Category: Statistics

Eur J Epidemiol. 2025 Feb 13. doi: 10.1007/s10654-024-01185-7. Online ahead of print.

ABSTRACT

Recent advancements in machine learning (ML) for analyzing heterogeneous treatment effects (HTE) are gaining prominence within the medical and epidemiological communities, offering potential breakthroughs in the realm of precision medicine by enabling the prediction of individual responses to treatments. This paper introduces the methodological frameworks used to study HTEs, particularly based on a single randomized controlled trial (RCT). We focus on methods to estimate conditional average treatment effect (CATE) for multiple covariates, aiming to predict individualized treatment effects. We explore a range of methodologies from basic frameworks like the T-learner, S-learner, and Causal Forest, to more advanced ones such as the DR-learner and R-learner, as well as cross-validation for CATE estimation to enhance statistical efficiency by estimating CATE for all RCT participants. We also provide a practical application of these approaches using the Preventing Overweight Using Novel Dietary Strategies (POUNDS Lost) trial, which compared the effects of high versus low-fat diet interventions on 2-year weight changes. We compared different sets of covariates for CATE estimation, showing that the DR- and R-learners are useful for the estimation of CATE in high-dimensional settings. This paper aims to explain the theoretical underpinnings and methodological nuances of ML-based HTE analysis without relying on technical jargon, making these concepts more accessible to the clinical and epidemiological research communities.

PMID:39946045 | DOI:10.1007/s10654-024-01185-7

Environ Sci Pollut Res Int. 2025 Feb 13. doi: 10.1007/s11356-025-36049-4. Online ahead of print.

ABSTRACT

Traditional drought indices, such as the Standardized Precipitation Index (SPI) and Standardized Runoff Index (SRI), often fail to capture the complexity of drought events, which involve multiple interacting variables. To address this gap, this study applies the Principle of Maximum Entropy (POME) copula to combine SPI and SRI into a Joint Deficit Index (JDI), offering a more complete assessment of hydrometeorological drought. We used machine learning models, including Random Forest (RF), Quantile Random Forest (QRF), Extreme Gradient Boosting (XGB), and Quantile Regression XGBoost (QXGB), to predict JDI, while also incorporating uncertainty analysis using the Uncertainty Estimation based on Local Errors and Clustering (UNEEC) method. This approach not only improves the accuracy of drought predictions but also quantifies the uncertainty of the models, enhancing reliability. Model performance, evaluated with R², RMSE, and MAE, showed XGB as the best performer, achieving R² = 0.93 and RMSE = 0.16. This integration of multivariate drought indices, machine learning, and uncertainty analysis provides a more robust tool for drought monitoring and water resource management in arid regions.

PMID:39946044 | DOI:10.1007/s11356-025-36049-4

J Interv Card Electrophysiol. 2025 Feb 13. doi: 10.1007/s10840-025-02008-9. Online ahead of print.

ABSTRACT

BACKGROUND: Pulmonary vein isolation (PVI) is a cornerstone of AF ablation. Posterior wall isolation (PWI) has become a frequently used adjunct to PVI. While there is data to suggest that PVI alone does not negatively impact left atrial function, the effect of PWI on left atrial mechanical function has not been definitively determined. Our aim was to determine if PVI plus PWI using a cryoballoon impacted left atrial mechanical function as measured by cardiac MRI.

METHODS: We studied 28 patients who underwent ablation for AF. Fourteen patients had PVI alone and 14 patients had PVI plus PWI. All patients had cardiac magnetic resonance (CMR) before and after ablation. The primary outcome was change in LA ejection fraction (LAEF) as measured by CMR.

RESULTS: There were no statistically significant differences in the average patient age, height, weight, type of AF, or frequency of concomitant diseases between groups. No statistically significant differences in LAEF, LA max volume, LA min volume, or LA stroke volume were identified between baseline and follow up CMRs for the PVI only group nor the PVI plus PWI group. When utilizing linear regression analysis to compare change in LAEF, LA max volume, LA min volume, and LA stroke volume before and after ablation between groups, no statistically significant differences were identified.

CONCLUSION: Cardiac MRI did not demonstrate a significant change in left atrial mechanical function as measured by left atrial ejection fraction after pulmonary vein isolation alone nor after PVI plus posterior wall isolation.

PMID:39946035 | DOI:10.1007/s10840-025-02008-9

Drugs. 2025 Feb 13. doi: 10.1007/s40265-025-02157-1. Online ahead of print.

ABSTRACT

BACKGROUND AND OBJECTIVE: Bimatoprost implant 10 µg (Durysta) is an intracameral biodegradable implant that releases bimatoprost to lower intraocular pressure (IOP). The purpose of this study was to prospectively collect effectiveness and safety data after administration of the implant in patients with open-angle glaucoma or ocular hypertension.

