The association of caregiver attitudes, information sources, and trust with HPV vaccine initiation among adolescents

ABSTRACT

This study described caregiver attitudes and the information sources they access about HPV vaccination for adolescents and determined their influence on human papillomavirus (HPV) vaccination initiation. An online survey was administered to 1,016 adults in July 2021. Participants were eligible if they were the caregiver of a child aged 9–17 residing in Mississippi, Arkansas, Tennessee, Missouri, and select counties in Southern Illinois. Multivariate logistic regression was used to estimate the association of caregiver attitudes and information sources with HPV vaccination. Information from doctors or healthcare providers (87.4%) and internet sources other than social media (31.0%) were the most used sources for HPV vaccine information. The highest proportion of caregivers trusted their doctor or healthcare providers (92.4%) and family or friends (68.5%) as sources of information. The HPV vaccine series was more likely to be initiated in children whose caregivers agreed that the vaccine is beneficial (AOR = 4.39, 95% CI = 2.05, 9.39), but less likely with caregivers who were concerned about side effects (AOR = 0.61, 95% CI = 0.42, 0.88) and who received HPV vaccination information from family or friends (AOR = 0.57, 95% CI = 0.35, 0.93). This study found that caregivers’ attitudes, information sources, and trust in those sources were associated with their adolescent’s HPV vaccination status. These findings highlight the need to address attitudes and information sources and suggest that tailored interventions considering these factors could increase HPV vaccination rates.

Introduction

Human papillomavirus (HPV) is a leading cause of six types of cancer and genital warts in both men and women.¹ HPV causes 91% of cervical and anal cancers, 75% of vaginal cancers, 70% of oropharyngeal cancers, 69% of vulvar cancers, and 63% of penile cancers.¹ More than 80% of men and women are estimated to acquire HPV before the age of 45.² Adolescents and young adults are especially at risk of contracting HPV. Prior to the introduction of the HPV vaccine in 2006, almost half of new infections occurred among adolescents and young adults between the ages of 15 and 24.³ However, the HPV vaccine has been shown to be highly effective in preventing HPV infection and associated conditions.⁴ Currently, HPV vaccination is recommended by the Centers for Disease Control and Prevention’s Advisory Committee for Immunization Practices at age 11 or 12 years, but can be started as early as age 9.⁵ The American Cancer Society and American Academy of Pediatrics recommend initiating HPV vaccination between the ages of 9 and 12.^6,7 Despite yearly improvements, rates of HPV vaccination still fall short of the Healthy People 2030 goal of 80% coverage by the age of 15, with only 75.1% of adolescents between 13- and 17-years-old having received at least one dose of the vaccine and 58.6% having received all necessary doses in 2021.^8,9

Initiation of HPV vaccination has been reported to be influenced by various individual- and caregiver-level sociodemographic factors, including geographic region, religion, and political affiliation.^9–11 For example, Franco et al. found that age and gender were individual predictors of completing HPV vaccination, with females and older teenagers being more likely to complete the HPV vaccination series than males and younger teenagers, respectively.¹¹ Young age has also been shown in previous studies as a common reason to postpone or refuse the HPV vaccine in other studies, with vaccination uptake and completion rates being higher in older children.¹² However, beyond sociodemographic factors, caregiver knowledge and attitudes may influence initiation of HPV vaccination. A recent systematic review by Peterson et al. reported that knowledge and attitudes about HPV and the vaccine can serve as barriers or facilitators for a child’s HPV vaccine uptake.¹⁰ Other studies have showed that parents with higher awareness and knowledge about HPV vaccination were more likely to initiate the vaccine.^13–15 A previous analysis we conducted showed that higher parental vaccine hesitancy was linked to decreased adolescent HPV vaccination uptake.¹⁶ In addition, parents with positive attitudes about the HPV vaccine have been reported to have higher odds of vaccinating their child, whereas those with barriers or concerns about safety had lower odds.^{13–15,17,18}

