Category: Statistics

Clinics (Sao Paulo). 2026 Apr 28;81:100987. doi: 10.1016/j.clinsp.2026.100987. Online ahead of print.

ABSTRACT

OBJECTIVE: To explore the application value of fecal Syndecan-2 gene Methylation (mSDC2) detection, combined detection of serum Carcinoembryonic Antigen (CEA) and Carbohydrate Antigen 72-4 (CA72-4) in screening of Colorectal Cancer (CRC) and precancerous lesions.

METHODS: A total of 196 participants were enrolled in this case-control study from March to December 2023, including 65 with CRC, 38 with advanced adenomas, 33 with non-advanced adenomas, and 60 controls. The sensitivity, specificity, and Odds Ratios (OR) for serum CEA, CA72-4, and fecal mSDC2 were evaluated.

RESULTS: The sensitivity of fecal mSDC2 for CRC was 86.2% (56/65), with a specificity of 96.7% (58/60), with an OR of 28.9 (95% CI 8.6-97.2, p < 0.001). The sensitivity in advanced adenomas was 34.2% (13/38). Serum CEA had a sensitivity of 56.9% (37/65) and a specificity of 96.7% (58/60) for CRC, with an OR of 5.7 (95% CI 2.0-16.4, p < 0.001). Combined detection of CEA and CA72-4 had a sensitivity of 69.2% (45/65) and a specificity of 81.6% (49/60). The triple combination (mSDC2 + CEA + CA72-4) achieved a sensitivity of 96.9% (63/65) and a specificity of 78.3% (47/60) in the CRC, and a sensitivity of 50% (19/38) for advanced adenomas. The combined detection had higher sensitivity than single detection, with statistically significant differences compared to serum-based detection (p < 0.001).

CONCLUSION: Combining fecal mSDC2 with serum CEA and CA72-4 increased sensitivity for CRC detection in this single-center study, at the cost of reduced specificity. Validation in larger, screening-intended cohorts with predefined thresholds is warranted.

PMID:42056830 | DOI:10.1016/j.clinsp.2026.100987

J Psychiatr Res. 2026 Apr 24;199:113-121. doi: 10.1016/j.jpsychires.2026.04.030. Online ahead of print.

ABSTRACT

OBJECTIVE: Depression is a leading cause of global disability, motivating the development of objective and scalable diagnostic approaches. Quantitative electroencephalography (QEEG) combined with machine learning (ML) and deep learning (DL) techniques has gained increasing attention for depression detection. This systematic review aimed to critically examine and descriptively compare the methodologies, performance metrics, and limitations of ML- and DL-based models applied to EEG data for depression detection.

METHODS: A systematic review was conducted in accordance with PRISMA 2020 guidelines. Seven electronic databases (PubMed, Scopus, IEEE Xplore, ScienceDirect, Web of Science, SAGE Journals, and MDPI) were searched for peer-reviewed studies published between 2020 and 2024. Eligible studies included human participants, used EEG signals for depression detection, and applied ML or DL algorithms. Extracted information comprised algorithm type, sample size, EEG acquisition parameters, validation strategies, and reported performance metrics, which were synthesized descriptively across studies. Risk of bias was assessed using the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool.

RESULTS: A total of 42 studies met the inclusion criteria, including 23 ML-based and 19 DL-based investigations. Reported classification accuracy ranged from approximately 76% to 100%. DL studies showed a higher mean reported accuracy than ML studies (93.92% vs. 90.78%); however, this difference was not statistically significant in the exploratory non-parametric comparison. Near-perfect performance values were frequently observed in studies with small sample sizes and subject-dependent or exclusively internal validation strategies, raising concerns regarding overfitting and limited generalizability. Studies relying on publicly available datasets tended to report more stable performance. QUADAS-2 assessment revealed recurrent risk-of-bias concerns, particularly in the domains of patient selection and index test conduct.

