Table 4

Bivariate and multivariate association between dichotomous correlates and engaging in sex without a condom at the last sexual encounter with a noncommercial partner

Correlate

Bivariate

Multivariate a

n = 161


n

Percent reporting condom nonuse at last sexual encounter

Prevalence ratio (95% CI)

P (χ 2)b

Adjusted odds ratio (95% CI)

P


Sociodemographic

Gender

Female

40

71.2

1.41

0.079

3.09

0.056

Male

123

50.6

(0.94–2.11)

(3.23)

(0.97–9.85)

Age group (years)

Older (20–24)

104

61.6

1.45

0.034

2.68

0.051

Younger (15–19)

59

42.5

(1.00–2.11)

(4.78*)

(1.00–7.23)

Geographic area

Urban

66

45.0

0.73

0.065

0.95

0.893

Rural

97

62.0

(0.53–0.99)

(3.56)

(0.43–2.11)

Currently in school

No

127

58.1

1.42

0.051

0.99

0.984

Yes

36

40.9

(0.97–2.09)

(4.02)

(0.27–3.56)

Work experience

Ever worked

132

55.8

1.16

0.520

N/A

N/A

Never worked

31

48.2

(0.74–1.81)

(0.42)

Condom attitudes

Perceived difficulty in obtaining a condom

High

23

80.4

1.58

0.017

4.17

0.084

Low

138

50.9

(1.18–2.12)

(6.17*)

(0.82–21.14)

Self-efficacy of using a condom with a new partner

Low

28

88.5

1.83

0.001

3.11

0.066

High

133

48.3

(1.46–2.30)

(27.38**)

(0.93–10.41)

Confidence to refuse sex without condom

Low

86

60.4

1.28

0.111

N/A

N/A

High

75

47.2

(0.95–1.72)

(2.64)

Sexual behaviors

Used a condom at first sexual encounter

No

103

78.3

3.75

0.001

12.92

0.001

Yes

60

20.9

(2.34–6.01)

(38.64**)

(4.96–33.62)

Had more than one sexual partner

Yes

111

54.4

0.99

0.958

N/A

N/A

No

52

54.9

(0.74–1.34)

(0.00)

Knowledge of HIV prevention methods

Effectiveness of condom in HIV prevention

Low

20

79.1

1.54

0.029

1.94

0.398

High

143

51.4

(1.13–2.10)

(5.06*)

(0.41–9.19)

Effectiveness of having one faithful partner in HIV prevention

Low

60

58.7

1.13

0.454

N/A

N/A

High

103

52.0

(0.83–1.54)

(0.57)

Exposure to HIV program

Ever heard of PILS (NGO)

No

66

66.0

1.42

0.018

3.27

0.010

Yes

97

46.5

(1.08–1.86)

(5.98*)

(1.35–7.90)


N/A, not applicable; 95% CI, 95% confidence interval; NGO, nongovernmental organization.

a. Three basic sociodemographic variables (gender, age group, and geographic area) and variables that attained P < 0.10 in the bivariate analyses were entered into the logistic regression model.

b. * P < 0.05. **P < 0.01.

Nishimura et al. BMC International Health and Human Rights 2007 7:8   doi:10.1186/1472-698X-7-8

Open Data