Spatial variation and determinants of hepatitis B birth dose vaccination in Nigeria: a population-based study

Abstract

Background

Hepatitis B virus (HBV) infection remains a major public health problem in Nigeria, with an estimated prevalence of 5–10%, with a sub-Saharan African regional average of approximately 6.1%. Timely administration of the hepatitis B birth dose (HepB-BD) is essential for preventing vertical transmission and its long-term complications; however, national coverage remains below recommended targets. Despite its importance, evidence on the geographic distribution and contextual determinants of HepB-BD uptake across Nigerian states is limited. This study examined the spatial variation and determinants of HepB-BD vaccination in Nigeria.

Methodology

A population-based cross-sectional study that analysed data from the 2018 Nigeria Demographic and Health Survey. Spatial analysis using Local Moran’s I was performed to detect clusters of high and low HepB-BD vaccination coverage across Nigeria. Using STATA 14, binary logistic regression was used to analyse data of 13,064 children aged 0–24 months to identify predictors of Hep-B-BD vaccination.

Results

Overall, 52% of children received HepB-BD, with significant regional variation. Children of mothers with secondary or higher education, compared with those with no education, had higher odds of vaccination (aOR: 1.92; 95% CI: 1.63–2.27). Similarly, children from the richest households, relative to the poorest, had higher odds of vaccination (aOR: 2.46; 95% CI: 1.88–3.23). Antenatal care attendance, compared with none (aOR: 2.76; 95% CI: 2.39–3.18), and health facility delivery, compared with home delivery (aOR: 2.09; 95% CI: 1.83–2.38), were also associated with higher odds of vaccination. Children in the North-East (aOR: 0.68; 95% CI: 0.55–0.84) and North-West (aOR: 0.64; 95% CI: 0.51–0.80), compared with North Central, had reduced odds. Spatial analysis revealed significant clustering, with cold spots concentrated in northern Nigeria, and global spatial autocorrelation confirmed this (Moran’s I = 0.220, p = 0.001).

Conclusion

HepB-BD coverage remains suboptimal and geographically clustered, with disparities associated with maternal, socioeconomic, and regional factors. Targeted, context-specific strategies may be required to improve equitable uptake across low-coverage states.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-026-27362-9.

Introduction

Hepatitis B virus (HBV) infection is a disease endemic to sub-Saharan Africa, where the average prevalence is estimated at 6.1% [1]. The West African subregion is considered a high-endemic area, with adult HBsAg prevalence typically around 8–10% in regional meta-analyses [2]. Notably, in Nigeria, the prevalence similarly ranges from approximately 5–10% [3–6]. HBV infection leads to significant morbidity and mortality. Many individuals who are chronic carriers of the hepatitis B infection eventually progress to end-stage liver failure and hepatocellular carcinoma [7]. In view of the significant burden of HBV infection in sub-Saharan Africa, the World Health Organization (WHO) recommends vaccination as the primary preventive strategy [8]. Currently, HBV vaccination has been integrated into the National Immunization Programme in approximately one-third of African countries, including Nigeria [9]. The importance of hepatitis B vaccination is underscored by its administration at birth, making it one of the first immunizations given to children. This early administration is crucial not only for providing protection against potential exposure to HBV but also because infants infected at this stage have a substantially higher risk of developing chronic carrier status and liver fibrosis later in life [10]. Studies have shown that about 90% of perinatally infected infants may develop chronic HBV infection, compared to less than 10% of infected adults who typically clear the virus and develop lifelong immunity [11]. Furthermore, the high endemicity of HBV significantly increases the likelihood of vertical transmission from mother to child, highlighting the critical need for early vaccination. With 70.9% of adults in Nigeria reportedly having poor knowledge of HBV [12], prioritizing the protection of infants born to women with potential HBV infection is imperative, both as a public health intervention and as a strategic measure to safeguard future generations against these preventable communicable diseases.

Despite the inclusion of the Hepatitis B birth dose (HepB-BD) in Nigeria’s Expanded Programme on Immunization (EPI), coverage remains suboptimal. According to the World Health Organization (WHO) and UNICEF Estimates of National Immunization Coverage (WUENIC) report, the national coverage of the Hepatitis B birth dose in Nigeria was estimated to be 52% as of 2023 [13]. Similarly, a study conducted in Benue State, Nigeria, reported coverage rates of 63% among infants born to HepB-positive mothers and 64% among those born to HepB-negative mothers, suggesting that maternal Hepatitis B status does not significantly influence the likelihood of vaccine administration at birth, despite the increased risk to the infant [14]. Several maternal and health system factors have been identified as determinants of receiving the HepB-BD, including younger maternal age, facility delivery, higher maternal educational status, antenatal care (ANC) attendance, previous delivery, and the strength of healthcare systems [14, 15]. Receiving the birth dose of HBV vaccine reflects a complex interplay of these factors, as shown by previous studies [14, 15]. However, the geographic variation in HepB-BD vaccination coverage across Nigeria remains understudied. Considering the substantial diversity in healthcare infrastructure, access to services, and maternal characteristics across states and regions, the national immunization coverage rate often masks subnational gaps. Assessing the spatial pattern of HepB-BD vaccination is therefore essential, as spatial analysis can identify clusters of low or high coverage, highlight regional disparities, and inform targeted, context-specific interventions. Spatial distribution further strengthens this approach by revealing geographic patterns that may not be evident from statistical models alone [16, 17]. While existing literature provides insight into determinants of birth dose uptake, it does not address how these factors relate to geographic and contextual differences. Therefore, this study aims to assess the spatial patterns and determinants of HepB-BD vaccination in Nigeria, identify key predictors, and examine regional disparities. The findings can provide critical evidence to inform national and subnational strategies for improving vaccination uptake and access across regions.

