Continuous Glucose Monitoring Systems Can Meet the Challenge of Glucose Management and Beyond in Individuals with Type 2 Diabetes: An Expert Multidisciplinary Position

Abstract

The increasing prevalence of type 2 diabetes (T2D) can be considered a global healthcare emergency, with far-reaching burdens on the health and well-being of people with diabetes, their carers and families, and the mounting costs within each national healthcare economy. Although application of diabetes technologies, such as insulin pumps, continuous glucose monitoring (CGM) systems, and a range of connected devices, is starting to have an impact on the outcomes of care for people with type 1 diabetes (T1D), similar application for people with T2D is lagging behind. This is a purely cost-based decision, since evidence from numerous randomized controlled trials (RCTs) and real-world studies has shown the significant clinical impact of diabetes technologies for people with T2D, whether they are on insulin therapy or not. Amongst available technologies, it is the lack of widespread access to CGM devices for people with T2D that is most pressing, as these systems have the potential to bring a quantum change in the way people with T2D and their healthcare professionals (HCPs) are supported to manage the adverse impact both of hyperglycemia and hypoglycemia. Central to improving diabetes care for people with T2D is the demonstration in many studies that CGM can actively support healthy behavioral changes to meal planning and physical activity, with concomitant improvements in mental health and quality of life. In this expert opinion, we review the significant evidence base on which application of CGM in people with T2D is founded, and make the case for wider access for every person with diabetes as early as possible after diagnosis, in order to mitigate the global impact of T2D.

Key Summary Points

The evidence base indicates the application of continuous glucose monitoring (CGM) systems on persistent basis for all people with type 2 diabetes (T2D) who are on insulin therapy, including those on basal insulin therapy.

Application of CGM systems episodically is important in special situations in T2D, particularly at the time of diagnosis and at any point when treatment escalation is proposed.

It is critical that CGM technologies (including CGM data management systems) are accessible for all medical practices that provide care for people with T2D, including family medicine, cardiology, nephrology, ophthalmology, and neurology services. This acknowledges the need for training within these specialties and integrated electronic health record systems.

Artificial intelligence (AI)-based tools must be employed in the primary care setting and in diabetes clinics (e.g., for diagnosis of early ophthalmic changes) to leverage the availability of CGM data for people with T2D.

Introduction

According to International Diabetes Federation (IDF) data from 2021, diabetes affects 10.5% of the world’s population aged 20–79 years, with almost half of these people unaware they have the disease [1]. IDF projections indicate that by 2045, 1 in 8 adults, or about 783 million people, will suffer from diabetes. Diabetes in 2019 was the direct cause of 1.5 million deaths, with 48% of all diabetes deaths occurring before the age of 70 [2]. A further 460,000 deaths from kidney disease were also attributed to diabetes [3]. Furthermore, elevated blood glucose levels are responsible for around 20% of deaths from cardiovascular causes [3]. In recognition of this morbidity and mortality, the United Nations (UN) has declared diabetes a twenty-first century epidemic.

Type 2 diabetes (T2D) is the most prevalent form of diabetes, accounting for more than 90% of cases. Its occurrence is influenced by socio-economic, demographic, environmental, and genetic factors. The most important of these are an ageing population, an increasing prevalence of overweight and obesity, decreasing levels of physical activity, and unhealthy eating habits. Up to 60% of individuals with T2D are overweight, while obesity is prevalent in a further 20%. In addition, T2D is diagnosed in 20–30% of patients with acute coronary syndromes, with 40% of these having impaired glucose tolerance [4]. Epidemiological data indicates that diabetes shortens life expectancy by an average of 6 years. Moreover, the average life expectancy of a 60-year-old with diabetes and with diagnosed cardiovascular disease is 12 years less than that of the general population [5].

In this review we identify how application of diabetes technologies in T2D may reduce morbidity and mortality for people with T2D, particularly through the use of continuing glucose monitoring (CGM) devices, used either on a daily basis or episodically, during periodic review or during treatment changes. The selection and discussion of studies in this expert position is in compliance with ethics guidelines. This article is based on previously conducted studies and does not contain any new studies with human participants or animals performed by any of the authors.

