Age nine is possible: Improving age 9 HPV initiation through a national quality improvement initiative during the COVID-19 pandemic

ABSTRACT

The American Cancer Society collaborated with a range of healthcare partners in 2020–2022 to implement quality improvement clinical interventions with the goal of improving HPV vaccination rates among adolescents’ ages 9–13. 2020 was the first cohort for which partners had been asked to submit HPV rate data for patients’ ages 9–12. At least 80% of the partners across all reported project years were able and willing to report HPV rates for these ages. Partners submitted HPV initiation rates at the beginning and end of the 12-month project year along with project activities, including evidence-based interventions (EBIs) implemented. Mean initiation rates for ages 9–10 significantly increased 4.1% during 2020 compared to non-significant rate increases of 2.6% and 2.0% for ages 11–12 and age 13, respectively. In 2021, ages 9–10 initiation saw a non-significant increase of 2.2%, whereas ages 11–12 and age 13 decreased non-significantly by 0.3% and 0.1%, respectively. The 2022 cohort saw significant initiation rate increases of at least 4% across all ages, potentially a promising result of the myriad back on track HPV vaccination campaigns designed to reverse the damage of the COVID-19 pandemic on adolescent immunizations. These findings demonstrate an effective adaptation of quality improvement in increasing HPV vaccination coverage among younger ages even during a national pandemic.

KEYWORDS:

• Adolescent immunization
• HPV vaccination
• initiation
• age 9
• quality improvement
• cancer prevention
• COVID-19

Introduction

More than 37,000 people in the United States develop cancers caused by the human papillomavirus (HPV) every year.¹ The HPV vaccine presents an extraordinary opportunity to prevent more than 90% of cancers and precancers caused by the virus.^1,2 The two-dose vaccination series offers effective and long-lasting protection before a person is exposed to HPV.³ The HPV vaccine is most effective when administered at younger ages, prior to exposure to the virus.^3,4 The CDC recommends routine vaccination for HPV at ages 11–12, though the series can be started as early as age 9.⁵ Growing evidence suggests myriad benefits to beginning the HPV vaccination series at age 9, ranging from an increased likelihood of on-time completion^6–8 to less stigma around the association with sexual activity and a greater focus on cancer prevention.⁹

In 2018, the American Academy of Pediatrics (AAP) updated its recommendations to include ages 9 and 10 as part of the recommended age range for initiation, a subtle but notable change in emphasis that recognizes the many advantages of earlier HPV initiation.¹⁰ In 2020, the ACS updated its guidelines¹¹ to encourage providers to start offering the HPV vaccine series at ages 9 or 10 to achieve higher on-time completion rates and ultimately prevent more cancers. Subsequently, intervention programming and data collection to promote age 9 initiation were built into national quality improvement cohorts. The AAP and the ACS both endorse initiation at age 9.

HPV vaccination coverage lags behind other adolescent vaccinations, such as meningococcal and tetanus, diphtheria, and pertussis (Tdap).¹² This lag is due to several factors, including the perceived link with sexual activity.¹⁰ Additionally, unlike other adolescent vaccines, HPV is rarely required for school attendance and requires a strong cancer prevention recommendation from a medical provider.¹³ The COVID-19 pandemic has only widened this gap,^14–16 further amplifying the importance of HPV initiation at age 9 as a tool to improve coverage and on-time completion rates at age 13.

ACS has successfully supported health systems to improve HPV vaccination rates since 2015.^17,18 This paper reports on 3 years of ACS quality improvement clinical intervention partnerships with healthcare partners to better understand the intentions and impact on starting the HPV series at age 9. Our analysis explores the impact of quality improvement interventions in health systems for HPV vaccine series initiation among adolescents ages 9–10. We looked specifically at capacity (i.e., can systems initiate and track at ages 9–10?), the unforeseen challenges posed by COVID, the role of health system type on HPV series initiation rate change for ages 9–10, and the influence of project aim statements focused on ages 9–10. These findings expand on existing literature by demonstrating an effective adaptation of quality improvement in increasing HPV vaccination coverage among younger ages even during a national pandemic.

Methods

Health system sample

In 2020–2022, the American Cancer Society (ACS) recruited federally qualified health centers (FQHCs), hospitals, health departments, and integrated delivery system (IDS) partners to participate in 12-month clinical intervention quality improvement (QI) projects to increase HPV vaccination rates among patients ages 9–13. All system partners had an existing relationship with an ACS field staff. To be eligible for participation, systems had to be routinely vaccinating their patient population in-house (i.e., not referring out for immunizations).

