Category: Statistics

BMC Med Educ. 2026 Jun 27. doi: 10.1186/s12909-026-09801-8. Online ahead of print.

ABSTRACT

BACKGROUND: Thoracentesis is an essential clinical procedure, but its teaching is often limited by patient safety concerns and insufficient opportunities for repeated practice. Virtual reality (VR) offers immersive, repeatable simulation-based training; however, its effectiveness for thoracentesis has not been rigorously evaluated.

METHODS: In this randomized controlled trial, 20 medical interns were randomly assigned to either a VR-based training group (n = 10) or a traditional training control group (n = 10). The VR group received theoretical instruction plus VR simulation practice (1 h/day, 4 days/week for 3 weeks), while the control group received the same theoretical instruction plus traditional mannequin-based practice. All participants completed a 4-week training program. Outcomes were assessed at baseline, week 2, week 3, and week 4 using a standardized 300-point scoring rubric (preoperative, intraoperative, and postoperative components). Statistical comparisons were made using t-tests and chi-square tests. The primary outcome was the final thoracentesis procedural score at week 4.

RESULTS: No adverse events occurred. Baseline characteristics and initial assessment scores did not differ significantly between groups. At week 2, the VR group scored significantly lower than the control group (p < 0.05), reflecting an initial learning curve. However, at week 3 and week 4, the VR group significantly outperformed the control group (p < 0.01). At the final assessment, 80% (8/10) of the VR group achieved scores ≥ 270 points (excellent), compared to only 10% (1/10) of the control group.

CONCLUSION: VR-based training may improve thoracentesis procedural skills after an initial adaptation period. VR appears to be a useful adjunct to traditional medical education, though these findings are preliminary due to the small sample size.

PMID:42374433 | DOI:10.1186/s12909-026-09801-8

BMC Health Serv Res. 2026 Jun 29. doi: 10.1186/s12913-026-14988-z. Online ahead of print.

ABSTRACT

BACKGROUND: Poor medication adherence remains a significant public health issue. Although pictograms are established as intuitive visual aids for medication guidance, a current evidence synthesis is lacking, particularly regarding their role in the rapidly advancing digital health landscape.

OBJECTIVE: To systematically evaluate the potential effect of pictograms on patients’ medication adherence and exploring implications for the digital health era.

METHODS: We searched PubMed, Web of Science, Embase, Cochrane Library, CNKI, Wanfang Data Knowledge Service Platform, VIP Network, and CBM databases for relevant studies on the effect of pictograms on patients’ medication adherence, from the inception of each database to April 27, 2025.

RESULTS: A total of 21 articles were included, comprising 19 English and 2 Chinese publications, all of which were randomized controlled trials. These studies varied in terms of study background, sample size, and drug regimens tested. All studies had methodological limitations, with pictographic interventions differing in complexity, duration, and adherence outcomes measured. Twelve studies (57.1%) reported statistically significant effects of pictographic interventions on patient adherence.

CONCLUSION: Pictograms serve as an effective adjunct to enhance medication adherence, particularly among patients managing complex medication regimens or possessing limited health literacy. Future efforts should focus on developing standardized, evidence-based pictogram libraries and integrating them into multimodal medication education systems. Incorporating pictograms into digital health platforms, such as those using AI and smart devices, represents a promising direction for enhancing their reach and impact.

PMID:42374421 | DOI:10.1186/s12913-026-14988-z

Cardiovasc Diabetol. 2026 Jun 29. doi: 10.1186/s12933-026-03277-y. Online ahead of print.

