Nested case-control study
- Study design
Inside a cohort that is already followed, those who fall ill and a few controls are picked, and only their stored samples are analysed.
What it is
A nested case-control study is carried out inside an existing cohort. Cases are cohort members who develop the disease; for each one, controls are chosen at random among members still free of disease at that moment. It is very useful when exposure is measured in biological samples stored at baseline, because only the samples of cases and controls are analysed: money is saved and exposure was measured before the disease.
How it is built
- Start from a cohort with data or samples stored at baseline.
- Identify the cases that appear during follow-up.
- For each case, choose at random controls still at risk on that date, often matched on age and sex.
- Measure exposure in the stored samples and calculate the OR.
Worked example
- A cohort of 6 000 adults stored frozen serum at baseline; over 8 years, 80 cases of colorectal cancer appear.
- For each case, 3 controls still free of cancer on that date are chosen: 240 controls.
- 320 samples are analysed instead of 6 000: 320 / 6 000 ≈ 5% of the total.
- Low vitamin D: 40 of 80 cases and 72 of 240 controls; not low: 40 cases and 168 controls.
- OR = (40 × 168) / (72 × 40) = 6 720 / 2 880 ≈ 2.3.
- Reading: low vitamin D measured years earlier was linked to 2.3 times the odds of cancer, at 5% of the laboratory cost.
Compared with neighbouring designs
| Feature | Classic case-control | Nested | Full cohort |
|---|---|---|---|
| Where participants come from | Hospital or community | A defined cohort | A defined cohort |
| When exposure is measured | After falling ill | Before falling ill (stored samples) | Before falling ill |
| Samples analysed | Cases and controls | Cases and controls | All |
| Cost | Low | Low | High |
Markers
- OR
- Biobank
When to use it
- Used in biobanks and large cohorts to study biomarkers, genes or past infections.
- A close variant is the case-cohort study, which compares cases with a random subcohort.
Common pitfalls
- Choosing controls among those who ended follow-up healthy, ignoring time: the advantage of sampling among those at risk is lost.
- Analysing without accounting for matching when controls were matched.
Test yourself
Think of the answer, then tap the question to see it.
Where do cases and controls come from in a nested study?
From one established cohort: cases are those who fall ill during follow-up and controls are members who were still healthy.
What advantage does it have over a classic case-control study?
Exposure was measured before the outcome (for example in stored serum), which avoids recall bias and secures temporality.
Why is it cheaper than analysing the whole cohort?
Because only the samples of the cases and a few controls per case are processed, not those of every participant.
Connections
Educational content on epidemiological and statistical methods; the examples use made-up numbers to teach the calculation. Not a clinical or legal guide: current Colombian rules are in the official sources.
Sources: Chapter 24: Including non-randomized studies on intervention effects (Cochrane, 2024); Principles of Epidemiology | Lesson 1 - Section 7 (Centers for Disease Control and Prevention, 2012).
Spotted a mistake?We are always improving for the community. Tell us what is wrong and our team will review it.Report a mistake
Representative model, for educational purposes only.