Optimizing presentation formats for SF-6Dv2 health status valuation: a qualitative study using cognitive interviews.

Area of Science:

• Health Economics
• Survey Methodology
• Psychometrics

Background:

• Discrete choice experiment (DCE) methods are crucial for health status valuation, such as the SF-6Dv2.
• Effective presentation strategies are vital for improving respondent understanding and reducing bias in DCE surveys, especially online.
• Optimizing DCE choice task presentation is essential for reliable health valuation data.

Purpose of the Study:

• To identify optimal presentation formats for DCE choice tasks.
• To evaluate the impact of different visual emphasis and attribute placement strategies on respondent understanding and choice behavior.
• To inform best practices for SF-6Dv2 health status valuation surveys.

Main Methods:

• Qualitative cognitive interviews were conducted with 40 South Korean adults.
• Participants used think-aloud and retrospective probing methods.
• Five presentation formats (non-emphasized, bold-underlined, yellow-highlighted, graphic circle, color-shaded) and two duration attribute placements (upfront vs. end) were evaluated.

Main Results:

• Presentation format significantly impacted readability, comprehension, and comparative judgment.
• The bold-underlined format was favored for its readability, quick comprehension, visual balance, and support for comparative judgment.
• Yellow-highlighted format was popular but criticized for visual burden; graphic format had scale and summation issues.
• Upfront duration placement was preferred by those prioritizing survival duration, while end placement was favored by those emphasizing quality of life.

Conclusions:

• While several formats improved comprehension, the bold-underlined format offers the best balance of comprehension enhancement and choice stability for SF-6Dv2 valuation.
• Presentation format choice can influence DCE response patterns, highlighting the need for careful survey design.
• Future DCE studies should consider the impact of visual presentation on data quality and respondent experience.