A novel concept on incremental cost-effectiveness ratio with censored cost data
The incremental cost-effectiveness ratio (ICER), defined as the ratio of the difference in cost between two treatments to the difference in effectiveness between those treatments, is the most widely used analytical tool in health economics evaluation research. Although the ICER has gained widespread acceptance, the theoretical foundation of ICER has not been rigorously examined in published studies. ICER estimates calculated by using the means of both cost and effectiveness measures can be highly variable. There is a strong need for a comprehensive statistical framework to ensure consistency in ICER estimations to have a thorough statistical inference on ICER and derive a consistent ICER estimator. We propose a novel method of calculating the ICER for definition of ICER with censored survival data and construct an ICER estimator with the vector of predictive covariates. We extend the methodology using the Cox proportional hazards model to censored cost-effectiveness data analysis, in particular, the estimation of ICER in presence of censoring dataset. To the best of our knowledge, this is the first attempt to incorporate Cox regression analysis in estimating the ICER. The application of our novel methodology to the analysis of the simulation data and the real Surveillance, Epidemiology and End Results (SEER) Medicare data are also described.
Jiang, Jing, "A novel concept on incremental cost-effectiveness ratio with censored cost data" (2014). Texas Medical Center Dissertations (via ProQuest). AAI3639429.