Individual, systemic and state factors associated with provider recommendation of HPV vaccination: Findings from NIS-Teen, 2020

ABSTRACT

The most important determinant of HPV vaccination uptake is healthcare provider recommendation, yet not all eligible patients receive HPV vaccination recommendations. We used data from the 2020 National Immunization Survey-Teen to determine individual, systemic, and state factors associated with provider recommendation of HPV vaccination. A total of 18,534 teens were included, with 81.4% receiving provider recommendations for HPV vaccination. HPV vaccination recommendation was higher among females compared to males (AOR: 1.57; 95% CI: 1.27–1.93), teens who received a well-child exam at 11 or 12 years compared to those who did not (AOR: 2.10; 95% CI: 1.61–2.74), and teens whose mothers had college, some college or at least a high school education compared to those with less than high school education. In addition to individual factors, provider recommendation of HPV vaccination was higher in hospitals (AOR: 1.51; 95% CI: 1.00–2.29) and STD/school/teen clinics (AOR: 2.47; 95% CI: 1.05–5.78) compared to public facilities. However, the odds of provider recommendation were lower when none of the teen’s providers ordered vaccines from the state compared to when all teen providers ordered vaccines from the state (AOR: 0.69; 95% CI: 0.52–0.93). Moreover, the state’s mean prevalence of provider recommendations of HPV vaccination was 7.2% lower (Coefficient: −0.072; 95% CI: −0.107 – −0.036) in states with high religious ideology compared to those with low religious ideology. Interventions to increase provider recommendation of HPV vaccination should take a multiprong and comprehensive approach that addresses religious and systemic barriers to HPV vaccination recommendation.

Introduction

Human papillomavirus (HPV) infection is the most prevalent sexually transmitted infection and is linked with several diseases and cancers.^1,2 The administration of HPV vaccination to eligible persons is an effective public health strategy in preventing the majority of HPV-associated cancers.³ More so, the recommendation of HPV vaccination by healthcare providers is the strongest indicator of HPV vaccination uptake.^4–6 According to the Centers for Diseases Control Advisory Committee on Immunization Practices (CDC ACIP), HPV vaccination should be offered to boys and girls at 11 or 12 years, with the option of recommending HPV vaccination as early as nine years.⁷ To reduce missed opportunities and boost HPV vaccination rates, the CDC ACIP further advises healthcare professionals (HCPs) to recommend HPV immunizations to eligible patients at every clinical encounter.⁸ However, despite these guidelines, providers do not consistently recommend HPV vaccines to eligible patients.^6,9–11 In the United States, the average HPV vaccination recommendation rate among unvaccinated teens was less than 50% in 2018.¹⁰ Moreover, the prevalence of provider recommendations for HPV vaccination vary across different states of the U.S.¹² The socio-ecological model asserts that individual behaviors are shaped by the complex interplay of various factors at both the individual and socio-environmental levels such as interpersonal relationships, organizational dynamics, community influences, and public policy considerations.¹³ These differences in HPV vaccination recommendation practices could be attributed to a combination of individual, state, and systemic factors.

The religious ideology of a state is an essential state-level factor that could impact the recommendation and uptake of HPV vaccination.¹⁴ Religious beliefs shape perceptions, attitudes, and behaviors and could potentially influence HPV vaccination recommendations.^14–16 Additionally, contextual factors such as the religious views of teenagers or their parents about sexual practices affect how healthcare professionals approach discussions about HPV and recommend HPV vaccinations.¹⁷ A study by Bodson et al. showed that religious young adult females were less likely to have heard about HPV and HPV vaccines or to have received HPV vaccination recommendations from their healthcare provider.¹⁸ In another study by Franco and colleagues, state religiosity was found to significantly influence HPV vaccination uptake, with highly religious states in the U.S. having the lowest HPV vaccination completion rates.¹⁹ Given the crucial role of religion in shaping attitudes toward public health interventions, it is important to examine the extent to which the religious ideology of a state influences HPV vaccination recommendations by healthcare providers. This could guide public health decisions and interventions aimed at addressing religious barriers to vaccination recommendations. In addition, this understanding could inform future Social Behavioral Change Communication (SBCC) strategies at the community level and in clinical settings, as SBCC could effectively improve HPV vaccination recommendations and subsequent uptake in religious settings.^20,21

Furthermore, factors at the systemic or organizational level, including the facility type and the ability of healthcare practices to order vaccines from state or local health departments, are critical determinants of provider recommendation of HPV vaccines to patients.^11,22 For instance, vaccination cost has been linked to vaccine recommendation and uptake, with uninsured patients less likely to receive HPV vaccine recommendations.^23,24 To mitigate this issue, the Vaccines for Children (VFC) program, available through state health departments, was created so that uninsured, low-income, and Medicaid-eligible teenagers can get vaccines at no cost.^25,26 However, not all providers are enrolled in the VFC program.²⁷ It remains unclear whether VFC-enrolled providers in practices with established systems for ordering vaccines through the health department experience increased rates of HPV vaccination recommendations. Thus, a comprehensive understanding of the role of systemic or organizational factors on HPV vaccination is pertinent in developing policies and guidelines to help increase HPV vaccination recommendation and, ultimately, uptake.

