Tag: statistics

Drugs Aging. 2026 Jul 31. doi: 10.1007/s40266-026-01324-7. Online ahead of print.

ABSTRACT

BACKGROUND: Loneliness is associated with high-risk medication use in older adults in cross-sectional studies, but the direction of this relationship is unclear.

OBJECTIVE: The aim of this study was to examine the association between loneliness and initiation of potentially inappropriate pain and psychotropic medications, and test for sex interactions.

METHODS: We conducted a retrospective cohort study of community-dwelling respondents to the Canadian Community Health Survey-Healthy Aging who were interviewed between December 1, 2008, and November 30, 2009, aged ≥66 years, and Ontario residents. Self-reported loneliness was defined at baseline as a score of ≥6 on the Three-Item Loneliness Scale. Survey responses were linked to health records; respondents were followed for 3 years to assess initiation of a potentially inappropriate pain or psychotropic medication, defined using the 2019 American Geriatrics Society’s Beers Criteria. We used weighted Cox proportional hazards regression models to estimate adjusted hazard ratios (HRs) and tested for sex interactions.

RESULTS: Of 2348 respondents (female 54.6%, mean age 75.4 years), 383 (12.3%) were lonely. Compared with those who were not lonely, lonely female respondents had higher rates of initiating potentially inappropriate pain medications (HR at 90 days: 1.57, 95% CI 0.96-2.27; 630 days: 1.52, 95% CI 1.05-2.12; 1080 days: 3.22, 95% CI 1.22-7.78) and potentially inappropriate psychotropic medications, but the latter estimates were imprecise. No association was found in males.

CONCLUSION: Lonely older females initiate potentially inappropriate pain and psychotropic medications at higher rates than those who are not lonely. Screening for loneliness could prioritize patients for medication reviews, possible deprescribing, and interventions that address root causes.

PMID:42536335 | DOI:10.1007/s40266-026-01324-7

Folia Microbiol (Praha). 2026 Jul 31. doi: 10.1007/s12223-026-01567-2. Online ahead of print.

ABSTRACT

Antibiotic resistance is the ability of microorganisms to survive and proliferate despite exposure to antibiotics that would normally inhibit or kill susceptible strains. This resistance can make antibiotics ineffective or diminish their ability to combat microorganisms, complicating the treatment of infections and potentially leading to serious complications. To combat antibiotic resistance, a comprehensive analysis of changes in the metabolic activities of microorganisms is crucial for understanding the underlying mechanisms. Proteus mirabilis is a critical pathogen, particularly as a common cause of urinary tract infections (UTIs), especially in women. Typically, treating P. mirabilis infections relies on antibiotic-based therapies. However, when faced with resistant strains, treatment options become limited. This research aims to simulate how antibiotic resistance develops in P. mirabilis when exposed to sub-inhibitory concentrations of ampicillin and to explore whether ampicillin-resistant strains display cross-resistance to other antibiotics through metabolomic approaches. For this purpose, P. mirabilis strains were gradually exposed to sub-inhibitory concentrations of ampicillin using the disk diffusion method, leading to the selection of resistant passages. Ampicillin and ceftazidime were then applied to both ampicillin-resistant passages and sensitive control strains, and differences in their metabolomic profiles were compared. The metabolomic data obtained from this study were supported by statistical and bioinformatics analyses to facilitate metabolic pathway mapping, comprehend metabolic alterations, and identify interactions among metabolites. Metabolomic studies in antibiotic resistance research provide valuable insights into identifying metabolic alterations in resistant microorganisms, understanding microbial responses to antibiotic exposure, clarifying the effects of antibiotics on metabolic pathways, and gaining a comprehensive perspective on the mechanisms of antibiotic resistance.

PMID:42536331 | DOI:10.1007/s12223-026-01567-2

Ann Nucl Med. 2026 Jul 31. doi: 10.1007/s12149-026-02255-4. Online ahead of print.