Search Strategy and Selection Criteria

The references used in this review on the application of CGM technology in T2D were identified through searches of medical literature databases (see below) for articles published from any point up to April 28, 2025. The search was performed using nested Boolean operators (AND/OR/NOT) to combine, expand, or limit results, selectively using the terms “continuous glucose monitoring”, “CGM”, “flash glucose monitoring”, “intermittently scanned continuous glucose monitoring”, “type 2 diabetes”, “T2D or T2DM”, “acute diabetes complications”, “acute diabetes events”, “diabetic ketoacidosis”, “time in range”, “time below range”, “hypoglycemia”, “time above range”, “hyperglycemia”, “macrovascular disease”, “microvascular disease”, “cardiovascular disease”, “CVD”, “chronic kidney disease”, “CKD”, “randomized controlled trial”, “randomized clinical trial”, “real world study”, “retrospective study”, “cross-sectional study”. Terms were used in strings of three operators.

English and foreign-language articles resulting from these searches and relevant references cited in those articles were reviewed. Conference abstracts were not searched on the basis that abstract presentations not culminating in a publication by the April 28, 2025 cutoff would not reflect substantiated outcomes at that point.

Databases and resources covered:

• MEDLINE/PubMed

• Embase

• CINAHL (Cumulative Index to Nursing and Allied Health Literature)

• Web of Science

• Scopus

• Google Scholar

Challenges Associated with Care of Individuals with Type 2 Diabetes

The timely diagnosis of T2D remains a challenge, since the condition does not present with clear symptoms for a significant period of time once hyperglycemia may be detectable. Consequently, people with new-onset T2D, or those at risk, do not actively seek medical help. It is for this reason that identifying people at risk and undertaking regular screening is essential for making an effective diagnosis of prediabetes or intermediate hyperglycemia [6] or T2D, as well as enabling early initiation of pharmacological and behavioral therapy [7]. Effective treatment from the moment of diagnosis and early intensification of therapy when treatment goals are not being met are key to preventing and delaying the development of the vascular complications of diabetes.

The largest group of people with T2D are those of advancing years with numerous comorbidities, in whom the benefits of lowering blood glucose levels must be weighed against the risk of hypoglycemia [8]. Frequent overlap between multiple disease processes and the drugs employed in treating them renders glycemic control, in this patient group, a therapeutic challenge [9].

People with T2D have up to a 10% greater risk of cardiovascular disease (CVD), a 53% higher risk of myocardial infarction, a 58% higher risk of stroke, and a 12% higher risk of heart failure when compared with people without diabetes [10]. Numerous studies confirm that the risk of CVD is increased by a factor of 1.6–2.3 in people with T2D, compared to those without [11–13]. Some of this risk is attributed to the impact of hypoglycemia on processes that are implicated in development of CVD. Studies in healthy individuals and in those with diabetes, particularly T1D, indicate that acute hypoglycemia leads to increased proinflammatory cytokines, mobilization of inflammatory leukocytes, and increased platelet reactivity [14–16]. Moreover, repeated hypoglycemia in T2D is also associated with enhanced coagulation, oxidative stress, vascular inflammation, endothelial dysfunction, and platelet activation [17], which are each associated with CVD.

Additionally, age-specific progressive reductions in activity, cognitive abilities, or organ changes can significantly affect the achievement of glycemic targets. Therefore, elderly people with T2D need their treatment regimens and glycemic goals to be individualized. In this elderly population, hypoglycemia constitutes a higher risk factor for cardiovascular disease, falls, fractures, the occurrence of cognitive disorders, and mortality. Reducing the risk of hypoglycemia is therefore a clear therapeutic priority, in order to ensure the safety of diabetes treatment and a good quality of life [8].

A major challenge in daily practice is the limited availability of specialist care and suboptimal collaboration between primary care and endocrinology [18]. The majority of people with T2D should be treated in primary care facilities [19]. Only individuals with type 1 diabetes (T1D) or other specific types of diabetes, as well as children and adolescents, pregnant women, or those planning a pregnancy, require a referral to a diabetes clinic for treatment [19].

