Category: Statistics

JMIR Med Educ. 2026 Jul 10;12:e70400. doi: 10.2196/70400.

ABSTRACT

BACKGROUND: The COVID-19 pandemic has placed a significant burden on health professionals (HPs). They face higher infection risks due to the nature of their work environment and patient care responsibilities. Their ability to access and apply reliable COVID-19 information affects their own preventive behavior and that of those around them. In this context, health literacy (HL) has become increasingly important. Despite extensive research, information to foster COVID-19-related HL in HPs remains limited.

OBJECTIVE: This systematic review aimed to identify, appraise, and synthesize intervention studies on the effectiveness of COVID-19-related HL interventions in HPs.

METHODS: Five electronic databases (eg, PubMed (MEDLINE), Embase), six clinical trials registries (eg, ISRCTN registry), one preprint server (MEDRXIV), published conference proceedings, and five gray literature databases (eg, opengrey.eu, ProQuest) were searched in May 2022 and updated in August 2025. Reference lists of included studies were screened manually. Two reviewers independently screened titles, abstracts, and full-texts according to eligibility criteria and extracted data; disagreements were resolved by discussion or consultation with a third reviewer. We included randomized controlled trials (RCTs), nonrandomized studies of interventions, and uncontrolled before-and-after studies evaluating the effectiveness of any COVID-19-related HL intervention. Primary outcomes include COVID-19-related HL, its four facets (access, understand, appraise, and apply COVID-19 information), and indicators (eg, COVID-19-related knowledge), assessed at postintervention and follow-up. When studies were sufficiently similar, random-effects meta-analyses were performed; otherwise, a narrative synthesis was provided. Risk of bias was assessed using validated tools based on study design, and the overall certainty of the evidence was evaluated by the GRADE (Grading of Recommendations, Assessment, Development, and Evaluation) approach.

RESULTS: We included 15 RCTs (2034 participants), 4 nonrandomized studies of interventions (291 participants), 74 uncontrolled before-and-after studies (327,298 participants), 5 ongoing studies, and 1 study with awaiting classification. Interventions targeted a broad range of health occupational groups. Intervention type, delivery mode, methods, settings, and comparator varied widely. No outcome measure explicitly referred to an HL model. Most studies aimed to enhance COVID-19-related knowledge and skills, and had a high risk of bias. COVID-19-related interventions may increase knowledge of vaccines (standardized mean difference 1.00; 95% CI 0.33 to 1.67, I2=24%), and the infection prevention control skills, such as donning and doffing of personal protective equipment (standardized mean difference 1.95; 95% CI 1.82 to 3.09, I2=46%), but the evidence remains very uncertain.

CONCLUSIONS: COVID-19-related HL interventions may promote HP’s short-term competencies in infection control. However, the evidence remains uncertain, primarily due to the low quality of studies, characterized by a high risk of bias. Interventions specifically designed to enhance the full COVID-19 HL operationalized by its four facets are lacking. High-quality RCTs with sufficient statistical power, grounded in HL theoretical principles, are needed to achieve precise understanding.

PMID:42430764 | DOI:10.2196/70400

Plast Reconstr Surg. 2026 Jul 9. doi: 10.1097/PRS.0000000000013317. Online ahead of print.

ABSTRACT

BACKGROUND: Triple nerve transfer (TNT) combines cross-face nerve grafts, masseteric-to-facial nerve transfer, and hypoglossal-to-facial nerve transfer to restore facial function in long-standing facial paralysis. However, its efficacy in patients with severe soft tissue laxity or atrophy may be limited. This study evaluates whether adding autologous fascia lata suspension (TNTF) to TNT improves functional and psychosocial outcomes.

METHODS: A retrospective study compared 22 patients with chronic unilateral facial paralysis, treated with either TNT alone (n=14) or TNT plus fascia lata suspension (TNTF, n=8). All patients completed the Facial Clinimetric Evaluation (FaCE) Scale pre- and postoperatively (≥12 months). Statistical analyses included Wilcoxon signed-rank tests, Mann-Whitney U tests, ANCOVA, and repeated-measures ANOVA to adjust for baseline differences.