ABSTRACT

Partial volume effects (PVE) in positron emission tomography (PET) imaging introduce quantification inaccuracies, particularly in small or heterogeneous lesions, necessitating precise correction methodologies. This study systematically evaluates the performance of statistical versus machine learning approaches in partial volume correction (PVC) across multiple PET reconstruction algorithms, including TrueX, TrueX + TOF, and Iterative + TOF. Phantom experiments were conducted across a broad range of lesion-to-background contrast ratios and lesion sizes to derive and validate exponential recovery coefficient (RC) fitting models. Additionally, advanced machine learning algorithms, including Random Forest, Support Vector Regression, and Gradient Boosting, were implemented to enhance PVC accuracy. The results demonstrate that Iterative + TOF reconstruction yielded the most consistent RC estimates, while machine learning-based PVC significantly outperformed traditional exponential fitting in minimizing residual errors. Among the machine learning models, Random Forest exhibited the lowest root mean square error (RMSE) and highest coefficient of determination (R²), indicating superior predictive accuracy and robustness. These findings underscore the potential of machine learning-driven PVC methodologies for standardizing PET quantification, thereby improving lesion characterization, therapy response assessment, and multi-center data harmonization.

PMID:42536328 | DOI:10.1007/s12149-026-02255-4

Clin Rheumatol. 2026 Jul 31. doi: 10.1007/s10067-026-08325-2. Online ahead of print.

ABSTRACT

PURPOSE: The clinical benefit of combining leflunomide (LEF) with tumor necrosis factor inhibitors (TNFi) in psoriatic arthritis (PsA) remains uncertain. We aimed to evaluate the efficacy and treatment durability of LEF-TNFi compared with non-LEF regimens (predominantly methotrexate (MTX)-TNFi and TNFi monotherapy).

METHODS: This retrospective cohort included 492 biologic-naive PsA patients initiating TNFi (2003-2020): LEF-TNFi (n = 85) versus non-LEF (n = 407). Multiple imputation addressed missing data, and propensity score matching (7 covariates; caliper 0.2 standard deviations of the logit-propensity score) addressed confounding by indication. Longitudinal outcomes were analyzed using linear mixed-effects models; treatment modification was evaluated via Cox models. Reasons for treatment modification were examined descriptively using a competing risks framework.

RESULTS: Substantial baseline imbalances (23 of 36 variables with standardized mean difference > 0.10) were eliminated by propensity score matching (0 of 7 matching covariates with SMD > 0.10; 96.9% of LEF patients retained). Post-adjustment, longitudinal disease activity trajectories did not differ significantly between groups (time-by-treatment interactions: Disease Activity Score in 28 joints (DAS28), p = 0.862; Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), p = 0.308). Overall treatment modification rates were similar (propensity score-matched hazard ratio (HR) = 1.16; 95% confidence interval (CI), 0.53-2.51; p = 0.709). Descriptively, LEF patients were more frequently subject to treatment modification for remission (10.6% vs. 5.9%) and less frequently for inefficacy (5.9% vs. 11.1%), although cause-specific hazard ratios did not reach statistical significance.

CONCLUSION: After propensity score adjustment, LEF-TNFi showed no detectable difference in disease activity trajectories or overall treatment persistence compared with MTX-TNFi and TNFi monotherapy. However, LEF-TNFi modifications were predominantly driven by achieved remission rather than inefficacy. Keypoints • Propensity score-adjusted analyses revealed no detectable difference in disease activity trajectories between the LEF-TNFi, MTX-TNFi, and TNFi monotherapy groups in psoriatic arthritis. • Competing risks analysis showed that LEF modifications were driven by remission rather than inefficacy, a clinical distinction obscured by standard composite endpoints. • These hypothesis-generating findings suggest that LEF may be a viable alternative to MTX as concomitant csDMARD therapy with TNFi in PsA.

PMID:42536327 | DOI:10.1007/s10067-026-08325-2

Clin Rheumatol. 2026 Jul 31. doi: 10.1007/s10067-026-08341-2. Online ahead of print.