ABSTRACT

BACKGROUND: Cardiovascular-kidney-metabolic (CKM) syndrome is recognized as a progressive pathophysiological continuum linking metabolic dysfunction, dysfunctional adiposity, chronic kidney disease, and cardiovascular injury. The atherogenic index of plasma (AIP) reflects lipid-related atherogenic burden, whereas novel adiposity indices, including body roundness index (BRI), weight-adjusted waist index (WWI), and a body shape index (ABSI), capture body-shape-related adiposity burden. However, the associations of AIP and AIP-based adiposity composite indices with mortality outcomes across the CKM spectrum remain unclear. This study aimed to evaluate the associations of AIP and integrative AIP-based composite indices, including AIP-BRI, AIP-WWI, and AIP-ABSI, with all-cause and cardiovascular mortality among individuals across CKM stages.

METHODS: We conducted a retrospective cohort analysis using prospectively collected data from 22,587 US adults in the National Health and Nutrition Examination Survey (NHANES) 1999-2018. Following the 2023 American Heart Association (AHA) Presidential Advisory, participants were classified into a hierarchical staging framework (Stages 0-4) to reflect the CKM disease continuum. Primary outcomes were all-cause and cardiovascular mortality, identified through linkage to the National Death Index. Integrative AIP-based composite indices were constructed by directly multiplying AIP by each adiposity index, including BRI, WWI, and ABSI, yielding AIP-BRI, AIP-WWI, and AIP-ABSI, respectively. These indices were evaluated as integrated exposure variables reflecting the combined burden of atherogenic dyslipidemia and adiposity-related body shape. All analyses incorporated complex survey weights to ensure national representativeness. Survey-weighted Cox proportional hazards models were used to calculate hazard ratios (HRs) and 95% confidence intervals (CIs). Model 2 adjusted for demographic and socioeconomic characteristics, lifestyle factors, blood pressure, and total cholesterol. Nonlinear associations were examined using restricted cubic splines (RCS). Incremental prognostic value was assessed using net reclassification improvement (NRI) and integrated discrimination improvement (IDI).

RESULTS: During follow-up, non-survivors exhibited significantly higher baseline integrative composites than survivors (P < 0.001). In the fully adjusted model (Model 2), which adjusted for demographic and socioeconomic characteristics, lifestyle factors, blood pressure, and total cholesterol, each standard deviation increase in the integrative composites was independently associated with a higher risk of all-cause mortality: AIP-BRI (HR, 1.09; 95% CI, 1.05-1.14), AIP-WWI (HR, 1.07; 95% CI, 1.02-1.11), and AIP-ABSI (HR, 1.07; 95% CI, 1.02-1.11). These point estimates were numerically slightly higher than that of stand-alone AIP (HR, 1.06; 95% CI, 1.01-1.10), but the differences should be interpreted cautiously. For cardiovascular mortality, the corresponding HRs were 1.17 (95% CI, 1.09-1.27) for AIP-BRI, 1.11 (95% CI, 1.03-1.21) for AIP-WWI, and 1.11 (95% CI, 1.02-1.20) for AIP-ABSI, compared with 1.10 (95% CI, 1.02-1.20) for AIP. RCS analyses revealed J-shaped associations, with a clinical risk threshold for AIP-BRI at 1.57. Subgroup analyses indicated that these associations were most evident in participants aged < 50 years. Adding BRI to AIP assessment yielded a statistically significant but modest NRI of 3.10% for cardiovascular mortality.

CONCLUSION: Across the CKM spectrum, higher AIP-based adiposity composite indices, particularly AIP-BRI, were associated with increased risks of all-cause and cardiovascular mortality, especially among younger individuals. However, their incremental predictive improvement was modest, suggesting that these indices may serve as supplementary exploratory markers rather than stand-alone tools for CKM risk stratification.

PMID:42374419 | DOI:10.1186/s12933-026-03277-y

BMC Med Educ. 2026 Jun 29. doi: 10.1186/s12909-026-09797-1. Online ahead of print.

ABSTRACT

BACKGROUND: Medical specialty choice shapes both physician identity and healthcare systems. Specialty choices are influenced not only by academic interest and aptitude but also by gender norms and socioeconomic pressures. While these dynamics are well documented in high income settings, fewer studies examine how perceived gender roles and income expectations jointly influence specialty preferences in LMICs contexts.

