Decisions turn on preferences and probabilities, yet the human mind is notoriously bad at computing probabilities. The mind is prone to error and the scientific cannon is there for a very good reason; to mitigate errors resulting from inaccuracy or imprecision.
Professor Richard Lilford, Professor of Public Health and Director of the Warwick Centre for Applied Health Research and Delivery, University of Warwick and an expert in a wide range of areas of health related research, from developing clinical trials and research ethics to assessing the cost effectiveness of treatments for the health service.
The Collaboration for Applied Health Research and Delivery(CAHRD) is led by LSTM in partnership with Warwick and a host of overseas collaborators. One of the themes explored by CAHRD is health economics research, contributing improved effectiveness, efficiency and equity within health systems. At this seminar presented by Professor Lilford to an audience of staff and students at LSTM, mitigations for inaccuracy in models used to measure the effectiveness of health service delivery were explored.
Research by Professor Lilford and collaborators into the evaluation of service delivery interventions to enhance patient safety in the NHS, has involved addressing concerns about the quality of existing health services research. The choice of study design used by researchers is particularly important when evaluating these interventions. Professor Lilford favours a controlled before and after design (such as the stepped wedge design) combined whenever possible with qualitative and quantitative methods.
Introducing Bayesian methods for the analysis of health economics data, Professor Lilford explained that these methods specify some prior probability that can then be updated in the light of new data. While useful in clinical research, this approach is also an essential underpinning concept in business research, where any type of cost effectiveness analysis is attempted. The argument in favour of this approach was illustrated with examples, attempting to effect change in the epistemology of management and policy research internationally.
Within the context of CAHRD, Diedre Hollingsworth, Assistant Professor of Epidemiology (joint appointment between LSTM and Warwick) is working with Professor Lilford, to explore the viability of causal modelling.
Professor Lilford has over ten years’ experience as Professor of Obstetrics and Gynaecology and over five years in the Department of Health. Currently he is Chair of Public Health at the University of Warwick. He additionally holds the position of Director for Warwick Centre for Applied Health Research and Delivery. Previously he was Professor of Clinical Epidemiology at the University of Birmingham, Vice-Dean for Applied Health Research and Director of the Primary Care Clinical Trials Unit. He directs the NIHR Collaboration for Leadership in Applied Health Research & Care for West Midlands (CLAHRC-WM); a NICE External Assessment Centre and is Co-Investigator on many other research grants