DECISION MAKING AND VALUE OF INFORMATION ANALYSIS IN A REAL-WORLD MULTI-CRITERIA CONTEXT

Purpose:

Decision making regarding reimbursement of new medical interventions typically involves ethical, social, legal and health economic aspects. In addition to cost-effectiveness, clinical and economic restrictions regarding for example adverse events, under-diagnosis, and budget impact, may also influence such reimbursement decisions. However, when applying standard Value of Information (VOI) analysis the value of additional research is estimated assuming all cost-effectiveness outcomes are equally relevant, and ignoring the effect of other considerations on these decisions. The purpose of this study is to determine if VOI results may be biased when the acceptability of potential intervention outcomes changes due to additional restrictions.

Method(s) :

In a simulation study we applied standard cost-effectiveness and VOI analysis for comparing a new hypothetical drug to a usual care drug. We determined the probability that the new drug was cost-effective, and the expected value of perfect information (EVPI). We then considered a real-world scenario in which the risk of adverse events for the new drug was restricted to a maximum of 3% and the additional cost was restricted to a maximum of €2,500 compared to the current drug. We re-estimated the cost-effectiveness and EVPI for the new drug given these restrictions.

Result(s):

For a willingness-to-pay of €20,000/QALY, the probability that the new drug was cost-effective compared to the current drug was 57%, with an EVPI of €1,866 per patient. Applying the adverse event restriction reduced the probability of exceeding this maximum event threshold from 34% to 2.3%, but also reduced the EVPI to €1,099. Similarly, restricting the additional cost reduced the probability of exceeding the cost threshold from 41% to 0%, but also reduced the EVPI to €765. Applying both criteria simultaneously resulted in a 3.3% and 0% chance of exceeding each of the respective thresholds, while reducing the EVPI further to only €290.

Conclusion(s):

When aspects other than costs and health effects influence the reimbursement decision of new medical technology this may lead to an upward bias in results from standard VOI analysis. Every additional restriction may prevent exploitation of potential benefits in cost-effectiveness outcomes found in a new study, thereby reducing the value of collecting additional evidence. When common real-world restrictions are in place, these should therefore be adequately incorporated in the VOI analysis to avoid incorrect research prioritization decisions.