Back to Annual Meeting

Back to Annual Meeting

APHA Scientific Session and Event Listing

Katherine J. Hoggatt, PhD, School of Public Health, University of Michigan, 109 Observatory Street, Ann Arbor, MI 48109, 734-647-3220, khoggatt@umich.edu

Two-phase designs have been proposed as an efficient way to collect detailed covariate or exposure data for a subsample nested within a well-defined Phase 1 group (e.g., a nested case-control study). Analyses that ignore the information in the known Phase 2 sampling fractions may yield inefficient estimates that are biased due to selection/response effects. In this talk, I will show the causal origins of the selection/response bias and will illustrate how a two-phase analysis can not only minimize the bias but can also improve the precision of the resulting estimates.

Learning Objectives: At the conclusion of the session, the participant (learner) in this session will be able to

• Understand the causal origins of selection/response bias

Presenting author's disclosure statement:

Any relevant financial relationships? No