CONCLUSIONS: Both ML and DL approaches demonstrate potential for EEG-based depression detection, but reported performance differences between them should be interpreted cautiously. Although DL studies tended to report higher accuracy values, this pattern was not statistically significant in exploratory analyses and was strongly influenced by sample size, validation strategy, and methodological design. Future research should prioritize larger and more diverse samples, subject-independent or external validation strategies, and standardized reporting frameworks to enhance methodological rigor and clinical applicability.

PMID:42056808 | DOI:10.1016/j.jpsychires.2026.04.030

Asia Pac J Clin Oncol. 2026 Apr 29. doi: 10.1111/ajco.70119. Online ahead of print.

ABSTRACT

AIMS: The prognostic value of clinical complete/near-complete response (cCR/near-cCR) relative to pathologic complete/near-complete response (pCR/near-pCR) after neoadjuvant therapy for locally advanced rectal cancer (LARC) remains incompletely defined.

METHODS: We retrospectively analyzed 180 non-metastatic LARC patients treated with total neoadjuvant therapy (TNT) or standard neoadjuvant chemoradiotherapy (nCRT) followed by total mesorectal excision. Agreement between cCR/near-cCR and pCR/near-pCR was assessed using Cohen’s κ, and associations with survival outcomes were evaluated using Kaplan-Meier and Cox models.

RESULTS: Preoperative cCR/near-cCR was observed in 89 patients (49.4%), whereas pCR/near-pCR occurred in 63 (35.0%), with moderate concordance (κ = 0.53, p < 0.001). Discordance between clinical and pathologic response classification was observed in 42 patients (23.3%). Compared with patients who achieved pCR/near-pCR, those without pCR/near-pCR had inferior 5-year distant metastasis-free survival (DMFS) (66.7% vs. 87.3%; hazard ratio [HR] 2.98, 95% confidence interval [CI] 1.44-6.09; p = 0.003) and locoregional recurrence-free survival (LRFS) (81.0% vs. 95.0%; HR 4.32, 95% CI 1.28-12.23; p = 0.02), while Overall survival (OS) did not differ significantly between groups (63.8% vs. 78.4%; HR 1.42, 95% CI 0.75-3.11; p = 0.06). cCR/near-cCR was associated with improved DMFS on univariable analysis; however, this association was attenuated after multivariable adjustment and did not remain independently significant. Prognostic separation by response was more pronounced in the TNT cohort. Among patients with pCR/near-pCR, baseline biopsy-derived lymphovascular invasion/perineural invasion was associated with numerically poorer outcomes, although these differences were not statistically significant.

CONCLUSION: pCR/near-pCR provides more consistent prognostic discrimination than cCR/near-cCR, and clinically relevant discordance persists in routine practice. Integrating baseline tumor biology with response assessment may refine post-treatment risk stratification, particularly after TNT.

TRIAL REGISTRATION: Not applicable. This study is a retrospective observational study and does not require trial registration.

PMID:42056758 | DOI:10.1111/ajco.70119

Clin Transl Sci. 2026 May;19(5):e70550. doi: 10.1111/cts.70550.

ABSTRACT

A frequently cited concern regarding patient-as-own-control trial designs in rare disease is the potential for placebo and related effects to inflate apparent treatment efficacy. Whether this concern is disqualifying or manageable has not been systematically evaluated. We reviewed meta-analyses quantifying placebo effect magnitude by endpoint type, reporter modality, and trial duration and evaluated statistical methods available for post-trial placebo adjustment in own-control designs. Placebo effects depend heavily on endpoint type. For objective endpoints (enzyme activity, serum biomarkers, imaging volumetrics)-which constitute the majority of primary endpoints in approved rare disease therapies-placebo effects are consistently small and in most meta-analyses statistically indistinguishable from zero (standardized mean difference [SMD] < 0.10). For subjective endpoints (patient-reported pain, caregiver-rated function), effects are larger (SMD 0.20-0.50) but well-characterized and correctable. Placebo effects peak early and decay over weeks, providing a temporal signature distinguishable from sustained pharmacological effects. Multiple complementary analytical methods-including temporal trajectory modeling, objective-subjective concordance analysis, Bayesian informative priors, extended run-in observation designs, and blinded outcome assessment-are available to quantify and manage placebo contributions. Importantly, the randomized controlled trial’s structural advantage in canceling placebo is itself degraded in small samples, where asymmetric allocation of placebo responders can distort between-arm comparisons. The placebo objection to own-control designs is manageable rather than disqualifying. For objective endpoints, correction is minimal. For subjective endpoints, a rich analytical toolkit supports credible decomposition of drug and placebo components. These findings support the broader adoption of own-control designs in rare disease clinical trials.

