Case-Control Studies

Case-control studies are defined by their starting point: the disease outcome. Cases are individuals who have developed the disease; controls are individuals from the same source population who have not. The investigator then asks retrospectively whether exposure was more frequent in cases than controls.

Because participants are selected by outcome status rather than exposure, incidence cannot be directly calculated. The measure of association is the odds ratio (OR): OR = (a×d)/(b×c) from a 2×2 contingency table. The null value is 1. For rare diseases, OR approximates RR (the rare disease assumption); for common diseases, OR overestimates RR.

Case-control studies are ideally suited to: rare diseases or conditions with long latency (cases are already identified); transient exposures; and situations where multiple exposures need to be evaluated for a single outcome. They are smaller and faster than cohort studies.

Case selection: incident (newly diagnosed) cases are preferred over prevalent cases. Prevalent cases have survived with disease longer, introducing length bias — severe or rapidly fatal cases are under-represented. Only one outcome per study can be examined.

Control selection requires that controls come from the same source population as cases. Index dates — the date on which the matched case developed the outcome — are assigned to controls to ensure exposure assessment periods are equivalent. This prevents artificial differences in exposure opportunity.

Exposure measurement must be identical in cases and controls: same questionnaire, same trained interviewers. Differential measurement leads to information bias. Recall bias — the most important form — occurs when cases are more likely to remember and report past exposures because they are seeking explanations for their illness. Blinding interviewers to case/control status reduces this risk.

Comparing cross-sectional to case-control: cross-sectional studies use prevalent cases and measure exposure and outcome simultaneously (suitable for stable conditions); case-control studies use incident cases and measure past exposures retrospectively (suitable for rare outcomes and transient exposures).