Intervention design

Clinical intervention QI projects ran from January through December of the respective cohort year. Local ACS field staff served as account managers for health systems participating in the cohort. ACS staff engagement included identifying health systems and clinical champions, building or joining a core QI team, facilitating the development of a project aim statement, supporting the selection and implementation of multiple evidence-based interventions (EBIs) through QI processes,¹⁹ sharing resources, and making connections between health systems. QI team composition varied by health system but often included a QI lead, provider champion, nurse and/or medical assistant, information technology, office managers, and immunization/patient care coordinators. National ACS QI and programming staff provided monthly technical assistance and training to ACS field staff on HPV rate data, evidence-based interventions, and QI strategies outlined by the Institute for Healthcare Improvement (IHI) Model for Improvement.²⁰ QI strategies included, but were not limited to plan-do-study-act (PDSA) cycle, which is an iterative model of testing small changes that could be scaled to expand and increase vaccination to younger target audience, process mapping to integrate younger age groups efficiently and systematically into the existing workflow, and root cause analysis to explore the causes of identified problems affecting ability to expand vaccination to 9- to 10-year olds. ACS field staff then shared the QI strategies and HPV rate data resources with their QI team throughout the project. This study was determined to be not human subjects research by the Institutional Review Board at the Morehouse School of Medicine (protocol #786449).

Data collection instrument

Health system partners submitted the data and reporting instruments at baseline and final. Baseline data covered the calendar year prior to the intervention. Final data covered the calendar year the intervention was implemented. The 2020 cohort was the first year that health system partners were asked to report HPV rate data for ages 9–10 and ages 11–12. Prior cohorts (2016–2019) reported on age 13 only. The baseline instrument included health system demographics, system background, rate data for HPV, meningococcal, and Tdap, and a project activity plan to guide QI strategy and EBI selection, including the development of an aim statement. The final instrument collected vaccination data, selections on the EBIs implemented, and other questions around implementation and outcomes. Data collection was done using the Shiny²¹ platform in 2020 and REDCap^22,23 in 2021–2022. Additional details about relevant data collected for the focus of this paper include the following:

Target patient population

System partners reported their active 9–13-year-old medical patient population at two time points using the following age groupings: ages 9–10, ages 11–12, age 13. Active patients were defined as those who had one or more medical visits at the health system in the last 12 or 24 months.

Adolescent vaccination rates

Systems reported system-level counts for the number of active patients who had ever received at least one dose of the HPV vaccination series (HPV initiation) and the number of active patients who had ever received both doses of the HPV vaccination series (HPV completion). Counts were grouped by patient age during the identified measurement period (i.e., pre-intervention year for baseline data or intervention year for final data) using the following categories: ages 9–10, ages 11–12, age 13.

Systems reported the 9–13-year-old active patient population for participating project sites using the same age groupings. Vaccination rates were calculated for HPV initiation and HPV completion for each age group by dividing the count of patients who had received the specified vaccine dose by the total active patient population. Meningococcal and Tdap vaccination rates were calculated for ages 11–12 and age 13.

Implementation strategies

System partners were offered a menu of EBIs that are documented as effective in increasing adolescent HPV vaccination rates, including among patients ages 9–10.^19,24 The process for selection of interventions varied depending on the gaps in processes for each partnering system. In some cases, ACS staff trained in quality improvement assisted partners in a gap analysis to select effective interventions. In other cases, system partners made improvements to interventions they were already implementing to increase effectiveness. Systems were encouraged to implement multi-component interventions in accordance with best practices for vaccination interventions.^25,26 Strategies were provider-focused, patient-focused, or system-focused, including (1) Training: general training of clinicians and staff on HPV vaccination as a cancer prevention tool as a part of the project; (2) Patient reminders and recall: sending reminders to members of a targeted population that vaccinations were due (reminders) or late (recall). Reminders and recalls were delivered by various methods including telephone, letter, postcard, and text message; (3) Provider prompts: creating systems to inform providers that individual patients were due for vaccination. Common methods to prompt providers included daily huddles, notes posted in patient charts, and electronic health record alerts; (4) Standing orders: creation of an official policy enabling patient vaccination without the need for examination or direct order from the attending provider; (5) Provider assessment and feedback: evaluation of provider performance in delivering the HPV vaccine to a client population and providing feedback on their performance. Systems indicated the EBIs they had selected using a yes/no binary scale. Additional factors affecting systems’ intervention selection included limited resources, existing priorities, staffing and staff time, leadership buy-in, and capacity. Systems also reported the number of providers and other staff who were trained/educated on HPV vaccination as part of the project. The addition of patients ages 9–10 did not alter the content or offerings of EBIs.

