Real-world evidence on clinical outcomes of people with type 1 diabetes using open-source and commercial automated insulin dosing systems: A systematic review

Artificial intelligence in the care of children and adolescents with chronic diseases: a systematic review

The integration of artificial intelligence (AI) and machine learning (ML) has shown potential for various applications in the medical field, particularly for diagnosing and managing chronic diseases among children and adolescents. This systematic review aims to comprehensively analyze and synthesize research on the use of AI for monitoring, guiding, and assisting pediatric patients with chronic diseases. Five major electronic databases were searched (Medline, Scopus, PsycINFO, ACM, Web of Science), along with manual searches of gray literature, personal archives, and reference lists of relevant papers. All original studies as well as conference abstracts and proceedings, focusing on AI applications for pediatric chronic disease care were included. Thirty-one studies met the inclusion criteria. We extracted AI method used, study design, population, intervention, and main results. Two researchers independently extracted data and resolved discrepancies through discussion. AI applications are diverse, encompassing, e.g., disease classification, outcome prediction, or decision support. AI generally performed well, though most models were tested on retrospective data. AI-based tools have shown promise in mental health analysis, e.g., by using speech sampling or social media data to predict therapy outcomes for various chronic conditions.

CONCLUSIONS

While AI holds potential in pediatric chronic disease care, most reviewed studies are small-scale research projects. Prospective clinical implementations are needed to validate its effectiveness in real-world scenarios. Ethical considerations, cultural influences, and stakeholder attitudes should be integrated into future research.

WHAT IS KNOWN

• Artificial Intelligence (AI) will play a more dominant role in medicine and healthcare in the future and many applications are already being developed.

WHAT IS NEW

• Our review provides an overview on how AI-driven systems might be able to support children and adolescents with chronic illnesses. • While many applications are being researched, few have been tested on real-world, prospective, clinical data.

Real-world efficacy and safety of open-source automated insulin delivery for people with type 1 diabetes mellitus: Experience from mainland China

BACKGROUND

Open-source automated insulin delivery systems are increasingly adopted yet predominantly discussed outside of Asia. This study aimed to describe efficacy and safety of android artificial pancreas (AAPS) in people with type 1 diabetes mellitus (T1DM) from mainland China.

METHODS

This real-world study recruited people who initiated AAPS for ≥ 3 months between 2019 and 2024. Key outcomes included glycated hemoglobin A1c (HbA1c) and metrics from continuous glucose monitoring, rates of diabetic ketoacidosis (DKA) and severe hypoglycemia.

FINDINGS

292 (male, 46·9 %) participants aged 25·7 (14·7, 35·0) years were included, with 183 (62·7 %) and 68 (23·3 %) using AAPS for 6 and 12 months. Prior-AAPS HbA1c was 7·6 ± 1·7 % with 44·5 % achieving < 7·0 %. After 3 months, mean HbA1c improved by -1·5 ± 2·0 % to 6·3 ± 0·8 % (P < 0.01), with 82·9 % achieving < 7.0 %. Time in range 3·9-10·0 mmol/L (TIR) improved to 78·8 ± 12·9 %, with 80·5 % achieving > 70 %, followed by time below 3·9 mmol/L of 3·9 (2·1, 6·1) %. After 12 months, HbA1c and TIR remained similar at 6·4 ± 1·0 % and 77·9 ± 12·2 %. No DKA and severe hypoglycemia was observed.

INTERPRETATION

Real-world data from mainland China highlights current uptake of open-source AAPS with potential glycemic benefits. No safety signals are seen. More support to enhance access and utilization of all AID systems in this region is warranted.

Automating insulin delivery through pump and continuous glucose monitoring connectivity: Maximizing opportunities to improve outcomes

The development of automated insulin delivery (AID) systems, which connect continuous glucose monitoring (CGM) systems with algorithmic insulin delivery from an insulin pump (continuous subcutaneous insulin infusion, [CSII]), has led to improved glycaemia and quality of life benefits in those with insulin-treated diabetes. This review summarizes the benefits gained by the connectivity between insulin pumps and CGM devices. It details the technical requirements and advances that have enabled this, and highlights the clinical and user benefits of such systems. Clinical trials and real-world outcomes from the use of AID systems in people with type 1 diabetes (T1D) will be the focus of this article; outcomes in people with type 2 diabetes (T2D) and other diabetes subtypes will also be discussed. We also detail the limitations of current technological approaches for connectivity between insulin pumps and CGM devices. While recognizing the barriers, we discuss opportunities for the future.

Beyond Expected Patterns in Insulin Needs of People With Type 1 Diabetes: Temporal Analysis of Automated Insulin Delivery Data

Background

Type 1 diabetes (T1D) is a chronic condition in which the body produces too little insulin, a hormone needed to regulate blood glucose. Various factors such as carbohydrates, exercise, and hormones impact insulin needs. Beyond carbohydrates, most factors remain underexplored. Regulating insulin is a complex control task that can go wrong and cause blood glucose levels to fall outside a range that protects people from adverse health effects. Automated insulin delivery (AID) has been shown to maintain blood glucose levels within a narrow range. Beyond clinical outcomes, data from AID systems are little researched; such systems can provide data-driven insights to improve the understanding and treatment of T1D.

Objective

The aim is to discover unexpected temporal patterns in insulin needs and to analyze how frequently these occur. Unexpected patterns are situations where increased insulin does not result in lower glucose or where increased carbohydrate intake does not raise glucose levels. Such situations suggest that factors beyond carbohydrates influence insulin needs.

