Skip to Content
MDPIMDPI

Advanced

VaccinesVaccines
PDF
  • Article
  • Open Access

26 October 2026

17 Pages

Level and Trends of Zero-Dose Prevalence Among Children in Mozambique: A Repeated Cross-Sectional Analysis of Three Household Surveys, 2011–2023

,
,
,
,
,
,
,
,
…
1
Instituto Nacional de Saúde de Moçambique, Marracuene District, EN1, Bairro da Vila—Parcela n° 3943, Matola C.P. 264, Maputo Province, Mozambique
2
Global Health and Tropical Medicine (GHTM), LA-REAL, Instituto de Higiene e Medicina Tropical, Universidade Nova de Lisboa, Rua da Junqueira 100, C.P. 1349-008 Lisbon, Portugal
3
Direcção Nacional de Saúde Pública, Ministério de Saúde, Maputo City C.P. 264, Mozambique
4
Department of Global Health, University of Washington, Seattle, WA 98105, USA
Vaccines2026, 14(11), 926;https://doi.org/10.3390/vaccines14110926
This article belongs to the Special Issue Vaccine Epidemiology and Population Health
Version Notes
Order Reprints

Abstract

Background: Despite the proven role of childhood immunization in reducing child mortality, global coverage targets remain unmet. Zero-dose (ZD) children—those who have not received any DTP vaccine dose—represent a critical challenge. Mozambique is among the lowest-coverage countries, with full immunization declining from 66% in 2015 to 38% in 2022–2023. This study determines the levels and trends of ZD prevalence in Mozambique. Methods: We analyzed data from three nationally representative surveys—DHS 2011, IMASIDA 2015, and DHS 2022/23—focusing on children aged 12–23 and 24–35 months. The analysis examined ZD prevalence trends across provinces and disparities by urban–rural residence, wealth index, and maternal education. Adjusted prevalence ratios (PRs) and 95% confidence intervals (CIs) were estimated using log-binomial regression. Results: ZD prevalence increased from 5.4% to 14.3% (2015–2022) among children aged 12–23 months, and from 6.4% to 16.0% among those aged 24–35 months. Key risk factors included rural residence (PR: 0.50; 95% CI: 0.29–0.88), maternal illiteracy (PR: 0.75; 95% CI: 0.61–0.92), and non-institutional delivery (PR: 0.20; 95% CI: 0.13–0.30). The wealth gap in ZD prevalence widened from 11.3 to 26.3 percentage points, and the urban–rural gap from 3.9 to 15.3 points. Children of mothers with no education had a 22.8% ZD prevalence versus 5.4% among those with secondary or higher education. Conclusion: Rising ZD prevalence and widening inequalities underscore the urgent need for targeted, equity-focused interventions. Strengthening health systems to address these inequities and improve routine immunization services is essential for better child health outcomes in Mozambique.

1. Introduction

Immunization is crucial in preventing and reducing child mortality rates [1,2]. Despite significant progress in increasing immunization coverage, the global target of 90% immunization rate remains unmet [2]. The World Health Organization (WHO) defines “zero-dose” (ZD) children as those who lack access to or are never reached by routine immunization services. These children are operationally defined as those who have not received their first dose of a diphtheria–tetanus–pertussis (DTP)-containing vaccine (DTP1) [3,4]. Overall, an assessment of nationally representative household surveys conducted during the years 2010 to 2019 from 92 low and middle-income countries (LMIC) found a 7.7% prevalence of ZD among children between 12 and 23 months [5] despite decades of increased global coverage for WHO-recommended vaccines [6].
Many developing countries continue to struggle with low vaccination coverage due to factors such as geopolitical barriers, socioeconomic disparities, conflict zones, and widespread misinformation about vaccines [7,8]. In particular, the COVID-19 pandemic disrupted the routine immunization services, leading to a sharp increase in the number of zero-dose children worldwide [9], especially in low- and middle-income countries (LMICs) [10]. For example, Nigeria has seen one of the highest numbers of zero-dose children during this period, with an estimated 2.2 million children missing all vaccines in 2020 [11]. Similarly, in the Democratic Republic of Congo, around 86,905 children missed the oral polio vaccine, while 74,860 did not receive a dose of the DTP or hepatitis B-Hib3 vaccines that same year [12].
In Mozambique, there has been documented COVID-19 pandemic-related maternal–child health services disruption, including child immunization, based on model estimates and routine health information system data [13]. However, these essential sources of information cannot provide the proportion of children in the population that did receive any of the recommended vaccines. This data gap presents a significant challenge for the country’s Expanded Program on Immunization (PAV) and broader public health strategies. Zero-dose children complicate immunization program performance, hinder immune response, and contribute to the persistence of outbreaks like poliomyelitis and measles outbreaks, as well as the risk of intussusception from the rotavirus vaccine [14].
This study aims to examine the proportion of zero-dose children across Mozambique’s 11 provinces from 1997 to 2023 and identify factors associated with being zero-dose, using data from three most recent nationally representative household surveys. We assessed the prevalence and trends of zero-dose children in Mozambique, among those aged 12–23 months and 24–35 months. The analysis explored disparities in zero-dose prevalence between urban and rural areas, as well as variations based on maternal education levels and wealth index. Additionally, we examined how household and maternal characteristics influence zero-dose prevalence. The findings will provide insights into the barriers preventing vaccination and inform health policies to improve childhood immunization uptake.

2. Materials and Methods

2.1. Study Design and Data Sources

We analyzed data from the three nationally representative household surveys: Demographic and Health Survey (DHS 2011), Inquérito de Indicadores de Imunização, Malária e HIV/SIDA em Moçambique (IMASIDA 2015), and DHS 2022/23. These surveys were designed to collect information on household, maternal, and child health characteristics using a nationally representative complex sampling design. The sampling design included stratification and two-stage sampling. Stratification was based on the 11 provinces and their urban/rural status, resulting in 21 strata. The first stage of sampling involved the demarcation of enumeration areas (ENU), from which clusters of 15–25 households were randomly selected. In the second stage, systematic sampling was applied within these clusters.
The surveys gathered data on household characteristics and maternal and child health status through questionnaires administered to women of reproductive age (15–49 years). These women provided information on their household’s sociodemographic, economic, and health status, including maternal and child health, environment, behavior, and prenatal and postnatal care. The survey response rates were high, at 98.9%, 94.5%, 94.3% for women interviewed in the 2011, 2015, and 2022/23 surveys, respectively. Detailed methodology regarding the participant selection process has been described elsewhere [15].