Aim statement development

At baseline, systems developed an aim statement to guide the clinical intervention(s). Aim statements help to answer the first questions in the Model for Improvement: “What are we trying to accomplish and how will we know if we’ve been successful?.” Each aim statement included the following: a specific numeric goal (e.g., to increase HPV initiation by 5%), the target population (e.g., adolescents ages 9–13), the intervention setting (e.g., all pediatric clinics in X system), and the time period (e.g., January – December 2020 for the 2020 cohort). A sample aim statement may look like: We aim to increase HPV initiation in adolescents ages 9–13 from 35% at baseline to 40% by December 31, 2020. Systems were instructed to be as specific as possible in crafting their aim statement relative to their baseline and capacity to influence vaccine rates and had ongoing support and coaching from their ACS field staff lead.

Data quality review and analysis

We reviewed all vaccination rate data for quality issues, including but not limited to instances where series completion rates exceeded series initiation rates, improbable vaccination rates, and improbable rate change between time points. Health systems that were unable to correct issues were excluded for vaccination rate analyses. Vaccination rate analyses were conducted separately by cohort to avoid masking effects and to acknowledge the different pandemic stages across 2020, 2021, and 2022. Rate change was assessed with two-tailed paired t-tests on mean baseline and final HPV initiation rates for each age group (9–10, 11–12, 13), a significance level of .05, and Cohen’s d for repeated measures. The same analyses were conducted for HPV completion, meningococcal, and Tdap vaccination rates but are not relevant to the focus of this paper. Stata/IC version 16.0²⁷ was used for all analyses.

A one-sample two-tailed t-test was used for each cohort to assess if HPV initiation 9–10 rate change differed by health system type. Health systems were categorized into FQHCs and Look-a-likes (LAL), which included any safety-net clinics, rural health clinics, or local government health departments providing vaccinations, and Hospital/IDS, which grouped hospital systems and integrated delivery systems together. Pediatric primary care systems in 2022 cohort were excluded from this analysis to maintain consistency in comparison.

Aim statements for all cohorts were combined and categorized for project focus on patients ages 9–10. Systems that did not submit aim statements were coded as missing. Aim statements were categorized using the following schema: inclusion of age 9, where statements indicated project focus would include the full 9–13 age range and/or focus explicitly on patients ages 9–10 for HPV initiation; or exclusion of age 9, where statements indicated project focus would be on patients ages 11–13 and/or HPV completion. A two-tailed one sample t-test collapsed across cohort year was conducted on HPV initiation rate change for ages 9–10 by aim statement focus to determine if systems’ project focus significantly impacted HPV initiation rates for patients ages 9–10 during the project. Cohorts were combined due to sample size. The aim statements of systems with excluded rate data were not included in this analysis. Chi-square analyses were performed on the top three most frequently reported EBIs and project focus by cohort year to determine if implementation strategies differed significantly by inclusion of age 9 initiation in the aim statement.

Results

Health system partner demographics

ACS engaged 64 health systems for quality improvement projects to improve HPV vaccination in 2020, 43 in 2021, and 52 in 2022 (Table 1). These cohorts had the potential to reach a collective 9–13-year-old active patient population of 288,629 in 2020, 289,612 in 2021, and 335,303 in 2022. Eighteen (42%) of the 43 systems in 2021 were returning systems from the 2020 cohort and another 23 (44%) of the 52 systems in 2022 had continued from the 2021 cohort. ACS supported a total of 159 HPV vaccination projects with health systems across all 3 years of programming.

Table 1. Characteristics of health systems participating in the clinical intervention quality improvement project.

