Why is the odds ratio the measure of choice in a case-control study?

1. A case-control study starts from a group of cases and a separately chosen group of controls. 2. The investigator decides how many controls to enrol, so the numbers in the study do not reflect any real population. 3. Risks and rates therefore cannot be computed, and neither can risk ratios or rate ratios. 4. The odds ratio survives because it depends only on the cross product of the four cells. 5. When the outcome is uncommon the odds ratio approximates the risk ratio closely, so it is an estimate rather than a more accurate measure. 6. Matching is a design choice used in some case-control studies but not all, so it is not what forces the odds ratio. _Source: CDC Principles of Epidemiology in Public Health Practice (3rd ed., US federal, public domain), Lesson 3 "Measures of Risk", section Odds ratio_