RESULTS: Both groups showed significant improvements in FaCE scores postoperatively (TNT: +18, p<0.001; TNTF: +24, p<0.01). The TNTF group had worse preoperative scores (median 25 vs. 37, p=0.006), but postoperatively, scores were comparable (49.1 vs. 55.1, p=0.095). The mean improvement was greater in TNTF (Cohen’s d ≈ 0.82, p=0.09), with significantly greater gains in oral function and social appearance (e.g., FaCE item Q15: +3 vs. +1, p=0.02). Correlation between EMG severity and baseline FaCE scores was significant (ρ ≈ 0.57, p<0.01).

CONCLUSION: TNT restores facial function, but adding fascia lata suspension improves static tone and symmetry in patients with severe soft tissue laxity, offering superior functional and psychosocial outcomes in complex cases.

PMID:42430759 | DOI:10.1097/PRS.0000000000013317

Kidney360. 2026 Jul 10. doi: 10.34067/KID.0000001265. Online ahead of print.

ABSTRACT

BACKGROUND: Clinical trials in rare glomerular disease may establish short-term treatment efficacy but are limited by small sample sizes and short study duration. Observational data are needed to assess longer-term outcomes and can provide insights into real-world prescribing practices. This study applied modern statistical methods to real-world data from the Cure Glomerulonephropathy (CureGN) network to compare the effectiveness of rituximab and calcineurin inhibitors on long-term primary membranous nephropathy outcomes.

METHODS: CureGN participants with biopsy-confirmed primary membranous nephropathy who initiated either of the two treatments at least 6 months after any previous immunosuppressant exposure were eligible. Inverse-probability-of-treatment weighting balanced covariates at treatment initiation. Inverse-probability-of-censoring weights accounted for censoring individuals if another immunosuppressant was started during follow-up. Outcomes included time from treatment initiation to composite kidney disease progression (40% decline in eGFR, kidney replacement therapy, or eGFR <15), proteinuria remission, and relapse following remission. Hazard ratios and differences in restricted mean survival times were estimated.

RESULTS: 325 treatment initiations across 250 unique participants were eligible, with median follow-up 53 months (25th-75th percentile: 24-79). Participants on calcineurin inhibitors had significantly higher risks of disease progression (HR=2.81; 95% CI: 1.16, 6.80). Hazard ratios for proteinuria remission (HR=0.77; 95% CI: 0.50, 1.17) and relapse (HR=1.42, 95% CI: 0.69, 2.92) had wide confidence intervals.

CONCLUSIONS: Rituximab was associated with better kidney function preservation than calcineurin inhibitors over long follow-up. Proteinuria remission and relapse results favored rituximab but did not reach statistical significance. Long-term treatment comparative effectiveness in rare diseases can be evaluated with real-world data.

PMID:42430748 | DOI:10.34067/KID.0000001265

Dermatol Surg. 2026 Jul 7. doi: 10.1097/DSS.0000000000005260. Online ahead of print.

ABSTRACT

BACKGROUND: Procedures have traditionally been delayed 6 to 12 months after isotretinoin because of concerns about abnormal scarring; evidence for fractional ablative CO2 laser is limited in higher Fitzpatrick skin types.

OBJECTIVE: To assess the safety and effectiveness of ablative fractional CO2 laser (AFCL) for atrophic acne scars during and after oral isotretinoin.

METHODS: This was a single-center retrospective cohort of 106 patients (188 AFCL sessions) grouped as concurrent isotretinoin, ≤90 days since discontinuation, 91 to 180 days, or >180 days (control). Outcomes included abnormal scarring, delayed re-epithelialization (>14 days), postinflammatory hyperpigmentation (PIH) at 1 and 3 months, and scar-score change. Overall group comparisons and adjusted patient-level and session-level models were performed.

