Factors associated with HPV vaccination initiation among United States college students

ABSTRACT

Human papillomavirus (HPV) remains the most common sexually transmitted infection (STI) in the U.S. despite widespread availability of a safe, effective vaccine. Although young adults are at greatest risk of HPV infection, extensive vaccine promotion and intervention efforts has been directed toward 11–12-year-olds. College students represent an ideal audience for HPV vaccine “catch-up;” however, research indicates inconsistent HPV vaccination rates within this demographic. An online survey assessing HPV and HPV vaccine knowledge and behaviors was distributed to all undergraduate college students at a large, public university in the Deep South region of the U.S. The primary outcome was receipt of HPV vaccination (binary response options of Yes/No). Logistic regression analyses were performed to determine predictors of HPV vaccination. Of the 1,725 who completed the survey, 47.0% reported having received at least one dose of HPV vaccine; overall series completion (series = 3 doses for this population) was 17.4%. The primary outcome was HPV initiation among college students, defined as having received at least one dose of the HPV vaccine. Results indicated substantial gaps in participants’ knowledge of their vaccination status. Provider and parental recommendations as well as social influences were shown to significantly impact student vaccination status, emphasizing the importance of incorporating these elements in future interventions, potentially as multi-level strategies. Future college interventions should address HPV and vaccination knowledge and the importance of provider and parental recommendations.

Introduction

With an estimated 79 million Americans currently infected and around 14 million new infections annually, human papillomavirus (HPV) is the most common sexually transmitted infection (STI) in the United States (U.S.). HPV is associated with genital warts as well as cancers of the cervix, vagina, vulva, anus, penis, and oropharynx.¹ HPV incidence is highest among individuals in their late teens and early twenties, who are likely to engage in sexual activity and are more likely to engage in risky sexual behaviors such as multiple sexual partners and unprotected sex.² Other potential risk factors among this population include low knowledge of HPV (including HPV-associated diseases) and limited awareness of opportunities for HPV prevention.^3–5

HPV-associated (HPVa) diseases are unique in that they are largely preventable through a safe, effective, and widely available vaccination. Initially approved for market use in the U.S. in 2006, the current HPV vaccine is a 2- or 3-dose series (depending on age of initiation) that protects against seven of the most common, high-risk strains of HPV (16, 18, 31, 33, 45, 52, and 52).⁶ The Advisory Committee on Immunization Practices (ACIP) recommends HPV vaccination for males and females between the ages of 9–26 years old, with initiation at 11–12 years old heavily promoted by many national public health organizations including the Centers for Disease Control and Prevention, the American Cancer Society, and the U.S. Department of Health and Human Services.^7,8 The FDA also recently approved HPV vaccination for males and females ages 27–45 years. While ACIP does not recommend catch-up vaccination for all adults within this range, they do recommend shared clinical decision-making to determine vaccination in this group.⁶ Despite these recommendations, uptake and completion of the HPV vaccine have been slow and uneven in the U.S.^9,10 According to the 2018 National Immunization Survey-Teen data, 51.1% of adolescents aged 13–17 years were up-to-date with the HPV vaccine series, and 68.1% had received ≥1 dose of HPV vaccine.¹¹ Missed opportunities for HPV vaccination beginning at ages 11–12 are well-documented.^9,10 Because many adolescents are not vaccinated within the targeted age of 11–12, public health efforts are needed to address the catch-up population, particularly young adults before they reach age 26. A recent study found that only 1 in 25 men and 1 in 4 women in the 18–26 age range had completed their HPV vaccine series.¹² Laz et al. found HPV vaccine initiation and completion rates to be 22.7% and 12.7%, respectively, among women aged 18–26 years.¹³

Certain geographic regions, specifically the Deep South (e.g., Alabama), fall significantly below national averages of HPV vaccination and far below the Healthy People 2020 goal of 80% coverage to achieve appropriate herd immunity.^11,14 The low rates of HPV vaccination in this region are especially concerning when considered in light of the area’s high rates of HPV cancer incidence and mortality.^11,15 With respect to cervical cancer, Alabama has the 2^nd highest incidence and the 4^th highest mortality in the U.S. rate of cervical cancer deaths; regarding oral cavity and pharynx cancer, Alabama ranks 5^th in incidence and 3^rd in mortality in the country.¹⁶ Other southeastern states such as Mississippi, Kentucky, Tennessee, and South Carolina also consistently rank in the top ten for these cancer incidence and mortality rates. Given these significant negative outcomes, the need for interventions emphasizing HPV prevention is particularly crucial for states in this region.

