A5077866593- Statistical Methods in Clinical Trials
- Statistical Methods and Bayesian Inference
- Advanced Causal Inference Techniques
- Health Systems, Economic Evaluations, Quality of Life
- Meta-analysis and systematic reviews
- Statistical Methods and Inference
- Probability and Statistical Research
- Data Analysis with R
- Global Peace and Security Dynamics
- Adolescent Sexual and Reproductive Health
- Mental Health Research Topics
- HIV/AIDS Research and Interventions
- Bayesian Methods and Mixture Models
- Advanced Statistical Modeling Techniques
- Musculoskeletal pain and rehabilitation
- demographic modeling and climate adaptation
- Primary Care and Health Outcomes
- Reproductive tract infections research
- Migration, Health and Trauma
- Spine and Intervertebral Disc Pathology
- Lymphoma Diagnosis and Treatment
- Health, Environment, Cognitive Aging
- Advanced Statistical Methods and Models
- Hydropower, Displacement, Environmental Impact
- Historical and modern epidemiology studies
University College London
2015-2024
MRC Clinical Trials Unit at UCL
2015-2024
LMU Klinikum
2024
Ludwig-Maximilians-Universität München
2024
London School of Hygiene & Tropical Medicine
2015-2023
Medical Research Council
2009-2023
Heidelberg (Poland)
2021-2022
University of Hohenheim
2021-2022
UC San Diego Health System
2021
Sharp HealthCare Foundation
2021
Simulation studies are computer experiments that involve creating data by pseudorandom sampling. The key strength of simulation is the ability to understand behaviour statistical methods because some 'truth' (usually parameter/s interest) known from process generating data. This allows us consider properties methods, such as bias. While widely used, often poorly designed, analysed and reported. tutorial outlines rationale for using offers guidance design, execution, analysis, reporting...
Ventilation-perfusion (V(dot)Q(dot) lung scanning and computed tomographic pulmonary angiography (CTPA) are widely used imaging procedures for the evaluation of patients with suspected embolism. has been largely replaced by CTPA in many centers despite limited comparative formal evaluations concerns about CTPA's low sensitivity (ie, chance missing clinically important embuli).To determine whether may be relied upon as a safe alternative to V(dot)Q(dot initial procedure excluding diagnosis...
Multiple imputation is a commonly used method for handling incomplete covariates as it can provide valid inference when data are missing at random. This depends on being able to correctly specify the parametric model impute values, which may be difficult in many realistic settings. Imputation by predictive mean matching (PMM) borrows an observed value from donor with similar mean; local residual draws (LRD) instead donor's residual. Both methods relax some assumptions of imputation,...
Adjustment for prognostic covariates can lead to increased power in the analysis of randomized trials. However, adjusted analyses are not often performed practice.We used simulation examine impact covariate adjustment on 12 outcomes from 8 studies across a range therapeutic areas. We assessed (1) how large an increase be expected practice; and (2) that prognostic.Adjustment known led increases most outcomes. When was set 80% based unadjusted analysis, median 92.6% (range 80.6 99.4%). Power...
Abstract Many clinical trials restrict randomisation using stratified blocks or minimisation to balance prognostic factors across treatment groups. It is widely acknowledged in the statistical literature that subsequent analysis should reflect design of study, and any stratification variables be adjusted for analysis. However, a review recent general medical showed only 14 41 eligible studies reported adjusting their primary variables. We show balancing groups leads correlation between If...
Identifying which individuals benefit most from particular treatments or other interventions underpins so-called personalised stratified medicine. However, single trials are typically underpowered for exploring whether participant characteristics, such as age disease severity, determine an individual's response to treatment. A meta-analysis of multiple trials, particularly one where individual data (IPD) available, provides greater power investigate interactions between characteristics...
BackgroundThe International Liaison Committee on Resuscitation has called for a randomised trial of delivery to cardiac arrest centre. We aimed assess whether expedited centre compared with current standard care following resuscitated reduces deaths.MethodsARREST is prospective, parallel, multicentre, open-label, superiority trial. Patients (aged ≥18 years) return spontaneous circulation out-of-hospital without ST elevation were randomly assigned (1:1) at the scene their by London Ambulance...
