Case-Control Statistics

Data are cross-tabulated with a TABLES <EXPOSURE> <DISEASE> command. This compiles the 2-by-2 table while calculating odds ratio and chi-square statistics. Fisher's test is calculated when an expected value in the 2-by-2 table is less than 5. With case-control data, relative risk statistics should be ignored.

Formulas and notation are based on the following 2-by-2 table cell labels:

 
Disease +
Disease -
 
Exposure +
a
b
n1
Exposure -
c
d
n2
 
m1
m2
N
 

p^1 = a / m1 (exposure proportion in cases)
p^2 = c / m2 (exposure proportion in controls)
o1^ = a / c (exposure odds in cases)
o2^ = b / d (exposure odds in controls)
OR^ = ad / bc (exposure odds ratio)

The above odds ratio represents an exposure odds ratio. The exposure odds ratio has the same interpretation as the disease odds ratio (e.g., an OR parameter of 1 indicates no association between the exposure and disease). Moreover, when the disease is rare (incidence < 5%), the odds ratio is approximately equal to the relative risk.