Search Results (23,207)

Introduction: Sudden cardiac death in children and adolescents is a devastating yet potentially preventable event. There is, therefore, an urgent need for early recognition of young patients at elevated cardiac risk. Electrocardiographic screening may play a central role in this effort. However, interpreting paediatric ECGs requires advanced diagnostic skills. Checklists have been advocated to mitigate errors in a number of complex fields, both medical and non-medical; however, their effectiveness in interpreting paediatric ECGs remains uncertain. Objective: To evaluate whether the use of a structured checklist improves primary care paediatricians’ performance in interpreting paediatric ECGs, and to assess the influence of professional experience on diagnostic accuracy. Methods: We conducted a prospective, parallel-group study involving primary care paediatricians in the Balearic Islands (Spain). Participants were randomly assigned to interpret paediatric ECGs, either using routine unstructured interpretation or with checklist support. Outcomes included diagnostic validity ratios, and appropriateness of referral to paediatric cardiologists. Performance was analysed according to checklist use or years of clinical experience. It was also analysed whether diagnostic accuracy varied according to whether the ECG was classified as normal, or abnormal with or without an increased risk of sudden cardiac death. Results: Thirty-one paediatricians completed the study, generating 310 ECG interpretations. Checklist use did not significantly improve sensitivity, specificity, or likelihood ratios for detecting ECG abnormalities associated with sudden cardiac death risk, nor did it increase appropriate referral rates. We report a trend towards higher specificity and likelihood ratios among paediatricians with fewer than 20 years of professional experience than among their more senior counterparts. Significantly, this study found true positive rates of normal ECGs, and of abnormal ECGs with risk of sudden cardiac death, to be around 90%, as opposed to true positive rates of around 60% for abnormal ECGs without risk of sudden cardiac death (91% vs. 85% vs. 59%, respectively). Conclusions: In this study, checklist support did not enhance diagnostic performance in paediatric ECG interpretation. This finding highlights the need for targeted efforts to improve diagnostic accuracy in this sensitive subset. We also report that mid-career paediatricians appear to achieve the highest ECG diagnostic accuracy, and we show a particularly high rate of correct interpretation of both simple ECGs and high-risk abnormal ECGs. Full article

Background/Objectives: Subarachnoid hemorrhage (SAH) is frequently complicated by cerebral vasospasm (VS). Clazosentan reduces VS, and fasudil hydrochloride hydrate (Fasudil) is widely used for VS prevention in Japan. However, the benefit of adding Fasudil to clazosentan remains unclear. We investigated whether adjunctive Fasudil was associated with VS-related and functional outcomes in clazosentan-treated aneurysmal SAH. Methods: We retrospectively analyzed data from the multicenter “Database of Cohort Study for Outcome of SAH in Japan,” collected from 2020 to 2024. Patients with aneurysmal SAH who underwent surgical clipping or endovascular coiling within 4 days of onset and completed 14 days of clazosentan treatment were included. Outcomes were compared between clazosentan plus Fasudil and clazosentan alone. Multivariable logistic regression assessed factors associated with angiographic vasospasm (AVS), cerebral infarction, and poor functional outcome (modified Rankin Scale 3–6) at discharge and 6 months. Results: Among 341 patients, 100 (29.3%) received adjunctive Fasudil and 241 (70.7%) clazosentan alone. AVS occurred in 58/329 (17.6%), cerebral infarction in 58/327 (17.7%), poor functional outcome at discharge in 142/341 (41.6%), and poor functional outcome at 6 months in 82/325 (25.2%). Adjunctive Fasudil was independently associated with higher odds of AVS (adjusted odds ratio [aOR] 2.41, 95% confidence interval [CI] 1.21–4.81), cerebral infarction (aOR 2.10, 95% CI 1.08–4.09), and poor 6-month outcome (aOR 2.88, 95% CI 1.08–7.71), but not with poor functional outcome at discharge (aOR 1.37, 95% CI 0.62–3.03). Conclusions: In clazosentan-treated aneurysmal SAH, adjunctive Fasudil use was not associated with additional benefit and was associated with higher odds of AVS, cerebral infarction, and poor functional outcome at 6 months. Full article