Currently in the U.S., HPV vaccination promotion overwhelmingly targets young adolescents (i.e., 11–12-year-olds), primarily through materials geared toward their parents and healthcare providers.^17–19 While the heavy focus on initiating HPV vaccination this population is important and recommened, it has led to the near exclusion of efforts targeted to older adolescents and young adults still within the ACIP recommended age range for HPV vaccination.⁶ Vaccinating through age 26 (or up to 45 on a case-by-case circumstance) provides an important window of opportunity to administer “catch-up” vaccines to those in their late teens and early 20s who have not yet initiated or completed the series, yet are at highest risk of contracting HPV.⁶ Previous studies have shown that missed opporuntity visits are common and have been accounted for up to two-thirds of visits for young adult women and nearly half of visits for pre-teens.^9,10,20,21 The percentage of adults aged 18 − 26 who have received one or more doses of HPV vaccine have nearly doubled between 2013 and 2018, but remains at 39.9%.²² Targeting this population, particularly in undergraduate/college settings, has strong potential for success. As they transition into young adulthood, individuals in this age group are beginning to take ownership of their health and starting to make medical decisions (e.g., vaccinations) for themselves.²³ Additionally, the cost of medical care is an infrequent barrier for the majority of these individuals, with the American College Health Association reporting that 95.5% of college students are insured.²³ College students also typically have access to health resources (including immunizations) on their campuses through student health services, so lack of convenience and access to care are less significant barriers for them than many other populations. In addition, starting at age 21, women receive counseling about Papanicolaou (Pap) testing. This conversation presents an excellent opportunity to discuss STIs and preventative measures, including the HPV vaccination series. When compared to their highly targeted 11–to 12-year-old counterparts, college students may be more likely to engage openly in dialogue about their sexuality, HPV, and potential sexually-associated risks with healthcare providers given that more than half of U.S. high school students report having sexual intercourse by 12^th grade.²⁴ Despite these favorable conditions, college students’ HPV immunization rates are significantly below the U.S. Department of Health and Human Services Healthy People 2020 goals, making them a high-risk population in significant need of intervention.²⁵

Understanding barriers to participation is an essential component in the process of designing effective interventions that increase rates of HPV vaccination and completion. Known barriers to HPV vaccination among college students include: insufficient or lack of a provider recommendation, cost, access to care, lack of knowledge of both HPV and the HPV vaccine, concerns about safety, low perceived susceptibility, and stigma associated with sexual activity.^26–29 Furthermore, a previous intervention aiming to improve HPV vaccination among college students have shown relatively low success, suggesting a need for more sensitive and focused formative work.³⁰ Understanding specific barriers for college students on a particular campus as well as assessing their HPV-related knowledge, attitudes, and perceived risk may help to inform future interventions and policies aimed at HPV-preventive behaviors, including “catch-up” vaccination among college students. The objective of the current study was to assess rates of HPV vaccine initiation and completion among college students as well as factors associated with vaccination.

Materials and methods

Data and sample

The final electronic survey was constructed using Research Electronic Data Capture (REDCap) software and conducted from August 2018-September 2018. The survey was distributed via e-mail to all undergraduate students at the University of South Alabama (USA) (n = 10,275). Eligibility criteria for survey inclusion were: undergraduate enrollment at USA with a valid, working USA-assigned e-mail address, and age 18 years or older. Valid University-issued e-mail addresses for currently-enrolled undergraduate students were obtained from the institution’s registrar office and participants were required to attest to that they were 18 years or older before they were able to enter the survey to further screen for eligibility. Because of the confidential nature of the survey, we are unable to know if any students younger than 18 years disregarded the survey after reading the attestation requirement. However, there were no data of individuals selecting they were under 18 years old in the recorded dataset. Since eligible students had to be 18 years or older, no participants of the current study would have been eligible for the change to a 2-dose HPV vaccine regimen that was approved in 2016; all students would have needed the full 3-dose series. Students were able to complete the survey only once, and those who participated were entered into a drawing for 1 of 100 15 USD gift card incentives. The study received exempt approval from the USA Institutional Review Board (Protocol 1288189–2).³¹

For race, the categories American Indian or Alaska Native, Asian, and Other were combined to form the “Other” option due to small number of respondents who identified as these races. Logistic regression models were used to determine factors related to receiving HPV vaccination. Survey response options to the question, “Have you ever received the human papillomavirus (HPV) vaccine?” were: Yes; No; Don’t know/Not Sure; No, but I plan to; and Not eligible due to age.

Measures

An electronic survey was constructed consisting of measures assessing the study’s five core themes: HPV vaccination history, HPV knowledge (analyzed and presented in a forthcoming paper), HPV vaccination knowledge and attitudes (analyzed and presented in forthcoming paper), and influences on HPV vaccination. Validated measures from previous undergraduate surveys on this subject were identified and adapted for the current study, combined with newly constructed measures.^28,32–35 The primary outcome of interest was HPV initiation among college students, and associated predictors of vaccination. Vaccine initiation was defined as having received at least one dose of the HPV vaccine series. Vaccine completion was defined as having received all doses in the HPV series – for this population, meaning all three doses. For the current study, the outcome variable of receipt of HPV vaccination was defined as vaccine initiation.

Analytical strategy

Although our preliminary descriptive analyses were restricted to the 1,725 students who completed the survey, there were small amounts of missing data. However, the percentage of missing data did not exceed 5% for any variable of interest. The characteristics of participants who completed the survey vs. non-completers were compared, but determined not to vary significantly (shown in Appendix Tables A, B, and C). Regarding the outcome of interest, “Did you ever receive the human papillomavirus (HPV) vaccine?” the current analyses were restricted to students who responded “Yes” or “No.” Students who responded, “No, but I plan to,” were grouped with, “No,” responses and those who responded, “Don’t Know/Not Sure” and “Not eligible due to age,” were excluded form the analyses. This process reduced the potential sample size to 1,289. A sensitivity analysis was conducted treating, “Don’t know/Not sure,” as another group using a multinomial logistic regression (results shown in Appendix Table D). With this sample size, we had 80% statistical power to detect unadjusted Odds Ratios of OR = 1.42 at the α = 0.05 significance level.