PMID:42056755 | DOI:10.1111/cts.70550

J Anat. 2026 Apr 29. doi: 10.1111/joa.70163. Online ahead of print.

ABSTRACT

The study was designed to compare femoral and acetabular morphometric parameters obtained via radiography and computed tomography (CT) in German Shepherd Dogs (GSDs), Labrador Retrievers (LABs), and Indian Pariah Dog, with the goals of identifying modality-specific differences, evaluating bilateral symmetry, and informing Orthopaedic implant planning. A total of 72 cadaveric specimens (24 per breed) underwent standardized radiography and CT. Fourteen morphometric parameters, including version angle (VA), femoral neck-shaft angle (FNA), femoral neck diameter (FND), canal flare index (CFI), and acetabular depth (AD), were measured bilaterally. All measurements were performed independently by two trained observers. Inter-observer reliability was assessed between the participating observers. Paired t-tests were used to compare radiographic and CT-derived values, while one-way ANOVA was employed to assess left-right limb symmetry. CT demonstrated greater sensitivity in detecting subtle morphometric variations, particularly in parameters involving cortical transitions and metaphyseal flare. In GSDs, FND (p = 0.004) and CFI (p = 0.028) were significantly higher on CT. LABs showed side-specific differences in right femoral FNA (p = 0.044) and FL (p = 0.020). In Indian Pariah Dogs, CT revealed significant discrepancies in FHO (p = 0.013) and FL (p = 0.020) on the left side, and FND (p = 0.007) and FNL (p = 0.034) on the right. No statistically significant bilateral asymmetry was found in any breed. Inter-observer reliability demonstrated excellent agreement (ICC range: 0.96-1.00; Cohen’s Kappa: 0.90-1.00) between observers. Radiography and CT demonstrated general agreement in most morphometric parameters; however, CT offered superior precision in measuring femoral neck dimensions, flare indices, and subtle angular deviations. Bilateral symmetry was preserved across all breeds, supporting the use of contralateral limbs as surgical references. While radiographs remain effective for routine morphometry and screening, CT is preferred for detailed preoperative planning and implant selection particularly in breeds with complex or variable proximal femoral anatomy. A combined imaging strategy enhances diagnostic accuracy and surgical outcomes in veterinary Orthopaedic practice.

PMID:42056735 | DOI:10.1111/joa.70163

J Child Psychol Psychiatry. 2026 Apr 29. doi: 10.1111/jcpp.70148. Online ahead of print.

ABSTRACT

BACKGROUND: Childhood adversity is widespread globally and is one of the strongest predictors of later psychopathology. However, the differential effects of type and timing of childhood adversities on childhood psychopathology remain unclear, highlighting the need to explore which life-course hypotheses (sensitive periods, accumulation of exposure, and/or recency of exposure) best explain these associations. Of particular importance, there is a lack of research in low- and middle-income countries (LMIC), where children experience higher rates of adversity relative to children in high-income countries (HIC).

METHODS: Participants included 787 children and their mothers from a South African birth cohort, the Drakenstein Child Health Study. Mothers reported child exposure to adversity from birth to 8 years of age across six adversity categories. We used the two-stage Structured Life-Course Modeling Approach (SLCMA) to examine life-course associations between childhood adversity exposures and internalizing/externalizing symptoms measured using the Child Behavior Checklist at age 8 years.

