Innovative Medical Device Design Significantly Reduces Surgeon Intra-Procedural Stress Levels (Ipsl)

Abstract

Existing methods for eliciting health state preferences vary considerably in approach, but have in common an aim to 'uncover' preferences by asking survey respondents to evaluate a sub-set of states, then using their responses to infer the relative importance of specific dimensions and levels. An alternative approach is to ask people to construct their own personal utility functions (PUFs) by directly asking them about the relative importance of dimensions, levels and interactions between them. We developed an innovative questionnaire designed to elicit PUFs from the general public via computer-assisted face-to-face interviews. We piloted the questionnaire with a convenience sample of n= 30 health outcomes professionals. The aims of this paper are to report the pilot results and discuss the feasibility of the PUF approach in valuing EQ-5D. MethOds: The approach is grounded in utility theory, and informed by swing-weighting techniques. 30 interviews were conducted in 2015 by a member of the study team. Respondents were encouraged to reflect on their answers and to change any previous responses. The approach is based on the premise that individuals construct their preferences in response to stated preference tasks. Results: Overall, pain/discomfort and anxiety/depression are the highest ranked dimensions and mobility is the lowest ranked dimension. High levels of internal consistency were observed. The ratings for intermediate improvements in each dimension show heterogeneity, both within and between participants. The validation exercise results demonstrate that the PUF algorithm works as intended. The task to identify the individual's location of dead within the descriptive system worked well. 20% of participants indicated that no EQ-5D health states are worse than dead. The data allow the reporting of individual respondents' PUFs. cOnclusiOns: The results of the pilot suggest the PUF approach is likely to be feasible and has potential to yield meaningful, well-informed preference data from respondents.