ABSTRACT
Missing data in longitudinal clinical trials is a complex and wide-ranging topic, in part because missing data may arise in many ways. Intermittent missing data occurs when subjects miss scheduled assessments but attend subsequent visits. Dropout (withdrawal, attrition) is when subjects miss all subsequent assessments after a certain visit. Intermittent missing data are generally less problematic than withdrawal/attrition because, having observed data bracketing the unobserved value makes it easier to verify assumptions about the missing data compared to when no subsequent values are observed.