Several bivariate logistic regression models were performed to obtain unadjusted Odds Ratios (OR) with 95% Wald-type Confidence Intervals (CI) and p-values. Variables that had statistically significant (p < .05) unadjusted ORs were entered into a multiple logistic regression model to obtain adjusted ORs with 95% Wald-type CIs and p-values. From this model, variables with p-values > 0.2 were omitted. After performing this analysis, a reduced model with ten significant variables (p < .05) was used for interpretation. These variables included: race, receiving a provider recommendation for the HPV vaccine, receiving a parent recommendation for the vaccination, having a primary care provider, frequency of primary care provider visits per year, having received the flu vaccine, whether a doctor’s encouragement would influence their decision to get the HPV vaccine, if many of their friends had received the HPV vaccine, importance of one’s partner having received the HPV vaccine, and perceived newness of the vaccine. Age (with ≤22 years as reference) was retained because age as a continuous variable was statistically significant in previous multiple logistic regression models. All analyses were performed using SAS PROC LOGISTIC.

Results

Of the 1,725 students who completed the survey (response rate = 16.8%), the majority were female (70.4%) and White (73.1%). This sample reflects the larger campus population, of which 64.7% are female and 62.7% are White. The participants mean age was 22.1 years old (±5.9 years). Most participants at the time of survey administration were upperclassmen: freshman (24.9%); sophomore (17.9%); junior (25.2%); senior (32.1%), likely contributing to a higher mean age. Rates of response by year are also comparable to the university’s current enrollment: freshman (25.8%); sophomore (20.3%); junior (23.1%); senior (30.8%). Close to half of the respondents, 809 students (47%), reported receiving at least one dose of the HPV vaccine. Although not included in the current analysis, regarding HPV vaccination receipt, 432 students (25.1% of all respondents) answered Don’t know/Not sure. Among those who reported receipt of the HPV vaccine, 11.4% stated they had received one dose; 14.8% reported receiving two doses; 37.3% responded they had received three doses; and 36.5% stated Don’t know/Not sure of how many doses they had received. When factoring in those who responded Don’t know/Not sure to having received the HPV vaccine at all, only 17.4% of all 1,725 participants reported 3-dose completion of the HPV vaccine series. Demographic and other behavioral characteristics are shown in Table 1.

Table 1.

Demographic characteristics for undergraduate students (n = 1725)

Variable		N (%)
Year (n = 1722)
	Freshman	428 (24.9)
	Sophomore	309 (17.9)
	Junior	433 (25.2)
	Senior	552 (32.1)
Age
		22.1 ± 5.9
Sex (n = 1707)
	Male	506 (29.6)
	Female	1201 (70.4)
Race (Check all that apply)
	Black or African American	342 (19.8)
	White	1261 (73.1)
	Other	122 (7.1)
Hispanic or Latino (n = 1723)
	Yes	76 (4.4)
	No	1647 (95.6)
State, territory, or country (n = 1725)
	Alabama	1217 (70.6)
	Other state, country or territory	508 (29.5)
Currently has health insurance (n = 1650)
	Yes	1485 (90.0)
	No	165 (10.0)
Currently has a primary healthcare provider (n = 1717)
	Yes	1120 (65.2)
	No	597 (34.8)
Where healthcare is received (If no PCP) (n = 593)
	Student Health Clinic	163 (27.5)
	Urgent Care	343 (57.8)
	Other	87 (14.7)
Frequency of healthcare provider visits (n = 1719)
	<1 time per year	360 (20.9)
	1 time per year	420 (24.4)
	2–3 times per year	783 (45.6)
	4+ times per year	156 (9.1)
Receive the flu vaccine (n = 1697)
	Yes	630 (36.6)
	No	650 (37.8)
	Sometimes	417 (24.2)
Received the HPV vaccine (n = 1721)
	Yes	809 (47.0)
	No	480 (27.9)
	Don’t know/Not sure*	432 (25.1)
Number of HPV vaccinations received (n = 809)
	1	92 (11.4)
	2	119 (14.8)
	3	300 (37.3)
	Don’t know/Not sure	294 (36.5)
Age at first HPV vaccination		14.9 ± 4.3
Age at second HPV vaccination		15.5 ± 3.7
Age at third HPV vaccination		16.1 ± 3.4

*Not included in logistic regression models.

Predictors of HPV vaccination

Respondent answers regarding attitudes and influences toward HPV and HPV vaccination are reported by number and percentage in Table 2. After statistically adjusting for all other variables, the 13 variable multiple logistic regression indicated that two variables concerning the number of vaccination doses and side effects of vaccination were not statistically significant. These variables were removed, and an 11 variable multiple logistic regression model was performed. Age (with ≤22 years as reference) was retained because age as a continuous variable was statistically significant in previous multiple logistic regression models. Table 3 presents the logistic regression model results. As seen in Table 3, a few predictor variables had small amounts of missing data with one predictor variable having 17 (1.3%) missing values. This reduced the sample size to 1,243 for the full model; however, this 3.6% missing data were not considered to large enough to bias the results.

Table 2.

