Summer Program Modules

**REGISTRATION FOR THE THE SUMMER PROGRAM
COMPARATIVE EFFECTIVENESS RESEARCH CLASS IS NOW OPEN**

Each July since 1999 the College of Public Health has trained hundreds of students, practitioners and researchers in applied biostatistical and epidemiological methods. This program brings highly regarded faculty from across the globe to provide hands-on education and training in a flexible short-course format. Registration information for the 2013 Summer Program will be available in early 2013.   

Attendees of the 2011 Summer Program Comparative Effectiveness Research course completed Modules 1-16, available on the CER Training Modules page, before attending the Summer Program.

Below are the 2011 Summer Program Modules, Modules 17-31. These modules are edited recordings of the 2011 Summer Program CER course discussions.   

Please see the bottom of the page for technical viewing requirements.

Module 17: Introduction to CER Using Observational Data

Instructor: Paul L. Hebert, PhD

This module, the first recording of the week-long Summer Program course, introduces Comparative Effectiveness Research. A historical context for CER and the foundations and fundamental challenges of this research are discussed. (26:56)

Module 18: Rubin’s Potential Outcome Framework

Instructor: Paul L. Hebert, PhD

This module reviews causal modeling. Important covered concepts include counterfactual outcomes, treatment heterogeneity, average treatment effects and treatment effects on the treated. (25:13)

Module 19: Can Quasi-Experiments Yield Causal Inferences?

Instructor: Matthew L. Maciejewski, PhD

This module outlines the quasi-experimental study. The benefits and limitations of quasi-experimental study designs are discussed. (33:49)

Module 20: Study Designs Appropriate for CER

Instructor: Matthew L. Maciejewski, PhD

This module reviews the elements of quasi-experimental studies. The lecture includes a discussion on cohort design and methods for preserving external and internal validity. (45:04)

Module 21: Defining the Treatment

Instructor: Paul L. Hebert, PhD

This module covers  identification of treatment in the observational CER study. This lecture also discusses the location of treatment data, specifically in Medicare datasets. (19:07)

Module 22: Does X Really Cause Y?

Instructor: Matthew L. Maciejewski, PhD

This module reviews the role of causality in quasi-experimental research. The effect of bias on quasi-experiments is also discussed. (16:53)

Module 23: Risk Adjustment

Instructor: Paul L. Hebert, PhD

This module describes methods for risk adjustment based on coding systems in administrative data. Covarites discussed include demographics, socio-demographics, comorbidity and severity. (42:43)

Module 24: Propensity Score Analysis for CER

Instructor: Matthew L. Maciejewski, PhD

This module outlines propensity score methodology. General principles guiding propensity score modeling are reviewed. A worked example illustrates  the process propensity score analysis and interpretation of the results. (67:32)

Module 25: Missing Data

Instructor: Paul L. Hebert, PhD

This module discusses the effect of missing data on research. Topics include types of missing data, missing data mechanisms and solutions to missing data problems. (78:42)

Module 26: Putting It All Together in a CER Analysis

Instructor: Paul L. Hebert, PhD

This module captures a discussion of the process to create a research publication. The example used is a work of the instructor. The discussion illustrates the application of study design, outcomse, covariates and methods principles in CER using Medicare data. (68:06)

Module 27: Instrumental Variable Exercise

Instructor: Paul L. Hebert, PhD

This module reviews instrumental variables exercises. The process of creating an instrument, testing the instrument’s correlation with treatment and confounding factors and methods for evaluting the instrumental variable estimate are discussed. (41:48)

Module 28: Unobserved Confounding Part 1

Instructor: Paul L. Hebert, PhD

This module begins with a discussion about appropriate use of propensity scores and instrumental variables and the influence of their fields of origin. Bias and efficiency, from the slides, are then reviewed. (40:26)

Module 29: Unobserved Confounding Part 2

Instructor: Paul L. Hebert, PhD

This module reviews randomization and selection bias. Worked examples for overcoming unobserved confounding and explanations of difference-in-difference estimators are inlcuded in this lecture. (44:36)

Module 30: Unobserved Confounding Part 3

Instructor: Paul L. Hebert, PhD

This module focuses on instrumenal variables. Properties and existing examples of good intrumental variables are discussed. (50:57)

Module 31: Unobserved Confounding Part 4

Instructor: Paul L. Hebert, PhD

This module revisits the existing examples of instrumental variables and presents a new worked example using instrumental variable analysis. (51:25)


Technical Information

To view a module, click the "View Module" button.

To print the slides corresponding to the lecture by clicking the "Print Slides" button.  

**Please note that to view instructional content, you will need to have the following free plug-ins installed on your computer:

- Microsoft Silverlight for viewing the modules, which is available here

- Adobe Reader for viewing printable slides and supplementary materials, which is available here

If you are still experiencing difficulty, please contact your network administrator.

 

Green Buckeye Certified CEPH CAHME