OBJECTIVES: Time trade-off (TTO) and discrete choice experiment (DCE) are commonly used preference-elicitation techniques. They can be administered using various modes of administration (MoA), including face-to-face interviews (F2F), unassisted, online surveys (UO), or online, remote-assisted interviews (RA). The objective of this study was to explore how the MoA affects the quality and reliability of preference-elicitation data.

METHODS: EQ-5D-5L health states were valued using TTO and DCE approaches by the UK general population (N=544). Participants were allocated to one of two study groups. Group A completed a F2F interview and a UO survey and Group B completed a RA interview and a UO survey. The feasibility of survey completion and the reliability and face-validity of data collected were compared across the different MoA.

RESULTS: Fewer respondents reported receiving sufficient guidance on the TTO tasks during the UO surveys compared with the two assisted MoA. However, respondents across all MoA typically reported receiving sufficient guidance on the DCE tasks.

TTO data were less reliable from the UO surveys compared with both assisted MoA, but there were no differences in DCE data reliability across all MoA. TTO data from all MoA passed face-validity checks; however, the UO surveys produced higher utility values for moderate and severe health-states than either assisted mode. Both F2F and RA modes provided comparably reliable data. MoA was confirmed as a significant predictor of observed differences in outcomes by regression analyses.

CONCLUSIONS: UO surveys are quicker to complete but yield less reliable TTO data than assisted surveys, in terms of logical inconsistencies. The reliability of DCE data is not affected by MoA. Both F2F and RA surveys produced similar quality data.