Research programme
Research programme (long-term projects)
The stability of the Group and its interests has meant that we have been able to invest time and resources in projects that have a long time pay-off. For example, in 1991 we embarked on a long-term evaluation of the picture archiving and communication systems (PACS) project at Hammersmith Hospital. This has now resulted in a stream of high quality publications, presenting the substantive results from the different facets of the evaluation.
Other examples of commitments to long-term projects deserve particular note. The first is an evaluation of the cost-effectiveness of the six English liver-transplant programmes. This work began in 1996 and has involved developing a detailed database on a cohort of patients from the six centres including information on clinical prognostic indicators, survival, quality of life and detailed resource usage. The database is now complete and initial reports are being produced for the Department of Health as research customer and peer-reviewed publication.
A further long-term example is our work as part of the MRC funded Multicentre Aneurysm Screening Study: a five-year follow-up study of screening men aged 65 or over for abdominal aortic aneurysms. Professor Buxton is a co-applicant in the study and HERG is responsible for the economic evaluation.
Key clinical areas
A long standing theme of work within HERG has related to the evaluation of screening programmes, reflecting both that they have been matters for national/central policy rather than simply local clinical judgement and that their discretionary nature makes them particularly susceptible to economic analysis. Research on breast cancer screening has been funded by the MRC to provide the necessary economic evaluation component to the MRC funded breast screening ‘frequency’ and ‘age’ trials. In parallel, research was undertaken by Brown and Bryan on different breast cancer screening modalities: the cost-effectiveness of two-view versus one-view mammography and of double-reading versus single-reading of mammograms.
Within the research on screening, our active collaboration with the Institute of Child Health (ICH), London, has seen the completion of a cost-effectiveness analysis of antenatal HIV testing in the UK. This project used decision analytic modelling techniques to assess the cost effectiveness of alternative strategies for antenatal HIV testing and the management of infected women and babies.
A final example of the main clinical areas in which we have worked is that of cardiac interventions. Building on our much earlier path-breaking study of the costs and benefits of the heart transplant programmes at Harefield and Papworth Hospitals, during this period of assessment we have worked on a wide range of major cardiac studies. In primary prevention Buxton and Wonderling undertook the economic evaluation of the British Family Heart Study. In secondary care, much of the work has related to alternative methods for revascularisation and treatment of angina. HERG has been integrally involved in the RITA trials (Randomised Intervention Treatment of Angina) since their inception and with trials at Papworth Hospital of transmyocardial laser revascularisation and most recently percutaneous myocardial laser revascularisation. Work on revascularisation techniques emphasises the need to understand the trade-offs patients are willing to make between the various differences that exist between therapies in terms of short-term and long-term mortality, morbidity/symptom relief and the process of delivery of care (hospitalisation and post-surgical ‘return-to–work’ time). We currently have research underway to explore these issues further.
Methodological themes
One of the themes running through this applied work across a variety of otherwise disparate clinical applications is that of making and testing methodological developments to improve our understanding of patient preferences. We have, for example, developed substantial experience in using conjoint analysis and have published a number of high quality contributions to the limited existing literature on health applications of this technique. From this body of research experience we are beginning to be able to demonstrate how valid and reliable are applications of the technique, to refine methods of analysis, and to show how this approach can best be used. In a similar manner we are concerned with improving techniques for economic modelling. Thus, in addition to our experience in using Markov chain models and Monte Carlo simulations (for example in modelling the costs and outcomes of heart transplant programme), we are exploring the use of discrete event simulation (DES) modelling. Despite its greater demands on modelling skills and computational resources, DES appears to offer significant advantages in terms of the ease with which it can directly utilise available data in a variety of forms, as compared with the more restrictive data requirements of Markov models. Other ongoing research led by Kennedy is evaluating the use of decision-aids to help reflect individual patient preferences in clinical decision-making.
Analysis of ’payback’ from research
A steadily expanding stream of research on the ‘payback’ from research itself is an extension to our focus on economic evaluation. The work, begun for the national Department of Health (DH) in 1993, has been further developed principally to advise the DH and NHS R&D policies. The first main journal publication (Buxton and Hanney, 1996) described the HERG payback framework that consists of a multidimensional categorization of benefits and a model of how to assess them.This pioneering work led to Professor Buxton accepting an invitation to spend three months in Canada looking at whether the approach could be adopted and further developed in the context of the portfolio of applied and more basic clinical research funded by the Alberta Heritage Foundation for Medical Research (AHFMR). This opportunity clearly demonstrated the international applicability and importance of the approach in the context of applied Health Services Research. It also indicated possible lines of research to better understand the process by which basic science research is fed through to application, and this was piloted in a study, in collaboration with the Policy Unit of the Wellcome Trust (see HERG research Report 31). The full range of approaches to the payback analysis was applied and further developed in a project for the Arthritis Research Campaign examining the long-term benefits of the research that it funds (Hanney et al. 2004, Wooding et al. 2004). This project was led by Dr Jonathan Grant formerly Head of Policy at the Wellcome Trust and now a Director of RAND Europe.
A further aspect of the work on the measurement of payback explored how the approaches might be used for routine monitoring of research impact. This led to our organising a workshop for the DH at Leeds Castle (November 1999, see Croxson et al, 2001). This was co-supported by Wellcome Trust, introduced by Sir John Pattison (DH Director of R&D), and had senior representation from MRC and ESRC and international inputs from the US Agency for Healthcare Research and Quality and AHFMR.
One major aspect of any routine monitoring system would be the collection of data about the first element of the multidimensional categorization: knowledge production. The Research Outputs Database (ROD) developed by the Policy Unit of the Wellcome Trust records bibliometric and funding data about all biomedical articles with at least one UK address that appear on the Science Citation Indices. The London Region of the NHS funded the NHS ROD Fellow at HERG to work on the NHS portion of the database. One of the main areas on which the Research Fellow, Teri Jones, has focused is to identify the journals in which NHS-funded research is published and the journals which clinicians regard as being important for their clinical practice. This is then compared with more tradition ways of assessing the impact of different journals (Jones et al., 2004).
HERG’s stream of payback work is increasingly seen as relevant in a range of national and international circumstances. It was used when HERG was commissioned to undertake the evaluation of the NHS R&D Implementation Methods Programme. It was identified as providing a model for assessing the impact of research in the 2003 report from the National Coordinating Centre for Health Technology Assessment and this has led to involvement in a project to assess the impact of the HTA Programme. Finally Steve Hanney and Martin Buxton have been consultants to the World Health Organisation’s Health Research Systems Analysis initiative. This included: working on the conceptual framework for the whole initiative (Pang et al, 2003); producing a background paper on the impact of research on health policymaking (Hanney et al, 2003); and being commissioned to produce a paper on the economic benefits from health research for a special edition of the Bulletin of the World Health Organisation focusing on the initiaive (Buxton et al. 2004).
