To evaluate whether empirical calibration of P-values using negative controls can effectively control Type I and Type II errors under unmeasured confounding bias in both simulated and real-world observational settings.

A simulation study was conducted under five settings reflecting different degrees of adherence to the U-comparability assumption-that is, the extent to which negative controls share the same unmeasured confounding structure as the exposure of interest. These included three primary scenarios (ideal, realistic, and violation of U-comparability) and two mixed scenarios reflecting partial violations. We varied sample size, the direction and strength of unmeasured confounding bias, and the number of negative controls. Based on UK Biobank data, the method was also applied to evaluate the association between hypertension and peripheral artery disease (PAD) in individuals with type 2 diabetes mellitus.

Standard logistic regression showed inflated Type I error rates across almost all settings, peaking at 44.2% under realistic U-comparability with a sample size of 20,000. In contrast, empirical calibration generally controlled Type I error close to the nominal 5% level and reduced bias by 80-100% under both ideal and realistic U-comparability. Type I error control improved with more negative controls, while Type II error control was influenced by whether the unmeasured confounding bias acted in the same or opposite direction as the true exposure-outcome effect. In the UK Biobank case study, 4 of 15 negative controls showed P < 0.05 after adjustment for measured confounders, indicating residual unmeasured confounding. After empirical calibration with 5, 10 or 15 negative controls, the association between hypertension and PAD remained statistically significant (calibrated P ≈ 0.004-0.006).

Empirical calibration of P-values can mitigate residual unmeasured confounding and reduce Type I error inflation in observational studies. Its performance depends on the validity and number of negative controls.

Diabetes mellitus has become an epidemic worldwide. Although several oral antidiabetic drugs are currently available in clinical practice, most of these drugs are often inadequate in restoring β-cell function or effectively preventing the onset of diabetic complications. In our previous study, we found that the Tujia ethnic medicine, Calanthe davidii showed potential antidiabetic and hepatoprotective properties. However, its active constituents and the mechanisms are still unclear.

To assess the anti-diabetic efficacy of alkaloids of C. davidii (CdA) in vivo, and further identify the active constituents and uncover their underlying mechanisms.

DPPH^• and OH^• scavenging assays, ferric-reducing power test, anti-lipid peroxidation test, and LPS-stimulated macrophage model were employed to verify the antioxidant and anti-inflammatory properties of CdA in vitro. The antidiabetic activity and preliminary mechanisms of CdA were elucidated in mice with STZ-induced diabetes. Various chromatographic and spectroscopic techniques were used to purify and characterize chemical constituents. ELISA, RT-qPCR, RNA-seq, immunofluorescence, and western blotting were explored to elucidate the mechanism of the active ingredients.

CdA significantly inhibited hyperglycemia, ameliorated pancreatic β-cell damage, mitigated redox imbalance and inflammatory response in diabetic mice, and notably attenuated liver injury in diabetic mice. The phytochemical study resulted in the acquisition of nine alkaloids. Among them, glucoindican was identified as the primary active constituent, which potently inhibited the NF-κB, MAPK, and PI3K/AKT cascades, consequently blocking the synthesis of pro-inflammatory cytokines to treat diabetes.

CdA possesses notable antidiabetic functions in vivo. The indole alkaloids are the main components of CdA, and the underlying mechanism mainly includes antioxidation and anti-inflammation. These findings provide evidence for the potential translation of C. davidii into a therapeutic agent for the treatment of diabetes.

There is limited understanding of the impact of glucose monitoring technologies on relationships with food and eating behaviours for those with type 2 diabetes mellitus (T2DM).

With previous reviews focused on T2DM treatment burden and self-management, this review aims to enhance understanding of the impact of flash or continuous glucose monitoring (FGM/CGM) on users' eating behaviours and relationships with food as this can inform important recommendations for CGM use amongst those with T2DM.

A systematic search was conducted across four databases: Scopus, Medline, CINAHL and PubMed, from October-November 2024, following PRISMA guidelines. Studies were quality appraised and qualitative data was synthesized using thematic synthesis.

