Is there a kink in consumers' threshold value for cost-effectiveness in health care?

Background: A reproducible observation is that consumers' willingness-to-accept (WTA) monetary compensation to forgo a program is greater than their stated willingness-to-pay (WTP) for the same benefit. Several explanations exist, including the psychological principle that the utility of losses weighs heavier than gains. We sought to quantify the WTP-WTA disparity from published literature and explore implications for cost-effectiveness analysis accept-reject thresholds in the south-west quadrant of the cost-effectiveness plane (less effect, less cost). <P>Methods: We reviewed published studies (health and non-health) to estimate the ratio of WTA to WTP for the same program benefit for each study and to determine if WTA is consistently greater than WTP in the literature. Results: WTA|WTP ratios were greater than unity for every study we reviewed. The ratios ranged from 3.2 to 89.4 for environmental studies (n=7), 1.9 to 6.4 for health care studies (n=2), 1.1 to 3.6 for safety studies (n=4) and 1.3 to 2.6 for experimental studies (n=7). Conclusions: Given that WTA is greater than WTP based on individual preferences, should not societal preferences used to determine cost-effectiveness thresholds reflect this disparity? Current convention in cost-effectiveness analysis is that any given accept-rejection criterion (e.g. $50 k|QALY gained) is symmetric - a straight line through the origin of the cost-effectiveness plane. The WTA-WTP evidence suggests a downward 'kink' through the origin for the south-west quadrant, such that the 'selling price' of a QALY is greater than the 'buying price'. The possibility of 'kinky cost-effectiveness' decision rules and the size of the kink merits further exploration. Copyright © 2002 John Wiley & Sons, Ltd.