METHODS: This phase IV, multicenter, prospective, observational, open-label, 18-month study (ARGOS) enrolled adult patients with open-angle glaucoma or ocular hypertension who were scheduled to receive the bimatoprost implant in one or both eyes. Data collected included IOP, use of topical IOP-lowering medications, treatment-emergent adverse events, and central corneal endothelial cell density. The primary endpoint was the proportion of primary (first-treated) eyes that received no additional (new) IOP-lowering treatment per standard medical care through month 6 after the implant administration.

RESULTS: A total of 217 patients (341 eyes) were enrolled, and 132 patients (60.8%) and 203 eyes (59.5%) completed the study. Most patients were on topical IOP-lowering medication before receiving the implant. After implant administration, the proportion of primary eyes that had received no additional treatment was 88.6% (95% confidence interval 86.6-90.6) at month 6 (primary endpoint) and remained high throughout the follow-up: 83.7% (95% confidence interval 80.2-87.3) at month 12 and 77.7% (95% confidence interval 73.4-82.1) at month 18. Intraocular pressure was reduced after implant administration, with mean changes in IOP from baseline at follow-up visits ranging from – 1.0 to – 2.0 mm Hg. The mean number of topical IOP-lowering medications used was also reduced, from 1.8 at baseline to 0.9 at month 12 and 1.0 at month 18. Increased IOP and dry eye were the most common ocular treatment-emergent adverse events. The mean percentage change in central corneal endothelial cell density from baseline at month 18 (central reading center evaluation) was – 3.47%. In qualitative interviews, most patients (84%, 21/25) reported overall satisfaction with their treatment outcomes.

CONCLUSIONS: The bimatoprost implant helped control IOP and decrease topical medication use. Throughout the 18 months after implant administration, an estimated 77.7% of eyes required no new added medication for IOP management. Patient-reported outcomes were favorable, and the safety profile of the implant was acceptable.

CLINICAL TRIAL REGISTRATION: ClinicalTrials.gov identifier NCT04647214, registered 23 November, 2020.

PMID:39946034 | DOI:10.1007/s40265-025-02157-1

Trop Anim Health Prod. 2025 Feb 13;57(2):59. doi: 10.1007/s11250-025-04308-8.

ABSTRACT

The purpose of this study was to examine the effects of anti-galactagogue (sage, parsley extract) and anti-inflammatory (echinacea extract) feed supplements on prolactin levels and fertility in the re-pregnancy of lactating Merino ewes. Eighty ewes were randomly assigned to one of two groups: control and treatment. In treatment group (Stop Lactin) group (n = 40) ewes were treated with Stop Lactin® 60 ml on day 0 orally by syringe. The control group (Control) (n = 40), ewes were not treated with any anti-galactagogue feed supplements. On day 0, a vaginal sponge containing 60 mg of medroxyprogesterone acetate was inserted for seven days. On day 7, all ewes received an intramuscular injection of 500 IU PMSG. The study results indicated no statistically significant differences between the Control group and Stop Lactin group in terms of conception rates (87.2% and 78.1%), late embryonic-early fetal mortality rate (20.6% and 8.0%), lambing rate (79.4% and 92%), twin rate (29.6% and 13.0%) and litter size (1.30 and 1.13). Estrus rates (97.5% and 82.1%, p = 0.050) tended to be higher and pregnancy rates (85.0% and 64.1%, p = 0.037) was found to be significant in the control group compared to the Stop Lactin group. The effect of treatment (P = 0.209) on serum prolactin levels and the interaction between treatment × day (P = 0.874) were both found to be insignificant. It is concluded that anti-galactagogue and anti-inflammatory feed supplements did not reduce prolactin concentration and did not improve fertility in lactating Merino ewes.

PMID:39945995 | DOI:10.1007/s11250-025-04308-8

Arch Osteoporos. 2025 Feb 13;20(1):23. doi: 10.1007/s11657-024-01479-z.

ABSTRACT

Through analyzing the data of the NHANES 2007-2020 cycle, this study concluded that high-intensity exercise 1-2 sessions a week can help maintain bone mass, and there is no significant difference from regular exercise more than 3 times a week.

PURPOSE: This study aims to explore the relationship between the various physical activity(PA) patterns and the risk of low bone mineral density(BMD) in Americans of working age.

METHOD: A total of 6482 participants aged 20-60 were selected from the National Health and Nutrition Survey (NHANES) conducted from 2007 to 2020. The PA data of the participants were obtained through individual interviews, and the participants were divided into four groups (inactive, insufficiently active, less frequent but sufficiently active(1-2 sessions a week and PA ≥ 150 min), and regularly active). Weighted logistic regression was used to analyze the correlation between PA patterns and the risk of low BMD. Subgroup analyses were applied to display the correlation between PA patterns and low BMD in different subgroups.