Thirteen studies met inclusion criteria and their findings were included in the synthesis. The review revealed that F/CGM enhanced participants' nutritional awareness, supported personalised dietary experimentation, and fostered intentional eating behaviours. However, data overload and emotional burden were also reported, with some users experiencing restrictive or distressing impacts on their relationship with food. These insights inform practical recommendations for CGM implementation with those with T2DM.

Further research should aim to identify factors that contribute to successful adaptation to CGM and explore the long-term effects of CGM use on food relationships.

Machine learning models are increasingly used to predict patients at risk of high health care usage for targeted interventions.

This study aimed to evaluate the predictive performance of multiclass ensemble models across different levels of health care usage and assess their potential application through real-world economic impact analysis.

A total of 4 previously developed binary classification models (base learners)-boosted trees, multivariate adaptive regression splines, multilayer perceptron, and logistic regression-were extended using a stacking ensemble approach. These base learner models generated individual-level predicted probabilities, which were used as inputs to build multiclass prediction models forecasting usage across defined strata: length of stay (LOS) of <7, 7-13, 14-29, and ≥30 days, and emergency department (ED) visits of <3, 3-4, 5-9, and ≥10 visits. In total, 3 ensemble algorithms were evaluated: random forest, boosted trees, and linear support vector machines. Ensemble models were trained on registry data from 2020-2021 and temporally validated on 2021-2022 data. Performance was assessed using multiclass area under the receiver operating curve, accuracy, and confusion matrix-derived metrics. Economic impact was estimated via Monte Carlo simulations using inpatient billing data, assuming a 20% cost reduction in the following year.

The models were trained on 108,886 patients and validated on 111,004 patients. Among all ensemble configurations, boosted tree models regardless of base learner achieved the highest performance, with multiclass area under the receiver operating curve scores of 0.6877 (95% CI 0.6927-0.7255) for LOS and 0.7601 (95% CI 0.7301-0.7654) for ED visits, and corresponding accuracies of 0.6522 (95% CI 0.6465-0.6579) and 0.7457 (95% CI 0.7405-0.7508), respectively. In the validation set, these models correctly assigned 30.3% of inpatient LOS and 39.8% of ED visits to the correct class, identifying 77% of future inpatient users and 73.9% of future ED users. Economic impact analysis for LOS identified the boosted tree model with logistic regression base learner as dominant, achieving a simulated average cost reduction of SGD $152 million (US $111 million), SGD $2.4 million (US $1.75 million; 1.5%) more than the next best model using a multilayer perceptron base learner.

Ensemble models can effectively predict multilevel health care usage and potentially generate meaningful cost savings when applied to real-world settings. These models may support targeted interventions and guide planning and budgeting in diabetes-related population health programs.

China's diabetes epidemic faces critical gaps in glycemic control, with only 50.1% of treated patients achieving hemoglobin A1C (HbA1c) targets in 2021. Conventional interventions struggle with scalability in primary care, particularly for vulnerable populations.

This study aimed to evaluate the use of a WeChat-based health education tool (the WeWalk mini program, the Bayu Health public account, and a WeChat group) for improving glycemic control in community-dwelling patients with type 2 diabetes mellitus.

This multicenter randomized controlled trial enrolled 600 adults with type 2 diabetes from 3 communities in Chongqing, randomly allocating participants 1:1 to either a 12-week WeChat-based intervention (n=300, 50%) or a control group (n=300, 50%) in September 2020. The control group received 4 face-to-face traditional health education sessions, whereas the intervention group participated in a digital program: a 4-week course followed by an 8-week practical implementation. At baseline and 12 weeks after the intervention began, both groups were examined in terms of HbA1c and fasting blood glucose (FBG) as the primary outcomes, as well as variables such as blood lipid profile, blood pressure, and physical fitness-related indexes as secondary outcomes. Longitudinal glycemic control was assessed through triplicate FBG measurements extracted from standardized electronic health records at the 2-year follow-up. Independent t tests or Mann-Whitney U tests were used to assess changes from baseline to follow-up between groups.

