Re: Why match in case-control studies?
16 February 2012
The answer to this is misleading. Unlike in cohort or experimental studies, matching in a case-control study does not of itself reduce confounding (paradoxically, it can introduce confounding where none was present). It can, however, make statistical adjustment for a suspected confounder more efficient [1]
1.Rothman KJ. Chapter 13, Matching. In: Rothman KJ. Modern Epidemiology. Little Brown and Company: Boston, 1986, p237-246.
Competing interests: None declared
MRC Lifecourse Epidemiology Unit, University of Southampton, MRC Lifecoursse Epidemiology Unit, Southampton General Hospital, Southampton SO16 6YD
Click to rate: