Category: Statistics

Rheumatology (Oxford). 2026 May 11:keag152. doi: 10.1093/rheumatology/keag152. Online ahead of print.

ABSTRACT

OBJECTIVES: To determine the prevalence and predictive correlates of arthritis and joint damage in systemic lupus erythematosus (SLE) patients in the Asia-Pacific Lupus Collaboration (APLC) cohort, and to determine their impact on health-related quality of life (HRQoL).

METHODS: SLE patient data (2013-2020) were collected from the prospective multinational APLC cohort. We defined arthritis according to SLE Disease Assessment Index (SLEDAI-2K) definition, persistent arthritis as arthritis in ≥ 2 consecutive visits, and joint damage according to the Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index (SDI) definition (deforming or erosive arthritis). HRQoL was measured by Short Form Survey (SF36). Descriptive statistics, univariable and multivariable Cox hazard models, and Kaplan-Meier analyses were performed.

RESULTS: During median 2.5(1.0-5.1) years of follow-up, 803/4106(19.6%) patients had arthritis at least once, and 18/3383(0.53%) accrued joint damage. Patients with arthritis were more likely to be female, Caucasian, current smokers at enrolment, and less like to have tertiary education; they also had higher overall disease activity, and lower physical and mental HRQoL. Kaplan-Meier analysis demonstrated that joint damage was more likely in patients with arthritis. Persistent arthritis and longer follow-up were risk factors for joint damage accrual; being from high-income countries was protective. Patients with joint damage also had worse physical HRQoL.

CONCLUSION: Arthritis in the APLC cohort was infrequent compared with other cohorts and was associated with smoking, higher overall disease activity, and damage accrual across multiple domains. Presence of arthritis significantly impacted physical and mental HRQoL. Joint damage was strongly predicted by persistent arthritis.

PMID:42114134 | DOI:10.1093/rheumatology/keag152

Eur Stroke J. 2026 May 6;11(5):aakag045. doi: 10.1093/esj/aakag045.

ABSTRACT

INTRODUCTION: Policy shifts towards home-based care are reshaping the management of stroke survivors, many of whom require long-term support. Home care, which encompasses social care for personal and household tasks and informal care provided by family and friends, plays a crucial role in post-stroke recovery and community reintegration. This study examined home care use up to 15 years post-stroke and its associated factors, and assessed unmet needs for assistance with activities of daily living (ADLs).

PATIENTS AND METHODS: Data from 7885 stroke survivors in the South London Stroke Register (1995-2022) were analysed at 3 months, 1, 5 and 15 years post-stroke. Descriptive analyses examined home care patterns and unmet needs. A Heckman selection model assessed factors associated with home care use while accounting for missing data.

RESULTS: On average, 75% of stroke survivors used home care across 15 years post-stroke, with 83% of care at 3 months and 87% at 15 years being provided through informal care. Home care use was more likely among those with functional dependency (29%; 95% CI, 22%-35%) and those living with family (24%; 95% CI, 21%-27%). Social care use was higher in people with greater dependency (48%; 95% CI, 42%-56%), those living alone (25%; 95% CI, 21%-29%), those with lower deprivation (10%; 95% CI, 1%-20%) or those with a Black ethnic background (6%; 95% CI, 2%-9%). Informal care use was higher among those living with family (12%; 95% CI, 9%-14%), those with moderate dependency (2%; 95% CI, 0.1%-5%) or females (6%; 95% CI, 3%-9%). Unmet needs in ADLs increased over time (12% at 3 months to 17% at 15 years) and were higher among those with moderate compared with severe functional dependency.

DISCUSSION AND CONCLUSIONS: Within home care, informal care remains the predominant long-term support for stroke survivors, persisting up to 15 years after stroke. Addressing health-related, socio-economic, ethnic and gender disparities in home care and unmet needs is essential for equitable community-based stroke care, and caution is needed when promoting home-based care models regarding the distributional impact of home care. Accurate measurement of home care is key to improving post-stroke care models and quality of life.

PMID:42114133 | DOI:10.1093/esj/aakag045

Eur Stroke J. 2026 May 6;11(5):aakag042. doi: 10.1093/esj/aakag042.

ABSTRACT

INTRODUCTION: Glomerular hyperfiltration has previously been associated with cardiovascular events and mortality but has scarcely been investigated in patients with stroke.