Estimands can be used in studies of healthcare interventions to clarify the interpretation treatment effects. The addendum ICH E9 harmonised guideline on statistical principles for clinical trials (ICH E9(R1)) describes a framework using estimands as part study. This paper provides an overview framework, outlined addendum, with aim explaining why are beneficial; clarifying terminology being used; and providing practical guidance decide appropriate study design, data collection, estimation...
A cross-sectional, observational study of outcomes for neonates with severe neonatal thrombocytopenia (SNT; platelet count <60 x 10(9) platelets per L) was performed to examine hemorrhage and use transfusions.Neonates who were admitted 7 NICUs developed SNT enrolled daily data collection.Among 3652 admissions, 194 (5%) SNT. The median gestational age 169 27 weeks (interquartile range [IQR]: 24-32 weeks), the birth weight 822 g (IQR: 670-1300 g). Platelet nadirs <20 10(9), 20 39 40 59 L 58...
Background: Recent research of lateral elbow tendinopathy has led to the use laboratory-amplified tenocyte-like cells. Hypothesis: Ultrasound-guided injection autologous skin-derived tendon-like cells are effective compared with other injectable therapies for treatment refractory patellar tendinosis. Study Design: Randomized controlled trial; Level evidence, 1. Methods: From 60 tendons in 46 patients tendinopathy, a 4-mm skin biopsy was sampled grow collagen-producing Patients were allocated...
Missing data due to loss follow‐up or intercurrent events are unintended, but unfortunately inevitable in clinical trials. Since the true values of missing never known, it is necessary assess impact untestable and unavoidable assumptions about any unobserved sensitivity analysis. This tutorial provides an overview controlled multiple imputation (MI) techniques a practical guide their use for analysis trials with continuous outcome data. These include δ ‐ reference‐based MI procedures. In...
Background Internet-accessed sexually transmitted infection testing (e-STI testing) is increasingly available as an alternative to in clinics. Typically this modality enables users order a test kit from virtual service (via website or app), collect their own samples, return samples laboratory, and be notified of results by short message (SMS) telephone. e-STI assumed increase access comparison with face-to-face services, but the evidence unclear. We conducted randomised controlled trial...
Aims/Hypothesis: Type 2 diabetes mellitus is associated with high levels of disease burden, including increased mortality risk and significant long-term morbidity. The prevalence differs substantially among ethnic groups. We examined the type diagnoses in UK primary care setting. Methods: analysed data from 404,318 individuals Health Improvement Network database, aged 0–99 years permanently registered general practices London. association between ethnicity 2013 was estimated using a logistic...
In this paper we study approaches for dealing with treatment when developing a clinical prediction model. Analogous to the estimand framework recently proposed by European Medicines Agency trials, propose 'predictimand' of different questions that may be interest predicting risk in relation started after baseline. We provide formal definition estimands matching these questions, give examples settings which each is useful and discuss appropriate estimators including their assumptions....
It has long been advised to account for baseline covariates in the analysis of confirmatory randomised trials, with main statistical justifications being that this increases power and, when a randomisation scheme balanced covariates, permits valid estimate experimental error. There are various methods available but it is not clear how choose among them.
Although new biostatistical methods are published at a very high rate, many of these developments not trustworthy enough to be adopted by the scientific community. We propose framework think about how piece methodological work contributes evidence base for method. Similar well-known phases clinical research in drug development, we define four research. These cover (I) proposing idea while providing, example, logical reasoning or proofs, (II) providing empirical evidence, first narrow target...
Abstract Introduction The massive-transfusion concept was introduced to recognize the dilutional complications resulting from large volumes of packed red blood cells (PRBCs). Definitions massive transfusion vary and lack supporting clinical evidence. Damage-control resuscitation regimens modern trauma care are targeted early correction acute traumatic coagulopathy. aim this study identify a clinically relevant definition based on outcomes. We also examined whether useful in that prediction...
Recent reviews have shown that while clustering is extremely common in individually randomised trials (for example, within centre, therapist, or surgeon), it rarely accounted for the trial analysis. Our aim to develop a general framework assessing whether potential sources of must be analysis obtain valid type I error rates (non-ignorable clustering), with particular focus on trials.A developed based theoretical results and case study recently published used illustrate concepts. A simulation...