Diabetes Management Outcomes Research Articles

Multiomics and Artificial Intelligence for Personalized Nutritional Management of Diabetes in Patients Undergoing Peritoneal Dialysis.

Managing diabetes in patients on peritoneal dialysis (PD) is challenging due to the combined effects of dietary glucose, glucose from dialysate, and other medical complications. Advances in technology that enable continuous biological data collection are transforming traditional management approaches. This review explores how multiomics technologies and artificial intelligence (AI) are enhancing glucose management in this patient population. Continuous glucose monitoring (CGM) offers significant advantages over traditional markers, such as hemoglobin A1c (HbA1c). Unlike HbA1c, which reflects an average glucose level, CGM provides real-time, dynamic glucose data that allow clinicians to make timely adjustments, leading to better glycemic control and outcomes. Multiomics approaches are valuable for understanding genetic factors that influence susceptibility to diabetic complications, particularly those related to advanced glycation end products (AGEs). Identifying genetic polymorphisms that modify a patient's response to AGEs allows for personalized treatments, potentially reducing the severity of diabetes-related pathologies. Metabolomic analyses of PD effluent are also promising, as they help identify early biomarkers of metabolic dysregulation. Early detection can lead to timely interventions and more tailored treatment strategies, improving long-term patient care. AI integration is revolutionizing diabetes management for PD patients by processing vast datasets from CGM, genetic, metabolic, and microbiome profiles. AI can identify patterns and predict outcomes that may be difficult for humans to detect, enabling highly personalized recommendations for diet, medication, and dialysis management. Furthermore, AI can assist clinicians by automating data interpretation, improving treatment plans, and enhancing patient education. Despite the promise of these technologies, there are limitations. CGM, multiomics, and AI require significant investment in infrastructure, training, and validation studies. Additionally, integrating these approaches into clinical practice presents logistical and financial challenges. Nevertheless, personalized, data-driven strategies offer great potential for improving outcomes in diabetes management for PD patients.

Knowledge and Practice in Insulin Therapy in Patients with Diabetes Mellitus

This study examines the knowledge, practices, and barriers affecting insulin therapy adherence in diabetes patients. Through a qualitative approach, semi-structured interviews were conducted with 100 participants currently on insulin therapy, with responses analysed thematically. Key findings highlighted four themes: healthcare support, psychological and social challenges, economic constraints, and knowledge gaps. While most participants demonstrated adequate knowledge of insulin administration and blood glucose monitoring, gaps were evident in insulin storage and injection site rotation, potentially impacting effectiveness. Psychological barriers such as injection anxiety, social stigma, and fear of long-term dependency significantly affected adherence. Additionally, financial burdens, especially among rural patients, and limited healthcare access were major obstacles. Findings emphasize the need for expanded support through education, counselling, and financial assistance to improve adherence. By addressing these barriers, healthcare providers can enhance diabetes management and patient outcomes.

Experimental and Theoretical Design on the Development of Matrix Tablets with Multiple Drug Loadings Aimed at Optimizing Antidiabetic Medication.