	2020 (n = 64)	2021 (n = 43)	2022 (n = 52)
Characteristic	n (%)	n (%)	n (%)
Patient population (ages 9–13) during intervention, total (median)	288,629 (2,170)	289,612 (2,811)	335,303 (1,994)
Participating clinics, total (median)	479 (4)	368 (5)	456 (4)
System reported on ages 9–10	51 (80%)	35 (81%)	44 (85%)
Project focus included ages 9–10	15 (23%)	17 (40%)	19 (37%)
System was funded	23 (36%)	14 (33%)	34 (65%)
Funding amount, mean	$27,400	$18,800	$20,221
System Type
FQHC	37 (58%)	21 (49%)	21 (40%)
Health Department	2 (3%)	1 (2%)	3 (6%)
Hospital/IDS	25 (39%)	16 (37%)	18 (35%)
Pediatric Private Practice	–	–	9 (17%)
Rural Health Clinic	–	5 (12%)	1 (2%)
Clinic Type
Rural	23 (37%)	16 (37%)	19 (37%)
Suburban	21 (33%)	18 (42%)	20 (39%)
Urban	39 (62%)	24 (56%)	35 (67%)
Other	11 (26%)	–	13 (25%)
U.S. Region
Northeast	7 (11%)	6 (14%)	9 (17%)
Midwest	12 (19%)	6 (14%)	5 (10%)
South	39 (61%)	28 (65%)	29 (56%)
West	6 (9%)	3 (7%)	9 (17%)
System patient population (ages 9–13) during intervention
<1000	18 (28%)	10 (23%)	11 (21%)
1001–2500	18 (28%)	8 (19%)	20 (39%)
2501–6000	11 (17%)	14 (33%)	12 (23%)
6001–10000	10 (16%)	7 (16%)	3 (6%)
>10000	7 (11%)	4 (9%)	6 (12%)
Reported 1+ EBI targeting:
Provider	52 (81%)	38 (88%)	51 (98%)
Patient	54 (84%)	36 (84%)	52 (100%)
System	39 (61%)	28 (65%)	28 (54%)

FQHCs represented the largest, though decreasing, proportion of participating systems during all 3 years of programming, comprising 58% of partners in 2020 (37 of 64 systems), 49% in 2021 (21 of 43 systems) and 40% in 2022 (21 of 52 systems). Hospitals or IDS partners consistently made up a little over a third of participating partners, or 39% in 2020 (25 systems), 37% in 2021 (16 systems), and 35% in 2022 (18 systems). Health departments, pediatric private practices, and rural health systems were a small portion of the partners engaged all 3 years. Historically, FQHCs were the target partners when the initial QI intervention program was piloted in 2015 for two reasons: ACS field staffs’ existing relationships with FQHCs and expressed need in FQHCs for this type of intervention. The intervention scope expanded in 2017 to include other types of health systems, such as integrated delivery systems (IDS), hospitals, and private practice. Participating health systems were spread across in 26 states in 2020, 14 states in 2021, and 21 states in 2022 (data not shown). Regional distribution was the highest in the South region, hosting 39 partners in 2020 (61%), 28 in 2021 (65%), and 29 in 2022 (56%). The Northeast and Midwest representation ranged from 10% to 19% of partners across all 3 years. The West region saw higher representation in 2022 with 9 systems (17%), when compared to 2020 and 2021, 6 systems (9%) and 3 systems (7%), respectively.

These health partners included 479 participating clinics in 2020, 368 clinics in 2021, and 456 clinics in 2022. The median number of clinics reported for all 3 years was 4, with the number of clinics per system ranging from a single clinic up to 61 clinics (data not shown in Table 1). Most participating systems during all 3 years of programming reported reaching a combination of urban, suburban, and rural locations for their clinic sites. Two-thirds of all participating systems reported urban clinics (62% in 2020, 56% in 2021, and 67% in 2022). More than one-third also reported suburban (33% in 2020, 42% in 2021, and 39% in 2022) or rural clinics (37% all 3 years). A quarter of systems reported other locations for clinics (26% in 2020 and 25% in 2022), which were most often school-based or mobile clinics.

Health system capacity to report vaccination data on ages 9–10 grew from 80% (51 out of 64 systems) and 81% (35 of 43 systems) in 2020 and 2021, respectively, to 85%, or 44 of the 52 participating systems in 2022. Participating health systems that reported including adolescents ages 9–10 in their aim also increased from 23% in 2020 to 40% and 37% in 2021 and 2022, respectively.