A diabetes consultation, as part of coordinated care within the primary care setting, is advisable when therapeutic goals are not being met (in order to intensify therapy with insulin), when diabetes or pharmacotherapy complications arise, and when comorbidities develop that make therapy more difficult. Coordinated care enables the standard of care to be elevated, thanks to the application of additional diagnostic tools, comprehensive and structured education, non-pharmacological interventions, as well as more efficient communication between the primary care physician and a diabetologist (Table 1) [19].

Table 1.

Challenges in daily practice when caring for people with T2D

Person profile	Delivery of care
Advanced years Limited learning potential Impaired cognitive abilities Low acceptance or limited ability to change lifestyle/nutritional habits High cardiovascular risk Risk of hypoglycemia (especially in association with insulin therapy) Suboptimal weight control Comorbidity with other chronic diseases Combination therapy	Late diagnosis Inertia in intensification of therapy Limited educational opportunities Limited time for medical appointments Suboptimal cooperation between diabetologist and primary care provider Insufficient coordination of the treatment process

T2D type 2 diabetes

Action Strategy for Type 2 Diabetes

Current therapeutic guidelines for the treatment of diabetes emphasize individualization of care, both in terms of glycemic targets and the route to achieving these [20]. The selection of a therapeutic strategy in a person with T2D should take into account any risk of hypoglycemia and its consequences (especially in the elderly and in patients with cardiovascular and/or nervous system disorders), the duration of the diabetes, the presence of major diabetic vascular complications, and of any serious concomitant diseases. The expectations and attitude of the person with T2D and their anticipated commitment to therapy must also be considered, as well as the benefit-to-risk ratio associated with the achievement of specific treatment targets [20]. Behavioral therapy is also now considered to be an essential first element of treatment for people with diabetes, regardless of diabetes type and age, where the priority is to maintain a healthy body weight [20]. This aspect of therapy focuses on health education and emotional support. It embraces issues related to healthy and balanced nutrition, regular physical activity that is tailored to the patient’s abilities and needs, smoking cessation and reduced alcohol consumption, along with reducing stress levels. Implementation of the principles of good nutrition and physical activity in people with diabetes can help to reduce excessive body weight, to achieve and maintain normal plasma glucose levels, with optimal serum lipid and lipoprotein levels, and optimal blood pressure readings [20].

According to the American Diabetes Association (ADA)/European Association for the Study of Diabetes (EASD) recommendations, the drugs of first choice used at the start of pharmacological treatment in patients with T2D are metformin, as well as sodium-glucose cotransporter type 2 (SGLT2) inhibitors and glucagon-like peptide 1 receptor agonists (GLP-1 RAs) [20]. In persons with documented atherosclerotic cardiovascular disease (ASCVD), heart failure, chronic kidney disease (CKD), or cardiovascular risk factors, treatment options with confirmed cardiovascular and nephroprotective effects should be selected, such as SGLT2 inhibitors in patients with CKD and/or heart failure, and GLP-1 RAs (where contraindications to the use of SGLT2 inhibitors are identified, as well as in patients with numerous risk factors). The prescribing of GLP-1 RAs or SGLT2 inhibitors is advised for patients suffering from obesity [20]. Since T2D is progressive in nature, initiation of insulin therapy may be required for some patients, with further intensification in order to achieve glycemic targets. Simplification of the insulin therapy regimen and/or consideration of less stringent glycemic targets is recommended for some individuals with T2D (e.g., those with a high risk of hypoglycemia, non-adherence with treatment, poor quality of life (QoL) as a result of their therapy, or those with a short, estimated survival time) [20].

In accordance with clinical guidelines, a key element in reducing the risk of diabetes complications is comprehensive management, which should include optimization both of antihyperglycemic treatment (by prioritizing the use of drugs with a proven organ-protective effect) and treatment of other cardiovascular risk factors (hypertension, lipid profile disorders, prothrombotic factors), as well as effective behavioral changes, particularly in terms of weight control [20–24].