Methods

We analyzed time series data on insulin on board (IOB), carbohydrates on board (COB), and interstitial glucose (IG) from 29 participants using the OpenAPS AID system. Pattern frequency in hours, days (grouped via k-means clustering), weekdays, and months were determined by comparing the 95% CI of the mean differences between temporal units. Associations between pattern frequency and demographic variables were examined. Significant differences in IOB, COB, and IG across temporal dichotomies were assessed using Mann-Whitney U tests. Effect sizes and Euclidean distances between variables were calculated. Finally, the forecastability of IOB, COB, and IG for the clustered days was analyzed using Granger causality.

Results

On average, 13.5 participants had unexpected patterns and 9.9 had expected patterns. The patterns were more pronounced (d>0.94) when comparing hours of the day and similar days than when comparing days of the week or months (0.3

Conclusions

Our study shows that unexpected patterns in the insulin needs of people with T1D are as common as expected patterns. Unexpected patterns cannot be explained by carbohydrates alone. Our results highlight the complexity of glucose regulation and emphasize the need for personalized treatment approaches. Further research is needed to identify and quantify the factors that cause these patterns.

Collapse

Key Words

• OpenAPS
• clustering
• data-driven
• insulin needs
• k-means
• machine learning
• multivariate time series
• open dataset
• temporal patterns
• type 1 diabetes

Collapse

MESH Headings Collapse

Grants Collapse

Affiliation(s)

Isabella Degen Interactive Artificial Intelligence Centre for Doctoral Training, School of Computer Science, Faculty of Science and Engineering, University of Bristol, 1 Cathedral Square, College Green, Bristol, BS1 5DD, United Kingdom, 44 7726100905
Kate Robson Brown University College Dublin President's Office, College of Engineering and Architecture, University College Dublin, Dublin, Ireland
Henry W J Reeve School of Mathematics, Faculty of Science and Engineering, University of Bristol, Bristol, United Kingdom
Zahraa S Abdallah School of Engineering Mathematics and Technology, Faculty of Science and Engineering, University of Bristol, Bristol, United Kingdom

Collapse

Collaborators Collapse

Challenges of Glycemic Control in People With Diabetes and Advanced Kidney Disease and the Potential of Automated Insulin Delivery

Diabetes is the leading cause of chronic kidney disease (CKD) and end-stage kidney disease in the world. It is known that maintaining optimal glycemic control can slow the progression of CKD. However, the failing kidney impacts glucose and insulin metabolism and contributes to increased glucose variability. Conventional methods of insulin delivery are not well equipped to adapt to this increased glycemic lability. Automated insulin delivery (AID) has been established as an effective treatment in patients with type 1 diabetes mellitus, and there is emerging evidence for their use in type 2 diabetes mellitus. However, few studies have examined their role in diabetes with concurrent advanced CKD. We discuss the potential benefits and challenges of AID use in patients with diabetes and advanced CKD, including those on dialysis.

Safety and Psychological Outcomes of Tandem t:Slim X2 Insulin Pump with Control-IQ Technology in Children, Adolescents, and Young Adults with Type 1 Diabetes: A Systematic Review

The Tandem t:slim X2 insulin pump is a second-generation automated insulin delivery system with Control-IQ technology. It consists of an X2 insulin pump, an integrated Dexcom sensor, and an embedded 'Control-IQ' algorithm, which predicts glucose levels 30 min in the future, adapting the programmed basal insulin rates to get glucose levels between 112.5 and 160 mg/dl (8.9 mmol/l). The system delivers automatic correction boluses of insulin when glucose levels are predicted to rise > 180 mg/dl (10 mmol/l). It has been commercially available since 2016. We reviewed the current evidence about the psychological, safety, and exercise-related outcomes of this device in children, adolescents, and young adults living with type 1 diabetes. We screened 552 papers, but only 21 manuscripts were included in this review. Fear of hypoglycemia is significantly reduced in young people with diabetes and their parents. Interestingly, diabetes-related distress is decreased; thus, the system is well accepted by the users. The sleeping quality of subjects living with diabetes and their caregivers is improved to a lesser extent as well. Despite the small number of data, this system is associated with a low rate of exercise-related hypoglycemia. Finally, evidence from the literature shows that this system is safe and effective in improving psychological personal outcomes. Even if further steps toward the fully closed loop are still mandatory, this second-generation automated insulin delivery system reduces the burden of diabetes. It properly addresses most psychological issues in children, adolescents, and young adults with type 1 diabetes mellitus; thus, it appears to be well accepted.

Refining Insulin on Board with netIOB for Automated Insulin Delivery

Automated insulin delivery (AID) systems enhance glucose management by lowering mean glucose level, reducing hyperglycemia, and minimizing hypoglycemia. One feature of most AID systems is that they allow the user to view "insulin on board" (IOB) to help confirm a recent bolus and limit insulin stacking. This metric, along with viewing glucose concentrations from a continuous glucose monitoring system, helps the user understand bolus insulin action and the future "threat" of hypoglycemia. However, the current presentation of IOB in AID systems can be misleading, as it does not reflect true insulin action or automatic, dynamic insulin adjustments. This commentary examines the evolution of IOB from a bolus-specific metric to its contemporary use in AID systems, highlighting its limitations in capturing real-time insulin modulation during varying physiological states.