Methodology

We conducted a secondary analysis of the Nigeria Demographic and Health Survey (NDHS), 2018. The NDHS is a nationally representative, population-based survey implemented by the National Population Commission (NPC) in collaboration with ICF International. It employs a stratified, two-stage cluster sampling design to collect information on demographic and health indicators, including maternal and child health, fertility, family planning, and immunization coverage. The NDHS employs a cross-sectional study design and data were obtained through structured household interviews using standardized questionnaires. The 2018 NDHS covered all 36 states and the Federal Capital Territory, enabling disaggregated analyses by geopolitical zone, state, and place of residence. In addition to the NDHS representativeness at the national level, the data are also representative at lower administrative levels, specifically, at urban-rural and state levels [18]. The NDHS data, along with the geographic coordinates which are made available on special request, allow for exploration of outcomes of interest by geographic locations specifically at state and urban-rural dichotomy levels. In the NDHS, information on child immunization were collected for children who were alive at the time of the survey. For our analysis, we used the Kids Recode data file. This study included a total 13,064 children representing weighted sample of Nigerian children aged 0–24 months.

Outcome variable

The outcome variable was receipt of the HepB-BD vaccination within 24 h of birth, in line with the World Health Organization recommendation [19]. In the NDHS data, the variable h50 (“received hepatitis B at birth”) assessed whether the child received HepB-BD. It had five response options: no (0), vaccination date on card (1), reported by mother (2), vaccination marked on card (3), and don’t know (98). For this study, HepB-BD vaccination status was recoded into a binary variable: vaccinated (1) and not vaccinated (0). The categories “vaccination date on card,” “reported by mother,” and “vaccination marked on card” were collapsed as “vaccinated,” while “no” and “don’t know” were recoded as “not vaccinated.” This approach is consistent with commonly used Demographic and Health Surveys (DHS) coverage analyses among children and has been applied in the literature [16, 20, 21]. In the primary analysis, uncertain responses (“don’t know”) were conservatively treated as non-receipt because their frequency was small. The proportion of “don’t know” responses in the analytic sample (children aged 0–24 months) was 14/13,064 (0.11%), which is below commonly used thresholds for creating a separate category. To further assess the robustness of this coding decision, a sensitivity analysis using multivariable logistic regression excluding “don’t know” responses was also conducted.

Independent variables

Variables included in this study as predictors of HepB-BD uptake, as well as their measurements, were guided by existing literature [14–17, 22], as presented in Table 1. In addition, household wealth quintiles constructed by DHS using an asset-based wealth index represent the relative socioeconomic position of households within the national survey sample.

Table 1.

Independent variables and measurement approach

Independent variables	NDHS measurement	Study measurements
Child’s sex	Male 1, Female 2	Male = 1, Female = 2
Parity	Numerically measured	1–4 = 1, 5 and above = 2
Mother’s age	Numerically measured	15–24 1, 25–34 2, 35 and above = 3
Mother’s marital status	Never in union = 0, married = 1, living with partner = 2, widowed = 3, divorced = 4, no longer living with partner/separated = 5	Married/living with partner = 1, Others = 2
Religion	Catholic = 1, other Christian = 2, Islam = 3, Traditionalist = 4, others = 96	Christians = 1, Islam and others = 2
Mother’s education attainment	No education = 0, Primary = 1, Secondary = 2, Higher = 3	No education = 0, Primary = 1, Secondary/higher = 2
Household wealth status	Poorest = 1, Poorer = 2, Middle = 3, Richer = 4, Richest = 5	Poorest = 1, Poorer = 2, Middle = 3, Richer = 4, Richest = 5
Place of residence	Urban = 1, Rural = 2	Urban = 1, Rural = 2
Region	North Central = 1, North East = 2, North West = 3, South East = 4, South South = 5, South West = 6	North Central = 1, North East = 2, North West = 3, South East = 4, South South = 5, South West = 6
Media exposure	Frequency of access to Newspaper or magazine/Radio/Television	Composite variable: Not exposed = 0, Exposed = 1 i.e. Access to any of the media
ANC reception	Number of ante-natal visits	Did not receive ANC = 0, Received ANC = 1 i.e. visits > 0
Distance to health facility	Big problem = 1, Not a big problem = 2	A big problem = 1, Not a big problem = 2
Place of delivery	Respondent’s home = 11, Other home = 12, government hospital = 21, government health centre = 22, government health post = 23, other public sector = 26, private hospital/clinic = 31, other private sector = 36, other = 96	Home = 1, Health facility = 2 i.e. government hospital = 21, government health centre = 22, government health post = 23, other public sector = 26, private hospital/clinic = 31, other private sector = 36

Statistical analysis

Data analysis was conducted using Stata Statistical Software: Release 14 (College Station, TX: StataCorp LP). All analyses incorporated DHS sampling weights, clustering, and stratification using the svyset command in Stata to account for the complex survey design. Univariable analysis was performed to describe respondents’ characteristics. At the bivariable level, cross-tabulations of the outcome variable with each background characteristic were conducted, and associations were assessed using the chi-square test.

