PhD in Biostatistics

Biostatistics involves the development and application of sound statistical and mathematical principles to research in the health sciences. Because original theoretical research in biostatistics flows from medical research, the foundations of methodological development must be firmly grounded in sound principles of statistical inference and a thorough knowledge of the substantive area that provides the source of the medical questions being addressed. Thus, the Department of Biostatistics encourages excellent methodological work that is motivated by sound science that includes but is not limited to active collaborations with other investigators.

Research collaborations for biostatisticians take place both within and across departments in the School of Public Health, as well as with other departments in the School of Medicine and the University at large. Areas of current research include the development of general methods that have wide applicability across different areas of health research, as well as more specific techniques for dealing with the underlying processes that give rise to the data of interest. A broad range of health topics addressed by students in this department includes chronic diseases such as cancer, genetic epidemiology, clinical research, and mathematical models for infectious diseases.

Graduates of the doctoral program in Biostatistics are employed in universities throughout the country, as well as in such dedicated research institutions such as the National Institutes of Health. In addition, graduates have pursued careers in the pharmaceutical industry, in which they are actively involved in the evaluation of new therapeutic strategies.

This program requires General GRE test scores. Please send them to institution code 3987 (there is no department code).

All courses are 1 unit unless otherwise noted.

The PhD degree requires a total of 15 course units. Course substitutions (other than those noted here) must be identified and approved by the student’s adviser and the DGS.

PhD Required Courses (7 course units)

PhD Elective Courses (8 course units)

Course offerings subject to change.

Research Experience

In a number of courses, students gain actual experience with various aspects of research including preparation of a research grant, questionnaire design, preparation of a database for analysis, and analysis and interpretation of real data. In addition, doctoral students can gain research experience by working with faculty members on ongoing research studies prior to initiating dissertation research, which includes but is not limited to BIS 699. During the summer following each year of course work, candidates are required to take a research rotation that is approved by the department and communicated to the DGS.

The Dissertation

The Department strives for doctoral dissertations that have a strong methodological component motivated by an important health question. Hence, the dissertation should include a methodological advance or a substantial modification of an existing method motivated by a set of data collected to address an important health question. The dissertation must also include the application of the proposed methodology to real data. A fairly routine application of widely available statistical methodology is not acceptable as a dissertation topic. Candidates are expected not only to show a thorough knowledge of the posed health question, but also to demonstrate quantitative skills necessary for the creation and application of novel statistical tools.

• Conduct independent research in the theory and/or application of biostatistics.
• Utilize critical thinking to select from a variety of analytical and computational tools to test statistical hypotheses, interpret results of statistical analyses and use these results to make relevant inferences from data.
• Demonstrate teaching and presentation skills to effectively communicate biostatistics theory and applications across a wide range of complex biomedical or public health problems.

• Enhanced Polygenic Risk Prediction Through Integrative Analysis of Sex-Specific Information and Local Genetic Correlations
  
• Causal Inference for Effects of Interventions in Groups
  
• Advancing Genetic Basis for Complex Phenotypes – Epidemiological Genetic Mediation Analysis, Gene-Environment Interaction Analysis and Their Applications
  
• New Statistical Methods for Mediation Analysis with Independent and Correlated Data
  
• Causal Inference for Infectious Disease Intervention Effects
  
• Bayesian Methods to Integrate Multimodal Biomedical Data for Biomarker and Subpopulation Identification
  
• Addressing Bias in Causal Effects Estimated Under Misspecified Interference Sets, With Application to HIV Prevention Trials
  
• Methods for Investigating Cell-Type Specificity and Genetic Regulation of DNA Methylation in Complex Traits
  
• Network Analysis of Disease Outcomes and Comparative Effectiveness Research via Mining Registry and Administrative Claims Data