bims-glumda Biomed News
on CGM data in management of diabetes
Issue of 2026–09–06
nineteen papers selected by
Mott Given
  1. Optimizing Continuous Glucose Monitoring for Older Adults with Diabetes: From Clinical Evidence to a Practical Implementation Pathway.
  2. Continuous Glucose Monitoring and Driving Safety: Clarifying the Inference From a Survey of Insulin-Treated Adults With Type 2 Diabetes.
  3. Timing of continuous glucose monitoring initiation and long-term glycemic trajectories in pediatric type 1 diabetes.
  4. Supporting continuous glucose monitoring for people with serious mental illness and type 2 diabetes: a co-design study for a structured intervention.
  5. Impact of Hybrid Continuous Glucose Monitoring on Glycemic Control in a Medical Intensive Care Unit.
  6. Associations of Hemoglobin Glycation Index with Glucose Variability and Time in Ranges in Patients with Type 2 Diabetes Undergoing Continuous Glucose Monitoring.
  7. Continuous Glucose Monitoring Reduces Severe Diabetes Distress Independently of Glycemic Improvement in Adults With Type 1 Diabetes: A Prospective Study in a Public Healthcare Setting.
  8. Post-transplant diabetes mellitus in kidney transplant recipients: insights from continuous glucose monitoring.
  9. EMBC Special Issue: PhyNet: A Physiology-Constrained Monotonic Neural Network for Safe and Explainable Blood Glucose Forecasting.
  10. Person-Reported and Continuous Glucose Monitoring (CGM) Recorded Hypoglycaemia in People With Insulin-Treated Type 2 Diabetes and Mild to Moderate Chronic Kidney Disease; Post Hoc Analysis From the Hypo-METRICS Study.
  11. Time in tight range (TITR): emerging evidence and clinical perspectives.
  12. Pre-Bedtime Whey Protein Ingestion Increases Blood Glucose Overnight in Adults With Type 1 Diabetes.
  13. Exercise timing or feeding status? Interpreting glycaemic responses to resistance exercise in type 1 diabetes.
  14. Deep learning for blood glucose prediction: Reproducibility challenges and factors affecting differential performance.
  15. Association of 14-day continuous glucose monitoring-derived postprandial hyperglycemic area with 6-month change in urinary albumin-to-creatinine ratio in patients with newly diagnosed type 2 diabetes mellitus.
  16. Effect of a continuous glucose monitoring-guided personalized lifestyle coaching intervention on glycemic outcomes and variability in patients with type 2 diabetes mellitus: a randomized controlled trial.
  17. Glucose management in children living with type 1 diabetes: results from the BETTER registry.
  18. Glycaemic Changes After Semaglutide Discontinuation in Adults With Type 1 Diabetes Using Automated Insulin Delivery: Analysis of the ADJUST-T1D Extension Study.
  19. When Every Alarm Matters: How Remote Diabetes Monitoring Is Transforming the Role of School Nurses Caring for Students With Type 1 Diabetes.
  1. Diabetes Technol Ther. 2026 Sep 03. 15209156261485635
    Optimizing Continuous Glucose Monitoring for Older Adults with Diabetes: From Clinical Evidence to a Practical Implementation Pathway.
    Tong Yang, Jinghao Cai, Jiaying Ni, Jingyi Lu, Jian Zhou.
      Older adults represent a growing proportion of people with diabetes and often experience hypoglycemia, glycemic variability (GV), multimorbidity, frailty, and treatment-related vulnerabilities that are not fully captured by glycated hemoglobin A1c (HbA1c) alone. Continuous glucose monitoring (CGM) provides time-resolved glucose profiles, trend information, alerts, and remote data sharing, making it well suited to safety-focused geriatric diabetes care.In this narrative review, we synthesize evidence on CGM features, randomized and real-world outcomes, patient and caregiver experience, barriers to sustained use, and practical implementation in older adults. Across studies, the principal benefits of CGM vary by diabetes type and treatment context. In older adults with type 1 diabetes, randomized evidence most consistently supports reduced hypoglycemia exposure and time below range, whereas in basal-insulin-treated type 2 diabetes, CGM primarily improves time in range and reduces hyperglycemia. CGM may also support individualized treatment adjustment by revealing nocturnal hypoglycemia, GV, and postprandial patterns that are missed by intermittent testing.Real-world studies associate CGM use with fewer acute diabetes events and hospitalizations in selected insulin-treated populations, but residual confounding limits causal interpretation. Implementation evidence shows that benefit depends on more than device performance. Usability, skin tolerability, education, alert burden, caregiver workflows, digital literacy, cost, and equitable access strongly influence sustained use.We propose a pragmatic geriatric CGM pathway with an explicit triage algorithm that emphasizes risk-based candidate selection, functional assessment, tailored device and support models, individualized alerts and glycemic goals, focused review of CGM metrics, structured education, and early follow-up. Current evidence supports CGM primarily as a safety and decision-support technology for selected older adults, particularly those using insulin or at elevated hypoglycemia risk. Future studies should include frail and cognitively impaired populations and prioritize severe hypoglycemia, falls, treatment burden, caregiver burden, acute care use, quality of life, and scalable delivery models.
