Lecture 9 - Case Control Study Flashcards
What type of study is a case control study?
Observational analytical study
What is the difference between the case and control group?
Case has even, outcome and effect
Control has no even, no outcome and no effect
What is a case control study?
Where cases (those with events) are selected, then controls are chosen and selected to match each case
In a case control study, how do you identify the case population and select the cases? (step1)
Identify the case population then select the cases (those with event) irrespective of exposure status
Example: Smoking (exposure) and lung cancer (event)
How do you identify the control population in a case control study? (step 2)
Do a nested case control if cases and controls share the same population
What is step 3 of case-control studies?
How is this done?
When you select you controls, Case-control matching
1 or multiple controls are match to each case based on some matching criteria ok similar age, same sex
Like Case 1 = Lung cancer 28yo M
Control 1 = No lung cancer 29yo M
To see examples check slide 11 and 12
What is a nested case control study?
When the populations that are being studied are the same (e.g Hypothesis = long term exposure to DDT (Exposure) associated with dementia (event). With dementia in City X and without dementia in City X)
What is the strength of a case control study?
Cheap (since event has already occured)
Efficient when the event is:
-rare (rare condition)
-conditions take long time to develop
Allows studying multiple exposures
Can have smaller sample size
What are the weaknesses of a case-control study?
Cant determine incidence and prevalence (num of cases and controls is fixed)
Not suitable for rare EXPOSURES (cases selected irrespective of exposure status)
Potential selection bias in controls
What is the general premise of a case-control study?
It selects the cases (those with the event) no the controls (those with the exposure)
Is there always recall bias in case-control studies?
Not always but it is possible
The actual bias depends on the actual study design and nature of the data of the research
