Table 1.

Hypothetical data example demonstrating relevant confounding magnitude in estimating an exposure-outcome association in presence of a low prevalent binary confounder.

A				B				C				D
total sample				total sample				confounder = 'no'				confounder = 'yes'
		exposure				exposure				exposure				exposure
		no	yes			no	yes			no	yes			no	yes
outcome	no	25	488	confounder	no	22	948	outcome	no	11	474	outcome	no	14	14
	yes	12	475		yes	15	15		yes	11	474		yes	1	1
		37	963			37	963			22	948			15	15