Missed Vaccination Opportunities Among U.S. Adolescents by Area Characteristics

Abstract

Introduction:

Three vaccines are recommended for U.S. adolescents: tetanus, diphtheria, and acellular pertussis (Tdap); meningococcal conjugate; and human papillomavirus. To understand disparities in vaccine availability and hesitancy, adolescent-, household-, and area-level characteristics associated with patterns of vaccine coverage are described.

Methods:

In 2020–2021, the authors generated national estimates among 8 possible combinations of vaccine coverage and identified associated characteristics using 2015–2017 National Immunization Survey–Teen for male and female adolescents aged 13–17 years (n=63,299) linked to area (ZIP Code) characteristics. Next, factors associated with missed opportunity for human papillomavirus vaccine (i.e., receipt of Tdap and meningococcal conjugate only compared with coverage of all 3 vaccines) were identified using logistic regression.

Results:

Most U.S. adolescents received all 3 vaccines (42.9%) or Tdap and meningococcal conjugate only (32.1%); fewer received no vaccines (7.7%) or Tdap only (6.6%); and the remainder received some combination of 1–2 vaccines. Missed opportunities for human papillomavirus vaccination were more likely among adolescents who were male, White race, uninsured, in middle-income households, and living in rural areas; and less likely among adolescents who were older, Medicaid-insured, whose parents completed surveys in Spanish, in poverty-level households, and living in high-poverty areas.

Conclusions:

A substantial number of U.S. adolescents are not fully vaccinated and coverage varies by vaccine type, population, and place. Providers should routinely stock all 3 vaccines and promote simultaneous, same-day vaccination to avoid missed vaccine opportunities. More research and interventions are needed to understand and modify patient, provider, payer, vaccine supply/storage, or other reasons for suboptimal coverage of all recommended vaccines.

INTRODUCTION

Three adolescent vaccines are routinely recommended in the U.S. The Advisory Committee on Immunization Practices (ACIP) recommends tetanus, diphtheria, and acellular pertussis (Tdap); meningococcal conjugate (MenACWY); and human papillomavirus (HPV) vaccines at ages 11–12 years.^1,2 Of the 3 vaccines, HPV coverage lags. In 2019, among adolescents aged 13–17 years, 1-dose coverage reached 90.2% for Tdap and 88.9% for MenACWY, but only 71.5% for HPV. Up-to-date coverage for MenACWY and HPV is even lower.³

Though many studies examine coverage of specific vaccines in isolation (e.g., missed HPV vaccination)^4–6 or simultaneous vaccination with ≥2 vaccines,^7–9 few studies examine all possible patterns of vaccine coverage to understand missed opportunities. Less is known about how many adolescents receive only Tdap, or HPV and Tdap, but not MenACWY vaccine, for example.¹⁰ Identifying characteristics associated with all patterns of vaccination by type of vaccine(s) may help inform whether efforts to increase complete vaccine coverage should focus more on vaccine availability and accessibility (policies and programs designed to ensure equitable access) or hesitancy (parent refusal or delayed acceptance). Further, despite evidence that multilevel characteristics such as rural–urban residence or area poverty are associated with vaccination,^11–13 to the authors’ knowledge, no nationally representative studies have examined how area-level characteristics relate to coverage of the 3 adolescent vaccines in isolation and combination.

Examining area-level characteristics is valuable to identify how factors beyond the individual are associated with vaccine coverage and is critical to identify the types of geographic areas needing intervention. Identifying area-level characteristics associated with specific vaccine coverage patterns is especially important because some places may have different vaccine gaps based on vaccine type. For example, some areas might benefit most from interventions designed to decrease vaccine hesitancy in general or specifically for HPV. Other places might benefit from targeted interventions to increase on-site availability of all vaccine types or specific vaccine types. This study describes adolescent-, household-, and area-level characteristics associated with coverage patterns among all 3 vaccines in the U.S.

