Bayesian multiple imputation for missing multivariate longitudinal data from a Parkinson's disease clinical trial

• Published in: Statistical methods in medical research Vol. 25; no. 2; p. 821
• Main Authors: Luo, Sheng, Lawson, Andrew B, He, Bo, Elm, Jordan J, Tilley, Barbara C
• Format: Journal Article
• Language: English
• Published: England 01.04.2016
• Subjects: Bayes Theorem; Creatine - therapeutic use; Humans; Longitudinal Studies; Parkinson Disease - drug therapy; Clinical trial; global statistical test; Markov chain Monte Carlo; item-response theory; latent variable; missing data
• ISSN: 1477-0334
• DOI: 10.1177/0962280212469358

In Parkinson's disease (PD) clinical trials, Parkinson's disease is studied using multiple outcomes of various types (e.g. binary, ordinal, continuous) collected repeatedly over time. The overall treatment effects across all outcomes can be evaluated based on a global test statistic. However, missing data occur in outcomes for many reasons, e.g. dropout, death, etc., and need to be imputed in order to conduct an intent-to-treat analysis. We propose a Bayesian method based on item response theory to perform multiple imputation while accounting for multiple sources of correlation. Sensitivity analysis is performed under various scenarios. Our simulation results indicate that the proposed method outperforms standard methods such as last observation carried forward and separate random effects model for each outcome. Our method is motivated by and applied to a Parkinson's disease clinical trial. The proposed method can be broadly applied to longitudinal studies with multiple outcomes subject to missingness.