Background: Emergency departments (EDs) operate under severe time pressure, diagnostic uncertainty, and resource constraints, making the consistent application of clinical protocols challenging. This scoping review aimed to map the barriers to protocol adherence and the strategies reported to support implementation in emergency care. Methods: PubMed/MEDLINE, Scopus, and Web of Science were searched for publications from January 2000 to June 2026. Following predefined eligibility criteria, 58 publications were included and charted according to clinician-related, guideline-related, patient-related, and organizational determinants, and we reported the implementation strategies. No formal design-specific risk-of-bias or certainty-of-evidence assessment was performed; therefore, the synthesis was intended to map the available evidence rather than establish the comparative effectiveness. Results: Commonly reported barriers included limited guideline knowledge and clinical experience, cognitive overload and occupational fatigue, poor guideline usability and workflow compatibility, patient communication difficulties and clinical complexity, overcrowding, staffing shortages, and limited organizational support. The reported strategies included education and simulation, audit and feedback, clinical decision support, workflow redesign, multidisciplinary collaboration, and leadership engagement. However, the heterogeneity in study designs, clinical settings, definitions of adherence, and reported outcomes precluded ranking these strategies or determining whether particular combinations were superior. Conclusions: Protocol adherence in EDs appears to be shaped by interacting clinician-related, guideline-related, patient-related, and organizational factors. The identified strategies may support implementation, but their relative and comparative effectiveness remains uncertain. Digital health and artificial intelligence should be considered priorities for prospective evaluation rather than established solutions. Full article

Background and Objectives: Digitalization studies often emphasize perceived benefits while overlooking professionals who cannot determine whether digital tools are useful. This study quantified digital hesitancy among anesthesia and intensive-care professionals, examined its relationship with professional role and experience, and compared it with reported barriers and technology use. Methods: A single-centre cross-sectional survey included 161 professionals (80.1% of 201 eligible departmental staff): 55 nurses and 106 physicians. A four-item Uncertainty Index measured inability to appraise digital benefits in patient safety, clinical decisions, management, and education. Barrier Burden and Resource-Efficiency indices were also calculated. Group differences were assessed using nonparametric tests, and predictors of high uncertainty were examined by multivariable logistic regression. Results: Nurses had substantially higher uncertainty than physicians (mean Uncertainty Index 1.60 vs. 0.49, difference 1.11, 95% CI 0.80 to 1.42; p < 0.001), particularly regarding educational and clinical benefits. Reported barriers were similar across professional groups, with technical problems being the most common. Any reported prior use of training simulators, irrespective of frequency or recency, was markedly lower among nurses than physicians (1.8% vs. 47.2%; absolute gap 45.4 percentage points, 95% CI 35.2 to 55.5; p < 0.001). After adjustment, physician status remained strongly associated with lower odds of high uncertainty, whereas overall seniority was not significant. In an exploratory within-nurse analysis, uncertainty was higher among more experienced nurses, reaching its highest level among those with more than 20 years of practice (Spearman ρ = 0.49, 95% CI 0.24 to 0.68; p < 0.001); the seniority bands were small (n = 7 to 18), so this gradient is hypothesis-generating. Conclusions: Digital hesitancy was concentrated among nurses and was distinct from shared technical barriers; within nurses it was higher at greater seniority. Because patient-safety outcomes and intervention effects were not measured, targeted simulator-based training and role-specific feedback should be regarded as hypotheses for prospective evaluation alongside, rather than established alternatives to, generic infrastructure improvements. Full article

Background: Recent transcriptomic studies classify small-cell lung cancer (SCLC) into molecular subtypes driven by ASCL1, NEUROD1, POU2F3, and YAP1. This study evaluated an immunohistochemistry (IHC)-based subtyping framework using real-world data and assessed its ability to predict chemoimmunotherapy outcomes. Methods: Proteomic expression profiles were quantified by IHC in 100 patients with SCLC. Clinicopathological trajectories and overall survival (OS) were analyzed using a parsimonious multivariate Cox model designed to mitigate overfitting. Results: Significant subclonal heterogeneity was captured by co-dominant hybrid phenotypes, including SCLC-AN (9%) and SCLC-AP (5%). In the treatment-adjusted, parsimonious multivariate analysis, individual continuous biomarker expressions showed an independent association with ASCL1 percentage (HR = 1.011, p = 0.017). Traditional extensive-stage disease (HR = 3.801, 95% CI: 1.669–8.657, p = 0.001) and synchronous bone metastases (HR = 1.968, 95% CI: 1.074–3.604, p = 0.028) remained the only robust clinical predictors of poor OS. When adjusted for therapeutic interventions, first-line chemoimmunotherapy showed a strong protective numerical trend, reducing mortality risk by 49% (HR = 0.512, p = 0.060). Within the immunotherapy subgroup (n = 19), the NE-low group had a 100% response rate and a numerically longer median overall survival than the NE-high group (49.4 vs. 21.4 months; log-rank p = 0.080). Conclusions: Our pilot evaluation provides a clinically feasible routine IHC framework for characterizing subclonal mosaicism in SCLC. Full article

Background: Rural and resource-constrained settings face major barriers to timely cancer screening, diagnosis, and treatment due to limited specialist availability and centralised service-delivery models. In the Eastern Cape, a largely rural province with a constrained oncology workforce, a decentralised–centralised hybrid model was introduced to improve access to cancer care. Nelson Mandela Academic Hospital serves as the central referral hub within this model. This study evaluates the implementation process and impact of this decentralised cancer care model on service utilisation, access, and continuity of care. Methods: A quantitative quasi-experimental pre–post implementation and quality improvement evaluation was conducted using retrospectively collected routine service utilisation and programme data from April 2023 to February 2025. The study assessed the impact of a decentralised oncology care model on access, service integration, and utilisation outcomes. Data from facility registers and district health information systems were managed using Microsoft Excel and analysed using Stata and IBM SPSS Statistics. Descriptive statistics, correlation analysis, and linear regression were used to compare pre- and post-implementation changes in patient volumes, screening coverage, referral completion, workforce capacity, gender distribution, and service uptake. The intervention decentralised screening, diagnosis, follow-up, and patient navigation services to district and satellite facilities while centralising specialised oncology care at referral centres to improve accessibility, efficiency, and continuity of care. Results: Cancer patient attendance increased substantially over the study period, from 355 patients in April 2023 to a peak of 1039 in April 2024, with a sustained upward trend (B = 18.03, p = 0.005), reflecting an average monthly increase of 18 patients. Female patients accounted for most visits, while male attendance showed a significant increasing trend (B = 8.03, p < 0.001). Service integration improved, with strong positive correlations between new patient registrations, follow-up care, palliative services, and inpatient admissions, indicating an expanding continuum of care. Breast and cervical cancers contributed the highest service burden, while cervical and lung cancers showed significant upward trends. Seasonal variation in attendance was observed, particularly during festive periods. From an implementation perspective, screening coverage for priority cancers increased by 18%, while 732,349 individuals were reached through community awareness initiatives. Access improved substantially, evidenced by a reduction of 56,400 km in cumulative patient travel distance over one year. Workforce capacity was strengthened through the training of 517 healthcare workers, and 1943 patients received structured navigation support. Referral efficiency and continuity of care improved, although persistent bottlenecks were observed in diagnostic and referral pathways. Conclusions: The decentralised–centralised oncology care model demonstrated improved cancer service utilisation, access, and continuity of care in a rural, resource-limited setting. However, increasing patient volumes and interconnected service demands place additional pressure on health system capacity. Sustained investment in workforce development, screening—particularly for cervical cancer—and system efficiency is required. This model provides a scalable and context-appropriate framework for strengthening oncology services in similar low-resource settings. Full article