2.2. Study Setting

Mozambique is a southeastern African country covering 801,590 km2, with a population of 34,631,765, the majority (61.2%) of whom live in rural areas. The population grows annually at a rate of 2.5% as of 2023 [16,17]. Children under the age of 5 make up 15.3% of the population, with 17.3% of them in their second year of life. Since gaining independence from Portuguese colonial rule in 1975, Mozambique has faced significant challenges, including a 16-year civil war that devastated its infrastructure and displaced over half of its population [18], and with increased frequency, it has been fustigated by cyclones and climate change disasters. Mozambique is classified as a low-income country (LIC) by the World Bank, with a gross domestic product (GDP) per capita of USD 1657.1 in 2023 [19]. According to the United Nations 2023 Human Development Index, the country ranked 183rd out of 193 countries.
Mozambique’s EPI began in 1979 with the introduction of Bacillus Calmette–Guérin (BCG), polio, DTP, and measles vaccines [20]. Over time, more vaccines were added to the national schedule, including an updated DTP vaccine (with additional immunogens for hepatitis B and Hemophilus influenzae type B) in 2008 and a pneumococcal conjugate vaccine (PCV) in 2013. In 2015, the rotavirus vaccine (RV), the injected polio vaccine (IPV), and the second dose of the measles vaccine were introduced. The measles and rubella (MRV) vaccine was introduced in 2019 [21]. Despite the broader range of vaccines available and improvement in each vaccine’s coverage and accessibility, overall full immunization continues to fall short of the EPI target of reaching full immunization coverage of 90%.

2.3. Outcome

The outcome of this study is the zero-dose (ZD) child status, stratified by children aged 12–23 months and 24–35 months. Two definitions of zero-dose are used. The first is based on the WHO’s current recommendations for routine childhood immunizations, which include BCG, polio, DPT and measles-containing vaccines to be administrated in the first year of life [16]. We define zero-dose as children who did not receive one or more of these four vaccines by the survey date, henceforth referred to as ZDWHO. The second definition follows the Gavi operational definition, which classifies zero-dose children as those who have not received the first dose of the diphtheria–tetanus–pertussis-containing vaccine (DTP1) [22]. The prevalence of zero-dose is defined as the proportion of children with zero-dose status, henceforth referred to as ZDGavi.

2.4. Covariates

We selected variables based on the literature on maternal–child determinants of health and the availability of data across all surveys. A total of 15 variables were chosen for potential association with the zero-dose status of Mozambican children. The variables included in this analysis were: (1) maternal age at the time of the survey, grouped as 15–24, 25–34 and ≥35 years old; (2) maternal educational level, defined as illiterate, primary and secondary or above; (3) marital status, which originally had six categories (single, married, living with a partner, separated, divorced, and widowed) and was recoded into three categories (single/never in a union, married/living with a partner, and divorced/separated/widowed); (4) mother’s occupation, defined as unemployed (not working/household domestic) and employed (professional, clerical, sales, skilled and unskilled manual, services, agriculture, self-employed, and all others); (5) geographic location, where the provinces were included separately and aggregated to reflect the three regions (northern, central, and southern) of the country; (6) area of residence, categorized as urban and rural; (7) wealth index, grouped as poorest, poor, middle, richer and richest; (8) total household members; (9) religion, categorized differently in each survey but regrouped into four standardized responses across all datasets: Catholic, Islamic, Protestant, and others, which also included no religion; (10) number of antenatal (ANC) visits (no visits, 1 to 3, and ≥4); (11) place of delivery (at home/other and at health facility); (12) sex of the child (male and female); (13) birth order (1, 2 to 3, and ≥4); and (14) health card possession (seen, not seen, and no card).

2.5. Data Processing and Analysis

The analysis was stratified by child age at the survey date, with age groups of 12–23 months and 24–35 months. Descriptive statistics were used to summarize child characteristics. We reported the prevalence of ZD children, along with 95% confidence intervals (CI), for each survey and per definition (ZDWHO and ZDGavi), accounting for the survey sampling design. Changes in ZD prevalence over time were estimated by comparing the ratios of ZD prevalence between 2011 and 2015, and between 2015 and 2022. For these ratios, the delta method was applied to calculate the 95% CI and p-value, testing whether the ratio differed significantly from the null hypothesis (ratio of 1.00).
We assessed the association between household, maternal, and child characteristics and ZD prevalence in the 2022/23 survey. Unadjusted and adjusted prevalence ratios (PR) were estimated using log-binomial regression, accounting for the survey design [23]. We assessed the convergence of the model, checking for software warnings or evidence of boundary estimates. As a sensitivity analysis, we fitted modified Poisson regression models with robust variance. The estimated prevalence ratios and corresponding confidence intervals were substantively unchanged. Given that potentially many of the characteristics included in the analysis were collinear, we evaluated collinearity using variance inflation factors (VIF). Variables with a VIF greater than 10 were removed from the model. The child vaccination card had the highest VIF (above 30), and the place of delivery and ANC use showed high collinearity, with a VIF close to 20. Therefore, we excluded both the child vaccination card and the ANC use from the model. Additionally, we report two adjusted regression models: one that includes the place of delivery and another that excludes it. The place of delivery serves as a proxy for healthcare access, and an opportunity for vaccination-related information to be provided to caregivers; the place of delivery may lie on the causal pathway between sociodemographic characteristics and childhood vaccination. All analyses were conducted in R version 4.4.1 [24] using the survey package [25].