While several studies have focused on sociodemographic characteristics, knowledge, and attitudes in relation to HPV vaccine uptake,^9–18 less attention has been paid to caregiver sources of information and trust in these information sources in states with low vaccination states. Recommendations from healthcare providers for HPV vaccination have been shown to increase rates of HPV initiation and completion,^15,18 and Fu et al. showed that children whose parents more highly trust healthcare providers have higher odds of having received the HPV vaccine, though that study only involved African American participants.¹⁹ While it is important to address disparities in healthcare, it is also crucial to understand health behaviors across all racial and ethnic groups to improve overall statewide public health outcomes. Additionally, compared to African American and Hispanic adolescents, HPV vaccination coverage is lower among non-Hispanic White adolescents,²⁰ suggesting a need for further research investigating trust and HPV vaccination among this population.

Some studies suggest that parents are most likely to use medical professionals and television for information.^15,21 Additionally, McRee et al. reported that parents who had heard of the HPV vaccine through the internet were more willing to vaccinate their child.²² Thompson et al. found that parents that viewed information about HPV on social media as credible were more likely to have vaccinated their child, as well as those that had higher internet verification skills.²³ While prior studies have shown the potential impact of healthcare providers, social media, and the internet on vaccination willingness, it is still unknown which information sources caregivers trust the most and how this trust affects HPV vaccination initiation. Additionally, prior studies have not investigated the association between caregiver use of and trust in family or friends, books, newspapers, or magazines, television or radio, government agencies, or religious organizations or leaders and HPV vaccination initiation. Understanding these factors could assist with the development of more effective, tailored interventions to increase HPV vaccine uptake. For example, this could guide providers, public health organizations, and governmental health agencies to utilize the most trusted and frequently used information sources for communicating about the HPV vaccine to increase the reach and impact of messaging, potentially improving HPV vaccination rates. Multi-level interventions that target a combination of individuals, healthcare providers, healthcare systems, community settings, and policy have shown promise in increasing HPV vaccination rates.²⁴

This study aimed to understand caregiver knowledge and attitudes regarding HPV in five states which we selected because they have vaccination rates below the national average and are in our surrounding area. There is a gap in the literature regarding which information sources caregivers trust the most, including family or friends, books, newspapers, magazines, television or radio, government agencies, and religious organizations or leaders, and how trust in information sources influences HPV vaccination initiation among states with low HPV vaccination rates. We sought to fill this gap by identifying the associations between caregiver use of and trust in various information sources and HPV vaccination initiation by the oldest child.To address these goals, we surveyed caregivers of children aged between 9 and 17 residing in five Midwestern and Southern states.

Materials and methods

Study design and data collection

In this study, we included participants residing in 5 Midwestern and Southern states with low or moderate HPV vaccination coverage according to data from the 2020 National Immunization Survey – Teen, where coverage refers to the approximate proportion of individuals who are up to date on the HPV vaccine.⁹ Adolescents were deemed up to date if they had received three or more doses of the HPV vaccine or if they had received two doses and the first dose was administered before their 15th birthday, and the interval between the first and second doses was at least five months minus four days, which is the minimum time between the first and second doses.⁹ States included were Mississippi, Arkansas, Tennessee, Missouri, and 42 select counties in southern or central Illinois. Shato et al. includes a more detailed description of the study population and map.¹⁶ These are all states with below average (Mississippi, Arkansas, Tennessee, Missouri) or average (Illinois) HPV vaccination coverage among teenagers between the ages of 13 and 17.⁹ None of these states have schools that require girls or boys to have received the HPV vaccine.²⁵ We selected these specific states because of their proximity to our location allowing us to potentially work to address the findings surrounding low vaccine rates in this region. An online survey was administered in July 2021, and participants were eligible if they were the parent or caregiver of at least one child between the ages of 9 and 17. Respondents were recruited through Qualtrics® Online Panels (Qualtrics, Provo, UT). The panel service emailed a survey link to potential participants residing in the geographic target area and ensured the sample was at least 20% non-White and 30% rural dwelling. These quotas were developed to ensure that the sample was representative of the states sampled. Our study was approved by the Washington University in St. Louis Institutional Review Board (IRB Protocol #202106066). Since our study was a brief cross-sectional survey, direct benefits and risks to participants were minimal.

