medications or diet. Participants were followed up by endocrinologists every 4 weeks and given advice on medication or diet adjustments based on the data from the rtCGM or SMBG.

The economic evaluation compared the cost-effectiveness of rtCGM and SMBG compared to SMBG alone.¹⁸

Outcomes

Changes from baseline of glycated hemoglobin (hemoglobin A1C) levels were evaluated as an outcome in 3 included SRs,^14,15 and in the included RCT.¹⁷

Time in range (TIR) is defined as the percentage of time in which blood glucose remains between specific target levels. TIR of 70 mg/dL to 180 mg/dL is considered a marker for glycemic control when using CGM.²⁰ A higher TIR of 70 mg/dL to 180 mg/dL is correlated with better long-term glycemic control. This outcome was reported in the SR by Aggarwal et al.¹² and in the RCT by Bergenstal et al.¹⁷

Glucose variability measured using coefficient of variation (CV) was reported in 2 SRs. It is calculated as standard deviation (SD) divided by mean glucose, and reported as a percentage.²¹ A CV of less than 36% is considered a stable glucose level.²²

Other outcomes reported in the included studies were hypoglycemic events (3 SRs,^12,14,15 1 RCT¹⁷), glucose variability (2 SRs^12,14), health-related quality of life (HRQoL) (1 SR¹⁴), patient satisfaction(1 SR¹⁵), percentage of time above 250mg/dL (1 SR¹²), insulin dose (1 SR¹⁴), and adverse events (1 SR¹²).

The other outcomes reported in the included RCT were hypoglycemic events (percentage of time < 50 mg/dL, < 60 mg/dL, or < 70 mg/dL) as well as percentage of time above 180 mg/dL.¹⁷

The cost-effectiveness analysis reported on incremental cost-effectiveness ratios (ICERs) of cost per QALY.¹⁸

Summary of Critical Appraisal

Additional details regarding the strengths and limitations of included publications are provided in Appendix 3.

Systematic Reviews and Meta-Analysis

All included SRs described their research question and inclusion criteria, and included components of population, intervention, comparators, and outcome.^12-16 In 3 SRs, the review protocol was established a priori and registered in PROSPERO.^13,14,16 Multiple electronic databases were searched in 3 SRs,^12,13,16 and a bibliographic search for additional studies was done in 3 SRs.^12-14 A detailed search strategy was reported in 4 SRs.^12,14-16 In all included SRs, study selection, data extraction, and quality assessment were performed independently by 2 reviewers, and any discrepancies were resolved with the help of a third reviewer.^12-16 Chang et al. also evaluated inter-rater agreement using Cohen’s kappa statistic.¹³ Assessment of quality of included studies was conducted by valid instruments such as Newcastle–Ottawa Scale,¹² modified Jadad scale,^12,14 Cochrane’s Risk of Bias tool,^13-15 and the National Heart, Lung, and Blood Institute Quality Assessment Tools.¹⁶ One SR conducted quantitative synthesis using an MA that was relevant to the current report.¹⁴ Dicembrini et al.¹⁴ conducted an MA with appropriate methods using fixed and random effect models. They also examined heterogeneity between the studies using the I² statistic. No heterogeneity was observed in the analysis (I²  = 0%) Publication bias was explored in 3 SRs using funnel plots and Egger’s test; however, none was detected.^13-15 Lastly, none of the included SRs received external funding or had potential conflicts of interest that could affect the findings of the reviews.^12-16

The SRs had several limitations. Two SRs did not identify any studies relevant to the current report; however, this could be due to limited scope of the reviews.^13,16 The scope of 1 SR was limited to perinatal women with diabetes,¹³ and the scope of the other SR was limited to patients with diabetes being managed by primary care providers.¹⁶ Two of the other SRs (which included relevant studies) did not conduct a comprehensive literature search.^14,15 Only 1 electronic database (MEDLINE) was searched; grey literature was not searched, and it was unclear whether reference lists were searched for additional publications.^14,15 Due to these limitations, it is possible that some relevant studies were not captured in the review and in the MA. The SR by Aggarwal et al. limited their search to studies published since 2018.¹² Although this ensured that the most recent evidence is captured, some previous studies relevant to the research question were not included. In this review, authors also mentioned that out of 60 “relevant” publications, only high-to-moderate quality studies were included in the review.¹² Since a list of excluded studies (and reason for exclusion) was not reported, it is unclear whether the findings of excluded studies differed from those included in the review. Aggarwal et al. also did not report results of all individual studies; only a selective summary of findings for each outcome was reported.¹² This meant that some results of included trials (e.g., hemoglobin A1C results of the MOBILE trial) were not reported in the SR. A list of excluded studies and reason for exclusion were not reported in any of the SRs.^12-16 In the SR with MA by Dicembrini et al.,¹⁴ results of outcomes for which an MA was not conducted were not reported in detail. Without details of results such as effect sizes or confidence intervals from each study, the certainty of those findings was difficult to ascertain. Furthermore, it was not clear whether any of the included primary studies were conducted in Canada, making the generalizability to Canadian settings unclear.^12-16

