Regression modeling for multivariate count data often struggles with assumption of overdispersion and correlation among response variables. To address these issues, this study proposes a new model called Multivariate Correlated Poisson Generalized Inverse Gaussian Regression (MCPGIGR), which integrates random effects through common shock variables and allows for flexible mean structures via a log-link function. This research develops a Maximum Likelihood Estimation (MLE) and Maximum Likelihood Ratio Tests (MLRT) to evaluate both simultaneous and partial significance of predictors. We conduct simulation studies to assess the consistency and performance of the proposed estimators. Furthermore, in an application to maternal and neonatal mortality across 38 districts/cities in East Java (Indonesia), MCPGIGR substantially improves model fit relative to a Multivariate Poisson Regression (MPR) baseline (AICc decreases from 2378.63 to 1924.60 for ). The proposed framework provides a practical and flexible tool for analyzing correlated, overdispersed multivariate counts in public health and related domains. The highlights of this research are: • The MCPGIGR model introduces a correlated multivariate count regression framework with exposure adjustment. • It provides robust parameter estimation and hypothesis testing via MLE and MLRT. • MCPGIGR demonstrates improved model fit and practical interpretability in public health applications.

This study developed and validated the Critical Right Scale (CRS) to measure emerging critical right attitudes and revised the Critical Social Justice Attitudes Scale (CSJAS-R), replicating its psychometric evaluation. A nationwide convenience sample of Finnish adults (n = 626) completed an online survey. Item screening used exploratory factor analysis with oblique rotation and loading and residual correlation criteria. Confirmatory factor analysis (CFA) and measurement invariance testing were conducted in lavaan using full information maximum likelihood. The final CRS consisted of five items with high reliability (α = 0.89, ω = 0.90) and good model fit in both male and female subsamples, with pooled-sample residual misfit judged minor given subgroup results. The CSJAS-R comprised six items with strong reliability (α = 0.88, ω = 0.89) and excellent fit. Both scales met configural and metric invariance; partial scalar invariance was achieved after freeing a small number of item intercepts. CRS scores were strongly associated with right-wing and conservative self-placement with higher scores concentrated among Finns Party and Christian Democrat voters, and weakly linked to perceived oppression. CSJAS-R scores were strongly associated with left-wing and liberal self-placement with higher scores concentrated among Left Alliance and Greens voters, and had a small-to-moderate association with justification of political violence. CRS and CSJAS-R were strongly negatively correlated (r = -0.62), indicating divergent validity. Both CRS and CSJAS-R demonstrated strong psychometric properties and distinct ideological profiles, providing validated tools for studying political attitude structures at opposing ends of the ideological spectrum.

We investigate maximum likelihood estimation for the drift parameters of stochastic Volterra processes in the ergodic regime. In our first result, we establish the equivalence of laws under general changes of drift and provide the corresponding Radon-Nikodym derivative. This allows us to develop a rigorous maximum likelihood estimation framework. As an application, we study the Volterra Ornstein–Uhlenbeck process in the ergodic regime, considering both continuous-time and high-frequency discrete-time observations. In both regimes, we prove the consistency and asymptotic normality of the maximum likelihood estimators. A key intermediate result, which may be of independent interest, is a uniform Birkhoff-type theorem under an asymptotic independence condition. This theorem yields a locally uniform Law of Large Numbers over the parameter space.

This paper proposes a novel low-rank approximation of the State-Space Model (SSM) with spatially correlated innovations for the analysis of multivariate spatio-temporal data. The SSM’s measurement equation is based on a linear coregionalisation model, which describes the cross-correlation between the observed variables, while the Matérn Gaussian innovation term in the state equation is modelled using the Stochastic Partial Differential Equation (SPDE) approach, allowing a finite-dimensional representation of the latent processes using basis functions defined on spatial meshes. Dimensionality reduction is achieved by appropriately reducing the number of nodes in the meshes. Inference on the model parameters is performed via Maximum Likelihood Estimation (MLE), implemented through the Expectation–Maximisation (EM) algorithm, which features closed-form updating formulas for most parameters and efficient numerical routines for the remainder. We derive theoretical results on the accuracy and convergence of the low-rank approximation and validate them through simulation studies. The EM algorithm and the likelihood derivatives required for inference are implemented in Python/JAX, enabling automatic differentiation and scalable execution across all available local CPU cores, with native support for GPU and TPU acceleration. By analysing a large bivariate air-quality dataset, we demonstrate that reducing the number of nodes by 75% enables model estimation to be 15.8 times faster with only a 15% increase in validation error. We also compare our approach with SPDE-INLA alternatives, demonstrating improved computational scalability while maintaining comparable predictive performance.