Measures

The primary outcome of this study was HPV vaccination initiation by the oldest child. Parents or caregivers, hereafter referred to as caregivers, were asked “Has child 1 (oldest child) been vaccinated against HPV?.” If caregivers answered “yes,” we counted this as the child having initiated HPV vaccination, while if caregivers answered “no,” “not yet, I haven’t decided,” “not yet, I am planning to,” or “don’t know,” we counted this as the child not having initiated HPV vaccination. Attitudes about the HPV vaccine were captured using a 4-point Likert scale ranging from strongly disagree to strongly agree. For analysis purposes, we dichotomized these attitudes (strongly agree, agree vs. disagree, strongly disagree). Knowledge about cancer prevention was assessed by asking “Do you think the HPV vaccine can prevent some cancers?” with response of options of “yes,” “no,” and “don’t know or not sure.” Caregivers were asked to select all sources from which they got information about the HPV vaccine and then assess how much they trusted the information they received from those sources, measured on a 5-point Likert-type scale (a great deal, a lot, a moderate amount, a little, none at all). The types of sources were: doctor or health care provider; family or friends; social media (e.g. Twitter, Facebook, Instagram, etc.); other internet sources; books, newspapers or magazines; television or radio; government agencies; religious organizations or leaders; other, with a chance to write in another source. Trust in information sources were dichotomized (a great deal, a lot, a moderate amount vs. a little or none at all).

Sociodemographic variables for the child and caregiver that were found to be significantly associated with HPV vaccine initiation in previous analyses using the survey data were included as covariates.¹⁶ The child’s demographic characteristics included gender (male or female) and vaccination history for the meningococcal conjugate vaccine and seasonal influenza vaccine (for the most recent season), and current age (9–12 years, 13+ years). The variables for the caregivers were evaluated categorically, including age (25–34 years, 35–44 years, or 45+ years), and gender (male or female). The meningococcal conjugate vaccine is required by sixth grade for Illinois, seventh grade for Arkansas, and eighth grade for Missouri.²⁶ We chose to include the meningococcal conjugate vaccine as a covariate because similar to the HPV vaccine, it is a routinely recommended vaccine for the same age group that is not always required by states, and so provides a comparator for adolescent vaccine uptake.

Analysis

Differences in initiation for HPV vaccination were compared between the above knowledge, attitude, and information source questions, using chi-squared tests. Significant variables were added together into a multivariable logistic regression model. The final model was adjusted for oldest child’s gender, age, vaccination with the meningococcal conjugate and the most recent season’s influenza, caregiver’s age, and caregiver’s gender. Odds ratios were reported for both the unadjusted and adjusted models. Statistical significance was assessed as p < .05. All analyses were conducted in SAS version 9.4 (SAS Institute, Cary, NC).

Results

Participant characteristics

1,016 caregivers reported on 1,645 children between the ages of 9 and 17. After excluding caregivers who had never heard of HPV (n = 90) from the analysis, data from the rest of the caregivers and their oldest child between 9 and 17 were analyzed (n = 926). The majority of the 926 caregivers were between 35 and 44 years old (53.1%) and were female (78.5%) (Table 1). The highest proportion of the 926 children were aged 13 years and older (70.4%) and had received the meningococcal conjugate vaccine (77.5%). Overall, most children (57.7%) had not initiated the HPV vaccine series. The median age of children at the time of receiving their first dose of the HPV vaccine (n = 392) was 14 years (interquartile range = 12–16), with a median time since receiving the first dose of the HPV vaccine of 2 years (interquartile range = 1–3).

Table 1.

Child and caregiver sociodemographic characteristics of respondents, stratified by initiation of HPV vaccine.