Importance of Continuous Glucose Monitoring for Glycemic Control in People with T2D

In recent years, we have seen the introduction of many new pharmacological therapies in the treatment of T2D that have had a significant impact on reducing the risk of complications from this disease [25]. Of equal importance in achieving this goal is the availability of new technological solutions in everyday diabetes practice, such as continuous subcutaneous insulin infusion (CSII) pumps and, more commonly, CGM systems. The use of these technologies, especially CGM devices, has a multifactorial impact on the challenges faced daily by people with T2D [26].

The biggest breakthrough in day-to-day diabetes self-management has been the introduction of CGM technology into diabetes care [27]. CGM systems were initially approved exclusively for ancillary use, as the glucose readings provided by the system did not allow for decisions to be made regarding insulin dosing. Over the years, as accuracy has improved, several of the available CGM systems enable insulin therapy to be adjusted, on the basis of the system’s readings, without the need for verification by glucometric means [28].

CGM systems consist of several components: a sensor placed on the arm/abdomen/upper buttock (depending on the system), a transmitter that operates in tandem with the sensor (either as a separate component or integrated into the sensor), and a reader or smartphone app that displays glucose data and, above all, that analyses the data generated by the sensor [28]. Glucose concentration is measured by the sensor in the interstitial fluid (ISF) and is based on an enzymatic-electrical reaction that is generated across the microfilament [29]. This reaction, initiated by the presence of glucose with glucose oxidase enzyme complexes covering a platinum or polymer electrode, generates an electrical impulse that is captured by the electrode. The intensity of the signal passed from the transmitter to the receiver is interpreted by an algorithm and translated into a glucose concentration value [30]. Glucose measurements are taken by CGM systems at intervals ranging from every minute to every 5 min (e.g., FreeStyle Libre 2 and FreeStyle Libre 3 System, Abbott Diabetes Care Ltd; Dexcom G6 and G7 System, Dexcom Inc.) [28], with ISF glucose values being displayed on a reader/smartphone app. Another method to assess the concentration of glucose in the ISF is photometric measurement (the Eversense system, Senseonics Inc.) [26]. The glucose present in the ISF reversibly binds to the fluorescent coating that covers the sensor, and the intensity of the light emitted by this coating varies with the glucose concentration, which is measured by photodetectors within the sensor and a specialized electronic circuit converts the result into a digital format, sending it via a transmitter to the mobile app [31, 32].

From a clinical perspective, the use of CGM systems allows for a significantly more comprehensive insight into glucose changes throughout the day [33], as summarized in Table 2. They provide the user not only with a point reading, as with the use of a blood glucose meter, but they also display an indication as to the speed and direction of glucose changes over the ensuing period, using visible trend arrows, as well as showing how glucose levels have fluctuated over the previous hours [34]. In a qualitative investigation of the impact of CGM for adults with T2D [35], users provided personal testimony of the impact of using CGM, for example: “I wasn’t really drawing connections [before CGM]. It was frustrating not being able to connect what I was doing and what my sugars were doing. I couldn’t really see the patterns. Now I have a better understanding. I see how food and even walking affects my sugar.” and “You really don’t know the effect everything is having [on your glucose levels] until you see it up close and in real-time. Before CGM, I was just guessing. I had somewhat of an idea, but I didn’t really know for sure what different foods did. CGM has brought so much insight into what different foods do to my system.” These statements were provided as part of the overall analysis under the theme—making the invisible visible [35].

Table 2.

Scope of information provided by a CGM system

Current glucose concentration value Trend (direction and rate of change) over the next 15–30 min Assessment of changes in glucose in the last hours/overnight period New parameters to assess diabetes control (% of time within range, % of time below/above target) Glucose alarms to warn about low/high glucose concentration levels Display of changes in glucose concentration for each day of sensor use Estimated HbA1c reading, average glucose concentration Number and duration of hypoglycemic events Evaluation of sensor activity The ability to enter additional information (meals/exercise periods/insulin dose) Information on dose, time of insulin administration (where integrated with a smart pen)

Using the FreeStyle Libre 2 system and the FreeStyle LibreLink app as an example [82]

CGM continuous glucose monitoring

Clinical Metrics Provided by CGM Devices

The data provided by a CGM system permits significantly more accurate adjustment of doses of insulin and other medications, which are based on observed values and trends in ISF glucose levels. Just as important, using a CGM system creates a better understanding by the person with diabetes of the impact of their daily decisions regarding food intake or physical activity on their glucose levels that is not possible using the information obtained from self-monitoring with a blood glucose meter [28, 36]. An additional advantage of CGM systems is the audible or vibrating alerts that users can receive when their glucose levels are too high or too low, thus enabling them to rapidly take remedial action [29, 37].