PATIENTS AND METHODS: We used pooled data from an individual patient data meta-analysis of prospective, cohort studies of stroke or TIA populations. For this analysis, we included participants from study sites that collected estimated glomerular filtration rate (eGFR) at stroke presentation. Using Cox proportional hazards regression, we investigated the risk of death, any stroke and vascular death according to glomerular hyperfiltration, defined as having an eGFR greater than the age- and sex-adjusted 95th percentile. We also investigated these outcomes according to eGFR as a continuous variable, modelled using fractional polynomials.

RESULTS: A total of 11,175 patients (mean age 70.7 years, 42% female) were included in the analysis, 554 (4.9%) with hyperfiltration. Compared to the normofiltration group (absence of hyperfiltration and eGFR ≥ 60 mL/min/1.73 m2), the hyperfiltration group had a higher rate of all-cause death, 147 per 1000 person-years (95% CI, 119-180) vs 61 (95% CI, 57-66). Compared to normofiltration, hyperfiltration was independently associated with the risk of death from any cause (adjusted hazard ratio [HR] 1.76; 95% CI, 1.46-2.11; P < .001) and the risk of vascular death (adjusted HR 1.68; 95% CI, 1.29-2.17; P < .001). There were non-linear associations of eGFR with risk of death and vascular death, with increasing risk at both low and high eGFR (Pnon-linearity < .001 for both).

DISCUSSION AND CONCLUSION: Glomerular hyperfiltration was associated with a 76% increased risk of death and a 68% increased risk of vascular death in multivariable models adjusted for age, sex and comorbidities. Glomerular hyperfiltration may be associated with adverse health outcomes, specifically in patients with ischaemic stroke. Further research is needed to confirm these findings.

PMID:42114132 | DOI:10.1093/esj/aakag042

Eur Stroke J. 2026 May 6;11(5):aakag039. doi: 10.1093/esj/aakag039.

ABSTRACT

INTRODUCTION: Sex-related disparities in acute ischaemic stroke (AIS) care have been widely reported. However, evidence from Italy remains limited. We aimed to evaluate sex differences in access to revascularisation treatments (RTs) and key time performance indicators in a large Italian cohort.

PATIENTS AND METHODS: We conducted a single-centre, retrospective, observational study including all adults admitted to the emergency department of a comprehensive stroke centre in Rome between January 2015 and December 2022 for suspected stroke. Clinical and demographic characteristics, comorbidities, presenting symptoms, RTs and stroke care time metrics were collected. Multivariable logistic and linear regression models, as well as restricted cubic spline analyses, were used to assess sex-related differences in RTs and time indicators, adjusting for relevant confounders.

RESULTS: Within the 9167 patients, 44.4% had AIS, and 48.2% were women. Among patients with AIS, women were older (P < .001), had higher NIHSS scores at onset (P < .001), a greater comorbidity burden (P < .001) and higher in-hospital mortality (P = .010). No significant sex differences were observed in intravenous thrombolysis and endovascular treatment administration rates. However, median door-to-needle time was 4 min longer in women than in men (P = .030). After adjustment, female sex was not significantly associated with RT administration or in-hospital mortality, but remained an independent predictor of longer DNT [adjusted β -8.242; 95% CI (-15.453 to -1.031); P = .025].

DISCUSSION AND CONCLUSIONS: Although access to RTs was comparable between sexes, women experienced longer DNT, indicating persistent in-hospital barriers to timely stroke care. These findings highlight the need for targeted interventions to achieve full equity in AIS management.

PMID:42114131 | DOI:10.1093/esj/aakag039

J Anim Sci. 2026 May 11:skag133. doi: 10.1093/jas/skag133. Online ahead of print.

ABSTRACT

Estimating individual dry matter intake (DMI) on pasture remains costly, labor-intensive, and episodic. We developed a deployable sequence model to predict daily, per-animal DMI using walk-on scale weights, metered water intake, animal metadata, and publicly available weather. A 3-layer long short-term memory (LSTM) network (hidden size = 256, dropout = 0.20) ingested 7-day windows of engineered covariates, including rolling statistics, first- and second-order differences, short lags, thermal ranges, and Temperature-Humidity Index (THI), scaled with a RobustScaler; categorical variables were encoded with learned embeddings. The prediction head was a compact multilayer perceptron, and outputs were inverse Box-Cox transformed to kilograms per day. Training used animal-identity-grouped splits, Huber loss, Adam optimization with gradient clipping, learning-rate scheduling, early stopping, and Optuna hyperparameter tuning. A single feature-scaler bundle was evaluated against National Academies of Sciences, Engineering, and Medicine (NASEM) equations and three strictly held-out external validations: regional drylot (Morgantown, WV), non-regional drylot (Hyplains, KS), and grazing (2023 Wardensville, WV). Using identical evaluation masks, the LSTM improved accuracy relative to NASEM (pooled RMSE 1.329 vs. 1.858 kg/d; R2 0.655 vs. 0.326), with the largest gains in grazing (LSTM RMSE 1.180, R2 0.507; NASEM RMSE 3.883, R2 -4.337). Iterative fine-tuning with a frozen encoder showed stable behavior under bias-only head updates across production systems. These results demonstrate that accurate, generalizable DMI prediction can be achieved from pragmatic inputs, enabling scalable evaluation of intake phenotypes and feed-efficiency traits directly in grazing systems.

