When people make choices, they may have multiple options presented simultaneously or, alternatively, have options presented 1 at a time. It has been shown that if decision makers have little experience with or difficulties in understanding certain attributes, these attributes will have greater impact in joint evaluations than in separate evaluations. The authors investigated the impact of separate versus joint evaluations in a health care context in which laypeople were presented with the possibility of participating in risk-reducing drug therapies.
In a randomized study comprising 895 subjects aged 40 to 59 y in Odense, Denmark, subjects were randomized to receive information in terms of absolute risk reduction (ARR), relative risk reduction (RRR), number needed to treat (NNT), or prolongation of life (POL), all with respect to heart attack, and they were asked whether they would be willing to receive a specified treatment. Respondents were randomly allocated to valuing the interventions separately (either great effect or small effect) or jointly (small effect and large effect).
Joint evaluation reduced the propensity to accept the intervention that offered the smallest effect. Respondents were more sensitive to scale when faced with a joint evaluation for information formats ARR, RRR, and POL but not for NNT. Evaluability bias appeared to be most pronounced for POL and ARR.
Risk information appears to be prone to evaluability bias. This suggests that numeric information on health gains is difficult to evaluate in isolation. Consequently, such information may bear too little weight in separate evaluations of risk-reducing interventions.