Clinical&Translational Science: Multi Analysis Clin&Trans Res

Introduction to multivariate regression; models that account for effects of multiple predictors on a single outcome, including linear and logistic regression and survival analysis. Prerequisite: CTS 6200 or Instructor permission.

Prereqs: CTS 6200 or equivalent; or instructor permission; Online synchronous; Open to Degree and PACE students

CTS 6250 covers multivariable analysis using linear regression, logistic regression, and Cox proportional hazards regression. This course is designed for fellows, junior faculty, and others wishing to develop skills in performing multivariable analysis. Though the course is designed to be very conceptual, the challenging nature of multivariable analysis requires students be comfortable with basic statistics. Datasets and software will be provided. Course objectives: Upon completion of the course, the student will: 1. Choose the correct multivariable regression method to test a hypothesis 2. Perform logistic, linear, and survival regression 3. Determine and perform the correct modeling strategy to adjust for confounding, prediction, etc 4. Assess if specific statistical assumptions are satisfied, and if not, determine the appropriate alternative analysis 5. Interpret the results of the analysis

Classes will provide a small group interactive seminar approach. Student analyses of sample datasets will be completed on a weekly basis. Prior to class participants will complete a homework assignment. Homework assignments may include answering questions, solving problems, and reading papers. Each class session will consist of a question and answer period, review of the previous homework, lecture, and a statistical laboratory. The course will use Stata statistical software, which is available to all UVM students for Microsoft Windows and MacIntosh OS through the UVM software download site. The required text for Spring 2023 is: Vittinghoff E, Glidden DV, Shiboski SC, McCulloch CE. Regression Methods in Biostatistics: Linear, Logistic, Survival, and Repeated Measures Models, 2nd ed. New York, NY: Springer Science+Business Media, Inc., 2012. ISBN 978-1461413523 Be sure to get the 2nd edition.

Student performance will be evaluated in three areas of the course: • 25% Participation and discussion • 25% Paper • 50% Student presentations Students are expected to attend class, read all assigned materials in advance of class meetings, generate weekly analyses, and participate in class discussions. Weekly analytical homework, readings, and journal articles will be presented and assigned each class period. The student will work with a partner to complete three special analyses during the course - one each of linear, logistic, and survival analysis. Each team will present their work to the whole class. Students will write a paper evaluating the analytic methods of a journal article handed out in class.