Randomized Controlled Trial

The study objectives were described, and the trial was registered with prespecified outcomes listed.¹⁷ The inclusion and exclusion criteria for the study were described and were appropriate for the objective. The study was conducted as a multi-centre, parallel group randomized trial. Participants were randomized by a predefined block allocation sequence. Study personnel were concealed to allocation until after the randomization. Randomization was done after a 2-week run-in period to adjust medications. Characteristics of study participants at baseline were reported. Potential confounders, such as duration of disease, body mass index (BMI), and age, were similar between the groups. Participants in both groups were followed up for the same duration of time. Simple outcome data for the main outcomes were reported. An adjusted analysis for a potential confounder (antidiabetic medication) was conducted and results reported.¹⁷

The RCT had several limitations. The trial was open-label, which means the participants and outcome assessors were aware of the intervention received. Since the outcomes were objective and not patient reported, the open-label design of the study is less likely to affect outcome bias. Approximately 30% of screened individuals were not enrolled in the study for various reasons (e.g., failed screening, not interested in participating). Ten patients withdrew from the study during the run-in period due to side effects of medication or for device-related reasons. Lastly, 7 patients (6.1%) withdrew after the study initiation due to side effects or time constraints. Study analysis was conducted excluding these patients. It is possible that the withdrawn patients would have had different outcomes from the study results. When describing the study findings, effect sizes and estimates of random variability such as confidence intervals were not reported for the comparative results between the groups. The details of statistical tests used to compare between the groups were unclear. The results of the subgroup analysis were not reported in a tabular form; interpretation of graphical presentation of those results were difficult to interpret. The study was funded by a pharmaceutical company and the study authors disclosed conflicts of interest related to pharmaceutical companies. The trial was not conducted in Canada, and the generalizability to Canadian settings was unclear.

Economic Evaluation

The economic evaluation had several strengths.¹⁸ The objectives and economic importance were described, and the interventions and comparators of interest were clearly reported. The form of analysis and perspectives were described along with the sources of input parameters in the analysis. Choice of model (Markov model) and the key parameters were justified. Currency, discount rates, and price data used to report the results were described. Clinical effectiveness data were obtained from a systematic review conducted by the authors, which was reported in the publication. Incremental cost-effectiveness results of the intervention compared to the control were reported. Results of the base-case analysis and various sensitivity and scenario analyses were reported. Conclusions were consistent with the data reported and were accompanied by the appropriate caveats.

One of the assumptions used in the model was that only 1 chronic comorbidity would be developed in each patient at each cycle. It is possible that patients could develop multiple comorbidities. The treatment and associated costs of medications considered in the model was unclear. It was unclear whether the patients using rtCGM or SMBG are taking insulin or oral hypoglycemic medications to manage their diabetes. Additional costs of comorbidities could increase the total annual costs of rtCGM and SMBG. The clinical effectiveness data were obtained from an SR, which included studies up to 6 months follow-up. Therefore, long-term clinical effectiveness is unclear and not considered in the economic evaluation. Lastly, the study was conducted in Spain from the perspective of Spanish National health Service.¹⁸ Based on the prevalence of T2D, differences in health care systems, and different willingness-to-pay (WTP) thresholds, generalizability to Canadian settings is unclear.

Summary of Findings

A summary of findings from the included studies are provided in the following paragraphs. Appendix 4 presents the main study findings and authors’ conclusions.

Clinical Effectiveness of rtCGM Versus SMBG for Adults With T2D

Among the studies included in the current report, 3 SRs and 1 RCT reported on the clinical effectiveness of rtCGM versus SMBG for adults with T2D. There was some overlap in the primary studies that were included in the SRs; to avoid duplication of data, results from each primary study are reported only once. When pooled estimates from the MA are available, results of relevant primary studies not included in the MA are presented separately, in addition to the results of the MA.

