Dear list member,
the next research seminar of The Open University Statistics Group will be held on Friday, March 2nd at the following venue: Alan Turing building, room 306, Walton Hall campus, Milton Keynes (MK7 6AA). The meeting will begin at 11 a.m., with tea and coffee available from 10.30 a.m., and usually finishes at approximately 12 noon. The meeting is free and open to all.
Dr John W. Stevens (Centre for Bayesian Statistics in Health Economics (CHEBS), University of Sheffield) will be speaking on the topic of "Bayesian Methods in Health Economics".
Abstract:
Decision about which new health technologies the NHS should use are made by the National Institute for Health and Clinical Excellence (NICE). Decisions are based on the lifetime population incremental cost-effectiveness ratio (ICER). Assessments are made using a decision analytic model because not all relevant evidence is contained within a single trial: not all interventions of interest may have been compared directly and lifetime costs and benefits extend beyond trial follow-up. The inputs to economic models are true population values and these are represented by probability distributions to represent uncertainty about their true values. Uncertainty in the output is derived by propagating input uncertainty through the model using probabilistic sensitivity analysis (PSA). After introducing the decision framework, this talk will discuss some statistical issues common to submissions to NICE, including: the synthesis of evidence from RCTs on all relevant interventions; the extrapolation of RCT evidence; and the role of structural uncertainty in addition to parameter uncertainty.
Kind Regards,
Dr Steffen Unkel (Department of Mathematics and Statistics, The Open University)
http://statistics.open.ac.uk
--
The Open University is incorporated by Royal Charter (RC 000391), an exempt charity in England & Wales and a charity registered in Scotland (SC 038302).
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
|