Drug Use Prevention Data, Missed Assessments and Survival Analysis

In prevention studies, researchers often investigate the incidence of initial drug experimentation or other drug use milestones and its relationship to individual attributes such as the level of parental monitoring or rebelliousness. In this case, survival analysis is the methodology of choice. Survival analysis methods deal efficiently with data from individuals who leave the study prematurely and do not return. However often individuals do return to the study. The application of survival analysis to a situation in which individuals miss assessments and later return is nonstandard. This article examines the use of multiple imputation as a methodology for utilizing information from all assessments.