Biostatistics

11 lectures, Fall (online)

Course description: This course provides an introduction to Clinical Research. The student will learn how clinical research studies are designed, how to analyze and interpret statistical comparisons, and will develop skills in critically reading the clinical research literature. The course consists of 11 lectures and includes the following topics. Study Design I: The Research Question; Observational Studies; Study Design II: Clinical Trials; Biostatistics I: Descriptive and Analytic Statistics - Univariate Techniques; Biostatistics II: Analytic statistics - Multivariate Techniques; Biostatistics III: Survival Analysis; Research Ethics; Comparative Effectiveness Research; Confounding and Casual Inference in Observational Studies; Healthcare and Research Informatics; Genetic/Genomic Issues in Clinical and Translational Research; Cost-Effectiveness Analysis. The course will culminate with a Final Exam. This course is open to all faculty, staff, and students at Einstein and Montefiore.

14 lectures; Winter/Spring

Course description: This seminar course aims to introduce students to clinical research with a focus on epidemiology and study design. The course uses an introductory clinical research text, along with a critical assessment of papers from the scientific (clinical and epidemiologic) literature, in order to learn about study designs: their strengths and weaknesses and how such studies are conducted. Topics to be covered include: basic epidemiology, measures of association, basic statistics, cohort studies, case control studies, clinical trials, causal inference, and research ethics. Follows up on concepts introduced in the Clinical Research 101 lecture series, and provides a deeper dive into study design and interpretation. This course is open to all faculty, staff, and students at Einstein and Montefiore.

Spring

Course description: This 4-week course will meet three times a week for combined lecture/lab sessions to intro-duce the basic concepts and methods of biostatistics. Concepts include: fundamentals of probability; foundations of statistical inference, confidence intervals, hypothesis tests, and sample size and power calculations. Students will gain familiarity with the freely-available statistical software, R, to explore and analyze data. This course is for students enrolled in the Graduate Programs in the Biomedical Sciences.

Spring

Course description: This 2-week course will meet three times a week for combined lecture/lab sessions to introduce the basic concepts and methods of regression analysis. Topics include linear regression, the analysis of two-way tables, One-way and Two-way ANOVA, permutation tests, logistic regression. Introduces and employs the freely-available statistical software, R, to explore and analyze data. This course is pre-requisite for Quantitative Skills for the Biomedical Researcher III. This course is for students enrolled in the Graduate Programs in the Biomedical Sciences.

Spring

Course description: This 2-week course will cover the statistical principles that are pertinent to the study of big –omic data sets being collected in biology. Students will learn about current statistical approaches, issues related to experimental design and reproducible research, and important case studies that illuminate some of the challenges of analyzing big data. This course is the third module of the Quantitative Skills for the Biomedical Researcher series, and builds upon the material covered in the first two modules. As part of the assessment, students will gain practical experience by conducting a mini big data research project while working in small teams. It is expected that students will have completed Quantitative Skills for the Biomedical Researcher I and II, or have acquired this material through other means (please consult the course leader if in doubt). Programming skills in R is mandatory. This course is for students enrolled in the Graduate Programs in the Biomedical Sciences.

Course description: The overall objective of this course is to learn the definitions and meaning of P-values, confidence intervals, hypothesis testing and type I and type II errors; identify the appropriate statistical methods to test for associations between study variables and to know and be able to assess the assumptions of each method; interpret statistical analyses of their own and those presented in other papers; evaluate the findings of analyses with regard to potential clinical implications; evaluate the limitations and strengths of analytic findings. This course is for students enrolled in the Clinical Research Training Program.

Biostatistics II builds on the knowledge of univariate, bivariate and multivariate analyses learned in the “Summer Intensive” course and expands on concepts related to multiple linear regression. Both the lecture and the lab will focus on multiple linear regression model building, interpretation of results, diagnostic tests of assumptions, assessing for interaction, and statistical adjustment for confounding. This course is for students enrolled in the Clinical Research Training Program.

Biostatistics III consists of a total of 14 lectures/labs which will be taught in two 7-week modules. The first module will cover logistic regression and the second module will cover survival analysis. The course objectives are to learn the basics and applications of logistic regression in assessing associations between exposure/explanatory variables and a dichotomous outcome variable; learn fundamental methods in analyzing time to event data using survival analysis, especially Cox proportional hazards modeling; use STATA software to conduct both logistic regression and survival analysis and to be able to interpret the statistical output related to these modeling techniques. This course is for students enrolled in the Clinical Research Training Program.

This course presents modern approaches to the analysis of longitudinal data. Topics include design of longitudinal studies, generalized linear models for correlated data (including generalized estimating equations, generalized linear mixed effects model), computational issues and methods for fitting models, and missing data issues. STATA statistical software will be used in the data analysis component of this course where the students will learn how analyze and interpret linear models for repeated measure continuous and discrete data. This course is for students enrolled in the Clinical Research Training Program.