2.6. Ethics

This research was conducted in accordance with the principles of the Declaration of Helsinki and local regulatory requirements. Human Ethics and Consent to Participate declarations: not applicable. Ethical review and approval were waived for this study due to the public availability of the data for analysis.

3. Results

3.1. Sample Description

The three surveys included a total of 5263 children aged 12–23 months and 5026 children aged 24–35 months (Table 1). Approximately three-quarters of children in both age groups lived in rural areas (73.2% and 71.8%, respectively). About one-third of children had illiterate mothers or caregivers (31.5% and 32.7%, respectively), and nearly two-fifths had a mother or caregiver under the age of 25 (42.9% and 37.5%, respectively). The majority of mothers or caregivers were either married or cohabiting (83.0% and 82.3%, respectively). While fewer than 10% of children were born to mothers who had no ANC, around one-third of the children (33.8% and 32.5%, respectively) were not born in institutional settings.
Table 1. Characteristics of mothers and households of children aged 12–23 months and 24–35 months. Frequencies are weighted according to survey design.
Of particular note, since 2011, the proportion of children aged 12–23 months without a vaccination card has increased nearly 2.84 times, rising from 10.8% to 30.7%. Similarly, the proportion of children aged 24–35 months without a vaccination card increased from 15.6% to 40.6%, a 2.60-fold rise (Table 1).

3.2. Prevalence of Zero-Dose and Time Trends

Between 2015 and 2022, the overall prevalence of ZDWHO children aged 12–23 months increased from 5.4% to 14.3% (Table 2), while the ZDGavi prevalence rose from 10.2% to 24.9% (Supplementary Table S1). These increases corresponded to a 2.64-fold (95% CI: 2.05 to 3.99) and 2.44-fold (95% CI: 2.01 to 2.96) rise, respectively (Table 3 and Table S2). Similarly, among the children aged 24–35 months, the ZDWHO prevalence increased from 6.4% to 16.0% (Table 2), reflecting a 2.52-fold (95% CI: 2.11 to 3.00) increase (Table 3), while the ZDGavi prevalence rose from 13.9% to 26.5%, a 1.91-fold (95% CI: 1.68 to 2.17) increase (Supplementary Table S1). Detailed prevalence estimates using the Gavi operational definition are provided in the additional material (Supplementary Table S1) and results of time multiplicative changes in zero-dose prevalence using the Gavi operational definition are provided in the additional material (Supplementary Table S2).
Table 2. Prevalence of zero-dose in percentage according to WHO definition (ZDWHO).
Table 3. Time relative changes in zero-dose prevalence using WHO definition.
From 2011 to 2015, the ZD prevalence among children aged 12–23 months remained relatively stable, with prevalence ratios of 1.09 (95% CI: 0.83 to 1.43) for ZDWHO and 1.08 (95% CI: 0.88 to 1.33) for ZDGavi (Table 3 and Table S2). However, for children aged 24–35 months, the ZDGavi prevalence significantly increased by 1.82-fold (from 7.7% to 13.9%) (Supplementary Tables S1 and S2), while the ZDWHO prevalence remained virtually unchanged, rising from 5.5% to 6.4% (Table 2).
The prevalence of ZDWHO among children without vaccination documentation remained virtually unchanged for children aged 12–23 months, from 43.6% (95% CI: 29.2 to 59.3%) in 2015 to 42.2% (95% CI: 35.0 to 49.7%) in 2022 (Table 2 and Table 4). For children aged 24–35 months, there was a notable increase from 28.4% (95% CI: 18.0 to 41.6%) in 2015 to 37.2% (95% CI: 32.1 to 42.7%), though this change was not statistically significant. Although at a higher prevalence level, the changes in ZDGavi followed a similar pattern to those in ZDWHO for children aged 12–23 months without vaccination documentation, from 2015 (57.5%, 95% CI: 43.3 to 70.6%) to 2022 (57.3%, 95% CI: 50.5 to 63.8%).
Table 4. Factors associated with the prevalence of zero-dose (WHO definition). Both models are adjusted.
Table S4 shows the adjusted associations with ZDGavi prevalence. Overall, compared to ZDWHO, the prevalence ratios for ZDGavi tend to be in the same direction but with weaker strength, with many confidence intervals including the null.

3.3. Change in Gaps of Zero-Dose Prevalence (ZDWHO and ZDGavi)

From 2015 to 2022, there was a differential increase in ZD prevalence across various categories, such as wealth quintiles, area of residence, and maternal education, resulting in widening gaps between these groups (based on both the WHO and Gavi definitions) (Table 2, Table 3 and Tables S1 and S2). Children from poorer households (compared to wealthier households), those living in rural areas (compared to urban areas) and children born to illiterate mothers (compared to those with mothers who had primary, secondary or higher education) experienced the highest increases in ZD prevalence.
The gap in ZDWHO prevalence between the 1st and 5th wealth quintiles among children aged 12–23 months grew from 11.3% (12.0 vs. 0.7%) in 2015 to 26.3% (28.2% vs. 1.9%) in 2022 (Table 2 and Table S5a). A similar trend was observed for children aged 24–35 months, with the gap increasing from 9.3% in 2015 to 26.5% in 2022 (Table 2 and Table S5b). The urban–rural gap in ZDWHO prevalence among children aged 12–23 months expanded from 3.9% (2.5% vs. 6.4%) in 2015 to 15.3% (3.2% vs. 18.5%) in 2022 (Table 2 and Table S6a). Among children aged 24–35 months, this gap grew from 3.3% (3.9% vs. 7.2%) to 15.8% (4.9% vs. 20.7%) (Table 2 and Table S6b). The gap in ZDWHO prevalence between children aged 12–23 months born to mothers with no education and those born to mothers with secondary or higher education increased from 5.2% to 17.4% (Table 2 and Table S7a). Similarly, among children aged 24–35 months, this gap expanded from 8.4% to 23.4% (Table 2 and Table S7b).
Similarly, the ZDGavi prevalence followed the same patterns of gap increase from 2015 to 2022 as ZDWHO prevalence (Tables S1, S5c,d, S6c,d and S7c,d), although ZDGavi had larger values. For instance, the urban–rural gap in ZDGavi prevalence among children aged 12–23 months expanded from 7.5% (4.6% vs. 12.1%) in 2015 to 21.9% (9.0% vs. 30.9%) in 2022 (Tables S1 and S6c).