Background/Objectives: First-line maintenance is established in advanced epithelial ovarian cancer (AEOC), but management remains uncertain for patients without a BRCA mutation or homologous recombination deficiency (HRD)-positive disease. We assessed real-world use of bevacizumab, poly(ADP-ribose) polymerase (PARP) inhibitors and active surveillance in France (FR), Germany (DE), Italy (IT), Spain (ES) and the United Kingdom (UK). Methods: Anonymised physician-reported charts were analysed from a retrospective, non-interventional online survey conducted by AplusA Healthcare Marketing Research. Eligible patients had stage III–IV AEOC, no known BRCA-mutated or HRD-positive disease, no progression after first-line platinum chemotherapy, and started maintenance therapy or active surveillance between November 2024 and March 2026. Treatment allocation was compared by chi-square testing. Multinomial logistic regression assessed whether country remained associated with strategy after adjustment for clinical and disease characteristics. Results: Among 3293 patients, 1295 (39.3%) received bevacizumab, 1087 (33.0%) niraparib, 158 (4.8%) rucaparib, 477 (14.5%) active surveillance and 276 (8.4%) other therapy. Treatment allocation differed by country (χ² = 725.4, df = 16, p < 0.001). Bevacizumab was most frequent in FR and least frequent in the UK; niraparib was most frequent in ES and the UK; active surveillance was most common in the UK and IT. In the adjusted model, country was the strongest factor associated with strategy (likelihood-ratio χ² = 550.1, df = 8, p < 0.001). Conclusions: First-line maintenance practice in AEOC without an actionable BRCA/HRD biomarker varies substantially across European countries. These differences may partly reflect national access, reimbursement, clinical practice and other unmeasured country-level factors. Outcome-based real-world studies are needed to clarify the best strategy. Full article

Background and Objectives: Dual-mobility cups are commonly used to reduce instability after total hip arthroplasty (THA) for femoral neck fracture (FNF), but their advantage over contemporary large-head standard cups remains uncertain. This registry study compared revision-related survival estimates up to 10 years among monoblock dual-mobility cups, modular dual-mobility constructs, and standard cups with femoral heads ≥ 36 mm in cementless THA for FNF. Materials and Methods: We retrospectively analyzed regional residents undergoing cementless, non-metal-on-metal primary THA for FNF. The cohort included 4827 THAs: 1220 monoblock dual-mobility cups, 408 modular dual-mobility constructs, and 3199 standard cups with ≥36 mm heads. Groups differed in age and sex distribution, with modular dual mobility used in older patients and both dual-mobility constructs used more often in women. Kaplan–Meier analysis estimated survival up to 10 years using aseptic failure and revision for dislocation/instability as endpoints. Cox models estimated adjusted hazard ratios, and cumulative incidence of aseptic failure was assessed with death as a competing event. Results: Kaplan–Meier-estimated 10-year aseptic failure-free survival was 94.4% for monoblock dual mobility, 96.8% for modular dual mobility, and 95.4% for standard cups (p = 0.815). Cup construct was not independently associated with aseptic failure: compared with standard cups, the HR was 1.12 for monoblock dual mobility (95% CI, 0.75–1.67) and 1.34 for modular dual mobility (95% CI, 0.71–2.52). Kaplan–Meier-estimated 10-year survival free from revision for dislocation/instability was 99.6%, 99.2%, and 99.3%, respectively (p = 0.620), with no independent effect of cup construct. Male sex increased the risk of both aseptic failure and revision for dislocation/instability. Periprosthetic fracture was the leading cause of aseptic revision in both dual-mobility groups, whereas failures were more evenly distributed in the standard cup group. Death-adjusted cumulative incidence of aseptic failure was similar across constructs (p = 0.919). Conclusions: Neither monoblock nor modular dual mobility improved revision-related outcomes compared with standard cups using femoral heads ≥ 36 mm. Late survival estimates, particularly for modular dual mobility, should be interpreted cautiously because of the limited number of patients remaining at risk. Full article

