Evaluating the Impact of Demand-Side Incentives (DSI) on Zero Dose Burden and Immunisation Coverage in Nigeria

BACKGROUND Immunisation prevents vaccine-preventable diseases and deaths, yet many children remain unvaccinated or under-immunised. According to the WHO/UNICEF Estimates of National Immunisation Coverage (WUENIC), approximately 25 million children missed routine vaccinations globally in 2021, including 18.2 million zero-dose children. More than 60% of these zero-dose children lived in ten countries, including Nigeria. Global coverage of the third dose of diphtheria-tetanus-pertussis vaccine (DTP3) declined from 86% in 2019 to 81% in 2021, reflecting disruptions associated with conflict, misinformation, the COVID-19 pandemic, and constraints in immunisation service delivery. The 2022 WUENIC estimates indicated a subsequent recovery in vaccination coverage.

Nigeria accounts for a substantial share of the global zero-dose burden. Estimates of DTP1 coverage differ across data sources: 68% in the 2022 Multiple Indicator Cluster Survey, 70% in the 2021 WUENIC estimates, and 90.95% in 2022 routine DHIS2 reporting. These differences highlight challenges in measuring immunisation coverage and identifying underserved populations. Geographic disparities persist across states, local government areas (LGAs), communities, and health facilities. Key barriers include caregiver knowledge and misconceptions, maternal education, limited access to health facilities, poverty, conflict, and gender-related constraints. Administrative coverage estimates can also be affected by outdated population denominators derived from the 2006 census.

Nigeria has implemented strategies to strengthen routine immunisation, including the Reach Every Ward (REW) strategy, adopted in 2005, and the Optimised Integrated Routine Immunisation Sessions (OIRIS) approach. These emphasise improving service access, outreach, supervision, microplanning, community engagement, accountability, and the use of data for action. Nevertheless, barriers to immunisation demand and service utilisation remain, particularly in poor, remote, and otherwise underserved communities.

Gavi, in partnership with Corona Management Systems, is implementing the Demand-Side Incentives (DSI) to Address Immunisation Challenges project in Cameroon, the Democratic Republic of Congo (DRC), Lesotho, and Nigeria. The project seeks to improve immunisation coverage, reach marginalised communities, promote timely vaccination, and strengthen caregivers' knowledge and perceptions of vaccination.

RATIONALE Zero-dose children are disproportionately concentrated in disadvantaged settings. The Equity Reference Group for Immunisation has highlighted conflict-affected areas, urban poor populations, remote rural communities, and gender-related barriers as priority equity concerns. Gavi's 2021-2025 strategy also emphasised reaching missed communities and strengthening equitable immunisation systems. Demand-side incentives, including cash and non-cash approaches, may help address financial and behavioural barriers to the uptake of health services, including vaccination. Evidence from previous implementations is relevant, but the appropriateness, effectiveness, and sustainability of incentive schemes must be assessed within local contexts.

Nigeria was selected based on its high zero-dose burden, low and uneven immunisation coverage, equity challenges, previous experience with DSI, and stakeholder interest in further implementation. Previous initiatives include conditional cash transfers linked to routine immunisation through New Incentives and the Subsidy Reinvestment and Empowerment Programme (SURE-P). Selection also considered the implementing team's country presence, implementation experience, and knowledge of local socioeconomic conditions. Corona Management Systems will collaborate with the National Primary Health Care Development Agency and relevant partners to adapt the intervention to national priorities and complement existing immunisation activities.

Within Nigeria, Oyo and Kwara states were selected following a feasibility assessment. The study will assess the effects of DSI interventions using a cluster randomised controlled trial (cRCT), with political wards serving as the units of randomisation.

OBJECTIVES OF THE STUDY

The objective of this cluster Randomised Controlled Trial (cRCT) is to evaluate the impact and effectiveness of demand-side incentives (DSI) in improving immunisation coverage and reducing the burden of zero-dose children in Nigeria. The study aligns with the Immunisation Agenda 2030, the Gavi 5.0 strategy, and the country's priorities in strengthening the immunisation programme. The specific objectives of this cRCT include:

• Assess the impact of demand-side incentives on increasing immunization coverage in the selected subnational regions of Nigeria, by December 2025.
• Assess the effect of demand-side incentives on reducing the zero-dose burden in selected subnational regions in Nigeria, by December 2025.
• Examine the effects of various demand-side incentive schemes on improving the timeliness of vaccination among children aged 0 to 23 months in the selected subnational regions of Nigeria, by December 2025.
• Explore the impact of different demand-side incentive schemes on enhancing the immunization knowledge and perception of caregivers with children aged 0 to 23 months in the selected subnational regions Nigeria by December 2025.

