BMJ Open Diabetes Research & Care最新文献

Introduction: Adults with diabetes require regular medical care which can be costly, but little is known about factors associated with delaying or forgoing medical care due to cost among US adults with diabetes.

Research design and methods: Data were from the 2009-2010, 2014-2015 and 2022-2023 cycles of the cross-sectional National Health Interview Survey and included participants age ≥18 years who self-reported a physician diagnosis of diabetes. Descriptive statistics were used to determine the prevalence and trends in delaying or forgoing medical care by sociodemographic and clinical characteristics and health insurance coverage. Logistic regression models were used to determine the OR for delaying or forgoing medical care associated with insurance status.

Results: Among US adults aged 18-64 years with diabetes, delaying or forgoing medical care due to cost decreased from 18.1% to 10.6% and from 14.6% to 10.2%, respectively, between 2009 and 2023. In 2022-2023, the prevalence of delaying medical care due to cost for adults aged 18-64 years was highest for non-Hispanic black adults (13.3%), those with a high school education or less and poverty income ratio <4.0 (12%-13%). In 2022-2023, uninsured adults ≥18 years were significantly more likely to delay medical care compared with those who were insured (adjusted OR (aOR) =7.5, 4.8-11.8, age 18-64 years (adjusted for sociodemographic and clinical characteristics)). Adults aged 18-64 years with Medicaid were significantly less likely to delay medical care compared with those who had private insurance (aOR=0.2, 0. 1-0.4).

Conclusions: There was a decreasing trend for delaying or forgoing medical care across all subpopulations, but adults with lower education and income and who were uninsured more often reported delays in medical care due to cost. The expansion of Medicaid may have reduced the likelihood of delaying or forgoing medical care due to cost among adults aged 18-64 years with Medicaid coverage.

Introduction: Environmental heavy-metal mixtures may contribute to diabetes risk, yet their combined effects remain understudied. We investigated the association between the Metal Mixture Inflammatory Index (MMII) and prevalent diabetes in US adults.

Research design and methods: We analyzed data from 23 288 participants in the 1999-2020 National Health and Nutrition Examination Survey. Survey-weighted logistic regression, restricted cubic splines (RCS), and stratified analyses assessed the relationship between MMII and diabetes. Least absolute shrinkage and selection operator (LASSO) regression identified key predictors, which were incorporated into a nomogram. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), calibration plots, and decision curve analysis.

Results: After multivariable adjustment, each 0.1-unit increase in MMII was associated with 2% higher odds of diabetes (OR=1.02; 95% CI 1.00 to 1.04; p=0.02). Participants in the highest quartile (Q4) had 26% greater odds than those in the lowest quartile (Q1) (OR=1.26; 95% CI 1.04 to 1.52; p=0.016). RCS analysis indicated a linear positive association between MMII and diabetes risk. Subgroup analyses revealed stronger associations among men, non-Hispanic white participants, former smokers, current alcohol users, and individuals without hypertension. The LASSO-based nomogram demonstrated excellent discrimination (AUC=0.869; 95% CI 0.863 to 0.875), good calibration, and net clinical benefit.

Conclusions: MMII is independently and linearly associated with diabetes risk. Metal exposures may enhance future risk stratification for diabetes. Prospective studies are warranted to confirm causal mechanisms.

Introduction: Hyperosmolar hyperglycemic state (HHS) is a life-threatening metabolic emergency with high mortality rate. Yet, there is no national system in the UK to monitor clinical practice or outcomes. To address this, we implemented and evaluated a multicenter surveillance system for HHS, assessing interhospital variations in management, outcomes, and barriers to guideline implementation.

Research design and methods: This mixed-methods observational study was conducted across 12 NHS hospitals between 2021 and 2024. A standardized data collection tool was developed, capturing demographics, biochemistry, treatment, and outcomes of HHS care. Adults meeting the Joint British Diabetes Societies criteria for HHS were included. Quantitative analyses were conducted to investigate care variations compared with guidelines among centers and identify predictors of HHS outcomes. In parallel, stakeholder interviews were analyzed thematically to explore implementation experiences. The Reach, Effectiveness, Adoption, Implementation, Maintenance framework guided evaluation.

