TABLE 2.

Logistic Regression on Condom Use in Relation to Sexual Identity and Sexual Behavior Before and After Adjusting for Demographic Characteristics and Other Covariates Among HBCU Male Students (n = 1,865)

	Unadjusted			Adjusted
Characteristic	OR	95% CI	p	OR	95% CI	p
Sexual identity
MSW	1.00	Reference		1.00	Reference
MSMW	0.61	0.44–0.85	.003	0.63	0.42–0.95	.029
Sexual behaviorsa
No, no	1.00	Reference		1.00	Reference
Yes, no	0.72	0.56–0.91	.008	0.68	0.51–0.90	.007
No/yes, yesb	0.48	0.34–0.68	< .001	0.54	0.36–0.81	.003
Age (years)
19 or younger	1.00	Reference		1.00	Reference
20 or older	0.57	0.43–0.75	< .001	0.67	0.45–0.98	.038
Campus living
On campus	1.00	Reference		1.00	Reference
Off campus	0.86	0.64–1.15	.303	0.94	0.65–1.36	.747
Total family income
< $30,000	1.00	Reference		1.00	Reference
$30,000–$75,000	1.37	1.09–1.72	.006	1.36	1.03–1.78	.028
> $75,000	1.28	0.98–1.68	.071	1.17	0.84–1.62	.343
Importance of religion
Extremely/very	1.00	Reference		1.00	Reference
Not at all/somewhat	0.84	0.68–1.04	.111	0.74	0.57–0.96	.023

Note. CI = confidence interval; HBCU = historically black colleges and universities; MSW = men who have sex with women; MSMW= men who have sex with both men and women; OR = odds ratio. Unadjusted ORs were based on logistic regression models without adjustment for any other covariate. Adjusted ORs were based on logistic regression models with adjustment for the covariates listed in the table and schools the male students attended.

^aOral sex more than once, anal sex more than once.

^bTwo categories (no, yes; yes, yes) were combined due to a limited sample size (n = 12) for the third category (no, yes).