A case-control study is an observational study that compares people who have a particular outcome or condition with people who do not.
Researchers typically look backward for differences in past exposures, behaviors, or characteristics that may be associated with the outcome.
Case-control studies are commonly used in medicine, public health, epidemiology, and other fields that investigate potential risk factors for diseases or other outcomes.
How do case-control studies work?
Researchers begin by identifying a group with the outcome of interest, called cases, and a comparison group without the outcome, called controls. Controls are ideally selected from the same population as the cases. Researchers then examine past information to determine whether certain exposures or characteristics were more common in one group.
A case-control study may involve:
Identifying participants with and without a particular outcome
Reviewing past exposures, behaviors, or characteristics
Using medical records, surveys, interviews, or other existing data
Comparing how frequently an exposure occurred in each group
Examining associations between potential risk factors and the outcome
For example, researchers might compare people with a particular disease to people without it and investigate whether they differed in previous environmental exposures, medications, or lifestyle factors.
Strengths of case-control studies
Case-control studies allow researchers to:
Study rare diseases or outcomes
Investigate outcomes that may take a long time to develop
Investigate multiple potential risk factors for the same outcome
Use existing records or previously collected information
Conduct research without following participants for long periods
Identify associations that may warrant further investigation
Case-control studies can be especially useful for exploring potential causes or risk factors when an outcome is uncommon or has a long period between exposure and development.
Limitations of case-control studies
Because researchers often rely on past records or participants’ memories, information about previous exposures may be incomplete or inaccurate. Participants may also remember past exposures differently depending on whether they have the outcome being studied, which can introduce recall bias.
Selecting an appropriate control group can also be challenging. If cases and controls are selected differently or do not come from comparable populations, selection bias may affect the results. Differences between cases and controls other than the exposure being studied may also influence the findings.
Because these studies are observational, they can identify associations but cannot by themselves establish that an exposure caused an outcome.