Tag: statistics

J Phys Chem Lett. 2026 Jul 17. doi: 10.1021/acs.jpclett.6c01412. Online ahead of print.

ABSTRACT

The quantitative prediction of biomolecular recognition is crucial to molecular science. The challenge is not merely structural determination but the prediction of (thermo)dynamic and kinetic observables arising from high-dimensional molecular ensembles, such as free energies, conformational distributions, and rate processes across different conditions. As the field shifts from structure-centric to ensemble-based descriptions, two complementary modeling strategies have matured: explicit energy-based approaches grounded in statistical mechanics and data-driven models that learn statistical representations of molecular configurations from large data sets. Physics-based methods, including molecular dynamics and free energy perturbation, estimate observables by sampling (Boltzmann-distributed) configurations under approximate molecular Hamiltonians, thereby providing mechanistic interpretability and thermodynamic consistency, albeit at non-negligible computational cost and with inherent force field limitations. In contrast, modern machine learning approaches rapidly generate structures and propose conformational ensembles without explicit thermodynamic weighting, by learning statistical patterns in structural and bioactivity data. While these methods often achieve high predictive performance, they do not inherently enforce thermodynamic consistency due to the lack of an explicit connection to a partition function and thus may produce configurations that are not physically realizable. We argue that, since physics-based simulations and machine learning provide complementary approximations to the underlying probability distribution associated with biomolecular recognition events, and they excel respectively in consistency with free-energy landscapes and state populations and in predictive accuracy, the central challenge for the coming decade will be integrating them into hybrid frameworks that are scalable and transferable.

PMID:42464806 | DOI:10.1021/acs.jpclett.6c01412

JMIR Mhealth Uhealth. 2026 Jul 16;14:e88887. doi: 10.2196/88887.

ABSTRACT

BACKGROUND: Use of effective contraceptive methods (ECMs) reduces maternal mortality. Person-centered counseling increases uptake, but barriers to high-quality counseling persist. Telehealth may improve access to comprehensive contraceptive care, but its effectiveness remains unclear.

OBJECTIVE: This study assesses the effectiveness and acceptability of tailored, interactive telehealth contraceptive counseling (TECC) compared with in-person counseling.

METHODS: We conducted a systematic review and meta-analysis of English-language randomized controlled trials (RCTs) comparing TECC with in-person counseling for women and girls of any age and setting. We searched MEDLINE, Embase, Web of Science, and the Cochrane Library from inception through October 15, 2025. Outcomes were use of ECM less than 6 months from intervention (primary outcome); use of ECM at 6-12 months; use of long-acting reversible contraception (LARC); choice of ECM; choice of LARC; satisfaction with counseling; and method switching. Two researchers assessed risk of bias (RoB) using Cochrane RoB 2, certainty of evidence using GRADE (Grading of Recommendations Assessment, Development, and Evaluation), and performed meta-analysis using a random-effects model. The protocol was registered with PROSPERO (International Prospective Register of Systematic Reviews) a priori (CRD42023404402). No specific funding was received.

RESULTS: Eight RCTs (RoB: low, n=4; moderate, n=2; and high, n=2) and 1 cluster RCT (moderate-to-high RoB) were included in the review, with 5353 participants across all included studies. Eight studies evaluated TECC as an adjunct to in-person care (hybrid model) and 1 as a standalone model, of which 7 and 1, respectively, contributed outcome data for the meta-analyses. Certainty of evidence was low to very low. The pooled effect of 4 studies of TECC on ECM use less than 6 months showed no clear evidence of an effect (relative risk [RR] 1.10; 95% CI 0.95-1.29). The pooled effect of 4 studies of TECC on choice of ECM (RR 1.07; 95% CI 0.96-1.18) likewise showed no clear effect. The pooled effect of 4 studies of TECC on ECM use at 6-12 months showed a small but statistically significant (P=.04) effect in favor of TECC, narrowly excluding the null (RR 1.07; 95% CI 1.002-1.130). Because of high statistical and clinical heterogeneity (I2=87%-96%), results for LARC choice (2 studies) and LARC use at 0-6 months (2 studies) and 6-12 months (2 studies) were narratively described. The evidence for these outcomes was very uncertain. Narrative analysis of satisfaction across 2 studies showed no difference in effect. There were no data on method switching to support the analysis.