3.4. Factors Associated with Zero-Dose Prevalence Among Children Aged 12–23 Months

Table S3 presents the unadjusted analysis for both prevalence definitions. Table 4 and Table S4 present the adjusted log-binomial regression analyses of factors associated with zero-dose prevalence according to the WHO and Gavi definitions, respectively. Among children aged 12–23 months, the adjusted model showed that children born to Islamic mothers had a 32% lower prevalence of WHO-defined zero-dose status than those born to Catholic mothers (PR: 0.68, 95% CI: 0.50–0.93). No significant differences were observed for other religious groups, and adjustment for place of delivery did not materially change the results.
Children living in urban areas had a 50% lower ZDWHO prevalence (PR: 0.50, 95% CI: 0.29 to 0.88) compared to those living in rural areas. This association remained virtually unchanged (PR: 0.56, 95% CI: 0.31 to 1.03) even after accounting for place of delivery. Additionally, when factoring in place of delivery, children’s province of residence (compared to Sofala) showed notable variations in ZDWHO prevalence. Children in Niassa (PR: 2.90, 95% CI: 1.34 to 6.27), Nampula (PR: 2.27, 95% CI: 1.02 to 5.05), and Zambézia (PR: 5.66, 95% CI: 2.82 to 11.38) exhibited higher prevalence rates (Table 4).
In the unadjusted model, wealth index quintiles were associated with ZDWHO prevalence, with higher quintiles exhibiting lower prevalence compared to the first quintile. After adjusting for place of delivery, the strength of this association diminished (Table 4). Maternal education level consistently influenced ZDWHO prevalence, regardless of the adjustment for place of delivery. Children born to mothers with primary education had a 25% lower ZDWHO prevalence (PR: 0.75, 95% CI: 0.61 to 0.92) compared to those born to illiterate mothers. Children born to mothers with secondary or higher education had a non-statistically significant 1.22 times higher ZDWHO prevalence (95% CI: 0.88 to 1.70) compared to those born to illiterate mothers. Children born in institutional facilities had an 80% lower prevalence of ZDWHO (PR: 0.20, 95% CI: 0.13 to 0.30) compared to those not born in institutional settings (Table 4).
For ZDGavi, although the associations are weaker than for ZDWHO, they follow the same direction and pattern.

3.5. Factors Associated with Zero-Dose Prevalence Among Children Aged 24–35 Months

After adjustment, factors such as area of residence, province, maternal education, and place of delivery remained statistically significant for ZDWHO prevalence (Table 4). Among children aged 24–35 months, those living in urban areas had a 49% lower ZDWHO prevalence (PR: 0.51, 95% CI: 0.30–0.86) compared to those in rural areas. When compared to Sofala, children in Zambézia, Tete and Nampula had significantly higher ZDWHO prevalence, with prevalence ratios of 4.09 (95% CI: 2.33 to 7.18), 3.44 (95% CI: 1.87 to 6.30), and 1.77 (95% CI: 092 to 3.40), respectively. Children born to mothers with secondary or higher education had a 65% lower ZDWHO prevalence (PR: 0.35, 95% CI: 0.18 to 0.70) compared to those born to illiterate mothers. Children of mothers with primary education showed an 11% lower prevalence (95% CI: 0.69 to 1.16). Additionally, children born in health facilities had a ZDWHO prevalence that was only 33% (PR: 0.33, 95% CI: 0.22 to 0.50) of that for children born at home or outside of health facilities (Table 4).

