Please join us as Michael Elliott presents on Injury Epidemiology and Child Passenger Safety.
Determining causation in the absence of randomized exposures or treatment remains a challenging task in epidemiology, and injury epidemiology is no exception. Besides issues of confounding, estimating causal effects in particularly challenging in injury settings due to issues such as sample selection, definition of exposures and outcomes, and other factors. We explore a setting which recently had controversy: use of booster seats for children in passenger vehicles. Dubner and Levitt, both in New York Times articles and in their book Superfreakonomics, challenged the prevailing wisdom that such seats were effective in reducing risk of injury and fatality in children otherwise restrained in passenger vehicles. We'll review a range of analyses with a variety of datasets to explore the causal effect of booster seats on risk of death among children aged 2 through 6 in passenger vehicles, considering issues of selection and confounding from a variety of perspectives, including combining multiple datasets, use of propensity scores and marginal structural models, and generalized instrumental variable analyses.
Michael Elliott is Professor of Biostatistics at the University of Michigan School of Public Health and Research Scientist at the Institute for Social Research. He received his Ph.D. in biostatistics in 1999 from the University of Michigan. Prior to joining the University of Michigan in 2005, he held an appointment as an Assistant Professor at the Department of Biostatistics and Epidemiology at the University of Pennsylvania School of Medicine, and prior to that as a Visiting Professor of Biostatistics at the University of Michigan School of Public Health and as a Visiting Research Scientist at the University of Michigan Transportation Research Institute. Dr. Elliott's statistical research interests focus around the broad topic of "missing data," including the design and analysis of sample surveys, casual and counterfactual inference, and latent variable models.
This event was webcast and recorded. TO SEE THE WEBCAST ARCHIVE, VISIT https://umichitam.adobeconnect.com/p1qsdbw80jc/ .