Introduction

Measles remains one of the most contagious vaccine-preventable diseases and continues to cause substantial morbidity and mortality worldwide. In 2022 alone, measles caused approximately 128,000 deaths globally, the vast majority among unvaccinated children in low- and middle-income countries (LMICs) [1]. Despite progress, 14.3 million children globally were classified as “zero-dose” in 2024—that is, they had not received any dose of a routine vaccine. This group remains a major driver of measles transmission through clustering in under-immunised communities [2,3]. Outbreaks in both LMICs and high-income countries underscore the continued threat posed by declining or stagnant immunisation coverage [4].

Zambia is no exception to these challenges. Between 2021 and 2024, the number of measles zero-dose children in Zambia rose from approximately 39,000–41,000, ranking the country among the top 12 in the Eastern and Southern Africa Region (ESAR) [5]. Nationally, measles elimination requires sustained ≥95% coverage with two doses of the measles-rubella (MR) vaccine, as outlined in the World Health Organization’s (WHO) Immunization Agenda 2030 (IA2030) [6]. However, routine immunisation coverage has stagnated, and inequities persist, especially in rural and socioeconomically disadvantaged communities. These gaps threaten Zambia’s ability to achieve measles elimination goals and mirror global concerns about stagnating immunisation progress [4,7].

Several studies have identified risk factors for zero-dose status, including rural residence, maternal absence or orphanhood, caregiver education level, and distance to health facilities [8–10]. However, many analyses rely on secondary datasets, lack disaggregated campaign-specific information, or fail to capture behavioural and social determinants of vaccine uptake. This limits the ability of policymakers to design context-specific interventions to close immunisation gaps. Locally generated evidence is especially critical as both Gavi’s equity agenda and IA2030 prioritise reaching zero-dose children as a global public health imperative [6,11].

In Zambia, the 2024 Measles–Rubella Supplementary Immunisation Activity (SIA) was reported to have reached 97% coverage administratively, vaccinating 165,000 zero-dose children through targeted microplanning, community engagement, and digital monitoring [12]. Yet, administrative coverage often overestimates true performance. Independent evaluation is necessary to validate coverage, quantify remaining zero-dose prevalence, and identify risk factors limiting access and uptake. To address these gaps, a Post-Campaign Coverage Survey (PCCS) was conducted following Zambia’s 2024 MR SIA. The survey was designed to provide an independent estimate of vaccination coverage among children aged 9–59 months, offering a validated measure beyond routine administrative reports. It also aimed to quantify the prevalence of zero-dose and under-immunised children, assess their overlap with RI performance, and evaluate the proportion reached during the campaign. In addition, the PCCS examined socio-demographic, geographic, and awareness-related factors associated with non-vaccination, generating insights into barriers to uptake and informing strategies to reduce missed opportunities for immunisation. By generating robust and nationally representative evidence, this study provides critical insights for strengthening measles control in Zambia and contributes to the global evidence base on strategies to reduce zero-dose children in LMICs.

Methods

Results

Zero-dose prevalence

Nationally, 11.97% (95% CI: 11.03–12.91%) of children were MR zero-dose, of whom 88.75% had not received Diphtheria, Pertussis and Tetanus first dose (DPT1) or other antigens, indicating systemic immunization gaps (Table 1, S2 Table). Prevalence was highest in Central (19.15%) and Western (17.71%) provinces, and lowest in Copperbelt (6.69%) (Fig 1). Rural prevalence (13.02%) exceeded urban (10.17%), with an adjusted odds ratio (OR) for urban residence of 0.76 (95% CI: 0.63–0.92), indicating higher odds of zero-dose status in rural settings (S1 Table). Children in large households (≥9 members) had a higher MR zero-dose prevalence (14.18%; 95% CI: 11.63–16.73%) compared to those in households with 1–4 members (12.29%; 95% CI: 10.72–13.86%) or 5–8 members (11.17%; 95% CI: 9.88–12.46%), though this association was not statistically significant (S1 Table). Stratified analysis showed similar patterns in both rural (≥9: 14.41%) and urban settings (≥9: 13.41%), confirming that the household size effect was consistent but not significant across residence types.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Key findings on MR zero-dose prevalence and adjusted associations, Zambia PCCS 2024 (n = 8,634).

https://doi.org/10.1371/journal.pgph.0005265.t001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Weighted zero-dose measles-rubella prevalence by province, Zambia, 2024 PCCS.

