Geospatial variations in the prevalence of children missing out of diphtheria-pertussis-tetanus vaccination in Nigeria: estimates from multiple household surveys Authors: Faith Eshofonie, Ezra Gayawan Published: 2026 ABSTRACT BackgroundVaccination remains one of the most effective public health measures, yet many children in Nigeria continue to miss essential vaccines, leaving them exposed to preventable diseases. Understanding the geographic and temporal patterns of missed vaccinations is important for designing targeted and equitable immunization strategies. This study investigates the geospatial variations in the prevalence of children who missed diphtheria-pertussis-tetanus vaccination in Nigeria.MethodsData were obtained from the Nigeria Demographic and Health Surveys conducted between 2003 and 2023/2024. Bayesian geostatistical models were fitted within the Integrated Nested Laplace Approximation (INLA) framework to estimate the prevalence of zero-dose and under-immunized children across states and survey periods. Exceedance probability maps were used to identify states with a high likelihood that the prevalence of zero-dose children exceeded ten percent.ResultsThe results show clear geographic clustering, with persistently higher prevalence of missed vaccinations in the northern regions compared to the south. Dropout between consecutive DPT doses has decreased over time, indicating improved follow-through once children start vaccination; however, each successive dose still shows a higher prevalence of missed vaccinations than the preceding one. Exceedance probability maps further highlight states such as Kebbi, Sokoto, Zamfara, and parts of Niger, Kwara, Borno, Yobe, Taraba, and Kogi, where the likelihood of zero-dose prevalence surpassing 10% remains high.ConclusionDespite improvements in childhood vaccination coverage, large geographic disparities persist, especially in northern states. The continued presence of states with a high probability of zero-dose prevalence suggests that national progress has not resulted in equitable gains. Strengthening local immunization systems, addressing regional barriers, and prioritizing targeted interventions are needed to ensure that all children benefit from routine vaccination services.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s12889-026-28173-8. Introduction Vaccination remains one of the most cost-effective public health interventions, saving millions of lives annually and reducing morbidity and mortality from infectious diseases [1]. Building on the momentum of the smallpox eradication campaign, the World Health Organization (WHO) launched the Expanded Programme on Immunization (EPI) in 1974 to ensure that all children worldwide benefited from life-saving vaccines [2]. Since its inception, EPI has evolved to include a wider range of vaccines beyond the initial focus on six childhood vaccine-preventable diseases, which were Bacillus Calmette-Guérin (BCG), diphtheria, pertussis, tetanus (DPT), poliomyelitis, and measles, extending protection to older children, adolescents, and adults [2]. Today, thirteen vaccines are recommended under the EPI framework, including Haemophilus influenzae type B (Hib), hepatitis B (HepB), pneumococcal disease, rotavirus, human papillomavirus (HPV), and COVID-19 for adults. Through systematic expansion of vaccine delivery, EPI has addressed gaps in immunization coverage and strengthened health outcomes globally [2]. The World Health Organization’s Immunization Agenda 2030 (IA2030) sets the current global framework, emphasizing equitable access, life-course immunization, and reaching zero-dose and under-immunized populations, with the overarching goal of ensuring that everyone, everywhere, at every age, fully benefits from vaccines for good health and well-being [3–5]. Among routine childhood vaccines, the diphtheria–pertussis-tetanus (DPT) vaccine serves as a critical benchmark of immunization system performance, with coverage of the third dose (DPT3) widely recognized as an indicator of the strength and continuity of routine immunization programs. DPT3 coverage not only reflects initial contact with health services but also indicates the system’s capacity to maintain follow-up, manage logistics such as cold chains, and provide routine care consistently, which has been shown to reduce the global burden of infectious diseases [6]. Gavi defines a zero-dose child as one who has not received a single dose of a DPT-containing vaccine [7]. Children who receive no DPT dose (zero-dose children), or those who start but do not complete the full three-dose schedule, commonly referred to as “dropouts” or “under-immunized”, remain at high risk of vaccine-preventable diseases. Globally, millions of children remain zero-dose or under-immunized, with