Six systems from 2020, two systems from 2021, and 7 systems from 2022 were excluded from all rate analyses due to unresolved data quality issues, resulting in a sample size of 58, 41, and 45 systems for the 2020, 2021, and 2022 cohorts, respectively. Of those that had no data quality issues, 51, 35, and 44 systems reported HPV initiation rates for patients ages 9–10 in the 2020, 2021, and 2022 cohorts, respectively.

In 2020, 23 systems (36%) received funding from a variety of sources, such as ACS region-specific funds, philanthropy organizations, and existing CDC funding mechanisms (mean amount of $27,400). Fourteen systems (33%) received funding from the same variety of sources in 2021 (mean amount of $18,800). Thirty-four systems (65%) received funding in 2022 (mean amount of $20,221).

Table 1 provides additional information about system partner characteristics for all cohorts.

Vaccination rate change

The mean HPV series initiation rates for ages 9–10 increased from baseline to final for all 3 years of program implementation. The 2020 cohort saw a statistically significant 4.1% point increase from 11.4% to 15.5% (p = .04, Cohen’s d = 0.29). The 2021 cohort's mean HPV initiation rate for ages 9–10 also trended positively, increasing by 2.2% points from 9.0% to 11.2%. In 2022, mean HPV initiation significantly increased 4.6% points from 10.0% to 14.6% for the 9–10 age group (p = .04, Cohen’s d = 0.51).

Rate change by age

For all cohorts, baseline initiation rates for age 13 and ages 11–12 are markedly higher than baseline initiation rates for ages 9–10. For example, during the 3 years of programming rates for age 13 ranged from 63.5% to 66.6%, cohort averages for ages 11–12 ranged from 47.4% to 51.1%, while average rates for 9–10-year-olds ranged from 9.0% to 11.4%.

HPV initiation rates for ages 9–10 increased the most during the 2020 and 2021 cohorts project compared to the other age groups in respective cohorts, which saw modest changes or maintained rates. Ages 11–12 and age 13 groups also saw overall drops in vaccination rates in 2021 and 2022. HPV initiation rates increased for all age groups in 2022. See Table 2.

Table 2. Mean HPV initiation rates by age group and cohort.

Age Group	Cohort	n	Baseline*	Final*	Rate Δ*	p	Cohen’s d
9–10	2020	51	11.4% ± 15.9	15.5% ± 18.7	4.1%	.04^†	0.29
	2021	35	9.0% ± 12.9	11.2% ± 15.5	2.2%	.24	0.20
	2022	37	10.0% ± 13.6	14.6% ± 15.4	4.6%	.04^†	0.51
11–12	2020	56	51.1% ± 21.4	53.7% ± 20.5	2.6%	.25	0.15
	2021	40	49.2% ± 20.7	48.9% ± 17.5	−0.3%	.90	0.02
	2022	45	47.4% ± 18.8	51.5% ± 17.0	4.1%	.02^†	0.35
13	2020	51	66.1% ± 21.1	68.1% ± 17.5	2.0%	.40	0.12
	2021	36	66.6% ± 18.8	66.5% ± 19.9	−0.1%	.97	0.01
	2022	45	63.5% ± 20.7	68.2% ± 14.9	4.7%	.04^†	0.32

*Vaccination rate, mean ± SD. † significant at p < .05.

Impact of implementation strategies on vaccination rate change

In each cohort year, systems implemented four EBIs, on average. Parent or patient education was the most implemented EBI in all 3 years and reported by 75.0% of systems in 2020 (n = 48), 79.1% of systems in 2021 (n = 34), and 86.5% of the systems in 2022 (n = 45) (data not shown). Examples of this education include posters in exam rooms with HPV vaccination guidelines for vaccination, flyers/handouts with HPV vaccine information, often translated into several languages, and verbal education with patients/parents about the vaccine (akin to recommendation). Patient reminders were the second most common EBI implemented in 2020 (n = 47, 73.4%) and 2022 (n = 41, 78.8%) and the third most common in 2021 (n = 29, 67.4%). Provider recommendation training was the other most frequently implemented EBI, ranking third in 2020 (n = 46, 71.9%) and 2022 (n = 40, 76.9%) and second in 2021 (n = 33, 76.7%). In 2022, we asked specifically about provider training for age 9 initiation in addition to other provider training content.

Rate change for ages 9–10 initiation was not significantly impacted by number or type of EBI selected for any cohort year (data not shown).