Exact values are in S1 Table.

https://doi.org/10.1371/journal.pgph.0005265.g001

Demographic risk factors

Measles-rubella (MR) zero-dose prevalence increased with age, peaking at 14.99% among children >36 months (OR = 1.60; 95% CI: 1.27–2.00) (Fig 2). No significant sex differences were observed between males (11.90%; 95% CI: 10.92–12.88%) and females (12.04%; 95% CI: 11.04–13.04%), with an adjusted odds ratio of 1.05 (95% CI: 0.89–1.23; p = 0.533) (S1 Table). Children whose mothers were absent (19.73%; OR = 1.74, 95% CI: 1.33–2.27) or deceased (24.17%; OR = 2.40, 95% CI: 1.23–4.68) had increased odds of being MR zero-dose (S1 Table). A small group with “unknown” maternal status showed zero vaccination (OR = 0.00), but this reflects unstable estimates due to very small sample size. Guardian education (e.g., Primary: 11.33%; 95% CI: 9.82–12.84%) and religion (e.g., Protestant: 12.50%; 95% CI: 11.44–13.56%) showed no statistically significant associations (S2 Table). Similarly, socioeconomic status (SES), derived from occupation, did not show a consistent pattern across low, medium, and high categories, and associations were not statistically significant. Muslim prevalence was unreliable (0.00%) due to small sample size (S2 Table).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Weighted percentage of MR zero-dose children by age and gender, Zambia, 2024 PCCS.

https://doi.org/10.1371/journal.pgph.0005265.g002

Barriers to vaccination

Lack of awareness of the campaign (42.58%) was the most cited reason for non-vaccination, followed by being too busy (19.02%) and inconvenient timing (7.69%) (Fig 3). Travel time of 1–2 hours (OR = 1.55; 95% CI: 1.20–2.00) and over 2 hours (OR = 3.20; 95% CI: 1.43–7.16) were associated with significantly higher odds of being MR zero-dose (S2 Table). The regression model also included travel means (mode of transport to the facility), but no significant associations were observed. CIs for >2 hours were wide due to small sample sizes.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Weighted reasons for non-receipt of measles-rubella vaccine among zero-dose children, Zambia, 2024 PCCS.

https://doi.org/10.1371/journal.pgph.0005265.g003

Discussion

Zambia’s national MR zero-dose prevalence of 11.97%, with high-burden provinces such as Central (19.15%) and Western (17.71%), reflects persistent geographic inequities and structural barriers, including limited access, low campaign awareness, and social vulnerabilities like maternal absence, common in LMICs. This analysis highlights three critical findings: first, clustering of zero-dose children in specific provinces and among older children; second, the dominance of systemic access barriers such as distance and awareness; and third, the overlap of MR zero-dose status with complete lack of other vaccines (88.75%), signalling service delivery failures across the immunisation system rather than isolated campaign shortfalls.

In regional context, Zambia’s 2024 burden (~41,000 zero-dose children) places it mid-range, higher than countries such as Botswana or Eswatini, but below large-burden nations like Ethiopia and Angola [5]. This comparative framing underscores the need for targeted but feasible interventions to reach IA2030 zero-dose reduction goals.

Older children and those with absent or deceased mothers were significantly more likely to be zero-dose, consistent with global evidence linking maternal absence/orphanhood to vaccination gaps [8]. These associations may reflect mechanisms such as reduced caregiver decision-making capacity, opportunity costs during agricultural seasons, or limited household support structures. By contrast, socioeconomic status and caregiver education were not significant, suggesting that structural determinants such as distance, service timing, and awareness are stronger drivers than demand-side constraints in this setting.

The co-occurrence of zero-dose MR with absence of other antigens emphasises that Zambia’s challenge lies in RI system failures. Weak microplanning, poor integration between SIAs and RI, and inconsistent card retention limit opportunities for catch-up. This aligns with global analyses indicating that SIAs often overestimate their independent contribution when RI remains weak [10,11].

Importantly, our interpretation of COVID-19–related disruptions requires clarification. Evidence shows that measles vaccination was strongly protective, with vaccinated children having markedly lower mortality (aOR ~0.06 compared to unvaccinated [15]) Zambia’s pandemic-era gaps therefore exacerbated measles outcomes by reducing coverage, not by conferring protection to unvaccinated children.

