An Evaluation of Statistical Methods with Missing Data in Small Clinical Trials

• Activity Number: 359 - Contributed Poster Presentations: Biopharmacutical Section
• Type: Contributed
• Date/Time: Tuesday, July 31, 2018 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Takayuki Abe* and Kazuhito Shiosakai and Manabu Iwasaki
• Companies: Yokohama City University, School of Data Science and Daiichi Sankyo Co., Ltd. and Yokohama City University, School of Data Science
• Keywords: rare disease; small clinical trials; multiple imputation; missing data; longitudinal clinical study

Abstract

In drug development for patients with rare disease, small clinical trials are usually conducted. It is critical to make a well-considered statistical analysis plan on how to deal with dropouts in such clinical trials, because the impact of missing data is relatively large. In this research we focused on longitudinal response with dropout in small clinical trials, where complex statistical analysis methods do not work well, due to small sample size relative to the number of time-points. We evaluated various imputation methods (single imputation; last observation carried forward (LOCF), and some multiple imputation (MI) models (marginal and conditional imputation)) under different missing mechanisms. An example motivated by an actual clinical trial was used in the evaluations. Simulation studies were performed for continuous responses under various settings including sample size, the number of time-points, hierarchical structures, missing mechanisms, missing fractions, and analysis models. The details of results will be shown in the presentation.