Hemoglobin A1C Levels

Hemoglobin A1C levels were reported in 2 SRs^14,15 and 1 RCT.¹⁷ A random effects MA¹⁴ of 4 RCTs (n = 429) with study durations ranging from 12 weeks to 26 weeks found that rtCGM use was associated with significantly lower hemoglobin A1C levels at the end of the study compared to SMBG (mean difference for rtCGM versus SMBG = –0.28% [95% CI, –0.43 to –0.13]).¹⁴ There was no heterogeneity between the studies (I² = 0%).

Two other RCTs included in the SR by Janapala et al.¹⁵ found no differences in change from baseline of hemoglobin A1C levels at 3 months between rtCGM and SMBG groups. The open-label RCT by Bergenstal et al.¹⁷ (n = 114) found that, at week 16, there was significant within-group improvement in hemoglobin A1C levels from baseline, but there was no difference in change in hemoglobin A1C levels between rtCGM and SMBG groups (P = 0.11).

It is possible that the pooled analysis in the MA detected between-group differences due to increased statistical power.

TIR of 70 mg/dL to 180 mg/dL

In the MOBILE trial (n = 175) included in the SR by Aggarwal et al.,¹² participants in the rtCGM group reported 15% (95% CI, 8% to 23%) more TIR of 70 mg/dL to 180 mg/dL at month 8 compared to the SMBG group. The between-group difference was statistically significant (P < 0.001).

Bergenstal et al.¹⁷ reported that rtCGM and SMBG were associated with a 18.21% and 11.55% increase, respectively, in TIR of 70 mg/dL to 180 mg/dL at week 16. However, there was no significant between-group difference (P = 0.13). The between-group effect estimate and 95% CI were not reported.

The MOBILE trial included participants on basal insulin,¹² whereas the trial by Bergenstal et al. included patients taking oral medications as well as insulin. It is possible that the difference in population and follow-up times along with the lower sample size (n = 114) in the Bergenstal trial accounts for the difference in findings. A subgroup analysis on patients on basal insulin was not reported by Bergenstal et al.¹⁷

Glucose Variability

One RCT included in the SR by Aggarwal and colleagues found that mean glucose values in the rtCGM group were significantly lower compared to those in SMBG group at month 8 (adjusted difference –26 [95% CI, –41 to –12]; n = 175).¹² The study was conducted among patients with T2D using basal insulin. The clinical significance of this finding is unclear.

Glucose variability measured using CV was reported in 2 SRs.^12,14 Aggarwal et al.¹² reported results from 1 RCT (n = 175) in which the CVs of rtCGM and SMBG groups at month 8 were 27% and 29%, respectively. Adjusted between-group differences indicated little to no difference between the groups (–1.8 [95% CI, –3.5 to 0]), and this was not statistically significant (P = 0.05). In the SR by Dicembrini et al.,¹⁴ results from 1 RCT (n = 158) found “no significant between-group difference” in the rtCGM (CV = 30%) and SMBG (CV = 29%) groups at 12 weeks (between-group effect estimates not reported).

Percentage of Time Above 180 mg/dL

Results from the RCT(n = 114) by Bergenstal et al¹⁷ showed that percentage of time above 180mg/dL decreased by a mean of 17.41% in the rtCGM group, compared to a 12.77% decrease in the SMBG group. The between-group comparison was not statistically significant.

Percentage of Time Above 250 mg/dL

Results from the MOBILE trial (included in an SR¹²) found that, at the 8-month follow-up, rtCGM use was associated with a significantly lower percentage of time above 250mg/dL compared to SMBG (adjusted difference = –16% [95% CI, –21% to –11%]; P < 0.001).

Hypoglycemic Events

Aggarwal et al.¹² reported results from a large (n = 36,080) observational study in which rtCGM initiators were compared with matched non-initiators using claims data. The study found that rtCGM initiation was associated with a greater reduction in hypoglycemia rate compared to non-initiation (adjusted rate difference = –4.0% [95% CI, –7.8% to –0.2%]). However, it was unclear whether all non-initiators were using SMBG to monitor their glycemic levels.

Two other SRs^14,15 reported results from multiple RCTs comparing rates or occurrences of hypoglycemic events between rtCGM and SMBG. There were no severe hypoglycemic events observed in either group in 2 RCTs.¹⁴ As for non-severe hypoglycemic events, the SRs reported that there was no significant difference between rtCGM and SMBG (across 5 RCTs).^14,15 It was noted that the studies included in 1 SR¹⁴ defined “hypoglycemic events” in different ways, ranging from self-reported events to measured blood glucose between 55 mg/dL and 70 mg/dL. The other SR¹⁵ did not provide definitions of hypoglycemia used in individual studies or numerical results from each study.