the majority concentrated in low- and middle-income countries [8]. These gaps highlight not only individual vulnerability but also broader challenges in health service access and program effectiveness [9]. Despite decades of global efforts to expand immunization, millions of children still miss out on life-saving vaccines. In 2024, an estimated 14.3 million children worldwide were zero-dose, having never received a single dose of DPT or any other routine vaccine [8]. In 2022, approximately 30.79 million children in the African Region received the first dose of DPT, compared with 30.44 million in 2021, representing about 350,000 additional children immunized. Coverage for the first and third doses of DPT was estimated at 80% and 72%, respectively, in 2022. DPT1 coverage declined from 83% in 2018–2019 to 81% in 2020, before plateauing at 80% in 2021–2022 [10]. With one of the highest global burdens of zero-dose and under-immunized children, Nigeria provides an essential basis for studying the spatial variations and changes in the pattern of zero-dose and incomplete vaccination coverage. Recent estimates suggest that Nigeria is home to over 2.1–2.3 million zero-dose children, the highest in Africa and the second highest globally [7, 11]. Studies show that although the prevalence of zero-dose children in Nigeria declined between 2003 and 2018 in many regions, rural areas and northern zones consistently experienced higher rates of incomplete vaccination compared to urban and southern zones [12]. Similarly, full vaccination coverage, considered as receiving all age-appropriate vaccines among young children, remained low in the country: 6.5% in 2003, 14.8% in 2008, 21.8% in 2013, and 23.3% in 2018 [13]. Distinguishing between zero-dose and dropout children is critical, as they reflect different types of health system failure. Zero-dose children often face barriers such as geographic isolation, caregiver education, poverty, social marginalization, and conflict, which prevent initial contact with immunization services. These challenges are frequently more pronounced in rural, remote, and conflict-affected areas. In contrast, dropout children may begin the vaccination schedule but fail to complete it due to supply-side constraints, including vaccine stockouts, inadequate health workforce, weak follow-up systems, or competing caregiver priorities [14]. Monitoring both indicators over time provides important insights into whether interventions are improving access, sustaining continuity of care, and reducing inequities across different regions. Achieving equitable immunization requires that vaccines are delivered to geographically, socially, or culturally marginalized populations, including displaced people, migrants, and those affected by conflict, political instability, or natural disasters. Geospatial approaches have become increasingly important for understanding inequalities in vaccination coverage and identifying underserved populations. Several studies have mapped vaccination coverage across Africa and within Nigeria, revealing clear geographic clustering and persistent disparities [15, 16], while some recent work has examined zero-dose prevalence at finer spatial scales [17, 18]. Many studies, however, focus on overall coverage or individual vaccines at a single survey wave, which can make it difficult to distinguish between failure to initiate vaccination and failure to complete the schedule. Evidence that simultaneously considers zero-dose prevalence, dropout between doses, and their spatial patterns across multiple survey waves remains limited. Understanding and mapping the distribution of missed vaccinations in Nigeria is therefore central to improving the reach and effectiveness of immunization programmes. Despite its potential to reveal inequalities and guide targeted interventions, the use of spatial evidence in health planning is often constrained by data gaps and methodological challenges [19]. Empirical studies in Africa have shown strong spatial clustering of vaccination coverage, with a clear north–south divide in Nigeria and strong links to poverty, remoteness, and aridity [15]. Between 2000 and 2016, DPT3 coverage rose in about 72% of second-level administrative units across Africa, but large within-country inequalities remained. By 2016, only Morocco and Rwanda had achieved at least 80% DPT3 coverage in all districts, while wide low-coverage zones (≤50%) persisted in the Sahel, Somalia, eastern Ethiopia, and Angola, largely driven by low DPT1 uptake and high dropout [16]. This study models the prevalence of children missing out on the DPT vaccine in Nigeria using Bayesian geostatistical models with data from five waves of the Nigeria Demographic and Health Survey (NDHS) conducted in years 2003, 2008, 2013, 