Diabetes Technology Trends: A Review of the Latest Innovations

Abstract

Context

Over the last decade, diabetes management tools such as continuous glucose monitors, automated insulin delivery systems, and connected insulin pens have experienced exponential growth. These technologies are more readily being adopted to manage diabetes due to increased availability. This mini-review provides information about recent innovations available in the United States for diabetes management to improve patient outcomes.

Evidence Acquisition

A systematic search was conducted using Medline, PubMed, ScienceDirect, and Embase databases, as well as the Cochrane Library to identify peer-reviewed articles published between 2014 and 2024, in English, and focused on treatment using technology in diabetes care.

Evidence Synthesis

Diabetes technology has significantly eased the burden of both glucose measurement and insulin delivery, which has, overall, improved diabetes management. Advancements in accuracy and glycemic outcomes have been demonstrated through rigorous clinical and observational trials, underscoring their potential to transform diabetes care. The literature suggests that the use of diabetes technologies promotes patient self-efficacy and enhances the quality of life for individuals with both type 2 and type 1 diabetes.

Conclusion

Diabetes technology has been shown to improve important aspects of diabetes care, from glycemic control to patient satisfaction and quality of life. It is important to assess the role of technology in type 1 and type 2 diabetes and individualize treatment goals and objectives.

In 2021, the International Diabetes Federation estimated that 537 million people were living with diabetes; approximately 5% to 10% had type 1 diabetes (T1D) and 90% type 2 diabetes (T2D) (1). Accurate measurement of glucose and insulin titration is crucial to optimize blood glucose management to prevent or delay diabetes complications. Today, advanced technologies are available to assist in glucose monitoring and manage diabetes more effectively, providing for improved health outcomes and quality of life (2).

Continuous glucose monitoring (CGM) systems have been commercially available since 1999 and over time have substantially improved glycemic control and enhanced the quality of life for people with diabetes (3, 4). Though first-generation CGM sensors struggled with accuracy, these devices have substantially improved over the last 2 decades (4). Accuracy is most commonly measured by the mean absolute relative difference (MARD) between the CGM trace and the precise blood glucose concentration values, and typically collected using laboratory-grade medical instruments in a hospital setting. An accurate MARD of self-monitoring of blood glucose (SMBG) by fingerstick falls within 5% to 10%, but CGM sensors failed to match this accuracy, resulting in provider hesitancy in incorporating this technology into patient care. CGM devices were approved for use with SMBG values via fingerstick to confirm the accuracy of CGM readings. Around 2014, Dexcom (Dexcom, Inc., San Diego, CA) achieved a 9% MARD with their updated algorithm in the G4 Platinum, and by 2024, the MARD of commonly prescribed CGMs ranged from 7.8% (Libre 3 sensor; Abbott, Alameda, CA (5)) to 10.6% (Medtronic Guardian 4 sensor; Northridge, CA (6)). Once these accuracy values were achieved, the FDA granted approval (4). A nonadjunctive CGM allows the patient much greater ease, convenience, less pain, and less blood waste compared with earlier adjunctive CGMs (7).

Randomized controlled trials (RCTs) have investigated the impact of CGM in multiple patient populations, including adults with T1D or T2D, children with T1D, and pregnant women with either T1D or gestational diabetes (8). While the end points vary by RCT, they have generally demonstrated that CGM use improves glucose control (defined by hemoglobin A1c [HbA1c], time in range [TIR], or both end points) compared with SMBG, and reduces the time spent in the hypoglycemic range (defined as a blood glucose <70 ßmg/dL) (8).

The advances in CGM technology and improved reliability helped smaller and safer automated insulin delivery (AID) systems be developed as a result of the integration of CGM technology and the delivery of rapid-acting insulin analogues via continuous subcutaneous insulin infusion pumps dictated by proprietor-specific algorithms. Today, each system has a unique algorithm that utilizes CGM-derived glucose values to automatically adjust insulin delivery through the insulin pump, including adjusting basal rates and insulin suspension and applying the sensitivity factor when corrective insulin is needed (9). AID systems have emerged as the most effective technological advancements for optimizing glucose control, and have significantly improved glycemic management for patients with T1D (10). There has been a significant surge in AID use in recent years, with numerous options available (11). In parallel, connected insulin pens, also known as smart pens, became available as an innovative technology, expanding diabetes treatment options for patients with T1D or T2D in optimizing management (12).