4. Discussion

Using three nationally representative surveys, we estimated the levels and trends of zero-dose (ZD) prevalence using two definitions: the minimal set of antigens recommended by WHO (ZDWHO), and the Gavi operational definition (ZDGavi). Additionally, we analyzed the predictors of ZD in Mozambique. Between 2015 and 2022, the prevalence of ZDWHO nearly tripled, reaching 14.3% among children aged 12–23 months and 16.0% among children aged 24–35 months. Furthermore, disparities in ZD prevalence widened between urban and rural areas, as well as across maternal education levels and household wealth quintiles.
Several factors contributed to the rise in zero-dose children, including the COVID-19 pandemic, Cyclone Freddy, and cholera outbreaks, as highlighted by Gavi and the WHO. These events led to an increase in the number of unimmunized children and polio resurgence, rising from 97,000 in 2019 to 750,000 by 2023, particularly in the provinces of Zambézia, Nampula, and Tete [26]. Another contributing factor was the financial crisis triggered by the “hidden debts” scandal, in which government officials borrowed undisclosed substantial sums of money (10% of the GDP) from international banks. This led to the suspension of International Monetary Fund aid and many bilateral and multilateral donors’ support, which resulted in a reduction of health sector financial support [27].
The findings from this analysis are similar to community health surveys conducted in some Sub-Saharan African countries prior to COVID-19, which reported lower zero-dose prevalence rates. For instance, South Africa recorded a national zero-dose prevalence of 9%, Zimbabwe reported 5%, and Malawi 3% [28,29,30,31]. Post COVID-19 pandemic, similarly, the ZD prevalence increased, as in Nigeria (42%) [32] and Angola (30%) [33], once more confirming the health service provision disruption driven by the COVID-19 pandemic [13].
Children in rural areas had the highest prevalence of ZD status. This is most likely due to distance and travel time challenges to access health services, particularly in rural areas [34]. Furthermore, research conducted in rural areas of Gaza and Zambézia provinces has identified several additional barriers to accessing healthcare, including financial constraints, lack of transportation, and poor service quality [35].
Provincial disparities in ZD prevalence were most pronounced in Zambézia, Tete, and Cabo Delgado, where high poverty levels, inadequate infrastructure, and, in Cabo Delgado’s case, ongoing conflict have disrupted routine health services. These findings align with previous studies highlighting the significant impact of geographic inequities on immunization coverage in Mozambique [36]. The combination of low vaccination rates and high zero-dose prevalence in these provinces has fueled recurrent outbreaks of polio, cholera, and measles, further threatening public health [21,26]. These provinces also experience some of the highest rates of malnutrition, diarrheal diseases, malaria, inadequate healthcare services, extreme weather events, and the lingering impacts of the COVID-19 pandemic. Moreover, since 2017, political instability in Cabo Delgado, compounded by natural disasters in the central and northern regions, has led to deaths, displacement and social instability. This includes a rise in orphaned children, insufficient healthcare personnel and resources, and increasing social challenges that exacerbate health service inequalities [37]. These challenges underscore the urgent need for targeted interventions to strengthen primary healthcare services, especially among children, across the country.
In response to the increasing prevalence of zero-dose children, the Mozambique EPI launched the first round of the Zero-Dose Children Recovery Initiative in early 2024. This initiative deployed mobile vaccination brigades across 74 districts in Nampula, Niassa, Cabo Delgado, Zambézia, Tete, Manica, and Sofala provinces. The campaign focused on mapping, sensitizing, and mobilizing caregivers to bring children to health units, while mobile brigades were deployed to vaccinate children aged 0 to 59 months in underserved areas [38].
This analysis revealed that the wealth index was an important predictor of zero-dose prevalence among Mozambican children. Specifically, children from lower-income families were likely to be zero-dose compared to those from better-off families. This finding is consistent with previous studies, which have shown that the likelihood of a child being zero-dose increases as family wealth decreases [1,7,39,40]. Although vaccines are provided free of charge in Mozambique, indirect costs, such as transportation expenses and missing family or outside work, are often borne by mothers and caregivers. These costs can create significant barriers to accessing maternal and child health services, including immunization [41,42].
The number of antenatal care (ANC) visits was a significant predictor of zero-dose status. Children whose mothers had never attended ANC visits exhibited a higher prevalence of being zero-dose compared to those whose mothers attended one to three or more ANC visits. This finding is consistent with previous studies, which have shown that mothers who do not attend health facilities during pregnancy are more likely to have zero-dose children [1,43,44]. A possible explanation for this result is Mozambique’s low adherence to ANC visits, which is partially attributed to limited healthcare service coverage in rural areas and long wait times at urban facilities [45,46]. A pilot study conducted in four maternity clinics in Mozambique found that scheduling ANC appointments in advance significantly increased women’s adherence to prenatal visits, highlighting the potential of appointment systems to improve attendance [45].
Although antenatal care was not retained in the final multivariable model because of its strong overlap with other maternal healthcare utilization variables, particularly place of delivery, it remains an important marker of children’s access to preventive health services. In our study population, 96.5% of women who delivered in a health facility reported at least one antenatal care visit, compared with 76.0% of those who delivered at home, indicating that these variables capture closely related dimensions of maternal engagement with the health system. Consequently, the effect of antenatal care is likely to be reflected, at least in part, through place of delivery in the adjusted analyses. This should not be interpreted as evidence that antenatal care is unimportant; rather, it highlights the close interrelationship between maternal healthcare utilization and childhood immunization. Strengthening antenatal care services and ensuring continuity of care through facility-based delivery remain important opportunities to identify and vaccinate children at risk of becoming zero-dose.
The place of delivery was also found to be associated with zero-dose status in children. Children born at home had a higher prevalence of being zero-dose compared to those born in health facilities. This finding aligns with previous studies, which show that children born at home are more likely to remain unvaccinated than those born in healthcare settings [1,40,47]. The literature identifies several factors contributing to home births in Mozambique, including overcrowded health facilities, perceptions of inadequate care, and challenges accessing healthcare services. These barriers often discourage women from delivering in healthcare settings, thereby limiting their children’s opportunities for timely vaccination [48].
The generally weaker associations observed for the Gavi definition of zero-dose may reflect outcome misclassification relative to the WHO definition. The WHO definition incorporates all routinely recommended vaccines administered during the first year of life, whereas the Gavi definition relies solely on the absence of a first dose of a diphtheria-containing vaccine. Consequently, some children classified as non-zero-dose according to the Gavi definition may nevertheless have received no other routine vaccinations, while others may have received selected vaccines despite missing the first diphtheria-containing dose. This imperfect agreement may reduce the ability of the Gavi definition to distinguish children with genuinely different levels of access to immunization services. From an epidemiological perspective, such non-differential outcome misclassification would be expected to attenuate associations with socioeconomic and healthcare-related determinants, resulting in weaker adjusted effect estimates than those observed using the WHO definition.
While this study significantly contributes to health evidence, its findings should be interpreted considering certain limitations. Firstly, although we did use repeated cross-sectional analysis at the individual level, the analysis is still unprotected from reverse causality between immunization status and the independent variables analyzed. Additionally, data on child immunization were collected from either vaccination cards or similar documents or based on maternal or caregiver recall. This reliance on memory may be subject to recall bias or influenced by the desire to provide socially acceptable responses, which could lead to either over- or underestimation of the number of unvaccinated children. In this study, maternal or caregiver reports were treated as equivalent to data from vaccination cards, potentially affecting the accuracy of the findings. Lastly, as this study was a secondary analysis, exploring key aspects such as knowledge and practices related to vaccination was impossible. This limitation restricted the ability to gain a deeper understanding of the factors associated with zero-dose status, thereby hindering a more comprehensive analysis of the barriers to immunization.