    Keywords:  continuous glucose monitoring; diabetes technology; geriatric care; hypoglycemia; older adults; time below range
    DOI:  https://doi.org/10.1177/15209156261485635
  2. J Diabetes Sci Technol. 2026 Sep 04. 19322968261483747
    Continuous Glucose Monitoring and Driving Safety: Clarifying the Inference From a Survey of Insulin-Treated Adults With Type 2 Diabetes.
    Ngoc-Nguyen Thi Nguyen.
      
    DOI:  https://doi.org/10.1177/19322968261483747
  3. Ann Pediatr Endocrinol Metab. 2026 Sep 03.
    Timing of continuous glucose monitoring initiation and long-term glycemic trajectories in pediatric type 1 diabetes.
    Dong Jun Ha, Hwa Young Kim, Jaehyun Kim.
       Purpose: The long-term impact of continuous glucose monitoring (CGM) initiation on glycemic trajectories in children and adolescents with type 1 diabetes (T1D) remains unclear. We assessed 5-year glycated hemoglobin (HbA1c) trajectories according to CGM initiation timing in Korean children and adolescents with T1D.
    Methods: We retrospectively analyzed patients diagnosed with T1D at ≤18 years between 2015 and 2024 at a single pediatric diabetes center. Patients were categorized by CGM initiation timing (<1 month, 1-<12 months, ≥12 months, or no-CGM use). HbA1c was assessed at prespecified intervals up to 60 months, and adjusted trajectories and rates of change were estimated with linear mixed-effects models accounting for repeated measurements within patients.
    Results: Among 216 patients (mean follow-up, 61.3 ± 34.8 months), 69.9% used CGM and 40.7% initiated within 1 month of diagnosis. Baseline HbA1c was higher in the early-CGM group than in the delayed/no-CGM group (12.5% [IQR, 11.0-13.9] vs 11.1% [IQR, 8.8-13.2]; p<0.001), but converged within 12 months. Median HbA1c at 60 months was 6.8%, 6.8%, 7.7%, and 7.3% in the <1-month, 1-<12-month, ≥12-month, and no-CGM groups; the proportion achieving HbA1c <7.0% fell from 86.4% at 9 months to 47.1% at 60 months in the no-CGM group, which alone showed a rising trajectory in mixed-effects models (+0.27% per year; p<0.001).
    Conclusion: Early CGM initiation after T1D diagnosis was associated with more favorable 5-year HbA1c trajectories compared with delayed initiation or no CGM use, supporting routine early CGM adoption in pediatric T1D management.
    Keywords:  Adolescent; Child; Continuous glucose monitoring; Glycated hemoglobin; Longitudinal studies; Type 1 diabetes mellitus
    DOI:  https://doi.org/10.6065/apem.2652270.135
  4. Front Health Serv. 2026 ;6 1807689
    Supporting continuous glucose monitoring for people with serious mental illness and type 2 diabetes: a co-design study for a structured intervention.
    Jennifer V E Brown, Olivia Matthias, Jessica Hendon, Ramzi A Ajjan, Najma Siddiqi, Ian Kellar, Peter A Coventry.
       Introduction: Individuals with serious mental illness (SMI), such as schizophrenia or bipolar disorder, are at a two- to threefold increased risk of developing type 2 diabetes and face significant health inequalities, including reduced life expectancy. Diabetes self-management in this population is challenging, and existing interventions are poorly suited to their needs. Continuous glucose monitoring (CGM) offers potential benefits but remains underused. This study aimed to co-design a logic model and programme theory for a structured CGM intervention tailored to adults with SMI and type 2 diabetes, as a resource to support the planned future co-production of the content and the implementation of the full intervention.
    Methods: Using experience-based co-design, this study engaged service users, carers, and healthcare professionals (HCPs) in workshops and discussions to identify challenges and facilitators for CGM use. Participants co-developed an outline of candidate intervention components, which were summarised in a logic model and programme theory. The process included synthesising existing evidence, exploring personas, and mapping CGM user journeys. The draft intervention outline was iteratively refined across multiple sessions.
    Results: Ten participants (four HCPs, four service users, and two carers) contributed to the co-design process. Barriers identified included limited technical support, accessibility challenges, and a lack of integrated physical and mental healthcare. Key components for a structured CGM intervention include personalised CGM training, flexible delivery methods, ongoing support, and peer networks. The proposed intervention was conceptualised in three phases: set-up, early days, and living with CGM. Participants prioritised flexibility and tailored intervention adjustments, emphasising the importance of addressing both physical and mental health needs concurrently.