Multiple studies have indicated a correlation between the sociodemographic characteristics of teenagers or their parents and the recommendation of HPV vaccination by healthcare providers.^12,28,29 For instance, sex, race/ethnicity, family income, and mother’s educational status have been associated with provider recommendation of HPV vaccination.²³ Also, teens who attend routine well-child exams at age 11–12, when the HPV vaccine is recommended, are more likely to receive HPV vaccine recommendation. More so, parent-child gender relationships, specifically among father-son pairs, are associated with higher HPV vaccine hesitancy due to safety concerns.³⁰ Furthermore, factors such as well-child visit attendance at the recommended age (11–12 years) for receiving HPV vaccination and ordering of vaccines from the state by healthcare providers are potentially modifiable through the implementation of proven strategies; thus, providing vital opportunities for increasing HPV vaccination recommendation. While factors such as adolescent age, sex, race/ethnicity, facility type, poverty status, mother’s educational level, and state religious ideology may not be directly modifiable, understanding how they impact HPV vaccination recommendation could guide future policies, guidelines, and tailored interventions to increase recommendation rates. Despite the importance of state- and systemic factors on healthcare providers’ recommendation of HPV vaccination, most studies tend to focus solely on individual-level factors and do not account for these broader-level factors in their analyses. Additionally, current research on religious ideology is only limited to its role in HPV vaccination initiation or completion and not on provider recommendation, an essential modifiable factor that impacts vaccine uptake. This understanding could also help guide resource allocation at the state level toward developing educational programs for healthcare providers in religious competence and educational interventions for adolescents and their parents to address religious barriers to HPV vaccination. Based on the socio-ecological model, we posit that HPV vaccine recommendation by an HCP is influenced by a combination of individual characteristics, and broader factors, including systemic barriers and the level of religiosity within a state. Therefore, this study uses nationally representative data to examine the individual, state, and systemic factors associated with provider recommendation of HPV vaccination.

Methods

Study design, setting, and participants

This cross-sectional study uses data from the 2020 National Immunization Survey-Teens (NIS-Teen). The NIS-Teen uses random digit dialing to select households of adolescents aged 13–17 years in the U.S. The survey administration involves two sequential phases. First, parents or caregivers of teens in the chosen households are contacted via phone to obtain information on their teen’s sociodemographic characteristics and immunization history and to obtain consent to contact their healthcare provider. Second, the teen’s healthcare providers are contacted to verify the vaccination history provided by the teen’s parents or caregivers and to complete a provider immunization history questionnaire. A detailed description of the methodology of the NIS-Teen is available online (cdc.gov/vaccines/imz-managers/nis/downloads/nis-teen-puf20-dug.pdf). We did not require ethical approval for this study since the NIS-Teen data is publicly available and fully anonymized. However, ethical approval for the NIS-Teen was provided by the National Center for Health Statistics Research Ethics Review Board. The study is reported following the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline.

Measures

Dependent variable

Provider recommendation of HPV vaccination

This was assessed using the survey question, “Had or has a doctor or other health care professional ever recommended that teen receive HPV shots? Participants with responses “Yes” or “No” were included in our analysis.

Prevalence of provider recommendation of HPV vaccination in a state

Provider recommendation of HPV vaccination in a state was operationalized as a percentage by calculating the number of teens who reported provider recommendation of HPV vaccination within a state divided by the total number of teens in the state.

Independent variable

Individual factors

The following individual-level factors were assessed based on the NIS-Teen survey: teen’s age, sex (male vs. female), race/ethnicity (Non-Hispanic White, Non-Hispanic Black, Hispanic, and Non-Hispanic Other), family’s poverty status (at or above poverty and income ≤ $75k, at or above poverty and income > $75k, and below poverty), mother’s education status (college graduate, some college, high school only, and less than high school), and teen’s clinic visit for a well-child exam or checkup at 11–12-year (yes vs. no). Also, we assessed the respondent’s relationship to the teen (Mother or female guardian, Father or male guardian, Grandparent, and Other family member/friend). The respondent’s relationship with the teen was used as a surrogate to measure the parent or family member present during a teen’s well-child exam.

