Continuous Glucose Monitoring Data Research Articles

Using Continuous Glucose Monitoring and Data Sharing to Encourage Collaboration Among Older Adults With Type 1 Diabetes and Their Care Partners: Qualitative Descriptive Study.

Persons with diabetes use continuous glucose monitoring (CGM) to self-manage their diabetes. Care partners (CPs) frequently become involved in supporting persons with diabetes in the management of their diabetes. However, persons with diabetes and CP dyads may require more communication and problem-solving skills regarding how to share and respond to CGM data. The purpose of this study was to describe the experiences of persons with diabetes and CPs who participated in the Share "plus" intervention, which addresses dyadic communication strategies, problem-solving, and action planning to promote sharing of CGM data among the dyad. Ten dyads participated in the Share "plus" telehealth intervention. Participants were interviewed during and after the Share "plus" intervention. Thematic analysis was used to analyze interview data. During postsession interviews, dyads described feeling a sense of shared responsibility yet viewed the persons with diabetes as ultimately responsible for the disease. Additionally, dyads shared that communication patterns improved and were able to recognize the negative aspects of previously established communication patterns. Dyads reported communication focused on hypoglycemia episodes while also differing in the frequency they reviewed CGM data and set alerts. Overall, dyads expressed positive reactions to the Share "plus" intervention. Share "plus" was helpful in promoting positive CGM-related communication among dyads and encouraged more CP support. CPs play an important role in supporting older adults with type 1 diabetes. Communication strategies help support dyad involvement in CGM data sharing and self-management among persons with diabetes.

Jointly Predicting Postprandial Hypoglycemia and Hyperglycemia Using Continuous Glucose Monitoring Data in Type 1 Diabetes.

The development of continuous glucose monitoring (CGM) systems has enabled people with type 1 diabetes mellitus (T1DM) to track their glucose trajectory in real-time and inspired research in personalised glucose prediction. In this paper, our aim is to predict postprandial abnormal-glycemia events. Different from prior research which focuses on hypoglycemia only, we make the first attempt to establish our problem as the joint prediction of hyperglycemia and hypoglycemia. On this basis, we propose a machine learning model that learns from the pattern of 1 hour past glucose and makes predictions for the two tasks simultaneously using a unified backbone. Key benefits of our methodology include 1) requiring only the CGM sequence as the input, thus making it more widely applicable than other counterparts using extra inputs such as the nutrition details, and 2) minimising the computational cost as the two tasks are unified into a single model. Our experiments on the openly available OhioT1DM dataset achieve state-of-the-art performance (Matthew's correlation coefficient of 0.61 for hyperglycemia and 0.48 for hypoglycemia). To encourage further study, we release our codes at https://github.com/r-cui/PostprandialHyperHypoPrediction under the MIT license.

Heterogeneous temporal representation for diabetic blood glucose prediction.

Background and aims: Blood glucose prediction (BGP) has increasingly been adopted for personalized monitoring of blood glucose levels in diabetic patients, providing valuable support for physicians in diagnosis and treatment planning. Despite the remarkable success achieved, applying BGP in multi-patient scenarios remains problematic, largely due to the inherent heterogeneity and uncertain nature of continuous glucose monitoring (CGM) data obtained from diverse patient profiles. Methodology: This study proposes the first graph-based Heterogeneous Temporal Representation (HETER) network for multi-patient Blood Glucose Prediction (BGP). Specifically, HETER employs a flexible subsequence repetition method (SSR) to align the heterogeneous input samples, in contrast to the traditional padding or truncation methods. Then, the relationships between multiple samples are constructed as a graph and learned by HETER to capture global temporal characteristics. Moreover, to address the limitations of conventional graph neural networks in capturing local temporal dependencies and providing linear representations, HETER incorporates both a temporally-enhanced mechanism and a linear residual fusion into its architecture. Results: Comprehensive experiments were conducted to validate the proposed method using real-world data from 112 patients in two hospitals, comparing it with five well-known baseline methods. The experimental results verify the robustness and accuracy of the proposed HETER, which achieves the maximal improvement of 31.42%, 27.18%, and 34.85% in terms of MAE, MAPE, and RMSE, respectively, over the second-best comparable method. Discussions: HETER integrates global and local temporal information from multi-patient samples to alleviate the impact of heterogeneity and uncertainty. This method can also be extended to other clinical tasks, thereby facilitating efficient and accurate capture of crucial pattern information in structured medical data.

Prolonged Use of an Automated Insulin Delivery System Improves Sleep in Long-Standing Type 1 Diabetes Complicated by Impaired Awareness of Hypoglycemia.