    Discussion: This study demonstrates the feasibility of co-designing interventions with individuals often considered "hard to reach." Our study highlights that people with SMI can take an interest in their physical health and see the potential for CGM to improve diabetes management, provided they receive appropriate and flexible support. Future work will develop and refine the intervention and evaluate its clinical and cost-effectiveness in diverse populations, aiming to address significant unmet needs in diabetes care for this vulnerable group.
    Study registration: https://osf.io/4ae3g/ (2 February 2024).
    Keywords:  co-design; continuous glucose monitoring; intervention development; serious mental illness; type 2 diabetes
    DOI:  https://doi.org/10.3389/frhs.2026.1807689
  5. Endocr Pract. 2026 Sep 04. pii: S1530-891X(26)01138-9. [Epub ahead of print]
    Impact of Hybrid Continuous Glucose Monitoring on Glycemic Control in a Medical Intensive Care Unit.
    Eileen R Faulds, Melanie Natasha Rayan, Molly McNett, Matthew Exline, Laureen Jones, Abigail Shoben, Chyongchiou J Lin, Brooke Lee, Amanie Rasul, Kathleen M Dungan.
       OBJECTIVE: Continuous glucose monitoring (CGM) offers real-time glucose data that may improve dysglycemia detection in critically ill patients; however, evidence supporting its clinical effectiveness in the intensive care unit (ICU) remains limited. The objective of this study was to evaluate whether a nurse-driven hybrid continuous glucose monitoring (CGM) plus point-of-care (POC) validation protocol improves glycemic outcomes compared with standard POC monitoring alone among medical ICU patients receiving continuous intravenous (IV) insulin.
    METHODS: In this single-center prospective intervention cohort study, 100 MICU patients receiving continuous IV insulin were monitored using a hybrid CGM+POC protocol and compared with 100 matched historical controls. The primary outcome was % time in glucose range (70-180 mg/dL) during continuous IV insulin therapy, calculated using POC values (at least every 4 hours) for between-group comparisons. Secondary outcomes included % time in additional glucose ranges, hyperglycemia (>250 and >400 mg/dL), and hypoglycemia (<70 and <54 mg/dL).
    RESULTS: Compared with controls, CGM-managed patients achieved an increased time in range (76%±18 vs 64%±21; p<0.001) with persistent improvement after adjusting for confounders and marked reductions in severe hyperglycemia (>250 mg/dL: 6%±11 vs 17%±18; >300 mg/dL: 2.7±0.6% vs 11.3±1.4%; p<0.001 for both). Hypoglycemic events were infrequent and similar between groups. A total of 3,590 temporally matched CGM-POC pairs were available for analysis.
    CONCLUSION: In this prospective MICU cohort, real-time hybrid CGM use improved glycemic control and reduced severe hyperglycemia without increasing hypoglycemia, supporting further evaluation of structured CGM integration in critical care practice.
    Keywords:  Continuous Glucose Monitoring (CGM); Critical Care; Glucose control; Hyperglycemia; Intensive Care Unit; Nursing; Time in range
    DOI:  https://doi.org/10.1016/j.eprac.2026.08.021
  6. Int J Med Sci. 2026 ;23(9): 2993-2999
    Associations of Hemoglobin Glycation Index with Glucose Variability and Time in Ranges in Patients with Type 2 Diabetes Undergoing Continuous Glucose Monitoring.
    Xiu-Yu Wei, Jun-Sing Wang.
       Background: Discordance between glycated hemoglobin (HbA1c) levels and blood glucose is commonly observed in clinical practice. The hemoglobin glycation index (HGI) has been proposed as a measure to quantify individual variations in glycation. Our study aimed to investigate the association between HGI and glucose variability assessed by continuous glucose monitoring (CGM) in patients with type 2 diabetes (T2D) treated with metformin monotherapy.
    Methods: Adults with T2D treated with metformin monotherapy (daily dose ≥ 1500 mg) who had HbA1c levels ≥ 7.0% (n = 100; mean age 54.0 ± 8.2 years; 52.0% female) undergoing CGM were analyzed. Predicted HbA1c was derived from linear regression of HbA1c on fasting plasma glucose, and the HGI was calculated as observed HbA1c minus predicted HbA1c. CGM metrics-including standard deviation, coefficient of variation, mean amplitude of glycemic excursions (MAGE), and time in ranges-were analyzed to evaluate their associations with HGI.
    Results: HGI was significantly associated with mean sensor glucose (β = 13.60, 95% CI 5.16-22.04, p = 0.002) and CGM-derived glucose variability metrics, including standard deviation (β = 7.53, 95% CI 4.49-10.58, p < 0.001), coefficient of variation (β = 2.22, 95% CI 0.29-4.15, p = 0.025), and MAGE (β = 22.14, 95% CI 14.59-29.69, p < 0.001). These associations remained significant after multivariable adjustment. HGI was also negatively associated with time in range (β = -7.29, 95% CI -12.79 to -1.79, p = 0.010) and positively associated with time above range (β = 7.35, 95% CI 1.69-13.01, p = 0.011).