The value of HbA1c as a marker for glycemic health can be compromised in situations where red blood cell dynamics are disrupted, such as in patients with CKD or end-stage renal failure, anemia, or blood transfusions [38]. Equally, glycation rates vary considerably between individuals, either with or without diabetes [38]. The introduction of CGM systems into diabetes care has enabled the practical use of new clinical parameters in the assessment of metabolic control in people with diabetes [20]. An international consensus, published in 2019 [39], identifies key parameters that are automatically generated by the systems, an analysis of which enables individualized treatment decisions to be made. The consensus also recommends ranges and targets that should be achieved by specific groups of people with diabetes. For the general population of people with T1D or T2D, it recommends achieving a time in range (TIR) within 70–180 mg/dL at least 70% of the time, with time below range (TBR) with glucose levels < 70 mg/dL no more than 4% of the time, with TBR readings below 54 mg/dL < 1% of the time. Individuals are also recommended to spend < 25% of the time above range (TAR) with hyperglycemia > 180 mg/dL and < 5% of the TAR where glycemia exceeds 250 mg/dL [39] (Fig. 1).

Fig. 1.

Target ranges and glycemic goals for people with T1D or T2D using CGM systems [39]

The recommended treatment target of > 70% TIR corresponds to an HbA1c level of around 7.0% (53 mmol/mol). This has been confirmed in studies describing %TIR and HbA1c simultaneously in different patient populations using different CGM systems. Each 10%-point increase in %TIR correlates with a reduction in HbA1c of approximately 0.5% [39]. It has also been demonstrated that a reduction in %TIR is associated with an increased risk of the development of microvascular and macrovascular complications associated with diabetes, including increased cardiovascular mortality in T1D [40–42] or T2D [41, 43]. Specifically in T2D, a cross-sectional study in adults with T2D [44] has reported that abnormal carotid intima-media thickness (CIMT), a marker of subclinical atherosclerotic disease, is associated with lower TIR, compared to those with normal CIMT (p < 0.001). Similarly, changes in grayscale median (GSM) indices of carotid artery wall plaque instability were associated with reduced TIR in adults with T2D, with no prior history of CVD [45]. Lower TIR in adults with T2D is also associated with increased brachial-ankle pulse wave velocity (BaPWV), which is a marker of increased arterial stiffness [46, 47]. In each of the studies discussed, the changes in risk were independent of HbA1c. Reduced TIR as a risk marker for poor cardiovascular health is supported by a large prospective cohort study showing that TIR during hospitalization is inversely associated with CVD mortality in people with T2D [43].

For older people with T1D or T2D and/or those at risk of severe hypoglycemia, the recommended daily percentage targets in specific glycemic ranges differ from those for the general population. This includes people with longer duration of diabetes or duration of insulin therapy, with impaired awareness of hypoglycemia (IAH), those with comorbidities, cognitive deficits, renal disease, joint disease, osteoporosis, fractures, and/or CVD. Ultimately, glycemic targets should be set individually for each person with diabetes, with priority given to reducing the duration of hypoglycemia < 70 mg/dL (< 3.9 mmol/l) [39].

We have observed in recent years a considerable expansion in everyday practice of clinical situations where the use of CGM systems is warranted [48–50]. In a group of adults with T1D, clinical guidelines clearly point to the validity and benefits of beginning their use from the moment of diagnosis [49, 50]. However, it can be actively proposed that CGM systems should also be used for all individuals with diabetes currently using a glucometer, regardless of whether they are reaching their glycemic targets or not. One of the many pieces of evidence that support such recommendations is the results of the COMISAIR study in T1D, which evaluated the use of four therapeutic models: insulin pens, insulin pumps—alone and in combination—with a CGM system [51]. The outcomes of this study indicate a significant and sustained difference in glycemic control (as measured by a reduced HbA1c) in those individuals using CGM systems, thus confirming that their value in achieving glycemic targets is greater than the insulin delivery method.

