Aim: To explore what happens to the expected value of perfect information (EVPI) as an increasingly large number of decision options is considered.
Methods: A stylised model of screening for a hypothetical cancer. The model was used to test different possible ages for ‘once in a lifetime’ screening under a variety of assumptions about model parameter uncertainty. Initial model runs only considered screening at ages which are multiples of 16 years, then multiples of 8 years were allowed, then 4 years, etc.
Results: Effects of more refined choice sets on the cost-effectiveness acceptability frontier (CEAF) and EVPI are shown. The CEAF collapses to zero for threshold incremental cost-effectiveness ratios at which screening at some age is preferred to no screening, while the EVPI stabilises at a non-zero figure.
Conclusions: The CEAF is likely to be highly dependent on the choice of options when these are a selection from a very large set of possible options. In contrast, the EVPI can be reasonably approximated by a model with a slightly limited choice set. Copyright © 2010 John Wiley & Sons, Ltd.