Attitudes toward and influences on HPV and HPV vaccination

Variable		N (%)
Influence of doctor encouragement on getting HPV vaccine (n = 1707)
	More likely	1367 (80.1)
	No change/Less likely	340 (19.9)
Influence of parental encouragement on getting HPV vaccine (n = 1707)
	More likely	1305 (76.5)
	No change/Less likely	402 (23.5)
Influence of CDC recommendation on getting HPV vaccine (n = 1703)
	More likely	1098 (64.5)
	No change/Less likely	605 (35.5)
Influence of partner/significant other encouragement on getting HPV vaccine (n = 1706)
	More likely	1308 (76.7)
	No change/Less likely	398 (23.3)
Influence of friend encouragement on getting HPV vaccine (n = 1702)
	More likely	1002 (58.9)
	No change/Less likely	700 (41.1)
Most important influence on whether or not I receive the HPV vaccine (n = 1632)
	My personal views and beliefs	461 (30.0)
	My healthcare provider’s recommendation	881 (57.3)
	My parent/guardian’s recommendation	223 (14.8)
	My partner/significant other	67 (4.4)
Many of my friends have received the HPV vaccination (n = 1704)
	Agree	343 (20.1)
	Disagree	185 (10.9)
	Don’t know/Not sure	1176 (69.0)
I know someone who has had HPV (n = 1702)
	Agree	336 (19.7)
	Disagree	617 (36.3)
	Don’t know/Not sure	749 (44.0)
I have talked about HPV with my friends (n = 1704)
	Agree	441 (25.9)
	Disagree	950 (55.8)
	Don’t know/Not sure	313 (18.4)
I have talked about the HPV vaccine with my friends (n = 1722)
	Agree	467 (27.4)
	Disagree	942 (55.2)
	Don’t know/Not sure	297 (17.4)
I have talked about HPV with a partner/significant other (n = 1702)
	Agree	366 (21.5)
	Disagree	1029 (60.5)
	Don’t know/Not sure	307 (18.0)
I have talked about the HPV vaccine with a partner/significant other (n = 1700)
	Agree	332 (19.5)
	Disagree	1047 (61.6)
	Don’t know/Not sure	321 (18.9)
It is important to me that my partner/significant other has received the HPV vaccine (n = 1707)
	Agree	627 (36.7)
	Disagree	480 (28.1)
	Don’t know/Not sure	600 (35.2)

Table 3.

Predictors of HPV vaccination

Variable		Unadjusted Analyses		Adjusted Analysis
		OR [95% CI]	p-value	OR [95% CI]	p-value
Age
	Age ≤22 (ref)	1.00		1.00
	Age >22	0.37 [0.28, 0.47]	<0.0001	0.94 [0.61, 1.44]	0.7792
Race
	White (ref)	1.00		1.00
	Black/African American	1.36 [1.02, 1.83]	0.0152	1.61 [0.98, 2.62]	0.0585
	Other	0.79 [0.50, 1.24]	0.0901	0.67 [0.32, 1.40]	0.2916
Received provider recommendation for HPV vaccine
	No (ref)	1.00		1.00
	Yes	13.76 [10.29, 18.41]	<0.0001	4.99 [3.30, 7.53]	<0.0001
	Not sure	0.31 [0.14, 0.70]	0.0049	0.14 [0.05, 0.38]	0.0001
Received parent recommendation for HPV vaccine
	No (ref)	1.00		1.00
	Yes	40.48 [27.60, 59.38]	<0.0001	18.03 [11.14, 29.18]	<0.0001
	Not sure	10.87 [6.55, 17.75]	<0.0001	8.72 [4.51, 16.86]	<0.0001
Have a Primary Care Provider (PCP)
	No (ref)	1.00		1.00
	Yes	1.33 [1.05, 1.69]	0.0186	0.94 [0.62, 1.42]	0.7534
Frequency of visits to PCP
	<1/year (ref)	1.00		1.00
	1 per year	1.51 [1.07, 2.12]	0.0183	1.03 [0.58, 1.86]	0.9108
	2–3 per year	1.53 [1.13, 2.06]	0.0057	1.06 [0.63, 1.78]	0.8271
	4+ per year	1.33 [0.86, 2.05]	0.2007	0.74 [0.35, 1.54]	0.4194
Received Flu Vaccination
	No (ref)	1.00		1.00
	Yes	1.79 [1.42, 2.26]	<0.0001	1.49 [1.01, 2.20]	0.0471
Influence of doctor encouragement on getting HPV vaccine
	No change/Less likely (ref)	1.00		1.00
	More likely	4.97 [3.71, 6.66]	<0.0001	1.92 [1.20, 3.06]	0.0064
Many of my friends have received the HPV vaccination
	Disagree (ref)	1.00		1.00
	Agree	10.11 [6.39, 15.99]	<0.0001	2.25 [1.08, 4.70]	0.0314
	Don’t know/Not sure	3.41 [2.31, 5.04]	<0.0001	2.38 [1.27, 4.44]	0.0067
It is important to me that my partner/significant other has received the HPV vaccine
	Disagree (ref)	1.00		1.00
	Agree	8.34 [6.07, 11.46]	<0.0001	2.85 [1.73, 4.69]	<0.0001
	Don’t know/Not sure	1.74 [1.30, 2.34]	0.0002	1.24 [0.77, 1.99]	0.3720
HPV vaccine is so new that I want to wait a while before receiving it
	Neutral (ref)	1.00		1.00
	Agree	0.23 [0.14, 0.37]	<0.0001	0.19 [0.09, 0.37]	<0.0001
	Disagree	3.73 [2.84, 4.90]	<0.0001	1.84 [1.18, 2.88]	0.0076

Demographic variables

Although not statistically significant, undergraduate students older than 22 years were less likely to be vaccinated for HPV. Participants who reported receiving the flu vaccine were almost 1.5 times more likely to have been vaccinated for HPV [AOR = 1.49; 95% CI: 1.01, 2.20; p = .0471].

