Case-Control Study Design


Case-Control Study Design - Definition

"A case-control study involves grouping people as cases (persons experiencing a health-related state or event) and controls [without the disease] and investigating whether the cases are more or less likely than the controls to have had past experiences, lifestyle behaviors, or exposures. In other words, what is it about their past that made them cases? The outcome is always identified before the exposure. Because a case-control study begins with the outcome and looks back at an antecedent variable or variables, it is retrospective in nature." (R.M. Merrill, Introduction to Epidemiology, 5th Ed. 2010)

CASES
Note
: The cases used can include all of the reported cases or a subset of this group within a specified region. You'll see as we read about these investigations, that the cases involved can be restricted by space (e.g., specific county), time (e.g., those exhibiting symptoms before or after a certain time), technique (e.g., laboratory confirmed), age (e.g., accurate recall), and other parameters. 

For example, in the 4th investigation (bottled water/Lisbon), Blake (as reported in Dworkin 2011) eliminated 12 cases from a pool of 47 for the following reasons:

(1) Individuals who had visited the spring associated with the study
(Concern: an intake of water from the spring rather than from the bottled water
may have exposed the individual to V. cholerae.)

(2) Individuals who are nine years of age or younger
(Concern: questionable ability to recall behavior in the past.)

(3) A cohabitant of an earlier victim of cholera
(Concern: Individual could have contracted cholera due to exposure within the home.)

(4) Nonresidents who had symptoms prior to arriving in Lisbon
(Concern: exposure occurred elsewhere.)

CONTROLS
These individuals, at the time of selection, do not exhibit the symptoms of the studied disease. They are, however, matched as well as possible, to cases in regards to demographics (e.g., age, gender, neighborhood). As noted above, exposure history does not enter into the selection of controls; if for some reason it does, the statistics would be erroneous. The goal is to compare two groups (ill vs. not ill) to see if those who are ill were exposed to a specific (pathogen) vehicle and those who are not ill, did not.* The conclusion is based on probabilities. How are controls found? It varies among investigations. Sometimes they are drawn from the same hospital as cases. Other times they are people of similar age, race and gender from the same neighborhood.

*"An epidemiologic assumption is that controls are representative of the general population in terms of probability of exposure and that controls have the same possibility of being selected or exposed as the cases. Controls drawn from a population of the same area or populace of the cases should reflect the same gender, age, and other significant factors. Controls from a general population are assumed to be normal and healthy and to reflect the well population from the area... Controls are typically draw from the same hospital or general population as the cases. They may also be drawn from the family, friends, or relatives of the cases."
(R.M. Merrill, Introduction to Epidemiology, 5th Ed. 2010, p.189)