ESTIMATION OF DISABILITY WEIGHTS IN THE GENERAL POPULATION OF SOUTH KOREA USING A PAIRED COMPARISON

Purpose: We investigated the feasibility of a paired comparison-only model to estimate disability weights in a national burden of disease study in South Korea. 

Method:

The study was conducted through two types of survey: a household survey and a web-based survey, in the same way as in the GBD 2010 disability weight study. A paired comparison, visual analogue scale (VAS), and standard gamble (SG) were used as valuation methods in the household survey, whereas a paired comparison and population health equivalence (PHE) were applied in the web-based survey. We described a total of 258 health states, with ‘full health’ and ‘being dead’ designated as anchor points. In the analysis, five models were considered: a paired comparison-only model; a hybrid model between a paired comparison and PHE; a Bradley-Terry model; a VAS model; and a SG model. Frequency distributions of the disability weights from each model were determined to estimate the validity of the health states. Furthermore, Pearson correlation coefficients were calculated to compare disability weights from our models to those of the GBD 2010 disability weight study.

Result:

A total of 2,728 individuals participated in the household survey and 3,188 in the web-based survey. Based on the distribution of disability weights of the health states for each model and the Pearson correlation coefficients, the paired comparison-only model based on probit regression was selected as the best model for estimating disability weight in South Korea and for maintaining simplicity of the analysis. When comparing the disability weights between the paired comparison-only model and the GBD 2010 disability weight study, a relatively similar pattern was noted.

Conclusion:

Disability weights can be estimated using paired comparison only by including ‘full health’ and ‘being dead’ in the list of health states. By using a simplified methodology of estimating disability weights, as in this study, more empirical evidence on the universality of disability weight is seen.