RESULTS: Maternal psychopathology, maternal adverse events, child food insecurity, and child exposure to community/domestic violence had the strongest associations with child psychopathology symptoms, with varying life-course models selected. The accumulation hypothesis best explained associations of maternal adverse events (partial R² = 2.3%) and child exposure to community/domestic violence (partial R² = 1.6%) with internalizing symptoms. The combined middle childhood sensitive period (age > 5-8) and recency hypotheses model best explained associations between maternal psychopathology and internalizing (partial R² = 7.0%) or externalizing (partial R² = 5.1%) symptoms.

CONCLUSIONS: We identified that different types and timing of childhood adversity confer differential risk for childhood psychopathology symptoms in this LMIC sample. Our work has implications for strategically-timed intervention and prevention strategies to improve mental health, which may need to be specifically designed for children in LMIC.

PMID:42056733 | DOI:10.1111/jcpp.70148

Ren Fail. 2026 Dec;48(1):2644784. doi: 10.1080/0886022X.2026.2644784. Epub 2026 Apr 29.

ABSTRACT

BACKGROUND: As the traditional risk factors cannot account for all hypertension (HTN) incidents, it is of great importance to determine other risk factors. This study aimed to identify long-term HTN risk associated with annual change of estimated glomerular filtration rate (eGFR) and prior adverse pregnancy outcomes (APOs) among women participating in a population-based study of Tehran Lipid and Glucose Study (TLGS).

METHODS: This study was performed using prospectively ascertained data of TLGS. A total of 2,404 women with recorded data of eGFR measurements and APO status participated. Data collection was conducted according to the standard guide of TLGS. Cox proportional-hazards regression models were used to estimate the hazard ratios (HRs) and their 95% confidence intervals (CIs) for the incidence of HTN.

RESULTS: A total of 2,404 women were enrolled. Adjusted model shows that, a one z-score positive eGFR change was associated with a reduced risk of developing HTN among women with a history of APOs (HR= 0.816, 95% CI: 0.708-0.939, p = 0.005). In contrast, no significant association was observed among women without a history of APOs (HR= 1.027, 95% CI: 0.913-1.156, p = 0.645). We also observed a statistically significant interaction between APO status and annual eGFR change for HTN incidence in total population (interaction p = 0.02).

CONCLUSION: The history of APOs is accompanied by alterations in kidney function in the long term. In women with a history of APOs, a positive change in eGFR levels was independently associated with a lower risk of HTN.

PMID:42056731 | DOI:10.1080/0886022X.2026.2644784

Womens Health (Lond). 2026 Jan-Dec;22:17455057261447666. doi: 10.1177/17455057261447666. Epub 2026 Apr 29.

ABSTRACT

BackgroundPregnant and postpartum incarcerated women are at risk of adverse health outcomes due to pre-incarceration risks and exposures during incarceration.ObjectivesThis study characterized maternal healthcare and neonatal outcomes among incarcerated pregnant women in Georgia from August 2020 to March 2025.DesignWe utilized a sequential mixed methods approach using data from a birth cohort of children exposed prenatally to incarceration in Georgia.MethodsWe analyzed qualitative data on maternal care experiences from 41 mothers and quantitative data on neonatal outcomes from caregivers of 84 children. Qualitative results informed exploratory tests of group differences for neonatal outcomes (i.e.,mode of delivery, neonatal complications, low birthweight) and breastfeeding initiation between (1) those who gave birth in the community compared to during incarceration; (2) those who were incarcerated in jail as compared to prisons.ResultsParticipants described low-quality maternal healthcare, inhumane treatment, and lack of safety and comfort, and the use of practices like solitary confinement and shackling. Fisher’s exact tests were significant for differences in breastfeeding initiation between those who gave birth in the community compared to those who gave birth during incarceration (p = 0.002), but breastfeeding initiation did not differ between those who were incarcerated in jail as compared to prison. Mode of delivery, neonatal complications, and low birthweight did not significantly differ between those who gave birth in the community compared to those who gave birth during incarceration nor those who were incarcerated in jail as compared to in prisons.ConclusionsIncarcerated pregnant and postpartum women in Georgia report low-quality maternal healthcare. Policy leaders should establish evidence-based policies for maternal healthcare within prisons and jails and consider community-based alternatives to incarceration for pregnant women.

