Multi-parameter evidence synthesis methods (MPES)
José Leal ‘s doctoral research has focussed on MPES: methods of incorporating all relevant evidence in decision models. This requires the combination of information from studies of different design (surveys, randomised and non-randomised studies), incorporation of elicited expert opinion, and estimation of unknown individual model parameters that are a mathematical function of other known parameters.
He has used a prostate cancer screening model to evaluate multi-parameter evidence synthesis methods applied to a cost-effectiveness model. This involves synthesising several prostate cancer autopsy studies to estimate the unknown histological prevalence of cancer by age group and ethnicity. The work also involves estimating the underlying rates of a Markov model and using histological prevalence, screening and clinical incidence data to calibrate, validate and check the consistency of the several sources of evidence and obtain measures such as sensitivity of the screening programme and lead-time bias. The work is being undertaken by José Leal and Jane Wolstenholme.
He has also applied the methods to estimate the cost-effectiveness of screening for medium chain acyl CoA dehydrogenase deficiency (MCADD), a rare disease resulting from an inborn error of metabolism. This involves synthesizing National Census data with data from published literature and pilot screening data, in order to estimate parameters such as sensitivity of the screening programme, clinical yield in the absence of screening and prevalence of disease. This work is being undertaken in collaboration with Professor Tony Ades, MRC Health Services Research Collaboration, University of Bristol.
See also Glaucoma treatments: meta-analysis