A5081409332- Meta-analysis and systematic reviews
- Health Systems, Economic Evaluations, Quality of Life
- Lung Cancer Diagnosis and Treatment
- Colorectal Cancer Treatments and Studies
- Statistical Methods in Clinical Trials
- Acute Myocardial Infarction Research
- Economic and Environmental Valuation
- Evaluation and Performance Assessment
- Diabetes Treatment and Management
- Delphi Technique in Research
- Diabetes and associated disorders
- Radiomics and Machine Learning in Medical Imaging
- Colorectal Cancer Screening and Detection
- Head and Neck Cancer Studies
- Diabetes Management and Research
- Reliability and Agreement in Measurement
- Eosinophilic Esophagitis
- Food Allergy and Anaphylaxis Research
- COVID-19 Clinical Research Studies
- Health Policy Implementation Science
- Animal testing and alternatives
- Acute Lymphoblastic Leukemia research
- Lung Cancer Treatments and Mutations
- Healthcare cost, quality, practices
- COVID-19 diagnosis using AI
University of Exeter
2015-2024
AstraZeneca (United Kingdom)
2024
Queen Mary University of London
2023
University of Leeds
2023
NHS Blood and Transplant
2023
Peninsula College of Medicine and Dentistry
2011-2020
Pearson (United Kingdom)
2014
Queensland University of Technology
2008-2014
University of Sheffield
2013
University of Bremen
2013
Egger's regression test is often used to help detect publication bias in meta-analyses. However, the performance of this and usual funnel plot have been challenged particularly when summary estimate natural log odds ratio (lnOR).To compare with a based on sample size (a modification Macaskill's test) lnOR as estimate.Simulation meta-analyses under number scenarios presence absence between-study heterogeneity.Type I error rates (the proportion false-positive results) for each their power it...
Abstract The trim and fill method allows estimation of an adjusted meta‐analysis estimate in the presence publication bias. To date, performance has had little assessment. In this paper, we provide a more comprehensive examination different versions number simulated scenarios, comparing results with those from usual unadjusted models two simple alternatives, namely use from: (i) largest; or (ii) most precise study meta‐analysis. Findings suggest great deal variability approaches. When there...
In meta-analysis, the presence of funnel plot asymmetry is attributed to publication or other small-study effects, which causes larger effects be observed in smaller studies. This issue potentially mean inappropriate conclusions are drawn from a meta-analysis. If meta-analysis used inform decision-making, reliable way adjust pooled estimates for potential required. A comprehensive simulation study presented assess performance different adjustment methods including novel application several...
Funnel plots are commonly used to investigate publication and related biases in meta-analysis. Although asymmetry the appearance of a funnel plot is often interpreted as being caused by bias, reality could be due other factors that cause systematic differences results large small studies, for example, confounding such differential study quality. can enhanced adding contours statistical significance aid interpreting plot. If studies appear missing areas low significance, then it possible...
Summary Between-study heterogeneity and publication bias are common features of a meta-analysis that can be present simultaneously. When both suspected, consideration must made each in the assessment other. We consider extended funnel plot tests for detecting bias, selection modelling trim-and-fill methods to adjust presence between-study heterogeneity. These applied two example data sets. Results indicate ignoring when assessing misleading, but test or may not powerful is large. It...
OBJECTIVE Monogenic diabetes, a young-onset form of is often misdiagnosed as type 1 resulting in unnecessary treatment with insulin. A screening approach for monogenic diabetes needed to accurately select suitable patients expensive diagnostic genetic testing. We used C-peptide and islet autoantibodies, highly sensitive specific biomarkers discriminating from non–type biomarker pathway diabetes. RESEARCH DESIGN AND METHODS studied diagnosed at age 30 years or younger, currently younger than...
When data needed to inform parameters in decision models are lacking, formal elicitation of expert judgement can be used characterise parameter uncertainty. Although numerous methods for eliciting opinion as probability distributions exist, there is little research suggest whether one method more useful than any other method. This study had three objectives: (i) obtain subjective characterising uncertainty the context a health technology assessment; (ii) compare two by same different ways;...