The goal of this implementation research study is to understand whether a package of community-based interventions can increase access to and uptake of the human papillomavirus (HPV) vaccine among very young adolescent girls in rural Ethiopia. The main questions this study aims to answer are:

• Can a package of community-based interventions increase delivery of routine HPV vaccination to girls aged 9-14 in rural Ethiopia?
• Is an intervention model that incorporates strategies that address gender norms and gender-specific barriers more effective at increasing delivery of routine HPV vaccination than a model that does not explicitly address gender?
• What is the acceptability, feasibility, cost, and potential for maintenance and scale of an integrated health intervention to delivery routine HPV vaccination in rural Ethiopia?

To evaluate the effectiveness of the intervention, researchers will compare HPV vaccination within regions where the new intervention model is being implemented (intervention areas) to regions where the new routine HPV vaccination delivery model is not being implemented (comparison areas) for approximately 1 year before the new intervention model is implemented and for approximately 1 year after the start of implementation to compare changes over time and between intervention and comparison areas.

The study will evaluate the effect of the intervention on HPV vaccination delivery using routine health facility data.

To understand acceptability, feasibility, implementation, and potential for scale, the study will enroll participants including health officials and providers, adolescents girls receiving HPV vaccination and other services in intervention areas, and their parents/caregivers.

These participants may be asked to participate in one or more of the following:

• Completing a short questionnaire after receiving health services, including HPV vaccination, and their knowledge and attitudes towards HPV vaccination and experiences of care
• Participating in an interview or a focus group discussion to discuss HPV vaccination services and the role of the intervention in delivering care

Background Cervical cancer is the second most common cancer in Ethiopia. Vaccinating very young adolescent girls is the most effective long-term strategy to reduce its risk. Ethiopia introduced the human papillomavirus (HPV) vaccine in 2018, initially targeting 14-year-old girls through school-based campaigns and updated its guidelines in 2021 to target girls aged 9-14 years. Despite vaccinating around eight million girls since the launch of the program, very young adolescent (VYA) girls, particularly those in rural, out-of-school, and hard-to-reach areas, remain underserved, prompting new strategies to more effectively address this age group. Integrating HPV vaccinations into existing adolescent health services offers a strategic opportunity to enhance vaccine accessibility and coverage by utilizing established healthcare platform. This approach can address both logistical challenges and health disparities, ultimately improving adolescent health outcomes and contributing to the prevention of HPV-related diseases. To address this, implementation research is needed to evaluate the effectiveness of this approach. If successful, it could be scaled and adapted to similar contexts, contributing to the global effort to reduce cervical cancer incidence and mortality.

Objectives We propose a type 3 hybrid effectiveness implementation research study to evaluate the effectiveness, feasibility, fidelity, acceptability, and scalability of integrating HPV vaccination into a very young adolescent services package to increase HPV vaccine coverage among girls aged 9-14 years in rural Ethiopia.

A secondary objective of the research is to assess the feasibility, acceptability and potential for scalability of integrated HPV vaccination with VYA services with Ethiopia's existing 'Smart Start' platform.

Methods The study will be conducted in four intervention and four comparison woredas in Oromia, Sidama, and Somali regions of Ethiopia. It will use a hybrid effectiveness implementation research design, comprising two main phases from July 2024 to December 2025. Phase 1, the rapid formative and design phase, will include key informant interviews (KIIs), a desk review of the literature and existing data sources and retrospective service statistics data, and co-design workshops in which a design prototype ('Model 0') will be refined and finalized as the initial implementation model ('Model 1'). Phase 2 will include a quasi-experimental effectiveness component comparing HPV vaccine delivery to girls aged 9-14 in intervention and comparison areas, along with implementation research activities to assess adoption, reach, implementation, activity-based costing, and readiness for integration and scale-up. In additional, phase 2 will include gender formative phase, in which a formal gender analysis will be conducted using data from KIIs, quantitative exit interviews (EIs) with girls and caregivers, qualitative in-depth interviews (IDIs) and focus group discussions (FGDs) with girls, caregivers, and health workers. At the end of phase 2, 'pause-and-reflect' gender co-design workshops will be conducted to refine 'Model 1' based on the results from the formal gender analysis. The resulting 'Model 2' intervention model will explicitly incorporate a gender-lens, with the objective of engaging very young girls (aged 9-14 years), their caregivers, health workers, and communities to reduce gender barriers faced by very young girls to improve self-efficacy and increase access to and uptake of HPV vaccination and other integrated VYA interventions.

A mixed methods design, incorporating both quantitative and qualitative research methods, will be used. A total of approximately 40 KIIs will be conducted in each of the three phases, with 10 KIIs in each intervention woreda and at the national level conducted in each phase. A total of approximately 60 in-depth interviews (IDIs) will be conducted with parents/guardians, girls, and health workers in each implementation phase (phases 2 and 3), with ~15 IDIs in each intervention woreda. Approximately 8 focus group discussions (FGD) with eligible girls and their parents/guardians will be conducted in phases 2 and 3 each, with approximately 1 FGD with girls and 1 FGD with parents/guardians per intervention woreda per implementation phase. The EIs will be conducted with approximately 400 VYA girls and 400 parents/guardians in each of phases 2 and 3 in intervention areas. In addition, anonymized routine monitoring data, including aggregated service delivery data collected by program monitoring systems as well as through the Ethiopian Ministry of Health DHIS2 (an open-source health information management system) platform and other existing data sources, will be extracted and analyzed from both comparison and intervention woredas. Of the primary data collection proposed in this study, only program monitoring data will be collected in comparison woredas. Phase 3 activities are dependent on the confirmation of anticipated additional co-funding.