METHODS

Study Sample

To quantify vaccination among adolescents aged 13–17 years, healthcare provider–verified vaccine records from the 2015–2017 National Immunization Survey–Teen (NIS-Teen) were linked to 2013–2017 American Community Survey 5-year estimates at the ZIP Code Tabulation Area level.¹⁴ A few records were excluded (1.44%, n=914) because of missing ZIP Codes. NIS-Teen is an annual, nationally representative survey of parents or guardians of eligible adolescents from all U.S. states and territories, identified using random-digit-dial samples of landline and cell phone numbers.¹⁵ Because ZIP Codes of participants’ residences are restricted variables, data were accessed at the Federal Research Data Center in Philadelphia.

Measures

First, 8 possible combinations of the 3 adolescent vaccines were defined: no vaccines (1), 1 vaccine only (3), any combination of 2 vaccines (3), or complete vaccination with all 3 vaccines (1). Second, adolescents with HPV missed opportunities were defined as those vaccinated with only Tdap and MenACWY but not HPV vaccine. For both variables, adolescents who received 1 dose of Tdap, ≥1 doses of MenACWY, or ≥2 doses of HPV vaccine were defined as vaccinated. Because the MenACWY booster dose is recommended at age 16 years, defining coverage by receipt of 2 doses would have truncated analytic sample to 17-year-olds.¹ Provider-verified data, and not self-report, were used to ensure accuracy and prevent recall bias; 51.12% of the records were excluded for this reason (n=66,212).

Adolescent and household characteristics included: adolescent age, race/ethnicity, sex, insurance status, household poverty/income threshold, U.S. region, language of survey, and survey year. At the ZIP Code level, area-level characteristics included: urban/rural status measured using the Rural–Urban Commuting Area classification¹⁶ and percentage of residents living in poverty. Covariate categorizations are shown in Table 1.

Table 1.

Adolescent, Household, and Area Characteristics of Adolescents Aged 13‒17 Years, National Immunization Survey–Teen, 2015‒2017

Characteristics	All eligible survey respondents N=63,299		Subset of respondents eligible for HPV vaccine missed opportunity analysisa N=47,162
	N (survey respondents)	Weighted %	N (survey respondents)	Weighted %
Year
2015	21,875	33.32	15,866	32.18
2016	20,475	33.33	15,253	33.48
2017	20,949	33.34	16,043	34.34
Age, years
13	12,968	19.75	10,129	19.89
14	13,252	19.88	10,196	20.02
15	12,770	21	9,445	20.05
16	12,811	20.51	9,338	19.56
17	11,498	18.86	8,054	17.09
Race/ethnicity
Non-Hispanic White only	38,729	52.87	28,785	52.36
Hispanic	11,718	52.87	8,959	23.44
Non-Hispanic Black only	5,957	13.86	4,376	13.63
Non-Hispanic other + multiple race	6,895	10.07	5,042	9.75
Sex
Male	33,285	51.06	24,742	50.65
Female	30,014	48.94	22,420	49.35
Insurance
Private insurance	37,905	53.81	29,004	55.34
Any Medicaid	18,838	35.68	13,798	35.11
Missing	254	0.35	156	0.33
Other insurance	3,718	5.63	2,670	5.53
Uninsured	2,584	4.54	1,534	3.69
Region
Northeast	12,073	16.48	10,128	18.58
Midwest	13,464	21.45	9,649	20.95
South	24,007	38.23	18,033	37.32
West	13,755	23.84	9,352	23.14
Area level poverty
Very low 1 (<5%)	8,631	13.06	7,029	13.52
2 (5%‒9.9%)	16,738	25.87	12,691	25.3
3 (10%‒19.9%)	22,475	25.48	16,124	27.24
Very high 4 (≥20%)	14,522	35.59	10,685	33.94
Language in which interview was conducted
English	57,169	87.28	42,691	87.69
Spanish	5,031	10.45	3,852	10.71
Other	1,099	2.27	619	1.6
Rural urban commuting area
Urban	52,052	89.54	40,065	91.82
Rural	10,328	10.46	6,476	8.18
Household poverty
Above poverty >$75,000	29,442	38.98	23,033	40.98
Above poverty ≤$75,000	20,003	33.38	14,210	32.45
Below poverty	11,584	21.48	8,307	20.66
Unknown	2,270	6.16	1,612	5.91