    Conclusion: These findings highlight a significant association between HGI and glucose variability from CGM, suggesting that HGI may serve as a complementary marker in diabetes management.
    Keywords:  continuous glucose monitoring; glycated hemoglobin; hemoglobin glycation index; type 2 diabetes
    DOI:  https://doi.org/10.7150/ijms.129813
  7. Can J Diabetes. 2026 Sep 04. pii: S1499-2671(26)00349-7. [Epub ahead of print]
    Continuous Glucose Monitoring Reduces Severe Diabetes Distress Independently of Glycemic Improvement in Adults With Type 1 Diabetes: A Prospective Study in a Public Healthcare Setting.
    Nathalia Costa Monteiro, Laís Prado Smith Lima, João Paulo Almeida Freitas, Maria de Lourdes Passos Machado, Fernanda Gonçalves Gama, Victória Dayane Santos Silva, Leonardo Machado Martins, Sophia de Santana Ferreira Lemos, Gutemberg Lemos Torres, Danielle Góes da Silva, Mariana Garcez Varela, Talita Letícia Trevisan, Mônica Sueli Vilela da Mota Silveira, Bruno Fernandes de Oliveira Santos, Nathalie Oliveira de Santana.
       OBJECTIVES: While technology improves glycemic control, its psychological impact in resource-limited settings is less explored. This study evaluated the impact of continuous glucose monitoring (CGM) on diabetes distress and glycemic outcomes in adults with type 1 diabetes (T1D) within a public healthcare system.
    METHODS: This prospective study enrolled 35 adults with T1D at a public center in Northeastern Brazil. Participants initiated CGM with monthly follow-ups. Diabetes distress was assessed at baseline and six months using Brazilian Type 1 Diabetes Distress Scale (T1DDS).
    RESULTS: Participants had a mean T1D duration of 13 ± 9.9 years and baseline HbA1c of 9.13 ± 1.87%. After six months, mean HbA1c decreased to 7.94 ± 1.28% (mean difference = -1.19%; p = 0.007). Time in range reached 54.4% within the first month and remained consistent throughout the follow-up. Total diabetes distress score decreased significantly from 2.86 ± 0.88 at baseline to 2.32 ± 0.85 at six months (mean difference = -0.28; p < 0.001; Cohen's dz = 0.76). The proportion of participants with severe distress dropped from 34.3% (n=12) to 14.3% (n=5) (p = 0.002). Major improvements in management and family/social domains. Psychosocial benefits were observed regardless of the magnitude of HbA1c reduction (r = -0.182, p = 0.384).
    CONCLUSIONS: CGM implementation reduced diabetes distress, particularly severe cases, and improved glycemic parameters in a resource-limited setting. The improvement in well-being independent of glycemic drop suggests that CGM is a valuable tool for both clinical and psychosocial management of adults with T1D in public health systems.
    Keywords:  Continuous glucose monitoring; Diabetes distress; Glycemic control; Low-resource settings; Type 1 diabetes
    DOI:  https://doi.org/10.1016/j.jcjd.2026.08.187
  8. Clin Kidney J. 2026 Sep;19(9): sfag252
    Post-transplant diabetes mellitus in kidney transplant recipients: insights from continuous glucose monitoring.
    Khaled Oweidat, Christopher K Farmer.
      
    DOI:  https://doi.org/10.1093/ckj/sfag252
  9. IEEE Trans Biomed Eng. 2026 Sep 02. PP
    EMBC Special Issue: PhyNet: A Physiology-Constrained Monotonic Neural Network for Safe and Explainable Blood Glucose Forecasting.
    Andrea Calzavara, Francesco Prendin, Giacomo Cappon, Simone Del Favero, Andrea Facchinetti.
       OBJECTIVE: Deep learning (DL) has become state-of-the-art for blood glucose (BG) forecasting in type 1 diabetes (T1D). However, its black-box nature raises safety and reliability concerns regarding its use for therapeutic decision-support. This study aims to: (1) highlight potential risks associated with standard DL-based BG forecasting, and (2) address them with PhyNet, a physiology-constrained monotonic neural network.
    METHODS: Two large-scale datasets (T1DEXI and MetaboNet, 848 subjects in total) were used to develop PhyNet-which enforces physiological consistency through a multi-branch structure and weight constraints-and compare it against six DL baselines (convolutional, recurrent, and transformer-based architectures). Models predicted BG levels up to 90-minute ahead using continuous glucose monitoring (CGM) data, carbohydrate intake, and insulin dosing, and were assessed for: (i) predictive accuracy with standard metrics, and (ii) adherence to physiological principles (i.e., carbohydrates increase BG, insulin lowers it) using explainable AI. Specifically, we evaluated model-predicted responses to varying carbohydrate and insulin intakes, and generated counterfactual explanations to identify model-recommended actions for avoiding adverse events.