2018, and 2023/2024. The data come from clusters selected randomly across the country during each wave of the survey and thus, it is essential to predict events throughout the country based on geostatistical models that use information from neighbouring locations to improve local estimates. Consequently, the main objective of the study is to produce spatial maps that reflect vaccination across the country and to track temporal trends in vaccination coverage to guide targeted immunization interventions. To assess vaccination progress, two complementary measures were considered: first, differences between consecutive doses were analyzed to identify where children were most likely to drop out after initiating vaccination, and second, coverage for each dose was compared across survey years to track changes over time. Although factors associated with low coverage are discussed to provide context, the main emphasis is on mapping and prediction rather than identifying associated factors. Exceedance probability maps were also produced to highlight areas where zero-dose prevalence is likely to fall below global targets, helping to pinpoint regions requiring intensified immunization efforts. Materials and method Data source The NDHS are nationally representative household surveys designed to provide information on fertility, maternal and child health, and related demographic indicators among others. Each survey employs a stratified, two-stage cluster sampling design based on the National Population Commission’s Enumeration Areas (EAs), stratified by state and urban–rural residence. During the 2003 survey, 365 clusters (165 urban and 200 rural) were selected from the 1991 census frame [20]. This was expanded to 888 clusters (286 urban and 602 rural) drawn from the 2006 census frame during the 2008 survey [21]. A total of 904 clusters were considered for the 2013 survey [22], while the 2018 survey sampled 1,400 clusters, 663 urban and 737 rural [23]. The 2023/2024 NDHS [24] used an updated digital sampling frame prepared for the planned first fully digital Population and Housing Census in Nigeria. A total of 1,400 clusters (701 urban and 699 rural) were selected from 74 strata across the 36 states and the Federal Capital Territory. Fieldwork was successfully completed in 1,380 clusters, with 20 clusters omitted due to insecurity in parts of Zamfara, Katsina, Borno, Kaduna, Niger, Benue, and Imo states. The study used data from children aged 12–23 months, the recommended age to complete routine childhood vaccinations. Vaccination outcomes were defined as follows: zero-dose children were those who had not received the first dose of DPT (DPT1), while dropout or under-immunized children were those who received an earlier dose (DPT1 or DPT2) but did not complete the full three-dose schedule (DPT2 or DPT3). To visualize the raw data, cluster-level bubble maps of DPT vaccination outcomes (DPT1–DPT3) were generated for each survey year, with each bubble representing a survey cluster and sized according to the observed proportion of unvaccinated children (Supplementary Figs. S.1–S.3). Statistical model A Bayesian spatio-temporal geostatistical model was fitted to estimate dose-specific prevalence of being unvaccinated. Let \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$y_{ijt}(h)$$\end{document} denote a binary indicator for child i in cluster j, survey year t, and dose \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$h \in \{1,2,3\}$$\end{document}, where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$y_{ijt}(h)=1$$\end{document} indicates that the child was unvaccinated for dose h. The outcome is modelled as:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} y_{ijt}(h) \sim \text {Bernoulli}(p_{ijt}(h)) \end{aligned}$$\end{document}with logit link:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} \text {logit}(p_{ijt}(h)) = \beta _0 + w(s_j,t) \end{aligned}$$\end{document}where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$w(s_j,t)$$\end{document} is a latent spatio-temporal Gaussian field capturing structured spatial dependence between nearby locations and temporal correlation across survey years. Spatial variation is modelled using a Gaussian random field with Matérn covariance structure, and temporal dependence is captured using an autoregressive (AR(1)) process [25, 26]. The model was estimated within a Bayesian framework using Integrated Nested Laplace Approximation (INLA). Posterior predictions were generated at observed cluster locations and projected onto a regular spatial grid to produce continuous maps of dose-specific prevalence of being unvaccinated across Nigeria. Vaccination progress was assessed using differences in posterior predicted prevalence across time and across vaccine doses, and