Category: Statistics

Eur J Cardiovasc Nurs. 2026 Feb 18:zvaf223. doi: 10.1093/eurjcn/zvaf223. Online ahead of print.

ABSTRACT

AIMS: To generally explore the relative importance of factors influencing medication adherence after percutaneous coronary intervention (PCI), and to provide a new perspective on the theoretical framework of medication adherence intervention.

METHODS AND RESULTS: A single-centre, cross-sectional study was conducted. Primary outcomes were indicators of medication nonadherence, including incorrect implementation behaviour, medication taking proportion <90% and discontinuation. Potential influencing factors included baseline characteristics at PCI and barriers to medication adherence after PCI. Binary logistic regression was used to analyse factors associated with the indicators. A total of 901 cases with valid data were identified. Incorrect implementation behaviour, medication taking proportion <90%, and discontinuation after PCI were 39.3%, 16.6%, and 8.9%, respectively. Rural residence, lacking medication necessary, and medication side effects were common factors associated with the above three indicators, and complex prescriptions, disturbing routine, age (at this PCI), number of coronary stents (up to this PCI), economic difficulties, memory decline, inconvenience, busy, and extensive travel were uncommon factors associated with these three indicators.

CONCLUSION: Exploring the relative importance of factors influencing medication adherence after PCI provides a new perspective for integrating multifaceted and tailored interventions as a new system.

PMID:41703427 | DOI:10.1093/eurjcn/zvaf223

Clin Trials. 2026 Feb 17:17407745251415538. doi: 10.1177/17407745251415538. Online ahead of print.

ABSTRACT

BACKGROUND: An estimand is a clear description of the treatment effect a study aims to quantify. The ICH E9(R1) addendum lists five attributes that should be described as part of the estimand definition. However, the addendum was primarily developed for individually randomised trials. Cluster randomised trials, in which groups of individuals are randomised, have additional considerations for defining estimands (e.g. how individuals and clusters are weighted, how cluster-level intercurrent events are handled). However, it is currently unknown if estimands are being used in cluster randomised trials, or whether the considerations specific to cluster randomised trials are being described.

METHODS: We reviewed 73 cluster randomised trials published between October 2023 and January 2024 that were indexed in MEDLINE. For each trial, we assessed whether the estimand for the primary outcome was described, or if not, whether it could be inferred from the statistical methods. We also assessed whether considerations specific to cluster randomised trials were described or inferable, how trials were analysed and whether key assumptions being made in the analysis (e.g. ‘no informative cluster size’) could be identified.

RESULTS: No trials attempted to describe the estimand for their primary outcome. We were able to infer the five attributes outlined in ICH E9(R1) in only 49% of trials, and when including additional considerations specific to cluster randomised trials, this figure dropped to 21%. Key drivers of this ambiguity were lack of clarity around whether individual- or cluster-average effects were of interest (unclear in 63% of trials), and how cluster-level intercurrent events were handled (unclear in 21% of trials for which this was applicable). Over half of trials used mixed-effects models or generalising estimating equations with an exchangeable correlation structure, which make the assumption that there is no informative cluster size; however, only one of these trials performed sensitivity analyses to evaluate robustness of results to deviations from this assumption. There were 14% of trials that used independence estimating equations or the analysis of cluster-level summaries; however, because no trials stated whether they were targeting the individual- or cluster-average effect, it was impossible to determine whether these methods implemented the appropriate weighting scheme and were thus unbiased.

CONCLUSION: The uptake of estimands in published cluster randomised trial articles is low, making it difficult to ascertain which questions were being investigated or whether statistical estimators were appropriate for those questions. This highlights an urgent need to develop guidelines on defining estimands that cover unique aspects of cluster randomised trials to ensure clarity of research questions in these trials.

PMID:41703418 | DOI:10.1177/17407745251415538

J Appl Clin Med Phys. 2026 Feb;27(2):e70509. doi: 10.1002/acm2.70509.

ABSTRACT

PURPOSE: The InTempo adaptive imaging system is an important component of the Accuray CyberKnife System, designed to enhance the system’s ability to track and correct tumor motion during treatment. However, a limitation of this feature is the reduction of available nodes for treatment planning. The impact of a reduced number of nodes on the quality of InTempo-based treatment plans has not previously been evaluated. This retrospective study aims to compare the dosimetry of CyberKnife plans with and without The InTempo path set for both prostate and lung stereotactic body radiotherapy (SBRT).

METHODS: This study included twelve consecutive prostate SBRT patients and twenty selected lung SBRT patients. The selection criteria for the 20 lung patients were motivated by being able to construct a data set representative of common treatment tracking methods and dose prescriptions. To evaluate the impact of InTempo imaging, treatment plans were re-optimized using the same optimization parameters and machine settings, except for the path set with the maximum number of nodes. To ensure a fair comparison, the study plans were prescribed using identical planning target volume coverage as the clinical treatment plans. Statistical analyses were performed using mean and standard deviation, dose metric plots, and a two-sided Wilcoxon signed rank test with multiple testing correction to compare dose metrics between different path sets.