PMID:42114127 | DOI:10.1093/jas/skag133

Genetics. 2026 May 11:iyag119. doi: 10.1093/genetics/iyag119. Online ahead of print.

ABSTRACT

In genetic association studies, permutation tests serve as a cornerstone to estimate p-values. This is because researchers may design new test statistics without a known closed-form distribution, or the assumption of a well-established test may not hold. However, permutation tests require a vast number of permutations which is proportional to the magnitude of the actual p-values. When it comes to genome-wide association studies where multiple-test corrections are routinely conducted, the actual p-values are extremely small, requiring a daunting number of permutations that may be beyond the available computational resources. Existing models that reduce the required number of permutations all assume a specific format of the test statistic to exploit its specific statistical properties. We propose Kernel-smoothed permutation which is a model-free method universally applicable to any statistic. Our tool forms the null distribution of test statistics using a kurtosis-driven transformation, followed by a kernel-based density estimation (KDE). We compared our Kernel-smoothed permutation to Naïve permutation using statistics from known closed-form null distributions. Based on three frequently used test statistics in association studies, i.e., t-test, sequence kernel association test (SKAT), and chi-squared test, we demonstrated that our model reduced the required number of permutations by a magnitude with similar or higher accuracy. Based on a real-world genome-wide association study (GWAS) analysis, we used Crohn’s disease cohort to further confirm that our model substantially outperforms the Naïve permutation.

PMID:42114112 | DOI:10.1093/genetics/iyag119

J Bras Nefrol. 2026 Jul-Sep;48(3):e20250199. doi: 10.1590/2175-8239-JBN-2025-0199en.

ABSTRACT

BACKGROUND: Chronic kidney disease (CKD) in dialysis patients compromises musculoskeletal health and reduces physical activity levels. The Dialysis Patient-Perceived Exercise Benefits and Barriers Scale (DPEBBS) was specifically developed to assess dialysis patients’ perceptions of exercise. This study aimed to translate, cross-culturally adapt, and evaluate the psychometric properties of the Brazilian version of the DPEBBS (EPAD).

METHODS: A cross-sectional study was conducted following the COSMIN guidelines. Psychometric properties assessed included reliability, internal consistency, and construct validity. Participants were recruited from the hemodialysis department of Unifesp. A total of 112 adults on dialysis completed the DPEBBS, the Short Form Health Survey-36 (SF-36), and underwent anthropometric evaluation. The interval between test-retest was one week. Descriptive and inferential analyses were performed to test validity and reliability.

RESULTS: The scale demonstrated high test-retest stability, with consistent mean scores across assessments. Internal consistency was strong, and reliability was supported by a low minimal detectable change and a high intraclass correlation coefficient. Convergent validity with the SF-36 Physical Functioning domain was weak but statistically significant (r = -0.326; p = 0.001), and the correlation with the General Health domain was weak and not statistically significant (r = -0.185; p = 0.052). Consistency analysis showed α = 0.885, ICC = 0.794, SEM = 4.96%, and demonstrated the absence of floor and ceiling effects.

CONCLUSION: The EPAD showed robust validity and reliability for dialysis patients. Despite adequate reliability and validity, this study has limitations, including a single-center sample. The EPAD may support individualized exercise counseling and rehabilitation planning in hemodialysis units.

PMID:42114104 | DOI:10.1590/2175-8239-JBN-2025-0199en

Ann Intern Med. 2026 May 12. doi: 10.7326/ANNALS-25-04761. Online ahead of print.