RESULT: After adjusting for confounding factors, the multiple logistic regression model showed that compared with inactive individuals, sufficiently active and regularly active individuals had a 35% (OR, 0.65; 95% CI, 0.46-0.92) and 24% (OR, 0.76; 95% CI, 0.62-0.93) lower risk of low BMD, respectively. Compared with regularly active adults, inactive adults had a 32% (OR, 1.32; 95% CI, 1.07-1.62) increased risk of low BMD, while sufficiently active individuals (OR, 0.85; 95% CI, 0.59-1.23) showed no significant difference compared with regularly active adults.

CONCLUSION: Compared with inactive adults, less frequent but sufficiently active adults have a lower risk of low BMD and showed benefits similar to those in regularly active groups. The sufficiently active pattern may become a new trend in modern working-age adults’ PA patterns.

PMID:39945926 | DOI:10.1007/s11657-024-01479-z

AMB Express. 2025 Feb 13;15(1):30. doi: 10.1186/s13568-025-01833-5.

ABSTRACT

Strengthening knowledge through surveillance and research is one of the main steps in treating antibiotic-resistant infections. The present study was therefore designed to analyse long-term antimicrobial resistance trends over the 12 years in Iran. Data from the patient’s clinical specimens were collected over 12 years from January 2012 until December 2023. Antimicrobial susceptibility patterns of gram-positive bacteria (GPB) were identified using the Kirby-Bauer disk diffusion method (DDM) and resistance proportions to most commonly used antibiotics were calculated for each GPB. All data were analysed using descriptive statistics. A total of 30,548 cultures were positive for GPB. Staphylococcus aureus (S. aureus) with 59.5% and coagulase-negative staphylococci (CoNS) with 26.6% frequently isolated from blood specimens. S. aureus was the frequently isolated organism in ≥ 6 years of age groups. S. aureus isolates showed the highest and lowest resistance rates to penicillin (91.8%) and vancomycin (1.1%), respectively. Nitrofurantoin (10.5%) was the most effective antimicrobial agent on Enterococcus spp. Time trends in antibiotic resistance revealed a sharp increase for several antibiotics: from 33.9% in 2013 to 54.5% in 2023 for gentamicin; from 51.2% in 2013 to 76.4% in 2023 for ceftriaxone; from 34.5% in 2013 to 54.8% in 2023 for imipenem; and from 66.7% in 2012 to 81.8% in 2023 for cefixime. The present research suggests that vancomycin and nitrofurantoin are effective antibiotics against infections caused by Staphylococcus and Enterococcus spp., respectively. This data revealed that the continuous monitoring of antimicrobial resistance (AMR) patterns is necessary for selecting the suitable drug.

PMID:39945925 | DOI:10.1186/s13568-025-01833-5

Thorac Cancer. 2025 Feb;16(3):e70002. doi: 10.1111/1759-7714.70002.

ABSTRACT

OBJECTIVE: This study was conducted to investigate the safety of neoadjuvant immunotherapy combined with chemotherapy in patients undergoing surgery for resectable stage III non-small cell lung cancer (NSCLC).

METHODS: Overall, 68 surgical patients with stage III NSCLC who underwent neoadjuvant therapy at the Thoracic Surgery Department of Beijing Chaoyang Hospital from June 2019 to September 2021 were included, including 19 patients who underwent neoadjuvant chemotherapy combined with immunotherapy and 49 who underwent neoadjuvant chemotherapy alone. Both groups of patients were diagnosed with NSCLC before treatment and had resectable stage III tumors. The surgical duration, blood loss volume, average postoperative hospital length of stay, intensive care unit length of stay, and complication rate were compared between the two groups.

RESULTS: The group treated with neoadjuvant chemotherapy combined with immunotherapy demonstrated higher values than the group treated with chemotherapy alone for surgical duration, blood loss volume, and rate of conversion to thoracotomy; however, the differences were not statistically significant. The incidence of postoperative complications in the group treated with neoadjuvant immunotherapy combined with chemotherapy was significantly higher than that of the group treated with neoadjuvant chemotherapy alone (p = 0.02).

CONCLUSION: Neoadjuvant immunotherapy combined with chemotherapy was safe and effective and did not increase the difficulty of surgery for NSCLC; however, it was associated with a higher incidence of complications than neoadjuvant chemotherapy alone (p < 0.05).

PMID:39940044 | DOI:10.1111/1759-7714.70002

Cardiooncology. 2025 Feb 12;11(1):16. doi: 10.1186/s40959-025-00315-8.

