Zhao is interested in statistical methods related to biomarkers. First, biomarkers have the potential to elucidate the key biological pathways through which treatment may affect clinical outcomes. However, measurement error in the biomarker measures can obscure this ability. She developed correction methods for the Cox model in the mediation analysis setting, when the mediator is measured with error. Second, biomarkers can be used for diagnosis of cancer and other diseases. She is interested in efficient designs of biomarker studies, such as sequential design and quota-sampling, with the aim to conserve biospecimens and reduce cost. Third, she is involved in studies with epigenetic biomarkers, which serve as mediators between environmental factors and many diseases.
Zhao is also focusing on multivariate failure time data analysis. Although the analysis of univariate failure time has been well developed, there are still big gaps in multivariate failure time methodology. In practice with rare diseases, we may not be able to observe the disease of interest in all study participants (i.e., these subjects are censored). However, we might observe other events during the study. If the observed events are related to the main study outcome, we may be able to use the data to fill in information about the main study outcome on those censored participants. For example, a study to predict time to stroke can benefit from information on time to coronary heart disease, given that these two outcomes are related. In addition, we might be interested in understanding how times to different diseases are related to each other, and how disease onset times relate within families. Zhao works on semiparametric models for bivariate failure times to address these issues.
Dr. Zhao received her Ph.D. in Biostatistics from the University of Washington in 2012. She was a postdoctoral fellow at the Fred Hutchinson Cancer Research Center from 2012 to 2014. She joined NIEHS in 2015.
- Carroll R, Lawson AB, Zhao S. Temporally dependent accelerated failure time model for capturing the impact of events that alter survival in disease mapping. Biostatistics. 2019;20(4):666-680.
- Prentice RL, Zhao S. Nonparametric estimation of the multivariate survivor function: the multivariate Kaplan-Meier estimator. Lifetime Data Anal. 2018;24(1):3-27.
- Carroll R, Lawson AB, Zhao S. A data-driven approach for estimating the change-points and impact of major events on disease risk. Spat Spatiotemporal Epidemiol. 2019;29:111-118.
- Prentice RL, Zhao S. Regression Models and Multivariate Life Tables. J Am Stat Assoc. 2021;116(535):1330-1345.
- Kim JI, Fine JP, Sandler DP, Zhao S. Accounting for Preinvasive Conditions in Analysis of Invasive Cancer Risk: Application to Breast Cancer. Epidemiology. 2022;33(1):48-54.
Related Scientific Focus Areas
This page was last updated on Tuesday, March 31, 2015