Recommendation variables

Individuals who had received a provider recommendation for HPV vaccination were almost 5 times more likely to have been vaccinated for HPV [AOR = 4.99; 95% CI: 3.30, 7.53; p < .0001] while those who reported being unsure if they had received a provider recommendation were over 8 times less likely to have been vaccinated for HPV [AOR = 0.14; 95% CI: 0.05, 0.38; p < .0001] compared to those who had not received a recommendation. Students who reported that a parent had recommended the HPV vaccination were 18 times more likely to have been vaccinated [AOR = 18.03; 95% CI: 11.14, 29.18; p < .0001], and those who were not sure if a parent had recommended the HPV vaccination were over 8 times more likely to have been vaccinated [AOR = 8.72; 95% CI: 4.51, 16.86; p < .0001] compared with students whose parents had not recommended the vaccine. Separate from provider recommendation, students were also asked a series of questions to gauge how influential some sources might be on their decision to get the HPV vaccine. Students who reported that a doctor’s encouragement would make them more likely to get the HPV vaccine were almost twice as likely to have been vaccinated [AOR = 1.92; 95% CI: 1.20, 3.06; p = .0064].

Social influence variables

Individuals who reported having friends who had received the HPV vaccination were over twice as likely to have been vaccinated [AOR = 2.25; 95% CI: 1.08, 4.70; p = .0314]. Additionally, those who reported not being sure if their friends had received the HPV vaccination were more than 2 times more likely to have been vaccinated [AOR = 2.38; 95% CI: 1.27, 4.44; p = .0067] compared with students who reported having friends not vaccinated. Subjects who felt it was important that their sexual partner had received the HPV vaccination were almost 3 times more likely to have been vaccinated [AOR = 2.85; 95% CI: 1.73, 4.69; p < .0001]. Individuals who agreed with the statement that the HPV vaccine was so new that they wanted to wait for a while before receiving it were almost 80% less likely to have received the vaccine [AOR = 0.19; 95% CI: 0.09, 0.37; p < .0001] while participants who disagreed with this statement were almost twice as likely to have been vaccinated [AOR = 1.84; 95% CI: 1.18, 2.88; p = .0076].

Discussion

The purpose of the current study was to evaluate college students’ HPV vaccination receipt and factors influencing HPV vaccination receipt, to inform future intervention efforts. The work presented here addresses the first step in that initiative: HPV vaccine receipt and factors associated with HPV vaccination in a sample of undergraduate students. In the current sample, 47% of participants reported having received the HPV vaccine, which is substantially lower than rates reported in other studies of college HPV vaccination. This is true for college populations in the same geographical region such as South Carolina, where one study found that 69.5% of the students surveyed reported having been vaccinated for HPV.²⁶ The results from Barrera et al. are comparable to other studies of college HPV vaccination in the country, including one at a small, Midwestern college where 60% of students reported having initiated the HPV vaccine series.²⁷ The rates of the current sample are well below both those from these studies and the national rates reported for 13–17-year-olds.¹¹ Interestingly, the remaining 53% of participants who did not respond ‘Yes’ to having received the HPV vaccine were almost evenly divided between reporting they had not received the HPV vaccine and responding Don’t know/Not sure. Notably, only 17% of all respondents in the current study reported completing all three doses. Low completion rates appear to be a concern for college students as, in their study, Rohde et al. indicated a completion rate of approximately 29% within their sample.²⁷ The concerningly low percentage of students who responded “Yes” to having received the HPV vaccination, nearly 17% below the national average, could be a sign of poor recall. Students were potentially vaccinated in their early teens and do not remember their vaccination history. For students to begin making informed decisions about their health it is necessary for parents to pass down the student’s health history.

Perhaps more striking than those failing to complete the series is that among those who reported having initiating the vaccine, over one-third stated they did not know or were not sure of the number of doses they had received. The substantial proportion of students who ‘Did not know/were unsure’ if they had been vaccinated may indicate that participants exhibited difficulty recalling their vaccination status due to series initiation during childhood, several years prior. If a student does not know if they have been vaccinated, then they cannot know if they need to complete the series. An additional challenge to “catch-up” vaccinations in the college population is variation in immunization information systems (IIS) and data sharing from state to state.³⁶ This creates an additional barrier to HPV vaccination for out-of-state college students who are unaware of their vaccination status. This presents a broader issue in care transition for students, less so the student’s willingness and ability to take ownership in their health but hesitancy to transition from their home provider to a university student health center. Being unaware of their vaccination status could also indicate that many students rely on their parents for health-related responsibilities such as details of their medical history and health decision making, as also evidenced by how strongly parent recommendation was associated with the likelihood of HPV vaccination. Empowering students to take on responsibility for and ownership of their health is an important step in future interventions to increase the likelihood that they will pursue and complete the HPV vaccination series or at least take the first step of determining their vaccination status. This limited knowledge also demonstrates another significant issue: students who are unsure of their HPV vaccination status likely do not understand the dangers associated with HPV, associated diseases, and the importance of prevention.³⁷

The low-reported completion rates indicate a potential gap in communication regarding HPV vaccine follow-up, which is a commonly seen problem in studies of broader populations.³⁸ Limited follow-up may also demonstrate low knowledge and/or awareness with respect to the number of vaccinations in the series that students need and/or have received. Another possible explanation for this failure to complete could be students’ relocation to college mid-series. For some or many students, this transition may make identifying new healthcare resources and transferring medical history or information inconvenient, cumbersome, and/or overwhelming. University health centers have an important role to play in engaging students for provision of general healthcare needs as well as providing more targeted health resources such as educating students on HPV vaccination availability and opportunities while away from their primary care provider.³⁹ However, these issues could be addressed through education, better communication, and helping students take a stronger role in advocacy of their healthcare.

