Short course: Global Sensitivity Analysis of Randomized Trials with Missing Data – Methods and Software
Daniel Scharfstein, ScD, Johns Hopkins Bloomberg School of Health, will present case studies that illustrate flexible methods and software (R and SAS) for conducting rigorous sensitivity analysis of randomized trials with missing data.
Missing outcome data are a widespread problem in randomized trials. The analysis of studies with missing data rely on untestable assumptions. As a result, it is important to evaluate the robustness of trial results to such assumptions (i.e., sensitivity analysis). In 2010, the National Academy of Sciences issued a report that recommended that sensitivity analysis “should be part of the primary reporting of findings from clinical trials.” While Chapter 5 of the report laid out a framework for conducting rigorous sensitivity analyses, such analyses are rarely reported. This is due, in part, to inadequate knowledge translation by statistical methodologists to principal investigators and their statistical collaborators as well as lack of software.
We will discuss flexible methods and software (R and SAS) for conducting rigorous sensitivity of randomized trials in which outcomes are scheduled to be measured at fixed points in time after randomization, the primary source of missing data is due to premature patient withdrawal, and outcomes may be missing prior to withdrawal. We present comprehensive case studies to illustrate the methods and software.
Masters level training in Statistics/Biostatistics; familiarity with R or SAS
Daniel Scharfstein, ScD
Daniel Scharfstein is professor of biostatistics at the Johns Hopkins Bloomberg School of Public Health. He was a member of the National Research Council panel that issued the report “The Prevention and Treatment of Missing Data in Clinical Trials.” He has been funded by FDA, PCORI, and NIH to develop, implement, and disseminate methods for conducting sensitivity analysis of randomized trials with missing data.