	Overall (N = 926)		No HPV Initiation (N = 534)		HPV series initiated (N = 392)
	N	%	N	%	N	%
Child’s age
9–12	274	29.59	223	41.76	51	13.01
13+	652	70.41	311	58.24	341	86.99
Child’s gender
Male	471	51.82	288	54.75	183	47.78
Female	438	48.18	238	45.25	200	52.22
Meningococcal conjugate vaccine
Vaccinated	718	77.54	379	70.97	399	86.48
Not vaccinated	90	9.72	66	12.36	24	6.12
Missing	118	12.74	89	16.67	29	7.40
Seasonal flu vaccine for the 2020–2021 season
Vaccinated	450	49.34	207	39.66	243	62.31
Not vaccinated	462	50.66	315	60.34	147	37.69
Caregiver age
25–34	209	22.69	151	28.44	58	14.87
35–44	489	53.09	269	50.66	220	56.41
45+	223	24.21	111	20.90	112	28.72
Caregiver gender
Male	198	21.55	114	21.63	84	21.43
Female	721	78.45	413	78.37	308	78.57
State
Arkansas	175	18.90	100	18.73	75	19.13
Illinois	68	7.34	35	6.55	33	8.42
Missouri	294	31.75	173	32.40	121	30.87
Mississippi	57	6.16	29	5.43	28	7.14
Tennessee	332	35.85	197	36.89	135	34.44
Caregiver race
Non-Hispanic White	697	76.09	410	77.65	287	73.97
Non-Hispanic Black	126	13.76	71	13.45	55	14.18
Some other race	93	10.15	47	8.90	46	11.86
Rurality (RUCA categorization)
Metropolitan	631	68.14	367	68.73	264	67.35
Micropolitan	158	17.06	88	16.48	70	17.86
Rural	137	14.79	79	14.79	58	14.80

Table abbreviations: RUCA = rural-urban commuting area.

Information needed for decision-making on HPV vaccination

Of the children who had not initiated the HPV vaccine series (n = 534), 19% (n = 102) of the caregivers had not yet decided whether to vaccinate their children. To make the decision to vaccinate their children, the highest proportion needed information on safety of the vaccine (53.9%, n = 55), followed by information on how well the vaccine works (47.1%, n = 48), and a doctor or healthcare provider recommendation (45.1%, n = 46) as shown in Table 2 out of the 102 respondents included.

Table 2.

Information needed to make a decision whether to receive the HPV vaccine among undecided caregivers (N = 102).

	N	%
What information do you need to make this decision?a
Information on safety of the vaccine	55	53.9
Information on how well the vaccine works	48	47.1
A physician’s recommendation	46	45.1
If it is safe right now during the COVID-19 pandemic	18	17.7
I want to know what other parents are doing	10	9.8
My child’s school to require it	6	5.9
Some other reasonb	14	13.7

^aAsked to caregivers who responded “Not yet, I haven’t decided” to the question “Has your oldest child been vaccinated against HPV?”

^bWrite in responses were provided such as child’s input and financial.

HPV vaccine attitudes and information sources

Table 3 shows the association between HPV vaccine attitudes and sources of information with HPV vaccination initiation. The majority of caregivers regardless of their child’s vaccination status agreed or strongly agreed that the HPV vaccine is effective (82.6%), beneficial for their children (80.3%), and they do or would do what their healthcare provider recommends about the vaccine (82.0%). Over half of caregivers (60.2%) agreed that HPV vaccine can prevent some cancers. There were significant differences between HPV vaccine attitudes and HPV vaccination initiation. For example, a significantly higher proportion of caregivers with children who had received at least one dose of the HPV vaccine agreed that the HPV vaccine can prevent some cancers compared with those caregivers of children who had not initiated the HPV vaccine series (76.8% vs. 47.9%, p < .001). In contrast, a significantly higher proportion of caregivers with children who had not initiated HPV vaccination, compared to their counterparts, agreed that the HPV vaccine had not been around long enough to be sure about its safety (53.1% vs. 31.2%, p < .001) and were concerned about the side effects of the HPV vaccine (64.5% vs. 38.5%, p < .001).

Table 3.

Caregiver HPV vaccine attitudes, information sources, and adolescent HPV vaccine initiation.

	Overall (N = 926)		No HPV Initiation (N = 534)		HPV series initiated (N = 392)
	N	%	N	%	N	%	p-valuea
HPV VACCINE ATTITUDES
HPV vaccine
… is effective							<.0001
Strongly disagree/disagree	161	17.42	129	24.20	32	8.18
Agree/Strongly agree	763	82.58	404	75.80	359	91.82
… is beneficial for my child(ren)							<.0001
Strongly disagree/disagree	182	19.70	157	29.46	25	6.39
Agree/Strongly agree	742	80.30	376	70.54	366	93.61
… follow HCP’s recommendations							<.0001
Strongly disagree/disagree	166	17.97	131	24.58	35	8.95
Agree/Strongly agree	758	82.03	402	75.42	356	93.61
… has not been around long enough to be sure it’s safe							<.0001
Strongly disagree/disagree	519	56.17	250	46.90	269	68.80
Agree/Strongly agree	405	43.83	283	53.10	122	31.20
… I am concerned about the side effects							<.0001
Strongly disagree/disagree	429	46.48	189	35.46	240	61.54
Agree/Strongly agree	494	53.52	344	64.54	150	38.46
Do you think the HPV vaccine can prevent some cancers?							<.0001
Yes	557	60.15	256	47.94	301	76.79
No	77	8.32	58	10.86	19	4.85
I don’t know/I’m not sure	292	31.53	220	41.2	72	18.37
HPV VACCINE INFORMATION
Sources of HPV vaccine information
Doctor or health care provider	809	87.37	443	82.96	366	93.37	<.0001
Family or friends	151	16.31	109	20.41	42	10.71	<.0001
Social media (e.g. Twitter, Facebook, Instagram, etc.)	98	10.58	63	11.8	35	8.93	.1608
Other internet sources	287	30.99	186	34.83	101	25.77	.0032
Books, newspapers, or magazines	112	12.1	68	12.73	44	11.22	.4864
Television or radio	137	14.79	92	17.23	45	11.48	.0149
Government agencies	75	8.1	38	7.12	37	9.44	.2005
Religious organizations or leaders	16	1.73	10	1.87	6	1.53	.6931
Trust in HPV vaccine information sources
Doctor or health care provider							<.0001
A great deal, a lot, or a moderate amount	856	92.44	474	88.76	382	97.45
A little or none at all	70	7.56	60	11.24	10	2.55
Family or friends							.0050
A great deal, a lot, or a moderate amount	634	68.47	346	64.79	288	73.47
A little or none at all	292	31.53	188	35.21	104	26.53
Social media (e.g. Twitter, Facebook, Instagram, etc.)							.0048
A great deal, a lot, or a moderate amount	271	29.27	137	25.66	134	34.18
A little or none at all	655	70.73	397	74.34	258	65.82
Other internet sources							.0349
A great deal, a lot, or a moderate amount	487	52.59	265	49.63	222	56.63
A little or none at all	439	47.41	269	50.37	170	43.37
Books, newspapers, or magazines							.0394
A great deal, a lot, or a moderate amount	586	63.28	323	60.49	263	67.09
A little or none at all	340	36.72	211	39.51	129	32.91
Television or radio							<.0001
A great deal, a lot, or a moderate amount	446	48.16	225	42.13	221	56.38
A little or none at all	480	51.84	309	57.87	171	43.62
Government agencies							<.0001
A great deal, a lot, or a moderate amount	505	54.54	252	47.19	253	64.54
A little or none at all	421	45.46	282	52.81	139	35.46
Religious organizations or leaders							<.0001
A great deal, a lot, or a moderate amount	356	38.44	176	32.96	180	45.92
A little or none at all	570	61.56	358	67.04	212	54.08

^aChi-Square test.

Regarding information sources (Table 3), the majority of the caregivers got information about the HPV vaccine from their doctor or healthcare providers (87.4%), followed by internet sources excluding social media (31.0%), and then from family or friends (16.3%). The highest proportion of caregivers trusted their doctor or healthcare providers (92.4%), family or friends (68.5%), as well as books, newspapers, or magazines (63.3%) as sources of information about the HPV vaccine. A significantly higher proportion of caregivers with children with at least one dose of the HPV vaccine received information about the vaccine from their doctor or healthcare providers (93.4% vs. 83.0%, p < .001). On the other hand, a significantly higher proportion of caregiver with children who had not initiated the HPV vaccine series, compared to their counterparts, got information from other internet sources (34.8% vs. 25.8%, p = .003) and from family or friends (20.4% vs. 10.7%, p < .001). Caregivers having at least a moderate amount of trust in any given information source was positively associated with HPV vaccine initiation for their oldest child.