RESULTS: No statistically significant differences were observed among the Prostate, Prostate_Short, and their corresponding InTempo path sets in at least 8 of the 14 evaluated plan metrics, including prostate clinical tumor volume (CTV) V40Gy(%), conformity index, and homogeneity index. For example, the mean prostate CTV V40Gy (%) for the Prostate, Prostate_Short, and their corresponding InTempo path sets was 90.8 ± 4.7, 89.4 ± 4.7, 90.2 ± 3.9, 91.0 ± 7.0, respectively. However, compared with the Prostate path set, the Prostate_InTempo path set exhibited a statistically significant reduction in delivery time (p = 0.0010), number of beams, and bladder V18Gy (%), along with a statistically significant increase in the number of imaging beams (p = 0.0010). Additionally, Prostate_Short demonstrated statistically significant reductions in delivery time and number of beams compared with the Prostate path set, while the number of imaging beams remained statistically equivalent. In contrast, the Reduced_Prostate and Reduced_Prostate_InTempo sets consistently resulted in inferior dosimetric outcomes, with several plans deemed unoptimizable due to insufficient node availability. For lung SBRT, statistically significant differences were observed in delivery time and the number of imaging beams between plans with and without InTempo. However, no statistical differences were found in dose distribution metrics between these two lung groups.

CONCLUSIONS: InTempo-compatible path sets do not significantly compromise plan quality for prostate or lung SBRT, provided adequate node availability. Specifically, the Prostate_InTempo and Prostate_Short path sets demonstrated a reduction in delivery time and an increase in adaptive imaging frequency compared with the Prostate path set. However, the Reduced_Prostate and Reduced_Prostate_InTempo result in inferior plan quality and reduced deliverability and should be used with caution. These findings support the selective use of InTempo imaging in SBRT planning without sacrificing dosimetric integrity.

PMID:41703417 | DOI:10.1002/acm2.70509

J Cardiovasc Med (Hagerstown). 2026 Feb 1;27(2):126-132. doi: 10.2459/JCM.0000000000001836. Epub 2026 Jan 19.

ABSTRACT

BACKGROUND: Takotsubo cardiomyopathy (TCM) is an acute form of left-ventricular systolic dysfunction triggered by emotional or physical stress, which can lead to refractory cardiogenic shock. In such cases, mechanical cardiovascular support, such as extracorporeal membrane oxygenation (ECMO), may be beneficial. However, the outcomes of ECMO in this population remain unclear.

OBJECTIVE: To evaluate the association between ECMO and in-hospital outcomes in patients hospitalized with TCM and cardiogenic shock.

METHODS: We conducted a retrospective cohort study using the National Inpatient Sample from 2016 to 2022 to evaluate outcomes in adult patients hospitalized with TCM and cardiogenic shock. ECMO use was identified using ICD-10 procedure codes. The primary outcome was in-hospital mortality. Secondary outcomes included hospital length of stay (LOS), total hospital charges (THCs), acute kidney injury, and bleeding complications. Propensity score matching with double adjustment using survey-weighted logistic and linear regression was used to adjust for confounders.

RESULTS: A total of 20 350 weighted hospitalizations were included, with 300 patients (1.5%) receiving ECMO. In the unadjusted analysis, ECMO was associated with higher in-hospital mortality (35.0 vs. 27.7%), longer LOS (19.4 vs. 12.1 days), and higher THCs ($761 206 vs. $254 690). After matching, 270 patients were identified in each group. ECMO was still associated with higher THCs ($630 317 vs. $372 195). In-hospital mortality remained higher in the ECMO group (32.5% vs. 26.7%), although not statistically significantly (P = 0.49).

CONCLUSION: Among patients with TCM complicated by cardiogenic shock, ECMO was not associated with a significant reduction in mortality. Further studies are warranted to improve patient risk stratification and clarify the clinical value of ECMO in this population.

PMID:41703407 | DOI:10.2459/JCM.0000000000001836

J Foot Ankle Res. 2026 Mar;19(1):e70135. doi: 10.1002/jfa2.70135.

ABSTRACT

BACKGROUND: Plantar fasciitis (PF) is a common cause of heel pain that affects the health-related quality of life of many individuals and has various treatment options. Two effective interventions are corticosteroid (CS) injections and dextrose prolotherapy (DP). This study aimed to compare the efficacy and safety of DP and CS in patients with PF systematically.

METHODS: Relevant studies, including those comparing DP and CS for treating PF, were identified by searching electronic databases until August 2025. The visual analog scale (VAS) pain score, foot function index (FFI), and plantar fascia thickness (PFT) were compared between the groups in the short term (0.5-1 month) and mid-term (3 months). Statistical analyses were performed via RevMan 4.5.1, and p < 0.05 was considered statistically significant.