Background/Objectives: This study aimed to compare the incidence, extent, and time to diagnosis of osteonecrosis of the femoral head (ONFH) between Femoral Neck System (FNS) and multiple cannulated screw (CS) fixation in patients with femoral neck fractures (FNFs) at a minimum follow-up of 24 months, as well as to identify factors associated with ONFH after FNS fixation. Methods: A retrospective cohort study was conducted at a single university hospital involving patients aged ≥18 years with isolated FNFs treated by closed reduction and internal fixation using either FNS or multiple CS. Reduction quality, fracture union, incidence and extent of ONFH (measured by Kerboul angle), and revision surgery due to ONFH or other causes were assessed as outcome measures. Logistic regression analyses were conducted to identify factors associated with ONFH after FNS fixation. Results: Eighty-seven patients were included, comprising 48 in the FNS group and 39 in the CS group. No significant differences were found between the groups in reduction quality, union rate, or time to union (p > 0.05 for all). The incidence of ONFH did not differ significantly between the FNS and CS groups (p = 0.314). However, the FNS group showed a significantly greater extent of ONFH (p = 0.031) and longer mean time to ONFH diagnosis (p = 0.001). No significant differences were observed in revision surgery rates due to ONFH (p = 1.000) or other causes (p = 0.624). Multivariate analysis identified fracture displacement (Garden classification stages III and IV) and poor reduction quality (“Broken S” by Lowell’s criteria) as significant predictors of ONFH after FNS fixation. Conclusions: ONFH following FNS fixation tended to exhibit a larger necrotic area and a delayed radiographic onset compared to CS fixation; however, the overall incidence and revision rates were comparable between implants. These differences may be due to initial fracture displacement and baseline severity, rather than the FNS implant itself. In the FNS group, the mean time to radiographic diagnosis of ONFH was 13.1 months, and initial fracture displacement and reduction quality were identified as risk factors for the development of ONFH following FNS fixation. Therefore, long-term and vigilant follow-up is essential after FNS fixation, particularly in patients with displaced fractures. Achieving optimal reduction quality and anatomical femoral neck alignment remains paramount to mitigating the risk of ONFH after FNS fixation. Full article

Large language model-enabled virtual patients may expand access to clinical skills practice by supporting explicitly defined practice tasks. This program evaluation examined medical students’ awareness, use, and perceptions of the McMaster Virtual SP Tool, a custom generative artificial intelligence tool developed to supplement clinical skills practice. We conducted an anonymous, single-institution cross-sectional survey of students across three cohorts at McMaster University. Quantitative responses were summarized descriptively, and free-text responses were analyzed using qualitative content analysis informed by task-aligned fidelity, deliberate practice, learner-centred feedback, and simulation instructional design. Thirty-five students responded; 23 (65.7%) were aware of the tool and 16 (45.7%) had used it. Among the 16 users, 15/16 (93.8%) found the tool at least somewhat easy to navigate; 13 (81.3%) would use it again; and 13 (81.3%) would recommend it. Across all respondents, 14/35 (40.0%) reported using AI-enabled virtual patients for OSCE preparation. Users valued its accessibility, independent low-stakes rehearsal, and usefulness for focused history-taking, question wording, and clinical reasoning. Perceived limitations included reduced human connection, nonverbal and emotional realism, physical examination practice, and feedback specificity. AI-enabled virtual patients may therefore be considered as adjuncts for selected cognitive and structural rehearsal tasks. Objective educational outcomes were not assessed. Full article

Objective: To evaluate the response of various oxygenation indices to the oxygen challenge test (OCT), prior to decannulation, in neonates supported with VA ECMO for respiratory indications, and to compare these responses between survivors and non-survivors. Design: Single-center retrospective observational study. Setting: Single, tertiary care, 40-bed Pediatric Intensive Care Unit in Miami, Florida. Patients: Neonates cannulated to VA ECMO for respiratory failure from 2012 to 2022 who had an OCT. Patients with congenital heart disease were excluded. Measurements and Main Results: A total of 63 neonates were included, of whom 13 patients died (20.6%). Oxygenation indices compared before and during OCT include PaO2, Delta PaO2, P/F ratios, Alveolar-arterial gradient, arterial/alveolar ratios, and shunt fraction. The Wilcoxon signed-rank test demonstrated increases in all oxygenation indices during OCT compared to pre-OCT, except for P/F ratios. There were no differences in oxygenation indices between survivors and non-survivors during OCT. However, after decannulation, all oxygenation indices were worse in non-survivors (p < 0.05). Linear regression analysis demonstrated that delta FiO2 on the ventilator, P/F before OCT, and ECMO flow during OCT significantly affected PaO2 response during OCT. Shunt fraction above 28.8% after decannulation demonstrated the highest discriminatory ability for mortality (AUC 0.853). Conclusions: The OCT response during ECMO is not a reliable predictor of survival in neonates on VA ECMO, and its value in assessing readiness for ECMO trial-off warrants further evaluation. Further studies investigating the role of oxygenation indices at the time of trial-off for predicting outcomes may be helpful. Full article

