The influence of the Ratio Bias phenomenon on the elicitation of Standard Gamble utilities

This paper tests whether logically equivalent risk formats can lead to different health state utilities elicited by means of the standard gamble (SG) method. We compare SG utilities elicited when probabilities are framed in terms of frequencies with respect to 100 people in the population (i.e., X out of 100) with SG utilities elicited for frequencies with respect to 1,000 people in the population (i.e., Y out of 1,000). We found that utilities were significant higher when success and failure probabilities were framed as frequencies type “Y out of 1,000” rather than as frequencies type “X out of 100”. This framing effect, known as Ratio Bias, may have important consequences in resource allocation decisions.