Case Control Studies

Question 0

Describe the limitations and assumptions inherent in case control study designs

Answer 0

Selection bias - have to ensure controls represent the general population and cases represent all cases
Recall bias - exposure status incorrectly determined
Not good for rare exposures

Question 1

Explain how to organise a case control study

Answer 1

Recruit disease free (controls) individuals and diseased (cases) individuals
Determine their exposure status to causative agent by asking them
Compare level of exposure in cases and controls
Calculate odds ratio (cross multiply) then error factor to determine confidence intervals

Question 2

Explain a nested case control study

Answer 2

Case control study is within a cohort study
Can calculate incidence rates, population for control sample defined, can collect more detailed information for minority of participants

Question 3

What would be the effect on the odds ratio if the cases understated more than the controls

Answer 3

Decreased odds ratio - underestimating the true effect

Question 4

What would be the effect on the odds ratio if the controls understated more than the cases

Answer 4

Increased odds ratio - overestimating the true effect

Question 5

What would be the effect on the odds ratio is both cases and controls understate randomly

Answer 5

Shrinkage to the null

Question 6

What is the difference between bias and confounding?

Answer 6

Bias - the effect someone else has on the study (a problem with the study)
Confounding - exists in the population e.g. age, sex