ObjectiveIn older adults with Type 2 diabetes mellitus (T2DM), the risk of delirium is significantly increased, driven by neuropathological alterations stemming from chronic insulin resistance. We utilized artificial intelligence and geriatric electronic health records to create an interpretable online machine-learning algorithm for predicting delirium risk. This tool facilitates prompt identification of high-risk elderly T2DM patients, enabling optimized interventions and improved clinical outcomes.MethodsThis retrospective cohort study identified older adults with T2DM using International Classification of Diseases (ICD) codes, with delirium defined by the Confusion Assessment Method for the intensive care unit (CAM-ICU). We extracted baseline demographics, vital signs, laboratory measurements, comorbidities and clinical severity scores. Candidate predictors for eight machine-learning algorithms were selected using least absolute shrinkage and selection operator regression and the Boruta method. Discrimination was assessed using accuracy, sensitivity, specificity and the F1 score. The final model was interpreted using SHapley Additive exPlanations (SHAP) and deployed as an online risk calculator.ResultsIntegrating dual feature selection methods identified 14 key predictors and the gradient boosting machine (GBM) model accurately predicted delirium risk in elderly patients with T2DM, demonstrating strong discriminatory performance with robust calibration in both internal and external validation. SHAP analysis highlighted the Glasgow Coma Scale, ICU length of stay and Sequential Organ Failure Assessment score as the predominant contributors to model predictions. The model was successfully deployed as an accessible online tool and the accompanying web-based calculator enables rapid, personalized risk assessment to support early intervention in ICU settings.ConclusionsThe GBM model showed strong performance in predicting delirium risk among elderly patients with T2DM, supporting clinically meaningful risk stratification. The accompanying web-based calculator enables rapid, individualized bedside assessment and may facilitate early identification of high-risk patients and timely intervention in ICU settings.

To investigate the synergistic interaction between fasting plasma glucose (FPG) and serum uric acid (SUA) on the severity of peripheral diabetic retinopathy (DR) and to develop a predictive nomogram for severe non-proliferative DR (S-NPDR) in type 2 diabetes mellitus (T2DM) patients.

This retrospective, single-center study included 900 T2DM patients admitted between January 2018 and December 2023. Patients were categorized into groups of No-DR (n = 300), mild to moderate NPDR (n = 350), and S-NPDR (n = 250). S-NPDR was compared against the combined No-DR and mild to moderate NPDR groups (n = 650). Independent predictors were identified via multivariable logistic regression. Additive interaction analysis assessed the synergistic effect of FPG and SUA. A predictive nomogram was developed and internally validated using 1000 bootstrap resamples.

Seven independent predictors of S-NPDR were identified in the final model: higher FPG (odds ratio [OR] = 1.701), SUA (OR = 1.062 per 10 µmol/L), urinary albumin-to-creatinine ratio (UACR; OR = 1.046 per 10 mg/g), high-sensitivity C-reactive protein (hs-CRP; OR = 1.191), lower subfoveal choroidal thickness (SFCT; OR = 1.330 per 10-µm decrease), lower serum albumin (ALB; OR = 1.153 per 1-g/L decrease), and the interaction between FPG and SUA (OR = 2.710, P < 0.001). A strong synergistic interaction was confirmed (adjusted OR for high FPG/high SUA = 17.56; relative excess risk due to interaction [RERI] = 12.58). The nomogram demonstrated excellent discrimination (area under the receiver operating characteristic curve [AUC] = 0.925) and good calibration (Hosmer-Lemeshow P = 0.418).

Concurrent elevations of FPG and SUA are strongly associated with an increased likelihood of S-NPDR. Patients with both markers elevated represent a high-risk group requiring intensive monitoring.

The validated nomogram enables personalized risk stratification, guiding targeted clinical management to prevent vision-threatening DR complications.