At multivariable level, binary logistic regression analysis was conducted to examine predictors of HepB-BD vaccination considering the binary nature of the outcome variable. All theoretically relevant covariates identified from the literature were included simultaneously in the multivariable model. Adjusted odds ratios (aORs) and 95% confidence intervals (CIs) were used to assess the strength of associations. A significance level of 5% was used to determine statistical significance. Variance inflation factors (VIF) were computed to assess multicollinearity, with a cut-off value of 5 [23]. Observations with missing values on covariates were minimal and were handled using complete-case analysis. In addition, state-level proportions of vaccinated children were extracted and exported for spatial analysis.

Spatial analysis

This study assessed the clustering of HepB-BD coverage in an effort to identify hotspots and cold spots of such coverage across Nigerian states. Administrative boundary data were obtained from the Global Administrative Areas (GADM) database version 4.1, using the Level 1 (state-level) shapefile corresponding to Nigeria’s 36 states and the Federal Capital Territory. State-level weighted proportions of vaccinated children were linked to the shapefile for spatial analysis. To first assess overall spatial dependence, Global Moran’s I statistic was computed using an inverse-distance spatial weights matrix constructed from the geographic centroids of the states. The spatial weights matrix assigns greater influence to geographically proximate states and lower influence to more distant states. Statistical significance was assessed using 999 Monte Carlo permutations at a 0.05 significance level.

Following the assessment of global spatial autocorrelation, Local Moran’s I (Local Indicators of Spatial Association, LISA) [24] was employed to identify specific clusters and spatial outliers of HepB-BD coverage across states. The analysis produced five categories: not significant, High–High, Low–Low, High–Low, and Low–High. High–High states represent hotspot states (high coverage surrounded by high-coverage neighbours), while Low–Low states represent cold spot states (low coverage surrounded by low-coverage neighbours). High–Low and Low–High states represent spatial outliers. Spatial analysis and visualization were conducted using Quantum Geographic Information System (QGIS) version 3.40 (QGIS Development Team), a free and open-source software for geospatial analysis and mapping.

Results

Background characteristics

A total of 13,064 mother–child pairs were included in the analysis. Slightly over half of the children were male (51.0%). Majority of the mothers were aged 25–34 years (49.6%), married or living with a partner (95.6%), and residing in rural areas (61.1%) (Table 2). About 44.0% of mothers had no formal education, 58.9% delivered at home, 25.1% did not receive antenatal care, and more than one-third (35.3%) of participants were from the North West region (Table 2). Of the children surveyed in the study, a little above half of the children (51.6%), received HepB-BD vaccine (Fig. 1).

Table 2.

Socio-demographic characteristics of study participants (n = 13,064)

Variable	n	%
Child's sex
Male	6,646	51.0
Female	6,418	49.0
Parity
1–4	8,568	65.4
5+	4,496	34.6
Mother's age (years)
15–24	3,911	29.7
25–34	6,476	49.6
35+	2,677	20.7
Mother's marital status
Married/living with partner	12,439	95.6
Others	625	4.4
Religion
Christians	5,224	37.0
Islam and others	7,840	63.0
Mother's education attainment
No education	5,664	44.0
Primary	1,950	14.5
Secondary/higher	5,450	41.5
Household wealth status
Poorest	2,971	21.0
Poorer	2,994	22.6
Middle	2,756	20.7
Richer	2,436	18.9
Richest	1,907	16.8
Place of residence
Urban	4,508	38.9
Rural	8,556	61.1
Region
North Central	2,308	13.9
North East	2,760	18.0
North West	3,828	35.3
South East	1,491	10.4
South South	1,300	9.2
South West	1,377	13.2
Media exposure
Not exposed	5,108	37.8
Exposed	7,956	62.2
ANC reception
Did not receive ANC	3,195	25.1
Received ANC	9,406	74.9
Distance to health facility
A big problem	3,914	28.1
Not a big problem	9,150	71.9
Place of delivery
Home	7,667	58.9
Health facility	5,397	41.1

Fig. 1.

Prevalence of HepB-BD vaccination in Nigeria

Spatial distribution of HepB-BD coverage

State-level HepB-BD vaccination coverage demonstrated substantial geographic variation across Nigeria. The exact weighted state-level coverage percentages are presented in Supplementary Table S1. Coverage ranged widely across regions, with higher proportions observed in southern states and lower proportions concentrated in parts of northern Nigeria (Fig. 2). In the South-West region, coverage ranged from 60% to 75% in Ogun and Ondo States. In the remaining South-Western states, Oyo, Osun, Ekiti, and Lagos HepB-BD coverage was higher, ranging from (75% to 90%). Similarly, in the South-East region, coverage ranged between 75% and 90% in Enugu, Anambra, Abia, and Imo States, while Ebonyi State recorded slightly lower coverage 60%–75%. In the South-South region, Edo State recorded between 75% and 90% coverage. Cross River State had lower coverage 45%–60%, while in Delta, Bayelsa, Rivers, and Akwa Ibom the proportion of children who received the birth dose was between (60% and 75%).