This mini-review highlights the evolution of diabetes technologies over time, describes the current landscape of diabetes technology, provides salient information about the impact of CGM and AID systems on glycemic control and quality of life in patients with T1D and T2D, and highlights disparities in diabetes technology utilization. Given its capacity to improve patient outcomes, this article aims to be a resource for health care practitioners in prescribing and implementing diabetes technology in practice.

Methods

A systematic search was conducted using Medline, PubMed, ScienceDirect, and Embase databases, as well as the Cochrane Library to identify published literature. The search was inclusive of articles published from January 2014 through March 2024. Keywords for the search included “T2D and glucose sensing” or “T2D and continuous glucose monitoring” or “T2D and CGM,” “T1D and glucose sensing” or “T1D continuous glucose monitoring” or “T1D and CGM,” “T1D and automated insulin delivery” or “T1D and AID” or “T1D integrated insulin pumps,” “T2D and automated insulin delivery” or “T2D and AID” or “T2D integrated insulin pumps,” “connected insulin pens and T1D” or “smart pens and T1D,” “connected insulin pens and T2D” or “smart pens and T2D,” “CGM and quality of life” or “continuous glucose monitoring and quality of life,” “automated insulin delivery and quality of life” or “AID and quality of life,” and “diabetes technology and disparities.” Articles were peer-reviewed, in English, and focused on treatment using technology in diabetes care.

CGM Mechanics and Utilization

CGM is one of the most clinically impactful recent advancements in diabetes care. The American Diabetes Association (2) and other professional societies, including the Endocrine Society, the International Society of Pediatric and Adolescent Diabetes (13), and the American Association of Clinical Endocrinology recommend CGM as the preferred method of care for patients with T1D due to the reduction in HbA1c levels and hypoglycemia. Similarly, the American Diabetes Association (2), the Endocrine Society, and the American Association of Clinical Endocrinology have emphasized that patients with T2D on insulin or not on insulin would benefit from CGM technology (14). Ultimately, CGM offers a convenient glucose monitoring method, allowing individuals with both T1D and T2D to track their glucose with minimal disruption in their daily routines (15).

CGM is a minimally invasive technology that most commonly relies on a glucose-oxidase–doped platinum electrode deposited on a needle, which is then inserted into the subcutaneous tissue. The glucose oxidase reaction conducted on the electrode produces gluconolactone, hydrogen peroxide, and an electrical current signal that is transformed into glucose reading. A calibration process that takes a few SMBG samples collected by the patient as a cross-reference is required by some (4). The CGM sensors deliver blood glucose readings every 1 to 5 minutes, providing an essentially continuous trace of blood glucose measurements that offers insights into glucose fluctuations that are not evident when monitoring SMBG 3 to 4 times daily or in HbA1c measurements.

When paired with dedicated CGM receivers or smart devices such as smartphones, smartwatches, or tablet computers, CGM devices directly display real-time, dynamic glucose readings transferred from the CGM via Bluetooth technology. This allows clinicians to download the data and generate an ambulatory glucose profile report, which includes graphics and statistics that provide insight into the patient's glucose patterns. Often, clinicians can jointly review the CGM report with the patient, enhancing shared decision-making and patient empowerment.

Today, the MARD of the most commonly used CGMs is equivalent or nearly equivalent to SMBG, leading to nonadjunctive labeling for different iterations of the Dexcom and Libre devices (16, 17). This has led to a significant expansion in CGM use. For instance, from 2010 to 2012 only 7% of the T1D Exchange Registry participants used CGM sensors (18). Yet in 2018, approximately 30% of patients with T1D in the T1D Exchange Registry were using CGMs, with individual studies demonstrating as high as 48% use for their patient cohort (19). The prevalence of CGM use among patients with T2D on a large scale is difficult to quantify because there is no registry for this patient population. However, we can look to smaller studies to gain insight into utilization over time in the T2D population. Pathak et al (20) evaluated the prevalence of CGM use in a T2D population by comparing CGM use before and after a commercial insurer expanded coverage of the device through a pharmacy benefit. Prior to coverage expansion, only 1.2% of T2D patients used a CGM and by the end of the study period (2016-2020), 14.9% of T2D patients were using a CGM.