Diabetes negatively impacts productivity, but the extent to which socio-economic factors influence this effect is unknown. This study examines how diabetes duration affects labour force participation and sick leave in Germany, focusing on socio-economic differences. We used self-reported data collected between 2009 and 2021 from the German Socio-Economic Panel Study, a longitudinal household survey. People with prevalent diabetes at baseline were excluded. To estimate the causal effect of diabetes duration on the outcomes, we employed marginal structural regression models for repeated measures, using stabilized inverse-probability-of-treatment-and-censoring weights to adjust for informative censoring, time-fixed (sex, age, socio-economic position, migration background) and time-varying confounding (body mass index, physical activity frequency, smoking status, previous outcome). We included interaction terms to assess diabetes-related productivity losses by subgroups of socio-economic position, sex, age and migration background. The analysis consisted of 35 906 observations from 18 456 individuals for the outcome labour force participation and 12 469 observations from 7244 individuals for the outcome sick leave days. A five-year increase in diabetes duration was associated with a labour force participation shortfall of 13.8% (95% confidence interval: 5.8; 21.1) and an increase of 6.8 sick leave days (-5.4; 19.0). Effects were more pronounced among individuals in lower socio-economic position and diminished with increasing socio-economic position. Diabetes-associated productivity losses predominantly affect people in low socio-economic position, reflecting a dual burden of higher diabetes prevalence and larger productivity losses.

Background Type 2 diabetes mellitus (T2DM) is a major global health concern, with microvascular complications, neuropathy, retinopathy, and nephropathy contributing substantially to morbidity. Selenoprotein P (SeP), a selenium-transporting antioxidant protein, has been implicated in metabolic dysregulation. This study examined the prevalence of microvascular complications in T2DM and explored their association with circulating SeP levels in a tertiary care center in South India. Materials and methods A cross-sectional study was conducted at Vinayaka Mission's Medical College, Karaikal (2022-2025), enrolling 80 adults with T2DM aged ≥40 years. Neuropathy was assessed using the 10 g monofilament test, retinopathy by fundus examination, and nephropathy by urine albumin-to-creatinine ratio (ACR). Serum SeP was quantified by enzyme-linked immunosorbent assay (ELISA). Analyses used SPSS version 26.0 (IBM Corp., Armonk, NY) with Pearson or Spearman correlation and independent t-test or Mann-Whitney U test, as appropriate. Results Among 80 participants (53.8% women; mean age: 58.84 ± 9.79 years), neuropathy was present in 45% (n = 36), retinopathy in 31.3% (n = 25), and nephropathy in 31.3% (n = 25). The mean SeP was 4.11 ± 1.47 µg/mL and tended to be higher with retinopathy (non-proliferative diabetic retinopathy {NPDR}: 4.52 ± 1.74; proliferative diabetic retinopathy {PDR}: 4.89 ± 1.26) than without (3.98 ± 1.34 µg/mL). Group differences were not statistically significant (retinopathy, p = 0.180; nephropathy, p = 0.187; and neuropathy, p = 0.120). Conclusion Microvascular complications are common in this T2DM cohort. Although not significant, higher SeP levels appeared to parallel retinopathy severity, supporting the further evaluation of SeP as a potential biomarker in larger or longitudinal studies.

Diabetic retinopathy (DR) is a leading cause of vision loss in individuals with diabetes, highlighting the need for timely screening. In Taiwan, limited ophthalmologic resources, especially in underserved areas, constrain screening coverage. This study evaluated the cost-effectiveness of artificial intelligence (AI)-assisted DR screening as an alternative strategy for early detection and improved resource allocation.

A Markov decision-tree model was constructed from the healthcare payer's perspective, using transition probabilities, costs, and quality-adjusted life years (QALYs) derived from domestic and international data. The model applied a 1-year cycle length, a 10-year time horizon, a 3% annual discount rate, and 10,000 Monte Carlo simulations. Incremental cost-effectiveness ratios (ICERs) were calculated for AI-assisted versus ophthalmologist-based screening, with probabilistic and one-way sensitivity analyses conducted to evaluate robustness. Statistical analyses were conducted using SPSS version 23.0, while cost-effectiveness analyses were performed using TreeAge Pro Healthcare 2021.