This study assessed changes in actigraphy-estimated sleep and glycemic outcomes after initiating automated insulin delivery (AID). Ten adults with long-standing type 1 diabetes and impaired awareness of hypoglycemia (IAH) participated in an 18-month clinical trial assessing an AID intervention on hypoglycemia and counter-regulatory mechanisms. Data from eight participants (median age = 58 years) with concurrent wrist actigraph and continuous glucose monitoring (CGM) data were used in the present analyses. Actigraphs and CGM measured sleep and glycemic control at baseline (one week) and months 3, 6, 9, 12, 15, and 18 (three weeks) following AID initiation. HypoCount software integrated actigraphy with CGM data to separate wake and sleep-associated glycemic measures. Paired sample t-tests and Cohen's d effect sizes modeled changes and their magnitude in sleep, glycemic control, IAH (Clarke score), hypoglycemia severity (HYPO score), hypoglycemia exposure (CGM), and glycemic variability (lability index [LI]; CGM coefficient-of-variation [CV]) from baseline to 18 months. Sleep improved from baseline to 18 months (shorter sleep latency [P < .05, d = 1.74], later sleep offset [P < .05, d = 0.90], less wake after sleep onset [P < .01, d = 1.43]). Later sleep onset (d = 0.74) and sleep midpoint (d = 0.77) showed medium effect sizes. Sleep improvements were evident from 12 to 15 months after AID initiation and were preceded by improved hypoglycemia awareness (Clarke score [d = 1.18]), reduced hypoglycemia severity (HYPO score [d = 2.13]), reduced sleep-associated hypoglycemia (percent time glucose was < 54 mg/dL, < 60 mg/dL,< 70 mg/dL; d = 0.66-0.81), and reduced glucose variability (LI, d = 0.86; CV, d = 0.62). AID improved sleep initiation and maintenance. Improved awareness of hypoglycemia, reduced hypoglycemia severity, hypoglycemia exposure, and glucose variability preceded sleep improvements.This trial is registered with ClinicalTrials.gov NCT03215914 https://clinicaltrials.gov/ct2/show/NCT03215914.

Fast-Acting Insulin Aspart in Patients with Type1 Diabetes in Real-World Clinical Practice: A Noninterventional, Retrospective Chart and Database Study.

This study utilized continuous glucose monitoring data to analyze the effects of switching to treatment with fast-acting insulin aspart (faster aspart) in adults with type1 diabetes (T1D) in clinical practice. A noninterventional database review was conducted in Sweden among adults with T1D using multiple daily injection (MDI) regimens who had switched to treatment with faster aspart as part of basal-bolus treatment. Glycemic data were retrospectively collected during the 26weeks before switching (baseline) and up to 32weeks after switching (follow-up) to assess changes in time in glycemic range (TIR; 70-180mg/dL), mean sensor glucose, glycated hemoglobin (HbA1c) levels, coefficient of variation, time in hyperglycemia (level1, > 180 to ≤ 250mg/dL; level2, > 250mg/dL), and time in hypoglycemia (level1, ≥ 54 to < 70mg/dL; level2, < 54mg/dL) (ClinicalTrials.gov Identifier NCT03895515). Overall, 178 participants were included in the study cohort. The analysis population included 82 individuals (mean age 48.5years) with adequate glucose sensor data. From baseline to follow-up, statistically significant improvements were reported for TIR (mean increase 3.3%-points [approximately 48min/day]; p = 0.006) with clinically relevant improvement (≥ 5%) in 43% of participants. Statistically significant improvements from baseline were observed for mean sensor glucose levels, HbA1c levels, and time in hyperglycemia (levels1 and 2), with no statistically significant changes in time spent in hypoglycemia. Switching to faster aspart was associated with improvements in glycemic control without increasing hypoglycemia in adults with T1D using MDI in this real-world setting.

A pregnancy-specific Glucose management indicator derived from continuous glucose monitoring in pregnant women with type 1 diabetes.