Background/Objectives: Hidradenitis Suppurativa (HS) is a chronic inflammatory skin disease associated with a significant impairment in patients’ quality of life (QoL). However, evidence on QoL assessment in HS remains fragmented, with heterogeneous instruments and domains reported. This scoping review aimed to map the assessment of QoL in individuals with HS, identify the instruments used, and describe the domains explored. Methods: This scoping review was conducted in accordance with the Arksey and O’Malley framework, as refined by Levac et al. and the Joanna Briggs Institute. A systematic search of six databases was performed between November 2025 and March 2026 to identify studies reporting QoL or patient-reported outcomes (PROs) in individuals with HS. Data were extracted and analyzed using a narrative synthesis approach. Results: A total of 21 studies met the inclusion criteria. HS was consistently associated with substantial QoL impairment across multiple domains, including physical, psychological, social, sexual, and occupational aspects. Symptom burden was a major determinant of reduced QoL, particularly pain, pruritus, malodor, and discharge. A wide range of instruments was used, with the Dermatology Life Quality Index (DLQI) being the most frequently applied. However, most tools did not adequately capture the multidimensional impact of HS. The use of HS-specific instruments was limited, and most studies adopted cross-sectional designs. Conclusions: QoL in HS is markedly compromised, but current assessment approaches are inconsistent and incomplete. Further research using comprehensive, disease-specific instruments and longitudinal designs is needed to better capture disease burden and support patient-centered care. Full article

Background: Trigeminal neuralgia associated with multiple sclerosis (TN-MS) is a secondary facial pain syndrome whose clinical and therapeutic features differ substantially from those of classical TN. Although the literature comparing treatment outcomes between TN-MS and classical TN is relatively abundant, evidence on the impact of treatment on quality of life in TN-MS is scarce. In patients with MS—already vulnerable owing to the course of their disease and the burden of its treatments—therapeutic success cannot be defined by pain scales alone, because even modest analgesic improvement and, in particular, a reduction in medication burden may represent meaningful gains. This review aims to reappraise the therapeutic goals in TN-MS. Methods: We provide a critical narrative review of the evidence, explicitly distinguishing direct evidence from indirect evidence extrapolated from the MS and classical-TN literature, across medical therapy, percutaneous procedures, stereotactic radiosurgery, and microvascular decompression (MVD). Results: First-line pharmacotherapy remains standard but is largely extrapolated from classical TN, and tolerability is frequently limited by sedative and motor adverse effects in already polymedicated patients. Percutaneous procedures and radiosurgery provide clinically meaningful relief in selected patients, but recurrence is common and patient-centered outcomes are poorly documented. Surgical exploration of the cerebellopontine angle—including MVD, neurolysis, and combined techniques—should not be categorically excluded on the basis of an MS diagnosis in carefully selected patients, although outcomes are generally less favorable than in classical TN and the direct evidence is limited and vulnerable to selection bias. Conclusions: Future studies should assess treatment impact on additional outcomes such as medication burden, quality of life, fatigue, cognition, and functional status. Full article