Systemic or organizational factors

Systemic or organizational factors assessed (facility type and ordering of vaccines from state) were measured at the individual level using the provider-completed immunization history questionnaire. The type of clinic adolescents attended was measured based on the provider’s self-report of their facility type and was categorized into public facilities, hospital facilities, private facilities, STD/school/teen clinics or other facilities, and mixed. The ordering of vaccines from the state by providers was used as a surrogate in assessing the participation of providers or practices in the VFC program. The VFC program which is administered through the state health department is available to all healthcare providers in the U.S. and aims to address cost-related barriers to HPV vaccination.²⁵ This was measured based on the question, “Does your practice order vaccines from your state or local health department to administer to children?” Some teens had multiple providers. To account for teens with multiple providers, this variable was further categorized into all providers, some providers, and no providers.

State factor

The state-level factor assessed was the religious ideology of the teen’s state. This was defined based on previous studies as the percent of the state that identified as highly religious and was classified as a state with high (≥60%), moderate (50–59%), and low (≤49%) religious ideology.^19,31

Statistical analysis

Survey weights were used to weigh the sampled individuals to be representative of the demographics of the U.S. population. The adjustments included accounting for non-response, non-resolution of telephone numbers, subsampling one age-eligible child per household, and multiple telephone lines in the home. Additionally, post-stratification adjustments were made based on the respondents’ sociodemographic characteristics and missing provider data. All variables included in our model were predetermined a priori based on the literature and relevance to the study objective. Thus, no formal model selection was conducted. The prevalence of provider recommendation of HPV vaccination was estimated using the survey weights. To assess the individual and systemic level factors associated with provider recommendation of HPV vaccination, survey-weighted multivariable logistic regression was conducted. We also assessed individual and systemic level factors associated with provider recommendation of HPV vaccines stratified by states with low, moderate, and high religious ideology. Exploratory analysis including the interaction between ordering vaccines from the state (a proxy for participation in the VFC program) and family income was conducted. In addition to adjusting for individual and systemic-level factors, all multivariable regression models included state religious ideology as a covariate. Furthermore, to determine association at the state level, we developed a separate model at the aggregate or ecological level. The prevalence of provider recommendations in a teen state was estimated for all teenagers. A regression model was used to determine the association between state religious ideology and the prevalence of provider recommendations at the state. The statistical analysis was performed using R version 4.2.3, and a p-value <.05 was considered statistically significant for all tests.

Results

A total of 18,534 teens were included in our final analyses with a mean (standard deviation) age of 15 (1.41) years. In our overall sample, 81.4% (80.2%-82.5%) reported receiving provider recommendations for HPV vaccination. Also, 49.2% were females, 52.2% were non-Hispanic Whites, 50.4% were above poverty with income greater than $75k, 44.1% had mothers with college graduate level education, 86.3% had a well-child visit at 11–12 years, and 68.5% had their mother or female guardian as the survey respondent. 77.9% of teens had only one healthcare provider, with 74.3% having all their healthcare providers order vaccines from the state. 55.4% of providers practiced in private facilities, 12.8% in public facilities, and 2.6% in STD/school/teen clinics. Also, 34.3% of teens were from high-religious states, and 31.5% were from low-religious states (Table 1).

Table 1.

Descriptive statistics in the overall study population – national immunization survey-teen, 2020.

Variable name	Categories	n (wt_n)	Prevalence (wt% [95% CI])
Individual factors
Doctor’s recommendation	No	3115 (3484851)	18.6 [17.5–19.8]
	Yes	15419 (15261169)	81.4 [80.2–82.5]
Age, mean (SD)		18534 (18746020)	15.0 [1.41]
Sex	Male	9636 (9526606)	50.8 [49.4–52.2]
	Female	8898 (9219413)	49.2 [47.8–50.6]
Race/ethnicity	Non-Hispanic White	11846 (9789528)	52.2 [50.8–53.6]
	Hispanic	2982 (4501498)	24.0 [22.6–25.5]
	Non-Hispanic Black	1527 (2470642)	13.2 [12.2–14.2]
	Non-Hispanic Other	2179 (1984351)	10.6 [9.8–11.4]
Poverty status	Above poverty > $75k	10402 (8891364)	50.4 [49.0–51.8]
	Above poverty ≤ $75k	5104 (5593308)	31.7 [30.4–33.1]
	Below poverty	2437 (3159604)	17.9 [16.7–19.1]
Mother’s education status	Less than high school	1349 (1995848)	10.6 [9.6–11.8]
	High school only	2683 (3778076)	20.2 [19.0–21.4]
	Some college	4880 (4712805)	25.1 [24.0–26.4]
	College graduate	9622 (8259290)	44.1 [42.7–45.4]
Well child exam at 11 or 12 years	No	1527 (1476262)	13.7 [12.5–14.9]
	Yes	9366 (9322785)	86.3 [85.1–87.5]
Relationship of respondent to teen	Mother or female guardian	13717 (12840620)	68.5 [67.1–69.9]
	Father or male guardian	4137 (4757594)	25.4 [24.1–26.7]
	Grandparent	403 (684868)	3.7 [3.1–4.3]
	Other family member/friend	274 (456641)	2.4 [2.0–3.0]
Number of providers	1	14129 (14610364)	77.9 [76.8–79.0]
	2	3631 (3534244)	18.9 [17.8–19.9]
	≥3	774 (601412)	3.2 [2.8–3.7]
Systemic/organizational factors
Facility type	Public facilities	2196 (2092979)	12.8 [11.8–13.9]
	Hospital facilities	2554 (2031367)	12.4 [11.6–13.3]
	Private facilities	8435 (9033337)	55.4 [54.0–56.8]
	Std/school/teen clinics or other facilities	445 (421572)	2.6 [2.2–3.1]
	Mixed	3144 (2738817)	16.8 [15.8–17.8]
Vaccines ordered from state	All providers	12359 (11675063)	74.3 [73.0–75.5]
	Some providers	2057 (1985414)	12.6 [11.7–13.6]
	No providers	1826 (2058012)	13.1 [12.1–14.1]
State factor
State religious ideology	Low religious ideology	5707 (5899469)	31.5 [30.5–32.6]
	Moderate religious ideology	6984 (6396298)	34.2 [33.3–35.1]
	High religious ideology	5477 (6426644)	34.3 [33.4–35.2]