ABSTRACT

BACKGROUND: Myocarditis is a potentially fatal condition, with a mortality rate of up to 50% in severe cases. Studies, including those by Nobel Laureate Honjo, have implicated autoantibodies against cardiac troponin I (cTnI) in driving cardiac inflammation in mice. Research has also identified autoantibodies under baseline conditions in some cancer models. However, data on the effects of recombinant cTnI on autoantibody production, myocardial inflammation, and contractile function remain limited. This study investigated cTnI-associated myocardial inflammation and autoantibody formation in both tumor-free and tumor-bearing mouse models.

METHODS: Female BALB/c mice were immunized with recombinant cTnI combined with adjuvants and compared to adjuvant-only controls. Cardiac function was assessed using gated cardiac MRI, including myocardial velocities, acceleration, deceleration, and standard volumetric parameters including ejection fraction (EF). Anti-cTnI autoantibodies were quantified using a custom-designed ELISA, while myocardial inflammation was assessed by analyzing T-cell subsets (CD4 + and CD8 +) in myocardial tissue samples. Baseline autoantibody reactivity was evaluated in tumor-bearing mice and tumor-free controls for comparison.

RESULTS: The left ventricular ejection fraction trended lower in the cTnI + adjuvant group (57.80 ± 1.7%) compared to controls (61.67 ± 4.1%), but the difference was not statistically significant (p = 0.073). Myocardial velocity, reflecting contraction speed, was significantly reduced in cTnI-treated mice (control:-1.2 ± 0.8 cm/s; cTnI:-1.05 ± 0.07 cm/s; p = 0.015). Anti-cTnI autoantibody levels increased significantly in cTnI-treated mice at 8 weeks (control:0.1 ± 0.02; cTnI:0.77 ± 0.28; p = 0.007). Additionally, the density of CD8 + T-cells in myocardial tissue was significantly higher in the cTnI group (control:2.2 ± 1.2 cells/mm²; cTnI:4.4 ± 2 cells/mm²; p = 0.013), indicating an enhanced cytotoxic T-cell response. The CD4/CD8 ratio was significantly lower in cTnI-treated mice (control: 8.2 ± 6.8; cTnI:3.1 ± 0.9; p = 0.029), further suggesting a shift toward a cytotoxic immune profile. Baseline autoantibody reactivity in tumor-bearing mice was not significantly different from controls (tumor-bearing: absorbance 0.049 ± 0.029; control: absorbance 0.068 ± 0.05 at 450 nm), indicating no inherent autoimmune reactivity in the tumor-bearing model.

CONCLUSIONS: Recombinant cTnI induces myocardial contractile dysfunction and promotes a cytotoxic immune response, supporting its role as an autoantigen in myocarditis. Advanced cardiac MRI revealed subtle functional impairments that EF alone could not detect. These findings highlight the potential for therapies targeting cTnI-induced autoimmunity, particularly in patients with ICI-associated myocarditis.

PMID:39940032 | DOI:10.1186/s40959-025-00315-8

BMC Health Serv Res. 2025 Feb 12;25(1):239. doi: 10.1186/s12913-025-12369-6.

ABSTRACT

BACKGROUND: Work organization significantly impacts occupational incidents and fatigue in hospital settings, particularly among nurses. This study aimed to evaluate the psychometric properties of instruments measuring work organization and fatigue and to examine their relationship with occupational accidents.

METHODS: A cross-sectional study was conducted in 2019 with 200 nurses working in hospitals in Qom, Iran using the stratified sampling method. Data were collected using three standardized tools: the Work Organization Questionnaire, the Fatigue Checklist, and a demographic information questionnaire. Structural equation modeling was employed to analyze the data, while instrument validity and reliability were assessed through Cronbach’s alpha, composite reliability, and average variance extracted (AVE). Analysis was performed using Smart PLS and SPSS V20.

RESULTS: The analysis revealed a significant relationship between work organization and occupational accidents (t = 3.22, p < 0.05). However, the relationships between work organization and fatigue (t = 0.03) and between fatigue and occupational accidents (t = 1.49) were not statistically significant. The Work Organization Questionnaire (WOAQ) demonstrated robust validity and reliability, making it suitable for assessing occupational risks in hospital environments. In contrast, the Fatigue Questionnaire (CIS) exhibited acceptable validity but insufficient reliability (Cronbach’s alpha < 0.7), highlighting the need for further refinement.

CONCLUSION: This study revealed that the Work Organization Questionnaire has acceptable validity and reliability, making it suitable for hospital settings, while the Fatigue Questionnaire requires further revision. It is recommended that hospital administrators optimize work schedules and provide fatigue management training, and policymakers utilize validated tools to reduce occupational risks and enhance workplace safety.

PMID:39940007 | DOI:10.1186/s12913-025-12369-6