RESULTS: Five RCTs and two cohort studies, with a total of 567 patients, were included in the meta-analysis. The analysis revealed that at the short-term follow-up (1 month), corticosteroid injections were more effective at reducing the VAS pain scores than dextrose prolotherapy for general VAS score (MD = 1.85, 95% CI [0.05, 3.64], p = 0.04), the VAS score at the first step in the morning (MD = 1.26, 95% CI [0.49, 2.02], p = 0.001), and the VAS score for pain while walking (MD = 1.85, 95% CI [0.68, 3.02], p = 0.002). Similarly, at the short-term follow-up (1 month), the analysis revealed a significantly greater reduction in the FFI score (MD = 18.81, 95% CI [0.06, 37.55]) and PFT (MD = 0.26 mm, 95% CI [0.07, 0.45]) in the CS group than in the DP group. At 3 months, the analysis revealed a significant decrease in the FFI score (p = 0.003) in the DP group compared with the CS group, whereas no significant difference was observed in the VAS scores or PFT.

CONCLUSION: In patients with plantar fasciitis, CS injections had greater efficacy than DP did in the short term; however, their efficacy became similar in the mid-term follow-up, with DP outperforming CS in terms of foot function. Further trials with standardized protocols and long-term follow-ups are needed to address potential biases.

PMID:41703400 | DOI:10.1002/jfa2.70135

J Gen Intern Med. 2026 Feb 17. doi: 10.1007/s11606-026-10257-1. Online ahead of print.

ABSTRACT

BACKGROUND: Most people with opioid use disorder (OUD) do not receive evidence-based treatment. To increase treatment rates, primary care clinics may choose to implement risk prediction tools available in the electronic health record (EHR) to identify patients with a high risk of OUD or overdose.

OBJECTIVE: To externally validate Epic’s cognitive computing model to predict the Risk of Opioid Abuse or Overdose (referred to as the Opioid Risk Score; ORS) in three large integrated health systems.

DESIGN: Prospective cohort study secondary to an ongoing clinical trial.

PARTICIPANTS: Patients (N = 704,764) aged 18-75 who had a primary care encounter during the study period (April 2021-December 2022) and did not have an OUD diagnosis at index.

MAIN MEASURES: Data were extracted from the EHR. The index date was defined as the first date within the study period where the patient met eligibility criteria and had an ORS calculated by the EHR. The binary outcome variable was whether the patient was diagnosed with OUD or experienced an opioid overdose within 12 months of the index date.

KEY RESULTS: Most patients were classified as low risk on ORS (99.6%). Few patients experienced an OUD diagnosis or overdose in the 12-month follow-up period (0.3%). The model correctly classified 185 of 2362 patients who experienced an event (sensitivity 0.0783, 95% CI 0.0675, 0.0892) and 699,926 of 702,406 patients who did not experience an event (specificity 0.9965, 95% CI 0.9963, 0.9966). Few patients with high ORS experienced the event (PPV 0.0694, 95% CI 0.0598, 0.0791). The model had excellent discrimination (c-statistic = 0.815) but was poorly calibrated, underestimating risk for patients who experienced the outcomes.

CONCLUSIONS: Epic’s ORS demonstrated excellent discrimination but very low sensitivity across three large integrated health systems. Health systems should exercise caution before implementing vendor risk prediction models without validating their use in their patient populations.

CLINICAL TRIAL NUMBER: Not applicable.

PMID:41703383 | DOI:10.1007/s11606-026-10257-1

Neuropsychopharmacol Rep. 2026 Mar;46(1):e70098. doi: 10.1002/npr2.70098.

ABSTRACT

M. K. Trivedi, A. Branton, D. Trivedi, S. Mondal and S. Jana, “Assessment of Cognitive-Motor Functions in Adults With Perceived Neuropsychological Problems Using NIH Toolbox After Remote Biofield Energy Treatment as Non-Pharmacological Intervention: A Randomized Double-Blind Placebo Controlled Trial,” Neuropsychopharmacology Reports 44, no. 4 (2024): 749-761, https://doi.org/10.1002/npr2.12482. The above article, published online on 13 September 2024 in Wiley Online Library (wileyonlinelibrary.com), has been retracted by agreement between the journal Editor-in-Chief, Tsuyoshi Miyakawa; the Japanese Society of Neuropsychopharmacology; and John Wiley & Sons Australia, Ltd. The retraction has been agreed upon as the study’s design, methods, results and conclusions are essentially the same as another article published elsewhere by the same author group in the same year, without any attribution to that article. Furthermore, the study contains physiologically implausible data and statistical anomalies. The editors consider the results and conclusions of this article to be invalid. The authors do not agree with the retraction.

