ABSTRACT
Missing values or incomplete data are commonly encountered in clinical trials. One of the primary causes of missing data is the dropout. Reasons for dropout include refusal to continue in the study; perceived lack of efficacy; relocation, adverse events; unpleasant study procedures; worsening of disease; unrelated disease; non-compliance with the study; the need to use prohibited medication; and death (DeSouza et al., 2009). Basically, there are three types of missingness mechanisms: (i) missing completely at random (MCAR), (ii) missing at random (MAR), and (iii) missing not at random (MNAR). Depending upon the missingness mechanisms, appropriate missing data analysis strategies can then be considered based on existing analysis methods in the literature. The purpose of this chapter is to provide an overview of statistical methods used under different missingness mechanisms. It should be noted that the method of missing data imputation has been criticized by using legal (statistical) procedure to illegally impute (make up) non-observed data. Other methods, such as the use of the concept of estimand and the derivation of valid statistical methods under incomplete data structure, are also discussed in this chapter.
