Min Zhang, PhD, joined the Department of Biostatistics at the University of Michigan (UM) as an Assistant Professor in 2008. Her methodological research has been focused on semiparametric methods, causal inference (comparative effectiveness analysis), survival data analysis, longitudinal data analysis, dynamic treatment regimes (personalized medicine), missing data and clinical trials.
She develops semiparametric methods to improve efficiency and robustness of statistical analyses for data obtained from clinical trials or observational studies (e.g., clinical registries, health claims data). Due to the nature of observational studies, accounting for potential confounding in a robust and efficient manner such that valid inferences can be obtained has always been an important yet challenging problem, especially in the presence of time-dependent confounders. Although confounding may seem less a concern for clinical trials, in reality complications often occur, for example, in cancer clinical trials, rendering properly accounting for confounding and robustness consideration necessary. For example, patients in a clinical trial may fail to take the assigned study treatment, optionally discontinue treatments, start a secondary treatment on his/her own, or violate the study protocol in other ways. Zhang’s research has been focused on developing more robust and efficient methods to handle these issues. She develops methodologies for various type of data including, for example, survival outcomes, longitudinal, clustered or hierarchical data.
Her collaborative research is mainly on the area of cardiovascular disease and she has been collaborating with physicians, surgeons and epidemiologists at the UM Cardiovascular Center (CVC) and the Department of Cardiac Surgery. In addition, since 2007 she continues to collaborate with investigators at Duke Clinical Research Institute on research on cardiovascular disease. Her research in cardiovascular disease focuses on understanding variation in health care practices and quality improvement as well as evaluating the effect and impact of clinical practices and patient characteristics in terms of clinical outcomes and health cost. Much of her research relies on national or statewide clinical registries (e.g., the Michigan Society of Thoracic and Cardiovascular Surgeons Quality Collaborative, the Interagency Registry for Mechanically Assisted Circulatory Support) and health claims data (e.g., medicare claims data and Blue Cross and Blue Shield claims data). Another main area of her applied research is in renal disease and solid-organ transplantation. Since 2009, she is a member at the UM Kidney Epidemiology and Cost Center (KECC) and has been working on, for example Scientific Registry of Transplant Recipients (SRTR) project and several projects involving patients with End Stage Renal Disease (ESRD).
- Ph.D, Statistics, North Carolina State University, 2008
- M.A., Ecology, Duke University, 2004
- B.S., Environmental Science (minor: Computer Science), Peking University, 2001
Zhang, B. and Zhang, M. (2018) C-learning: a New Classification Framework to Estimate Optimal Dynamic Treatment Regimes. Biometrics, in press.
Zhang, B. and Zhang, M. (2018) Variable Selection for Estimating the Optimal Treatment Regimes in the Presence of a Large Number of Covariates. Annals of Applied Statistics, in press.
LaPar, D.J., Likosky, D.S., Zhang, M., Theurer, P., Fonner, C.E., Kern,J.A., Bolling, S.F., Drake, D.H., Speir, A.M., Rich, J.B., Kron, I.L., Prager, R.L., Ailawadi, G. (2018). Development of a Risk Prediction Model and Clinical Risk Score for Isolated Tricuspid Valve Surgery. The Annals of Thoracic Surgery, pii: S0003-4975(18)30036-5. doi: 10.1016/j.athoracsur.2017.11.077.
Liang, Q., Ward, S., Sinha, S., Pagani, F.D., Zhang, M., Kormos, R., Aaronson, K.D., Althouse, A., Kirklin, J.K., Naftel, D., Miller, M.A., and Likosky, D.S. (2018). Linkage of Medicare Files to the Interagency Registry of Mechanically Assisted Circulatory Support (INTERMACS): The Penetration of INTERMACS in US Durable Mechanical Circulatory Support Device Implant Activity. The Annals of Thoracic Surgery 105(5):1397-1402.
He, Z., Zhang, M., Lee, S., Smith, J.A., Kardia, S.L.R., Diez Roux, A.V., Mukherjee, B. (2017). Set-Based Tests for Gene-Environment Interaction in Longitudinal Studies. Journal of the American Statistical Association, in press, http://dx.doi.org/10.1080/01621459.2016.1252266.
Likosky, D.S., Zhang, M., Paone, G., Collins, J., DeLucia, A., Schreiber, T., Theurer, P., Kazziha, S., Leffler, D., Wunderly, D.J., Gurm, H.S., Prager, R.L. (2016) Impact of Institutional Culture on Rates of Transfusions During Cardiovascular Procedures: The Michigan Experience. American Heart Journal, 174:1-6. (doi: 10.1016/j.ahj.2015.12.019. PMID: 26995363)
He, Z., Zhang, M., Lee, S., Smith, J.A., Guo, X., Palmas, W., Kardia, S.L.R., Roux, A.V.D., Mukherjee, B. (2015). Set-based tests for genetic association in longitudinal studies. Biometrics, 71(3):606-15.
Zhang, M. (2015). Robust methods to improve efficiency and reduce bias due to chance imbalance in estimating survival curves in randomized clinical trials. Lifetime Data Analysis, 21(1),119-137. (doi: 10.1007/s10985-014-9291-y PMID: 24522498)
Zhang, M. and Wang, Y. (2013) Adjusting for observational secondary treatments in estimating the effects of randomized treatments. Biostatistics, 14(3),491-501 (PMID: 23349243)
Zhang, M. and Schaubel, D. E. (2012). Double-robust semiparametric estimator for differences in restricted mean lifetimes in observational studies. Biometrics, 68, 999-1009 (doi: 10.1111/j.1541-0420.2012.01759.x, PMID: 22471876, PMCID: PMC:3432755)
- American Statistical Association
- International Biometric Society, ENAR
- International Society of Clinical Biostatistics (ISCB)