RESULTS: Abnormal scarring occurred in 3/106 patients (2.8%) with no significant between-group difference (p = .762). Delayed re-epithelialization >14 days occurred in 11/106 patients (10.4%) and was numerically highest in the concurrent isotretinoin group (6/31 [19.4%]); however, the overall four-group comparison was not significant (p = .243), and concurrent isotretinoin was not statistically significant in an adjusted session-level generalized estimating equation model (OR: 4.79, 95% confidence interval: 0.79-29.13; p = .089). PIH at 3 months was higher in the ≤90-day group versus controls (adjusted OR: 6.03, 95% confidence interval: 1.55-23.52; p = .010). Concurrent isotretinoin was not associated with increased PIH or abnormal scarring. Higher density predicted delayed re-epithelialization.

CONCLUSION: AFCL during low-dose isotretinoin or within 6 months after isotretinoin discontinuation was not associated with increased abnormal scarring. Because delayed re-epithelialization was numerically more frequent during concurrent isotretinoin and density was associated with slower healing, conservative density selection and careful counseling remain appropriate. PIH risk was elevated when treatment occurred within 90 days after discontinuation.

PMID:42430738 | DOI:10.1097/DSS.0000000000005260

J Med Internet Res. 2026 Jul 10;28:e101388. doi: 10.2196/101388.

ABSTRACT

BACKGROUND: Parenting interventions can improve parental and child outcomes across diverse settings. However, less is known about how family size, including the number of children, shapes baseline conditions, and how intervention effects unfold over time. Most studies also focus on average treatment effects, with limited attention to heterogeneity across family contexts and trajectories of change.

OBJECTIVE: This study examined whether the number of children was associated with baseline differences in parental and child outcomes, moderated immediate postintervention effects, and shaped postintervention trajectories over 6- and 12-month follow-up periods.

METHODS: We conducted secondary analysis of a pragmatic cluster randomized controlled trial evaluating a universal digital-human parenting intervention delivered through the preschool system in China (N=541). Families were categorized by the number of children (1, 2, 3, or more). We examined (1) baseline differences in parental and child outcomes, (2) moderation of intervention effectiveness at immediate postintervention, and (3) trajectories of change over 6- and 12-month follow-up periods using mixed-effects models.

RESULTS: Of the 541 enrolled families, 494 were included in the complete-case baseline analysis. Compared with 1-child families, 2-child families, and families with 3 or more children, reported lower levels of baseline early learning and stimulation and proactive parenting, as well as greater endorsement of corporal punishment and higher parenting stress. We found no statistically significant evidence that the number of children moderated immediate postintervention effects. In intervention-group trajectory analyses, 2-child families showed greater improvement in early learning and stimulation at the 6-month follow-up (b=3.966, 95% CI 1.468-6.463). Families with 3 or more children showed a similar pattern (b=5.749, 95% CI 0.536-10.962), although estimates for this subgroup were less precise because of the small sample size. This subgroup also showed larger but more variable reductions in selected child behavioral outcomes over follow-up.

CONCLUSIONS: Family size might not always be associated with short-term intervention effectiveness but was associated with divergence in longer-term trajectories. These findings suggest that caregiving demands are relevant for the sustainability of intervention effects. By integrating baseline differences, short-term effects, and longitudinal trajectories within a single framework, this study highlights the importance of moving beyond average treatment effects to more dynamic, context-sensitive evaluations. Designing parenting interventions, particularly scalable digital-human programs, that incorporate sustained and context-responsive support may be critical for addressing variation in family structure and enhancing long-term effectiveness.

PMID:42430727 | DOI:10.2196/101388

JMIR Res Protoc. 2026 Jul 10;15:e94010. doi: 10.2196/94010.

ABSTRACT

BACKGROUND: Kidney transplant recipients present reduced physical function and a high prevalence of cardiometabolic complications, which increase cardiovascular risk and compromise long-term graft outcomes. Resistance training has demonstrated beneficial effects in this population; however, previous interventions have shown heterogeneity in load prescription and have not incorporated objective monitoring of movement velocity. Velocity-based resistance training (VBT) allows precise regulation of exercise intensity and fatigue, potentially improving the safety and individualization of exercise prescription in clinical populations.

OBJECTIVE: This study aims to evaluate the effects of a 12-week VBT program on renal function and metabolic health in kidney transplant recipients and to compare 2 different load-control strategies based on movement velocity.