In 2 trials^12,17 (1 reported in an SR), rtCGM use was associated with a significant reduction in the percentage of time below 70 mg/dL. The MOBILE trial (included in the Aggarwal et al.¹² SR), found that at 8 months, rtCGM use was associated with a significant reduction in the percentage of time below 70 mg/dL. However, the effect was small (adjusted difference for rtCGM versus SMBG = –0.24% [95% CI, –0.42 to –0.05]), and the clinical significance is unclear. Similarly, in the RCT by Bergenstal et al.¹⁷ at week 16, rtCGM was favoured over SMBG in reducing the percentage of time below 70 mg/dL (P < 0.005).

Bergenstal et al. also found that rtCGM use was associated with significantly greater reductions in percentage of time below 60 mg/dL and below 50 mg/dL, compared to SMBG use. It was noted that in the SMBG group, the percentage of time below 70 mg/dL, 60 mg/DL, and 50 mg/dL increased from baseline at 16 weeks.¹⁷

Insulin Dose

The SR by Dicembrini et al.¹⁴ provided results from 2 RCTs that evaluated total insulin dose between rtCGM and SMBG groups. The trials found no difference between the groups (data not reported).

Quality of Life and Patient Satisfaction

HRQoL outcomes were evaluated in 2 RCTs, as reported in 2 SRs.^14,15 There was no “meaningful” difference in HRQoL measures between the rtCGM and SMBG groups in 1 RCT (data not reported).^14,15 One SR¹⁴ reported that “SMBG (was) better” as found in the second RCT (data not reported). Due to gaps in reporting, the clinical meaningfulness of these findings is uncertain.

Janapala and colleagues¹⁵ reported that rtCGM use was associated with a “significant reduction” in BMI, calorie intake and postprandial glucose level, as well as an increase in weekly exercise duration (1 RCT). However, no numerical data were provided.

Adverse Events

A summary of adverse events was reported in 1 SR¹² (from 1 RCT). No serious adverse events were observed in the trial by Price et al. (n = 70). In both the rtCGM and SMBG groups, there was 1 incidence of device-related skin irritation, and in the SMBG group there was an incidence of a disease-related hypoglycemic event. In the RCT by Bergenstal et al. (n = 114),¹⁷ there were no observed adverse events in either group.

Occurrence of adverse events was not evaluated or reported in 2 SRs.^14,15

Clinical Effectiveness of rtCGM Versus SMBG for Pediatric Patients With T2D

No relevant evidence regarding rtCGM versus SMBG for pediatric patients with T2D was identified; therefore, no summary can be provided.

Cost-Effectiveness of rtCGM Versus SMBG for Patients With T2D

The included cost-effectiveness analysis found that rtCGM use is associated with an ICER of 180,533€ per QALY compared to SMBG.¹⁸ The results suggested that rtCGM is not a cost-effective option at a WTP threshold of €20,000 to €25,000 per QALY. The authors concluded that the cost of rtCGM is considerably higher than SMBG.

The cost-effectiveness acceptability curve similarly showed that rtCGM is not cost-effective compared to SMBG in T2D. At the estimated WTP threshold (€20,000 to €25,000 per QALY) in the base case, the probability of rtCGM becoming the cost-effective option is 0%. The threshold in which there is 70% probability of rtCGM becoming cost-effective is approximately €220,000.

Limitations

Two of the included SRs did not identify any relevant primary studies for inclusion, due to their scope being limited to perinatal women with diabetes¹³ or patients with diabetes being managed by primary care providers.¹⁶ The SRs that identified relevant primary studies had serious limitations as described in earlier sections.^12,14,15 The evidence was also heterogeneous across the studies in terms of devices used for rtCGM and medications or insulin used to manage glycemic levels. Another major limitation is the relatively shorter follow-up periods in the included trial. The length of rtCGM use in included primary studies (where reported) ranged from 48 hours (Cosson et al.)¹⁵ to 9 months (MOBILE trial).¹² No evidence was identified regarding the clinical effectiveness or cost-effectiveness of rtCGM in pediatric patients with T2D. Lastly, it was not clear whether any of the included primary studies in the SRs were conducted in Canada^12-16 The RCT was not conducted in Canada. Thus, the generalizability to Canadian settings is unclear. The cost-effectiveness study¹⁸ was conducted in Spain; therefore, its applicability to Canadian settings is unclear as well.

