What This Document Is
These are lecture notes from an Introduction to Epidemiology course for Nutritionists (M 6120) at Columbia University, specifically covering the critical concepts of confounding, bias, and interaction in epidemiological studies. The notes focus on understanding study validity – both internal and external – and how these concepts impact the interpretation of research findings. It directly relates to an upcoming paper assignment where students will critically evaluate existing nutrition research.
Why This Document Matters
This document is essential for nutritionists and students of nutrition who need to understand and interpret epidemiological research. It’s particularly valuable when evaluating the strength of evidence supporting nutritional recommendations or when conducting their own research. Understanding confounding and bias is crucial for determining whether observed associations between diet and health outcomes are genuine or due to other factors. The notes are designed to prepare students for a paper requiring critical analysis of published studies.
Common Limitations or Challenges
These notes provide a foundational overview of complex concepts. They do *not* offer detailed statistical methods for controlling confounding or bias, nor do they provide a comprehensive guide to study design. Users will still need to consult textbooks, additional readings, and potentially statistical software documentation to fully apply these concepts in practice. This preview does not cover all methods to control confounding, only introduces the concepts.
What This Document Provides
The full document includes:
* Definitions of internal and external validity.
* An explanation of confounding, including criteria for identifying confounders.
* Discussion of how to determine if an observed association is likely due to confounding or chance.
* An overview of methods to control confounding, including randomization, restriction, and matching.
* Guidance on applying these concepts to critically evaluate research papers.
* Specific examples to illustrate confounding (e.g., grey hair and mortality).
This preview provides a high-level overview of the topics covered and their relevance to nutritional epidemiology. It does *not* include detailed statistical formulas, specific examples of data analysis, or practice problems.