METHODS: This sequential mixed methods study combined a cross-sectional online survey with semi structured interviews. The quantitative phase included 155 medical students (response rate 79.5%) and examined associations between demographic characteristics and specialty preferences. The qualitative phase comprised 34 interviews purposively sampled across all five cohorts. Interview data were analyzed thematically using Braun and Clarke’s six step framework, following an inductive and reflexive approach guided by Social Cognitive Career Theory to explore how students understood and negotiated gender norms, income potential, and workload intensity.

RESULTS: Quantitative analysis showed no statistically significant associations between income-based specialty preference and gender, GPA, family income, hometown, or high school background (all p > 0.05). High income specialties were selected by 60 students (39%), low-income specialties by 26 (17%), and other fields by 69 (44%), with similar gender distribution across these groups (p = 0.81). In contrast, qualitative analysis revealed pronounced gendered patterns. Many female students chose specialties they perceived as low intensity, often citing anticipated caregiving responsibilities and work life balance concerns as reasons for avoiding surgery, emergency medicine, and other high intensity fields. Male students frequently reported social and familial pressure to pursue prestigious or high-income specialties, prioritizing financial stability and provider roles even when these conflicted with personal interests. Across both genders, themes highlighted the centrality of perceived income, prestige, mentorship inequities, and cultural narratives of sacrifice in shaping career compromises.

CONCLUSION: Specialty choice among medical students in LMICs reflects not only personal interest but also structural inequities linked to gender norms and income expectations. Although quantitative analysis of demographic factors did not independently predict specialty preferences, qualitative findings revealed powerful social pressures that shape career compromises. Targeted, gender sensitive career counselling, equitable mentorship, and financial support are needed to enable students to choose specialties that align with their values and competencies.

PMID:42374415 | DOI:10.1186/s12909-026-09797-1

BMC Med Inform Decis Mak. 2026 Jun 27. doi: 10.1186/s12911-026-03652-5. Online ahead of print.

ABSTRACT

BACKGROUND: The substantial heterogeneity among septic patients with new‑onset atrial fibrillation (NOAF) complicates intensive care unit (ICU) management, and the absence of a parsimonious subphenotype classifier has impeded the implementation of personalized therapeutic strategies. In this study, we aim to identify the subphenotype of septic patients with NOAF, develop a concise classifier model, and reveal the differential efficacy of heart-rate control therapies across these subphenotypes.

METHODS: This retrospective study utilized data from the Medical Information Mart in Intensive Care (MIMIC)-IV and MIMIC-III database. Consensus clustering based on hierarchical clustering was employed for subphenotype derivation. Nine supervised classifiers were employed to construct the subphenotype classifier, including random forest (RF), extreme gradient boosting (XGBoost), support vector machine (SVM), partial least squares (PLS), neural network (NN), naïve Bayes (NB), linear discriminant analysis (LDA), least absolute shrinkage and selection operator (LASSO), and adaptive boosting (AdaBoost.M1). Subphenotype-specific propensity score-derived stabilized inverse probability of treatment weights was used to weight comparisons of intensive care unit and hospital lengths of stay and discharge dispositions across subphenotypes.

RESULTS: Among 1535 septic patients with NOAF, three distinct subphenotypes emerged: Phenotype A was the mildest subgroup; Phenotype B exhibited pronounced metabolic acidosis, highest anion gap, severe renal impairment, and elevated severity scores; Phenotype C showed intermediate laboratory values and score profiles with notable hyperchloremia and inflammatory markers. A parsimonious model established by SVM showed the best effectiveness. HR-control therapy was associated with subphenotype-specific reductions in hospital and ICU lengths of stay and more favorable discharge dispositions in Phenotype A, while more favorable survival outcomes were observed across all three subphenotypes in IPTW-weighted analyses.