Real-World Evidence of Omnipod® 5 Automated Insulin Delivery System Use in 69,902 People with Type 1 Diabetes

Background: The Omnipod® 5 Automated Insulin Delivery System was associated with favorable glycemic outcomes for people with type 1 diabetes (T1D) in two pivotal clinical trials. Real-world evidence is needed to explore effectiveness in nonstudy conditions. Methods: A retrospective analysis of the United States Omnipod 5 System users (aged ≥2 years) with T1D and sufficient data (≥90 days of data; ≥75% of days with ≥220 continuous glucose monitor readings/day) available in Insulet Corporation's device and person-reported datasets as of July 2023 was performed. Target glucose setting usage (i.e., 110-150 mg/dL in 10 mg/dL increments) was summarized and glycemic outcomes were examined. Subgroup analyses of those using the lowest average glucose target (110 mg/dL) and stratification by baseline characteristics (e.g., age, prior therapy, health insurance coverage) were conducted. Results: In total, 69,902 users were included. Multiple and higher glucose targets were more commonly used in younger age groups. Median percentage of time in range (TIR; 70-180 mg/dL) was 68.8%, 61.3%, and 53.6% for users with average glucose targets of 110, 120, and 130-150 mg/dL, respectively, with minimal time <70 mg/dL (all median <1.13%). Among those with an average glucose target of 110 mg/dL (n = 37,640), median TIR was 65.0% in children and adolescents (2-17 years) and 69.9% in adults (≥18 years). Subgroup analyses of users transitioning from Omnipod DASH or multiple daily injections and of Medicaid/Medicare users demonstrated favorable glycemic outcomes among these groups. Conclusion: These glycemic outcomes from a large and diverse sample of nearly 70,000 children and adults demonstrate effective use of the Omnipod 5 System under real-world conditions.

[Position paper: Open-source technology in the treatment of people living with diabetes mellitus-an Austrian perspective : Technology Committee of the Austrian Diabetes Association]

People living with diabetes mellitus can be supported in the daily management by diabetes technology with automated insulin delivery (AID) systems to reduce the risk of hypoglycemia and improve glycemic control as well as the quality of life. Due to barriers in the availability of AID-systems, the use and development of open-source AID-systems have internationally increased. This technology provides a necessary alternative to commercially available products, especially when approved systems are inaccessible or insufficiently adapted to the specific needs of the users. Open-source technology is characterized by worldwide free availability of codes on the internet, is not officially approved and therefore the use is on the individual's own responsibility. In the clinical practice a lack of expertise with open-source AID technology and concerns about legal consequences, lead to conflict situations for health-care professionals (HCP), sometimes resulting in the refusal of care of people living with diabetes mellitus. This position paper provides an overview of the available evidence and practical guidance for HCP to minimize uncertainties and barriers. People living with diabetes mellitus must continue to be supported in education and diabetes management, independent of the chosen diabetes technology including open-source technology. Check-ups of the metabolic control, acute and chronic complications and screening for diabetes-related diseases are necessary and should be regularly carried out, regardless of the chosen AID-system and by a multidisciplinary team with appropriate expertise.

Enhancing equity in access to automated insulin delivery systems in an ethnically and socioeconomically diverse group of children with type 1 diabetes

INTRODUCTION

Manufacturer-supported didactic teaching programmes offer effective automated insulin delivery (AID) systems onboarding in children and young people (CYP) with type 1 diabetes (T1D). However, this approach has limited flexibility to accommodate the needs of families requiring additional support.

RESEARCH DESIGN AND METHODS

Evaluate the efficacy of an inperson manufacturer-supported didactic teaching programme (Group A), in comparison to a flexible flipped learning approach delivered virtually or inperson (Group B). Retrospective analysis of CYP with T1D using continuous glucose monitoring (CGM), who were initiated on AID systems between 2021 and 2023. Compare CGM metrics from baseline to 90 days for both groups A and B. Additionally, compare the two groups for change in CGM metrics over the 90-day period (∆), patient demographics and onboarding time.

RESULTS

Group A consisted of 74 CYP (53% male) with median age of 13.9 years and Group B 91 CYP (54% male) with median age of 12.7 years. From baseline to 90 days, Group A lowered mean (±SD) time above range (TAR, >10.0 mmol/L) from 47.6% (±15.0) to 33.2% (±15.0) (p<0.001), increased time in range (TIR, 3.9-10.0 mmol/L) from 50.4% (±14.0) to 64.7% (±10.2) (p<0.001). From baseline to 90 days, Group B lowered TAR from 51.3% (±15.1) to 34.5% (±11.3) (p<0.001) and increased TIR from 46.5% (±14.5) to 63.7% (±11.0) (p<0.001). There was no difference from baseline to 90 days for time below range (TBR, <3.9 mmol/L) for Group A and Group B. ∆ TAR, TIR and TBR for both groups were comparable. Group B consisted of CYP with higher socioeconomic deprivation, greater ethnic diversity and lower carer education achievement (p<0.05). The majority of Group B (n=79, 87%) chose virtual flipped learning, halving diabetes educator time and increasing onboarding cadence by fivefold.

CONCLUSIONS

A flexible virtual flipped learning programme increases onboarding cadence and capacity to offer equitable AID system onboarding.

Comparative Effectiveness of Hybrid Closed-Loop Automated Insulin Delivery Systems Among Patients with Type 1 Diabetes

BACKGROUND

Clinical trials have demonstrated the efficacy and safety of hybrid closed-loop (HCL) systems, yet few studies have compared outcomes in the real-world setting.

METHOD

This retrospective study analyzed patients from an academic endocrinology practice between January 1, 2018, and November 18, 2022. The inclusion criteria were diagnosis code for type I diabetes (T1D), >18 years of age, new to any HCL system [Medtronic 670G/770G (MT), Tandem Control IQ (CIQ), or Omnipod 5 (OP5)], and availability of a pump download within three months. The outcomes included %time in range (TIR) of 70 to 180 mg/dL, %time below range (TBR) <70 mg/dL at 90 days, and HbA1c for 91 to 180 days.