AI-assisted screening incurred higher costs ($10,077.34) than ophthalmologist-based screening ($8282.06) but provided greater health benefits (7.60 vs. 6.34 QALYs). The ICER was $1429.19/QALYs, well below willingness-to-pay threshold ($33,983, 2024 Taiwan per capita gross domestic product), demonstrating high cost-effectiveness.

AI-assisted DR screening is a cost-effective approach that may enhance access, especially in regions with limited specialist availability. By enabling earlier detection and reducing reliance on ophthalmologists, AI-based screening has the potential to improve both efficiency and equity in healthcare delivery. These findings support its integration into national screening programs and emphasise the importance of local data in informing policy decisions.

Diabetes-related distress refers to the emotional strain experienced by individuals who live with diabetes and face continuous demands of treatment and self-management. Elevated distress is associated with poor adherence, impaired metabolic control, and diminished quality of life. This study aimed to describe diabetes-related distress among Spanish adults, analyse its associations with sociodemographic and clinical variables, identify predictors of higher distress, and define patient profiles with distinct psychosocial patterns.

A cross-sectional analytical study was conducted with 201 adults (n = 135 type 1, n = 66 type 2 diabetes) from two public healthcare centers in Seville, Spain. Participants completed the Spanish version of the 17-item Diabetes Distress Scale and a questionnaire covering demographic and clinical variables. Statistical analyses included non-parametric tests, logistic regression, cluster analysis, and discriminant analysis to identify predictive factors and patient groupings.

Among participants, 39.3% presented low or no distress, 22.4% moderate distress, and 38.3% high distress, with an average score of 2.72. Higher distress was significantly associated with younger age, type 2 diabetes, and combined pharmacological treatment. Diabetes type was the strongest predictor of distress. Cluster analysis revealed three profiles: older adults with type 2 diabetes and low distress, middle-aged adults with type 1 diabetes and moderate distress, and younger adults with type 1 diabetes and high distress.

Diabetes-related distress is a heterogeneous phenomenon influenced by age, diabetes type, and treatment regimen. Regular assessment and individualised psychoeducational and emotional interventions may improve psychological well-being and adherence, particularly among younger adults and those under complex therapies.

Physical activity is an integral component in the treatment and management of Type 2 diabetes. It is recommended that healthcare professionals (HCPs) promote physical activity to patients with Type 2 diabetes; however, they report many challenges to doing this in practice.

This systematic review is aimed at identifying HCPs' barriers and facilitators to promoting physical activity among adults with Type 2 diabetes.

A mixed-methods systematic review (MMSR) was conducted using a convergent integrated approach, following the Joanna Briggs Institute methodological procedures. MEDLINE, PubMed, PsycINFO, CINAHL, Web of Science and grey literature were searched from 1951 to 2020, with updates in 2024. Studies were excluded if they lacked HCPs' views on Type 2 diabetes care or were not in English, but no study design restrictions were applied. Barriers and facilitators were coded to the Theoretical Domains Framework (TDF); each domain was ranked for importance, and an inductive thematic synthesis was conducted.

A total of 29 primary studies were included in the review. Barriers were reported across 11 TDF domains and facilitators across eight. Key findings within these TDF domains include insufficient time and resources to promote physical activity, a lack of organisational support, lack of knowledge and skills, challenges related to patient comorbidities/complications and HCPs' perceptions of their roles and responsibilities in promoting physical activity.

The findings highlight the need for change at HCP, organisational and environmental levels. Key improvements include clearer roles for promoting physical activity, system-embedded prompts and the integration of disease-specific modules into medical training.

To develop a chain mediation model to elucidate the relationship among physical performance, instrumental activities of daily living (IADL), regular exercise, and cognitive function among older adults who are comorbid with diabetes mellitus and hypertension (OA-DM&HTN).