CONCLUSION: Cluster analysis revealed three clinically relevant subphenotypes among septic patients with NOAF, each demonstrating distinct clinical outcomes and heterogeneous responses to heart rate control therapy.

PMID:42374411 | DOI:10.1186/s12911-026-03652-5

BMC Oral Health. 2026 Jun 27. doi: 10.1186/s12903-026-09059-y. Online ahead of print.

ABSTRACT

BACKGROUND: Oral health findings in pediatric chronic kidney disease (CKD) are heterogeneous, and CKD-related metabolic alterations may contribute to plaque mineralization and oral hygiene outcomes. Yet, the relationship between routinely measured serum biochemical parameters and standardized caries/oral hygiene indices in children with CKD is not well defined. This study aimed to evaluate the associations between serum biochemical parameters and oral health indices in pediatric CKD.

METHODS: This cross-sectional study included 34 children aged 6-14 years with diagnosed CKD. Demographic and available clinical data were recorded, and routinely measured serum biochemical parameters reflecting renal function, mineral metabolism, acid-base status, hematologic status, and nutritional/metabolic profile were obtained from routine laboratory records. Oral examination included caries assessment using Decayed, Missing, and Filled Teeth (dmft/DMFT) index for primary/permanent dentition and International Caries Detection and Assessment System (ICDAS) II indices, as well as oral hygiene evaluation using the Simplified Oral Hygiene Index (OHI-S), Debris Index (DI), and Calculus Index (CI). Developmental enamel defects were evaluated, and soft tissue lesions were examined. Pearson correlation analysis and Mann-Whitney U tests were performed. Statistical significance was set at p < 0.05.

RESULTS: Most serum biochemical parameters were not significantly associated with caries experience or oral hygiene indices (p > 0.05). Serum magnesium showed weak-to-moderate negative correlations with CI (r = – 0.383, p = 0.026) and OHI-S (r = – 0.362, p = 0.035), indicating lower calculus accumulation and lower OHI-S scores with higher magnesium levels. In contrast, ALP demonstrated a weak-to-moderate positive correlation with CI (r = 0.386, p = 0.024).

CONCLUSIONS: In this pediatric CKD cohort, routine serum biochemical parameters showed limited cross-sectional associations with caries experience and oral hygiene indices. However, the observed associations of serum magnesium and alkaline phosphatase with calculus accumulation suggest that mineral metabolism-related markers may be linked to plaque mineralization. These findings should be interpreted cautiously in view of the cross-sectional design, small sample size, and clinical heterogeneity of the study group.

PMID:42374410 | DOI:10.1186/s12903-026-09059-y

BMC Med Inform Decis Mak. 2026 Jun 29. doi: 10.1186/s12911-026-03662-3. Online ahead of print.

ABSTRACT

BACKGROUND: Large language models (LLMs), a form of artificial intelligence, are increasingly being utilized in healthcare to support patient education and information delivery. The aim of this study was to perform a comparative analysis of five different LLMs (i.e., ChatGPT-4o, Claude 3.7 Sonnet, Gemini 2.5 Pro Preview, DeepSeek-V3, and Microsoft Copilot) in terms of accuracy, completeness, and readability, based on their responses to frequently asked questions in preoperative patient education for mitral valve surgery (MVS).

METHODS: A standardized questionnaire comprising seven frequently asked questions by patients prior to MVS was developed. Prompting procedures and model parameters were fully reported to support reproducibility. These questions were presented to each LLM in an identical manner. The responses were evaluated by two academic experts in cardiac surgery using structured assessment criteria across three main dimensions: accuracy, completeness, and readability. For the readability analysis, the Simplified Measure of Gobbledygook (SMOG) Index and the Flesch-Kincaid Grade Level (FKGL) scale were utilized.

