S4) Case-Control Study Design

Question 1

How does one conduct a case-control study?

Answer 1

• Identify a group of cases
• Identify a suitable group of non-cases (controls)
• Ascertain previous exposure status of everyone
• Compare level of exposure in cases and controls

Question 2

Identify 4 benefits of case-control studies

Answer 2

• Quick
• Cheap
• Require less detailed information
• Used for all outcomes (selected on the basis of having the outcome or not)

Question 3

How does one analyse a case-control study?

Answer 3

Question 4

How does one calculate the error factor for a case control study?

Answer 4

Question 5

How does one calculate the 95% confidence interval for a case-control study?

Answer 5

Confidence interval = (OR/ef) to (OR x ef)

Question 6

What is the ratio of controls to cases?

Answer 6

Usually, 4 - 6 x as many controls as there are cases because precision of an OR is affected by the number of controls and also cases

Question 7

Identify 3 primary issues for case-control studies

Answer 7

• Selection bias
• Information bias
• Confounding

Question 8

How can one deal with selection bias?

Answer 8

• Cases selected should be representative of all cases
• Controls selected should be representative of the population from which the cases came

Question 9

How can one deal with confounding?

Answer 9

• Minimise confounding in selection (matching)
• Adjust for confounding in analysis (adjusting)