Special types of case-control studies
Nested case-control studies
Definition: an
important type of case-control study where cases and controls are drawn from the population
in a fully enumerated cohort. Baseline characteristics of these patients are obtained by
interview, physical examination, and pertinent lab or imaging studies. The patients are then
followed to determine the outcome. Those patients who develop the condition of interest
become cases in a case-control study, and those who do not develop the condition become
eligible for the control group of a study. The cases and representative sample of controls
are studied, and data from the two groups are compared using analytic methods appropriate for
case-control studies [1].
This type of study is particularly useful when measurement of exposure is expensive.
Case-cohort studies
Definition: a
variant of the case-control design in which controls are drawn from the same cohort as the
cases regardless of their disease status. Cases of the disease of interest are identified,
and a sample of the entire starting cohort (regardless of outcomes) forms the controls. This
provides an estimate of the risk ratio.
This type of design can be more efficient to perform when you have several types of cases that
you wish to compare to the same pre-defined cohort (comparison) group, If investigators are
only interested in one type of case either a nested case-control (in you have a cohort study)
or a standard case-control study is more efficient.
Density case-control studies
Definition: controls are sampled concurrently with cases (i.e. each time a
case is identified control must also be identified). This variant allows controls to
represent the person-time experience that generated the cases.
Occasionally "density" or "risk set sampling" case control studies may be selected. This design
provides an estimate of the rate ratio.