Conclusions and Implications for Decision- or Policy-Making

Five SRs^12-16 and 1 RCT¹⁷ were included in this report to summarize the evidence regarding the clinical effectiveness of rtCGM compared to SMBG in adult individuals with T2D. One economic evaluation was identified regarding the cost-effectiveness of rtCGM compared to SMBG in people with T2D.¹⁸ No evidence was identified regarding the clinical effectiveness of rtCGM in pediatric patients with T2D.

In adults living with T2D, evidence from an MA¹⁴ showed that glycemic monitoring with rtCGM is associated with improved hemoglobin A1C levels compared to SMBG. Evidence from 2 RCTs regarding the outcome TIR of 70 mg/dL to 180 mg/dL was mixed, with 1 trial finding results favouring rtCGM and the other finding no difference between the groups. rtCGM was also found to be favoured over SMBG in maintaining a stable blood glucose level, as indicated by decreased percentages of time in either extreme, including time below 50 mg/dL, 60 mg/dL, and 70 mg/dL, as well as time above 250 mg/dL. There was little to no difference in glucose variability (measured using CV), hypoglycemic events, or insulin dose between rtCGM and SMBG. Due to lack of reporting of data in the SRs, the clinical effectiveness of rtCGM versus SMBG in quality-of-life outcomes is uncertain. In the 2 RCTs (1 reported in an SR) that reported adverse events, rtCGM appeared to be relatively safe. One incidence of skin irritation was reported in both the rtCGM and SMBG groups in an RCT of 70 participants. The identified evidence is uncertain due to risk of bias within the included studies, indirectness due to possible limited generalizability to Canadian settings, and heterogeneity across the studies.

Evidence from a cost-effective analysis conducted in Spain suggested that glycemic monitoring with rtCGM was not cost-effective compared to SMBG alone in individuals living with T2D.¹⁸

A recent CADTH report²³ on the clinical effectiveness and cost-effectiveness of rtCGM compared to SMBG in T1D found that rtCGM is favoured over SMBG for glycemic monitoring in adults and pediatric patients. rtCGM was also found to be cost-effective compared to SMBG at a WTP threshold of $50,000 per QALY. Another CADTH report on evidence-based guidelines regarding the use of rtCGM in all patients with diabetes is currently being produced.

Future large-scale clinical trials with longer periods of follow-up are warranted to evaluate the effectiveness of rtCGM in T2D. Economic evaluations from a Canadian perspective, in light of recent trials, could help inform decision-makers in a Canadian setting.

References

1.Diabetes in Canada: backgrounder. Ottawa (ON): Diabetes Canada; 2020: https://www.diabetes.ca/DiabetesCanadaWebsite/media/Advocacy-and-Policy/Backgrounder/2020_Backgrounder_Canada_English_FINAL.pdf. Accessed 2022 Jul 11.

2.Saeedi P, Petersohn I, Salpea P, et al. Global and regional diabetes prevalence estimates for 2019 and projections for 2030 and 2045: results from the International Diabetes Federation Diabetes Atlas, 9(th) edition. Diabetes Res Clin Pract. 2019;157:107843. PubMed

3.Bullard KM, Cowie CC, Lessem SE, et al. Prevalence of diagnosed diabetes in adults by diabetes type - United States, 2016. MMWR Morb Mortal Wkly Rep. 2018;67(12):359-361. PubMed

4.Wu Y, Ding Y, Tanaka Y, Zhang W. Risk factors contributing to type 2 diabetes and recent advances in the treatment and prevention. Int J Med Sci. 2014;11(11):1185-1200. PubMed

5.Farkas B, Corbett C, Hofmeister M, et al. Glucose monitoring technologies for the management of insulin-dependent diabetes. Calgary (AB): The Health Technology Assessment Unit, University of Calgary; 2019: https://www2.gov.bc.ca/assets/gov/health/about-bc-s-health-care-system/heath-care-partners/health-authorities/bc-health-technology-assessments/glucose-monitoring-hta.pdf. Accessed 2022 Jul 28.

6.Lin R, Brown F, James S, Jones J, Ekinci E. Continuous glucose monitoring: a review of the evidence in type 1 and 2 diabetes mellitus. Diabet Med. 2021;38(5):e14528. PubMed

7.Gomez AM, Henao Carrillo DC, Munoz Velandia OM. Devices for continuous monitoring of glucose: update in technology. Med Devices (Auckl). 2017;10:215-224. PubMed

8.Farrell K, Loshak H. Intermittently scanned and real-time continuous glucose monitoring for people with diabetes. Can J Health Technol. 2022;2(7). https://www.cadth.ca/sites/default/files/pdf/htis/2022/RC1436%20rtCGM%20vs%20isCGM%20Diabetes_Final.pdf. Accessed 2022 Aug 09.

9.Shea BJ, Reeves BC, Wells G, et al. AMSTAR 2: a critical appraisal tool for systematic reviews that include randomised or non-randomised studies of healthcare interventions, or both. BMJ. 2017;358:j4008. PubMed

10.Downs SH, Black N. The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Community Health. 1998;52(6):377-384. PubMed

11.Higgins JPT, Green S, editors. Figure 15.5.a: Drummond checklist (Drummond 1996). Cochrane handbook for systematic reviews of interventions. London (GB): The Cochrane Collaboration; 2011: http://handbook-5-1.cochrane.org/chapter_15/figure_15_5_a_drummond_checklist_drummond_1996.htm. Accessed 2022 Jul 22.

12.Aggarwal A, Pathak S, Goyal R. Clinical and economic outcomes of continuous glucose monitoring system (CGMS) in patients with diabetes mellitus: a systematic literature review. Diabetes Res Clin Pract. 2022;186:109825. PubMed

13.Chang VYX, Tan YL, Ang WHD, Lau Y. Effects of continuous glucose monitoring on maternal and neonatal outcomes in perinatal women with diabetes: a systematic review and meta-analysis of randomized controlled trials. Diabetes Res Clin Pract. 2022;184:109192. PubMed

14.Dicembrini I, Mannucci E, Monami M, Pala L. Impact of technology on glycaemic control in type 2 diabetes: A meta-analysis of randomized trials on continuous glucose monitoring and continuous subcutaneous insulin infusion. Diabetes Obes Metab. 2019;21(12):2619-2625. PubMed

15.Janapala RN, Jayaraj JS, Fathima N, et al. Continuous glucose monitoring versus self-monitoring of blood glucose in type 2 diabetes mellitus: a systematic review with meta-analysis. Cureus. 2019;11(9):e5634. PubMed

16.Kieu A, King J, Govender RD, Ostlundh L. The benefits of utilizing continuous glucose monitoring of diabetes mellitus in primary care: a systematic review. J Diabetes Sci Technol. 2022:19322968211070855. PubMed

17.Bergenstal RM, Mullen DM, Strock E, Johnson ML, Xi MX. Randomized comparison of self-monitored blood glucose (BGM) versus continuous glucose monitoring (CGM) data to optimize glucose control in type 2 diabetes. J Diabetes Complications. 2022;36(3):108106. PubMed

18.García-Lorenzo B, Rivero-Santana A, Vallejo-Torres L, et al. Cost-effectiveness analysis of real-time continuous monitoring glucose compared to self-monitoring of blood glucose for diabetes mellitus in Spain. J Eval Clin Pract. 2018;24(4):772-781. PubMed

19.Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. J Clin Epidemiol. 2009;62(10):e1-e34. PubMed

20.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. PubMed

21.Umpierrez GE, P Kovatchev B. Glycemic variability: how to measure and its clinical implication for type 2 diabetes. Am J Med Sci. 2018;356(6):518-527. PubMed

22.Danne T, Nimri R, Battelino T, et al. International consensus on use of continuous glucose monitoring. Diabetes Care. 2017;40(12):1631-1640. PubMed

23.Barbara AM, Grobelna A. Real time continuous glucose monitoring for people living with type 1 diabetes. Can J Health Technol. 2022;2(8). https://canjhealthtechnol.ca/index.php/cjht/article/view/RC1442/RC1442. Accessed 2022 Aug 09.

Appendix 1: Selection of Included Studies

Figure 1: Selection of Included Studies

Appendix 2: Characteristics of Included Publications

Note that this appendix has not been copy-edited.

Table 2: Characteristics of Included Systematic Reviews and Meta-Analyses

CGM = continuous glucose monitoring; CSII = continuous subcutaneous insulin infusion: DM = diabetes mellitus; FGM = flash glucose monitoring; HRQoL = health-related quality of life; MDI = multiple daily injections; n = number; NR = not reported; NRS = non-randomized study; RCT = randomized controlled trial; rtCGM = real-time continuous glucose monitoring; SD = standard deviation; SMBG = self-monitoring of blood glucose; T1D = type 1 diabetes mellitus; T2D = type 2 diabetes mellitus; TIR = time in range

^aOnly partial results from the unpublished trial (NCT01237301) were available when this SR was conducted. The results of that trial were not included in the meta-analysis. Therefore, the trial has been included separately in the current report (Bergenstal et al., 2022¹⁷)

Table 3: Characteristics of Included Randomized Controlled Trial

CGM = continuous glucose monitoring; DPP4 = dipeptidyl peptidase 4; GLP = Glucagon like peptide; n = number; RCT = randomized controlled trial; rtCGM = real-time continuous glucose monitoring; SD = standard deviation; SMBG = self-monitoring of blood glucose; T2D = type 2 diabetes mellitus; TIR = time in range; TZD = thiazolidinediones.

