Frontiers in clinical diabetes and healthcare最新文献

Metabolic dysfunction-associated fatty liver disease (MAFLD) is rapidly emerging as a major public health challenge in Vietnam, driven by rising rates of obesity, type 2 diabetes, and lifestyle changes. Although it contributes significantly to morbidity, mortality, and economic burden, MAFLD remains under-recognized within national health strategies and is largely neglected in non-communicable disease (NCD) frameworks. This narrative review synthesizes epidemiological, clinical, and health systems data on MAFLD in Vietnam, drawing on studies published between 2015 and 2024 alongside international guidelines to evaluate their relevance in the local healthcare setting. The evidence indicates that MAFLD affects more than one-quarter of urban adults and is steadily increasing in rural populations. Barriers to effective management include limited diagnostic capacity, lack of standardized guidelines, insufficient awareness among clinicians, and underdeveloped multidisciplinary care models. Complications are exacerbated by the coexistence of hepatitis B virus infection and the growing burden of cardiovascular comorbidities. Together, these factors heighten disease severity and accelerate progression to cirrhosis and hepatocellular carcinoma. Given these challenges, urgent multisectoral action is needed. We propose a strategic national roadmap that incorporates MAFLD into NCD policy, expands primary care screening using simple non-invasive tools, and invests in health workforce training to improve early detection and risk stratification. Strengthening multidisciplinary collaboration and leveraging digital health technologies can enhance patient engagement and access to care. Finally, regional cooperation and Vietnam's participation in international clinical trials are essential to accelerate innovation and policy response. In conclusion, MAFLD represents a neglected but pressing public health issue in Vietnam. Proactive and coordinated strategies are required to reduce its long-term health and socioeconomic impact and to position Vietnam as a regional leader in addressing metabolic liver disease.

Introduction: Young adults living with type 1 diabetes face unique challenges as they transition to greater independence, balancing diabetes management in all other dimensions of life. In Sweden, the transfer from pediatric to adult diabetes care at the age of 18 adds to these challenges. This study aimed to translate and evaluate the Swedish version of the T1DAL (Type 1 Diabetes and Life) self-report questionnaire for young adults living with type 1 diabetes.

Method and materials: The T1DAL questionnaire was translated into Swedish and was completed by 191 young adults aged 18-25 who were registered at a diabetes clinic in three Swedish hospitals. An expert group tested content validity. To determine the number of underlying factors, a parallel analysis (PA) was conducted. The questionnaire's latent structure was further examined through exploratory factor analysis, in which the items were constrained to a four-factor solution as recommended and found in the original version.

Results: The content validity index of the total score was 0.94. The response distribution analysis revealed the presence of floor or ceiling effects. An EFA with a four-factor solution was conducted, yielding a Model Fit Measure with a χ² of 326.68 and df = 249, resulting in a cmin/df of 1.31, an RMSEA of 0.04, and a TLI of 0.92. Internal consistency was assessed for the subscales suggested by the factor structure, based on the items that loaded onto each factor. Cronbach's alpha values ranged from 0.75 to 0.89, indicating acceptable to high internal consistency. The four-factor solution explained 45.04% of the total variance.

Conclusion: The Swedish T1DAL questionnaire showed good factorial validity and reliability. The Swedish version requires further testing with potential item reduction; however, it is still expected to be valuable in assessing health-related quality of life among young adults living with type 1 diabetes.

Clinical implications: The Swedish T1DAL questionnaire, particularly its domains related to emotional experiences, managing diabetes effectively, and peer relationships outlines the characteristics of young adulthood and can be used to empower the target group, and is expected to be feasible to implement in clinical practice.

Background: With the prevalence of Type 2 Diabetes (T2D) projected to increase, understanding its potential consequences on healthcare systems is crucial for adequately preparing society to address this growing challenge. In 2019, Malmö, Sweden's third-largest city, joined the Cities for Better Health global initiative to tackle the multifaceted challenges associated with T2D, including its significant economic burden on the healthcare system and the broader community. Understanding the economic burden of T2D on the healthcare system will facilitate optimisation of the initiatives undertaken by the programme. Therefore, this study aimed to estimate the costs associated with primary care, hospital care, and work absenteeism due to diabetes-related complications among people with T2D residing in Malmö.

Methods: In order to estimate the cost for the City of Malmö, we expanded a model (Andersson et al., 2020) developed to estimate the cost of T2D on a national level, using retrospective data from 1997-2016. The costs were estimated by using NordDRG weights and national reference prices. Primary care costs for Region Skåne were added to the model. Data on healthcare utilisation, work absence, and socioeconomic factors were collected from Swedish national and regional registers. The method was expanded to include Malmö-specific adjustments for demographics, employment, and education, as well as regional primary care costs.