RESULT

Of the 176 participants, 47 were MT, 74 CIQ, and 55 OP5. Median (25%, 75%) change in HbA1c was -0.1 (-0.8, 0.3), -0.6 (-1.1, -0.15), and -0.55 (-0.98, 0)% for MT, CIQ, and OP5, respectively, (P = .04). TIR was 70 (57, 76), 67 (59, 75), and 68 (60, 76)% (P = .95) at 90 days while TBR was 2 (1, 3), 1 (0, 2), and 1 (0, 1)%, respectively, (P = .002). The %time in automated delivery was associated with TIR and change in HbA1c. After controlling other factors including %time in automated delivery, HCL type was not an independent predictor of change in HbA1c nor TIR but remained a significant predictor of TBR.

CONCLUSION

There were significant reductions in HbA1c in CIQ and OP5. TIR was similar across pumps, but TBR was highest with MT. The %time in automated delivery likely explains differences in change in HbA1c but not TBR between HCL systems.

Beneficial effects of automated insulin delivery over one-year follow-up in real life for youths and adults with type 1 diabetes irrespective of patient characteristics

AIM

To investigate glycaemic outcomes in youths and adults with type 1 diabetes with either MiniMed™ 780G or Tandem t:slim X2™ control-IQ automated insulin delivery (AID) systems and to evaluate clinical factors that migrate, mitigate the achievement of therapeutic goals.

MATERIALS AND METHODS

This retrospective, real-world, observational study was conducted in a specialized university type 1 diabetes centre with patients observed for 3-12 months post-initiation of an AID system. Primary outcomes were the percentage time in the target glucose range [TIR70-180 mg/dl (3.9-10 mmol/L)] as measured by continuous glucose monitoring, mean glucose management indicator (GMI) and glycated haemoglobin (HbA1c) levels.

RESULTS

Our study cohort consisted of 48 adolescents and 183 adults (55% females) aged 10-77 years. The mean (95% confidence interval) TIR70-180 mg/dl after 30 days was higher than baseline and by 14% points after 360 days with 71.33% (69.4-73.2) (n = 123, p < .001). HbA1c levels decreased by 0.7% and GMI by 0.6% after 360 days. The proportion of time spent <70 mg/dl (3.9 mmol/L) was not significantly different from baseline. During follow-up, 780G users had better continuous glucose monitoring results than control-IQ users but similar HbA1c levels, and an increased risk of weight gain. Age at onset influenced TIR70-180 mg/dl in univariate analysis but there was no significant relationship after adjusting on explanatory variables. Baseline body mass index did not influence the performance of AID systems.

CONCLUSIONS

This analysis showed the beneficial effects of two AID systems for people with type 1 diabetes across a broad spectrum of participant characteristics. Only half of the participants achieved international recommendations for glucose control with TIR70-180 mg/dl >70%, HbA1c levels or GMI <7%, which outlines the need to maintain strong educational and individual strategies.

(Hybrid) Closed-Loop Systems: From Announced to Unannounced Exercise

Physical activity and exercise have many beneficial effects on general and type 1 diabetes (T1D) specific health and are recommended for individuals with T1D. Despite these health benefits, many people with T1D still avoid exercise since glycemic management during physical activity poses substantial glycemic and psychological challenges - which hold particularly true for unannounced exercise when using an AID system. Automated insulin delivery (AID) systems have demonstrated their efficacy in improving overall glycemia and in managing announced exercise in numerous studies. They are proven to increase time in range (70-180 mg/dL) and can especially counteract nocturnal hypoglycemia, even when evening exercise was performed. AID-systems consist of a pump administering insulin as well as a CGM sensor (plus transmitter), both communicating with a control algorithm integrated into a device (insulin pump, mobile phone/smart watch). Nevertheless, without manual pre-exercise adaptions, these systems still face a significant challenge around physical activity. Automatically adapting to the rapidly changing insulin requirements during unannounced exercise and physical activity is still the Achilles' heel of current AID systems. There is an urgent need for improving current AID-systems to safely and automatically maintain glucose management without causing derailments - so that going forward, exercise announcements will not be necessary in the future. Therefore, this narrative literature review aimed to discuss technological strategies to how current AID-systems can be improved in the future and become more proficient in overcoming the hurdle of unannounced exercise. For this purpose, the current state-of-the-art therapy recommendations for AID and exercise as well as novel research approaches are presented along with potential future solutions - in order to rectify their deficiencies in the endeavor to achieve fully automated AID-systems even around unannounced exercise.

Psychosocial Outcomes Among Users and Nonusers of Open-Source Automated Insulin Delivery Systems: Multinational Survey of Adults With Type 1 Diabetes

BACKGROUND

Emerging research suggests that open-source automated insulin delivery (AID) may reduce diabetes burden and improve sleep quality and quality of life (QoL). However, the evidence is mostly qualitative or uses unvalidated, study-specific, single items. Validated person-reported outcome measures (PROMs) have demonstrated the benefits of other diabetes technologies. The relative lack of research investigating open-source AID using PROMs has been considered a missed opportunity.

OBJECTIVE

This study aimed to examine the psychosocial outcomes of adults with type 1 diabetes using and not using open-source AID systems using a comprehensive set of validated PROMs in a real-world, multinational, cross-sectional study.