SD = Standard Deviation; wt_n = Weighted number; wt% = Weighted percentage; CI = Confidence Interval.

In our stratified analyses (Table 2), the prevalence of provider recommendation of HPV vaccination was higher among females (84.7%) compared to males (78.2%), and higher among teens who received a well-child visit at 11–12 years (84.5%) compared to those who did not (71.7%). Also, the prevalence of provider recommendation of HPV vaccination was highest among teens living above poverty with income greater than $75k (86.3%) compared to those above poverty with income less than or equal to $75k (78.6%) and those living below poverty (74.7%) (Table 2).

Table 2.

Descriptive statistics of individual, systemic, and state-level factors stratified by provider recommendation – national immunization survey-teen, 2020.

Variable name	Categories	Provider Recommendation of HPV Vaccination
		Yes (N = 15419) (wt% [95% CI])	No (N = 3115) (wt% [95% CI])	p-value
Individual factors
Age, mean (SD)		15.0 [1.40]	14.7 [1.40]	<.001
Sex	Male	78.2 [76.5–79.9]	21.8 [20.1–23.5]	<.001
	Female	84.7 [83.1–86.2]	15.3 [13.8–16.9]
Race/ethnicity	Non-Hispanic White	84.0 [82.8–85.1]	16.0 [14.9–17.2]	<.001
	Hispanic	76.3 [72.7–79.4]	23.7 [20.6–27.3]
	Non-Hispanic Black	79.0 [75.7–82.0]	21.0 [18.0–24.3]
	Non-Hispanic Other	83.4 [79.9–86.3]	16.6 [13.7–20.1]
Poverty status	Above poverty > $75k	86.3 [85.0–87.6]	13.7 [12.4–15.0]	<.001
	Above poverty ≤ $75k	78.6 [76.2–80.7]	21.4 [19.3–23.8]
	Below poverty	74.7 [71.3–77.8]	25.3 [22.2–28.7]
Mother’s education status	Less than high school	70.1 [64.8–74.9]	29.9 [25.1–35.2]	<.001
	High school only	75.2 [72.0–78.1]	24.8 [21.9–28.0]
	Some college	81.6 [79.5–83.6]	18.4 [16.4–20.5]
	College graduate	86.8 [85.5–88.1]	13.2 [11.9–14.5]
Well child exam at 11 or 12 years	No	71.7 [67.5–75.6]	28.3 [24.4–32.5]	<.001
	Yes	84.5 [83.0–86.0]	15.5 [14.0–17.0]
Relationship of respondent to teen	Mother or female guardian	83.5 [82.2–84.7]	16.5 [15.3–17.8]	<.001
	Father or male guardian	77.6 [75.0–80.0]	22.4 [20.0–25.0]
	Grandparent	76.9 [68.0–83.9]	23.1 [16.1–32.0]
	Other family member/friend	69.5 [59.3–78.1]	30.5 [21.9–40.7]
Number of providers	1	81.4 [80.1–82.6]	18.6 [17.4–19.9]	.069
	2	80.4 [77.4–83.1]	19.6 [16.9–22.6]
	≥3	88.1 [83.2–91.7]	11.9 [8.3–16.8]
Systemic/organizational factors
Facility type	Public facilities	70.9 [67.0–74.5]	29.1 [25.5–33.0]	<.001
	Hospital facilities	86.3 [84.0–88.4]	13.7 [11.6–16.0]
	Private facilities	83.5 [81.9–85.0]	16.5 [15.0–18.1]
	Std/school/teen clinics or other facilities	70.9 [67.0–74.5]	29.1 [25.5–33.0]
	Mixed	86.3 [84.0–88.4]	13.7 [11.6–16.0]
Vaccines ordered from state	All providers	81.7 [80.3–83.0]	18.3 [17.0–19.7]	.193
	Some providers	83.8 [80.1–86.9]	16.2 [13.1–19.9]
	No providers	79.4 [75.7–82.6]	20.6 [17.4–24.3]
State factor
State religious ideology	Low Religious ideology	84.0 [81.3–86.3]	16.0 [13.7–18.7]	<.001
	Moderate religious ideology	83.2 [81.5–84.7]	16.8 [15.3–18.5]
	High religious ideology	77.3 [75.4–79.1]	22.7 [20.9–24.6]