    RESULTS: At a 30-minute horizon, predictive accuracy was similar across models (RMSE: 19.39-21.00 mg/dL; Time Gain: 10.45-13.35 min). Despite this, only PhyNet consistently captured the physiological effects of carbohydrates and insulin, yielding 0% unsafe recommendations versus up to 64.3% for baselines.
    CONCLUSION: Standard DL models can achieve state-of-the-art performance while failing to respect physiology, posing clinical risk. PhyNet preserves accuracy while enhancing physiological fidelity, supporting safer integration into T1D technologies.
    SIGNIFICANCE: Evaluating physiological consistency alongside predictive accuracy is essential for responsible clinical translation of DL-based BG forecasting.
    DOI:  https://doi.org/10.1109/TBME.2026.3730324
  10. Diabetes Obes Metab. 2026 Sep 01.
    Person-Reported and Continuous Glucose Monitoring (CGM) Recorded Hypoglycaemia in People With Insulin-Treated Type 2 Diabetes and Mild to Moderate Chronic Kidney Disease; Post Hoc Analysis From the Hypo-METRICS Study.
    HypoRESOLVE Consortium
       BACKGROUND AND AIMS: People with diabetes and chronic kidney disease (CKD) may be at increased risk of iatrogenic hypoglycaemia. We compared rates of sensor-detected (SDH) and person-reported hypoglycaemia (PRH) in people with insulin-treated Type 2 diabetes (T2D) with and without CKD in the Hypo-METRICS study.
    MATERIALS AND METHOD: Insulin-treated T2D patients with > 1 hypoglycaemic episode within the last 3 months wore a blinded continuous glucose monitor (CGM) and recorded PRH on the bespoke Hypo-METRICS app for 10 weeks. SDH was defined as per consensus guidelines. The participants were grouped based on estimated glomerular filtration rate (eGFR), that is CKD 1 (eGFR ≥ 90 mL/min; n = 117), CKD 2 (eGFR 60-89 mL/min; n = 140), and CKD 3 (eGFR 30-59 mL/min; n = 56). We used descriptive statistics, Kruskal-Wallis rank sum tests, and a negative binomial models to control for confounders.
    RESULTS: In total, 313 participants with T2DM were included. Median age increased with declining eGFR (p < 0.001). Median (IQR) events per week for SDH < 3.9 mmol/L were 1.72 (0.66-3.43) in CKD 1, 2.12 (0.85-4.55) in CKD 2, and 2.22 (1.09-4.57) in CKD 3 (p = 0.12). For SDH < 3.0 mmol/L, these rates were 0.20 (0.00-0.50), 0.21 (0.10-0.61), and 0.20 (0.00-0.71), respectively (p = 0.88). PRH events had medians of 1.05 (0.40-2.12), 1.31 (0.61-2.42), and 1.28 (0.68-2.25) for CKD 1, CKD 2, and CKD 3 (p = 0.24).
    CONCLUSION: In people with insulin-treated T2DM and CKD stages 2 or 3, the rates of SDH and PRH do not significantly differ from those in people without CKD.
    TRIAL REGISTRATION: ClinicalTRials.gov: NCT04304963.
    Keywords:  Type 2 diabetes; clinical trial; continuous glucose monitoring (CGM); diabetic nephropathy; hypoglycaemia; insulin therapy
    DOI:  https://doi.org/10.1111/dom.71247
  11. Diabetol Int. 2026 Oct;17(4): 72
    Time in tight range (TITR): emerging evidence and clinical perspectives.
    Takayuki Ohno.
      Continuous glucose monitoring (CGM) has enabled increasingly stringent assessment of glycemic control beyond conventional time in range (TIR; 70-180 mg/dL). Time in tight range (TITR), defined as the percentage of time with glucose levels between 70 and 140 mg/dL, was proposed in 2023 as a more stringent CGM-derived metric. This review summarizes current evidence regarding the clinical significance and limitations of TITR. TITR correlates strongly with HbA1c and TIR, and may be more sensitive than TIR to changes in mean glucose when glycemia approaches the normal range. Recent cohort studies suggest that lower TITR is linearly associated with higher risks of all-cause and cardiovascular mortality, while higher TITR is associated with lower incidence of diabetic retinopathy. TITR has also shown associations with favorable outcomes in COVID-19 pneumonia. However, its interpretation requires caution because glycemic variability can influence TITR differently according to mean glucose levels. In individuals with type 1 diabetes treated with multiple daily injections, higher TITR may be accompanied by increased time below range, suggesting that indiscriminate pursuit of higher TITR could increase hypoglycemia risk. Further prospective studies are needed to establish optimal TITR targets and determine whether TITR provides prognostic value beyond established CGM metrics across diverse populations.