Glucose management indicator (GMI) is a core metric derived from continuous glucose monitoring (CGM) and is widely used to evaluate glucose control in patients with diabetes. No study has explored the pregnancy-specific GMI. This study aimed to derive a best-fitting model to calculate GMI from mean blood glucose (MBG) obtained from CGM among pregnant women with type 1 diabetes mellitus (T1DM). A total of 272 CGM data and corresponding laboratory HbA1c from 98 pregnant women with T1DM in the CARNATION study were analysed in this study. Continuous glucose monitoring data were collected to calculate MBG, time-in-range (TIR), and glycaemic variability parameters. The relationships between the MBG and HbA1c during pregnancy and postpartum were explored. Mix-effect regression analysis with polynomial terms and cross-validation method was conducted to investigate the best-fitting model to calculate GMI from MBG obtained by CGM. The pregnant women had a mean age of 28.9±3.8years, with a diabetes duration of 8.8±6.2years and a mean body mass index (BMI) of 21.1±2.5kg/m2 . The HbA1c levels were 6.1±1.0% and 6.4±1.0% during pregnancy and at postpartum (p=0.024). The MBG levels were lower during pregnancy than those at postpartum (6.5±1.1mmol/L vs. 7.1±1.5mmol/L, p=0.008). After adjusting the confounders of haemoglobin (Hb), BMI, trimesters, disease duration, mean amplitude of glycaemic excursions and CV%, we developed a pregnancy-specific GMI-MBG equation: GMI for pregnancy (%)=0.84-0.28* [Trimester] + 0.08 * [ BMI in kg/m2 ] + 0.01 * [Hb in g/mL] + 0.50 * [MBG in mmol/L]. We derived a pregnancy-specific GMI equation, which should be recommended for antenatal clinical care. ChiCTR1900025955.

Importance of continuous glucose monitoring in the treatment of diabetes mellitus

Background: Emerging evidence suggests that controlling both glycemic variability and hemoglobin A1c is necessary to prevent complications associated with diabetes mellitus. Hence, continuous glucose monitoring (CGM) is crucial for effectively managing diabetes.Current Concepts: There are two primary types of CGM. Retrospective CGM only allows the reviewal of glycemic data after the monitoring period, whereas personal CGM allows real-time monitoring. Personal CGM can be further categorized into real-time CGM and intermittently scanned CGM. To interpret CGM data, time in range (TIR) is considered the standard parameter. A TIR of 70–180 mg/dL for more than 70% of the period has been established as a typical target for both type 1 and type 2 diabetes. Other parameters such as time below range, time above range, coefficient of variation, and glucose management indicator should also be reviewed. Importantly, numerous clinical studies have demonstrated the efficacy of CGM in both type 1 and type 2 diabetes.Discussion and Conclusion: A wealth of clinical evidence supports the application of CGM in diabetes, confirming its effectiveness across various treatment stages. CGM has emerged as a compelling therapeutic option in instances when other treatment choices remain limited. With a growing body of clinical evidence, the widespread adoption of CGM in diabetes management appears inevitable. However, challenges related to user comfort, cost, the need for extensive data interpretation, and necessary system improvements remain unaddressed. Further research is required to validate the appropriate usage and frequency of CGM through costeffectiveness analyses.

Changes in the glycaemia risk index and its association with other continuous glucose monitoring metrics after initiation of an automated insulin delivery system in adults with type 1 diabetes.

To evaluate the glycaemia risk index (GRI) and its association with other continuous glucose monitoring (CGM) metrics after initiation of an automated insulin delivery (AID) system in patients with type 1 diabetes (T1D). Up to 90 days of CGM data before and after initiation of an AID system from 185 CGM users with T1D were collected. GRI and other CGM metrics were calculated using cgmanalysis R software and were analysed for 24 hours, for both night-time and daytime. GRI values were assigned to five GRI zones: zone A (0-20), B (21-40), C (41-60), D (61-80) and E (81-100). Compared with baseline, GRI and its components decreased significantly after AID initiation (GRI: 48.7 ± 21.8 vs. 29 ± 13; hypoglycaemia component: 2.7 ± 2.8 vs. 1.6 ± 1.7; hyperglycaemia component: 25.3 ± 14.5 vs. 15 ± 8.5; P < .001 for all). The GRI was inversely correlated with time in range before (r = -0.962) and after (r = -0.961) AID initiation (P < .001 for both). GRI was correlated with time above range (before: r = 0.906; after = 0.910; P < .001 for both), but not with time below range (P > .05). All CGM metrics improved after AID initiation during 24 hours, for both daytime and night-time (P < .001 for all). Metrics improved significantly more during night-time than daytime (P < .01). GRI was highly correlated with various CGM metrics above, but not below target range, both before and after AID initiation.

Real-World Evidence Analysis of a Hybrid Closed-Loop System.