Fig. 2.

Distribution of HepB-BD vaccination coverage across Nigerian States

In the North-Central region, Nasarawa State and the Federal Capital Territory (FCT) recorded relatively high coverage levels 75%–90% compared to Kogi, Benue, and Plateau States which each recorded coverage between 60% and 75%. Kwara State had between 45 and 60% of children covered while Niger State had the lowest coverage in the region (30–45%) (Fig. 2).

In the North-East region, coverage was particularly low in Yobe and Gombe States (15%-30%). Bauchi and Taraba States had coverage between 30% and 45%, while Borno and Adamawa States had 45%–60% of the children covered. The North-West region had states with the lowest coverage levels. Kebbi, Sokoto, and Zamfara States had between 15% and 30% of their children receive the vaccine. Kaduna State had slightly higher coverage (45%–60%), while Kano, Katsina, and Jigawa States recorded coverage between (30% and 45%).

Global spatial autocorrelation analysis demonstrated significant clustering of HepB-BD coverage across Nigerian states (Moran’s I = 0.220, expected I = − 0.028, z = 8.71, p = 0.001), showing that states with similar vaccination coverage levels tend to be geographically clustered rather than randomly distributed.

Following the significant global spatial dependence, Local Indicators of Spatial Association (LISA) analysis was conducted to identify the specific locations of clustering (Fig. 3). Overall, the Local Moran’s I results revealed that not all states showed statistically significant spatial autocorrelation. While several states demonstrated no significant clustering, distinct hotspots and cold spots of HepB-BD coverage were identified. In the northern region of the country, some states were identified as cold spots, indicating significant clustering of low vaccination coverage. In the North-West region, Kebbi, Sokoto, Zamfara, Jigawa, and Katsina States were identified as cold spots. Similarly, in the North-East region, Yobe and Bauchi States were also revealed as cold spots, suggesting that low HepB-BD coverage is spatially concentrated in these areas. In contrast, Ogun State in the South-West and Kogi State in the North-Central were identified as hotspots suggesting that these were states where high coverage of HepB-BD was clustered. No states were identified as significant spatial outliers.

Fig. 3.

Local Moran’s I analysis of HepB-BD vaccination in Nigeria

Factors associated with HepB birth dose vaccination

In assessing the association between respondents and HepB-BD vaccination, Table 3 shows that only child’s sex was not statistically significantly associated with HepB-BD (p = 0.669). Other respondents’ characteristics including parity (p < 0.001), mother’s age (p < 0.001), mother’s marital status (p < 0.001), mother’s religion (p < 0.001), mother’s educational attainment, (p < 0.001) and household wealth status (p < 0.001) were significantly associated with HepB-BD vaccination. Other characteristics significantly associated with HepB-BD vaccination the include place of residence, region, media exposure, ante-natal care attendance, distance to health facility and place of delivery (Table 3). Specifically, among the vaccinated children, 51% were male while 49% were female. In respect of parity, while 72% of HepB-BD vaccinated children were to mothers who had 1–4 children, 28% were to mothers with five or more children. Among vaccinated children about 26% were children of mothers aged 15–24 years, 53% were to mothers aged 25–34 years and about 21% were to mothers aged thirty-five years and above. In addition, among the vaccinated children, 95% were to mothers who were married or living with partner. Of the vaccinated children, 52% were born to mothers who practiced Christianity, while 48% were born to mothers who practiced Islam or other religions.

Table 3.

Bivariate analysis of factors associated with hepatitis B birth-dose vaccination in Nigeria (n = 13,064)