^aIncludes adolescents who received all 3 vaccines (complete vaccination; tetanus, diphtheria, and acellular pertussis [Tdap] vaccine, meningococcal conjugate [MenACWY], and human papillomavirus [HPV]); and adolescents with a missed opportunity for HPV vaccine (received 2 vaccines [Tdap and MenACWY] and did not receive HPV vaccine).

Statistical Analysis

NIS-Teen data for 2015–2017 were combined, and sampling weights for the provider-verified only data were recalculated¹⁵ to generate nationally representative estimates. This approach reduced bias related to nonresponse, noncoverage, and overlapping samples of mixed telephone users. Analysis occurred in 2020–2021.

Patterns of vaccine coverage across all 8 possible combinations were described using weighted percentages and associated 95% CIs and were illustrated using a proportional Venn diagram.¹⁰ Weighted percentages were calculated for all adolescents and separately by populations defined by adolescent-, household-, and area-level characteristics. Next, factors associated with an HPV missed opportunity compared with those receiving all 3 vaccines (complete) were identified using unadjusted and multivariable logistic regression with sampling weights. The model included state of residence as a random effect to account for impact of state programs on vaccine accessibility and coverage. The University of Utah IRB deemed this study exempt (IRB#00080885).

RESULTS

Table 1 lists characteristics of the 63,299 adolescents in the 8-category analysis and the 47,162 in the HPV missed opportunity analysis. Figure 1 diagrams the weighted percentages and 95% CIs of vaccine coverage for the U.S. adolescent population. Many adolescents received all 3 vaccines (42.9%) or Tdap and MenACWY only (32.1%); fewer received no vaccines (7.7%) or Tdap only (6.6%); and the remainder received some combination of 1–2 vaccines.

Figure 1.

Vaccine coverage of U.S. adolescents aged 13‒17 years: weighted percentages and 95% CIs (n=63,299).

HPV, human papillomavirus vaccine; Tdap, tetanus, diphtheria, and acellular pertussis vaccine;

MenACWY, meningococcal conjugate vaccine.

Table 2 lists the weighted percentages (and 95% CIs) of adolescents who received vaccines separately for the top 5 populations, measured by prevalence of vaccine receipt, for all combinations of the 3 vaccines; and specifically, for the populations defined using area-level measures. This illustrates the 5 populations with the highest vaccine coverage, separately for every vaccine pattern and vaccine patterns for each dimension of the measured area-level characteristics. The complete ranked order list is provided in Appendix Table 1. Complete vaccination was most common among adolescents with parents responding in Spanish, who lived in the Northeast, and were Hispanic. The following populations were the most likely to have received no vaccines: adolescents who were uninsured or missing insurance information, and those who lived in rural areas. The populations most likely to have a missed HPV opportunity were: adolescents who were aged 13 years (youngest age group), lived in very low-poverty areas, and male. To examine the role of state policy environment, vaccine patterns based on residence in a state with a MenACWY school entry requirement were also described. Thirteen states and the District of Columbia had this requirement before 2015.¹⁷ Appendix Table 2 demonstrates that adolescents living in states with school entry requirements, when compared with those with no such requirement, were slightly more likely to be completely covered and less likely to have no vaccines.

Table 2.

