Course Objectives

Classes for ID 207 will be held Monday through Friday from 8:30 am to 5:00 pm.

Course Objectives

At the completion of the course students will be:

1)    Critical consumers of the public health and medical literature by understanding the basic principles and methods of epidemiology, including disease (outcome) measures, measures of association, study design options, bias, confounding, and effect modification.

2)    Able to interpret descriptive epidemiologic results in order to develop hypotheses about possible risk factors for a disease.

3)    Able to design valid and efficient studies to address public health and clinical problems.

4)    Able to organize, summarize, and display quantitative data.

5)    Comfortable with statistical methods for calculating summary estimates, measures of variability, and confidence intervals.

6)    Aware of and able to manipulate probabilities and the Normal and Binomial distributions.

7)    Able to carry out and interpret a variety of tests of significance, including two-group comparisons using t-tests, Wilcoxon tests, chi-square tests, Fisher exact tests, log-rank test, McNemar’s test, ANOVA, and Kruskal-Wallis Test.

8)    Familiar with power and sample size calculations.

9)    Familiar with basic principles and uses of linear and logistic regression models for clinical research.

10)  Able to carry out simple data analyses using the JMP program.

Texts and Reading Materials

There are two statistical textbooks and one epidemiology textbook for this course available online or at the Harvard COOP in Cambridge, MA (1400 Mass Ave, Cambridge, MA).

Biostatistics Text: There are two statistical textbooks available, but neither book is required for this course, since the material presented during lecture and the accompanying notes will be sufficient for the course. However, students may find it helpful as adjunct reading to get one book or the other. The first suggestion is Fundamentals of Biostatistics, by B. Rosner, from Duxbury Press. This text is very comprehensive and would serve as a useful reference throughout an investigator’s research career. However, it is very detailed and may be hard to navigate when first learning the material.

The other textbook, Basic and Clinical Biostatistics, by B. Dawson and R.G. Trapp, McGraw-Hill, may also be useful for alternative explanations of the methods covered in class. It is more reader-friendly but less comprehensive.

Epidemiology Text: The epidemiology textbook is Rothman, KJ. Epidemiology: An Introduction. (Second Edition) Oxford University Press, New York. 2012. Other class readings material will be available on the course website, including journal articles, citations, and weblinks. Please review the syllabus to determine which readings are required and which are optional.

Other textbooks that a student may already own are probably also acceptable, since the material covered during the course is basic and included in most introductory texts.

Statistical computing will be done using the JMP program.

Outcome Measures

Class Participation:  Class participation and discussion add greatly to the written notes that are provided. Students are expected to attend and participate in all classes.

Homework:  There will be two graded assignments that will each count toward 10% of your grade (20% for both combined). Each student must hand in their own assignment with their name on it, but they may work in groups or take advantage of the office hours to help solve the problems.

Data Analysis Exercise: All students are required to perform a statistical analysis on an epidemiologic data set. The data set will be available for analysis in computer labs during the third week of the program. All students will be required to submit a summary of their analysis that describes the analysis plan and summarizes the results. This summary should be no longer than two pages in length. In addition, no more than 3 tables can accompany each summary. Students must hand in their own assignment with their name on it and reflecting their own work on this project, but individuals may discuss the results with other students in the course. The summary and tables will be handed in at the same time as the take-home final exam (described below) and  will count toward 25% of the grade.

Final Exam:  During the last week of the course, each student will receive a take-home exam with approximately one week to complete it. This exam must be done individually with no collaboration with others. The exam will count toward 25% of the grade.

Student Projects:  Each student is required to design a research study proposal. The development of these proposals is a central activity of this course. In the month preceding the course, each student will submit a one-page description of a research question, background, and very rough design of a study that addresses a research question of interest to that student. At the end of the first week, we will have a student project workshop where students will have 15 minutes to discuss their initial ideas with faculty. There will, in addition, be 30-minute individual “office hours” with faculty to help students develop their proposals. At the end of the third week, there will be a class where several students will describe their proposals in 30-minute PowerPoint presentations. All students will submit written 4-page proposals for grading and comments. These proposals will count toward 20% of your grade.