Background: Diabetes is a growing global health crisis that requires effective therapeutic strategies to optimize treatment outcomes. This study aims to address this challenge by developing and characterizing extended-release polymeric matrix tablets containing metformin hydrochloride (M-HCl), a first-line treatment for type 2 diabetes, and honokiol (HNK), a bioactive compound with potential therapeutic benefits. The objective is to enhance glycemic control and overall therapeutic outcomes through an innovative dual-drug delivery system. Methods: The tablets were formulated using hydrophilic polymers, such as Carbopol® 71G NF and Noveon® AA-1. The release kinetics of M-HCl and HNK were investigated through advanced mathematical models, including fractal and multifractal dynamics, to capture the non-linear and time-dependent release processes. Traditional kinetic models (zero-order, first-order, Higuchi equations) were also evaluated for comparison. In vitro dissolution studies were conducted to determine the release profiles of the active ingredients under varying polymer concentrations. Results: The study revealed distinct release profiles for the two active ingredients. M-HCl exhibited a rapid release phase, with 80% of the drug released within 4-7 h depending on polymer concentration. In contrast, HNK demonstrated a slower release profile, achieving 80% release after 9-10 h, indicating a greater sensitivity to polymer concentration. At shorter intervals, drug release followed classical kinetic models, while multifractal dynamics dominated at longer intervals. Higher polymer concentrations resulted in slower drug release rates due to the formation of a gel-like structure upon hydration, which hindered drug diffusion. The mechanical properties and stability of the matrix tablets confirmed their suitability for extended-release applications. Mathematical modeling validated the experimental findings and provided insights into the structural and time-dependent factors influencing drug release. Conclusions: This study successfully developed dual-drug extended-release matrix tablets containing metformin hydrochloride and honokiol, highlighting the potential of hydrophilic polymers to regulate drug release. The findings emphasize the utility of advanced mathematical models for predicting release kinetics and underscore the potential of these formulations to improve patient compliance and therapeutic outcomes in diabetes management.

A CROSS-SECTIONAL STUDY TO INVESTIGATE COST-RELATED NON-ADHERENCE AMONG DIABETIC PATIENTS IN TWIN CITIES

Background: Medication nonadherence is a significant global health issue, particularly among diabetic patients, and is associated with worsened health outcomes and increased healthcare costs. In Pakistan, where diabetes prevalence is high, socioeconomic disparities exacerbate adherence challenges. Cost-related barriers, forgetfulness, and complex medication regimens are commonly cited reasons for nonadherence. Understanding the influence of socioeconomic factors such as income, education, and insurance coverage on medication adherence is essential to developing effective strategies for improving diabetes management. Objective: To investigate the correlation between socioeconomic factors (income levels, education, and insurance coverage) and medication adherence among diabetic patients in Rawalpindi and Islamabad. Methods: A cross-sectional study was conducted involving 300 diabetic patients from Rawalpindi and Islamabad. Data were collected using a validated self-administered questionnaire consisting of six sections, including demographics, adherence challenges, and coping strategies. Statistical analysis, including chi-square tests, was performed using SPSS version 23.0 to identify significant associations (p<0.05). Ethical approval was obtained from the university's ethical review board. Results: Among the 300 participants, 45% reported nonadherence to medication regimens, primarily due to financial concerns (37.3%) and forgetfulness (50.3%). The majority (61%) were aged between 30 and 50 years, and 79.3% sought alternative financial support, while 80.3% were aware of community resources. Significant associations were observed between age and key adherence factors, including cost-related challenges (p<0.001). Despite awareness and financial assistance, adherence barriers persisted. Conclusion: This study highlights the critical impact of socioeconomic factors on medication adherence among diabetic patients in Rawalpindi and Islamabad. Addressing nonadherence requires a multifaceted approach, including enhanced financial support, personalized counseling, and innovative interventions such as healthcare reminders, to improve diabetes management outcomes.

Robust diabetic prediction using ensemble machine learning models with synthetic minority over-sampling technique

This paper addresses the pressing issue of diabetes, which is a widespread condition affecting a huge population worldwide. As cells become less responsive to insulin or fail to produce it adequately, blood sugar levels rise. This has the potential to cause severe health complications including kidney disease, vision impairment and heart conditions. Early diagnosis is paramount in mitigating the risk and severity of diabetes-related complications. To tackle this, we proposed a robust framework for diabetes prediction using Synthetic Minority Over-sampling Technique (SMOTE) with ensemble machine learning techniques. Our approach incorporates strategies such as imputation of missing values, outlier rejection, feature selection using correlation analysis and class distribution balancing using SMOTE. The extensive experimentation shows that the proposed combination of AdaBoost and XGBoost shows exceptional performance, with an impressive AUC of 0.968+/-0.015. This outperforms not only alternative methodologies presented in our study but also surpasses current state-of-the-art results. We anticipate that our model will significantly improve diabetes prediction, offering a promising avenue for improved healthcare outcomes in diabetes management.