METHODS

Adults with type 1 diabetes completed 8 validated measures of general emotional well-being (5-item World Health Organization Well-Being Index), sleep quality (Pittsburgh Sleep Quality Index), diabetes-specific QoL (modified DAWN Impact of Diabetes Profile), diabetes-specific positive well-being (4-item subscale of the 28-item Well-Being Questionnaire), diabetes treatment satisfaction (Diabetes Treatment Satisfaction Questionnaire), diabetes distress (20-item Problem Areas in Diabetes scale), fear of hypoglycemia (short form of the Hypoglycemia Fear Survey II), and a measure of the impact of COVID-19 on QoL. Independent groups 2-tailed t tests and Mann-Whitney U tests compared PROM scores between adults with type 1 diabetes using and not using open-source AID. An analysis of covariance was used to adjust for potentially confounding variables, including all sociodemographic and clinical characteristics that differed by use of open-source AID.

RESULTS

In total, 592 participants were eligible (attempting at least 1 questionnaire), including 451 using open-source AID (mean age 43, SD 13 years; n=189, 41.9% women) and 141 nonusers (mean age 40, SD 13 years; n=90, 63.8% women). Adults using open-source AID reported significantly better general emotional well-being and subjective sleep quality, as well as better diabetes-specific QoL, positive well-being, and treatment satisfaction. They also reported significantly less diabetes distress, fear of hypoglycemia, and perceived less impact of the COVID-19 pandemic on their QoL. All were medium-to-large effects (Cohen d=0.5-1.5). The differences between groups remained significant after adjusting for sociodemographic and clinical characteristics.

CONCLUSIONS

Adults with type 1 diabetes using open-source AID report significantly better psychosocial outcomes than those not using these systems, after adjusting for sociodemographic and clinical characteristics. Using validated, quantitative measures, this real-world study corroborates the beneficial psychosocial outcomes described previously in qualitative studies or using unvalidated study-specific items.

Real-World Glycemic Improvements with Hybrid Closed Loop Pumps in Youth with Type 1 Diabetes

Objective

The use of hybrid closed-loop insulin delivery systems, specifically the t:slim X2 insulin pump with Control IQ (CIQ), has demonstrated improvement in glycemic control in clinical trials and real-world settings. We sought to describe changes in glycemic control with use of CIQ in minority and nonminority youth. Research Design and Methods. This was a retrospective study of youth with type 1 diabetes (T1D) using CIQ over a 12-month period. Medical record data, pump data, and hemoglobin A1c (HbA1c) were collected from the visit prior to starting CIQ and at each clinic visit up to 12 months after starting CIQ. Continuous glucose monitor (CGM) data and HbA1c trajectory over time were compared to baseline and between minority and nonminority youth.

Results

The study included 136 patients of whom 21 were minority youth (non-Hispanic Black and Hispanic), 50% were male, with median age of 13.3y, and median diabetes duration of 4.9y. After starting CIQ, baseline median HbA1c for the nonminority group decreased from 7.8% to 7.1% (p < 0.001), baseline median HbA1c for minority youth decreased from 9.8% to 7.8% (p=0.03), and the percentage of patients meeting target HbA1c <7% increased from 26% to 45%. Both nonminority and minority youth had a significant increase in time in range and decrease of average CGM glucose (p < 0.05).

Conclusions

HbA1c levels decreased in both minority and nonminority youth within 12 months of starting CIQ, and more patients reached the HbA1c target of less than 7%. Disparities in HbA1c between minority and nonminority youth remained and additional studies are warranted to improve this.

Hybrid Closed-Loop Therapy in Adults With Type 1 Diabetes and Above-Target HbA1c: A Real-world Observational Study

OBJECTIVE

We explored longitudinal changes associated with switching to hybrid closed-loop (HCL) insulin delivery systems in adults with type 1 diabetes and elevated HbA1c levels despite the use of intermittently scanned continuous glucose monitoring (isCGM) and insulin pump therapy.

RESEARCH DESIGN AND METHODS

We undertook a pragmatic, preplanned observational study of participants included in the National Health Service England closed-loop pilot. Adults using isCGM and insulin pump across 31 diabetes centers in England with an HbA1c ≥8.5% who were willing to commence HCL therapy were included. Outcomes included change in HbA1c, sensor glucometrics, diabetes distress score, Gold score (hypoglycemia awareness), acute event rates, and user opinion of HCL.

RESULTS

In total, 570 HCL users were included (median age 40 [IQR 29-50] years, 67% female, and 85% White). Mean baseline HbA1c was 9.4 ± 0.9% (78.9 ± 9.1 mmol/mol) with a median follow-up of 5.1 (IQR 3.9-6.6) months. Of 520 users continuing HCL at follow-up, mean adjusted HbA1c reduced by 1.7% (95% CI 1.5, 1.8; P < 0.0001) (18.1 mmol/mol [95% CI 16.6, 19.6]; P < 0.0001). Time in range (70-180 mg/dL) increased from 34.2 to 61.9% (P < 0.001). Individuals with HbA1c of ≤58 mmol/mol rose from 0 to 39.4% (P < 0.0001), and those achieving ≥70% glucose time in range and <4% time below range increased from 0.8 to 28.2% (P < 0.0001). Almost all participants rated HCL therapy as having a positive impact on quality of life (94.7% [540 of 570]).

CONCLUSIONS

Use of HCL is associated with improvements in HbA1c, time in range, hypoglycemia, and diabetes-related distress and quality of life in people with type 1 diabetes in the real world.

