Interpretation of Epidemiological Findings Pt.2

Question 1

What is confounding?

Answer 1

effect of an extraneous variable that wholly or partially accounts for the apparent effect of the study exposure or that masks an underlying true association

Question 2

What are ways to potentially identify confounders?

Answer 2

subject matter knowledge. stratification, test by: associated with exposure in the source population; associated with the outcome in the absence of the exposure; not a consequence of the exposure, compare crude and adjust statistical estimates

Question 3

List examples of potential confounders.

Answer 3

birth order and Down’s syndrome (maternal age), alcohol consumption and lung cancer (smoking), playing games of cognition and reduced levels of dementia

Question 4

What is effect modification?

Answer 4

strength of association between of the primary exposure on an outcome differs depending on the level of a third variable

Question 5

Should you control for effect modification?

Answer 5

No

Question 6

How do you present effect modification?

Answer 6

present a stratified analysis

Question 7

How to statistically test for effect modification?

Answer 7

Breslow-Day test
Q test
Interaction terms in regression models

Question 8

What term is considered equivalent to effect modification?

Answer 8

interaction

Question 9

What is the meaning of the term synergism?

Answer 9

effect modifier potentiates the effect of the exposure

Question 10

What is the meaning of the term antagonism?

Answer 10

effect modifier diminishes the effect of the exposure

Question 11

List examples of effect modification.

Answer 11

post menopausal women using HRT/not > risk reduction for endometrial cancer (effmod: only for BMI less 30)