We analyzed real-world evidence to assess the performance of the mylife CamAPS FX hybrid closed-loop system. Users from 15 countries across different age groups who used the system between May 9, 2022, and December 3, 2022, and who had ≥30 days of continuous glucose monitor data, and ≥30% of closed-loop usage were included in the current analysis (N = 1805). Time in range (3.9-10 mmol/L) was 72.6 ± 11.5% (mean ± SD) for all users and increased by age from 66.9 ± 11.7% for users ≤6 years old to 81.8 ± 8.7% for users ≥65 years. Time spent in hypoglycemia (<3.9 mmol/L) was 2.3% [1.3, 3.6] (median [interquartile range]). Mean glucose and glucose management indicator were 8.4 ± 1.1 mmol/L and 6.9%, respectively. Time using closed-loop was high at 94.7% [90.0, 96.9]. Glycemic outcomes from the present real-world evidence are comparable to results obtained from previous randomized controlled studies and confirm the efficacy of this hybrid closed-loop system in real-world settings.

A Retrospective Cohort Study of Racial/Ethnic and Socioeconomic Disparities in Initiation and Meaningful Use of Continuous Glucose Monitoring Among Youth With Type 1 Diabetes.

Continuous glucose monitor (CGM) use improves type 1 diabetes (T1D) outcomes, yet children from diverse backgrounds and on public insurance have worse outcomes and lower CGM utilization. Using novel CGM data acquisition and analysis of two T1D cohorts, we test the hypothesis that T1D youth from different backgrounds experience disparities in meaningful CGM use following both T1D diagnosis and CGM uptake. Cohorts drawn from a pediatric T1D program were followed for one year beginning at diagnosis (n = 815, 2016-2020) or CGM uptake (n = 1392, 2015-2020). Using chart and CGM data, CGM start and meaningful use outcomes between racial/ethnic and insurance groups were compared using median days, one-year proportions, and survival analysis. Publicly compared with privately insured were slower to start CGM (233, 151 days, P < .01), had fewer use-days in the year following uptake (232, 324, P < .001), and had faster first discontinuation rates (hazard ratio [HR] = 1.61, P < .001). Disparities were more pronounced among Hispanic and black compared with white subjects for CGM start time (312, 289, 149, P = .0013) and discontinuation rates (Hispanic HR = 2.17, P < .001; black HR = 1.45, P = .038), and remained even among privately insured (Hispanic/black HR = 1.44, P = .0286). Given the impact of insurance and race/ethnicity on CGM initiation and use, it is imperative that we target interventions to support universal access and sustained CGM use to mitigate the potential impact of provider biases and systemic disadvantage and racism. By enabling more equitable and meaningful T1D technology use, such interventions will begin to alleviate outcome disparities between youth with T1D from different backgrounds.

A Markov Model of Gap Occurrence in Continuous Glucose Monitoring Data for Realistic in Silico Clinical Trials

Background and objectiveContinuous glucose monitoring (CGM) sensors measure interstitial glucose concentration every 1-5 min for days or weeks. New CGM-based diabetes therapies are often tested in in silico clinical trials (ISCTs) using diabetes simulators. Accurate models of CGM sensor inaccuracies and failures could help improve the realism of ISCTs. However, the modeling of CGM failures has not yet been fully addressed in the literature. This work aims to develop a mathematical model of CGM gaps, i.e., occasional portions of missing data generated by temporary sensor errors (e.g., excessive noise or artifacts). MethodsTwo datasets containing CGM traces collected in 167 adults and 205 children, respectively, using the Dexcom G6 sensor (Dexcom Inc., San Diego, CA) were used. Four Markov models, of increasing complexity, were designed to describe three main characteristics: number of gaps for each sensor, gap distribution in the monitoring days, and gap duration. Each model was identified on a portion of each dataset (training set). The remaining portion of each dataset (real test set) was used to evaluate model performance through a Monte Carlo simulation approach. Each model was used to generate 100 simulated test sets with the same size as the real test set. The distributions of gap characteristics on the simulated test sets were compared with those observed on the real test set, using the two-sample Kolmogorov-Smirnov test and the Jensen-Shannon divergence. ResultsA six-state Markov model, having two states to describe normal sensor operation and four states to describe gap occurrence, achieved the best results. For this model, the Kolmogorov-Smirnov test found no significant differences between the distribution of simulated and real gap characteristics. Moreover, this model obtained significantly lower Jensen-Shannon divergence values than the other models. ConclusionsA Markov model describing CGM gaps was developed and validated on two real datasets. The model describes well the number of gaps for each sensor, the gap distribution over monitoring days, and the gap durations. Such a model can be integrated into existing diabetes simulators to realistically simulate CGM gaps in ISCTs and thus enable the development of more effective and robust diabetes management strategies.