The First Regulatory Clearance of an Open-Source Automated Insulin Delivery Algorithm

Open-source Automated Insulin Dosing (OS-AID) algorithms are made publicly accessible so that every facet of their operation can be understood. Currently, commercial AID algorithms are kept proprietary trade secrets, despite the role they take in making life and death decisions for people living with type 1 diabetes. Loop was the second OS-AID algorithm, developed initially by Nate Racklyeft and Pete Schwamb. In 2018, the nonprofit organization Tidepool (Palo Alto, CA) announced the launch of the "Tidepool Loop" initiative with the aim to generate real-world evidence and obtain regulatory clearance. By the end of 2020, the U.S. Food and Drug Administration received Tidepool's application for an interoperable automated glycemic controller based on Loop. After 2 years, the FDA approved the Tidepool Loop on January 23, 2023.

UK's Association of British Clinical Diabetologist's Diabetes Technology Network (ABCD-DTN): Best practice guide for hybrid closed-loop therapy

This best practice guide is written with the aim of providing an overview of current hybrid closed-loop (HCL) systems in use within the United Kingdom's (UK) National Health Service (NHS) and to provide education and advice for their management on both an individual and clinical service level. The environment of diabetes technology, and particularly HCL systems, is rapidly evolving. The past decade has seen unprecedented advances in the development of HCL systems. These systems improve glycaemic outcomes and reduce the burden of treatment for people with type 1 diabetes (pwT1D). It is anticipated that access to these systems will increase in England as a result of updates in National Institute of Health and Care Excellence (NICE) guidance providing broader support for the use of real-time continuous glucose monitoring (CGM) for pwT1D. NICE is currently undertaking multiple-technology appraisal into HCL systems. Based on experience from centres involved in supporting advanced technologies as well as from the recent NHS England HCL pilot, this guide is intended to provide healthcare professionals with UK expert consensus on the best practice for initiation, optimisation and ongoing management of HCL therapy.

A Systematic Review of Commercial Hybrid Closed-Loop Automated Insulin Delivery Systems

INTRODUCTION

Several different forms of automated insulin delivery systems (AID systems) have recently been developed and are now licensed for type 1 diabetes (T1D). We undertook a systematic review of reported trials and real-world studies for commercial hybrid closed-loop (HCL) systems.

METHODS

Pivotal, phase III and real-world studies using commercial HCL systems that are currently approved for use in type 1 diabetes were reviewed with a devised protocol using the Medline database.

RESULTS

Fifty-nine studies were included in the systematic review (19 for 670G; 8 for 780G; 11 for Control-IQ; 14 for CamAPS FX; 4 for Diabeloop; and 3 for Omnipod 5). Twenty were real-world studies, and 39 were trials or sub-analyses. Twenty-three studies, including 17 additional studies, related to psychosocial outcomes and were analysed separately.

CONCLUSIONS

These studies highlighted that HCL systems improve time In range (TIR) and arouse minimal concerns around severe hypoglycaemia. HCL systems are an effective and safe option for improving diabetes care. Real-world comparisons between systems and their effects on psychological outcomes require further study.

Efficacy, Safety, and Satisfaction with the Accu-Chek Insight with Diabeloop Closed-Loop System in Subjects with Type 1 Diabetes: A Multicenter Real-World Study

Aim: To evaluate the efficacy, safety and satisfaction of the closed-loop system Accu-Chek^® Insight with Diabeloop™ (DBLG1™) in adults with type 1 diabetes (T1D) in real-world conditions. Methods: Patients with T1D using DBLG1 for at least 3 months were included. Glucometric parameters were analyzed at baseline, 1, 2, and 3 months after starting DBLG1. HbA_1c was measured before and at 3 months. Technical issues and acute complications were recorded and patients completed a satisfaction questionnaire. Results: Sixty-two patients were included (43 women; age 44.2 ± 11 years; diabetes duration 24.6 ± 12 years; 40 used flash and 22 continuous glucose monitoring (CGM); 45 were on insulin pump and 17 on multiple daily injections). A significant improvement was observed in the CGM-derived glucose metrics early in the first month: Time in range (%TIR) 70-180 mg/dL (54.86 ± 17 vs. 72.23 ± 10.11); time above range level 1 (%TAR1) 180-250 mg/dL (26.26 ± 13.3 vs. 19.48 ± 6.78), time above range level 2 (%TAR2) > 250 mg/dL (12.02 ± 13.09 vs. 6.14 ± 5.23), time below range level 1 (%TBR 1) 54-70 mg/dL (5.73 ± 11.5 vs. 1.67 ± 1.3), time below range level 2 (%TBR2) < 54 mg/dL (1.18 ± 1.97 vs.0.44 ± 0.49), %CV (38.66 ± 7.53 vs. 29.63 ± 3.74), median glucose (168.57 ± 36 mg/dL vs. 154.63 ± 17.55 mg/dL), and %GMI (7.37 ± 0.91 vs. 7.02 ± 0.42). Also, HbA_1c decreased significantly (7.45% ± 1.05% vs. 6.95% ± 0.7%). No acute complications or serious adverse events occurred. Similar improvement was observed regardless of prior therapy or the glucose monitoring system used. Three patients discontinued DBLG1 and 21 experienced technical issues. Overall, patient satisfaction was high. Adjustments of the settings were modified in general in the direction of greater aggressiveness. Conclusions: A significant improvement in glycemic control without serious adverse events and a high degree of patient satisfaction were observed in this first real-world study evaluating the closed-loop system, Accu-Chek Insight with Diabeloop.