CONCLUSIONS: Current evidence suggests, with low certainty, that adjunct TECC, when delivered alongside in-person care, shows little to no effect on contraceptive use compared with in-person care. For use at 12 months, method choice, LARC use, and satisfaction compared with in-person care, the evidence is very uncertain. Future research should prioritize adequately powered evaluations of standalone models and assess how tailoring, timing, and delivery influence effectiveness, including long-term use and method switching.

PMID:42464800 | DOI:10.2196/88887

J Clin Psychopharmacol. 2026 Jul 17. doi: 10.1097/JCP.0000000000002229. Online ahead of print.

ABSTRACT

OBJECTIVE: To evaluate the effects of Withania somnifera extract (WSE), or Ashwagandha, on symptoms of schizophrenia and to investigate decreases in proinflammatory markers as a possible mechanism of action.

METHODS: Schizophrenia subjects with a history of a recent exacerbation were randomly assigned to receive double-blind adjunctive treatment either with WSE or placebo for 12 weeks. Outcomes were assessed using the Positive and Negative Syndrome Scale (PANSS), the Clinical Global Impression Scale (CGI), and the Perceived Stress Scale (PSS). Inflammatory markers were measured at baseline and 12 weeks.

RESULTS: Among 47 randomized subjects, 24 received WSE and 23 received placebo. WSE treatment was not associated with a greater reduction in symptoms on the PANSS, the CGI, or the PSS. We did not find a significant effect of WSE on inflammatory markers and we did not find an association between changes in inflammatory measures and changes in symptoms.

CONCLUSION: This clinical trial failed to find that WSE resulted in a statistically significant improvement of clinical measures or inflammatory markers. Disruptions from the COVID pandemic, the small sample size, heterogeneity, and baseline differences in psychopathology between the groups were important limitations of the study. However, the pattern that we observed did not indicate that the findings would have been different with a larger sample size or that the findings were influenced by the baseline differences.

PMID:42464773 | DOI:10.1097/JCP.0000000000002229

J Endovasc Ther. 2026 Jul 17:15266028261453269. doi: 10.1177/15266028261453269. Online ahead of print.

ABSTRACT

BACKGROUND: Dual-layer micromesh (DLMM) carotid stents were developed to enhance plaque scaffolding and reduce peri-procedural embolization during carotid artery stenting (CAS). Although Roadsaver is widely used in contemporary European practice, comparative evidence against first-generation stents (FGSs) and other DLMM devices remains limited. This study systematically reviews the European experience with Roadsaver and performs an exploratory meta-analysis of 30-day clinical outcomes and in-stent restenosis (ISR).

METHODS: A systematic search of PubMed, Embase, and Medline identified European studies published between January 1, 2014 and December 31, 2024 reporting transfemoral CAS using the Roadsaver stent. Adult patients with extracranial, de novo carotid stenosis-symptomatic or asymptomatic-were included. Comparative analyses were limited to studies reporting paired outcomes for Roadsaver versus FGSs or C-Guard. Primary endpoints were 30-day stroke and death; the secondary endpoint was ≥50% ISR at approximately 12 months, as defined by each study. Pooled effects were calculated using Peto odds ratios with random-effects DerSimonian-Laird models.

RESULTS: Twenty-two studies comprising 3539 Roadsaver procedures were included in the qualitative review; 5 studies met criteria for quantitative pooling. Across all studies, 30-day stroke and death occurred in 1.6% and 0.5% of patients, respectively. Meta-analysis showed no statistically significant differences in 30-day stroke or death between Roadsaver and FGSs, with low heterogeneity. In-stent restenosis comparisons similarly demonstrated no significant differences between Roadsaver and FGSs or between Roadsaver and C-Guard; however, ISR analyses were limited by marked heterogeneity in definitions, imaging modalities, and follow-up timing.

CONCLUSIONS: Roadsaver DLMM stenting in European practice is associated with low 30-day stroke and death rates. Exploratory pooled analyses did not identify significant differences between Roadsaver and FGSs or between DLMM devices; however, the small number of comparative studies, variability in ISR assessment, and predominantly observational evidence base substantially limit the certainty of these findings. Larger, rigorously designed comparative studies, and randomized trials are needed to determine whether clinically meaningful differences exist between carotid stent platforms.Clinical impactIn this paper, transfemoral CAS with the Roadsaver dual-layer micromesh stent was associated with low 30-day stroke and death rates in European practice. Comparative analyses did not demonstrate significant differences versus first-generation stents or between DLMM devices; however, the observed safety profile, combined with mechanistic data on enhanced plaque scaffolding and embolic protection, supports the use of Roadsaver in routine CAS, including complex anatomies. These findings reinforce DLMM technology as a contemporary standard option rather than an experimental alternative, while underscoring the need for standardized ISR assessment and adequately powered randomized trials to guide device selection in carotid revascularization.