Variable	Total n(%)	Not Vaccinated n (%)	Vaccinated n (%)	χ² (p-value)
Child’s sex
Male	6,646 (51.0)	3,238 (50.8)	3,408 (51.3)	0.256 (0.669)
Female	6,418 (49.0)	3,134 (49.2)	3,284 (48.7)
Parity
1–4	8,568 (65.4)	3,799 (58.7)	4,769 (71.7)	244.4 (< 0.001)
5+	4,496 (34.6)	2,573 (41.3)	1,923 (28.3)
Mother’s age (years)
15–24	3,911 (29.7)	2,148 (33.7)	1,763 (25.9)	101.4 (< 0.001)
25–34	6,476 (49.6)	2,923 (45.9)	3,553 (53.1)
35+	2,677 (20.7)	1,301 (20.4)	1,376 (20.9)
Mother’s marital status
Married/living with partner	12,439 (95.6)	6,111 (96.5)	6,328 (94.7)	23.16 (< 0.001)
Others	625 (4.4)	261 (3.5)	364 (5.3)
Religion
Christians	5,224 (37.0)	1,573 (21.6)	3,651 (51.6)	1254.5 (< 0.001)
Islam and others	7,840 (63.0)	4,799 (78.4)	3,041 (48.4)
Mother’s education attainment
No education	5,664 (44.0)	4,079 (65.1)	1,585 (24.2)	2463.7 (< 0.001)
Primary	1,950 (14.5)	889 (13.7)	1,061 (15.2)
Secondary/higher	5,450 (41.5)	1,404 (21.2)	4,046 (60.6)
Household wealth status
Poorest	2,971 (21.0)	2,214 (32.4)	757 (10.4)	2219.1 (< 0.001)
Poorer	2,994 (22.6)	1,839 (29.2)	1,155 (16.4)
Middle	2,756 (20.7)	1,265 (20.5)	1,491 (20.8)
Richer	2,436 (18.9)	742 (12.4)	1,694 (24.9)
Richest	1,907 (16.8)	312 (5.4)	1,595 (27.6)
Place of residence
Urban	4,508 (38.9)	1,401 (24.2)	3,107 (52.7)	1120.6 (< 0.001)
Rural	8,556 (61.1)	4,971 (75.8)	3,585 (47.3)
Region
North Central	2,308 (13.9)	876 (11.8)	1,432 (15.9)	1720 (< 0.001)
North East	2,760 (18.0)	1,749 (22.7)	1,011 (13.6)
North West	3,828 (35.3)	2,588 (48.2)	1,240 (23.2)
South East	1,491 (10.4)	371 (5.2)	1,120 (15.4)
South South	1,300 (9.2)	470 (6.7)	830 (11.5)
South West	1,377 (13.2)	318 (5.5)	1,059 (20.4)
Media exposure
Not exposed	5,108 (37.8)	3,423 (52.8)	1,685 (23.7)	1174.7 (< 0.001)
Exposed	7,956 (62.2)	2,949 (47.2)	5,007 (76.3)
ANC reception
Did not receive ANC	3,195 (25.1)	2,504 (40.5)	691 (10.5)	1503.7 (< 0.001)
Received ANC	9,406 (74.9)	3,672 (59.5)	5,734 (89.5)
Distance to health facility
A big problem	3,914 (28.1)	2,408 (35.0)	1,506 (21.5)	293.4 (< 0.001)
Not a big problem	9,150 (71.9)	3,964 (65.0)	5,186 (78.5)
Place of delivery
Home	7,667 (58.9)	5,074 (80.0)	2,593 (39.1)	2251 (< 0.001)
Health facility	5,397 (41.1)	1,298 (20.0)	4,099 (60.9)

Furthermore, majority of vaccinated children (61%) had mothers with secondary or higher education. About 53% of HepB-BD vaccinated children were from households in the richer and richest wealth categories, while 21% were from households with middle wealth status. Additionally, 53% of vaccinated children resided in urban areas. Regionally, 23.2% of vaccinated children were from the Northwest while about 20.4% were from the Southwest. Most vaccinated children had mothers who were exposed to media, and approximately 90% had mothers who received antenatal care during pregnancy. Also, 79% of vaccinated children were born to mothers who did not consider distance to a health facility a problem, and about 60.9% were delivered at a health facility (Table 3).

Binary logistic regression was employed to explore the predictors of HepB-BD vaccination in Nigeria. Among the predictors examined, child’s sex, parity and Mother’s marital status had no significant association with HepB-BD uptake (Table 4). The odds of children being vaccinated with HepB-BD increased as mothers age increased. For instance, compared to children of mothers aged 15–24 years, 27% and 36% increases were observed for those aged 25–34 years (aOR:1.27; p = 0.002) and 35 years and above (aOR:1.36; p = 0.002). Compared to children of mothers who were Christians, those of mothers who practised Islam and other religions had 32% reduced odds of being vaccinated with HepB-BD (aOR:0.68; p < 0.001).

Table 4.

Binary logistic regression analysis of factors associated with hepatitis B birth-dose vaccination in Nigeria (n = 13,064)

Variable	Adjusted Odds Ratio	95% CI	p-value
Child’s sex
Male (ref)
Female	1.000	0.908–1.102	0.994
Parity
1–4(ref)
5+	0.927	0.811–1.059	0.266
Mother’s age (years)
15–24(ref)
25–34	1.268	1.092–1.471	0.002
35+	1.357	1.119–1.645	0.002
Mother’s marital status
Married/living with partner(ref)
Others	1.121	0.900-1.397	0.306
Religion
Christians (ref)
Islam and others	0.682	0.573–0.811	< 0.001
Mother’s education attainment
No education (ref)
Primary	1.361	1.148–1.612	< 0.001
Secondary/higher	1.919	1.625–2.266	< 0.001
Household wealth status
Poorest (ref)
Poorer	1.219	1.034–1.438	0.019
Middle	1.284	1.052–1.569	0.014
Richer	1.581	1.258–1.988	< 0.001
Richest	2.465	1.884–3.225	< 0.001
Place of residence
Urban (ref)
Rural	0.834	0.707–0.985	0.033
Region
North Central (ref)
North East	0.678	0.545–0.844	< 0.001
North West	0.639	0.509-0.800	< 0.001
South East	0.638	0.491–0.829	0.001
South South	0.613	0.481–0.782	< 0.001
South West	1.201	1.068–1.350	0.001
Media exposure
Not exposed (ref)
Exposed	1.226	1.084–1.387	0.001
ANC reception
Did not receive ANC (ref)
Received ANC	2.757	2.388–3.184	< 0.001
Distance to health facility
A big problem (ref)
Not a big problem	1.303	1.138–1.492	< 0.001
Place of delivery
Home (ref)
Health facility	2.085	1.826–2.381	< 0.001