Background: Celiac disease is a chronic autoimmune enteropathy whose only established treatment is a strict, lifelong gluten-free diet (GFD). Beyond physical symptoms, the disease and its dietary management affect emotional, social, and economic functioning. Objectives: This study aimed to assess health-related quality of life (HRQoL) in adult patients with celiac disease and to identify factors associated with their daily functioning and adherence to the GFD. Methods: A cross-sectional survey was conducted between February and March 2026 among 101 adults with a self-reported diagnosis of celiac disease made by a gastroenterologist, recruited through online support groups. Data were collected using an author-designed sociodemographic questionnaire, the Gastrointestinal Quality of Life Index (GIQLI), and the Celiac Dietary Adherence Test (CDAT). Analyses (Mann–Whitney U and Kruskal–Wallis tests, repeated-measures ANOVA, Spearman correlations with false-discovery-rate correction, and multivariable linear and logistic regression with prespecified sensitivity analyses) were performed in R 4.5.2, with α = 0.05. Results: The most frequently reported difficulties were eating away from home, the high cost of the GFD, and limited product availability; the most common symptoms after inadvertent gluten exposure were abdominal pain, bloating, fatigue, diarrhea, and headache. HRQoL was lowest in the emotional and physical domains. Better dietary adherence was significantly correlated with higher HRQoL across the total score and all domains except the social domain, whose weak internal consistency in this sample (α = 0.57) precludes substantive interpretation (total GIQLI: ρ = −0.50; p < 0.001), and in the multiple regression, adherence remained the strongest independent correlate of HRQoL (β = −0.36; adjusted R² = 0.41). Longer time on the diet was independently associated with better adherence (adjusted OR = 0.55 per duration category; p = 0.017). In a small vocational education subgroup (n = 8), poorer adherence and lower HRQoL were observed relative to participants with higher education, though only the HRQoL difference survived correction for multiple testing (FDR-adjusted p = 0.011 vs. 0.060 for adherence); given the size of this subgroup, this comparison is exploratory and hypothesis-generating. Higher HRQoL was also associated with milder symptoms, better-rated medical care, greater product availability, and higher household income. Conclusions: In this sample, HRQoL scores were lowest in the emotional and physical domains; because the design was cross-sectional and included no comparison group, these data characterize the distribution of HRQoL within the sample and do not establish that celiac disease reduces HRQoL. Dietary adherence was the strongest correlate of patient functioning, alongside symptom burden, self-rated quality of medical care, and product availability. Because the design is cross-sectional, these associations do not establish direction, and whether comprehensive health education and psychological support improve outcomes requires prospective and interventional evaluation. Full article

Background: The ASA Physical Status (ASA-PS) classification and the Charlson Comorbidity Index (CCI) are common pre-operative scoring tools. Language models could automate structured pre-operative scoring, but direct comparisons require paired inference because all models are evaluated on the same patients. Methods: In this retrospective single-center concordance analysis, 101 consecutive adult orthopedic patients were independently rated by two clinicians; the rounded mean for ASA-PS and arithmetic mean for CCI formed a clinician-derived composite reference. The cohort contained no ASA-PS IV-V patients. Six model configurations received identical prompts. Agreement was assessed using quadratic weighted kappa, ICC(2,1), exact and adjacent agreement, MAD, RMSE, and Bland–Altman limits. Post hoc between-model comparisons used 10,000 patient-level paired bootstrap replicates with Benjamini–Hochberg correction. Results: Inter-clinician weighted kappa was 0.713 for ASA-PS and 0.914 for CCI. GPT-5.2 reached kappa 0.884 for ASA-PS and 0.970 for CCI. In paired analyses, GPT-5.2 had significantly higher quadratic weighted kappa than every other tested model for both outcomes and significantly higher ICC for CCI after multiplicity correction. Phi4 and deepseek-r1-70B were not significantly different from inter-clinician agreement for CCI kappa or ICC; equivalence was not tested. Conclusions: Among the six evaluated model configurations, GPT-5.2 achieved significantly higher agreement with the clinician-derived composite reference than the other tested models for both ASA-PS and CCI in post hoc paired analyses with multiplicity correction. Locally deployable phi4 and deepseek-r1-70B showed CCI agreement estimates that were not statistically distinguishable from inter-clinician agreement, although equivalence was not tested. These findings are limited to the evaluated models and study cohort. Full article