RESULTS: The ChatGPT-4o and Gemini 2.5 Pro Preview models received statistically significantly higher scores than Claude 3.7 Sonnet and Microsoft Copilot for both accuracy (median 5 for ChatGPT-4o and Gemini 2.5 Pro Preview vs. 4 for Claude 3.7 Sonnet and Microsoft Copilot, p < 0.001) and completeness (median 5 for Gemini 2.5 Pro Preview vs. 3 for Claude 3.7 Sonnet, p < 0.001). Claude 3.7 Sonnet achieved the highest readability scores, with significantly lower SMOG (10.90 for Claude 3.7 Sonnet vs. 12.24 for ChatGPT-4o, p = 0.006) and FKGL (8.0 for Claude 3.7 Sonnet vs. 9.04 for ChatGPT-4o, p = 0.004) scores, indicating simpler and more comprehensible sentence structures. Significant differences were observed among the evaluated models across all three assessment dimensions (p < 0.001 for all comparisons).

CONCLUSIONS: The LLMs represent valuable supplementary tools in patient education processes. However, their implementation in clinical practice must be carefully evaluated, particularly with regard to accuracy and completeness. This study highlights the potential applicability of ChatGPT-4o and Claude 3.7 Sonnet models for preoperative patient education in MVS, while emphasizing that all LLMs should be used under the supervision and guidance of healthcare professionals. For LLMs to be reliably utilized in the medical field, improvement in medical accuracy and standardization are essential.

PMID:42374405 | DOI:10.1186/s12911-026-03662-3

BMC Vet Res. 2026 Jun 27. doi: 10.1186/s12917-026-05653-y. Online ahead of print.

ABSTRACT

The efficacy and safety of pharmaceutical products are critically dependent on maintaining appropriate storage temperatures throughout the supply chain, from manufacturer to end-user. In veterinary medicine, this includes storage within clinics and during ambulatory services. The Federal Capital Territory (FCT) of Nigeria experiences high ambient temperatures, posing a significant risk to drug stability. This study aimed to investigate drug storage temperatures in veterinary clinics that also dispense and retail veterinary pharmaceuticals within the FCT and to assess their compliance with manufacturers’ recommended storage conditions. A cross-sectional study was conducted across 23 veterinary clinics in four Area Councils of the FCT (AMAC, Bwari, Gwagwalada, Kuje). A structured questionnaire was used to gather data on storage infrastructure, including refrigerator use, alternative power supply, and temperature monitoring devices. The temperature of drug storage areas was recorded over two weeks (March 1st-15th, 2025) using an HTC-2 thermometer (Guangdong, China). Data were analyzed using descriptive statistics. While all 23 (100%) clinics possessed a refrigerator, only 60% had an alternative power supply. Critically, 60.9% (n = 14) of clinics lacked any form of temperature monitoring device in their storage areas. The mean ambient storage temperatures in all clinics exceeded the recommended maximum of 30 °C for pharmaceuticals stored in cabinets, with some clinics recording temperatures above 36 °C. During ambulatory services, 87% of Clinicians used drug storage boxes, but 30% parked their vehicles in areas without shade, potentially exposing drugs to high temperatures during transport. This study reveals a significant gap in compliance with manufacturer drug storage temperature recommendations among veterinary clinics in the FCT. The lack of temperature monitoring and exposure to excessive ambient temperatures may compromise drug quality and pose a risk to patient safety and treatment efficacy.

PMID:42374404 | DOI:10.1186/s12917-026-05653-y

BMC Med. 2026 Jun 29. doi: 10.1186/s12916-026-05022-4. Online ahead of print.

ABSTRACT

BACKGROUND: Non-randomized studies of interventions (NRSIs) provide important evidence on harms, especially for rare adverse events that randomized controlled trials (RCTs) are often underpowered to detect. Evidence synthesis is therefore needed to integrate findings across study designs and to inform a comprehensive assessment of harms. However, synthesizing evidence from RCTs and NRSIs remains methodologically challenging. We examined how evidence from RCTs and NRSIs is synthesized in practice and how conclusions were drawn when findings conflict.

