2018/05/18 by Andrea Gabrio, Michael J. Daniels, Gabrio, Andrea +3
Economics, Econometrics and Finance · #Economic and Environmental Valuation #FOS: Computer and information sciences #Health Systems, Economic Evaluations, Quality of Life #Methodology (stat.ME) #Pharmaceutical Economics and Policy
paper · pdf · doi:10.48550/arxiv.1805.07147
openalex publication_date 2018/05/18 · openalex created_date 2022/10/06 · openalex updated_date 2026/07/28
Trial-based economic evaluations are typically performed on cross-sectional\nvariables, derived from the responses for only the completers in the study,\nusing methods that ignore the complexities of utility and cost data (e.g.\nskewness and spikes). We present an alternative and more efficient Bayesian\nparametric approach to handle missing longitudinal outcomes in economic\nevaluations, while accounting for the complexities of the data. We specify a\nflexible parametric model for the observed data and partially identify the\ndistribution of the missing data with partial identifying restrictions and\nsensitivity parameters. We explore alternative nonignorable scenarios through\ndifferent priors for the sensitivity parameters, calibrated on the observed\ndata. Our approach is motivated by, and applied to, data from a trial assessing\nthe cost-effectiveness of a new treatment for intellectual disability and\nchallenging behaviour.\n