Results: The prevalence of T2D in Malmö was 5.4%, and diabetes complications were: diabetic retinopathy (49.9%), diabetic kidney disease (19.1%), angina pectoris (13.7%), ischaemic heart disease (10.9%), and myocardial infarction (10.5%). Total excess costs for T2D in primary care were €12.7 million. The lowest primary care excess costs were in the age group 16-34 and the highest in the age group 65-74. Estimated overall hospital-based costs for T2D were €38.8 million, and costs related to macrovascular and microvascular complications were €18.1 million and €16.4 million, respectively. Estimated total cost due to absence from work related to T2D complications was €15.4 million. The complication costs were higher for men, except for neuropathy.

Conclusions: These findings may support city-level healthcare planning and preventive interventions, as Malmö is facing substantial costs both in monetary terms and in reduced quality of life. Younger persons increasingly develop diabetes complications, which needs to be considered when allocating resources for primary prevention, treatment of complications, and municipality costs within a near future.

Aim and objectives: To investigate the Cutaneous Signs of Insulin Resistance, namely acanthosis nigricans (AN) and acrochordon (AC), in individuals with Central Obesity (CO-CSIR) as physical predictors of metabolic syndrome (MetS), underlying adipose tissue pathology, and consequent pathophysiological traits in South Asians.

Methods: In this study, 371 participants were recruited in a tertiary care facility and grouped based on the presence of cutaneous signs (AC and/or AN) and central obesity. Each participant was assessed for MetS, T2D, as well as other demographic, biochemical, and radiological parameters. Additionally, we conducted transcriptome profiling in adipose depots for selected individuals to investigate whether there are modules of co-expressed genes that show a correlation with cutaneous sign(s) and MetS/T2D, in order to decipher the link between these signs and metabolic derangement.

Results: ANOVA analyses revealed significant differences among groups with varying cutaneous signs and W:H ratios, particularly highlighting the combined predictive capability of these markers. Post hoc tests further confirmed these findings, showing substantial differences in MetS, T2D, and HOMA-IR between these groups. Sensitivity-specificity analyses demonstrated that CO-CSIR provides a more balanced and accurate prediction of MetS status compared to either CO or CSIR alone. Furthermore, in predicting MetS status based on the number of MetS components (from 5 to ≥1), it also performed well. WGCNA analysis in visceral fat revealed modules of co-expressed genes significantly correlated with AC and MetS, indicating a link between the adipose tissue molecular pathology and the cutaneous signs.

Conclusion: CO-CSIR is a promising physical sign for predicting MetS and the underlying adipose tissue-driven dysmetabolism in South Asians.

The rapid advancement of digital health technologies-such as continuous glucose monitors, automated insulin delivery systems, and telehealth platforms-has transformed diabetes management. However, a persistent digital divide continues to amplify health disparities based on socioeconomic status, geography, and age. Generational gaps are especially notable. "Digital immigrants" (older adults) often face significant barriers to technology adoption, such as reduced digital literacy, lower smartphone ownership, and difficulties using remote care platforms. Individuals aged 80 and above have much lower uptake of continuous glucose monitors and telemedicine, even when cost is not a barrier. Health literacy and language challenges further increase the impact and reduce the use of digital health tools among older adults. In contrast, "digital natives"-younger generations with strong technological skills-are well positioned to help bridge this divide. This mini-review summarizes current evidence on disparities in digital diabetes care and proposes an innovative, student-led solution. We advocate for empowering digital native health sciences students to lead digital health literacy initiatives and serve as technology mentors for both patients and clinicians. By establishing student-led digital literacy centers, academic institutions can promote intergenerational collaboration. This approach can turn the generational divide from a barrier into an opportunity. Ultimately, student-led initiatives offer a sustainable, community-based pathway to equitable adoption of digital diabetes technologies and improved health outcomes.

Background and purpose: Diabetes mellitus and dyslipidemia are major risk factors for atherosclerosis. Hypoechoic plaques, which indicate vulnerable or unstable plaques, may rupture and lead to ischemic stroke, cognitive impairment, increased adverse cardiac events, and even death. This study aimed to investigate the correlation between plasma lipid levels and the characteristics of atherosclerotic plaques in adult patients with type 2 diabetes mellitus.

Methods: A retrospective analysis was conducted on adult patients with type 2 mellitus who were hospitalized in the Department of Endocrinology at Affiliated Hospital of Hebei University between January 2017 and December 2021.Patients were categorized into two groups based on arterial ultrasound results. Statistical analyses were performed to compare plasma lipid levels and plaque characteristics across the groups.