This recommendation also applies to individuals with T2D on intensive insulin therapy, which is defined as at least three insulin injections per day [52]. The 2024 guidelines of the Polish Diabetes Association (PTD) emphasize that people with diabetes who are being treated with MDI or CSII should use CGM systems, which improve the safety and effectiveness of insulin therapy and enhance the convenience and quality of diabetes care [20].

The most significant application of CGM systems concerns those patient groups treated with less intensive insulin therapy models. The current joint ADA/EASD guidelines on managing hyperglycemia in patients with T2D indicate that the introduction of CGM with the initiation of insulin therapy, irrespective of its intensity, should be considered for simple insulin therapy (basal insulin 1 × daily) [53].

A number of published clinical studies, systematic reviews, and meta-analyses also indicate that the use of CGM systems is justified and can yield measurable glycemic benefits for people with T2D whether treated with insulin or on non-insulin treatment regimens [54–58]. For individuals with T2D on non-insulin therapies that incorporate sulfonylurea (SU) drugs, the risks for hypoglycemia are significant [59, 60] and use of CGM has been shown to reduce episodes of low glucose associated with SUs [61]. Wider use of CGM has been proposed for glycemic management in this group of individuals [62]. A similar case for the use of CGM in individuals on steroid therapy has also been made, with the goal of managing hyperglycemia in this group of individuals [63]. A significant observation has been that individuals with T2D being treated with GLP-1 RAs gain significant additional glycemic benefit if a CGM system is initiated alongside GLP-1 RA therapy or added in [64, 65]. Retrospective analysis show that, when initiated together, GLP-1 RA with CGM results in a greater reduction in HbA1c at 6 months for adults with T2D compared to a matched cohort initiating GLP-1 RA alone (− 2.43% vs − 2.06%, p < 0.001) [65]. When CGM is initiated in adults with T2D already established on GLP-1 RA therapy, HbA1c was reduced by − 1.5% after 6 months, compared to baseline prior to CGM initiation [64]. Reductions in HbA1c were greatest for individuals with the highest baseline HbA1c. Importantly, addition of CGM together with GLP-1 RA therapy was shown to be cost-effective for adults with T2D, using a US payer model [66].

Along with these glycemic benefits, the RELIEF retrospective studies in people with T2D in France, using the French national SNDS health system database, showed that using CGM devices is associated with considerable reductions in acute diabetes events (ADEs) that lead to hospitalizations for diabetic ketoacidosis (DKA) or severe hypoglycemia [67–70]. This includes reductions in hospital admissions for ADEs in people with T2D on intensive insulin therapy [67], on basal insulin only [68], or on non-insulin therapies [70]. Significantly, outcomes published in 2024 using data from the Swedish National Diabetes Register (NDR) and the National Patient Register (NPR) show that using CGM devices in people with T2D on either intensive or non-intensive insulin therapies is associated with significant reductions in hospital admissions for long-term cardiovascular complications of T2D, including nonfatal acute myocardial infarction and nonfatal stroke for individuals on intensive insulin therapy, and for heart failure in adults on basal insulin [71]. Clinical observations in individuals with either T1D or T2D demonstrate that CGM systems act as an educational tool—allowing the person with diabetes to understand significantly more about the impact of medication, food intake, or physical or occupational activity on glycemic changes [72, 73]. The knowledge gained allows people with diabetes to make better daily decisions, and gives them a sense of security and control over their disease, which contributes to an improved QoL, as well as a greater commitment to self-management of their condition [53]. The distribution of these outcomes across adults with T2D on intensive insulin therapy, on basal-insulin therapy, or on non-insulin therapies is shown in Fig. 2.

Fig. 2.

Evidence-based impact of using CGM technology in the care of adults with T2D. ADEs acute diabetes events, CGM continuous glucose monitoring, DSMES diabetes self-management and support

Continuous use of the systems is also recommended for people with T2D in many clinical situations that are not directly related to the treatment regimen, but rather to the type of diabetes complications they suffer (e.g., existence of high cardiovascular risk, renal or neuropathic complications, general cognitive impairment, and high risk of hypoglycemia) [37, 74, 75].