Descriptive analysis, using frequency, percentages, cross-tabulations, graphs and multivariable and time series analyses will be performed to analyze and present the quantitative data. Qualitative data will be systematically transcribed and analyzed using thematic analysis, in which the data will be structured into emergent codes (inductive). The study protocol has been reviewed and approved by the Institutional Review Board of the Ethiopian Public Health Association and the Research Ethics Board (REB) of Population Services International.

Expected outcomes The expected outcome of this study is to provide evidence on the effectiveness, acceptability, feasibility, and scalability of integrating HPV vaccination into the existing adolescent health services package to increase HPV vaccine coverage among girls aged 9-14 years in rural Ethiopia. Qualitative findings will offer insights into the acceptability of this approach among parents/guardians, adolescents, and health providers, while assessments of scalability and readiness for integration will provide crucial information on feasibility and cost-effectiveness. Quantitative findings will provide robust data on HPV vaccination rates and identify factors influencing coverage.

The study's key outputs will include data on HPV vaccine and other VYA service delivery among girls aged 9-14 in intervention and comparison areas. Study results will be disseminated at a national workshop and at scientific conferences, and the findings will be submitted for publication in peer-reviewed journals. De-identified study data may be made publicly available.

Dependent on securing co-funding, the study is expected to generate evidence on key gender-based norms, attitudes, and behaviors that limit girls' self-efficacy and serve as barriers to access to HPV vaccination and other VYA health services, as well as accompanying opportunities and proposed solutions for improving girls' self-efficacy, health outcomes, and positive outcomes across non-health sectors. Co-funding will support the testing of a gender-integrated HPV/VYA model, thereby generating evidence on the incremental effectiveness of gender-integrated programming on HPV vaccination models among 9-14-year-old girls in rural Ethiopia. Specifically, the phased implementation approach supported by the co-funding will allow evaluation of an initial model (model 1) and, additionally, of an iterative model (model 2) that addresses gender-related determinants of vaccination uptake and negative attitudes and perceptions related to gender, women and girl's autonomy in the implementation design.

Exclusion Criteria for key informant interviews

- Does not provide informed consent

Inclusion Criteria for focus group discussions and in-depth interviews:

• Girl: aged 9-14 years, female
• Caregiver: Parent/legal guardian of eligible 9-14-year-old girl (above); 18 years of age or older

Exclusion Criteria:

• Girl: Does not provide informed assent, including to audio-recording, Parent/legal guardian does not provide consent
• Caregiver: Does not provide informed consent, including to audio-recording

Inclusion Criteria for exit interviews:

• Girl accompanied by adult (18+) parent/legal guardian: Aged 9-14 years old; Female; Received a VYA health services supported by the program*, including HPV vaccination, on the date of recruitment
• Girl accompanied by an/other adult(s) (18+): Aged 9-14 years old; Female; Received a VYA health services supported by the program*, including HPV vaccination, on the date of recruitment
• Girl accompanied by an/other minor(s) only: Aged 9-14 years old; Female; Received a VYA health services supported by the program*, including HPV vaccination, on the date of recruitment
• Girl completely unaccompanied: Aged 9-14 years old; Female; Received a VYA health services supported by the program*, including HPV vaccination, on the date of recruitment
• Caregiver: Parent/legal guardian or caregiver of eligible 9-14-year-old girl (above); Accompanying eligible 9-14-year-old girl (above) during care/service visit; 18 years of age or older

Exclusion Criteria:

• Girl accompanied by adult (18+) parent/legal guardian: Does not provide informed assent; Parent/legal guardian does not provide consent; Requires urgent medical treatment or referral
• Girl accompanied by an/other adult(s) (18+): • Does not provide informed assent; Caregiver does not provide permission; Requires urgent medical treatment or referral
• Girl accompanied by an/other minor(s) only or completely unaccompanied: Does not provide informed assent; Requires urgent medical treatment or referral
• Caregiver: Does not provide informed consent

The primary aim of this study is to evaluate the impact of the intervention on delivery of HPV vaccinations to rural adolescent Ethiopian girls. No IPD will be used to evaluate the primary outcome for the effectiveness component of this study; rather, the effectiveness component will be assessed using aggregated service statistics data, which is inherently de-identified. These data will be made publicly available, as will other primary data collected in the study (quantitative exit survey data). Qualitative data, particularly from key informants, may be more difficult to completely de-identify; therefore, the investigators will make the decision during inspection of qualitative interview transcripts as to whether all qualitative data can be de-identified adequately to make pubicly available.