Students who had received the influenza vaccine were significantly more likely to have initiated the HPV vaccination series. Receiving annual vaccination indicates proactive and preventive health behaviors and healthcare engagement; these individuals are likely more health-conscious, informed, and more willing to seek out preventive health services. Also, more visits to healthcare settings (such as for flu vaccination) provide increased opportunities for provider recommendation and education on the importance of HPV vaccination.^40–43 These additional contacts also offer convenient occasions for follow-up vaccine administration. It is important to note that young adults in the collegiate age group are generally healthy and less likely than both younger and older age groups to seek preventative health appointments. Children are seen frequently for immunization visits and older adults are likely to have annual health examinations.

Keeping with previous literature, we found that students in the current study who had received a healthcare provider recommendation were much more likely to be vaccinated for HPV.⁴⁴ Provider recommendation has been consistently demonstrated as one of the most influential factors for vaccine uptake.²⁸ Healthcare providers are typically a trusted source of medical guidance for patients, and people are willing to follow health-related suggestions, trusting their provider’s expertise. An unexpected finding in the current analysis was that students who reported not knowing if their provider recommended the vaccine were extremely less likely to be vaccinated. This finding has been reported in other recent literature and requires more indepth investigation, potentially through in-depth interviews.^45,46 Our initial interpretation is that this may further emphasize how influential a provider’s vaccination recommendation is – shown here by its absence. It is possible that lacking remembrance of a clear recommendation indicated to students that the vaccine must not be important or else their provider would have given a clear, strong recommendation. It is also important to note that participants are being asked to recall a provider recommendation that possibly occurred a decade prior. Further study will assist in clarifying such nuances. Interestingly, the current findings demonstrated that students in our sample far more influenced by parental encouragement than provider recommendation, a much greater discrepancy than previously observed in the literature.^32,47 Students who reported that a parent had recommended the HPV vaccination were 18 times more likely to have been vaccinated. However, mean age of first vaccination was 14.9, indicating that series initiation began while still living with their guardian. This suggests that parents may have increased influences in high school students before they leave to live independently as undergraduates. Even though this sample has moved away from their family unit for a collegiate education, parental guidance remains an important factor in their healthcare decisions. Continued educational outreach to parents, in addition to new efforts focused on young adults, remains a crucial part of increasing vaccination rates in this population.

Some of our findings regarding social influences have not been reported in the U.S. or have been presented inconsistently in previous studies.^{29,32,48–50} Some of these novel findings regarding HPV vaccination include increased likelihood of vaccination among students who reported having friends who have received the HPV vaccine. In addition to a higher likelihood of vaccination among those who agreed it was important that their partner/significant other had received the HPV vaccine. These findings may be interpreted as the actions of the students in our survey potentially being influenced by their friends and/or partners. It may also be true that students in this population tend to associate themselves with people who have perceptions and values similar to their own (i.e., “like-minded individuals”). Further study on how social influences, particularly peer influences, relate to HPV vaccination in this age group and the corresponding directional flow(s) will help us to better understand the most effective ways to utilize these findings as strategies to promote vaccination in the future.

Previous studies have found concerns about vaccine safety to be a predictor of vaccination status and associated with unvaccinated students.^28,29,32 As expected, students who said they wanted to wait to receive the vaccine because it was too new were less likely to be vaccinated.⁵¹ However, the perception that the vaccine is too new is likely the result of low vaccine knowledge since it has been available in the U.S. since 2006 and has almost two decades of safety data.^52,53 Comparatively, students who did not believe the vaccine was too new were more likely to have received the HPV vaccine.

Increasing HPV vaccination is the only primary preventative measure for lowering HPV infection rates, especially in the high-risk undergraduate population. In order to understand the most effective way to convey the information on HPV and vaccinations to college students and promote HPV vaccine receipt, researchers and practitioners must gather detailed information to inform interventions. Future studies should conduct focus groups to elicit specific feedback about the most effective outreach strategies for college populations, as recent research has indicated this audience is receptive to social media as a tool for distributing HPV information.³¹ More information is needed to determine the most effective messaging content, including thoughtful consideration of whether messaging would be most effective emphasizing HPV-associated genital warts or HPV-associated cancers.⁵⁴

Certain limitations should be acknowledged regarding the current study findings. Most participants were residents of Alabama (70.6%) and there was an underrepresentation of racial/ethnic groups, as is common in survey research; therefore, the sample is unlikely to be representative of all college students in the United States. Another limitation of the study is the inability to verify the participant’s HPV vaccination status and number of doses received (e.g. using electronic health record data or provider-verified documentation). It is also important to note, though, that college students do not all fall in the 18–26-year-old age range, and conversely, not all 18–26-year-olds are college students, so focusing on undergraduate populations does not encompass all young adults. Finally, as with all survey research, there is the risk of self-report, recall, social desirability, and other biases, although measures were taken wherever possible to reduce these (e.g., the inclusion of reversal items in the questionnaire design).

However, the study is strengthened by its considerable sample size (n = 1,725), far exceeding those of previous studies examining HPV knowledge and behaviors in college populations, which have ranged from n = 190–817, except for an instance when data was compiled from multiple institutions (n = 2,397).^23,55–58 Additionally, the current study population was distributed fairly evenly, demonstrating adequate representation of each class year. The overall college enrollment rate for young adults ages 18–24 in the U.S. had steadily increased to 41% in 2018 and was projected to rise, indicating that college students are a substantive population of young adults with whom to begin these intervention strategies.⁵⁹

