Modifying the Composite Time Trade-Off Method to Improve Its Discriminatory Power

Area of Science:

• Health Economics
• Psychometrics
• Decision Science

Background:

• Health state utilities are crucial for cost-effectiveness analysis (CEA) of health technologies.
• The composite time trade-off (cTTO) method is widely used for eliciting utilities, especially for the EQ-5D-5L instrument.
• Standard cTTO has limitations including respondent non-trading (NT), utility censoring at -1, and poor correlation of negative utilities with severity.

Purpose of the Study:

• To investigate modifications to the cTTO method to mitigate its limitations.
• To improve the discriminatory power of cTTO for health state utility elicitation.
• To enable the capture of utilities below -1 and enhance the correlation with health state severity.

Main Methods:

• Online survey of 478 students valuing 10 EQ-5D-5L health states.
• Three arms: Standard cTTO with NT/censoring exploration (Arm A); modified cTTO with alternative trade-offs to overcome loss aversion and unify tasks (Arms B & C).
• Modified cTTO in Arms B & C aimed to elicit utilities < -1 and improve negative utility representation.

Main Results:

• Modified cTTO (Arms B & C) showed significantly less non-trading (4-4% vs 10% in Arm A).
• More strictly negative utilities were observed (38-40% vs 25%), with more utilities ≤ -1 (18-30% vs 10%).
• The most severe state (55555) utility dropped substantially (e.g., to -2.52 in Arm C from -0.53). Modified methods showed better alignment of negative utilities with severity but increased task difficulty and inconsistencies.

Conclusions:

• Modifications to the cTTO method can enhance its discriminatory power for health state utility elicitation.
• The standard cTTO may overestimate utilities, particularly for severe health states.
• Improving cTTO's ability to capture nuanced negative utilities may require increasing task complexity.