METHODS: This pilot randomized controlled trial will include adult kidney transplant recipients with stable graft function. Participants will be randomly assigned (1:1) to either a maximal velocity group, in which sets will be terminated at a 20% velocity loss threshold, or a constant submaximal velocity group, in which participants will perform repetitions at 50% of the participant’s individual maximal velocity. Both groups will complete 3 supervised training sessions per week for 12 weeks with real-time velocity monitoring. Primary outcomes will include renal and metabolic health domains assessed through venous blood analysis. Serum creatinine will be the prespecified hierarchical primary renal end point, and high-density lipoprotein cholesterol will be the prespecified hierarchical primary metabolic end point. Estimated glomerular filtration rate will be calculated using the Chronic Kidney Disease Epidemiology Collaboration equation. Secondary outcomes will include blood pressure, body composition, muscular strength, metabolic syndrome criteria, and the force-velocity profile. Data will be analyzed using analysis of covariance and linear mixed-effects models following a predefined hierarchical inferential strategy.

RESULTS: The study was initiated in September 2025. Participant recruitment and the intervention phase have been completed. All 14 participants completed the 12-week training program, and no participants were lost to follow-up. Preintervention and postintervention data collection has been completed according to the study protocol. The study database has been cleaned and locked, and statistical analyses are currently underway. Publication of the primary study results is anticipated in late 2026.

CONCLUSIONS: This study introduces the implementation of VBT in kidney transplant recipients. The findings are expected to provide evidence on the feasibility and potential benefits of this approach and may support the integration of exercise professionals into multidisciplinary transplant care teams.

TRIAL REGISTRATION: ClinicalTrials.gov NCT07370727; https://clinicaltrials.gov/study/NCT07370727.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/94010.

PMID:42430723 | DOI:10.2196/94010

JMIR Med Inform. 2026 Jul 10;14:e91249. doi: 10.2196/91249.

ABSTRACT

BACKGROUND: Psychiatric clinical notes in electronic health records (EHRs) provide rich longitudinal information that can support clinical decision-making. Using historical medical data can enable earlier identification of mental illness, better characterization of disease trajectories, and more personalized treatment planning. Natural language processing (NLP) transforms these unstructured notes into analyzable representations for research and care.

OBJECTIVE: This study aims to systematically summarize NLP methodologies for psychiatric clinical notes, compare major modeling paradigms and application areas, and highlight emerging large language model (LLM) trends, key challenges, and future research directions.

METHODS: Following the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) guidelines, a literature search was conducted for articles on NLP methods based on psychiatric clinical notes published from January 2021 to December 2025 in Ovid MEDLINE, Ovid EMBASE, PubMed, Scopus, Web of Science, the ACM Digital Library, and ScienceDirect. This scoping review analyzed NLP methods applied to psychiatric clinical notes, focusing on major trends, identifying suitable features for traditional machine learning (ML)-based models, applications of pretrained language models (PLMs), and key challenges. Approaches were categorized as rule-based, traditional ML, hybrid, deep learning (DL), and LLM-based methods across information extraction and text classification tasks.

RESULTS: In total, 101 studies were eligible for inclusion. Rule-based methods (n=36) and hybrid approaches (n=34) remained the most widely used techniques, largely favored for their interpretability in handling nuanced, subjective clinical notes. These were followed by DL (n=15), traditional ML (n=10), and LLM-based approaches (n=6). Traditional ML studies relied heavily on engineered features, which could be grouped into 5 broad categories: domain knowledge features, lexical and statistical features, vector-based semantic features, emotion-related features, and temporal features. PLMs improved performance mainly through domain adaptation and task-specific fine-tuning, enhancing the handling of psychiatric language, medical terminology, and clinical note structure. LLM-based studies, although still limited in number, indicated a growing shift toward generative and reasoning-based applications.

CONCLUSIONS: Hybrid NLP approaches remain dominant, combining domain rules with ML for extraction and classification. DL approaches continue to advance, with domain adaptation supporting medical terminology and clinical semantics. LLMs may further automate complex workflows via zero-shot capabilities and reasoning, alongside growing interest in temporal modeling and multimodal integration. Key future needs include improved generalizability across institutions, privacy protection, and careful attention to ethical implications in clinical deployment.