Health system differences on age 9 initiation

We found no significant differences across all 3 years of interventions in HPV initiation changes for ages 9–10 by health system type, though hospital/IDS systems tended to see greater rate change than FQHCs. A lack of statistical power may account for this. See Table 3 for specific values.

Table 3. Mean HPV initiation rates for ages 9–10 by system type and cohort.

Cohort	System Type	n	Baseline*	Final*	Rate Δ*	p	Cohen’s d
2020	FQHC/LAL	29	14.6% ± 19.4	18.3% ± 19.5	3.7% ± 13.6	.80	0.07
	Hospital/IDS	21	7.4% ± 8.6	12.1% ± 17.7	4.7% ± 15.3
2021	FQHC/LAL	20	9.2% ± 12.8	9.4% ± 13.2	0.2% ± 13	.22	0.43
	Hospital/IDS	15	8.8% ± 13.5	13.5% ± 18.4	4.8% ± 6.3
2022	FQHC/LAL	17	8.7% ± 12.4	11.5% ± 13.2	2.8% ± 7.5	.20	0.47
	Hospital/IDS	14	11.1% ± 12.2	18.4% ± 16.5	7.3% ± 12

*Vaccination rate, mean ± SD.

2020 cohort

FQHC/LAL mean HPV initiation rates for ages 9–10 increased from 14.6% at baseline to 18.3% at final, an increase of 3.7% points. Hospital/IDS systems’ initiation rates for ages 9–10 increased 4.7% points from 7.4% at baseline to 12.1% at final. These mean rate changes did not differ significantly by health system type.

2021 cohort

FQHC/LAL mean HPV initiation rates for ages 9–10 increased slightly from 9.2% at baseline to 9.4% at final, an increase of 0.2% points. Hospital/IDS systems’ initiation rates for ages 9–10 increased 4.8% points from 8.8% at baseline to 13.5% at final. Though not statistically significant, Hospital/IDS trended toward slightly greater rate change, particularly for the 2021 cohort. This result is discussed further below in the context of varied impact of COVID-19 on different health systems.

2022 cohort

FQHC/LAL mean HPV initiation rates for ages 9–10 increased from 8.7% at baseline to 11.5% at final, an increase of 2.8% points. Hospital/IDS systems’ initiation rates for ages 9–10 increased 7.3% points from 11.1% at baseline to 18.4% at final. Though not statistically significant, Hospital/IDS saw greater rate change, as with the 2021 cohort. This result is discussed further below in the context of varied impact of COVID-19 on different health systems.

When looking across all 3 years of programming, we observed dropping baseline rates at FQHCs, from 14. 6% in 2020, to 9.2% in 2021, and 8.7% in 2022. In contrast, Hospital/IDS partners saw gradually increasing baseline rates for HPV initiation among ages 9–10, from 7.4% in 2020, to 8.8% in 2021, and 11.1% in 2022.

Impact of aim statement focus

Age 9 initiation

Fifty aim statements across 2020–2022 included age 9 initiation and 85 aim statements excluded it. Excluded aim statements focused explicitly on any combination of ages 11–13 and/or focused their project on HPV completion. Nine systems were missing aim statements (n = 6 systems from 2020 cohort; n = 3 systems from 2021 cohort).

Systems with a project focus that included ages 9–10 had greater HPV initiation rate change for ages 9–10 (M = 4.9%) than systems with a project focus that excluded ages 9–10 (M = 3.0%), though not significantly so (p = .41) – see Table 4. Systems missing initiation rates at either time point were excluded from the analysis. Baseline initiation rates for ages 9–10 differed significantly for systems that included ages 9–10 (M = 14.6%; SD = 15.9) and systems that excluded ages 9–10 (M = 6.1%; SD = 10.3), (p < .001, Cohen’s d = 0.66).

Table 4. Mean HPV initiation rate change for ages 9–10 by aim statement, combined cohorts.

Aim Statement Focus	n	Baseline*	p	Cohen’s d	Final*	p	Cohen’s d	Rate Δ*	p	Cohen’s d
Included Ages 9–10	47	14.6% ± 15.9	.00^†	0.66	19.5% ± 17.8	.00^†	0.66	4.9% ± 14.4	.41	0.16
Excluded Ages 9–10	67	6.1% ± 10.3			9.2% ± 13.8			3.0% ± 9.4

Systems missing 9–10 rates at any time point were dropped. *Vaccination rate, mean ± SD. †significant at p < .01

Implementation strategy selection

The relationship between projects with an age 9 focus and the implementation of parent/patient education was significant in 2021, X² (1, N = 38) = 6.61, p < .01. All 15 systems with an age 9 project focus chose to implement parent/patient education compared to 65% (N = 15) of systems without an age nine project focus. Aim statement focus did not significantly impact implementation for any other EBIs in 2021 or the other project years.

