PhD, Johns Hopkins Bloomberg School of Public Health, 2007
ScM, North Carolina State University, 1999
OverviewMy collaborative research focuses on studies of critically ill patients whom are treated within the intensive care unit (ICU), through the OACIS group at Johns Hopkins School of Medicine. Specifically, I am interested in how different baseline and ICU treatments are associated with patient outcomes such as mortality and physical function while the patients are in the ICU. In addition, there is growing knowledge that survivors of critical illness carry the burden of this illness for prolonged periods after ICU and hospital discharge. Therefore, I have been working with long-term follow-up studies among these survivors assessing physical function, mental health and quality of life. Several statistical challenges in these studies include: how to model the intensive longitudinal data generated while the patients are in the ICU (daily or twice daily measurements) when often the exposures are time-varying, and the competing risk of mortality both in and out of the ICU which is especially challenging when the outcome of interest is not a survival event but a continuous or binary outcome.
I am the lead statistician on the TREAT study (Treating Parents to Reduce NICU Transmission of Staphylococcus Aureus Trial).
I also work on reviewing and developing statistical methods that leverage prognostic baseline variables to improve precision for estimating the average treatment effect within randomized trials. This is joint work with Michael Rosenblum and several others.
TEACHING: I teach Analysis of Longitudinal Data and Multilevel Statistical Models in Public Health.
In the past, I worked on multi-stage models for Alzheimer's disease prevalence (work with Ron Brookmeyer), we have developed a web-based tool to project the prevalence of Alzheimer's disease globally, for the US or for other user-specified regions.
View my CV