Patterns of Vaccine Coverage Among Adolescents Aged 13‒17 Years, National Immunization Survey–Teen, 2015‒2017 (n=63,299)

Variable	No vaccination	Vaccination patterns reflecting coverage of 1 or 2 vaccines in combination						Complete vaccination (receipt of all 3 vaccines)
		HPV only	Tdap only	MenACWY only	Tdap and MenACWY only (HPV vaccine missed opportunity)	HPV and MenACWY only	HPV and Tdap only
	Weighted percentages (95% CI)
Total	7.7 (7.3, 8.1)	0.7 (0.5, 0.9)	6.6 (6.2, 7)	4.3 (4, 4.6)	32.1 (31.4, 32.7)	3.5 (3.3, 3.8)	2.1 (1.9, 2.3)	42.9 (42.2, 43.6)
Rank order of population groupsa
1 (highest percent)	Uninsured 13.8 (11.6, 15.9)	Other language survey 2.7 (0.1, 5.3)	Rural 11.1 (10, 12.2)	Other language survey 12.1 (8.8, 15.4)	Age 13 years 39.9 (38.3, 41.4)	Other language survey 16.2 (11.8, 20.5)	Other language survey 3.8 (1.6, 6)	Spanish survey 57.3 (54.4, 60.3)
2	Rural 12.4 (11.2, 13.6)	Insurance missing 1.5 (0, 3.6)	Insurance missing 8.5 (2, 14.9)	Uninsured 10.3 (8.1, 12.4)	Area poverty 1 (very low) 37.5 (35.8, 39.3)	Unknown poverty 7 (4.7, 9.4)	NH other/multiple race 2.7 (1.9, 3.5)	Northeast 51.8 (50.4, 53.2)
3	Insurance missing 11.4 (6.2, 16.5)	Age 17 years 1.1 (0.5, 1.7)	>Poverty, ≤$75K 8.1 (7.4, 8.8)	Age 17 years 6.6 (5.7, 7.5)	Male 36.1 (35.1, 37)	Age 17 years 6.4 (5.5, 7.2)	Female 2.6 (2.2, 2.9)	Hispanic 49.8 (47.8, 51.8)
4	West 9.4 (8.1, 10.7)	West 1.1 (0.5, 1.7)	Age 13 years 7.8 (6.8, 8.8)	Unknown poverty 6 (4.4, 7.5)	>Poverty, >$75K 35.9 (34.9, 36.9)	<Poverty 5.7 (5, 6.4)	Area poverty 4 (very high) 2.6 (2, 3.1)	Insurance missing 48 (37.9, 58.2)
5	>Poverty, ≤$75K 8.9 (8.2, 9.6)	Hispanic 1 (0.5, 1.6)	South 7.7 (7.1, 8.3)	<Poverty 5.9 (5.1, 6.6)	Age 14 years 35.9 (34.3, 37.4)	Spanish survey 5.6 (4.4, 6.9)	West 2.5 (1.9, 3.2)	Female 47.7 (46.6, 48.8)
Estimates for area-level variables
Rural	12.4 (11.2, 13.6)	0.8 (0.5, 1.2)	11.1 (10, 12.2)	5.6 (4.8, 6.5)	30 (28.4, 31.6)	3.6 (2.6, 4.5)	2.4 (1.9, 2.9)	34.1 (32.4, 35.7)
Urban	7.2 (6.8, 7.6)	0.7 (0.5, 0.9)	6.1 (5.7, 6.5)	4.2 (3.9, 4.5)	32.3 (31.6, 33.1)	3.5 (3.2, 3.8)	2.1 (1.8, 2.3)	43.9 (43.1, 44.7)
Area poverty 1 (very low)	5.6 (4.7, 6.4)	1 (0.1, 1.8)	4.9 (3.9, 5.8)	3.3 (2.6, 3.9)	37.5 (35.8, 39.3)	2.2 (1.7, 2.6)	1.7 (1.3, 2)	44 (42.2, 45.8)
Area poverty 2	6.8 (6.2, 7.5)	0.7 (0.4, 1)	6.4 (5.7, 7)	4.1 (3.6, 4.6)	34.4 (33.1, 35.6)	2.9 (2.5, 3.4)	1.8 (1.4, 2.2)	42.9 (41.5, 44.3)
Area poverty 3	8.4 (7.8, 9)	0.6 (0.4, 0.8)	7.6 (6.9, 8.3)	5 (4.5, 5.6)	31.6 (30.5, 32.8)	3.4 (2.9, 3.9)	2.1 (1.8, 2.5)	41.2 (39.9, 42.4)
Area poverty 4 (very high)	8.6 (7.6, 9.7)	0.6 (0.4, 0.9)	6.1 (5.4, 6.9)	4.2 (3.6, 4.7)	27.8 (26.4, 29.2)	5 (4.3, 5.7)	2.6 (2, 3.1)	45.1 (43.5, 46.7)

