A5002389217- Health Systems, Economic Evaluations, Quality of Life
- Statistical Methods and Inference
- Global Cancer Incidence and Screening
- Colorectal Cancer Screening and Detection
- Advanced Causal Inference Techniques
- Insurance, Mortality, Demography, Risk Management
- Statistical Methods and Bayesian Inference
- Statistical Methods in Clinical Trials
- demographic modeling and climate adaptation
- Genetic factors in colorectal cancer
- Meta-analysis and systematic reviews
- Cancer Risks and Factors
- Global Health Care Issues
- Pregnancy and preeclampsia studies
- Economic and Financial Impacts of Cancer
- Multiple and Secondary Primary Cancers
- Birth, Development, and Health
- Chemotherapy-induced cardiotoxicity and mitigation
- Asthma and respiratory diseases
- Acute Lymphoblastic Leukemia research
- Probability and Statistical Research
- Gastric Cancer Management and Outcomes
- Healthcare Policy and Management
- Acute Myocardial Infarction Research
- Genetic Associations and Epidemiology
Karolinska Institutet
2016-2025
University of Leicester
2016-2025
Norwegian Institute of Public Health
2024-2025
Cancer Registry of Norway
2018-2025
Duke University
2023
University of Alabama at Birmingham
2023
Glenfield Hospital
2023
NIHR Leicester Biomedical Research Centre
2023
Institut polytechnique de Grenoble
2023
Paris Cardiovascular Research Center
2023
To compare the performance of different meta-analysis methods for pooling odds ratios when applied to sparse event data with emphasis on use continuity corrections.Meta-analysis side effects from RCTs or risk factors rare diseases in epidemiological studies frequently requires synthesis rates. Combining such can be problematic zero events exist one both arms a study as corrections are often needed, but, these influence results and conclusions.A simulation was undertaken comparing several...
The use of individual participant data instead aggregate in meta-analyses has many potential advantages, both statistically and clinically. <b>Richard D Riley colleagues</b> describe the rationale for an meta-analysis outline how to conduct this type study
A dramatic improvement in the survival of patients with chronic myeloid leukemia (CML) occurred after introduction imatinib mesylate, first tyrosine kinase inhibitor (TKI). We assessed how these changes affected life expectancy CML and life-years lost as a result between 1973 2013 Sweden.
Royston and Parmar (2002, Statistics in Medicine 21: 2175–2197) developed a class of flexible parametric survival models that were programmed Stata with the stpm command (Royston, 2001, Journal 1: 1–28). In this article, we introduce new command, stpm2, extends methodology. New features for stpm2 include improvement way time-dependent covariates are modeled, these effects far less likely to be over parameterized; ability incorporate expected mortality thus fit relative models; superior...
There has been a recent growth in the use of Bayesian methods medical research. The main reasons for this are development computer intensive simulation based such as Markov chain Monte Carlo (MCMC), increases computing power and introduction powerful software WinBUGS. This enabled increasingly complex models to be fitted. ability fit these led MCMC being used convenient tool by frequentists, who may have no desire fully Bayesian. Often researchers want ‘the data dominate’ when there is prior...
<h3>Context</h3>Mortality among human immunodeficiency virus (HIV)–infected individuals has decreased dramatically in countries with good access to treatment and may now be close mortality the general uninfected population.<h3>Objective</h3>To evaluate changes gap between HIV-infected population.<h3>Design, Setting, Population</h3>Mortality following HIV seroconversion a large multinational collaboration of seroconverter cohorts (CASCADE) was compared expected mortality, calculated by...
Attacks of wheezing induced by upper respiratory viral infections are common in preschool children between the ages 10 months and 6 years. A short course oral prednisolone is widely used to treat with who present a hospital, but there conflicting evidence regarding its efficacy this age group.We conducted randomized, double-blind, placebo-controlled trial comparing 5-day (10 mg once day for 24 20 older children) placebo 700 60 months. The presented three hospitals England an attack...