SD = Standard Deviation; wt_n = Weighted number; wt% = Weighted percentage; CI = Confidence Interval.

Furthermore, the prevalence of HPV vaccination recommendation was highest among HCPs who practice in hospitals (86.3%) or mixed facilities (86.3%), compared to those in private facilities (83.5%), public facilities (70.9%) and in STD/school/teen clinics (70.9%). Moreover, the prevalence of provider recommendation of HPV vaccination was highest in states with low religious ideology (84.0%) compared to states with moderate religious ideology (83.2%) and states with high religious ideology (77.3%) (Table 2).

The results of our survey-weighted multivariable regression analysis (Table 3) revealed that the odds of receiving a provider recommendation of HPV vaccination increased by 19% (Adjusted odds ratio (AOR): 1.19; 95% CI: 1.10–1.28) for each year increase in teen’s age. Compared to males, female teens had 57% higher odds (AOR: 1.57; 95% CI: 1.27–1.93) of provider recommendation of HPV vaccination. The odds of provider recommendation of HPV vaccination were 44% lower (AOR:0.56; 95% CI: 0.40–0.78) among teens living below poverty compared to those living above poverty with income greater than $75k. In addition, compared to teens whose mothers had less than a high school education, the odds of provider recommendation of HPV vaccination was 2.3 times higher (AOR: 2.26; 95% CI: 1.52–3.37) among teens whose mothers were college graduates, 66% higher (AOR: 1.66; 95% CI: 1.12–2.47) among teens whose mothers had some college education, and 64% higher (AOR: 1.64; 95% CI: 1.10–2.46) among teens whose mothers had only high school education. Teens who received a well-child exam at 11 or 12 years had over two times higher odds (AOR: 2.10; 95% CI: 1.61–2.74) of receiving HPV vaccination recommendations from their provider. The odds of HPV vaccination recommendation were 47% lower (AOR:0.53; 95% CI: 0.42–0.67) when the survey respondent was the father or male guardian compared to when respondents were teens mother or a female guardian.

Table 3.

Association between individual and systemic-level factors with provider recommendation of HPV vaccination – national immunization survey-teen, 2020.

Variable name	Categories	AOR	95% CI	p-value
Individual factors
Age		1.19	1.10–1.28	<.001
Sex	Male	Ref	Ref	Ref
	Female	1.57	1.27–1.93	<.001
Race/ethnicity	Non-Hispanic White	Ref	Ref	Ref
	Hispanic	0.85	0.65–1.12	.250
	Non-Hispanic Black	0.77	0.54–1.10	.152
	Non-Hispanic Other	1.00	0.73–1.37	.999
Poverty status	Above poverty > $75k	Ref	Ref	Ref
	Above poverty ≤ $75k	0.78	0.60–1.01	.062
	Below poverty	0.56	0.40–0.78	.001
Mother’s education status	Less than high school	Ref	Ref	Ref
	High school only	1.64	1.10–2.46	.016
	Some college	1.66	1.12–2.47	.012
	College graduate	2.26	1.52–3.37	<.001
Well child exam at 11 or 12 years	No	Ref	Ref	Ref
	Yes	2.10	1.61–2.74	<.001
Relationship of respondent to teen	Mother or female guardian	Ref	Ref	Ref
	Father or male guardian	0.53	0.42–0.67	<.001
	Grandparent	0.81	0.38–1.73	.580
	Other family member/friend	0.68	0.35–1.33	.262
Systemic/organizational factors
Facility type	Public facilities	Ref	Ref	Ref
	Hospital facilities	1.51	1.00–2.29	.049
	Private facilities	1.18	0.85–1.64	.331
	Std/school/teen clinics or other facilities	2.47	1.05–5.78	.037
	Mixed	1.39	0.93–2.08	.107
Vaccines ordered from state	All providers	Ref	Ref	Ref
	Some providers	0.88	0.59–1.31	.524
	No providers	0.69	0.52–0.93	.014

In addition to adjusting for individual and systemic-level factors, the multivariable regression model included state religious ideology as a covariate.