^aAppendix Table 1 provides full rank ordering.

HPV, human papillomavirus vaccine; Tdap, tetanus, diphtheria, and acellular pertussis; MenACWY, meningococcal conjugate; NH, Non-Hispanic.

In the multivariable model (Table 3), the following characteristics were independently associated with increased odds of having a missed HPV opportunity: rural residence, living in the Midwest or South (versus West), and being uninsured or having other insurance (versus private insurance). Characteristics associated with decreased odds of missed opportunity included: female sex, older age, more recent survey year, all racial/ethnic populations (versus White), Medicaid or missing insurance information (versus private insurance), survey completion in Spanish or other language (versus English), and high area-level poverty. Results related to household SES indicated a divergent pattern. Compared with the highest-income households (i.e., ≥$75,000 annual income), adolescents living in above-poverty households with <$75,000 annual income experienced a higher odds of an HPV vaccine missed opportunity whereas those in households below poverty were less likely to have a missed opportunity.

Table 3.

Correlates of Having a Missed Opportunity for Human Papillomavirus (HPV) Vaccine (n=47,162)

	HPV vaccine missed opportunity (Tdap and MenACWY only vs all 3 vaccines)a
Characteristics	Unadjusted OR (95% CI)	AOR (95% CI)
Year
2015	ref	ref
2016	0.90 (0.86, 0.94)	0.90 (0.86, 0.94)
2017	0.76 (0.73, 0.79)	0.75 (0.72, 0.79)
Age, years
13	ref	ref
14	0.81 (0.77, 0.86)	0.80 (0.76, 0.85)
15	0.65 (0.61, 0.69)	0.63 (0.59, 0.67)
16	0.63 (0.59, 0.67)	0.59 (0.55, 0.62)
17	0.53 (0.5, 0.57)	0.49 (0.46, 0.52)
Race/ethnicity
Non-Hispanic White only	ref	ref
Hispanic	0.59 (0.56, 0.62)	0.83 (0.78, 0.88)
Non-Hispanic Black only	0.84 (0.80, 0.89)	0.85 (0.80, 0.90)
Non-Hispanic other + multiple race	0.77 (0.73, 0.83)	0.81 (0.76, 0.87)
Sex
Male	ref	ref
Female	0.62 (0.60, 0.64)	0.60 (0.58, 0.63)
Insurance
Private insurance	ref	ref
Any Medicaid	0.71 (0.68, 0.74)	0.84 (0.80, 0.89)
Missing	0.60 (0.43, 0.85)	0.68 (0.48, 0.96)
Other insurance	1.03 (0.95, 1.12)	1.13 (1.04, 1.23)
Uninsured	1.23 (1.12, 1.36)	1.68 (1.51, 1.87)
Area level poverty
Very low 1	ref	ref
2	0.94 (0.88, 1.00)	0.96 (0.90, 1.02)
3	0.90 (0.85, 0.95)	0.97 (0.91, 1.03)
Very high 4	0.72 (0.68, 0.77)	0.91 (0.84, 0.97)
Language in which interview was conducted
English	ref	ref
Spanish	0.42 (0.39, 0.44)	0.49 (0.45, 0.92)
Other	0.67 (0.58, 0.78)	0.79 (0.67, 0.92)
Rural urban commuting area
Urban	ref	ref
Rural	1.20 (1.12, 1.28)	1.12 (1.05, 1.20)
Household poverty
Above poverty >$75,000	ref	ref
Above poverty ≤$75,000	0.95 (0.91, 0.99)	1.08 (1.02, 1.13)
Below poverty	0.66 (0.63, 0.70)	0.91 (0.85, 0.98)
Unknown	0.75 (0.69, 0.81)	1.00 (0.92, 1.10)
Region
Northeast	1.09 (1.03, 1.16)	0.95 (0.89, 1.01)
Midwest	1.41(1.33, 1.49)	1.21 (1.14, 1.28)
South	1.56 (1.49, 1.64)	1.44 (1.36, 1.51)
West	ref	ref

