Table 3.

Absolute frequencies, outcome prevalences, exposure prevalences, crude and pooled prevalence ratio (PR) estimates, and relative confounding for the analysis of the original data using asthma as the outcome, maternal smoking as the risk factor and social class as confounder (situation 2 original).

First stratum: High social class
	Asthma		No	All
	N	Prev.	N	N	Exp. prev. = 26.6%
Mother smokes	37	25.7%	107	144	PR = 1.19
No	86	21.6%	312	398	M-H weight = 22.85
All	123	22.7%	419	542
Second stratum: Low social class
	Asthma		No	All
	N	Prev.	N	N	Exp. prev. = 43.3%
Mother smokes	122	42.8%	163	285	PR = 1.24
No	129	34.6%	244	373	M-H weight = 55.87
All	251	38.1%	407	658
Combined strata: High and low social class
	Asthma		No	All	Exp. prev. = 35.8%
	N	Prev.	N	N	PR (crude) = 1.33
Mother smokes	159	37.1%	270	429	PR (M-H) = 1.22
No	215	27.9%	556	771	Confounding = -7.9%
All	374	31.2%	826	1200	P-value(het)*= 0.8

* P-value for testing heterogeneity of the prevalence ratios across strata.