Epidemiological Study Designs for Longitudinal Data

• Activity Number: 279
• Type: Roundtables
• Date/Time: Tuesday, August 11, 2015 : 7:00 AM to 8:15 AM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Jonathan Schildcrout*
• Companies: Vanderbilt University
• Keywords: longitudinal data ; epidemiology ; outcome dependent sampling ; study design ; mixed models

Abstract

Longitudinal data are widely available for research purposes using pre-existing data warehouses (e.g., cohort studies, administrative databases, electronic medical records). In this roundtable, I will lead a discussion regarding opportunities and considerations for conducting cost-efficient retrospective sub-studies to address longitudinal data analysis questions. Often, most of the data required to address such questions are contained in these rich resources; however, one or several key exposure variables must still be ascertained. The primary challenge is that exposure ascertainment is expensive and limits the sample size of the sub-study. Strategies for targeting the sampling scheme toward the most informative subjects will be discussed; analysis procedures for valid inferences will be described; and handouts will be provided that will include, among other material, URLs for references and software.