exceedance probability was defined as the posterior probability that zero-dose prevalence exceeds 10%, used to identify areas with high levels of zero-dose prevalence for targeted immunization interventions. Details of the model specification, prior choices, validation procedures, and sensitivity analyses are provided in the Supplementary File. Model validation Model predictive performance was assessed using a spatial holdout validation approach, in which selected states were excluded from model fitting and used for out-of-sample prediction. Model accuracy was evaluated using root mean squared error (RMSE) and the correlation between observed and predicted prevalence across the held-out locations. The model demonstrated strong predictive performance across all DPT doses, with high correlation and low prediction error, indicating good agreement between observed and predicted values. Visual inspection of observed versus predicted values further confirmed the model’s ability to capture spatial patterns in vaccination coverage (Fig. 1).Fig. 1Scatter plots of observed versus predicted prevalence of DPT1, DPT2, and DPT3 for selected states that were used as holdout validation Scatter plots of observed versus predicted prevalence of DPT1, DPT2, and DPT3 for selected states that were used as holdout validation Results State-level trends from raw survey data Overall, vaccination coverage improved across Nigeria from 2003 to 2023/2024, but progress was uneven. Southern states and parts of the North Central region showed steady reductions in zero-dose and under-immunized children, while a few northern and southwestern states experienced slower improvements or recent increases. Figure 2 illustrate these patterns.Fig. 2State-level proportion of vaccination outcomes across Nigerian states by survey year, with panels for DPT1 (zero-dose), DPT2, and DPT3 State-level proportion of vaccination outcomes across Nigerian states by survey year, with panels for DPT1 (zero-dose), DPT2, and DPT3 Spatial distribution of missed DPT vaccinations Zero-dose prevalence was consistently highest in the North West and parts of the North East throughout the survey period, exceeding 75% in some areas in 2003, and gradually declining by 2023/2024 (Fig. 3). DPT2 and DPT3 follow similar spatial patterns, with coverage improving over time but completion of DPT3 remaining challenging in some northern states (Figs. 4–5). The associated uncertainty estimates for each dose are provided in the Supplementary File (Supplementary Figs. S4–S6).Fig. 3Prevalence of zero-dose children in Nigeria (Zero-dose is proxied by those who have not received the first dose of DPT vaccine) Prevalence of zero-dose children in Nigeria (Zero-dose is proxied by those who have not received the first dose of DPT vaccine) Fig. 4Prevalence of children who have not received the second dose of DPT vaccine (under-immunized) Prevalence of children who have not received the second dose of DPT vaccine (under-immunized) Fig. 5Prevalence of children who have not received the third dose of DPT vaccine (under-immunized) Prevalence of children who have not received the third dose of DPT vaccine (under-immunized) Trends in vaccination dropout within doses Across survey years, dropout within individual doses showed both improvements and setbacks. For zero-dose (DPT1), northern states such as Borno, Yobe, and Bauchi saw fluctuations, whereas central states (Kaduna, FCT, Plateau) generally improved. DPT2 and DPT3 follow similar trends, although dropout was higher for DPT3, indicating that completion of the full series remains more challenging (Figs. 6, 7 and 8).Fig. 6Changes in the prevalence of zero-dose between time periods Changes in the prevalence of zero-dose between time periods Fig. 7Changes in the prevalence of children missing out of the second dose of DPT between time periods Changes in the prevalence of children missing out of the second dose of DPT between time periods Fig. 8Changes in the prevalence of children missing out of the third dose of DPT between time periods Changes in the prevalence of children missing out of the third dose of DPT between time periods Trends in vaccination dropout across doses Dropout between consecutive doses (DPT1–DPT2, DPT2–DPT3) decreased nationally from 2003 to 2023/2024. For DPT1–DPT2, average dropout fell from 0.121 to 0.045, and for DPT2–DPT3, from 0.109 to 0.067 (Table 1). Early dropout hotspots included northern and central states, while southern states showed consistently lower dropout. By 2023/2024, completion improved across most regions, though later doses remained more challenging than initiation. Figures 9–10 present these patterns.Table 1Summary statistics of differences between consecutive DPT vaccination doses across survey yearsYearDPT1 - DPT2DPT2 - DPT3MeanMinMaxMeanMinMax20030.121−0.2130.4880.109−0.0820.61220080.079−0.1200.4570.096−0.0910.47120130.059−0.0810.3500.077−0.0990.36620180.076−0.0780.4390.088−0.0710.5462023/20240.045−0.1250.3270.067−0.1810.404 Table 1Summary statistics of differences between consecutive DPT vaccination doses across survey yearsYearDPT1 - DPT2DPT2 - DPT3MeanMinMaxMeanMinMax20030.121−0.2130.4880.109−0.0820.61220080.079−0.1200.4570.096−0.0910.47120130.059−0.0810.3500.077−0.0990.36620180.076−0.0780.4390.088−0.0710.5462023/20240.045−0.1250.3270.067−0.1810.404 Summary statistics of differences between consecutive DPT vaccination doses across survey years Fig. 9Differences in coverage between DPT1 and DPT2 for each survey year Differences in coverage between DPT1 and DPT2 for each survey year Fig. 10Differences in coverage between DPT2 and DPT3 for each survey year Differences in coverage between DPT2 and DPT3 for each survey year Exceedance of zero-dose prevalence Persistent hotspots exceeding the 10% zero-dose threshold were concentrated in the North West, parts of the North East, and selected central states (Kebbi, Sokoto, Zamfara, Niger, Kwara, Borno, Yobe, Taraba, Kogi) throughout 2003–2023/2024 (Fig. 11). Despite national improvements, these areas indicate ongoing challenges in reaching children with the first dose.Fig. 11The exceedance probability for zero-dose children (threshold = 10%) The exceedance probability for zero-dose children (threshold = 10%) Sensitivity analysis Sensitivity analyses comparing alternative model specifications confirm the robustness of these hotspots (Figs. 12–13). Positive differences (red) indicate locations where the alternative model predicted higher probabilities than the baseline, while negative differences (blue) indicate lower probabilities.Fig. 12Difference in exceedance probabilities for zero-dose DPT prevalence above 10% between the IID temporal model and the baseline AR(1) model Difference in exceedance probabilities for zero-dose DPT prevalence above 10% between the IID temporal model and the baseline AR(1) model Fig. 13Difference in exceedance probabilities for zero-dose DPT prevalence above 10% between the prior-relaxed model and the baseline AR(1) model Difference in exceedance probabilities for zero-dose DPT prevalence above 10% between the prior-relaxed model and the baseline AR(1) model Compared with the baseline model, exceedance probabilities under the IID specification remained strongly correlated (r = 0.94), indicating that the overall spatial pattern of high-risk areas was largely preserved despite removing temporal dependence. Using an 80% exceedance threshold to define high-risk areas, approximately 10% of locations changed classification under the IID model. Prior relaxation produced virtually identical exceedance probabilities to the baseline (r = 0.99), with only 0.03% of locations reclassified. The results show that the regions at the highest risk of zero-dose DPT1 are consistently identified. Removing temporal dependence caused only minor changes, and relaxing the model priors had almost no effect, showing that the overall pattern of high-risk areas is reliable. Discussion Despite decades of routine childhood immunisation efforts, uneven immunisation coverage persists in Nigeria, particularly in the North-West and North-East regions. National estimates suggest that DTP1 and DTP3 coverage reached 70% and 62% respectively in 2023 [27, 28], indicating improvement compared with earlier years. However, subnational estimates from this study show that many of the same states that fared poorly in 2003 continue to lag behind in 2024, indicating that progress has not been evenly distributed. The prevalence of zero-dose children remains particularly high in northern states, especially in the North West and North East. Data from the 2018 Nigeria Demographic and Health Survey show that full immunisation coverage was lowest in the North West (13%) and North East (18%), compared with southern regions, with maternal education, household wealth, and region of residence influencing immunisation status [29]. More recent trends from 2018 to 2024 indicate a sharp rise in zero-dose prevalence in the North West, from 25.7% in 2021 to 47.4% in 2024, while South-Eastern states maintained coverage around 90% [30]. These patterns highlight persistent inequities across regions. Several factors likely contribute to low vaccination uptake in the North, including socioeconomic disadvantage, limited access to health facilities, logistical challenges, weak reminder systems, security concerns, low health literacy, and cultural factors [31]. Addressing these barriers will require targeted interventions, such as promoting maternal education, improving access to health services, implementing outreach strategies