^aHPV missed opportunity was defined as having received 2 vaccines (tetanus, diphtheria, and acellular pertussis [Tdap] vaccine and the meningococcal conjugate [MenACWY]) compared to the referent group of adolescents who received all 3 vaccines (Tdap, MenACWY, and HPV).

The primary analysis of missed opportunities focused on receipt of <2 HPV doses; in additional analyses, 0 versus 1 HPV dose was also examined. Among those receiving Tdap and MenACWY only (i.e., HPV missed opportunity), most received 0 HPV doses (72.7%, 95% CI=71.5%, 73.8%) and fewer received 1 HPV dose (27.3%, 95% CI=26.2%, 28.5%). To examine whether correlates differed based on missed opportunity definition (i.e., receipt of 0 or 1 dose), additional multivariable models were conducted with the ref group being complete vaccination (Tdap, Men, and ≥2 HPV vaccine doses). Results were generally similar to the primary analysis (data not shown).

DISCUSSION

This study illustrates that there are clear gaps in U.S. adolescent vaccine coverage that differ by vaccine type and for diverse populations defined by adolescent, household, and area characteristics. Less than half (42.9%) of adolescents received all 3 recommended vaccines, a substantial portion (32.1%) had a missed HPV opportunity, many (7.7%) had received no vaccines, and, notably, 1 in 6 (17.3%) had some other type of vaccine pattern.

This study was able to assess how vaccine coverage patterns varied by area-level characteristics by using restricted-use ZIP Code–level NIS-Teen data. Though uncommon in the overall U.S. population, several patterns were more prevalent in particular types of areas. For example, a higher percentage of adolescents in rural areas (12.4%) received no vaccines compared with adolescents in urban areas (7.2%). Rural adolescents were also most likely to have received Tdap only. Multivariable regression demonstrated that residents of rural areas were also more likely to have missed HPV opportunities. This is especially concerning given high rates of HPV-associated cancers in rural areas¹⁸ and the relative scarcity of HPV vaccine interventions in rural communities.¹⁹ Disparities by area-level poverty were also observed. More adolescents in high-poverty areas (8.6%), compared with low-poverty areas (5.6%), were completely unvaccinated. The gradient for area-level poverty was especially striking for Tdap and MenACWY only; missed HPV opportunities were far more common in low-poverty areas. These results build upon previous work identifying area-level disparities in specific vaccine coverage.^3,13,20 Collectively, this growing literature highlights the importance of using place-based data in vaccine surveillance and lends urgency to developing policies and programs that target outreach to specific types of geographic regions.