PMID:42464753 | DOI:10.1177/15266028261453269

Eur J Psychotraumatol. 2026 Dec;17(1):2692768. doi: 10.1080/20008066.2026.2692768. Epub 2026 Jul 17.

ABSTRACT

Background: Childhood maltreatment (CM) is associated with sleep disturbance, but observational findings may be affected by confounding, shared familial liability, and reverse-direction explanations. We conducted bidirectional two-sample Mendelian randomization (MR) to examine genetic evidence linking composite CM liability with six sleep-related phenotypes.Methods: Genetic instruments for CM were obtained from a large European-ancestry multi-cohort GWAS meta-analysis. CM was defined as a composite phenotype encompassing multiple abuse and neglect subtypes and ascertainment modes. Summary statistics for chronotype, daytime sleepiness (DS), insomnia symptoms, sleep apnea (SA), daytime napping (DN), and sleep duration (SD) were derived from large European-ancestry GWASs. Inverse-variance weighted MR was the primary method, with weighted median and MR-Egger sensitivity analyses. Benjamini-Hochberg correction was applied and reported as P(BH).Results: In forward MR, genetically proxied CM was associated with higher DS (OR = 1.04, 95% CI 1.01-1.07; P(BH) = 0.030), insomnia symptoms (OR = 1.08, 95% CI 1.01-1.15; P(BH) = 0.030), and DN (β = 0.07, 95% CI 0.03-0.09; P(BH) = 0.002). Little evidence was observed for chronotype, SA, or SD. In exploratory reverse-direction MR, genetic liability to DN (β = 0.16, 95% CI 0.05-0.27; P(BH) = 0.012) and insomnia symptoms (β = 0.21, 95% CI 0.07-0.35; P(BH) = 0.012) was associated with CM liability.Conclusions: This study provides suggestive genetic evidence linking composite CM liability with selected sleep-related phenotypes, particularly self-reported DS, insomnia symptoms, and DN. Reverse-direction findings may reflect shared genetic liability, gene-environment correlation, reporting-related mechanisms, or broader familial pathways rather than temporal effects of adult sleep traits on CM. Given modest effects and uneven sensitivity support, findings should be interpreted cautiously.

PMID:42464740 | DOI:10.1080/20008066.2026.2692768

Ren Fail. 2026 Dec;48(1):2700069. doi: 10.1080/0886022X.2026.2700069. Epub 2026 Jul 17.

ABSTRACT

Protein-energy wasting (PEW) is common in incident hemodialysis patients and linked to poor outcomes. The uric acid/HDL-cholesterol ratio (UHR) and intact parathyroid hormone (iPTH) relate to metabolic, inflammatory, and nutritional disturbances, but their value for predicting PEW in incident hemodialysis is unclear. This retrospective multicenter study included 863 incident hemodialysis patients. PEW was defined according to the International Society of Renal Nutrition and Metabolism criteria. UHR and iPTH were evaluated using ROC analysis, multivariable logistic regression, and ten machine learning models. Restricted cubic spline, mediation, and trajectory analyses were performed, with SHAP for interpretability. PEW was identified in 59.2% of patients. Feature selection across four machine learning approaches consistently identified UHR, iPTH, and eGFR as key predictors. Higher baseline and cumulative UHR were independently associated with a lower risk of PEW. Trajectory analysis showed that both rapidly increasing and decreasing UHR patterns with higher mean levels were linked to the lowest PEW risk. Nonlinear associations were observed, with an inverted U-shaped relationship between UHR and PEW and an S-shaped relationship between iPTH and PEW. Mediation analysis indicated that iPTH accounted for approximately 10-13% of the association between UHR and PEW. Among all models, XGBoost achieved the best performance (AUC = 0.801). A web-based Shiny tool was developed for individualized PEW risk assessment. UHR and iPTH were associated with PEW risk within specific ranges. Machine learning models integrating these markers showed favorable predictive performance and may assist individualized risk assessment in clinical practice.

PMID:42464731 | DOI:10.1080/0886022X.2026.2700069

Yi Chuan. 2026 Jul;48(7):701-713. doi: 10.16288/j.yczz.25-201.