Quality of life and psychological well-being among children and adolescents with diabetes and their caregivers using open-source automated insulin delivery systems: Findings from a multinational survey

BACKGROUND

Open-source automated insulin delivery (AID) systems have shown to be safe and effective in children and adolescents with type 1 diabetes (T1D) in real-world studies. However, there is a lack of evidence on the effect on their caregivers' quality-of-life (QoL) and well-being. The aim of this study was to assess the QoL of caregivers and children and adolescents using open-source AID systems using validated measures.

METHODS

In this cross-sectional online survey we examined the caregiver-reported QoL and well-being of users and non-users. Validated questionnaires assessed general well-being (WHO-5), diabetes-specific QoL (PAID, PedsQL) and sleep quality (PSQI).

RESULTS

168 caregivers from 27 countries completed at least one questionnaire, including 119 caregivers of children using open-source AID and 49 not using them. After inclusion of covariates, all measures but the PAID and one subscale of the PedsQL showed significant between-group differences with AID users reporting higher general (WHO-5: p = 0.003), sleep-related (PSQI: p = 0.001) and diabetes-related QoL (PedsQL: p < 0.05).

CONCLUSIONS

The results show the potential impact of open-source AID on QoL and psychological well-being of caregivers and children and adolescents with T1D, and can therefore help to inform academia, regulators, and policymakers about the psychosocial health implications of open-source AID.

Closed-loop insulin delivery: update on the state of the field and emerging technologies

INTRODUCTION

Over the last five years, closed-loop insulin delivery systems have transitioned from research-only to real-life use. A number of systems have been commercialized and are increasingly used in clinical practice. Given the rapidity of new developments in the field, understanding the capabilities and key similarities and differences of current systems can be challenging. This review aims to provide an update on the state of the field of closed-loop insulin delivery systems, including emerging technologies.

AREAS COVERED

We summarize key clinical safety and efficacy evidence of commercial and emerging insulin-only hybrid closed-loop systems for type 1 diabetes. A literature search was conducted and clinical trials using closed-loop systems during free-living conditions were identified to report on safety and efficacy data. We comment on emerging technologies and adjuncts for closed-loop systems, as well as non-technological priorities in closed-loop insulin delivery.

EXPERT OPINION

Commercial hybrid closed-loop insulin delivery systems are efficacious, consistently improving glycemic control when compared to standard therapy. Challenges remain in widespread adoption due to clinical inertia and the lack of resources to embrace technological developments by health care professionals.

Open-Source Automated Insulin Delivery in Type 1 Diabetes

BACKGROUND

Open-source automated insulin delivery (AID) systems are used by many patients with type 1 diabetes. Data are needed on the efficacy and safety of an open-source AID system.

METHODS

In this multicenter, open-label, randomized, controlled trial, we assigned patients with type 1 diabetes in a 1:1 ratio to use an open-source AID system or a sensor-augmented insulin pump (control). The patients included both children (defined as 7 to 15 years of age) and adults (defined as 16 to 70 years of age). The AID system was a modified version of AndroidAPS 2.8 (with a standard OpenAPS 0.7.0 algorithm) paired with a preproduction DANA-i insulin pump and Dexcom G6 CGM, which has an Android smartphone application as the user interface. The primary outcome was the percentage of time in the target glucose range of 70 to 180 mg per deciliter (3.9 to 10.0 mmol per liter) between days 155 and 168 (the final 2 weeks of the trial).

RESULTS

A total of 97 patients (48 children and 49 adults) underwent randomization (44 to open-source AID and 53 to the control group). At 24 weeks, the mean (±SD) time in the target range increased from 61.2±12.3% to 71.2±12.1% in the AID group and decreased from 57.7±14.3% to 54.5±16.0% in the control group (adjusted difference, 14 percentage points; 95% confidence interval, 9.2 to 18.8; P<0.001), with no treatment effect according to age (P = 0.56). Patients in the AID group spent 3 hours 21 minutes more in the target range per day than those in the control group. No severe hypoglycemia or diabetic ketoacidosis occurred in either group. Two patients in the AID group withdrew from the trial owing to connectivity issues.

CONCLUSIONS

In children and adults with type 1 diabetes, the use of an open-source AID system resulted in a significantly higher percentage of time in the target glucose range than the use of a sensor-augmented insulin pump at 24 weeks. (Supported by the Health Research Council of New Zealand; Australian New Zealand Clinical Trials Registry number, ACTRN12620000034932.).

Practical Guidance on Open Source and Commercial Automated Insulin Delivery Systems: A Guide for Healthcare Professionals Supporting People with Insulin-Requiring Diabetes

As increasing numbers of people with insulin-managed diabetes use automated insulin delivery (AID) systems or seek such technologies, healthcare providers are faced with a steep learning curve. Healthcare providers need to understand how to support these technologies to help inform shared decision making, discussing available options, implementing them in the clinical setting, and guiding users in special situations. At the same time, there is a growing diversity of commercial and open source automated insulin delivery systems that are evolving at a rapid pace. This practical guide seeks to provide a conversational framework for healthcare providers to first understand and then jointly assess AID system options with users and caregivers. Using this framework will help HCPs in learning how to evaluate potential new commercial or open source AID systems, while also providing a guide for conversations to help HCPs to assess the readiness and understanding of users for AID systems. The choice of an AID system is not as simple as whether the system is open source or commercially developed, and indeed there are multiple criteria to assess when choosing an AID system. Most importantly, the choices and preferences of the person living with diabetes should be at the center of any decision around the ideal automated insulin delivery system or any other diabetes technology. This framework highlights issues with AID use that may lead to burnout or perceived failures or may otherwise cause users to abandon the use of AID. It discusses the troubleshooting of basic AID system operation and discusses more advanced topics regarding how to maximize the time spent on AID systems, including how to optimize settings and behaviors for the best possible outcomes with AID technology for people with insulin-requiring diabetes. This practical approach article demonstrates how healthcare providers will benefit from assessing and better understanding all available AID system options to enable them to best support each individual.