AOR = Adjusted Odds Ratio; CI = Confidence Interval; Ref = Reference.

In addition, the odds of provider recommendation of HPV vaccination were 31% (AOR:0.69; 95% CI: 0.52–0.93) lower when none of the teen’s providers ordered vaccines from the state compared to when all teen providers ordered vaccines from the state. Also, compared to providers in public facilities, those in hospital facilities and std/school/teen clinics had 18% (AOR: 1.51; 95% CI: 1.00–2.29) and 147% (AOR: 2.47; 95% CI: 1.05–5.78) higher odds respectively, of recommending HPV vaccination.

Additional analyses of the association between individual and systemic level factors with provider recommendation of HPV vaccines stratified by states with low, moderate, and high religious ideology are shown in Supplemental Tables S1–3. Also, interaction between ordering vaccines from the state and family income was statistically not significant.

Furthermore, as shown in Table 4, the state’s mean prevalence of provider recommendations of HPV vaccination decreased by 7.2% (Coefficient: −0.072; 95% CI: −0.107 – −0.036) in states with high religious ideology compared to those with low religious ideology.

Table 4.

Association between state religious ideology and the prevalence of provider recommendation of HPV vaccination in a state – national immunization survey-teen, 2020.

Variable name	Categories	Coefficient	95% CI	p-value
State religious ideology	Low religious ideology	Ref	Ref	Ref
	Moderate religious ideology	−0.017	−0.051–0.017	.323
	High religious ideology	−0.072	−0.107 – −00.036	<.001

AOR = Adjusted Odds Ratio; CI = Confidence Interval; Ref = Reference.

Discussion

Consistent with previous reports, our study found that about four out of every five teenagers in the U.S. receive recommendations for HPV vaccination from their HCPs.^12,32 Since the CDC ACIP guidelines for HCPs to recommend HPV vaccines to eligible patients at every clinical encounter, provider recommendation rates have been on the rise.^8,10 However, caution is needed when interpreting this finding as recommendation rates reported in our study is among a population of both vaccinated and unvaccinated adolescents. A previous study found that recommendation rates are low among unvaccinated adolescents.¹⁰ More so, Sonawane et al. reported that even though the HPV vaccination recommendation rates nearly doubled over a six-year period, they were still below 50% among unvaccinated adolescents in 2018.¹⁰ While more HCPs may be recommending HPV vaccination to their patients, there is still a vital opportunity to enhance HPV vaccination uptake by boosting recommendation rates among those who have not yet been vaccinated. Various provider-based interventions have successfully improved providers’ self-efficacy in counseling HPV vaccine-hesitant patients and recommending HPV vaccines. For instance, in Texas, where almost 60% of HCPs reported no formal training in HPV vaccination promotion and counseling, such training was associated with enhanced provider self-efficacy in counseling hesitant patients and higher recommendation rates.^11,33 Also, the use of reminders or prompts are practical approaches in minimizing missed opportunities in recommending HPV vaccination to eligible patients; thus, their routine use by HCPs should be promoted to increase HPV vaccination completion.^34,35 In addition, healthcare practices should incorporate regular communication training for healthcare providers to increase their self-efficacy in making strong HPV vaccination recommendations to patients.³³

In addition, our study points to the critical role of religious ideology in HPV vaccination recommendations. We observed that states with high religious ideology had lower recommendation rates than states with low religious ideology. This is the first study to our knowledge to indicate an association between religious ideology and HPV vaccination recommendation, but it is not surprising given the permeating effect of religion in shaping societal norms and behaviors. Expectedly, healthcare providers have a professional obligation to provide evidence-based medical interventions, including vaccinations, regardless of the religious inclinations of the state in which they practice.^36,37 However, HCPs, as members of society, are not immune from the influence of religion and occasionally see their beliefs in conflict with the recommendations of medical interventions.³⁸ In a national study of healthcare providers in the U.S., a substantial number of providers reported that they did not believe they had an obligation to recommend an intervention they considered objectionable for religious or moral reasons.¹⁵ Addressing the potential tension between religion and science calls for a multiprong approach. On the one hand, healthcare providers would benefit from continued medical education in cultural and religious competence to enhance their skills in dealing with patients’ cultural and religious needs.³⁹ Concurrently, religious leaders and organizations as important stakeholders could play critical roles in shifting societal beliefs and values and should be involved in communicating the benefits of HPV vaccination as a safe and effective strategy for cancer prevention. Therefore, it is necessary to explore interventions that foster the understanding of religious leaders about the painstaking process involved in HPV vaccine research, development, and monitoring and encourage them to become state advocates of its safety and efficacy.