ABSTRACT

Suicide is defined as an intentional act of ending one’s own life. Suicide attempt (SA) is a significant risk factor for suicide death. Research on SA has progressed from socio-psychological perspectives to the molecular and genetic levels. While the biological mechanisms underlying genome-wide association studies (GWAS) identified risk loci remain largely unclear. To investigate the potential risk mechanisms, we constructed a systematic analytic pipeline using brain protein quantitative trait locus (pQTL) datasets (Banner, N=152; ROSMAP, N=376), a brain expression quantitative trait locus (eQTL) datasets (N=452), and SA GWAS summary statistics (N_case=35,786, N_control=779,392). We performed proteome-wide association study (PWAS), Mendelian randomization (MR), Bayesian colocalization analysis, transcriptome-wide association study (TWAS), and multi-marker analysis of genomic annotation (MAGMA) to systematically identify and screen for novel genetically supported candidate proteins related to the biological mechanism of SA in the brain. For functional annotation, we used the GeneMANIA to bulid a functional prediction network integrating co-expression, physical interactions, and pathway colocalization to identify core proteins. PWAS identified three brain proteins whose genetically predicted abundance was significantly associated with SA. Among them, GMPPB was prioritized as putative causal protein, supported by MR analysis (false discovery rate, FDR<0.05) and Bayesian colocalization analysis (posterior probability PPH₄≥0.8). Specifically, higher genetically predicted GMPPB protein levels were associated with increased risk of SA. Although our analyses primarily relied on datasets from European-ancestry populations, the shared genetic architecture across populations and the generalizability of genome-wide data analytical approaches suggest that our findings may still provide useful insights into the biological mechanisms underlying SA and help inform the development of intervention strategies and genetic counseling in Chinese populations.

PMID:42464726 | DOI:10.16288/j.yczz.25-201

Int J Urol. 2026 Jul;33(7):e70549. doi: 10.1111/iju.70549.

ABSTRACT

OBJECTIVES: This post-marketing surveillance study evaluated the real-world safety and effectiveness of cabozantinib in Japanese patients with advanced renal cell carcinoma (aRCC).

METHODS: This prospective, observational study enrolled patients with histologically or cytologically confirmed aRCC across 102 sites in Japan. Patients received cabozantinib as monotherapy or in combination with nivolumab according to approved label regimens and were monitored for 26 weeks. The primary outcome was the incidence of adverse drug reactions (ADRs) related to hepatic failure/dysfunction and pancreatitis. Secondary outcomes included grade 3 or higher ADRs, treatment discontinuation due to ADRs, and investigator-assessed objective response rate (ORR) using Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1.

RESULTS: Of 388 enrolled patients, 385 were included in the analysis (median age, 71.0 years; 70.1% men). Patients receiving monotherapy (n = 322) and those receiving combination therapy (n = 63) reported ADRs related to hepatic failure/dysfunction (16.15% and 25.40%), pancreatitis-related ADRs (2.80% and 4.76%), grade 3 or higher ADRs (24.84% and 38.10%), and ADRs leading to treatment discontinuation (13.98% and 30.16%), respectively. The most common grade 3 or higher ADRs were hypertension (3.73%), reduced appetite (3.11%), and hand-foot syndrome (3.11%) in the monotherapy group, and diarrhea (4.76%) in the combination therapy group. The ORR was 32.9% for patients receiving monotherapy and 41.3% for those receiving combination therapy.

CONCLUSIONS: Cabozantinib, as monotherapy or in combination with nivolumab, demonstrated safety and effectiveness in Japanese patients with aRCC in real-world settings, consistent with previous literature. No new safety concerns were identified. These findings support the use of cabozantinib-based regimens in Japanese clinical practice.

TRAIL REGISTRATION: Japan Registry of Clinical Trials (JRCT), registration number: jRCT2031210003.

PMID:42464667 | DOI:10.1111/iju.70549

Catheter Cardiovasc Interv. 2026 Jul 16. doi: 10.1002/ccd.70760. Online ahead of print.

ABSTRACT

BACKGROUND: Neoatherosclerosis is a major mechanism of late and very late stent failure. While plaque rupture (PR) has traditionally been considered the dominant substrate, the prevalence and characteristics of other neoatherosclerotic plaque complications remain insufficiently defined. We aimed to investigate the morphological and clinical differences between neoatherosclerotic PR and plaque erosion (PE) using optical coherence tomography (OCT).

