Home Statistical Dictionary Case-Control Study
Case-Control Study
An observational study design that starts by identifying individuals who already have an outcome of interest (cases) and a comparable group who do not (controls), then looks backward to compare their prior exposure to a suspected risk factor.
In Plain English
A case-control study works backward compared to most study designs: instead of starting with an exposure and following people forward to see what happens, you start with the outcome, people who already have a disease or condition (cases) and a comparable group who don't (controls), and then look back in time to compare how much each group was previously exposed to a suspected cause. This backward-looking structure makes case-control studies especially efficient for studying rare outcomes, you don't need to follow a huge cohort for years waiting for a handful of rare cases to develop, you can go find people who already have the rare outcome directly and compare their history to controls, which is exactly why case-control studies are a workhorse design in epidemiology for investigating rare diseases.
Definition
A case-control study identifies a group of individuals with the outcome of interest (cases) and a comparable group without it (controls), typically matched or statistically adjusted on relevant characteristics, then retrospectively compares the two groups' prior exposure to a suspected risk factor, estimating the association via the odds ratio rather than relative risk, since case-control sampling (fixing the number of cases and controls in advance, rather than sampling from the full exposed and unexposed populations) does not allow direct estimation of disease incidence or relative risk. Case-control studies are particularly efficient for investigating rare outcomes, since they require identifying and interviewing far fewer people than a prospective cohort study would need to accumulate a comparable number of cases, but they are especially vulnerable to recall bias (cases and controls may recall or report past exposures differently, often because cases have spent more time reflecting on possible causes of their condition) and to selection bias in how controls are chosen, both of which require careful design and interpretation to guard against.
Properties
- Case-control studies cannot directly estimate disease incidence or relative risk, because the number of cases and controls is fixed by the researcher's sampling design rather than reflecting their natural proportions in the underlying population, the odds ratio is the association measure case-control data can validly support, and under the rare-disease assumption it closely approximates the relative risk.
- Recall bias is a particularly well-documented threat to case-control study validity, cases, having already experienced the outcome, often recall and report past exposures more thoroughly or differently than controls, who have no similar motivation to scrutinize their history, a systematic difference that can create or distort an apparent exposure-outcome association.
- The choice of an appropriate control group is one of the most consequential and difficult decisions in case-control study design, controls should represent the exposure distribution of the population that produced the cases, a poorly chosen control group (for example, drawn from a hospital population with its own health-related selection factors) can introduce substantial selection bias into the resulting odds ratio.
At a Glance