^aReported as T2D onset age in the publication.

Appendix 3: Critical Appraisal of Included Publications

Note that this appendix has not been copy-edited.

Table 5: Strengths and Limitations of Systematic Reviews and Meta-Analyses Using AMSTAR 2⁹

AMSTAR 2 = A MeaSurement Tool to Assess systematic Reviews 2; GRADE = Grading of Recommendations, Assessment, Development, and Evaluations; NRS = non-randomized study; PRISMA = Preferred Reporting Items for Systematic Reviews and Meta-Analyses; RCT = randomized controlled trial; rtCGM = real-time continuous glucose monitoring.

Table 6: Strengths and Limitations of Clinical Study Using the Downs and Black Checklist¹⁰

AGP = ambulatory glucose profile; CGM = continuous glucose monitoring; DM = diabetes mellitus; RCT = randomized controlled trial.

Table 7: Strengths and Limitations of Economic Evaluation Using the Drummond Checklist¹¹

ICER = incremental cost-effectiveness ratio; MA = meta-analysis; QALY = quality-adjusted life years; T2D = type 2 diabetes mellitus.

Appendix 4: Main Study Findings and Authors’ Conclusions

Note that this appendix has not been copy-edited.

Table 8: Summary of Findings by Outcome — Hemoglobin A1C Levels

CI = confidence interval; MA = meta-analysis; n = number; NR = not reported; RCT = randomized controlled trial; rtCGM = real-time continuous glucose monitoring; SD = standard deviation; SMBG = self-monitoring of blood glucose; SR = systematic review; T2D = type 2 diabetes mellitus; vs. = versus.

Table 9: Summary of Findings by Outcome — Time in Range

CI = confidence interval; n = number; RCT = randomized controlled trial; rtCGM = real-time continuous glucose monitoring; SD = standard deviation; SMBG = self-monitoring of blood glucose; SR = systematic review; T2D = type 2 diabetes mellitus; TIR = time in range; vs. = versus.

Table 10: Summary of Findings by Outcome — Glucose Variability

CI = confidence interval; MA = meta-analysis: n = number; RCT = randomized controlled trial; rtCGM = real-time continuous glucose monitoring; SD = standard deviation; SMBG = self-monitoring of blood glucose; SR = systematic review; T2D = type 2 diabetes mellitus; vs. = versus.

Table 11: Summary of Findings by Outcome — Percentage of Time > 180 mg/dL

CI = confidence interval; MA = meta-analysis: n = number; RCT = randomized controlled trial; rtCGM = real-time continuous glucose monitoring; SD = standard deviation; SMBG = self-monitoring of blood glucose; SR = systematic review; T2D = type 2 diabetes mellitus; vs. = versus

Table 12: Summary of Findings by Outcome — Percentage of Time > 250 mg/dL

SMBG = blood glucose monitoring; CI = confidence interval; n = number; RCT = randomized controlled trial; rtCGM = real-time continuous glucose monitoring; SMBG = self-monitoring of blood glucose; SR = systematic review; T2D = type 2 diabetes mellitus; vs. = versus.

Table 13: Summary of Findings by Outcome — Hypoglycemic Events

CGM = continuous glucose monitoring; CI = confidence interval; MA = meta-analysis: n = number; NR = not reported; RCT = randomized controlled trial; rtCGM = real-time continuous glucose monitoring; SD = standard deviation; SMBG = self-monitoring of blood glucose; SR = systematic review; T2D = type 2 diabetes mellitus; vs. = versus.

Table 14: Summary of Findings by Outcome — Quality of Life

CMG = continuous glucose monitoring; CI = confidence interval; DTSQ = diabetes treatment satisfaction questionnaire; HRQoL = health-related quality of life; MA = meta-analysis: n = number; NR = not reported; RCT = randomized controlled trial; rtCGM = real-time continuous glucose monitoring; SD = standard deviation; SMBG = self-monitoring of blood glucose; SR = systematic review; T2D = type 2 diabetes mellitus; vs. = versus.