5. Conclusions

Between 2015 and 2022, Mozambique saw a troubling threefold increase in the prevalence of zero-dose children (ZDC), with Zambézia and Cabo Delgado provinces experiencing the most significant surge. Antenatal care visits and institutional deliveries were identified as key factors associated with a reduction in ZD prevalence, emphasizing the critical role of these services in ensuring immunization uptake. Other significant factors linked to zero-dose status included birth order, possession of a vaccination card, place of delivery, province of residence, religion, and wealth quintile. These findings underscore the persistent disparities in healthcare access and service delivery, which must be addressed to effectively reduce the prevalence of ZD children.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/vaccines14110926/s1, Table S1—Prevalence of zero-dose in percentage according to Gavi operational definition (ZDGavi), Table S2—Time multiplicative changes in zero-dose prevalence using the Gavi operational definition, Table S3—Unadjusted log-binomial regressions, Table S4—Factors associated with the prevalence of zero-dose (Gavi definition), Table S5a–d—presenting contrasts in zero-dose prevalence by household wealth index for both the WHO and Gavi definitions among children aged 12–23 and 24–35 months, based on regression models including interactions between survey year and wealth quintile, Table S6a–d—present contrasts in zero-dose prevalence between rural and urban residence for the same age groups and zero-dose definitions, using regression models with interactions between survey year and place of residence, Table S7a–d—show contrasts in zero-dose prevalence according to maternal education level for both age groups and zero-dose definitions, estimated from regression models including interactions between survey year and maternal education, Table S8a–d—present multivariable regression analyses examining the associations of antenatal care visits, place of delivery, and maternal education with WHO-defined and Gavi-defined zero-dose status among children aged 12–23 and 24–35 months.

Author Contributions

Conceptualization, A.C., O.A. and E.G.; methodology: A.C., E.E.F., E.H.C., A.P.J., B.M., N.d.D. and O.A.; validation: O.A., N.d.D. and K.S.; formal analysis, A.C., E.E.F., E.H.C., A.P.J. and O.A.; investigation, A.C., E.E.F., E.H.C., A.P.J., B.M., L.N., A.J., L.S., B.B. and O.A.; resources, A.C., E.G., S.C. and O.A.; data curation, A.C. and O.A.; writing—original draft preparation, A.C., E.G., A.P.J., C.C., E.E.F., E.H.C., A.J., C.C. and L.N.; writing—review and editing, A.C., E.G., E.E.F., E.H.C., A.P.J., C.C., B.M., L.N., A.J., L.S., B.B., S.C., K.S., N.d.D. and O.A.; visualization, A.C., S.C. and K.S.; supervision, O.A. and K.S.; project administration, A.C. and E.G.; funding acquisition, A.C. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Bill & Melinda Gates Foundation [INV-008001] and the Brazilian Ministry of Health/DECIT/CNPq [Grand Challenges PALOP]. Under the grant conditions of the Foundation, a Creative Commons Attribution 4.0 Generic License has been assigned to the Author Accepted Manuscript that may result from this submission. It is also co-funded by national funds through FCT—Fundação para a Ciência e a Tecnologia, I.P., under the R&D Unit Global Health and Tropical Medicine (UID/04413/2025) and the Associate Laboratory in Translation and Innovation Towards Global Health, REAL (LA/P/0117/2020).

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki. The Demographic and Health Survey was approved by the Comité Nacional de Bioética para a Saúde (CNBS) of Mozambique, under approval reference/code 78/CNBS/2021, on 22 December 2021. Ethical review and approval were waived for this secondary analysis due to the public availability of the anonymized data.

Data Availability Statement

The data used for the analysis is publicly available on the MEASURE/DHS website.

Acknowledgments

The authors would like to thank the parents and guardians who provided survey responses. Special thanks to Augusto Nhacuoanga, the project manager, Claude Pirmez, Maria Julia Gomes de Moura, Gabriel Antonio Rezende de Paula and Júlia Assiat for their support and important discussions during the project implementation. The authors also wish to honor the memory of Marta Cassocera (in memoriam), whose vision and dedication helped initiate and promote this important movement of secondary data analysis.

Conflicts of Interest

The authors declare no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

Abbreviations

The following abbreviations are used in this manuscript:
ANCAntenatal Care
BCGBacillus Calmette–Guérin
CIConfidence Interval
CNBSComité Nacional de Bioética para a Saúde (National Bioethics Committee for Health)
COVID-19Coronavirus Disease 2019
DHSDemographic and Health Survey
DTPDiphtheria–Tetanus–Pertussis
DTP1First Dose of Diphtheria–Tetanus–Pertussis-Containing Vaccine
ENUEnumeration Area (Área de Enumeração)
EPIExpanded Program on Immunization
GDPGross Domestic Product
IMASIDAInquérito de Indicadores de Imunização, Malária e HIV/SIDA em Moçambique
IPVInjected Polio Vaccine
LICLow-Income Country
LMICLow- and Middle-Income Country
MRVMeasles and Rubella Vaccine
PAVPrograma Alargado de Vacinação (Expanded Program on Immunization)
PCVPneumococcal Conjugate Vaccine
PRPrevalence Ratio
RVRotavirus Vaccine
VIFVariance Inflation Factor
WHOWorld Health Organization
ZDZero-Dose
ZDCZero-Dose Children
ZDGaviZero-Dose According to Gavi Operational Definition
ZDWHOZero-Dose According to WHO Definition