While CGM sensors have demonstrated benefits for diabetes patients, barriers to use still exist. In a real-world observational study of patients in the T1D Exchange Registry explored these disparities and found that adult patients were more likely to be CGM users than pediatric patients (19). Race/ethnicity was also a variable in differential CGM use with non-Hispanic White patients having a higher rate of CGM use (49.5%) than non-Hispanic Black (17.7%) or Hispanic (38.4%) patients. Another variable was insurance type in which 57.2% of patients with private insurance used a CGM as opposed to 33.3% of patients with public insurance, including Medicaid and Medicare (19).

CGM and Quality of Life Measures

Speight et al (21), perhaps, said it best, “it is now appreciated that ‘adding life to years’ is as important to many people as ‘adding years to life.’” Diabetes distress results from an elevated mental load associated with the everyday management of diabetes, the cognitive and psychological capacities of diabetes patients to cope with these demands, and their concerns about their disease and its consequences (22). In 2017, the DIAMOND RCT published their findings from 158 patients with T1D randomized to CGM or 4× daily SMBG monitoring (23). Both the control and intervention groups were assessed with 3 diabetes-specific quality of life surveys: Diabetes Distress Scale, Hypoglycemia Fear Survey, and Hypoglycemic Confidence Scale. The investigators found that the CGM group reported significantly higher hypoglycemia confidence and significantly lower perceived diabetes-related distress than their SMBG counterparts (23). Similar findings were reported in another trial in which patients assigned CGMs reported improved general emotional well-being, as defined by higher scores on the Diabetes Treatment Satisfaction Questionnaire and World Health Organization-5 Well-Being Index and confidence in managing their hypoglycemia at 6 months compared with the SMBG arm (24).

These benefits are also seen in medically complex patient populations. In the CONCEPTT trial, a multinational study of pregnant patients with T1D and those planning pregnancy, 325 participants were randomized to CGM plus SMBG or SMBG alone (25). Participants in the CGM group demonstrated a higher degree of satisfaction over time with glucose monitoring compared to the SMBG group. In addition, the CGM group engaged in less hypoglycemia-avoidant behaviors over time, while behaviors in the SMBG group remained unchanged (25). This finding is important because strict glycemic control is paramount during pregnancy to avoid complications for both the mother and fetus.

Lastly, the data sharing capability with CGM systems is another feature that has improved the quality of life of patients with diabetes. A sample of adults with T1D (N = 302) were surveyed to assess the impact of sharing data in real-time from a Dexcom G5 or G6 CGM system with a support person (26). Almost 90% of participants said that data sharing contributed to improved hypoglycemic confidence. Participants also expressed that the data sharing capabilities resulted in improved overall well-being and reduced diabetes distress.

CGM and Improved Glycemic Outcomes in T1D and T2D

The nearly instantaneous blood glucose feedback delivered by CGM devices helps patients with insulin dose decision making. This includes adjusting a dose for physical activity and for dietary intake, enabling more accurate, personalized glucose management strategies that have resulted in improved glycemic control and diabetes-related outcomes in patients with both T1D and T2D.

Recent RCTs in patients with T2D receiving intensive insulin therapy have demonstrated the efficacy and safety of CGM use in lowering glycated hemoglobin levels without increasing the risk of hypoglycemia (27). A recent open-label RCT conducted across 26 European diabetes centers examined the impact of CGM technology as a substitute for SMBG in individuals with T2D using intensive insulin therapy (28). The study concluded that CGM technology was a safe, effective replacement for SMBG and demonstrated statistically significant decreases in time in hypoglycemia (50%) and nocturnal hypoglycemia (52%) at 12 months compared with baseline. Similarly, a meta-analysis of 12 RCTs of CGM use vs SMBG in adults with T2D found that the CGM group improved glycemic control regardless of whether they were treated with insulin or with oral agents alone (29).