Table 15: Summary of Findings by Outcome — Insulin Dose

CMG = continuous glucose monitoring; CI = confidence interval; MA = meta-analysis; n = number; NR = not reported; RCT = randomized controlled trial; rtCGM = real-time continuous glucose monitoring; SD = standard deviation; SMBG = self-monitoring of blood glucose; SR = systematic review; T2D = type 2 diabetes mellitus; vs. = versus.

Table 16: Summary of Findings by Outcome — Adverse Events

n = number; RCT = randomized controlled trial; rtCGM = real-time continuous glucose monitoring; SMBG = self-monitoring of blood glucose; SR = systematic review.

Table 17: Summary of Findings of Included Economic Evaluation

ICER = incremental cost-effectiveness ratio; QALY = Quality-adjusted life-year; rtCGM = real-time continuous glucose monitoring; SMBG = self-monitoring of blood glucose; T1D = type 1 diabetes mellitus; T2D = type 2 diabetes mellitus; vs. = versus; WTP = willingness-to-pay.

Appendix 5: Overlap Between Included Systematic Reviews

Note that this appendix has not been copy-edited.

Table 18: Overlap in Relevant Primary Studies Between Included Systematic Reviews

^a,bThese were separate publications of the same trial

Appendix 6: Real-World Evidence from the Literature Search

Note that this appendix has not been copy-edited.

The following publications were identified because they provide information related to real-world evidence for real-time continuous glucose monitoring for people living with type 2 Diabetes.

Gavin JR, Bailey CJ. Real-world studies support use of continuous glucose monitoring in type 1 and type 2 diabetes independently of treatment regimen. Diabetes Technol Ther. 2021 09;23(S3):S19-S27. PubMed

Da Silva J, Bosi E, Jendle J, et al. Real-world performance of the MiniMedTM 670G system in Europe. Diabetes Obes Metab. 2021 Aug;23(8):1942-1949. PubMed

Van Der Linden, van der Linden J, Welsh JB, Hirsch IB, Garg SK. Real-time continuous glucose monitoring during the coronavirus disease 2019 pandemic and its impact on time in range. Diabetes Technol Ther. 2021 Mar;23(S1):S1-S7. PubMed

Appendix 7: References of Potential Interest

Note that this appendix has not been copy-edited.

Previous CADTH Reports

Flash glucose monitoring and continuous glucose monitoring for people with diabetes in acute care settings. (CADTH reference list: summary of abstracts). Ottawa (ON): CADTH; 2021: https://www.cadth.ca/sites/default/files/pdf/htis/2021/RB1562%20CGM%20and%20FGM%20in%20Acute%20Care%20Final.pdf. Accessed 2022 Jul 28.

Continuous glucose monitoring for patients with all diabetes types: clinical effectiveness and guidelines. (CADTH rapid response report: summary of abstracts). Ottawa (ON): CADTH; 2018: https://www.cadth.ca/sites/default/files/pdf/htis/2018/RB1289%20Continuous%20Glucose%20Monitoring%20Final.pdf. Accessed 2022 Jul 28.

Review Articles

Rosenberg K. Self-monitoring of blood glucose can lead to improved glycemic control. Am J Nurs. 2022 02 01;122(2):50. PubMed

Kesavadev J, Misra A, Saboo B, et al. Time-in-range and frequency of continuous glucose monitoring: recommendations for South Asia. Diabetes Metab Syndr. 2022 Jan;16(1):102345. PubMed

Babu RN, Pravallika MYL, Kumar NDP, Bhargavi MB. Continuous glucose monitoring devices: a systematic review. J Glob Trends Pharm Sci. 2020;11(2):7562-7568.

Oser TK, Litchman ML, Allen NA, et al. Personal continuous glucose monitoring use among adults with type 2 diabetes: clinical efficacy and economic impacts. Curr Diab Rep. 2021 12 09;21(11):49. PubMed

Additional References

Jackson MA, Ahmann A, Shah VN. Type 2 diabetes and the use of real-time continuous glucose monitoring. Diabetes Technol Ther. 2021 03;23(S1):S27-S34. PubMed

Meng R, Meng R, Gu T, Yang F, Liu J, Sun Q, Zhu D. Performance evaluation of the Glunovo® continuous blood glucose monitoring system in Chinese participants with diabetes: a multicenter, self-controlled trial. Diabetes Ther. 2021 Dec;12(12):3153-3165. PubMed