Psychometric characteristics of the Persian version of the State Self-Compassion Scale in patients with cardiovascular diseases.

Neuroimaging studies in humans and translational animal models consistently demonstrate ECT-induced hippocampal volume increases. However, evidence linking hippocampal volume changes to clinical improvement in depression has been inconsistent. This systematic review with meta-analysis extends previous work to investigate whether hippocampal volume changes following the completion of an ECT-index series are associated with the clinical course of depression. We conducted a systematic review and meta-analysis following PRISMA guidelines. Studies were eligible if they assessed hippocampal volume and included at least three imaging timepoints, with two performed after completion of the ECT-index series (t1=pre-ECT, t2=post-ECT, t3=follow-up). Standardized mean change (SMC) was calculated for hippocampal volume and depressive symptom severity across timepoints. Pooled estimates were derived using random-effects models with restricted maximum likelihood (REML) estimation to account for between-study heterogeneity. Meta-regression analyses were performed to evaluate associations between volumetric changes and trajectories of depressive symptoms post-ECT. Fifteen studies (N=447 patients) were included in the systematic review, with six studies (N=151 patients) contributing complete volumetric data for meta-analysis. Hippocampal volumes increased significantly from t1 to t2 and largely returned to baseline at t3. Depression severity decreased substantially from t1 to t2 and remained stable at follow-up. Meta-regression analyses indicated no significant associations between changes of hippocampal volume and depressive symptoms from post-ECT to follow-up. Sensitivity analyses confirmed robustness of the findings. Hippocampal volume increases following ECT are transient and not associated with the clinical course of depression.

Privacy concerns on short-form video platforms and creativity among communication students in China: A longitudinal mediation model involving TikTok use motives and IT identity.

Allogeneic hematopoietic stem cell transplantation (allo-HSCT) remains the only curative treatment for many hematologic malignancies, yet patient outcomes vary significantly. Patient expectations influence recovery in other medical contexts, yet their role in allo-HSCT remains unclear. This pilot study examined whether pre-transplant treatment expectations predict psychological and immunological outcomes post-transplant. In this prospective, single-center observational cohort study, 42 patients undergoing allo-HSCT were assessed at baseline (T0), discharge (T2), and six months post-transplant (T3). Questionnaires measured illness-related disability (PDI, primary endpoint at T3), treatment expectations (TEX-Q), quality of life (FACT-Leu), depression (PHQ-9), and anxiety (GAD-7). Immunological markers, including inflammatory markers were collected at T0 and T3. Baseline-adjusted regression analyses with full-information maximum likelihood estimation were used. P-values were corrected for multiple comparisons using a false discovery rate approach. Baseline expectations were associated with psychological outcomes at hospital discharge and immunological and inflammatory markers at six-month follow-up: For instance, negative impact expectations were associated with higher disability (β=0.522, p<0.001), depression (β=0.693, p=0.009), anxiety (β=0.737, p=0.003), and lower quality of life (β=-0.576, p<0.001) at T2. Benefit expectations were associated with higher lymphocyte counts (β=0.453, p<0.001) and lower CRP levels at T3 (β=-0.28, p=0.011). Positive impact expectations were associated with more favorable T-cell subsets. Pre-transplant expectations may influence psychological and immune recovery following allo-HSCT. Addressing expectations could enhance outcomes and should be explored in future intervention studies.

Community-based interventions to improve tuberculosis treatment outcomes: A meta-analysis.