Adjusted and unadjusted logistic regression

In the unadjusted model (Table 4), those who agreed that the HPV vaccine was effective (OR = 3.58, 95% CI = 2.37, 5.41) and beneficial (OR = 6.11, 95% CI = 3.91, 9.55) were far more likely to have initiated the HPV vaccine series for their oldest child. Those who followed the doctor’s recommendations were also more likely to have initiated the vaccine series (OR = 3.32, 95% CI = 2.22, 4.94). Caregivers who thought that the vaccine was too new to determine safety (OR = 0.4, 95% CI = 0.31–0.54) and those who were concerned about side effects (OR = 0.34, 95% CI = 0.26, 0.45) were less likely to have initiated the HPV vaccine. Those who believed that the HPV vaccine could prevent some cancers were almost four times as likely to have initiated the series (OR = 3.59, 95% CI = 2.08, 6.19) then those who did not.

Table 4.

Unadjusted and adjusted logistic regression for association between HPV vaccine attitudes and information sources and HPV vaccination initiation.

	Unadjusted OR (LCL, UCL)	Adjusteda OR (LCL, UCL)
HPV VACCINE ATTITUDES
HPV vaccine
… is effective
Strongly agree or agree	3.58 (2.37, 5.41)	0.70 (0.35, 1.41)
Disagree or strongly disagree	Ref	Ref
… is beneficial for my child(ren)
Strongly agree or agree	6.11 (3.91, 9.55)	4.39 (2.05, 9.39)
Disagree or strongly disagree	Ref	Ref
… follow HCP’s recommendations
Strongly agree or agree	3.32 (2.22, 4.94)	1.14 (0.63, 2.06)
Disagree or strongly disagree	Ref	Ref
… has not been around long enough to be sure it’s safe
Strongly agree or agree	0.40 (0.31, 0.54)	0.74 (0.50, 1.09)
Disagree or strongly disagree	Ref	Ref
… I am concerned about the side effects
Strongly agree or agree	0.34 (0.26, 0.45)	0.61 (0.42, 0.88)
Disagree or strongly disagree	Ref	Ref
Do you think the HPV vaccine can prevent some cancers?
I don’t know/I’m not sure	1.00 (0.56, 1.79)	0.51 (0.23, 1.11)
Yes	3.59 (2.08, 6.19)	1.06 (0.49, 2.27)
No	Ref	Ref
HPV VACCINE INFORMATION
Sources of HPV vaccine information
Doctor or health care provider
Yes	2.89 (1.83, 4.57)	1.66 (0.93, 3.00)
No	Ref	Ref
Family or friends
Yes	0.47 (0.32, 0.67)	0.57 (0.35, 0.93)
No	Ref	Ref
Other internet sources
Yes	0.65 (0.49, 0.87)	0.64 (0.43, 0.94)
No	Ref	Ref
Television or radio
Yes	0.62 (0.43, 0.91)	0.70 (0.43, 1.16)
No	Ref	Ref
Trust in HPV vaccine information sources
Doctor or health care provider
A great deal, a lot, or a moderate amount	4.83 (2.44, 9.57)	1.35 (0.55, 3.34)
A little or none at all	Ref	Ref
Family or friends
A great deal, a lot, or a moderate amount	1.51 (1.13, 2.00)	0.98 (0.66, 1.44)
A little or none at all	Ref	Ref
Other internet sources
A great deal, a lot, or a moderate amount	1.33 (1.02, 1.72)	1.27 (0.85, 1.90)
A little or none at all	Ref	Ref
Television or radio
A great deal, a lot, or a moderate amount	1.78 (1.36, 2.31)	1.48 (0.99, 2.21)
A little or none at all	Ref	Ref

Table abbreviations: LCL = lower confidence limit, OR = odds ratio, Ref = reference, UCL = upper confidence limit.

^aAdjusted for all other variables in the table in addition to child gender, meningococcal conjugate vaccine uptake, seasonal flu vaccine uptake for the 2020–2021 season, child age, caregiver age, and caregiver gender.

Caregivers who reported getting HPV vaccination information from their doctor or health care provider were nearly three times as likely to have initiated the series (OR = 2.89, 95% CI = 1.83, 4.57). Those who got their information from family or friends (OR = 0.47 95% CI = 0.32, 0.67), other internet sources (OR = 0.65, 95% CI = 0.49, 0.87), or television or radio (OR = 0.62, 95% CI = 0.43, 0.91) were less likely to have initiated the series. Having at least a moderate amount of trust in HPV information from a doctor or health care provider, family or friends, social media, other internet sources, books, newspapers, or magazines, television or radio, government agencies, or religious organizations or leaders was associated with increased odds of having initiated the HPV vaccine series.

In the adjusted model (Table 4), the HPV vaccine series was more likely to be initiated in children whose caregivers agreed that the HPV vaccine is beneficial (AOR = 4.39, 95% CI = 2.05, 9.39), but less likely with caregivers who were concerned about side effects (AOR = 0.61, 95% CI = 0.42, 0.88). Caregivers who received their information from family or friends (AOR = 0.57, 95% CI = 0.35, 0.93) or from other internet sources (AOR = 0.64, 95% CI = 0.43, 0.94) were less likely to have initiated the HPV vaccine. All other effects that were significant in bivariate analysis were attenuated in the adjusted model and no longer significant.

Discussion

In this study, caregiver attitudes, information sources, and trust in information sources were associated with HPV vaccine initiation by the oldest child. We found that among adolescents that were vaccinated, their caregivers were significantly more likely to express positive attitudes about the HPV vaccine and use healthcare providers as an information source, while the caregivers of unvaccinated children were significantly more likely to express concerns about side effects. Our findings regarding caregiver attitudes are consistent with the literature.^{13–15,17,18} Novel results from our study include the findings that caregivers who received information about the HPV vaccine from family or friends or internet sources other than social media were less likely to have initiated the HPV vaccine series for their child. Additionally, we found that higher trust in information sources including family or friends, television or radio, social media, other internet sources, government agencies, and religious organizations or leaders was associated with HPV vaccine initiation, which has not been found in other studies.

In the adjusted model, we observed a notable attenuation of the associations previously identified in the unadjusted model. For example, beliefs regarding the effectiveness of the HPV vaccine, which appeared highly influential in the unadjusted model, lost significance after adjusting for covariates, allowing us to see whether there is an independent effect of that variable after controlling for all other covariates. The fact that several covariates lose statistical significance upon adjustment suggests that some covariates such as sociodemographic factors or highly trusted information sources may act as mediating factors in many of the vaccine attitude associations.

Given the high trust in and use of health care providers as an information source for HPV, there is an opportunity for public health interventions to make use of them. Health care providers could increase knowledge and positive attitudes about the HPV vaccine by adequately counseling caregivers of adolescents. Our findings that almost half of caregivers who had not yet decided whether to vaccinate their child would like a doctor or healthcare provider recommendation highlight the opportunity for health care providers to consistently provide strong recommendations for HPV vaccination and allowing time to clear up any misconceptions about the vaccine could improve HPV vaccine coverage. A doctor’s recommendation has been shown to be associated with vaccine initiation even in studies conducted in states with high HPV vaccination coverage such as California^27,28 and North Carolina.²⁹ This would align with recommendations in the literature for providers to initiate the discussion about the HPV vaccine by using clear, concise, and strong messages followed by motivational interviewing techniques to encourage caregivers who show hesitancy toward vaccination.³⁰ Furthermore, providers could be used to disseminate facts as a trusted messenger through other sources of information such as television, radio, and social media.³¹ This in turn could potentially increase HPV vaccine uptake.