ABSTRACT

BACKGROUND: Hashimoto’s thyroiditis (HT) and Sjögren’s syndrome (SS) frequently co-occur clinically, implying a potential association between the two diseases. However, their genetic association remains unclarified. To investigate whether there is a genetic link between HT and SS, we performed a two-sample bidirectional Mendelian randomization (MR) analysis.

METHODS: Data for HT were obtained from the IEU Open GWAS project, consisting of 15,654 cases and 379,986 controls. Data for SS were sourced from FinnGen Release 11, with 2981 cases and 439,424 controls included. We adopted the inverse variance-weighted (IVW) method plus four complementary robust MR methods to evaluate the genetic association between HT and SS. Comprehensive sensitivity analyses were implemented to verify the robustness of the MR estimates. Furthermore, a reverse MR analysis was performed to explore the potential for reverse association.

RESULTS: After rigorous screening, seven single nucleotide polymorphisms (SNPs) were selected as instrumental variables (IVs) for HT, while six SNPs served as IVs for SS. Positive MR analysis revealed a statistically significant causal effect of HT on SS, with an IVW odds ratio (OR) of 1.2188 (95% confidence interval (CI): 1.0755-1.3813; P = 0.0019). This finding was further validated by the weighted median and weighted mode methods (P < 0.05). Conversely, inverse MR analysis identified a statistically significant causal effect of SS on HT, with an IVW OR of 1.2154 (95% CI: 1.1387-1.2972; P < 0.001). This result was corroborated by the weighted median, MR-Egger, weighted mode, and simple mode methods (P < 0.05). Sensitivity analysis confirmed the robustness of these findings.

CONCLUSIONS: This study identifies a bidirectional genetic association between genetically predicted HT and SS in European populations. The observed relationship is likely attributable to shared autoimmune predisposition. These results may offer a useful reference for exploring their underlying pathogenesis. Key Points • HT and SS often coexist in clinical practice; nevertheless, the exact genetic relationship between 3 them remains to be elucidated. • A two-sample bidirectional MR approach was employed to evaluate the genetic relationship between HT and SS. • Our MR study identified a bidirectional genetic association between genetically predicted HT and SS in European populations.

PMID:42536326 | DOI:10.1007/s10067-026-08341-2

Endocrine. 2026 Jul 31;91(1):241. doi: 10.1007/s12020-026-04701-9.

ABSTRACT

PURPOSE: Glucagon-like peptide-1 receptor agonists (GLP-1 RAs), used to treat type 2 diabetes and obesity, may improve metabolic health before conception. However, their association to hypertensive disorders of pregnancy (HDP) after periconceptional or early pregnancy exposure remains unknown.

METHODS: We performed a meta-analysis to investigating association between GLP-1 RAs preconception or first trimester of gestation exposure and HDP risk. Cochrane Central Register of Controlled Trials databases, ClinicalTrials.gov, PubMed, Scopus, and EMBASE databases were searched from inception through December 15, 2025. All eligible studies were observational cohorts. Case reports, reviews, editorials, and studies that lacked HDP data were excluded. We pooled odds ratios with 95% confidence intervals using a random-effects Mantel-Haenszel model. ROBINS-I tool was used to evaluate risk of bias.

RESULTS: Of 75 records identified, 3 retrospective cohort studies met inclusion criteria with 10,880 pregnancies (4942 exposed to GLP-1 RAs and 5938 unexposed). All the studies were conducted in the United States between 2014-2025 and evaluated the semaglutide, liraglutide, dulaglutide, tirzepatide, exenatide, lixisenatide, and albiglutide exposure. Two studies showed a lower HDP risks among exposed pregnant, whereas one study found a higher risk. In the pooled analysis, GLP-1 RA exposure showed a HDP risk with an OR 0.91 (OR 0.91, CI 0.57-1.47) with no statistical significance.

CONCLUSION: Periconceptional or first-trimester exposure to GLP-1 RAs are not significantly associated with HDP risk. This available evidence is limited and indicating the need for large prospective studies to establish a possible association between GLP-1 RAs are not significantly associated with HDP risk.

