Centre for Statistics in Medicine
20 years experience in medical statistics
80+ current trials
Team of medical statisticians, methodologists and systematic review specialists
Using STATISTICs to conduct MEDICAL RESEARCH
Primary medical research
We provide statistical input at all stages of clinical trials and other clinical and preclinical research projects. Our input helps to ensure that researchers ask clear questions that can be answered. We are involved from grant application and design to publication, and collaborate with clinical trial units and researchers across the UK. We are also involved in the NIHR's Research Design Service, offering free pre-submission advice on research proposals.
A few of our current projects are highlighted at the bottom of this page. Please visit our website for a more extensive list of examples of the over 100 medical studies we are currently involved with.
Prognosis is the forecast or estimate of the risk of something happening in the future. In medicine, we are interested in predicting a person's future health state using their current characteristics, such as their weight, history, blood markers, or X-ray results. We may be interested in whether they will develop a particular disease, whether they will have recovered from a current disease, or whether their pain levels will have changed, for example.
Prognostic models are mathematical models that relate a person's characteristics now to the risk of a particular future outcome. Doctors use them to get objective estimates of the probability that something will happen, to use alongside other clinical information. We use clinical data sets to build prognostic models that predict an individual's risk of a certain disease, given that individual's characteristics. We also test previously published prognostic models with new data, which is called external validation.
One of the challenges in medical research is the sheer volume of information and existing studies available. We combine and assess published clinical results in systematic reviews and meta-analyses, helping clinicians to access the most up-to-date knowledge in the literature. We are particularly involved in oncology systematic reviews, through our collaboration with CRUK, and in haematology reviews, through our collaboration with the NHS Blood Trust.
We carry out applied statistical research on the design, conduct, and reporting of medical research. Medical research must be based on sound methodology if it is to be useful, usable, and reproducible. We want our research to be practically relevant for clinicians and medical researchers, helping them to carry out excellent research. Our work ultimately benefits patients and improves patient care.
If you are interested in pursuing a Master's or DPhil project focusing on any of the below areas of medical statistics, please get in touch.
The results of a study are only as good as the analysis plan used. We are particularly interested in:
- The effects of missing data, sample size, and other factors in prognostic modelling
- Using propensity score analysis in observational studies to balance confounders and estimate treatment effects
- Optimising the methodology used in randomised controlled trials
- Optimising the methodology used in systematic reviews
Research studies must be reported fully. If a publication is missing key information, then it cannot be understood by its readers, cannot be reproduced, and cannot be included in the systematic reviews that inform clinical decision-making. Reporting guidelines help researchers to include every necessary detail in their publications.
CSM members are involved with working groups to develop new guidelines for particular kinds of studies. We investigate how well reporting guidelines are used and what actions improve their uptake by researchers and journals.
We also work closely with the EQUATOR Network, a central repository for reporting guidelines that educates the research community on good reporting.
Selected research projects
The DELTA2 project aims to develop guidance for specifying the target difference (or effect size) in the sample size calculation for a randomised clinical trial (RCT). This guidance is aimed at researchers and funders of RCTs. It is hoped that the guidance will help improve the design of future RCTs.
Selected clinical trials
6MP: 6-mercaptopurine (6MP) and low-dose methotrexate in patients with known BRCA-defective tumours
ART: Anal squamous cell carcinoma: Investigation of functional imaging during chemoradiotherapy
CHARIOT: ATR inhibitor and radiotherapy in oesophageal cancer
CORKA: Community-based rehabilitation after knee arthroplasty
IntReALL HR: International study for treatment of high risk childhood relapsed acute lymphoblastic leukemia 2010
LASER: Laminar airflow in severe asthma for exacerbation reduction
PATH2: Platelet-rich plasma in Achilles tendon healing
For more clinical trials, visit OCTRU.
For our full list of primary biomedical research, visit the CSM website.
RSS at Oxford
We coordinate the Oxford chapter of the Royal Statistical Society. We welcome all those with an interest in statistics, academic or non-academic, to join our programme of events. Read more here
Christl Donnelly joins us on Wednesday 26 April 2017 at the Department of Statistics, 4pm-5pm. More information here
Denise Lievesley joins us on Wednesday 28 June 2017 at Green Templeton College, 4pm-5pm. More information here