生物识别与生物统计学杂志

An Analysis of Selection Models For Non-ignorable Dropout: An Application to Multi-centre Trial Data

Abstract

Ali Satty

A common problem encountered in statistical analysis is that of missing data, which occurs when some variables have missing values in some units. The present paper deals with the analysis of longitudinal continuous measurements with incomplete data due to non-ignorable dropout. In repeated measurements data, as one solution to a problem, the selection model assumes a mechanism of outcome-dependent dropout and jointly both the measurement together with dropout process of repeated measures. We consider the construction of a particular type of selection model that uses a logistic regression model to describe the dependency of dropout indicators on the longitudinal measurement. We focus on the use of the Diggle-Kenward model as a tool for assessing the sensitivity of a selection model in terms of the modeling assumptions. Our main objective here is to investigate the influence on inference that might be exerted on the considered data by the dropout process. We restrict attention to a model for repeated Gaussian measures, subject to potentially non-random dropout. To investigate this, we carry out an application for analyzing incomplete longitudinal clinical trial with dropout by using a practical example in the form of a multi-center clinical trial data.