If continuous use of the CGM system is not possible, episodic use may be considered in a number of situations where a person with diabetes may experience metabolic changes, with altered glycemic control, at regular intervals [73, 76] (Table 3). Continuous monitoring of changes to glucose levels in such situations makes it possible to identify adverse changes and their causes, and to implement appropriate management quickly [76]. In all individuals, the use of CGM systems in the initial period after diagnosis of T2D seems particularly appropriate [77, 78]. Although this usually is not the moment to begin insulin therapy, the possibility of visualizing and understanding how the behavior of the person with newly diagnosed T2D impacts their glucose levels (e.g., food consumption and physical activity) creates an opportunity to permanently influence and develop behavioral habits that better promote health and thus inhibit the progression of their diabetes [76].

Table 3.

Examples of clinical situations where CGM systems can be used in the management of people with T2D

Continuous use	Episodic use
As part of insulin therapy (regardless of the treatment regimen) In persons with risk factors, e.g.,  Cardiovascular complications  Chronic kidney disease  High hypoglycemia risk  Visual impairment  Cognitive/mental disorders  Comorbidities that significantly affect glycemic control  Risk associated with occupational hazards	At diabetes diagnosis Significant changes to treatment pattern Identification of factors related to treatment or patient management that contribute to suboptimal diabetes control Acute conditions that affect glycemia The need for education/re-education Improvement of patient motivation/involvement in self-management For the purpose of risk assessment

CGM continuous glucose monitoring, T2D type 2 diabetes

Artificial Intelligence-Driven Clinical Decision Support Systems

The management of insulin and non-insulin therapy in T2D is an area that is ripe for the application of artificial intelligence-driven clinical decision support systems (AI-CDSS). Such tools are a natural development in T2D, since the management of persons with T2D is laid out in comprehensive guidelines that are driven by clear treatment algorithms [53]. Combining these decision paths with real-time CGM data on short- and long-term glycemic performance creates an opportunity for evidence-based treatment recommendations to be delivered to a person with T2D, for immediate action.

This has been directly tested in a small-scale randomized controlled trial (RCT) in T2D [79], in which an AI-CDSS was used to support basal insulin titration and dosing decisions through smart-speaker conversations, during which the person with T2D shared their most-recent insulin use and fasting plasma glucose (FPG) values. The voice-based AI-CDSS would then offer insulin dosing instructions based on the information shared. After 8 weeks, in comparison with a control group receiving usual care from their diabetes healthcare professional (HCP), the AI-CDSS users were able to optimize basal insulin doses in a mean of 15 days, compared to > 56 days for the usual care group, with 81% of AI-CDSS users achieving the target FPG compared to 25% in the usual care group. The benefits for people with T2D of using AI-CDSS are potentially large, and currently limited by the small number of available applications. Benefits to healthcare providers will be associated with reduced time spent on routine care tasks and improved confidence in daily self-management for people with T2D.

Conclusions

In view of the benefits discussed throughout, our expert opinion is summarized below:

1. Employ CGM systems on a continuous basis for all people with T2D who are on insulin therapy. The use of CGM in hypoglycemia detection and prevention is a proven attribute here.

2. Employ CGM systems episodically in special situations (e.g., at the time of diagnosis of T2D), regardless of the treatment model. The use of CGM as a behavioral modification tool can have high value in this setting.

3. While recognizing that healthcare provision is subject to regional variation, wherever possible the care of people with T2D should be coordinated within a collaborative framework between primary care teams and specialist diabetologists using available technological solutions.

4. CGM technologies (including CGM data management systems) should be introduced into specialized medical practices that provide care for people with T2D (e.g., family medicine, cardiology, nephrology, ophthalmology), in order that outcomes may be improved for people with T2D across the spectrum of comorbid conditions.

5. Introduce CGM technologies into the care of people with diabetes within a hospital setting [80, 81], subject to applicable inpatient guidelines and with appropriate training of medical teams.

