PH717 Module 11 - Confounding and Effect Measure Modification
Contents
Page 1
Introduction
Learning Objectives
Page 2
What Is Confounding?
Conditions for Confounding to Occur
Effects of Confounding
Page 3
Down Syndrome and Maternal Age
Page 4
Three Methods for Minimizing Confounding in the Study Design Phase
Randomization in a Clinical Trial
Strengths of Randomization
Limitations of Randomization to Control for Confounding
Restriction of Enrollment
Drawbacks of Restriction
Matching Compared Groups
Advantages of Matching
Drawbacks of Matching
Page 5
Adjusting for Confounding in the Analysis
Standardization to Control for Confounding
Age-Specific Mortality Rates
An Operational Way to Identify a Confounding Factor
The 10% Rule for Confounding
Page 6
Stratification to Control for Confounding
Residual Confounding
The Mantel-Haenszel Pooled Estimate
The Magnitude of Confounding
Analyzing the Physical Activity-CHD Problem with Standardization
Page 7
Other Sources of Confounding
Confounding by Indication
Reverse Causality
Page 8
Effect Measure Modification
A Combination of Effect Measure Modification and Confounding
Effective Measure Modification is a Judgment
Summary of Stratified Analysis
Some Hypothetical Practice Problems
return to top
