Abstract

The study by Pasenggong et al represents a pivotal advancement in quantifying health utility in ophthalmology by mapping the Impact of Vision Impairment (IVI) questionnaire to EQ-5D-5L utility scores in Thai patients with visual impairment. Their use of the adjusted limited dependent variable mixture model aligns well with the known nonnormal distribution of utility scores, particularly in domains in which ceiling effects are frequent.

Their choice of model 2—incorporating IVI domain scores with sociodemographic and clinical variables—yielded the best predictive performance, notably with an root mean square error of 0.137 and mean absolute error of 0.101. The ability of adjusted limited dependent variable mixture model to account for multimodality and skewness in utility score distributions offers strong justification for its use in cost-utility analysis involving vision-specific measures.