Case Control Studies

Interviewer Bias

If the interviewer knows who the cases are, this may influence the way the information is interpreted

Are case control studies suitable for looking at rare diseases?

Yes

Disadvantages of case control studies (5)

1) Can't calculate prevalence as the sample isn't random 2) Can't assess incidence as the data is from one point in time 3) Bias can be a problem 4) Can't assess causation as the data is collected after the outcome occurred 5) Can only study association between exposure(s) and one outcome

How to define a control group (3)

1) Controls must not have the disease under study, or a diseased related to the exposures under study 2) Same eligibility criteria as case group 3) Controls need to be as similar to case group as possible, i.e. from same hospital/community

What is a case control study? (3)

1) Patients are selected based on the presence or absence of disease 2) An exposure of interest is then measured in both groups and compared 3) The effect of exposure on the risk of disease is calculated using OR

How to define a case group (3)

1) Precise case definition criteria 2) Eligibility criteria 3) Source of cases

Advantages of case control studies (3)

1) Reduces number of people needing to be studied 2) No follow up over time = relatively cheap and easy 3) Allows multiple exposures to be studied

Selection Bias

Cases and controls are not representative of people with/without the disease

Response Bias

Exposure information is reported differently between case and control groups (demand characteristics)

Systematic Sampling

Sampling sequentially based on the nth person

Random Sampling

Sampling where people are identified from a list of all potential participants

Matched Sampling

Sampling where the control is paired with the case (i.e. age, sex, hospital)

Recall Bias

Selective memory of issues related to exposure/outcome (i.e. those who are ill may remember the exposure better)