CGMs also assist patients with T1D in making more precise insulin-dosing decisions (30), reducing the risk of hypoglycemia and enhancing overall glucose management (10). It also allows them to make informed decisions about food intake, physical activity, and medication (31). A study by Bolinder et al (32) demonstrated that a factory-calibrated CGM system reduced hypoglycemia by 38% in patients with T1D. In another recent investigation, adolescents and young adults with T1D, using either multiple daily injections or insulin pump therapy, were recruited for an 8-week, open-label, randomized, crossover clinical trial (33). Participants (n = 29) were randomly assigned to monitor their daily glucose levels using either the Dexcom G6 CGM system or SMBG. The primary outcome, time in target glucose range (70-180 mg/dL) as recorded by CGM, was significantly higher during the CGM period than in the control period. The use of CGM also reduced mean sensor glucose and time spent above the target range. Moreover, HbA1c levels were reduced by an average of 3% (8.9 mmol/mol) in the CGM period (33).

Finally, a retrospective analysis of patients with T1D (n = 131) or T2D (n = 176) using the flash FreeStyle Libre system observed a significant reduction in HbA1c levels within 3 months, with reductions sustained for 12 months (34). A significant benefit of the FreeStyle Libre system was reported in patients with a starting HbA1c greater than 7.5%.

Other Benefits Associated With CGM Use

Hannah et al (35) performed a retrospective analysis of United States health care claims data. Results demonstrated an association between the initiation of the Dexcom G6 CGM system for individuals with T2D using intensive insulin therapy and reductions in the proportion of individuals with more than 1 diabetes-related emergency department visit (30% decrease) and inpatient admission (41.5% decrease), and reductions in costs related to health care resource utilization ($341/person/month). The authors concluded that broader use of CGM could amplify both patient care benefits and decreased costs.

In the CONCEPTT trial, the pregnant patients and those planning for pregnancy with T1D who were randomized to the CGM plus SMBG arm achieved a lower HbA1c than the SMBG only arm (25). Arguably, the most impactful finding from this RCT is the subgroup of pregnant women in the CGM arm who delivered babies with fewer health complications. This was defined as a decreased incidence of large for gestational age infants, fewer neonatal intensive care admissions for >24 hours, fewer incidences of neonatal hypoglycemia necessitating treatment with dextrose, and reduced total length of hospital stay (25). Of note, there was no difference between the pregnant women in the control and intervention groups with respect to hypertensive disorders, preeclampsia, cesarean delivery, gestational age, or preterm delivery.

Currently Available CGMs in the United States

Table 1 provides a comprehensive overview of the latest CGM technology products. It includes key product details such as brand name and model, warm-up time or duration required for the sensor to become operational after insertion, and wear time. It also indicates the CGM's compatibility with specific software for data analysis and visualization, as well as its integration capabilities with AID systems and connected pens.

AID Systems and Quality of Life

Advances and broader availability of CGM has enabled the advances in AID systems that revolutionized diabetes management. The most advanced systems have an algorithm that uses glucose values derived from a CGM and automatically adjusts insulin delivery through an insulin pump (9). A 2015 review article of patients with T1D found that 20% to 30% of participants reported severe diabetes distress that impacted their management of diabetes, glycemic control, and quality of life (36). A research group in France postulated that hybrid closed-loop (HCL) AID systems would improve the quality of life of patients with T1D because it would automate many of the daily tasks required for diabetes management (22). This multicenter study assessed participants prior to initiating an HCL system and at 3- and 6-month follow-up visits postimplementation. Assessment tools evaluated diabetes distress, quality of life, stress, anxiety, depression, fear of hypoglycemia (behavior and worry scales), physical activity, sleep, and fatigue; diabetes distress and quality of life were primary outcomes. At baseline 64% of adults reported severe diabetes distress, which decreased to 50% at 3 months of follow-up and remained 50% at 6 months. The quality of life global score significantly improved between baseline and 3 months postimplementation and remained stable at 6 months (22).

Other psychosocial benefits of AID devices included reduced anxiety, improved sleep secondary to improved overnight glucose control, less restrictive eating habits, and “time off” from the stress/responsibilities of diabetes care (37). Furthermore, when asked, most patients in AID studies stated that they would continue using this technology or recommend it to others. These findings are important because the patient's perspective must also be considered, given that diabetes is a chronic disease that impacts every aspect of their life. While it is well-reported that AID systems improve glycemic control and limit hypoglycemia, as clinicians, if we can pass on the knowledge that patients like these devices and mitigate diabetes distress, perhaps we can “add life to years.”