sensitive to local contexts, and strengthening routine immunisation systems in high-risk areas. Dropout between consecutive doses, calculated from differences in predicted prevalence for DPT1, DPT2, and DPT3, reveals that the primary challenge lies not in starting vaccination but in ensuring children complete the series. Although children make initial contact with the health system, maintaining follow-through for subsequent doses remains difficult. Structural barriers such as recurring vaccine stock-outs, limited cold-chain capacity, and weak follow-up systems further impede completion [32]. As a result, children may receive the first dose but fail to progress, reducing the overall effectiveness of the program. Beyond logistical issues, factors including caregiver awareness, trust in health services, and clinic accessibility influence completion, especially in rural settings. In northern Nigeria, additional obstacles such as mobility restrictions, vaccination myths, gender-related decision constraints, and limited confidence in providers further depress uptake [33], while negative experiences with health workers, unclear appointment records, and fear of side effects discourage continued participation. Together, these challenges help explain why some northern states persistently exceed the 10 percent zero-dose threshold and continue to experience high dropout. While prevalence maps indicate where vaccination coverage is low, they do not capture the certainty that a state exceeds a critical threshold. In our study, exceedance probability maps complement prevalence estimates by highlighting areas where zero-dose prevalence is likely to surpass 10%. Northern states consistently exceed this threshold. By quantifying uncertainty, this probabilistic approach distinguishes states near the threshold from those where zero-dose prevalence is almost certainly high. This helps policymakers identify true hotspots. Combining prevalence and exceedance maps enables programmatic planning to focus on areas where social barriers, dropout, and systemic challenges sustain under-immunization. The policy implications are as follows. First, Nigeria’s immunisation strategy should shift from a uniform national approach to a geographically differentiated model, with high-risk northern states receiving intensified outreach, flexible service delivery, and culturally sensitive community engagement. Second, focus must extend beyond initiation to completion of the DPT vaccination series through strengthened follow-up, improved vaccine stock management, and integration with maternal and child health services [32, 34]. Third, spatially informed monitoring tools, such as exceedance probability maps, should guide programmatic decisions to ensure that resources target areas with the highest likelihood of persistent under-coverage. Limitations include reliance on household survey data, where vaccination status is partly based on caregiver recall in the absence of vaccination cards, which may introduce recall bias. In addition, the geographic coordinates of DHS clusters are randomly displaced, which can affect the precision of spatial estimates, particularly at finer geographic scales. The analysis focused on spatial patterns and did not adjust for socioeconomic or demographic differences across states, such as maternal education, household wealth, or place of residence, which are known to influence vaccination uptake. Important geospatial variables, such as distance or travel time to health facilities, are not directly available in DHS data and were therefore not included. While DHS provides self-reported measures of access to care, these do not capture actual geographic accessibility. Factors related to service delivery, such as vaccine stock availability, staffing, and outreach activities, were also not captured in the data. As a result, some variation in the estimates may reflect influences that were not measured. Future research should combine household survey data with external geospatial data on healthcare access and local service delivery, such as mobile vaccination campaigns and community outreach initiatives, to better evaluate how these factors influence dropout and improve equitable coverage [34–36]. Conclusions National vaccination averages can hide states where under-immunization persists. The persistence of zero-dose and dropout clusters reflects geographic disadvantages, social inequalities, fragile health services, and limited trust. Effective strategies must prioritize equity, target persistent hotspots, and address both initiation and completion of vaccination schedules to achieve meaningful improvements in childhood immunization coverage in Nigeria [37, 38]. Supplementary Information Supplementary Material 1. Supplementary Material 1.