**1. Introduction**

The expansion of routine immunization is heralded as a global success story [1], enabling greater survival and improved child health worldwide [2]. Nevertheless, an estimated 25 million children were un- or under-vaccinated in 2021 [3], with many facing compounding barriers in vaccine access, availability, and demand. The ongoing COVID-19 pandemic has contributed to at least some of today's gaps in childhood vaccination [3], with estimates of under-one children without any doses of the diphtheria–tetanus–pertussis vaccine (no-DTP) rising from 10% prevalence in 2019 to 14% in 2021 [3]. Communities with high levels of unvaccinated or "zero-dose children" often face myriad vulnerabilities [4–7], such as residing in highly remote areas or informal settlements in cities [7–9]; being affected by displacement and/or prolonged conflict or unrest [7,8]; longstanding poverty and/or societal neglect [4]; or some constellation of these factors. Subsequently,

**Citation:** Fullman, N.; Correa, G.C.; Ikilezi, G.; Phillips, D.E.; Reynolds, H.W. Assessing Potential Exemplars in Reducing Zero-Dose Children: A Novel Approach for Identifying Positive Outliers in Decreasing National Levels and Geographic Inequalities in Unvaccinated Children. *Vaccines* **2023**, *11*, 647. https://doi.org/10.3390/ vaccines11030647

Academic Editors: Ahmad Reza Hosseinpoor, M. Carolina Danovaro, Devaki Nambiar, Aaron Wallace and Hope Johnson

Received: 2 February 2023 Revised: 9 March 2023 Accepted: 10 March 2023 Published: 14 March 2023

**Copyright:** © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

optimally identifying where and how to better reach zero-dose children will likely require a combination of context-specific strategies and broader investments to address persisting structural challenges.

Over the last few years, a growing body of research has sought to assess characteristics of zero-dose children and their families or households, as well as potential drivers of high zero-dose prevalence at different geographic levels [4–8,10–17]. Past work has found that zero-dose children experience a higher odds of missing or lacking access to other types of primary care services [6,11,12], while their mothers were more likely to have no antenatal care visits and not deliver at a health facility [11,12]. Lower levels of household wealth, educational attainment, and measures of women's empowerment also have been associated with higher levels of zero-dose children [4,10,16,17]. Gender-based inequalities, which span from differential rates of immunization by infant gender and gender-related barriers related to who can seek or provide vaccination services [16,18], emphasize the complex yet crucial role that gender plays in a country and/or community [19]. Prior studies have found ethnic disparities [15], as well as differences by religious affiliation [13,20], among children who have received no doses of DTP, though the exact nature of these relationships varied by country. Quantifying these risk factors and determinants of zero-dose children can provide critical program inputs, spanning from identifying key barriers to service access [7,18] to honing in on what sociocultural forces may be negatively affecting vaccine sentiments and trust [18–20]. However, exclusively focusing on zero-dose risk profiles and factors associated with higher rates of unvaccinated children may miss important lessons around successful approaches to addressing inequalities in childhood immunization. Accordingly, also understanding what has worked to improve childhood vaccination can inform program and policy adaptations tailored for reaching zero-dose children.

Positive outlier, or so-called 'positive deviance', methodologies have been used at the unit or organizational level in healthcare settings [21–23], as well as for more populationlevel contexts [24–28], to generate or strengthen the evidence base around what works to improve key health priorities. While the exact approaches toward this type of research and synthesis vary, they usually espouse a shared premise: knowledge and implementation strategies around achieving success or progress exist from places or contexts where such success or progress have been previously attained [21]. As a result, identifying and then examining what contributes to exceptional performance or progress can offer actionable insights into what policies and practice could be adapted for similar impact elsewhere. For instance, the *Good Health at Low Cost* case studies first in 1985 [26] and then in 2013 [27], sought to synthesize how and why countries or regions achieved substantial advances in several health indicators compared to their peers with similar income and demographic profiles; in 2018, the World Bank took a similar approach for understanding rapid progress on universal health coverage measures and facilitating shared learning opportunities across countries [28]. In 2016, the Global Burden of Disease study developed analyses to compare country-level performance on various health metrics relative to changes in sociodemographic development [29–31]; such findings emphasized that important health program and policy lessons could be learned from countries where achievements exceeded expected levels or trends on the basis of sociodemographic improvements alone. Lastly, the Exemplars in Global Health (EGH) program has sought to synthesize key lessons and strategies used by countries that attained exceptional progress in health—exemplars through mixed-methods research and engagement with partners [32–35]. As highlighted by past and current work on positive outliers, such analyses can foster opportunities for cross-country learning and exchange around successful policy or programmatic approaches for a given health challenge. With more learning agendas and priority-setting around zero-dose children for both national and global initiatives (e.g., Immunization Agenda 2030 [IA2030] [36] and Gavi 5.0 [37]), adopting a positive outlier lens toward country progress in reducing zero-dose children could further inform key immunization program and policy efforts.

With this study, we develop a novel approach for identifying positive outliers in reducing zero-dose children over time. This analysis currently takes a geographic focus, one of many important dimensions of inequality, by comparing patterns in both national and subnational declines in the percentage of under-one children with no doses of DTP (no-DTP) from 2000 to 2019 among 56 low- and lower-middle-income countries (LMICs). Based on this approach, identified 'exemplar' countries or subnational locations that substantially reduced zero-dose children could be targeted for further examination into the policy or program factors behind such gains.

### **2. Materials and Methods**

## *2.1. Data*

We used estimates of DTP1 among children under 1 year of age at the national and second administrative levels from the Institute for Health Metrics and Evaluation (IHME). The methods used to estimate DTP1 at different geospatial resolutions are detailed elsewhere [38]; in brief, DTP1 coverage estimates were derived from georeferenced household surveys and modeled using Bayesian geostatistical methods for 106 countries at the first and second administrative levels from 2000 to 2019. We opted to use these spatially modelled estimates over alternative sources (e.g., administrative data) to maximize both the potential number of countries included and comparability of estimates across locations. We subtracted DTP1 estimates from 100% to reflect the percentage of under-one children with no doses of DTP, or no-DTP prevalence—a commonly used indicator for zero-dose children [10,36].

For this analysis, we focused on 56 LMICs (Table 1). These countries were selected on the following criteria: (1) designation of low- or lower-middle income for fiscal year 2020 by the World Bank [39] or having received support from Gavi, the Vaccine Alliance as of 2018 [40]; (2) availability of both national and subnational no-DTP estimates at the second administrative level from 2000 to 2019; (3) not being classified as a post-transition middle-income country by Gavi and inclusion as part of Gavi's zero-dose segmentation country groups [41]. Supplementary Table S1 includes the full list of initially considered countries and those excluded from the current analysis.

**Table 1.** Included countries for identifying potential exemplars in reducing zero-dose children. \* Gavi-supported indicates that the country received Gavi support as of 2018 or had a dedicated country hub page. \*\* Countries with national and subnational DTP1 estimates (for both first and second administrative units) as modeled by the Institute for Health Metrics and Evaluation. Supplementary Table S1 provides the list of initial countries considered but excluded due to not meeting inclusion criteria.



#### **Table 1.** *Cont.*


**Table 1.** *Cont.*
