1.Developments in evaluating biomarkers: stratified medicine and predictive markers
Medical interventions come at a price, no drug is without side-effects, and pharmacotherapy can sometimes be extremely toxic and very costly. There is heterogeneity in the response to treatment, and researchers have tried to identify biomarkers that could in identifying those most likely to respond to and benefit from therapy, enabling others to avoid ineffective and potentially harmful therapy. Predictive molecular tests and biomarkers may be used as treatment selection markers, targeting drugs to subgroup profiles, and bringing stratified medicine closer to everyday clinical practice. Substantiating claims about putative predictive markers requires the development of new methods and modifications in existing approaches to test and biomarker evaluation.
2.Developing, evaluating and applying clinical prediction rules
Numerous articles in the medical literature develop and report a clinical prediction rule, which is also referred to as a prognostic model, prognostic index, or risk score. The aim of a prediction rule is to accurately predict future clinical outcomes for the individual patient, using available clinical information and patient characteristics. A well-known example is the Nottingham Index, which is used to predict mortality risk in patients with breast cancer based on factors such as tumour size and grade. A reliable prediction rule can inform individualised treatment strategies and aid patient-counselling. However, prediction rules have often been criticised as their accuracy is often not reliably validated, and very few find a role within clinical practice. In this session we invite talks on how more reliable and clinically useful prediction rules can be developed and validated, and look to stimulate debate on when prediction rules are suitable for application in clinical practice.
3.Evaluating tests for monitoring and surveillance
Monitoring and surveillance can be defined as the periodic measurement that guides the management of a chronic or recurrent condition. It forms a major part of the management of long term illnesses, which is an important and costly element of health care. Despite the substantial impact and cost it entails, monitoring has been neglected as an area for research. Moreover, in many cases chronic disease remains poorly controlled despite the investment in monitoring. For example, in a UK study, only 14% of 21,024 newly diagnosed patients with hypertension had met target blood pressure after 12 months. While poor monitoring may be an expensive waste; good monitoring can clearly improve patient outcomes. For example, a recent systematic review of Warfarin self-monitoring has shown a one third reduction in mortality with no increase in haemorrhage rates. This session aims to presenting the latest research carried out in monitoring and surveillance as well as exploring the latest most relevant questions in the field.
4.Systematic reviews Guideline development, and practice recommendations related to choice and use of tests
Clinical guidelines often aim to contain recommendations about the use of both diagnostic tests and healthcare interventions. The framework and methods for systematically reviewing the literature and producing guidelines on the use of interventions is well established and evaluated. However, those producing guidelines on the use of tests often encounter problems including (amongst others) challenges arising from the complex nature of test technologies, the lack of adequate primary studies, and problems in applying evidence hierarchies to test evaluations . This session will review these problems focusing on experience gained by the NICE Clinical Guidelines development programme, and look at recent developments.
5.Economic evaluations of tests and the use of modelling
Like other health care interventions, particularly new drug therapies, cost-effectiveness is increasingly being suggested as the metric by which population-level decisions on provision and reimbursement of tests should be made. Modelling is a key tool in providing evidence on cost-effectiveness. Indeed for tests, modelling also has an important role in establishing effectiveness too, particularly where an assessment of impact on patient outcomes is sought. Early experience suggests there are clear challenges requiring innovative responses and further methodological work. However, given the recognised problems associated with alternatives particularly direct empirical observation of effects on patient outcome, modelling is an approach to test evaluation which must continue to be pursued. In this spirit, for those who are unfamiliar with health economics and modelling, this session seeks to introduce how modelling approaches can assist in test evaluation and will high-light the challenges and potential solutions, including areas for methodological research.
6.Industry and regulatory perspectives on the development and evaluation of diagnostic tests
This session will focus on the evaluation of test technologies within industry, with illustrations from a keynote speaker leading a development team for new products. We also are hoping to have discussions of the perspectives of a regulatory authority on their approach to approval of test technologies. We would welcome submissions of abstracts looking at challenges and approaches of industry to test evaluation.
7.Design and analysis of primary studies evaluating tests
We are welcoming abstracts which advance our thinking and practice looking at the analysis and design of primary studies of tests, whether evaluating test accuracy or any other aspect of test performance.