    Keywords:  CGM; Hypoglycemia; TBR; TIR; TITR
    DOI:  https://doi.org/10.1007/s13340-026-00923-4
  12. Diabetes Obes Metab. 2026 Sep 03.
    Pre-Bedtime Whey Protein Ingestion Increases Blood Glucose Overnight in Adults With Type 1 Diabetes.
    Sam Zhao, Giang M Dao, Cara Schofield, Sara Vogrin, Declan T Hennessy, Carmel E Smart, Dessi P Zaharieva, David N O'Neal, Clinton R Bruce, Greg M Kowalski, Dale J Morrison.
      
    Keywords:  Type 1 diabetes; continuous glucose monitoring (CGM); dietary intervention; hypoglycaemia
    DOI:  https://doi.org/10.1111/dom.71307
  13. Diabetologia. 2026 Aug 31.
    Exercise timing or feeding status? Interpreting glycaemic responses to resistance exercise in type 1 diabetes.
    Stuart R Gray, Ameenah Al-Awadhi, Ebaa Al Ozairi.
      
    Keywords:  Afternoon exercise; Blood glucose; Continuous glucose monitoring; Fasted exercise; Fed exercise; Glycaemic response; Hyperglycaemia; Hypoglycaemia; Resistance training; Timing of exercise; Type 1 diabetes
    DOI:  https://doi.org/10.1007/s00125-026-06850-y
  14. PLOS Digit Health. 2026 Sep;5(9): e0001633
    Deep learning for blood glucose prediction: Reproducibility challenges and factors affecting differential performance.
    Baiying Lu, Biratal Wagle, Zhaohui Liang, Yanjun Cui, Temiloluwa Prioleau.
      Blood glucose prediction is a critical component of next-generation diabetes technologies, such as artificial pancreas systems, where reliable performance is essential for safety and effectiveness. Although deep learning methods have achieved promising advances in this area, a critical gap remains in understanding the reproducibility and generalizability of these methods. To contextualize the gap, this study reviewed 67 recent papers that proposed a deep learning method for glucose prediction to identify key reproducibility challenges. Next, we adopted a standardized framework, encompassing technical, statistical, and conceptual reproducibility evaluations, to experimentally assess the reproducibility of eight representative deep learning methods. To achieve this, we reimplemented and evaluated these eight deep learning methods using over 1.36 million continuous glucose monitoring samples (5,061 days) from 128 individuals with type 1 diabetes across three public datasets: OhioT1DM, DiaTrend, and T1DEXI. We found that even though these models demonstrated good technical and statistical reproducibility, their conceptual reproducibility-the ability to generalize to datasets with different diabetes management patterns-was limited. Further analyses revealed that each model's overall prediction performance was strongly influenced by individual glycemic control, with higher prediction errors observed among participants with lower time with blood glucose in the target range (70-180 mg/dL). This study identified key reproducibility challenges associated with current blood glucose prediction methods within type 1 diabetes populations, highlighting the need for increased transparency, dataset diversity, standardized evaluation practices, and code accessibility to ensure reproducible and reliable models for blood glucose prediction.
    DOI:  https://doi.org/10.1371/journal.pdig.0001633
  15. Front Endocrinol (Lausanne). 2026 ;17 1890463
    Association of 14-day continuous glucose monitoring-derived postprandial hyperglycemic area with 6-month change in urinary albumin-to-creatinine ratio in patients with newly diagnosed type 2 diabetes mellitus.
    Tianchi Yu.
       Background: Early kidney involvement may already be present at the diagnosis of type 2 diabetes mellitus (T2DM), whereas glycated hemoglobin A1c (HbA1c) and fasting plasma glucose do not fully characterize postprandial exposure. We evaluated the association of continuous glucose monitoring (CGM)-derived postprandial hyperglycemic area (PPHA) with 6-month change in urinary albumin-to-creatinine ratio (UACR) in adults with newly diagnosed T2DM.
    Methods: This single-center retrospective study consecutively screened 381 adults with newly diagnosed diabetes who underwent professional CGM from January 1, 2020, to December 31, 2024. After predefined exclusions, 297 participants with valid 14-day CGM and baseline and 6-month UACR measurements were analyzed. PPHA was the daily mean threshold-excess area above 10.0 mmol/L during 0-4 h after each main meal, calculated from valid monitoring days beginning on the first complete day after all components of the initial treatment regimen had been started. The primary outcome was ΔlnUACR [ln(6-month UACR + 1)-ln(baseline UACR + 1)]; exploratory UACR worsening was secondary. Multivariable regression, restricted cubic spline analyses, treatment-interaction analyses, and sensitivity analyses were performed.