In the aftermath of the COVID-19 pandemic, medical-religious partnerships with faith-based organizations (FBOs) proved to be an effective strategy in addressing community mistrust and facilitating the distribution of COVID-19 vaccines in partnership with state entities.⁴⁰ Consequently, FBOs have the potential to assume crucial roles by leveraging their network within religious communities to address religious barriers and contribute to the development of tools to aid providers in delivering strong HPV vaccination recommendations.⁴¹ Faith-based interventions have been shown to be effective in improving community health behavior, such as HBV testing and vaccination, intent to undergo yearly mammograms, and reducing HIV stigma and mistrust.^42–44 Moreover, findings from a previous study indicate that SBCC strategies, including faith-based mobilization, community engagement, and social change communication, are vital for integrating religion and health.²⁰ When strategically deployed, the SBCC approach has been noted to help stimulate community dialogue, address mistrust in the healthcare system, reduce HPV vaccine-related stigma, and promote HPV vaccination.^20,21 Therefore, in states with high religious ideology, resources should be deployed to implement SBCC interventions, educate religious leaders as community HPV vaccination change agents, and enhance healthcare providers’ religious and cultural competence in clinical settings.

Furthermore, this study supports the goal of the VFC program – a federally funded program that became operational in the U.S. in 1994.²⁵ The VFC program ensures uninsured teens and individuals from low socioeconomic backgrounds can access HPV vaccines at no cost.²⁵ In the U.S., practices can stock up on vaccines through the VFC program at no cost or by private purchase.^25,45 Participating providers in the VFC program receive fees solely for administering the vaccines, which are reimbursed mostly through insurance payments. In contrast, providers who procure vaccines privately recover costs by billing insurance for both the cost of the vaccine and its administration. Enrolling in the VFC program requires HCPs to undergo training and establish a system to handle, store, and administer all VFC vaccines. Our finding demonstrates that teenagers whose providers participate in the VFC program and order vaccines from the state health departments are more likely to receive HPV vaccine recommendations. However, while all states in the U.S. participate in the VFC program, not all physicians or medical practices are registered as VFC providers due to dissatisfaction with reimbursement for vaccine administration from Medicaid and CHIP, challenges with record keeping, and the complexity of maintaining separate vaccine stocks.^45–47 Consequently, there are disparities in patient access to this vital initiative.²⁷ Also, at present, only 34 states in the U.S. explicitly permit pharmacists to participate in the VFC program.⁴⁸ These hints at the need to broaden the enrollment of HCPs, including pharmacists, in the VFC program to expand patients’ access to the HPV vaccination. Addressing systemic hurdles experienced by practices in stocking HPV vaccine and receiving reimbursement would boost enrollment in the VFC program, HPV vaccine recommendations, and, ultimately, uptake. Medicaid should incentivize providers by improving reimbursement rates and the timeline for reimbursing providers. Moreover, the vaccine record system for VFC-participating providers should be simplified. In addition, we found that teens whose providers practice in sexually transmitted (STD) or school-based clinics were over twice as likely to recommend HPV vaccines compared to those in public facilities. The prioritization of immunization and sexually transmitted disease preventive services at such facilities (STD clinics and school-based health centers) may partly explain the high HPV vaccination recommendation rate seen., This finding is also pertinent given that most U.S. states currently do not require or mandate HPV vaccination for school enrollment, a policy that has negatively influenced providers’ recommendation of HPV vaccination.^49,50 Worthy of note is that Rhode Island, where HPV vaccination is required before seventh grade, has the highest HPV vaccination rates in the U.S.⁵¹ HPV vaccination interventions would need to be developed to mitigate the particular challenges in states with and without HPV vaccine mandates. In such jurisdictions where HPV vaccination is mandated, interventions should address barriers to accessing care and increase routine well-visits at ages 11–12 to enhance compliance rates. On the other hand, in states without HPV vaccination mandates, in addition to addressing healthcare access barriers, HCPs should receive continuous medical education and be equipped with tools such as fact sheets and scripts to provide strong recommendations and address the concerns of HPV vaccine-hesitant patients and parents. Furthermore, our study revealed that more adolescents receive care from private facilities (55%) compared to hospital-based settings (12%). Moreover, recommendation rates in private facilities are similar to those in hospital settings, which were found to have the highest recommendation rate. Therefore, strategies that increase recommendation rates at private facilities will reach a large proportion of adolescents and significantly boost HPV vaccination uptake. Overall, our findings suggest that systemic level factors significantly influence HPV vaccination recommendation and should be factored into interventions to increase provider recommendations of HPV vaccination.