PMID:42056725 | DOI:10.1177/17455057261447666

Diabetes Obes Metab. 2026 Apr 29. doi: 10.1111/dom.70795. Online ahead of print.

ABSTRACT

AIMS: To assess the efficacy and safety of alpha-lipoic acid (ALA) and pregabalin, both as mono and combination therapy, for treating painful diabetic peripheral neuropathy (DPN) in patients with type 2 diabetes mellitus, with the hypothesis that pregabalin monotherapy is non-inferior to combination therapy.

MATERIALS AND METHODS: A phase 4 randomized, active-controlled, open-label, multicentre trial was conducted over 12 weeks to investigate changes in visual analogue scale (VAS) pain scores from baseline as a primary efficacy endpoint. A total of 151 eligible subjects were randomly assigned to ALA (480 mg/day), pregabalin (150 mg/day), and combination groups in a 1:1:1 ratio.

RESULTS: The pregabalin monotherapy group showed a VAS change of -19.73 ± 18.94 mm, while the combination group showed -23.28 ± 18.15 mm at Week 12. The least square mean (LSM) difference between the two groups was 3.46 mm (95% CI: [-4.94, 11.87]), demonstrating that pregabalin monotherapy is non-inferior to combination therapy. Safety analysis revealed no significant differences across treatment groups. Cluster analysis revealed statistically significant differences in VAS scores between the pregabalin monotherapy and combination therapy groups at 12 weeks in cluster 1, characterized by a relatively shorter duration of DPN, and the LSM difference between both groups was 14.79 mm [4.59, 24.99] (p = 0.0055).

CONCLUSIONS: The pregabalin monotherapy demonstrated non-inferiority compared to the combination therapy in alleviating DPN pain. Cluster analysis supported the identification of patient groups where combination therapy could be more effective, but future comprehensive studies are required for further verification.

TRIAL REGISTRATION: ClinicalTrials.gov, NCT04846673.

PMID:42056717 | DOI:10.1111/dom.70795

Diabetes Obes Metab. 2026 Apr 29. doi: 10.1111/dom.70835. Online ahead of print.

ABSTRACT

BACKGROUND: Diabetes mellitus is widely regarded as a risk factor for cancers. There is considerable controversy over whether maternal diabetes can cause cancer. This study aims to comprehensively assess and quantify the association between maternal diabetes and the risk of cancers in mother-offspring.

METHODS: The PubMed, Embase, and Web of Science databases were searched up to September 30, 2025, to explore the impact of maternal diabetes on cancer in mothers and their offspring. The primary outcome was the risk of cancers in the mother or offspring, presented as risk ratios (RRs) with 95% confidence intervals (CI). The I² statistic is used to assess heterogeneity among studies, thereby guiding the selection of random-effects or common-effects models.

RESULTS: The 81 studies involving 44 917 447 mother-offspring. Maternal diabetes was associated with increased risks of any cancers in mothers and offspring. In studies adjusted for multiple confounders, the risk of offspring suffering from haematological malignancies (RR, 1.37; 95% CI, 1.23-1.52; I² = 1.0%) has significantly increased, especially leukaemia (1.34; 1.22-1.47; 0.0%). However, the risk of solid tumours (1.17; 1.09-1.24; 70.1%) in mothers increases significantly, especially, head and neck (1.34; 1.23-1.47; 35.1%), respiratory system (1.33; 1.05-1.67; 0.0%), gastrointestinal (1.30; 1.16-1.45; 45.5%), and gynecologic (1.16; 1.04-1.29; 64.1%). Maternal pre-gestational diabetes was more strongly associated with the risk of most cancers in offspring than gestational diabetes (1.60 [1.14-2.14] vs. 1.10 [1.02-1.18]; subgroup difference p = 0.0046).

CONCLUSION: Maternal diabetes is associated with an increased risk of cancers in mothers and offspring. Further high-quality large-sample studies are needed to clarify and consolidate potential causal relationships.

PMID:42056716 | DOI:10.1111/dom.70835