Furthermore, results show that the higher the education attainment of mothers, the higher the odds of children being vaccinated with the HepB-BD. For instance, compared to children of mothers with no education, the odds increased by 36% and 92% respectively to children of mothers who had primary education (aOR:1.36; p < 0.001) and secondary or higher education (aOR:1.92; p < 0.001). There is a positive relationship between household wealth status and the odds of HepB-BD vaccination. Relative to those born in poorest household, the odds of children being vaccinated with HepB-BD increased by 22%, 28%, and 58% for children born in the household with poorer (aOR:1.22; p = 0.019), middle (aOR:1.28; p = 0.014) and richer (aOR:1.58; p < 0.001) household wealth status respectively. Children in the richest household wealth status category had a little above two folds of the odds of those born in the poorest household wealth status (aOR:2.46; p < 0.001 (Table 4).

Likewise, children in rural areas had 17% lower odds of being vaccinated with HepB-BD compared to those in urban areas (aOR:0.83; p < 0.033). Region wise, except for children from the South West region of Nigeria who had about 20% higher odds of being vaccinated (aOR:1.20; p = 0.001), compared to children from the North Central, children from other regions such as North East (aOR:0.68; p < 0.001), North West (aOR:0.64; p < 0.001), South East (aOR:0.64; p = 0.001) and South South (aOR:0.61; p < 0.001) had between 32 and 39% reduced odds of being vaccinated with HepB-BD. In addition, children born to mothers exposed to the media had 23% higher odds of being vaccinated with HepB-BD compared to those of mothers who had no media exposure. (aOR:1.23; p = 0.001)

Similarly, children of mothers who received ante-natal care had about 3-fold higher odds of vaccination than those whose mother did not receive antenatal care (aOR:2.76; p < 0.001). Also, while 30% higher odds of vaccination was observed for children of mothers who reported that distance to health facility was not a big problem compared to those who considered it a big problem (aOR:1.30; p < 0.05), children delivered at a health facility had 2-fold higher odds of vaccination than those delivered at home. (aOR:2.08; p < 0.001) (Table 4).

Sensitivity analysis

We conducted a sensitivity analysis by re-running the multivariable logistic regression with “don’t know” responses excluded. A total of 14 (0.11%) children had “don’t know” responses for HepB-BD vaccination status. In the primary weighted analysis, overall coverage was 51.6%. Excluding these responses in the sensitivity analysis resulted in a weighted coverage of 51.64%, representing a negligible absolute difference of 0.04% points. Re-estimation of the weighted multivariable logistic regression model excluding “don’t know” responses produced virtually identical adjusted odds ratios and confidence intervals (Supplementary Table S2). There were no meaningful changes in the magnitude, direction, or statistical significance of the adjusted associations compared with the primary multivariable logistic regression model (Table 4). The full sensitivity model coefficients are provided in Supplementary Table S2.

Discussion

This study examined the spatial distribution and determinants of HepB-BD vaccination in Nigeria, revealing critical disparities that highlight underlying geographic, socio-economic, maternal, and healthcare related factors. In this study, only slightly more than half (51.6%) of Nigerian children aged 0–24 months received the HepB-BD, a figure that falls short of WHO’s target of timely birth dose coverage ≥ 90% within 24 h of life [25]. The result is however consistent with earlier reported rate of 52% provided by WHO for HepB-BD coverage in the country [13] but lower than the coverage reported in a study from Benue, Nigeria [14].

The present study builds on previous literature by providing spatial evidence of childhood immunization coverage across Nigeria. The spatial analysis revealed a marked north–south divide in vaccine coverage within the country, with states in the southern regions, the Southwest and Southeast, reporting substantially higher vaccination rates (60–90%) than their northern counterparts. In contrast, many states in the Northwest and Northeast recorded much lower coverage levels, with rates ranging between 15% and 45%. These geographic disparities align with findings from earlier studies that reported similar trends in other immunization programs, often attributed to regional differences in health infrastructure, maternal education, and cultural beliefs [26, 27]. The observed spatial disparities highlight regional divides and geographic inequities often concealed by national averages. If left unaddressed, these gaps may perpetuate pockets of zero-dose and under-immunized children, particularly in the Northwest and Northeast, where the burden is disproportionately concentrated. These inequities may be due to well documented factors such as limited access to health facilities that can administer vaccines within 24 h of birth, lower maternal education and awareness, and reduced availability or utilization of facility-based deliveries services, findings that are consistent with our results in this study. For instance, in the northern states, cultural preferences for home births and the use of traditional birth attendants may limit access to timely postnatal services, thereby reducing the likelihood of HepB-BD uptake [28]. These findings have important programmatic implications, as they can guide the strategic allocation of resources to low-coverage regions and support the design of context-specific interventions, including outreach services, community engagement initiatives, and investments in health infrastructure in affected regions. These findings highlight the continued relevance of ongoing equity-focused strategies aimed at ensuring that no child is left behind, in alignment with the Immunization Agenda 2030.