SCOPE OF THE STUDY Thematic Scope

The cluster Randomised Controlled Trial (cRCT) aims to evaluate the impact and effectiveness of demand-side incentives (DSI) in improving immunisation coverage and reducing the burden of zero-dose children in Nigeria. Key areas of focus for the study include:

1. A review of existing demand-side incentives and their impact on immunisation coverage and equity.
2. Behavioural and social drivers of vaccination uptake and vaccination service utilisation by the target population.
3. Structural determinants of zero-dose children in different settings (e.g., distant rural areas, urban poor settings, conflict-affected regions, areas with gender-related barriers).
4. Identification of primary challenges and opportunities through a baseline evaluation to inform the design and implementation of the Demand-Side Incentive (DSI) Programme to strengthen immunisation programmes; and
5. Evaluation of the impact of demand-side incentives on vaccination coverage, reduction of zero-dose children, improvement of vaccination service timeliness and shifts in behavioural and social drivers of vaccination uptake.

Geographic Scope The study will be conducted in selected Oyo and Kwara State in Nigeria. These areas were identified during the feasibility assessment based on the following criteria: low vaccination (low Penta 3 coverage rate), high burden of zero-dose children (low Penta 1 coverage rate or absolute number of zero-dose children), the absence of any current demand-side incentives intervention, and sustained stakeholder buy-in. The study will specifically target poor and underserved communities, including those in remote and hard-to-reach areas.

Chronological Scope The study will run from August 2024 till December 2025 and will enrol children eligible for routine immunisation in the intervention locations.

METHODOLOGY Study Type This study will employ a pragmatic, parallel cluster Randomised Controlled Trial (cRCT) design to evaluate the impact of the Demand-Side Incentives (DSI) on zero-dose burden and immunisation coverage in intervention LGAs in Nigeria. A pragmatic, parallel cRCT design is a type of research design that aims to evaluate the effectiveness of an intervention in real-world settings, providing insights into how the intervention works in routine practice.

Study Design The study will employ a cRCT design, focusing on the smallest administrative unit in Nigeria, known as a "political ward" or "ward," each with an estimated population of 10,000. These wards will serve as the sub-district level for the study, with each selected ward considered a cluster. The study is structured as a four-arm cRCT, with a total of 480 clusters (120 per arm). These clusters will be randomly divided into four groups: three intervention arms (360 clusters) and one control arm (120 clusters).

Enrolment will occur at the cluster level, meaning all children aged 0-23 months living in a selected cluster will be allocated to the same arm as the cluster. The study will be implemented across an estimated 49 Local Government Areas (LGAs), each typically including an average of ten wards, in selected sub geographies of Nigeria.

The study locations have been chosen for their low immunisation coverage, thereby focusing on areas with a high number of children receiving zero doses of essential vaccines. The sample size has been calculated to enable impact analysis on Zero Dose in these regions.

The interventions for each arm were selected based on findings from a feasibility evaluation conducted with national and sub-national stakeholders. This evaluation considered the local context of immunisation challenges and identified incentives that have broad stakeholder support across the Expanded Programme on Immunization (EPI) and the community.

Study Area and Population The study will be conducted in selected wards within identified Local Government Areas (LGAs) in Oyo and Kwara states of Nigeria. These areas have been chosen due to their immunisation coverage challenges and low performance indicators. The target population for this study will include key stakeholders involved in designing and executing immunisation programme strategies, community-level participants, and caregivers of infants aged 0 to 23 months.

Sampling Strategy

To ascertain the optimal sample size for each cluster (ward) to detect a 15% effect size, the WebPower package was employed within the R programming language. This package facilitated the computation of statistical power analysis for Cluster Randomized Trials (cRCT) with two arms-specifically, comparing the control arm against each implementation arm. The calculation considered the specified effect size, the number of subjects per cluster, intra-class correlation, significance level, and power.

Using an Intra-Class Correlation (ICC) of 0.16, based on relevant literature, and adopting a 95% confidence level, various sample sizes were stimulate. The results depicted revealed that over 80% power could be achieved with sample sizes ranging from 100 to 400. Considering practical implementation feasibility, a sample size of 100 was opted for with 20% margin in case of drop out, making it 120.