We evaluate the performance of targeted maximum likelihood estimation (TMLE) for estimating the average treatment effect in missing data scenarios under varying levels of positivity violations. We employ model- and design-based simulations, with the latter using undersmoothed highly adaptive lasso on the "WASH Benefits Bangladesh" data set to mimic real-world complexities. Five missingness-directed acyclic graphs are considered, capturing common missing data mechanisms in epidemiological research, particularly in one-point exposure studies. These mechanisms include also not-at-random missingness in the exposure, outcome, and confounders. We compare eight missing data methods in conjunction with TMLE as the analysis method, distinguishing between non-multiple imputation (non-MI) and multiple imputation (MI) approaches. The MI approaches use both parametric and machine learning models. Results show that non-MI methods, particularly complete cases with TMLE incorporating an outcome-missingness model, exhibit lower bias compared to all other evaluated missing data methods and greater robustness against positivity violations across. In comparison MI with classification and regression trees (CART) achieve lower root mean squared error, while often maintaining nominal coverage rates. Our findings highlight the trade-offs between bias and coverage, and we recommend using complete cases with TMLE incorporating an outcome-missingness model for bias reduction and MI CART when accurate confidence intervals are thepriority.

The main tools for making clinical decisions based on efficient use of resources are economic evaluation studies that allow the assessment of both the costs and benefits of different therapeutics, with appropriate guidelines for preparing reports. This study aimed to develop a checklist of consumable cost elements to be considered in estimates for micro-costing studies in peritoneal dialysis (PD). Four stages were conducted, followed by data analysis and interpretation. Three stages were carried out to develop the direct cost elements questionnaire: 1st - designing the first version of the checklist; 2nd - evaluating and expanding it using the Delphi method; 3rd - conducting two expert panels; and 4th - applying the questionnaire to professionals from 18 Latin American countries. Inclusion criteria: professionals with at least one year of clinical and/or administrative experience in PD. A discrete probability distribution adjustment was performed. Distribution lots were considered according to the category of cost elements for each country. The maximum likelihood estimation method was applied, and the statistical classification of the adjustments was assessed using the Akaike Information Criterion. A total of 596 questionnaires, comprising seven dimensions and 41 elements, were validated. From the results of each batch, it was possible to segment the elements into three choice options, with the probability of evaluating an element as very important, thus allowing for the classification of the cost elements. The checklist favors more equitable economic dimensioning in comparative studies, making it possible to compare economic values in PD across countries, while considering the appropriate cost elements.

The Rasch Rating Scale Model (RSM) is widely applied in questionnaire analysis, yet many Rasch software packages require installation and advanced technical expertise. This study presents RaschOnline, a web-based application deployed on Google App Engine that enables full RSM analysis through a browser interface. The system implements a streamlined three-step workflow—data upload, joint maximum likelihood estimation, and automated reporting. RaschOnline produces item and person parameter estimates, standard errors, and fit statistics, and provides essential diagnostic visualizations, including Wright maps, KIDMAP displays, category average plots, person outfit plots, and item dimension plots. All computations are performed on cloud-based infrastructure, ensuring platform independence and eliminating local software requirements. A simulation study demonstrates that RaschOnline yields item difficulty estimates and fit statistics comparable to those generated by Winsteps. These findings indicate that RaschOnline offers an accessible and scalable alternative for Rasch analysis in applied research and education.

Mupirocin-based nasal decolonization to prevent Staphylococcus aureus surgical site infections: A meta-analysis of randomized control trials.