    Results: UACR worsening occurred in 88 participants (29.6%). ΔlnUACR and the frequency of UACR worsening showed an increasing descriptive pattern across PPHA quartiles, with significant overall between-group differences (both P < 0.001). After adjustment for age, sex, body mass index, systolic blood pressure, HbA1c, fasting plasma glucose, estimated glomerular filtration rate, and baseline lnUACR, each 5 mmol·h·L-¹·d-¹ increase in PPHA was associated with a 0.08 higher ΔlnUACR (95% confidence interval [CI], 0.05-0.12; P < 0.001) and higher odds of UACR worsening (odds ratio [OR], 1.73; 95% CI, 1.27-2.36; P < 0.001). The association was approximately linear (P for nonlinearity=0.932), remained directionally consistent after simultaneous adjustment for recorded chronic comorbidities, and showed no statistically significant interaction with initial metformin, insulin, sodium-glucose cotransporter 2 inhibitor, or glucagon-like peptide-1 receptor agonist therapy (all interaction P>0.05).
    Conclusions: Higher CGM-derived PPHA was associated with greater 6-month change in UACR after adjustment for fasting plasma glucose and HbA1c. PPHA may complement conventional glycemic measures in early kidney-risk assessment; external validation is required.
    Keywords:  albuminuria; continuous glucose monitoring; diabetic kidney disease; postprandial hyperglycemia; type 2 diabetes mellitus
    DOI:  https://doi.org/10.3389/fendo.2026.1890463
  16. J Korean Acad Nurs. 2026 Aug;56(3): 393-407
    Effect of a continuous glucose monitoring-guided personalized lifestyle coaching intervention on glycemic outcomes and variability in patients with type 2 diabetes mellitus: a randomized controlled trial.
    Youngran Yang, Youngmin Park, Heung Yong Jin.
       Purpose: This study examined the effects of continuous glucose monitoring (CGM)-guided personalized lifestyle coaching on glycemic outcomes and glycemic variability in patients with type 2 diabetes mellitus (T2DM).
    Methods: This three-arm randomized controlled trial was conducted in South Korea from June 2023 to November 2024. Participants were assigned to CGM alone (Intervention I, n=34), CGM-guided personalized lifestyle coaching (Intervention II, n=34), or standard diabetes education (control, n=34). The intervention lasted 3 months, with assessments at baseline and at 3, 6, and 12 months. Data were analyzed using generalized estimating equations.
    Results: At 3 months, glycated hemoglobin (HbA1c) and fasting plasma glucose (FPG) levels were significantly lower in both intervention groups than in the control group. Compared with the control group, HbA1c was lower by 0.78 percentage points in Intervention I (p=.008) and by 1.12 percentage points in Intervention II (p<.001). At 3 months, FPG was lower than in the control group by 34.05 mg/dL in Intervention I (p=.001) and by 34.77 mg/dL in Intervention II (p<.001). At 6 months, TBR70 (time below range <70 mg/dL) was lower in Intervention I than in Intervention II (adjusted mean difference=-2.18, p=.010), whereas TAR180 (time above range >180 mg/dL) was lower in Intervention II than in the control group and Intervention I by 7.85 and 8.28 percentage points, respectively (both p=.004).
    Conclusion: CGM-guided personalized lifestyle coaching improved glycemic control and reduced hyperglycemic exposure in patients with T2DM. Real-time glucose data may support individualized lifestyle counseling and sustained diabetes self-management. This study was retrospectively registered with the Clinical Research Information Service (CRIS) of the Republic of Korea (KCT0008872) on 16 October 2023.
    Keywords:  Blood glucose self-monitoring; Diabetes mellitus; Glycemic control; Randomized controlled trial
    DOI:  https://doi.org/10.4040/jkan.26036
  17. Horm Res Paediatr. 2026 Sep 02. 1
    Glucose management in children living with type 1 diabetes: results from the BETTER registry.
    Virginie Messier, Mélanie Henderson, Meranda Nakhla, Judith Simoneau-Roy, Meryem Talbo, Domitille Dervaux, Nathalie Kinnard, Andréane Vanasse, Monia Rekik, Remi Rabasa-Lhoret, Anne-Sophie Brazeau.
       AIMS: The Behaviors, Therapies, TEchnologies and hypoglycemic Risk in Type 1 diabetes (BETTER) registry collects patient-reported outcomes and experiences on individuals living with type 1 diabetes in Canada. The objective of our study was to describe the management of type 1 diabetes reported by parents of children and young adolescents (< 14 years of age) enrolled in the BETTER registry.
    MATERIALS AND METHODS: We conducted a cross-sectional evaluation using self-reported data collected from April 2019 to October 2024. Parents of children and adolescents < 14 years self-registered and provided written online informed consent. As of October 2024, 699 children and adolescents < 14 years of age and living with type 1 diabetes were enrolled.
