Table 2.

Determinants of consistent condom use among sexually active women, at the time of resumption of sexual intercourses. Univariate analysis and multivariate logistic regression (DITRAME Plus, Abidjan, 2001–2005)

Variables	N = 776	Women having used condom at resumption of sexual activity		Multivariate analysis §
		%	p*	aOR#	95% IC	p**
HIV status			0.886			0.934
HIV-positive	421	28.3		1.02	0.63 – 1.65
VIH-negative	355	28.7		1
Age (in years)			0.134			0.109
18–19	50	32.0		1
20–24	261	32.2		1.10	0.56 – 2.18	0.773
25–29	268	24.6		0.77	0.38 – 1.54	0.461
30–34	133	24.1		0.85	0.40 – 1.84	0.694
35 and above	64	35.9		1.65	0.71 – 3.82	0.240
Education level			<0.001			0.051
No education	254	20.5		1
Primary school	299	29.1		1.31	0.85 – 2.02	0.225
Secondary school and above	223	36.8		1.78	1.11 – 2.85	0.016
Religion			<0.001			0.017
Christian	456	34.2		1
Muslim	261	18.0		0.54	0.35 – 0.83	0.005
Animist or no religion	59	30.5		0.91	0.49 – 1.67	0.761
Cohabiting with her partner			0.008			0.050
No	231	35.1		1
Yes	545	25.7		0.68	0.47 – 1.00
Declaring at least one co-spouse			0.025			0.072
No	644	30.1		1
Yes	132	20.5		0.63	0.39 – 1.04
Living in shared housing †			0.328			0.199
No	302	30.5		1
Yes	474	27.2		1.27	0.88 – 1.82
Woman having disclosed her status to her partner			0.314			393
No	231	26.0		1
Yes	545	29.5		1.21	0.77 – 1.89
Partner tested for HIV before the resumption of sexual activity			0.016			0.115
No	702	27.2		1
Yes	74	40.5		1.52	0.90 – 2.58

Chi-2 tests.

^§14 women whose data are not complete were excluded for analysis.

^#aOR, adjusted odds ratio. CI, confidence interval.

^**Wald test.

^†Typical housing in Abidjan with several houses organized around a yard, where inhabitants share kitchen and restroom and live in crowded accommodations