TABLE 3.

ORs of 14–26 year old women’s HPV initiation and completion by their characteristics

Variable	Received 1 dose OR (95% CI)	P value	Completed 3-dose series OR (95% CI)	P value
Age, y	0.84 (0.75–0.95)	.005	0.80 (0.68–0.93)	.005
Race/ethnicity
White	Referent		Referent
Black	0.49 (0.23–1.05)	.067	0.41 (0.16–1.06)	.066
Hispanic (born in the US or moved >5 years ago)	0.48 (0.25–0.93)	.030	0.46 (0.21–1.04)	.063
Hispanic (moved to the US ≤5 years ago)	0.23 (0.07–0.84)	.026	0.14 (0.02–1.09)	.060
Education
Did not graduate high school	Referent		Referent
High school graduate	3.37 (1.68–6.77)	.001	3.03 (1.26–7.30)	.013
Gravidity
<2	Referent		Referent
≥2	0.42 (0.23–0.78)	.006	0.44 (0.19–0.99)	.047
Had abnormal Papanicolaou test result	—	—
Noa	—	—	Referent
Yes	—	—	4.08 (1.40–11.95)	.010

Data are based on multivariable logistic regression analysis. Two logistic regression models were used. Dependent variables included the following: “Received 1 dose or not (yes vs no)” and completed 3-dose series or not (yes vs no). Independent variables include the following: age in years (continuous variable), race/ethnicity (categorical), education (categorical), gravidity (categorical), and abnormal Papanicolaou test (categorical). Because the response “had abnormal Papanicolaou test result” is dependent on the response “had at least 1 Papanicolaou test,” we used the response “had abnormal Papanicolaou test result” in both multivariable models. Area under the receiver-operating characteristics curve includes 0.76 and 0.79 for the vaccine initiation and completion logistic regression models, respectively. The P value for Hosmer-Lemeshow goodness of fit test = .547 and .964 for the vaccine initiation and completion models, respectively.

CI, confidence interval; HPV, human papillomavirus; OR, odds ratio; US, United States.

^aIncludes participants without a history of Papanicolaou testing.