Harnessing Digital Initiatives for Improved Health Outcomes in Diabetes Management: An Observational Patient Program.

Patients with diabetes have easy access to a wide range of digital applications that may help with self-management and lower barriers; however, robust evidence of their effectiveness remains somewhat elusive. Zyla is a medical artificial intelligence (AI)-based personalized care management app that assists the treating physician in improving the standard of patient care by offering the patients comprehensive and individualized care. This preliminary evaluation of data collected through the Zyla app aims to understand the impact of diabetes disease outcomes among patients subscribed to this app. This was a retrospective, observational program conducted through the Zyla app in the calendar year 2020. The Zyla app's objective is to assist the treating physician in improving the standard of patient care by giving them the choice of assembling a personalized team (consisting of clinical nutritionists, physiotherapists, and counselors over a virtual platform) that can offer patients comprehensive and individualized care. Data on parameters like glycated hemoglobin (HbA1c), fasting blood sugar (FBS), post-prandial glucose (PPG), serum creatinine (SC), total cholesterol (TC), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C) and low-density lipoprotein cholesterol (LDL-C) were collected through the Zyla app. Clinical outcomes assessed were the change from baseline to last reported levels of the mentioned parameters and are reported using descriptive analysis. The glycemic control parameters, HbA1c (change from baseline (CFB): -1.08), FBS (CFB: -15.93), and PPG levels (-18.42), were significantly lower (P<0.0001) at the last assessment compared with baseline. For the lipid profile, levels of TGs (P<0.0001) and TC (P = 0.0037) were significantly lower compared with baseline, while HDL-C levels were comparatively higher (CFB: 0.68) and LDL-C levels were lower (CFB:11.60), however non-significant. Serum creatinine was also lower compared to baseline (CFB: -0.25); however, the difference was not statistically significant. A significant improvement in all glycemic parameters was seen with the use of the Zyla app along with numerical improvements in kidney function parameters and cholesterol status among patients. These preliminary findings warrant further rigorous studies to validate the impact of medical apps in the management of diabetics in India.

Picking the low-hanging fruit: the cost-effectiveness of opportunistic diabetes screening during tuberculosis contact investigations on the Texas–Mexico border

ObjectiveThere is a known association between type 2 diabetes (diabetes) and tuberculosis (TB), and TB clinics have become hubs for new diabetes diagnosis among active patients with TB. However, despite...

Exploring the perceptions and experiences of diabetes management and self-regulation in young adults using a continuous glucose monitoring device

Young adults face many pressures as they start being independent and responsible for their own actions and decisions. The use of continuous glucose monitoring has been shown to improve medical outcomes in diabetes management, but the mechanism of these improvements needs further investigation. This study aimed to examine the self-regulatory experiences and behaviours of a group of young adults using continuous glucose monitoring to manage type 1 diabetes. An exploratory qualitative approach was taken, using six virtual semi-structured interviews with young adults living with type 1 diabetes and using continuous glucose monitoring. A thematic analysis was used to identify themes from the gathered data. In this study, three themes demonstrated reciprocal determinism in that the observation of the continuous glucose monitoring results led to changes in the behaviours of the participants. The concepts of feedback and feedforward were also evident in the information provided by continuous glucose monitoring and the adjustment of the participants to achieve their personal blood glucose goals. The results of this study thus support the fact that self-regulation of those living with type 1 diabetes can be improved using continuous glucose monitoring as a tool for observation of blood glucose fluctuations and can, as a result, lead to changes in diabetes management behaviours and health outcomes. Continuous glucose monitoring provides the opportunity to observe the consequences of diabetes management behaviours, which activates the self-regulation processes of feedback and feedforward. This process has the potential to improve diabetes management behaviours in this vulnerable group.

Assessing the impact of social determinants of health on diabetes severity and management.

