Tag: statistics

Trop Med Int Health. 2026 May 19. doi: 10.1111/tmi.70168. Online ahead of print.

ABSTRACT

OBJECTIVE: To analyse the epidemiological characteristics, spatial distribution and predictors of TB-related mortality in Risaralda from 2020 to 2024.

METHODS: A retrospective observational study was conducted using secondary data from the National Public Health Surveillance System (SIVIGILA). Descriptive statistics, spatial analysis (choropleth maps) and binary logistic regression were used to identify sociodemographic and clinical factors associated with TB case fatality. Model performance was evaluated using ROC curves, sensitivity, specificity and 5-fold cross-validation.

RESULTS: A total of 2081 confirmed TB cases were analysed, with a case fatality rate of 13.2%. Mortality was higher among individuals ≥ 65 years (OR: 3.87), hospitalized patients (OR: 7.33), HIV-positive individuals (OR: 3.49), those with undernutrition (OR: 1.89), males (OR: 1.45) and Indigenous persons (OR: 1.67). The logistic model showed good discrimination (AUC: 0.82), with sensitivity of 79.2% and specificity of 73.8%. Cross-validation confirmed model stability (AUC: 0.81). Spatial analysis revealed high mortality in rural municipalities with low case burden, suggesting health inequities.

CONCLUSION: TB mortality in Risaralda is influenced by age, comorbidities and social determinants, with pronounced territorial disparities. These findings support the integration of spatial and predictive tools in TB surveillance to strengthen targeted interventions and reduce mortality in high-risk populations.

PMID:42153348 | DOI:10.1111/tmi.70168

Database (Oxford). 2026 Jan 15;2026:baag026. doi: 10.1093/database/baag026.

ABSTRACT

Cancer metastasis involves complex molecular mechanisms that cannot be fully explained by individual gene expression profiles. Previous studies have shown that correlations between RNA and miRNA expression can capture metastatic behaviour more effectively than expression of individual genes. However, no publicly available databases provide systematic analysis of RNA-miRNA correlations specific to cancer metastasis. We developed an efficient computational method to identify differential correlations between miRNAs and RNAs that are specific to individual tumour samples. Using data from The Cancer Genome Atlas (TCGA), we computed differential correlations for tumour samples across 9 cancer types and 21 metastatic sites, encompassing ~200 million RNA-miRNA pairs. Statistical analysis identified RNA-miRNA pairs with site-specific correlations using Mann-Whitney U-tests. MetaCancerDB contains RNA-miRNA correlation networks for 9 primary cancer types and 21 metastatic sites. Site-specific correlations showed distinct patterns, with lung metastasis displaying the most conserved correlations across cancer types. Survival analysis revealed that specific RNA-miRNA pairs are prognostic for patient outcomes in a metastatic site-dependent manner. MetaCancerDB provides a comprehensive resource for exploring RNA-miRNA correlations in cancer metastasis. The database enables researchers to identify molecular signatures specific to metastatic sites and can serve as a foundation for developing predictive biomarkers. MetaCancerDB is freely available for academic purposes.

PMID:42153344 | DOI:10.1093/database/baag026

Biostatistics. 2026 Jan 20;27(1):kxag009. doi: 10.1093/biostatistics/kxag009.

ABSTRACT

A live clinical question is: which patients benefit from intensive care unit (ICU) transfer? Motivated by this question we address the problem of estimating conditional average treatment effects (CATE) in the presence of unmeasured confounding, by leveraging instrumental variables (IVs). CATE-learners (eg R-learner) developed for settings without unmeasured confounding can readily incorporate IVs by substituting the observed exposure with first-stage predictions from a regression of treatment on IVs and covariates. Such predictions may be obtained via flexible data-adaptive methods (eg statistical or machine learning procedures) to alleviate concerns about model misspecification. However, the large regularization bias typical of data-adaptive predictions may propagate into the CATE estimates, resulting in poor accuracy. Neyman-orthogonal learners have therefore been developed, which prevent this by “insulating” the resulting CATE estimates against bias and estimation errors in these predictions. However, synthetic data simulations reveal that previously proposed Neyman-orthogonal learners for IV regression perform poorly. We remedy this problem using infinite-dimensional targeted learning, which strategically tailors first-stage predictions to perform well in their ultimate task: delivering accurate, precise CATE estimates. The resulting targeted Neyman-orthogonal learner is easy to construct based on arbitrary, off-the-shelf learners. It can handle continuous or discrete exposures, and arbitrary types and numbers of IVs and covariates. Simulation studies and a re-analysis of the benefits of ICU transfer show substantial enhancements in performance, underscoring the importance of the proposed IV-learner.

