Truncation by death and unmeasured confounding are ubiquitous challenges in epidemiological studies of older adults. Motivated by the Project on Optimal Vitamin D in Older Adults (PROVIDO), a study composed of five observational cohorts of older adults with the goal of improving the design of vitamin supplementation trials, I will describe three statistical methods developed to address truncation by death or unmeasured confounding. Results from simulation studies and application of the methods to PROVIDO will be discussed, including the potential public health significance of the findings.
Speaker: Michelle D. Shardell, PhD, Professor of Epidemiology & Public Health, University of Maryland School of Medicine
To obtain a Zoom link, please email [email protected].