Results: In our cohort, a total of 218 HHS episodes were included. Median patient age was 77 years; 84.4% had type 2 diabetes, with a high comorbidity burden. The median hospital stay was 10.3 days, and the mortality rate was 16.1%. Significant interhospital variation was observed in insulin dosing, glucose monitoring, and time to discharge. Multivariate analysis identified older age and elevated sodium as independent predictors of mortality. The Digital Evaluation of Ketosis and Other Diabetes Emergencies (DEKODE)-HHS model demonstrated feasibility, high user engagement, and potential for integration into routine quality improvement structures. Qualitative findings revealed barriers, including diagnostic misclassification and resource constraints, to the adoption of the DEKODE-HHS model. However, they also highlighted the educational impact and system usability once the model was adopted.

Conclusions: The DEKODE-HHS model represents the first UK multicenter surveillance initiative for HHS. It identifies variation in practice and outcome predictors while highlighting systemic barriers to guideline adherence. This model provides a scalable framework for continuous quality improvement in HHS management and may inform future updates to national guidance.

Introduction: Dyslipidemia is a major risk factor for cardiovascular disease in type 2 diabetes mellitus (T2DM), but its association with kidney disease progression remains incompletely defined. Although low high-density lipoprotein cholesterol (HDL-C) has been linked to diabetic nephropathy, evidence regarding hard kidney outcomes is limited. We examined the associations between HDL-C and kidney events in patients with T2DM, in comparison with other lipid parameters.

Research design and methods: A total of 1,033 patients with T2DM from the Fukushima Cohort Study were included. Participants were followed for kidney events, defined as a ≥50% decrease in estimated glomerular filtration rate (eGFR) or onset of kidney failure requiring kidney replacement therapy, and all-cause mortality over a median follow-up period of 5.3 years. Lipid parameters HDL-C, triglycerides (TG), low-density lipoprotein cholesterol, non-HDL-C, and TG/HDL-C ratio were categorized into quartiles and evaluated using Cox proportional hazards models, adjusted for age, sex, smoking history, history of cardiovascular disease, body mass index, systolic blood pressure, eGFR, hemoglobin A1c, and proteinuria.

Results: The median patient age was 66 years, 56% were men, and the median eGFR was 68.6 mL/min/1.73 m². After multivariable adjustment, patients in the lowest HDL-C quartile (<42 mg/dL) had significantly higher risks of kidney events (adjusted HR 2.61, 95% CI 1.32 to 5.14) and all-cause mortality (adjusted HR 2.27, 95% CI 1.16 to 4.42) than the reference group (HDL-C 49-58 mg/dL). A U-shaped association was observed between HDL-C and all-cause mortality. Subgroup and sensitivity analyses were consistent. No significant associations were observed for other lipid parameters with either kidney events or mortality.

Conclusions: Low HDL-C levels were independently associated with kidney events and all-cause mortality in patients with T2DM. Future studies are warranted to clarify whether interventions targeting HDL-C can improve kidney disease progression in this high-risk population.

Trial registration number: UMIN000040848.

Introduction: Cardiovascular disease (CVD) is a common complication and major cause of mortality in people with type 1 diabetes (T1D). This study quantifies the prevalence of CVD among commercially insured adults with T1D in the USA from 2017 to 2021, overall and among age-defined and sex-defined subgroups.

Research design and methods: We used Merative MarketScan nationwide commercial insurance claims database (2017-2021) to identify adults ≥20 years with T1D (International Classification of Diseases, 10th Revision (ICD-10) codes). CVD ascertainment was based on ICD-10 codes for myocardial infarction, atrial fibrillation, ischemic heart disease, heart failure, peripheral artery disease, and stroke. Comorbidities included hypertension, obesity, hyperlipidemia, retinopathy, neuropathy, nephropathy, severe hypoglycemia, and diabetic ketoacidosis. Annual prevalence and age-specific and sex-specific prevalence of CVD were calculated overall and by comorbidities. Logistic regression was used to examine associations between sex, prevalent comorbidities, and odds of CVD.

Results: The sample size ranged from n=21 748 in 2017 to n=13 294 in 2021. Among adults with T1D (mean (SD) age (48.51 (13.95) years in 2017 and 46.80 (13.04) years in 2021; 47% female), the prevalence of CVD ranged from 18.18% (95% CI 17.77 to 18.66%) in 2017 to 20.58% (95% CI 19.91 to 21.24%) in 2021. In 2021, among those aged 20-39 years, 40-64 years, and 65+years, the prevalence of CVD was 4.97%, 20.41%, and 52.94%, respectively. The age-adjusted prevalence of CVD was higher in males than females (21.93% vs 19.07%). Age, sex, and all comorbidities were independently associated with CVD. Odds of CVD were highest among those with hypertension (adjusted OR 3.15, 95% CI: 2.77 to 3.57).