Access barriers were central with children living >2 hours from a facility had a prevalence of 26% versus 10% within 15 minutes representing ~16 percentage-point difference (aOR 3.20). Awareness gaps also remain striking, with 42.6% of caregivers of zero-dose children reporting they had not heard of the campaign. Together, these patterns suggest that reducing travel time and improving health worker–led communication could yield greater gains than socioeconomic-targeted messaging alone. Similar patterns of access-related and caregiver-related determinants of under-vaccination have been documented in India, Ethiopia, and the Democratic Republic of the Congo, reinforcing the generalisability of these findings across diverse LMIC settings [16–21].

Generalisability is high for LMIC settings with dispersed rural populations, weak outreach systems, and heavy reliance on SIAs. In dense urban contexts with abundant fixed sites, demand-side determinants (e.g., confidence, complacency) may be more salient, suggesting that Zambia’s lessons apply most strongly to rural, outreach-dependent systems.

This study makes three novel contributions: (i) it quantifies the overlap of MR zero-dose with all-antigen zero-dose nationally, (ii) it demonstrates the primacy of access and awareness barriers after adjusting for stratification, and (iii) it identifies older children and those without maternal caregivers as particularly high-risk groups requiring tailored catch-up strategies.

To close the remaining immunisation gaps, Zambia should prioritise strengthened, proactive communication led by health workers and tailored to local contexts, particularly in high-burden areas where awareness remains low, as supported by recent evidence showing that targeted health-worker–driven communication significantly improves uptake [22]. Furthermore, expanding proximity-based delivery strategies, such as regular mobile and outreach sessions, would reduce long travel times and improve accessibility, consistent with studies demonstrating the effectiveness of mobile teams in tracing unvaccinated children and increasing RI coverage [23]. In parallel, closer integration of Supplementary Immunisation Activities (SIAs) with routine immunisation (RI) is essential, ensuring that all SIA posts implement defaulter tracing and provide on-site catch-up for routine antigens, supported by improved card verification. Additionally, adopting geospatial microplanning approaches to generate and update hotspot maps quarterly would facilitate more precise targeting of older children and households without maternal caregivers, aligning with findings from recent geospatial analyses showing the value of high-resolution mapping for identifying zero-dose clusters [24,25]. Digital tools, including mobile applications for defaulter tracking and AI-supported hotspot identification, can also enhance microplanning efficiency, building on evidence from both Zambia and other LMICs that digital health tools increase coverage and strengthen planning [26], as well as Nigeria’s documented success in reducing zero-dose gaps [27]. Moreover, adapting Zambia’s cholera control coordination model to measles, leveraging community engagement and strengthened surveillance, could further improve RI performance in rural hotspots [28]. Collectively, embedding these equity-focused reforms within national immunisation strategies would enable SIAs to serve as more effective platforms for reinforcing RI, addressing systemic gaps, and accelerating progress toward measles elimination.

Limitations

The study has several limitations. Although 88.3% of vaccination status assessments relied on caregiver recall, several mitigation strategies were used to reduce recall bias. Interviewers followed WHO-standardised probing procedures, captured card images where available, and used event-anchoring to contextualise timing. These steps minimise but do not eliminate recall bias; therefore, the direction of potential misclassification is uncertain. However, the approach reflects real-world immunisation documentation challenges in Zambia. Small sample sizes in some subgroups, particularly for deceased mothers and certain provinces, limit the precision of stratified estimates. Additionally, district-level variations may be masked by provincial-level reporting, potentially obscuring important local patterns that could inform targeted interventions.

Conclusion

This study set out to estimate the burden of measles–rubella zero-dose children following the 2024 supplementary immunisation activity and to identify the key factors associated with non-vaccination. The survey showed that a substantial group of children remained unreached, and these children were disproportionately concentrated in rural settings, specific high-burden provinces, older age groups, and households without maternal caregivers. The analysis also highlighted the influence of structural barriers, particularly limited awareness of the campaign and long travel distances to vaccination points, which undermined uptake despite the national effort.

Overall, the findings demonstrate that achieving measles control in Zambia requires strategies that strengthen RI while directly addressing the contextual barriers that limit access and awareness. Approaches centred on community engagement, improved communication, and enhanced service reach, especially for hard-to-access populations, will be essential for reducing zero-dose children and advancing progress toward national and global measles elimination goals.

Supporting information

Acknowledgments

The authors extend their sincere gratitude to the staff of the Zambia Ministry of Health, WHO Zambia Country Office, UNICEF Zambia, and the Zambia Statistics Agency (ZAMSTATS) for their invaluable technical and logistical support throughout the 2024 Post-Campaign Coverage Survey. We also acknowledge the dedication of the data collectors and community health workers whose efforts made this study possible.