Unfortunately, these benefits are not equally accessible across the diabetes patient population. In a sample of 109 pediatric and adult T1D providers practicing in the United States, both racial/ethnic-mediated bias and insurance-mediated bias influenced whether providers prescribed a diabetes technology for both CGM and AID (38). Disparities in technology utilization and access are not unique to the United States. In a retrospective observational analysis of 1631 adult patients with T1D in the United Kingdom, patients were evaluated according to socioeconomic deprivation, using the Index of Multiple Deprivation 2019 (scale: decile 1, the most deprived 10% of neighborhoods in England to decile 10, the least deprived 10%) (39). The researchers used the index as a frame of reference to divide the data into quintiles. Not surprisingly, they found that significantly fewer people used diabetes technology in the most deprived quintile than in the least deprived quintile. The greatest technology disparity was observed with continuous subcutaneous insulin infusion, in which only 16% of the most deprived quintile used pumps, as opposed to 32% of the least deprived quintile (P < .001) (39).

AID Systems and Insulin Pump Technology

The first AID system was developed in 1963 comprised an intravenous glucose monitor and 2 intravenous syringe pumps: (1) insulin for hyperglycemia and (2) glucagon for hypoglycemia (9). This system never made it to market because it was the size of a large backpack. The first commercial AID system was developed in the 1970s, but the size and complexity of the machine limited its use to the inpatient setting. Throughout the 2000s, increasingly smaller, accurate, and reliable continuous subcutaneous insulin infusion pumps and interstitial CGM systems were developed which allowed for regular use of AID systems for patients with diabetes. At the time, the most advanced AID technologies commercially available were considered HCL systems, where the algorithm independently adjusted basal rates and suspended insulin depending on the CGM readings, but users were still required to bolus for prandial insulin (9).

Recent clinical trials demonstrate that AID systems are safe and help patients with T1D significantly decrease HbA1c levels, maintain enhancements in glycemic levels, and increase time in target glucose range of 70 to 180 mg/dL with very low occurrence of hypoglycemia (40, 41). Beck et al (42) published a meta-analysis of RCT outcomes among 369 patients with T1D ranging from 2 to 72 years old using the Tandem Control IQ AID system, which showed beneficial effects on glycemic control in all patient groups. Studies assessing glycemic outcomes of AID software systems from Medtronic and Insulet have shown similar trends in T1D (41, 43-45).

A significant area of current research is the development of a closed-loop AID system, also called an artificial pancreas. This system would have feed-forward insulin for meals, so the user need not announce food, and would have a precise understanding of the rate of change of blood glucose, so exercise, missed meals, etc., are also accounted for algorithmically. Deshpande et al (46) reported that the use of continuous adaptation scheme in predictive modeling resulted in consistent improvement across the entire glucose range without increasing the risk of hypoglycemia. Specifically, for the scenario with induced insulin resistance combined with no bolus before the meal, the percent time spent in the target range of 70 to 180 mg/dL was higher (53.5% vs 48.9%), and the percent time in the tighter range of 70 to 140 mg/dL overnight was also significantly improved (70.9% vs 52.9%) (46). Similarly, in a recent study, participants used an interoperable artificial pancreas system and an algorithm on an unlocked smartphone for 48 hours (47). During the AID period in this study, the percent time in closed-loop mode and connected to the CGM was 92.7% and 99.6%, respectively. Investigations revealed that participants and their parents/guardians were satisfied with the automated system. In the same framework, Aiello et al (48) presented the design of an Artificial Pancreas with Data-Driven Learning of Multi-Step-Ahead Blood Glucose Predictors. The approach they developed provided accurate forecasts of future glucose and demonstrated good closed-loop performance (9). While closed-loop technology is not yet available from a licensed manufacturer, it is an area of active research and development.

Several new AID systems are expected to come to market in the coming years with a target glucose level as low as 87 mg/dL, features that allow more adjustments for exercise, boluses based on food type to include fats and proteins, as well as offering the systems through the pharmacy benefits in the United States. This may provide greater opportunity for use by more people with diabetes.