Conclusion: In this sample of US commercially insured adults with T1D, CVD prevalence remained stable at ~20% from 2017 to 2021. Early detection via improved screening and targeted management of comorbidities are key preventive strategies.

Aim: Genetic variants associated with gestational diabetes mellitus (GDM, n=14 SNPs) were recently classified into two groups: type 2 diabetes predominant effects (Class-T, three SNPs) and GDM-predominant effects (Class-G, eight SNPs; three SNPs unclassified). We aimed to compare the effects of GDM-associated variants on glucose levels (fasting glucose and 2-hour post-OGTT) measured during versus post pregnancy.

Research design and methods: We calculated genetic scores (GS) by class (T_GS and G_GS) and overall (All_GS) in 10 225 pregnant women and 4763 women post pregnancy (mean 10.5 years post pregnancy) from eight datasets representing four ancestrally-diverse cohorts: Exeter Family Study of Childhood Health, Genetics of Glucose Regulation in Gestation and Growth, Hyperglycaemia and Adverse Pregnancy Outcome, and Finnish Gestational Diabetes. We used linear regression models adjusted for ancestry principal components to investigate associations between standardized GS and glucose levels during or post pregnancy. Analyses were performed separately in each dataset and then combined using inverse-variance weighted random-effects meta-analyses.

Results: All_GS was associated with fasting glucose both during and post pregnancy (β (95% CI), in mmol/L per 1 SD higher GS=0.06 (0.04 to 0.08) during vs 0.06 (0.04 to 0.07) post pregnancy). All_GS was also associated with 2-hour post-OGTT glucose levels during pregnancy but not after (0.10 (0.04 to 0.15) during vs 0.01 (-0.04 to 0.07) post pregnancy). Both G_GS and T_GS showed consistent associations with fasting glucose during and post pregnancy (0.06 (0.04 to 0.08) during and 0.05 (0.03 to 0.07) post pregnancy for G_GS; 0.02 (0.01 to 0.02) during and 0.02 (-0.001; 0.05) post pregnancy for T_GS). G_GS showed weak evidence of association with 2-hour glucose levels during pregnancy (0.06 (-0.002 to 0.11)) and no association with 2-hour glucose levels post pregnancy (-0.03 (-0.08 to 0.03)). However, T_GS was associated with 2-hour glucose during pregnancy and post pregnancy (0.10 (0.04 to 0.16) and 0.06 (0.01 to 0.12)).

Conclusion: Consistent associations with fasting glucose levels during and after pregnancy may suggest that biological pathways underlying GDM genetic susceptibility to fasting hyperglycemia are not pregnancy specific. However, the results for All_GS and 2-hour glucose provide evidence that some genetic associations with postprandial glucose may be stronger in pregnancy and should be followed up in larger samples.

Introduction: Achieving optimal glycemic control remains challenging for many patients with diabetes. Text message-based interventions offer a scalable approach to enhance management. This systematic review and meta-analysis evaluated the impact of texting interventions on glycemic control in adults with diabetes.

Research design and methods: We searched EMBASE, PubMed, and Cochrane CENTRAL for randomized controlled trials comparing texting interventions to standard care in high-income countries. The primary outcome was the between-group difference in hemoglobin A1c (HbA1c) change from baseline. Risk of bias and overall quality of evidence were assessed using the Cochrane and Grading of Recommendations Assessment, Development, and Evaluation tools respectively. Results were pooled using an inverse variance random-effects model. Heterogeneity was evaluated using the I² statistic.

Results: Over 3 months of follow-up (14 trials, n=1,460 intervention, n=1,487 control), texting interventions were associated with a 0.29-unit greater reduction in percent HbA1c over control (95% CI 0.14 to 0.45, p=0.0001, I²=57%). At 6 months (20 trials, n=2,332 intervention, n=2,371 control), texting was associated with 0.19-unit greater HbA1c reduction (95% CI 0.07 to 0.30, p=0.001 I²=45%). At 12 months (seven trials, n=2,038), there was a non-significant benefit associated with texting. Among studies with a mean baseline HbA1c ≥8.6%, texting was associated with 0.48- and 0.36-unit greater HbA1c reductions at 3 (p=0.004) and 6 (p=0.004) months, respectively. Subgroups were not significantly different.

