Correlates of human papillomavirus vaccine series completion among young adult female initiators

Abstract

Incomplete human papillomavirus (HPV) vaccination is a public health concern. The objective of this study was to examine the correlates of vaccine series completion among 18–26 year old US women using the Behavioral Risk Factor Surveillance System (BRFSS) data. Using BRFSS data collected during 2008–2010, we conducted multivariable logistic regression analysis to examine the correlates of HPV vaccine completion among HPV vaccine initiators. Among 656 women (18–26 years old) who initiated the HPV vaccine, the overall weighted vaccine series completion rate was 60.7%. It was 32.9%, 65.3%, and 69.9% in 2008, 2009, and 2010, respectively. Black and Hispanic women were less likely to complete the series compared with white women. Higher income, having a college degree and completion of the study in a more recent year were associated with higher completion rates. Thus, the reasons for HPV series non-completion may be multifactorial. Interventions targeting 18–26 year old female vaccine initiators with low income and education, and minority backgrounds may improve HPV vaccine series completion.

Keywords: :

• Human papillomavirus
• HPV Vaccine
• Vaccine initiation
• Vaccine completion
• Correlates
• Young adult females
• Behavioral Risk Factor Surveillance System

Introduction

Human papillomavirus (HPV) infection is the most prevalent sexually transmitted infection in the US.¹ Four HPV strains are responsible for 70% of cervical cancer (HPV 16 and 18) cases and 90% of genital warts (HPV 6 and 11) in women.^2,3 The Advisory Committee on Immunization Practices (ACIP) recommends routine HPV vaccination for US girls 11–12 y of age and “catch-up” vaccination for those 13–26 y of age and not previously vaccinated.^4,5 In spite of the proven efficacy of the HPV vaccine in protecting against precancerous lesions and genital warts in sexually active adolescents and young women,^4,6,7 recent national data demonstrate low HPV vaccine uptake among 18–26 y old women in the US (23% for initiation and 13% for completion).⁸ Thus, 10% of young adult women did not complete the vaccine series, and therefore, may not have achieved the required immunity for protection.⁹

Correlates of HPV vaccine initiation and series completion have been widely investigated.^8,10-16 In these studies and our earlier study,¹⁰ completion was measured as a proportion of the total number of women who answered questions on HPV vaccination. It is important to understand why women who initiate the vaccine do not complete the series, and whether there are racial and ethnic differences in vaccine completion. Several studies have examined the correlates of HPV vaccine completion among young adult women initiators.^8,17,18 However, two of these studies were based on data collected over 5 y ago, one examined correlates in 11–26 y old women¹⁷ and the other was based on members of a managed care organization.¹⁸ Only one study, based on nationally representative data collected in 2010, examined the correlates of HPV vaccine completion.⁸ No studies have examined annual trends of completion among initiators in the US. Moreover, there is a need to identify correlates for 18–26 y old women vaccine initiators as they (> 18 y olds) are ineligible for the Vaccine for Children (VFC) program and do not need parental consent for their vaccination. The objective of this study was to examine the correlates of vaccine series completion among 18–26 y old women who had initiated the vaccine using data collected through the Behavioral Risk Factor Surveillance System (BRFSS) study.

Results

Based on data from 2008–2010, 97.5% (2632/2700) of 18–26 y old women responded to the questions on HPV vaccination. Among these, 656 initiated the HPV vaccine. The weighted HPV completion rate was 60.7% (95% CI 56.3–65.0%). Table 1 shows HPV vaccine completion status among initiators by sociodemographic characteristics. Ethnic differences were noted with higher completion rates among whites as compared with black and Hispanic women. Completion rates were also higher among initiators who had at least some college hours and those with an annual household income > $75,000. Women residing in the Northeast, Midwest, and the West had higher completion rates than women living in the South.

