Table 4.

Multivariable Logistic Regression Analysis of Predicting Condomless Sex

Variable	βa	OR (95% CI)	P Value
Having heterosexual intercourse, yes	−0.98	0.37(0.21 ~ 0.70)	0.002
More types of intimate partner violence experienced b	0.46	1.58(1.31 ~ 1.92)	< 0.001
Ever having anal sex, yes	0.56	1.75(1.06 ~ 2.84)	0.027
Condom use at first sex, yes	−1.65	0.20(0.14 ~ 0.27)	< 0.001
Attitudes towards condom use points	−0.14	0.87(0.80 ~ 0.95)	0.001
Condom use self-efficacy points	−0.18	0.84(0.78 ~ 0.90)	< 0.001

^aUnstandardized β coefficients were calculated from the multivariable logistic regression model

^bTypes of intimate partner violence experienced included physical, verbal, and sexual types