Adverse Social Determinants of Health (SDoH) are considered major obstacles to effective management of type-2 diabetes. This study aims to quantify the impact of SDoH factors on diabetes management outcomes. We quantified the joint impact of multiple SDoH by applying a self-control case series method-which accounts for confounding by using individuals as their own control-to electronic health record data from an academic health system in Maryland. We found a consistent increase in HbA1c levels associated with SDoH across alternative study designs. The estimated total contributions of SDoH ranged 0.014-0.065 across the alternative designs. Transportation issues demonstrated particularly significant contributions, with estimates of 0.077-0.144. When assuming SDoH's risk window to be ±45 days, for example, the total contribution was estimated to be 0.065 (95% CI [0.010, 0.120]) increase in HbA1c and the transportation issues' contribution 0.134 (95% CI [0.020, 0.249]). Our result suggests that reducing transportation barriers may be an effective SDoH intervention strategy for diabetes management; however, the clinical impact of such interventions warrants further investigation.

The impact of patient support and health education on diabetes management and glycemic control.

Diabetes mellitus is a chronic and complex medical condition that can lead to significant morbidity and mortality. Maintaining an adequate blood glucose level is important for patients with diabetes, and to improve glycemic control, patients need proper support and health education, which are essential components of comprehensive diabetes care. We used a rigorous approach based on the PRISMA and Cochrane Handbook principles, specifically focusing on randomized controlled trials (RCTs) published in English from 2005 onwards. The statistical analysis was conducted using the RevMan software. Pooled risk ratios were calculated for dichotomous data, whereas mean differences were calculated for continuous data. Heterogeneity and publication bias were also evaluated. From an initial pool of 544 records, 368 studies were examined after eliminating duplicates. Ultimately, 24 studies were deemed suitable based on the inclusion criteria. These studies involved 2437 participants in the intervention group and 2305 people in the control group. The quality assessment indicated that 41.7% of the studies were categorized as low risk, 16.7% as high risk, and 41.7% had certain concerns regarding bias. The analyses revealed noteworthy decreases in HbA1c levels in the intervention group at several time points, particularly showing improvements after 3 months. Egger's regression indicated the presence of possible publication bias. The results emphasize the crucial impact of health education and mentorship interventions on enhancing glycemic control in individuals with type 2 diabetes. Interventions focused on empowering patients proved to be especially effective in enhancing diabetes management outcomes.

Criteria for Personalised Choice of a Continuous Glucose Monitoring System: An Expert Opinion.

Despite the growing evidence supporting the outpatient use of continuous glucose monitoring (CGM) for improving glycaemic control and reducing hypoglycaemia, there is a need for a detailed understanding of the specific features of CGM devices that best meet individual patient needs. This expert opinion, based on a comprehensive literature review and the personal perspectives of clinicians, aims to provide thehealthcare professionals (HCPs) with a comprehensive framework for selecting CGM devices. It evaluates the current state of CGM technology, categorizing features into essential features, major drivers of choice, and additional/useful features. Moreover, the practical model presented outlines a patient's journey with CGM, emphasising the importance of aligning device features with patient needs. This includes understanding the patient's lifestyle, clinical conditions, and personal preferences to optimize CGM use and improve diabetes management outcomes.

The Impact of Telemedicine during the COVID-19 Pandemic on Diabetes Management and Pregnancy Outcomes in Women with Gestational Diabetes Mellitus (GDM)

Background/Objectives: Gestational diabetes mellitus (GDM) can lead to various complications for both the mother and the child. Many factors influence the onset of the disease including GDM in a previous pregnancy, overweight and obesity, as well as the increasing age of women who become pregnant. The aim of this study was to assess the impact of telemedicine during the COVID-19 pandemic on diabetes management and pregnancy outcomes in women with gestational diabetes mellitus (GDM). Methods: A retrospective analysis was conducted. We compared two groups of GDM women from the pandemic and pre-pandemic periods in terms of pregnancy course and obstetric outcomes. Results: It was observed that women with GDM in the pandemic group were significantly more often overweight and significantly less often obese before pregnancy than women in the pre-pandemic group. GDM was diagnosed significantly earlier but in later pregnancies in the pandemic group than in the pre-pandemic group. The course of pregnancy in both groups was similar and the percentage of pregnancies with the delivery of neonates &gt; 4000 g did not differ significantly. The number of caesarean sections and spontaneous deliveries was also similar in both periods. No differences were found in anthropometric parameters in newborns and neonatal and maternal complications. Conclusions: The occurrence of the COVID-19 pandemic and the necessity of employing telemedicine tools in the medical care of women with GDM did not significantly impact diabetes management and the pregnancy outcomes achieved.