Understanding caregiver knowledge and attitudes, in addition to the sources of information they use and how much they trust this information about HPV, is key to improving rates of HPV vaccine initiation and completion. Our work was done in collaboration with the HPV Cancer Prevention Program at St. Jude Children’s Hospital in Memphis, Tennessee, which is working regionally to increase HPV vaccination and prevent HPV-related cancers. This multi-state engagement positions us strategically to incorporate the findings from our current study into broader interventions. To improve vaccination rates, it is essential to develop effective parental outreach and education programs that address the knowledge and attitudes of caregivers. These programs should be designed to provide accurate information about HPV vaccination, its benefits, and potential side effects. The outreach and education programs should be implemented at multiple levels, including schools, healthcare facilities, and community centers. Our data also allows for examination of differences between state vaccination rates to determine whether interventions need to be tailored by state.

Our study uncovered noteworthy differences between the trust in and use of certain sources as sources of information for HPV vaccination. For example, even though trust in social media, books, newspapers or magazines, government agencies, and religious organizations or leaders was significantly associated with vaccine initiation, few caregivers reported obtaining information about HPV vaccination from these sources. These results offer insights into the complex landscape of information sources and trust mechanisms related to HPV vaccination, showing the need to properly leverage information sources that have the potential to be most persuasive.

Media sources including television networks, newspapers, social media, and online news stories have been shown to lack comprehensive information about the HPV vaccine, often missing key details such as about vaccine safety and efficacy.^32–35 Additionally, studies have reported that written materials regarding the HPV vaccine are often difficult to read and generally requires a reading level above eighth grade.^36–40 Health literacy refers to an individual’s ability to obtain, understand, and effectively use health information and services to make informed decisions about their health. Over one-third of adults in United States have been shown to have limited health literacy skills, with men, non-whites, and people with low socioeconomic status being more likely to have basic or below-basic health literacy.⁴¹ Given the prevalence of poor health literacy in the United States, the difficulty of reading these materials could explain why their use was not associated with an increase in HPV vaccine uptake in this study. Following recommendations by the American Medical Association and National Institutes of Health for patient education materials to be written at or below the sixth grade level and ensuring that information about the HPV vaccine is accurate and thorough could make these sources more impactful for HPV vaccination rates.^42,43

However, these findings also have limitations. Since this study was cross-sectional, we cannot be sure what the directionality of these associations are, or whether additional variables exist affecting both the factors we assessed and HPV vaccine uptake. Data about HPV vaccination was also self-reported, and thus is subject to misclassification. Given the specific Midwestern and Southern geographic regions we surveyed in the United States, these results may not be generalizable to other parts of the country or world. There may also have been response bias, where participants felt inclined to report more socially acceptable attitudes and levels of trust in or use of information sources. Additionally, it is possible that children that recently turned 9 may not have had the opportunity to receive the HPV vaccine yet.

The results of this study offer insight on the association between caregiver attitudes, information sources, and trust in information sources with HPV vaccine initiation among adolescents aged 9–17 years. These findings highlight the need to address caregiver attitudes and use of information sources concerning the HPV vaccine. Interventions to improve HPV vaccination rates that target these factors may be more successful.

Funding Statement

This project was supported by the HPV Research Group as part of the St. Jude Children’s Research Hospital and Washington University in St. Louis Implementation Sciences Collaborative, the American Lebanese and Syrian Associated Charities (ALSAC) of St. Jude Children’s Research Hospital, the National Cancer Institute under Grant [3P30CA091842-19S4], and Siteman Cancer Center.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Author contributions statement

Conception and design: A.A., T.S., A.R.B., H.M.B., L.M.K., V.L.S.T., and M.I.S. Analysis and interpretation of the data: A.A., T.S., S.H., and M.I.S. Drafting of the paper: A.A., T.S., and S.H. Revising it critically for intellectual content: A.A., A.R.B, H.M.B, L.M.K., V.L.S.T., and M.I.S. All authors approved of the final version to be published and agree to be accountable for all aspects of the work.

Data availability statement

Please contact the corresponding author at akila@wustl.edu for access to any of the data associated with this study.