Results: 1) Statistically significant differences were observed among the two groups in terms of gender, hypertension, age, duration of diabetes mellitus, plaque location, triglycerides (TG),total cholesterol (TC), Apolipoprotein A1 (Apo A1),very-low-density lipoprotein (VLDL), VLDL/apolipoprotein B(ApoB), high-density lipoprotein cholesterol (HDL)/ApoA1 (P<0.05). 2) Univariate and multivariate regression analyses revealed that VLDL,VLDL/ApoB and HDL/ApoA1 were correlated with plaque stability, the higher the levels of VLDL,VLDL/ApoB and HDL/ApoA1,the more likely they were to be hypoechoic plaque group (OR>1, P<0.05). 3) VLDL, VLDL/ApoB and HDL/ApoA1 showed predictive value for determining whether patients with type 2 diabetes had stable plaques. The area under the receiver operating characteristic (ROC) curve for VLDL, VLDL/ApoB and HDL/ApoA1 respectively were 0.789,0.779 and 0.728.

Conclusion: In clinical practice, the characteristics of atherosclerotic plaques and lipid profiles should be jointly evaluated to guide targeted treatment and effectively reduce the risk of atherosclerotic cardiovascular disease.

Objective: Diabetes mellitus (DM) is a prevalent chronic disease that can lead to severe microvascular complications. Among these, diabetic nephropathy (DN) remains a leading cause of end-stage renal disease worldwide. Glycemic variability, reflecting fluctuations in blood glucose, has been suggested as a potential predictor of DM complications. This study aimed to investigate whether visit-to-visit HbA1c variability contributes to the development and progression of DN in patients with DM.

Methods: In this retrospective cohort study, 228 patients were selected from 2,000 individuals diagnosed with DM between January 2007 and December 2017. A total of 80 patients without DN at baseline (ODN) and 148 patients with DN at baseline (WDN) were included in the study. HbA1c was measured 2-4 times per year over 3-5 years. Mean, standard deviation (SD), and coefficient of variation (CV) of HbA1c were calculated. Annual urea, creatinine, and albumin/protein levels were recorded. Logistic regression identified independent risk factors.

Results: DN developed in 47 (58.8%) patients in the ODN group, whereas progression occurred in 44 (29.7%) patients in the WDN group. In the ODN group, higher HbA1c mean, SD, CV, hypertension, and albuminuria were significantly associated with DN onset (p<0.05). Logistic regression analysis confirmed HbA1c variability and hypertension as independent predictors. No significant association was found between HbA1c variability and DN progression.

Conclusions: Variability in HbA1c is linked to the onset of DN but not its progression. These findings highlight the need for strategies targeting glycemic stability in DM management. Larger, multicenter prospective studies are warranted to confirm these results.

Introduction: Prediabetes is a highly prevalent metabolic condition that significantly increases the risk of developing type 2 diabetes and cardiovascular disease. Despite its clinical importance, over 80% of individuals with prediabetes remain undiagnosed. Voice analysis has emerged as a non-invasive, accessible method for disease screening, with prior work showing promising results in detecting hypertension and type 2 diabetes from acoustic features. This study investigates whether voice-based machine learning models can identify individuals with prediabetes and evaluates the generalizability of these models across populations.

Methods: Participants were recruited from clinical sites in India and a community college in Canada. All participants recorded the same spoken phrase multiple times daily via a mobile app, and glycemic status was assessed using HbA1c levels. Voice recordings were preprocessed to remove silence and trimmed to exclude potentially uninformative sections. A total of 167 acoustic features were extracted from each sample using Librosa, scipy, and parselmouth. Features were averaged per participant. Sex-specific models were developed under six experimental configurations varying by dataset balance (age/BMI-matched vs. unbalanced) and BMI inclusion. Feature selection was conducted using L1-regularized logistic regression (LASSO), and SMOTE was applied during training to address class imbalance. Twelve machine learning classifiers were evaluated using leave-one-subject-out cross-validation (LOSO-CV) on the India dataset. Final models were tested on a holdout India subset and the independent Canada dataset.

Results: In cross-validation, the best female model (XGBoost, balanced, no BMI) achieved a balanced accuracy of 0.78, and the best male model (Random Forest, balanced, no BMI) achieved 0.68. However, holdout set testing identified different optimal configurations for generalization: the male XGBoost model trained on an unbalanced dataset outperformed the cross-validated model. In the Canada dataset, models failed to generalize effectively, with several configurations unable to correctly identify prediabetic participants.

Discussion: Voice-based prediction models show potential for prediabetes screening in controlled populations, but their performance declines when applied across geographic or demographic boundaries. These findings highlight the need for more diverse training data and population-specific model tuning to support real-world applicability.