Andrew R Willan1. 1. Program in Child Health Evaluation Sciences, SickKids Research Institute and Department of Public Health Sciences, University of Toronto, Toronto, Canada. andy@andywillan.com
Abstract
BACKGROUND: The results of the HOPE study, a randomized clinical trial, provide strong evidence that 1) ramipril prevents the composite outcome of cardiovascular death, myocardial infarction or stroke in patients who are at high risk of a cardiovascular event and 2) ramipril is cost-effective at a threshold willingness-to-pay of $10,000 to prevent an event of the composite outcome. In this report the concept of the expected value of information is used to determine if the information provided by the HOPE study is sufficient for decision making in the US and Canada. METHODS: and results Using the cost-effectiveness data from a clinical trial, or from a meta-analysis of several trials, one can determine, based on the number of future patients that would benefit from the health technology under investigation, the expected value of sample information (EVSI) of a future trial as a function of proposed sample size. If the EVSI exceeds the cost for any particular sample size then the current information is insufficient for decision making and a future trial is indicated. If, on the other hand, there is no sample size for which the EVSI exceeds the cost, then there is sufficient information for decision making and no future trial is required. Using the data from the HOPE study these concepts are applied for various assumptions regarding the fixed and variable cost of a future trial and the number of patients who would benefit from ramipril. CONCLUSIONS: Expected value of information methods provide a decision-analytic alternative to the standard likelihood methods for assessing the evidence provided by cost-effectiveness data from randomized clinical trials.
BACKGROUND: The results of the HOPE study, a randomized clinical trial, provide strong evidence that 1) ramipril prevents the composite outcome of cardiovascular death, myocardial infarction or stroke in patients who are at high risk of a cardiovascular event and 2) ramipril is cost-effective at a threshold willingness-to-pay of $10,000 to prevent an event of the composite outcome. In this report the concept of the expected value of information is used to determine if the information provided by the HOPE study is sufficient for decision making in the US and Canada. METHODS: and results Using the cost-effectiveness data from a clinical trial, or from a meta-analysis of several trials, one can determine, based on the number of future patients that would benefit from the health technology under investigation, the expected value of sample information (EVSI) of a future trial as a function of proposed sample size. If the EVSI exceeds the cost for any particular sample size then the current information is insufficient for decision making and a future trial is indicated. If, on the other hand, there is no sample size for which the EVSI exceeds the cost, then there is sufficient information for decision making and no future trial is required. Using the data from the HOPE study these concepts are applied for various assumptions regarding the fixed and variable cost of a future trial and the number of patients who would benefit from ramipril. CONCLUSIONS: Expected value of information methods provide a decision-analytic alternative to the standard likelihood methods for assessing the evidence provided by cost-effectiveness data from randomized clinical trials.
Authors: Lusine Abrahamyan; Andrew R Willan; Joseph Beyene; Marjorie Mclimont; Victor Blanchette; Brian M Feldman Journal: J Gen Intern Med Date: 2014-08 Impact factor: 5.128
Authors: Rena Conti; David L Veenstra; Katrina Armstrong; Lawrence J Lesko; Scott D Grosse Journal: Med Decis Making Date: 2010-01-04 Impact factor: 2.583
Authors: Aileen R Neilson; Hanne Bruhn; Christine M Bond; Alison M Elliott; Blair H Smith; Philip C Hannaford; Richard Holland; Amanda J Lee; Margaret Watson; David Wright; Paul McNamee Journal: BMJ Open Date: 2015-04-01 Impact factor: 2.692
Authors: Manuel Gomes; Edmond S-W Ng; Richard Grieve; Richard Nixon; James Carpenter; Simon G Thompson Journal: Med Decis Making Date: 2011-10-19 Impact factor: 2.583