PMID:42430721 | DOI:10.2196/91249

J Cachexia Sarcopenia Muscle. 2026 Aug;17(4):e70331. doi: 10.1002/jcsm.70331.

ABSTRACT

BACKGROUND: Body composition is a major determinant of health, yet the role of fat-free mass, a key component of body composition, in mortality remains unclear.

METHODS: A PRISMA-guided systematic review and meta-analysis (PROSPERO: 321722) of observational cohort studies in community-dwelling adults was conducted. PubMed was searched from inception to 25 October 2025, and Web of Science and Embase were searched from inception to 31 July 2025. Eligible studies assessed fat-free mass using computed tomography, dual-energy X-ray absorptiometry (DXA), bioelectrical impedance analysis (BIA) or anthropometry. Studies involving hospitalized participants were excluded. Maximally adjusted effect estimates were pooled using random-effects models to calculate summary risk ratios (RRs) and 95% confidence intervals (CIs). Publication bias was assessed using graphical and statistical methods; influence analyses evaluated robustness; E-values quantified potential unmeasured confounding; and meta-regression explored between-study heterogeneity.

RESULTS: Of 7741 screened records, 49 studies met the inclusion criteria (1 149 807 participants; 83 798 deaths). Low versus high fat-free mass was associated with higher all-cause mortality (RR: 1.42, 95% CI: 1.30-1.55). Trim-and-fill analysis indicated publication bias (adjusted RR: 1.31, 95% CI: 1.20-1.44). Sensitivity analyses confirmed robustness (leave-one-out RR range: 1.39-1.43; E-value: 2.19). Associations were consistent across age (p_{difference; within-study} = 0.133), geographic region (p_difference = 0.983) and cause of death (p_{difference for cardiovascular diseases; within-study} = 0.240, p_{difference for cancer; within-study} = 0.136). Effect sizes varied by sex (men RR: 1.56, 95% CI: 1.32-1.85; women RR: 1.25, 95% CI: 1.11-1.39; p_difference < 0.0001) and measurement method: strongest for calf circumference (RR: 2.19, 95% CI: 1.50-3.20), moderate for DXA (RR: 1.52, 95% CI: 1.29-1.79) and weakest for BIA (RR: 1.23, 95% CI: 1.07-1.41).

CONCLUSIONS: Low fat-free mass is associated with a 42% higher risk of all-cause mortality in community-dwelling adults. Routinely assessing fat-free mass provides clinical value in identifying high-risk individuals and informing preventive care strategies.

PMID:42430205 | DOI:10.1002/jcsm.70331

J Med Internet Res. 2026 Jul 10;28:e85290. doi: 10.2196/85290.

ABSTRACT

BACKGROUND: Online health communities (OHCs) are important channels for families of children with autism spectrum disorder to obtain health information and psychosocial support. Differences between an open forum platform and the physician-patient consultation platforms may shape caregiver decisions, yet comparative evidence from China remains limited. A large language model (LLM) provides a scalable approach for systematic content annotation in large OHC datasets.

OBJECTIVE: This study proposes and validates a standardized LLM-assisted annotation framework under a unified classification schema and compares topic distributions and poster identities across an open forum platform (Baidu Tieba) and physician-patient consultation platforms (Chunyu Doctor and Haodf).

METHODS: We implemented an LLM-assisted annotation framework. A unified taxonomy of topics and poster identities was first developed through human open coding. Poster identities in Baidu Tieba were annotated through a double-blind manual procedure. For topic classification, interannotator and human-LLM agreement were evaluated on a manually labeled subset to benchmark models of varying sizes. The best-performing LLM was selected for full-dataset topic annotation, followed by statistical and cross-platform analysis.