Prior studies have examined individual-level factors associated with provider recommendations of HPV vaccines with inconsistent results across studies.²³ Most of these studies fall short in accounting for state-level and systemic-level factors in their analysis. After controlling for state and system-level factors, we found that older teenagers, females, those whose mothers have at least high school, and those who received 11–12-year-old well-child exams or checkups were more likely to report HPV vaccination recommendations from their providers. It is worth noting that most adolescent immunizations are given during well-child visits, and studies indicate that adolescents who receive such preventive visits are likely to initiate HPV vaccination.^52,53 Healthcare providers should utilize proven strategies such as the combination of phone calls, SMS, e-mail, or patient portal message reminders ahead of clinic appointments to minimize no-shows at routine well-child checks and increase HPV vaccination recommendation and uptake.^22,54 Additionally, our study revealed that teens from families living below the poverty or those with fathers or male guardians as survey respondents had lower odds of receiving HPV vaccination recommendations from their providers. This finding resonates with a previous study that found higher rates of HPV vaccine hesitancy among male teens whose fathers were survey respondents.³⁰ However, while survey respondents may reflect the parent or family member usually present during a teen’s well-child checkup, this finding should be interpreted cautiously. Survey respondents may not always be family members in a consulting setting; therefore, they may not influence vaccination recommendations. Further investigations are needed to determine the role of respondents’ gender-based relationship with teens on vaccination decisions, as this may inform future educational interventions. Moreover, even after years of rolling out guidelines for eligible boys and girls to receive HPV vaccination recommendations from their healthcare providers, our results indicate that rates of provider recommendations of HPV vaccination for male adolescents still lag behind females. These findings confirm existing disparities in provider recommendation of HPV vaccination and highlight the need for tailored interventions to ensure universal access to care, enhance providers’ self-efficacy in HPV vaccination recommendation, and ultimately, increased initiation and completion of the HPV vaccination series. Embedded in the socio-ecological model, our findings suggest that interventions to increase provider recommendations can be implemented across multiple levels – individual, systemic, and state levels. At the individual level, providers can receive training and tools to improve their self-efficacy in counseling vaccine-hesitant parents. Simultaneously, changes should be made at the systemic level to reduce vaccine costs and improve the overall vaccination process. Furthermore, health departments can collaborate with religious organizations to develop communication tools to assist HCPs in addressing religious concerns regarding the vaccine among vaccine-hesitant parents.

Our study had a few limitations. This was a cross-sectional study limiting our ability to infer causality. Given the cross-sectional nature of our study design, there may be reverse causality in the association between ordering vaccines from the state and recommending HPV vaccines. Future longitudinal studies are warranted to address these limitations. Also, our study may be prone to recall bias since teens or their parents had to recollect receiving a provider recommendation over time. In addition, though less likely, there may be potential social desirability bias if teenagers or their parents report receiving a provider recommendation based on the overall societal perception of vaccination. Although religion has a far-reaching influence on individual beliefs, the assessment of religiosity as operationalized in our study was at the state level. As such, caution is required when interpreting the association between a state’s religious ideology and HPV vaccination recommendations to avoid making conclusions at the individual level. Also, our study did not account for potential residual confounding that may arise from the heterogeneity in religious beliefs within religious faith as well as within and across states. This would be pertinent for future research on religiosity and HPV vaccination recommendations. Despite these limitations, our study is the first to concurrently assess individual, systemic, and state-level factors associated with HPV vaccination recommendation. Also, this study uses a nationally representative study, increasing the generalizability of our findings.

In conclusion, we found that HPV vaccination recommendations were lower in states with high religious ideology; however, provider recommendation was higher among adolescents whose providers order vaccines from the state health departments. Also, individual-level factors such as older age, being female, higher educational level of the teen’s mother, and receipt of well-child exams increased provider recommendation, while living below poverty and having a father or male guardian as respondent decreased provider recommendation of HPV vaccination. Increasing the HPV vaccination rate would require a multiprong and tailored approach that targets broader factors together with individual-level determinants of provider recommendation of HPV vaccination.

Funding Statement

The study was funded by the National Cancer Institute (P30CA016672 to S. Shete), the Betty B. Marcus Chair in Cancer Prevention (to S. Shete), the Duncan Family Institute for Cancer Prevention and Risk Assessment (S. Shete), and the Cancer Prevention Research Institute of Texas (grant RP170259 to S. Shete). The funders were not involved in the study design, analysis, interpretation of data, or manuscript writing.

Disclosure statement

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

Author contributions

Conceptualization: I.O., M.C., and S.S.; Data curation: R.T. and S.S.; Formal Analysis: R.T.; Methodology: I.O., M.C., R.T., and S.S.; Writing—original draft: I.O. and M.C.; Writing—review and editing: R.T. and S.S.; Funding acquisition: S.S.; Supervision: S.S. All authors have read and agreed to the published version of the manuscript.

Supplementary data

Supplemental data for this article can be accessed on the publisher’s website at https://doi.org/10.1080/21645515.2023.2239678.