Investigating the Health Diet Needs of Diabetic Elderly in Daily Life: A Systematic Review

Diabetes has increasingly affected populations over the decades, particularly among elderly groups. Maintaining a healthy diet is crucial for managing diabetes, yet elderly people with diabetes often exhibit poor dietary habits. Therefore, this paper aims to investigate the dietary needs of elderly individuals with diabetes to provide comprehensive guidance for improving their dietary habits and overall diabetes management. A systematic review approach was adopted in this study, three reference databases, Scopus, Web of Science, and PubMed were consulted; 28 articles were screened in the final data code and analysis. Results found that four important needs among diabetic elderly include healthy diet knowledge, dietary suggestions, glucose monitoring, and physical activity. Those findings reveal the significance of diabetes management, underscoring the need for targeted educational interventions. Personalized nutrition plans and user-friendly glucose monitoring technologies are essential for improving dietary adherence and blood glucose control. Barriers to regular physical activity, including mobility issues and fear of injury, highlight the need for tailored, enjoyable exercise programs. Addressing these issues through holistic and individualized approaches can enhance diabetes management and health outcomes for this population. The study emphasizes the need for healthcare providers to develop and implement personalized nutrition plans, user-friendly glucose monitoring devices, and tailored exercise programs to support better diabetes management in the elderly population. Future research should focus on creating and testing these user-friendly products to further enhance care for this vulnerable group.

Psychological and Clinical Challenges in the Management of Type 1 Diabetes during Adolescence: A Narrative Review.

Adolescence, a critical period of physical and psychological development, presents unique challenges in type 1 diabetes (T1D) management due to endocrinological changes, reduced therapeutic adherence, and elevated susceptibility to psychological issues such as depression, anxiety, and eating disorders. This narrative review explores the impact of psychological and behavioral factors on glycemic control in adolescents with T1D. We examine the prevalence and influence of mental health disorders, lifestyle factors, harmful behaviors, and social dynamics on diabetes management and glycemic outcomes. Strategies for improving metabolic control are also reviewed, including cognitive behavioral therapy, technological devices, and educational interventions. The importance of tailored psychological support, family involvement, and targeted interventions to improve adherence to treatment and glycemic control in adolescents with T1D should be emphasized.

Continuous glucose monitoring among nurse practitioners in primary care: Characteristics associated with prescribing and resources needed to support use.

Continuous glucose monitoring (CGM) can improve health for people with diabetes but is limited in primary care (PC). Nurse Practitioners (NPs) in PC can improve diabetes management through CGM, but NPs' interest in CGM and support needed are unclear. We describe behaviors and attitudes related to CGM for diabetes management among NPs in PC. This cross-sectional web-based survey of NPs practicing in PC settings used descriptive statistics to describe CGM experience and identify resources to support prescribing. We used multivariable regression to explore characteristics predicting prescribing and confidence using CGM for diabetes. Nurse practitioners in hospital-owned settings were twice as likely to have prescribed CGM (odds ratio [OR] = 2.320, 95% CI [1.097, 4.903]; p = .002) than private practice; those in academic medical centers were less likely (OR = 0.098, 95% CI [0.012, 0.799]; p = .002). Past prescribing was associated with favorability toward future prescribing (coef. = 0.7284, SE = 0.1255, p < .001) and confidence using CGM to manage diabetes (type 1: coef. = 3.57, SE = 0.51, p < .001; type 2: coef. = 3.49, SE = 0.51, p < .001). Resources to prescribe CGM included consultation with an endocrinologist (62%), educational website (61%), and endocrinological e-consultations (59%). Nurse practitioners are open to prescribing CGM and can improve diabetes management and health outcomes for PC patients. Research should explore mechanisms behind associations with CGM experience and attitudes. Efforts to advance CGM should include educational websites and endocrinology consultations for NPs in PC.

