Dropout is a common problem in longitudinal clinical trials and cohort studies, and is of particular concern when dropout occurs for reasons that may be related to the outcome of interest. For example, HIV studies are often longitudinal in nature, and it is well documented that many subjects have missing observations due to death or disease progression, leading to concerns of non-ignorable dropout [1]. Dropout is not ignorable when missingness depends on the values of the unobserved outcomes, even after conditioning on the available data [2]. In this scenario, standard longitudinal data analyses can produce biased results.
展开▼