Conclusion

The need for increased HPV vaccination is great among at-risk populations in the U.S., where HPV and associated diseases remain critical. In the current college population, rates of both HPV vaccine initiation and completion were low. Additionally, the rates reported here are substantially lower than those reported in previous studies of other college students and national averages of adolescents and young adults. These findings underscore the need for a greater focus on young adults, particularly college students, for specific, targeted interventions to increase HPV vaccination. The high proportions of students who reported limited knowledge of their HPV vaccination status illustrates a need to educate and encourage this population to take ownership of their health and medical decisions and to advocate for themselves as they transition into adulthood. The observed influence of parent and physician recommendations on increased vaccine uptake emphasizes the importance of incorporating these elements in future interventions, potentially as multi-level strategies. The effectiveness of peer influences – such as the value that individuals place on the vaccination status of friends and partners – should also be explored when exploring new strategies to encourage HPV vaccination in this population. Future work in this area should focus on filling the significant gap in promoting HPV vaccination among young adults, particularly college students. Materials and interventions targeted at this age group are greatly needed to appeal to their questions and concerns as young adults. College students should be engaged in the development process to achieve the most appropriate, effective health communications materials and intervention strategies. Researchers should solicit students’ input and feedback regarding messages, content, distribution channels, and components to generate substantial information that resonates with the target population and has the greatest likelihood of success.

Appendices.

Appendix Table A.

Characteristics of completers vs. non-completers

	Completers N (%)	Non-Completers N (%)
Year
Freshman	428 (24.9)	81 (30.0)
Sophomore	309 (17.9)	55 (20.4)
Junior	433 (25.2)	69 (25.6)
Senior	552 (32.1)	65 (24.1)
Missing/Not Reported	3	52
Age (Means ± SD)	22.1 ± 5.9	21.6 ± 6.8
Missing/Not Reported	15	55
Sex
Male	506 (29.6)	76 (28.3)
Female	1201 (70.4)	193 (71.7)
Missing/Not Reported	13	53
Race
Black or African American	342 (19.8)	87 (27.0)
White	1261 (73.1)	160 (49.7)
Other	122 (7.1)	75 (23.3)
Hispanic or Latino
Yes	76 (4.4)	12 (4.4)
No	1647 (95.6)	259 (95.6)
Missing/Not Reported	2	51
State, territory, or country
Alabama	1217 (70.6)	209 (64.9)
Other state, country or territory	508 (29.5)	113 (35.1)
Currently has health insurance
Yes	1485 (90.0)	171 (85.9)
No	165 (10.0)	23 (11.6)
Missing/Not Reported	75	128
Currently has a primary healthcare provider
Yes	1120 (65.2)	132 (66.7)
No	597 (34.8)	66 (33.3)
Missing/Not Reported	8	124
Frequency of healthcare provider visits
<1 time per year	360 (20.9)	48 (24.2)
1 time per year	420 (24.4)	58 (29.3)
2–3 times per year	783 (45.6)	71 (35.9)
4+ times per year	156 (9.1)	21 (10.6)
Missing/Not Reported	6	124
Receive the flu vaccine
Yes	630 (36.6)	70 (35.2)
No	650 (37.8)	76 (38.2)
Sometimes	417 (24.2)	48 (24.1)
Missing/Not Reported	28
Received the HPV vaccine
Yes	809 (47.0)	83 (34.7)
No	480 (27.9)	47 (23.6)
Don’t Know/Not sure*	432 (25.1)	69 (34.7)
Missing/Not Reported	4	123

Appendix Table B.

Demographics characteristics based on HPV vaccination status (completers only N = 1721)

Variable	No (N = 480)	Yes (N = 809)	Don’t Know/ Not Sure (N = 432)	p-value
Age (Mean ± SD)	25.0 ± 8.5	20.9 ± 3.3	21.3 ± 5.2	<0.0001
Missing/Not Reported	10	2	3
Year				0.0016
Freshman	97 (20.3)	212 (26.3)	118 (27.3)
Sophomore	7816.3)	137 (17.0)	94 (21.8)
Junior	119 (24.8)	209 (25.9)	104 (24.1)
Senior	185 (38.6)	249 (30.9)	116 (26.9)
Missing/Not Reported	1	2	0
Sex				<0.0001
Male	150 (31.6)	147 (18.4)	209 (48.6)
Female	325 (68.4)	651 (81.6)	221 (51.4)
Missing/Not Reported	5	11	2
Race				0.0865
Black or African American	81 (16.9)	178 (22.0)	82 (19.0)
White	362 (75.4)	584 (72.2)	313 (72.4)
Other	37 (7.7)	47 (5.8)	37 (8.6)
Hispanic or Latino				0.7527
Yes	18 (3.8)	37 (4.6)	20 (4.6)
No	460 (96.2)	772 (95.4)	412 (95.4)
Missing/Not Reported	2	0	0
Currently has health insurance				0.0156
Yes	405 (87.9)	724 (92.2)	355 (88.1)
No	56 (12.1)	61 (7.8)	48 (11.9)
Missing/Not Reported	19	24	29
Insurance Type				0.2765
Private	376 (92.8)	666 (92.1)	325 (91.8)
Public	20 (4.9)	50 (6.9)	25 (7.1)
Other	9 (2.2)	7 (1.0)	4 (1.1)
Missing/Not Reported	75	86	78
PCP				0.0092
Yes	298 (62.5)	556 (68.9)	264 (61.3)
No	179 (37.5)	251 (31.1)	167 (38.7)
Missing/Not Reported	3	2	1
Frequency of healthcare provider visits				0.0061
<1 time per year	113 (23.7)	139 (17.2)	108 (25.0)
1 time per year	109 (22.9)	202 (25.0)	107 (24.8)
2–3 times per year	206 (43.2)	387 (47.9)	190 (44.0)
4+ times per year	49 (10.3)	80 (9.9)	27 (6.2)
Missing/Not Reported	3	1	0

Appendix Table C.