RESULTS: When metrics were arithmetically averaged across all annotation tasks, the best-performing LLM achieved agreement levels comparable to human annotation (accuracy=79.18%, SD 0.20%; κ=0.736, SD 0.003; F₁-score=0.727, SD 0.006), approaching interannotator agreement (accuracy=81.65%; κ=0.767; F₁-score=0.758), demonstrating strong stability and scalability. Full-dataset analysis yielded 3 main findings. First, model performance increased with parameter scale but plateaued beyond 14B, indicating diminishing marginal returns from further scaling. Second, clear cross-platform differences in poster identity were observed: the open forum platform was dominated by family members of patients (caregivers: 2377/3516, 67.61%), with substantial participation from commercial rehabilitation practitioners (commercial posters: 427/3516, 12.14%), resulting in a more heterogeneous participation structure. Third, topic distributions reflected both shared high demand for resource-related information and differentiated help-seeking pathways: both platform types demonstrated consistently high demand for resource recommendation and evaluation; the open forum platform was primarily characterized by diagnosis-related discussions (1183/7535, 15.70%), whereas the physician-patient consultation platforms were centered on intervention-related consultation (2864/7687, 37.26%).

CONCLUSIONS: The LLM-assisted annotation framework proposed in this study enables reliable large-scale annotation of OHC data while maintaining high human-LLM agreement and operational stability. Midsized models (eg, 14B) demonstrated favorable cost-performance efficiency. The findings reveal 2 key aspects: the open forum platform exhibits a complex participation structure, and the influence of commercially affiliated actors should not be overlooked; users on both platform types show sustained demand for resource-related information but follow different help-seeking pathways, emphasizing diagnostic exploration and professional intervention, respectively. These results suggest that platform structure and governance mechanisms may shape caregivers’ information access and decision-making. The framework provides a transparent, reproducible, and cost-effective approach for OHC research. All data were deidentified and handled in accordance with relevant platform policies and ethical standards.

PMID:42430199 | DOI:10.2196/85290

JMIR Form Res. 2026 Jul 10;10:e84089. doi: 10.2196/84089.

ABSTRACT

BACKGROUND: Understanding how medical students learn is critical for improving teaching strategies in clinical education. Despite the widespread use of learning style frameworks, such as visual, aural, read/write, and kinesthetic (VARK), evidence from sub-Saharan Africa remains limited, and the use of learning style approaches is debated in the literature. In clinical and health sciences education, aligning teaching with learners’ preferences can enhance knowledge retention, procedural competence, and ultimately the quality of patient care.

OBJECTIVE: This study aimed to determine the predominant learning style preferences of clinical students at a Nigerian medical school and to examine how demographic and academic factors influence these preferences, with explicit attention to implications for clinical pedagogy.

METHODS: A cross-sectional survey was conducted among 200 clinical students (400-600 level) at Niger Delta University between October 2021 and December 2021, using the validated VARK inventory (version 7.8). Descriptive statistics summarized distributions, and the Pearson chi-square tests or Fisher exact tests assessed bivariate associations with sex, age group, and year of study. A multivariable modeling strategy was prespecified but not performed due to the categorical structure of the primary outcomes, sparse cells for some modality categories, and the sample size limitations for multinomial modeling.

RESULTS: Of 200 participants (mean age 25.1, SD 3.9 y; n=107, 53.5% male), 105 (52.5%) preferred unimodal learning, and 95 (47.5%) preferred multimodal learning. Kinesthetic (n=121, 60.5%) and auditory (n=110, 55%) were the most common dominant preferences, followed by read/write (n=68, 34%) and visual (n=36, 18%). Visual preference was significantly higher among male participants (χ21=4.49; P=.03). Read/write preference varied by year of study (χ22=8.29; P=.02). No significant associations were found with age. The pedagogical implications for clinical teaching were discussed, including bedside instruction, skills laboratory, simulation, small-group teaching, and audio-visual learning resources.

CONCLUSIONS: Clinical students in this Nigerian setting predominantly favored kinesthetic and auditory learning, with nearly half reporting multimodal preferences. Medical educators should adopt blended instructional designs that include hands-on, discussion-based, and audio-visual elements to better prepare students for clinical practice. These insights can inform faculty development, curriculum design, and national medical education policies to foster adaptive, learner-centered training that improves clinical competency and readiness for professional service.

PMID:42430124 | DOI:10.2196/84089