Formulating Prior Information for Clinical Trials
The Centre for Bayesian Statistics in Health Economics at the
University of Sheffield invites applications for a 3-year Research
Associate position. The successful applicant will take part in a
project to study the formulation of prior knowledge for use in
Bayesian statistical analysis of clinical trials. Starting salary
will be on the RA1A or RA1B scale as appropriate, in the range £17,451
to £21,290.
The successful candidate will have an MSc or PhD in
Statistics. For an appointee with an MSc the project will offer the
opportunity to obtain a PhD. The following additional knowledge or
experience will be an advantage but are not essential. · Knowledge of
Bayesian statistics · Experience in medical statistics and the
analysis of clinical trials · Knowledge of economics and/or the
economic evaluation of medicines. More important than any such
additional knowledge or experience is intellectual ability and a
keenness to become involved in a multidisciplinary project of very
substantial importance to the pharmaceutical industry and health
funding agencies. Closing date for applications: Monday, 3rd
December, 2001.
Contact the Centre for application details.
Email [log in to unmask] Telephone +44 (0)114 222 3754.
CHEBS, Department of Probability and Statistics, The University of
Sheffield, Sheffield S3 7RH.
The Project
==========
The process of development of new drugs passes through a number of
well-defined phases of experimentation. In Phases 1 and 2, drugs are
tried out on volunteers to test whether they appear to be effective
and to establish safe doses. Experiments in Phases 1 and 2 are
usually relatively small. In Phase 3, however, larger trials are
generally carried out under controlled and randomised conditions to
provide clear statistical evidence of efficacy and safety. There is
an increasing trend also to use Phase 3 trials to examine
cost-effectiveness, by gathering information on utilisation of a
range of medical resources as well as on efficacy, the latter often
being measured by a variety of endpoints, not all of them clinical.
Phase 3 experimentation is expensive and time-consuming. Bayesian
statistical methods offer the possibility of saving both cost and
time by reducing the size of samples required. This can be achieved
if substantive prior information can be identified to supplement the
information gained in the trial data. In principle, such information
exists in the experiences gained in Phase 1 and 2 experiments, and
also in the expert knowledge of pharmacologists who understand the
drug's mode of action. However, a key purpose of Phase 3 trials is
to provide objective evidence to regulatory agencies who will approve
the drug for sale and use, and so any additional prior information
will have to be rigorously justified.
This project will explore the formulation of prior information to
supplement Phase 3 trials. Techniques to elicit expert knowledge in
reproducible and defensible ways will be considered, as well as ways
to ensure that potential biases in the use of Phase 1 and/or 2 data
can be eliminated. The successful applicant will work with leading
researchers in Bayesian statistics, health economics and medical
statistics, and with participating pharmaceutical companies.
The Centre for Bayesian Statistics in Health Economics
==================================
The Centre has recently been formed as a joint research initiative of
the Department of Probability and Statistics and the School of Health
and Related Research (ScHARR) at the University of Sheffield. The
Department of Probability and Statistics is one of the leading
research centres for Bayesian Statistics in the UK, and has a
long-established reputation for innovative research. The School of
Health and Related Research, particularly through the Rapid Reviews
and Sheffield Health Economics Groups, is one of the leading centres
for economic evaluation of medicines and medical interventions. It
is one of only six units recognised by the National Institute for
Clinical Excellence (NICE) for conducting economic appraisals.
The Centre was formed to combine these two key areas of expertise,
and so to promote the application of Bayesian Statistics in Health
Economics. The Director is Professor Tony O'Hagan (Department of
Probability and Statistics). Other senior personnel include
Professor Ron Akehurst, Chris McCabe and Jim Chilcott (ScHARR).
----
Professor A O'Hagan Email: [log in to unmask]
Department of Probability and Statistics
University of Sheffield Phone: +44 114 222 3773
Hicks Building
Sheffield S3 7RH, UK Fax: +44 114 222 3759
----------- http://www.shef.ac.uk/~st1ao/ ------------
|