Royal Statistical Society Medical Section 
Half day meeting on Health economics, evidence synthesis and decision science in health 
care resource allocation

Date and Time: November 28th 2006 at 2pm
Venue: The Royal Statistical Society, 12 Errol Street, London

1. Simon Thompson, MRC Biostatistics Unit, Cambridge
Aortic aneurysms: from randomised trials to a national screening programme
Randomised trials have shown that ultrasound screening for abdominal aortic aneurysms 
(AAA) is effective in reducing mortality.  However, evidence on long-term cost-effectiveness 
comes principally from decision models.  I will discuss some of the issues involved in building 
a reliable model, including parameter estimation and uncertainty, structural uncertainty, and 
model validation.  This research has informed the recent decision to initiate a national AAA 
screening programme in the UK.

2. Nicky Welton and Tony Ades, Department of Social Medicine, University of Bristol
Expected Value of Information Calculations in Practice: Interventions to Increase Uptake of 
Breast Cancer Screening

As interest has increased in the use of formal Bayesian methods in Cost-Effectiveness 
Analysis, there has also been a growing interest in Expected Value of Information (EVI) to 
prioritise and design research aimed at reducing uncertainty in decision models. In this talk 
we illustrate some of the practical issues to consider when carrying out such an analysis. We 
develop a simple cost-effectiveness analysis of breast cancer screening as a typical case 
study, and use data from a 1999 cluster randomised 2x2 factorial trial of interventions to 
increase uptake. An EVI analysis is developed which shows that, based on the evidence 
available beforehand, the trial was cost-effective, but that after incorporating the results of the 
trial it would still be cost-effective to carry out research that further reduced decision 
uncertainty. We identify key conceptual and technical issues: the relation between the target 
interventions and the previous evidence, the distinction between variation and uncertainty, 
and methods for correlated parameters. We discuss the advantages of using EVI methods 
over current methods of research prioritisation, but also consider some potential limitations to 
their use in practice.

3. Norman Waugh, Department of Public Health, University of Aberdeen

Health technology assessment for UK policy makers

Health technology assessment poses three main questions;
does it work?
at what cost?
is it worth it?
The UK HTA Programme has supported the work of NICE and others, by commissioning 
technology assessment reports (TARs) from seven academic units around the UK. These 
provide the NICE Technology Appraisal Committee with an independent assessment of the 
evidence. The manufacturer provides their view of the evidence and their own economic 
model. Clinical and patient groups also provide submissions. This presentation will provide 
an overview of the process and its problems, mainly from the perspective of someone who is 
a producer of TARs, for NICE and other UK bodies. 


Dr Stephen Walters
Senior Lecturer in Medical Statistics
Medical Statistics Group
ScHARR, University of Sheffield                 
Regent Court, 30 Regent St, Sheffield S1 4DA
PHONE 0114 2220730 
FAX   0114 2724095