An Exploratory Application of Eye-Tracking Methods in a Discrete Choice Experiment

Area of Science:

• Health economics
• Decision science
• Behavioral economics

Background:

• Discrete choice experiments (DCEs) are widely used for benefit-risk tradeoff analysis.
• Understanding respondent decision-making strategies in DCEs is crucial for accurate preference elicitation.
• The impact of risk communication on choices within DCEs requires further investigation.

Purpose of the Study:

• To explore the utility of eye-tracking methods in understanding decision-making strategies within DCEs.
• To investigate how different risk communication formats influence respondents' choices and attention.
• To validate self-reported attribute nonattendance (ANA) using objective eye-tracking data.

Main Methods:

• Two versions of a breast screening DCE were developed, varying risk communication (percentages vs. icon arrays and percentages).
• Eye-tracking equipment recorded respondents' eye movements at 1000 Hz.
• DCE choices were analyzed using a heteroskedastic conditional logit model; eye-tracking data (pupil size, saccades, dwell time) were analyzed using OLS regressions.

Main Results:

• No significant difference in attribute attention (fixations) was found between risk communication formats.
• Respondents exhibited more horizontal saccades when alternatives were presented in columns.
• Eye-tracking data confirmed self-reported ANA, showing significant reductions in dwell time for risk attributes (e.g., probability of cancer detection).

Conclusions:

• Eye-tracking offers valuable insights into respondent decision-making processes in health care DCEs.
• Risk communication strategies can impact decision-making, though attention to attributes did not differ significantly in this study.
• Self-reported ANA may not be a reliable indicator of actual nonattendance to cost attributes in DCEs.