Lec 9-10. Counfounding& effect modification Flashcards
What 3 aspects do epidemiologists evaluate on their study before declaring a real true association?
check for confounding/effect modification, bias and statistical significance.
What is a confounding variable?
3rd variable related to study subjects that distorts an association (RR/OR/HR) between exposure and outcome
Describe the relationship of a confounding variable and the exposure and outcome of interest
exposure—> outcome
^confounder ^
What is confounding by indication?
When a variable is a risk factor for a disease among non exposed persons and is associated with exposure of interest in population from which the cases derive, without being an intermediate step between the pathway of exposure and disease. Very common in observational studies
When does residual confounding arise?
2 ways, Mixing of effects or confusion effects
What does it mean by mixing of effects?
Exposure and outcome association distorted by mixing them with another 3rd factor that is also associated with outcome
What does it mean by confusion of effects?
Effects of exposure distorted, effect of 3rd factor mistaken for effect of exposure
To be a confounder a 3rd variable must be what?
Independently associated with exposure, Independently associated with outcome, and not directly in casual-pathway linking exposure to outcome.
What is the main way (and its sub categories) researchers use to attempt to control for counfounders before a study starts?
Study design stage and the subs are Randomization, Restriction, and matching
What is the main way (and its sub categories) researchers use to attempt to control for counfounders after a study starts?
Analysis of Data Stage and the subs are stratification (w/ weighting) and multivariate statistical analysis(regression analyses)
What is randomization and what are the strengths and weaknesses?
Letting random people in study. Strength: with big sample size will make groups equal. Weakness: Sample size may not be big enough, lots of unknown confounders.
What is Restriction and what are the strengths and weaknesses?
Set pre-specified categories of confounder. Strength: convenient, doesn’t impact negatively internal validity. Weakness: Narrow sample size, must be narrow for no confounders, can negatively impact EXTERNAL validity.
What is Matching and what are the strengths and weaknesses?
One person type in both groups. Strengths: Intuitive. Weaknesses: hard, time consuming, expensive, doesnt control for any confounders.
What is Stratification and what are the strengths and weaknesses?
Comparing exposure and outcome with in categories with in confounding variables. Strengths: Intuitive,straight forward. Weaknesses: Impractical for simultaneous control of multiple confounders.
What is Multivariate and what are the strengths and weaknesses?
Math factoring out effects of multiple Confounder’s. Strengths: Control multiple confounders, ORs can be obtained and interpreted . Weaknesses: People need to know how to interpret data.