Barriers to Uptake of Open-Source Automated Insulin Delivery Systems: Analysis of Socioeconomic Factors and Perceived Challenges of Caregivers of Children and Adolescents With Type 1 Diabetes From the OPEN Survey

Background

As a treatment option for people living with diabetes, automated insulin delivery (AID) systems are becoming increasingly popular. The #WeAreNotWaiting community plays a crucial role in the provision and distribution of open-source AID technology. However, while a large percentage of children were early adopters of open-source AID, there are regional differences in adoption, which has prompted an investigation into the barriers perceived by caregivers of children with diabetes to creating open-source systems.

Methods

This is a retrospective, cross-sectional and multinational study conducted with caregivers of children and adolescents with diabetes, distributed across the online #WeAreNotWaiting online peer-support groups. Participants—specifically caregivers of children not using AID—responded to a web-based questionnaire concerning their perceived barriers to building and maintaining an open-source AID system.

Results

56 caregivers of children with diabetes, who were not using open-source AID at the time of data collection responded to the questionnaire. Respondents indicated that their major perceived barriers to building an open-source AID system were their limited technical skills (50%), a lack of support by medical professionals (39%), and therefore the concern with not being able to maintain an AID system (43%). However, barriers relating to confidence in open-source technologies/unapproved products and fear of digital technology taking control of diabetes were not perceived as significant enough to prevent non-users from initiating the use of an open-source AID system.

Conclusions

The results of this study elucidate some of the perceived barriers to uptake of open-source AID experienced by caregivers of children with diabetes. Reducing these barriers may improve the uptake of open-source AID technology for children and adolescents with diabetes. With the continuous development and wider dissemination of educational resources and guidance—for both aspiring users and their healthcare professionals—the adoption of open-source AID systems could be improved.

Large-Scale Data Analysis for Glucose Variability Outcomes with Open-Source Automated Insulin Delivery Systems

Open-source automated insulin delivery (AID) technologies use the latest continuous glucose monitors (CGM), insulin pumps, and algorithms to automate insulin delivery for effective diabetes management. Early community-wide adoption of open-source AID, such as OpenAPS, has motivated clinical and research communities to understand and evaluate glucose-related outcomes of such user-driven innovation. Initial OpenAPS studies include retrospective studies assessing high-level outcomes of average glucose levels and HbA1c, without in-depth analysis of glucose variability (GV). The OpenAPS Data Commons dataset, donated to by open-source AID users with insulin-requiring diabetes, is the largest freely available diabetes-related dataset with over 46,070 days’ worth of data and over 10 million CGM data points, alongside insulin dosing and algorithmic decision data. This paper first reviews the development toward the latest open-source AID and the performance of clinically approved GV metrics. We evaluate the GV outcomes using large-scale data analytics for the n = 122 version of the OpenAPS Data Commons. We describe the data cleaning processes, methods for measuring GV, and the results of data analysis based on individual self-reported demographics. Furthermore, we highlight the lessons learned from the GV outcomes and the analysis of a rich and complex diabetes dataset and additional research questions that emerged from this work to guide future research. This paper affirms previous studies’ findings of the efficacy of open-source AID.

An Open-Source Web Portal for Managing Self-reported Data and Real-world Data Donation in Diabetes Research: Feasibility Study. (Preprint)

Background

People with diabetes and their support networks have developed open-source automated insulin delivery systems to help manage their diabetes therapy, as well as to improve their quality of life and glycemic outcomes. Under the hashtag #WeAreNotWaiting, a wealth of knowledge and real-world data have been generated by users of these systems but have been left largely untapped by research; opportunities for such multimodal studies remain open.

Objective

We aimed to evaluate the feasibility of several aspects of open-source automated insulin delivery systems including challenges related to data management and security across multiple disparate web-based platforms and challenges related to implementing follow-up studies.

Methods

We developed a mixed methods study to collect questionnaire responses and anonymized diabetes data donated by participants—which included adults and children with diabetes and their partners or caregivers recruited through multiple diabetes online communities. We managed both front-end participant interactions and back-end data management with our web portal (called the Gateway). Participant questionnaire data from electronic data capture (REDCap) and personal device data aggregation (Open Humans) platforms were pseudonymously and securely linked and stored within a custom-built database that used both open-source and commercial software. Participants were later given the option to include their health care providers in the study to validate their questionnaire responses; the database architecture was designed specifically with this kind of extensibility in mind.

Results

Of 1052 visitors to the study landing page, 930 participated and completed at least one questionnaire. After the implementation of health care professional validation of self-reported clinical outcomes to the study, an additional 164 individuals visited the landing page, with 142 completing at least one questionnaire. Of the optional study elements, 7 participant–health care professional dyads participated in the survey, and 97 participants who completed the survey donated their anonymized medical device data.

Conclusions

The platform was accessible to participants while maintaining compliance with data regulations. The Gateway formalized a system of automated data matching between multiple data sets, which was a major benefit to researchers. Scalability of the platform was demonstrated with the later addition of self-reported data validation. This study demonstrated the feasibility of custom software solutions in addressing complex study designs. The Gateway portal code has been made available open-source and can be leveraged by other research groups.