In case–control studies, the outcome has already occurred at the time of investigation. Data on exposure to a potential risk factor is collected retrospectively, usually from records or by administration of questionnaires.

Figure 9a illustrates the basic design of a case–control study. In a case–control study, we start off by recruiting a group of people who already have the disease of interest (i.e. the cases). We then use a group of people drawn from the same population who do not have the disease of interest (i.e. the controls). The cases and the controls are then compared for the prevalence of the risk factor of interest.

(Note: The investigator chooses the proportion of cases to controls depending on the required sample size; the most efficient study design is a 1:1 ratio. In general, ratios of over 4 controls to 1 case do not offer much additional benefit).