Aim statement change for returning systems

2020–2021 cohorts

Eighteen systems participated in 2020 and 2021. Of the 13 systems that did not include initiation at ages 9–10 in their 2020 aim statement, 5 systems (38.5%) incorporated it, either on its own or with ages 11–13, in 2021.

2021–2022 cohorts

Twenty-three systems participated in 2021 and 2022. Of the 12 systems that did not include initiation at ages 9–10 in their 2021 aim statements, 3 systems (25%) incorporated it, either on their own or with ages 11–13, in 2022.

Discussion

Between 2020 and 2022, the American Cancer Society partnered with health systems to support 159 12-month-long HPV vaccination quality improvement projects. Despite the moderating effects of the COVID-19 pandemic on routine adolescent health care and vaccination, healthcare systems across a range of sizes and types demonstrated eagerness to train providers to recommend the HPV series starting at age 9, as well as the ability to both monitor and improve HPV initiation rates as early as age 9 for the first time using a range of provider-, patient-, and system-focused evidence-based interventions. ACS’s existing QI framework was successfully adapted to include HPV vaccination for ages 9–10. Participating systems reported increased capacity to report data on ages 9–10 and project aim statements that focused on vaccination for ages 9–10 also increased between 2020 and 2022. Specifically, in the subset of systems that participated in at least 2 years of the interventions, at least a quarter of systems adjusted their project focus to include initiation at ages 9–10 in the second project year. Neither health system type nor aim statement focus significantly impacted average vaccination rate change among partners reporting ages 9–10 data, suggesting that intervention impact on vaccination rates can be beneficial even when not directly targeting the 9–10-year-old population. We did find that 2021 systems with an age 9 focus were more likely to implement parent/patient education than 2021 systems without an age 9 focus. Given the general trend of no other significant impact of aim statement on project focus, this is likely an anomaly unique to 2021.

This clinical quality improvement project spanning 3 years was significantly impacted by the COVID-19 pandemic. The pandemic impacted the 2021 cohort especially, in terms of the number of systems participating and vaccination initiation rates in FQHCs/LALs as compared to hospital/IDS sites. Though the patient population remained steady from 2020 to 2021, the number of participating systems decreased between 2020 and 2021. Mean initiation rate change for ages 9–10 was also lower for FQHCs/LALs in 2021 compared to 2020. Starting baseline rates for FQHC/LALs partners dropped from 2020 to 2021 and then again from 2021 to 2022. Hospitals/IDS initiation baseline rates and rate changes held year to year, suggesting hospital and IDS partners were better resourced in the face of the pandemic than FQHCs. This data supports the well-documented challenges faced by publicly supported health systems during the first year of the pandemic, including a substantial drop in patient population^28,29 and a decrease in the number of routine pediatric vaccine ordering and doses.¹⁵ Additionally, while we observed higher rate changes among IDS/hospital partners, we were unable to detect significant differences in vaccination rate change between the two types of health systems. This either means that our intervention impacted FQHCs and hospitals similarly or that our sample size was insufficient to detect significant differences.

Notably, an increasing percentage of participating focused on initiation at ages 9–10 each year. This may be attributed to the sizable number of systems continuing for multiple years and/or the influence of ACS’s staff and guideline guidance. The large proportion of systems that provided data on age 9 initiation demonstrated the ability of partners to monitor patient vaccination by multiple age groups.

