VALUATION OF SF-6DV2 HEALTH STATES USING TIME TRADE-OFF AND DISCRETE CHOICE EXPERIMENT WITH DURATION APPROACHES IN CHINA
Author(s)
Wu J1, Xie S1, He X1, Chen G2, Brazier JE3
1Tianjin University, Tianjin, China, 2Monash University, Melbourne, Australia, 3University of Sheffield, Sheffield, UK
Objectives: To generate the value set of Short-Form Six-Dimension (SF-6Dv2) using both composite time trade-off (C-TTO) and discrete choice experiments with duration (DCETTO) approaches in China. Methods: A large representative sample of Chinese general population stratified by age, sex, education and proportion of urban/rural population, was recruited from eight provinces/municipalities in China. All respondents were asked to completed 8 C-TTO tasks and 10 DCETTO tasks during face-to-face interviews, with the order of C-TTO and DCETTO tasks randomized. OLS model, Random-effect model, Fixed-effect model and Tobit model were used for C-TTO data, and conditional logit model and mixed logit model were used for DCETTO. Acceptability assessed by self-reported difficulties on understanding/answering, monotonicity of logical ordering of model coefficients, and consistency between predicted values and observed C-TTO values using intraclass correlation coefficient (ICC), mean absolute difference (MAD) and mean squared difference (MSD) were evaluated and compared between two approaches. Results: A total of 3320 respondents (50.3 % males; range from 18-90 years) were recruited. For C-TTO, the predicted health state values ranged from -0.240 to 1 using random-effect model, with 931 (4.97%) out of all 18,750 states considered as worse than dead (WTD). The corresponding range was -0.535 to 1 using conditional logit model for DCETTO, with a higher WTD percentage (8.50%, N=1,593). No significant difference was observed in self-reported difficulties on understanding/answering between two approaches. DCETTO presented minor nonmonotonicity with the coefficients in two dimension-levels (the first- and second-level in Mental-health and Vitality), which was not observed for C-TTO. Both approaches values are highly consistent (ICC=0.981, MAD=0.0808, MSD=0.0094) while DCETTO values were slightly lower than C-TTO. Conclusions: Both C-TTO and DCETTO approaches present relatively good acceptability, monotonicity and consistency for eliciting utility values in China. Minor issues of nonmonotonicity did present for DCETTO. A systematic difference was found between value sets between C-TTO and DCETTO.
Conference/Value in Health Info
2020-05, ISPOR 2020, Orlando, FL, USA
Value in Health, Volume 23, Issue 5, S1 (May 2020)
Acceptance Code
PP3
Topic
Methodological & Statistical Research, Patient-Centered Research
Topic Subcategory
Health State Utilities, Patient-reported Outcomes & Quality of Life Outcomes, PRO & Related Methods
Disease
No Specific Disease