Innovative Connected Pens

Insulin pens have been studied for their accuracy, convenience, and potential for dose titration. Over the past decade, this technology has evolved with the addition of new features such as Bluetooth connectivity, bolus dose calculators, and integration with mobile apps and CGMs. Connected pens are aesthetically similar to traditional insulin pens but use Bluetooth technology to allow data related to insulin injections and resulting glucose values to be recorded and stored. This data can be analyzed for insulin dose adjustments by the individual with diabetes and the provider. Sy et al (49) found that connected pens have the potential to improve medication adherence and glycemic control and help address other barriers to effective diabetes management, including calculation of carbohydrate and corrective insulin dosing. Additionally, these devices can provide valuable insights regarding the interaction of insulin, food intake, and physical activity, ultimately enabling optimization of insulin regimens for individuals with diabetes (49).

A retrospective cohort analysis was done using real-world data collected from the InPen (Medtronic) system used by individuals with T1D or T2D (50). In the overall population and the T1D population, taking insulin at least 3 times/day and missing <20% of the doses were associated with better glycemic control. Similarly, in adults with T2D, missing <20% of the prescribed doses was the significant factor that determined improved glycemic outcomes (50). An observational study found that reliable insulin dose data from the NovoPen (Novo Nordisk, Plainsboro Township, NJ) connected pens significantly contributed to increased TIR, significantly decreased time spent in level 1 hypoglycemia, and bolus insulin dosing was significant compared with baseline for 94 individuals with T1D (51).

Despite the benefits offered by connected pens, few insulin users currently employ these devices, attributable to various factors such as limited awareness among health care professionals, inadequate initial training for prescribers, barriers to technology access, inadequate insurance coverage, and challenges in device setup (52). With the increasing digitization of health care systems, connected insulin pens have the potential to facilitate a data-driven approach to diabetes management for individuals who are not using insulin pumps or AID systems (53).

Clinical trials evaluating innovative technologies in AID, connected insulin pens, and insulin pumps have yielded encouraging outcomes in diabetes management. Table 2 provides an overview of a few selected trials conducted in the last few years (51, 54-60). These trials have shown improvements in glucose levels, TIR, HbA1c levels, and reduction in missed bolus doses (MBD) in individuals across various age groups when they utilized technology for diabetes management.

Conclusion

Since their initial iterations, significant strides have been made in improving glucose sensing mechanisms, AID systems, and integrated pens. These improvements have been designed to simplify daily diabetes management and improve the quality of life of diabetes patients. These devices have been largely successful on both fronts. Current guidelines issued by professional societies endorse consideration of these devices when clinically appropriate (2). In particular, advancements in CGM and AID systems have improved glycemic outcomes and quality of life for many individuals.

Despite recommendations by professional societies and ample data supporting improvements in diabetes care offered by these technologies, they remain underprescribed despite their benefits on diabetes-related outcomes, particularly in glycemic control and hypoglycemia and the benefits in patient's quality of life. Technology will continue to change and improve. Thus, the authors encourage readers to consider these options for each patient.

Acknowledgments

The authors thank Christine G. Holzmueller, MS, for contributing to the editing of the manuscript.

Supplement Sponsorship

This article appears as part of the supplement “Guidebook for Providers on Comprehensive Diabetes Care,” sponsored by the University Hospitals Mary B. Lee Chair in Adult Endocrinology endowment, and by the Ratner Family Fund.

Funding

B.H., Deborah and Ronald Ratner Fund and University Hospitals (awarded the Mary B. Lee Chair in Endocrinology).

Disclosures

B.H. received support as the study site PI (08/31/23-08/31/26) from, Jaeb Center for Health Research, Inc.; and support as the study site PI (04/18/2024-04/30/2027) from Diasome Pharmaceuticals. N.B. consults for Abbott Diabetes, Embecta, Sanofi Diabetes, and Ascensia Diabetes. She has received grant support from Tandem Diabetes (1/2023-4/2024). E.L.L., E.R., and C.L. have nothing to declare.

Data Availability

Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.

References

Abbreviations

 
• AID

  automated insulin delivery

 
• CGM

  continuous glucose monitoring

 
• HCL

  hybrid closed-loop

 
• HbA1c

  hemoglobin A1c

 
• MARD

  mean absolute relative difference

 
• RCT

  randomized controlled trial

 
• SMBG

  self-monitoring of blood glucose

 
• TIR

  time in range

 
• T1D

  type 1 diabetes

 
• T2D

  type 2 diabetes