6. Use available artificial intelligence (AI)-based tools in the primary care setting and in diabetes clinics (e.g., for diagnosis of early ophthalmic changes).

The use of CGM systems and emerging technologies in people with T2D is an important element in increasing the likelihood of achieving recommended glycemic control targets and of avoiding/inhibiting the development of disease complications. Published research results and the daily observations of clinicians confirm its multifactorial impact on the therapeutic process. This tool, which enables patients to optimize their therapy and achieve more effective glycemic control, while reducing the likelihood of many key risk factors in cardiovascular complications (hyperglycemia, hypoglycemia, glycemic variability, reduction of HbA1c values), can also significantly change the everyday behavior of patients, motivating them to become more involved in diabetes self-management (Table 4). The impact of using CGM for all individuals with T2D is further demonstrated by the clear association of CGM use with reduced hospital admissions for acute diabetes events and for long-term complications of diabetes, highlighted in large registry studies. Thus, CGM systems are a solution that is fully in line with the recommendations of the comprehensive and multifactorial management of patients with T2D.

Table 4.

Anticipated clinical benefits of CGM systems for people with T2D

Effective glycemic control	Treatment optimization
Reduction of HbA1c levels Increased TIR Reduction in time spent in the hyper- and hypoglycemic ranges	More complete monitoring of changes and trends in glucose concentration Adjustment of the model/dose/type of insulin therapy (intensification/simplification) In-depth periodic assessment of the glycemic change profile

Reducing risk factors for cardiovascular complications

Education/motivation/commitment

Reduction of hyperglycemia/hypoglycemia/glucose concentration variability/reduction of HbA1c Preventing the onset/progression of microvascular and macrovascular diabetic complications (such as myocardial infarction, stroke, PVD, CKD, retinopathy, inter alia) Supporting weight control

Understanding the impact of medication/meals/physical activity on changes to glucose concentration Increase sense of security and control over disease Mitigate fear of hypoglycemic events Increase patient motivation to make dietary and lifestyle changes The ability to engage in physical activity safely Improvement of cooperation between patient and doctor Improvement in QoL Possibility of maintaining/developing occupational activity Possibilities to support relatives in controlling the disease

Benefits to the healthcare system
Reductions in period of hospitalization/outpatient visits/emergency team interventions due to acute or chronic diabetes complications Reduction in costs incurred in treating acute or chronic diabetes complications Reduction in the use and cost of blood glucose meter strips Optimization of pharmacological treatment costs Possibilities of effective use of remote advice in diabetes care

CGM continuous glucose monitoring, CKD chronic kidney disease, PVD peripheral vascular disease, T2D type 2 diabetes, TIR time in range

Author Contributions

Malgorzata Mysliwiec, Leszek Czupryniak, Ryszard Gellert, Piotr Jankowski, Agnieszka Mastalerz-Migas, Marek Rekas, Krzysztof Strojek and Mieczyslaw Walczak each contributed to the concept and development of the manuscript, and were each involved in writing and reviewing serial drafts of the manuscript. This process was supported by the medical writer identified in the acknowledgements. All named authors meet the International Committee of Medical Journal Editors (ICMJE) criteria for authorship for this article, take responsibility for the integrity of the work, and have given their approval for this version to be published.

Funding

Funding for this publication was provided by Abbott Diabetes Care who supported the editorial assistance disclosed above and also the Rapid Service Fee.

Declarations

Conflict of Interest

Malgorzata Mysliwiec has received honoraria from Abbott, Ascencia, Dexcom, Medtronic. Leszek Czupryniak has received honoraria from Abbott, Abbott, Ascencia, Medtronic, Roche. Ryszard Gellert has received personal fees from AstraZeneca. Piotr Jankowski has received personal fees and travel grants from Boehringer Ingelheim. Krzysztof Strojek has received fees for activities on behalf of AstraZeneca, Medtronic, Roche and Abbott. Agnieszka Mastalerz-Migas, Marek Rekas and Mieczyslaw Walczak declare no conflicts of interest.

Ethical Approval

This article is based on previously conducted studies and does not contain any new studies with human participants or animals performed by any of the authors.