    RESULTS: Children had a mean age of 8.8 ± 3.1 years old with a diabetes duration of 2.3 ± 2.7 years, 44.9% were female, and 85.0% were of European ancestry. More than 80% used a continuous glucose monitoring system and 38.3% used an insulin pump. Most recent HbA1c ≤ 7% was reported by 15.9% of participants. At least 1 episode of level 2 hypoglycemia (glucose levels < 3.0 mmol/L, with ability to self-treat) in the last month was reported by 72.5% of participants, with a median number of episodes (interquartile range) of 5 (2, 10). The occurrence of level 3 hypoglycemia (low glucose, requiring help from another person, use of glucagon, hospitalization, or loss of consciousness) in the last 12 months was reported by 8.6% of participants. Among these, the median number of episodes was 1 (1, 3).
    CONCLUSIONS: Our results indicate that, despite a wide use of continuous glucose monitoring, the occurrence of hypoglycemia is still high and the proportion of children and adolescents reaching the HbA1c target is low.
    DOI:  https://doi.org/10.1159/hrp/adaag016
  18. Diabetes Obes Metab. 2026 Aug 31.
    Glycaemic Changes After Semaglutide Discontinuation in Adults With Type 1 Diabetes Using Automated Insulin Delivery: Analysis of the ADJUST-T1D Extension Study.
    Eslam Montaser, Kanchan Bhasin, Davida Kruger, Halis Kaan Akturk, Andrew Ahmann, Janet K Snell-Bergeon, Viral N Shah.
       AIMS: To evaluate changes in glycaemic control following semaglutide discontinuation in adults with Type 1 diabetes and obesity who completed the ADJUST-T1D randomised controlled trial.
    MATERIALS AND METHODS: Continuous glucose monitoring (CGM) data were collected during a 12-week extension phase after the ADJUST-T1D trial. Analyses included participants previously randomised to semaglutide who discontinued treatment (n = 16) and participants previously randomised to placebo who did not receive a glucagon-like peptide-1 receptor agonist during follow-up (n = 22). Changes in CGM-derived metrics from Week 26 to the extension phase were compared using Wilcoxon rank-sum tests with Benjamini-Hochberg adjustment. Analysis of covariance (ANCOVA) adjusting for Week 26 values was performed as a sensitivity analysis.
    RESULTS: Compared with placebo, participants who discontinued semaglutide experienced a greater decline in time in range (70-180 mg/dL) (median change -3.6% vs. 1.5%; adjusted p = 0.044). Semaglutide discontinuation was also associated with greater increases in glucose standard deviation (6.0 vs. 1.2 mg/dL; adjusted p = 0.044) and coefficient of variation (3.1% vs. 1.1%; adjusted p = 0.044), indicating worsening glycaemic variability. Time above range (> 180 mg/dL) increased numerically but did not remain significant after adjustment (adjusted p = 0.064). No significant between-group differences were observed in hypoglycaemia-related CGM metrics. ANCOVA analyses showed consistent findings.
    CONCLUSIONS: In adults with Type 1 diabetes and obesity, semaglutide discontinuation was associated with a greater decline in time in range and increased glycaemic variability compared with placebo over 12 weeks. These findings suggest that glycaemic benefits achieved during treatment may not be fully sustained following withdrawal.
    Keywords:  GLP‐1 receptor agonist; Type 1 diabetes; automated insulin delivery; continuous glucose monitoring; glycaemic variability
    DOI:  https://doi.org/10.1111/dom.71287
  19. J Sch Nurs. 2026 Sep 03. 10598405261483764
    When Every Alarm Matters: How Remote Diabetes Monitoring Is Transforming the Role of School Nurses Caring for Students With Type 1 Diabetes.
    Debby Syahru Romadlon, Diana Barsasella, Rudy Kurniawan, Renny Wulan Apriliyasari, Safiruddin Al Baqi, Andi Safutra Suraya, Muh Gunawan.
      Remote diabetes monitoring has transformed Type 1 Diabetes (T1D) management in schools, yet its impact on school nursing practice remains poorly understood. This study explored how remote diabetes monitoring is reshaping the role of school nurses caring for students with T1D in Indonesia. Using an Interpretive Description approach, semistructured interviews were conducted with 20 school nurses from 15 schools between November 2025 and February 2026. Data were analyzed inductively. Five themes emerged: living in a state of constant digital vigilance, negotiating responsibility in real-time care, managing parental expectations in an era of shared data, redefining professional boundaries and clinical decision-making, and transforming the school nurse role in the digital era. Findings indicate that remote diabetes monitoring extends school nursing beyond episodic care, fostering more proactive, data-driven, and continuously connected roles. Training, policy development, and clearer role expectations are needed as digital health technologies become increasingly integrated into school-based diabetes care.
    Keywords:  clinical decision-making; continuous glucose monitoring; professional role transformation; remote diabetes monitoring; school nursing; type 1 diabetes
    DOI:  https://doi.org/10.1177/10598405261483764
This page is being maintained by Thomas Krichel. It was last updated on 2026‒09‒06 at 0:05.