Selection bias: neighbourhood controls and controls selected from those presenting to a Health Unit in a case control study of efficacy of BCG revaccination

Table 2 Distribution of biological features, contact with Tb and socio-economic characteristics in HU and neighbourhood controls

Variables	Health Unit Control		Neighbourhood		P value
	N°	%	N°	%
Sex					0.063
Male	210	38.1	209	43.8
Female	341	61.9	268	56.2
Tb contact					0.520
Yes	72	13.1	56	11.7
No	479	86.9	421	88.3
Head of family employed					0.026
Yes	400	72.6	315	66.2
No	151	27.4	161	33.8
Income of the head of family					0.152
<1 MW*	31	5.63	33	6.92
1 to <2MW	103	18.69	88	18.45
2 to <5MW	45	8.17	30	6.29
≥ 5MW	28	5.08	12	2.51
No reported or not known	344	62.43	314	65.83
Piped water					<0.001
Yes	503	91.5	467	98.5
No	47	8.5	07	1.5
Washing machine					0.007
Yes	134	24.3	81	17.3
No	417	75.7	386	82.7
Frigidaire					0.929
Yes	507	92.0	429	91.9
No	44	8.0	38	8.1
Videocassette					0.088
Yes	221	40.1	163	34.9
No	330	59.9	304	65.1

*MW: the income level of the head of the family was relative to the minimum wage (MW). MW<1 was the baseline. 1–1.9+ = US$100–US$199/month; 2–4.9 = US$200–US$499/month; 5+ = US$>=US$500/month.

ISSN: 1471-2288