The identification of high-coverage hotspots and low-coverage cold spots suggests the presence of geographically structured determinants of HepB-BD uptake. The clustering of low coverage across several North-West and North-East states suggests that vaccination inequities are geographically concentrated rather than randomly distributed. These regions have historically experienced health system constraints that includes limited facility density and prolonged insecurity, which has been associated with disruptions in maternal and child health service delivery, including immunization programs [29, 30]. While causal inferences cannot be drawn from this study, the spatial pattern found aligns with wider contextual conditions that may impact vaccination delivery and uptake within these regions. Addressing HepB-BD inequities therefore requires attention to geographically structured health system capacity, service accessibility, and regional contextual determinants.

On multivariable analysis for the determinants of HepB-BD, the study found maternal education as a significant predictor of HepB-BD uptake, a finding consistent with global literature. Mothers with secondary or higher education had significantly higher odds of their children receiving vaccination at birth, a finding in keeping with studies from Ghana [31], Nigeria [32, 33] and India [34]. Educated mothers are more likely to understand the importance of vaccination, navigate healthcare systems effectively, and challenge misinformation. Equally, we also found that wealth index was associated with the uptake of HepB-BD, as children from wealthier households had higher odds of HepB-BD vaccination. This finding is in keeping with previous studies and patterns observed in other low-and middle-income countries [34–36]. Economically disadvantaged households are less likely to seek healthcare, even for free services such as vaccination [32], due to financial barriers including transportation costs and other indirect health expenses. There is a need to design programmes and interventions that can reduce the indirect costs associated with accessing healthcare, particularly for these households. Such approaches may include deploying mobile immunization services to bring vaccines closer to families and providing transportation support or incentives, which may contribute to improved vaccination uptake among economically disadvantaged populations. The odds of receiving HepB-BD were lower among rural residents compared to their urban counterparts, possibly reflecting the urban–rural disparities in access to health services, higher health literacy, and healthcare infrastructure. This divide highlights systemic inequities that urgently require policy attention [37]. To mitigate these inequities, there is need to strengthen primary healthcare infrastructure, expand services, and enhance community health education to improve timely vaccination.

Facility-based delivery and ANC attendance were also identified as predictors of HepB-BD vaccination. Mothers who delivered in health facilities were twice as likely to have their children vaccinated compared to those who did not. This finding is not surprising, as mothers who deliver in health facilities are more likely to have their newborns vaccinated immediately after birth, since HepB-BD is often administered before discharge. In contrast, mothers who deliver at home may experience delays in accessing vaccination services or miss the critical window entirely. Furthermore, the low vaccination coverage observed in the Northwest and Northeast regions from the spatial analysis, along with the lower odds of HepB-BD vaccination among mothers with non-facility-based deliveries, supports the association between health facility delivery and vaccination uptake. This inference is supported by Olubodun et al. [38], who reported low facility delivery rates in these regions. The overlap may reflect the immediacy of vaccination opportunities in health facilities, which facilitate timely HepB-BD administration and reduce the likelihood of missed doses among facility-delivered newborns. These findings highlight the need to strengthen strategies that promote facility-based deliveries particularly in the regions of the country with low facility delivery.

Similarly, ANC attendance increased the odds of vaccination nearly threefold, consistent with previous studies [39, 40] and supporting evidence that ANC is a critical opportunity for educating mothers and initiating immunization plans. Strengthening health education during ANC visits by providing clear vaccination information and, where feasible, introducing written vaccination plans for pregnant women could serve as practical reminders and help improve the timeliness and uptake of the HepB-BD vaccine. Perception of distance to health facilities also showed a significant association, as mothers who did not consider distance a barrier were more likely to have their children vaccinated. This finding is consistent with earlier studies [41–43], which documented that physical accessibility and service availability are key enablers of maternal and child health service utilization.

Additionally, religion was associated with the likelihood of receiving HepB-BD, as children born to Muslim mothers were significantly less likely to receive the birth dose compared with those of Christian mothers. This disparity may reflect differences in health-seeking behaviour, cultural beliefs, and trust in the health system across religious groups. Lower engagement with formal healthcare services and greater exposure to vaccine-related misconceptions in some Muslim communities may contribute to reduced uptake, consistent with historical vaccine hesitancy in parts of northern Nigeria linked to distrust in Western medicine and concerns about vaccine safety [44]. Targeted faith-based outreach and vaccination communication, alongside sustained engagement with religious, traditional, and opinion leaders, may help address these disparities, as such leaders are highly influential within their communities [45, 46]. Likewise, children born to mothers exposed to the media had higher odds of receiving HepB-BD compared to those whose mothers had no media exposure, consistent with previous studies [47, 48]. The media may serve as an important source of health information for mothers regarding the benefits of vaccination. Overall, media exposure appears to be an important factor associated with HepB-BD uptake within the broader socioeconomic context.