Table 1. HPV vaccine completion rate among initiators by different characteristics among 18–26 y old US women, 2008–2010

	Total	Completed 3-dose series
	n = 656	% (95% CI)	P value
Age, year, n (%)			0.652
18–21	328	59.8 (53.5–65.7)
22–26	328	61.9 (55.2–68.1)
Race/ethnicity			< 0.001
White	454	67.7 (62.5–72.5)
Black	63	37.6 (24.7–52.5)
Hispanic	95	47.7 (35.8–59.8)
Other	43	54.1 (35.0–72.0)
Marital Status, n (%)			0.192
Single, never married	499	62.3 (57.2–67.1)
Married, divorced,	157	55.1 (45.4–64.4)
widowed, separated
Education, n (%)			< 0.001
≤ HS	256	49.6 (42.3–56.9)
Some college hours	221	62.9 (55.2–70.0)
College graduate	171	74.5 (65.5–81.8)
Annual Household Income, US$, n (%)			0.015
< 50,000	226	52.4 (44.2–60.5)
50,000–75,000	143	59.4 (49.5–68.6)
> 75,000	88	73.3 (63.3–81.4)
Missing value	162	63.3 (54.4–71.5)
Have healthcare coverage, n (%)			0.192
No	74	57.2 (42.9–70.3)
Yes	582	61.2 (56.5–65.7)
Length of time since last routine medical check-up			0.122
≤ 1 y	522	62.7 (57.6–67.4)
≥ 2 y	134	53.2 (42.8–63.4)
Influenza vaccination in the past season			0.498
No	415	59.5 (53.6–65.1)
Yes	241	62.9 (55.3–69.8)
HPV completion, weighted % (95% CI)
Overall	656	60.7 (56.3–65.0)	< 0.001
2008	169	32.9 (25.8–41.0)
2009	229	65.3 (57.3–72.6)
2010	258	69.9 (62.7–76.2)
Region			0.012
Northeast	367	67.1 (52.3–67.6)
Midwest/West^a	151	67.3 (57.6–75.7)
South	138	47.0 (37.4–56.8)

^a Midwest and West were combined together to balance the sample size for each region.

Variables that met the screening criteria for inclusion in the multivariable logistic regression model were race/ethnicity, education, marital status, income, healthcare coverage, history of routine medical check-up, year of study, and region of residence. Variables with P > 0.10 (marital status, healthcare coverage, and region of residence) were excluded from the final model. Among those who initiated the HPV vaccine, black and Hispanic women were less likely to complete the 3-dose series (Table 2). Women with a college degree, annual household income of > $75,000, and a routine medical check-up within the past year were more likely to complete the 3-dose series. Women vaccinated in 2009 and 2010 were more likely to complete the series compared with women who were vaccinated in 2008. In multivariable analyses by race/ethnicity, almost similar results were observed in white women. However, among minority women, only year of vaccination was associated with vaccine series completion.

Table 2. Correlates of HPV vaccine series completion among 18–26 y old US women vaccine initiators, 2008–2010

Characteristics	Completed 3-dose series among initiators
	Overall (n = 656)		White (n = 454)		Black/ Hispanic/others (n = 202)
	Adjusted odds ratio (95% CI)	P value	Adjusted odds ratio (95% CI)	P value	Adjusted odds ratio (95% CI)	P value
Race/ethnicity
White	Reference		-		-
Black	0.35 (0.17–0.72)	0.004	-		-
Hispanic	0.45 (0.23–0.86)	0.015	-		-
Other	0.56 (0.24–1.29)	0.175
Education,
≤ HS^a	Reference		Reference		Reference
Some college hours	1.52 (0.95–2.43)	0.078	1.56 (0.87–2.79)	0.133	1.51 (0.63–3.63)	0.357
College graduate	2.32 (1.29–4.17)	0.005	2.24 (1.06–4.74)	0.034	2.75 (0.94–8.04)	0.064
Annual Household Income, US$
< 50,000	Reference		Reference		Reference
50,000–75,000	0.94 (0.51–1.71)	0.832	1.13 (0.53–2.40)	0.745	0.59 (0.17–2.05)	0.399
> 75,000	1.92 (1.01–3.65)	0.045	2.40 (1.11–5.21)	0.027	1.23 (0.35–4.25)	0.745
Missing value	1.59 (0.92–2.64)	0.099	1.86 (0.91–3.82)	0.090	1.39 (0.55–3.53)	0.480
Length of time since last routine medical check-up
≥ 2 y	Reference		Reference		Reference
≤ 1 y	1.82 (1.06–3.14)	0.030	1.78 (0.94–3.37)	0.075	1.63 (0.49–5.36)	0.421
Study year
2008	Reference		Reference		Reference
2009	3.76 (2.22–6.37)	< 0.001	4.10 (2.15–7.84)	< 0.001	2.97 (1.08–8.19)	0.036
2010	4.64 (2.75–7.84)	< 0.001	4.16 (2.25–7.67)	< 0.001	5.66 (1.99–16.10)	0.001