Critique of Literature Seminar

One Critique of Literature Seminar is scheduled for the last week of the course. The class will be divided into smaller groups to review selected papers. The papers discussed on this day cover a wide variety of topics. All papers will be located on the course website.

Additional Information

Classes will be recorded each day on video and accessible on the course website.

Students who do not wish to use the JMP program may use any alternative package. However, the instructor and teaching assistants may not be able to provide any help or guidance. The student will still be responsible for carrying out all of the same analyses for homework assignments and the final exam.

Week 1

Monday
Overview of Program
Biostatistics: Introduction, Philosophy, Vocabulary (Rosner 1)
Epidemiology: Introduction, Outcome Measures (Rothman 1-3)
Probability: Concepts, Terminology, Rules (Rosner 3)
Optional Office Hours

Tuesday
Graphical Displays of Data (Rosner 2)
Measures of Effect (Rothman 1-3)
Summary Statistics (Rosner 2)
Computer Lab: Summary Statistics & Graphs

Wednesday
Normal Distribution and Central Limit Theorem (Rosner 5,6)
Study Design: RCT (Rothman 4)
Tests, P-values and Confidence Intervals (Rosner 6,7)
Optional Office Hours

Thursday
Two-Sample T-Test (Rosner 8)
Study Design: Cohort Studies (Rothman 4)
Confidence Intervals for Comparing Two Means (Rosner 8)
Homework 1 Distributed
Optional Office Hours

Friday
Study Design: Case Control Studies (Rothman 4)
Two-Sample Wilcoxon Rank Sum Test (Rosner 9)
Project Workshop
Computer Lab: Crude Analysis – Continuous Outcome

Week 2

Monday
Binomial and Poisson Distributions (Rosner 4)
Study Design: RCT Redux and Time Trend Designs (Notes)
Inference on One Proportion (Rosner 7)
Optional Office Hours

Tuesday
Comparing Two Proportions (Rosner 10)
Confounding & Bias (Rothman 5,8)
Fisher’s Exact Test and Chi-Square Test (Rosner 10)
Computer Lab: Crude Analysis – Binary Outcome

Wednesday
Stratified Analysis (Rothman 8)
Effect Modification (Rothman 9)
Power and Sample Size (Rosner 8,10)
Computer Lab: Power and Sample Size
Homework 1 Due
Homework 2 Distributed

Thursday
Contingency Tables (RxC) (Rosner 10)
Project Workshop
Chi-Square Trend Tests (Rosner 10)
Computer Lab: Stratified Analysis

Friday
Paired Binary Data: Kappa & McNemar’s Test (Rosner 10)
Matching (Notes)
Paired Continuous Data: T-Test & Wilcoxon (Rosner 8,9)
Optional Office Hours

Week 3

Monday
Analysis of Variance & Krusal-Wallis Test (Rosner 12)
Test Evaluation & Screening (Notes)
Pairwise Comparisons (Rosner 12)
Optional Office Hours
Homework 2 Due
Take-Home Final Distributed

Tuesday
Pearson and Spearman Correlations (Rosner 11)
Overview of Regression in Clinical Research (Rothman 10)
Linear Regression (Rosner 11)
Computer Lab 1: Analysis of Clinical Data Exercise

Wednesday
Logistics Regression (Rosner 13)
Propensity Scores (Notes)
Overview of Causal Inference (Notes)
Optional Office Hours

Thursday
Clinical Prediction Rules (Notes)
Computer Lab Demonstration: Regression analysis
Critique of the Literature
Optional Office Hours

Friday
Presentation of Project Designs
Presentation of Project Designs
Wrap-up
Optional Office Hours