A total of 656 participants were investigated with the Mini-Mental State Examination, the Short Physical Performance Battery, the Instrumental Activities of Daily Living, and a questionnaire on regular exercise frequency between January and September 2022. Sequential multiple mediation models were conducted to analyze the data.

The average age of the participants was 73.47 ± 7.40 years, and 49.24% (n = 323) of participants were female. The average cognitive function score was 22.36 ± 6.14, and 32.62% (n = 214) of participants exhibited cognitive impairment. Cognitive performance exhibited significant associations with demographic factors, including gender, age, marriage status, educational background, and income level (p < 0.05). Chain mediation analysis indicated that physical performance directly predicted cognitive function (β = 0.525, 95% CI: 0.000-1.050); physical performance had indirect effects mediated by IADL (β = 0.917, 95% CI: 0.635-1.230) and regular exercise (β = 0.076, 95% CI: 0.003-0.180). A significant chain-mediating effect involving both IADL and regular exercise was also observed on the relationship between physical performance and cognitive function (β = 0.034, 95% CI: 0.002-0.071).

Physical performance is a significant predictor of cognitive function, and it can also affect cognitive function through the independent or chain-mediating effects of IADL and regular exercise among OA-DM&HTN. Therefore, to delay cognitive decline among OA-DM&HTN, it is essential to provide tailored functional training, encourage improvement in IADL, and promote regular exercise among OA-DM&HTN.

Vitamin D is a pleiotropic molecule involved in various physiological processes beyond skeletal health, including immune modulation and metabolic regulation. This prospective observational biomedical study aimed to assess the impact of short-term high-dose vitamin D supplementation on selected metabolic parameters in adult patients with type 2 diabetes mellitus (T2DM) and low serum 25-hydroxyvitamin D [25(OH)D] concentrations.

Thirty patients were enrolled and assigned to receive either 10,000 IU/day of cholecalciferol (Group A) or a significantly lower dose (960 IU/day, Group B) for 12 weeks based on recruitment order (odd/even identification numbers). The primary endpoints were changes in parathyroid hormone (PTH), fasting blood glucose (FBG), calcium, phosphorus, and glycated haemoglobin (HbA1c).

A strong, statistically significant negative correlation between changes in 25(OH)D and PTH (Spearman r = -0.69052, p = 0.0044) was also observed. In the high-dose group, 25(OH)D increased from 17.2 to 31.8 ng/mL (median change 13.3 ng/mL), while PTH decreased from 3.27 to 2.76 pmol/L (median change -0.27 pmol/L). In the lower-dose group, 25(OH)D increased from 18.5 to 28.2 ng/mL (median change +8.1 ng/mL). The increase in 25(OH)D was significantly greater in the high-dose group than in the lower-dose group (median change +13.3 vs +8.1 ng/mL, p = 0.015). Within the observed range, patients with larger increases in 25(OH)D tended to show greater reductions in PTH. Other metabolic markers (HbA1c, FBG, calcium, and phosphorus) remained stable over 12 weeks.

These findings support the effectiveness and safety of high-dose vitamin D supplementation in correcting vitamin D deficiency and reducing PTH levels in patients with T2DM while highlighting the need for longer-term studies to evaluate its broader metabolic effects.

Childhood obesity has emerged as a major global public health concern. Recent reports indicate that the worldwide obesity rate among children and adolescents has climbed to 8.5%. Obesity-related conditions present serious health risks. The causes of childhood obesity are multifactorial, with growing attention from researchers on the link between maternal gestational diabetes mellitus (GDM) and an elevated risk of obesity in offspring. This association can have lasting effects throughout childhood. Consequently, understanding the causes of childhood obesity and implementing effective preventive and therapeutic measures are crucial for managing and mitigating the condition and its associated health complications. This study aims to thoroughly explore the relationship between maternal GDM and the heightened risk of childhood obesity, examining developmental stages, underlying pathophysiological mechanisms, and contributing factors. By developing and applying targeted management strategies during a child's growth and development, the objective is to effectively curb and improve obesity outcomes among children born to mothers with GDM.