References

  1. Eryurt, M.A.; Yalçin, S.S. Zero-dose children in Turkey: Regional comparison of pooled data for the period 1990 to 2018. BMC Infect. Dis. 2022, 22, 421. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  2. Greenwood, B. The contribution of vaccination to global health: Past, present and future. Philos. Trans. R. Soc. Lond. B Biol. Sci. 2014, 369, 20130433. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  3. Wonodi, C.; Farrenkopf, B.A. Defining the Zero Dose Child: A Comparative Analysis of Two Approaches and Their Impact on Assessing the Zero Dose Burden and Vulnerability Profiles across 82 Low- and Middle-Income Countries. Vaccines 2023, 11, 1543. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  4. Indicator Metadata Registry Details. Available online: https://www.who.int/data/gho/indicator-metadata-registry/imr-details/7792 (accessed on 10 December 2024).
  5. Cata-Preta, B.O.; Santos, T.M.; Mengistu, T.; Hogan, D.R.; Barros, A.J.D.; Victora, C.G. Zero-dose children and the immunisation cascade: Understanding immunisation pathways in low and middle-income countries. Vaccine 2021, 39, 4564–4570. [Google Scholar] [PubMed]
  6. Bosch-Capblanch, X.; Banerjee, K.; Burton, A. Unvaccinated children in years of increasing coverage: How many and who are they? Evidence from 96 low- and middle-income countries. Trop. Med. Int. Health 2012, 17, 697–710. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  7. Sato, R. Zero-Dose, Under-Immunized, and Dropout Children in Nigeria: The Trend and Its Contributing Factors over Time. Vaccines 2023, 11, 181. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  8. World Health Organization. State of the World’s Vaccines and Immunization; World Health Organization: Geneva, Switzerland, 2009; Available online: https://iris.who.int/handle/10665/44169 (accessed on 10 December 2024).
  9. Castro-Aguirre, I.E.; Alvarez, D.; Contreras, M.; Trumbo, S.P.; Mujica, O.J.; Peraza, D.S.; Velandia-González, M. The Impact of the Coronavirus Pandemic on Vaccination Coverage in Latin America and the Caribbean. Vaccines 2024, 12, 458. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  10. Evans, B.; Keiser, O.; Kaiser, L.; Jombart, T. Analysis of global routine immunisation coverage shows disruption and stagnation during the first two-years of the COVID-19 pandemic with tentative recovery in 2022. Vaccine X 2023, 15, 100383. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  11. Sato, R. Geospatial and Time Trend of Prevalence and Characteristics of Zero-Dose Children in Nigeria from 2003 to 2018. Vaccines 2022, 10, 1556. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  12. Muhoza, P. Routine Vaccination Coverage—Worldwide, 2020. MMWR Morb. Mortal. Wkly. Rep. 2021, 70, 1495–1500. [Google Scholar] [PubMed]
  13. Augusto, O.; Roberton, T.; Fernandes, Q.; Chicumbe, S.; Manhiça, I.; Tembe, S.; Wagenaar, B.H.; Anselmi, L.; Wakefield, J.; Sherr, K.; et al. Early effects of COVID-19 on maternal and child health service disruption in Mozambique. Front. Public Health 2023, 11, 1075691. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  14. Wigley, A.; Lorin, J.; Hogan, D.; Utazi, C.E.; Hagedorn, B.; Dansereau, E.; Tatem, A.J.; Tejedor-Garavito, N. Estimates of the number and distribution of zero-dose and under-immunised children across remote-rural, urban, and conflict-affected settings in low and middle-income countries. PLoS Glob. Public Health 2022, 2, e0001126. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  15. The DHS Program—Available Datasets. Available online: https://dhsprogram.com/data/available-datasets.cfm (accessed on 12 April 2025).
  16. Population Pyramids of the World from 1950 to 2100. PopulationPyramid.net. Available online: https://www.populationpyramid.net/world/2024/ (accessed on 18 January 2025).
  17. IDS Relatório de Indicadores-Chave, INE. Available online: https://ine.gov.mz/web/guest/d/ids-relatorio-de-indicadores-chave (accessed on 18 January 2025).
  18. Chissaque, A.; Guimarães, E.; Limaymanta, C.H.; Conjo, C.; Capece, B.P.S.; Gonçalves, L.; de Deus, N.; Craveiro, I. Post-independence health research productivity in Portuguese-speaking African countries: A bibliometric analysis of 43 years of research and higher education in Mozambique. Heliyon 2024, 10, e35767. [Google Scholar] [PubMed]
  19. World Bank Open Data. Available online: https://data.worldbank.org (accessed on 18 January 2025).
  20. Human Development Report 2020|United Nations iLibrary. Available online: https://www.un-ilibrary.org/content/books/9789210055161 (accessed on 18 January 2025).
  21. Cassocera, M.; Chissaque, A.; Martins, M.R.O.; de Deus, N. 40 years of immunization in Mozambique: A narrative review of literature, accomplishments, and perspectives. Cad. Saúde Pública 2020, 36, e00038320. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  22. Bergen, N.; Cata-Preta, B.O.; Schlotheuber, A.; Santos, T.M.; Danovaro-Holliday, M.C.; Mengistu, T.; Sodha, S.V.; Hogan, D.R.; Barros, A.J.D.; Hosseinpoor, A.R. Economic-Related Inequalities in Zero-Dose Children: A Study of Non-Receipt of Diphtheria-Tetanus-Pertussis Immunization Using Household Health Survey Data from 89 Low- and Middle-Income Countries. Vaccines 2022, 10, 633. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  23. Skov, T.; Deddens, J.; Petersen, M.R.; Endahl, L. Prevalence proportion ratios: Estimation and hypothesis testing. Int. J. Epidemiol. 1998, 27, 91–95. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  24. The R Project for Statistical Computing. Available online: https://www.r-project.org/ (accessed on 18 January 2025).
  25. Lumley, T. Complex Surveys: A Guide to Analysis Using R; John Wiley & Sons, Inc.: Hoboken, NJ, USA, 2010. [Google Scholar]
  26. Mozambique’s Grand Plan to Reach More than 750,000 Unimmunised Children in 2024. Available online: https://www.gavi.org/vaccineswork/mozambiques-grand-plan-reach-over-750000-unimmunised-children-2024 (accessed on 11 January 2025).