Simulating Realistic Continuous Glucose Monitor Time Series By Data Augmentation.

Simulated data are a powerful tool for research, enabling benchmarking of blood glucose (BG) forecasting and control algorithms. However, expert created models provide an unrealistic view of real-world performance, as they lack the features that make real data challenging, while black-box approaches such as generative adversarial networks do not enable systematic tests to diagnose model performance. To address this, we propose a method that learns missingness and error properties of continuous glucose monitor (CGM) data collected from people with type 1 diabetes (OpenAPS, OhioT1DM, RCT, and Racial-Disparity), and then augments simulated BG data with these properties. On the task of BG forecasting, we test how well our method brings performance closer to that of real CGM data compared with current simulation practices for missing data (random dropout) and error (Gaussian noise, CGM error model). Our methods had the smallest performance difference versus real data compared with random dropout and Gaussian noise when individually testing the effects of missing data and error on simulated BG in most cases. When combined, our approach was significantly better than Gaussian noise and random dropout for all data sets except OhioT1DM. Our error model significantly improved results on diverse data sets. We find a significant gap between BG forecasting performance on simulated and real data, and our method can be used to close this gap. This will enable researchers to rigorously test algorithms and provide realistic estimates of real-world performance without overfitting to real data or at the expense of data collection.

Association between glycaemia risk index (GRI) and diabetic retinopathy in type 2 diabetes: A cohort study.

To investigate the association between a new composite metric, glycaemia risk index (GRI), and incident diabetic retinopathy (DR). A total of 1204 adults with type 2 diabetes without DR at baseline were included between 2005 and 2019 from a single centre in Shanghai, China. GRI was obtained from continuous glucose monitoring data at baseline. Cox proportion hazard regression analysis was used to assess the association between GRI and the risk of incident DR. During a median follow-up of 8.4 years, 301 patients developed DR. The multivariable-adjusted hazard ratios (HRs) for incident DR across ascending GRI quartiles (≤14 [reference], 15 ~ 28, 29 ~ 47 and > 47) were 1.00, 1.05 (95% CI 0.74-1.48), 1.33 (95% confidence interval [CI] 0.96-1.84) and 1.53 (95% CI 1.11-2.11), respectively. For each 1-SD increase in GRI, the risk of DR was increased by 20% (HR 1.20, 95% CI 1.07-1.33) after adjustment for confounders. In patients with type 2 diabetes, higher GRI is associated with an increased risk of incident DR. GRI has the potential to be a valuable clinical measure, which needs to be further explored in future studies.

1821-PUB: Relationship between Time in Range and Glucose Management Indicator in Adolescents and Children with Type 1 Diabetes Mellitus

Objective: To investigate the relationship between time in range (TIR, 3.9-10.0mmol/L) and glucose management indicator (GMI) in adolescents and children with type 1 diabetes mellitus (T1DM) and explore the impact of coefficient of variation (CV) on their relationship. Methods: Data derived from continuous glucose monitoring (CGM) and other clinical data (including age, duration of T1DM, and laboratory-measured HbA1c) were obtained from the annual follow-up of the Guangdong T1DM Translational Medicine Study. The patients under 18 years old were included. The patients were divided into CV≤36% group and CV&amp;gt;36% group by the attainment of CV. The relationship between TIR and GMI of both groups was assessed with correlation coefficient. Further, patients were divided into 4 groups by the interquartile range of CV. The linear regression model was used to calculate the TIR predicted value corresponding to the same GMI in 4 groups. Results: The 56 eligible datasets collected from May 2014 to August 2021 were included. The median age, duration of T1DM, and laboratory-measured HbA1c were 14.00 (12.00, 16.00) years, 4.15 (1.66, 5.29) years, and 9.00 (7.65, 11.23) %, respectively. The median TIR, GMI, CV, and valid number of days the CGM device was worn were 60.73 (42.59, 77.72) %, 7.37 (6.78, 8.56) %, 30.38 (24.87, 34.94) %, and 3.39 (2.97, 4.84) days, respectively. TIR and GMI were highly linear correlated (R2=0.89, p&amp;lt;0.001), and a significantly higher Spearman’s correlation coefficient was observed in the CV≤36% group than in CV&amp;gt;36% group ((R2=0.92, p&amp;lt;0.001) vs. (R2=0.63, p=0.004)). When the GMI was 7%, the corresponding TIR predicted values gradually decreased with the increase of CV, which were 75.66% (CV≤24.87%), 73.48% (24.87%&amp;lt;cv≤30.38%), 69.16% (30.38%34.94%). Conclusions: TIR and GMI were highly linear correlated in adolescents and children with T1DM. With constant GMI, the less the glycemic fluctuation, the higher the TIR.&amp;lt;/cv≤30.38%), 69.16% (30.38% Disclosure M.Lei: None. J.Lv: None. X.Mai: None. H.Deng: None. C.Wang: None. D.Yang: None. X.Yang: None. W.Xu: None. J.Yan: None. Funding Science and Technology Planning Project of Guangzhou (202102010154)

First Real-World Experience With Bigfoot Unity: A 6-Month Retrospective Analysis.