ABSTRACT

Human health and natural systems are intrinsically linked-stable natural systems enable healthy human life. Health systems aim to promote, restore, and maintain health. Health systems may promote human health while having detrimental effects on natural systems, contributing to the transgression of planetary boundaries, such as biosphere integrity, climate change, and the introduction of new entities like microplastics. To date, the health guideline field lacks methods to assess the impacts of health interventions on planetary boundaries. The GRADE (Grading of Recommendations Assessment, Development and Evaluation) Working Group established the Planetary Health Project Group in 2023 to develop formal GRADE guidance for integrating planetary health into guideline recommendations to address this gap.

Guided by the concepts of planetary health and planetary boundaries and following established methods for GRADE guidance development, the project group conducted iterative case study analyses, expert workshops, and a 2-round global Delphi consensus process. Four case studies were selected for application of this guidance before recommendations were finalized. The GRADE Working Group approved the official guidance.

The Planetary Health Project Group presents 7 domains of guidance for incorporating planetary health aspects into the guideline development process, including highly desirable items and optional items. Highly desirable items include formally addressing planetary health in public health and health system guidelines and explicitly justifying its exclusion where it is not addressed. Judgments within the evidence-to-decision (EtD) framework should systematically integrate included evidence across the prioritized planetary boundaries and equity. This guidance aims to support guideline developers and policymakers in making evidence-based, trustworthy recommendations to protect individual and planetary health, while maintaining thoroughness and feasibility for guideline developers within the GRADE approach.

PMID:42114098 | DOI:10.7326/ANNALS-25-04761

Bioinformatics. 2026 May 11:btag289. doi: 10.1093/bioinformatics/btag289. Online ahead of print.

ABSTRACT

MOTIVATION: Phylogenetic networks represent complex biological scenarios that are overlooked in trees, such as hybridization and horizontal gene transfer. Although numerous methods have been developed for phylogenetic network inference, their scalability is severely limited by the computational demands of likelihood optimization and the vastness of network space. Composite (or pseudo-) likelihood approaches like SNaQ have improved computational tractability for network inference, but they remain inadequate for datasets of sizes routinely handled by tree inference methods.

RESULTS: Here, we introduce SNaQ.jl, a new standalone Julia package with the composite likelihood inference originally implemented within PhyloNetworks.jl as well as new scalability features that enhance computational efficiency through (1) parallelization of quartet likelihood calculations during composite likelihood computation, (2) weighted random selection of quartets, and (3) probabilistic decision-making during network search. Through a simulation study and empirical data analysis, we show that this new version of SNaQ.jl (version 1.1) improves average runtimes by up to 499% on average with no change in function parameters or method accuracy.

AVAILABILITY AND IMPLEMENTATION: SNaQ.jl is a new open source Julia package available at https://github.com/JuliaPhylo/SNaQ.jl.

PMID:42114082 | DOI:10.1093/bioinformatics/btag289

Neurol Clin Pract. 2026 Jun;16(3):e200612. doi: 10.1212/CPJ.0000000000200612. Epub 2026 Apr 6.

ABSTRACT

OBJECTIVES: The aim of this study was to evaluate whether the treatment effects of tolebrutinib on confirmed disability worsening (CDW) diverge from its effects on relapse prevention compared with other disease-modifying therapies (DMTs) for relapsing multiple sclerosis (MS).

METHODS: We extracted published effect estimates for annualized relapse rate (ARR) and CDW confirmed at 24 weeks from all phase 3 trials with teriflunomide as the active comparator: ASCLEPIOS (ofatumumab), OPTIMUM (ponesimod), ULTIMATE (ublituximab), EVOLUTION (evobrutinib), and GEMINI (tolebrutinib). When duplicate trials were available, pooled estimates were derived. Log-transformed estimates were used in a weighted linear regression of CDW vs ARR, with bubble size reflecting statistical precision. Tolebrutinib was excluded from the regression fit but displayed for comparison.

RESULTS: Across 4 DMTs other than tolebrutinib, a strong linear association was observed between treatment effects on ARR and CDW (R² = 0.997), indicating that disability benefit was generally proportional to relapse reduction. By contrast, tolebrutinib deviated from this relationship, with a hazard ratio for CDW of 0.71 (95% CI 0.53-0.95) despite a relapse rate ratio of 1.03 (95% CI 0.85-1.25).

DISCUSSION: Tolebrutinib was the only therapy to show a benefit on CDW without a measurable effect on relapses, highlighting a dissociation between disability worsening and relapse suppression not observed with other DMTs.

PMID:42114075 | DOI:10.1212/CPJ.0000000000200612