Biostatistics

• SPH BS 853: Generalized Linear Models with Applications
  This course introduces statistical models for the analysis of quantitative and qualitative data, of the types usually encountered in health science research. The statistical models discussed include: Logistic regression for binary and binomial data, Nominal and Ordinal Multinomial logistic regression for multinomial data, Poisson regression for count data, and Gamma regression for data with constant coefficient of variation. All of these models are covered as special cases of the Generalized Linear Statistical Model, which provides an overarching statistical framework for these models. We will also introduce Generalized Estimating Equations (GEE) as an extension to the generalized models to the case of repeated measures data. The course emphasizes practical applications, making extensive use of SAS for data analysis.
• SPH BS 854: Bayesian Methods in Clinical Trials
  Bayesian statistical methods use prior information or beliefs, along with the current data, to guide the search for parameter estimates. In the Bayesian paradigm probabilities are subjective beliefs. Prior information/ beliefs are input as a distribution, and the data then helps refine that distribution. The choice of prior distributions, posterior updating, as well as dedicated computing techniques are introduced through simple examples. Bayesian methods for design, monitoring analysis for randomized clinical trials are taught in this class. These methods are contrasted with traditional (frequentist) methods. The emphasis will be on concepts. Examples are case studies from the instructors' work and from medical literature. R will be the main computing tool used.
• SPH BS 855: Bayesian Modeling for Biomedical Research & Public Health
  The purpose of this course is to present Bayesian modeling techniques in a variety of data analysis applications, including both hypothesis and data driven modeling. The course will start with an overview of Bayesian principles through simple statistical models that will be used to introduce the concept of marginal and conditional independence, graphical modeling and stochastic computations. The course will proceed with the description of advanced Bayesian methods for estimation of odds and risk in observational studies, multiple regression modeling, loglinear and logistic regression, latent class modeling including hidden Markov models and application to model-based clustering, graphical models and Bayesian networks. Applications from genetics, genomics, and observational studies will be included. These topics will be taught using real examples, class discussion and critical reading. Students will be asked to analyze real data sets in their homework and final project.
• SPH BS 856: Adaptive Designs for Clinical Trials
  An adaptive design is a clinical trial design that allows modification to aspects of the trial after its initiation without undermining the validity and integrity of the trial. Adaptive designs have become very popular in the pharmaceutical industry because they can increase the probability of success, considerably reduce the cost and time of the overall drug development process. With a recent rapid development in this area, there is a high demand for statisticians proficient in designing and conducting adaptive clinical trials. Students will learn different (both frequentist and Bayesian) adaptive designs and gain hands-on experiences on adaptive randomization, adaptive dose-finding, group sequential, and sample-size reestimation designs.
• SPH BS 857: Analysis of Correlated Data
  The purpose of this advanced seminar is to present some of the modern methods for analyzing tricorrelated observations. Such data may arise in longitudinal studies where repeated observations are collected on study subjects or in studies in which there is a natural clustering of observations, such as a multi-center study of observations clustered within families. Students start with a review of methods for repeated measures analysis of variance and proceed to more complicated study designs. The course presents both likelihood-based methods and quasi-likelihood methods. Marginal, random effects and transition models are discussed. Students apply these methods in homework assignments and a project.
• SPH BS 858: Statistical Genetics I
  This course covers a variety of statistical applications to human genetic data, including collection and data management of genetic and family history information, and statistical techniques used to identify genes contributing to disease and quantitative traits in humans. Specific topics include basic population genetics, linkage analysis and genetic association analyses with related and unrelated individuals.
• SPH BS 859: Applied Genetic Analysis
  Statistical tools such as linkage and association analysis are used to unravel the genetic component of complex disease. Investigators interested in the genetic analysis of complex traits need a basic understanding of the strengths and weaknesses of these methodologies. This course will provide the student with practical, applied experience in performing linkage and association analyses, including genome-wide analyses. Special emphasis is placed on understanding assumptions and issues related to statistical methodologies for genetic analysis to identify genes influencing complex traits. Students will use specialized genetics software for homework assignments.
• SPH BS 860: Statistical Genetics II
  This course covers current topics in statistical genetics, with emphasis on how statistical techniques can be used with various types of genetics data for mapping genes responsible/contributing to complex human diseases. Topics such as genetics map functions, gene mapping in experimental organisms, advanced linkage analysis methods, statistical approaches for the analysis of genome-wide high density SNP scans in unrelated and family samples will be discussed.
• SPH BS 861: Applied Statistics in Clinical Trials II
  This course covers a variety of biostatistical topics in clinical trials, including presentation of statistical results to regulatory agencies for product approval, analysis of safety data, intent-to-treat analyses and handling of missing data, interim analyses and adaptive designs, and analyses of multiple endpoints. Upon completion of the course, students will be able to make and defend decisions for many study designs and for issues faced when analyzing efficacy and safety data from clinical trials. Students will also be able to present, in a written format following standard guidelines accepted by the clinical trials' community, results of such efficacy and safety analyses to the medical reviewers and statistical reviewers of regulatory agencies.
• SPH BS 940: Culm Exp Biost
  

Back to full list of Courses