Examining the interplay of self-efficacy and emotional autonomy on diabetic control index

AimThis study aims to explore the relationship between self-efficacy, emotional autonomy, and diabetic control index in adolescents with type 1 diabetes. MethodsFrom January to September 2023, this cross-sectional study involved 112 adolescents aged 14–18 with type 1 diabetes who were seeking medical check-ups at the Medical Faculty Hospital in Konya, Turkey. Data collection included demographic information, the Diabetes Management Self-Efficacy Scale for Adolescents with Type 1 Diabetes, the Emotional Autonomy Scale, and diabetic control index values based on their most recent Hemoglobin A1c levels measurement. Data analysis was performed using SPSS statistical software, with statistical significance set at a p-value <0.05. ResultsThe results showed a good overall self-efficacy score, indicating a balanced level of confidence in diabetes management. The emotional autonomy findings revealed a moderate level of independence, emphasizing a healthy process of differentiation from parents while acknowledging them as individuals. The ability of adolescents to openly discuss their diabetes may be associated with more favorable glycemic outcomes. However, high emotional autonomy was significantly associated with high Hemoglobin A1c levels. ConclusionThis study highlights positive outcomes in adolescent diabetes management, with strong self-efficacy and balanced emotional autonomy. Open communication about diabetes correlates with favorable glycemic outcomes, but caution is necessary as high emotional autonomy is linked to elevated Hemoglobin A1c levels. The importance of receiving parental support in managing diabetes, especially in adolescents with high emotional autonomy should be highlighted.

Implementing HbA1c monitoring in sub-Saharan Africa: Lessons learnt from a pilot project in Mozambique

AimsWe aimed to evaluate lesson learnt from a pilot project in Mozambique focused on point-of-care (POC) HbA1c testing for diabetes management in primary health care facilities. MethodsOver a three-year period, several health centers were equipped with POC HbA1c testing machines. The evaluation involved 12 months of data collection, interviews with patients and staff, and regular supervision visits. ResultsThe project screened over 22,000 individuals and provided HbA1c testing to 2362 diabetes patients. Among the analyzed results, 65.7 % had HbA1c levels below 7 %, 10.9 % between 7 % and 8.5 %, and 23.4 % above 8.5 %. POC testing showed advantages such as reduced costs and improved workload management. ConclusionsLimited access to HbA1c testing remains a challenge in African countries. The pilot project demonstrated the feasibility of POC HbA1c testing and highlighted the need for increased efforts to make it more widely available, leading to improved diabetes management and patient outcomes.

Proposed prototype for a quantitative biosensor for insulin detection using a liquid crystal

In this paper, we present a novel approach for label-free detection of human insulin using 5CB liquid crystal (LC). Monitoring insulin levels in the body is essential for treating diseases like diabetes, which are caused by insufficient production or inefficient use of insulin. Contemporary techniques of detection rely on glucose detection, and limited information is available on the direct detection of insulin. A simple method using a common room temperature LC 4-cyano-4-pentyl biphenyl (5CB) interacting with different concentrations of insulin, observed through a polarising optical microscope (POM), is reported in this paper. An analysis of observed textures revealed the detection of insulin as low as 25 μM concentration. The distinguishable features of the observed textures provide a scale for estimating insulin concentration. The effect of different pH and salt concentrations was studied in order to understand the optimal working condition of the proposed biosensor. Image J analysis demonstrated a positive correlation (R2 equal to 0.9553) between grey index (GI) values and human insulin concentrations. RGB analysis and interference studies confirmed the biosensor’s specificity in insulin detection. To correlate the textures to the happenings at the molecular level, docking studies were carried out. In addition to this, properties such as aromaticity, H-bond, hydrophobicity, interpolated charges, ionizability and solvent accessible area (SAS) in the minimum energy binding sites of 5CB-LC and human insulin chains were measured to know more about the interaction properties. This novel strategy shows promise for improving insulin detection sensitivity and accuracy, ultimately leading to better diabetes management and healthcare outcomes.

