Summer Institute for Training in Biostatistics (SIBS)
 Introduction to Biostatistics   (3 credits) Consider the following three experimental scenarios: In a clinical trial designed to test the efficacy of a new drug, participants are randomized to either a control arm (e.g., a standard drug or placebo) or a treatment arm, and carefully monitored over time. After the study ends, the two groups are then compared to determine if the differences between them are "statistically significant" or not. By formulating an explicit mathematical model, an investigator wishes to describe how much variation in a response variable, such as blood pressure in a group of individuals, can be deterministically explained in terms of one or more "statistically significant" predictor variables with which it is correlated. In a longitudinal study of a cohort of individuals, the strength of association between a disease such as COPD (Chronic Obstructive Pulmonary Disease), and exposure to a potential risk factor such as smoking, is estimated and determined to be "statistically significant." The overall goal of this course is to provide students an overview of fundamental statistical concepts, and a practical working knowledge of the basic statistical techniques they are likely to encounter in applied research. In complex dynamic systems such as biological organisms, it is important to distinguish genuine - or "statistically significant" - sources of variation, from purely "random chance" effects. This first course is an introduction to the basic but powerful techniques of statistical analysis - techniques which formally implement the fundamental principles of the classical scientific method - in the general context of biomedical applications. Students will learn how to: formulate a hypothesis about some characteristic of a variable quantity measured on a population (e.g., mean cholesterol level, survival time after disease diagnosis) classify different designs of experiment that generate appropriate sample data (e.g., randomized clinical trials, case-control studies); investigate ways to explore, describe and summarize the resulting observations (e.g., visual displays, numerical statistics); conduct a rigorous statistical analysis (e.g., by comparing the empirical results with a known reference obtained from Probability Theory), and finally; infer a conclusion (i.e., whether or not the original hypothesis is rejected) and corresponding interpretation (e.g., whether or not there exists a genuine "treatment effect"). These important biostatistical techniques form a major component in much of the currently active research that is conducted in the health sciences, such as the design of safe and effective pharmaceuticals and medical devices, epidemiological studies, patient surveys, and many other applications. Lecture topics and exams will include material on: Exploratory Data Analysis of Random Samples Probability Theory and Classical Population Distributions Statistical Inference and Hypothesis Testing Survival Analysis Regression Models For more detailed course information, please see the Introduction to Biostatistics - Course Syllabus.