This study has several important strengths. First, it used nationally representative data from the 2018 NDHS, enhancing generalizability to Nigerian children aged 0–24 months. Second, beyond conventional regression analysis, this study utilised spatial statistical methods to quantify and visualize geographic disparities and identify specific states driving clustering patterns. This approach strengthens inference by demonstrating that observed regional disparities are not random but geographically structured. Third, the multivariable logistic regression model was adjusted for various factors, reducing confounding and allowing the identification of independent predictors of HepB-BD uptake. Importantly, the use of sensitivity analysis including “don’t know” vaccination responses demonstrated that outcome misclassification did not significantly affect the results, thereby enhancing the robustness of our findings. Finally, the state-level analysis provides policy-relevant insights by identifying specific geographic cold spots and hotspots, which may support more targeted immunization strategies.

Despite these strengths, the study has several limitations that should be acknowledged. First, the cross-sectional design of the NDHS limits causal inference. Second, vaccination status was derived from vaccination cards and maternal recall, which may introduce recall bias. Third, spatial regression models were not applied, as the primary aim was to quantify clustering and identify geographic patterns rather than to estimate spatially lagged effects. Fourth, some potentially relevant contextual factors were not available in the NDHS dataset; thus, residual confounding by unmeasured factors such as facility-level characteristics, physical access to health facilities, regional insecurity and sociocultural context may persist. Finally, although the 2018 NDHS provides the most recent complete nationally representative dataset with detailed vaccination information at the time of analysis, coverage patterns may have evolved since data collection. Therefore, findings should be interpreted within the context of the survey period.

Policy implications

The findings of this study highlight the need to move beyond uniform national strategies toward geographically targeted and equity-focused approaches. Given the clustering of low HepB-BD coverage in the Northwest and Northeast, programmatic efforts should prioritize these cold-spot regions through state-specific microplanning and resource allocation aligned with spatial risk profiles. Rather than broadly expanding infrastructure, efforts should focus on improving timely birth-dose delivery within existing primary healthcare facilities, particularly by strengthening coordination between maternity services and immunization units to ensure administration before discharge. In settings with high rates of home delivery, structured postnatal follow-up systems and community-linked notification mechanisms may help reduce missed opportunities for timely vaccination.

Integration of vaccination education into antenatal care should go beyond awareness creation to include structured birth preparedness counselling that integrates a vaccination plan for delivery. Community health workers and mobile outreach services may be strategically deployed in identified low-coverage clusters to address geographic access barriers. Concerning the observed religious disparities, partnerships with trusted religious and traditional leaders should be institutionalized within immunization programming rather than implemented as ad hoc outreach activities. Communication strategies should be data-driven and culturally tailored, using local media channels in cold-spot regions to address context-specific misconceptions.

Finally, because household wealth remains a significant determinant of HepB-BD uptake, linking immunization services with existing social support and maternal health financing schemes may provide a pathway for reducing indirect access barriers while advancing equity in birth-dose coverage.

Conclusion

HepB birth-dose vaccination coverage in Nigeria remains suboptimal, with significant geographic and socioeconomic disparities. Coverage was associated with maternal education, antenatal care utilization, household wealth, media exposure, and regional location. Spatial analysis demonstrated significant clustering of vaccination coverage across states, with identifiable low-coverage clusters concentrated in parts of northern Nigeria and localized high-coverage clusters in select southern and north-central states. Efforts to improve coverage may benefit from context-specific and geographically targeted strategies that strengthen maternal education, antenatal care utilization, and health facility delivery, particularly in states identified as low-coverage clusters.

Abbreviations

ANC

Antenatal care

BD

Birth dose

CI

Confidence interval

DHS

Demographic and Health Surveys

EPI

Expanded Programme on Immunization

HBV

Hepatitis B virus

HepB-BD

Hepatitis B birth dose

ICF

Inner City Fund (ICF International)

NDHS

Nigeria Demographic and Health Survey

NMEP

National Malaria Elimination Programme

NPC

National Population Commission

OR

Odds ratio

UNICEF

United Nations Children’s Fund

VIF

Variance inflation factor

WHO

World Health Organization

WUENIC

WHO and UNICEF Estimates of National Immunization Coverage

Authors’ contributions

OAO conceptualized the study. OAO and BA developed the research proposal. HO extracted and analyzed the data and presented the results. OAO also participated in data analysis and interpretation of findings. OAO, BA, FOF, and TO drafted the manuscript. TO, EO, AO, and ATF contributed to the discussion, critical review, and revision of the initial manuscript. All authors read and approved the final manuscript.

Funding

No funding was secured for this study.

Data availability

The datasets analyzed are publicly available from the DHS Program upon registration and approval of a data request. Additionally, anyone interested can freely access the data online at https://dhsprogram.com.

Declarations

Ethics approval and consent to participate

Since this study involved secondary analysis of an existing, publicly available dataset, additional ethical approval was not required. However, permission to use the dataset was obtained from the MEASURE DHS program https://dhsprogram.com following registration and approval of the data request. All data used in this analysis are de-identified and publicly accessible upon reasonable request from the MEASURE DHS website.

Consent for publication

“Not applicable” in this section.

Competing interests

The authors declare no competing interests.