  27. Centro de Integridade Pública (CIP). Custos e Consequencias Das Dividas Ocultas Para Mocambique; Centro de Integridade Publica and Chr. Michelsen Institute: Maputo, Mocambique, 2025; Available online: https://macua.blogs.com/files/cip-custos-e-consequencias-das-dividas-ocultas.pdf (accessed on 10 January 2025).
  28. The DHS Program—Zimbabwe: DHS, 2015—Final Report (English). Available online: https://dhsprogram.com/publications/publication-fr322-dhs-final-reports.cfm (accessed on 18 January 2025).
  29. Office/Malawi, NS, ICF Malawi Demographic and Health Survey 2015–16. 2017. Available online: https://dhsprogram.com/publications/publication-fr319-dhs-final-reports.cfm (accessed on 18 January 2025).
  30. The DHS Program—South Africa: Standard DHS, 2016. Available online: https://dhsprogram.com/methodology/survey/survey-display-390.cfm (accessed on 19 January 2025).
  31. Cata-Preta, B.O.; Santos, T.M.; Wendt, A.; Hogan, D.R.; Mengistu, T.; Barros, A.J.D.; Victora, C.G. Ethnic disparities in immunisation: Analyses of zero-dose prevalence in 64 countries. BMJ Glob. Health 2022, 7, e008833. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  32. ZDLH_Nigeria_Situation_Analysis_2023.pdf. Available online: https://zdlh.gavi.org/sites/default/files/2023-12/ZDLH_Nigeria_Situation_Analysis_2023.pdf (accessed on 19 January 2025).
  33. Zero-Dose Children in Angola Are Being Protected Through a Nationwide Vaccination Campaign|UNICEF. Available online: https://www.unicef.org/stories/zero-dose-children-in-angola-protected-through-vaccination-campaign (accessed on 19 January 2025).
  34. dos Anjos Luis, A.; Cabral, P. Geographic accessibility to primary healthcare centers in Mozambique. Int. J. Equity Health 2016, 15, 173. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  35. Schwitters, A.; Lederer, P.; Zilversmit, L.; Gudo, P.S.; Ramiro, I.; Cumba, L.; Mahagaja, E.; Jobarteh, K. Barriers to Health Care in Rural Mozambique: A Rapid Ethnographic Assessment of Planned Mobile Health Clinics for ART. Glob. Health Sci. Pract. 2015, 3, 109–116. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  36. Egger, E.-M.; Salvucci, V.; Tarp, F. Evolution of Multidimensional Poverty in Crisis-Ridden Mozambique. Soc. Indic. Res. 2023, 166, 485–519. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  37. When Debt and Terrorism Intersect: The Case of Mozambique|ISS Africa. Available online: https://issafrica.org/iss-today/when-debt-and-terrorism-intersect-the-case-of-mozambique (accessed on 20 January 2025).
  38. OMS Apoiou a Recuperação de Cerca de 860 Mil Crianças Com Zero Dose e Sub-Imunizadas em 74 Distritos de Moçambique|OMS|Escritório Regional Para a África. 2025. Available online: https://www.afro.who.int/pt/photo-story/oms-apoiou-recuperacao-de-cerca-de-860-mil-criancas-com-zero-dose-e-sub-imunizadas-em (accessed on 20 January 2025).
  39. Acharya, P.; Kismul, H.; Mapatano, M.A.; Hatløy, A. Individual- and community-level determinants of child immunization in the Democratic Republic of Congo: A multilevel analysis. PLoS ONE 2018, 13, e0202742. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  40. Johri, M.; Rajpal, S.; Subramanian, S.V. Progress in reaching unvaccinated (zero-dose) children in India, 1992–2016: A multilevel, geospatial analysis of repeated cross-sectional surveys. Lancet Glob. Health 2021, 9, e1697–706. [Google Scholar] [CrossRef] [Scilit]
  41. Munguambe, K.; Boene, H.; Vidler, M.; Bique, C.; Sawchuck, D.; Firoz, T.; Makanga, P.T.; Qureshi, R.; Macete, E.; Menéndez, C.; et al. Barriers and facilitators to health care seeking behaviours in pregnancy in rural communities of southern Mozambique. Reprod. Health 2016, 13, 83–97. [Google Scholar] [CrossRef] [Scilit]
  42. Child Poverty in Mozambique—Multiple Overlapping Deprivation Analysis|UNICEF. 2019. Available online: https://www.unicef.org/mozambique/en/reports/child-poverty-mozambique-multiple-overlapping-deprivation-analysis (accessed on 20 January 2025).
  43. Farrenkopf, B.A.; Zhou, X.; Shet, A.; Olayinka, F.; Carr, K.; Patenaude, B.; Chido-Amajuoyi, O.G.; Wonodi, C. Understanding household-level risk factors for zero dose immunization in 82 low- and middle-income countries. PLoS ONE 2023, 18, e0287459. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  44. Mohamoud, S.A.; Ali-Salad, M.A.; Bile, A.S.; Singh, N.S.; Mahmud, A.J.; Nor, B. Determinants and prevalence of zero-dose children in Somalia: Analysis of the 2020 Health Demographic Survey data. PLoS Glob. Public Health 2024, 4, e0002612. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  45. Steenland, M.; Dula, J.; de Albuquerque, A.; Fernandes, Q.; Cuco, R.M.; Chicumbe, S.; Gudo, E.S.; Sequeira, S.; McConnell, M. Effects of appointment scheduling on waiting time and utilisation of antenatal care in Mozambique. BMJ Glob. Health 2019, 4, e001788. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  46. Caregivers in Mozambique Share the Barriers They Face in Vaccinating Their Children. Available online: https://www.gavi.org/vaccineswork/caregivers-mozambique-share-barriers-they-face-vaccinating-their-children (accessed on 20 January 2025).
  47. Nchinjoh, S.C.; Saidu, Y.; Agbor, V.N.; Mbanga, C.M.; Muteh, N.J.; Njoh, A.A.; Ndoula, S.T.; Nsah, B.; Edwige, N.N.; Roberman, S.; et al. Factors Associated with Zero-Dose Childhood Vaccination Status in a Remote Fishing Community in Cameroon: A Cross-Sectional Analytical Study. Vaccines 2022, 10, 2052. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  48. Long, Q.; Madede, T.; Parkkali, S.; Chavane, L.; Sundby, J.; Hemminki, E. Maternity care system in Maputo, Mozambique: Plans and practice? Cogent Med. 2017, 4, 1412138. [Google Scholar] [CrossRef] [Scilit]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Article Metrics

Citations

Article Access Statistics

Journal Statistics
Multiple requests from the same IP address are counted as one view.
Post-Campaign Evaluation of Diphtheria Vaccination Coverage in Somalia Following the 2025–2026 Outbreak Response
Previous
Calcium-Containing Squalene Emulsion as a Novel Vaccine Adjuvant Delivery System
Next