Health-related characteristics based on HPV vaccination status (completers only N = 1721)

	No (N = 480)	Yes (N = 809)	Don’t Know/ Not Sure (N = 432)	p-value
Receive the flu vaccine				<0.0001
No	220 (46.1))	260 (32.3)	169 (40.8)
Yes	156 (32.7)	342 (42.5)	132 (31.9)
Sometimes	101 (21.2)	202 (25.2)	113 (27.3)
Missing/Not Reported	3	5	18
Received provider recommendation for HPV vaccine				<0.0001
No	278 (57.9)	101 (12.5)	277 (65.5)
Yes	140 (29.2)	700 (86.6)	126 (29.8)
Not sure	62 (12.9)	7 (0.9)	20 (4.7)
Missing/Not Reported	0	1	9
Received parent recommendation for HPV vaccine				<0.0001
No	421 (87.7)	160 (19.8)	231 (53.6)
Yes	36 (7.5)	554 (68.5)	47 (10.9)
Not sure	23 (4.8)	95 (11.7)	153 (35.5)
Missing/Not Reported	0	9	1
Influence of doctor encouragement on getting HPV vaccine				<0.0001
More Likely	301 (63.2)	718 (89.5)	348 (81.5)
No change	160 (33.6)	79 (9.9)	74 (17.3)
Less likely	15 (3.2)	5 (0.6)	5 (1.2)
Missing/Not Reported	4	7	5
Many of my friends have received the HPV vaccination				<0.0001
Agree	60 (12.7)	271 (33.8)	11 (2.6)
Disagree	94 (19.8)	42 (5.2)	49 (11.5)
Don’t know/Not sure	320 (67.5)	488 (60.9)	367 (85.9)
Missing/Not Reported	6	8	5
It is important to me that my partner has received the HPV vaccine				<0.0001
Agree	83 (17.5)	456 (56.8)	87 (20.3)
Disagree	208 (43.9)	137 (17.1)	134 (31.3)
Don’t know/Not sure	183 (38.6)	210 (26.1)	207 (48.4)
Missing/Not Reported	6	6	4
HPV vaccine is so new that I want to wait a while before receiving it				<0.0001
Agree	123 (26.0)	26 (3.3)	86 (20.2)
Neutral (ref)	172 (36.4)	159 (19.9)	211 (49.7)
Disagree	178 (37.6)	614 (76.8)	128 (30.1)
Missing/Not Reported	7	10	7

Appendix Table D.

Multinomial model: predictors of HPV vaccination (no vaccination reference response)

	Received HPV Vaccination		Don’t Know/Not Sure
Variable	OR [95% CI]	p-value	OR [95% CI]	p-value
Age (≤22 ref)
Age >22	0.82 [0.56, 1.21]	0.3110	0.42 [0.29, 0.60]	<0.0001
Race (White ref)
Black/African American	1.67 [1.09, 2.56]	0.0192	1.17 [0.78, 1.76]	0.4413
Other	0.91 [0.48, 1.75]	0.7856	0.98 [0.55, 1.74]	0.9441
Received provider recommendation for HPV vaccine – No (ref)
Yes	4.23 [2.88, 6.21]	<0.0001	0.63 [0.43, 0.92]	0.0176
Not sure	0.16 [0.06, 0.41]	0.0002	0.16 [0.08, 0.33]	<0.0001
Received parent recommendation for HPV vaccine – No (ref)
Yes	15.75 [9.97, 24.90]	<0.0001	2.95 [1.73, 5.02]	0.0001
Not sure	9.37 [5.18, 16.96]	<0.0001	13.86 [8.05, 23.86]	<0.0001
Have a Primary Care Provider (PCP) – No ref
Yes	0.98 [0.68, 1.42]	0.9344	1.17 [0.84, 1.62]	0.3574
Frequency of visits to PCP – <1/year (ref)
1 per year	1.07 [0.64, 1.79]	0.7977	1.05 [0.67, 1.65]	0.8384
2–3 per year	1.39 [0.88, 2.20]	0.1617	0.81 [0.54, 1.22]	0.3099
4+ per year	0.96 [0.49, 1.88]	0.9108	0.51 [0.26, 0.98]	0.0446
Received Flu Vaccination – No ref
Yes	1.22 [0.86, 1.74]	0.2593	1.16 [0.85, 1.59]	0.3588
Influence of doctor encouragement on getting HPV vaccine – No change/Less likely (ref)
More likely	2.13 [1.39, 3.27]	0.0005	2.51 [1.73, 3.64]	<0.0001
Many of my friends have received the HPV vaccination – Disagree (ref)
Agree	1.99 [1.00, 3.96]	0.0490	0.34 [0.15, 0.76]	0.0090
Don’t know/Not sure	1.68 [0.96, 2.94]	0.0716	1.98 [1.26, 3.11]	0.0032
It is important to me that my partner has received the HPV vaccine – Disagree (ref)
Agree	2.59 [1.65, 4.08]	<0.0001	1.30 [0.82, 2.04]	0.2623
Don’t know/Not sure	1.23 [0.81, 1.88]	0.3332	1.05 [0.73, 1.50]	0.8069
HPV vaccine is so new that I want to wait a while before receiving it – Neutral (ref)
Agree	0.21 [0.11, 0.39]	<0.0001	0.72 [0.48, 1.09]	0.1182
Disagree	1.91 [1.28, 2.83]	0.0014	0.63 [0.43, 0.91]	0.0133

Funding Statement

Research support was provided, in part, from Health Resources & Services Administration (HRSA) grant T0BHP30027 (PI: Perkins, A).

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.