Eloise Kaizar

The Inaugural Ross-Royall Symposium:
From Individuals to Populations
Wood Basic Sciences Auditorium
Friday, February 26, 2016
8:30 am – 6:00 pm
Eloise Kaizar, B.A., B.S., M.S., Ph.D.
Associate Professor, Department of Statistics, The Ohio State University
Title:
Estimating beyond the trial-represented population by incorporating studies with
self-selected treatments
Abstract:
Although randomized controlled trials are considered the 'gold standard' for
clinical studies, the use of exclusion criteria may impact the external validity of
trial results. It is unknown whether estimators of effect size are biased by
excluding a portion of the target population from enrollment. We propose to use
observational data to estimate the bias due to enrollment restrictions in the
presence of non-constant treatment effects. We find that we can estimate the bias
of our estimates of average treatment effect due to certain kinds of confounding,
even for mis-specified models. While data availability for such methods is
currently limited, by augmenting randomized controlled trial data with
observational data, we can reduce our reliance on homogeneity assumptions when
making inference for broad populations.