^a HS, High School; Adjusted by variables listed in this Table.

Discussion

Our study showed that correlates of vaccine series completion among initiators and correlates of HPV vaccine completion among those who responded to HPV vaccination questions reported in the literature are different.^8,10-16 Moreover, the evidence suggests racial differences in correlates of vaccine completion between white women and minority women. In this study, we observed that after HPV vaccine initiation, income, minority background and a college degree are important predictors of series completion. On the other hand our earlier study showed that age, marital status, college degree, income, healthcare coverage, routine medical check-up, influenza vaccination in the past 12 mo, and region were significant correlates of vaccine completion among women who responded to HPV vaccination questions. These results suggest that a subset of correlates need to be taken into account for interventions targeting series completion among young adult women vaccine initiators.

Our findings show that among initiators, the rate of HPV vaccine series completion in 2010 (69.9%) is similar to that observed by Laz et al.’s study using 2010 National Health Interview Survey data.⁸ In addition to their findings that black women were less likely to complete the HPV vaccination series, we also observed that Hispanic women were less likely to be complete the series. Chou et al. observed similar racial disparities among women 11–26 y of age.¹⁷ We also observed that college graduation, high-income, and routine health check-up in the past year were associated with higher completion of the HPV vaccine series. The differences in correlates of vaccine completion between our study and Laz et al.’s study could be due to sample size (656 vs. 408) and study year (2008–10 vs. 2010). Another study based on 18–26 y old female members of a managed care organization observed a separate set of correlates i.e., neighborhood education level, provider specialty, and healthcare utilization variables.¹⁸ These studies support the existence of racial disparities in regimen completion, along with other sociodemographic correlates.

Our finding that minority women had lower rates of HPV vaccine completion is a major public health concern. It is well documented that black and Hispanic women have higher rates of cervical cancer compared with white women.¹⁹ Similarly, minority women have been shown to have disproportionately higher prevalence of both high and low-risk HPV types.¹ Strategies to improve the HPV vaccine series completion among minority women are critically needed to lower the burden of HPV-related diseases and cancers. Moreover, as lower income is also associated with lower series completion, extension of the federal VFC program for 19–26 y old women may increase vaccine completion in this age group. We also observed higher HPV vaccine series completion among those who had undergone a routine medical check-up in the past year, which may provide a window of opportunity for physicians to discuss the importance of completing the HPV vaccine series.

We observed that completion of the 3-dose series has increased from 2008 to 2010, irrespective of race/ethnicity. Previous studies also support this notion. Chao et al.,¹⁸ Chou et al.,¹⁷ and Laz et al.⁸ based on 2006–07, 2007–08, and 2010 data observed completion rates of 47.1%, 31.2%, and 64.5%, respectively. While Chao et al. observed a higher completion rate than Chou et al., this discrepancy may be due to differences in study populations, as the former study was based on members of a managed care organization.