Several barriers likely contribute to observations of disparities by area-based characteristics. For example, in addition to vaccine hesitancy, provider- and healthcare system–level barriers potentially lead to unequal vaccine availability and accessibility. Vaccines for Children (VFC) Program provides no-cost ACIP-recommended vaccines to low-income, underinsured, and uninsured children and adolescents in the U.S. However, spatial access to VFC providers varies.²¹ In rural areas, hospital closures have increased shortages of primary care providers.²² Further, providers in rural and high-poverty areas may not be consistently stocking and storing all 3 vaccines, or alternatively, may not be offering all 3 vaccines. Financial barriers at the health system and clinic level, including vaccine stocking fees, storage costs, and reimbursement practices, may lead to less vaccine availability. For example, up-front vaccine costs can be a barrier, especially for small clinics such as those in rural areas. The Centers for Disease Control and Prevention price list indicates that single doses of Tdap, MenACWY, and HPV cost $20, $100, and $196, respectively.²³ Rural clinics may keep a low inventory of more expensive vaccines to avoid expired doses and also may carry a larger inventory of Tdap because it is recommended for children and adults of all ages with repeat boosters (i.e., greater demand). Payment reforms, increasing provider enrollment in VFC, and other interventions such as delivery in alternative settings (e.g., school-based health centers, mobile health clinics)^24,25 are needed to ensure greater availability and accessibility of all vaccine types.

Several adolescent and household characteristics were associated with vaccine coverage. A substantial percentage of uninsured adolescents received no vaccines. Adolescents with parents or guardians who completed the survey in Spanish were more likely to be completely vaccinated. This finding may reflect the positive impact of the VFC program among Spanish-speaking households.^26,27 The authors are unaware of literature exploring why vaccination coverage may differ for adolescents with parents who speak languages other than English; future studies should examine potential explanations including vaccine access, social desirability, less hesitancy, and higher perceived risk.

Adolescents who were 13 years old were very likely to have received only Tdap, whereas by age 17 years, this pattern was far less common. This may indicate that Tdap is offered prior to offering of the other 2 vaccines, perhaps because it is more likely to be in stock. In addition, although few U.S. adolescents received only the MenACWY vaccine (4.3%, 95% CI=4.0%, 4.6%), MenACWY-only vaccination was quite common among the other-language group (12.1%, 95% CI=8.8%, 15.4%), uninsured adolescents (10.3%, 95% CI=8.1%, 12.4%), and in households below the federal poverty line (5.9%, 95% CI=5.1%, 6.6%). For the other-language group, this could reflect increased willingness for MenACWY vaccine among immigrants coming from high meningitis incidence countries in Latin America, Asia, and sub-Saharan Africa. It’s unclear why uninsured and poverty-level adolescents would have high MenACWY only vaccine coverage, especially given its higher cost when compared with Tdap. It should be noted that the smaller sample sizes for this vaccine pattern result in relatively wide CIs for the uninsured and other-language populations.

Nearly one third of adolescents had a missed opportunity for HPV vaccine and both area and household SES were associated with missed opportunity. Area poverty showed a generally linear association but the association with household SES was more complex. Compared with those with the highest household income, those above the poverty line but with slightly lower income were more likely to have a missed opportunity, whereas those with the lowest household income (i.e., living below the federal poverty line) were less likely to have a missed opportunity (i.e., U-shaped pattern). These results may reflect the positive impact of public programs such as VFC on low-income populations.^26,27 These findings could also indicate a tendency for vaccine hesitancy among groups with higher SES.²⁸ However, studies on the association of various SES indicators and vaccine hesitancy are mixed.^29,30 Additional interventions are needed to increase HPV vaccination in the identified populations, including among higher-income households, in low-poverty areas, and in rural areas. For HPV vaccine specifically, psychosocial factors such as greater parental knowledge of HPV and its spread, higher parental perceived susceptibility of HPV infection, as well as greater adolescent knowledge are associated with increased first dose receipt.³¹ Much remains unknown about how these and other psychosocial factors might vary among the populations and types of places identified as having high likelihood of HPV missed opportunity.