<b>Objective</b> To compare four potential screening strategies, and subsequent interventions, for the prevention treatment of type 2 diabetes: (<i>a</i>) diabetes to enable early detection treatment, (<i>b</i>) impaired glucose tolerance, intervening with lifestyle interventions in those a diagnosis tolerance delay or prevent diabetes, (<i>c</i>) as but pharmacological (<i>d</i>) no screening. <b>Design</b> Cost effectiveness analysis based on development evaluation probabilistic,...
Abstract Meta‐analysis of individual patient data (IPD) is the gold‐standard for synthesizing evidence across clinical studies. However, some studies IPD may not be available and only aggregate (AD), such as a treatment effect estimate its standard error, obtained. In this situation, methods combining AD are important to utilize all evidence. paper, we develop assess range statistical in meta‐analysis continuous outcomes from randomized controlled trials. The take either one‐step or two‐step...
In population-based cancer studies, cure is said to occur when the mortality (hazard) rate in diseased group of individuals returns same level as that expected general population. The fraction (the proportion patients cured disease) interest and a useful measure monitor trends survival curable disease. There are 2 main types model, mixture model non-mixture with most previous work concentrating on model. this paper, we extend parametric incorporate background mortality, thus providing...
Abstract Relative survival is frequently used in population‐based studies as a method for estimating disease‐related mortality without the need information on cause of death. We propose an extension to relative flexible parametric model proposed by Royston and Parmar censored data. The provides smooth estimates excess rates using restricted cubic splines log cumulative hazard scale. approach has several advantages over some more standard models, which adopt piecewise approach, main being...
When multiple endpoints are of interest in evidence synthesis, a multivariate meta-analysis can jointly synthesise those and utilise their correlation. A random-effects must incorporate estimate the between-study correlation (ρ B ). In this paper we assess maximum likelihood estimation general normal model generalised for bivariate (BRMA). We consider two applied examples, one involving diagnostic marker other surrogate outcome. These motivate simulation study where properties from BRMA...
<b>Objective</b> To determine whether detection of invasive cervical cancer by screening results in better prognosis or merely increases the lead time until death. <b>Design</b> Nationwide population based cohort study. <b>Setting</b> Sweden. <b>Participants</b> All 1230 women with diagnosed during 1999-2001 Sweden prospectively followed up for an average 8.5 years. <b>Main outcome measures</b> Cure proportions and five year relative survival ratios, stratified history, mode detection, age,...
Summary The 5‐year relative survival of adults diagnosed with acute myeloid leukaemia ( AML ) was less than 10% during the 1970s and 1980s in E ngland. This population‐based study estimated ‘cure’ for 48 380 adult patients ngland 1971–2006. Relative cure mixture models were used to produce estimates percentage ‘cured’. ‘Cure’ defined as proportion a group survivors whom there is no excess mortality compared general population. ‘cured’ increased aged under 70 years at diagnosis 1971–2006, but...
Often multiple outcomes are of interest in each study identified by a systematic review, and this situation separate univariate meta-analysis is usually applied to synthesize the evidence for outcome independently; an alternative approach single multivariate model that utilizes any correlation between obtains all pooled estimates jointly. Surprisingly, rarely considered practice, so paper we illustrate benefits limitations provide helpful insight practitioners. We compare bivariate...
Competing risks are a common occurrence in survival analysis. They arise when patient is at risk of more than one mutually exclusive event, such as death from different causes, and the these may prevent any other event ever happening. There two main approaches to modelling competing risks: first model cause-specific hazards transform cumulative incidence function; second directly on transformation function. We focus approach this paper. This paper advocates use flexible parametric framework....
Simulation studies are conducted to assess the performance of current and novel statistical models in pre‐defined scenarios. It is often desirable that chosen simulation scenarios accurately reflect a biologically plausible underlying distribution. This particularly important framework survival analysis, where simulated distributions for both event time censoring time. paper develops methods using complex when generating times practice. We describe general algorithm involving numerical...