PMID:42536323 | DOI:10.1007/s12020-026-04701-9

Drug Saf. 2026 Jul 31. doi: 10.1007/s40264-026-01697-z. Online ahead of print.

ABSTRACT

BACKGROUND: Nonsteroidal anti-inflammatory drugs (NSAIDs) and paracetamol have been associated with neutropenia and agranulocytosis with inconsistent results.

OBJECTIVE: To investigate the risk of neutropenia in association with frequently used NSAIDs compared to paracetamol.

METHODS: We conducted a cohort study using the UK-based Clinical Practice Research Datalink (CPRD) GOLD. In three pairwise comparisons, we compared the risk of neutropenia including agranulocytosis (defined by Read codes) between new NSAID users (diclofenac, ibuprofen, and naproxen) and new paracetamol users (active comparator) during a maximum follow up of 60 days. Secondary and tertiary outcomes additionally included (1) in-patient diagnosed agranulocytosis and (2) laboratory values indicating neutropenia. Both were additionally restricted to only agranulocytosis. We applied propensity score-fine stratification to control for measured confounding and quantified incidence rates (IRs) as well as hazard ratios (HRs) with 95% confidence intervals (CIs).

RESULTS: Our weighted cohorts included 1,003,314 (paracetamol) to 2,207,612 (diclofenac) patients. Weighted IRs of neutropenia (primary outcome) were between 2.1/10,000 person years (PYs) and 2.3/10,000 PYs. HRs for the primary outcome ranged between 1.00 (95% CI 0.51-1.94) and 1.12 (95% CI 0.57-2.23) for NSAIDs versus paracetamol. The secondary and tertiary outcome yielded reduced HRs between 0.53 and 1.03, and even lower HRs when restricted to agranulocytosis (HR between 0.19 and 0.51).

CONCLUSION: Our results indicate no risk of neutropenia for NSAIDs when compared to paracetamol. An increased risk of agranulocytosis in association with paracetamol is possible, but residual confounding by frailty may at least partially explain this association.

PMID:42536322 | DOI:10.1007/s40264-026-01697-z

J Urban Health. 2026 Jul 31. doi: 10.1007/s11524-026-01117-8. Online ahead of print.

ABSTRACT

Extreme heat events increasingly threaten public health, particularly in rapidly urbanizing areas like Maricopa County, Arizona. This study addresses gaps in identifying communities most vulnerable to extreme heat and heat waves by creating a Heat Vulnerability Index (HVI) that integrates often-overlooked populations. Utilizing US census data, satellite imagery, chronic illness prevalence rates, and unhoused population data, this HVI assesses vulnerability across census tracts in Maricopa County’s diverse urban-rural landscape. Principal components analysis identified nine factors influencing heat vulnerability: (1) socioeconomic disadvantage; (2) isolation; (3) elderly populations; (4) chronic illness; (5) environmental risks; (6) African American race and language barriers, (7) Native American and unemployment status; (8) lack of housing and male; and (9) mobile home residents. Model validation found that heat-related mortality rate increased with heat vulnerability. Despite statistical limitations from data resolution and timeframe, this study integrates unhoused data into vulnerability assessments, emphasizing the need for equitable approaches that include underserved communities to address extreme heat vulnerability.

PMID:42536312 | DOI:10.1007/s11524-026-01117-8

Updates Surg. 2026 Jul 31. doi: 10.1007/s13304-026-02781-w. Online ahead of print.