Increasing adolescent vaccination is challenging because teens infrequently visit providers.² School entry requirements and school-based clinics²⁵ are strategies with great potential to increase coverage for all recommended vaccines. Similar to prior studies,^8,10 this study documented that Tdap only was more common than MenACWY only. School entry requirements may partially explain this. Whereas Tdap is required for school entry in all 50 states; as of 2021, only 35 states also require MenACWY. Only 3 states and the District of Columbia require HPV vaccine for either female adolescents only or both sexes; however, enforcement and effectiveness of these requirements vary widely.¹⁷ This analysis indicated that slightly more adolescents were fully vaccinated in states that had MenACWY requirements. Ultimately, multisector and multipronged solutions are needed to improve vaccine coverage and school entry requirements and school clinics are some effective strategies.³²

Once adolescents are seen by a provider, simultaneous (i.e., same-day) vaccination of all 3 vaccines can help ensure complete coverage. Bundling of adolescent vaccines during the same visit is an efficient strategy recommended by ACIP, the American Academy of Pediatrics, and National HPV Vaccination Roundtable.^33–37 Estimations suggest HPV vaccine 1-dose coverage could reach 91%–93% if administered simultaneously with other vaccines.^5,38 A recent study illustrated that the practice of injection limiting (i.e., providers only offering 1 immunization per adolescent visit) leads to a high likelihood of incomplete HPV immunization.⁹ Vaccine bundling and simultaneous vaccinations have increased in recent years.^7,10 In 2019, however, 16% of adolescents in Oregon experienced injection limiting at the point of care and never received >1 immunization per visit from age ≥9 years.⁹ Simultaneous administration is timely given the recent approval of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) vaccine for children and adolescents. ACIP recently removed restrictions around co-administration of other vaccines with the SARS-CoV-2 vaccine; thus, there will be additional opportunities to increase adolescent HPV vaccine coverage.³⁹ Another strategy to achieve higher on-time HPV coverage specifically is the recent American Cancer Society recommendation encouraging providers to start offering the HPV vaccine at age 9 years.⁴⁰ More research is needed to determine whether bundling vaccine recommendations or initiating recommendation for HPV prior to other adolescent vaccines will be most effective at increasing HPV coverage. Other recommended strategies, such as standing orders for all vaccines, are also needed.³⁶ The goal was to assess overall adherence to recommended vaccines, and this study did not specifically measure simultaneous vaccination as others have done.^7–9 Additional studies are needed to assess how the practice of injection limiting impacts coverage⁹ and to understand the extent to which simultaneous vaccination can increase coverage.

Limitations

This study faces several limitations, including smaller sample size for some groups, such as those responding in a language other than Spanish or English and those who received only HPV vaccine, resulting in wide CIs for some estimates. The sample was relatively small for each state and the American Community Survey data are pooled multiyear estimates that can have relatively large margins of error at the ZIP Code level. The NIS-Teen response rate was relatively low and could vary by region and area. The survey was available in Spanish and English. For participants speaking other languages, interviewers used real-time telephonic translation; however, data about language spoken were unavailable. However, this study has several strengths. Analysis of restricted-use ZIP Code–level data is notable because it allowed for identification of the types of areas where additional education, outreach, and delivery efforts are needed. Additionally, nationally representative data were used, disparities among 8 vaccine patterns (rather than single vaccines in isolation) were examined, and provider-verified data were used to ensure accuracy.

CONCLUSIONS

Given the 3 vaccines’ divergent patterns, it is critical to identify and address specific challenges related to the availability, accessibility, delivery, storage, recommendation and communication by providers, and parental acceptance of vaccines for each vaccine type. Evidence-based⁴¹ policy and program solutions are needed, and should be targeted to populations and places in need, especially rural communities. Although growing evidence indicates the importance of area-level factors and geospatial processes on vaccine coverage,^13,42,43 more research is needed to identify programs and policies that solve observed geographic disparities. The large number of HPV missed opportunities will likely result in thousands of HPV-related pre-cancers and cancers in the U.S. Ultimately, these results lend urgency to ongoing efforts to increase complete vaccination and avoid missed opportunities among U.S. adolescents.