The strength of this study is that it includes three years of data collected from 12 states; however, some limitations exist. Not all states within the US conducted the HPV vaccination uptake survey among young adult women in BRFSS study. We also do not know whether the women completed the 3-dose regimen within the CDC recommended timeframe. In addition, the multivariable analyses stratified by race/ethnicity were based on small number of observations and, therefore, we may not have enough power to make conclusive statements. Finally, the BRFSS survey is based on self-reported data which may be subject to recall bias. In spite of these limitations, this study provides important information regarding the correlates of HPV vaccine completion among a nationally representative sample of women who initiated the vaccine.

Although the completion rate of HPV vaccine regimen increased during 2008–10, minority women who initiated the vaccine were less likely to complete the 3-dose series. Awareness of the necessity to complete the series regimen, extension of VFC program for 19–26 old women, and interventions targetting women with lower education levels and minority backgrounds may improve vaccine series completion among initiators.

Materials and Methods

The BRFSS, conducted by the Centers for Disease Control and Prevention (CDC) since 1984, is a cross-sectional telephone health survey among adults from all 50 US states, the District of Columbia, and US territories. More than 400,000 interviews are conducted in this survey each year, making it the largest ongoing health survey in the world. Details of the survey methods have been published elsewhere.²⁰ Since the adult HPV module was introduced in 2008, we limited our study to 2008–2010 data in which 12 US states participated (Connecticut, Delaware, Kansas, Massachusetts, Minnesota, Nebraska, Oklahoma, Pennsylvania, Rhodes Island, Texas, West Virginia, and Wyoming). The response rates ranged from 43.4% to 66.9% in different years. We conducted a secondary analysis based on women who were included in our earlier study on geographic variations in HPV uptake in the US.¹⁰ For this study, however, we restricted our sample to women aged 18–26 y who initiated HPV vaccine (≥ 1 dose). This study did not require approval from the University of Texas Medical Branch Institutional Review Board because the BRFSS is a publicly available de-identified database.

This study focused on survey questions related to HPV vaccination status, number of doses received, socio-demographic characteristics, and health care utilization. The main outcome of interest was HPV completion (3 doses) among those who initiated the vaccine. HPV initiation was based on the question: “A vaccine to prevent the human papilloma virus or HPV infection is available and is called the cervical cancer or genital warts vaccine, HPV shot, (GARDASIL or CERVARIX). Have you ever had the HPV vaccination?” The response options were “yes” or “no.” Those who responded “yes” were considered HPV vaccine initiators and were further asked “How many HPV shots did you receive?” Those who had reported that they received three shots were considered HPV vaccine completers. The main exposures of interest in this study were sociodemographic characteristics and healthcare utilization variables. Age, race/ethnicity, education, marital status, annual household income, last routine medical check-up, influenza vaccination in the past year, and health care coverage were categorized for the purpose of analysis. Region of residence was categorized as Northeast, Midwest, West, and South.²¹

We used STATA 12 svy commands (STATA Corporation, College Station, TX) for data analysis by incorporating probability sampling weights in conjunction with strata and primary sampling units (psu) generated by BRFSS complex survey design. Poststratification weight was used to correct for the complex BRFSS study design, and bias originated from nonresponse and non-telephone coverage. We used Chi-square tests for bivariate comparisons and multivariable logistic regression models to examine the correlates of HPV vaccine completion. Variables were screened for inclusion in an initial multivariable model. Candidate variables with a P value of ≤ 0.20 with the dependent variable were included in the initial multivariable model based on all women. After that, variables with P > 0.10 were excluded from the overall model. Similar independent variables were used in race/ethnicity based multivariable models. Adjusted odds ratio and 95% CIs for each of the correlates were reported.

Abbreviations:
HPV	=	Human papillomavirus
BRFSS	=	Behavioral Risk Factor Surveillance System
ACIP	=	Advisory Committee on Immunization Practices
VFC	=	Vaccines for Children

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Funding sources

Dr. McGrath is supported by a research career development award (K12HD052023: Building Interdisciplinary Research Careers in Women’s Health Program -BIRCWH) from the Office of Research on Women’s Health (ORWH), the Office of the Director (OD), the National Institute of Allergy and Infectious Diseases (NIAID), and NICHD at the National Institutes of Health.

10.4161/hv.29633