ABSTRACT

Bethesda IV thyroid nodules remain a major diagnostic challenge because cytology cannot reliably distinguish benign from malignant follicular-patterned lesions, often leading to diagnostic surgery for ultimately benign disease. This study evaluated whether ultrasound radiomics combined with machine learning could improve preoperative risk stratification in this setting. We conducted a retrospective monocentric study including 69 surgically treated patients with Bethesda IV thyroid nodules and definitive histopathology. Ultrasound images acquired between 2019 and 2020 were manually segmented, and radiomic features were extracted using LIFEx software. After preprocessing and removal of non-informative features, the dataset was split into training (n = 48) and a held-out test set (n = 21), isolated prior to any modelling procedure. Dimensionality reduction was performed with principal component analysis analysis (20 components, 99.5% explained variance). Three supervised machine-learning models were developed and compared: K-nearest neighbors (KNN), Random Forest (RF), and Extreme Gradient Boosting (XGBoost). Internal validation was performed by leave-one-out cross-validation on the training set, with StandardScaler and PCA fitted inside each fold. Final performance was assessed once on the held-out test set. All metrics are reported with 95% confidence intervals. Model performance was assessed using cross-validation and leave-one-out cross-validation, with evaluation of accuracy, F1-score, sensitivity, specificity, and ROC-AUC. Final histology showed malignancy in 39/69 nodules (56.5%). On the held-out test set, RF (n = 50 estimators, depth = 10) achieved the best overall performance: AUC 0.735 (95% CI 0.513-0.956), sensitivity 0.714 (95% CI 0.454-0.883), specificity 0.571 (95% CI 0.250-0.842), and accuracy 0.667 (95% CI 0.454-0.828). KNN (k = 3) achieved AUC 0.602 (95% CI 0.359-0.845). XGBoost did not generalise to the test set (AUC 0.378, 95% CI 0.094-0.661). Leave-one-out cross-validation estimates were modest across all models (AUC range 0.397-0.588), with confidence intervals overlapping 0.50, reflecting the limited statistical power of the training set at n = 48. Ultrasound radiomics combined with Random Forest shows preliminary discriminative ability for distinguishing benign from malignant Bethesda IV thyroid nodules. The wide confidence intervals observed highlight the need for a larger prospective validation cohort. These findings establish a methodological framework and provide sample size benchmarks for a powered confirmatory study.

PMID:42536308 | DOI:10.1007/s13304-026-02781-w

Updates Surg. 2026 Jul 31. doi: 10.1007/s13304-026-02785-6. Online ahead of print.

ABSTRACT

A catecholamine-induced hypertensive crisis in patients with pheochromocytoma has recently been defined as systolic/diastolic blood pressure ≥ 180/120 mmHg. However, the consequences of observing blood pressure values above these thresholds during laparoscopic adrenalectomy have never been evaluated. The aim of this study was to identify factors associated with intraoperative catecholamine-induced hypertensive crisis (main objective) and postoperative cardiovascular complications (secondary objective). A multi-institutional retrospective cohort study from 01/01/2000 to 12/31/2016 was performed in eight university hospitals to identify independent factors associated with catecholamine-induced hypertensive crisis while adjusting for the clustering of patients within hospitals (generalized linear mixed model for hierarchical analysis). Logistic regression analysis was used to analyze postoperative cardiovascular complications on day 30. A total of 1056 patients underwent adrenalectomy for pheochromocytoma, of whom 894 were included in the final analysis. During adrenalectomy, 122 patients (13.6%) presented with at least one episode of a catecholamine-induced hypertensive crisis. Cardiovascular complications were observed in 32 patients (3.6%) on the 30th postoperative day. Tumor diameter (OR 1.011, 95%CI 1.001-1.021; p = 0.035) and any biological profile with high adrenaline (OR 1.602, 95%CI 1.051-2.441; p = 0.028) were significantly associated with the occurrence of intraoperative catecholamine-induced hypertensive crisis. The occurrence of intraoperative catecholamine-induced hypertensive crisis (OR 2.4, 95%CI 1.603-3.707; p = 0.033) remained significantly associated with cardiovascular complications at 30 days postoperatively. Catecholamine-induced hypertensive crisis is a relevant criterion for evaluation during laparoscopic adrenalectomy for pheochromocytoma because it is associated with an increased incidence of postoperative cardiovascular complications.

PMID:42536306 | DOI:10.1007/s13304-026-02785-6