lecture 27 - confounding

Question 1

what is stratification

Answer 1

calculating measure of association for each stratum of potential confounder and comparing them

Question 2

what is the basic recipe for stratification

Answer 2

1. calculate measure of association between exposure and outcome - before done anything to control for confounding
2. divide your potential confounder into strata (levels)
3. for each stratum calculate the measure of association between the exposure and outcome (stratum specific measure of association)
4. compare the stratum specific measures of association to the pre measure of association

Question 3

what are the pros to stratification

Answer 3

• easy for small number of potential confounders with limited strata
• can evaluate impact of confounding
• can identify effect of modification

Question 4

what are the limitations to stratification

Answer 4

• can leave residual confounding
• not feasible when dealing with lots of potential confounders with many strata

Question 5

what is multivariable analysis

Answer 5

statistical method for estimating measure of association whilst controlling for multiple potential confounders

• can work in situations when stratification cant

Question 6

when might standardisation be used

Answer 6

• when age structures are different
• when disease risk varies by age

Question 7

how much chance indicates confounding

Answer 7

general guideline is if controlling or confounding changes the measure of association by 10% or more from the crude value

Question 8

what is effect modification

Answer 8

the association between exposure and outcome differs across strata of the effect modifier