The Bigfoot Unity Diabetes Management System, a smart pen cap system cleared by the U.S. Food and Drug Administration in May 2021, incorporates continuous glucose monitoring data, real-time glycemic alerts, and clinician-directed dose recommendations. This study analyzed real-world clinical outcomes data for an initial cohort (n = 58, from 13 clinics) managing multiple daily injection insulin therapy using the pen cap system for 6 months. We examined glycemic control, including hypoglycemia events and interaction with and use of the pen cap system. In a cohort mainly consisting of adults with type 2 diabetes and an average age of 62 years, the results demonstrate close adherence to established glycemic targets, including a relatively short amount of time spent in the hypoglycemic range.

128-LB: Accessing Glucose Fluctuation—The Role of FPG, PPG, and HbA1c

Background: Time in range (TIR) has been widely used clinically to investigate the glucose fluctuation in diabetes management. Studies have been done in exploring the relationship between TIR and HbA1c to estimate TIR without multi-invasive measurements. Beyond the univariate correlation model, this study incorporated fasting plasma glucose (FPG), postprandial glucose (PPG) obtained from glucose tolerance test, which can to some extent reflect glucose fluctuations, and revealed the advantage and the potential of the multi-variate linear or no-linear model to estimate TIR. Method: This study involved 399 Chinese T2DM patients. TIR were calculated from continuous glucose monitoring (CGM) data with a coverage of at least 70%. FPG, PPG were measured by oral or mixed-meal glucose tolerance test. Univariate, multivariate linear regression and non-linear random forest (RF) regression were used in regression analysis. Model performance was assessed using MAE, RMSE and MSE. Results: Regression analysis between glucose tolerance test data and CGM suggests that FPG were strongly linked to trough glucose while PPG were strongly correlated with peak glucose. Regression analysis revealed a stronger correlation between TIR and a triad of FPG, PPG, HbA1c than single HbA1c variable. And the performance of RF model is superior (correlation = 0.79, MAE = 14.77), among multivariate linear (0.73, 16.67) and univariate linear (0.62, 21.29) model. Conclusion: Our study shows that FPG and PPG are closely associated to trough and peak glucose in CGM data, as they provide a more complete picture of glucose fluctuations compared to HbA1c alone. The TIR estimation model that based on FPG, PPG, and HbA1c outperformed univariate model. However, the performance of RF based non-linear model suggests a more complex relationship between those parameters, indicating personalized models for different populations is needed in TIR estimating without long-term invasive glucose monitoring. Disclosure J. Ma: None. X. Su: None. L. Feng: Employee; Hua Medicine. B. Ding: None. R. Yan: None. R. Sun: None. Y. Zhang: None. Y. Duan: None. Funding Nanjing Municipal Health Commission (ZKX22038)

392-P: Optimizing Duration and Threshold of Sensor Glucose for Identifying Symptomatic Hypoglycemia in Insulin-Treated Diabetes—Hypo-METRICS Study