METHODS: The meta-epidemiological study included systematic reviews indexed in PubMed between 1 January 2017 and 31 December 2024 that synthesized evidence from both RCTs and NRSIs for the same outcome. We evaluated methodological practices across four synthesis scenarios. For reviews that combined RCTs and NRSIs in a meta-analysis, we assessed key methodological components of the review process. For reviews that meta-analyzed RCTs and NRSIs separately, we assessed qualitative agreement between RCTs and NRSIs based on the magnitude, direction, and statistical significance of the estimates. When qualitative disagreement was observed, we further evaluated whether the review conclusions were reasonable, taking into account the certainty of evidence and the heterogeneity of the estimates.

RESULTS: Of 42,341 records screened, 195 systematic reviews were included. 49 (25.1%) conducted only qualitative syntheses of both RCTs and NRSIs. 11 (5.6%) meta-analyzed only RCTs, with NRSIs synthesized qualitatively; and 7 (3.6%) meta-analyzed only NRSIs, with RCTs synthesized qualitatively. Among the 91 reviews (46.7%) that combined RCTs and NRSIs in a single meta-analysis, important methodological gaps were identified: 72.5% included NRSIs at moderate or high risk of bias, 49.5% used unadjusted estimates, and 53.8% did not conduct subgroup analyses by study design. Separate meta-analyses for RCTs and NRSIs were conducted in 37 reviews (19.0%), of which 67.6% showed qualitative disagreement between the two study designs, and 20.0% were judged to have inappropriate conclusions according to our assessment criteria.

CONCLUSIONS: Systematic reviews synthesizing RCTs and NRSIs for harms frequently overlook essential methodological considerations and often draw conclusions without adequately addressing conflicting findings across study designs. These practices risk compromising the credibility of harm assessments used in clinical, regulatory, and policy decision-making.

PMID:42374394 | DOI:10.1186/s12916-026-05022-4

BMC Oral Health. 2026 Jun 29. doi: 10.1186/s12903-026-08789-3. Online ahead of print.

ABSTRACT

BACKGROUND: The cross-modal correlations between radiomics and pathomics, as well as their clinical translational applications in oral tumors and cysts, remain unclear. Here, we proposed a novel radiomic feature selection strategy guided by radio-pathomic correlations in Ameloblastoma (AM) and Odontogenic Keratocyst (OKC) to enhance their preoperative differentiation.

METHODS: We automatically extracted radiomic and pathomic features from CBCT scans and multi-resolution (25, 50, 100, 200 μm) whole slide image-derived cell density maps, respectively. Subsequently, radio-pathomic associations were evaluated by correlation analysis at two levels, directly between features, and between latent factors derived from features via factor analysis, separately in AM and OKC. Additionally, we compared the diagnostic performance of machine learning models using our proposed pathomics-guided feature selection strategy against traditional selection approaches.

RESULTS: At the feature level, correlation analysis identified one and seven significant feature pairs in AM and OKC, respectively (all |ρ| > 0.50, q < 0.05), suggesting that radiomic morphological feature weres strongly correlated with pathomic textural features reflecting tissue complexity. At the factor level, one significant factor pair was revealed in AM (ρ = 0.50, q = 0.007) and another in OKC (ρ = – 0.41, q = 0.04). Additionally, classification performance was enhanced by our proposed strategy across all six models, with an average area under the receiver operating characteristic curve (AUROC) improvement of 0.036 and individual gains ranging from 0.016 in Logistic Regression to 0.063 in Lasso.

CONCLUSION: Significant cross-modal correlations between radiomic and pathomic features were identified in AM and OKC. Leveraging these associations, our proposed pathomics-guided radiomics showed the potential to improve the accuracy of preoperative differentiation between the two lesions, although these improvements reached statistical significance only in some models and still require further validation before clinical translation.

PMID:42374392 | DOI:10.1186/s12903-026-08789-3