Case-control study
Which prior exposures differ between cases and sampled controls?
When to use this project
Use a case-control study when selecting people by outcome status makes investigation efficient, especially for an uncommon outcome or costly exposure ascertainment. Controls must represent the exposure experience of the population that produced the cases.
How it differs
A cohort starts with eligibility and follows events; a case-control study samples cases and controls then reconstructs relevant exposures. A nested case-control study additionally draws controls from a defined cohort’s event-time risk sets. Control sampling determines the interpretation of the odds ratio.
Decisions before you start
- Choose incident or prevalent cases and validate the outcome.
- Define cumulative, incidence-density or other control sampling before selecting controls.
- Specify matching, exposure reference dates and recall or measurement bias; matching alone does not remove confounding.
How the flows work together
Each flow below links to its diagram, steps and tables. Flow order organizes the project; it does not imply that all work is strictly sequential. Data assessment, analysis planning and protocol refinement may inform one another.
1. Case-control overview
Overview · Question through exposure association reporting
Use this overview to state the purpose and link to detailed decisions. Keep a versioned protocol and decision register, and add evidence references for the rationale.
2. Case and control selection
Selection · Define cases and sampling frames
The source population links cases to valid controls. Control sampling and index-date assignment precede analysis and determine what the association estimates. Matching is optional and must be reflected in the analysis.
3. Exposure and covariate ascertainment
Data · Measure comparable prior exposure histories
Use comparable exposure windows and ascertainment procedures in cases and controls. Assess recall, recording quality and changes in exposure caused by early symptoms; a lag may reduce but cannot prove removal of reverse causation.
4. Sampling and analysis plan
Statistics · Account for matching and sampling
An odds ratio is not automatically a risk ratio. Interpret it in light of control sampling and distinguish unmatched summaries from conditional analyses of matched sets. Sparse or uninformative matched sets affect precision.
5. Bias and sensitivity plan
Statistics · Assess selection and measurement errors
Consider participation and source-population selection separately from exposure measurement and confounding. Sensitivity analyses should probe plausible mechanisms, rather than only changing a significance threshold.
Equations and statistical interpretation
These explanations describe notation and assumptions. Flow does not calculate these values.
Unmatched exposure odds ratio
a and b are exposed and unexposed cases; c and d are exposed and unexposed controls. This is an unadjusted two-by-two summary.
Matched designs require an analysis that respects matched sets. With cumulative sampling, the outcome odds ratio approximates a risk ratio only under a rare-outcome condition; incidence-density sampling supports a rate-ratio interpretation under its sampling assumptions.
Published example
Doll and Hill (1950), Smoking and carcinoma of the lung: preliminary report
Inspect the selection of hospital cases and controls and retrospective smoking ascertainment. Consider how hospital admission and recall could affect the exposure comparison.
This publication illustrates a study design, not an endorsement of every method or a reproduction of its results. Follow its protocol, supplement and update history when studying the example.
Hypothetical scenario
Investigators want to examine whether prior solvent exposure is associated with a rare neurological disorder using incident cases and controls from the same catchment population.
- Case-control overview: Define the disorder, source population and etiologic exposure period.
- Case and control selection: Identify incident cases and select controls by a documented scheme from the population giving rise to them.
- Exposure and covariate ascertainment: Measure pre-index solvent exposure using comparable records or interviews in both groups.
- Sampling and analysis plan: Choose an analysis compatible with control sampling and any matching.
- Bias and sensitivity plan: Assess participation, recall, exposure misclassification and residual confounding.
Deliver a populated protocol, decision log, data specification, and planned outputs. Enter numerical findings only after analysis and verification. Keep decisions made before seeing results distinguishable from later amendments.
Sources
- STROBE checklist for case-control studies
- RECORD checklist for studies using routinely collected health data
- HMA–EMA Catalogue of RWD studies: registration and protocol guidance
- Pearce (2016): Analysis of matched case-control studies
- Rodrigues and Kirkwood (1990): case-control sampling and measures of relative incidence
Source check: 2026-09-17. Record the version of each reporting or appraisal tool used in your actual protocol.
Compare all Observational and RWE studies project types · Open editor in a new tab