Many episodes of hypoglycemia defined by the ATTD consensus are asymptomatic. We investigated if optimizing the glucose threshold and duration for sensor detected hypoglycemia (SDH) improves precision and sensitivity for identifying symptomatic hypoglycaemia from continuous glucose monitoring (CGM) data. We analysed 10 weeks of blinded CGM (Libre 2) and FitBit data from 435 participants [217 type 1, 218 type 2] with intact hypoglycemic awareness (Gold score &amp;lt;4). They self-reported symptomatic hypoglycemia on the Hypo-METRICS smartphone app. We used particle Markov chain Monte Carlo optimisation to generate the threshold and duration of SDH that maximizes detection of symptomatic hypoglycemia across our population and assessed impact of sleep status and diabetes type. The optimized threshold and duration were 70mg/dl for 27 min, with 2% gain of precision but 8% loss of sensitivity vs level 1 SDH (70mg/dl &amp;gt;15 min). Both level 1 SDH and our optimized definition had 3-fold increased sensitivity vs level 2 SDH (54mg/dl &amp;gt;15 min), with minimal loss of precision (Table). Our data validate ATTD level 1 hypoglycemia. Overall precision for detection of symptomatic hypoglycemia remains low despite optimization. Disclosure P.Divilly: None. U.Pedersen-bjergaard: Advisory Panel; Novo Nordisk A/S, Sanofi, Vertex Pharmaceuticals Incorporated. R.J.Mccrimmon: Advisory Panel; Sanofi, Speaker's Bureau; Novo Nordisk A/S. J.K.Mader: Advisory Panel; Novo Nordisk A/S, Abbott Diabetes, Roche Diabetes Care, Eli Lilly and Company, Sanofi, Medtronic, Becton, Dickinson and Company, Pharmasense, embecta, Research Support; A. Menarini Diagnostics, Abbott Diabetes, Roche Diabetes Care, Dexcom, Inc., Profusa, Inc., Speaker's Bureau; Novo Nordisk A/S, A. Menarini Diagnostics, Abbott Diabetes, Roche Diabetes Care, Eli Lilly and Company, Sanofi, Boehringer Ingelheim Inc., Becton, Dickinson and Company, Ypsomed AG, Viatris Inc., Servier Laboratories, Medtrust, Stock/Shareholder; Decide Clinical Software GmbH. M.Evans: Advisory Panel; Zucara Therapeutics, Pila Pharma, Dexcom, Inc., Other Relationship; Novo Nordisk, AstraZeneca, Abbott Diabetes, Speaker's Bureau; Eli Lilly and Company. S.A.Amiel: Advisory Panel; Medtronic, Other Relationship; Sanofi, Novo Nordisk. P.Choudhary: Advisory Panel; Medtronic, Novo Nordisk, Dexcom, Inc., MannKind Corporation, Insulet Corporation, Research Support; Abbott Diabetes, Speaker's Bureau; Sanofi, Lilly. Hypo-resolve consortium: n/a. Z.Mahmoudi: Employee; Novo Nordisk. G.Martine-edith: Other Relationship; Novo Nordisk A/S. D.Boiroux: Employee; Novo Nordisk A/S. N.Zaremba: None. U.Soeholm: Employee; Novo Nordisk A/S. M.Gomes: Employee; Novo Nordisk A/S. A.A.Vaag: None. B.E.De galan: Research Support; Novo Nordisk. Funding Innovative Medicines Initiative 2 Joint Undertaking (777460)

970-P: Effectiveness of Continuous Glucose Monitoring Use for Diabetes Management in an Academic Endocrinology Practice

The achievement of individualized glycemic targets in patients with diabetes is often a significant challenge. Empowering patients with technological tools in the ambulatory setting can pave the way for improved glucose control. We retrospectively examined the records of 94 patients managed in an academic endocrinology practice who were referred for management of diabetes over a 12-month period. Forty-one patients (18 females and 23 males, average age 57.7 years), who had at least two hemoglobin A1c measurements and two 14-day periods of personal continuous glucose monitoring (CGM) data, were included in the analysis. Nine patients had type 1 and 32 patients had type 2 diabetes. Eight patients were not on any insulin therapy, 8 patients were on basal insulin, 18 were using multiple daily insulin injections, and 7 were managed with insulin pump therapy. A multidisciplinary management approach was utilized. The pharmacologic regimen, including both insulin and other medications, was adjusted every 2-4 weeks via in-person visits and remote data upload as necessary. The average A1c was 8.4% at initial visit and 7.1% at the most recent follow-up, a decrease of 1.3% (p=0.0002). Review of CGM data showed that the average time-in-range (70-180 mg/dl) increased from 63.6% at first CGM download to 74.1% at a mean follow-up of 301 days (p=0.0097), while the amount of time in the hyperglycemic range (time-above-range, &amp;gt;180 mg/dl) decreased from 33.5% to 22.1% (p=0.0079). Time-below-range (&amp;lt;70 mg/dl) remained statistically unchanged, signifying no worsening of hypoglycemia. In conclusion, in ambulatory patients with diabetes and suboptimal glucose control, frequent and tailored adjustments in insulin and noninsulin medications aided by CGM use with close follow-up can lead to significant improvement in glycemic control. The latter, if sustained, can translate to fewer complications and greater patient satisfaction in the long-term. Disclosure K.N.Grennan: None. A.A.Rizvi: None. A.Kathuria: None.