PMID:42153341 | DOI:10.1093/biostatistics/kxag009

J Rural Health. 2026 Mar;42(2):e70167. doi: 10.1111/jrh.70167.

ABSTRACT

PURPOSE: This study examined whether the Supportive Services for Veteran Families (SSVF) program has a differential impact on healthcare utilization and costs among rural and urban Veterans experiencing housing instability. Understanding geographic variation in program effects is important for guiding implementation and resource allocation within the Veterans Affairs (VA) healthcare system.

METHODS: We conducted a retrospective cohort study using national VA administrative data from October 1, 2018 to September 30, 2023. Veterans were classified as SSVF participants or eligible non-SSVF Veterans based on indicators of housing instability. A target trial emulation framework with inverse probability weighting was used to adjust for baseline differences. Rurality was defined using Rural-Urban Commuting Area codes applied to Veterans’ residential addresses recorded in VA administrative data. Weighted longitudinal models estimated quarterly changes in healthcare utilization and costs and tested whether effects varied by rurality.

FINDINGS: SSVF enrollment was associated with reductions in inpatient utilization and costs and modest increases in outpatient visits, resulting in overall decreases in total VA healthcare costs. Emergency department use showed small reductions among urban Veterans and little measurable change among rural Veterans. Overall patterns of healthcare utilization and spending were similar across rural and urban Veterans, and statistical tests did not indicate significant rural-urban differences in SSVF effects.

CONCLUSIONS: SSVF participation was associated with shifts away from inpatient care and toward greater outpatient engagement among Veterans experiencing housing instability. These patterns were observed among both rural and urban Veterans, suggesting that the healthcare benefits of housing stabilization programs are similar across geographic settings.

PMID:42153335 | DOI:10.1111/jrh.70167

J Anim Sci. 2026 May 18:skag159. doi: 10.1093/jas/skag159. Online ahead of print.

ABSTRACT

Selenium (Se) is known to improve gut health in animals, yet research on the effects of different dietary Se compounds on the intestinal health of broilers remains limited. Therefore, this study evaluated the effects of sodium selenite (SS), selenium-enriched yeast (SY), selenomethionine (SM), and nano-selenium (NS) on gut microbiota and their metabolites, intestinal antioxidant capacity, immune response, and gut morphology in broilers, and investigated the potential molecular mechanisms by which Se influences intestinal function in broilers. A total of 360 1-day-old male yellow-feathered broilers with an average body weight of 37.00 ± 0.17 g were randomly assigned to five treatments, each comprising six replicates with 12 chicks per replicate. Broilers received either a basal diet or a basal diet supplemented with SS, SM, SY, and NS at 0.5 mg Se/kg for 56 days. Data were analyzed using one-way ANOVA with Tukey’s post-hoc test for multi-group comparisons, and statistical significance was set at P < 0.05. Supplementation with SY increased ileal concentrations of secretory immunoglobulin A by 74.87% and interleukin-10 (IL-10) by 54.90%, enhanced ileal activities of total superoxide dismutase (T-SOD) by 123.55% and catalase by 197.20%, and elevated cecal acetate by 35.67% and total short-chain fatty acids (SCFAs) by 28.78%, as well as ileal ursodeoxycholic acid concentration by 154.05% (P < 0.05). Dietary SS elevated ileal IL-10 concentration by 44.72% and glutathione peroxidase activity by 93.74% while reducing tumor necrosis factor-alpha level by 26.46% (P < 0.05). Supplemental NS increased cecal concentrations of acetate by 45.56%, propionate by 85.94%, and total SCFAs by 39.68% (P < 0.05). Compared with the SS, SY supplementation improved jejunal total antioxidant capacity by 81.08% and ileal T-SOD activity by 84.22% (P < 0.05). Additionally, dietary Se supplementation increased the abundances of potentially beneficial bacteria, including Lactobacillus, Parabacteroides, Akkermansia, and UCG_005 (P < 0.05). Genes such as CCR9, CD28, MUC2, HTR6, KCNK5, and SLC9A3 were up-regulated, while GIP, SSTR2, SST1, and CRHR1 were down-regulated by SS or SY supplementation, indicating involvement in intestinal function. In summary, SS and SY improved intestinal antioxidant and immune functions in broilers, whereas SY and NS enhanced cecal SCFAs production. Moreover, Se supplementation modulated the cecal microbial community in broilers.