PMID:41703379 | DOI:10.1002/npr2.70098

Psychon Bull Rev. 2026 Feb 17;33(3):68. doi: 10.3758/s13423-025-02821-3.

ABSTRACT

Threat detection is compromised across the schizophrenia spectrum, often revealed by paranoia and delusions. Threat difficulties extend to nonclinical populations with liability toward schizophrenia. A key source of these difficulties may be due to hyper-sensitivity to social stressors in real-world environments. In a large, nonclinical sample (N = 161), we measured the influence of social context to threat detection in social interactions. Social interactions were captured in naturalistic videos and validated as threatening or nonthreatening. Deep learning models were employed to re-render the videos by parsing different amounts of social context depicted in these interactions. Then, we measured how threat detection was influenced by individual variability in schizotypal and autistic traits as a function of social context. Individuals with high schizotypal traits showed reduced threat discrimination ability in the presence of more social context, but better threat detection when the interactions were primarily reduced to body kinematics. The effect was more pronounced in individuals higher on suspicious tendencies and odd belief traits in schizotypy, and social communication traits in the autism spectrum. These results suggest that disruptions from social context may underlie threat detection difficulties across the schizophrenia spectrum.

PMID:41703359 | DOI:10.3758/s13423-025-02821-3

Ann Biomed Eng. 2026 Feb 17. doi: 10.1007/s10439-026-04022-y. Online ahead of print.

ABSTRACT

AI is widely recognized as a tool that biomedical scientists, engineers and clinicians can, and should, use. However, what do we mean by a tool? I take the example of convolutional neural networks that learn latent statistical associations from images, but those associations can be used to different ends. I focus on two different uses in the field of medical diagnostics, what I call human-AI “outskilling” and human-AI “newskilling”. Outskilling is a prosthetic human-AI activity to outperform human capacities (in Greek: prosthesis, adding) in tasks that experts can nevertheless perform well. I study computer-aided diagnostics (CADx) to detect polyps as an example of AI outskilling, which carries the risk of deskilling without a proven gain in meaningful outcomes. I term the second use “newskilling,” a human-AI activity that brings forth something new (in Greek: poiesis) by using latent statistical associations to discover variables that human inference cannot detect. I study the example of AI deriving clinically relevant variables from retinal fundus images to derive “retinal age gaps” as an example of human-AI newskilling. There are two major conclusions based on this distinction: the design of AI uses, and the discernment of how and when to use them.

PMID:41703356 | DOI:10.1007/s10439-026-04022-y

Inflamm Res. 2026 Feb 17;75(1):37. doi: 10.1007/s00011-025-02137-x.

ABSTRACT

BACKGROUND: Acute kidney injury (AKI), characterized by rapid renal dysfunction and high mortality, is critically driven by ferroptosis, an iron-dependent form of cell death. While PTEN-induced kinase 1 (PINK1) and sirtuin 3 (SIRT3) are implicated in mitochondrial homeostasis and ferroptosis regulation, their mechanistic interplay in AKI remains unclear. This study investigated the role of emodin, a natural anthraquinone, in alleviating AKI via SIRT3-mediated PINK1 deacetylation and ferroptosis suppression, focusing on mitochondrial integrity, transferrin (TF) interaction, and redox balance.

MATERIALS AND METHODS: Male C57BL/6 mice (n = 6/group), PINK1⁻/⁻, and SIRT3⁻/⁻ mice were pretreated with emodin (40-160 mg/kg, 3 days) before LPS-induced AKI (15 mg/kg). Human renal tubular HK-2 cells were treated with emodin (10-40 µg/ml) and Erastin (0.4 µM, 24 h). Assays included RNA sequencing, immunoprecipitation-mass spectrometry (IP-MS), histopathology (H&E/PAS/PB-DAB staining), ROS/Fe²⁺/GSH quantification, and immunoblotting. Statistical analysis used ANOVA and Student’s t-test.

RESULTS: Emodin reduced serum creatinine and urea in AKI mice, alongside decreased tubular injury and apoptosis. RNA-seq identified ferroptosis as the central pathway, with emodin upregulating PINK1 expression. IP-MS revealed emodin disrupted PINK1-TF binding via SIRT3-mediated deacetylation, reducing Fe²⁺ accumulation and restoring GPX4 levels. In SIRT3⁻/⁻ and PINK1⁻/⁻ models, emodin’s protective effects were abolished, confirming pathway dependency.

CONCLUSION: Emodin mitigates AKI by activating the SIRT3/PINK1 axis, suppressing ferroptosis through cytoplasmic PINK1 deacetylation and TF interaction disruption. These findings highlight SIRT3/PINK1 as a therapeutic target and emodin as a potential agent for AKI management.

PMID:41703345 | DOI:10.1007/s00011-025-02137-x