METHODS: This retrospective single-center study included patients with stent failure undergoing OCT-guided percutaneous coronary intervention between January 2020 and December 2025. Patients with neoatherosclerosis-related stent failure were classified according to OCT findings as PR or PE. Clinical presentation, angiographic characteristics, OCT-derived plaque morphology, and long-term outcomes were compared.

RESULTS: Among 103 screened patients, 44 met the inclusion criteria; 28 had PR, 14 had PE and 2 had instent eruptive calcific nodule. PR patients more frequently presented with STEMI and demonstrated significantly higher levels of macrophage infiltration (57.1% vs. 23.1%, p = 0.043), cholesterol crystals (60.7% vs. 28.6%, p = 0.050), lipid arc ≥ 180° (80% vs. 42.9%, p = 0.031), and thin-cap fibroatheroma (35.7% vs. 7.1%, p = 0.048). In contrast, layered plaque was markedly more prevalent in PE (92.9% vs. 48.3%, p = 0.004), and minimal in-stent lumen area was smaller in PE lesions (p = 0.039). Postprocedural thrombus protrusion was significantly more common in PR (81% vs. 27.5%, p = 0.005). Although not statistically significant, all-cause and cardiac mortality were numerically higher in PR patients during follow-up.

CONCLUSIONS: Neoatherosclerotic PR and PE represent distinct mechanisms of stent failure with different inflammatory profiles, plaque morphology, and clinical presentations. PR is associated with higher lipid burden and inflammatory activity, whereas PE is characterized by layered neointima and lower inflammatory features. OCT-based plaque characterization may improve risk stratification and support personalized treatment strategies in late stent failure.

PMID:42464582 | DOI:10.1002/ccd.70760

Diagn Interv Radiol. 2026 Jul 17. doi: 10.4274/dir.2026.263984. Online ahead of print.

ABSTRACT

PURPOSE: To evaluate the real-world multimetric performance of four commercially available computed tomography (CT)-based artificial intelligence (AI) solutions for acute intracranial hemorrhage (AIH).

METHODS: Patients who underwent non-contrast brain CT for suspected AIH in our emergency room between February and March 2024 were screened. After applying the inclusion and exclusion criteria, 436 CT scans were included in the final analysis. Three neuroradiologists established the ground truth for AIH and hemorrhage volume. For detection performance, the area under the receiver operating characteristic curve (AUROC), area under the precision-recall curve (AUPRC), and Brier score were calculated based on the available probability score, whereas sensitivity, specificity, precision, and F1 score were calculated based on binary classification. Bland-Altman analysis was performed to assess volumetric agreement for AIH between each algorithm’s calculations and the neuroradiologists’ measurements.

RESULTS: A total of 436 patients (mean age, 62 years ± 20; male patients, 209) were enrolled. The AUROC (0.96 to 0.99) and sensitivity (0.85 to 0.92) were high across all solutions, with no statistically significant differences in pairwise comparisons (P > 0.05). However, solution B demonstrated the highest AUPRC [0.98, 95% confidence interval (CI): 0.94, 1.00] and the lowest Brier score [0.02 (95% CI: 0.02, 0.03)]. In binary performance, both solutions B and D exhibited significantly higher specificity (1.00 and 0.99), precision (0.90 to 0.98), and F1 score (0.87 to 0.94) than the other solutions (P < 0.05). For volumetric agreement of AIH, solution D showed the lowest mean difference [-0.87 mm³ (95% CI: -1.47, -0.27)] and the narrowest limits of agreement (-13.4 to 11.6) relative to the neuroradiologists’ measurements.

CONCLUSION: In a real-world emergency setting, all four commercially available CT-based AI solutions for AIH demonstrated uniformly excellent performance; however, meaningful differences emerged in confirmatory performance and volumetric agreement. These distinct, algorithm-specific trade-offs provide practical guidance for selecting and integrating appropriate AI solutions to improve AIH diagnosis and management workflows.

CLINICAL SIGNIFICANCE: The algorithm-specific performance trade-offs identified in this study suggest that no single AI solution is universally optimal; solutions with superior confirmatory performance may reduce unnecessary notifications in high-volume emergency settings, whereas those with more consistent volumetric agreement may better support treatment planning and longitudinal monitoring. A structured, multimetric evaluation aligned with institutional priorities is essential for evidence-based AI procurement in acute stroke imaging.

PMID:42464524 | DOI:10.4274/dir.2026.263984