Our findings contribute to growing evidence that health systems implementing QI initiatives among adolescents ages 9–10 can improve HPV series initiation rates significantly. Vaccination rate change for ages 9–10 exceeded those of ages 11–12 and age 13 in 2020 and 2021 and were comparable to older ages in 2022. Initiation rates for ages 9–10 have been lower than 11–12-year-old rates,^6,9,30,31 highlighting the vast opportunity for QI initiatives in primary and pediatric settings. The HEDIS Immunization Measure for Adolescents (IMA) is the industry benchmark for vaccination by age 13 with 2020 HPV completion rates ranging from 29.2% to 39.9% depending on insurance type.³² To raise HPV vaccination to levels on par with Tdap and meningococcal rates, immunization practitioners have demonstrated that starting at 9 provides more time to deliver all recommended Advisory Committee on Immunization Practices (ACIP) adolescent vaccines by age 13 (i.e., “on time”).^6–8 Additional rationale for formalizing HPV vaccination starting at 9 includes the sheer volume of vaccinations now needed for adolescents, with the other two control vaccines, flu, and now COVID it could mean 5 vaccines at one visit. Vaccinating early would decrease the number of shots needed for 11–12-year-olds.

We were unable to determine any effect of unique evidence-based interventions on vaccination rate change. We believe this is due to two reasons. First, each system selects from a package of intervention options based on their goals and unique situation. This customizable approach to intervention enables systems to tailor interventions to their needs, but it can mean that detecting the role or impact of individual interventions can be difficult. Second, we may need a larger sample size to account for the large number of EBIs systems choose from. Further research is needed to help better understand the volume of impact different EBIs have alone and in combination.

During the 3 years of interventions covered in this paper, ACS actively prioritized engaging rural-reaching health systems due to the rural urban divide in HPV and adolescent vaccination rates.³³ Our efforts to increase the volume of health systems that reported rural-reaching clinics were limited. Further efforts are needed to find ways to successfully recruit and engage rural reaching health systems to continue to address this vaccination divide.

Systems that developed aim statements focused on ages 11–13 still saw positive rate change for HPV initiation among adolescents ages 9–10. This suggests that health systems may still impact HPV vaccination initiation rates starting at 9 regardless of whether they formally included ages 9–10 in their aim statement. Prior quality improvement projects on HPV vaccination have demonstrated that projects focused on HPV vaccination may also have carryover effects on the rates of the other two adolescent vaccinations, meningococcal and Tdap.¹⁷ Further exploration is needed to see if a similar effect is happening with vaccination rates among ages 11–12, even when a system is aiming to target ages 9–10.

Findings from the 2022 cohort offered some indication that pandemic recovery was underway. Not only was the cohort larger in number of participating systems and patient population (nearly on par with the 2020 cohort, which had been recruited in fall 2019), but mean initiation rate change was comparable across all age groups above and beyond the respective rate change seen in 2021 or 2020. Multiple back on track vaccination campaigns,³⁴ many of which were spearheaded by the ACS, may have contributed to this upswing. This project is ongoing in 2023 with 32 health system partners participating. We will continue to track HPV vaccination rates and impact as the United States moves out of the acute stage of the COVID-19 pandemic.

Limitations

Small sample sizes for types of health systems may have limited the ability to detect statistical significance. HPV initiation rates for ages 9–10 at baseline were significantly different at baseline between systems focusing on ages 9–10 and those focusing on an older age group, suggesting existing differences between the two groups.

Conclusions

ACS experience demonstrates that inclusion of age 9 in data collection and intervention efforts is critical to driving up initiation rates in the crucial window before age 13 when cancer prevention is most effective. Systems agreeing to implementation of a QI HPV project are prepared to train providers on the importance of initiating the HPV vaccination series at age 9; monitor and collect age 9 initiation data; and have achieved increases in age 9 initiation rates as a result. As healthcare systems look to catch up adolescent patients on missed vaccines while ensuring provision of on-time delivery of vaccination to new cohorts, a focus on age 9 initiation serves as an effective intervention category.

Abbreviations

HPV	=	Human Papillomavirus
ACS	=	American Cancer Society
QI	=	quality improvement
EBIs	=	evidence-based interventions

Acknowledgments

The authors would like to recognize the ACS Patient Pillar Community Impact staff leading HPV quality improvement work in addition to ACS’s valued health system community partners working to prevent HPV cancers.

Disclosure statement

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

Supplementary material

Supplemental data for this article can be accessed on the publisher’s website at https://doi.org/10.1080/21645515.2023.2284359.

Additional information

Funding

The American Cancer Society’s HPV vaccination work is supported in part by the Centers for Disease Control and Prevention Cooperative Agreement [#5NU38OT000283-05]. Additional funding to support portions of personnel and program implementation efforts is provided by Merck Sharp & Dohme Corp, The National Football League (NFL), Genentech, Sanofi, and Lyda Hill Foundation.