Although the association between depressive symptoms and cardiovascular disease (CVD) has been extensively studied, evidence for a long-term causal relationship remains sparse. This research employed advanced causal inference techniques to evaluate this longitudinal effect and its potential reversibility. We analyzed data from 37,668 participants across three prospective cohorts: CHARLS (China), HRS (USA), and KLoSA (South Korea). Applying the Longitudinal Targeted Maximum Likelihood Estimation (LTMLE) method across five time points, we assessed the causal effect of depression (defined by CES-D scale cutoffs) on self-reported physician-diagnosed CVD. Subgroup analyses were conducted by gender and age. Multiple sensitivity analyses were conducted to validate the robustness of the findings. Across all cohorts, the risk of CVD significantly increased with longer follow-up durations under persistent depressive symptoms. For example, in CHARLS, the adjusted odds ratio (OR) increased from 1.570 (95% CI: 1.398-1.798) at Year 2 to 2.097 (95% CI: 1.659-2.651) by Year 9. Further analysis of different exposure sequences of depressive symptoms revealed that the risk of CVD increased gradually with a greater cumulative number of waves with depressive symptoms, whereas it decreased correspondingly with more waves without depressive symptoms, demonstrating a pattern consistent with reversible association. This multi-cohort study provides evidence for a longitudinal causal relationship between depressive symptoms and CVD, showing temporal cumulative effect and a risk pattern consistent with reversible association. These results highlight the need to integrate mental health care into CVD prevention.

The effect of perceived stress on turnover intention among Chinese geriatric caregivers of nursing institutions: The mediation of psychological resilience and coping style.

Although full-information maximum likelihood (FIML) estimation is widely used for diagnostic classification models (DCMs), its computational efficiency deteriorates sharply in high-dimensional settings. This scalability challenge is increasingly critical as DCMs are applied to large-scale assessments, psychological testing and longitudinal studies involving many attributes. We propose a composite marginal likelihood (CML) estimation approach via expectation-maximization (EM) algorithm (CML-EM) for higher-order DCMs (HO-DCMs) as an alternative. The central premise is that, because response probabilities depend only on the attributes specified by the Q-matrix and because of the conditional independence assumption of HO-DCMs, the full likelihood can be partitioned into low-dimensional subsets of items and attributes. This reduces both attribute and response spaces in the E-step to that of each subset, resulting in substantial computational gains that become more pronounced with larger sample sizes and numbers of attributes. We also introduce a subset-construction procedure that ensures both efficiency and feasibility of CML-EM and present two methods for attribute classification. Simulation results demonstrate that CML-EM is significantly faster than FIML while maintaining accurate parameter recovery and acceptable classification performance. The practical utility of the method is further illustrated through an empirical application to a high-dimensional personality assessment.

This cross-sectional study examines the effects of multidimensional social support on anxiety and depression symptoms in colorectal cancer patients, with particular focus on the mediating role of psychological resilience in this association. We recruited patients from the Shantou Central Hospital between October 2023 and December 2024 using consecutive recruitment. Participants completed validated instruments: the Perceived Social Support Scale (PSSS), Connor- Davidson Resilience Scale (CD-RISC), and Hospital Anxiety and Depression Scale (HADS). Structural equation modeling (SEM)-based path analysis with maximum likelihood estimation was performed to analyze the mediation pathways after establishing bivariate correlations through Pearson's analysis, using composite scores of the scales (rather than latent variables). We recruited 268 Chinese colorectal cancer patients (51.49% male, mean age predominantly 45-65 years, 34.33% at Stage Ⅲ and 31.72% at Stage Ⅳ) who received surgery alone or combined with chemotherapy/radiotherapy. The cohort demonstrated moderate resilience levels (58.62 ± 12.14), with mean perceived social support scores of 42.38 ± 10.72 and HADS scores of 22.40 ± 7.62. Significant correlations emerged: social support positively correlated with resilience (r = 0.62, P < 0.01) and negatively with anxiety/depression (r=-0.54, P < 0.01), while resilience showed inverse associations with anxiety/depression (r=-0.67, P < 0.01). The SEM revealed excellent model fit (CFI = 0.97, RMSEA = 0.04). Social support was associated with lower anxiety/depression both directly (β=-0.41, P < 0.001) and indirectly via an association with greater resilience enhancement (β=-0.29, P < 0.01), with resilience accounting for 29.17% of the variance in the social support-anxiety/depression relationship. Our findings suggest psychological resilience acts as a partial mediator in the pathway between perceived social support and mental health among colon cancer patients. The results highlight the importance of integrating psychosocial interventions that simultaneously strengthen external support systems and cultivate internal resilience capacities.