PMID:42153328 | DOI:10.1093/jas/skag159

Brief Bioinform. 2026 May 4;27(3):bbag240. doi: 10.1093/bib/bbag240.

ABSTRACT

We discuss the drawbacks of using homogeneity and completeness, statistics defined for the evaluation of clustering accuracy, for evaluating plasmid binning tools accuracy, motivated by the recently published paper “Circling in on plasmids: benchmarking plasmid detection and reconstruction tools for short-read data from diverse species,” by Teixeira et al.

PMID:42153317 | DOI:10.1093/bib/bbag240

Stat Med. 2026 May;45(10-12):e70575. doi: 10.1002/sim.70575.

ABSTRACT

Bang and Zhao address an important methodological gap by proposing meta-analytic methods for cost-effectiveness analysis (CEA) based on separate pooling of incremental costs (ΔC) and incremental effectiveness (ΔE), with summaries displayed on the cost-effectiveness plane. We share the aim of expanding the toolkit for CEA synthesis; however, we believe that several issues need clarification before the proposal can serve as a general blueprint. First, the two motivating examples appear not to satisfy the key combinability criteria described by Shields and Elvidge. Second, the proposal does not incorporate the data harmonization steps recommended for the meta-analysis of economic evaluations by Bagepally et al. Third, separate univariate pooling does not retain the within-study association between ΔC and ΔE, which complicates the joint interpretation of the pooled ICER and its uncertainty. As a possible way forward, we propose a two-stage framework in which a structured combinability assessment and data harmonization precede any quantitative synthesis, and discuss why joint modeling of (ΔC, ΔE) is preferable to separate univariate pooling. In our view, such a framework provides a more defensible route to pooled ICERs, joint uncertainty summaries, and decision-relevant quantities such as cost-effectiveness acceptability curves (CEACs).

PMID:42153314 | DOI:10.1002/sim.70575

Int J Med Robot. 2026 Jun;22(3):e70174. doi: 10.1002/rcs.70174.

ABSTRACT

INTRODUCTION: Endoscopic retrograde cholangiopancreatography (ERCP) is technically challenging, with high operator radiation exposure, physical strain, and a steep learning curve, necessitating a robotic solution. This study evaluated robotic-assisted ERCP feasibility and efficacy in an animal model.

METHODS: In this randomised controlled crossover study, three experienced endoscopists performed 24 ERCP procedures (12 robotic-assisted vs. 12 manual) on four beagles using the YunSRobot system. Primary outcomes included papilla positioning maintenance, cannulation time, and success rate.

RESULTS: Robotic-assisted ERCP demonstrated superior papilla positioning (95.74% vs. 77.25%, p < 0.01), indicating more endoscopic stability during cannulation. The successful cannulation rate was identical in both groups (83.3%). The robotic group required fewer attempts (2.0 vs. 3.5) and experienced no scope dislocations (0 vs. 5). No severe adverse events occurred in either group.