898-P: Real-World Glycemic Outcomes of &amp;gt;3,300 Children and Adolescents with Type 1 Diabetes Using the Omnipod 5 Automated Insulin Delivery (AID) System with Cloud-Based Data Management

The Omnipod 5 AID System is a novel tubeless hybrid closed-loop system that allows for personalized therapy through customizable glucose targets from 110-150mg/dL in 10mg/dL increments. The System enables automatic upload of data for all users initiating the system, facilitating unprecedented access to evaluate real-world outcomes. Continuous glucose monitoring (CGM) and insulin data from Omnipod 5 users with type 1 diabetes (T1D) aged &amp;lt;18y in the US with ≥90 days of data available in the cloud-based data management system were included. Data from 3,369 users with sufficient CGM data (≥75% of days with ≥220 readings) and using the lowest target (110mg/dL) for ≥50% of the time, aged &amp;lt;6y (n=182), 6 to 12y (n=2,050), and 13 to 17y (n=1,137) were available at the time of analysis. Users were aged mean±SD 10.9±3.4y with median 146 days of system use. Outcomes are shown in the Table for each age group. Median time in target range (70-180mg/dL) was 74.0%, 70.7%, and 67.1% for each age group, respectively. Time &amp;lt;70mg/dL in each age group was low: median 2.7%, 1.8%, and 1.2%, respectively. These results are the first to demonstrate that in over 3,000 children and adolescents using the Omnipod 5 System in a real-world setting, highly favorable glycemic outcomes are achievable and are similar to those first reported in pivotal trials. Disclosure J.Sherr: Advisory Panel; Bigfoot Biomedical, Inc., Insulet Corporation, Medtronic, Vertex Pharmaceuticals Incorporated, Cecelia Health, StartUp Health T1D Moonshot, Consultant; Bigfoot Biomedical, Inc., Insulet Corporation, Medtronic, Lilly, Research Support; Insulet Corporation, Medtronic, NIH - National Institutes of Health, Juvenile Diabetes Research Foundation (JDRF), Speaker's Bureau; Insulet Corporation, Zealand Pharma A/S, Lilly, Medtronic. D.Desalvo: Consultant; Dexcom, Inc., Research Support; Insulet Corporation. L.M.Huyett: Employee; Insulet Corporation, Stock/Shareholder; Insulet Corporation. K.Snow: Employee; Insulet Corporation. J.J.Mendez: None. I.Hadjiyianni: Employee; Insulet Corporation, Stock/Shareholder; Insulet Corporation, Eli Lilly and Company. T.T.Ly: Employee; Insulet Corporation. Funding Insulet Corporation

591-P: Elite Cyclists with Type 1 Diabetes (T1D) Show Acceptable Glycemic Excursions during a Time-Trial Performance under High-Definition Transcranial Direct-Current Stimulation (HD-tDCS)

Patients with T1D are typically challenged by glycemic fluctuations during exercise. We recently applied HD-tDCS - a neuromodulation enhancer - both in healthy and T1D elite cyclists to improve endurance performance. Here, in a double-blind, counterbalanced, randomized order, nine elite cyclists (28 ± 3.5 years; BMI: 20.8 ± 1.3 kg/m2; VO2peak: 65.3 ± 1.7 mL/min/kg) with T1D (no complications) underwent either HD-tDCS (F3, F4) or control (SHAM) and completed a constant-load trial (CLT) at 75% of the 2nd ventilatory threshold plus a 15-km cycling time-trial (TT). Comparing the two conditions by real-time continuous glucose monitoring (CGM), subjects showed similar glycemic variability in any part of the session (Figure: bold line = mean; dotted lines = min &amp; max), with a significant effect of time (F=26.32; P&amp;lt;0.0001), but no interaction stimulation x time (F=0.08; P&amp;gt;0.99) nor effect of stimulation (F=0.077; P=0.79). Further, CGM data were stable for the 3 days preceding and following the tests. Given that glycemic control becomes of critical interest not only for health but also for performance goals, these findings suggest that HD-tDCS can be safely used as a performance improvement device in athletes with T1D, and possibly in a wider population of active T1D-subjects. Disclosure R.Codella: None. G.Gallo: None. A.Meloni: None. L.Luzi: Advisory Panel; Eli Lilly and Company, Medtronic, Research Support; Gelesis, Speaker's Bureau; A. Menarini Diagnostics, Amgen Inc., Boehringer Ingelheim and Eli Lilly Alliance, Eli Lilly and Company, Novo Nordisk, Novartis. L.Filipas: None.