This study compared the effects of 3 different approaches to handling missing data (listwise deletion of participants with missing data, mean imputation, and full information maximum likelihood [FIML]) when predicting functional independence trajectories over 10 years in older adults after traumatic brain injury (TBI). Twenty-three TBI Model Systems (TBIMS) inpatient rehabilitation facilities in the United States. Adults who sustained a complicated mild, moderate, or severe TBI at age 60 years or older and needed inpatient rehabilitation. They had to meet all eligibility criteria and have one or more functional independence measure (FIM) scores at 1, 2, 5, or 10 years post-TBI from the TBIMS national database. Retrospective analysis of observational data using hierarchical linear models. FIM total scores at 1, 2, 5, and 10 years post-TBI. Different missing data approaches led to drastically different findings. Model comparisons supported a quadratic effect of time only in the listwise deletion model and found no other significant predictors. Linear trajectories were found in the mean imputation and FIML models. For both these models, older age, underrepresented minority status, unemployment at injury, longer posttraumatic amnesia duration, and pre-injury limitations all predicted lower overall FIM trajectories. However, when compared with the mean imputation model, the FIML-estimated b-weights were larger with smaller P-values. Years of education significantly predicted higher overall FIM trajectories in the mean imputation model but not the FIML model, likely because of the artificial shrinking of the estimated b-weight standard errors in mean imputation. History of mental health treatment predicted lower FIM trajectories only in the FIML model. These findings show that it is critical to use appropriate modern methods to handle missing data because the method can affect outcome trajectory shape and identification of relevant predictor variables. Using older methods for handling missing data, such as listwise deletion, greatly reduces predictive ability, resulting in less generalizability and imprecision in longitudinal rehabilitation research.

ABSTRACT Hydraulic infrastructure design often analyses the region-specific rainfall characteristics through intensity–duration–frequency (IDF) curves. Unavailability of fine-scale data may lead to inaccuracies, particularly in capturing extremes for short durations, common in regions such as northeast India. The study focuses on disaggregating coarser temporal resolution rainfall data (4 h) to finer resolution data (15 min) using the multiplicative random cascade model (MRCM). The model framework utilizes 15 min point rainfall data and incorporates a weighted approach along with two micro-canonical models (a baseline model and a dependent model) for parameter estimation using maximum likelihood estimation. The dependent model outperforms the baseline model, and the output is utilized to derive IDF curves using generalized extreme value distribution, which are validated using observed data. By leveraging the cascade model, the study aims to enhance the temporal resolution of rainfall data and generate detailed IDF curves in data-scarce regions – crucial for effective hydrological planning.

For accurate disease characterization using positron emission tomography (PET), it is desirable to image multiple radiotracers in a single scan. Conventional PET methods cannot do this due to the indistinguishable annihilation photons produced by different radiotracers. One approach is to label one radiotracer with a positron+prompt-gamma ($\beta^+\!\!-\!\!\gamma$) isotope producing triple coincidences, and another with a pure positron-emitting ($\beta^+$) isotope producing double coincidences. However, $\beta^+\!\!-\!\!\gamma$ emitters present challenges in correctly identifying the two annihilation photons, or equivalently, assigning the correct line-of-response (LOR) to triple-photon coincidence events. Here, we propose a maximum likelihood estimation (MLE) framework leveraging spatial, timing, and energy information to determine the most probable LOR. Simulation studies validated the method: simulations showed over 96\% and 94\% accuracy for LOR assignment of $\beta^+\!\!-\!\!\gamma$ emitters $^{22}$Na and $^{124}$I point sources, respectively. Furthermore, simulated phantom imaging of $^{22}$Na or $^{124}$I distributions alongside a $\beta^+$ emitter demonstrated that MLE LOR assignment achieved comparable image quality-measured by contrast recovery coefficient (CRC) and cross-talk ratio (XR)-to benchmark methods, where the prompt gamma was identified using an energy threshold ($\geq 650$ keV) for $^{22}$Na and as the highest-energy photon for $^{124}$I.