CONCLUSIONS: Robotic-assisted ERCP is feasible, safe, and effective, offering significantly improved endoscopic stability with comparable cannulation success to manual techniques.

PMID:42153283 | DOI:10.1002/rcs.70174

Med Teach. 2026 May 1:1-10. doi: 10.1080/0142159X.2026.2663870. Online ahead of print.

ABSTRACT

INTRODUCTION: Although peer support is known to foster well-being among students in health professions education, the mechanisms driving informal help-seeking behavior remain insufficiently understood. Ongoing psychological distress and persistent barriers to faculty-based support increase the need for accessible resources for these future health care professionals. This study explored whether students reach out when experiencing performance pressure and which opportunities they identify to strengthen their peer support network.

METHODS: This sequential, mixed-methods study was conducted at the Radboud University Medical Center, the Netherlands and spanned a six-year period through two distinct phases of data collection (2018-2024). In Part 1, 26 students in Medicine and Biomedical Sciences completed one-on-one interviews. Qualitative analysis identified six key barriers to informal help-seeking; ‘Stigma’ (S), ‘Harm’ (H), ‘Inutility’ (I), ‘Ego’ (E), ‘Load’ (L), and ‘Disapproval’ (D), together forming the SHIELD-framework. In Part 2, 946 students completed a survey developed from Part 1 findings. A mixed-format question design and advanced statistical analyses were used to examine latent patterns in their informal help-seeking behavior.

RESULTS: Across both study parts, most students concealed their internal struggles, creating a gap between experience and expression. Their motivations were consistently reflected in the SHIELD-framework across gender, program and year of study. ‘Stigma’ and ‘Disapproval’ captured fear of judgment. ‘Harm’ and ‘Load’ reflected concerns about burdening others or worsening personal distress. ‘Inutility’ referred to doubts about the usefulness of disclosure, while ‘Ego’ related to students’ self-image. Many students described suppressing difficult thoughts and emotions, keeping stress confined to their inner world.

DISCUSSION: Students’ tendency to shield their stress reflected a complex interplay of personal and contextual barriers. Our SHIELD-framework offers a practical and conceptual handhold for students and educators to better understand these dynamics. Supporting a shift from shielded stress to shared support requires intentional efforts that normalize vulnerability and foster meaningful peer connection.

PMID:42153275 | DOI:10.1080/0142159X.2026.2663870

Histol Histopathol. 2026 May 19:25092. doi: 10.14670/HH-25-092. Online ahead of print.

ABSTRACT

According to recent studies, periostin expression is significantly higher in cardiac disease and tumor tissues in the majority of cancers compared with normal tissues. Periostin was reported to be expressed in the stroma of Warthin’s tumors. In the present study, we focused on the periostin expression in Warthin’s tumor and malignant lymphomas, which require differentiation from these tumors. This study included 26 patients (28 specimens; 14 men and 12 women) with a mean age of 68.4 (range 51-85) years who underwent parotid tumor resection at the Division of Otorhinolaryngology, Tohoku Medical and Pharmaceutical University, between May 2016 and March 2022, and were clinically and pathologically diagnosed as Warthin’s tumors and malignant lymphomas. Formalin-fixed, paraffin-embedded sections from the surgical specimens were immunostained with anti-periostin antibodies to evaluate the expression and distribution of periostin. Histologically, the tumors were diagnosed as Warthin’s tumors in 13 cases (15 specimens) and malignant lymphomas in 13 cases (13 specimens). An increased expression of periostin was found in 24 (12 Warthin’s tumors and 12 malignant lymphomas) of the 28 samples (85.7%) in the stroma of both Warthin’s tumors and malignant lymphomas. Three distinct patterns of periostin expression were observed: negative, superficial, and infiltrative. Statistically significant differences were found between periostin expression patterns and the histological classification of the tumors. Our results suggest that periostin may be involved in the pathogenesis of Warthin’s tumors and malignant lymphomas, potentially serving as a novel biomarker for these conditions.

PMID:42153264 | DOI:10.14670/HH-25-092