A cohort study is one in which two or more groups of subjects are tracked, to determine the effect of an event (like radiation exposure) or condition (like heart disease) over time. In a cohort study the two subject groups must be as similar as possible, with the exception of the specific aspect being investigated, in order for useful comparisons to be made when the research period is concluded.
Cohort studies are valuable for following how a disease develops differently in different people, or for determining different outcomes after a single event. Cohort studies often make up longitudinal research, which are conducted over a long period of time, like studies on aging. However, because it can be difficult to assemble two or more groups of people with a great degree of similarity, researchers have to account for and disclose confounding, or interfering, factors that may influence the final results.
Cohort studies can track forward in time (prospective), or begin with a date in the past, following the historic development of changes to the present (retrospective).
David A Grimes and Kenneth F Schulz. "Epidemiology Series. Cohort Studies: Marching Towards Outcomes." Lancet 2002; 359: 341–45.