Case-Control Study
Definition: In a case-control study, 2 groups, known as cases and controls, are selected based on the presence and absence, respectively, of a disease/outcome of interest. The groups are then queried about various exposures that may have been a source of disease. Associations between exposures and outcomes are measured using odds ratios, which estimate the relative risk.
Steps
- Establish a case definition – what constitutes a case of disease?
- Identify cases of a disease
- Select a matching non-diseased population for controls
- Ask about exposures.
Note: The order of these events is crucial.
Cautions
- Sampling must be independent of exposure
- Controls are selected as a representative sample of the population that gave rise to the cases
Advantages to a Case-Control Study
- Great for rare diseases (requires smaller sample than cohort study)
- Relatively fast
- Relatively inexpensive
Disadvantages to a Case-Control Study
- Cannot calculate prevalence
- Inefficient for rare exposures
- Can only study one outcome
- Increased susceptibility to bias
- Sampling assumptions (selection bias)
- It is crucial to select cases and controls before gathering any information about exposures
- Recall/information bias (potential error in recalling exposure
- Case-patients may recall events differently that controls
- Case-patients may recall events differently that controls
- Sampling assumptions (selection bias)
Example: A local health department notices an increase in the number of cholera cases reported in Metrotown residents. In order to find out what is happening, a case-control study is conducted.
- A case definition for cholera in Metrotown is established, which includes a time frame for onset of symptoms.
- Controls are selected using a random sample of Metrotown residents. Because the health department does not know the cause of the disease, they do not match based on any criteria for fear that matching (even based on age and/or sex) would make them unable to analyze these criteria for risk or association.
- Once cases are identified and controls are selected, staff from the health department interview study participants to determine potential exposures. Odds ratios are then calculated to determine if any of the exposures that they considered were potentially associated with the disease.