Proactive Identification of Patients with Diabetes at Risk of Uncontrolled Outcomes during a Diabetes Management Program: Conceptualization and Development Study Using Machine Learning.

The growth in the capabilities of telehealth have made it possible to identify individuals with a higher risk of uncontrolled diabetes and provide them with targeted support and resources to help them manage their condition. Thus, predictive modeling has emerged as a valuable tool for the advancement of diabetes management. This study aimed to conceptualize and develop a novel machine learning (ML) approach to proactively identify participants enrolled in a remote diabetes monitoring program (RDMP) who were at risk of uncontrolled diabetes at 12 months in the program. Registry data from the Livongo for Diabetes RDMP were used to design separate dynamic predictive ML models to predict participant outcomes at each monthly checkpoint of the participants' program journey (month-n models) from the first day of onboarding (month-0 model) up to the 11th month (month-11 model). A participant's program journey began upon onboarding into the RDMP and monitoring their own blood glucose (BG) levels through the RDMP-provided BG meter. Each participant passed through 12 predicative models through their first year enrolled in the RDMP. Four categories of participant attributes (ie, survey data, BG data, medication fills, and health signals) were used for feature construction. The models were trained using the light gradient boosting machine and underwent hyperparameter tuning. The performance of the models was evaluated using standard metrics, including precision, recall, specificity, the area under the curve, the F1-score, and accuracy. The ML models exhibited strong performance, accurately identifying observable at-risk participants, with recall ranging from 70% to 94% and precision from 40% to 88% across the 12-month program journey. Unobservable at-risk participants also showed promising performance, with recall ranging from 61% to 82% and precision from 42% to 61%. Overall, model performance improved as participants progressed through their program journey, demonstrating the importance of engagement data in predicting long-term clinical outcomes. This study explored the Livongo for Diabetes RDMP participants' temporal and static attributes, identification of diabetes management patterns and characteristics, and their relationship to predict diabetes management outcomes. Proactive targeting ML models accurately identified participants at risk of uncontrolled diabetes with a high level of precision that was generalizable through future years within the RDMP. The ability to identify participants who are at risk at various time points throughout the program journey allows for personalized interventions to improve outcomes. This approach offers significant advancements in the feasibility of large-scale implementation in remote monitoring programs and can help prevent uncontrolled glycemic levels and diabetes-related complications. Future research should include the impact of significant changes that can affect a participant's diabetes management.

Personalized Diabetes Management with Digital Twins: A Patient-Centric Knowledge Graph Approach.

Diabetes management requires constant monitoring and individualized adjustments. This study proposes a novel approach that leverages digital twins and personal health knowledge graphs (PHKGs) to revolutionize diabetes care. Our key contribution lies in developing a real-time, patient-centric digital twin framework built on PHKGs. This framework integrates data from diverse sources, adhering to HL7 standards and enabling seamless information access and exchange while ensuring high levels of accuracy in data representation and health insights. PHKGs offer a flexible and efficient format that supports various applications. As new knowledge about the patient becomes available, the PHKG can be easily extended to incorporate it, enhancing the precision and accuracy of the care provided. This dynamic approach fosters continuous improvement and facilitates the development of new applications. As a proof of concept, we have demonstrated the versatility of our digital twins by applying it to different use cases in diabetes management. These include predicting glucose levels, optimizing insulin dosage, providing personalized lifestyle recommendations, and visualizing health data. By enabling real-time, patient-specific care, this research paves the way for more precise and personalized healthcare interventions, potentially improving long-term diabetes management outcomes.