A total of 92.7% (556/600) of the participants completed the 12-week follow-up visit. The WeChat-based intervention demonstrated superior glycemic control outcomes, with intervention participants achieving a 0.59% greater HbA1c reduction than controls (-0.03% vs 0.56%; P<.001) and significant improvements in FBG levels (-0.69 vs 0.00 mmol/L; Δ=0.69; P=.001). Subgroup analysis revealed that WeChat-based health education was significantly effective in patients with diabetes with a disease duration of <10 years, educational level of junior high school or lower, and annual family income of <CN ¥50,000 (US $7172.10) . These benefits persisted throughout the 2-year follow-up, where the intervention group maintained lower FBG levels (6.87 vs 7.35 mmol/L; P=.001).

WeChat-based health education was beneficial for glycemic control in primary health care settings. However, the sustained efficacy and feasibility of this approach require further investigation.

The aim of this review was to examine the association between diabetic peripheral neuropathy (DPN) and deficiency of fat-soluble vitamins A, D, and E. Vitamins A, D, and E are involved in crucial neuroprotective mechanisms through the regulation of gene expression and antioxidant activity. Vitamin A appears to actively regulate the expression of nerve growth factor (NGF), essential for the development, survival, and regeneration of peripheral neurons. Vitamin D exhibits immunomodulatory effects, reduces the production of pro-inflammatory cytokines, and enhances neuroplasticity. Similarly, vitamin E, through its potent antioxidant action, minimises damage by reactive oxygen species (ROS), which are responsible for demyelinating lesions. Vitamin supplementation studies have shown a potential symptom improvement or reversal. These findings underline the necessity for early detection and correction of vitamin deficiencies as an integral component of a comprehensive strategy for the prevention and management of DPN.

Aurora Kinase A (AKA) is overexpressed in prostate cancer and promotes disease progression through oncogenic pathways and immune modulation. AKA inhibition faces challenges in clinical trials, and alternate methods, such as degradation, may overcome some of these barriers. We evaluated SP-2-067, a novel selective AKA degrader, in C4-2 prostate cancer cells using viability assays, Western blot analysis, and siRNA knockdown. SP-2-067's dose- and time-dependent effects on AKA, related proteins (AUNIP and NINEIN) and immune checkpoint (IC) proteins (PD-L1 and LAG3) were quantified. In silico analysis of TCGA datasets examined the clinical link of the AKA axis and its correlation with immune cell infiltration in prostate cancer. SP-2-067 demonstrated potent dose-dependent anti-proliferative effects. The compound selectively degraded AKA over Aurora Kinase B (AKB) with sustained effects at 72 h. Treatment induced marked PD-L1 upregulation (~ 20-fold at 3 h) and concurrent LAG3 downregulation. siRNA experiments confirmed these effects were AKA-dependent. TCGA analysis revealed AKA, AUNIP, and NINEIN overexpression in prostate cancer correlates with distinct immune-cell infiltration. SP-2-067 is a novel, selective AKA degrader that modulates IC expression by upregulating PD-L1 and downregulating LAG3, indicating its potential for immunomodulation through AUNIP- and NINEIN-mediated mechanisms. Additionally, AKA, AUNIP, and NINEIN are linked to a distinct immune cell profile, presenting an opportunity to reshape the tumor immune microenvironment and support combinatorial strategies with IC inhibitors in prostate cancer.

Liver cancer ranks as the fourth most common malignant tumor and the second leading cause of cancer deaths in China. Guangxi is a high-risk region, with a crude incidence rate of 41.65/10⁵, significantly exceeding the national average. At a regional scale, liver cancer incidence in Guangxi exhibits a distinct zonal distribution, which shows notable spatial coupling with cadmium (Cd) concentrations in rice grains, though the underlying mechanism remains unclear. This study systematically investigated this relationship by collecting hepatocellular carcinoma incidence and rice grain Cd data from 44 counties/cities, analyzing blood Cd levels in 105 patients and 105 healthy controls, and comparing paired blood (n = 316) and rice samples (n = 216) from five typical areas. Results demonstrated a significant spatial coupling and correlation between rice Cd distribution and liver cancer incidence. Liver cancer patients had elevated blood Cd levels. High consistency between blood Cd and rice grain Cd levels identified rice consumption as the primary exposure pathway. This study suggests a link between rice cadmium levels and liver cancer in Guangxi, potentially contributing to its geographical distribution. This provincial-scale study provides fundamental data for understanding the chronic hepatocarcinogenic effects of Cd exposure via rice.