Conclusion: Text messaging interventions are associated with modest improvements in glycemic control over 3-6 months, particularly in patients with poorer baseline HbA1c. These effects may be meaningful at scale and support texting as a potential adjunct to routine diabetes care. Benefits appear to diminish by 12 months, underscoring the need for high-quality trials focused on long-term impact and intervention optimization.

Prospero registration number: CRD42023416462.

Background: Diabetic foot ulcer (DFU) is a severe complication of diabetes mellitus, often characterized by a chronic disease course and a high recurrence rate, posing significant challenges to patient management. Accurately predicting DFU recurrence is essential for enhancing patient care and outcomes through timely treatment and intervention. This study aimed to develop a machine learning (ML) model to predict the 3-year recurrence risk in patients with DFU.

Methods: A total of 494 patients with DFU were included and assigned to a training set and a test set at a 4:1 ratio. Four feature selection methods-least absolute shrinkage and selection operator, minimum redundancy maximum relevance, Fisher score and recursive feature elimination-were applied to the training set, and intersecting features were selected to construct the final feature set. Seven ML algorithms, including logistic regression, support vector machine, random forest, gradient boosting decision tree, AdaBoost, extreme gradient boosting (XGBoost) and light gradient boosting machine, were employed to develop predictive models. The models' parameters were optimized using fivefold cross-validation. Model performance was evaluated using the area under the receiver operating characteristic curve (AUROC). The best-performing model was calibrated using Platt scaling, with calibration performance assessed by the Brier score. ML model interpretability was enhanced using SHapley Additive exPlanations (SHAP) analysis.

Results: The XGBoost model demonstrated superior predictive performance, achieving an AUROC of 0.924 (95% CI 0.867 to 0.967). Following calibration with Platt scaling, the model exhibited a Brier score of 0.096, indicating good calibration. SHAP analysis identified key risk factors that aligned with existing literature and clinical expertise, further validating the model's interpretability and clinical relevance.

Conclusion: The XGBoost model demonstrated strong predictive accuracy and clinical relevance in assessing DFU recurrence risk. However, further multicenter validation with a larger sample size is needed to improve its generalizability and clinical applicability.

Introduction: Prednisolone is widely prescribed to hospitalized patients for a range of conditions. Up to 40% of hospitalized patients treated with prednisolone will experience hyperglycemia. Current guidelines recommend management of acute hyperglycemia in hospitalized patients with subcutaneous basal-bolus insulin (BBI), but the optimum treatment strategy has not been defined. We aimed to assess the performance of an individualized subcutaneous BBI regimen for management of prednisolone-associated hyperglycemia in hospitalized patients.

Research design and methods: This cross-sectional study included 23 adult inpatients prescribed subcutaneous BBI based on total daily insulin requirements estimated from a 24-hour intravenous insulin infusion and 24 historical controls who were prescribed a standard, institutional weight-based subcutaneous BBI to treat prednisolone-associated hyperglycemia. The primary endpoint was the mean 24-hour point-of-care (POC) glucose concentration on day 1. Exploratory end points included proportion of glucose measurements within target glucose range, SD of glucose, and stress hyperglycemia ratio (SHR).

Results: There was no significant difference in mean POC glucose on day 1 between participants prescribed an individualized insulin regimen and patients receiving a standard body weight-based BBI regimen (10.7±3.4 vs 11.9±3.2 mmol/L, p=0.07). Proportion of glucose measurements within the target glucose range was higher (52.0±4.8 vs 37.0±4.5%, p=0.0007) and SD for glucose lower (3.1±1.5 vs 4.0±1.6, p=0.04) on day 1 of individualized BBI insulin. Over 2 days, there was an increase in glucose SD in both groups, but no significant difference in mean glucose and SHR between groups.

Conclusions: Individualizing a subcutaneous BBI regimen for management of prednisolone-associated hyperglycemia was associated with a modest reduction in mean POC glucose, an increased proportion of blood glucose measurements within the target range, and reduced short-term glycemic variability.

Trial registration number: ACTRN12618001211257.